A TA appointment is a fixed number of hours, but TA work expands to fill whatever time you give it. Forum questions arrive all day, grading always takes longer than planned, and research gets the leftovers. It is common: in a survey of graduate TAs at one university, 62% said they worked overtime in their teaching role, and 72% said teaching left them too little time for research. This page is about keeping the job inside its hours: knowing where your time actually goes, guarding your research time, setting limits students can see, and raising the alarm early when the work doesn't fit.
Defaults that work. If you only take four things from this page:
- Log your real hours by task for the first few weeks. Your estimates will be low.
- Check the forum and email at set times, and tell students when to expect answers.
- Protect research blocks the way you protect section times.
- Tell your instructor the first week you're well over, with your log, not at the end of term.
Know where the time goes
- Log your hours by category for the first three or four weeks: section, preparation, office hours, grading, forum and email, meetings. A spreadsheet or notes file is enough.
- Expect your guesses to be wrong. People reliably underestimate how long their own tasks take. The fix is to plan from last week's log, not from what you hoped.
- Agree a budget with your instructor early: roughly how many hours go to each task, and how fast grading should come back. Revisit it after a few weeks with your log.
An example week
A made-up example, to show the shape; your course will differ.
| 20-hour appointment | Hours | 10-hour appointment | Hours |
|---|---|---|---|
| Attend lecture | 3 | Lead one lab | 2 |
| Lead two sections | 3 | Office hours | 2 |
| Section prep, time-boxed | 2 | Preparation | 1 |
| Office hours | 3 | Staff meeting | 1 |
| Staff meeting | 1 | Grading, one block | 3 |
| Grading, two 3-hour blocks | 6 | Forum, once a day | 1 |
| Forum and email, two short checks a day | 2 | ||
| Total | 20 | Total | 10 |
In an exam week, the grading row grows and something else has to shrink. Plan which one in advance.
Keep TA work in its blocks
- Batch the forum and email. Check at set times, for example twice a day, and close it in between. Interruptions make people work faster but with more stress, and every switch back to research costs time to get going again.
- Post a response window students can see: "within one business day," and no replies late at night. Expectations nobody stated become 24/7 by default.
- Set a cutoff before deadlines and exams, such as no new questions answered after 8 pm the night before, and answer repeated questions once, for everyone.
- Time-box preparation. Decide how long section prep gets, and stop there.
- Use quiet office hours for grading or forum work you can drop the moment a student arrives.
- Set a grading budget per submission, time a sample, and grade by question in long blocks. Grading Faster Without Cutting Corners has the details.
Protect your research time
- Put research blocks on your calendar and treat them like a section you're teaching: at least one long block on days with no TA duties.
- End each TA block at a clear stopping point. Write a one-line "next step" for the task you're leaving. Research on task switching finds that unfinished tasks keep pulling at your attention, and a defined stopping point helps you let go.
- Don't do TA work in the middle of research time, even "just one quick answer." The switch costs more than the answer.
Plan the heavy weeks
In the first week, mark the exam weeks and project deadlines on your calendar. Move research goals away from those weeks, plan extra office hours as sign-ups rather than open-ended blocks, and agree with your instructor which tasks shrink when grading peaks.
Warning signs, and what to do
| Sign | What to do |
|---|---|
| Over your hours two weeks in a row | Bring your log to your instructor and agree what gets cut or changed |
| Most of your TA work happens at night and on weekends | Reset your check times and response window, and post them |
| No real research block in two weeks | Re-block your calendar, and raise it with your instructor or advisor |
| Grading disputes feel personal | Use a written regrade process and bring your instructor in on hard cases |
| Exhausted after every workday, dreading your inbox | Talk to your advisor, and use your campus counseling service. Graduate students report anxiety and depression at far higher rates than the general population; you are not alone. |
If the work truly doesn't fit the hours, that is a workload problem for your instructor and department to fix, not a time-management failure on your part. Working With Your Instructor covers what the contract says and how to raise it.
Saying no
Requests outside your role will come: an extra section, a new assignment to write, a favor for another course. "I'm at my hours this week. If it's a priority, which of these should I set aside?" is a complete and professional answer.
Further reading
Guides
- UC Berkeley GSI Teaching & Resource Center, Time Management Strategies: a teaching log, set email times, cutoffs before exams.
- University of Michigan CRLT-Engin, GSI/IA Resource Packet (PDF): includes a time-management worksheet.
- University of Wisconsin–Milwaukee, Teaching Assistant Workload: logging time and reviewing it with your supervisor.
- University of Waterloo CS, TA time allocation form (PDF): an example of agreeing hours per task up front.
Research
- "Dedicated and overworked: Graduate Teaching Assistants at the London School of Economics" (2022) (PDF): a survey of 173 graduate TAs.
- Leroy, "Why is it so hard to do my work? The challenge of attention residue when switching between work tasks" (2009).
- Rubinstein, Meyer and Evans, "Executive control of cognitive processes in task switching" (2001).
- Mark, Gudith and Klocke, "The Cost of Interrupted Work: More Speed and Stress" (CHI 2008) (PDF).
- Buehler, Griffin and Ross, "Exploring the 'planning fallacy'" (1994): why we underestimate our own tasks.
- Evans et al., "Evidence for a mental health crisis in graduate education" (Nature Biotechnology, 2018).
- Levecque et al., "Work organization and mental health problems in PhD students" (Research Policy, 2017).
Related pages: Working With Your Instructor, Grading Faster Without Cutting Corners and Answering Questions Online.