Mentoring Graduate Students as Teachers

While faculty support graduate students in their research as mentors, they also often model the specific skills of teaching, especially as graduate students begin to teach in different settings. At Princeton, for instance, graduate students often work as Assistants in Instruction (AIs) for undergraduate courses taught by faculty; they mentor undergraduates in labs; teach courses through the Prison Teaching Initiative; and have opportunities to collaboratively teach courses with faculty, as well.

Course Planning and Implementation

  • Share the course’s learning objectives and explain how you see the different components of the course supporting them.
  • Involve your AIs in substantive decisions you make in the course before and during the semester: taking them through the pros and cons of different decisions can model the process of course design and teaching, and also help them feel like they are active and valued members of the teaching team. Try to engage them in substantive pedagogical as well as administrative ways. 
  • Invite your AIs to give a guest lecture in your course. It can be helpful to meet in advance of the lecture to mentor them in the process: make sure their questions are answered, give them the opportunity to talk through their plans or outline, and offer any suggestions and best practices you have for lecturing. Afterwards, find time for another meeting to debrief: what went especially well? What might they think about for the next time they give a lecture? At what points could you see that students were particularly engaged?
  • Offer to observe your AI’s teaching. You might offer to observe your AI’s precept to give them feedback on their interactions with students in a smaller class setting.
  • Involve AIs in any course meetings when you plan to receive support: for instance, the Canvas team and McGraw staff overall are always happy to meet with AIs along with the instructor of the course, and often AIs have helpful perspectives and insights into your students’ experiences in the course. To plan a meeting regarding your Canvas course, feel free to write to [email protected].
  • Meet regularly during the semester as a team of AIs to offer them a chance to support each other as teachers of the course and share activities and approaches that have worked well, and also to hear about their sense of their students’ experiences in the course. 
  • Encourage AIs to ask for regular feedback from their students, for instance, through mid-semester surveys. This feedback can help them retool their teaching throughout the semester, which is especially important during their early experiences as instructors.
  • Encourage AIs to consider the Collaborative Teaching Initiative, which offers a chance for more formal and sustained mentorship of a graduate student, from the process of course design through the semester of teaching.


Grading can be frustrating and time-consuming for instructors, especially for AIs who have less teaching experience. The following suggestions and strategies may help you guide your AIs in developing responsible grading practices and becoming more confident and efficient graders.

  • Seek input from your AIs when designing prompts and assignments.
  • Meet regularly with your AIs to discuss grading and other teaching practices.
  • Determine and share the grading scale for your course. If a student meets your expectations, what grade will they earn? If they exceed them? 
  • Establish grading criteria and create rubrics with your AIs in order to make grading more consistent and efficient. Encourage your AIs to discuss these criteria in class, precept, or lab. It can be helpful to have a norming session after the first major assignment in which you compare notes on your feedback and ensure that you are grading consistently across sections.
  • Suggest that your AIs grade exams horizontally–that is, recommend that they grade one question across a set of exams, which will also make grading more consistent and efficient. Digital tools such as Gradescope can make such grading across a section and course with multiple AIs easier. Consider grading anonymous submissions, which has been shown to have some benefits in terms of equitable assessment. 
  • Guide your AIs in commenting meaningfully on student work. A meaningful comment on writing may engage with a student's ideas and the organization of those ideas, responding to the thesis, argument, or use of evidence. In a lab-based setting, you might offer specific suggestions for commenting on a student’s data analysis or methodology and for grading quantitative problem sets. For instance, make transparent any practices you have around awarding partial credit, which can vary significantly in different academic contexts.
  • Establish procedures for settling grading disputes, clarifying to your AIs how they should handle any issues that come up, and when to involve you as a resource and next step for students in disputing a grade. 
  • Offer suggestions for dealing with disappointed students. In your experience, have you found it useful to request that students wait 24 hours before coming to discuss their grades, to ask that students explain their concerns in writing, or to focus a meeting with a disappointed student on helping them prepare for the next assignment or exam?
  • Help AIs establish appropriate, professional relationships with their students. AIs may feel caught between being a peer and an authority figure to their students, especially when grading. Share with them your model of interacting with undergraduates.
  • Encourage AIs to discuss assignments or exams in class, making general comments on the class performance and calling attention to any analytical difficulty–difficulty solving certain types of problems, for instance—that students are having. Make suggestions for how AIs can help students use their graded assignments or exams to complete the next assignment more productively or study for the next exam more effectively.

References and Additional Resources

Austin, A. E. (2002). Preparing the next generation of faculty. Journal of Higher Education, 73(1), 94-122.

Marincovich, Michele, Jack Prostko, and Frederic Stout, Editors. The Professional Development of Graduate Teaching Assistants. John Wiley & Sons, Incorporated, 1998.

Von Hoene, L. & Mintz, J. (2001). “Research on Faculty as Teaching Mentors: Lessons learned from a study of participants in UC Berkeley’s Seminar for faculty who teach with graduate student instructors.” To Improve the Academy: 77-93.