Monday, August 11, 2008


There are some clear Conflicts of Interest (COI) that exclude one from reviewing proposals (and, though not enforced to the same extent, papers) by particular people: advisors, graduate/postdoctoral advisees, close colleagues, relatives. These are obvious COIs, but more and more I find myself encountering a COI gray zone: former undergraduate students.

The number of ex-undergrads I have taught in classes or advised as interns or research students of various sorts and who are now proposal-writing faculty increases every year. Of course, not all (or even most) of them are writing proposals in my same field of research, but some of them are.

When asked to review a proposal by a former undergrad I had advised and/or taught, I used to write to the program director who solicited my review and say "I advised X as a summer intern" or "I advised Y's senior thesis" and ask "Do you still want me to do the review?", and they always replied "If you feel that you can be objective, please do the review; if you can't, don't."

So, the decision is up to me.

I don't really see why former undergrads, especially those who did research with me, aren't official COIs. If I work closely with a student on a research project for a year or more, how is that different from, say, a former MS student, which is an official COI relationship? I want (most of) my former students to do well; does this cloud my judgment?

The semi-COI is most obvious in cases of advisor-advisee relationships, but in some cases I have strong, possibly not-objective opinions about former students who only took classes from me. In one case that I can recall, I was placed in a situation of evaluating a former undergraduate student who had been extremely rude and obnoxious in two of my classes. I guess he's brilliant, but he's a jerk, and I did not feel I could be objective in that particular case.

So, I do the review or not, and, if I do the review, I write the potential COI information in the box provided for this purpose on the reviewer website and let the powers-that-be sort it all out. I assume that I am not wasting my time writing a review that will be excluded owing to a semi-COI.


Drugmonkey said...

I think the NIH uses the standard of "any" mentoring relationship, there is nothing special about undergrads. The default COI rules also feature a time limit, 3 years. Still, lots of gray area where the individual has to make the call.

Trouble is that the bad actor won't see themself as in conflict and those that worry are probably fairest.

Candid Engineer said...

I don't know what the norm is in your field, but in mine, professors have extremely minimal contact with undergraduates who are working in their labs. Undergrads are typically overseen by grad students or postdocs, and often the professor wouldn't even know who these kids are. Perhaps the agencies soliciting your review don't want to include undergrads as COI knowing that, in many cases, there really isn't much conflict.

Kim said...

I've always listed my undergrad mentees as COIs. (I don't have grad students; I've done research and published with the undergrads; I've written letters of recommendation for grad fellowships and even, in one case, for jobs post-PhD. In those cases, there's no way I could be objective.)

Anonymous said...

I think the director is right.

You're the only one who can judge your impartiality in this case.

Ms.PhD said...

You're not wasting your time, because they are so desperate for what they consider "qualified" reviewers, that COIs are often ignored. I know this is true for paper reviews, I can only assume it's true for grants too.

I think it's disgusting that you made a point of raising the issue - which you didn't have to do - and they didn't automatically exclude you on the spot.

The whole point of COI is that you can't objectively judge whether you can be objective. You're subjective by definition (we all are).

Anonymous said...

If you have to ask if it's a COI, then it IS a COI.

Anonymous said...

If you have to ask if it's a COI, then it IS a COI.

One can only wish. In real life COI rules are sometimes drafted by bureaucrats with little knowledge about actual research. In one case, the rules forbade a potential reviewer because twenty years ago we had published one many-authored paper together.

Anonymous said...

A lot of these rules just don't seem to address the spirit of the issue. Of course someone you mentored as an undergrad is a COI (to me, it is not a semi in reality, just in the letter of the rules you're dealing with in this situation).

and a default of 3 years like drugmonkey mentioned is absurd; how is it NOT COI for my graduate advisor to review something of mine 4 years after I finish?


Douglas Natelson said...

What is the half-life of a COI relationship? I've been out of college for over 15 years. Would it be a COI for the prof who was my advisor on my junior independent work back in 1991 to review my grant proposals? Three years may be too short, but there has to be some sensible time limit for COI.