The ever insightful Flavia of Ferule and Fescue has a post up about taking candidates to dinner and some of the issues for hosts. It's well worth reading. She talks about how important it is to share meals with candidates, and makes the point that meals are a time when department members sell the department.
I'd like to talk about some of the nuts and bolts.
First, as a candidate, you should know you're going to be fed. (At least, you should be.) But it never hurts to carry a couple granola or protein bars with you just in case. If you have a food allergy or are a vegetarian, I think you should let the department know as they're planning. Speaking for my department, we have enough people who are vegetarians around that we pretty much assume we need to go somewhere with decent vegetarian options. But I'm guessing a lot of people in my department would be totally thrown by a request for kosher options. (I'm not sure we have any kosher eateries here, but I'm guessing I'd try a restaurant with good vegetarian options and hope for the best.)
We have a fair number of restaurant choices in our small city, but a lot of them have drawbacks. For example, the best Thai place is a hole in the wall that doesn't exude class, and gets a little loud, so we probably don't take many candidates there. Still, you can guess that within whatever budget constraints the school sets, the faculty are going to try to take you to one of the better places around. If it sucks, then local dining out places probably suck in general. That might be good to know.
As a way of selling the department, meals are difficult. For one thing, after a long day of classes, meetings, extra stuff for candidate visits, and so on, some folks will find it really difficult to go out to dinner. Folks with kids, for example, are likely to find it especially difficult. So don't feel offended if only a few people take you out. And those who are out are probably really tired, and may not sparkle as they would on a weekend evening.
You can tell a lot about department friendliness by watching the interactions. Do people know about each other's families? About their work? Their hobbies?
It's got to be a balancing thing, though; you want to work with folks who are friendly, and care about you as a person, but you probably don't want to work with folks who are too involved in your off campus life.
So, there's lots for you to be looking for as you socialize. But, of course, people are also looking at you.
Some departments probably care if you have a drink. Some expect it, and will be worried if you don't (but probably not hugely worried, since they'll realize you might be nervous). But some probably will be a bit uncomfortable if you do. If I have any advice, it would be to have a drink if that's your preference, but only one (because, as Flavia notes, drinks are often paid for by the faculty hosts, and not the institution). I don't think anyone in my department much cares, though at the last few dinners I've had on candidate visits, we've all tended to go with hot tea, in part, at least, because it's so cold that we want something to help warm us up.
Follow the leader on orders. If everyone else orders lobster and you want lobster, go ahead.
We do send our candidates to a meal with faculty who are on the tenure track but not yet tenured. We do this explicitly because we think we treat our tenure track folks decently, and we think that our review and tenure processes are strong and fair, and we hope that candidates will ask questions about the review and tenure process and get that sense from talking to the tenure track folks. We also hope the tenure track folks will tell them about how relatively low rents are and how much there is to do in the community, and so on.
I don't know if every department or campus does this. (I would think in smaller departments, it might be difficult because they don't have anyone untenured.) But I had a friend in grad school who had a campus interview where she ate with tenure track folks, and they told her in no uncertain terms that they were all trying to get out and that it was a bad place to be. So she took another offer, and was happy she had the option.
If you get the opportunity to talk privately with tenure track folks, ask about the review process and how it works. That will probably give you a good sense of how healthy the process is, or not.
I was at one meal recently and an adjunct from our department came. In a way, that's probably healthy, no? It means that our adjuncts don't feel completely abused by the department and want to participate in our hiring process. But in a way it felt weird, too, though I'm not sure I can put my finger on why. (It's probably something some folks would perceive as my snobbishness about adjunct faculty and how they aren't full members of the department. And maybe there's something in that. They won't, for example, ever work in the personnel committee, review colleagues, or be subject to the same review process [their review process is different]). (I'm pretty sure the candidate wasn't aware that this person was an adjunct; I don't think it was mentioned, and there was lots to talk about more generally.)
What about making an impression? At a recent candidate dinner, there were five of us faculty folks: one silverback, one silverback in training, one so not alpha male, and two of us women. The candidate quickly figured out that the so not alpha male wasn't a silverback or silverback in training, and basically focused hir attentions for the evening totally on the silverback and silverback in training. It was rude, I thought. But the silverback and silverback in training found the candidate wonderfully warm and friendly, I'm sure.
I've decided not to put in a response sheet on this candidate. I only met hir at the dinner, really, and then we didn't much interact. I was unimpressed at the way the candidate treated the women and non alpha male. On the other hand, I recognize that it's a really valuable life-skill to court silverbacks and silverbacks in training; I happen to be dismally bad at that, and to generally have moderately poor relations with silverback types. So maybe this candidate is just so much more savvy and skilled than I am that zie will do great. But even with my poor skills, I know that putting down on a piece of paper that the candidate was rude to the women and the non alpha male will 1) be totally denied by the silverback types (who would mansplain that they're feminists and there's no sexism anymore, not in our department, anyway), and 2) will only make them cranky at me.
(I haven't talked to the other woman or the so not alpha male since, and I probably wouldn't bring it up when I do.)
Anyway, I guess I'd suggest not turning away from someone just because you perceive them to have less power than someone else, while at the same time making the person you perceive to have the most power feel like you find them fascinating. Good luck with that.
I mentioned earlier a response sheet. Here's how my department works the decisions.
Just before the candidate visits, each department member receives a small packet on the candidate. The packet includes the candidate's letter of application, CV, campus visit schedule, and a response sheet. The candidate does hir visit, and everyone is encouraged to participate as much as they can, to see the talk, go to meals, meet in the informal meet and greet (during an open half hour in the department coffee room), and so on. Then the chair sends everyone a reminder to please turn in the response sheet by a given deadline.
On it, we're asked to write our names, to check off our interactions with the candidate, and then to answer a few specific questions about our impressions. What would we think of the candidate as a teacher? a scholar? and a colleague? Then finally, we're asked to check either yes or no, do we think we'd like to hire this candidate. And right next to that question, there's a statement that this isn't a vote for one candidate over another, but just about this candidate. So if there are three candidates visiting, I could vote "yes" for each of them, indicating that I think that each is someone we could do well to hire.
Then the committee meets. Diverse committees diversely handle it, but the idea is that the committee should read all the responses about each candidate, taking into account the respondent's interactions and so on (so someone who went to the talk weighs more heavily on the scholarly question than someone who didn't, though it's not nearly so numerical). Then the committee makes a recommendation to the chair.
Then the chair takes that information (with the written reasoning), and writes hir recommendation to the dean, and on up the line. If we're lucky, things move up the line and back down quickly, and the chair is able to make a happy phone call to the candidate to make an offer. (We have an administrator who is sometimes slower than we'd like to make decisions, alas. So sometimes things don't move up and down nearly as quickly as we'd like.)
Obviously, we'd like to be able to make our offer quickly, and to have the candidate accept it with joy. Usually, the candidate needs some time to think about it and perhaps negotiate.
I'd like to end with one final thought about my department's hiring practices. Sometimes, our first choice candidates don't accept our offer. Then we may move to our second choice or invite another candidate to campus. Because we made that first offer based on a preference, often not a hugely overwhelming preference, we're usually pretty happy with the next candidate. What I'm getting at is, here at least, someone who's waited a bit and realizes that they probably weren't the first choice shouldn't feel unwelcome at all. It's likely that a large number of people put "yes" on the should we hire this person question (because if people had put "no" then we would have moved to another campus visit).
We may really love both candidates, but the committee has to choose one to recommend, and so it does. But the second person is probably a great fit, too. Still, we can only hire one person. At this point in the decision, it's almost never the case that one person bombed the visit. Rather, it's usually that one person impressed more people a little more.
I guess I have one more post in mind, after all.