Being a Good TA: Advice from Undergrads & Profs

[StatusX]

Can anyone give me some tips on being a good TA? This is my second year doing it, and both times the attendance has dropped off a lot after the first few classes. I realize this is common, but it seems to be dropping off for me more than other TAs of the same class (down to one or two students, whereas most others still have 5-10).

I never really went to section as an undergrad, so I don't really know what they expect. I've been trying to do mostly example problems, but I do get bogged down in theory here and there. I've always liked explaining stuff, but I feel like if people don't have the same interest in the material as me, they won't like my approach to it (which emphasizes a deeper understanding rather than memorizing formulas). I'd like to be a professor one day, so this is something I want to improve at (plus it's really awkward when one or zero students shows up). Anyone have any advice? Undergrads, what do you like about your favorite TAs?

 

[Topher925]

I only teach labs but what I found works is to just be strait to the point. Usually, I just tell the students what they need to know and how to do it so we all can get out of there at a descent time.

As for basic teaching practices, just make sure you speak clearly and you make concepts as simple as possible to understand. This usually involves organizing your thoughts before hand and preparing yourself to present them.

 

[Mike H]

My three semesters as a TA "benefited" from the fact that I had to give a quiz nearly every week, which in the end added up to about 10% of the student's final grade for the lecture, so I'm figuring that's not in play with your current situation.

What is it that you're TAing? Is it an introductory level class, or an upper-division course that is intended for students majoring in the field? I found that when I was a TA for introductory general chemistry, I frequently had students asking good, germane questions for which the detailed answers could take more time than I had allotted. I would ask them, if they were really interested, to either stay after class or come to my office hours. Sometimes they would, sometimes they wouldn't. More involved discussions of fine points can make more sense in the context of an upper-division course, where the students will ideally require less of your time in explaining the mechanics of solving problems, and more about understanding the material and formulating how to attack a problem they may encounter.

I generally structured my recitation/discussion section with the following model: a brief mention of what was covered in section last week, followed by a few words about the current week's lectures, and maybe following up on something that wasn't covered in detail in lecture. There'd be a few example problems (I always had four or five worked out in detail from the text or occasionally an old exam or my own undergraduate text), time for questions, and then the quiz. Now, the example questions should be thoroughly worked out and checked through, not just sketched out so you don't get tripped up or get tempted to digress extensively.

I hope some of this was helpful. I was a bit nitpicky about things as a TA, but it helps to have a plan in mind for each week and try to hold to it. Sometimes you'll have to skip over things or simply not have time, but you can actually get a good deal covered if you go in there with a plan of action.

 

[Moonbear]

What type of class are you a TA for? Are you giving the primary lectures, or is it a recitation section?

One thing that's important is that students need to feel they are getting something out of showing up for the class. That could be that they get points for assignments turned in or pop quizzes given, or it could be that you give them really useful information for passing their exams.

If you're teaching a recitation section (that's somewhat what it sounds like), be careful of your tendency to get "bogged down in theory". If the students are just trying to understand how to solve a problem, and you're giving them another lecture on theory without ever getting to the point of how to solve the problem, that's not going to help them, and not inspire them to show up.

When you get asked a question, do you organize your thoughts and take them through it little step by little step? That's a very helpful thing.

What about your level of interaction? Do they just say, "I didn't understand problem 5.17" and you then talk about it for 10 minutes, or do you ask them questions back? Think about the way we do homework help here, and apply that to your real class setting. Rather than just going through the problem for them and rambling on about theory, ply their brains for that information. For at least some of them, it's stuck in there somewhere, and your job is to help them find it and use it. It also keeps them awake.

Actually, take it a step further, when they ask a question, tell them to put on the board what they have so far, and have the rest of the class jump in if they see where a mistake starts.

 

[buffordboy23]

I realize that this is an old post, but I will be a TA in my school's physics department come next fall and have thought much about how to proceed with this role of responsibility.

From my experience during recitations, I saw that the common approach is that the students show up and ask questions about their homework (which is eventually turned in as a grade) and the TA offers hints or works through the problems. I am turned off by this approach because the students need not seriously think about their homework before coming to class.

I offer the following model and would appreciate any feedback. This model rests on the the premise that physics is not an easy subject and that the student must "practice" to become more efficient at solving problems. Practice encompasses the following: understanding the conceptual aspects of the subject, thinking about one's approach to solving problems, and studying a variety of interesting problems.

With that said, when students show up to my recitation I will not devote any time to the assigned homework. Instead, the students will be organized into groups and will receive a worksheet aligned with the content of their professor's lectures to complete. This worksheet will have simple conceptual questions and a few applicable problems that require more time. As the students work, I will proceed around the room to answer group questions. By doing this, I will see where the common problems in conceptual understanding are and can target these briefly by reviewing those answers. The same applies to the harder problems, but in this case I will review how to solve these by how to go about thinking about the problem, offering problem solving strategies (e.g. generalization, adding auxiliary elements, etc.), and requesting student feedback.

Questions regarding homework will be answered outside of class by email or during my office hours and only if the student has showed considerable thought about the problem. If these requirements are met, I will ask questions that help to guide their thinking process not solve the problem outright as I have seen some TAs do.

 

[cristo]

With that said, when students show up to my recitation I will not devote any time to the assigned homework. Instead, the students will be organized into groups and will receive a worksheet aligned with the content of their professor's lectures to complete. This worksheet will have simple conceptual questions and a few applicable problems that require more time. As the students work, I will proceed around the room to answer group questions. By doing this, I will see where the common problems in conceptual understanding are and can target these briefly by reviewing those answers.

It's not a bad idea, but I suspect you will start to have fewer and fewer students turning up to your classes. You'll be basically doubling the amount of work they have to do for your course, and I can't imagine many will like that!

 

[buffordboy23]

Cristo,

Thanks for the response. Your right, there needs to be this incentive for some students. I thought about this after my original post too. I guess the time that I devote to homework would be secondary to everything else planned and would occur during the last 15 minutes or so of a 50 minute recitation.

From my experiences, the recitations that I attended were not helpful at all. It soon became apparent that the students were relying upon the TA to get their homework done. The TA would practically go through every homework problem (even simple procedural problems and whose answers were in the back of the book) and then we would turn our homework in afterward so the students all did very well on their homework scores. This is bad in my opinion because the student who worked hard to solve all of the problems prior to the recitation is never distinguished from the other students--for these students to distinguish themselves, it can then only be through tests--and the lazy student is rewarded, never learns the subject well, and acquires a poor foundation for harder subjects in the future. I yearned to learn something fresh or the current subject in a different light, but my needs were not met.

It's these experiences that prompted me to visualize a different route. Yes, the students would be doing more work but this work would only be during the recitation. If we don't get through all of the planned material, then it does not matter. I want to focus on common misconceptions that are directly applicable to the assigned homework problems. As a simple example, the teacher may assign a problem that involves the gravitational potential energy of a satellite in orbit. Many students may make the wrong assumption and use U=mgh. Therefore, in recitation let's not solve the professor's assigned problem outright. Instead, let's build an understanding of when to use which form of gravitational potential energy by looking at when and how the calculations of both forms diverge from some acceptable accuracy.

 

[physics girl phd]

At the University of Colorado at Boulder, recitation time is a bit "lab-like" in nature -- and they use the http://www.phys.washington.edu/groups/peg/tut.html" pretty extensively.

One interesting thing I read about recently was http://prst-per.aps.org/abstract/PRSTPER/v5/i1/e010102" (see the link to a journal article). It seems like an interesting idea... and you might be able to convince students their grades will be higher by showing them the data. I know a university that did a similar thing during recitation by having a reusable white-board for each group of students.

Still, to get students to take recitation time seriously, you probably do need to have it built into the grade somehow. You may want to talk to the professor you are working under to see if this is possible (you will have to work this out beforehand). In the case of concept mapping, you can have students make a concept map, then work through a homework problem using their concept map (perhaps showing that the problem can be solved by a few different methods, and having students share their group's map with the rest of the class) -- giving groups points for the map, and the problem/discussion.

 

[buffordboy23]

Physics Girl,

Good post. Thanks for the resources. I will look into them some more. These resources promote student learning because they require the student to play with the physics. The students are actively engaged and this is key to building understanding. Its akin to solving a tough problem. At first, you have a rough idea of how to develop the solution, and by trial-and-error of different ideas, the problem becomes clearer and the solution is more obtainable based on your new understanding. In a traditional lecture or recitation, the students are passive learners. They listen rather than do, so understanding tends to be at its most shallow level.

 

[Choppy]

This seems appropriate...

http://www.phdcomics.com/comics/archive.php?comicid=5

 

[Howers]

The problem I find is that TAs think they are professors. They aren't, and they shouldn't lecture in tutorials. They should just write full solutions to some of the problems because this is what students need. We already get theory from out textbooks and lecture notes, what we don't get are full solutions. We don't need a TAs interpretation of the material or some problem he pulls out from a grad text. Don't try to make classes more theoretical or more interesting because you think people are bored. They won't listen to you anyway. What they will do is look at your solutions to problems come time when they start assignments.

 

[whybother]

This past year I had two tutorials, one where I was expected to, somewhat, lecture, and the other which was a question and answer period.

If it's just supposed to be Q&A, attendance will really depend on how well you answer the questions that students bring to class to ask. At the beginning of the term, no one asked any questions, so I just had to choose problems from their textbook that I thought would be illustrative for their problem sets (my courses were math, not physics, but the basic idea is the same here) and solved them. Once they got a little more courage up, students brought questions they wanted solved with them. As long as you can solve whatever they ask, they seem to keep coming back with more. I remember not going to tutorials as an undergrad if they TA wasn't prepared or didn't know the material well himself. Even if I don't know how to do a problem right away, faking confidence really seems to help get me through it. Students need to believe you know the material 100% it seems, for them to take you seriously.

My other tutorial I had to cover material their prof didn't get to in class and it was difficult. I had great attendance though, but that's just because my prof did really incomplete lectures, but asked challenging problems on the midterms on material that he never would cover. Most TAing isn't like that though... hopefully you work for people who are less lazy than that guy. Profs who use TAs to fill in for their inability/disinterest in teaching are the worst... although, they do ensure a great attendance rate in tutorials.

 

[Moonbear]

When I was a student, I admit I usually skipped recitation sections unless they had quizzes. Usually, they were completely unhelpful since it was just the TA solving problems for us, and not always doing them correctly or explaining why they were doing what they were doing.

I think a recitation would work best if you make sure you can work through all the problems the students are assigned AND can explain the reasoning for each step. BUT, don't just go ahead and solve the problems for the students. Instead, let them either work together in teams or as an entire class (i.e., ask volunteers to go to the board) and let the students reinforce learning by teaching each other. I got a lot more out of classes where that type of learning happened. Your role would then be more of a facilitator to help if they get stuck, or explain why a step was done if a student knows to do it, but can't explain why they did it (maybe it was trial and error for them).

One possible way to do it would be to have students get together in groups for the first half of class and work together on any problems they were stuck on. Then, go around and see if there are still any problems they couldn't figure out as a group, or that they all were confused about, and spend the second half of class focusing on those problems that they can't figure out on their own, maybe only taking time to re-teach the major concepts on one or two problems that way rather than trying to guess at what students need and boring them with a repeat lecture on the stuff they could have solved on their own.

 

[sx70]

Excellent advice from an excellent mathematics teacher

http://www.math.uiuc.edu/~reznick/ciu.html

 

[Moonbear]

One interesting thing I read about recently was http://prst-per.aps.org/abstract/PRSTPER/v5/i1/e010102" (see the link to a journal article). It seems like an interesting idea...

Somehow I missed this before. Concept mapping isn't very new at all, but still remains a solid method for learning. We were using that over 15 years ago when I first started teaching freshman biology, and still use it now in our PBL course for med students. So, yes, I would agree that teaching students to use concept mapping is very useful...and a great study skill for them to develop.

 

[maze]

Concept mapping sounds good, but the paper presents a hugely flawed experiment. They let students decide whether or not to come to the concept mapping class, and those who did got better scores. Self-selection bias...

To do a proper experiment they would have to run 2 classes of separate students, with everyone required to attend their class. In one class they do concept mapping, and in the other class they spend equal time doing a more traditional physics review. Then see who gets better scores.