Pre-interview stuffs to do

For people who have interviews approaching, here's a short, preparatory (and probably mostly self-evident) checklist for the campus visits:

• If you haven't done this yet, check to see if it is traditional in the department for prospective students to contact potential advisers in advance of the interviews. In some departments (including the three I've been part of), it's really your future adviser who grants you admission, and the department just rubber-stamps their decision. In other places, it's common for students to join without knowing exactly who their adviser will be. Write your potential adviser, if it's appropriate.
• Read the departmental website thoroughly. Try to determine the requirements for achieving candidacy, what kinds of coursework students take, whether there are lab rotations or teaching requirements, and whether there are department-wide policies on how graduate students are funded.
• Read thoroughly the lab websites and personal websites of faculty with whom you are interviewing. Read their CVs. What are their academic backgrounds, and who were their advisers? What fields have the potential advisers published in, and which fields have they published in recently? Do they have tenure, and if not, when will they be up for it? (For outdated websites, you might want to supplement their publication list with searches on Google Scholar, PubMed, arXiv, Web of Knowledge/Science, Scopus, etc.).
• Scan the specialties of the faculty members in the department. Get a sense of the breadth of the department.
• For each faculty member with whom you have a decent chance of working, read the abstracts of all of their papers that you find vaguely interesting. Read thoroughly at least a few papers that relate to work you might do with them. If you don't have access to these papers through a university, it's perfectly fine to email the professor to ask for a "reprint" (which will really just be a pdf). It's better not to do this at the last minute.
• Scan the pages and publications of the grad students, techs, and postdocs in the labs you might join. 
• Figure out who the department chair is.
• Write a cribsheet with this information somewhere and take it with you to interviews. This is really helpful if you're interviewing at a bunch of places back to back.
• Review your personal statement, and use it to develop a list of questions in your field that you find interesting and another list of potential projects. These ideas don't have to be genuinely viable at this point, but they show you're curious and thinking about research, and you can use them as a springboard in conversation to talk about more realistic directions.
• Write down the questions you have about the department, research, etc. (See the posts on meetings with potential lab members and potential advisers for more ideas.)
• Take a notepad and pen with you to your interviews.

Bonus points:

• If your potential advisers are funded by the NSF, look up their current and past grants on the NSF's Fast Lane Award Search. If they're funded by the NIH, check out the NIH RePORTER database. Most private foundations have their own databases too. Sources of funding are usually listed on CVs.
• Look up your potential advisers' most cited publications in one of the citation databases mentioned above to get a semi-objective sense of what they're known for. Also look at who's citing some of their more recent publications--which hypotheses are being debated? Who are their collaborators? Pay attention to any Science, Nature, Cell, PNAS, and Phys Rev Letters publications.

 Are there other preparations that people have found helpful?

Following up after interviews

Thank everyone with whom you met for >15 min! This means graduate students, postdocs, faculty, and administrators.

A simple email should suffice. Following classic thank-you form, it's good to highlight what you particularly enjoyed about the conversation.

If you're thanking a potential adviser, the email should be a bit longer and touch on the research projects that interest you and (if true) the positive interactions you've had with other lab members. If the lab is your first choice, it's okay to say so. You can also gently remind the potential adviser about any outstanding questions he said he'd get back to you on.

If you want to write your letters by hand on your prettiest stationery, that's okay*--as long as they go out immediately. Assuming you have already written a decent note of some kind, the most important thing is that you send it quickly, ideally within 2-3 days of your visit. Snail mail has an obvious disadvantage here. Where I did my PhD, we gave feedback on the candidates several days after the campus visit. It's best if you can beat this deadline.

The whole thanking-people-for-meeting-with-you thing is a generally good habit for life.

*Keep in mind that a lot of researchers travel frequently and don't check their mailboxes that often. I'm one of those people who prefers having all my correspondence in one, easily searchable place (e.g., my gmail inbox, supplemented by tags and threads). Email might make a lot more sense.

What to wear

Disclaimer: I haven't seen any randomized, controlled studies that test the hypotheses (read: opinions) below.* That said, I've experienced both urban and rural academic environments in different parts of the country and in Europe, and I'm a fairly socially astute person. I've also worked in very formal professional environments. What I write below applies to science programs in the United States.

Looking too nice makes you look clueless. Looking too sloppy makes you look clueless too. The goal should be to dress slightly more cleanly, neatly, and smartly than the people with whom you'd be working.

Women are probably best off wearing slacks and a blouse +/- a sweater. Men should wear slacks and a shirt. No tie, unless one really makes you happy. Flat shoes are good for both sexes. Try to avoid conspicuous brands.

This stuff shouldn't matter, but it kind of does: I remember a grad student making light fun of an applicant's precociously blingy handbag. (This might not have happened if the applicant hadn't already displayed an attitude to match.)

Remember, you can dress almost** however you want when you're a grad student.

If you've never dressed for a professional job before, get someone else to judge your outfit. Ask them if you're "dressy casual."

If you're visiting a lab that works with a lot of doctors or industry types, people might dress more nicely, but probably not enough to affect the recommendations above. I've wondered about the labs at Mayo Clinic, which requires its medical residents to wear suits every day.

Lastly, if it's cold out, take a coat. If it might rain, take an umbrella. Wear shoes that allow you to walk several blocks to a restaurant. When you're famous and have tenure, it's okay to show up to give a talk in January in the Midwest without so much as a sweatshirt (true story). For interviews, you want people to focus on your words and fantastic attitude, not your social awareness or thermoregulatory abilities.

*The disclaimer applies to the entire blog, come to think of it.
**This is an interesting topic for another post.

The campus visit - part three

Perhaps the most exciting part of the visit is chatting with potential labmates.

You'll probably spend more time every day interacting with other grad students and postdocs in your lab than with your advisor. Your labmates are the people you'll first go to when you have an idea that might be the awesomest thing ever or really stupid and you need to check. They're probably following the literature more closely than your adviser, and they'll be the ones who forward you cool articles related to your work. They'll almost certainly be more familiar with the methods you're trying to use.

It's really important that you at least slightly like your labmates, on average.

Speaking from personal experience, a single curmudgeonly or even pathological labmate is endurable if you're surrounded by an otherwise happy, creative, and amiable group of people.

Talk to as many potential labmates as possible. Talk to them about their research, and see if the conversation takes off. You can also ask about the same topics you discuss with your potential adviser. How did they arrive at their current projects? How would they describe their adviser's mentoring style? What conferences have they been to? How's the funding situation? With whom do they collaborate in the lab, in the department, and at other schools?

The point isn't to search for dirt about the potential adviser or anyone else. That said, if a potential labmate makes any suggestive comments, follow up as tactfully as you can. Face-to-face conversations are practically the only option in this case. I was once in the awkward position of meeting with a potential advisee of a professor (not my adviser) who was severely negligent as a scientist and a mentor. The advisee didn't ask about the professor's advising habits or working style. I compromised by suggesting a few pseudo-reasons why another program might be a better fit.

My scientific development might owe more to a superfantastic labmate than to my adviser. This labmate was an extremely clever older grad student who decided early on to involve me in her brainstorming. She was generous and encouraging, and we wrote some kickass papers together.

Science is intensely collaborative, and the more you like your collaborators, the more fun it all is.