Graduate Teaching vs Research Assistantship: What Nobody Tells You
Most PhD applicants spend months agonizing over which schools to apply to, then accept an offer without fully understanding whether the funding attached to it is a teaching assistantship or a research assistantship. That distinction shapes your schedule, your advisor relationship, your dissertation timeline, and your job prospects in ways that dwarf the school's ranking. Same tuition waiver on paper. Similar stipend. Same 20-hours-a-week expectation on the surface. But completely different worlds underneath.
Here's the honest version of that comparison.
What TAs and RAs Actually Do
Teaching assistants are funded by the university department. You run discussion sections, hold office hours, grade assignments, and sometimes teach your own recitation. In STEM fields, that typically means supervising a lab section for 25 to 35 undergraduates each week. In humanities, you're likely reading undergraduate essays on Tuesday nights. Either way, you're earning your stipend through pedagogical work that is separate from your dissertation.
Research assistants are funded by a faculty member's grant. Your job is to advance that professor's research program — running experiments, building code, collecting survey data, writing literature reviews, or whatever the active grant requires. The key promise: this work should connect directly to your own thesis. You're theoretically getting paid for something you'd be doing anyway.
Both positions average around 20 hours per week and come with a tuition waiver. But the funding source, the power dynamic, and the opportunity cost are completely different situations.
| Teaching Assistantship (TA) | Research Assistantship (RA) | |
|---|---|---|
| Funded by | University/department budget | Faculty advisor's grant |
| Primary duties | Teaching, grading, office hours | Research tied to advisor's project |
| Tuition waiver | Yes, typically full | Yes, typically full |
| Alignment with your research | Low | High (in theory) |
| Funding stability | High | Depends on grant renewals |
| Best fit | Early PhD years, teaching careers | Later years, research careers |
The Money Breakdown
The financial gap between TAs and RAs is real, but smaller than the internet makes it sound. Cornell University's published 2025–26 graduate school stipend rates set the minimum 9-month stipend at $35,661 for both TA and RA appointments across all appointment types on the Ithaca campus. Same floor. At the Cornell Tech campus in New York City, cost-of-living adjustments push all appointment types to $59,029 annually.
Where the difference emerges is above that floor. RAs funded by large federal grants — NSF, NIH, DOE, industry partnerships — often earn well above the departmental minimum. A biomedical engineering RA with an NIH-backed advisor might earn $6,000 to $9,000 more per year than the department's baseline TA rate. In fields with limited external funding (comparative literature, philosophy, art history), RA and TA stipends often converge because professors simply don't have discretionary grant money to supplement.
The real financial variable isn't TA vs. RA — it's your field, your advisor's grant portfolio, and whether the university supplements pay at all.
Both positions come with health insurance at most R1 institutions. Both are treated as employee compensation, meaning Social Security and Medicare taxes apply. International students face an extra layer: stipend income and fellowship income are taxed differently, and treaty provisions vary by country. If you're on an F-1 visa, get advice from your international student office before assuming your net take-home matches the advertised stipend.
Career Trajectories: Where Each Path Actually Leads
This is where the choice gets genuinely consequential, and where most guides give you a non-answer.
For faculty positions at research universities, an RA accelerates your publication record. You're working on research full-time (in the best-case scenario), building proximity to your advisor's network, and generating the papers that search committees prioritize. But here's what advisors often don't volunteer: you still need teaching experience to be a competitive candidate at most institutions. Hiring committees at research universities look at your publications first; committees at liberal arts colleges look at teaching first. Swarthmore, Williams, and Grinnell explicitly want evidence that you can design a course, manage a classroom, and mentor undergraduates through difficult material. A CV with zero formal teaching experience raises a quiet red flag, regardless of your research output.
For industry and government careers, neither appointment type matters much to employers. What they evaluate is your technical output: papers you published, software you built, methods you developed, problems you solved. A data scientist at a biotech firm or a policy analyst at a federal agency doesn't ask whether you funded your degree through teaching or through a professor's grant. Your portfolio and your skills make the case.
The common mistake is treating TA vs. RA as a permanent identity. Most PhD students hold both types of appointments at different points in their degree. The Survey of Earned Doctorates consistently shows that median time-to-degree in STEM fields — where RA funding dominates — runs about 5.7 years, compared to 7.3 years in humanities, where TAs are far more common. RA funding correlates with faster completion, though field-specific differences make a clean causal comparison impossible.
The Time Problem Nobody Warns You About
A 20-hour-per-week TA appointment sounds manageable. But a demanding lab section, three simultaneous sets of papers to grade, and office hours stacked during midterm week can consume 30-plus hours in a rough stretch. Add two graduate seminars, and your dissertation sees nothing from October through December. This pattern is how "five-year PhD programs" turn into seven-year programs.
RAs have a different time problem. The work aligns with your research — theoretically — but you're accountable to your advisor's deadlines, not your own chapter timeline. If your advisor has a grant review in March and needs a conference paper submitted by February 15th, your chapter three waits. This arrangement feels genuinely productive when advisor interests and student interests converge perfectly. It can feel suffocating when they don't.
The misalignment problem is more common than people admit. A professor whose grant funds studies on machine learning fairness might admit a student interested in natural language processing. The student accepts the RA because it pays more and sounds research-adjacent. Two years in, they realize they've spent 800 hours on someone else's research agenda and their own dissertation is 40 pages long. The TA path would have been slower financially, but at least the dissertation direction would have stayed their own.
Hidden Risks in Each Role
Neither option is without landmines.
TA risks:
- Scope creep. Some departments treat TAs as inexpensive instructional labor and quietly overload them, especially in large intro courses.
- Dissertation stagnation in years three and four, precisely when output matters most for the job market.
- Repetitive assignments. TAing the same introductory course three semesters in a row builds no new skills after the first run.
RA risks:
- Grant dependency. If your advisor's R01 or NSF award isn't renewed, your funding can disappear mid-semester. In fields where NIH success rates hover around 20%, this is not a remote scenario.
- Research mismatch. The grant work and your dissertation topic don't always align as promised. You end up doing the professor's research while your own dissertation stalls.
- Power concentration. Financial dependence and research dependence in the same person is a meaningful concentration of risk. One difficult conversation with an advisor can jeopardize both your funding and your academic progress simultaneously.
One student quoted on HappySchools put it bluntly: "During RA you are a slave of your professor." That's too strong. But the underlying dynamic — that a single relationship controls your income, your research direction, and your degree timeline all at once — is worth taking seriously before you sign on.
How to Choose: A Real Decision Framework
The right call depends on three things: your year in the program, your career direction, and your advisor situation.
If you're entering year one or two without a committed advisor, a TA is usually the smarter default. You don't yet know whose research you want to join, and a department-funded stipend gives you runway to explore before locking in. You'll trade some research momentum for flexibility, but that trade is worth it early on.
If you have a committed advisor with active, multi-year funding, take the RA. Alignment between paid work and dissertation research is the closest thing to a productivity cheat code in graduate school. You're compressing two separate time commitments into one.
If your advisor's grant situation is unclear, ask directly before accepting: "When does your current grant end? Do you have a renewal application pending, or fallback funding if it isn't renewed?" A good advisor answers this without hesitation. Vagueness here is its own kind of answer.
Use this simple decision tree:
- Do I have a dissertation advisor confirmed?
- No → TA while you find one
- Yes → move to step 2
- Does their research align with my dissertation topic?
- No → TA to stay intellectually independent
- Yes → move to step 3
- Is their funding stable for at least two more years?
- No → TA, or negotiate a written fallback plan
- Yes → RA is the right call
My honest opinion: students are too quick to chase the RA label because it sounds more prestigious. It isn't. A TA who finishes a dissertation with a coherent argument and a teaching portfolio has better odds on the academic job market than an RA who published two papers in their advisor's subfield and never taught a class. Get both experiences across your PhD. That's the strongest possible profile, and it's achievable with a bit of intentional planning.
Bottom Line
- Start with a TA if you're entering a program without a confirmed advisor. The university-backed stability outweighs the marginal pay difference in early years.
- Switch to an RA once you have a clear dissertation direction and an advisor with demonstrably funded, multi-year grants. That's when the alignment between paid work and your own research actually delivers.
- Ask hard questions about grant timelines before accepting any RA offer. Funding uncertainty is the most under-discussed risk in graduate school, and nobody volunteers that information unprompted.
- Get at least one year of formal teaching experience regardless of your career goals. Research universities expect it on the academic market, and the communication skills transfer well into any career.
The students who navigate this well aren't the ones who picked the "right" type of assistantship. They're the ones who understood what each one actually cost — in time, in autonomy, in risk — and planned accordingly.
Frequently Asked Questions
Is a research assistantship always better than a teaching assistantship?
No, and the assumption that it is leads a lot of first-year students into situations they didn't anticipate. RAs can accelerate dissertation completion and research output when the project aligns well with your thesis. But they concentrate significant risk in one advisor relationship and depend on grant renewals that aren't guaranteed. TAs offer stable, department-backed funding and develop skills that matter on the academic job market.
Can I switch between TA and RA during my PhD?
Yes, and many students do — intentionally. The standard pattern is a TA appointment in years one and two while coursework and advisor selection happen, followed by a transition to RA once grant funding and dissertation direction are locked in. Some students also return to TA positions if an advisor loses funding mid-degree. Ask your department's graduate coordinator how transitions are handled in your program before you need the answer urgently.
Do TAs and RAs receive the same tuition waiver?
At most research universities, yes. Both appointments typically include a full tuition waiver and health insurance coverage. The difference is the funding source: department budgets cover TA waivers, while faculty grants cover RA waivers. According to Cornell University's 2025–26 published stipend schedule, the minimum 9-month stipend is identical for both TA and RA appointments at $35,661 on the Ithaca campus.
What happens to my funding if my advisor's grant runs out?
This is the scenario most students don't ask about until it happens. If a grant expires or a renewal isn't awarded, your advisor may lack the budget to continue your RA appointment. Outcomes vary: some departments absorb the student into a TA slot, others offer emergency fellowships, and in some cases there's a genuine gap. Before accepting any RA offer, ask your advisor directly about their grant's end date and what contingency plans exist. That conversation tells you a lot about how the relationship will go.
Does having only TA experience hurt my chances at a research-focused job?
It depends on what "research-focused" means. For R1 faculty positions, a strong publication record carries more weight than funding type. For industry research roles, nobody asks. Where TA-heavy profiles can look thin is at research-intensive labs or postdoc positions where advisors want to see evidence of sustained, independent research output. The fix is simple: do both, at some point in your degree.
Are stipends from TAs and RAs taxed the same way?
Generally yes for U.S. citizens and permanent residents — both are treated as employee compensation and appear on a W-2. The difference arises with fellowship income, which is sometimes reported differently and may be exempt from FICA taxes. International students on F-1 visas may be exempt from Social Security and Medicare taxes for the first five calendar years of U.S. residence, regardless of appointment type, though the specifics depend on their home country's tax treaty with the United States.
