Our situations are a bit different - I'm in science, while she'll be looking for programs in performing arts, and will likely be put in charge of a full course rather than a lab section associated with a course. What isn't likely to be too different, though, is that she'll be entering with little or no preparation in how to teach a class, aside from maybe a few days in a "crash course" seminar (like I had in August).
I didn't actually find that seminar to be very helpful, mainly because it seemed geared toward students who would be teaching introductory courses in which they were mainly responsible for choosing the course layout - developing their own syllabus, readings, assignments, etc. I don't do that - we get pre-written labs that roughly parallel what the students are learning in lecture, and maybe do a little additional lecturing on the side. So attending a seminar where we learned "think-pair-share" and other active learning techniques was neat in theory, but not really applicable to a structured setting like a lab course. (We did have one short session on teaching labs, but I was so annoyed by the fact that the invited TA wouldn't shut up and stop talking over the professor who was running the thing that I didn't get much out of it.)
Anyway, the gist of it is that I've either been relying on advice from other TAs or learning on my feet. This is both good and bad; the good is, you find out really quickly what works and what doesn't, but the bad part is, my first class of the week usually gets shorted in terms of quality of teaching because they end up being the guinea pigs. My TA group does get together and go through the labs and experiments beforehand, but there's a big difference between when we do the lab and when the students have to do it; we often find that something that makes sense to us is totally incomprehensible to the class. This doesn't do much for my nerves. I apparently don't show it, but I get really nervous about standing in front of a bunch of people and being responsible for them learning something properly. I'm always worried that I won't teach something in a way that makes sense, or that I'll get something wrong and then be embarrassed about it in front of a bunch of people who aren't all that much younger than me.
Fortunately, I realized something early on: The students don't know when you make a mistake, they don't usually remember it unless it was something really stupid, and for the most part, they're not out to get you. Some of them really want to learn and some of them just want to pass and finish their science requirement. A few will want to go on in the field, and you want to try and do the best job you can for their sake, so they don't have to relearn it later - but you shouldn't let it totally consume your time or attention.
So here are a few more of the things I've learned so far:
- Make sure you know what you're going to teach ahead of time. If you need a powerpoint, don't wait until five minutes before class to put it together; if you're talking about a reading, read it; if you're doing a lab, try it out so you know where your students are likely to screw up.
- If you're in grad school and you have a TAship, someone thinks you're competent enough to be in front of a class. Try not to doubt your ability to teach.
- Don't give them the answers, but don't not answer questions. Lead them; say "How would you solve this problem?" Make them go through their process and point out where they might be going wrong.
- Group work sucks - there's no way to get around that. Group work in class works better than group work on projects, though - if you're not around to watch, someone will always end up doing all the work and someone else will slack off. Be cautious about assigning out-of-class group work.
- Anecdotes and analogies keep people interested. (Students have much more fun learning about impact cratering when you let them throw pumpkins off the roof of the building, for example.)
- Unless you're really doing a terrible job, your salary doesn't depend on their end-of-semester evaluations. Not everyone is going to like you all the time, but as long as you make sure that they understand what you're teaching (ask them!), you don't have to be the "cool" TA. (Although that helps...)
- Don't stress over the ones that choose to fail - and by this I mean people who just don't make any effort, not the ones who do the work and ask for help and still don't do well (keep plugging for those). You're responsible for letting someone know that they're doing badly, but they need to come to you for help. Don't spend all your time chasing them around.
- Don't take yourself too seriously! Expect courtesy and respect from your students, and give them the same, but don't forget that you're all just students - joke a little, sympathize when they have midterms, ask how their plays or sports competitions or concerts went.