What are issues that arise in advising grad students and young colleagues, or about which you have sought advice?

Question

I'll start off the thread below with some thoughts on coauthorship, and I hope others will chip in on that and other issues that might fall under the category of mentorship.

Answer 1

Coauthorship: Economics papers are very often coauthored, and coauthors often have productive collaborations that span multiple papers (and years).  But questions often arise about whether someone should be a coauthor.

Early in a project, should someone who you had a productive conversation on the subject be a coauthor?  How about if you are now working with someone else on it, or if someone approaches you with an idea related to a conversation you already had with someone else?

Later in a project, how about when someone offers a helpful suggestion: when is an acknowledgment enough?

How about if you are thinking about something that builds on a paper you already coauthored; are you obliged to invite your previous coauthor(s) in?

I have been asked about things like this many times, and of course the answers depend on the circumstances.  But I think two (somewhat contradictory) general principles guide the questions I ask and the discussions that follow.

First, be kind to your friends and colleagues. (It isn’t always clear what course of action this suggests, when e.g. one person suggests a topic to you that you have already talked about with another, or when you have fruitful preliminary conversations with many people…). But friends and colleagues are important, both because you want to be the kind of person who is a good friend and colleague yourself, and because collaborations often improve your work.

Second, seek an equilibrium that allows work to be accomplished, expeditiously.  Bringing old coauthors along for the ride sometimes hampers new efforts (not least by being unappealing to new collaborators). Ideas at an early stage are not yet (or, I think, shouldn’t be) long-term private property; they have to be developed in non-obvious ways to get a project started, and then a lot of work has to be done to develop them. Most often, your coauthors are readily identifiable as those who have participated in the development of the ideas. Acknowledge freely the useful conversations that don’t meet this test, but don’t feel obliged to offer coauthorship unless it seems likely to make the work better or faster, or unless the conversation was really pivotal in your subsequent development of the idea.

These thoughts certainly don’t determine what to do in each circumstance, but they help think about what to do. (And if the coauthor in question is someone who cares about your success, like your advisor, then give more weight to the second principle about getting the work done expeditiously; while if the coauthor is a peer or less senior than you, maybe the first principle deserves more weight…)

Comments

I would also like to add to this list how people decide what is worthy of coauthorship from a research assistant - where there is a paid contract for work provided but often the extra enticement of the possibility of coauthoring.

---

Agreed, it would be very helpful to answer this: both from the perspective of the professor and from the perspective of the RA

Answer 2

Have someone else read what you think is a mostly final draft of your paper. That’s one of the things colleagues are for. Or better, have your wife or daughter read it (at least if the latter isn’t too busy.)

Corollary, you can’t proofread your own paper. No one can. Your brain sees what you meant—not what you typed.

Further result, if you didn’t grow up speaking a Germanic or Romance language it’s nearly impossible for you to get all the English grammar right...at least not for years and years. Get someone to edit for you.

Comments

Or your husband?

---

Your supposed corollary goes a bit too far. While there are obvious disadvantages to doing one's own proofreading, no one can really do without it. And there are ways to get some distance to one's own writing; mostly waiting and doing something else. It is rather easy to find mistakes after being on holidays. Another great time for finding one's own mistakes is one day after having submitted the paper to a journal...

---

Totally. I meant you should have someone proofread after you proofread yourself. And I also follow your rule of waiting. I put a paper down for at least 24 hours before a final go-over.

---

Yep, or your husband. I was just referencing my personal situation.

Answer 3

As a grad student, I'm curious about more senior folks' perspective on this question:

What approaches can grad students take to best facilitate an adviser/mentor relationship? I often find myself stuck in the question of what point in my research is the right time to check in -- should it be once I've hit a roadblock or before I get to that point? What if I don't have a question about what to do next, but just feel like it's been too long since I've stuck my head in their office to say, "Research still going well. Here's what I've done"?

Or do you as the senior person take the initiative as an adviser to maintain regular check-ins? If so, why? If not, why not?

Answer 4

This is an answer to the question below of "As a grad student, I'm curious about more senior folks' perspective on this question:  What approaches can grad students take to best facilitate an adviser/mentor relationship?":

I'm not senior, but I've had a very good relationship with my advisers, and I'm still working with one of them a lot.

First, advisers vary. Some will be happy to chat without much plan (though this seems to be rare), whereas others prefer to talk about concrete issues. Some will take it as their responsibility to make you see them whenever they think you should, whereas others view themselves as resource that you can go to if you'd like to, but that don't approach you. Because of that variation, it's also very hard to draw inference from what an adviser thinks about you, unless you compare to how the adviser is with other students (i.e. you need to account for adviser fixed effects).

Second, you don't only want to talk to your adviser when you have a problem and you're stuck. That just sends a bad signal, and chances are, your adviser won't be able to solve the problem in a short meeting if you haven't succeeded even after giving it multiple serious attempts.

Actually, I think what advisers are most useful for is to prevent you putting a lot of work into things that are not worth doing. So, basically, tell your advisers what your plans are, both for the big picture of the project, as well as for how you are planning to approach the project. Chances are that they will have good suggestions about asking the research question in a more productive manner, about literature that relates to your question that you were not aware of, or about methods that you might not know.

So you want your adviser's advice about the things which, from your point of view, are unknown unknowns. That's why you want to keep your adviser updated about your progress and, more importantly, plans. And you don't just want to tell him / her "research is still going well", because there's a chance that what you might think is good work will not be appreciated by other economists (for PhD projects, that chance is actually fairly high, and the chance that it could be done in a way that would be more appealing to other economists is usually close to 1).

Answer 5

To continue the discussion about how students should manage their advisers, I certainly agree that you shouldn't only talk to your adviser when you have a problem.

I've found that a mixture of frequent short meetings and occasional long meetings works best for me (and hopefully for my students).  In Boston, where we drank coffee indoors, I would have a scheduled coffee with anyone who wanted to join every day, in the area outside my office.  Sometimes we might talk for an hour, but sometimes after checking that everyone was still alive we would all take our coffee back to our offices after 15 minutes. In California, where we drink coffee outside under the oaks, it always takes an hour, so I have two scheduled coffees a week.  These informal meetings allow people to get quick answers to quick questions and keep things moving, and to hear what their colleagues and mine are up to.

Real progress reports take longer, one-on-one meetings, and these I try to get scheduled in advance. I do work with my door open, however, so sometimes someone sticks his head in my door to ask a quick question or to make an appointment, and a short conversation nevertheless turns into a long one.

A large part of research consists of conversations. Have them often, with your colleagues, and with your adviser...