31

I am head TA in a 300-student class, with 12 other TAs. Currently I do the following things to motivate my TAs:

  1. We have an internal Slack channel called #shoutouts where I mention TAs every time I notice them going above and beyond.

  2. I post Piazza statistics to the Slack channel every week so that we can see who is posting the most, and then rank TAs in order of how much they posted (currently, a lot of students say that Piazza is very helpful)

  3. I post grading statistics to the Slack channel so that people can see who grades the most (and least) submissions, and what is the average grade they give.

  4. I have a feedback form for students to give feedback to the course staff, and I send a message to the #shoutouts channel every time a student says a TA has been especially helpful.

  5. I tell TAs they are getting evaluated by the instructor (which I think a lot of the TAs didn't know).

However, I have been told (by someone not involved with the class) that I am "overquantifying" things. Also, even with all of this, there are four people who I would describe as "bad TAs" (and five people who I would describe as excellent TAs, and three people who I would describe as good TAs).

What are your best ways of motivating TAs?

  • 185
    I'm a TA. All the ways you are causing some sort of competition between TAs would drive me nuts and be extremely de-motivating. If a TA isn't doing enough, let them know privately. Creating a public ranking amongst TAs strikes me as really unpleasant. – Bamboo Mar 8 '18 at 21:39
  • 50
    The public ranking kind of comes across as being similar to that restaurant manager in Offices space. "You don't have enough flair on your uniform". Sure stats are good in private, but ranking a job that should be teamwork seems counter productive. – scrappedcola Mar 8 '18 at 22:12
  • 37
    Are you trying to desperately get rid of half your TAs? Because 2 & 3 is how you get rid of the "bad" ones. If you want them to do equal work, assign them equal work. If they are able to choose how much they work, you end up with an unequal share, that is just how it works, especially if they can not compete (Why bother becoming #1 if you do not have the time, better do nothing) What is the relevance of showing the students the average grade the TAs give? At best its irrelevant to the students and at worst it gives them an emotional cop-out ("It wasn't that bad, it's the stupid TA"). – Polygnome Mar 9 '18 at 0:42
  • 55
    Do you evaluate yourself publicly with the same paternalistic enthusiasm you apply to the others? – Dan Fox Mar 9 '18 at 6:23
  • 58
    This question is the poster child of "The road to hell is paved with good intentions"... – DonFusili Mar 9 '18 at 7:51

10 Answers 10

50

However, I have been told (by someone not involved with the class) that I am "overquantifying" things. Also, even with all of this, there are four people who I would describe as "bad TAs" (and five people who I would describe as excellent TAs, and three people who I would describe as good TAs).

I presume that your good TAs are always on top of the lists you publicly post, and your "bad" TAs are on the bottom? Why do you think this public shaming will motivate them?

Here is what I would do:

  • Centrally, I would stop posting things that make some TAs feel good at the expense of making others feel bad. The shoutouts are nice, but for statistics etc. I would only post who was on top (or maybe the top-3 or something), and only if it is not always the same people. Try your best to rotate your statistics in a way that everybody gets some love.
  • Assign work. It appears to me that TAs are somewhat free in choosing their own work (since you give stats on who graded how many assignments). This naturally leads to some people slacking off and others, more or less happily, picking up their tasks. A simple solution to that is to make up a schedule.
  • Validate that your "bad" TAs are objectively not doing a good job, and are not just weaker in comparison to your exceptionally great other TAs.
  • If you feel your bad TAs are actually objectively lacking, use the central management principle of praising in public (which you already do, to some extent) and criticising in private. Talk to the TAs that you are not satisfied with in private and explain what you think they should do to improve. Give actionable tips - "you should know more about the subject" may be objectively true, but will also not change in the duration of the course. My impression is that with your statistics you attempt to shame people into improving without having to have an awkward conversation with the underperformers. Unfortunately, this is not how management works.
  • Above all, it pays to keep in mind that not everybody is the same and has the same constraints etc. No matter what you do, not everybody will go above and beyond. If 5 out of 12 people regularly go above and beyond and most others are also doing a good job, you already have a strong team. Don't downplay this, not to yourself and definitely not to the team.
  • Thanks for the suggestions! I used to post the top 3 Piazza people only, but then some people accumulated much more Piazza posts than others, so they were always top of the ranking. Now that I post everyone every week, it is easier to calculate how many posts they are making in that week specifically (because I can just look at last week's and do subtraction). I guess I could keep it in a separate spreadsheet. – Head TA Mar 8 '18 at 22:20
  • 1
    Everyone except one TA has been positively mentioned in the #shoutouts channel at least a few times (in part because there are a lot of different reasons you can get recognized). Some of the "bad TAs" actually have a lot of mentions, because the reason they are bad is because they are inconsistent (sometimes go above and beyond and sometimes just avoid work). There is one TA who is a real outlier and was mentioned in #shoutouts much more than anyone else. – Head TA Mar 8 '18 at 22:22
  • @HeadTA What subject do you teach? By the mentality used in assessing them, I suspect it's something related to business or management, and you're trying to apply some theory principles to your team... By the many comments, and answer by Dmitry, I hope you've already realised that it doesn't work to create too much internal competition. Treating people just like capital, as simple minded factors of production doesn't work. There needs to exist some respect and dignity. I wish you success in the near future, though. – An old man in the sea. Mar 11 '18 at 0:10
  • @Anoldmaninthesea. Nah, those were business theories a good bunch of decades ago (taylor e.g.). I'd bet on something technical were people are interested in numbers and statistics – DonQuiKong Mar 11 '18 at 13:07
74

I think the best way to motivate a person in academia is to give them a bit of job security, some funding which enables them to follow their interests, and a great team to work with. I am not sure if you can realistically give "your(sic!) TAs" any of this.

In absence of solid motivating factors, you are trying to engage the team in some form of internal competition, which may motivate some people and completely put off others. For example, I personally would be puzzled as to why should I compete for some statistics or evaluations which look rather meaningless unless it has any effect on my promotion or chances of future employment.

Most people I worked with responded well to:

  • smart goals
  • positive example
  • cutting the crap (including crap meetings)

In my experience, teaching is rather a pleasant activity and there is a lot of gratification in the process of working with students. An excessive amount of paperwork, bureaucracy and management can only spoil the fun.

33

I have heard beatings can work well:

There's a famous expression attributed to Captain Bligh of the stricken HMS Bounty, who was overthrown in a mutiny. While flogging sailors for small misdemeanours, he is said to have declared: "The beatings will continue until morale improves."

  • 3
    This is exactly the point... – paul garrett Mar 9 '18 at 16:02
  • 1
    Bligh was in fact considered a very moderate user of disciplinary measures by the standards of the day. His image as a tyrant is largely down to a concerted campaign by powerful relatives of some of the mutineers, amplified by hollywood's script writers. – StephenG Mar 11 '18 at 18:35
  • We have received flags about this post being inappropriate. Readers: this post is a joke. Please do not beat your TAs, or students, or advisors, no matter how much you want to. If this concerns you please read the fantastically titled "Stack Overflow: where we hate fun" blog post. – eykanal Apr 17 '18 at 14:35
14

I have been told (by someone not involved with the class) that I am "overquantifying" things.

That person is correct. Besides the quantifying thing and publicly showing these numbers, what do you expect? 12 excellent TA's? Not gonna happen. There are always high, middle and low profiles in a group of people. Sometimes people out perform (themselves), sometimes only for a limited time frame.

In stead of posting all these statistics, try talking with your TA's. Aknowledge there effort, give feedback, give advice. That's how you motivate people.

  • For example the reputation points system and badges on stackoverflow; One colleague and I are competing for the most reputation points, because we like to. Other colleagues don’t. And that is fine. The moment my boss would grade us, compare us, and enforce us to have x amount of reputation points. That is the moment my colleague and I will dislike the reputation point system. – roel Mar 15 '18 at 8:57
12

You have good intentions, so please forgive me for being blunt, but right now, the best thing you can do to motivate them is to stop doing all the things you're currently doing to motivate them.

The good TAs are going to be the ones who like teaching and thus don't need to be motivated. The main downside of teaching is that it takes away from research time, and you're currently making that worse. I imagine that "being in competition with my coworkers" would have also made the list of downsides if any of my supervisors had ever done that.

There are, however, things you can do to improve how the team functions together. The one potentially useful useful thing in your list is the grading statistics. Not because it can be used to rank TA quality, but because it's important that everyone is grading according to the same standards so no student gets an unfair advantage (or disadvantage) based on who graded their work. So keep those statistics, and if you see a few outliers, you might have everyone meet up and coordinate how harshly things should be graded. One meeting before each major assignment would probably suffice.

8

My 2 cents.

Probably won't be a popular opinion here, but I've been head TA numerous times, along with being a TA (or equivalent) and this is what I have personally found.

  • A good TA is by their nature a good TA. It's an inherent property. Incentives or disincentives will never transform a bad TA into a good one. A good TA can however get burnout if working conditions are bad, or if they consistently feel that their efforts aren't acknowledged or appreciated.
  • Giving good TAs space is good, and the tools to do the job. They don't need someone looking over their shoulder. However, completing an arduous task well, and being told they did "a great job" will go a long way.
  • Acknowledging that certain tasks are a pain will definitely motivate TAs, while an attitude that such tasks have to be completed because they're "part of the job" will not. On the same note, making it clear that, as head TA, you are not merely offloading responsibilities, but making a significant contribution yourself will hugely motivate good TAs. Bad TAs will see you doing work as an excuse to swing the lead, but as I have already said, they're a lost cause anyway.
  • Attempting to reduce busy-work by TAs will always be appreciated. Are there tasks that have to be completed by TAs that don't really aid students? Can you ask the professor/lecturer that such tasks be removed? Are the rubrics for corrections too convoluted? Can tasks be streamlined? Reducing the amount of work have to be done on terrible software like Blackboard will help.
  • Depending on the institution there may be some way to have some monetary bonus at the end of the semester given to good TAs. That's not "a prize to one TA" but rather some bonus to all the members on the team that pulled their weight. It doesn't have to be much. It doesn't have to be a competition, just an acklowledgement that your team did good, and that their work is appreciated. A distinction, if you will, between the bad TAs who are there just for the money or because they are forced to be, and those who really care about doing a good job in the first place.
  • Just to reiterate: there is no cure for a bad TA. You can cajole them, you can monitor them, you can encourage them: there is no point, they will only work as long as you are monitoring them, and no longer.
  • 7
    "A good TA is by their nature a good TA. It's an inherent property." No, they have to learn how to do it. – Anonymous Physicist Mar 9 '18 at 21:51
  • 3
    @AnonymousPhysicist enthusiasm, dedication, dependability can't be learned (or at least in the time frame that we're talking). These are the most important aspects, besides actual knowledge of the subject that's being taught, of course. – Stumbler Mar 9 '18 at 22:19
  • 4
    Those things are definitely learned. I would guess the time scale is highly variable. And they are much more important than knowledge of the subject. – Anonymous Physicist Mar 10 '18 at 0:46
  • 7
    One can learn how to teach, one can learn how to act professionally, and many people do both. – Dan Fox Mar 10 '18 at 9:45
  • Yea, I'm going to overnight 24 years of behavioral habits in 10 weeks of managing you. Does that sound realistic Anon? – Skyler Mar 15 '18 at 20:37
7

Not everyone is motivated by the same things, and you may be encouraging behavior you don't actually want. Let's examine your current assumptions that I'm guessing you hold if you think posting charts with names is effective: * Everyone wants to be the best among their peers * Everyone wants to be known as the best * Everyone feels they can be the best * Doing the most of some activity means you are the best at it

Here's the truth though: * Not everyone is interested in competing * Not everyone enjoys public recognition * If there's a best, there must be a worst, and it's probably going to be the same people in the top 50% and bottom 50% on a regular basis * Making more posts or slapping a grade on the most papers does not mean you are actually doing your job better

People who don't like competition will ignore your graphs. People who dislike being publicly recognised as being better than their peers will actively try to score lower. People who feel they can't compete even if they wanted to will come up with an excuse to ignore your graphs. Those who do like all those things will behave in a way that strictly improves their metrics, even if it is detrimental to the students (think many short posts with a minimum of effort rather than long, thoughtful, helpful posts).

There's almost no way posting your metrics will improve performance. You need to very carefully consider which behaviors you want to reinforce, how you will reinforce that behavior for each person. Being publicly declared the most helpful TA will thrill someone or mortify them. There is no single tactic that will motivate every person. You'll need a combination of talking to your TAs and trial and error to figure out what works.

I would strongly recommend avoiding tracking how many of X each person does. Your issue isn't that some people are grading more assignments than others in the same amount of time; the problem is some people are taking long breaks. And why wouldn't they? Grading isn't fun, and a break has an immediate and guaranteed reward of not having to grade during the break. Instead, change your process so one person taking breaks doesn't cause another person to do more work, or limit the breaks to specific times when the whole group breaks together.

In short, focus on the behaviors that you want to increase/decrease, then figure out what each person considers a reward. Then give out those rewards when you get the behavior you want.

There are whole books on this topic if you really want to get into it.

  • 2
    Not everyone motivated by same thing. Very good observation +1 – mathreadler Mar 12 '18 at 7:39
4

Give them important things to do. Show them how their work helps students learn.

Grading and "posting" do not necessarily help students learn.

This is more the responsibility of the instructor and institution, rather than a "head TA".

1

It can be difficult for TAs to feel invested in a course; often what works best is to sit down with the TA at the beginning of the semester, ask them what they want to get out of the TAship that semester. If the TA can feel involved with course development -- whether that means that they get to design one of their own lectures -- or have some form of input to make it a worthwhile experience for themselves, then that can mean more for retention and performance than anything quantitative.

0

Only my opinion, but I don't see many posts that push giving TAs helpful ideas, suggestions for developing their teaching skills. Intended for those TAs who happen to be doing the teaching parts of it all. Many who have never taught anything before to anyone can feel hopelessly lost at times, as will their students. Suggesting ideas to them for designing exams, marking papers, calculating grades, and all that could go a long way for providing assistance to any TA. Plus, their students would likewise benefit from a trained instructor. When TAs or anyone start to feel competent and confident in their work, that pays for itself at least in part.

protected by Alexandros Mar 11 '18 at 20:38

Thank you for your interest in this question. Because it has attracted low-quality or spam answers that had to be removed, posting an answer now requires 10 reputation on this site (the association bonus does not count).

Would you like to answer one of these unanswered questions instead?

Not the answer you're looking for? Browse other questions tagged or ask your own question.