date
stringlengths 10
10
| nb_tokens
int64 60
629k
| text_size
int64 234
1.02M
| content
stringlengths 234
1.02M
|
|---|---|---|---|
2023/01/30
| 1,089
| 4,902
|
<issue_start>username_0: I started my PhD with my present supervisor a couple of years back. Her profile looked interesting. However, suddenly her previous work started receiving comments on publicly available domains (pubpeer). It appears that almost 14 of her papers that came out of her lab are under scrutiny. Moreover, one of the papers has already received a retraction.
Though none of these papers have my name in them, my concern is that the profile of my supervisor could affect my career in the long run. Is it a wise idea to switch labs before it gets too late for me? In the worst case scenario, if I continue to work under her, how badly can this affect my profile?<issue_comment>username_1: This is a judgment call only you can make. You should think about more things than whether association with your advisor will taint your future.
Can you accurately evaluate the public criticism of her previous work? Do the complaints ring true to you? Do they seem more like professional jealousy? Will you be comfortable working with her?
If you are nearly done and the actual results in your thesis have been honestly attained and clearly not subject to any of the issues raised by the public scrutiny of your advisor's other work, then finish and move on.
If you are just beginning your own work and can imagine a better place to do it, perhaps at a loss of a year or two, then consider looking for such a place.
Upvotes: 3 <issue_comment>username_2: Science operates on a certain level of trust. While principles like transparency about data and detailed, replicable descriptions of methodology can help build reliability into science, you cannot bypass a need for trust because even there you must trust that the data presented are the real, unaltered data, and that the methodology was actually performed as stated. There are rarely enough resources available to check and re-check every claim made.
There are a few ways in which fraud can be revealed, though. Image manipulation is one example. Other examples are statistical (ir)regularities in data that suggest manipulation or fabrication, or p-values or test statistics that are impossible given certain structural aspects of the original data, like points on an integer scale.
Most other types of fabrication are impossible to extract from the published record, though. We all just have to trust. For me, that means that I would *definitely* extend suspicions of someone known to have a past record far beyond their known problem papers. I cannot comprehend the current hesitancy of journals and institutions when clear fraud is present in published papers, the sort raised on PubPeer and in particular by people like [<NAME>](https://twitter.com/MicrobiomDigest) who are dedicated to uncovering discoverable fraud.
The issues these folks raise are not issues of scientific controversy or accidental errors, but in most cases are clear evidence of intentional manipulation. That is, there is no plausible explanation besides fraud. Even if the motivation is claimed to approach innocence like "well, this is representative of what we saw under the microscope, I just did it to make a nicer picture for the journal" or "yeah but the main results of the paper are from a quantitative analysis so these images we photoshopped to show the result don't actually impact the results".
You'll have to decide what level of misconduct has occurred, if any, and act accordingly. If it were me, I'd be gathering up the available evidence and walking directly to my advisor's office and confronting them. At that point, I wouldn't be concerned with our relationship: if there are credible accusations that they cannot defend in a meeting, I would not feel comfortable having a professional relationship with them.
That said, I mentioned that journals and institutions have been hesitant, especially when involving powerful or notable people (e.g., Nobel prize winners and directors of large research programs). Punishments, when present, have often been silly, like brief (1-2 year) bans on research funding, or a stern warning along with some retractions. These punishments do not fit the offenses committed. I hope this will change in the future, but in the meantime, it sadly appears that a profitable academic strategy is to have one of these fraudsters promote your work within the community of people that for some reason still trust them. Maybe you'll get hired to work with their previous advisor who trained them to cheat, or with their former student who also benefited from some deceptively beautiful papers in fancy journals. I wouldn't feel comfortable with that, but I also recognize that blowing up your current PhD training to go elsewhere is going to cause you harm in the short term (if nothing else, by some lost time) and that your actions alone will not be sufficient to change the system overnight.
Upvotes: 3
|
2023/01/30
| 1,971
| 8,288
|
<issue_start>username_0: I don't understand the reception that [this question](https://academia.stackexchange.com/q/192977/152405) has gotten.
What I understand is, due to the dating app, there is an influx of male students in classes who become the top %, and they attract all the female students there either through the pretext of homework help, or, general peacocking of intelligence.
The issue I have is, wouldn't it be the same in any class that there are some students who excel and rise in academic hierarchy compared to others? And wouldn't these students be attractive to other academically inclined students anyways?
And ultimately, whatever happens, there is a limit to number of people who can sit in class/ a person can interact with, so shouldn't it all cancel out in the end?<issue_comment>username_1: Almost no way of meeting prospective partners is inherently unethical. It will always depend on how you go about "meeting/attracting" the prospective partners.
The unethical aspect comes into play if the way you go about meeting new people makes them **or others around them that you do not want to attract** uncomfortable, objecified, treated with sexism, etc.
That is what has been happening in the case in question. Behaving like this is shitty in no matter which environment, but when you do this at a place of education (where in a worst case scenario people have even payed a lot of money to learn) and you are not only harrassing others but in doing so distract them from learning or even worse might drive them to drop the course subject altogether, you are causing even more harm.
From your profile I gather that you are a man, and it is often hard to grasp for men that unwanted attention is neither flattering nor otherwise else perceived positively or wanted by most women\*.
Upvotes: 3 <issue_comment>username_2: It's unethical if it is predatory. It is unethical if you become a distraction to serious students. It is unethical if you affect the "curve" lowering the grade of others, though the instructor shares the blame for that.
But beyond the ethics, it is creepy. Do you want a reputation as a creep?
And beyond that, it is suboptimal for your own education, trading grades for learning. You are not taking advantage of opportunities that might advance your career, which has some ethical concerns as well. It is an example of "Self Defeating Behavior". Someone looking at your transcript would, perhaps, wonder why you were taking such low level courses as an advanced student. How will you answer?
Upvotes: 1 <issue_comment>username_3: You ask for an ethical perspective, so here is one.
Attending a university course with the sole intention to flirt with classmates is considered unethical by many people because it is an attempt to manipulate the classmates. A situation, which should occur in a learning environment, is manipulated into taking place in a dating environment, and even worse, the "flirter" is putting themselves into a position of power by using their advance in knowledge when they offer "private tutoring" to attractive students as indicated in the linked question.
<NAME> wrote:
>
> So act that you use humanity, whether in your own person or in the person of any other, always at the same time as an end, never merely as a means.
>
>
>
which can be reformulated to "rational beings must always also be treated as ends themselves, requiring that their own reasoned motives must be equally respected" ([Wikipedia](https://en.wikipedia.org/wiki/Kantian_ethics)). By manipulating a university course into a dating platform, one clearly disrespects the own reasoned motives by the other students (i.e. to study within a learning environment), and therefore this is considered unethical.
Upvotes: 3 <issue_comment>username_4: It's unethical because it's happening at a lecture/tutorial, i.e. a place of learning. Students attend lectures/tutorials to learn, not to meet partners (although that can be a side effect). Getting suggestive comments at a lecture or tutorial is very distracting from the original goal, and because it is also unrelated to that goal, it is unwelcome.
If you use a dating app or go to your local bar to meet people, and can safely say "I got really good grades in this course", more power to you. You could reverse the above scenario - instead of meeting partners as a side effect, maybe you'll get to teach a more junior student as a side effect. But the key point remains: the people you meet this way are interested in developing the relationship. The same cannot be said for most students at a lecture/tutorial.
As for:
>
> The issue I have is, wouldn't it be the same in any class that there are some students who excel and rise in academic hierarchy compared to others? And wouldn't these students be attractive to other academically inclined students anyways?
>
>
>
Yes - it's obvious who the good students are, and good students tend to flock together. But the goal of these students when they get together remains to do well in the course, not to become life partners (although that can be a side effect).
Upvotes: 1 <issue_comment>username_5: I started this as a comment to @sursula's answer, but its too long. However, I don't want to be seen as taking away from their experience and advice.
---
In a situation where it is known that there are a large number of individuals in a class who are there with the express purpose of finding younger women to impress with the aim of dating, then any woman will have to treat any interaction with a man connected with that class through that lens:
What is the motivation this person talking to me? Do I need to be sending "i'm not interested" signals to them? Are they really interested us helping each other study, or do they just want an opportunity to be spend time alone with me? Do they really want to help me, or do they just want to demonstrate how clever they are? This takes mental and emotional energy away from the class. Even if they are attracted to the person in question, even having to ask the question to themselves in circumstances where that sort of thing wasn't their purpose detracts from the main purpose.
To an extent women (and all of us) have to have these things in the back of our minds on any interaction with anyone. But when it is known that a large number of individuals are their only for the purpose of dating, then hypervigilance is required.
Coming back to the point @sursula made in the comments to their post: one such interaction is not on its own a problem, but when there are repeated such interactions even if (particularly if) they come from different people each time, this can be a major problem. That is even if no individual's behaviour is harassment, the collective effect on all the individuals on the recipient amounts to harassment. They are harassed, even if no individual is harassing them.
Furthermore, the type of showboating/showing off behaviour described in the question is damaging to the learning environment. For example, there are many reasons I might want to ask the class a question. I might want to gauge the level of understanding in the class of something. For I might be encouraging a particular train of thought. I might want to enter into a conversation with a student that doesn't know the answer where we reason our way together to the answer, thus demonstrating to them, and the rest of the class, the thought processes involved. None of these cases is helped by a smart-alec too senior for the class who just knows the answer.
Upvotes: 2 <issue_comment>username_6: There may be different opinions on whether an individual taking an easy course with the purpose of impressing members of the other sex is ethical or whether it is a waste of time.
But what surely is unethical is advertising someone else's course as a place to hit on women. It's not their own course so they have no business advertising a change in its purpose. It is negative publicity and will ruin the course experience for most of the women, and convince others not to take it. It is really surprising that the university does not do anything against it. The university is effectively endorsing the fact that some of their courses are turned into dating events.
Upvotes: 1
|
2023/01/30
| 892
| 4,016
|
<issue_start>username_0: My university has recently moved to a new location. Last semester, we were informed the new campus would be ready come start of this spring semester. The semester has just began, and we have been informed only half of the campus will be open. Upon visiting the campus, it is a mess. The entire campus is a construction site, windows are shattered, and elevators are breaking mid-use. The worst part of this situation is that today I was informed there will be a 2-week delay minimum for labs to start and that it will potentially be longer. There was no request for the students to agree to this, no tuition reimbursement or anything of this kind. I am wondering if this can be considered fraud? The course is officially listed to last the entire semester, and the full tuition is applied as such. Do I have any recourse? For context, this is a U.S. university, and the course I am mentioning is solely a lab course. This is also a mandated course this semester.<issue_comment>username_1: You are asking a legal question for which we are not qualified to give a binding answer. However, I doubt that it would be considered fraud in the US if they have made a "good faith" effort to provide needed services. Fraud normally implies intent to defraud, which is unlikely here. Building programs are difficult to do on a strict schedule and often overrun due to things like weather and pandemics and supply chain issues.
A lawyer might also ask whether you, yourself, did due diligence before you signed up for a course under these conditions.
Instead of thinking of it as a legal question, try to find out what they are willing to do to keep you on track for your degree. That is the important thing. Go talk to the department head about your needs and how they can be met.
Upvotes: 4 <issue_comment>username_2: There are a few issues you should consider before going all the way to an allegation of fraud. Firstly, most universities allow you to drop your courses without incurring the fees for the course; this is usually allowed for a limited period even after you have started the course. Such an action would allow you to cancel one or all of your courses now that you know that the quality of service is below what you were expecting. Secondly, even if your university does not have a mechanism for dropping courses without financial penalty, at worst this would constitute a breach of contract; unless there is some compelling evidence of an intention to defraud you, a mere failure to provide a service at an expected level of quality is unlikely to amount to fraud.
In regard to the legal situation here, this is a case where a vendor has sold a service but now cannot deliver the service at the expected quality. They most likely have mechanisms available to give you a refund for individual courses if you don't wish to continue with those courses at this time. An argument could be made for a broader breach of contract (and perhaps some non-criminal breaches of commercial law) in relation to your enrolment in the degree program, if they are now unable to supply courses for that program at an appropriate level of quality, to allow you to complete your degree. If you wish to pursue this beyond merely cancelling your present enrolment in courses, you could speak to a lawyer to see what kind of case you would have for breach of contract.
Rather than jumping straight to an extremely serious legal allegation, I recommend you step back and ask yourself what outcome you want to achieve in this situation. You could decide to pause your progression in your degree program and do something else for a year while they finish their construction work. You could decide to cancel your enrolment and transfer to another university at the next intake. You could decide to grin-and-bear the construction noise and other inconveniences and push through your program regardless. Determine the outcome you want and then seek that outcome in the most co-operative way you can.
Upvotes: 3
|
2023/01/31
| 539
| 2,162
|
<issue_start>username_0: Here's an example of the kind of paper I'm thinking of: [Quantum gravity phenomenology at the dawn of the multi-messenger era—A review](https://www.sciencedirect.com/science/article/pii/S0146641022000096?via%3Dihub). The scope of the review is grand (1578 references cited) and pretty much every expert in the field is an author (100+ authors total).
When every expert in the field is an author, who peer reviews the article?<issue_comment>username_1: As this has still no answer here are a few possible explanations:
* One logical answer would be that the authors suggested potential
peer-reviewer in their submission letter because they would know
which other experts exist in the field that are (for whatever reason) not part of this consortium.
* If a paper is written by almost all the top experts of the field than a high quality of the paper is already guaranteed as they will kind of "review" each other anyway as nobody wants to put their established name on a paper that they do not fully agree with.
* In rare cases top level papers are only reviewed by the editors. I
think that this [happened](https://www.nature.com/articles/427584c) for example for the DNA structure paper.
* As mentioned in the comments this post might also help: [How does peer-review work for "gigantic" collaboration projects?](https://academia.stackexchange.com/q/63659/17254)
Upvotes: 2 <issue_comment>username_2: This seems to be a hypothetical scenario. In reality there are always experts that are not on the paper.
Take the example posted in the OP. This is a typical review paper produced by a COST action, an EU funding schemes aimed at building international collaborations. As such, the author list is very EU centric. I can easily think of a few names of US based scientists that could referee this particular paper.
I posit that this will generally be the case. There will always be experts that were not part of this group effort. Sometimes because of geographical reasons, sometimes because of sociological reasons, and sometimes because some people just actively avoid this type of large group activity.
Upvotes: 3 [selected_answer]
|
2023/01/31
| 616
| 2,562
|
<issue_start>username_0: My original PhD plan was to do a pilot study, followed by a main study based on the results of the pilot study. However, the complexity of the project is growing, and I've realised that I probably can't complete the PhD within my allotted 3 years if I do both.
The pilot study by itself would be worth at least two, maybe three papers. I believe it would provide enough novel data to earn a PhD. But I'm curious - have you heard of people completing their PhD based only on a pilot study in Psychology?
Also, is this a bad strategy for getting an academic job post-PhD? I already have a few publications under my belt (including a journal article from the literature review for this PhD), but not much of a funding record at this stage. Will the lack of a main study affect the perception of my PhD's quality?
Thanks a lot!<issue_comment>username_1: As this has still no answer here are a few possible explanations:
* One logical answer would be that the authors suggested potential
peer-reviewer in their submission letter because they would know
which other experts exist in the field that are (for whatever reason) not part of this consortium.
* If a paper is written by almost all the top experts of the field than a high quality of the paper is already guaranteed as they will kind of "review" each other anyway as nobody wants to put their established name on a paper that they do not fully agree with.
* In rare cases top level papers are only reviewed by the editors. I
think that this [happened](https://www.nature.com/articles/427584c) for example for the DNA structure paper.
* As mentioned in the comments this post might also help: [How does peer-review work for "gigantic" collaboration projects?](https://academia.stackexchange.com/q/63659/17254)
Upvotes: 2 <issue_comment>username_2: This seems to be a hypothetical scenario. In reality there are always experts that are not on the paper.
Take the example posted in the OP. This is a typical review paper produced by a COST action, an EU funding schemes aimed at building international collaborations. As such, the author list is very EU centric. I can easily think of a few names of US based scientists that could referee this particular paper.
I posit that this will generally be the case. There will always be experts that were not part of this group effort. Sometimes because of geographical reasons, sometimes because of sociological reasons, and sometimes because some people just actively avoid this type of large group activity.
Upvotes: 3 [selected_answer]
|
2023/01/31
| 3,293
| 13,323
|
<issue_start>username_0: Summary
-------
Similarly to how [gerrit](https://academia.stackexchange.com/users/1033/gerrit) asked "[Is there any research on the prevalence of academic theft?](https://academia.stackexchange.com/questions/27472/is-there-any-research-on-the-prevalence-of-academic-theft)", I am wondering if there are any surveys or studies on the forms and prevalence of "systemic corruption" (definition below) in Academia. I refer in particular to corruption where people systematically ignore the rules, sometimes for "good reasons" (i.e. not for their direct benefit), but in such a way that it somehow justify others' abuses, ignoring the same rules for their own benefit.
As a member of Academia (in particular while being in Latin America), I would like to make my modest contribution to reduce the prevalence of such corruption, if only by showing the example and teaching about it to students. Are there any documentation around which can help in such an endeavor?
I list below Wikipedia's definition and some examples of such systemic corruption, along with more details about my motivation in asking such a question.
---
Definition of Systemic Corruption
---------------------------------
The [wikipedia page about Corruption](https://en.wikipedia.org/wiki/Corruption) defines various types of corruption, among which "Systemic corruption, defined as
>
> corruption which is primarily due to the weaknesses of an organization or process. It can be contrasted with individual officials or agents who act corruptly within the system.
>
>
>
---
Examples of systemic corruption in academia
-------------------------------------------
(disclaimer: some of those I witnessed, and some others I was suggested to perform by the hierarchy and/or secretaries):
1. Requesting money for one item and using it for another one. If requesting travel or salary money for person A so that it can be used to pay person B because the rules prevent or complicate paying to B, it can be done for reasons that both consider "good" (e.g. sending a student to a conference) but open the door to such rules to be abused (e.g. paying someone for a job they did not do).
2. Requesting a letter of invitation for a longer period of time than the actual visit, "to simplify things". When justifying a (paid) trip to various places/countries in the same trip, for distinct research projects, one could consider to ask a single invitation letter for the whole period rather than 3 distinct ones for the respective periods, to "simplify things" for the administration which has to decide whether to allow the trip or not, but it opens the door to people having the university to pay for their (extended) vacations.
3. Adding colleagues as co-authors to publications "because they need it" or asking colleagues to add one as a coauthor. [A colleague was recently suggested to do so by the director of their department to compensate their lack of publications after a period of intense administrative responsibilities, arguing that "everybody does it" (which I strongly disagree with).] One can be "generous" with co-authorship as a general principle, (among other reasons, because measuring each contribution is problematic, and expecting the n co-authors of a publication to contribute exactly the same would be complicated to enforce), but it seems to be one step on the path to adding (a) co-author(s) after the article has been completely redacted as part of a "collusion deal" to increase publication numbers.
4. Shortening classes in agreement with the students without informing the administration. It is hard to maintain attention for 1h30, so it is a common practice to take a break in the middle. I saw some instructors deciding for the class to start 30mns late (arguing that "the students are always late anyway) and teaching only for 1h, and others ending the class 30mns early. It might make sense in term of student attention, but it might also open the door to abuses.
5. Skipping classes (in agreement with the students) without informing the administration. Some universities explicitly require professors to name an official replacement for their teaching duties during periods for which they have been allowed to travel (and check that the teaching duties of such replacement do not intersect with those of the traveling professor). When setting-up such "replacement", it seems common to request a "virtual replacement" where the replacement is just in name, and the students are "freed" of classes during the corresponding period. Such practice might seem justified for short periods of time (e.g. 1 week), but could be abused by being used for much longer periods of time, in a way detrimental to the student's learning.
6. Self-Plagiarism: when using a similar technique in two (or more) distinct publications (for distinct problems), duplicate the technical proof with minor corrections rather than "refactoring" it into a technical lemma in the first publication, to be cited in the later ones. Failing to do so might be just blamed to bad redaction, but it can be abused in order to make a publication "more technical" and novel than it really is.
Those are just examples which come to mind or which I witnessed, I am sure that there are many others. I am curious as to how prevalent such examples are, and of resources to help (young and less young) academics to
* convince their peers at an institution to ban/reduce/fight some of such practices, locally considered "normal" and wide-spread, with arguments that such practices are globally NOT considered that normal, and definitely not that wide-spread in academia at the world level; or at least
* convince students and young researchers NOT to adopt the local customs, with some arguments about what is accepted and what is not accepted in other institutions (where they could land a job in the future).
---
Motivation for the question
---------------------------
I witnessed various instances (some benign, some more grave) of such "systemic corruption" in academia in Latin America, but this gives only anecdotal evidence. I discussed it with some international colleagues, which also only gives anecdotal information about their prevalence:
1. (some) colleagues from Latin America *do not consider as corruption per se* (for them, "corruption" rather corresponds to what the wikipedia page refers to as "Petty Corruption" and "Systemic Corruption"); while
2. (some) colleagues from Europe react by saying that *such corruption exists in Europe too* (with which I would agree although I think they do not understand the difference in prevalence) and
3. (some) colleagues from North America condemn such corruption (but admit that some of it exist also in some "bad" places in North America).
I understand at least some of the mechanisms behind such corruption: the people practicing it are mostly doing so in order to get around inadequate legislation and not for direct personal benefit, a type of legislation that they cannot change on their own. While understandable, this is highly detrimental to society in various ways:
* in some cases the amount (mental, time, money, etc.) energy wasted in multitudes getting around a single "stupid" rule by out weights by far the amount of energy which would be required to design and switch to a better rule, *if only people could coordinate*;
* in all cases, having most people disobeying a rule (without abusing it) opens the door for a few individuals to abuse disobeying such a rule (each decides to which point to disobey the rule), which is later used to justify even stricter rules, which more people will try to get around of in order to simply doing their job, in an endless vicious cycle.<issue_comment>username_1: Certainly there is corruption but it takes at least two different forms. I can only speak for the US, however.
There is minor corruption that is probably pretty common with people cutting corners, even so minor as taking home a ream of paper for their home printer when it will be used "mostly" for work. Padding of expense accounts for travel might qualify. This might not be a managed issue as the individual occurrences are small enough that policing them is more expensive than letting it go. But, I'd expect that a few people are fired over such issues. I don't see universities (reputable universities) as corruption machines that make such things easy.
The other kind of corruption is when a university is actually founded and organized around a corrupt purpose, say the financial or political benefit of the founders. *Trump University* was shut down and fined over such things. A few (not all) "religious" colleges in the US have a similar reputation. Even some technical "for profit" schools have run afoul of the law. And some schools have badly abused the student loan system to their own benefit. But these sorts of things are relatively localized, and aren't representative of the higher education system in general.
One feature of higher education that some consider corrupt is the tight relationship between some university presidents and the board of trustees resulting in very high salary. But this is done in the open, generally speaking, so may not be "corrupt" in the classical sense. Undesirable, perhaps.
Upvotes: -1 <issue_comment>username_2: I was not able to find up-to-date data on the academic corruption worldwide, the closest result would probably be [Transparency.org's](https://www.transparency.org/en/gcb/global/global-corruption-barometer-2017) Global Corruption Barometer. As is typical for social studies, there are multiple caveats to the methodology; numbers may be hard to interpret: for example, Belgium has a low rate of bribery, yet the perception of corruption by institution (Table 2) is high overall and exceeding that of Bangladesh, which has a much higher bribery rate ([from the 2013 report](https://images.transparencycdn.org/images/2013_GlobalCorruptionBarometer_EN_200525_112757.pdf)). That is to say, you are completely correct in noting that the same practice may be perceived as normal in one country and completely off-limits in another, and that would heavily affect the implications of any analysis.
To approach the systematic aspect of corruption in academia in particular, I would start with [this highly cited paper](https://doi.org/10.1207/S15327930pje8001_5) (Rumyantseva, 2005), which attempts to build a taxonomy of academic corruption. It distinguishes between administrative corruption and education-specific corruption, and further divides the latter into cases falling under "academic corruption" and "corruptions in services". However, it does not discuss prevalence of these cases and generally sticks to more obvious, even blatant examples throughout the text.
[This review](https://www.alexandria.unisg.ch/254858/) (Denisova-Schmidt, 2018) and references therein offer a more detailed view on the prevalence of corruption cases. This includes high-profile public figures in the educational sector:
>
> In 2016, the Ministers of Education of Armenia, Bosnia and Herzegovina, Croatia, Russia, and Ukraine were all implicated in conflicts of interest. In addition, some or all the deputy Ministers of Education in Armenia, Azerbaijan, Croatia, Moldova, Serbia, and Ukraine, as well as some members of the cabinets in Armenia and Kazakhstan, have also been accused of having conflicts of interest.
>
>
>
In Russia, rampant corruption and plagiarism have prompted a formation of a vigilante group within the academic community - [Dissernet](https://www.dissernet.org/). This is not to single out Russia or ex-USSR countries, although the situation is certainly problematic there. It is not a coincidence two of the authors I have cited so far have Slavic names. Many developing countries are experiencing similar growth pains, too. For an example more relevant to you, a [high-level overview of this transition in Latin America](https://eric.ed.gov/?id=ED511961) (Segrera, 2010) is concerned with reasons behind these pains.
For the **advice on fighting corruption**, I will refer to Denisova-Schmidt again:
>
> In order to combat this corruption, the faculty should present their assignments and expectations more clearly to the students, stipulating their educational and cultural backgrounds.
>
>
>
She then stresses:
>
> Faculty members should serve as role models, however. If they also cheat, they might not be able to demand the opposite behaviour from their students.
>
>
>
And
>
> General research on corruption suggests not fighting corruption in general but rather focusing on specific malpractices [...]
>
>
>
Coming from a somewhat similar background, I may only add to this advice one thing: do not put much effort in combating corruption where you can not provide a positive example. Pick your fights. Keep serving as a role model. But lecturing anyone on morals is not going to be particularly effective unless you can show your example *works*, that one does not *have* to offer bribes to be able to put bread on the table. As long as people keep trying, as long as they hope for a better future, there is progress. Do not demand it to be "all or nothing", because you will keep getting nothing. Offer encouragement and forgiveness.
Upvotes: 2
|
2023/01/31
| 1,914
| 8,583
|
<issue_start>username_0: My team has been trying to replicate a study published in a journal.
After a month of effort my team is convinced that the details provided in the paper are incorrect and the study cannot be replicated with the information provided in the paper.
We now want to publish the discrepancies in the paper along with proposed modifications.
I want to know:
1. What article type (Commentary, Letter, Note, etc ) would fit for the proposed submission?
2. Should it be submitted to the same journal or could it be submitted to any similar interest journal?<issue_comment>username_1: I don't think we can answer your first 'article type' question, as different journals call different types of submission by different names. Look at the journal's instructions to authors and see what they say should be submitted to comment on previously-published work, and follow that. Or ask the journal directly if it's unclear.
In terms of the 'where' question, you can of course submit it to any relevant journal. But if it's a narrow criticism of a single paper's methods, it's unlikely that any journal other than the original publisher of the article would think it's sufficiently noteworthy. If the result you're questioning is particularly high-profile, or the method is widely- and incorrectly-used, there may be more interest from other journals, but it doesn't sound like that's the case here.
Upvotes: 3 <issue_comment>username_2: First of all, as the comments have already indicated, I, and most reviewers, would be skeptical that the paper in question has various errors/discrepancies. Maybe there are some notation issues, different understandings, or other typesetting glitches, but these errors are different than what you *seem to be* implying, which is that the paper's claims are "wrong". You should make it clear in your question whether you think this original paper is simply flawed, or whether you think this original paper has serious issues and should be retracted or corrected (I get the impression that you're implying the latter).
If there are errors, you need to make the distinction between a technical error which is mostly harmless and due to imperfect human understanding, as opposed to an intentional or negligent error on the part of the study's authors. Just correcting some notation/equation issues is not significant enough to warrant publication by itself, at least in the fields I am familiar with (e.g. various areas of engineering). On the other hand if you think there was intentional or negligent errors on the part of the original study's authors, and you have some evidence to back up this claim, then it sounds like you should contact the journal's editors with your concern and evidence.
Ultimately, to warrant publication, it is expected to make a significant new contribution. Correcting some typos or pointing out problems in previous papers shows that you understand the material, but that by itself is not a significant new contribution. If the paper you're correcting is especially noteworthy, I suppose the correction might warrant publication by itself as a letter length paper. But you really need to think in terms of making a significant new contribution, as opposed to your current mindset which *comes across as more* "this paper in the literature had various problems in it" (which may very well be true, but saying that by itself is not a contribution).
Lastly, you raise the question of reproducibility: well in a lot of engineering, it's basically impossible to reproduce a study unless you also have the code. The code might not be available for a number of reasons (e.g. funding might impose that the funding agency has sole access to the code for some time; or maybe the authors have a follow-up in mind and have kept the code private intentionally until the follow-up is completed). Also, the authors might not reply to you for various reasons (e.g. they're on vacation, or have too many emails and missed yours). This is all to say, a lot of the literature is, unfortunately, NOT reproducible (at least in the strict sense of what 'reproducible' means).
You should focus on your paper's contribution (not the contribution of previous papers), and if some paper in the literature gets some parts wrong, then of course you should present the correct way in your paper, but it's not needed to call out specific authors/papers as being 'wrong' or 'bad'. If your way is truly better, rest assured that other people will (eventually) notice this as well, and tend to cite your paper more.
Upvotes: 2 <issue_comment>username_3: Journals publish less-than-perfect papers so often that it's seldom even noteworthy. I can think of several often cited papers from the 20th century with glaring typos in the math formulas, and/or with gaps in proofs, which were fixed in later papers.
As your first step, you should contact the article's "authors", using the contact information in the paper, or on their departmental web site. You should include all the authors, not just the designated "corresponding author", unless there are too many of them. You should include all your detailed and specific criticism of their paper in your initial e-mail, rather than a "teaser".
Example of a bad e-mail: "I found problems in your paper, please contact me so we can discuss." - many people would assume this to be spam/crank e-mail, and not even bother responding.
Example of a better e-mail: "You state in your paper that because water is wet, therefore the sky is blue. But I see a gap in your argument because you assume that the quadratic variance term is zero." Unless you have reasons to suspect intentional misconduct, you should assume an honest mistake, and treat the authors like you'd want to be treated if you were in their shoes.
Optional: "We further proved a weaker result, that the sky is blue-green, and found some counterexamples of the sky being green." If you've made some non-trivial advances, then you may choose to share them with the authors or not. Could you use them in your own publication? Are you worried that the authors would steal it?
The authors should immediately acknowledge the receipt of your e-mail, and within a few weeks either explain to you why this is not a problem, or send a correction / retraction / clarification to the original journal, graciously thanking you for finding the problems. It would be unusual for the journal to decline to publish a correction submitted by the authors. Or they may even argue that the problems are unimportant, but you may disagree.
If the authors blow you off or you're otherwise still not satisfied, then the next step may be to post your criticism on [pubpeer](https://pubpeer.com/). Other people may comment on your criticism there, and help you improve it. Or the authors may respond to your satisfaction this time. You may decide to skip this step and go straight to the letter to the editor.
The editor should acknowledge its receipt right away, solicit feedback from the authors and the reviewers, and hopefully publish your letter, together with any rebuttal from the authors. But they may totally blow you off, especially if it's not a reputable journal.
The next step may be to write a little paper on your own. Frankly, a paper focusing mostly on the problems in another paper is unlikely to be of interest to many people. You should try to include some new results if possible, so people reading it won't feel that it's a total waste of their time. Posting a preprint on arxiv or the like may be more realistic than a journal.
Upvotes: 4 <issue_comment>username_4: I do not think these kind of omissions are publishable unless the paper is highly cited and the results are misleading.
If this is in computer science, a significant portion of the papers are in the same state. Replication therefore requires guesswork. Although this makes the paper a low quality one, there are many of these even in top journals. General response to these are simply contact the author to obtain additional information or source code of the project. I suspect most of these authors think that explanation that they provide is enough, but since only they have complete knowledge, explanation given is often too little. Even when there is an algorithm, it generally does not cover entire method.
To summarize, while this trend of low quality, very hard or impossible to replicate papers do slow down the speed of research conducted, it is considered normal and you will not be able to get back anything for the time you have lost.
Upvotes: 1
|
2023/01/31
| 666
| 2,377
|
<issue_start>username_0: Suppose that we have a citation style in which given names of authors are abbreviated in the bibliography, that is, either
>
> <NAME>.: An interesting observation.
>
>
>
or
>
> <NAME>: An interesting observation.
>
>
>
In this case, how should one handle an author who publishes under a name that includes a nickname in quotes? Something like
>
> <NAME>. “Jack” Miller
>
>
>
or even worse,
>
> Walter ‘The Wizard’ Brown
>
>
>
Is there any style guide that covers such cases?<issue_comment>username_1: The example is of someone who has adopted an Anglicized name, probably because the person is tired of mis-pronunciation and maybe in this case, because the name is one associated with uncultured Bavarians. A similar practice is much more prevalent among Chinese Americans, who then face the problem under which name to publish.
I would just use the name exactly as they write it, since I am unaware of any rules regarding the use of alternative names. I would however restrain from calling it a nick-name, because I do not think that it is a capricious decision.
Upvotes: 0 <issue_comment>username_2: Personally, I would stick with formal names and leave anything you see as a nickname off of the bibliographic reference. So *<NAME>* or *Brown, W.* is enough. The extra common name isn't needed to get to the paper, after all. And a single initial for such a name looks a bit odd.
In the body of the paper you might actually spell out their "moniker" if you like.
If an editor suggests otherwise, then follow their advice for that publication.
Upvotes: 2 <issue_comment>username_3: I consult the *APA Style Expert* and one of their expert Stefanie tells me "A nickname in quotes in an author's name is represented by an initial for the nickname in quotes in the reference." in APA
In your example, that would be:
<NAME>. “Jack” Miller -> M<NAME>. "J."
Walter ‘The Wizard’ Brown -> <NAME>. 'T.'
Original:
>
> There are cases where you do not need to invert the name and keep the full name of an author. For example, you do not need to abbreviate and invert names that consists of 'an inseparable multipart name' like '<NAME>' or '<NAME>'. This is said in Section 9.8, Page 300 of Publication Manual of the American Psychological Association, 7th Ed. (2020).
>
>
>
Upvotes: 3 [selected_answer]
|
2023/01/31
| 3,585
| 15,819
|
<issue_start>username_0: Being an experienced reviewer, I know that rejection can be frustrating, but also fruitful and a good learning experience. There are many kinds of reviews (good, bad, constructive, opinionated, short, missing the points, etc.) and as many kinds of responses (revision, rebuttal, abandonment, etc.). My question focuses on a specific case, a bit different from those [considered here previously](https://academia.stackexchange.com/q/5865/94523).
I received a notification about a grant proposal (I'd prefer not to say which agency I submitted to), already being evaluated as a fundable or excellent one, and quite critical to my career (really a lot of stuff depends on it, and one can see this from my CV). One review was plain positive (no criticism at all), and one was grossly negative. The two reviews had almost diametrically opposed conclusions. (There were only two!)
*My problem:* The negative review made several *false claims* (the evidence was easy to pinpoint) and put up a range of *straw men* (rather generic rejection arguments pinpointing irrelevant or inappropriate details). For example, they criticized the oldest of my papers I had cited and used this as evidence that I was insufficiently experienced in this field, but they overlooked the next seven papers I had also published in the same field (some in prestigious journals).
At this point, I suspect the reviewer was not knowledgeable of the topic or even had a certain conflicts of interest. In either case, they IMHO should have declined the review request.
However, the evaluation committee explained that they had to particularly take into account the negative review to inform their decision (without giving a reason), and interestingly, added further negative criticism.
*What can I do about this notification?* Should I contact the funding agency, write a rebuttal with the details, and request a third review and a revision? Should I think about legal action because there might be gross negligence or reputational damage at play? I don’t have any idea, because talking to funding agencies is like trying to fly to Mars. Re-submission, even if possible, will take another year because it will be treated like a new application. It is, at least, unclear whether there are other options.<issue_comment>username_1: You can of course contact the program manager, but in all of my experience, not very much will come out of it.
I don't know whether you have been part of funding panels in the part and so have experienced the dynamics of people debating proposals. What often happens in cases such as yours is that the two reviewers of the proposal present their cases for whether they think the proposal should or should not be funded. If there are opposing opinions, the panel has to come to some kind of conclusion of what to recommend, and others often jump in to read all or parts of the proposal with an eye towards the specific points of criticism. They may not have the time or inclination to write a full review of the proposal, but they will be able to participate in the discussion and, in your case, they appear to have thought that the negative review needed to be given some weight, and the discussion found other points to critique as well that then ended up in the panel summary.
Program managers are typically reluctant to substitute their own opinion for that of the panel (for good reasons). As a consequence, it may be useful to the program manager to know if a reviewer was completely off topic, if only to know that they shouldn't invite that person again as a reviewer. But I would suspect that your chances of getting another shot with this proposal are rather slim.
Because this is so, my perspective on these things is to not dwell too much on the reviewer's failings -- because that is something I cannot change, and dwelling on things I cannot change do not help me move through life. Rather, I try to assume that it was my own fault that the reviewer did not get what I was trying to say, and use that as an opportunity to question how I presented what I wrote, and how I can improve in the future. In other words, I use these occasions as learning opportunities.
Upvotes: 4 <issue_comment>username_2: I'm sorry this is happening to you. Unfortunately, you have no realistic recourse. If this is truly destructive of your career, the best use of your time is to figure out what you are going to do next in a different career. The race doesn't always go to the swift, the battle to the strong, nor the academic career to the most qualified.
I'm not saying you should accept the bad reviewer's behavior. But my experience is that your acceptance doesn't matter. Academia doesn't correct its mistakes for individuals. Even if it comes out next year that this reviewer has an email chain saying "I don't like Mario and I'm super dumb, so I'm going to shoot his proposal down," the best you will get is an opportunity to apply again. At that point the money will have been already distributed to the other applicants. In my experience, the best use of you time is to move on, unfortunately. Whatever that means for you.
I hope that you overestimate the negative effects this will have for your career.
Upvotes: 4 <issue_comment>username_3: As others have said, there is not much you can do against a dishonest reviewer, if they already have support of the grant panel. Consider resubmitting your idea to another funding agency if you can.
One thing though -- time may be of essence. Your reviewers may have left those unjustified comments because they misunderstood your work or did not make a proper effort to understand it. But it could also be that they actually understood the ideas, and purposefully decided to shoot it down, to delay your progress and to exploit the ideas themselves. Keep an eye for similar ideas to appear in preprints, grant applications, etc. Try to protect your ideas by publishing them at least as preprints, but under your name, before your reviewers publish them under theirs.
Upvotes: 3 <issue_comment>username_4: **Contacting the program manager**
You can, and probably should, contact the program manager at the funding agency to point out the flaws in the negative review. This is an easy action to take, with little downside that I can see, other than the effort you would spend and the possible disappointment if it doesn't produce the outcome you want.
If you do contact the program manager, I recommend:
* Focusing on *factually false* statements the reviewer made. Matters of fact are objectively verifiable, and a conscientious program manager would likely regard errors of fact as one of the only reasonable grounds for overruling a reviewer and soliciting a new review.
* Do not waste your time disputing subjective matters. If the reviewer makes a subjective objection not based on a factual error that you perceive as a straw man attack or as misguided for some reason, I don't think you stand a chance of getting the proposal's rejection reconsidered based on such a flaw, however frustrating it might be for you. Subjective disagreements are too common, everyone who gets rejected thinks the subjective criticism against them is misguided, and at the end of the day, the standard response to someone objecting to such criticism is that, well, if the reviewer didn't understand what you were proposing then that means you just didn't explain it well enough — try harder next time.
* If you can get your proposal and the review looked at by some colleagues with a good reputation, who agree with your views that the review contains falsehoods, and those colleagues are willing to write a letter supporting your objections (about factual matters!) and have you attach it to your letter of rebuttal, I think that might be helpful. It isn't guaranteed to help, but it won't hurt. If you do this, be sure to state clearly any conflicts of interest that those colleagues have which might influence their opinion (for example, that they are your collaborators or work in the same department as you).
**The idea of filing a lawsuit**
I can't say if this idea makes sense or not. Neither can anyone else here who isn't a lawyer (let alone a lawyer in your particular jurisdiction, which you haven't specified). Please don't listen to some of the nonsense I see in the comments, written by people with no legal expertise, about what constitutes defamation, gross negligence etc. If you want to know whether a lawsuit stands a chance of success, *consult a lawyer*.
On the academia side of things, I think it's true that a lawsuit will likely have effects on your reputation and affect the way you are looked at by some of your colleagues. On the other hand, the funding agency may also have reasons to fear for its own reputation from a lawsuit (or the threat of one, or even the threat of you going public with the incompetence being shown in the evaluation you got), so this argument cuts both ways. So, I don't think you've given us enough information to completely rule out the idea of a lawsuit.
That being said, I think you should consider whether your position that the reviewer and other people who played a role in your proposal's rejection are sabotaging your career (and should be held accountable for that, including using the legal system if needed) is really that reasonable. The funding agency has no responsibility to help you with your career. It is offering grants and has a process for evaluating proposals asking for funding, which you have voluntarily agreed to take part of. If your career is so dependent on a single grant that it will be irreparably harmed by this rejection, I think there's a strong argument to be made that this is the fault of your employer and the flawed system they have for evaluating academics. Or possibly — I hope this is not the case, but can't rule it out — it's your own fault for reaching a vulnerable stage of your career with a track record that is still kind of marginal and that necessitates a 100% success rate to keep your career going. (To be clear, I'm not saying this to be mean or to criticize you, but simply showing you what possible objections can be raised to the logic of your arguments. If I don't raise these arguments now, it's quite likely someone else will raise them later, so it's better for you to be prepared.)
Anyway, I'm sorry about your grant rejection. Good luck trying to get this situation fixed!
Upvotes: 6 [selected_answer]<issue_comment>username_5: I don't see the following relevant point being raised in either the question or the answers: **the program probably has its rules and regulations regarding appeals** (of course without details it's impossible to say).
These may be vague, and not very helpful to you. Harsh reality is that the peer-review process does not encourage second-guessing the peers reviewing a rejected proposal, as it's assuming their expertise. This is only mildly annoying with publications (just submit elsewhere) but less so when, say, applying for a grant from a national institution is virtually the only career route you can take, and a time-limited one at that (e.g., several European countries have such institutions and competitions for young scientists - and without a particular grant at this early stage, it can be challenging to pursue an academic career later on). In my particular country (details omitted as well), the appeal may only be raised against a formal failure in the due process, not against a disagreement with reviewers. The reviewers are bound to academic honesty by the grant authority, but it can be tricky to show that they were negligent, and that their rejection is not their honest expert opinion. And, frankly, I admit that when being a rejected party myself in the past, my judgement might have been sometimes clouded. But for what is worth, I sympathize with the dread of missing something you see as crucial for your career.
**However**, I do know people who successfully appealed such decisions. This is more about a yearly grant proposal cycles, not a one-off call for grants. The agencies I am aware of, all publish information about awarded grants, **including success rate of appeals**. This is not something on the top of their web pages, but this information is often available. If your particular grant proposal is a recurring one, this info can give you the overall feeling, whether an appeal is worth your time. You may also want to try to contact people that were successful with their appeals, but that depends on your networks (see below).
I personally appealed a rejection once, having a strong feeling that a reviewer was biased and not honest (personal favorite - the main objection was that the research I planned in that grant *was not published yet*). The appeal was overturned, as the reviews were deemed within bounds of an expert opinion (the publication thing - should I publish this research earlier, I would score more points for my academic achievements, therefore the reviewer was right to hold that against me). On the other hand, a colleague of mine was able to conclusively show that their reviewer did not read the whole proposal (raising points that were explicitly addressed beyond abstract and such), and their appeal was upheld.
I emphasize that I was able to acquaint myself with my colleague's case by sheer coincidence, as I would not contact a total stranger that went through with an appeal to ask them about that. So I'd advise to talk to your colleagues, who will know more about the context of your grant proposal - and perhaps know someone that can be more helpful in your specific case.
Finally, I strongly second @username_4 in that if you contemplate legal action, *talk to a lawyer*, better yet, *one that has specific experience with academia in your country*.
Upvotes: 2 <issue_comment>username_6: Aside from contacting the appropriate portfolio manager, as described in a previous (excellent!) answer, another action you can take is to examine the roster of reviewers, and if you're aware of any *real* conflicts in the reviewer pool, let the program officer know.
Note that being very expert in the area is not a conflict.
If a conflict is identified, then that would probably mean that particular reviewer would be disqualified from reviewing a resubmission. If the program manager knows for a fact that a person you deem to be in conflict trashed your application, there may be fixes available to them -- maybe even a re-review.
Keep in mind that you will likely not find out who reviewed your grant. Also, it may be the case that a person you *think* is in conflict may even have been your biggest proponent during the review process. My impression from being on Study Section is that reviewers tend to get very provincial about their subfields, and like to see money ending up in the area, even if that money doesn't go to them. (Note that this is clearly not a desirable approach for a reviewer to have! I'm just letting you know that I've seen people I think may have been behaving in that way)
Of course, the go-to option is a resubmission addressing the reviewer's points. Be diplomatic, but if you don't believe the points are valid, and believe you can convince a study section that you're right, then give it a go.
Lastly, while the situation comes up over and over again, having important parts of your career success hanging on the outcome of one grant is a very stressful situation (though, to some extent, this is simply the unpleasant nature of the beast). Many departments have informal or formal faculty mentorship arrangements that can really help investigators fine-tune their approaches to grantsmanship. If such a program exists for you, by all means, ping this situation off of a mentor.
Upvotes: 2
|
2023/02/01
| 600
| 2,701
|
<issue_start>username_0: Some conference papers have misleading names, for example, 'journal of physics conference series' with both journal and conference in the name. 'journal of physics conference series' is almost like a journal because it publishes full manuscripts on their website but it also has conference in its name because it collects submissions from a conference that really takes place somewhere. PROCEEDINGS OF THE ROYAL SOCIETY A is just a journal rather than a conference. In such cases how to tell if a publication is really a journal or a conference?
To tell if something is a conference or a journal doesn't really matter that much but there is a follow up question. If it's considered a conference with a full manuscript, it is possible to publish roughly the same contents on another normal journal. If it's considered a journal then of course there's no reason to publish it on another journal once again.<issue_comment>username_1: Once a paper appears it is unlikely that *another reputable publisher* will consider republishing it. There are several reasons, including (lack of) novelty and copyright restrictions.
If it isn't clear whether a paper appeared at a conference (which really shouldn't happen), then you need to ask the publisher. But the difference may be immaterial in some fields.
Note that some conferences publish papers in a "proceedings" and then publish a (perhaps edited) version in a journal that is somehow connected to the conference. Others just treat the proceedings as the final word and consider those proceedings to be the equivalent of a journal. This is common in computing when a conference is sponsored by one of the large membership organizations. The ACM, for example, has a worldwide spread of conferences.
Upvotes: 2 <issue_comment>username_2: Papers published in Journal of Physics: Conference Series are usually quite easy to identify as conference papers from the PDFs. On the website, the volume of the "journal" sometimes directly names the conference, but sometimes you only find out by clicking through to the volume.
One way to tell if a publication venue is a conference proceeding or a journal is to look it up on Scopus. [Here](https://www.scopus.com/sourceid/130053) is the page for Journal of Physics: Conference Series, which is identified as a proceeding. Compare with the case of [Journal of Physics A: Mathematical and Theoretical](https://www.scopus.com/sourceid/5700165208), which is identified as a journal. I'm not sure if this would identify cases where a journal would publish proceedings of some specific conference as a special issue, but again, special issues/volumes are usually clearly labeled.
Upvotes: 2
|
2023/02/01
| 614
| 2,613
|
<issue_start>username_0: I am dyslexic and often struggle with reading and comprehension. Screen readers, such as VoiceOver on Macs, can be helpful in addressing these difficulties.
However, mathematics in arXiv papers can present a challenge for screen readers. A partial solution is ar5iv, which converts arXiv papers into HTML, improving accessibility (for example, see [this ar5iv paper](https://ar5iv.labs.arxiv.org/html/2212.07286)).
However, I have yet to find a screen reader or configuration of a screen reader that will read ar5iv papers in a natural, fluid way. For example, VoiceOver says 'maths' after every MathML equation, interrupting the reading flow.
**Question:** Is there a screen reader or configuration that can effectively read arXiv or ar5iv papers, with mathematics included, in a more natural and seamless manner?<issue_comment>username_1: Once a paper appears it is unlikely that *another reputable publisher* will consider republishing it. There are several reasons, including (lack of) novelty and copyright restrictions.
If it isn't clear whether a paper appeared at a conference (which really shouldn't happen), then you need to ask the publisher. But the difference may be immaterial in some fields.
Note that some conferences publish papers in a "proceedings" and then publish a (perhaps edited) version in a journal that is somehow connected to the conference. Others just treat the proceedings as the final word and consider those proceedings to be the equivalent of a journal. This is common in computing when a conference is sponsored by one of the large membership organizations. The ACM, for example, has a worldwide spread of conferences.
Upvotes: 2 <issue_comment>username_2: Papers published in Journal of Physics: Conference Series are usually quite easy to identify as conference papers from the PDFs. On the website, the volume of the "journal" sometimes directly names the conference, but sometimes you only find out by clicking through to the volume.
One way to tell if a publication venue is a conference proceeding or a journal is to look it up on Scopus. [Here](https://www.scopus.com/sourceid/130053) is the page for Journal of Physics: Conference Series, which is identified as a proceeding. Compare with the case of [Journal of Physics A: Mathematical and Theoretical](https://www.scopus.com/sourceid/5700165208), which is identified as a journal. I'm not sure if this would identify cases where a journal would publish proceedings of some specific conference as a special issue, but again, special issues/volumes are usually clearly labeled.
Upvotes: 2
|
2023/02/01
| 476
| 1,972
|
<issue_start>username_0: According to [this Wikipedia article](https://en.wikipedia.org/wiki/List_of_sovereign_states_by_research_and_development_spending), Sweden, Austria, Switzerland, Germany, Denmark, Belgium, and Finland are the top spenders in the R&D sector by PPP per capita.
Does this mean that they pay a higher salary to the PostDoc researchers?
If not, is there any metric to know this?<issue_comment>username_1: R&D spending does not equate to PostDoc salary. A substantial part of the spending is done by private companies in their own facilities, and private companies typically don't hire PostDocs. Some of the spending goes to stuff (supplies, machinery, laboratories, etc.) rather than people. Salaries in many of these countries are not driven by the available budget. If more money becomes available the universities just hire more PostDocs, rather than pay them better.
If you want to know the salary, you just look at the salary. But do take the cost of living into account. Not at the national level, but at the local level of the university you are interested in.
Upvotes: 4 <issue_comment>username_2: Postdocs are more or less government employees and so in a lot of EU countries, postdoc salaries are publically available so a little bit of digging will usually tell you the exact salary of any position you see advertised.
However, if you only want a quick rule of thumb to compare different European countries I would go with nominal GDP per capita. Yes, cost of living is higher in the wealthier countries as well, but academic salaries usually overcompensate for this (when comparing countries in Europe). The standard of living a postdoc salary buys you in Switzerland or Norway is much higher than in Greece or Portugal.
If cost of living differs a lot within a country the academic salaries usually do not fully account for these differences so living in an expensive city will make you relatively poorer.
Upvotes: 3 [selected_answer]
|
2023/02/01
| 1,070
| 4,377
|
<issue_start>username_0: Currently, we have a discussion in our work group as to whether one may use images from WikiMedia Commons in Case Reports. I have a general understanding of what types of CC license one may use or not. However, this specific case is confusing me:
[](https://i.stack.imgur.com/RgmqW.png)
Said file has been uploaded under multiple licenses. How is this possible?
Is it correct that I can use that image for a scientific paper (provided I give proper credit) which is going to published under a "NonCommercial 4.0 International (CC BY-NC 4.0) " license?
Any input is greatly appreciated.
Background: the image is necessary for a case report in Medicine. Regrettably, it is about a plant that is not growing in winter. Thus, I need to find an alternative way of obtaining a picture of that plant.
I know there are related topics: [Can I use images from Wikipedia in my paper?](https://academia.stackexchange.com/questions/168479/can-i-use-images-from-wikipedia-in-my-paper)
Yet, in this case, the image has multiple licenses.<issue_comment>username_1: >
> Said file has been uploaded under multiple licenses. How is this possible?
>
>
>
If you own something, you can license it however you want as long as you haven't agreed to some terms that prevent you from releasing it under other terms (usually referred to as an 'exclusive license'). A fairly common scenario would be releasing something under a fairly permissive license that lets others use the content for anything except commercial purposes, and then also licensing to specific users for commercial purposes for a fee.
Often journal publications occur with terms that either give the full rights to the journal and then re-license the content back to the others allowing them certain privileges, or keeping the ownership with the original authors but offering the journal an *exclusive* license that promises A) That you haven't already licensed the work for certain purposes, and B) That you can't in the future.
It seems unlikely to me that this image under either of these licenses will be useful for your purpose if you plan to submit your paper to a typical medical journal, and [I don't see compatibility with CC BY-NC 4.0](https://creativecommons.org/share-your-work/licensing-considerations/compatible-licenses), but I'm not going to pretend to provide legal advice on that, I'm no lawyer. It also doesn't seem like it is truly *necessary* to have an image of a plant for a case report - I'd look for something in the public domain or something you can obtain directly from the rights holder.
---
It looks like [this SE Q&A disagrees with me, suggests it's fine to use a CC-BY-SA image in a work released as CC-BY-NC, and references a dispute of this that led to a court case](https://opensource.stackexchange.com/questions/9216/using-cc-by-sa-images-in-a-cc-by-nc-sa-document). Personally, I'd still probably want clearer permission for something like this (as it seems it was still unclear enough to have a case over) or to work around the issue entirely.
Upvotes: 3 <issue_comment>username_2: IANAL, but I don't see any issue with your use of the image, provided you keep some things in mind. First, once a license is granted it can't be replaced by a more restrictive license. Both of the licenses permit what you'd like to do.
Just note, however, that your "non commercial" restriction won't apply to that image, since it already has a permissive license. The NC clause will apply to other things in your work, however, but others are still free to use the image "commercially" due to the earlier license(s).
You have to attribute the image to its source, of course. And, if you modify the image you have to adhere to the "share alike" clause which you agree to in using the image. It would be proper to put a note in the paper concerning the license on the image which both gives you rights to use it and notifies others of their rights.
---
Caveat: I can (barely) envision a situation in which that image is so essential to your larger work that you would need to apply the more permissive license to your own work. That seems unlikely, however, but would apply if the report you write could be fairly considered to be a work derived from the image. Unlikely, but not impossible.
Upvotes: 1
|
2023/02/01
| 1,133
| 4,628
|
<issue_start>username_0: I am a high school student planning to do a research project involving human participants. This is for a research class at my school, and one of my teachers told me that I need IRB approval. My research mentor has a computer science PhD.
I want to improve on existing machine learning models for detecting dyslexia. I want to collect data from both dyslexic and non-dyslexic participants.
My school does not have an IRB so I am planning to send it to an IRB of a surrounding university.
My question is, how can I get my project reviewed by an IRB? Is it possible for me to send IRBs to multiple universities at the same time? I know you can submit to a different IRB if one IRB rejects you, but I could not find any answers to my question.<issue_comment>username_1: Although it might vary from country to country, my experience is that many universities are willing to undertake ethical review of research projects ***for a fee***.
To prevent one from shopping around for a favorable review, it is often a condition that you declare on your submission that you have not previously submitted a similar application to another Human Research Ethics Committee (HREC) or Institutional Review Board (IRB). You might, as a consequence, need to consider carefully which institution is the best one for your submission, and perhaps do so in consultation with an appropriate person from your own institution.
Maintaining an IRB or HREC is an expensive exercise and a number of smaller institutions can't afford to do so. In addition, by providing a review service to non-insitutional members, a university that does have ad IRB can assist those people who are not affiliated with any formal institution to meet the demands of journals that insist that authors have previously had their project reviewed for conformance with accepted ethical standards.
You will see, for example, that the National Health and Medical Research Council of Australia discusses "non-affiliated" applications on [this page](https://www.nhmrc.gov.au/research-policy/ethics/human-research-ethics-committees).
Upvotes: 3 <issue_comment>username_2: >
> how can I get my project reviewed by an IRB?
>
>
>
You pay them. University IRBs may or may not handle cases from outside the university, but there are [commercial IRBs](https://hso.research.uiowa.edu/session-3-commercial-irbs-who-are-they-and-what-do-they-do) that do this all the time.
It is likely that your project would qualify for a "minimal risk" exemption (e.g., if you are having participants complete simple tasks and recording their responses and reaction times, then the odds of a participant being harmed are negligible). But, even this would be take months and cost at least $1,000.
The possible complication is that you would presumably be recording diagnoses (e.g., which participants are dyslexic) and publishing data based on this dataset. I do not know if this would be considered a "medical record," but I suspect it might be, in which case you might actually need a full IRB protocol. This would probably be a no-go: I seriously doubt any IRB would approve a study in which the principal investigator is a high school student. (And even if they did, this would be very expensive and time consuming).
>
> one of my teachers told me that I need IRB approval.
>
>
>
This may or may not technically be true. For sure federally-funded research requires an IRB (either a full protocol or a formal exemption), and if you want to publish this in a "real" science journal, they will require it too. If this is for a science competition, they may have a policy requiring one (though they probably just say "at your own risk"). If none of the above apply, you probably don't actually need IRB approval.
That said, the above discussion should hopefully make it clear that there are lots of legal landmines here. While the odds of you harming someone and getting sued are very small, it could really mess up your life if it happened. Your teacher advisor is probably (rightly) concerned about this (I suspect "you'd need IRB approval for that" is an elegant way of saying "no freaking way!").
>
> I want to improve on existing machine learning models for detecting dyslexia. I want to collect data from both dyslexic and non-dyslexic participants.
>
>
>
Bear in mind that machine learning models are data-hungry, and collecting data is tedious. Even getting data from 50 participants would likely require a heroic amount of work on your part, and it's unlikely you could do much machine learning based on such a small dataset.
Upvotes: 3 [selected_answer]
|
2023/02/02
| 632
| 2,620
|
<issue_start>username_0: I’m fairly new to the research world and currently working with a university research institute. My PI put me on a project for a lab he is collaborating with from another research institute. The work I’m doing is going on her grant that she is working on, and my PI is writing a portion of it. He told me I was going to be on her grant and payroll for my contribution. He also said she will be paying 20% of my salary when her grant gets approved.
I’m confused about what this means and how grant salaries work in regards to collaborations. Do they just pay a percentage of my institutional base salary (IBS) to make up for the total salary? Or is this in addition to my current IBS that I get from my PI?
I have asked him to clarify, but perhaps I’m still not understanding how this works, and I don’t think I got a clear answer.<issue_comment>username_1: It's most likely that you will be working 80% on your advisor's grant and 20% on the collaborator's grant and that your salary won't change.
There are pros and cons to such an arrangement for the student:
Pro:
* You'll get to work with and learn from another researcher.
* You may get additional opportunities to publish.
Con:
* You'll be working for two different bosses who may have different expectations.
* Working on two different projects at the same time will require more mental effort to keep track of both projects. For example, you might have to read and absorb twice as many journal articles as before.
I've had some advisees work 10 hours per week as TAs and 10 hours per week as RAs, but I'd avoid asking a student to work on two different research projects for 10 hours per week because of the difficulty of keeping track of two separate projects.
Upvotes: 4 <issue_comment>username_2: This professor is not saying that the other PI will be putting you on their payroll. When collaborators receive a grant, there is often a primary PI (seems to be the person from the other institute) and co-PIs or secondary PIs. These folks typically receive sub-grants or secondary grants. This could be from the primary PI's institution or directly from the grantmaker - in this case it seems that the former is happening, so your professor is saying that the other PI is "paying your salary." This is true in some sense, but, the important point is that the funds end up in your PI's account and from your perspective you are paid as usual.
The point the professor was probably trying to make with the comment is that you will be spending 20 percent of your time on the project in this proposal if it gets funding.
Upvotes: 3
|
2023/02/02
| 715
| 3,294
|
<issue_start>username_0: Can I provide two plots made using two different softwares in a scientific paper? For example, one is plotted using Origin and another is plotted using Python?<issue_comment>username_1: It is not unusual to have plots created with different software if they are used to display different kind of data. But I would not plot the same kind of data in the same paper once with one software and once with another (unless there is a very convincing reason to do so, in case the reason should also be provided).
Upvotes: 3 <issue_comment>username_2: **Q:** Can I provide two plots made using two different softwares in a scientific paper?
**A:** Unless a publisher (and their review process) tells you otherwise, you can do whatever you want for figures.
Ideally, the plots in your paper will be consistent formatting with each other (e.g., same tick styles, label fonts).
Personally, I have seen papers where plots come from different programs, and I could tell because the figures mismatched each other with style.
As long as the formatting of the plots does not distract from the science you are presenting, I would not be worried about the format.
*Thus, I think consistent and readable formatting is more important that the program used for your plots.
I would encourage you to focus on your plotting skills rather than the program used to create your plots.*
Upvotes: 0 <issue_comment>username_3: Using different applications for different plots is not uncommon, especially when there are collaborations among different groups which are accustomed to different applications.
For instance, I have a number of papers published with various groups across the world. I'm used to plotting graphs by using the workflow Matlab->matlab2tikz, whereas other coauthors of mine are more confident with Origin. For diagrams, I usually work with Inkscape, but other groups may have other preferences.
Most of the time, we didn't have any issue, but there have been rare occasions in which a reviewer (you know, that reviewer #2, the nitpicking one) asked us to make the diagrams more uniform across a paper. This request was typically not referred to plots but to circuit diagrams, and it was more related to the symbols used rather than the software used, but indeed it's easier to guarantee uniformity when the same software is used (in cases like this one, when I'm the first author, I usually volunteer to redraw everything in a uniform way with the same software).
Upvotes: 1 <issue_comment>username_4: It will depend on how the journal operates. Many journals will be able to accept images you create. Others will want the data for a graph to typeset it themselves. Others will accept images sometimes for some types of figures, and want to plot the data themselves for others. For images, there will be image file formats they can accept, and others they cannot. They will also have opinions on what format and resolution looks best.
Also, these things may change over time. A journal that acquires new publishing software may find that new sets of restrictions exist.
Contact the journal to find out what they can deal with, and what restrictions there are. They probably have a standard response for such questions, because they probably get it frequently.
Upvotes: 0
|
2023/02/02
| 887
| 3,545
|
<issue_start>username_0: Almost 2.5 months ago, I approached a professor in Japan to be my host supervisor for JSPS fellowship. I also sent him my research proposal and he asked me to do some correction.
In the mean time, I got busy with multiple other works (research paper writing, examinations etc) and therefore it is taking almost 2 months to give him my updated proposal.
My questions are,
1.Is it normal to contact host after such a long time?
2.If he has already decided to host someone else already (due to my delay), is there still scope for him to host another JSPS fellow (i.e. me ?)<issue_comment>username_1: Firstly, it is important to understand that there are 2 categories of JSPS Postdoc: one for Japanese applicants and another for foreign applicants.
UPDATE: The previous text was incorrect: according to last year's [FAQ posted on JSPS site](https://www.jsps.go.jp/english/e-fellow/faq/data/FAQ_e202212.pdf), it says there is only a limit in terms of simultaneous applications by one host researcher, but there is no limit to the number of Postdocs that a researcher can host:
>
> Q2002: If I, as a host researcher, am already hosting a fellow, may I
> also apply to host another fellow? A: Yes, you may. There is no limit
> placed on the number of individual fellows that a researcher may host.
> However, the hosting of multiple fellows should be done in a way that
> is not disruptive to the hosting system. As many researchers must be
> supported under the program with limited financial resources, limits
> are placed on the number of applications that may be submitted at one
> time. Please check the Application Guideline in advance.
>
>
>
Nevertheless, this fact does not have a significant effect on the rest of my answer: just because a professor is allowed to receive multiple people does not mean that they will necessarily accept anybody. Some professors may decide that they only have time, resources and patience to accept one single person, or even no one at all.
In fact, I know one case where a professor had already accepted one JSPS applicant, and refused to accept anyone else from a different category, unless the applicant was strongly recommended by some trusted fellow colleague.
To be blunt, you should not have taken so long to reply. At the very least, within the first month you should have sent a short email to apologize for the delay, and give an approximate timeframe for submitting the revisions. By waiting so long, you are at risk of sending a strong message to the professor that you have given up, or that you are not taking the JSPS application very seriously, or that you have moved on to something else. Just because the professor has discussed the proposal with you in the past, this does not give you any excuse to neglect your duty to keep him informed about your situation. In any case, you should urgently contact the professor, give a sincere apology for the delay, and ask if he is still able to accept you as an applicant.
The only person who can confirm whether you can still be accepted as an applicant is the host professor, so you should contact the person as soon as possible.
Upvotes: 2 <issue_comment>username_2: The other answer is not correct, the [JSPS application guidelines](https://www.jsps.go.jp/english/e-fellow/application-23.html) explicitly say that
>
> Each applicant (host researcher) may submit up to three applications. However, if an applicant submits more than one application, s/he should assign them a priority order.
>
>
>
Upvotes: 2
|
2023/02/02
| 789
| 3,004
|
<issue_start>username_0: I am an junior undergraduate in the U.S, and Interested in HEP "Theoretical High Energy" in Physics. I have had some exposure in research (i.e. Quantum Field Theory, General Relativity), besides having a GPA around *3.67*/*4.0*.
I am currently working with a faculty, on Experimental *Particle Physics*, but would like to study its theoretical side, in the future, as well.
Further, I have tried to reach out in my university about this topic, in particular. I could not find a research vacancies, Unfortunately.
What should I do to secure a place in research? If I am unable to do something related to theoretical HEP *( while I am still an undergraduate)* is it possible to do that in grad school?
Edit:
Thank you all for your recommendations! I have tried to reach out a faculty again, and he said he could be my instructor over the summer!<issue_comment>username_1: You get to do research in the way you tried: by asking around.
Given the highly mathematical nature of this field, I’m sure you realize that very few faculty will want to invest time and money to hire an undergraduate when these resources can be better invested in graduate students (especially students with a much broader exposure to mathematics than you).
In other words, it’s unlikely that a summer position for an undergraduate would be productive to a faculty member in this area.
I’m not sure why you worry about getting into grad school: it’s putting the cart before the horse. Just complete your degree and apply to graduate programs at the same time as anyone else.
It’s likely you may have to complete a year of courses before anyone in THEP commits to you anyways, so you should concentrate on doing well and look at experimental HEP as a way of broadening your perspective on this field.
Upvotes: 1 <issue_comment>username_2: Consider it from another angle. "There is a single light of science, and to brighten it anywhere is to brighten it everywhere." [<NAME>imov](https://www.brainyquote.com/quotes/isaac_asimov_101762#:%7E:text=Isaac%20Asimov%20Quotes&text=There%20is%20a%20single%20light%20of%20science%2C%20and%20to%20brighten,is%20to%20brighten%20it%20everywhere.)
Find any aspect of research that any of your profs will let you participate in. At your level, pretty much any sort of research will be good for you. It will teach you things. It will look good on your CV. It will build relationships with other researchers. You will see how these things are done, what the mundane aspects are, and so on. You might even get your name on an article as a co-author. And the prof involved will be a good source of letters of recommendation when you apply for grad school.
And don't be concerned if you get to do some mundane task. For example, extracting data and making graphs or some such. Or standing around running some bit of equipment or something. Science is not all glamor and AHA! moments. A big part of it, even for the most senior people, is turning the crank.
Upvotes: 2
|
2023/02/03
| 3,212
| 12,685
|
<issue_start>username_0: I need to resubmit a revised version of my paper to a journal. While submitting, I was asked to provide a list of responses to the reviewers' comments. I've started wondering about an accepted way of organizing these responses.
Should it be a text in free form? Or a bullet-point list?
Given the reviewers' comments in a free form, how would they know to which comments I provide a response? Should I copy the text of the reviewers?
Also, as for responding to changes in the manuscript, what tense is usually used? For example, how should I say that "The introduction has been/is/was modified ..."?<issue_comment>username_1: You can decide all of these things according to your personal preference. Enjoy.
Upvotes: 1 <issue_comment>username_2: Although there are no rules (unless there are rules... read the decision email properly!), I'd always copy the review text, normally without leaving anything out (the reviewer may not like the impression that you try to sweep something under the carpet), and respond to any specific thing directly after. For example:
>
>
> >
> > Symbol x has not been defined.
> >
> >
> >
>
>
> Done now on p. 4, l.11.
>
>
>
> >
> > I suggest shortening and re-organising Chapter 4.
> >
> >
> >
>
>
> We have removed the last paragraph, however we think that the current
> organisation is fine because (...), and the reviewer hasn't explained
> what kind of re-organisation they want.
>
>
>
(End example.)
As reviewer and editor I find this way of doing it most easy to handle. I have seen authors adding something like "Thank you for this important remark" to every single response, which is bad as it makes reading more difficult. But starting with a general "We'd like to thank the reviewers for their valuable suggestions that led to an improvement of the paper" note can't hurt. Regarding the use of tenses I'm in favour of "The introduction has been modified", but then I'm not a native speaker. Also language errors in the response to reviewers are hardly important.
Upvotes: 6 [selected_answer]<issue_comment>username_3: The general rule, similar to any document, is to make it easier for the reader to find the necessary information.
To this end, I use a different color for reviewer comments and my responses. I use a bullet list to highlight key facts. Further, I include relevant background from the paper in my responses; this ensures my responses are self-contained.
Upvotes: 2 <issue_comment>username_4: While I agree with username_2's answer, I would like to focus on one specific aspect of your question which I believe is the crux---you mention that the reviewer's comments are in the form of free text (not structured as a list).
My technique for this situation is to convert the free text into a list and then address the list items.
Say reviewer 1 submits:
>
> Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do
> eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad
> minim veniam, quis nostrud exercitation ullamco laboris nisi ut
> aliquip ex ea commodo consequat. Duis aute irure dolor in
> reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla
> pariatur. Excepteur sint occaecat cupidatat non proident, sunt in
> culpa qui officia deserunt mollit anim id est laborum.
>
>
>
and reviewer 2 submits:
>
> Sed ut perspiciatis unde omnis iste natus error sit voluptatem
> accusantium doloremque laudantium, totam rem aperiam, eaque ipsa quae
> ab illo inventore veritatis et quasi architecto beatae vitae dicta
> sunt explicabo. Nemo enim ipsam voluptatem quia voluptas sit
> aspernatur aut odit aut fugit, sed quia consequuntur magni dolores eos
> qui ratione voluptatem sequi nesciunt. Neque porro quisquam est, qui
> dolorem ipsum quia dolor sit amet, consectetur, adipisci velit, sed
> quia non numquam eius modi tempora incidunt ut labore et dolore magnam
> aliquam quaerat voluptatem. Ut enim ad minima veniam, quis nostrum
> exercitationem ullam corporis suscipit laboriosam, nisi ut aliquid ex
> ea commodi consequatur? Quis autem vel eum iure reprehenderit qui in
> ea voluptate velit esse quam nihil molestiae consequatur, vel illum
> qui dolorem eum fugiat quo voluptas nulla pariatur?
>
>
>
You can break the sentences into a list like so:
>
> R1.1 Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do
> eiusmod tempor incididunt ut labore et dolore magna aliqua.
>
>
> A: Your response
>
>
> R1.2 Ut enim ad minim veniam, quis nostrud exercitation ullamco
> laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor
> in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla
> pariatur.
>
>
> A: Your response
>
>
> R1.3 Excepteur sint occaecat cupidatat non proident, sunt in culpa
> qui officia deserunt mollit anim id est [here I used square brackets
> to clarify some term that the reviewer used that doesn't make sense
> out of context].
>
>
> A: Your response
>
>
> R2.1 etc
>
>
>
The power of this method is that it allows you to swap around the order of the reviewer's comments, or to group comments together, or even to group together a comment from reviewer 1 with a comment from review 2. It is highly flexible. The only rule is that you must include all text of the reviews. This technique has been highly valuable whenever I received a long rambling review.
Upvotes: 4 <issue_comment>username_5: I agree with the answer by [<NAME>](https://academia.stackexchange.com/users/110247/) --- it is best to copy the entire referee report and respond directly to all comments that require a response. I highlight all the referee comments in a different colour (and use black for my own responses and contextual information) so that it is clear which parts are from the referee and which parts are my responses. To add further to that answer, I will note that I use a practice of giving a simple bold statement of the general tenor of my response (agree, disagree, partially agree, revised, no revision, etc.) and then I elaborate more on what I think of the referee suggestion, what I have done, and why. This helps the reviewer see up-front where I have revised things and where I disagree with them or chose not to revise. Typically I have four types of main responses:
>
> **Agree - revised:** Used when I agree with the referee and have revised accordingly.
>
>
> **Partially agree - revised:** Used when I partially agree with the referee and have revised accordingly.
>
>
> **Disagree - no revision:** Used when I disagree and see no way to improve the paper from the comment.
>
>
> **Disagree - revised anyway:** Used when I disagree but the point raised in the disagreement allows me to improve the paper (e.g., tighted up my own argument to head off a misunderstanding).
>
>
>
Here is an example of what a response to referees might look like (using italicisation of referee comments, since I can't change their colour):
>
> *The paper is well-written and I like the writing style of the author. However, I am concerned that the introductory parts lack sufficient literature review and they author does not do a good job of making it clear what parts of their work are novel and what parts are existing knowledge. I recommend that the author provide a more comprehensive literature review and be more specific about the novel aspects of the contribution of the paper.*
>
>
> **Agree - revised:** I am glad the referee liked the writing style of the paper. I have now revised the introductory section (Sec. 1, pp. 1-3) to provide a more comprehensive literature review. Specifically, I have added reference to James *et al* (1981), Harrington (1983), <NAME> Wang (1990), <NAME> Evans (1991), <NAME> Watts (1992) and Bonjorno (2013). I have also revised aspects of the paper to be more explicit about the main contribution of the paper. I have added a paragraph discussing the contribution of the paper on p. 3 and I have also revised Section 3 of the paper to say which of the theorems I give are new and which are repititions or variations from existing literature. This new material should allow the reader to see which parts of the paper are novel contributions and which parts are contextual information using results from previous literature.
>
>
> ...
>
>
> *I also found the presentation of the simulation analysis on p. 16 to be unusual.
> Usually a simple barplot would be used. The violin plot used in Figure 4 is unusual in this area and it has not been used in other literature in the field. I recommend using a simple barplot instead.*
>
>
> **Disagree - no revision:** I have constructed a barplot of the data to see if it is an improvement over the violin plot used in Figure 4. While I appreciate the suggestion of the referee, my view is that a mere barplot of the data obscures important information about the shape of the distribution that is discussed in pp. 16-17. The existing violin plot already has an embedded barplot that shows the median, quantiles, etc., and this is noted in the caption for the figure. The use of the violin plot shows the estimated shape of the distribution in greater detail than a barplot would, and this is necessary for my later discussion (e.g., my discussion of bimodality, etc.). I am aware that other literature in this field has typically used a barplot for this kind of simulation analysis, but I consider the use of a violin plot (with an embedded barplot) to be an improvement. For this reason I have not revised this figure.
>
>
>
Upvotes: 3 <issue_comment>username_6: Unless the journal specifies a particular format, it's almost entirely up to you.
Here's a formula that has worked well for me. The goal is to convey that a) you've taken the feedback seriously and b) show how the manuscript has changed in response, so that the decision-makers can quickly and easily accept it--hopefully!
**Start with a short summary** of the reviewers' comments and the changes made in response to them. This should highlight any major themes: "All three reviewers asked about X, so we performed experiment Y and analysis Z; two reviewers asked us to comment on A in light of B, so we've added a new section to the Discussion on page 9". This does not need to be comprehensive, but should serve as an overview for the editor.
Next, go **point-by-point through each reviewer's comments**. I reproduce whatever they've written (in a distinctive font/color), and then write my response to that point below it in another style. These responses vary in length: a simple "fixed" is fine for comments about typos or minor errors. A major point might require an entire page, possibly with a bespoke figure. If you have changed the manuscript in response to address this comment, **include the location of the changes** (e.g., "Discussion, p. 9--10"). If the change is 1--2 paragraph or less, **quote the revised text in the response** too.
If the review is "free-form" text, break it into discrete "issues" however you see fit, but do not remove or paraphrase text and try to keep the points in the same order. Some people number them, using R*X*.*Y* to refer to the *y*th comment raised by Reviewer #*x*.
**Write something for every comment.** If you feel certain that the reviewer is incorrect, it may be tempting to rant about their apparent ignorance. I would avoid doing this unless it's an objective, factual error. Instead, consider the possibility that there's a germ of a good idea, badly expressed, or that the writing failed to properly convey your own brilliance. At the very least, your text can often be revised to make points more clearly.
Try to **make each Reviewer's section self-contained**. If a reviewer repeatedly returns to the same point, it's fine to occasionally say "As mentioned above in R1.4", especially within a single reviewer's section. If several reviewers ask for a small change, I would just list them in each section. Prefer repeating "We now include the participant's ages, heights, and weights on page. 9 [quote]" to "See R1.3." For longer responses, references are fine but if multiple reviewers have the same major comment, address it in your overview. Some of your responses may rely on specific data from references. It's sometimes helpful to include relevant quotes or figures from those. Give the citation as well, obviously.
Here's a [complete example](https://doi.org/10.1371/journal.pbio.3000834.r004) of this format from one of our papers. A growing number of journals now publish peer review files, so it might be worth checking whether examples are available from your field or target journal.
Upvotes: 1
|
2023/02/03
| 302
| 1,311
|
<issue_start>username_0: If a postdoc receives a conditional offer of employment, is it still a good idea for them to tentatively accept any further interview offers from another university which come during the waiting period, basically as a backup?
I suppose if one is committed to the first position, it won't offend them if the candidate is theoretically still interviewing?<issue_comment>username_1: It is fine to continue interviewing until you accept an offer. There shouldn't be any ethical issue with that. You might leave yourself vulnerable to the withdrawal of an offer (some situations, some places) or not.
You can even continue the negotiation (better salary) if you like, though that carries risks as well. You can also ask for an extension of the offer. But you aren't bound by the acts of others; the offer.
Don't worry about "offending" anyone. That is an orthogonal issue. You have a responsibility to yourself until you make some commitment to another. Only then does it bind you in any way.
Upvotes: 2 <issue_comment>username_2: Of course you can continue interviewing. Even if you have an offer they could still withdraw it (happened to me!). And who knows ... maybe you will even get a better offer.
But once you signed the contract you should stop interviewing elsewhere.
Upvotes: 2
|
2023/02/03
| 768
| 3,467
|
<issue_start>username_0: We recently asked several manufacturers if they could send us a sample of their material(s) for us to test towards their suitability for a process we developed. They sent us free samples. Once we identify the most suitable material we plan on ordering more from that material.
After receiving the materials, we decided to expand our testing towards optimizing said process and we have enough content for a publication. I am now wondering if we should (or have to) thank the manufacturers in the acknowledgements for their provision of free materials, even though we did not explicitly ask them for materials for the purpose of the publication.<issue_comment>username_1: I suggest that you ask each of them whether they would like to be acknowledged or not. Some might not want to be named, especially since you didn't make your intention about the use clear. Many corporations are (very) sensitive about being named in any way that might affect their brand.
If some say yes and some no, then you have a dilemma and you might have to be clear in an acknowledgement that there were others who prefer not to be named.
In general, however, it is good to acknowledge such contributions to a study but also to make the use clear and that the research will likely result in publication.
Upvotes: 3 <issue_comment>username_2: You should definitely mention this in the paper. There is a conflict-of-interest concern here so that you should be as transparent as possible. This arrangement is very common with pharmaceutical manufacturers and academic research.
Whether you do it in the acknowledgement or within the methods is really up to you, I'd probably put it within the materials and methods, as part of the ordinary recognition of where materials come from. E.g.:
>
> Basket-weaving materials (straw, Intl Straw Corp, Des Moines, IA, USA; wool, Sheeps Down Under, Wagga Wagga, NSW, Australia) were donated by the manufacturers.
>
>
>
In addition, you may want to outline the scope of the conflict in a separate disclosure section, and note that while materials were provided for free the companies involved had no input on the design, execution, or presentation of the results, assuming that's all true. Remember, "conflict of interest" doesn't mean your work is corrupted, it just means that interests conflict, and the best way to defuse any perceived issues is to be transparent.
You may also want to address whether this was an appropriate use of the materials from the perspective of the manufacturers. Presumably they expected you were going to use the materials to decide whether to purchase more: they're providing them as a sales tactic. Instead, you've created a product (research paper) out of the freely given materials, and that product could reflect positively or negatively on the people providing you materials. As you can imagine, someone could be quite upset if you've used their free gift to explain to everyone that their product is inferior to their competitors. If you've signed any agreement about the transfer of materials I'd first check whether that agreement has any statements about how the materials can and can't be used, and if it's unclear or unstated, it's probably worth reaching out for permission, or just asking to pay for the value of the samples if that's feasible. <NAME> recommends in a comment to consult with your institution's legal team and I think that's a very good idea.
Upvotes: 4
|
2023/02/03
| 3,964
| 16,530
|
<issue_start>username_0: I am a graduate student in some subdiscipline of mathematics and have had the following impressions. One of the senior faculty members (a professor) at my (not too famous) university mentioned, when talking about famous people in the field (high positions at top universities, etc.), that it is very important "not to rub such a person the wrong way". Another senior faculty member mentioned that it is very important to be political when citing the same type of people and giving very much credit and in particular that one does not criticize errors in their papers (in one case the error was anyway corrected in a subsequent paper).
To what extent are they right in saying/doing this? Can such a person ruin your reputation/career? Are there examples of this?
I understand that the answer to these questions are in principle "yes", but would like to have an elaboration if possible and also some examples of this.<issue_comment>username_1: My experience (in North America) in dealing with superstars is that most of them do not care about being cited: they know where they are in the queue, others know where they are in the queue, and citations do not matter to them.
(There are exceptions of real obnoxious superstars but they are thankfully rare.)
The more toxic people tend to be amongst the wannabes below superstar level. Some of these guys feel their contribution are unfairly ignored or not recognized properly. These are a dangerous tribe and they will not hesitate to step on others to climb up the ladder of success.
I do not think it is more likely that famous rather than just regular obnoxious people will actively seek to ruin the career of random persons. It’s simpler for most brilliant people to just ignore the plebs.
This being said, I would double and even triple check before pointing an error in *any* paper: I have found over the years that in most cases it was I who did not understand the work properly (which can be quite poorly written). It is unlikely that these errors are *serious* as the work of such superstars is quickly read by others and such mistakes quickly corrected.
Upvotes: 5 <issue_comment>username_2: I may not answer for Mathematics specifically, but I would be generally less pessimistic about the politics in academia. But there are many layers to peel here.
First, "do not try to challenge famous researchers" may be a useful heuristic. They became famous for a reason, and lots and lots of resources were spent following in their footsteps. Challenging some of their results without a strong attack is unwise, and going after foundational principles of any field will cause an instinctive rejection by the community, as it have many times in the past. On the other hand, no guts, no glory - if you DO have a strong attack on a problem, it may very well be the road to fame. Planning research around having a breakthrough is not particularly feasible, however.
Second, in some parts of the world, challenging the authority has broader cultural implications. This is far from universal, but (allegedly) there are places where it may well be a career suicide.
Third, if the previous point does not apply, I would call "giving very much credit" either a poorly worded or poorly understood (either by you or a person who stated that) approach. See, "famous authors were all wrong" implies that anything based on that paper is invalid, and that stance loses immediately. Instead, try to reconcile the development of the field with your newfound results. It may not be that important to pay tribute to specific **people**, but failure to recognize the value of a significant fraction of **research** done in the subfield would easily lead to the paper being rejected, ignored, or even labeled as crankery. Starting off by antagonizing the very community receiving this work is very counter-productive, and not being sufficiently specific with the critique may, unfortunately, trigger certain heuristics in fellow researchers that will place your paper straight into the trash bin. "Giving credit" may have the same presentation as "reconciling the differences", but a very different driving force behind it. The latter tries to be impartial even towards oneself, the former makes scientific merit highly subjective and, plainly speaking, feels *wrong*.
And finally, there is certainly a component of having to work around personalities in academia. The higher up the ladder, the more political it gets. In that sense, yes, famous people wield a lot of power, and if you try to get a big project going, you *have* to consider who will be in the committee, what they believe has merit and what does not. There are not enough funds going around to pursue every venue meaningfully, and attempts to divert more resources than just venture capital from what seems to be working great otherwise would not be received favorably. Convincing other researchers that what you are working on is important is a substantial part of being an academic, and this is especially significant in mathematics as [this acute observation](https://academia.stackexchange.com/a/167041/145124) by <NAME> attests.
Upvotes: 3 <issue_comment>username_3: #### Be charitable to *all*, without special dispensations for people of high status
It is inherently difficult to predict the future behaviour of other people you don't know, but the mere fact that there is a reputation that is spread around warning you against valid criticism of the work of others is itself worrying; is an indicator of unheathy behaviour and scientific corruption in academia. Irrespective of whether vindictive behaviour responding to a correction is likely to occur, I don't think it is healthy to approach academic work in a political manner where you formulate your own behaviour in order to accommodate the presumed or suspected vindictiveness of others, based on their power and status in the profession. If others *are not* going to act vindictively then there is nothing to worry about, and if they *are* going to act vindictively then you should be bold and fearless and proceed in the spirit of open scientific criticism anyway. Therefore, I would counsel against any attempt to speculate on or predict vindictive behaviour by high-level academics; just be faithful to the spirit of scientific inquiry and professional courtesy and take these as your guide for behaviour.
Academics with high status should not get any special treatment here, but obviously you ought to show some appreciation for the fact that they have made valuable contributions to the discipline and put any errors in perspective. You should also approach critiques of the work of *any author* with a level of good faith and charitableness appropriate to professional research. People who do lots of work in a difficult field are going to make lots of errors in the course of doing this work, and occasionally these errors will sneak past discerning eyes, including in the peer review process. If you find errors in the works of other mathematicians (whether they are famous or not) you should correct those errors as appropriate and proceed using the corrected version of their work. Typically if you are citing a work and using it but correcting an error then you would have a footnote noting the error and the correction but you would not make a big fuss about it or be harsh on the author --- just make the appropriate correction, make it clear to your reader, and proceed on the basis of the corrected work. Undertake your corrections in a calm and non-judgmental way that shows respect for the fact that the totality of the work of these other authors has positive value and merit despite occasional mistakes. You should exhibit this level of appreciation and charity when dealing with errors *from any of the practitioners in your profession, irrespective of their status*.
As to the question of the power of famous academics, they have quite a bit of power and discretion over others within academia (e.g., decision-making ability over hiring, promotions, reviewing, etc.), but this power is curtailed by ordinary protections for other workers, the tenure system, etc. A vindictive professor might be able to exact punishment on a lower academic for a perceived slight, but it would harm their own reputation if it became known that they act in such an unprofessional way.
Upvotes: 3 <issue_comment>username_4: As an addendum to the answers already given, it pays to be mindful that some academics can hold positions of various boards or committees that can either open doors to researchers or firmly close them with little explanation or oversight.
Or they may have the ear of the people who sit on those boards\committees.
This is particularly true when it comes to allocating grants or other forms of funding.
Notable academics can ensure that doors are opened to likeminded people, or people who they feel are expanding a particular field in the right direction. Often this is the direction that the academic is trying to take it. They can close them if they feel that you have the wrong mindset or are trying to take the field in the wrong direction.
This shouldn't be the case, and there are many academics who will be agnostic and open doors purely on merit. Knowing which type a person is can be important.
Upvotes: 2 <issue_comment>username_5: I like to think of the formal "powers" any senior person in a field might have and how that could enable them to harm another researcher. I come from the social sciences but I'm not sure that anything I say works much differently in other fields.
Such a senior person may be involved in various aspects of the process of hiring new colleagues in their department, either via serving on the search committee, serving as department chair, or merely being an influential voice in faculty meetings. So if you seek employment at such a person's institutions, they may hold some sort of bias against you. This could be out of malice or merely due to the underlying disagreement leading them to conclude you don't know what you're talking about.
They could also have various roles in the publication process, like serving as editor, associate editor, or peer reviewer at various important journals (or conferences). Of course, a peer reviewer does not know the identity of authors in many fields but I'm not sure about mathematics. At any rate, this person's influence in the editorial process could potentially close off some small number of avenues for publication.
I'm only familiar with the US system, but part of the process for gaining tenure here is that outside peers are invited to review your file. At many institutions that use these peers, even mild criticism from these peers can be deeply problematic for the candidate's chances at tenure and promotion. Should you be so unlucky, they could cause harm in that way. But the chances of them being invited for this purpose are low and I believe it would be considered inappropriate if you had some sort of public intellectual dispute with them.
The more informal influence such a person could have is by generally using their social status in the discipline to cast doubts about you or your work. I'd say that in general talking negatively to peers about another researcher would come off as very uncouth among the people I spend any time around. On the other hand, criticizing *work* can be very appropriate under certain circumstances although I'd admit it is not often that I find myself in situations that would open the floor for someone to start badmouthing some non-high profile research.
Besides the above, there's always the potential of the more juvenile and pernicious forms of rumoring and attempts at retribution that any person might try to do. But I don't think many (or any) people have the power to sort of push the red button that destroys some junior person's career. In fact, the most likely person to be able to do that is a person in a graduate student's department. Even then, we know many cases where apparent bad faith by an advisor/mentor/etc. fails to completely derail a researcher's education/career.
Upvotes: 2 <issue_comment>username_6: Personal view on these issues is that the problems fall into several categories:
* Economic Issues: When many agencies ([NSF](https://www.nsf.gov/pubs/policydocs/pappg22_1/pappg_3.jsp#IIIB), [NIH](https://grants.nih.gov/grants/peer-review.htm), ect...) choose external reviewers, they are often attempting to find known "subject matter experts" in the field, and therefore, a famous or highly cited author in the field is "probably" more likely to be chosen. Therefore, a famous or highly cited author is more likely to have influence over your financial funding future.
* Social [Network Effect](https://en.wikipedia.org/wiki/Network_effect): Generally, famous academics have more friends, ties, and influence in the community (similar to social media followers) and whether good or bad, their comments often carry disproportional weight. If a famous author criticizes your work, it can be more damaging than if a "nobody" criticizes your work.
* [Social Multiplier Effect](https://en.wikipedia.org/wiki/Social_multiplier_effect): Peer effects are captured by the sum of friends' efforts in some activity. An individual pays a price for deviating from a norm. In the latter, as an individual gains more peers who have a certain attribute, they experience greater utility for adopting this attribute. "Act like us, we'll all do better."
* [System Gaming](https://en.wikipedia.org/wiki/Gaming_the_system): Not all, yet many academic positions have become extremely competitive, and as the [Guardian notes "academics is a highly incentivized game"](https://www.theguardian.com/higher-education-network/2018/feb/16/performance-driven-culture-is-ruining-scientific-research). When faced with this competition, many turn to gaming the system. This is a well known criticism of [author level metrics](https://en.wikipedia.org/wiki/Author-level_metrics#Criticism). However, the same applies with a social network. The equivalent of "don't insult them, they're a 'high-level' player."
* Viral Infectiousness: Related to the network effect above, a famous criticism has more pathways to infect someone's career. And if it represents something good for gossip, or viral sharing, then it may spread pervasively. "did you hear what so-and-so said about them?"
* [Rational Ignorance](https://en.wikipedia.org/wiki/Rational_ignorance): In the face of evaluating potential threats to their academic career, a busy academic can't reasonably be expected to evaluate all the possible ways critiquing someone might be damaging. So you get rules of thumb, 'very important "not to rub such a person the wrong way"'
* Anonymous Denial: From many sources of funding, the reviews, and even the denial of funding, may be completely anonymous. Even if there is a conflict of interest, from the person being denied's perspective, that can never be identified. Per the economic section above, famous are already more likely to be chosen to review, and there's no way for the submitter to note that the only harsh review was from someone they criticized.
"Are there examples of this?" While neither of the examples below are specifically of the "famous" case, there are numerous examples of the type of behavior that then leads to the pervasive fear of criticizing anyone of influence:
* [Universities to Fear](https://academicjobs.fandom.com/wiki/Universities_to_fear) (Anonymous Issues with Schools) - Many are complaints about interviews, yet several illustrating the social issues above.
* [Academia is Killing My Friends](https://academiaiskillingmyfriends.tumblr.com/) (Anonymous Stories of Academic Abuse)
An [Inside Higher Ed Article](https://www.insidehighered.com/advice/2021/11/18/truths-about-academic-career-people-often-dont-share-opinion) from <NAME>, also has several concrete examples of what leads to this behavior, especially the "Egos are huge. People may turn on you" section.
Note the pattern in the sites, there also appears to be a strong trend in current academia that it is not safe to talk publicly. The following article notes this directly:
* [The Young Academic's Twitter Conundrum (The Atlantic, Olivia Bateman)](https://www.theatlantic.com/education/archive/2017/05/the-young-academics-twitter-conundrum/525924/) - “You can use the story, but don’t mention my name. I don’t want any trouble.”
Upvotes: 2
|
2023/02/03
| 1,368
| 5,614
|
<issue_start>username_0: I am a finishing graduate student after, what I personally would say, a rather mediocre PhD (mathematics).
There has been some personal issues (mental health issues) and so I could not perform to the utmost of my abilities. The PhD will, however, still very likely be finished. The mental health issues have improved.
Regarding the question of continuing with research my supervisor has claimed that there is an agreement with another university to trade finishing PhD-students for post-docs and my impression was that a post-doc was essentially a guarantee. (However other faculty members do not seem to believe that I am cut out for this and the argument would be that my supervisor, having had no prior graduate students before me, is not experienced enough to accurately decide this).
So I ask:
When is it not a good idea to continue with research after the PhD? Are there any traditional red-flags?<issue_comment>username_1: I'd say the traditional red flags are:
* Poor performance in PhD
* Discouragement from supervisor (or anyone more experienced than you are and are familiar with your work)
While these are obvious, I'd also add:
* Unfamiliarity with the non-academic job market
If you pursue postdocs while unfamiliar with the non-academic job market, it could be a sign that you are doing postdocs without having really thought about it, and you are avoiding the "real world" not because it is unsuitable, but because you don't know what it's like.
See also the answer I wrote to a [related question](https://academia.stackexchange.com/questions/125785/getting-a-job-with-a-phd-in-pure-mathematics).
Upvotes: 2 <issue_comment>username_2: I suggest that you don't consider a postdoc as an end in itself. It isn't a career. It is a useful stepping stone for many who want an academic career and need to both stay relevant and increase their productivity *while they seek a permanent position*.
I also suggest that you try for permanent positions and see what might be available to you, considering the overall life of an academic, both teaching and research. But also, follow up on postdoc opportunities in case the job hunt is difficult, as it is for many at the moment.
Don't turn down a postdoc leaving you with *nothing*, however. If you wan to stay in academia then it is a good and potentially productive interim step. But the real goal is a permanent, tenure track, position (or the equivalent if this isn't the US).
When the job market is tough, people have to make some compromises in the short term so that they can continue to work toward long term goals.
---
And if you want an industrial position instead, search in that market and use a postdoc as a filler.
Upvotes: 0 <issue_comment>username_3: First of all, a postdoc position is not the only way to continue research. There are many research institutes except universities that can contribute to your CV even more than a postdoc.
As for your question, I think it is not a good idea to start as a postdoc if
* **The lab is too small.** For instance, if it is only two postdocs and the supervisor, of course depending on the area, the productivity might not be what you expect.
* **The collaboration histroy of the lab is not rich.** If the previous postdocs have a portfolio of publications only with the lab, then chances are you won't have collaboration and networking opportunities.
* **The university is a lower-ranked university.** This might be a bit controversial but it is what it is. If a university is not very high-ranked researchwise, then the chances are you will not find many opportunities to improve yourself during your postdoc.
* **The supervisor's publication record is subpar.** I think the main reason to work as a postdoc is doing research, and the output of the research is published papers. Considering that, if your supervisor has a subpar publication record, then they either do more teaching than research or they simply cannot obtain good results. In both cases, your postdoc won't be as productive as it should be.
I also want to note that postdoc is a risky position. It is generally considered as a mere internship position. The longer you continue to be a postdoc, the lower your chances to land a permanent job. So, if you are not absolutely sure that your postdoc will benefit your career dearly, then it is better to avoid it.
Upvotes: -1 <issue_comment>username_4: First, here's hoping your mental health issues can be improved. Best wishes.
You should think about the post doc as a stepping stone to an academic career. If you are keen on being an academic of some kind then pursue it.
If you are thinking that academia and you have spent enough time together, there is no shame in it. If that's the case, seek a job outside academia. There are lots of PhDs in math working in various industries. From finance to aircraft design to pharmaceuticals to you name it. Your PhD will be proof that you can handle complicated ideas and produce creative solutions. It won't be a "snap" to get a job, but it is certainly possible.
This is not necessarily an easy decision. But consider. If you do switch to industry of some kind, then a year or two in a post doc position means that amount of time you could have spent getting experience outside academia. You could be making contacts, learning the new context, etc. And quite likely earning a paycheck at least as large as the typical post doc.
So the decision is, stay in academia or leave academia. Inside, look for a post doc. Outside, look for a job that will make use of your talents and skills.
Upvotes: 0
|
2023/02/04
| 1,407
| 5,572
|
<issue_start>username_0: [Sci-hub](https://en.wikipedia.org/wiki/Sci-Hub) and [ResearchGate](https://en.wikipedia.org/wiki/ResearchGate) both provide (among other things) online access to journals that sit behind publisher paywalls. For example, I have just discovered, not unhappily, most of my own papers on both sites despite the copyright being held by the original publishers and despite none of the publishers apparently having allowed free access to the papers.
Yet, despite their similarities ...
* Sci-hub seems to be regarded by publishers as a rogue site that ought
to be shut down (consider, for example, multiple news stories about
[Elsevier's attempts](https://www.nature.com/articles/nature.2017.22196) over the years to recover damages from Sci-hub or to prevent its websites from being [DNS indexed](https://torrentfreak.com/domain-registry-takes-sci-hubs-se-domain-name-offline-230126/) or [publicized](https://www.science.org/content/article/twitter-shuts-down-account-sci-hub-pirated-papers-website))
* ResearchGate is either well-regarded, or at least tolerated .
Have I understood the situation correctly, and if so, what has led to the current state of affairs?
**Added**: To remove doubt, I'm not asking why it might be acceptable to publish a preprint with the permission of the publisher but unacceptable to publish the final paper without the permission of the publisher. I'm interested why comparable behavior on the two sites is apparently treated, and regarded, differently.<issue_comment>username_1: **ResearchGate is legal, Sci-Hub is not**
How is that possible when they both "provide access to paywalled papers"? That's when you need to know what exactly is copyrighted and cannot be shared.
[Elsevier sharing policy](https://www.elsevier.com/about/policies/sharing) (which is typical of publishers)
>
> **Preprint**
> Authors can share their preprint anywhere at any time.
>
>
> **Accepted Manuscript**
> Authors can share their accepted manuscript:
>
>
> Immediately
>
>
> * via their non-commercial personal homepage or blog
> * by updating a preprint in arXiv or RePEc with the accepted manuscript
> * via their research institute or institutional repository for internal institutional uses or as part of an invitation-only research collaboration work-group
> * directly by providing copies to their students or to research collaborators for their personal use
> * for private scholarly sharing as part of an invitation-only work group on commercial sites with which Elsevier has an agreement
>
>
> After the embargo period
>
>
> * via non-commercial hosting platforms such as their institutional repository
> * via commercial sites with which Elsevier has an agreement
>
>
>
It should be obvious that ResearchGate generally does not violate these guidelines (and if they do, a DMCA takedown request gets them to remove the offending copyrighted material).
This is in stark contrast with Sci-Hub, which is a pirate website that treats copyright laws with contempt.
**Also**: the methods Sci-Hub uses to get their papers are dubious at best (it's possible some sympathetic researchers donate their institutional credentials to Sci-Hub, but that still almost surely violates their institutional policies) and [outright illegal at worst](https://skeptics.stackexchange.com/questions/50490/does-sci-hub-use-malware-and-phishing-to-obtain-researchers-login-credentials). ResearchGate's papers are acquired legally.
The two websites are not comparable.
Upvotes: 2 <issue_comment>username_2: Sci-hub allows access to pay-walled articles. They are like the 'robin hood' of academia. Hence, they have many fans, especially among institutions who do not have the funds to subscribe to journal databases and proponents of open access. Further, people who access sci-hub tend to have access to pay-walled articles. I for example use sci-hub to get an easier access to articles; my university's library access sucks. Hence, publishers do not really care much. This is because those who can pay are already paying, and those who cannot, they are never a customer.
Researchgate has a bad reputation. It has questionable practices. For example, it spammed my co-authors using my name hoping to increase their subscribers. I also suspect that many subscribers who follow me are fake.
**Additions**: Elsevier is a profit driven company; search for their annual revenue. ResearchGate allows publishers or copyright holders (or authors) to request their works to be removed from ResearchGate. Further, many publishers, such as IEEE, allow preprints to be posted as long as preprints have a blurb that indicate their copyright owner. Hence, ResearchGate arguably is nicer to publishers and/or do not infringe on publishers' rights. On the other hand, sci-hub is setup to get around a pay-wall, meaning it is a threat to publishers.
Upvotes: 2 <issue_comment>username_3: One answer to your distilled form of the question (from a comment)
>
> why is ResearchGate apparently tolerated by publishers whereas Sci-hub is relentlessly pursued, at least by Elsevier?
>
>
>
might be the following. ResearchGate, at least in my experience, do not put full-texts of paywalled articles on their website without consent from at least one of the authors. So it is the author who is at fault if this is done in violation of their agreements with the journal. It's really not much different to an author putting the published version on their personal website - the publisher doesn't go after the website host if this happens.
Upvotes: 2
|
2023/02/04
| 742
| 3,362
|
<issue_start>username_0: I submitted a paper to a journal a few months ago and applied for a research grant one year ago. Currently, my grant application has not yet been awarded, but I anticipate that it will be granted in approximately two months. The exact date, however, is unknown.
I intend to include my grant number in the acknowledgments section of my paper, as per the grant policy. However, my grant number is currently not available as my application is still in progress. Last week, I received a report from the journal indicating that only minor revisions are required and I have one month to submit my revisions.
**Question 1**: What should I do if my paper is accepted before I receive my grant number?
**Question 2**: Is it possible to ask the journal to add my grant number after the online version of my paper is published?
My questions may appear illogical or unethical, but let me explain how grants work in my location.
Applying for a grant is like playing a game. Once your grant application is successfully awarded, you are given a specific time period, often 1 or 2 years, to "finalize" it. If you cannot finalize your awarded grant within this time frame, you will not receive any funding.
To finalize a research grant, you must acknowledge the grant with a grant number in your published papers. The number of papers and the time frame in which they must be published are specified in the grant policy. If you cannot meet these requirements, you will not receive any financial support and will likely have a negative reputation in the academic community.
So I am following the rules and trying to meet them. If I were a very smart person who could produce many papers in the future, I wouldn't be worried about this matter.<issue_comment>username_1: You list grant numbers in the acknowledgments because the grant supported the work presented in the paper. But since you have not been awarded the grant, it cannot have supported the work in question. Consequently, you should not list the grant in the paper.
Upvotes: 5 <issue_comment>username_2: You should not acknowledge support from a grant for work done prior to the awarding of said grant, irrespective of the date you applied for the grant. The policy of all grants I know applies only for work done *during* a period covered by the grant.
In your situation it would be incorrect to put a grant number in your revised paper.
Upvotes: 3 <issue_comment>username_3: One possibility is to ask the journal for more time to submit the revisions (which they are likely to agree to since they probably have plenty of papers happy to have their publication dates bumped up slightly earlier). Then, if your grant application is accepted *and* you start receiving grant money while you are still working on the revisions, then the grant is supporting the paper and it is acceptable to acknowlege the grant in the revised paper.
I don't know if this is really a great choice, but it seems like an ethical choice to me (presumably you have plenty of other work you can be doing while you're waiting to do the revisions).
It's certainly not ethical (as others have pointed out) to list a grant that you haven't received or that didn't contribute to the paper; and the journal won't change any published version after it's published (that would be unethical on their part).
Upvotes: 2
|
2023/02/04
| 2,632
| 11,267
|
<issue_start>username_0: I'm currently a first-year Ph.D Student. My field is applied mathematics. Thankfully, I was able to publish my first paper. In our institution, students usually spend two years before having their first article published.
I heard from one of my colleagues, that his supervisor, a professor of pure mathematics in our department told him that that my field of research is "easy" and that literally anyone can have as many papers as they want published.
This isn't the first time I've heard this being said by this professor, and frankly it's beginning to bother me. Especially since I spend so much time and effort in my work. Don't get me wrong, I love my field, but I can't control the way this makes me feel, especially that having this information around the department and in conferences is not a good thing, since academics, other than the ones working in the same area, and especially pure mathematicians in our department, tend to believe whatever they hear, especially when it comes from a colleague-professor. I'm thinking of talking to the professor, in a conference or if I have the occasion, and try to have a conversation with them in the subject and why not convince them that this is not the case.
Is this the right way to deal with this problem?<issue_comment>username_1: I am currently doing my Ph.D. in pure Mathematics and I have frequently heard from professors of pure Mathematics that few fields (not all) in applied mathematics, as well as pure mathematics, are comparatively easy for publishing papers. So, based on that they have formulated their opinions. So, I would simply suggest you stay away from negative comments and look for positive criticism from colleagues. Also, focus on well being of yourself, your work and prioritize your mental peace.
I would also like to share quite opposite experience. I was asked about the impact factor of the journal that I recently published in by an applied mathematics professor and it was around 0.9 since we have journals with less IF compared to applied mathematics journals. They said that since I was at an early stage of my research that it was okay to publish in such journals but try to publish at better places in the near future, say for example in journals with an Impact Factor of more than 3. I was a little bit uncomfortable and try to explain to them that there are not many such journals in my field but they were quite adamant. So, I was a little confused and had a talk with my supervisor who cheered me up and explained to me how there are differences in the ranking of journals field-wise.
Upvotes: 3 <issue_comment>username_2: My advice is to ignore them and do your own work with supportive faculty. It is a common enough feeling among pure mathematicians (my field) that applied math is somehow "less" than pure math rather than different. The goals and techniques of applied math are not the same so they can't really be compared in that way.
A professor of, say, anthropology could make the same comments and they wouldn't bother you, I'd guess, though a conversation with them might be more fun and you'd learn something, perhaps.
But, you aren't going to convince them. The problem is that they haven't tried it and, perhaps, don't care a lot about applications, or the "real world" in general as a subject of research. Let it go.
My dissertation was so esoteric that I figured it would never be applied to anything useful. Beautiful and fun, but not useful outside pure math. It took about 40 years IIRC to be proven wrong.
As a student you are unlikely to have any positive impact and an argument with a professor isn't a good start to a career. Let the faculty argue amongst themselves about such things.
Upvotes: 5 <issue_comment>username_3: #### The level of difficulty of your work does not imply its value
You don't need to do anything here at all --- just keep researching in your field and publish papers that advance knowledge in that field. Take solace in the fact that the [labour theory of value](https://en.wikipedia.org/wiki/Labor_theory_of_value) is false, so the value of your research is not determined by the ease with which one can produce papers in that field. If this professor is correct in his evaluation then that just means that there is an unmined area of research where progress and publication of results can be made with relative ease. That is all the more reason that researchers *should* pursue this area of research, not a reason to ignore it.
Since you are just starting out, you should trust that the peer review system in your field will operate appropriately, and if you have a paper accepted for publication then that will generally mean that it is contributing something valueable, irrespective of how easy it was to do the work and get the paper accepted.
As a secondary point to put you at ease, you needn't worry about whether you are as good a mathematician as a senior professor, or working on problems that are as hard as those done by a senior professor, when you are only a first-year PhD student. I'll save you the suspense --- you're not. It is perfectly reasonable for you to do research in this field of interest to you, learn that field, and then move on to other work later in your career in the natural way that research interests evolve over time. You will probably find that, after a decade or two, you will have some good research experience across several areas of mathematics and you will have encountered problems of varying levels of complexity.
Upvotes: 5 <issue_comment>username_4: Looking at this from a slightly different angle, you said
>
> I can't control the way this makes me feel
>
>
>
In *every* field of human experience, the single biggest lesson that any therapist can help their clients absorb is that ***this statement is wrong and dangerous***.
Everyone who is not a sociopath has imposter syndrome to some extent, because everyone devalues their own skills to some extent. (That's not exaggeration, by the way - this is literally part of how sociopathy is defined and recognised.) So the fact you're experiencing this makes you *normal*. The problem then is that regardless of this being normal, it's unhealthy for you.
You can't control what other people say or think. To some extent you might be able to guide them in other directions, if you have the energy and willpower and they're receptive to it. If you're in a context where there are defined behavioural rules and they go so far as to break them, you can ask other people to intervene. But if you don't or they aren't, you simply can't change what they do.
The lesson from therapists is learning how to put aside criticism from people who don't matter. Does this person's opinion truly matter? Answer: no, not really, because you don't have to work with them. And anyone you do have to work with will be valuing your skills over his. It's not easy to do, because humans aren't wired that way. We're not cats. But it's a key mental health skill.
For myself, I'm an engineer. We've recently had to solve some problems with improving linearity fit methods to position measurements. We needed to work with someone with mad skills in applied maths for this, because they needed to appreciate that this was solving a real problem, with real data which contains errors. We didn't want a theoretician.
The professor is right that you can publish as many papers in applied maths as you want. Where he's wrong is thinking that it's easy. The reason you can publish so many papers is that there are so many real applications which would benefit from your skills, and they aren't easy because otherwise they'd already be solved!
Sure, maybe a purist can look down on this. But at the end of the day, you will have done something that makes a difference somewhere. Can he say the same? Or is he just jealous of the fact that he can't, and putting you down through his own imposter syndrome?
Upvotes: 3 <issue_comment>username_5: *Don't get me wrong, I love my field, but I can't control the way this makes me feel, especially that having this information around the department and in conferences is not a good thing, since academics, other than the ones working in the same area, and especially pure mathematicians in our department, tend to believe whatever they hear, especially when it comes from a colleague-professor.*
There is a lot in this sentence and between its lines.
Looks like you - and your pure math colleague, possible his professor too - haven't really yet made that mental and emotional shift to a graduate school mode of thinking about your work.
Grad school isn't about competing with other PhDs in your school not those in other schools. Grad school is about simply trying to advance your own and the world's knowledge of math in a single application of interest from where you were yesterday.
So:
1. Stop preening yourself - privately or publicly - on getting a first paper a year ahead of most others. This is just a bit of luck on your part - maybe your area is topical at present, maybe you have a respected supervisor, maybe there was a lull on submissions, etc. If you make odious comparisons, you invite others (and their supervisors) to wait in the long grass for you. Just look at how Dr <NAME> (3 year PhD, chem eng, steam condensation, Urbana-Champaign 1959) turned out . . . You don't want to be like him, I take it.
2. Stop using nearby colleagues' *apparent* progress as a gauge for your own or as emotional/social goalposts: it's time for you to start to build a circle of purely socially friendly acquaintances outside your own school for relaxation, a game of squash, a hike or whatever. We all need external human supports and we all have a desire to support/nurture others in the same way.
3. Under **no** circumstances bar abuse of *other* people before your own eyes should you challenge a member of academic staff. If the mutterings - and make sure they are real, not imagined - get to a crazy point then you go to *your* supervisor and leave it with him.
4. Whether as a result of parental/teacher influence, our own inhibitions or just popular culture, we often tend to think that our human worthiness in personal relationships is reflected by our career progress. Of course this is nonsense - but it is a nonsense that takes a bit of determination to refute. It wouldn't be out of order for you to start looking at the faces in the city crowd once in a while and testing your conviction - as a person, not as an ace researcher - in yourself.
Incidentally, I personally believe it's fair to say that pure math is more challenging conceptually than applied is; and this may have implications for the difficulty of publishing papers. Then again, many pure mathematicians may not be stimulated by the human benefits of applying existing concepts in the industry, medicine, social sciences, etc.
Yet, as username_2 alludes, all pure math tends to find application within a generation of its discovery. So pure and applied math need each other for justification and advancement. In such a situation relations between people involved in these respective fields need to have (and sincerely show) more appreciation of the other's challenge.
Upvotes: 3
|
2023/02/05
| 909
| 3,610
|
<issue_start>username_0: I just got a lecturer position at UK university. However, coming from the clinical field, I have no idea what items I need to negotiate. I know the research part of it but my colleague told me it is not only lab space but I also have to negotiate equipment and consumables.
I will do teaching but I have just realized that I need to negotiate the teaching package and I have no idea what things I can negotiate
I have read many things but it is only for US universities.
I'd appreciate some guidance on this as I negotiate a startup package at a UK university.<issue_comment>username_1: How much towards consumables and equipment you might reasonably expect is going to be discipline specific. How much teaching you are expected to do is going to be discipline and institution specific. Things will also be dependent on what level you are coming in at and whether the university in question is a teaching or research focused institution.
For an overview you might like to consult [Acton SE, Bell AJ, Toseland CP, Twelvetrees A. A survey of new PIs in the UK. Elife. 2019 Sep 20;8:e46827](https://elifesciences.org/articles/46827).
In addition to lab/office space, other things you might want to consider is large equipment, cash (which would be used for consumables, travel, computers, small equipment), PhD students and teaching relief.
In terms of large equipment, the approach is usually - if you absolutely need it to do your research, we'll help you find a way to get access to it. This might be buying something or promising you access to someone else's, it might be promising to put up the match funding an equipment grant would require.
In terms of students, preliminary data for your first grant will almost certainly come from a PhD student. Many places will allocate a studentship to a new lecturer. If this isn't the case, you'll need to find out what the situation is with access to PhD programs like DTPs (doctoral training programs), are you allowed to apply to them? Can it be arranged such that you are co-supervisor if a certain level of experience is required? Is there a way to prioritise your applications?
Cash is generally very limited. I got £10k ten years ago, and that was fairly typical, and things havn't changed much since. Although I did also get a PhD studentship that came with £5k/year research costs.
Finally teaching. In most places you will eventually be expected to spend 40-50% of your time teaching. For a sense of scale a year is approx 1600 working hours, so 40% is 640 hours. We get 2 hours prep for each hour teaching plus 20 minutes a student marking time, so that's at least 100 contact hours. But few research focused places will require that in year one. So you can negotiate how long after you start your job you'll have to start teaching, and how quickly your load will increase. So I negotiated 0 hours in my first year, 20 in my second, 50 in my third, 100 in my fourth.
An final important thing to consider is what your probation (read tenure) criteria will be. How is success going to be judged and over what time scale?
One last thing.
Get. Everything. In. Writing.
Do not rely on anyone's say so. Particularly be aware of heads of department promising things they don't actaully have the power to deliver.
Upvotes: 2 <issue_comment>username_2: There is normally no negotiation for positions of Lecturer in the UK. Your salary is determined by HRs and you normally start at the bottom of the grade. There is normally no startup package.
Congratulations on getting this position. Welcome to the UK academia.
Upvotes: 0
|
2023/02/05
| 1,192
| 4,646
|
<issue_start>username_0: A Christian pastor claims to be a Stanford professor who holds three PhDs in comparative theology, history, and eschatology.
However, his profile is not anywhere on the university website or anywhere in Google (rather than in his YouTube channel). However, he started a theological seminary and claims to be affiliated with Stanford University.
Below there is the certificate given in his seminary. I am more skeptical about this because the logo used below is the [Stanford Medicine](https://med.stanford.edu/) logo. I don't know why they used the medicine logo in a theology certificate. And on the official Stanford website, they don't mention a *Department of Philosophy in Religious Studies.*
Is this an authentic or fake one?
[](https://i.stack.imgur.com/kbq2y.jpg)
### Update
I e-mailed Stanford about this and they confirmed that this is a fake one:
>
> No, Stanford does not have a department of philosophy in religious studies and we don’t offer a degree in Bible ministry.
>
>
><issue_comment>username_1: I see several red flags.
1. On Stanford's website, there is a Department of Philosophy and a Department of Religious Studies. There is no "Department of Philosophy in Religious Studies". Even if there was a combined department in the past, it should be an "and" and not "in".
2. There is no date saying when the honorary degree was awarded
3. A typo on the word 'title' incorrectly spelled as 'titile'
Upvotes: 7 [selected_answer]<issue_comment>username_2: >
> Is this an authentic or fake one?
>
>
>
It looks highly suspect, as @NebUzer and @username_3 explain, so it's probably inauthentic. But - you can always just **Ask Stanford**. [Here](https://philosophy.stanford.edu/academics/undergraduate-program-philosophy/combined-major-philosophy-and-religious-studies) is a page of their program in philosophy and bible studies, or you could just go up to the department of philosophy. Write their secretariat or general contact address and inquire about this degree.
Another option is Stanford's ["Cerificates and Verifications"](https://studentservices.stanford.edu/my-academics/request-transcripts-certifications/certifications-and-verifications) service. While that's student-oriented, they might still be willing to help.
Upvotes: 5 <issue_comment>username_3: Adding to red flags raised by others:
1. This degree was granted to someone currently 35 years old. This is *very young* to get such a degree “honoris causa”, which is a kind of lifetime achievement award. I suppose it is possible to get such a degree by making a significant donation, or by being truly brilliant, but in all circumstances I know the university will put up some sort of public elogium when the degree is awarded.
2. My own degree does not give my date of birth but does indicate the precise date on which the degree was awarded: this is missing from the image.
Upvotes: 4 <issue_comment>username_4: Note that your first two sentences already show that this person is not who they claim to be. All professors at a proper university have a website that can be found through the university/ department website. It might not say much beyond the name and department of the university but every department will list every single professor. So if you can't find their website on the Stanford university pages there are not a professor at Stanford.
Additionally having three PhDs is highly suspicious. It is common trope in fiction to have brilliant scientists with PhD in multiple fields but this almost never happens in practice. There certainly are brilliant people who have done signficiant scientific work in multiple fields but they don't hold multiple PhDs (there are multiple questions on this site from people asking about doing a second PhD, the universal answer is: don't do it). The only real exception is honorary PhD titles, people can and do hold multiple of these but when these are mentioned they are always referred to as honorary, leaving that word out would be academic dishonesty.
Upvotes: 3 <issue_comment>username_5: It's a fake, top to bottom. Stanford does not confer honorary degrees, and this policy has been in place since at least [the 2000-2001 academic year](https://web.stanford.edu/dept/registrar/bulletin0001/pdf/GraduateDegrees.pdf).
Upvotes: 3 <issue_comment>username_6: As several folk have already pointed out, people with 3 PhDs are rare; so are professors at Stanford with no publications to their credit. Maybe you could use Google Scholar to check to see whether <NAME> has ever published anything.
Upvotes: 3
|
2023/02/06
| 871
| 3,507
|
<issue_start>username_0: I am a first year PhD student working in molecular simulation and GPU research.
I have poor grades in my bachelor's (49%).
Will this poor grade create a roadblock for me in the future if I want to work as a research professor in academia?
E.g., [This advertisement](https://www.mpg.de/19723474/postdoc-position-mass-spectrometry-based-proteomics) asks for transcripts of records for a PostDoc position.<issue_comment>username_1: **No**
Only the grades from your highest degree matters, and even then, [not a lot](https://academia.stackexchange.com/questions/90/university-rank-stature-how-much-does-it-affect-ones-career-post-ph-d).
>
> We only care about the quality and visibility of your results. Publish strong papers and give brilliant talks at top conferences. Convince well-known active researchers to write letters raving about your work. Make a good product and get superstars to sell it for you. Do all that, and we'll definitely want to hire you, no matter where you got your degree.
>
>
>
[Your poor BS grades can sink your application to PhD programs](https://academia.stackexchange.com/questions/38237/how-are-ph-d-applications-evaluated-in-the-us-particularly-for-weak-or-borderl), but given that you're already a PhD student, you've passed that hurdle. Do well in the PhD and your BS grades should cease to matter.
Upvotes: 6 [selected_answer]<issue_comment>username_2: I've been asked for **undergraduate transcripts** when applying for **tenure-track professor jobs** at universities in the **US and Canada**. It's not very common. Out of probably 60 applications, I've been asked for such a transcript four times. So probably **<10%** of my applications, although it might be more common for teaching-focused jobs than research-focused jobs (I've applied to about equal number of both). Three of the times I was asked for it was for predominantly undergraduate institutions, which might make sense. I don't know why they ask for it, but presumably the search committee will consider it. But how much consideration it might be given will probably depend on the attitude of each particular committee member.
Being asked for graduate transcripts is much more common, although I'm still unsure how much they are considered in deciding on whether to offer you the job.
If you do apply to an academic job that asks for an undergraduate transcript, you could probably make it into a positive: you could say something in your teaching statement (or diversity statement) about your struggles in succeeding in your undergraduate academic work and how you could use that experience to help struggling students.
I would like to mention here that, in my experience, the people who did best in classwork do not necessarily turn out to be the best researchers and many successful researchers were not the stars of their classes. In classwork, there is essentially only one path to success, while there are many different paths to success in research. Some succeed in research by having marvelous people skills, building networks of excellent collaborators and efficient research teams. Others succeed by single-mindedly obsessing over important technical problems and solving them. Others get really good at understanding the dynamics of funding agencies and how funding is distributed (nearly all successful independent researchers need some savvy with funding). So don't let some poor grades make you think you can't succeed at research. Best of luck!
Upvotes: 2
|
2023/02/06
| 1,021
| 4,562
|
<issue_start>username_0: I have encountered another weird situation as a referee. A couple of months ago, I submitted a referee report for an article, asking for major revisions. (This is theoretical physics, in which getting only a single referee report for published paper is not uncommon, even in top journals.) Last week, I was contacted by the journal again, with the news that the authors had submitted a substantially revised version, and the editor wanted me to review it again. (The journal involved here is not a particularly prestigious one, but it is generally a respectable one and has been published by World Scientific for several decades.)
So I went to the journal's online platform and started looking at the referee reports, the authors' replies, and the updated manuscript. However, I quickly realized that both referee reports were by me! More precisely, an editor had split my lengthy report into two, so that it looked like the journal had gotten opinions from two scientists. My comments were chopped up and parceled out between the two separate reports, with a little bit of added rewriting so that each one made sense on its own, more or less. (I was also annoyed that the editor added to one of "my" reports a suggestion that the authors check their English using the free version of Grammarly—something I would never in a million years suggest!)
So what should I do? The paper is actually greatly improved and can probably be published with little or no further editing needed. I don't really want to penalize the authors. On the other hand, I am pretty angry at the editor(s). A colleague suggested that I should just write to the editors, calling them out on their dishonesty and announcing that I will not be refereeing for them in the future. I suppose that's a reasonable approach to take, but it seems harsh. Is there an easier tack I can take, perhaps?
Or—on the off chance—does anyone think that what the editor did was actually fine, and that I am overreacting?<issue_comment>username_1: This is pretty clearly unethical. One of the recent training videos I've seen (at another publisher) actually threatened employees with termination if they modified reviewer comments, illustrating the seriousness of what happened. For this reason (& because it's rather unfair to the authors) I would not suggest declining to review the resubmission - whoever did it needs to know that what they did was unethical.
I cannot tell if it's the publisher who did it or an editorial board member who did it, but it's probable that what happened was intentional and not your report being too long (since Editorial Manager ought to be able to handle long reports). On the other hand, it is a rather weird thing for an editor to do, since they'd have to register a second reviewer (which requires a profile, email addresses, etc), split your review in two, and then submit it. It's not a simple process in Editorial Manager and requires significant effort. Furthermore, the benefit is unclear. Many of the best physics journals are happy to accept papers based on only one review, so it's not clear what is the point of making it seem like there are two reviewers. It could conceivably be a way to bypass a strict journal policy of requiring two reviews per paper, but even then it seems pretty weird to invite you to review the revision, since you are sure to notice.
Two ways to approach this:
1. You could file a complaint with the journal. Since there's a real chance it's the publisher that is at fault here, I would avoid complaining to the journal office and instead contact the editor-in-chief / some other senior editorial board member to ask them to investigate.
2. Alternatively, I happen to know some editorial consultants at this publisher personally. They would be well-placed to investigate this - they oversee the editorial processes & are not directly involved with managing journals. I could ask them to investigate for you. If you choose this option, I'll need some identifying information for the paper (which journal it is, the article ID, etc). If you choose option #1 and the editor-in-chief or board member complains to the publisher, chances are these editorial consultants will be involved as well.
Upvotes: 6 <issue_comment>username_2: I recommend turning down the review, telling them why, asking your name to be taken off their referee list until they clean up their act, and sending an email along with your original review document to the Editor in Chief, suggesting they fix their section editor.
Upvotes: 3
|
2023/02/06
| 1,922
| 8,195
|
<issue_start>username_0: I'm pretty sure almost all mathematicians have been in a situation where they found an interesting problem; they thought of many different ideas to tackle the problem, but in all of these ideas, there was something missing—either the "middle" part of the argument or the "end" part of the argument. They were stuck and couldn't figure out what to do.
>
> 1. In such a situation what do you do?
> 2. Is the reason for the "missing part" the incompleteness in the theory of the topic that the problem is related to? What can be done to find the "missing part"?
>
>
>
For tenure-track/tenure professors, maybe this is not a big deal because they have "enough" time and can let the problem "stew" in the "back-burner" of their mind, but for limited time positions, for example, PhD, postdoc, etc., where the student/employee has to prove their capability to do "independent" research, so that they can be hired for their next position. I think for these people it is quite a bit of a problem, because they can't really afford to spend a "lot" of time thinking about the same problem.
I'm particularly interested in the answer in the realms of mathematics, but I would like hear suggestions from "non-mathematicians" as well.<issue_comment>username_1: I usually have multiple related strands of a project going on at the same time. I will focus on most promising one, but if that stalls I switch to the next most promising. That can be difficult, especially if you already spent a lot of time on a given attempt. However, as you already mentioned, in an early career position you cannot afford to beat your head against the same wall over and over again. The reassuring part is that I found that very often the work I have done in those abandoned strands often came back as useful at a later point in time for another project.
Upvotes: 3 <issue_comment>username_2: Generic answer, not specific to maths problems. But it should help a little.
When you are stuck on a problem, do any part you know how to do, then look at the overall problem once more.
If you run out of areas you know how to do, then go looking in "the usual sources" for possible methods to approach it. Textbooks on the subject. Also web pages, journal articles, etc. Requests for hints from colleagues.
If this cycle gets stuck, then consider alternate approaches. That is, completely different methods of solving the problem.
Sometimes a "brute force" method can give you a hint as to the more elegant results. For example, if you were doing an integral. (I know, but it's something I know how to do.) You could do things like solving a simplified version for practice. Or you could do some numerical work to get an idea what the closed form should look like. When solving differential equations you can use tricks like "solution generation" to get solutions to "nearby" versions of the equation that may give you hints. There will be similar things that can be done in your actual research work.
Some folks may not know what I mean by "solution generation." Suppose you have to solve a differential equation that looks like so.
y" = f(x,y)
Suppose that the actual f() you are working with is resisting your efforts. As a practice run, choose a different f() that is not drastically different to yours, but that has a closed form solution you already know. That may give you some hints what the answer to your problem is.
Upvotes: 1 <issue_comment>username_3: First, you need to understand that not every math problem is tractable, and many more are not going to be solvable in a fixed time period. Some aren't "ripe" enough in the general course of things for it to be likely that insight will occur, though strange things can happen.
If you work on only the Riemann Hypothesis, you are likely to fail, though you'd be famous if you succeed. The time between the statement and published proof of Fermat's Last Theorem was over 350 years.
For tractable problems, a "vacation" from the problem, even if for an hour or two (or overnight) can sometimes (sometimes!) help. But for truly difficult things, the field might even need to mature for a while and a lot of collaborative discussion might be needed.
So, what you do, when really stuck is to move on to one of your other, hopefully many, lines of inquiry that might result in something that will advance your career while the hard problem "ripens".
Keep a notebook of things that seem interesting and might be followed up in the future. It isn't quite "work in progress", but "future possibilities". Review this notebook periodically and make notations to it if you have gained any insight. Turn to it when you need to set some other project aside, or finish another project.
And, when you leave a problem for an extended period, don't forget to record any partial insights you have when you set it aside. That is, it goes into your future work notebook.
Upvotes: 4 <issue_comment>username_4: Give up and work on a different problem.
I have in my files (some paper some electronic) notes on about twice as many abandoned projects as I have papers. And these are only the ones that got far enough for me to have extensive notes. Counting the problems I've thought about seriously at all, I'm probably batting .050 or so.
It's generally possible to write (and publish) a survey paper on what you learned to work on the problem, or a paper on your partial results. The survey paper can be helpful if well-written; the partial results paper probably not. Either could be used to satisfy bureaucrats counting papers, but research mathematicians are unlikely to give either much weight.
Upvotes: 3 <issue_comment>username_5: 1. Consider successively more special cases of the problem until you find one that you can solve. Work with the typical example instead of the general case.
2. Consider more general cases of the problem. Yes, sometimes proving a more general statement is easier than proving a very specific one.
3. Change the question. Just because you set out to solve a certain problem does not mean you need to stick to this plan. If you found a plausible technique that just doesn't fit your intended application, find a different application that it does fit and pretend this was your plan all along.
4. Zoom out. Consider the problem in a wider context. Think about what it is that you are really trying to do when you abstract away from the details. Instead of thinking in terms of technical details, think in terms of analogies. For example: what I need in my proof is to find the correct analogue of hypermodular widgets on finite simple groups in the context of locally compact groups.
5. Talk to people about your problem.
6. Take a break.
7. Talk to more people about your problem.
Upvotes: 6 [selected_answer]<issue_comment>username_6: I would break down the parts of the system or function that you are already confident in and create a visual diagram of it, see how the parts interact, and the "missing bit" might just jump out at you.
I would create a few diagrams of your function or system, and I would follow some kind of universal diagramming tool to make it easy, like UML.
Because I'm a software engineer I need to create hugely complex system with no way to hold the whole system in my head at once, I need to break it down into bitesize pieces from a high level all the way down to the minutia. I would approach understanding a system in the same way though, whether it's software, a social system, a maths problem, they're all just "systems" that have components that interact.
That said, I would create a hierarchy of "Component" diagrams with an "Activity" diagram for each component. The system itself is a Component so make a top level activity diagram for that too.
The point is to break down what is a huge problem into smaller parts, define how the parts interact and then you can work on one part at a time.
Hope this inspires you :)
EDIT: Just came across "Category Theory" which looks like a mathsy flavour of UML, should help you map out your missing bits :)
Upvotes: 1 <issue_comment>username_7: Explain the problem to your dog. Sometimes the solution will jump out at you.
Upvotes: 1
|
2023/02/06
| 1,690
| 7,348
|
<issue_start>username_0: I see everyone in my university learning about machine learning and artificial intelligence. I understand they might have a genuine interest in it but is it the only subject that computer science is heading towards?
I have a deep interest in low-level computer science and have done projects in which I have designed compilers, assemblers and an RTOS running on microcontrollers.<issue_comment>username_1: I usually have multiple related strands of a project going on at the same time. I will focus on most promising one, but if that stalls I switch to the next most promising. That can be difficult, especially if you already spent a lot of time on a given attempt. However, as you already mentioned, in an early career position you cannot afford to beat your head against the same wall over and over again. The reassuring part is that I found that very often the work I have done in those abandoned strands often came back as useful at a later point in time for another project.
Upvotes: 3 <issue_comment>username_2: Generic answer, not specific to maths problems. But it should help a little.
When you are stuck on a problem, do any part you know how to do, then look at the overall problem once more.
If you run out of areas you know how to do, then go looking in "the usual sources" for possible methods to approach it. Textbooks on the subject. Also web pages, journal articles, etc. Requests for hints from colleagues.
If this cycle gets stuck, then consider alternate approaches. That is, completely different methods of solving the problem.
Sometimes a "brute force" method can give you a hint as to the more elegant results. For example, if you were doing an integral. (I know, but it's something I know how to do.) You could do things like solving a simplified version for practice. Or you could do some numerical work to get an idea what the closed form should look like. When solving differential equations you can use tricks like "solution generation" to get solutions to "nearby" versions of the equation that may give you hints. There will be similar things that can be done in your actual research work.
Some folks may not know what I mean by "solution generation." Suppose you have to solve a differential equation that looks like so.
y" = f(x,y)
Suppose that the actual f() you are working with is resisting your efforts. As a practice run, choose a different f() that is not drastically different to yours, but that has a closed form solution you already know. That may give you some hints what the answer to your problem is.
Upvotes: 1 <issue_comment>username_3: First, you need to understand that not every math problem is tractable, and many more are not going to be solvable in a fixed time period. Some aren't "ripe" enough in the general course of things for it to be likely that insight will occur, though strange things can happen.
If you work on only the Riemann Hypothesis, you are likely to fail, though you'd be famous if you succeed. The time between the statement and published proof of Fermat's Last Theorem was over 350 years.
For tractable problems, a "vacation" from the problem, even if for an hour or two (or overnight) can sometimes (sometimes!) help. But for truly difficult things, the field might even need to mature for a while and a lot of collaborative discussion might be needed.
So, what you do, when really stuck is to move on to one of your other, hopefully many, lines of inquiry that might result in something that will advance your career while the hard problem "ripens".
Keep a notebook of things that seem interesting and might be followed up in the future. It isn't quite "work in progress", but "future possibilities". Review this notebook periodically and make notations to it if you have gained any insight. Turn to it when you need to set some other project aside, or finish another project.
And, when you leave a problem for an extended period, don't forget to record any partial insights you have when you set it aside. That is, it goes into your future work notebook.
Upvotes: 4 <issue_comment>username_4: Give up and work on a different problem.
I have in my files (some paper some electronic) notes on about twice as many abandoned projects as I have papers. And these are only the ones that got far enough for me to have extensive notes. Counting the problems I've thought about seriously at all, I'm probably batting .050 or so.
It's generally possible to write (and publish) a survey paper on what you learned to work on the problem, or a paper on your partial results. The survey paper can be helpful if well-written; the partial results paper probably not. Either could be used to satisfy bureaucrats counting papers, but research mathematicians are unlikely to give either much weight.
Upvotes: 3 <issue_comment>username_5: 1. Consider successively more special cases of the problem until you find one that you can solve. Work with the typical example instead of the general case.
2. Consider more general cases of the problem. Yes, sometimes proving a more general statement is easier than proving a very specific one.
3. Change the question. Just because you set out to solve a certain problem does not mean you need to stick to this plan. If you found a plausible technique that just doesn't fit your intended application, find a different application that it does fit and pretend this was your plan all along.
4. Zoom out. Consider the problem in a wider context. Think about what it is that you are really trying to do when you abstract away from the details. Instead of thinking in terms of technical details, think in terms of analogies. For example: what I need in my proof is to find the correct analogue of hypermodular widgets on finite simple groups in the context of locally compact groups.
5. Talk to people about your problem.
6. Take a break.
7. Talk to more people about your problem.
Upvotes: 6 [selected_answer]<issue_comment>username_6: I would break down the parts of the system or function that you are already confident in and create a visual diagram of it, see how the parts interact, and the "missing bit" might just jump out at you.
I would create a few diagrams of your function or system, and I would follow some kind of universal diagramming tool to make it easy, like UML.
Because I'm a software engineer I need to create hugely complex system with no way to hold the whole system in my head at once, I need to break it down into bitesize pieces from a high level all the way down to the minutia. I would approach understanding a system in the same way though, whether it's software, a social system, a maths problem, they're all just "systems" that have components that interact.
That said, I would create a hierarchy of "Component" diagrams with an "Activity" diagram for each component. The system itself is a Component so make a top level activity diagram for that too.
The point is to break down what is a huge problem into smaller parts, define how the parts interact and then you can work on one part at a time.
Hope this inspires you :)
EDIT: Just came across "Category Theory" which looks like a mathsy flavour of UML, should help you map out your missing bits :)
Upvotes: 1 <issue_comment>username_7: Explain the problem to your dog. Sometimes the solution will jump out at you.
Upvotes: 1
|
2023/02/06
| 269
| 1,126
|
<issue_start>username_0: Is the article processing charge paid only by the corresponding author or can it be split among all coauthors, with separate bills and receipts?<issue_comment>username_1: It depends on the publisher. For example, the IEEE does allow it:
<https://supportcenter.ieee.org/app/answers/detail/a_id/1385/kw/payment/related/1>
Upvotes: 1 <issue_comment>username_2: Yes (at least in my experience). I’ve done this in two ways.
1. The publisher (*v.g* the American Physical Society) or its representative has a payment website where partial payments can be done. The software keeps track of the amounts of various institutions until the full amount is reached. The invoice includes the names of the various parties and the amounts they paid.
2. One person or grant pays the entire fee and my grant reimburses this person or grant through normal institution-to-institution invoicing or bank transfer.
In all instances I’m familiar with, publishers have recognized that there may be more than one person paying the APC so option 1 is now the most common (at least with larger publisher in physics).
Upvotes: 0
|
2023/02/06
| 342
| 1,454
|
<issue_start>username_0: I recently applied to US universities and contacted professors I mentioned in the application. In one particular case (the most important one), all three professors are not accepting students. Would the application get rejected straight away?
My research area as mentioned in my SOP is “*Deep Learning and its applications in Computer Vision*”.
I guess the area I mentioned in my SOP is neither too broad nor too restrictive. Does it affect the application in any way?<issue_comment>username_1: It depends on the publisher. For example, the IEEE does allow it:
<https://supportcenter.ieee.org/app/answers/detail/a_id/1385/kw/payment/related/1>
Upvotes: 1 <issue_comment>username_2: Yes (at least in my experience). I’ve done this in two ways.
1. The publisher (*v.g* the American Physical Society) or its representative has a payment website where partial payments can be done. The software keeps track of the amounts of various institutions until the full amount is reached. The invoice includes the names of the various parties and the amounts they paid.
2. One person or grant pays the entire fee and my grant reimburses this person or grant through normal institution-to-institution invoicing or bank transfer.
In all instances I’m familiar with, publishers have recognized that there may be more than one person paying the APC so option 1 is now the most common (at least with larger publisher in physics).
Upvotes: 0
|
2023/02/07
| 3,120
| 13,410
|
<issue_start>username_0: I was shocked to learn that the participation fee for a large annual European conference almost didn't change when the host country changed from one of the most expensive Western European countries to one of the most inexpensive Eastern European countries. I couldn't see any explanation other than the organizers making a huge profit, given that the number of participants was a few hundred and the participation fee was about 500 Euro per participant.
So, my question is: How common is it for conference organizers in academia (that is, the responsible institution and/or members of the local organizing committee) to derive a monetary profit by deliberately charging a too high participation fee and pocketing the difference between the collected participation fee and actual expenses and/or by getting a kickback from the venue or a hotel?
I'm asking about making an unduly good profit that is clearly not commensurate with the time and effort spent by the organizers - say, a profit at least a few times larger than what the time spent by the organizers is worth according to their paying rate at their institution.
To put it simply, do conference organizers in academia make good money?<issue_comment>username_1: While there are certainly some people who organize conferences for profit, every reputable conference that I have ever dealt with (as either organizer or participant) has been run with the intent of keeping fees as low as possible while breaking even.
Large conferences have many costs beyond the venue:
* Large conferences often have paid staff, whose salaries are a major expense.
* If you are working with a society, there may be fees for publications, as well as fees to support society staff.
* Hotels can be a gamble: they give you discounts, but it's often required to be linked to a block of pre-reserved rooms by a contract, and if attendees don't use all the rooms in the block, the conference will end up responsible for the cost of the rooms.
* Finally, given the unpredictability of attendance, a yearly conference often banks money in more profitable years to avoid bankruptcy in years where it takes a loss (this is one of the services that societies provide as well).
Corruption such as you describe (organizers taking kickbacks or lining their own pockets) is certainly possible of course, and I am sure it happens in some cases. I would expect, however, that it is relatively rare, particularly given that conference finances are usually run through a larger organization (whether society or university) with accounting processes designed to prevent corruption rather than via the personal finances of a poorly supervised group of organizers.
Upvotes: 6 <issue_comment>username_2: Many conferences in engineering are organized by the relevant professional body, e.g. IEEE, IChemE, etc.
These organizations have permanent staff for administration of membership services, CPD programmes, publications and conferences.
Seeing how these bodies have over the past 40 years been so quick to capitalize on their membership's economic potential in relation to credit cards, I would expect that the admin staff for conferences would seek to make a good margin on these events - if only to justify their own staffing numbers, salaries and benefits. They can always justify the profit by saying it may be used to subsidize small attendance CPD or support the organization's regional network.
It is rare to see conference entrepreneurship along the lines of Prof <NAME>' ex-colleague. Yet individual or small group organization of special events (*à la* ["Lorenzo's Oil"](https://en.wikipedia.org/wiki/Lorenzo%27s_Oil)) occasionally happen. I doubt if profit is the motive given the stresses of doing such events.
National conferences offer far less opportunity for hotel and block-flight swindles with most delegates making their own way (often piling into their PI's car) to the university venue, staying in student accommodation and 'dining' in the campus' cafeterias.
It wouldn't be so bad if the "profit" made on first world attendees were used to subsidize those from third-world countries who would otherwise not be able to come without saving for ~ 3 years.
Upvotes: 2 <issue_comment>username_3: I organized several big conferences and my answer is no.
You would be shocked to hear the prices for a conference venue, even if the conference takes place in a university. (OK, then the university gets the money, but not the organizing group.) Venues outside a university are even more expensive.
A second important point is the catering. Assume, for example, the catering costs only 20 Euros per participant per day (which is realistic for a Western European country - not for real meals, only sandwiches, coffee, soft drinks) and there are 500 participants. Then you have 10.000 Euros per day, 30.000 euros for a three-day conference, only for a little bit of catering that people hardly notice.
Then you need a lot of helpers. The usual university staff are not enough to handle this. You need additional help, external staff you have to be paid per hour and conference days are long, let alone the days and weeks before.
In Western Europe, the invited speakers often get their travel costs reimbursed, i.e. the conference organizers must pay their flights and hotel bills. Of course you don't want to spend a fortune on this, but if you want them to come back one day, you won't choose the cheapest hotel either ...
And last but not least, there are more costs you would not think of in the first place, e.g. for promoting the conference or the information material for the participants.
No, conference organizers do not make a fortune with a conference. Perhaps they have moved to Eastern Europe because they could not afford Western Europe any more, due to the current rise in prices.
Upvotes: 5 <issue_comment>username_4: The opportunity for achieving excessive profit is limited by market forces.
A large annual European conference is organized in the name of a specific global or European society, by an organization that's local to the venue.
The society is thus negotiating with a different local partner every year and they cannot afford the risk of damaging their brand by partnering with an incompetent or borderline corrupt organization who won't deliver on the expectations every now and then. The society looks at more than one target country, more than one potential partner for every annual conference that's less than 3 and more than 1 years ahead (assuming that it's desirable if they can eventually announce the next year's venue to the participants of the current year's conference).
The society will have some general, rather stable requirements, both technical and economic (non-paying participants, maximum participation fees, maximum student fees, venue characteristics), based on which all interested partners eventually prepare a formal proposal, the key part of which is the budget. The society then applies selection criteria which might not be formulated in advance, but they invariably include the proposed participation fee (lower is better). One candidate partner might offer a modern venue in an attractive city, another might be able to present a budget which appears both lean and not terribly cheap (perhaps because their university management or another local party appears to be effectively subsidizing the event, such as by charging less than the usual rate for the venue, as an investment into their own "brand" by hosting a large and prestigious conference). Everybody wants to be invited at a discount!
Market forces:
* The society has a monopoly over their own brand, but other credible parties can establish annual conferences independent of the society.
* The local organizers provide only limited accountability to the society and any profit or loss (intended or unintended) is theirs, but other credible parties may get in touch with the society and start competing for organizing the conference 2 or 3 years from now (and eventually prepare their own proposals), if they know how to do so more efficiently.
* The participants have the most freedom of all as they (or their sponsors) are ultimately paying the bill and there are *so many conferences* to choose between.
I'll conclude with two opinions for which I have no data.
* Where the profits are eventually modest, the potential for any corruption (of which the local organizer, or the society, would be the victims) is also modest.
* For academics who organize conferences, the career/fame/connection benefits probably way exceed any potential monetary benefits - especially if the conference is a prestigious one.
Upvotes: 2 <issue_comment>username_5: A lot of conferences are put out by nonprofit organizations that release financial statements, so you can check yourself by looking at the numbers.
The Society for Neuroscience is the principal neuroscience professional organization. They have an annual meeting with about 30,000 attendees.
Here's their 2022 financial statement: <https://www.sfn.org/-/media/SfN/Documents/NEW-SfN/About/Annual-Report/Consolidated-Financial-Statements/SFN-Consolidated-2022-FS.pdf>
They report annual income related to the meeting of around $7.7 million, compared to expenses of $5.1 million. There was additionally over $4 million in membership dues, and it's likely many members sign up specifically for the meeting each year (the cost of membership is less than the difference in meeting cost for members vs nonmembers). You can see in 2021 the meeting attendence was much lower and membership was as well.
However, as a non-profit, the proceeds go to fund the society's other programs. This particular organization is also run by professional nonprofit organizers, who it should be noted likely care little about the goals of the organization and instead want their numbers to reflect good fiscal management so they can build a resume and get their next job on the field.
My assumption is that these big conferences are not the norm. Much smaller conferences take in very little revenue, are organized by volunteers rather than professionals, and likely use nearly all of their funds for conference expenses. If there are leftover funds they are likely used to fund the next meeting.
Upvotes: 4 <issue_comment>username_6: It's like any other service -- they charge what people are willing to pay for the service they provide. If the fee they charge is too high, not many conferences would use their services, and if the fee were too low, they wouldn't be able to stay in business because they wouldn't make money. The fact that conference planners exist means that there is a happy medium.
Will they become wealthy? I don' think so. Conference planning is hard work, and if they managed to become wealthy, they're probably ready to retire.
That said, there are some perks to being a well reputed conference planner that might make it an attractive lifestyle for some people. For example, if you like to travel, it's probably a big plus. While it's still work, many hotels and resorts plugged into the conference circuit will provide conference planners with free visits to come evaluate their facility. Flights are a business expense, which can also support the desire to travel.
Upvotes: 1 <issue_comment>username_7: It depends on the conference.
>
> participation fee for a large annual European conference almost didn't change when the host country changed from one of the most expensive Western European countries to one of the most inexpensive Eastern European countries
>
>
>
This isn't that unusual. First, you need to realize that the registration fee is rarely the most significant expense of attending a conference: once you've paid for accommodation, travel and food, the registration fee is not necessarily that large, and is rarely a deciding factor in participating.
In Eastern Europe, some labs or institutes organize conferences specifically to make a profit. "Everybody" knows this and people still go there because *[noblesse oblige](https://en.wikipedia.org/wiki/Noblesse_oblige)*, i.e. people recognize that some are more fortunate than others and they know the leftover profits will be used to support legitimate activities of the lab or the institute.
Nevertheless, organizing a conference (especially a large conference) actually costs a lot more than people think. In particular, beyond the organizing committee, there might be administrative support needed to collect registration, deal with invitations, visas etc. This staff does not necessarily work for free, especially in countries with fewer resources.
In my experience the big society conferences are the most outrageously priced as registration pays to support the expenses of the society officers, and I have no sympathy for this kind of racket.
On the other hand, what I learned from my organizing conferences and in discussion with friends who have organized conferences, is that it can difficult to understand the constraints faced by the organizers. The overwhelming majority of organizers are doing their best, and the organizers - especially in developing or emerging countries - are more worried about loosing their reputation than loosing or making money: if the product is disappointing, it is the last time someone will show up at an event organized there.
Upvotes: 2
|
2023/02/07
| 1,631
| 6,734
|
<issue_start>username_0: From [an answer to another question](https://academia.stackexchange.com/a/172932/84834):
>
> Unsurprisingly, both of these "jobs" cost money. Education needs teachers, research needs researchers, and both need administrative support staff. Buildings and other infrastructure need to be maintained, and many other small and large costs accrue. Sadly, neither of these "jobs" actually generates substantial money directly - in Sweden education is free (students pay no tuition, even at our private university), and with the exception of the rare patent- or spinoff-generating research, even world-class publications do not pay salaries or utility bills. At Chalmers, just like at most universities (at least in Europe), all our core functions are substantial loss leaders.
>
>
> Hence, Chalmers needs income streams. **The government pays us for graduating students, which is sufficient (more or less) to cover the "education job" at least on bachelor and master level.**
>
>
>
This sounds suspiciously like the conflict-of-interest that is regularly levelled at open access publishing - publish more papers (even if they're terrible), get more revenue; graduate more students (even if they're terrible), get more revenue.
How does the government prevent this conflict of interest from turning universities into "predatory universities"?
I'm tagging this question with `sweden`, although it should also apply to other countries that pay universities for graduating students (are there such other countries?).<issue_comment>username_1: Here in England (?and Wales), there's a government agency called the Office for Students which can withdraw a university's power to award degrees if it's dissatisfied with that university in any of various ways, one of those ways being the rigour of the university's degree programmes. (In its current incarnation, I believe the OfS can do this even if the university's degree-awarding powers were granted in the first place by primary legislation.)
Upvotes: 1 <issue_comment>username_2: My perspective is the US, but the ideas are, I think, pretty common. Note that such things as "paying for graduates" doesn't happen in a vacuum. I can't start up a university, claim to graduate thousands (thousands, I say) and send a bill to some government.
There are several guardrails, both regulatory a repetitional. Universities who get public funding are also subject to public regulation at various levels. In the US it is mostly (not all) from the state level.
More broadly, however, reputable universities strive to have qualified faculty with advanced degrees. The reputations are public. When bad things happen, they are likely to be revealed since people are allowed to speak up and complain both inside and outside the institution.
In short, there is pressure to "do the right thing" at all levels of institutions that must operate in the public sphere.
In the US, some private, usually for-profit, institutions are less bound, but they aren't likely to directly feast on the public dole. Student loans are a problem, however, as they are too loosely regulated and the system is open to abuse. These institutions can (and sometimes are) predatory.
In general, though, the reputations of too many people depend on the reputation of the university they work at or attend. That, in itself, generates the pressure to do the right thing, though perhaps not the "perfect" thing.
Upvotes: 0 <issue_comment>username_3: In addition to [username_1's answer](https://academia.stackexchange.com/a/193263/82972), in the UK we also have the concept of [External Examiners](https://en.wikipedia.org/wiki/External_examiner). At the end of each semester a leading academic in the same field from a comparable university will come to the university. They will look at our syllabi, check the exam papers, read a selection of the exam answers, read some dissertations, talk to students, examine our administrative processes. Only with their signature can the marks be made official and degrees be awarded. As they are from one of our "competitors", it is in their interests to be rigours, but not unfair (because someone will be external examining their degrees).
In addition, there is idea of accreditation, as outlined in the comments. Universities must be accredited by the government to receive funding for giving degrees. The difference from OASPA, SCI, COPE etc is that you are still allowed to publish in a journal even if it doesn't belong to one of those organisations. In the UK you are not allowed to give a degree unless you are accredited by OfS. There are additional, voluntary accreditations in some subjects that might reassure students that that course is particularly high quality (generally things like Engineering, accounting, journalism, medicine etc).
However, the final thing to realise that teaching undergraduates isn't that profitable either. Generally the only activity that brings in income substantially above what it costs is educating overseas students. And here there very much are accusations that standards have been lowered to make more money.
Its not clear to me however, how the government paying universities to graduate students is any more open to corruption than the students themselves paying.
Upvotes: 2 <issue_comment>username_4: In Sweden the quality of higher education and research is evaluated by a government agency called [Universitetskanslersämbetet](https://www.uka.se/) (UKÄ), or "Swedish Higher Education Authority". Reports with results for various institutions can be found [here](https://www.uka.se/vara-resultat/hogskolekollen) (only in Swedish, as far as I'm aware). If a program or institution fails to uphold quality standards they may lose accreditation for the specific field or even degree. Note that this is true both for public and "private" institutions.
It's also worth noting that Riksdagen (the parliament) sets a ceiling or upper limit to how much money a given institution can receive for, e.g., graduating students and ["full year achievements"](https://sv.wikipedia.org/wiki/Hel%C3%A5rsstudent_och_hel%C3%A5rsprestation). Basically, a university that admits too many students is on the hook financially. This naturally limits the potential revenue, and negotiations to raise this ceiling probably take into account factors like demand and quality. For places like Chalmers with typically more applicants than spots, education quality and employability have long been selling factors that they are incentivized to keep in place. But you can also find institutions that have a reputation for having (some) watered-down programs. This is the case in most countries, I suppose.
Upvotes: 4 [selected_answer]
|
2023/02/07
| 467
| 1,886
|
<issue_start>username_0: I am currently retaking a calculus course that I failed, and noticed that all the quizzes are exactly the same from when I took it. I do have the solutions to the quizzes, as I have looked over them when I completed them in the past. Does it breach any rules if I study the solutions and get everything right? Am I risking expulsion/punishment?<issue_comment>username_1: It would seem that this is clearly on your professor, as students repeating a calculus class is a common situation. However, you would need to read the academic misconduct rules of your university carefully, as these rules can be counter-intuitive.
It might be best sending an email to your professor stating that you took the class before, have the quizzes, and apparently, the quizzes are not new, but that you have of course been reviewing the materials.
Upvotes: 3 <issue_comment>username_2: This is a problem for the professor(s) running the course and not for you. Perhaps they have made a conscious decision to operate in this way. Perhaps they are just lazy. They might have the attitude that if you can do **these problems** then you know enough to advance.
While regulations about what materials can and can't be used while sitting the exam, I'd object to any regulations that forbid you any materials whatever prior to the start of the exam. If those regulations exist they are honey-traps that shouldn't exist. "Gotcha. I see you read page 45 in the text when I was clear you weren't allowed to go past page 44." Bizarre.
If you are learning the material from your study it is the purpose of the course. It isn't a game to be played with arcade rules, especially [Calvinball](https://calvinandhobbes.fandom.com/wiki/Calvinball) rules.
But if you think you need more of a challenge in the course, talk to the professor and ask how you can get it.
Upvotes: 3 [selected_answer]
|
2023/02/07
| 1,447
| 6,342
|
<issue_start>username_0: My question is about alternative salary streams for academics in fields with strong industry collaboration/interest. What is a typical "bonus" that a tenured academic can expect on top of their salary grade?
I am finishing up my PhD in Engineering in the UK. Given that my PhD (and I am sure many others in similar areas) has transferable skills for the industry (e.g. data science/finance/engineering), I could leave academia for a high paying job. Like many, despite my interest in science, it is hard to justify the ~3x paycut I need to take to continue on with a postdoc after borderline poverty for so many years. And it is hard to justify turning down a job that makes more than a full professor (according to the publicly available salary spine for academics) in my first year.
I have heard that it is common for professors in applied fields to have alternative income streams, through, for example, consulting, industry bonuses, spin-outs, and tutoring services. How common are these alternate streams, and how much (on average) can one expect to make on top of their base salary?
I hope to stay in science but can't help think I'm making an ideological choice over a practical one if I do. Many thanks for any data/perspective!<issue_comment>username_1: You can not expect additional income on top of your academic salary. It may be possible to earn extra income, but you should not treat such opportunities as granted.
1. Consulting, industry bonuses, spin-outs, and tutoring services are time-consuming enterprises. Academic work is very time-consuming itself, often taking more than nominal 40 hours per week. The demands of academic teaching, research and administration combined can take 50, 60 or sometimes 80 hours per week. Academic workload can be overwhelming. Some academic, particularly in early career, find that they can not fully meet the expectations of their University/Department and have some minimal time to take care of their own health and wellbeing. An attempt to squeeze an extra work may be detrimental to you own health.
2. Consulting and industry bonuses come easier to people who are already established in their career, such as famous University professors. It may be difficult to secure/negotiate such income streams when you are a fresh academic who has yet to prove themselves to the world.
3. If/when you are successful securing extra income through consultancy, University may (and often will) require a cut of your cake. The exact way how it works depends on country, University and personal circumstances, but if your consultancy or side project goes really well, you may find it easier to focus on it full-time, rather than try to balance two jobs.
Upvotes: 2 <issue_comment>username_2: Generally speaking, a university that pays you full time expects, fairly enough, that you will give them full time effort.
That said, there are some minor things and one (at least) major option. If you do a bit of consulting to a firm they might give the university a grant that you can draw on for such things as travel and equipment. This enhances your regular work and might also benefit a student or two. The university isn't likely to object provided that you don't neglect normal duties. Some such consulting might be in conjunction with an advanced student or two.
Tutoring is probably off limits if you are asking about tutoring the university's own students.
You can earn a bit of money by writing books. A few (very few) reap large income but most only earn small rewards for a limited time and the publisher may want regular new editions. The first edition is fun, the later ones are a bit of drudge work. If it is coordinated with your normal work it is probably acceptable and the university won't (in any case I know of) claim a share. Some people teaching writing even write novels while holding full time positions. I've known one or two millionaires based on publishing key/major books. Most of us are content with a few thousand per year for a few years.
The major possibility is in the realm of patents. Most places the university insists that they be the patent owner but is likely, in such cases, to share monetization with those who developed it. This is probably to your advantage, since they also take on the task of defending the patent against claims, which can be very expensive.
Upvotes: 1 <issue_comment>username_3: I have a somewhat difference experience with respect to that of the answers given by username_2 and username_1. Intermittently, along the years, I did a bit of consulting with the industry, mainly electronic design and consulting about measuring systems, and my experience is that *if one wants to commit a significant amount of time of their career to consulting* (which is not my case), there's room for that and to have a significant alternative salary stream.
Indeed, you have to be willing to sacrifice your research output, and this could be difficult if you are at the beginning of your career or you don't have a large group, but I'm aware of people who made this choice.
In my experience, industry consultancies can be usually framed in two ways:
1. Private consultancies, if your university allows them. These are the most remunerative, but you have to check the limits imposed by your university and indeed the tax regulations of your country to understand how to declare this secondary source of revenue.
2. Consultancies obtained through university contracts. In this case, the industry pays the university and not you directly, but then the university gives you part of the money after having taken some overhead (which keeps the university happy). You can then choose to keep the money as a research fund or to transfer it to your personal account.
However, if you want the consultancy income to be significant you need to be able to:
1. Advertise yourself, and grow connections within the industry. The world out there should be able to discover you.
2. Actually solve the industry problems within a reasonable time frame. This is the most difficult part because frequently the *Ivory Tower*'s solution are not practical for the industry, and the academic's time scale is not compatible with the industry needs. And if you're not able to fulfil the industry needs, you'll soon run out of contracts.
Upvotes: 0
|
2023/02/07
| 1,159
| 4,848
|
<issue_start>username_0: This is a follow-up from [Patching things up with advisor? He has been avoiding me since I found a simple answer to a problem he thought would be challenging](https://academia.stackexchange.com/questions/187724/patching-things-up-with-advisor-he-has-been-avoiding-me-since-i-found-a-simple)
I'm a math PhD student in the western world. Last year, I solved a question that my advisor hadn't solved himself. This led to conflict between us, in which we he would tell me things like "*solving this problem won't get you a job*", "you should consider industry", etc.
I *applied for postdocs* independently, without substantial support from him. He *did write me a recommendation* though. I managed to *generate interest from some top people* in my field, and got a postdoc I wanted.
I recently again got a result that supersedes his previous work. This is related to the above problem as well. There are *minor flaws* that can be ironed out easily. My advisor *magnified those errors*, told me that it is all completely wrong and far from a solution, and told me to forget about this problem and do something completely outside of my current expertise. When I asked him if I *should publish* my results when I managed to iron out the flaws, he told me I should completely forget about it for now. He then suggested five different things I should do instead of this problem.
The troubling part is that my advisor first blurted out a way to iron out the tiny flaws. Then he went back on his own comment and said that this strategy would never work, and I should forget about it.
I don't trust my advisor's judgement anymore.
1. Should I change advisors? This is the last semester of my PhD.
2. Should I not publish these results? They are important, but not enough for the very top journals. However, publishing them will obviously destroy ties with my advisor forever.
3. Should I completely change fields? This would reduce conflict with my advisor.<issue_comment>username_1: My suggestion, if you can manage it is: to finish your degree as soon, and as simply as you can, and move on. You don't need to publish the "new" results at the moment, and will need to consider whether he has any authorship claim to the work or only an acknowledgement. You don't need *Euler*'s permission to extend his work (or anyone's).
Develop an independent career, and new collaborative opportunities. Let referees judge the quality of the paper you submit, rather than depending on someone who may prioritize their own career over yours.
Once you finish the degree, Changing fields is sub-optimal at this moment, but future studies may lead naturally to that, as it might for anyone.
Upvotes: 3 <issue_comment>username_2: It's a difficult and unpleasant situation. Your current advisor's behavior is unusual, but I have seen this before (long time ago).
Here are my suggestions:
Postdocs (in math) normally start in August-September. Your dissertation defense is likely to be in May or during Summer. My suggestion is to start writing your paper (which supersedes your advisor's results) now. You do not need your advisor's permission to write one. I assume that you have enough results for your dissertation without that work. Writing takes time (including polishing the presentation). Since it is already February and you will be busy with finalizing your dissertation, chances are that you will not even finish writing before the Summer starts. After your thesis defense and after everything related to your degree is finalized, finish writing your paper, put it on the arXiv and send to a journal of your choice.
Caveat 1: arXiv requires an endorsement if you are not an established mathematician. However, given that you got your postdoc, chances are that you already have one or more papers posted, in which case you do not need an endorsement (which usually comes from your advisor). If this is your first arXiv submission, just submit to a journal and wait until your postdoc starts before putting the paper on the arXiv.
Caveat 2: If you are not 100% sure that your proofs are solid, after your defense send the paper to your postdoctoral advisor and ask for an input (before submitting the paper to arXiv or/and a journal).
After you have submitted the paper, email a copy to your current advisor, saying in in the email that you have managed to iron all the kinks in the proof, thanking for suggestions of future directions, all the help, etc. There is a chance that this will not help and you will have spoiled your relationship with PhD advisor forever. You still have your new postdoctoral advisor whom you will be asking for future recommendation letters (as well as other mathematicians in your area).
This situation is unpleasant but (I believe) it should not prevent you from publishing the results.
Upvotes: 2
|
2023/02/07
| 590
| 2,536
|
<issue_start>username_0: I'm applying for a summer internship at a big biology research lab in my city. This is such a fantastic opportunity and it would be such a tremendous help careerwise to get this experience. The application asks for 2 letters of recommendation from "individuals familiar with your academic performance." I've asked 4 professors and have gotten one yes, one no, and 2 haven't answered. The application states that the letters of recommendation must be attached to the application so I need to receive the letters of recommendation. Because of this, I think some of my professors are hesitant to write them.
With the deadline 2 weeks away I'm getting kind of desperate, would it be frowned upon if I get a former lab partner to write my second letter of recommendation seeing as they are an "individual familiar with [my] academic performance?" They have seen me in a lab setting and are familiar with my work ethic and technique. I'm not exactly best friends with them but I'm sure they would help me out in a pinch. Is there anyone else I can ask?<issue_comment>username_1: You can ask, but it is sub-optimal. They may know about your performance, but are not in a good position to evaluate it in such a case. Moreover, peer evaluations might be suspect in any case, if friendship relationships can be assumed to exist.
If "haven't answered" means that they haven't responded to emails, then it is time to go visit them in person. You will be much more likely to get a response that way and you might even learn how strong it will be. Be sure to inform them about deadlines if they agree.
Upvotes: 2 <issue_comment>username_2: **No.** A letter of recommendation evaluates your suitability for the program by comparing you to others in the field. An undergraduate student, even if dispassionate (which is debatable), does not have the academic maturity to make such an assessment. While I'm sympathetic that "a bad letter is better than no letter," a letter from an undergraduate might not be accepted at all.
I think you should consider other options:
* Follow up with the professors who haven't responded, or to other professors. Even a graduate student, while borderline, is certainly better than nothing.
* If the "no" was definitely due to discomfort about providing the letter directly to you, perhaps they would agree to provide it in a sealed (i.e., with an institutional seal) envelope? You could then attach the envelope to your application. A little unusual, but shouldn't be a problem.
Upvotes: 2
|
2023/02/08
| 951
| 4,173
|
<issue_start>username_0: Recently one of my paper was rejected after review. The editor didn’t mention about the option of resubmission, just said:
>
> It is challenging for us to envision a straightforward revision process to fully address these concerns, and therefore I am sorry that we can't offer to further consider this paper in its current shape.
>
>
>
I’d like to know whether I can understand this as a possibility of potential resubmission after substantial changes of my paper upon the editor and the reviewers’ request or not.
I have a couple of good experiences with resubmission to the same journals, but at that time, editors always gave a clear option to resubmit my paper. This time, I’m not sure about whether the door to this journal is completely closed or not.<issue_comment>username_1: Does the communication from the editor state *explicitly* the paper is rejected or is this your interpretation of the situation after reading the comments?
I am asking because a student I know concluded their submission had been rejected whereas in fact the editor had nowhere said so.
If the editor does not state that the submission is rejected, a resubmission is possible, but the submission in its current form has limited chances of eventually being accepted without major changes.
I suspect the referees did not recommend publication for qualitative rather than specific reasons, something like “it’s not really that novel” rather than “good idea but this section needs to be expanded”. It is not impossible to reorganize a manuscript to address such “soft” objections, but definitely not easy.
Upvotes: 2 <issue_comment>username_2: I'm sorry to say but this reads like the most negative reaction of the editor possible. The editor doesn't see how even a substantial revision of the paper would lead to a path to publication. They can't forbid you to resubmit the paper however many times you want but their point of view is that you would need to change it enough to make it essentially a completely new paper to make it worthwhile to resubmit.
Upvotes: 6 <issue_comment>username_3: This seems just a polite way to say no to me.
Upvotes: 5 <issue_comment>username_4: I read this as "Maybe you could revise this in a way it could be acceptable, but we don't see it happening." In other words, the editor is strongly discouraging you from resubmitting without closing the door entirely. In my opinion, it's poor editing to not just say outright, in cases like this, that resubmissions wouldn't be accepted, rather than dangle a slim hope of eventual acceptance in front of the authors. But my feeling is it's pretty common. Editors who do this think they're being more generous by keeping an open mind, when really they're inviting you to waste your time on a resubmission that they don't want to see.
(The qualification "in its current shape" seems particularly ill-conceived. Of course the editor "can't offer to further consider this paper *in its current shape*", because this round of consideration is done now. The thing an author wants to know is whether a resubmission would be considered.)
Upvotes: 3 <issue_comment>username_5: This sounds like "no amount of mere *editing* will fix this paper: the data is bad in some foundational way that would require additional research to fix."
As such, I think your comment "The reason that we are considering resubmit the paper after a major surgery is that all the suggestions are available for us to do by performing extra experiments that will take at least 3 months…" seems like a good path forward.
You won't be merely editing and resubmitting: you will be doing months of additional work and essentially submitting a new paper based on the new findings. Sure, the new paper might have much of the same *wording*; but the important foundations will have been changed.
I'd write to the editor to ask whether they would find those additional months of time and investment to be a worthy change, though: their wording is vague. They might have some other, unstated reason to decline you, and they might then give a firmer no, rather than a "maybe if you make those radical changes".
Upvotes: 2
|
2023/02/08
| 862
| 3,791
|
<issue_start>username_0: I applied for a tenure-track assistant professor position in the US, was selected as a semi-finalist and had my first round of interview. The next step is that they will choose around three finalists to be invited for an on-campus interview. Does anyone have an idea how long it would take to be notified after the first-round interview in case one is selected as finalist? Thanks!<issue_comment>username_1: Does the communication from the editor state *explicitly* the paper is rejected or is this your interpretation of the situation after reading the comments?
I am asking because a student I know concluded their submission had been rejected whereas in fact the editor had nowhere said so.
If the editor does not state that the submission is rejected, a resubmission is possible, but the submission in its current form has limited chances of eventually being accepted without major changes.
I suspect the referees did not recommend publication for qualitative rather than specific reasons, something like “it’s not really that novel” rather than “good idea but this section needs to be expanded”. It is not impossible to reorganize a manuscript to address such “soft” objections, but definitely not easy.
Upvotes: 2 <issue_comment>username_2: I'm sorry to say but this reads like the most negative reaction of the editor possible. The editor doesn't see how even a substantial revision of the paper would lead to a path to publication. They can't forbid you to resubmit the paper however many times you want but their point of view is that you would need to change it enough to make it essentially a completely new paper to make it worthwhile to resubmit.
Upvotes: 6 <issue_comment>username_3: This seems just a polite way to say no to me.
Upvotes: 5 <issue_comment>username_4: I read this as "Maybe you could revise this in a way it could be acceptable, but we don't see it happening." In other words, the editor is strongly discouraging you from resubmitting without closing the door entirely. In my opinion, it's poor editing to not just say outright, in cases like this, that resubmissions wouldn't be accepted, rather than dangle a slim hope of eventual acceptance in front of the authors. But my feeling is it's pretty common. Editors who do this think they're being more generous by keeping an open mind, when really they're inviting you to waste your time on a resubmission that they don't want to see.
(The qualification "in its current shape" seems particularly ill-conceived. Of course the editor "can't offer to further consider this paper *in its current shape*", because this round of consideration is done now. The thing an author wants to know is whether a resubmission would be considered.)
Upvotes: 3 <issue_comment>username_5: This sounds like "no amount of mere *editing* will fix this paper: the data is bad in some foundational way that would require additional research to fix."
As such, I think your comment "The reason that we are considering resubmit the paper after a major surgery is that all the suggestions are available for us to do by performing extra experiments that will take at least 3 months…" seems like a good path forward.
You won't be merely editing and resubmitting: you will be doing months of additional work and essentially submitting a new paper based on the new findings. Sure, the new paper might have much of the same *wording*; but the important foundations will have been changed.
I'd write to the editor to ask whether they would find those additional months of time and investment to be a worthy change, though: their wording is vague. They might have some other, unstated reason to decline you, and they might then give a firmer no, rather than a "maybe if you make those radical changes".
Upvotes: 2
|
2023/02/08
| 2,293
| 10,052
|
<issue_start>username_0: I recently messed up my presentation in front of the audience because I couldn't answer or understand their questions well enough (even after I asked them to clarify their question). I keep ruminating about the moment and can't focus on work. I'm wondering if you have any strategies to overcome the embarrassment.<issue_comment>username_1: At the moment you just need to move on and recognize that not everything will always be perfect. Take a deep breath, have a culturally appropriate beverage.
For the future, when you don't understand a question and can't say something sensible, ask the questioner to "take it off line" and to either discuss it with you later or to contact you at their convenience.
Upvotes: 6 <issue_comment>username_2: Public speaking is hard for almost everyone. From a previous question of yours, it seems you also find it hard. Fortunately, like many human endeavors, one can get better at it and ultimately more comfortable.
First, training on public speaking and presentations may be available at your institution. You should ask around in your department, school, or a graduate advisor office if there is such training available. It could be anywhere from an hour presentation up through a multi-day short course with specific instructor feedback on short presentations that you have done.
Second, practice practice practice. This could be something like [Toastmasters](https://www.toastmasters.org), or just be part of your normal research group activities. Where I did my postdoc, grad students were expected to present a slide or more on what they had done lately at the weekly group meeting. They got lots of practice standing in front of the group and trying to explain things while getting questions on it. This also meant they had to figure out what the questions where about, since they were not always clear from their fellow students. It also meant that by the time they got to a conference talk, just about every question that could be asked about the work had already been asked of them.
Third, teach a class or review section. While public speaking practice in general, it also is a situation where you need to think on your feet as you get asked questions by students.
Public speaking is one of the biggest fears of just about everyone. You can get better at it, with time and effort.
Upvotes: 4 <issue_comment>username_3: With the exception of ultra-rare congenital superstars, we've all been there. username_1 and Jon give good technical advice. One thing that's helped me on the emotional side in similar circumstances is going to the end-of-conference social event and seeing that the folks with whom my talk had bombed didn't think any less of me personally, we were all still friends.
Upvotes: 3 <issue_comment>username_4: I agree with others that it's not as big an issue as you're probably making it at the moment, and you'd be well served by taking a deep breath, relaxing about it, and just moving forward. Everyone in academia has attended their share of bad talks, and it's nothing you can't bounce back from.
That said, I'd also recommend you try to turn the rumination into reflection. -- The distinction here is that "rumination" implies dwelling on the issue, continually poking hurt feelings without any sort of forward progress, while "reflection" is ideally an objective (emotionally neutral) process by which you attempt to discern how you might better your performance next time.
In particular, it sounds like the Q&A section was the hard part for you. You probably want to reflect (rather than ruminate) about why you had issues with it. Were the questioners using unfamiliar terminology? Were they asking about a sub-area which you're unfamiliar? Were they coming at the issue from a different perspective? (e.g. a pure mathematician and an experimental physicist attending the same astrophysics talk will have very different ways of looking at things) Or was it just that you were nervous and flustered, and couldn't make sense of things at the time?
One suggestion is to approach trusted colleague(s) who were in attendance at the talk and see how they interpreted the questions. Were they able to understand the questions? If not, keep in mind you do sometimes get befuddled questions from audience members who are deep in the Dunning–Kruger sweet spot for the topic. On the other hand, if your colleague(s) did understand the questions, discuss things with them and reflect to figure out what you were missing at the time. What was it about the question which caused you not to be able to interpret it at the time? What extra thing did you need to know to be properly able to understand the question?
Hopefully, objective reflection (rather than rumination) will point you to what you were missing at the time, and that will present an obvious course of action. (e.g. If you were simply nervous, practice more to gain confidence. If you didn't understand the terminology or perspective used, read more widely from that angle. If it was a sub-area you aren't familiar with, become more familiar with that area.)
---
As a final note, keep in mind that questions which are formed during a talk are often predicated by missed understandings (this includes situations where it's missed because the speaker forgot to mention it). Instead of spending a bunch of time (and awkwardness) getting the questioner to more precisely define the question, it's often worth simply identifying the part of the talk (e.g. the slide) the question is about and then simply re-iterate or clarify the point at issue. This will either (1) answer the question (2) prompt the person to formulate a more precise follow-on based on your recent phrasing or (3) let the questioner realize that the question isn't worth it. -- I've seen more than one senior professor use this approach to address questions they obviously didn't fully understand.
Upvotes: 4 <issue_comment>username_5: Probably no-one apart from you is going to remember in a few days time, so learn from the experience and think about how you could do it better next time.
The era we live in seems to be replete with this attitude that we need to constantly beat ourselves up all the time when we mess things up, which is definitely not healthy in the long run. So you should go easy on yourself, relax and then identify if there is anything you could do to avoid a similar situation happening again.
Upvotes: 4 <issue_comment>username_6: There are several reasons for which one does not understand a question well enough:
1. The person asking the question uses terms/concepts coming from an other field or outside the knowledge of the speaker.
2. The person asking the question is not familiar with the subject of the topic hence the question might not have a precise/formally correct sense.
3. Some concepts are named in a way by a community, an other one by an other community and we can have a "language" issue.
There are also reasons for which the speaker cannot answer the question.
1. The question is actually subtle, and the answer could be the scope of a paper.
2. The question is quite broad.
3. The question does not make sense (I did not say it was the case in the context of the mentioned talk, and this does not have to be taken in a condescending way). This can happen when the audience is not familiar with the topic and can of course happen even if you are familiar with it.
4. Some questions are almost impossible to answer, like: "this makes me think about (replace by anything you want). Is there a connection between these two concepts?" Of course you can answer if there is one but it will be hard to say that the two concepts are unrelated. It is not impossible that someone in a few years will indeed make the connection.
5. Some questions require some thinking process but you do not have the time during a talk, like: "do you have an example satisfying this but not that".
6. Some are unexpected (both in terms of timing and content) and even with a good preparation, we cannot anticipate everything.
7. Lack of expertise in some parts of what was presented, which can happen very easily in a joint work, no matter how good one is.
These were reasons for which the situation of not answering a question can happen. Now the thing is what to when this happens. There is no miracle solution, but here are some possible approaches:
1. If you are for example in the situation of 5. (the story with example), you can still give your opinion about the question: what makes you think that such an example can exist, some names of authors that may have explored these kind of things.
2. Try to write down the questions that were asked just after the talk in order to not to forget. For the next talks, it will help to see the kind of questions you can have and try to anticipate.
3. Observe actively the speaker's behavior when you are yourself in the audience of a talk and especially during the question time.
Upvotes: 6 [selected_answer]<issue_comment>username_7: The question is:
>
> I keep ruminating about the moment and can't focus on work. I'm wondering if you have any strategies to overcome the embarrassment.
>
>
>
Absolutely - give another talk as soon as possible!
While @username_1's "[culturally appropriate beverage](https://academia.stackexchange.com/a/193319/69206)" humorously suggests diluting your memory of the experience with something (potentially) other than tea, what really works is to dilute it and make it less relevant by adding more memories of talks that go better.
As soon as possible, schedule another talk, perhaps with a more familiar audience, perhaps with a slightly simpler topic, and welcome questions and answer some.
Distract yourself by having a more immediate and more positive experience, such that this one occupies less space in memory.
While other answers focus on improving, best way to overcome falling off of a bicycle is to get back on the bicycle and put some bicycling distance between you and where you fell.
Upvotes: 3
|
2023/02/08
| 687
| 2,369
|
<issue_start>username_0: I recently attended a talk and the talk was recorded and posted to YouTube, and the slides are available on the author's webpage. I would like to cite this talk in some work I am doing, and there are no other references for this material. How should I cite this?
This is the talk <https://www.youtube.com/watch?v=meOQXoN0QRc>
And these are the slides <https://web.ma.utexas.edu/users/dafr/Anomaly.pdf><issue_comment>username_1: You can cite is pretty much as you do here along with author and title, but add a date at which you accessed them as you should for any internet (or other volatile) reference. Alternately cite a reference at the wayback machine.
For a talk, give the date and venue of the presentation for things you visit "live".
Upvotes: 3 <issue_comment>username_2: Absent a specific rule, I would cite the video using "<NAME>" as your author, "<NAME> | Feb. 7, 2023 | What is an anomaly?" as the article name (because that's how it's labeled, even though it duplicates other information), *Mathematical Picture Language* as the publication/journal, and Youtube.com as the publisher. Note the date of the presentation as if it were the publication date (which it looks like it actually was). List the link and the "retrieved on" date. Do this all according to the the citation system (MLA, APA, etc) that you are using. For the slides, I'd probably treat U Texas as the publisher.
In any case, the important thing is to give credit *and* make it possible for someone else to find it. The technicalities for web materials are not fixed like they are for books, so just try to be as consistent as possible.
Upvotes: 4 [selected_answer]<issue_comment>username_3: username_1 is correct, especially about using the Wayback Machine for the pdf slides.
In LaTeX with biblatex, your entries would be:
```
@online{freed_youtube,
title = {What is an anomaly?},
date = {2023-02-07},
author = {<NAME>},
url = {https://www.youtube.com/watch?v=meOQXoN0QRc},
urldate = {2023-02-09},
organization = {The Mathematical Picture Language Project}
}
@online{freed_slides,
title = {What is an anomaly?},
date = {2023-02-07},
author = {<NAME>},
url = {https://web.archive.org/web/20230209103515/https://web.ma.utexas.edu/users/dafr/Anomaly.pdf},
urldate = {2023-02-09}
}
```
Upvotes: 2
|
2023/02/09
| 847
| 2,605
|
<issue_start>username_0: Because of [reasons](https://github.com/Langenscheiss/bibitnow/issues/59), I'm curious to know whether there are any simple examples of papers where two or more authors have the same full name.
(To be clear: I am *only* interested in *exact* full-name matches. Instances where authors share a last name are so common as to be completely unremarkable, and instances where the first name matches but additional initials distinguish the authors are also reasonably common and *not* what I'm looking for.)
Does anybody have any particular pointers that can be useful?<issue_comment>username_1: Here's an example with co-first authors who share the same name:
* [<NAME>, <NAME>, <NAME>, <NAME>, <NAME>, <NAME> & <NAME> "Comparative proteomic analysis reveals novel insights into the interaction between rice and Xanthomonas oryzae pv. oryzae." *BMC Plant Biology* **20**, 563 (2020)](https://doi.org/10.1186/s12870-020-02769-7).
From physics I can only think of papers where two authors share initials and last name, but don't have the full name written out. An example, though perhaps not so simple, is [this one](https://doi.org/10.1103/PhysRevLett.112.091804) from the ATLAS collaboration. Its author list includes two instances each of "<NAME>", "<NAME>", "<NAME>", "<NAME>", "<NAME>", "<NAME>", "<NAME>", "<NAME>", "<NAME>", "<NAME>", "<NAME>", "<NAME>", "<NAME>", "<NAME>", "<NAME>", and "<NAME>".
Upvotes: 6 [selected_answer]<issue_comment>username_2: A recent example:
Otto , P., & Otto , P. (2022). Impact of Academic Authorship Characteristics on Article Citations. REVSTAT-Statistical Journal, 20(4), 427–447. <https://doi.org/10.57805/revstat.v20i4.382>
which includes the following footnote:
>
> To the best of our knowledge, this is the first published article where both authors share their surname and given name, while working at the same university. Thus, the two authors are largely indistinguishable, which highlights the importance of individual author identifiers like ORCID.
>
>
>
Upvotes: 5 <issue_comment>username_3: They do not share the *full* name, only the last name:
[<NAME>, <NAME>, <NAME>: *On power-law relationships of the Internet topology* ACM SIGCOMM Computer Communication Review, Volume 29, Issue 4O,ct. 1999, pp 251–262](https://dl.acm.org/doi/10.1145/316194.316229)
What is also amusing, is that the "Faloutsos" last name is extremely rare in Greece. According to [this](http://faloutsos.com/) they are siblings.
Upvotes: -1
|
2023/02/09
| 718
| 3,105
|
<issue_start>username_0: So, I'm about 2 months away from submitting my PhD Qualifying Report (QR).
I've spent the last 2 years (part time) doing a thorough literature review and writing significant amounts of software to analyse a relevant public dataset that is relevant to where I want to take my thesis and I believed hadn't been thoroughly analysed before. I'm about 2/3s of the way through writing a paper based on this analysis.
Now, to my horror, I found a paper tonight doing *exactly* the analysis I've spent the last year doing. I hadn't found it because it wasn't published in a journal, and doesn't seem to be indexed on Google. I found it deep inside a login-only area of the website of the organisation that owns the dataset.
So... Obviously, I'm going to email my supervisors in the morning to try and work out what to do but I guess my question is: is this salvageable? Is the work I've done essentially worthless as far as my QR is concerned? Obviously I've learned a lot, but I'm not sure if that matters for the QR.
Is my only option to pivot, pull 2 months of all-nighters and hope I can scrape through the QR on a different premise?<issue_comment>username_1: What you describe is fairly common. You can't always know what parallel research is going on while you do your own study. In "hot" areas of research it is much more common.
It may be that for the purposes of your degree, the advisor and the institution generally will judge that you are fine and that your work has value enough to let you proceed. It might not result in publishable work, of course.
It might also be that you have learned enough in what you have done so that you have enough that goes beyond what has already been done and still do something publishable. Your advisor can help with that judgement.
An outside possibility is to try to establish a collaborative relationship with the authors of the report you found. Perhaps together you can find a suitable advance that won't set you back.
But in the worst case, you will need to take up a different track. Hopefully the university will help in avoiding that outcome.
---
I know of a case in which two people, known to each other, finalized theses that were almost identical in ideas (not words). It turned out that they had worked completely independently of one another. It took a year of inquiry, but both were awarded degrees after it was decided there was no hint of plagiarism. Both their advisors were well known and known to each other. The problem they solved was considered very important at the time. The importance of the problem led to their independent interest in working on it.
Upvotes: 4 [selected_answer]<issue_comment>username_2: From the point of view of your thesis committee you might need to give them only original work. But from the scientific point of view you just stumbled by chance on a collaborator who actually already completed their task, so once you finish your analysis you can compare the results and methods, and hopefully write a great paper together. Maybe that is an argument you can put forward.
Upvotes: 1
|
2023/02/09
| 594
| 2,642
|
<issue_start>username_0: I’m applying to schools for my Masters in Computer Science and I’m writing my statement of purpose. I don’t really know what I want to do for my project/thesis, but in their prompts, the schools ask for a description of tentative research plans.
I could be interested in researching the intersection of CS with topics/disciplines like psychology, kinesiology, sports, and humanitarian work. That really is me just spitballing though.
I don’t know if simply stating that I’m interested in studying the intersection of CS and one or multiple of these topics would be sufficient, or if I should attempt to come up with more specific project ideas so I can elaborate more.<issue_comment>username_1: What you describe is fairly common. You can't always know what parallel research is going on while you do your own study. In "hot" areas of research it is much more common.
It may be that for the purposes of your degree, the advisor and the institution generally will judge that you are fine and that your work has value enough to let you proceed. It might not result in publishable work, of course.
It might also be that you have learned enough in what you have done so that you have enough that goes beyond what has already been done and still do something publishable. Your advisor can help with that judgement.
An outside possibility is to try to establish a collaborative relationship with the authors of the report you found. Perhaps together you can find a suitable advance that won't set you back.
But in the worst case, you will need to take up a different track. Hopefully the university will help in avoiding that outcome.
---
I know of a case in which two people, known to each other, finalized theses that were almost identical in ideas (not words). It turned out that they had worked completely independently of one another. It took a year of inquiry, but both were awarded degrees after it was decided there was no hint of plagiarism. Both their advisors were well known and known to each other. The problem they solved was considered very important at the time. The importance of the problem led to their independent interest in working on it.
Upvotes: 4 [selected_answer]<issue_comment>username_2: From the point of view of your thesis committee you might need to give them only original work. But from the scientific point of view you just stumbled by chance on a collaborator who actually already completed their task, so once you finish your analysis you can compare the results and methods, and hopefully write a great paper together. Maybe that is an argument you can put forward.
Upvotes: 1
|
2023/02/09
| 1,589
| 6,686
|
<issue_start>username_0: After a career in the industry, I started to work in research as a graduate student/research assistant, and I am learning to adjust my working and thinking style. That process is a little painful, but I also enjoy broadening my perspective.
I am genuinely thankful to work in a great team with bright and friendly people. Everyone is younger than me, including the Head of the department. I can imagine writing a dissertation someday if I find an interesting topic.
The thing that puzzles me most is the process of writing papers. My Head of the department and two colleagues that have more experience seem to be especially good at it. While I seem to be good (or very good? I get good feedback.) at writing sections of papers that I get assigned, the initial idea-finding and structuring work elude me completely. I listen to these guys talking about paper structures on topics I am familiar with, and it is as if they speak in another language. So I decided to play on "easy mode" and structure a paper that I had planned to be the lead author on in the classical approach, following the steps:
1. This is the problem.
2. This is how I show it's a problem.
3. This is how I solve the problem.
4. This is how I show that I solved the problem.
And be done with it. So I called a meeting with my colleagues and the Head of the department and presented it (poorly). In the same session, my experienced colleague presented his paper idea and had already picked a conference with a date and deadline and a Call for Papers. And not only that. The Head of the department went off and looked for stories to tell around my topic and how to use it in connection with historical data that we got and how that would distinguish us in the eyes of reviewers and that it would be best to present practical, applicable research results to go with it. Now he is much better at this game than I (and everyone else in the room), and his advice is excellent. But I also feel pretty overwhelmed and discouraged. I intended to keep it simple and not overextend myself on my first try, and he raised the bar and expectations by a factor of three. I bet he didn't even notice because it comes to him as quickly as breathing.
My wife, who is also a professor, says I should not give up and not be discouraged.
So I want to know:
* How do I learn that way of thinking?
* What are the questions I need to ask myself?
* Are there checklists or so online?
* How can I avoid burning out on my first paper?
* What do I miss?<issue_comment>username_1: My papers have a specific structure; i.e., I think of writing a paper as filling in a 'form'. Consequently, writing a paper becomes easy because its organization is given. For each section and its corresponding paragraphs, I know exactly what type of information to gather and also their flow. All I have to decide is which information or arguments or facts to use for each section and paragraph.
Upvotes: 2 <issue_comment>username_2: There Are Only Two Good Ways To Learn To Write Research Papers
--------------------------------------------------------------
**1. Read Research Papers**
If your current path requires publications or original research, you have to be reading academic/research papers anyway. When you do, pay attention to how they are written. They have structure associated with them, and they really don't vary all that much.
When it comes time to write something up, you will have a venue in mind, so go read (or at least skim) some of the recent papers from that venue. Read them specifically for how they are organized and structured.
**2. Write Research Papers**
The only way to truly get good at something is to do it. But there are built in guard rails in most academic settings to prevent you from writing your first paper "blind."
One will be the publishing guidelines of whatever venue you select. Much of this will be devoted to minutiae that you're really not worried about right now-- margin spacing, fonts, captioning of diagrams, etc. But in my discipline, those documents are usually structured like the academic papers themselves. I've been known to use them as templates.
Another will be writing workshops or perhaps even full fledged writing courses given by your university. Depending on your prior writing experience (or lack) these can be very helpful.
Finally, it's not reasonable to expect a starting graduate student to produce a flawless research paper, as the sole author, the first time. That can happen, but that is not the expectation. My cohort of graduate students generally co-wrote our first papers with our advisors in some form or another, even if it is just "Student writes draft, advisor gives feedback, lather, rinse, repeat." This is part of what you are learning.
By the *end* of your program, you should be much more accomplished at this, and need minimal guidance. But at the beginning, yes, it can seem daunting.
Upvotes: 2 <issue_comment>username_3: There is a good deal of literature to suggest that to be able to write good academic papers one needs to master multiple skills. The good news is that the skills are both identifiable, separable, and learnable. I have found that it has been helpful to have a guide to the various skills involved, and also helpful to have a way of practising the skills individually rather than trying to master all of them simultaneously. By identifying and separating the various skills, I have found the process of writing much less overwhelming than I used to, although I would hardly claim to be an expert.
I wish that I could claim to be the originator of the ideas in the previous paragraph, but in fact I have learned them by first discovering (in my university library), and then obtaining, a copy of the book "Writing Your Journal Article in Twelve Weeks: A Guide to Academic Publishing Success" by <NAME> (ISBN: 9780226499918), and by working steadily through the book not just once, but several times.
I commend the book to you.
Upvotes: 1 <issue_comment>username_4: Throughout the course of my Ph.D., I've been collecting resources, books, articles and videos on improving academic writing. I collected links to all of them in this [Notion page](https://username_4.notion.site/Improving-your-academic-writing-42da2498f8f14e3a9c9e20a068cbe472).
All of these have helped me enormously. Not only in improving my writing, but also in gaining clarity of thinking, and in developing research ideas. It has been a long journey, but with enough time dedicated to writing regularly, drafting, and editing, you will improve. ...and you might even find that academic writing is enjoyable! Good luck!
Upvotes: 2
|
2023/02/10
| 1,852
| 7,469
|
<issue_start>username_0: [Athena Swan charter](https://www.advance-he.ac.uk/equality-charters/athena-swan-charter) is a UK organization promoting gender equality in academia, among other things. University departments apply for gold / silver / bronze Athena Swan certificates, while providing an action plan of promoting gender equality.
In STEM Departments, typical goals in an action plan may include percentage targets for:
1. female PhD students
2. women in Lecturer / Senior Lecturer position
3. female speakers in seminars
The question: **is this legal in the UK to set such recruiting goals?**
By the 2010 Equality Act, positive discrimination, i.e. hiring someone based on their protected characteristic, such as age, race or gender, when they are not the most suitable candidate, is illegal in the UK, see [here for some examples](https://www.davidsonmorris.com/positive-discrimination/), including a successful litigation case.
Do not goals (1), (2), (3) link directly to positive discrimination, hence form an illegal practice in the UK?
(3) is more subtle, as no actual employment is involved, but it also seems to be a more blunt case of positive discrimination, as having more female speakers than the average in the field (whatever this average is), provided that the ability is equally distributed among the genders, will most likely imply that some better, male speakers will not be invited.<issue_comment>username_1: You are making the wrong assumption here that aiming for some minimum percentage of women would involve selecting a weaker women over a stronger man.
I am involved with running a conference series in CS/Math, where the expectation is that at least a third of the invited speakers are women (the area is so overwhelmingly male that this is already a lofty standard). Whenever I was on the programme committee, this was achieved just by reminder all PC-members to not forget about nominating great female speakers. There are a few subconscious biases around in our society that make it easy to overlook women, and just taking a moment to consciously think about "Aren't there some women around who give great talks/have done some fascinating work recently?" can lead to the realization that of course, yes, there are. Not once I have witnessed a situation where people thought Bob would be the better choice, but then invited Alice because she was a woman.
Similarly, when it comes to Athena-Swan-related questions like "What can our department do to increase the percentage of women faculty?", the two top approaches I've heard people voice are: A) "Let's make sure that our department provides a welcoming work environment for women." B) "Let's make sure that highly qualified women applicants know about A)." Not once have I seen any indication of "Let's just hire some mediocre woman!" as being part of the strategy.
When it comes to hiring decisions themselves, there are a number of well-documented subconscious biases around which can lead to people assessing female applicants as weaker than they are. Attempting to avoid those is simultaneously going to work towards ensuring that appointments are as merit-based as possible, while also increasing the percentage of women.
Upvotes: 6 <issue_comment>username_2: IANAL and I am not qualified to provide a binding answer to a legal question.
However, from a purely general point of view, a University may set up a goal of reaching a certain proportion of under-represented staff. Your question seems to imply that the only way of achieving this goal is through (positive) discrimination, that is not necessarily true.
Athena Swan encourages universities to identify why certain groups are under-represented and make actions to remove barriers. These actions often end up being useful for both the under-represented group and for other candidates as well. For example, it may not be easy for women to secure a job or promotion in academia, if a job description specifies that candidates have to secure X grants in N years after their PhD.
Women are more likely to take an extended career break due to maternity leave, and therefore may find it more difficult to meet such time-restricted criteria. As an example, relaxing the criteria to take in account career breaks helps women to compete more fairly for the positions. But it also helps men who had career breaks due to the health issues or family issues etc. At the end, everyone benefits from such adjustments and no discrimination happens.
Upvotes: 3 <issue_comment>username_3: There are two main factors at play here. Firstly, as suggested by some comments above, there are ways of attempting to increase the number of women in various roles, such as ensuring a supportive environment or combating the perception of some departments as a 'male-dominated' area. These do not directly impact on the hiring process itself at all.
However, even in the hiring process, there is a [provision in the Equality Act](https://www.legislation.gov.uk/ukpga/2010/15/part/11/chapter/2) which is broadly supportive of a range of types of 'positive action' to redress historic disadvantages due to protected characteristics. This includes targeted support, promotion of the role to disadvantaged groups, and, in the limit of otherwise tied applicants (which is arguably very rare), taking the protected characteristic into consideration during hiring decisions. There are a number of limitations on this, so they can't e.g. have women-only openings (officially or unofficially) but there is a range of positive actions available which are still legal to increase the representation of disadvantaged groups.
Upvotes: 4 <issue_comment>username_4: I think others have done a pretty good job of answer the question about 1 & 2 (although I'd note that the PhD students are not employees, and therefore don't fall under the ban on positive discrimination for employees): Its not illegal to set recruitment targets as long as your proposed means for achieving those recruitment targets don't involve privileging individual women over individual men.
However, for your third point, seminar invites, (and perhaps PhD students, I don't know), the law (Equalities Act 2010, paragraph 158(2) )explicitly allows for positive action:
>
> This Act does not prohibit P from taking any action which is a proportionate means of achieving the aim of—
>
> (a) enabling or encouraging persons who share the protected characteristic to overcome or minimise that disadvantage,
>
> (b) meeting those needs, or
>
> (c) enabling or encouraging persons who share the protected characteristic to participate in that activity.
>
>
>
Upvotes: 3 <issue_comment>username_5: There's a relevant older post:
[Is it lawful for a fellowship linked to a permanent faculty position at a British university in the STEM field to only be available to females?](https://academia.stackexchange.com/questions/78591/is-it-lawful-for-a-fellowship-linked-to-a-permanent-faculty-position-at-a-britis)
The top [answer](https://academia.stackexchange.com/a/78592/16078)
>
> In the UK, there are a number of fellowships exclusively for women (e.g., L'Oréal WISE fellowship). The Equality Act of 2010 allows for positive action outside of recruitment which these fellowships likely fall under. The Equality Act of 2010, however, does not allow for positive action in recruitment (apart from giving preference for equally qualified applicants).
>
>
>
Upvotes: 2
|
2023/02/10
| 545
| 1,878
|
<issue_start>username_0: I am writing in a description of three items in a thesis on Particle Physics, and I can do it in one line in theory. If there were just two items, e.g. A and B, which produce results C and D, I could write it like so:
>
> When using A (B) to estimate the upper limit, the best fit result is C (D)...
>
>
>
where A is associated with result C, B is associated with result D.
I wondered if there was a convention for three items that you'd ever come across? Or would this just be 2/3 separate sentences? It's just the following don't look right, I wondered if there was a convention people had seen in publications previously?
>
> Ex.1. When using A (B) (C) to estimate the upper limit, the best fit result is D (E) (F)...
>
>
>
>
> Ex.2. When using A (B (C)) to estimate the upper limit, the best fit result is D (E (F))...
>
>
><issue_comment>username_1: I wouldn't use parenthesis for this at all, but rather use 'respectively':
>
> When using A, B or C to estimate the upper limit, the best fit result is D, E or F respectively.
>
>
>
This becomes a bit harder to read if you have many objects to list, but for three I think it would be fine.
Upvotes: 7 [selected_answer]<issue_comment>username_2: Another option besides the good answer from username_1 would be:
>
> When using A (B, C) to estimate the upper limit, the best fit result is D (E, F)...
>
>
>
or, based on Stef's comment,
>
> When using A (respectively B, C) to estimate the upper limit, the best fit result is D (E, F)...
>
>
>
It is slightly less clear than a more explicit sentence would be, but is a little more concise and might work well in certain contexts where brevity is required or the construction will be used frequently.
I would not use either of the suggestions in the original post, and I think they would lead to confusion.
Upvotes: 3
|
2023/02/10
| 694
| 2,872
|
<issue_start>username_0: I have heard lots of stories about rejection based on unfair reviews, etc. But this is a new one for me, and I wanted to seek the Academia SE's advice...
We recently submitted a paper (let's call it A). Paper A got reviewed (Major Comments) and we revised the manuscript and resubmitted (let's call it A.R1). Recently we got the review of the revision and it was "rejected". I can live with the rejection, but here is where things get messy and complicated... The attached review and rejection were based on a completely different paper with a completely different title and a completely different set of authors (let's call it paper B). So basically the rejection was based on the rejection of paper B that a reviewer had rejected.
Now comes the even weirder part, the review attached to A.R1 states that paper B is rejected. But Paper B was actually published in the same journal around the same time frame! So, it seems like B got an acceptance based on A.R1's reviews.
I am really mind-boggled by this incident. I did reach out politely to seek more information, but the associate editor was not any help and denied everything. How to proceed?
Field: Civil, Energy, Petroleum
EDIT - July 08, 2023:
Emailed the editor and the principal editors twice in the past 4 months without a response. Was finally able to get the Elsevier chat to contact the "Journal Manager".
side note: Its weird how Springer has an ethics complaints Email, but ElSevier does not.<issue_comment>username_1: Editors, reviewers, editors-in-chief, and journal staff are all people, not machines. Mistakes happen. In a case where the evidence that a mistake happened is clear, lay out your case and go up the chain one step at a time until someone listens. If the associate editor in charge is too proud or scared to admit an error, then go to the editor-in-chief. If that person is too busy to be helpful, go to the journal admins, and from there to the publisher's publication board.
Upvotes: 5 <issue_comment>username_2: That looks like they clearly made a mistake. It's hard to believe that the associate editor when contacted wouldn't be of any help. I'd expect "sorry, this was the wrong review, here we send the right one", or "the reviewer may have mixed things up, we try to solicit another one". If the situation is as I get from what you write, next thing to do is contacting the main editor. Also trying once more to get something better out of the AE may be worthwhile.
A worthwhile addition from the comments by @Dawn:
"I would also have another party read your email to the AE before sending to make sure it is very clear. Perhaps they were unhelpful because they were confused about the nature of the mistake."
(Not sure whether this adds much to the answer of @username_1 but I was asked to put it here so that is that.)
Upvotes: 6 [selected_answer]
|
2023/02/11
| 737
| 3,277
|
<issue_start>username_0: I am a PhD student working at the intersection of applied mathematics and machine learning. I'm still early in my PhD, but I'm unhappy because my current research direction is too focused on applications as opposed to theory. I consider myself a mathematician and would like to prove theorems primarily. My advisor has mostly given me projects that involve applying methods he's previously developed to new problems and is heavy in coding (which I do not enjoy very much).
I told my advisor that I wanted to do more theory, and while he seemed receptive, he hasn't offered me many options. This seems to stem from several factors:
1. He will oversell how much theoretical work there is in my current projects. He will often comment that we showed such and such, but these results are superficial.
2. He seems reluctant to offer me "hard" problems: he will sometimes discourage me from exploring something as it would be challenging to produce results.
3. I do not have a strong sense of what I would like my research direction to be and therefore am unable to come up with problems of my own: I believe that my advisor will be receptive if I come up with an impactful problem to address. I feel like I am in a chicken-and-egg situation.
I know that this is not a lack of ability on his part as he's done some very theoretical work previously and because he's advised (albeit stronger) students who have produced interesting theoretical work. I therefore have two questions:
1. How do I properly communicate these concerns, given that I have already tried to?
2. How can I learn to formulate my research direction and find problems that are both impactful and that I find interesting?<issue_comment>username_1: You have communicated, actually, but aren't quite satisfied with the results. You can keep hammering if you think that might work, but there is another option. You are currently working toward things that are quite applied as you note yourself. If you are successful at it then it will get you a doctorate. From then on you can control your own research trajectory.
So, one option is to just finish expeditiously and then work with others who have a theoretical focus as your career progresses. There is no bar to that. In other words, work toward the long term, knowing that you may not get all you want in the short term.
Aside from that, keep a notebook of ideas that arise from your current studies, especially those with a more theoretical focus and that you can follow up on later or in spare moments now. And make connections with theoreticians along the way. You don't need to depend solely on your advisor for ideas, though you have to satisfy them about your progress.
Your career will span many (many) years. You will have many opportunities along the way to think new thoughts and pursue new pathways.
Upvotes: 2 <issue_comment>username_2: I would recommend an open conversation with your supervisor. You may want to start from what you are doing now and suggest a more theoretical approach as well as a path towards a publication of the results.
I believe that you should have a good time working with your supervisor. You should "click" together. If that is not the case you may want to consider changing it.
Upvotes: 0
|
2023/02/11
| 2,142
| 9,144
|
<issue_start>username_0: My imposter syndrome is focused around the academic job market (which is different than other posted questions), and it comes in two parts:
(1)
My feelings of being an imposter have been exacerbated by poor treatment and abuse from my PhD advisor over the years who has, in moments in frustration, remarked derisively that I have the worst technical ability of any student in the lab, do not fit in, and would have better off with an advisor working in more applied topics. The only reason why he kept me employed as his student was because I worked hard, said 'yes' to every thing he asked me to do and basically failed to set any personal boundaries, allowing him to step all over me. These repeated remarks have been extremely cruel and I have thought about quitting and even ending my life several times. That being said, I know deep down that there is some truth to his comments.
Fast forward several years, and I am privileged enough to be interviewing at some very prestigious universities for academic positions (top 5 places). But I am fully aware that this is largely due to my brilliant and extremely controlling advisor who has been the driving force behind the success of a lot of my work.
While my other prestigious letter writers have raved about me and written exceptionally strong letters, I can't help but feel that they do not know the truth: which is that the image that they see of me is one that is carefully crafted by my advisor. But once I graduate and leave the influence of my advisor I will stumble and not meet their expectations.
(2)
I am exceedingly anxious about interviewing at these top 5 institutions due to a fear of being looked down upon by faculty who are clearly, and quantitatively more intelligent and brilliant than me. I have heard of stories where faculty will just ignore you at dinner if they've figured out you're not at their level of intellect.
At most places however, even at slightly less prestigious institutions (top 5-15 places) I do not get this anxiety. The faculty are more down to earth and I can typically find faculty who are on a similar 'competency plane' as me, who I can relate to. But that is not the case at the most top institutions.
I am also quite introverted. I am never the center of attention (in fact I hate it), and I do not naturally fit the mold of the social and intellectual butterfly that can wow people at dinners.
Even if I get an offer from these prestigious institutions, I am hesitant about taking the offer given that I may always be a "small fish" in a big pond for the rest of my career. And I believe I would be happier at a slightly less prestigious institution.
Logically speaking, I wonder if it even makes any sense to interview at these prestigious institutions aside from social obligation to my letter writers + nobody in their right mind would actually cancel interviews at these places.
It is a popular trend to suggest that everyone has imposter syndrome and everyone belongs. But the cold harsh truth is that people do get found out as imposters and those people don't get tenure!
I understand how this might seem like a vent from a privileged person, but I would love to hear other people's thoughts about this general topic of dealing with imposter syndrome as I think it can apply to a lot of people.<issue_comment>username_1: Let me just clarify a couple of misconceptions that you seem to have from what you have written:
* There is no need to "wow people at dinners". All workplaces have more extroverted and more introverted people, and it is by no means true that the person that does most of the talking/jokes is the most liked by everyone.
* There is no need to go to the top institution. You should work at a place where you are comfortable and will enjoy yourself. Forget about statements like "nobody in their right mind would actually cancel interviews at these places". Those statements are wrong.
That said, I will refer you to [the existing question](https://academia.stackexchange.com/q/11765/112007) for dealing with impostor syndrome. Your success was achieved by you, and not by your advisor. All the controlling in the world cannot make a person successful if that person does not have the ability.
I would go to the interviews (also at the top places) and see how they make you feel about the place. If it feels too competitive or that people look down on others, you can always reject and go to a place that is a better fit for you. But why not give it a try? Even if the interview is uncomfortable, if you do not take the job then it will just be a story to tell.
Upvotes: 3 <issue_comment>username_2: *Firstly, I'm sorry to hear that you had previously thought about ending your life. If you ever feel this way again, please talk to your family and friends and make sure you are well looked after. If you feel too self-conscious to talk to people you know, contact a counsellor, or even hit me up with an email and I'd be happy to chat. Actual or perceived lack of ability or success in academic work should never be a life-ending event.*
---
Setting aside the issue of supervisory abuse, and concentrating solely on your feeling of being an imposter, there are a few things I notice about your description of your candidature experience:
* Your supervisor is a brilliant researcher;
* During your candidature, your supervisor pushed you hard and was not shy about pointing out your deficiencies and the magnitude of these deficiencies;
* You were highly diligent and willing to do additional work, beyond what would normally be expected;
* Like all other PhD students you were a novice research with areas of strengths and weaknesses (weakness on technical skills; strength on diligence), including some holes in your knowledge and ability;
* By the end of your candidature, all relevant referees have a high opinion of you, and your supervisor is also taking action to ensure that others have a good image of you.
All of this evidence points to a pretty common situation in academia, which is that an incoming PhD student is weak and deficient as a researcher during the early part of their candidature, but then they work hard to get better and they become competent (often without perceiving how much more skilled they have become). Consequently, their relationship with their supervisor goes from one involving heavy criticism of their work, to a gradual easing of that, to a later relationship involving praise for their skills and abilities. While the particulars of your case may involve poor treatment (which I wouldn't presume to know about), the above aspects of your experience suggest the standard situation in which you were bad at things but then got a lot better.
What you describe in your post does indeed sound like "imposter syndrome" to me, since your own view of your work conflicts sharply with the assessments of the referees that have observed your work.\*\* Two of the primary things that people with imposter syndrome forget is (1) that *you get better at things* during periods in which you work on becoming better at them; and (2) that there are expectations for competence that differ at every stage of our training and careers. The fact that you were previously told off for being deficient at a thing does not mean that you are permanently deficient at that thing, or deficient in general. Similarly, the fact that you are not as good a researcher as your brilliant and experienced supervisor does not mean that you are deficient relative to the career stage you are at. Your description says that several years after your previous PhD candidature you are now getting positive evaluations of your work, both from your supervisor and other referees. Assuming that those people are competent to judge your work, this suggests that you are at the level of competency you need to be at for your career stage, and if anything, you are probably at the top end of competency in that cohort. (Bear in mind that academic spots at prestigious universities are insanely competitive; to be considered you probably need to be in the top 1-2% of PhD graduates, so if you're even in the ball-park then you are probably somewhere at the top end of your cohort.)
Finally, I notice that you have said that you know deep down that there is truth to your supervisors comments. His negative comments on your skills might well have been true at the time, but now you've gotten better at research. If what your supervisor says is indeed true, then presumably this means he was correct in pointing out your research deficiencies during your training and he is *also correct* in now having such a high opinion of you that he is taking action to arrange strong letters of recommendation for a prestigious academic position. In other words, why believe your supervisor speaks the truth only when he says bad things but not later when he says (or thinks) good things?
---
\*\* And I am one of the people on this site who is a huge sceptic of the frivolous overuse of the term "imposter syndrome" to describe anyone who has been told they are bad at a thing.
Upvotes: 1
|
2023/02/11
| 1,094
| 4,714
|
<issue_start>username_0: I am a BSc student of Mathematics in Austria. I've been studying more advanced courses on the side (diff geo) to further my knowledge. I would like to attend contact/office hours of some professors to ask them questions about math and get to know them.
Is it normal in Europe to go to office hours weekly and talk mathematics in their area of research? (of course being prepared with good questions and things I've tried to solve them)
Do professors like being visited or view it as extra work?
Is there a difference between the EU and US office-hour-culture?
I appreciate any personal experience as well.<issue_comment>username_1: In the UK office hours are normally reserved for students currently enrolled at the University. Often they are restricted to students attending courses of this professor. Sometimes separate office hours are allocated for students studying at a specific course (aka module) of this professor.
Meeting general public at office hours seems a reasonable idea, but for some reason I never heard of it while in the UK.
Upvotes: 4 <issue_comment>username_2: I work in the UK but have studied in Italy.
When I was a student, I'd shoot an email and ask for an appointment to talk about XYZ. At worst, I was told that it was not possible, but it happened rarely.
Now that I'm sitting at the other side of the desk, I'd only expect students attending my courses to show up during office hours, but I'm also happy to see those who would like to talk more generally about academic topics relevant in my field (economics). Office hours do take quite a bit of my time, so I never complain if only a few students show up (or none at all). When overcrowded, I give priority to those students who informed me in advance and/or have more relevant questions about the courses I'm currently teaching.
My own strategy is to publish my office hours on the course webpage (only visible to enrolled students), setting different times and/or days for different courses (when feasible with other duties). For everyone else, I ask them to request an appointment by email. So, were you coming to see me, you'd probably fall in this last category and I'd be happy to see you as time allows.
My colleagues have similar arrangements.
I've had a few students who knocked on my door to ask *"What's up with economics these days?"* without showing that they did some independent Googling beforehand. While I'd politely answer, I don't find that particularly enjoyable. I'd much rather see and talk to students who show up to say *"I've seen [this] and [that] are on these days, but I think [this different but related thing] is more interesting, what do you think?"* or *"I'm interested in [this problem] but I'm having troubles {starting, finding literature, thinking formally, understanding this paper}."*
In conclusion, I'd say that office-hour culture in Europe is indeed different than the American one, although I've never experienced the latter myself and I've only heard from friends about it. But if you wrote to me with a set of clear questions in mind (they need not be formally clear, they only should be motivated by some goal), I'd be more than happy to schedule an appointment. I'd expect colleagues in any other discipline to do the same (but YMMV).
Upvotes: 3 <issue_comment>username_3: There are positive and negative things to say, and it will hugely depend on what courses somebody teaches, and how many students from these courses use the office hours and for what reason.
It is a pleasure to talk to curious students who have a proper interest in the field, so these are in principle always welcome. However it is in fact working time, and I can use office hour time for something else in case there is no student. As there are always too many things to do and never enough time, we won't complain for sure if nobody turns up to office hours either (I work in Italy and for the courses I'm currently teaching mostly there aren't many students who turn up, so in fact it's better if students let me know in advance that they will come because then I have it on my radar and won't forget the office hour because nobody turned up three weeks in a row).
Or on the other hand there may be so many students that more time is needed than reserved, and this is of course hard to manage as we all have lots of other stuff to do.
So by all means use the office hours, even if it's not about material specifically covered in the course taught by the same lecturer; however have a look at how much traffic there is and try to not overdo it (if every student came once a week, it would be impossible to handle unless course sizes are tiny).
Upvotes: 5 [selected_answer]
|
2023/02/11
| 2,594
| 10,624
|
<issue_start>username_0: My course has 12-15 pop quizzes scattered throughout the semester. The quizzes are unannounced (thus, *pop*). I also have some at the beginning of the class, some at the end, and others in the middle. Most are 10-15 minutes in length.
One of my students has a university-approved accommodation for 50% extra time. I'm struggling with how to implement this accommodation. To be clear -- I'm very much on-board with providing accommodations. I am **not** trying to dodge the requirement.
I don't feel it would be helpful to single out the student in any way. Therefore, I don't want to collect the quiz papers and then have the student leave the room to get the extra 5 or 6 minutes of time. And, they would have to leave the room, as the distraction of the lecture would make the extra time fairly useless. Further, that student would then also miss the next bit of lecture.
Because the quizzes are random, I don't want to notify the student ahead of time to, for instance, "swing by my office to take the quiz ahead of time" as word will get out about the existence of a quiz that day. Ditto for "show up at class early" for a quiz that starts at the beginning of class.
I suppose that a quiz at the end of class could be handled with the extra time handled immediately after class. But, what do I do about the other quizzes?
Has anyone else faced a similar problem? Any suggestions?<issue_comment>username_1: I give pop quizzes. They are just as you've described.
To resolve the logistics nightmare, perhaps, one takes a step back.
What is the purpose of pop quizzes, and what do they mean to you?
* attendance
* continuous assessment
* knowledge ringfencing.
In my case, from a pedagogy viewpoint, they, PQ, are part of the course scaffolding.
Thus, they are formative assessments, yet they go further.
The implication of this are:
1. PQ assists the students in reinforcing learning during the class discourse or refreshing or 'rejuggling' knowledge from the previous class session.
2. PQ are feedback snapshots (to me as the lecturer) of what the students get or are getting from the class.
3. PQ are sometimes 'just-for-fun' and, in those instances, merely count for 'attendance'.
For me, typically students get marks for the best half or 2/3 of their PQ for the term/semester. Invariably, I take the best half or 2/3, counting for marks towards their final assessment. This is clearly written in their study guide. It is also discussed during our 'contract' class: typically, the first class of the term/semester where we go over the study guide.
If the *disability* is adamant that you must allocate extra time for the student?
* factually write down the logistic 'nightmare' as you've articulated here;
* also, document for the disability unit, your 'best of half or 2/3' marks of PQ
* allow the student to only attempt half or 2/3 of the PQ questions, except when the PQ is only one Q!
These should fly with the *disability unit* and with the said student as well. However, there are no guarantees as some might rigidly follow the 'text' and not the rationale.
While at it, keep track of the trend. Do your learning analytics. The *patterns* might come in handy later, or just contribute to your reflective teaching.
Upvotes: 2 <issue_comment>username_2: The other answer here by [username_1](https://academia.stackexchange.com/users/162770/) gives a good articulation of some pedagogical advantages to pop quizzes, so I can see why you might want to continue these. Here are two things you should do:
* Rather than trying to alter the logistics of the pop quiz, you could try to find an "academically equivalent" accommodation that does not involve extra time. For example, you might proceed with a system where the student is only required to do 2/3 of the questions (you would need to give an instruction prior to the pop quiz as to which questions apply). Alternatively, you might proceed as normal with the pop quiz but give some other marking accommodation to the student (e.g., quizzes are redeemable against some other work). Remember that you could also speak to the student to see if they have any ideas or preference about how to proceed. There are many options you could consider that would not involve giving extra time but would involve some "equivalent" accommodation.
* Regardless of what you come up with, go and talk to the Disability Unit at your university that gave this accommodation and ask them to assist with a variation of their present accommodation that meets the logistical requirements of your pop quizzes. Their job is to assist students with diability accommodations but also to fit within the structure and logistics of the courses determined by the lecturer, so feel free to put the onus on them to find a solution *within the constraints of your preferred assessment structure*. The Diability Unit will look at the course accommodations holistically, so if there is no sensible accommodation in the pop quizzes, that does not necessarily matter and could potentially be dealt with using some other accommodation.
Whatever solution you come up with, it would be best for this method to be formalised into the disability accommodation granted to the student, which means it will need a variation that specifies the accommodation (or lack thereof) for the pop quizzes.
Upvotes: 0 <issue_comment>username_3: As other answers have said, consult with your disability office regarding regulations. I am writing separately to note that I do not agree with the sentiment that it is likely to be okay to allow the student to do only a fraction of the questions, or to grade only their best questions. My suspicion is that you will not be able to make true pop quizzes work and that, while you can make frequent quizzes work, you will need to change your course infrastructure around them.
For example, here are the [policies at my department](https://lsa.umich.edu/content/dam/math-assets/math-document/Policies/Mathematics_Department_Policy_on_Testing_Accommodations%20(2).pdf), which I believe are very similar to the policies throughout the University of Michigan. Here is the key part:
>
> **Quizzes.** The logistics for providing testing accommodation on an in-class quiz can be challenging. Please work with the student to find a reasonable accommodation. Here are some examples of approaches that have worked in the past. In all of these examples, suppose your class has exactly one student who requires testing accommodations, and that that student’s VISA [Verified Individualized Services and Accommodations] form indicates that the student is to be given time-and-a-half on all timed assessments.
>
>
> 1. Design a twenty-minute quiz that will be given first thing in class. Have the extended-time student work on the quiz for ten minutes prior to class in your office and then complete the quiz with the rest of the class.
> 2. Design a twenty-minute quiz that will be given last thing in class. Have the extended-time student take the quiz in class and then work on the quiz for ten minutes in your office after class is over.
> 3. Have the extended-time student complete the entire quiz in office hours and work quietly on something else during the in-class quiz.
> 4. Have the extended-time student take the quiz in the TAC [Testing Accommodations Center]. Time things so the student will arrive at class just as the other students complete their quizzes.
>
>
> Here are some examples of what is not acceptable (i.e., if you do any of them, you will likely be engaging in discrimination that violates federal law). In all of these examples, suppose your class has exactly one student who requires testing accommodations, and that that student’s VISA form indicates that the student is to be given time-and-a-half on all timed assessments.
>
>
> 1. Design a twenty-minute quiz, and then give everyone in the class thirty minutes to complete it.
> 2. Design a quiz with 6 questions, but ask the time-and-a-half student to complete only 4 of the 6 questions.
> 3. Design a twenty-minute quiz, give everyone in the class 20 minutes to complete it, and then adjust the final score on the extended-time student’s quiz by a factor of 1.5.
>
>
>
Upvotes: 3 <issue_comment>username_4: There is a lot here that is unknown, so this is a bit speculative, even orthogonal. Perhaps some of the ideas can help.
* If this arises only seldom, consider not using pop quizzes at all for the term. The problem goes away and you also get the chance to analyze whether your normal practice has value.
* Consider giving the quizzes as normal with no time adjustments, but not including them in the grading. Give feedback on the answers to each student to help guide them in the proper direction. You can give "exceptional" students "exceptional feedback", where "exceptional" can be interpreted broadly, not just for disability.
* Have a three way conversation with yourself, the student, and the disability office to get everyone's input. It might be important to the student not to be obviously singled out to their peers, or not. This helps you avoid making assumptions that might not be valid.
* If you need quizzes at all, move them to the end of the class (but see the note above). This lets you adjust the time as stated in the policy with minimal disruption. It does, however, assume that you and the student are available for a few extra minutes.
* Orthogonally, instead of pop quizzes, work with the students to take better notes. In particular, they can use note cards during the lecture to try to capture the, say, three most important ideas in the class. Spend a few minutes at the end of each class asking for student comments (from the note cards) on what those important ideas are. Those same note cards can be used at the start of the next class rather than the end of the current one in the same way and form a basis for later study by the student.
* In the case that pop quizzes are used as a goad to get them to do homework and study regularly, collect note cards at the start of each class about their experience since the previous class. What they learned from study, what issues they have. This takes time out of the equation. One particular item might be "What is the most important question you have at this moment about the course material?". Note that I haven't used this myself and would need to think about it more to refine it. But those note cards are then the "ticket to play" for the student. You can use them to guide the lecture, after scanning a few, and also to give individual feedback.
Upvotes: 1
|
2023/02/12
| 1,384
| 5,626
|
<issue_start>username_0: Many entries in the [DOI catalogue](https://www.doi.org) contain incorrect information. Is there any central way to report such entries (to the DOI team or the publisher\*)?
Example:
When importing a work into a [reference management software](https://en.wikipedia.org/wiki/Reference_management_software) by its DOI this results in:
```
title={This Is An Example},
journal={Big Journal},
author={<NAME>.},
year={2000}
```
But correct would be:
```
title={This Is An Example},
journal={Big Journal},
author={<NAME>. & <NAME>.},
year={2000}
```
\* I am aware of the possibility to write each publisher an email. But I am looking for a faster solution<issue_comment>username_1: Assuming the metadata retrieval method you're using uses [CrossRef](https://en.wikipedia.org/wiki/Crossref) behind the scenes you can contact them. This is the case, for example, when adding references by DOI in [JabRef](https://docs.jabref.org/collect/add-entry-using-an-id) and [Zotero](https://www.zotero.org/support/adding_items_to_zotero#add_item_by_identifier).
(If the method doesn't use CrossRef's services I suppose you can still verify that the problem is present in their metadata before contacting them.) According to the [FAQ](https://www.crossref.org/faqs/#q6general):
>
> **I’ve found a problem within your metadata - how do I get you to fix it?**
>
>
> While we aren’t able to correct the metadata provided by our members, report any metadata issues to our [support staff](https://support.crossref.org/hc/en-us/requests/new?ticket_form_id=360001642691) and we’ll contact the responsible member and ask them to make corrections.
>
>
>
Upvotes: 2 <issue_comment>username_2: I can understand your frustration at trying to create an accurate bibliography, only to find that the data you believed would be correct actually turns out not to be! To answer your question, however, requires an understanding of how, and where, your problem might have arisen.
The first thing to understand that there is no *DOI catalogue*. As Wikipedia explains, the [Digital Object Identifier](https://en.wikipedia.org/wiki/Digital_object_identifier) (DOI) is "a persistent identifier or handle used to uniquely identify various objects". It's primary purpose is to provide a persistent link to a digital "object" such as a journal article, despite the fact that the actual location of the journal article on a publisher's website might change. One consequence of this is that it becomes the responsibility of a publisher to inform the maintainers of the DOI when a link to which a DOI resolves (for example, when one browses to
`https://doi.org/nn.ddddd/thisisthedoilabelgivenbythepublisher`) is changed.
Moreover, when a publisher first publishes an article and notifies the DOI maintainers, the primary thing that they are doing is telling the DOI maintainers the [Uniform Resource Locator](https://en.wikipedia.org/wiki/URL) (URL) to which the DOI should resolve.
The question that underpins your problem is not really about a non-existent "DOI catalogue" (which at best might be conceived of as being the database that links a DOI to a URL) but instead, a question about how, and from where, your [reference management software](https://en.wikipedia.org/wiki/Reference_management_software) (which I'll call RMS for conveniece) is acquiring the metadata about an article, and how errors might be introduced in the process.
There are several possibilities, so much more likely than others. Here I simply list some of the possibilities with different possible variations:
* **Scenario 1**: When you enter a DOI into your RMS it uses <https://doi.org> to resolve the DOI to a publisher's URL but the resolution is incorrect and the reference management software ends up on a page for the wrong article. In my experience, this is just about the least likely of the various possible errors. Nonetheless, if it occurs, it is almost certainly the publisher who has made an error by incorrectly notifying the DOI maintainers of the correct URL and so it is the publisher whom you would have to contact to correct your problem.
* **Scenario 2.** When you enter a DOI into your RMS it uses <https://doi.org> to correctly resolve the DOI to a publisher's URL where the journal article appears. Several things can now go wrong.
+ **2a:** The publisher has published incorrect metadata on the page (e.g., omitting an author's name or incorrectly stating the page numbers of the article). In this case, you would have to contact the publisher.
+ **2b:** The publisher has published the correct metadata and you try to import the bibliographic information directly from the correct web-page but your RMS does not scrape the information from the page correctly. In this case, you would need to contact the maintainers of your RMS.
+ **2c:** You choose not to have your RMS scrape information directly from the web-page but instead choose to export the bibliographic information from the page using one or other available export formats such as \*.ris . The export process might either go correctly, and export the correct data, or fail and export incorrect data. The only way of finding out which has occurred would be to examine the exported metadata file using a text editor. If the exported data is incorrect, contact the publisher. If the exported data is correct, then the error will have occurred at the final stage of importing the metadata file into your RMS. The problem should be reported to your RMS producer.
There are, of course, yet more possibilities, but these are the most likely ones.
Upvotes: 4
|
2023/02/13
| 2,673
| 11,056
|
<issue_start>username_0: Sometimes I see articles, even high-quality ones, use illustrations (e.g. imagine a simple Powerpoint depiction of a certain process).
I'm now writing my thesis and found some interesting illustrations that I would like to use in the *Definitions* section of my thesis.
However, I'm not quite sure if that would turn out to be beneficial:
* pro: It would break up the wall of text a bit and provide a more intuitive way of grasping a definition.
* con: My supervisor (who probably would not bother to read the whole section anyway) glances over it and doesn’t like illustrations and marks me down.
I don't really want to ask my supervisor about this, since he is a highly decorated senior professor and head of faculty. I'd rather use my “capital” wisely on methodological questions.
What is your view on that?<issue_comment>username_1: If it is like putting a picture of a film star next to an article about them, then you should not do it.
If they are merely "interesting illustrations" to "break up the wall of text", then you should probably not use them.
If they are positively useful, show things that cannot be shown in text, or make something much clearer than text, then you should use them.
(Edited to add: As others have said, if you want to use an illustration from someone else's publication, you need to think about copyright. And in many cases, it is better to make your own illustrations.)
Upvotes: 7 [selected_answer]<issue_comment>username_2: I must push back on the frame of your question: I think you should ask this question of your advisor for the following reasons:
First, learning to write an academic paper or thesis is part of what you are there to learn as a Master's candidate, and part of what your advisor is there to teach you.
Second, as you yourself point out, your advisor is the person (or one of the few people) to be evaluating your work. It is fine to cast a wide net for advice when developing your overall style as an academic writer. But it would be foolish to take our advice over your advisor's advice when you are writing a paper for the narrow audience of your advisor and/or your committee.
That said: The valid use of drawings, figures, diagrams, plots, illustrations, etc is to convey information efficiently and effectively. They can also be beautiful in their own right; placed strategically to break up text, etc. But if they are not there to convey information, they should probably not be there.
Upvotes: 5 <issue_comment>username_3: >
> he is a highly decorated senior professor and head of faculty
>
>
>
Highly decorated senior professors and heads of faculty are not demi-gods.
They are human like everyone else and can be great people or assholes. This is how you should assess them and act accordingly. Since he is your advisor he has a duty to help you (he does not need a PhD candidate like someone just starting).
>
> I'm now writing my thesis and found some interesting illustrations
>
>
>
Since you wrote an introduction, its readers will benefit from a crisp, helpful picture (which as Napoleon put it, is better than 100 PowerPoint bullets).
Upvotes: 4 <issue_comment>username_4: The other answers make good points, but some textbooks also have (a few) pictures that are not strictly related to the text, often at the beginning of chapters. If done with style, this can be fine.
But you are not a textbook author and your thesis is not a textbook. It is also your first major article, so you may not have the judgement to know what is appropriate and what is not. Also, a thesis is a different type of document.
Nevertheless, you might consider adding a personal touch, for example in the dedication or perhaps even in the introduction. But if you are really worried that your supervisor might not like it, you should probably leave it out. Even if they don't have a problem with it, you run the risk of always thinking about whether it might have been a problem and contributed to the grade and other doubts.
Upvotes: 2 <issue_comment>username_5: Textbooks often have pictures. For example Munkres' Topology:
[](https://i.stack.imgur.com/dJPy9.png)
[](https://i.stack.imgur.com/tkKhg.png)
[](https://i.stack.imgur.com/fVSq3.png)
Consult textbooks and papers related to your masters topic for what a "professional" illustration looks like.
Upvotes: 5 <issue_comment>username_6: I would say a thesis is **making an argument, with evidence**. The question is: **do the illustrations help to build the argument?**
If they are decoration, like in a magazine, clearly not. If the text can be fully understood without them, like in an online news story which the publication requires to have an illustration, even if it's a stock photo, then the photo isn't adding anything.
Your reader does not need images to help them read the piece, because you can assume they are able to read at a good standard, and your piece does not need to sell itself on a newsstand, bookstore or online clickbait.
But if the argument is aided by the image, then maybe there is a case to include them. A diagram or graph presents information in a different way to text, and that would be absolutely fine IMHO. A picture of an experimental setup or historical source material may help the reader understand the context better. A cartoon could make a point in an ironic way that the text could not do. **Think about what the purpose of each image is, and what you are saying with it.**
In general, though, I would be sparing in using images. The reader is there to read, not to view a photo album. It is hard to make a good argument with primarily images ('eye of the beholder' and all that). Also, you may run into copyright and plagiarism issues if too much of the content is not your own.
Obviously there are some fields which are very graphical - eg art or fashion - where your source material is all image-based and you have to use that material. For those different expectations might apply.
Upvotes: 3 <issue_comment>username_7: There's a difference here that I think needs to be drawn between "illustrations" and "diagrams".
Diagrams are technical drawings that add useful information to your thesis that is better conveyed visually than textually. I would also include charts in this category. Include as many of these as you need, they're a great way to effectively convey information.
Illustrations, however, tend to be less textual and more descriptive. So, you might show an illustration of what a design might look like in context. Now, whether you include these or not is largely dependent on what your thesis is. If you're doing a design-based course like Architecture or Fashion, then sure, include illustrations. Other than that, I probably wouldn't, except perhaps as an appendix if you really want to.
The key difference here is whether you're actually adding information to the paper or merely illustrating it in a pretty fashion.
---
That being said, it's your thesis. You're not submitting it for publication to a journal so sans explicit instructions from the university to the contrary, you're free to make stylistic decisions to suit yourself. I, for one, chose to break my dissertation into chapters and put amusing, albeit relevant, epigraphs at the start of each. Not very "professional", I'll admit, but it made the whole thing a bit nicer to read and I liked them. No reasonable professor is going to care about a stylistic choice here or there, they're only going to care about the content. Within reason of course.
Upvotes: 2 <issue_comment>username_8: There is something in OP which I don't think the existing answers address.
>
> I'm now writing my thesis and found some interesting illustrations that I would like to use [...]
>
>
>
The other answers talk about the benefits of having figures, etc., in your thesis, but these should be figures that you create yourself. There are several drawbacks to adding illustrations that you merely found.
* You will probably need to get permission to do this (unless the image comes with a CC or similarly permissive licence).
* You will certainly need to attribute the image (even if the original creator does not require this, the presumption is that anything in your thesis is your own work unless specified not to be, and you must not appear to take credit).
* Since they are not your own work, they will not get credit and so can't really benefit you - but they could still harm you if your supervisor/examiner thinks they are inappropriate.
So for these reasons, I would not use them. But if you can create your own (and I mean create, not just redraw) images that are helpful to the reader, then by all means do so.
Upvotes: 3 <issue_comment>username_9: I think that some of the previous answers have tried to be fair and still protective of you.
But maybe they miss the point of why you are applying the illustrations - and this is no surprise as all you say is that they are in the *Definitions* section of the thesis.
As I read it, you may be introducing illustrations to make it easier for the *general* reader in your subject - as opposed to those actively researching in a small field within it - to quickly "get" the basic concepts that you are introducing at the start of the thesis. And there is no doubt about it but a picture *is* worth a thousand words in getting many concepts. In physical sciences, schematic diagrams are often used to put across processes that are in reality much more detailed and complex though the latter aspects are simply to make the process more effective.
I don't see much wrong in this exercise as long as your diagrams are expressly labelled as schematics and the added complexities are referred to at least in the accompanying text.
But you really must discuss your purpose with these illustrations with your supervisor so that he understands why you are putting these in your *Definitions* section. Noting that the attention span of supervisors is quite short, you would be well advised to have a short and simple phrasing prepared before you present your reason for these graphics.
Upvotes: 1 <issue_comment>username_10: Depends from your filed of study. Literature (for example) do not use many pictures. Physics (again, for example as field) believe that "a picture is value 1000 words".
My personal opinion: pictures that explains the concepts/ideas/tests/methodology(etc.) are very welcome as are capable to say something that may be difficult to express in words. However every picture that you add in your manuscript should be properly described.
This is what I recommend to my students when they write their thesis with me:
<https://francescolelli.info/thesis/simple-writing-rules-that-can-improve-the-quality-of-your-thesis/>
The second bullet talks about how to describe pictures and tables.
Upvotes: 0
|
2023/02/13
| 2,881
| 11,943
|
<issue_start>username_0: So, I am a physics student at UC Berkeley. However, at one point, just due to its utility and my own passion for the subject, I wanted to double major in math. Around a year ago, I had very bad personal circumstances which led to notably poor marks in two of the three courses I took that semester, namely two C+ marks. One of these was in an upper division math course, which, today I feel good about in terms of how I understand it, given that I use it almost daily. Due to this fact, I wonder if I should retake this course for the sake of graduate school applications. This question also extends to other courses I haven’t performed that well in.
As of now, my cumulative GPA is a 3.0 (have been receiving B- or B in most technical courses, unfortunately), which almost says no to any graduate school (especially because at the moment, research experience simply is not enough to compensate). This is largely a result of the fact that I have immense exam stress, rather than the fact that I don’t understand my courses, but of course graduate schools would not care about this or see this just by seeing my GPA. I have calculated that if I were to receive the A mark in the remainder of my courses up until graduation, I could end with a 3.7. However, that could be higher if a retook a course or two, one of which is that upper division math course. But would graduate admission officers really consider a low mark in a course that isn’t directly in my field of study, is something I’ve wondered.
I know GPA is not all that matters in these applications, but as I understand it, it is a big deal. In any case, I plan to take the GRE (at least in my subject).
Any advice is greatly appreciated.<issue_comment>username_1: If it is like putting a picture of a film star next to an article about them, then you should not do it.
If they are merely "interesting illustrations" to "break up the wall of text", then you should probably not use them.
If they are positively useful, show things that cannot be shown in text, or make something much clearer than text, then you should use them.
(Edited to add: As others have said, if you want to use an illustration from someone else's publication, you need to think about copyright. And in many cases, it is better to make your own illustrations.)
Upvotes: 7 [selected_answer]<issue_comment>username_2: I must push back on the frame of your question: I think you should ask this question of your advisor for the following reasons:
First, learning to write an academic paper or thesis is part of what you are there to learn as a Master's candidate, and part of what your advisor is there to teach you.
Second, as you yourself point out, your advisor is the person (or one of the few people) to be evaluating your work. It is fine to cast a wide net for advice when developing your overall style as an academic writer. But it would be foolish to take our advice over your advisor's advice when you are writing a paper for the narrow audience of your advisor and/or your committee.
That said: The valid use of drawings, figures, diagrams, plots, illustrations, etc is to convey information efficiently and effectively. They can also be beautiful in their own right; placed strategically to break up text, etc. But if they are not there to convey information, they should probably not be there.
Upvotes: 5 <issue_comment>username_3: >
> he is a highly decorated senior professor and head of faculty
>
>
>
Highly decorated senior professors and heads of faculty are not demi-gods.
They are human like everyone else and can be great people or assholes. This is how you should assess them and act accordingly. Since he is your advisor he has a duty to help you (he does not need a PhD candidate like someone just starting).
>
> I'm now writing my thesis and found some interesting illustrations
>
>
>
Since you wrote an introduction, its readers will benefit from a crisp, helpful picture (which as Napoleon put it, is better than 100 PowerPoint bullets).
Upvotes: 4 <issue_comment>username_4: The other answers make good points, but some textbooks also have (a few) pictures that are not strictly related to the text, often at the beginning of chapters. If done with style, this can be fine.
But you are not a textbook author and your thesis is not a textbook. It is also your first major article, so you may not have the judgement to know what is appropriate and what is not. Also, a thesis is a different type of document.
Nevertheless, you might consider adding a personal touch, for example in the dedication or perhaps even in the introduction. But if you are really worried that your supervisor might not like it, you should probably leave it out. Even if they don't have a problem with it, you run the risk of always thinking about whether it might have been a problem and contributed to the grade and other doubts.
Upvotes: 2 <issue_comment>username_5: Textbooks often have pictures. For example Munkres' Topology:
[](https://i.stack.imgur.com/dJPy9.png)
[](https://i.stack.imgur.com/tkKhg.png)
[](https://i.stack.imgur.com/fVSq3.png)
Consult textbooks and papers related to your masters topic for what a "professional" illustration looks like.
Upvotes: 5 <issue_comment>username_6: I would say a thesis is **making an argument, with evidence**. The question is: **do the illustrations help to build the argument?**
If they are decoration, like in a magazine, clearly not. If the text can be fully understood without them, like in an online news story which the publication requires to have an illustration, even if it's a stock photo, then the photo isn't adding anything.
Your reader does not need images to help them read the piece, because you can assume they are able to read at a good standard, and your piece does not need to sell itself on a newsstand, bookstore or online clickbait.
But if the argument is aided by the image, then maybe there is a case to include them. A diagram or graph presents information in a different way to text, and that would be absolutely fine IMHO. A picture of an experimental setup or historical source material may help the reader understand the context better. A cartoon could make a point in an ironic way that the text could not do. **Think about what the purpose of each image is, and what you are saying with it.**
In general, though, I would be sparing in using images. The reader is there to read, not to view a photo album. It is hard to make a good argument with primarily images ('eye of the beholder' and all that). Also, you may run into copyright and plagiarism issues if too much of the content is not your own.
Obviously there are some fields which are very graphical - eg art or fashion - where your source material is all image-based and you have to use that material. For those different expectations might apply.
Upvotes: 3 <issue_comment>username_7: There's a difference here that I think needs to be drawn between "illustrations" and "diagrams".
Diagrams are technical drawings that add useful information to your thesis that is better conveyed visually than textually. I would also include charts in this category. Include as many of these as you need, they're a great way to effectively convey information.
Illustrations, however, tend to be less textual and more descriptive. So, you might show an illustration of what a design might look like in context. Now, whether you include these or not is largely dependent on what your thesis is. If you're doing a design-based course like Architecture or Fashion, then sure, include illustrations. Other than that, I probably wouldn't, except perhaps as an appendix if you really want to.
The key difference here is whether you're actually adding information to the paper or merely illustrating it in a pretty fashion.
---
That being said, it's your thesis. You're not submitting it for publication to a journal so sans explicit instructions from the university to the contrary, you're free to make stylistic decisions to suit yourself. I, for one, chose to break my dissertation into chapters and put amusing, albeit relevant, epigraphs at the start of each. Not very "professional", I'll admit, but it made the whole thing a bit nicer to read and I liked them. No reasonable professor is going to care about a stylistic choice here or there, they're only going to care about the content. Within reason of course.
Upvotes: 2 <issue_comment>username_8: There is something in OP which I don't think the existing answers address.
>
> I'm now writing my thesis and found some interesting illustrations that I would like to use [...]
>
>
>
The other answers talk about the benefits of having figures, etc., in your thesis, but these should be figures that you create yourself. There are several drawbacks to adding illustrations that you merely found.
* You will probably need to get permission to do this (unless the image comes with a CC or similarly permissive licence).
* You will certainly need to attribute the image (even if the original creator does not require this, the presumption is that anything in your thesis is your own work unless specified not to be, and you must not appear to take credit).
* Since they are not your own work, they will not get credit and so can't really benefit you - but they could still harm you if your supervisor/examiner thinks they are inappropriate.
So for these reasons, I would not use them. But if you can create your own (and I mean create, not just redraw) images that are helpful to the reader, then by all means do so.
Upvotes: 3 <issue_comment>username_9: I think that some of the previous answers have tried to be fair and still protective of you.
But maybe they miss the point of why you are applying the illustrations - and this is no surprise as all you say is that they are in the *Definitions* section of the thesis.
As I read it, you may be introducing illustrations to make it easier for the *general* reader in your subject - as opposed to those actively researching in a small field within it - to quickly "get" the basic concepts that you are introducing at the start of the thesis. And there is no doubt about it but a picture *is* worth a thousand words in getting many concepts. In physical sciences, schematic diagrams are often used to put across processes that are in reality much more detailed and complex though the latter aspects are simply to make the process more effective.
I don't see much wrong in this exercise as long as your diagrams are expressly labelled as schematics and the added complexities are referred to at least in the accompanying text.
But you really must discuss your purpose with these illustrations with your supervisor so that he understands why you are putting these in your *Definitions* section. Noting that the attention span of supervisors is quite short, you would be well advised to have a short and simple phrasing prepared before you present your reason for these graphics.
Upvotes: 1 <issue_comment>username_10: Depends from your filed of study. Literature (for example) do not use many pictures. Physics (again, for example as field) believe that "a picture is value 1000 words".
My personal opinion: pictures that explains the concepts/ideas/tests/methodology(etc.) are very welcome as are capable to say something that may be difficult to express in words. However every picture that you add in your manuscript should be properly described.
This is what I recommend to my students when they write their thesis with me:
<https://francescolelli.info/thesis/simple-writing-rules-that-can-improve-the-quality-of-your-thesis/>
The second bullet talks about how to describe pictures and tables.
Upvotes: 0
|
2023/02/13
| 873
| 3,352
|
<issue_start>username_0: I was a reviewer of an article that was finally published some time ago in a high impact journal.
I plan to apply to an open competition for a teaching assistant position in a subject very related to the topic of the paper. Extremely high qualifications are not normally required for these positions, but every point counts.
To apply I must send my CV to a university. I will mention the review in my CV. Is it okay in this case to indicate the topic or the name of the paper?
My motivation for doing so is that the recognition received from being contacted by a high-impact journal to review a paper on one of the main topics of the course (of course, the paper goes much deeper) should have some value for a teaching assistant position. Although the subject is widely used as a practical tool, there are very few works of theoretical analysis on it.
NOTE: I couldn't find legal details on it in the ReviewerHub website.<issue_comment>username_1: It is common for people to state on their CVs *which journals* they reviewed for, but I would generally advise against explicitly disclosing the paper you reviewed. This is particularly the case for junior folks -- you can get the positives you are looking for by just providing the journal name.
Upvotes: 5 <issue_comment>username_2: ### Ethics, Norm
Except where it is explicitly stated, it is not so much of a *legal* restriction. However, you are bound by #ethics.
Ethics also covers what the norm is. A key consideration is whether the journal's review process is:
* double-blinded review
* single-blinded review; or
* open peer review (transparent review)
Except for open peer review, as a reviewer, you are bound by confidentiality. You cannot disclose your review outside of the review process.
### ReviewerHub
However, what you can do, since you indicated reviewerhub,
1. Go to your [reviewer hub profile](https://reviewerhub.elsevier.com/reviewer/reviews/history). You'll have the list of manuscript titles and the date you completed your review; alternatively
2. You can get a report for a specific manuscript in your [ReviwerHub Report](https://reviewerhub.elsevier.com/reviewer/rewards-reports/reports)
The first page will have an '*Elsevier Reviewer Recognition*' stamp and the journal's name. Subsequent pages will include the first page details and the list of the manuscript's title, the review's completion date, and the review's revision cycle.
better still
3. Within your [ReviewerHub Rewards and Report](https://reviewerhub.elsevier.com/reviewer/rewards-reports/certificates), download a copy of your 'recognised Reviewer Certificate' for the specific journal.
That should suffice.
You do not need to include the contents of your review: and **you should not**.
**Open-peer review (Transparent review):**
Some journals allow you to include your review activities in [Publon (now part of WoS)](https://publons.com/wos-op/). If the journal subscribes to [open review (transparent review)](https://doi.org/10.12688/f1000research.11369.2), you are at liberty to point to your review.
Read more on [open (transparent) peer review](https://doi.org/10.3138/jsp.44-4-001).
Some journals like [BMJ](https://blogs.bmj.com/bmjopen/2021/09/21/bmj-open-peer-review/), [F1000 Research](https://f1000research.com/) follow open-peer review.
Upvotes: 4
|
2023/02/13
| 2,574
| 10,865
|
<issue_start>username_0: It has recently been proposed that our college's math department move away from giving remote, online exams to our online students, and begin mandating that all students take (at least) an in-person final exam. This was the format for *years* before the pandemic, so this would be a return to something all\* faculty felt good about. The proposal is meant to address widespread cheating via Chegg, etc. [Online proctoring software is a solution that has its [own](https://www.reddit.com/r/cheatonlineproctor/) [issues](https://www.npr.org/2022/08/25/1119337956/test-proctoring-room-scans-unconstitutional-cleveland-state-university), so I specifically want to discuss in-person, human-proctored exams.]
There has been some push-back from administration on this, citing that mandating in-person exams seems incongruous with the online course model. ["*A student signs up for an online course and then finds out they have to take an in-person test.*" However, the proposal includes advertising this to students before they ever sign up, and makes provisions for students to take the exam at an off-site proctoring facility as needed.]
I am wondering how common this practice is these days and if there is any agreement (e.g. general trend in higher ed) that online courses should have no in-person components (e.g. final exams). If your college has changed its tune on this (moving to give proctored exams to online students), I'd also appreciate hearing how that move affected students.<issue_comment>username_1: At the FernUniversität Hagen, the German (or rather NRW state) long distance university founded in 1974, where I worked 1980 - 1984, we had all examinations supervised. In rare occasions, (inmates), by persons of trust, but most course examinations were taken in person, usually in rooms rented from another university. We bundled several exams in order to save costs (e.g. beginning Math with second year Math) and had them supervised by a person competent to answer questions. Students needed to travel about 1-2 hours maximum by car. We (scientific employees) did not question this model because at this time, cheating on exams in Mathematics was common, to the extent that state examinations for future teachers were done under much stricter conditions.
Therefore, I agree with you that in person exams are helpful. Once students get it into their heads that cheating happens (whether that is true or not) they are much more likely to engage in cheating themselves because they do not want to be honest losers. They want to be honest in their great majority, but do not want to be severely disadvantaged by it. My impression at a US private university now is that students do not think that anti-cheating software is successful.
In some classes, it is certainly possible to restructure examinations so that they can be given as open-book, open-online, but if it is a class that is a standard undergraduate class, then paid help is much more possible.
We are having the same problem already with homework for a long time. If you have a large class, then having students explain their homework becomes impossible, especially since random samples would lead to accusations of picking on students.
I personally give online, open-book exams in my upper-division classes. Nobody interferes with this choice. If I give a coding project, I would include a clause that students need to defend their code. If they cannot explain their code then they get no points for the explanation, which implies that they get a failing grade on the code.
Upvotes: 5 <issue_comment>username_2: Let me give the purely cynical answer to this question.
If you require an in person final you will have fewer students. For a typical US institution (i.e. not a private non-profite with a huge endowment), if you have fewer students, your university will have less money, either directly in terms of tuition, or indirectly because government subsidies are in part based on your number of students. If you have less money, you will be able to hire fewer faculty and may even have to lay off some, or not replace them when they leave. Furthermore, having less money means your working conditions will be worse in various ways.
Requiring an in person final reduces students in two ways:
1. Some students will find it inconvenient to actually come to campus for the final. They will be able to get a degree from somewhere else that allows them to stay at home for their final; it makes sense for them to choose this option.
2. Because they do not have the option of cheating, more students will fail, and hence fewer students will continue to take further classes towards their degree. Note that most universities have no reputation to speak of, so unless cheating becomes widespread enough that it becomes widely known in society at large, the cheating itself has no impact on the reputation of the university or its ability to attract students.
Upvotes: 2 <issue_comment>username_3: Not a direct answer, but a better more viable solution to the underlying question is too simply write more thoughtful and original exams. They don't have to be completely original questions, but changing the wording and a few numbers in each question on an old exam will go a long way to prevent cheaters from Googling the answers. You could even distribute different versions of each exam with similarly worded, but different questions to prevent classmates from sharing answers. This is a better solution as it requires a minimal time commitment from both the professors and students and is less invasive than proctoring software.
Upvotes: -1 <issue_comment>username_4: From a student perspective, generally no. I have taken online classes since circa 2004, some with proctored in-person tests at outside universities, some remotely using proctoring software, and some where I traveled upwards of 90 minutes to take the in-person test at the university I was attending online.
The language used to describe the course goes a long way. Consider calling the class a hybrid class if in-person attendance is required for part of it. At least in the US, online options are fully online so you could attend from the other side of the planet if desired. Hybrid classes allow for flexibility on from where to attend class, but also come with the expectation of some in-person experiences. In-person classes are 100% in person (even if recorded). **A hybrid class clearly sets the in-person expectation up front** as opposed to burying it somewhere in the course description.
Asking a student to change their location for a single test adds barriers to their one-and-done chance to showcase their skills. It is not a fair assessment of a student's knowledge in that subject. The travel alone is enough to disrupt one's ability to focus and can add undue stress, plus taking a test in a new environment can increase test anxiety.
If in-person tests are a requirement for reducing cheating, consider strongly recommending students to show up in-person more than once or twice a term. This will encourage students to a) learn the commute, b) develop good patterns around commuting, and c) learn the environment in which they are being tested. Once every other week for in person discussions or group exercises or even just lecture will do the trick. Even students who plan to take the test at different in-person location will benefit from having made the trip a handful of times before having a test placed in front of them. Instead of asking students to change their patterns for what is often a high stress moment already, **encourage habits so that the high stress moment is about the stress you are measuring (their knowledge) and not the stress about the change in their patterns.**
As I mentioned, I have had to travel for these one-off in-person tests. I prefer them to putting proctoring (aka, spy) software on my personal computer. That said, I've had tests were I had to travel to the university the day before and get a hotel for a night because the test schedule didn't work with my schedule (work, access to the shared vehicle, accounting for commute traffic, etc.) even though it was a financial hardship. I've had tests were I arrived to find that the test room was very very cold and I did not have a jacket handy. I've had tests where the HVAC was broken, causing "whining" sounds in the room that we all just had to deal with. These issues were noticeably distracting, and could have been mitigated by knowing the environment (including travel) in which I was testing before starting the test.
Upvotes: 4 <issue_comment>username_5: When I was involved (which is many years ago, using a snail-mail format) we had different *subjects* in the *course* handled differently. So no student could complete the *qualification* without supervision, but could complete *elements of the course* without supervision.
And although I wasn't involved in other programs, there were other programs where other kinds of requirements were in place for *prior* or *subsequent* qualification. So it would be reasonable, for example, to have no supervision for a 'certificate', but require supervision for a 'diploma', with a simple conversion exam.
That's all in the far past, but taking it as a template, it would be reasonable to offer some of your exams online, or to offer all of your exams online without conferring a headline qualification.
Upvotes: 2 <issue_comment>username_6: Yes, having a mandatory in-person exam for an online course is a reasonable and recommended practice by at least some experts in online teaching. For example, consider this work:
**<NAME>, and <NAME>. *Teaching online: A practical guide*. Routledge, 2017.** This was used as the official text in an online-teaching certification for my university; note that this predates the COVID pandemic.
Ch. 5 presents sample syllabi for both online and blended-course (hybrid) options. The syllabus for the "fully online version" version includes a single, in-person, proctored exam for 25% of the course grade (p. 131):
>
> 4. Proctored exam
>
>
> There is a mandatory proctored exam that must be completed no later
> than April 23. To schedule your proctored exam with an approved
> proctor near you, see the classroom link for “Exam Appointments.”
>
>
>
Again, this is for the "fully online version" of the course. (The blended/hybrid syllabus that comes after expands to include required in-person meetings on alternate weeks.)
So it seems that the best-practice recommended by these online teaching experts is that at least one in-person proctored exam is necessary to maintain course integrity. I would suggest that with the rise of new cheating services in the last few years (e.g., Chegg, ChatGPT), this is likely more important than it was a few years ago.
Upvotes: 4
|
2023/02/13
| 888
| 3,816
|
<issue_start>username_0: This is about a post-review rejection. This is one of the top journals in our field. The reviewers found our work interesting but sent a long list of controls and additional experiments to do. The editor rejected it, saying it would take a lot of time. However, most of the concerns were addressable and we appealed. The editor said he'd be happy to reconsider and we can submit, **via email**, a point-by-point response letter along with a new version of the manuscript incorporating all the changes.
We took the time to do so and sent everything he asked us for via email, but there's been no response in 5 weeks. There's been no acknowledgment that he received it. My advisor sent a follow-up a week ago just to make sure he received the email, but there's been no response to that either. The portal doesn't show any indication of a reconsideration. The editor doesn't have any automated response indicating he is on a vacation, and according to the journal website, he doesn't seem to have moved to another job.
I know that the situation is unusual (especially with the emailing part), but what are my options? Should I reach out to someone above him? Resubmit the manuscript on the portal? (Which is what we probably would have in the first place if he didn't explicitly instruct us to email him the files)
I have never had this long of a waiting period before we heard back from the editor about whether they are sending my submission to reviewers. (Usually takes 2-3 weeks for me). But this time I don't even have an acknowledgment of the submission in 5 weeks.<issue_comment>username_1: I'm reading the OP as you sent the revised manuscript to the editor via email. If so, this is not a good idea - papers should be submitted via the journal's editorial management system (EMS). If you send a manuscript via email, the editor would have to submit it to the EMS for you. It's not *that* hard to do, but it's still annoying, especially if one is not a full-time staff of the journal.
However, this doesn't explain why the editor has not answered you for five weeks.
The best thing to do would be to keep contacting the journal until someone responds. Personally, I'd prefer contacting the journal office. Full-time staff of the journal ought to know what to do, or who to contact.
You could also just resubmit the manuscript. A properly-configured EMS will flag your manuscript as a duplicate (since it has already been submitted before), and then whoever is looking at the manuscript will in all likelihood assign it to the editor who originally handled the paper. Still not ideal (since it might be viewed as rather impertinent), but if your communication with the editor was via the EMS, those messages will be logged and can probably be accessed by a third party, and they should know what to do.
Upvotes: 3 <issue_comment>username_2: Here is an answer based on my comments:
1. Contact journal's editor-in-chief, describe the situation (very politely, without assigning any blame) and ask for an advice on how to proceed. Make sure, you include the current editor in the correspondence.
2. Given your username, I assume that you are still a graduate student. Then, make sure that you involve your graduate advisor in the process, maybe make them co-sign the email to the editor-in-chief. Also, ask your advisor if they happen to personally know the current editor and editor-in-chief. The knowledge of their personalities might alter the cause of action.
3. Regarding your question "I am worried that the current Editor might consider this as going over his head and be rejected as a result." - I can only say that it is impossible to predict, but given that the current editor did not respond to your emails for 5 weeks, going to the next level is appropriate.
Upvotes: 1
|
2023/02/14
| 825
| 3,611
|
<issue_start>username_0: I was accepted to a fantastic [U.S.] program last year I was surprised to get an offer from but because they didn't offer me any funding and my life was still in a whirlwind due to the pandemic, I deferred. I reached out about a month ago reiterating my happiness in being accepted and then I asked whether I could be reconsidered for funding. This is the part where I messed up. I said that I was looking forward to working with their faculty, which they took as planning to enroll. They responded by saying that they were happy to hear that I was enrolling for the fall, and they also cc'd one of my POI. When they responded, I realized that I hadn't chosen my words well.
The program itself is exactly what I am looking for in an MA program, but I really need funding to make it work. If had received any last year, I would have probably figured out how to move to that city and started the program on a part-time basis.
The bigger issue, if you haven't guessed it, is that I applied to a few other programs this cycle with the hopes of receiving funding. My question is, I haven't signed or submitted anything, I have only sent that email saying that I was interested in being considered for more funding. I'm guessing I should just let things be, and I honestly don't know if I will be accepted to any of those other programs.
Now I'm really worried that I've inadvertently committed myself and will be in a tricky situation if I do get a funded offer. Do grad programs take such statements to heart, or do they understand that grad applicants tend to change their minds about a million times in the process? I also would hope that graduate programs would be understanding of the fact that students want to avoid creating huge debt for themselves?<issue_comment>username_1: This is a genuine mistake, so send an email **as soon as possible** to explain the misunderstanding. Admission committees are fully aware that MA programs cost a lot of money and that many people couldn't enroll without funding.
As soon as you are clearing the misunderstanding quickly enough so that it is not a burden for the admission committee, there should be no problem.
Upvotes: 2 <issue_comment>username_2: >
> I really need funding to make it work
>
>
>
Be direct and honest about this. If you can only attend the program if funding is available, there's no need to be cagey about it. Either there is funding available for you and the program is a possible option (pending your other applications), or there isn't funding available and it isn't an option for you.
Don't just let it sit. You should respond, take responsibility for the confusion that you were accepting admission (not necessarily because it's your fault, it's just the polite thing to do), and clarify that while you are excited and happy to be accepted, you are asking about funding opportunities because without funding you will be unable to attend.
Now, generally US masters students are not funded. There are exceptions, and if you require it then it only makes sense for you to pursue those occasional exceptions, but usually available funding through TAships and RAships are held for PhD students. Training PhD students is seen as somewhat of a public good, as it's part of training the next generation of academic researchers. Training masters students, on the other hand, is seen more as part of the individual education for those students to take jobs with higher salaries outside of academia. Therefore, it's expected that masters students will pay their own way as an investment in future earnings.
Upvotes: 2
|
2023/02/14
| 912
| 3,868
|
<issue_start>username_0: I recently submitted a paper to a (not very selective) journal (Sage publications). After a week of submission, the status changed to "awaiting reviewer selections". I took this to mean that the paper has passed the desk review, since the editor is selecting reviewers. However, after a few days, I got a desk reject notification.
Is it possible to get a desk reject after the "awaiting reviewer selection" stage? Is it possible that the journal didn't find reviewers, hence rejected the paper? The rejection email gave only general feedback, hence I can't infer the exact reason.
Further, the journal submission guidelines indicated that the authors should suggest three potential reviewers. However, there was no specific field to submit this information, hence I didn't include it anywhere. Given this context, I have three questions:
1. Was my paper desk rejected because I didn't submit reviewer names? In general, can a paper be rejected for this/similarly trivial reason?
2. Do you include reviewer names in the title page/cover letter, if there is no specific field given to enter this information?
3. Can I re-submit to the same journal with the reviewer names included? Will it offend the editor?<issue_comment>username_1: >
> Was my paper desk rejected because I didn't submit reviewer names? In general, can a paper be rejected for this/similarly trivial reason?
>
>
>
Yes, that is possible. It depends on how many papers are submitted and how much space the journal has. It is not uncommon for a journal to have to reject 90% or more of the submissions. In those cases it helps everybody to reject early.
An alternative reason could be that some associate editor took control over a bunch of submissions including yours, pushed the "awaiting reviewer selections" button for all of them, and when it came time to select reviewers for your paper (s)he looked at it a bit more carefully and decided: "nope, that is not worthwhile".
>
> Do you include reviewer names in the title page/cover letter, if there is no specific field given to enter this information?
>
>
>
I would have just contacted the journal and asked.
>
> Can I re-submit to the same journal with the reviewer names included? Will it offend the editor?
>
>
>
I would contact the editor and ask.
Upvotes: 1 <issue_comment>username_2: >
> Is it possible to get a desk reject after the "awaiting reviewer selection" stage?
>
>
>
Clearly it is, as your example illustrates.
More specifically, though: You are looking for all sorts of formal reasons why the paper might have been rejected, or maybe why they *wrongly* rejected it. But I don't think that's the right way of thinking about it. If an editor *requires* you to name three reviewers but can't find the three names anywhere, then they will generally tell you so rather than making up other reasons to reject a paper. The same applies to most of the other hypotheticals you are asking about: If an editor *needs* some information, it is easier to just *ask* for it than to come up with phony reasons to reject the paper.
My take is that the editor assigned the paper to reviewers and then one of two things happened: (i) They took a closer look at the paper and realized that there is no point to ask anyone to spend time on it; (ii) one of the reviewers wrote back and said something along the lines of "If you absolutely need me to, I can spend the time and write a full review. But I've looked at this paper and it will not stand a chance to make it past reviewers; you might as well save everyone the time and effort and reject it outright because of A, B, and C".
As for re-submitting to the same journal: You can always try, but it's rarely successful in practice. Take whatever feedback you got, make the paper better, and submit it somewhere else.
Upvotes: 4 [selected_answer]
|
2023/02/14
| 758
| 3,126
|
<issue_start>username_0: As a new hobby, I am wanting to read more and stick to the facts only.
I recently became interested in the public education system and many sites and YouTube videos claim that the modern American public education system is based on the Prussian system of indoctrination after Prussia lost to Napoleon.
None of them provide sources. How do I go about in my research to find sources for these claims?
Are there scholarly websites that publish articles on such events?<issue_comment>username_1: >
> Was my paper desk rejected because I didn't submit reviewer names? In general, can a paper be rejected for this/similarly trivial reason?
>
>
>
Yes, that is possible. It depends on how many papers are submitted and how much space the journal has. It is not uncommon for a journal to have to reject 90% or more of the submissions. In those cases it helps everybody to reject early.
An alternative reason could be that some associate editor took control over a bunch of submissions including yours, pushed the "awaiting reviewer selections" button for all of them, and when it came time to select reviewers for your paper (s)he looked at it a bit more carefully and decided: "nope, that is not worthwhile".
>
> Do you include reviewer names in the title page/cover letter, if there is no specific field given to enter this information?
>
>
>
I would have just contacted the journal and asked.
>
> Can I re-submit to the same journal with the reviewer names included? Will it offend the editor?
>
>
>
I would contact the editor and ask.
Upvotes: 1 <issue_comment>username_2: >
> Is it possible to get a desk reject after the "awaiting reviewer selection" stage?
>
>
>
Clearly it is, as your example illustrates.
More specifically, though: You are looking for all sorts of formal reasons why the paper might have been rejected, or maybe why they *wrongly* rejected it. But I don't think that's the right way of thinking about it. If an editor *requires* you to name three reviewers but can't find the three names anywhere, then they will generally tell you so rather than making up other reasons to reject a paper. The same applies to most of the other hypotheticals you are asking about: If an editor *needs* some information, it is easier to just *ask* for it than to come up with phony reasons to reject the paper.
My take is that the editor assigned the paper to reviewers and then one of two things happened: (i) They took a closer look at the paper and realized that there is no point to ask anyone to spend time on it; (ii) one of the reviewers wrote back and said something along the lines of "If you absolutely need me to, I can spend the time and write a full review. But I've looked at this paper and it will not stand a chance to make it past reviewers; you might as well save everyone the time and effort and reject it outright because of A, B, and C".
As for re-submitting to the same journal: You can always try, but it's rarely successful in practice. Take whatever feedback you got, make the paper better, and submit it somewhere else.
Upvotes: 4 [selected_answer]
|
2023/02/15
| 923
| 4,181
|
<issue_start>username_0: I submitted an article 5 months ago to a journal. Since then, the article has been jumping between "*Waiting for Reviewer Assignment*" and "*Contacting Potential Reviewers*" (10 changes of status now). The article is quite technical and multidisciplinary, so I understand that finding reviewers is hard. Still, this had never happened to me and I wonder how to proceed now.
I wrote an email to the generic address listed in the journal page (*journalname@publishername*), obtaining first an automatic response:
>
> (...) We are currently experiencing a high volume of requests and will answer your inquiry as soon as possible. Please allow additional time for the processing of recently-submitted manuscripts and for the posting of recently-accepted manuscripts as Accepted Articles. (...)
>
>
>
And then a fairly generic one, stating that the handling editor is presently working to secure reviewers, sending invitations "daily". In my question I asked whether they had secured at least one reviewer, but I did not receive a concrete answer, so I assume they have not.
What is the best course of action now? I have several options, from writing directly one of the editors (I do not know which one, one of the options would be to write the editor of a previous article we submitted and published in the same journal), to withdrawing the manuscript. How can I respectfully convey the journal the message that I would prefer not to withdraw the article\*, but that I will do so if they do not find any reviewers soon?
---
\*Note: this is actually the third article in a series submitted to this journal (the previous two were published) and I honestly think that it is an ideal fit, so I would prefer not to withdraw it.
Note 2: The article waited almost a month (25 days) awaiting associate reviewer assignment.<issue_comment>username_1: Although there are many courses of action you might take, perhaps consider asking yourself what do you want to achieve:
* to get your paper published sooner; or
* to motivate the journal to improve the publishing practices.
These goals are not necessarily aligned. If you want to achieve the second, you need to look for the action which matters for the journal, and perhaps consider withdrawing your paper and resubmitting it elsewhere. If you prefer the first, you may proceed with various complaints within the journal hierarchy (up to the Editor-in-Chief), which often improves the chances that the handling Editor might focus more on your manuscript, but usually does not impact the way how the process is (dis)organised generally.
Upvotes: 1 <issue_comment>username_2: If you are already corresponding with an individual, such as an editor, you could offer to provide a list of names of potential reviewers along with their qualifications. Don't include your collaborators unless you mark them as such.
Some papers, perhaps yours, are difficult to review simply because finding the necessary competence combined with willingness is difficult. Withdrawing your paper and submitting it elsewhere probably won't speed up the process if a new editor faces the same difficulty.
Upvotes: 3 [selected_answer]<issue_comment>username_3: I was in this exact boat with a top flight journal in one of my fields. I waited five months with reasonably consistent contact with editors, and they could not secure a referee for my paper even with several suggested referees on my part (I assume they were all swamped). I eventually withdrew it and had success in another journal. There's no guarantee that will work in your case, but different journals have different pools of referees they can pull from.
My paper was a bit more mathematical than a standard mathematical physics paper, was somewhat lengthy for how technical it was (~20 pages), and was extremely novel. I think all three played a role with regards to the challenges of securing a referee. You may face similar issues due to the nature of your paper based on its description. Reach out to an editor directly to see if you can get a better view as to what is going on and see if there is anything you can do to help expedite the process.
Upvotes: 1
|
2023/02/15
| 1,014
| 4,512
|
<issue_start>username_0: I am in my first year of PhD studies within science in Europe, and my interest and goal in the future is to to pursue an academic career, where I can find a permanent position in Europe. I am wondering if (and if yes, how much does) attending physically (international -) conferences, workshops, schools etc affect the application of a candidate searching for a post-doc position in Europe? Do you have a feeling for how many attended conferences etc are considered sufficiently low to be not invited for an interview of a post-doc position?<issue_comment>username_1: Attendance alone means little. But attendance is an opportunity to start to build a network that can serve you well later. If you, and your work, are known to many people and you are seen as a helpful contributor to the field, then your prospects improve.
So, it isn't the number that is so important, but how much you take advantage of the opportunity. You can and should follow up with people you meet and interact with.
The time between sessions is just as important as the sessions themselves as you get to meet people informally and chat over ideas.
So, zero is a bad number, and ten may be a good number or not. If you are invisible you gain little in the networking area, though you may learn something in the sessions.
At some point, going to the next conference lets you reinforce connections that you've already made. This can get you into a circle of collaboration, meeting friends/colleagues of those you met in the past. So, in that sense, more is better than less. But you have to work at it.
Upvotes: 4 <issue_comment>username_2: Just being physically present doesn't do anything.
Networking and meeting people who can hire you or speak well of you to others is valuable.
There's no magic "number of conferences", you aren't earning points that you later trade in. If you never attend a conference and conference attendance is typical in your field, it might look odd on a CV. If you never attend a conference and conference papers are the primary research dissemination mechanism in your field, not attending basically means not producing any research, so that's surely going to sink your chances.
For networking, though, it's going to be somewhat stochastic. You could meet the person that eventually hires you at the first conference you attend, or it could be that no one you meet is hiring when you are ready to apply.
Your PhD advisor should help you decide which conferences to attend and hopefully you can attend some together and they can assist with introductions.
Upvotes: 6 [selected_answer]<issue_comment>username_3: For your own professional and personal development, I think it's good to attend at least one conference each year (in Europe that could mean one national and one international every other year or depending on your funds it could be a European one one year and an intercontinental one the other year. Always apply for any travel fellowship you can too, to start building a funding record - even when your lab has money to send you somewhere!
But attending/sitting in a conference isn't going to do anything. Always submit an abstract, try to present a poster or, of possible and if your project is at that stage, select that you also want to be considered for a talk selected from the abstracts. This ensures that you are developing your academic/scientific skills and get feedback on your work - which will improve your science. Plus you get to see other people's posters and get new ideas as well - it's a lot easier to interact with others when you are presenting work yourself - posters are THE academic conversation starter).
Talk to people at breakfast, lunch, coffee breaks, social hours, dinners, poster sessions (yes conferences are exhausting), so you naturally start building a network in your area/field.
Of course, if you want to change fields at some point, or scope out adjacent fields for your postdoc, you can also see (and discuss with your advisor) if you can attend a conference in your "new"/desired field. But make sure you then have some conference experience under your belt, so you feel a bit more comfortable talking to strangers and senior scientists.
I remember that I found it intimidating for the first couple of years (I think everyone does), but everything becomes easier and better with practice and when you know a few faces (and yes, people do get offered postdoc positions or at least invitations or hints to apply).
Upvotes: 3
|
2023/02/15
| 662
| 2,923
|
<issue_start>username_0: I have a prof who is teaching between five universities, full-time at one and adjunct at the others. This semester alone they're teaching over 35 credit hours, which is double a regular full-time teaching load.
They are not organized and sometimes don't give helpful feedback.
Who can I report them to? This is too much of a load to properly teach.<issue_comment>username_1: Any administrator, such as a department head or dean, would do. It is possibly contrary to their full-time contract. But, before you make any report, make sure your information is accurate. Hearsay can be very damaging.
It is possible, however, that at some institutions this would be within the bounds, if unusual. It is hard for me to see how they could be fulfilling their contract with so much outside employment, even if it isn't teaching elsewhere.
Ask for confidentiality if you decide to report it. It is up to the institution to investigate and issue appropriate sanctions if warranted. It is even possible hat they have permission and some special need. But it is also possible that you need a better institution if this is within bounds.
Upvotes: 3 <issue_comment>username_2: Well… it is quite possible in the current academic context that some instructors need to stitch together multiple jobs.
Is this person tenured or tenure-track? It would be unusual for this person to be tenure-track as such a workload would affect other conditions required for tenure (v.g committee work, supervision, research etc). If this person is tenured it will be harder for the university to do much unless the instructor is in breach of contract, i.e. obviously does a poor job or is explicitly forbidden to hold a position outside of their current institution.
Keep in mind it is entirely possible for one to have a full-time limited-time contract position at one place without this restricting work outside of the contract. In particular, it could well be that this person does not want to abandon courses taught elsewhere so that, once the contract runs out, they have multiple cushions for a soft landing.
Ultimately the only thing you can do is provide constructive but critical comments at the course evaluation. While these are not necessarily available to the Dean, they at least point to the issue.
Upvotes: 4 [selected_answer]<issue_comment>username_3: There seem to be two different issues. First, you say that there are problems that you have experienced with aspects of the instructor's teaching; for example, that the instructor does not give good feedback. Additionally, you indirectly assert that the *cause* of the poor teaching is the number of hours that the person is working.
What is it you are really concerned about: the poor quality teaching, or what you infer to be the cause? I suggest that you focus on the problem and its possible solutions and not on your beliefs about the cause.
Upvotes: 4
|
2023/02/16
| 1,720
| 7,014
|
<issue_start>username_0: **TL; DR:** At least for the next few weeks, I'm going to be overwhelmed with reading a lot of research papers one after the other, sometimes even at the same time. How to deal with this? Are there any tricks to reading many research papers simultaneously?
---
I am a senior year undergraduate in Computer Science.
This semester, I took three of research-based courses in Computer Systems.
1. A course named "Topics in YYY"
2. A seminar course
3. A UG Research Project course [This is essentially "my" project, with a professor and a PhD student to advise and support me]
In 1, I am usually expected to read 2 papers per week.
In 2 as well, I have to read, review and present 2 papers per week.
For 3, initially there was less reading and more of brainstorming and experimentations. However, now that we're going to be writing a research paper based on our initial findings, there are a lot of "related" work papers that I have to go through.
In 1 also, initial reading was easy, but as the course is progressing the readings are getting advanced.
This is probably a result of poor planning on my part. I should have taken some programming-related or basic courses instead.
But the end result is, at least for the next few weeks, I'm going to be overwhelmed with reading a lot of research papers one after the other, sometimes even at the same time.
I could drop either 1 or 2 but that means running away from the responsibility; and my transcript will show the withdraw grade, so that's not really cool either.
How to deal with this? Are there any tricks to reading many research papers simultaneously?<issue_comment>username_1: There probably is no way of developing a speed-reading skill on such a short notice. You could approach the papers in a structured way.
Have a table (with more rows than shown) of data that you fill out for each paper:
| Aspect | Paper 1 | Paper 2 |
| --- | --- | --- |
| Method | | |
| Novel results | | |
You can consume the papers as a categorisation task where you look for the relevant information which you can report and forget or leave for later reading. The task of reading a whole paper is replaced with filling a table with data and moving on. **Obviously this is not the same as reading and understanding a paper** but would make your current situation manageable.
You don't need to fully comprehend every paper but will still become familiar with them and understand where further reading is needed.
A good order of reading a paper would then be: Introduction -> Conclusion -> Results -> Methodology. Literature review can be skipped unless you are interested in context.
Upvotes: 2 <issue_comment>username_2: My suggestion might turn out to be inappropriate to your domain of expertise but in my own area(s) it has proved very useful. Moreover, my method has nothing to do with so-called *speed reading*.
Many papers are published as essentially standalone entities. The authors assume a certain vocabulary on the part of the reader in a particular field, but do not assume that the reader knows very much about the sub-field in question. As a consequence, the introductory parts of the article, which might be several pages long, provide history, background and motivation for the research. Finally, after presenting the method and results, another long section is devoted to the discussion.
Basically, I skip the introduction and discussion in a great many papers and proceed instead, directly to the method and results. I know enough of my research areas to feel reasonably confident about my own assessment of the usefulness and validity of various research methods (notwithstanding that they have already survived the scrutiny of peer-review). If the methods survive my own assessment, I go ahead and read the results. Then, rather than reading what is often a long discussion that recalls many aspects of the introduction, I spend time *thinking*. To my mind, thinking often pays dividends that far surpass the time spent reading. To be more specific, I spend time considering whether the results pass a simply plausibility test and what kind of conclusions I can draw from the results in light of the various limitations (and of course positive aspects) of the methods.
And that's it! I'm not, of course, suggesting that you never read the introductory parts of a research paper, or the discussion. Rather, I'm suggesting that you can use your time more effectively by reading the parts of a paper that are *most likely* to tell you something new.
Having written this, I realize that my suggestion is perhaps least applicable to mathematics. Mathematicians seem to have a knack of getting to the meat of a research paper without too much flim-flam at the front. The results, instead, are allowed to speak for themselves.
Upvotes: 0 <issue_comment>username_3: I share with you something that I already posted in this conversation:
[How to write a "Related Work" section in Computer Science?](https://academia.stackexchange.com/questions/68164/how-to-write-a-related-work-section-in-computer-science/193404#193404)
Perhaps [how to read scientific paper faster (taking into account IMRaD structure)](https://francescolelli.info/thesis/read-scientific-papers-quickly-and-effectively/) is the article that will interest you the most. However, this is what I usually suggest to my students. You may want to start from [this link to my website contains a selection of videos that can help you in conducting a related work](https://francescolelli.info/tutorial/related-work-literature-review-survey-paper-a-collection-of-resources/), depending on where you are in your journey.
You also may want to consider the following tips for reading, selecting and categorizing the works that you want to include:
* Not all papers are made equal: [heuristics for understanding if a
paper is worth reading](https://francescolelli.info/tutorial/related-work-literature-review-survey-paper-a-collection-of-resources/) or you should move forward to the next
one.
* Not all journal and conferences are made equal: [understand how
to select good venues (conference or journal)](https://francescolelli.info/thesis/how-scientific-venues-work-an-heuristic/) where you can
search for good publication.
* Quantity may be important: learn how to [read a scientific paper faster](https://francescolelli.info/thesis/read-scientific-papers-quickly-and-effectively/) and more effectively.
* Keep your resources in an orderly manner: [master the right features
in MS-Word for handling the related work and managing the growing complexity of
the task](https://francescolelli.info/thesis/how-to-use-references-in-word-a-few-tips-and-suggestions-for-your-thesis/).
* A literature review is about put some order in the chaos: understand [how to categorize the papers that you are reading](https://francescolelli.info/thesis/how-to-use-the-literature-review-for-your-research/). In this way it will be beneficial for your readers.
Upvotes: 0
|
2023/02/16
| 1,753
| 7,109
|
<issue_start>username_0: I am finishing PhD studies in a field of physics, and I want to do something similar (but more advanced) for a postdoc. However after browsing Linkedin, Researchgate, Nature Careers and some others, I found that there aren't relevant postdocs.
But there are plenty of interesting PhD programs that would be a nice continuation of what I did.
I know that doing two PhDs is generally discouraged here, but my only alternative are boring corporate jobs that boil down to making powerpoints adjacent to what I studied.<issue_comment>username_1: This is not a direct answer, as I do not have enough experience to say anything about a second PhD. But there is something to be said about finding a postdoc position.
In my experience, many people (me included) did not find a postdoc by looking for advertised positions. Instead, you can reach out to potential supervisors directly, usually by email, and ask whether they have any opening for a postdoc position\*.
It helps if you have met them before or have any sort of connection, but it is not necessary. Usually, you'd want to include a short summary of who you are, what you would like to work on, and why you want to work with them.
In your case, you are already in a good position: you have already identified several groups that are looking for people to work on topics you are interested in. You should reach out and simply ask if they are also looking for postdocs to work on the same topic: in many cases, it may be possible that they can convert the PhD position into a postdoc.
\*: I appreciate this may differ depending on the country and the research field.
Upvotes: 6 [selected_answer]<issue_comment>username_2: I’m not sure what you are trying to achieve with another PhD. Really. For one, living on a student salary is not something most people want to extend if they can avoid it.
There aren’t enough faculty (or even postdoc) positions in physics for every graduating PhD student, so why would you think your chances of landing such jobs will increase once you have spent another 4 years working on another similar degree? In other words, what are the odds that doing another PhD will improve your situation rather than simply delay your entry to the private sector, which is where the vast majority of physics jobs are?
Realistically you will not find a private sector job doing supersymmetry or string theory but [the most recent available study (2015)](https://www.aip.org/sites/default/files/statistics/phd-plus-10/PhysPrivSect.pdf) of the American Society asked
>
> (p)hysicists who worked in the private sector … to identify the most rewarding aspects of their jobs. They wrote that they found their jobs intellectually stimulating and challenging, and that they enjoyed regularly working with smart and interesting people
>
>
>
and that
>
> (t)he types of careers commonly pursued by physicists in the private sector involved solving complex problems, managing projects, and writing for a technical audience.
>
>
>
It is also the case that, in “hot” fields like quantum information, private companies can’t hire physics graduates fast enough. In Canada, a country which historically lacks the US in terms of private sector jobs for physicists, the federal government expects that private sector jobs in quantum technologies will [increase 200-fold](https://ised-isde.canada.ca/site/national-quantum-strategy/en/national-quantum-strategy-consultations-what-we-heard-report#s0) in the next 20 years or so. That’s 20000%, i.e. that’s really 200 times. I can’t find a similar document for the US but or Europe but the X-factor must be in the same range. Universities have difficulties attracting faculty in this area because they can’t match offers from … you guessed it: the private sector.
All of this to say that suggesting private sector jobs are “boring” is a bit of a broad exaggeration.
If you are hell-bent on doing another degree, consider instead a Masters in a cognate field (computer science or even an MBA) where the timelines are shorter and you can complement or highlight the skill set you picked up in your PhD.
Upvotes: 4 <issue_comment>username_3: In general, it is very rare to see/have someone do a second PhD. Depending on the country you are in, you are not even allowed to do one in a similar area, only in a different field. So sporadically you come across someone who has a PhD in physics and theology, for example.
Apart from that, I would also suggest that you think about why you want to do another PhD. Presumably because you like what you are doing right now in terms of day-to-day work. But what are your longer term dreams/desires/wishes? Where do you see yourself in 5 years or 10 years? Do you want to stay in academia? If so, then a postdoc would seem like a logical follow up as it would prepare you towards independence or a staff scientist position. Maybe you want to teach mostly?
I'd say: grant yourself the time to explore options. There are plenty of physicists super happy doing exciting work outside of academia, so talk to them and see if their life/work would fit you or not. Same for academia.
An definitely e-mail labs you are interested in and offer to write your own grants to support your salary etc. to increase your chances of landing a postdoc if that's what you want.
Upvotes: 2 <issue_comment>username_4: A second PhD is a strange enough choice that I think it will not look nearly as bad on your resume as people think.
I knew someone that a PhD in physics and then got a PhD in philosophy and ended up in a successful career in philosophy of science.
A professor I knew did a PhD in both math and in biology. Also my PI when I was an undergrad I've heard claims that he was chosen because he had masters in chemical engineering (I think), which made him stick out for our engineering focused school.
I would say if your new PhD is sufficiently different, it will make you certainly look a lot more unique to other candidates and will also give you the option of transitioning to something that's a better fit for you.
Overall though, if you can manage to change fields by getting a postdoc position in the new field of interest, I think this is a lot more natural of a progression.
Upvotes: 2 <issue_comment>username_5: Аndrea is right, in that sometimes the funds for hiring a PhD student may be converted into hiring a postdoc. With this circumstance in mind, you indeed should continue your search.
Nonetheless, as a back-up option, you still may consider doing a second PhD in an adjacent area. My friend, who had received his PhD in solid state physics in the Ukraine, found it difficult to get a permanent position in the US. So he made a decision to do his second PhD, this time in material science, and at a well-known American school. Given his excellent background, he did it relatively easily and quickly. Soon thereafter, he obtained a permanent job at a respected government lab.
So, if your financial and family circumstances permit you to do this, you still may consider the 2nd PhD as a back-up option.
Upvotes: 2
|
2023/02/16
| 328
| 1,254
|
<issue_start>username_0: I am a student of English literature.
Since 2018, I have regularly published short stories and literary articles in national and international newspapers. The number is around 20.
Should I mention these published write-ups in my academic CV? If yes, then what is the right method of mentioning it in the CV?<issue_comment>username_1: You should absolutely mention successful work in your CV if it's relevant to the position/program that you are applying to.
With so many, it is best to mention the number of submissions and the nature of the work and periodicals and individually cite only a couple of the most recent or relevant ones.
Upvotes: 4 <issue_comment>username_2: **Yes.** To build upon [username_1's answer](https://academia.stackexchange.com/a/193523/33210), list them under a section such as *Public outreach*, *Media outreach*, or *Popular press*.
I would look at the CV's for the faculty (usually listed on their website) to see if/how they list these articles.
As a caveat, the articles are not peer-reviewed so do not try to present them as such.
Some people will value them on your CV, others will not.
As long as you do not lie, this should not hurt you.
Only help you or be netural on your CV.
Upvotes: 4
|
2023/02/16
| 1,516
| 6,399
|
<issue_start>username_0: I am doing my first postdoc in pure mathematics, which will come to an end in four months. I have been applying for numerous positions, of course, but to no avail. It is about time (has been some time?) that I considered a contingency plan.
Now a rule of thumb is, I hear, that once you get an industry job, you do not simply get back into academia. Even outside of the US (which I am).
I am interested in applying for jobs in the next application cycle in case I am not offered any job this time. Given the rule of thumb, would I be more likely to be offered an academic position if I stayed unemployed (except for odd jobs, if need be) after the termination of the current contract than I would if I got a full-time job in industry?<issue_comment>username_1: It is not a matter of you not being offered jobs after going to industry. It is more that people who went to industry don't want to go back to academia. So, don't worry. Take a job if you get an offer and see what the appeal is. That is better than being unemployed.
Upvotes: 6 <issue_comment>username_2: >
> once you get an industry job, you do not simply get back into academia
>
>
>
I think you're missing the causal structure of this concept. The issue is not that having an industry job is somehow disqualifying for academic ones, but rather that it's extremely difficult to produce the work that is valued in academia (in particular, publishing research) while not employed in academia.
Can you improve the pace at which you publish in academia while being unemployed relative to when you were a post doc, enough to get another job in academia? Maybe you think you can, but I think it's very low probability. There's also the practical issue that being unemployed means you don't have a paycheck to pay for the things you need.
Upvotes: 6 [selected_answer]<issue_comment>username_3: I wonder whether you have really considered all of your options. If your job search has been focused too narrowly it is harder to get any position. If you are willing to teach, primarily, rather than do research there are many more options (in the US, at least). Nearly every US college needs math faculty since even non-math majors take some math courses. And many, while small, still have active faculty.
So, broaden your job search if you think it is a bit narrow.
But if you do leave the academic market for a while it is harder to get back for (at least) two reasons. In industry you make much more money than academics and it can be hard to give that up once you get used to it and raise your cost of living. The other reason is that it is difficult, in most industry positions, even research, to do things that are relevant to academia. Much industry research is product focused, not knowledge based.
One option, if you can afford to do it is to take on some academically relevant project to fill a gap year. Writing a book is a possibility for many. It could be a textbook or something more focused on your research area. Among other things, it gives a reason for a gap.
And, keeping contact with potential collaborators is also possible.
Upvotes: 4 <issue_comment>username_4: In addition to the answers already given, I would like to point out that there is at least one other option.
If you can afford to be unemployed for a year, you might be able to obtain some sort of "affiliate" status with your current university that can allow you to effectively have an additional year of unpaid postdoc in your current position.
This isn't a very good solution (and a terrible precedent), so you definitely shouldn't do it for more than one year. It's worth considering, however, within the space of possibilities.
Upvotes: 2 <issue_comment>username_5: Working in a company provides you with useful experience, not just in terms of CV, but of personal growth too. You get to see problems and situations from a different perspective. If you want to learn about leadership and effective management, a company job may help you a lot and turn out to be useful even if you decide to go back to academia. Moreover ...
@username_3 "The other reason is that it is difficult, in most industry positions, even research, to do things that are relevant to academia. Much industry research is product focused, not knowledge based." Yes, but that depends on the irrelevance of a lot of academmic research. Experiencing how mathematics or physics or whatever are actually used in practice gives sense for which problems and techniques are actually relevant, as well as providing insight into their deep motivation, beyond the formalism. Often deep, new results emerge from concrete problems , rather than from academic ruminations.
Upvotes: 1 <issue_comment>username_6: I was in exactly your position nine months ago (applied math rather than pure math). I was geographically limited by where I looked for faculty positions, and while I got interviews, they never led to offers.
As mathematicians we have an advantage compared to other sciences: our research is often not limited by a lack of funding. Not sure what side of pure math you do, but my guess is that to get work done all you need is pencil, paper and laptop. Get yourself a day job, and treat math as a hobby. Work on hard problems that you were too afraid to work on before, perhaps for lack of progress or a perception that "this problem is for people with job security". Publish as unaffiliated or, if you don't want to list your home address as your institution, incorporate and register a non-profit research centre that only exists on paper (I'm looking to do this right now, if only because it's an adventure in its own right). Or, as others have suggested here, try to get a job where publication is expected.
If a professor position is your dream, that's fine. But don't let the dream tie you down. You might find, like I did, that there are plenty of interesting careers outside of academia, many of which pay more (either initially, or have a longer runway of earnings potential), allow great work-life balance and provide superior benefits. And besides, there's no rule saying that you can't go back later. Industry experience is valuable, and depending on the industry you enter and the role you get, you might find that what you do is transferable if you decide to throw your hat in the ring for a tenure track job along the line.
Upvotes: 2
|
2023/02/16
| 2,278
| 9,522
|
<issue_start>username_0: Some years ago, I was stiffed for $100, but I let it go and took steps in my business model to be sure it never happened again. However, my colleague, is being stiffed for $600, an amount that is pretty painful to ignore.
**Situation**
1. The university provides a list of approved editors to students. (The editors are *not* university employees.)
2. Student pays deposit for editing their doctoral proposal.
3. When invoiced for the balance, student ignores the editor. They will not
respond to texts, emails, or phone calls.
4. Student is still in the doctoral process.
5. Student is also a prof at a different university.
Is any part or parts of what has happened an ethics violation that can be reported? If yes, to who (the chair? DSEM? student's advisor? Other?)
**Clarifications**
***Editing Context***
This is a response to questions from viewers of this post who might find it helpful to have more context. Please take note this is in a U.S. context.
* DSEM = Dissertation Scholarship Editorial Manager. This person approves the readiness of a study for the next step at different stages (i.e., ready to write the proposal, ready to defend the proposal, ready to defend the final, ready for dean's review).
* At the start of their writing, all students are advised to plan for an editor because in the end, all final doctorals must be submitted to DSEM with an editor's signed certificate of compliance with APA and university standards.
* Some students take this immediately to heart, starting with the doctoral justification. Others do nothing until either the chair or DSEM insists on it.
* Chairs who take their role seriously will know early on which students need immediate editor help to ensure that DSEM approves the work at every stage. Other chairs will "pass the buck" and let DSEM reject the manuscript over and over.
* Although students do not need to use an editor from the approved list, it's high risk because students do not know what they do not know about the expected standards or process and therefore how to vet an editor's qualifications.
* In addition to proofreading and copyediting for APA standards and university Handbook standards, it is *very* typical to provide coaching on how to synthesize, "connect the dots" within and across chapters, write transitions, and more.
* On occasion, the university does contract directly with an approved editor on a student's behalf. This is not the case in this situation.
***Debtor's Academic Experience and Credentials***
To the degree that a LinkedIn profile can be trusted, the debtor claims to be a
* graduate with a PsyD in Industrial Organization at the same university related to this post;
* current adjunct instructor at a different university and a technical college, and an adjunct professor at a community college;
* past dean of academics at a business school; and a
* past faculty program advisor at a different university where they provided subject-matter expertise in "... global financial markets, legal environments, ethics, ... ."<issue_comment>username_1: Never
-----
Failing to pay one's personal debts may be immoral or illegal, or even a violation of personal or business ethics, but it has nothing to do with **academic ethics** - even if the debt is for academic-related services. And any other kind of ethics is outside the scope of this site.
Academic ethics might be relevant to the question of *whether* it's ethical to use a third-party editor for academic work, but I don't see it as having anything to do with the financial side of such an arrangement.
If your colleague tries to "report" this person to the student's advisor, employer, or any official at either university, I would expect them to say: "That is a matter between you and X, it has nothing to do with me or the university. Goodbye."
(The other alternative, if they have a strict interpretation of FERPA, would be: "I can neither confirm nor deny that there is any student named X affiliated with this university. Goodbye.")
Actually, in some jurisdictions it is *illegal* for a creditor to reveal the existence of a debt to any third party for purposes of collecting it. So your colleague must tread very carefully.
By analogy, look at what happens when a student owes money to the university itself (overdue tuition, library fines, what have you): the university might withhold any further educational services (by preventing the student from taking classes), but I have never heard of them invoking any kind of academic ethics procedure. If they don't do it for money owed directly to them, they surely wouldn't do it for money owed to an unrelated third party.
Upvotes: 5 <issue_comment>username_2: Whilst there are certainly ethical issues around paying debts, the ethics processes in universities are not really there as a mechanism for adjudicating ordinary commercial disputes like unpaid invoices. A claim for an unpaid invoice can be pursued in the small claims court in your jurisdiction and that is the appropriate forum for it. Sometimes invoices are unpaid because there is a dispute about the delivery of the service pertaining to the invoice, and these are the types of issues that are resolved in small claims court. It is unlikely that a university would want to wade into the details of a dispute over service provision and payment, and so it is unlikely that the university would consider this issue to fall within the scope of its ethics policy.
Upvotes: 5 <issue_comment>username_3: It may not be directly actionable by the university. However if the student is failing to respond, it is perfectly justifiable to directly contact their advisor at the university they're a student at, their departmental head at the university they're a student at, and their departmental head at the university they're a faculty member at. Phrase it as "I'm very sorry, but I can't get hold of them and perhaps they've moved house and you're the only other contact point I can think of" or something like that, which makes it all sound reasonable and polite. But be very clear that you're trying to contact them because they owe you money for work on their course and they haven't responded to you.
As a faculty member, it also becomes easier to visit their actual university to demand payment in person.
Upvotes: 3 <issue_comment>username_4: **At any point when the student attests to the school that the unpaid work is their own**
This answer has come to me after considering the answers and the comments from other posters, as well as my understanding that students must attest to the work being their own.
Here is my line of thinking on when failing to pay the editor (or a fact checker or anyone else who was hired to help the student) crosses the line from a contract dispute to an ethics violation.
1. The editor has provided services that the student accepts (in essences, approves).
2. However, the student cannot attest to "owning" the effort until the service has been paid for.
3. To attest that all of the work is theirs is true only when they have, indeed, paid for all of the work. If they have not paid for the work, then they cannot ethically claim ownership of the entire work.
Depending on the institution, the student may not be asked to attest to the "their" work until they pass their final oral defense. In others, they might be attesting to "their work" at several major gates in the process.
Upvotes: 0 <issue_comment>username_5: I'm responding as a dissertation and faculty editor with various types of relationships with universities (approved editor, designated editor, and on the payroll).
First, this is not merely a case of ethics: theft of services is a crime and your friend could simply report it to the police, which would undoubtedly get the debtor's attention. In fact, a simple letter stating that intention might be enough to get the bill paid.
The ethics involved here are more general rather than pertaining specifically to academia. Along the lines of **"If you see something, say something."** If your friend does not report the debtor to the university, that sends him the message that he can get away with theft. Assuming that the debtor is willfully refusing to pay his bill (and not, say, incapacitated), he has a character flaw that is likely to be repeated in future contexts. Any current or future employer would not be well served by employing him. In that sense, it's a service to the university and society to report him.
I agree with username_3 that the editor who is a victim of theft would be justified in reporting the incident to the student's department, especially since the university maintains a list of approved editors. The editor does have in some sense a relationship with the university despite not being on the payroll. It's in the university's interest to stop incidents like this to keep this DSEM system moving smoothly. Otherwise this debtor could very well cheat his next editor.
Holding the debtor accountable is the right thing to do morally, and anyone who is aware of the situation should exert whatever influence they have to correct it. I see no reason why the student's advisor could not have an informal conversation with him about the bill.
I would try the following solutions, in this order:
(1) Letter to the student stating the intention to take action to collect the bill.
(2) Speak with the student's advisor.
(3) Letter to the department chair.
(4) Report it to the police.
Upvotes: 2 [selected_answer]
|
2023/02/17
| 579
| 2,419
|
<issue_start>username_0: I am referring to PhD admissions in US here. I've been always told that SoP's are important and require some commitment in writing. But in the meantime I hear a lot of such voices that PhD programs typically receive tons of applications like 500, so professors in committee would only spend a matter of 5mins-ish long reading about a single application. I am wondering which is the case? I really want to express myself thoroughly in my SoP but I guess a lot of valuable information would not passed to the committee if only under a fast read. (Or is it that I am underestimating their reading speed?)<issue_comment>username_1: The world is what it is, and admissions committees are made up of individuals with their own ideas and priorities. So, you do the best you can on every required element of any application.
If a department really got that many applications there might be some initial weeding out process that reduces the number to a more reasonable level, at least for the first pass. Some applications are incomplete, some flawed. GPA can be a gate some places. Those weeded out may get cursory looks later, provided the more likely pool doesn't yield enough good candidates. Even a quick read of an SoP can be useful in avoiding quick rejection.
So, you write the best SoP you can, detailing academic and career goals and some plan, perhaps, for meeting them. If you do a poor job of this it lowers your expectations.
But, writing an SoP has another objective. It forces you to think about those goals and make them somewhat explicit. *Know Thyself*. A generally useful thing to do.
Upvotes: 2 <issue_comment>username_2: When I was on my department's graduate committee (which dealt with admissions), I read ~120 applications per year. About half of these are "obviously" not competitive for some reason or other (typically low GPAs) and you might not read all of it very carefully, but I read the SoPs of all competitive applications and I believe that my colleagues did the same. I read them quicker than I would the PhD theses of my students or a scientific paper, but I read them front to back. And they mattered to me and my colleagues.
The message here is: If your application isn't competitive, then you probably don't need to bother with a good SoP. But if your application is competitive, you *also* need a good SoP. Poor SoPs can and do sink applications.
Upvotes: 4
|
2023/02/17
| 1,651
| 7,439
|
<issue_start>username_0: *(I am posting this for advice regarding somebody else’s problem, but for simplicity, let me pretend it's me. To clarify misunderstandings in the comments: I am asking on behalf of my partner and can supply any additional detail as needed; I am fully informed.)*
I have conducted a research near the start of my PhD studies, which I have since quit. This was collaborative work with another PhD student (who has also left since then) and my/our supervisor. We had an agreement regarding the authorship of the paper: I, as the person having done most of the work, would have been first, followed by my classmate, and finally our supervisor.
Some time has passed since then, I have quit the programme, so did my classmate, but we have only got around to publishing the paper now.
Today, I received an email from my supervisor that since both of us students have left the faculty and are not planning on continuing our studies there, he proposes a change to the authorship of the article, making himself the first author, as (I paraphrase) *even though I have done most of the work, it would have not been possible without his financial and infrastructural support.* There is some truth in this argument, but as I said, authorship was already established.
I do not think it is fair to alter the order of authors merely because we are no longer affiliated with the institution; we *were* at the time of writing and the relevant research work.
The paper is already finished and is just waiting to be submitted. I am willing to take part in creating a second revision if requested by the journal upon review (and have made this clear to my supervisor).
---
Naturally, I have begun negotiating with my supervisor (and the 3rd author) in the matter, but I would like to ask: **if my supervisor refuses to give up the place of the first author** (I must add that I believe he in fact did very little work), **what are my options to resolve the situation?**
---
**UPDATE** To clarify further confusion in the comment section: the paper in question is a complete, finished manuscript, ready for submission. The reason this whole thing is happening *now* (several months after we have finished working on it) is that it has already been submitted to another journal previously and was eventually rejected. This is a resubmission to another journal.
There has been no change to the contents of the paper. The supervisor has done no additonal work (such as ‘picking it up as a draft and completing it’). He had very little input in the entire process.
I believe he wishes to change the author ordering because we no longer ‘need’ publications (having quit the academic career at least for the time being), whereas he does. It would not be the first time he claims other people’s work as his own, but this is not relevant for this post.
---
**UPDATE 2** Unfortunately (and unsurprisingly to me) it does not seem that our (former) is cooperative in the question. Upon reaching out, he offered to alter the name order, moving me up to the place of second author, but still keeping himself as first, *if and only if I agree to handle the entire submission process.* In addition, he has brought up omitting me as an author altogether?? That said, I do not wish to abuse SE for a blog sharing my experience with this problem. I will continue attempting to reach an agreement with my supervisor (and the 3rd author) and potentially escalate the problem to my uni’s ethics office or similar if there is no way to handle this locally. I plan to include update the post with a final conclusion to let future readers possibly facing such a problem see how it can be resolved.<issue_comment>username_1: If the work has all been done, and the first author will continue to take responsibility for the paper throughout the submission process (which may include multiple rounds of revisions, some with new analyses or even new experiments), there is no reason for the authorship order to be altered.
If the original first author is not willing to participate in all these steps, there will eventually develop some gray area where even if that person did most of the original work, they can't really be responsible for the sum total of the paper anymore. In that case, it may be more sensible to have someone else take the first author position. It would not be fair, though, for that author to be somehow *prevented* from participating. Also, if I were advising the supervisor rather than the student in this case, I'd strongly encourage them to let the student keep first authorship. It's just not good business sense to be a jerk.
Ultimately, if there is an authorship dispute the primary recourse is to oppose publication of the paper. Assuming everything in the publishing process works how it's supposed to work, papers can only be published when every author agrees with publication, including authorship order. At that point, it's basically between the parties to decide whether their place in the authorship order is worth not having a paper published at all.
(as a side note: present affiliation doesn't really matter at all; I'd expect your affiliation to be listed as the place you did the work originally, possibly with an added affiliation for your current location if still in academia, otherwise probably nothing else [Affiliation on a paper written mostly in previous position](https://academia.stackexchange.com/questions/16866/affiliation-on-a-paper-written-mostly-in-previous-position) [Changing affiliation on publication](https://academia.stackexchange.com/questions/11151/changing-affiliation-on-publication) )
Upvotes: 5 [selected_answer]<issue_comment>username_2: From the information given, I see no reason that would justify a change of authorship from your original agreement for you to be the first author. The fact that the paper was previously submitted with you as first author, and the supervisor has not made any additional contribution since then, is strong reason to keep the original authorship order. The provision of finance and infrastructure are generally considered to be relatively minor contributions to a paper in their own right, and would not usually warrant first authorship in a case where another person does the bulk of the substantive work on the paper. You might also be interested to note that some journals are moving towards a system where particular contributions of authors are specified (see e.g., the [credit system](https://www.elsevier.com/authors/policies-and-guidelines/credit-author-statement) at Elsevier), though authorship order is still important in many fields.
Regarding the process to solve this disagreement, the first step (which you are already taking) is to discuss the matter with your former supervisor and see if he will agree to retain the original authorship agreement. If there is still an impasse then you should propose to have the matter adjudicated by an independent academic or the ethics office of the university, with each of the authors agreeing to be bound by the decision. In this case the matter could be reviewed and considered by an academic who is independent from you and your supervisor, taking account of all relevant circumstances and contributions. If your supervisor does not agree to this then you could refer the matter to the Department and/or ethics office unilaterally and seek an adjudication on the matter.
Upvotes: 3
|
2023/02/18
| 506
| 2,175
|
<issue_start>username_0: Is it correct to remove a co-author from a resubmitted manuscript if the testing protocol was changed and the original experiments of the coauthor are no longer in the paper? The experiments were replaced, the result however is the same.
The co-author developed the method for this assay and worked on the project for at least 6 months. Can their work just be ignored by telling them that their measurements are not in the paper anymore and noone should be an author of the paper if they just measured „similar things“. This all went about behind the back of this co-author.<issue_comment>username_1: Others might have a different opinion on this. My opinion is that it is completely wrong to do so and the expunged person should take action if it is available.
They worked in good faith on the project, presumably. They offered ideas and contributed to conversations. Some of those conversations might, in principle, have cause the group to actually change direction. It is impossible to say, after the fact, that no contributions to *the work* defined broadly were made.
I'd treat it as a profound professional insult and would, myself, treat no one in that way, nor work with people who do.
Upvotes: 3 <issue_comment>username_2: That is really wrong as @username_1 says. This is evident, since the authors went behind the co-author's back (I mean if this is ok - why hide it?).
I will add that this kind of behavior, can have far-reaching consequences. Not only will you lose faith with your (presumably no longer) coauthor, most academic sub disciplines are small (say, 50 labs). Word gets around, and this kind of behavior will make it more difficult for a lab to get new students, new collaborators and in some cases hurt the PI's career (if there's a consistent pattern).
In addition, if you're submitting to the same journal (or a journal within the same discipline), then the editor may find out and react accordingly. Furthermore, the removed co-author can easily block the publication (if they're feeling angry enough) by emailing the editor and letting them know. A decent editor would prevent these kinds of shenanigans.
Upvotes: 2
|
2023/02/18
| 1,844
| 7,891
|
<issue_start>username_0: I am currently writing my masters thesis. I'm fully aware of the academic guidelines when it comes to citations.
Yet, I have a more nuanced question that maybe more stylistic in nature: when explaining a general concept, rather than specific research findings, I often use academic books as my source. Based on my understanding, I then introduce the concept in my thesis. When doing this, what do you think about citations such as "for more details see (Author, 1998)" at the end of the concept introduction part? Would you cite this in the text or rather with a footnote? Would you skip the "for more details" completely?
I do realize that a lot of researchers (which I am not, so I might not have this privilege as a masters student) tend to not cite "broader" concepts at all. That is, they introduce it in their own words without citing any sources.
Would you say I could to this as well? Even though I fostered my understanding of a concept from a book, I am able to comprehensively introduce it without looking at the source in my own words.
I know this question might come off as being obsessed with insignificant details, yet I am still wondering what the more senior guys on here think.<issue_comment>username_1: If the concept is very well known in your field, then it is not essential to cite anything. But in your case, where you are a master's student and learned about it from a book, I think you should cite the book(s).
For your second bullet point, I would cite it in the text, not a footnote, but it probably depends on whether footnotes are commonly used in your field. I would drop "for more details," but it is up to you really.
Upvotes: 3 [selected_answer]<issue_comment>username_2: In a master's thesis there is of course more flexibility around citation styles than for peer-reviewed articles, wherein these are currently quite strict, conservative, brief, specific, and univocal. I would say that the usual practice when writing peer-reviewed papers is to stick with a single in-text citation practice, such as "this is the concept or study - this is the paper containing it". That is not the same as to say that this is the best practice nor the optimal scenario.
In my humble opinion, this -usual- method is quite unfortunate and limiting, because this style is picked out of convenience for the author, and not for the reader. I suppose that in the fields of Humanities it is more common than in Scientific-Technological fields to find different citation styles with embedded instructions for the potential audience, because their narrative styles tend to be more sophisticated and empathic (less "machine-like" so to speak) due to the authors' training in narrative styles, which is not as common in other fields.
Supposedly, when an author provides a reference, it is because they feel that the reader may want to expand information on that particular topic. However, in academic circles nowadays, citations are used for various other purposes, such as demonstrating (or sometimes just showing off) the author's research capabilities, proving that the background literature has been checked, abbreviating otherwise lengthy explanations, computing research interest indexes, etc. All this can be seen as a perversion of the original purpose of a citation, which is indeed a deference of the author towards the reader put in place so to facilitate understanding.
Ideally, whenever a citation is added, the writer could provide a -necessarily brief- guide on its nature, and other metadata, such as the specific section being referred to, or the preferred edition, or the suggested way of browsing the referred material. Thus, is should be common practice (although sadly it is not) to provide various types or references, such as not only peer-reviewed papers, but also books, book chapters, technical manuals, press clippings, magazine articles, lists of authors, websites of various kinds, or multimedia content, preferably with accompanying footnotes. However, in some circles, using references that are not strictly peer-reviewed articles can be perceived as unprofessional, specially in those contexts that monetize scientific research the most, or in those that take great pride in their attention to rigor. There are, nonetheless plenty of examples of horrible papers written in a very rigorous style (e.g., [these](https://www.theatlantic.com/ideas/archive/2018/10/new-sokal-hoax/572212/))
Of course, the most time-saving (some would say "lazy") way to add a reference is to just drop a number in between brackets and then leave the rest of it to the audience. In my view, it should be common practice to provide two or more references per concept, instead of just one per concept, or to provide, for further study, a list of authors who delved deeply into the topic (for instance in the fields of Economics or Philosophy), instead of a narrow selection of their articles.
Furthermore, you must bear in mind that not all readers will read an article in its entirety, and even those who do will usually not check all of the references in detail; this is particularly true for scientific reviewers due to a scarcity of time and/or lack of incentives, although of course some reviewers will not need to run a deep check if the topic is narrow and they are sufficiently well-versed in the field. Sometimes they will not even feel the real need to read an article from a particular author because they have already met the author in person at several conferences, and they can read between the lines through your bibliography.
The truth of the matter is that, when later in your career you are doing cutting-edge science which goes beyond the state-of-the-art (something that everyone claims to be doing), most likely no reviewer will be sufficiently well-versed into what you are writing, so in the long term, it pays off to respect your audience as much as possible and to show off as little as possible.
Unluckily, the pressure that the pay-per-publish paradigm, the rise of predatory journals, and the scarcity of resources for research exert on both authors and reviewers prevents this from happening to a large extent. Some prominent sources are already talking about "the crisis of peer-review" ([see this for instance](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1420798/)). As a consequence, Research Assessment is a matter that is currently under scrutiny in order to prevent the current drift towards increasingly closed science (as opposed to open science) and hopefully steer it towards more fruitful lands. You will find some more information on these matters in the [Agreement on Reforming Research Assessment](https://coara.eu/agreement/the-agreement-full-text/) at CoARA (recently signed by the European Commission) or in the San Francisco DORA ([also endorsed by the EC](https://research-and-innovation.ec.europa.eu/news/all-research-and-innovation-news/commission-signs-agreement-reforming-research-assessment-and-endorses-san-francisco-declaration-2022-11-08_en)).
Notwithstanding all of the above, when citing for a master's thesis you naturally want to be perceived as a professional writer so to earn a good grade and minimize your risks, so it is more likely that you will act more conservatively and less innovatively by strictly following common practice, although that is ultimately a matter for you to decide. Anyway, I would suggest to only provide lists of authors if you know their works well, and not because you just saw them linked in some Wikipedia article, since doing this will stand out to the eyes of the trained researcher. It is also advisable to always study beforehand the preferred citation styles by your educational institution, your teachers, your advisor/s, and your potential audience, and to always stick with them (if, of course, they match your own values).
Upvotes: 0
|
2023/02/18
| 758
| 2,924
|
<issue_start>username_0: Last year, I submitted a paper to the conference XX'22 in January. We made an Arxiv preprint.
And the paper got rejected. We got very good reviews, except there were some imprecisions that we needed to improve.
Some months later we submit to YY'23 and the paper got accepted in November 2022. We wrote blogs about the paper, tutorials, we released software...
A few weeks ago, we presented the paper in the YY'23 conference.
Now I am a reviewer of XX'23. And I have to review a paper whose main claim is the same that we made in our submission in XX'22. It has been on Arxiv since then and has now been published.
The keywords are the same. The paper title has the exact same bigram. Their method is different, but with the same theoretical principles.
One of the experiments has the same evaluation methodology as ours.
And, obviously, there is no citation.
The main contribution of this paper is being novel about this technique that is very simple.
**As a reviewer, what should I do?**
Deadlines:
* Paper submission XX'22: Jan 2022
* Paper notification XX'22: May 2022
* Conference XX'22: July 2022
* Preprint in arxiv: Jan 2022
* Paper submission YY'22: Aug 2022
* Paper notification YY'22: Nov 2022
* Conference YY'23: Feb 2023
* Paper submission XX'23: Jan 2023
Note that the XX'23 submission is before the YY'23 presentation<issue_comment>username_1: I would recommend you do now what a peer reviewer is supposed to do: Give feedback on the scientific substance of the manuscript. You can explain in your review pretty clearly that there is previous work that would need to be taken into account in the discussion of the literature, and that the authors should explain how their work contributes new knowledge. Also, if appropriate, you can express your concern that there are similarities between your paper and the manuscript that suggest that the authors plagiarized. Then it is up to the editors what to do with this information.
It also would make sense to disclose to the editors as clearly as possible what happened from your perspective, like you do it in your question, and that you are an author of the earlier paper. Do not put this info into the review for the authors, though, in order not to violate the anonymity of the review process.
Upvotes: 7 [selected_answer]<issue_comment>username_2: It is relevant that your paper was published in YY23 and that it was on arxiv earlier- you should highlight the YY23 paper in the review, tell the area chair that you are the author of the YY23 paper, and let the area chair decide if it is too similar (even plagiarism?) or no longer novel, or that the publication is so recent that it doesn't affect novelty.
While it stings, it is irrelevant that your paper was rejected from XX22 since the authors of the 'new' paper had no access to rejected papers. Mentioning the XX22 submission will hurt your credibility.
Upvotes: 3
|
2023/02/18
| 433
| 1,801
|
<issue_start>username_0: I was having an email discussion regarding my postdoctoral application with my potential host supervisor. At one point, he asked for recommendation letter from my PhD supervisor.
When I contacted my PhD supervisor, he asked me to forward the above-mentioned email correspondence to him. So, I forwarded my email discussions with postdoc host to my PhD supervisor. The mail didn't contain anything related to postdoctoral project, but only on official matters like interview timings etc. My supervisor has sent his recommendations to the postdoc host in the same email track. I wanted to know is it unprofessional from my part?
Recent Update: I came to know that my PhD supervisor wrote a simple email instead of a formal letter with his signature and stamp. Moreover, he mentioned only my first name with wrong spelling (e.g. 'Dr. Jahn', instead of Dr. <NAME>). He also asked the postdoc host if he could collaborate with him in the same email. The email doesn't contain anything bad about me but the structure of it looked naive to me. Can anyone suggest anything regarding that?<issue_comment>username_1: It seems unlikely to me that there was anything confidential in what you sent and therefore little reason to worry.
It may have given your advisor some context in which to write a better letter, actually.
I suggest relaxation.
Upvotes: 3 <issue_comment>username_2: For the future you may consider putting your supervisor in CC in one of the email you send to the other professor.
>
> Dear X,
>
>
> Thank you for your interest in my application.
>
>
> My supervisor @Y is happy to provide you with a reference letter in the coming
> days.
>
>
> Best regards,
> Z
>
>
>
This is my preferred method to share email communication with others.
Upvotes: 1
|
2023/02/18
| 344
| 1,471
|
<issue_start>username_0: I was going through one advertisement for the Tenure track position. The job advertisement has written something like this: What does it mean?
Effective Date of Appointment: August 23, 2023 (Subject to Budgetary Approval)<issue_comment>username_1: It means that the job might be offered or not, depending on money being available. It is an ad for a "possible" position, but not a definite one.
This is probably necessitated by uncertainty of various kinds, such as legislative action at a US State university.
They can't make a definite offer, most likely, until the uncertainty is resolved, but they can look at candidates.
If the place and the position are interesting enough then it might be worth considering, but don't give up other opportunities "hoping" it will work out.
Upvotes: 2 <issue_comment>username_2: It means *somewhere* some budget isn’t finalized so there a chance nobody is hired because the money for the position never materialized, or the date of appointment may be pushed back because not *all* the money materialized.
This is a pretty standard disclaimer clause and so not much to worry about. It does happen that positions are pulled because of budgetary issues but *rarely*, and when this happens the university has the black eye since such actions are likely to impact future recruitment efforts. I mean: if you advertise X years in a row and the position is always pulled, who will apply in year X+1?
Upvotes: 3
|
2023/02/19
| 1,672
| 7,572
|
<issue_start>username_0: I am currently working (not in academia) after having graduated from college recently. I hope to apply to PhD programs at the end of the year.
I worked with a professor on a paper about a year ago and it got published in a top-5 AI conference. I was hoping to ask them to write my letter of recommendation when I apply for PhD programs. In fact, my first choice would be to work in their lab doing research similar to what I did with them earlier.
I wanted to explore the idea in my paper a little further and see if I could improve it on my own time. My paper presented a new AI algorithm to do a task that performed better than the baseline method. While doing this exploration, I realized that the results obtained in my paper with my novel algorithm could also be obtained with a slightly different, trivial baseline that behaves randomly. The baseline used in the paper behaved uniformly.
So, basically, my novel algorithm does just as well as a baseline method performing randomly. Nothing in the paper that got accepted to the conference is wrong, but it seems like the new algorithm I presented actually isn't all that effective.
What do I do? I'm obviously hesitant to tell the professor I worked with since they would be my strongest recommender and I want to work in their lab. I also know of other groups that want to use this algorithm. Should I tell them that they might as well use this random baseline?<issue_comment>username_1: If the results in the paper are correct, I don't see an issue here.
It sometimes happens that a simple algorithm can do the work of a previously known complicated approach.
If you can **prove** that the simple baseline method outperforms your previous complicated approach, then any reasonable researcher would see this as a net positive.
This is the point of research - you discover something, it works, then you continue exploring it and discover that something else works better. If I were the professor I would laud you for discovering a simpler approach to solve a problem!
Upvotes: 5 <issue_comment>username_2: If the original paper was strictly correct, then it should be left alone.
What I would do is continue to investigate your complex algorithm with an eye to determine when it outperforms your second randomized baseline.
Is there some specific type or subset of data where the complex algorithm outperforms baseline 2? What types of inputs break baseline 2, and how does your complex algorithm respond to those? Is there a modification to your algorithm that can make it beat baseline 2 in those cases?
If you can answer those questions, you can publish another paper saying that although simple baseline 2 beat your algorithm, it still works better in these cases or with these changes.
That way you're reporting on improvements and evolution of your algorithm. This is how many things work: AI and artificial neural networks were first proposed in the 1960's and they clearly didn't work very well until multiple decades of advances got them to where they are now.
You also want to report the work in this way to your professor. You can say that it appears our algorithm doesn't work well against baseline 2, but preliminary results show that these changes would fix the problem, and here's a plan for further investigation (if you hire me). That's very impressive as it shows the skills of independence, drive, and "stick-to-it-iveness" needed for a PhD.
Even if your complex algorithm is unsalvageable, you now have a new task: do better than baseline 2. Your next paper could be our complex algorithm A was defeated by baseline 2, but now complex algorithm B does much better than A and 2. If you keep going down paths like this, you now have a career in research.
Upvotes: 3 <issue_comment>username_3: #### Write a new paper showing this result
It sounds to me like you have a topic for a valuable new paper. In your previous paper you showed that there is a novel method X that solves a problem better than existing method A. In your next paper you can show that method X unfortunately does not perform better than existing method B, so it does not appear to be effective after all. Both results are publishable and both advance the discipline. Contrary to your question title, both results are still "substantial" --- they introduce a new method and show how it compares with various other methods.
Upvotes: 2 <issue_comment>username_4: I suspect some of the other answers are (over)generalizing from other fields. In AI, publishing two papers on something that doesn't work is indeed unfortunate. All the worse if the baseline propagated by the second paper is something that you should have tried while writing the first paper. Fairly or not, publishing a second paper with the "trivial baseline that behaves randomly" is a little like saying "Breaking news: I just realized that the algorithm I was hocking last year, and some of you bought, is total snake oil!" Not a great look, even if you are acting in good faith.
I would consider posting an updated version of the paper on the arXiv. Not a new paper, nor an erratum, but simply a new version with a brief new section showing the new results. You can then write to the other researchers who you know are using your method, and point them to this new version. This is probably the cleanest way to fulfill your ethical obligations while drawing minimal attention to this unfortunate incident.
Upvotes: 2 <issue_comment>username_5: Remember that your previous work was examined by peer researchers during the peer review, and if none of them suggested to try the random baseline you described here, then it means either this random baseline is not that common in your field (in which case your random baseline is a new contribution to the field, or at least to make it more known, so that other researchers will always think about using this strong baseline), or that probably it's common, but the reviewers also missed that (in which case, you shouldn't feel too bad, since it's not only you who missed this baseline).
At any case, one thing that you shouldn't do is to not mention this to your advisor. You should definitely mention this and discuss the implication. Sometimes, when a method that seems good gets outperformed by a simple baseline, it indicates that the thought process of coming up with the model might have something that needs fixing. And this may or may not spark new discussions and new models/algorithms.
I understand that having our work be defeated by a simple baseline feels devastating. We spent much of our time proudly working and proving and testing that method to work, and wouldn't like to see that the method is apparently not that good. But that's okay, we can treat this as getting new knowledge that perhaps our the mental model of how we approach the problem might be incomplete. We can think of this as a possible new direction where we might contribute to.
So, in terms of actions, the normal course of action would usually be:
1. Mention this random baseline with your advisor and discuss it
2. Think of another new method that improves over this random baseline
3. In the paper describing this newer method, also mention the fact that this baseline is quite strong and other researcher should use it. (or, if the task is kinda solved with this random baseline, then you've shown that it's not an interesting task, and maybe move on with other more interesting tasks)
Step 2 and 3 would probably be suggested by your advisor too after you do step 1. =)
Upvotes: 0
|
2023/02/19
| 1,695
| 6,976
|
<issue_start>username_0: While flipping through my thesis to look up details in writing a postdoc application, I realized that I messed up my reaction rate quantification of one protein covalently bonding with another on a gel. The error I made here is that I assigned too large an intensity value for an unreacted protein control.
So here's the thing: I've already submitted my thesis to the library and none of my supervisors or examiners caught this mistake. For instance, I had to really pause and look close at the bands itself to notice that the unreacted band was actually smaller in size before going back to review my own work. Now this result was part of a bigger experiment that simply didn't work and will not be published. This was one interesting result that meant to show how two parts of a protein reacted similarly, despite one being heavier. If this result was not there, it would not upend the key conclusion of my thesis, although I would have 2/3 less of a figure.
What should I do at this point? Should I tell my supervisor immediately or file a correction? Or should I just go on with my life with the notion that no one will likely read my thesis? I feel like my advisor is glad to have me out of the lab, but I don't want to be accused of fabricating data years down the line.<issue_comment>username_1: Not to worry. Things like that happen. Tell your adviser.
Your university may have a way to add a note to your dissertation on file explaining the (small) correction. If so, take advantage of it. If not, count on the fact that it's very unlikely that years from now someone will read your thesis and draw conclusions from this data.
Upvotes: 5 <issue_comment>username_2: If it is such a small error, just write an extra piece of paper titled "Errata" in which you describe the error very shortly and what would be the correct thing and put it in the thesis in your library. Alternatively, put it on your homepage. Honestly, if it is such a small error, akin to a typo, no one really cares, but rather expects it to occur. NO single thesis is without these small errors! I would regard myself already as happy, if there would be not more than one of these errors per page, which makes a lot of errors for a 120 page or so thesis. Because that is, what will happen, when you read your thesis again. You will certainly find the next error. And then the next one. And another one too. And so on. So stop reading your thesis! :)
Upvotes: 3 <issue_comment>username_3: I am a bit worried that your response to having made an invalid assumption is to think:
>
> should I just go on with my life with the notion that no one will likely read my thesis
>
>
>
What did you just spend the last 4 years (or more) of your life doing? Your thesis topic is (in theory) the culmination of years of work learning how to apply the scientific method rigorously. If your hope is that no one will read it.... What have you done with your time? Most of the professors that I know in biochemistry post doctorate work, and into the majority of the career was related to the work they did in their thesis.
Why am I worried about this?
----------------------------
2 Reasons.
1. What have you been wasting your life on that you hope no one will read your thesis? It should be a major accomplishment that you are proud of, that will help you as you start your career as a scientist.
2. As you apply to your first job with a PhD, Every single person you interview with will want to talk about your thesis work. **It is the capstone of your PhD.**
When you talk to them about your thesis, which do you think sounds better?
"I made a mistake in my thesis, but I assumed no one would notice it and am glad that it is over. You should hire me because that is what I bring to the table... A person that makes mistakes and hopes no one will notice".
Or "While going over my PhD thesis, after it had been accepted by my universities library, I realized that I had made a mistake. And I then went thru the (however challenging process it is) to file a correction. I am not perfect, but I am honest, I admit when I am wrong and fix my mistakes." It becomes a talking point that you can use when someone asks you "Have you ever made a mistake and what did you do?"
What am I saying?
-----------------
I don't want to be judgy... But I worry that your question was "Should I hide the mistake I made, or not" instead of "How will this mistake effect me/ what can I do to mitigate the impact it has". It makes me worried because one of these is an ethical question: Should I be dishonest and hide a mistake I made... Or should I not. And one is the question of someone who made a mistake and wants to know what they can do to mitigate the impact it has on their career.
At the end of the day, making a mistake on your thesis, while it would have been better to not make a mistake, is not a big deal. You followed your schools process and multiple people looked at your paper and didn't see the mistake. You are at the start of your career **no one expects you to be perfect**. But ignoring the mistake and hoping that no one will notice is a big deal... I can guarantee that your thesis will get at least cursory glance by anyone who is really interested in hiring you as a postdoc.
Upvotes: 0 <issue_comment>username_4: If I were you, I'd "take control" of your thesis by uploading the source files (assuming it's in a text-based source format like LaTeX) to GitHub or similar, put a label on the version that got published by the institution, then make a new commit with the corrections and put a description in a `CHANGES.md` file. Get the issue out in the open - this won't be the last time you make a mistake in public, and dealing with it openly & honestly, but not necessarily making a big deal about it, is a good habit to get into.
You might need to make sure you *have the right to publish your own work* like this, by the way, by checking with your institution.
Upvotes: 1 <issue_comment>username_5: There are definitely provisions like this in place at all academic institutions.
This is especially the fact that your error is **NOT A TYPO** but a **CHANGE IN VALUE** which means a slightly different outcome, as you pointed out will reflect in the figure - the others on here have not recognized the type of error, but just dismissed it as trivial.
**DON'T assume that no one will read it or just let it fly and hope for the best is a good attitude given the TYPE OF ERROR - as explained above**
Now that you have noticed the error, write it down and what the new outcome would be had the error been captured. Then formulate this in a correction to thesis (ERRATA) section as an appendment, and then speak to your ex-supervisor to get the course of action needed to fix the submission.
Doing this is a better course of action, especially if you're planning a post-doc position that relies on credibility and honesty of academic research. period!
Upvotes: 0
|
2023/02/20
| 2,436
| 10,651
|
<issue_start>username_0: I am reviewing the MsC. dissertation of a friend of mine before it is submitted to the evaluation board. I started looking for formatting errors, badly written parts, malformed citations, typos, orphaned references, factual errors, logical gaps, and other stuff like that. One of the authors that she most cite is, of course, her advisor.
However, when I checked all that stuff, I found that the works of her advisor are very sloppy, containing a lot of errors, typos, incomplete sentences, misnumbered pages, misnumbered sections, inconsistent spacing and margins, mixing text with two different fonts in the same paragraph for no reason, misplaced images obscuring parts of the text and even sentences that were left painted with red or yellow to be reviewed before publication, but that were published anyway without further review.
The worse is that her advisor works contains a lot of malformed citations. Those includes things like: typos in cited authors' names; opening quotation marks without closing them (so I can't tell for sure where the quote ends); closing quotation marks that didn't open anywhere; quoted parts that differ significantly from the quoted work when I double check; citations from works that do not show up in the references section; cited text parts that don't appear anywhere in the cited work (maybe misquoted something from paper A as being from paper B); papers in the reference section that aren't cited anywhere in the text; quoting things in the middle of the text without telling who or what is being quoted; telling that someone is being cited without actually providing a citation, etc. It is so sloppy, that it certainly even contains a lot of unintended plagiarism not due to malice, but due to the severe lack of competence and discipline on properly quoting stuff and on editing, formatting and reviewing documents.
Luckily, this is not my work and I don't have any relation with her advisor. Also, my job is reviewing her paper, not her advisor's papers. But, if I were in her place, what should I do if I am obliged to cite a lot of very low-quality work? Trying to argue or denounce or whatever with the advisor is probably just a way to ensure being dropped from the program, specially near the scheduled date for the deadline when the text is in its final development steps. Also, it is likely that this might cause some trouble inside the evaluation board that if composed by competent people would likely perceive all of those (since I did, surely could them). Further, even after the degree is received, the fact that the Master's dissertation is based in so much low-quality works is likely to cause problems further down the road. What should I do?<issue_comment>username_1: Your question is very dependent on the locale and the time.
Over the last decade, in most of the world, the formal standards for scholarly work have increased a lot. People who are now reaching retirement age published in an era where there was no word-processing for the general public and students graduating even with a doctorate routinely employed other people to write up their thesis. Their advisors often had their thesis set, an expensive and cumbersome procedure, and there is at least one famous dissertation (or a dissertation by a famous mathematician) that was so marred by typos that its significance was only recognized later. Even if authors typed their own papers using something like [troff](https://en.wikipedia.org/wiki/Troff), they made many mistakes. You can still see this in LaTeX-set papers where many people still do not know how to put text components into an equation.
The expectation on the quality of citations has also increased significantly. Thirty years ago in computer science, we used a lot of abbreviations and because there was not a lot of information on the Internet, we sometimes cited by (short-term) memory and got things wrong.
The acceptance and availability of modern electronic tools also varied quite a bit according to countries. In addition, the quality of work has increased a lot in certain countries over the last couple of decades.
Regarding your question: First make sure that you are applying contemporary standards to contemporary works. Only if the advisor's current papers are full of problems of the type you are describing, then maybe taking an action would make sense.
Second, what do you expect as the outcome of a complaint? Do you want the titles of the advisor nullified or do you want her to be fined? Presumably, the advisor is subject to evaluation and their evaluators will look at current publications and react to them. What you describe is only feasible in low-quality outlets for which no or very small credit would be given.
Third, your friend is stuck with the advisor. Is your friend really basing their results on bad papers or rather on poorly written papers? If your friend uses previous work, then your friend is held responsible for a critical reading of this previous work, independent on how well or poorly it is written.
Fourth, if your friend is forced to use invalid or shoddy work as a foundation of the friend's thesis, then indeed her degree might not be worth a lot. In this case, and only in this case, your friend should reconsider the relationship, even if it means graduating later. However, it might just be best for your friend to get the degree "and run". In industry or government, the title might be the only things that counts.
Upvotes: 5 [selected_answer]<issue_comment>username_2: You are making too much of this. Writers are not responsible for the typesetting or grammar in the papers that they cite. Citing a paper does not mean endorsing all aspects of it. Presumably your friend is citing her advisor because her new work builds on the advisor's previous work; so long as the parts being built upon are technically sound, and your friend's paper is itself well written, there is really nothing to do here.
Upvotes: 6 <issue_comment>username_3: You only need to worry about the quality of the advisor's work to the extent that it impacts on the quality of your friend's work. For example, if the advisor gives an ambiguous quotation from another paper (without closing quotes to indicate where the quote ends), that is only going to matter if your friend also uses that quote, in which case your friend should go back to the primary source to use the corrected quotation (citing the original source of the quote). If there are mis-numbered pages that affect the ability to clearly cite the source then your friend should note this problem for the reader in their citations (e.g., with a footnote) so that the reader can find the relevant pages for the citation.
In cases where the cited work of the advisor contains an error relevant to the point being cited, your friend should note the error and act accordingly (e.g., correcting spelling or grammatical mistakes with appropriate notation of changes). As a reviewer of your friend's work you can point out instances where you think the material being cited is erroneous and suggest appropriate changes. However, it is not necessary to point out errors in cited work that do not impact on the work you are reviewing. Many of the issues you raise with the cited work (e.g., aesthetic problems with fonts, etc.) are not things that are going to impact the other paper.
From what you have described, it sounds like some of the works being cited are working papers, rather than properly peer reviewed work. (I would certainly hope that some of those problems would not appear in a final version published in a journal.) In such a case the bibliography will make clear which references are working papers. It is of course possible that the level of mistakes in the cited work is so severe that you might legitimately question whether that work is trustworthy to cite at all. If you form the view that it is not then you can make this known in your review, though your friend is probably constrained by the fact that he is working with this advisor.
Upvotes: 4 <issue_comment>username_4: Almost none of this matters. To my mind, there are three reasons that you cite something:
(1) To reference specific facts established in that work.
(2) To acknowledge the precedence of previous research in the area.
(3) To point the reader to other sources in the area which might interest them.
For (1), if the source establishes your fact correctly, it basically doesn't matter if the typesetting, formatting or grammar are unclear. If there are material errors in the source, then I would indicate how to fix them in as non-judgmental language as I can.
For example: "See SOURCE for the complete table of X's. Note the typo in the third line of the table: The 2 should be a 3," or "The equivalence of conditions Y and Z is Theorem 1 in SOURCE, but it appears to us that the authors of SOURCE only prove one direction of the implication, so we provide a proof of the converse." If the badly written material is brief and easy to write well, you can also do "Theorem 1 was proved by SOURCE; since the proof is short, we include it for completeness." Note that there is no need to directly criticize SOURCE in the last example.
If you need to directly quote material which has a typo, you can use [sic](https://en.wikipedia.org/wiki/Sic) or square brackets to fix it, but I would consult with your advisor about how to do this without seeming rude; I think this is something where norms differ from field to field.
For (2), I think this doesn't matter at all. "This research area was pioneered by SOURCE" -- if SOURCE is badly written, this sentence is still true and you have done your job acknowledging precedence.
For (3), this can be a bit awkward. If I am pointing the reader to important earlier sources or surveys, then I don't want to leave out major papers, but I also want to give them sources which they will find useful. I usually opt for praising rather than criticizing: "For background on TOPIC, see SOURCE1, SOURCE2 or SOURCE3; the author particularly recommends the clear exposition of SOURCE3."
Upvotes: 3 <issue_comment>username_5: >
> I am reviewing the MsC. dissertation of a friend of mine
>
>
>
You did not mention the country, but at least in Europe an MSc dissertation is a formality and there is nothing (or almost nothing) new in it.
You have to write it, and then forget it. Nobody cares about details and the reviewers have several of them to go through (which, otoh, means that a neat, readable dissertation will help). But nobody is going to check details such as the citations contents.
Upvotes: 2
|
2023/02/20
| 1,079
| 3,990
|
<issue_start>username_0: I'm writing my thesis. For example, this paper on arxiv.
<https://arxiv.org/abs/1409.0473>
I'm using Overleaf. I used `Export Bibtex Citation` on arxiv and paste the information into the `.bib` file. The result is shown as in the graph.
But my prof says I should add the detailed information of the references such as pages, number, and volume.
I'm very confused, how should I do this? Are this information contained in the Bibtex Citation from arxiv?
[](https://i.stack.imgur.com/CEW1d.png)
The exported information is
```
@misc{https://doi.org/10.48550/arxiv.1409.0473,
doi = {10.48550/ARXIV.1409.0473},
url = {https://arxiv.org/abs/1409.0473},
author = {<NAME> <NAME> <NAME>},
keywords = {Computation and Language (cs.CL), Machine Learning (cs.LG), Neural and Evolutionary Computing (cs.NE), Machine Learning (stat.ML), FOS: Computer and information sciences, FOS: Computer and information sciences},
title = {Neural Machine Translation by Jointly Learning to Align and Translate},
publisher = {arXiv},
year = {2014},
copyright = {arXiv.org perpetual, non-exclusive license}
}
```<issue_comment>username_1: Maybe your boss is not aware that your article is available in a standard way / in a journal?
From the link you provided I exported the citation the way GoogleScholar suggests it, I think in this way it becomes clear that this is an arXiv article.
```
@article{bahdanau2014neural,
title={Neural machine translation by jointly learning to align and translate},
author={<NAME> <NAME>yun and <NAME>},
journal={arXiv preprint arXiv:1409.0473},
year={2014}
}
```
Keep in mind: What information is being displayed in your references of the compiled PDF in what style/format depends on your bibtex settings. I think in your file bibtex items of class "misc" are simply not displayed with all the information you want to see.
Upvotes: 1 <issue_comment>username_2: TLDR: You should include detailed information about the paper you cite, and you have not done this. This is why your prof is confused.
---
If the screenshot is how the citation looks in your thesis, then it's obvious why your advisor is confused.
You have shown only the authors' names, title, and year. Anybody reading your thesis will be confused how they are expected to search for the paper, and how they can be sure that they found the same paper that you meant.
This is why, when citing a journal article, you include not only authors' names, title and year, but also journal title, volume, number, and preferrably pages. This will give readers a very clear indication which work has been cited.
If you cite a paper from arXiv, you should include information that will explicitly say that this is an arXiv preprint, and give readers some obvious way to find the paper on arXiv. An arXiv url perfectly fits this purpose, or alternatively you can use any wording that will convey the same information, e.g. "digital preprint, arXiv #1409.0473", or something similar. Your prof should be able to tell which style is preferrable for your thesis.
Note that arXiv url is already present in your bibtex record, so it is just a matter of TeX or BibTeX configuration to have it shown in the citation. My TeXfu is rusty, but I remember tinkering a lot with configs to show the citations exactly as I see fit in my thesis some 10 years ago.
And regarding the comments by your prof, most probably they thought that this is a citation from a journal, and therefore asked to include the said information. Without anything indicating that this is an arXiv publication, it is quite natural to think that this is a journal citation. If you tell them that this is an arXiv paper, they most probably would not require journal name etc., but would require an arXiv url or similar information as I have explained above.
Upvotes: 2
|
2023/02/20
| 814
| 3,330
|
<issue_start>username_0: I have modified scientific open-source code licensed with the GPLv3 license. If I run some simulations on my PC with this modified code and later on I want to publish a paper with the results of my simulations, would this be considered "conveying" according to the GNUv3 definition (and therefore triggering the obligation to share the code under the same license)?
I am all for open source and sharing the work but as I will be having this discussion with my PhD advisor I want to know for certain which are our obligations or not.<issue_comment>username_1: No, the output of a GPL-covered program is not covered by the same license as the source code, apart from very special cases.
* Reference 1: [In what cases is the output of a GPL program covered by the GPL too?](https://www.gnu.org/licenses/gpl-faq.html#WhatCaseIsOutputGPL)
* Reference 2: [Is the output of an open source program licensed the same?](https://opensource.stackexchange.com/questions/5478/is-the-output-of-an-open-source-program-licensed-the-same)
Upvotes: 6 [selected_answer]<issue_comment>username_2: I'm not a lawyer, etc.
I've simply never heard of this being a problem or heard it suggested as a consideration.
The work you obtained under the terms of GPLv3 was the software you used. Your paper is not a combined work, derived work, what have you of that software, as long as the paper doesn't actually include code from that software. Any software you made based on the GPL code is those things.
If you write a paper reporting on results from software, but no code, you have no obligation related to the GPL regarding the paper. If you distribute your code separately to be checked, which as a subjective point I think you should as a matter of basic academic integrity, then that *code* is definitely covered under GPLv3 and you have the corresponding obligation to license it under those terms. I prefer not to license code via GPL myself but have had to for a few projects because of this dynamic.
Upvotes: 3 <issue_comment>username_3: I'm in agreement with the [answer of username_1](https://academia.stackexchange.com/a/193622/75368) and [the answer of Anonymous M](https://academia.stackexchange.com/a/193623/75368), but point out that you need to either keep the modified code to yourself or have the license extended to it.
While it is fine, theoretically, to keep the code to yourself, it might make any external verification of your results difficult and possibly call them in to question.
For results, in these times, that depend on particular computer code, it is frowned upon to keep that code confidential. How serious that "frowning" can be depends on the importance and significance of the results. So people are "encouraged" to publish code.
Your choice, of course, but be aware of the possible future effect.
There is a subtlety here, of course. The real question isn't whether your results trigger GPL, but whether the GPL license requires that you actually *publish* any and all modifications of GPL code, or can a person/entity make a modification and not publish it at all. If they do publish it is clearly GPLd, and a possible interpretation is that it is even if not published, but that would be moot. After all, you wouldn't be "licensing" the code to others if it is held close.
Upvotes: 4
|
2023/02/20
| 535
| 2,356
|
<issue_start>username_0: I was applying for postdoctoral position in a reputed lab in a foreign country and they have asked for recommendation from my PhD supervisor. Although my PhD supervisor has given an informal recommendation by email, he has not given formal recommendation yet.
When I asked, he said that our country has some diplomatic problems with the host country (nothing serious in reality), so he cannot give a formal recommendation unless he knows the exact potential job description in that country.
Now, if I have to ask the host lab, I cannot tell all these 'diplomatic' stuff (which I also believe is pointless and my supervisor is just being paranoid). So, is it good if I just ask them the exact project I would be working on and tell them that my PhD supervisor will only give recommendations after getting such information?<issue_comment>username_1: Actually, I think *you* need to see an "exact potential job description" before you even consider a job. If the employer is unwilling to give that then there are all sorts of red flags, especially for foreign nationals. I suggest you ask for a complete description and see no reason why you shouldn't share it with your advisor.
Perhaps they are being paranoid, but perhaps they are trying to either protect you or to write the most appropriate letter that they can. Such knowledge is useful to both you and your advisor.
Upvotes: 4 <issue_comment>username_2: Here's my general approach to dealing with these sorts of situations:
You are interacting with two people whose intentions you do not fully understand and with whom you apparently don't feel at ease just being honest with. (For example, you don't want to tell the host lab person about the 'diplomatic' stuff.) Let them figure it out themselves in an email to everyone:
>
> Dear Lab Head and Host Person,
> I asked Lab Head about a letter of recommendation, and he suggested first seeing a complete job description in order to tailor his letter to the purpose. @LabHead: Is this what you were asking for? @HostPerson: Would you be so kind to provide what LabHead is looking for?
> Sincerely
> Possible future postdoc
>
>
>
My reading from your post is that you are confused what everyone wants and why. But these people know, and they've got experience with how these things work. Let them sort it out!
Upvotes: 3
|
2023/02/20
| 2,061
| 8,295
|
<issue_start>username_0: Is there a "statute of limitations" on plagiarism? After being accused in one course, can my work in past courses be re-examined? I believe I unknowingly plagiarized in a past course.<issue_comment>username_1: Actually, anyone reading your work can accuse you of plagiarism if it is noticed that you copy the ideas of others without attribution.
Ordering an investigation, however, is a different matter and would depend on local rules and customs. But universities like to protect their own reputation and so some "due diligence" is required. This would be especially true for graduate degrees, I'd expect.
But once you have been accused and not exonerated, then it would be natural for people to look for other instances. In a serious case, such as a thesis, you would bear considerable risk.
Edited to add: No, there isn't a time limit or statute of limitations. No statutes at all in most places. It is what it is unless corrected. If the work can be seen you can be accused. This is distinct, of course, from local disciplinary rules.
Upvotes: 5 <issue_comment>username_2: Anyone might accuse you at any time, even decades or centuries past your death. However, the burden of the proof is to the accuser.
Accusing someone of plagiarism is a serious issue, and normally, the people who does the accusation have proofs sustaining the accusations.
Once accused, you are expected to defend from the accusation. If you are actually guilty, it is likely that there is nothing you can do to defend yourself and the best you can do is to apologize and recognize your error. However, the academia tends to be unforgiving and ban you forever from working on anything seriously even if you actually apologize.
If you are innocent, defending from a plagiarism accusation is hard. There are a lot of cases where innocent people had no real chance to defend themselves from unfair accusations of this type. What you should try to do is to prove that there was no plagiarism at all and that you properly quoted everyone you should quote, which is a very tiring, stressful and cumbersome process.
Sometimes, someone else plagiarized your work and the accuser thought it was the other way around. In this case, you should accuse that someone else as part of your defense.
Sometimes a defendant defends by counter-attacking and plays the "uno reversal card": The accuser is actually who is plagiarizing work! If the defendant is innocent, this is probably the case, but proofing it might be hard. If the defendant is guilty, this would be part of their strategy to steal other people work.
In cases of a work with several authors that is accused of plagiarism, this might create a serious issue as possibly, only one of the authors did the plagiarism and the others were simply trusting him. If the other authors aren't able to clear themselves, everyone would be considered guilty.
It seems that you previously plagiarized stuff as part of a work done in the context of a previous discipline. If you are a grad student and plagiarized due to not actually knowing how to properly quote stuff or as a result of a very sloppy work, the teacher, while still giving an F or a zero to your work, might then just choose to rub that under the rug. I once caught a student who made exactly that and what I told him was simply "quote all those things correctly and resubmit your work as soon as possible, otherwise there is no other grade than a zero", I also gave him a 10-minute informal lecture about how to properly quote stuff. He did fixed his work and resubmitted, which is a much better outcome than just ruining his academic life.
Upvotes: 3 <issue_comment>username_3: I believe that the process varies widely by institution and country. In my experience working for a public university in the United States, I needed to file a report for each student accused of plagiarism with the university. I need to provide documentation of plagiarism, an explanation of why this constituted plagiarism, and course materials about the assignment in question. After this process, the university would contact the student to see if they wanted to dispute the accusation or accept a failing grade, with a requirement to take a remedial course regarding plagiarism in order to graduate. This was a thorough, time-consuming process so I did not do it unless it was a clear violation of plagiarism. For students who seemed to have issues with citations, for instance, I would give them a warning that they needed to follow proper rules or else they would be reported to the university. Again, this process can vary widely, but in my instance as a professor, I only reported plagiarism that occurred within the classes I taught while the classes were in session. Unfortunately, I usually report two to three students for plagiarism each year.
Upvotes: 3 <issue_comment>username_4: At my institution each department has a plagiarism officer, who deals with cases reported by faculty according to the procedures set out in university regulations. I am aware of at least one case where the plagiarism officer conducted a review of a students previous work because of the nature of their offense (after it had been investigated and punished appropriately). So retrospective investigations can and do happen, if the institutions regulations allow it (and there is good reason).
Upvotes: 1 <issue_comment>username_5: There is certainly no accepted “status of limitations”. Researchers now have old papers retracted because plagiarism is discovered by other unrelated academics sometimes years after the fact.
As an example [a German rabbi has recently been found to have plagiarized part of his 1992 dissertation](https://www.jta.org/2023/02/16/global/german-rabbi-at-center-of-ongoing-scandal-plagiarized-his-dissertation-newspaper-finds). This is not a university example but you can find a large number of miscellaneous examples by searching for [the keyword “plagiarism”](https://retractionwatch.com/?s=plagiarism) on the [Retraction Watch website](https://retractionwatch.com/) and find those related to university, including some cases where degrees were withdrawn, as this case of [a US Senator](https://www.nytimes.com/2014/10/11/us/politics/plagiarism-costs-degree-for-senator-john-walsh.html?_r=0).
So: it does happen that people look back on old work, and it does happen that degrees are revoked as a result of plagiarism.
Upvotes: 2 <issue_comment>username_6: There are no real rules, but as you allude, things can get fairly complex, as there are two fairly conflicting goals of Academic Honesty procedures in US schools.
One goal is to defend the value of the diploma from a given school. If employers figure out that a school graduates a large number of cheaters, they will (appropriately??) weight the value of a diploma from that school in a downward direction. Toward that goal, finding every instance of dishonesty in a student's history at that school may be a valuable exercise.
The other goal would be to modify the behavior of a student caught cheating, providing the opportunity of learning from the experience and learning how not to cheat. With that in mind, if one delves too deeply into past behavior, one might find a veritable well of plagiarism, bringing the level of penalty up to permanent separation, thus removing the pathway of modifying behavior.
These goals may well conflict. The ideal medium would be to identify cheating every time and immediately upon the event, but that really can't happen.
Practically, it would be difficult to gather all of a student's past (and passed!) assignments and put them under a microscope. It's also a big ask for faculty members.
Note that for *faculty* caught plagiarizing, it's fairly likely that their life work will be put under a microscope, that that's probably appropriate.
Upvotes: 2 <issue_comment>username_7: In Germany in 2013, 33 years after writing her doctoral thesis, the education minister <NAME> was stripped of here doctorate by her <NAME>ater, Heinrich Heine University in Düsseldorf. She subsequently [resigned her post](https://www.bbc.com/news/world-europe-21395102). Your country may differ but educational institutions and employers can have long memories in this regard.
Upvotes: 2
|
2023/02/20
| 1,356
| 5,653
|
<issue_start>username_0: About 4 years ago, before I started my PhD, I published a paper which used a method to improve an existing result. There were 2 papers on the existing result and method which I used for my work and appropriately cited. A couple of days ago I discovered another paper on arXiV by the same authors, which dealt with the same problem, which I did not cite for reasons I can't remember. Also, I realize that they address the same problem that I worked on, in a small sub-section of their arXiV paper.
Based on the references that I used, I gathered that the problem I solved had not been looked into in the works of theirs that I cited and made a statement about this fact. Although, the method that I have used is different from theirs, it still makes my statement erroneous. I feel that I am thoroughly screwed and the damage is irreparable. How do I go about resolving this issue? I am willing to let everybody involved on my paper know about it, including the authors whose papers that I have not cited. Is there anything more to be done?<issue_comment>username_1: It's difficult to tell precisely as a bystander, but this just doesn't sound like a particularly serious inaccuracy. From your description, your previous publication make no claims about the other authors' paper you've more recently found.
>
> I realize that they address the same problem that I worked on, in a small sub-section of their arXiV paper.
>
>
>
So you missed a small sub-section of a preprint in your literature search. This type of thing happens all the time, at least in my part of math. Yes, you could wish that you had spotted it, but it sounds like your method was still a novel approach to the problem. You're certainly not "screwed" with "irreparable damage" to the paper.
If you like, in the future you can add a bit of hedging to novelty statements going forward. I sometimes include this kind of language:
>
> To the best of the authors' knowledge, the method we will discuss here is the first to solve the problem of fizzing buzzes.
>
>
>
In the grand scheme of things it actually doesn't matter, except in acknowledging that literature searches are hard and that you haven't intentionally neglected something a reader or reviewer happens to know about.
Upvotes: 4 <issue_comment>username_2: I am assuming that when you say *published*, you mean *published in a peer-reviewed journal*. On the other hand, the paper you found was not published. It was on ArXiv. I argue that because of this, you should not feed bad at all. You weren't scooped, since that paper wasn't published.
ArXiv is a non-archival venue. There is no peer-review process, and some academics (myself included) are extremely wary of citing ArXiv papers that haven't been published elsewhere. If they were published elsewhere, I cite that version, not the ArXiv version. The main purpose of ArXiv in my opinion (which is how I use it), is to make your work accessible to a broad audience. Some folks treat their ArXiv papers as published, and expect to receive academic credit for them. I disagree with this viewpoint on a moral/ethical basis, as it undermines the entire purpose of the academic publishing process.
Now, let's pretend that the paper was published and that you simply missed it. Is it truly identical? It seems very unlikely that the result, method, techniques, *everything* is the same.
I wouldn't worry about it. Let me tell you a story that happened to me not too long ago. We had a result (call it A) that we wanted to publish. When we were in the process of submitting it, a paper with a result B was "published" on ArXiv. Result B generalized our result. However, given the argument I made above, we went ahead and submitted our work, and it got accepted. Later that year, *another* result C, was published in another venue, which was a special case of our work. They were all valuable contributions in their own way, even though the only one that "mattered" was Result B (the most general one). Joke's on them though - we published an even more general result earlier this year!
This is how science works. We build on each other's work, and if a topic is interesting, you'll have a lot of duplications. As long as you're operating in good faith, citing sources appropriately and acknowledging effort, you're doing it right.
Upvotes: 3 <issue_comment>username_3: Personally, while not a huge issue, I do feel like other answers are downplaying this a little more than it should be. In some sense it has become a chronic issue, to the point where I feel more than half of the results I see have been done before and few people in the field seem to have any idea that it was done before. Reviewers don't notice, colleagues are excitedly talking about the novelty of the approach, etc... and I'm like, "but xxxx did this back in the 30s." The harm isn't in this happening once, it is that over time entire fields seem to suffer a collective loss of memory.
So while I wouldn't beat yourself up about it, it might be worth publishing an erratum as a service to those that come after you trying to determine the novelty of their own work (they might be reading your work and not have seen the other one yet, and then may never read the other work). Of course with the caveats that 1) the journal you published in is amenable to such minor corrections and 2) your field and coauthors are similarly amenable to the idea. Personally, I respect authors willing to correct their works more than those that do not, but I admit that I know others that view any sort of correction, no matter how minor, as a black mark on the work.
Upvotes: 2
|
2023/02/20
| 610
| 2,654
|
<issue_start>username_0: I have an interview scheduled for tomorrow for a faculty position. However, today my flight got canceled. I tried to get another flight today, but unfortunately I could not.
I informed the university and requested to do my interview via Zoom, to which they agreed.
I booked the flight and hotel two weeks ago and the university had informed me they would reimburse me. Because the flight got canceled, I will get a refund from the airline. However, I will not get any refund from the hotel (the hotel price was almost $400). The university had given me 3-4 options to choose from for the hotel, each of which was a 4/5 star hotel, so it was expensive.
Now the question is: can I ask for a reimbursement for the hotel from the University? Will it be a good gesture?<issue_comment>username_1: I think it is your responsibility to try to get any possible refund for the hotel (that is, contact the hotel and discuss your circumstances, even if their cancellation policy would seem to not allow a refund). Then, it's the responsibility of the institution to still reimburse you as if you had traveled. That assumes that the cause for the cancellation is entirely outside your control, which in this case it sounds like it is.
This is one good reason that institutions should pay these costs themselves up-front rather than reimbursing interviewees afterwards.
It is possible that some arcane rules (especially if you are dealing with a government-affiliated institution) will somehow prevent you from being reimbursed, but even in that case there is nothing wrong with making the request. If anyone holds it against you for asking, that may not be a place you want to work anyways.
Upvotes: 6 <issue_comment>username_2: When the institution effectively forced you to choose a high-price hotel, they were clearly willing to foot the bill and had already budgeted for it. It would be extremely unreasonable for them to force you to pay for something you never expected and might not be able to afford, or would have chosen a less expensive option if it came out of your own pocket.
So you should definitely request reimbursement, and I think they're likely to provide it. In the very least, you should insist on the difference between the high price you're actually being charged and a cheap hotel you might have chosen if you knew you'd have to cover the bill yourself.
Furthermore, what's the worst that could happen? If they say no, you're no worse off than you currently are. If you think this might color their hiring decision, I think that's unlikely. That decision comes from academics, not the bean counters.
Upvotes: 2
|
2023/02/21
| 1,248
| 5,493
|
<issue_start>username_0: Previously I was working with professor A. During that time I feel lost: progress moved much slower than I expected and it's quite difficult to learn new skills required in the lab, which cannot intrigue my interest and push me to read paper at all. Also, lab environment makes me feel uncomfortable sometimes and there's a lack of funding in the lab currently. So I thought I may still need to tolerate these for one year until I formally start my project.
Based on these, I left the group and chose to work with another professor B with enough funding. And I can also start my project immediately. However, different accidents occurred that I forgot to take them into account before I made the decision. One of the most serious consequences is that I screwed up friendship with others. I regret my decision a lot now. This causes me in serious depression.
Was I too hasty and short-sighted then? Only considering to start my project as soon as possible while ignoring almost everything else. I didn't talk to my advisor in depth about these and members in the group.
I can't stop myself thinking on these everyday recently and can't focus on work. I'm always thinking if I persist on staying there for more time, it may be better. I don't know how to solve my issue. I think one of the biggest problem of me is that I did not evaluate myself carefully and correctly. A large part of decision was based on my illusion rather than facts.
Recently I searched online about changing advisor, and I found most of the reason people change is because their advisors are terrible. But this is not the case for me. Both advisors are nice and I'm interested in the research topic. The previous advisor has more communication with students. I'm even thinking go back to the previous advisor, but this may not be a good choice.
I don't know what I can do. Can anyone help?<issue_comment>username_1: First, feeling bad for a bit because you think you made a wrong decision (darn, I bought the Ferrari, when I shoulda, coulda, got the Lambo) isn't the same thing as [Clinical Depression](https://en.wikipedia.org/wiki/Major_depressive_disorder). For the former, you just work out a solution that fits, perhaps at some cost.
For the latter, however, professional medical help is probably warranted.
In either case, however, it is probably worth seeking at least short term advice from a professional to get a diagnosis and some suggestions for improvement even for the "simple" case.
For the simple case (Ferrari/Lambo) I can offer a few suggestions.
Sit down, perhaps with your significant other if appropriate. Make a list of your goals. Make a list of your options. For each option list how the contribute or not to your goals. For each, try to come up with alternative actions that help meet the goals better.
Don't assume that you have to get it perfectly right. Remember that you will have future decision points and opportunities to make directional changes in your career. Don't rule out big changes if they are feasible and support your goals. Don't assume you are stuck.
See who around you, both personal and professional can help you achieve your goals. Perhaps from another professor can help. Talk to those who have mentored you, perhaps.
Upvotes: 2 <issue_comment>username_2: I think your depression will fade if you take action on the problems. If I understand you correctly, you have two problems: which advisor you should work with, and what happened with your friends. For the first problem you can talk to the advisors directly (one or the other or both), or to some other trusted confidant, like another professor. As username_1 said, knowing what your goals are is a good starting point. Discuss your goals with the advisors or a another professor.
For the second problem, talk directly with your friends. Tell them you sense tension between you and you want to understand their point of view and explain why you did what you did.
My sense is that your depression is caused by perseverating on negative thoughts (Who is displeased with me? What are they thinking? What will the consequences be?) inside your head instead of taking action to solve the problems. I believe that both of your problems are solvable by thinking them through logically and talking with the people involved.
You didn't do what you did out of ill will. Many or most people make wrong decisions even when they do think carefully. Perhaps they miss some key fact or they lack some important insight that they only realize later. This is human. When that happens, you need to forgive yourself quickly and immediately move to problem-solving mode. List the possible solutions and start working on them. If it turns out that you offended someone, apologize and tell them you want to restore the relationship. The faster you take action, the faster these problems will be behind you. Don't allow your perseverating brain to trap you in inaction.
Some strategies that can help your depression: (1) You're stuck in a loop of repeating negative thoughts. Visualize yourself stepping out of the loop. (2) Don't believe everything your anxious and depressed mind is telling you. Shift to the part of your mind that knows that almost all problems are solvable! Say to yourself, "These are only feelings. They're not reality. I can take action to change my reality."
The problems you've described are solvable. I believe they will be behind you within the next two weeks.
Upvotes: 2
|
2023/02/21
| 1,033
| 4,383
|
<issue_start>username_0: I got a job offer from a UK university as a lecturer. I accepted the written offer and started start-up package negotiations.
I did not get anything cause it is a small university so I had a counteroffer to get my salary higher but of course, it did not work out and got a response that this is a very late time to negotiate salary.
Is there something else I should negotiate at this point? or just accept it?
Note for those who ask:
I received advice from one of the directors at the university that they do not provide a start-up package so I did not negotiate and accepted the offer. then I had a meeting with the dean and he asked me to provide him with what I needed. I provided him with a list of things I need and I get a response from the dean that it is very late to negotiate these things.<issue_comment>username_1: Negotiation is a very cold game: if you are not willing to walk away, then there is no negotiation and you will just have to accept whatever the other side is offering.
It can still pay to talk: You may be able to get things that are cheap to the employer but valuable to you. For example the university may have a reserved number of spots at a child care facility and these represent a "sunk cost" for the university. Or you might be able to convince them to invest in you, e.g. allow you to teach fewer courses in the first couple of semesters, so you can develop them as good courses. As a consequence after those initial semesters they have a lecturer with a portfolio of quality courses.
As to your question: If the response came from the university level, then you are just one of many. They will have seen these attempts at negotiation many many times before, and will have forgotten your very existence the moment they pushed the sent button and moved on to the next thing that needs to be done. This negotiation is important to you, but for them it is just item 367 of the 8,563 items that need to be done that day. So don't overthink your response. Keep it short, to the point, and polite.
Upvotes: 4 <issue_comment>username_2: >
> I accepted the written offer and started start-up package negotiations
>
>
>
Caveat that I'm not familiar with how things work specifically in the UK so maybe there is some difference there, but from a US perspective at least, a negotiation is about getting to an offer you can accept. Your only leverage in a negotiation over a job is to say or imply "I will not accept this offer; here are things you could change in the offer that would make me accept it". From there, you might be offered what you asked for, or there might be further counter-offers or discussions leading to a mutual agreeable offer, or else one of the parties walks away.
You accepted the offer, so your negotiating is done. From the perspective of the university, you've agreed to the terms they stated. It feels odd to them that now, after having accepted this offer, you've asked for more. They may wonder how many more times in the future you will ask for something additional that they can't provide.
It seems there are some communication barriers between you and the dean/directors: when they've asked what you "need", they clearly had something different in mind than you. I can't begin to guess, so you should probably ask them to clarify.
Again, things may differ in the UK, but in the US I would interpret a position without available start-up funds to be effectively a teaching-only position, where the institution doesn't have interest (at least not the sort of interest that will separate them from their money) in your research program, but is rather hiring you primarily to teach. Alternatively, they view you as sufficiently advanced in your career that you are not in need of start-up funds as you are expected to have or obtain extramural grant funding.
Upvotes: 3 <issue_comment>username_3: In the UK there is typically no negotiations for the entry-level academic posts, such as Lecturer. You may be able to negotiate you starting salary a bit (one or two spinal points) but that's about it.
There is nothing wrong in trying, as any University may be able and willing to make exceptions to attract exceptional candidates. But if you are not prepared to walk away, you do not have much leverage for the talks and are at disadvantage.
Welcome to UK adademia and good luck!
Upvotes: 3
|
2023/02/21
| 1,374
| 5,085
|
<issue_start>username_0: I am a Postdoc, recently I submitted my first single-author publication (physics). I had one journal in mind and briefly discussed the idea with several seniors, and nobody really objected. Now I started to realize that the journal I targeted is most probably not perfectly matching the topic and on top of that is too high-ranking for my work, which is simply an improvement and extension of already existing methods.
I suppose that I might get rejected, but will this have negative impact for me? Should I contact the editor already now? Currently the article is in its second revision stage, and the anonymous reviewer was not too enthusiastic about my work in the first place (which probably is also due to the issue I am describing), but also I did not get rejected right away. My fear is that a low-impact article in a high-impact journal might shed a bad light on me and my possible future career. So should I interrupt the process, or wait and see what happens?<issue_comment>username_1: Let the journal's consideration take its course. Having a paper rejected has no negative impact for you. (Indeed, the rejection will be unknown to most people.) The *only* negative effect is the time lost while you wait. But this is outweighed by the benefit if the paper is accepted (even if that likelihood is small).
If your papers are always accepted by the first journal you send them to, then you are aiming too low.
Upvotes: 3 <issue_comment>username_2: You asked several questions in one.
1. I suppose that I might get rejected, but will this have negative impact for me?
2. Should I contact the editor already now?
3. ... should I interrupt the process, or wait and see what happens?
Starting with **#3**. You should keep calm and enjoy the process unfolds.
* A targeted journal need not perfectly match. Though there should be some matching. Hence why journals post their 'aim', 'scope', and 'about'.
* Another indicator of #match is alignment between your reference list, authors and journals.
Invariably **#3** answers **#2**. You don't have to contact the editor already.
Before addressing **#1**, seemly you have a **#4**.
* The 'fear' should be having ('low-impact) article(s) in shaky or low-impact journals. That is what would 'shed a bad light on' a possible future academic career.
* Having a shot at discipline's foremost high-ranking journal is a target every researcher should attempt. With the right mindset (well-conditioned 'thick skin'), there's much to be gained from the review process. Yes, some reviewers can be obnoxious and condescending. Looking beyond the (review) text and looking within the review text (reading in-between the line) is key to improving one's manuscript (and making adjustment to the research work where need be.
As to your **#1**, a rejection does not impact negatively.
In between, rejection is a fabric we all wear with pride as researchers. Fearing rejection as academia is fearing to excel.
Contribution to knowledge comes in many ways. *Novelty* is not only in 'new frontiers'. Novelty can come from 'improvement and extension of already existing methods'. The shift in the existing method might (go ahead to) make a groundbreaking impact on the discipline. Discussing 'novel/novelty' might take us to discuss #philosophy, which might distract us from your question. [For a few pointers to the philosophy of novel/novelty, see [James, 1979](https://books.google.co.za/books?id=-0NgRF-y0IIC); [Miller, 1950](https://doi.org/10.2307/2021254); [Cohen, 2017](https://doi.org/10.7554/eLife.28699)]. This quote from Cohen, 2017 might lead to other nuances of novelty. I would venture into those here.
>
> “(T)he primary novelty of this work is the ability to make a prediction about drug sensitivity. Reviewers felt that the predictive ability would be very hard to generalize, however, reducing the impact of this novel feature. This concern about novelty… was the driving factor in this decision.” -excerpt from a rejection letter received by the author
>
>
>
Novel/novelty might also lead us into the terrain of positivism, interpretivism, pragmatism, critical realism and the rest of research philosophy. It appears novel/novelty might have been part of the driver of <NAME> in his ferocious 'development' of critical realism.
Be it as it may, since your 'article is in its second revision stage', go through the process, develop strong rebuttal skills and insightful communication. You'll need them in your 'future career'. The world is your oyster.
Parting shot: Your research work has already caught the attention of the Editor. You've scaled through ***desk reject***. Don't abort!!!
[<NAME>. (1979). Some problems of philosophy (Vol. 7). Harvard University Press](https://books.google.co.za/books?id=-0NgRF-y0IIC)
[<NAME>. (1950). Novelty and Continuity. The Journal of Philosophy, 47(13), 369-378](https://doi.org/10.2307/2021254)
[<NAME>. (2017). How should novelty be valued in science?. Elife, 6, e28699](https://doi.org/10.7554/eLife.28699)
Upvotes: 2
|
2023/02/21
| 571
| 2,463
|
<issue_start>username_0: So I have started putting my (LaTeX generated, classic thesis .cls file) resume out there in job applications and this one question has been bugging me. Are Applicant Tracking Systems (ATS) able to read LaTeX generated resumes well? This question is inspired by [this post](https://www.reddit.com/r/cscareerquestions/comments/8gi9hl/psa_i_got_a_lot_more_interview_offers_with_a_docx/) on reddit.
From my understanding, ATS systems can be used to auto detect keywords in resumes that determine if the applicant makes it to later stages in the hiring process. However, when I upload my resume, the auto-complete feature (where the company's system extracts text from my resume to autofill the application) always does a terrible job of extracting the information. It makes me wonder if these systems have a hard time reading LaTeX generated resumes. Does anyone have experience with this?<issue_comment>username_1: If an ATS software tries to parse your application the process can be roughly separated into two steps.
First it needs to extract the raw text from your pdf file. Depending on whether your pdf file was generated by Acrobate Writer, MS Word, LaTeX or some other method the internal structure of the pdf might be somewhat different but any not completely terrible pdf reader will get the text out of the pdf no matter how the pdf was generated.
Second it needs to transform the raw text into useful information and 'understand' it. This is hard for a software and the formatting of the document (a table, running text, different text boxes) can make this more or less difficult, depending on the specific software. Any ATS software might fail on your CV here but that has nothing to do with the fact that you used LaTeX but rather that they tried to use software for a task that was to difficult for their software. If that happens making the same CV with MS word or some other software won't change anything, but changing the formatting might.
Upvotes: 1 <issue_comment>username_2: I think this might be a serious problem. Many students in STEM use latex and want to get a non-academic job.
Two issues I can think of are non-standard font encoding and the use of glyphs to get proper ligatures. My guess is \usepackage[T1]{fontenc} fixes the first and \input{glyphtounicode} fixes the second.
The safe thing is to bite the bullet and submit a Word document. To fit in. A sad state of affairs.
Upvotes: 3 [selected_answer]
|
2023/02/22
| 1,401
| 5,553
|
<issue_start>username_0: I have been an full time appointed research professor most of my career (until 2017) in UK and Australia. I still work in a small business as CEO working exclusively with academic clients. I also have one 'staff' university based appointment but at senior Lecturer level to do a particular part time role. I have also been appointed Visiting Professor at 3 UK universities and adjunct professor at one Australian university. All 4 of these are current and official. Can I use the term professor across all of my roles? Is there anywhere that verifies or precedents? Be glad of all and any advice.<issue_comment>username_1: I would look at it from the point of view of what happens if you get it wrong.
* If you choose *not* to use the professor title and someone finds out that you are entitled to use it, then pretty much nothing happens.
* If you choose to use the professor title and someone finds out that you are *not* entitled to use it, then that is embarrassing.
I was in a situation where it was unclear whether or not I could use a professor title. I chose not to, and a colleague in the same situation chose to use the title. A couple of years later, the situation was clarified and the colleague had to remove all mentions of her/him as a professor. Consensus among colleagues is that (s)he acted in good faith; the law was just unclear. Still it is embarrassing and I am very happy about my choice not to use the professor title.
Upvotes: 3 <issue_comment>username_2: Many Universities appoint Assistant Professors (~Lecturer in the UK) and Associate Professors (~Senior Lecturer in the UK), but if you hold such posts you normally do not put *Prof Surname* on your card, as this title is reserved for (full) Professors.
Similarly, a role of a Visiting Professor normally does not compare with actually being a Professor.
There are surely people working in industry who try to impress their peers with a fancy title, but it is quite easy to verify whether it is deserved or self-proclaimed.
Upvotes: 4 <issue_comment>username_3: The traditional etiquette is that Professor is a Rank or title, not a job. Once you have that rank, it stays with you for life, the same way military titles or royal honors stay with you for life.
The difficulty is that in recent years, universities have muddied the water by giving people job titles with the word professor in them, when the person themself is not a full professor in rank, in order to more closely mirror the situation in the US.
Prof is also a title that is entirely based on etiquette - it is not legally protected in any way, so in theory, any one is free to use it.
I would look at it it this way:
* You can still be a professor, even if you currently hold a contract as a Senior Lecturer at a university. This is not relevant.
* While you might call yourself a professor on the basis of a visiting professorship (and I know people who do), I think many people would look down on this.
* So really it comes down to the status of your "research professorship". I'm not really sure that I know what the status of a research professorship is, it's not really a title I've come across before. If you think this was a "full" professorship, then I would feel free to use the title. Things you might think about is whether you were entitled sit on the academic senate of your university? did you have an inauguration? Would you have been entitled to emeritus status if you had stayed? If it is something that you feel was equivalent, then I'd feel free to continue to use the title. If not I wouldn't. I'd note that there is no real right or wrong answer here, and it will be a judgement call.
If in doubt, its probably worth considering @MaartenBuis's advice that its better to undersell than oversell.
Upvotes: 4 <issue_comment>username_4: We are all forgetting the obvious solution to OP's "problem".
In UK/IRL you have two kinds of professor.
(1) A **departmental chair** - meaning an appointment to certain positions, e.g. Head of Department or any other endowed professorship within the department, which automatically carry an academic rank and salary of Professor and will entitle one to have Professor <NAME> printed on his/her door and stationary.
(2) A **special chair** - meaning an attainment of distinction (usually for research and/or teaching) over a period of 10 - 20 years within a field of their department and which entitle an attainee to use the *title* Professor before their name but which does not automatically give them any increase in academic rank or salary.
As OP has attained some distinction in his career, I am surprised that he/she did not "negotiate" a special chair for themselves when being appointed to Senior Lecturer rank in the UK university. As the SL job is the main one, use of Professor with their name would then be free of problems, I think.
So, OP, please have a word with your HoD at the main job.
Upvotes: 0 <issue_comment>username_5: It's not a protected title... <NAME> could call himself <NAME> if he wanted to.
Tradition would dictate that once you've held a *full* profressorship in any role, you keep the title for life regardless of what role you currently hold.
However, in my experience, using this title in a non-academic context would be a bit of a faux pas and just come off as pretentious. Most people with doctorates don't insist on being called <NAME> in a professional context anymore, they just use their name and that *is* an actual, protected, title.
Upvotes: 2
|
2023/02/22
| 1,130
| 4,817
|
<issue_start>username_0: I was born and educated in the UK but have spent the entirety of my career working in the US in government and higher education. I'm not, and never have been, a faculty member but rather part of the 'academic staff' supporting researchers with technical and domain specific knowledge. Due to changing personal circumstances, I've recently been considering returning to the UK but was surprised at the disparity in salaries between my role at a US institution and similar positions advertised in the UK.
I know this question has been asked before with regard to [academic salaries](https://academia.stackexchange.com/questions/182128/uk-faculty-salaries-are-much-lower-than-us-is-there-any-upside) but how do UK universities attract and retain staff? Even within the UK, it seems like there are much better paying options in industry than a university. At my current institution in the US, we have a hard enough time recruiting staff and our salaries are 2 to 3 times higher than those I have seen posted in Britain.
What strategies or incentives do British institutions use to recruit staff that offsets such low salaries? Or do they rely on individuals being mission and prestige oriented to choose a role in a University over a for-profit corporation?
Alternatively, am I just spoiled by higher US salaries and UK academic staff positions are actually well compensated given the prevailing wage in their region?<issue_comment>username_1: I think the answer to this very much depends on the sorts of role you are referring to.
If you are talking about positions such as Research Software Engineers, Scientific programmers, or Scientific Officers/Staff Scientists that might, or for example, running a core facility, a large part of the answer is "we don't". Firstly, we make far less use of this sort of people than many other countries do, often relying instead on that most English of ideas of the interested amateur muddling along as best they can. But when we do have such positions, we struggle to recruit to them and turnover is high.
If you are talking about research technician type roles then my experience is that people are either young (and don't stay long) or for the people that do these jobs, it's often the household's second income.
In all cases, it's getting increasingly difficult to recruit and retain good people.
Why do people do it? There is a combination of motivations. Partly, we do rely on the whole "vocation"/"mission oriented". Partly, we rely on the fact that for many specialties, industrial positions are pretty concentrated on a small (expensive) part of country, whereas academia is fairly well distributed and people have a variety of reasons for wanting to live away from where the industrial jobs are. Partly, I think people have a (misguided) idea that an academic job is more secure than an industrial one. Partly, it is the case that salaries for everything, from waitstaff to CEOs, are higher in the US. I don't know about academic staff, but the pay of a senior lecturer at a research-intensive Russell Group university is in the top 10% of salaries in the country, although this varies for other categories of university.
Upvotes: 5 <issue_comment>username_2: Partly it's because universities in the UK tend to have quite generous employee benefits. Annual leave is typically greater than average (40+ days in total for long serving staff isn't uncommon). Pensions traditionally have also been pretty good. For technical and administrative staff in particular, the work environment can be less demanding, working weeks can be as little as 35 hours for full time employees and there's generally not massive pressure to be constantly busy. There are also usually a lot of other perks and discounts to take advantage of if they so wish.
You also have to remember the UK generally has lower salaries than the US across all sectors. This is somewhat made up by things like socialised healthcare, higher sick pay and annual leave regulations. The cost of living is also generally lower.
Upvotes: 4 <issue_comment>username_3: I looked hard at returning to industry after my PhD, before going for postdoc roles.
I would have been on about ⅓ more salary in the private sector, and saved another 10% of my postdoc pay that goes on my commute. But financial motivation isn't everything; especially in academia plenty of people are content to make enough to afford a decent lifestyle, rather than trying to be rich.
The working environment would have been very different - academia suits a lot of people, though of course not everyone. There would have been a lot more travel and a lot less flexibility in the industry roles I looked at - not great with a young child. Having worked in both, the pressures are very different in industry.
Upvotes: 1
|
2023/02/23
| 766
| 3,218
|
<issue_start>username_0: I'm preparing a research statement for a permanent position in math, but the number of pages have a strict constraint of two. So it seems I'm unable to fit in the bibliographic references in the research statement. **Assuming that the the two pages *do* count the bibliographic references (correct me if I'm wrong!)**, I'm *linking* relevant research, instead of *citing* them using the
>
> \href{}{}
>
>
>
command in LaTeX: so for example, I'm writing this in my LaTeX:
>
> \href{website}{this result is proved}
>
>
>
Thus, when the reader clicks on "this result is proved", s/he will be automatically directed to the "website", that has the article in question.
Is this *okay*, or will it be taken as an *unusual practice*? It's difficult to fit in the research in only *two* pages, but I can't see how I can also fit in the bibliographic references in the same.<issue_comment>username_1: Usually the page limit includes the bibliography in a research statement or proposal (except if explicitly stated otherwise). Keep in mind that people, especially if they want to focus on something for longer than 20 minutes, still like to print things before they read it! So a link will not necessarily work for your reviewer.
It is not unusual to link things in a proposal, but rather as an extra feature instead of the only way to access the reference. You do not know, if you referee even has access to your original PDF!
What I have seen and done myself in the past is the use of footnotes for references. Footnotes are usually allowed with smaller typeface (even down to 8pt for example in MSCA applications) and can thus save you a lot of space. You should always provide a form of citation in the text if you are referencing work.
Upvotes: 4 <issue_comment>username_2: If there's a tight page limit for a posted ad, my personal assumption would be that there are a lot of anticipated applicants, and that applications are very likely to go through some sort of initial screening and sorting process. If this is the case, you shouldn't expect the person/people doing this sorting to link out of your document. There is a fair chance your link will go unclicked.
Upvotes: 2 <issue_comment>username_3: I suspect from reading the comments this answer is going to be unpopular, but:
If there is no style guide telling you how to format citations, you can format them any way you like, subject to the constraint that you provide enough information for readers to find the source material if they want to. Your href-based proposal passes this test, so yes, it's OK.
Upvotes: 1 <issue_comment>username_4: **No**, you cannot use links to references because:
* the style doesn't follow the norms of standard academic citations.
* those references will not work on printed out materials.
If you do not know whether the bibliography counts as part of the page limit, **contact the place you are applying to!**. We can only speculate and that's not really helpful when it comes to applying for faculty positions. The stakes are high, so you have to make sure you understand any and all formatting requirements! And if they are not specified you must contact them to clarify.
Upvotes: 1
|
2023/02/23
| 723
| 3,161
|
<issue_start>username_0: I am currently an undergrad looking to apply for pure math PhD programs sometime in the future. The thing is, I do not wish to apply for grad program right after graduation but want to work in the industry a few years after graduation.
I am wondering (1) If general GRE test scores matter so much for math PhD applications, (2) how much fo math GRE test scores matter, and (3) up to how many past years' score would be counted, i.e. do some grad school admissions say that they require GRE scores "in the last 4 years" or else you have to take it again.
My question is not intended to garner answers related to whether or not it would be a good strategy to apply few years after graduating instead of applying for it right after graduation. I have already thought deeply about my life and what I want to do. Thank you.<issue_comment>username_1: Usually the page limit includes the bibliography in a research statement or proposal (except if explicitly stated otherwise). Keep in mind that people, especially if they want to focus on something for longer than 20 minutes, still like to print things before they read it! So a link will not necessarily work for your reviewer.
It is not unusual to link things in a proposal, but rather as an extra feature instead of the only way to access the reference. You do not know, if you referee even has access to your original PDF!
What I have seen and done myself in the past is the use of footnotes for references. Footnotes are usually allowed with smaller typeface (even down to 8pt for example in MSCA applications) and can thus save you a lot of space. You should always provide a form of citation in the text if you are referencing work.
Upvotes: 4 <issue_comment>username_2: If there's a tight page limit for a posted ad, my personal assumption would be that there are a lot of anticipated applicants, and that applications are very likely to go through some sort of initial screening and sorting process. If this is the case, you shouldn't expect the person/people doing this sorting to link out of your document. There is a fair chance your link will go unclicked.
Upvotes: 2 <issue_comment>username_3: I suspect from reading the comments this answer is going to be unpopular, but:
If there is no style guide telling you how to format citations, you can format them any way you like, subject to the constraint that you provide enough information for readers to find the source material if they want to. Your href-based proposal passes this test, so yes, it's OK.
Upvotes: 1 <issue_comment>username_4: **No**, you cannot use links to references because:
* the style doesn't follow the norms of standard academic citations.
* those references will not work on printed out materials.
If you do not know whether the bibliography counts as part of the page limit, **contact the place you are applying to!**. We can only speculate and that's not really helpful when it comes to applying for faculty positions. The stakes are high, so you have to make sure you understand any and all formatting requirements! And if they are not specified you must contact them to clarify.
Upvotes: 1
|
2023/02/23
| 456
| 1,917
|
<issue_start>username_0: Our professor gave us a long exam worth 166 points. It was composed of 130 items 10 of which were identification and 120 were multiple choice. A lot of the questions required ethical analysis as out subject was ethics. He only gave us 30 mins to answer the exam. It this even legal?
Location: Philippines<issue_comment>username_1: I can't speak specifically about the Philippines, but in no country I have come across does the law provide any special provisions for things that are and aren't allowed in Universities. This particularly applies to situations where any judge deciding on a matter would have to decide if a professor's judgement about matters within that professors subject expertise or on issues of pedagogy, as the professor will be regarded as being in a better place to know than the judge.
Exceptions to this exist in some countries for equality issues related to "protected characteristics", such as race, gender, sex, religion or disability. In other countries, universities are subject to general law about discrimination that might apply to specific situations.
However, unless discrimination is an issue (and there is no suggestion it is here), no country is going to have laws on what is and isn't a fair way to assess someone.
Upvotes: 2 <issue_comment>username_2: I have no idea what "points" mean here and I don't want to know. But in any case 120 multiple choice questions in 30 minutes means you have 15 seconds to tick each answer. Yes, that's too much, if you are expected to complete them all; even reading out all the questions could take longer.
Upvotes: 2 <issue_comment>username_3: **Yes, it's legal**
There is no law in the Philippines or elsewhere that dictates length of an exam or time limits on the exam.
Some professors purposefully give an exam that no one can finish and then grade on a curve. That is possibly what happened here.
Upvotes: 2
|
2023/02/23
| 1,059
| 4,150
|
<issue_start>username_0: I am co-organizing a conference that will take place virtually on Mar 13, 2023. We invited the speakers and they confirmed their participation in Nov, 2022. Now, we sent them a form that collects their talk info + bios + photos, and a consent form to sign in Jan. and some of them have not submitted their info yet nor emailed us the signed consent form until now.
We sent a reminder on Feb 2 and they have not submitted the data yet! - I know that professors are busy but also this is putting a lot of stress on us, "the organizers", the registration for the event ends on Mar 9!
It is a volunteer conference so we offer no honorarium to the speakers and maybe that is why they are not taking it seriously but we (the organizers) are taking our volunteering work seriously and we care about the details!
I am uncomfortable asking them again to submit their info. What advice do you have for me?
Those who are late on submission are old professors and probably busy, and I am a graduate student.
And also, how to write an effective but gentle reminder email to the professors who are late on submission?
Here is what I came up with:
>
> Hello all,
>
>
> Hope this email finds you well. We are very excited that (conference
> name) is approaching and we cannot wait to meet you all!
>
>
> Could you please fill out these forms and email us the consent form if
> you have not done so at your earliest convenience ideally before the
> end of Feb? - Registration closes Mar 9.
>
>
> Thank you.
>
>
> BR, xxx
>
>
><issue_comment>username_1: Send one more and give a deadline (firm). "If we don't have your material by -firm date- you won't appear in the program."
If the recipients don't see the consequences of being later then there is little incentive (from their standpoint) to be "ideal". Make it clear.
Upvotes: 4 <issue_comment>username_2: As many reminders as it takes to get the information! I would keep a few things in mind. Decide on a firm deadline. Put it in the subject line. “Deadline 2/28: Submit program info for Name of Conference” Then send this to each person by name. When the email goes to many people (“Dear All”), it seems that many people are late. If I get a personal note, I get the impression that everyone else has submitted. Good luck.
Upvotes: 6 [selected_answer]<issue_comment>username_3: The way busy people read email is often a triage system. People make a quick decision on whether the email requires a response immediately, requires a response later, or doesn't require a response. For busy people with a lot of emails, the "requires a response later" category can quickly slide into "never going to get around to it."
So your strategy should be to make it extremely clear that your email *does* require a response immediately. Don't worry about being overly polite -- often including additional extraneous words beyond your request simply makes it harder for the person to determine what you need, which makes them more likely to decide that the email is not something they need to respond to right away. Just state what you need simply firmly and make it clear there is time pressure. Furthermore, personalizing emails adds pressure to respond.
>
> Dear Prof X,
>
>
> We are excited for your upcoming talk at the Y conference.
>
>
> The deadline for title/abstract/bio submission is Z. Please make sure to submit your material by the deadline so that we can include your information on the website.
>
>
> Thank you,
>
>
> Nadine
>
>
>
Upvotes: 3 <issue_comment>username_4: It seems you think it is rude to tell them about a deadline. What is ten times ruder is not telling them about the deadline, they miss it, and as a result can't speak on the conference.
So tell them. Using words that are hard to ignore.
Upvotes: 2 <issue_comment>username_5: I would suggest being more direct. Explain it as your problem (like the printer of the program being a stickler for a deadline). But make very clear what you need doing and by when. Otherwise many will not even realise the need. If you want additional background put that after the call to action.
Upvotes: 2
|
2023/02/24
| 1,326
| 5,767
|
<issue_start>username_0: I am in a very confusing situation. I am almost 3 years into my PhD project, and just today my direct supervisor suggested to me to read a paper in the best journal a PhD student would want to publish in.
After going through it, I realised that they have used a method that I spotted during the 5th month of my PhD. I did multiple experiments with that method with success, but at some point my supervisor told me that they didn't want to talk about that analysis anymore. I kept working on it without telling them.
Today, I could not believe that my supervisor was suggesting an idea they have rejected before. I just can't. I double checked the results from back then, and there are multiple significant results. I am not sure if I can keep working with them anymore.
How should I proceed from here?<issue_comment>username_1: Your supervisor made a mistake. If they had supported you in following this research thread, it would likely have been successful, and you would have been that much further on your research path.
This made you frustrated and disappointed with your advisor, and now you do not want to work with them anymore. I recognize your frustration and very much empathize with it. I had a supervisor shoot down one of my ideas; I got back to it nearly four years later when I was a professor, and it turned out to be highly successful.
However, allow me to offer a few perspectives on your situation.
First, it is extremely difficult to tell what ideas are good, and what ideas aren't. You have the benefit of hindsight now because you saw the journal version and your own experiments (which you hid from your advisor for some reason, more on that in a bit).
Advisors are human beings, and make judgment errors. They are operating with limited time/financial/computational/human resources, and will inevitably make some big mistakes on the way. The main thing that you should do is to *communicate* with them. If you have a lot of faith in an idea: tell your advisor! If you ran experiments despite their advice and they worked out - tell them.
Now, if their reaction is to be mad at you for not listening to them, rather than being happy for being proven wrong, then yes - there is a major issue.
If you were uncomfortable communicating with them which is why you hid your experiments - there's a major issue.
You shouldn't fault your advisor for misjudging a research direction. You should absolutely question how it is handled from now on.
If there are still open problems/directions to take here - you are in an excellent position to explore them I believe. Take the opportunity to learn a lesson in better communication with your advisor (something both of you should clearly do better in), and move forward if you can.
Upvotes: 6 <issue_comment>username_2: The universal answer to most questions on this forum applies to your case as well: "Talk to your adviser".
You suggested a method a while ago, he thought it wasn't good, now he's changed his mind. Reasonable people will then have a conversation of the kind "I thought you didn't like this method! What made you change your mind?" Your adviser might have good reasons. Or they respond "I did? I have no recollection of it, but in any case, I must have been confused at the time". The point of it all being that you should have a *conversation*; you might learn something from it, and moreover spending your time trying to read the tea leaves what precisely might be on your adviser's mind is just a waste of everyone's time!
Upvotes: 4 <issue_comment>username_3: It happens to lots of people. You'll find it mentioned frequently in PhD memes and comic strips (I remember a phdcomics strip joked about it, but I can't find it now).
If you are planning on remaining in academia and supervising students after your PhD, you'll probably be on the other side of this situation at some point in your career.
It'll happen to you in other non-academic situations as well.
You already have some good results, so you're closer to something publishable than your advisor would have expected if you started these experiments from scratch -- this is a *good* thing! If I were your advisor, I would be quite pleased at being proven wrong.
To avoid the same in the future, you can try to understand what convinced your advisor that this line of inquiry would be productive, and modify the way you sell your preliminary results. Being able to convince others that your preliminary results have merit and that you should receive support on fleshing those out is a skill that you should learn when you're in grad school, because it is a useful skill to have after you graduate.
I see you've mentioned in other comments that your supervisor doesn't remember what you discussed in previous meetings. This is very normal. Remember that they probably have other students to supervise, and classes to teach. These are not trivial responsibilities, and could make it hard to remember every single detail that you discussed with them.
I would often find myself spending a not-insignificant chunk of time recapping what was discussed in previous meetings with my committee members. I found that maintaining a document that summarized every meeting was helpful in jogging their memory and reducing the time spent to remind them about things we'd already discussed.
Upvotes: 3 <issue_comment>username_4: Go your own way now.
They didn't acknowledge you at first, and now they are furthering their career, with the method you discovered.
The only ethical way to resolve that, is that you become an entrepreneur and escape your university. Your supervisor needs to suffer. Don't let them take credit for your rejected ideas again. You are smart, you can do it.
Upvotes: -1
|
2023/02/24
| 481
| 2,117
|
<issue_start>username_0: One manuscript has been submitted after the 1st revision, and I am the 4th author. I didn't contribute to the work initially, so my name was not there on the first submission. Whereas during addressing comments from reviewer, I contributed and hence my name added as 4th author during re-submission. However, there is a mistake about my affiliation in the re-submitted manuscript. The manuscript is yet to be accepted. If it will accept, then can I change my affiliation? If so, in which stage I can do that? Is it possible during proof-reading? I am afraid if we inform the editor about this before acceptance, he won't take it well and reject the manuscript due to this silly mistake. As this particular query is not available in the 'Guide for Authors' or anywhere in the journal, I can't understand what should I do.<issue_comment>username_1: Yes, you can. Affiliations change, sometimes in the middle of submission. The "author information" side does not affect the "accept on the merits" side.
Upvotes: 2 <issue_comment>username_2: Yes, you can usually change your affiliation after acceptance, during the proofs stage.
There's no need to inform the editor that there is an error in your affiliation, but you should inform them that there is another author. If they're paying attention, they should notice themselves, and investigate (because extra authors is a potentially fraught thing that need to be clarified). If you inform the editor, then you can head off the editor's request for clarification if/when they discover it.
Upvotes: 1 <issue_comment>username_3: Yes, you can definitely change it later on. Most important thing is that you will need to be on top of this with the main (first/last) author who is taking care of the submission and proofreading. It's rare that the middle authors concern themselves with the final stages of getting the manuscript ready for publication, so make sure you see a next version of the manuscript where it is correct or send them this information ASAP and then again when you hear the manuscript is accepted for publication etc.
Upvotes: 0
|