Wednesday, August 04, 2010

Family Ties

Based on my own experiences at various universities, the accounts of colleagues at yet other universities, and what I have gleaned from the instructions that accompany tenure dossiers that I am sent to evaluate, it seems that different institutions have different policies regarding how publications and grants are counted for tenure. For example, policies seem to vary as to whether publications and grants are considered in the tenure evaluation if they stem from Ph.D. research and involve the former adviser(s) as co-authors or co-PIs.

The variations I have seen are (1) everything counts; (2) only those items that post-date the start of the tenure-track appointment count; and (3) only those items that post-date the start of the tenure-track appointment and that do not have ex-advisers as co-authors count. Everything goes on the CV, but what is actually considered in the tenure evaluation may vary.

I was thinking about this recently because I have a new project and collaborative grant with one of my former graduate students, now a professor. We got the grant a couple of years after she completed her Ph.D. and it is a completely new research project, only slightly related to her Ph.D. research. We just started the project, but of course we hope that publications and perhaps future grants will result.

I told my ex-student/now-colleague to find out exactly what the policy is at her university re. what counts for tenure. I am sure there will be a way to explain that this is a new project, even though it involves an old adviser. I am not worried at all that our collaboration will harm my former student's tenure case in the future because she has other grants and independent projects, so she's going to have a strong record and an impressive reputation for her own work no matter what the policy of her university.

But I wondered: In a more marginal case or in a case of a very strict definition of what counts for tenure, is it possibly a bad thing to collaborate with a former student before they get tenure?


ChemProf said...

I think there is always the danger that it will be considered that the former student isn't actually capable of carrying out independent projects, and is depending upon the former mentor in writing the grants etc. Sadly, especially if the former student is female. And yet collaboration is usually encouraged, so this is a catch 22.

Odyssey said...

My postdoc advisor always drummed it into his trainees that collaborating with former advisors prior to tenure was a very bad thing. Thirteen years later I'm not so sure. Certainly in the case of a strong candidate with projects and funding of their own it doesn't hurt one iota. In the case of a marginal candidate though I think it could hurt more than help. Tenure candidates never want to be viewed as running a satellite lab of a former advisor.

Anonymous said...

I wonder if it's more acceptable to collaborate with former postdoc advisers than former pHd advisers? The idea being that by the postdoc a scientist should be independent enough that the postdoc adviser-advisee relationship is much more collaborative than top-down.

But I also don't see the harm in some cases of finishing the pHd research, moving in other more independent directions for postdoc and first few years of tenure track, and then striking up a renewed collaboration with the old pHd adviser. If the collaboration is fruitful and producing good science, why should that be counted as a strike against one?

qaz said...

I think the key may well depend on whether the junior collaboration team-member in question has other projects as well. In my experience (which may be particular to my institution, but seems to be similar at other institutions that I've seen), whether or not collaboration counts depends on whether there is independence as well.

The logic is something like this: if you have 5 collaboration papers and no independent papers, then you don't have anything that counts. But if you have 5 collaboration papers and 3 independent papers, then you have 8 papers that count. This is rarely stated in such an explicit manner, but this seems to be the attitude in tenure discussions that I've seen.

A similar logic seems to apply to major/minor papers and journal/conference papers.

Comrade PhysioProf said...

My institution does not have any formal policy on this, and it is basically in the discretion of each person at the various levels of the decision-making hierarchy how to weigh the qualifications of the candidate and whether to vote in favor of tenure. Do other institutions have explicit written policies on this?

Anonymous said...

Why would you want to collaberate with a former student? They are now your academic competition! The way I see it: if you are not at my university anymore, you are dead to me.

Average Professor said...

It feels beyond weird for me to say this, but ditto CPP. At my institution, everything goes on the vita, but individuals at points in the process are free to weight things however they want. It's important for the candidates to be as forthcoming as possible about the circumstances of those projects and publications, just so that nobody that cares about those details misinterprets something.

Also, this is one of those cases where the answer SHOULD be simple but of course isn't. What would be nice is if it were just true that researchers should work with whomever it makes sense for them to work with to do interesting and innovative things.

Anonymous said...

"If you are not at my university anymore, you are dead to me."

So you only collaborate with academics from your own university? What an odd policy... What reason do you have for this?

Anonymous said...

I'm really hoping that Anonymous posting at 08/04 10:30 am is trying to be funny.

Anonymous said...

I talked to my department mentor about this. He told me publications and even grants with my former grad advisor would not be likely to be seen as negatives, assuming I'm doing good stuff on my own as well.

Anonymous said...

Because this "policy" (official or otherwise) varies so much from one institution to another, it may also be noted as a negative by external reviewers no matter what the home institution's policy is. This happened to me and while the comment was negated by the promotion committee, it is still in my file. Something to weigh in the discussion.

Anonymous said...

What about writing publications and grants with biological parents pre-tenure? There's a guy in my department who has 1/3 of his publications with his dad and 1/3 with his thesis advisor.

a physicist said...

The rule of thumb I've always heard is that publications where you collaborate with someone senior to you, generally don't count, period. Former advisors are more senior to you, but any other senior collaborators are also a strike.

But I think this advice goes well with qaz's point: if you have some great independent publications, then anything extra is a nice bonus. The key is to have a core of work that is clearly your own group's work, where it's unambiguous it came from your group.

The main point of this overly harsh attitude is not that collaborations are bad. Rather, you're trying to demonstrate you can do good science all by yourself. If all of my tenure-case papers are with Einstein and Feynman, even if neither were my former advisors, wouldn't everybody wonder if I had any ideas of my own? You'd have to know the situation in detail to be able to give me credit. My co-authors Einstein and Feynman could make my contribution clear. But others could suspect that I'm just the technician carrying out their ideas.

Yeah, yeah, I realize this is not a great situation to be in, and if Einstein wanted to collaborate with me pre-tenure I'd be a fool to say no! I'm just saying, this is my perception of how it is: you have to have a core body of work that clearly originates with you. I would not recommend collaborating with Einstein if it detracts from your ability to publish independent papers.

Ann said...

Independence is a big issue in the tenure process. Provided an individual has some projects on which he or she can demonstrate leadership, collaboration with a former mentor should not hurt, and could help. It looks bad though if all of the best work is with more senior co-authors.

Anonymous said...

I think it is pretty typical that it will raise red flags if too many of someone's papers are collaborative with ANY more senior and/or established scientist. I saw this come up and result in rejection of one candidate for tenure. The issue was raised not about papers written with advisors of any sort, but with a co-author who actually was of comparable seniority to the candidate himself. This co-author was nonetheless suspected of being the "idea man", and because the candidate had very few papers without a more established author of some kind, it was argued that he was just carrying out the ideas of others. This led to much heated discussion and not all agreed with it but combined with other factors I believe this argument was critical to sinking this person's tenure case.

So what I tell junior people is that it is ok to collaborate with advisors or anyone else, but there has to be some set of material that you clearly "own" where no one will doubt that your contribution is central. That's consistent with FSP's view of her protege.

I had many papers with my postdoc advisor and I got tenure no problem. I think the key was I had enough substantive papers without him as well, and some of the ones with him were on things that were clearly not in his normal portfolio so my contribution was fairly obvious.

Female Computer Scientist said...

This is a somewhat tangential question, but how important are single-author publications for tenure? I always assumed that was field-dependent.

In my subfield it's unusual to see many single-author papers, except perhaps for review articles or book chapters. (And I read neither of those counted much for tenure.)

Ms.PhD said...

I think this is a moving target. My first postdoc advisor said that he always assumed anything done for the PhD was the work of the PI, and he only evaluated assistant professor candidates based on the postdoctoral publications.

But my understanding is that postdoctoral work is attributed to the postdoctoral PI, and tenure is only evaluated based on independent publications.

However, all of that goes out the window depending primarily, so far as I can tell, on one thing: how famous the PI/collaborators are.

If the person is more famous/senior than you, they will get the credit for your work. No matter what you say. Unless that person is gracious and generously writes you a glowing letter saying it was all your idea and they only contributed one minor thing.

But in my experience, famous people are rarely gracious or generous.

However, all of THAT goes out the window if the people on the committee know the famous person and/or want to be further associated with them. Then it becomes something like, "Oh my god, you know OPRAH! What's she like??!!" And it doesn't matter how much the candidate did or did not do. They are friends with Oprah. They are inducted into Teh Club.

GMP said...

@ FCS:

What counts are the publications that were *primarily* done in the tenure track faculty's group.

In my field, that means: if the first author is my student or postdoc and the last author (i.e. lead senior author) is me, then this paper is considered MY paper and such papers are the most important ones for tenure. Collabprative papers where some other seniopr faculty is last author (or however else the lead senior person is denoted) are worth less.

I have had a couple sole-author papers on TT, but by-and-large students and postdocs are supposed to be first authors.

I was also explicitly advised to sever ties (in terms of publications) with former advisor as soon as possible.

Unbalanced Reaction said...

Can I adopt you as my former PI?

Like many have already commented, here at PermaU it all depends on that year's tenure committee. Oh, how wonderful it would be to have some black and white policies!

Anonymous said...

What about papers that don't involve students or postdocs? This is going to be mostly theoretical and/or mathematical work most likely (as you kind of need students/postdocs to get lab work done once you're a PI IMO).

In these situations it's not uncommon to see many PI's as authors, and even a junior faculty member the last author.

Will a last-author paper where the other authors (first, middle) are senior faculty count against you, whereas a last-author paper where the other authors are all students will not?

Anonymous said...

About single author papers (response to Female Computer Scientist): I'm sure this is field dependent. In my field(s) they are not critical.

About book chapters and reviews and things: I actually think these things can be very helpful. They may or may not "count" in the sense of a hard number of "real" publications, but I do think that if single authored they can go some way towards establishing a junior person's ownership of some material.

For example, if the junior person has papers with grad or postdoc advisor on topic X, raising doubt about whether she is really the master/idea person/owner of topic X and driving force behind those papers, she can help to establish ownership by writing a single-author review or book chapter. This would carry some weight with me as a tenure review committee member and I think I'm not alone.

Anonymous said...

A colleague of mine took a sabbatical right after getting tenured to spend time at her PhD uni writing up papers with her PhD adviser. Her university discouraged her from publishing with her adviser even though she had great data. Fortunately in their case, the data still publishable after 6 years, but I thought it was silly to stipulate this sort of thing. I understand the independence thing, but there's good work that's not getting out! She's brilliant by the way and clearly can work on her own too..

Anonymous said...

I had this problem at time of hiring. Most of my papers were with rather senior people, so some evaluators assumed I just went along for the ride. Luckily some hiring committee members made informal inquiries upon which they were informed by said senior co-authors that those papers were based on my ideas and plan of action. End result: I got a nice TT position.

Another female science professor said...

I have seen this concern inconsistently applied - depending on whether the person was 'liked' or not (ugh). The concern, does however have some merit. We hired someone nearly 15 years ago at the associate level who had been working with her advisor (as a post doc) for >= 10 years. We thought she could hit the ground running given the # pubs she had. We discovered she could not setup and run a lab herself. She is still an associate prof.