Within our department, there are graduate students who are interested in enhancing their teaching credentials either because they believe that such enhanced credentials will give them a leg up on the traditional academic job market, or because they see themselves ultimately pursuing a teaching focused career. Moreover, due to sabbaticals etc., our department sometimes needs graduate students to serve as the instructor of record for courses both during the standard fall and spring academic semesters, as well as during our summer and winter online sessions. While the 691G course I created for incoming for graduate students provides some introduction to the principles of active learning via the physics education research literature, there is no mechanism currently in place at UMass for such students to get additional formal training in teaching. To help meet this need, I shifted my TA usage, from primarily serving as a student resource in class and during office hours, to more of an apprenticeship model giving GTAs both an under-the-hood look at running a large-enrollment introductory course and experience in front of such a classroom. I specifically want GTAs to get experience in this particular environment as such courses are both generally part of the workload for instructional faculty and an experience that most early-career research-focused faculty lack. In this model, the GTA still serves as an in-class assistant, but instead of out-of-class office hours, they meet with me for two hours after each class to help with the logistics of running such a course and planning the next day’s lesson.
When graduate students have served as primary instructors for our department in the past, they often do excellently in front of the classroom. Their lectures are well thought out, the figures well designed, the problems carefully considered, and the exams fair and well built. Some of our graduate students have even explored some active learning and alternative grading techniques in their own courses. I believe this expertise comes from their many years both as a student and as serving as in class GTAs for courses at UMass Amherst.
However, almost all our graduate students serving as primary instructors have also stumbled with what I call under-the-hood management of such horses. For example, one postdoctoral scholar serving as a primary instructor a few years ago scheduled his night exams 4/6 PM to 8:00 PM. His rationale for this time slot was that the more common 7:00 to 9:00 PM was a bit late; he said he would not have liked to take an exam that late at night, and so it seems unreasonable to ask his students to do so as well. While I fundamentally agree with this reasoning, such a choice greatly increased the difficulty that this postdoctoral scholar had to deal with regards to exam administration. At UMass Amherst, the 7:00 to 9:00 PM slot is reserved for night exams. Consequently, exams scheduled during this time slot take precedence over most other university activities classes labs etcetera. In contrast, exams scheduled outside of this window have a lower priority than other university activities requiring the instructor who scheduled such an exam to give the makeup. While giving makeup exams in a small class of approximately 20 or so is not a significant challenge, the logistical complexity increases dramatically with class sizes of several hundred. If this postdoctoral scholar had engaged in some sort of deeper, more formal training with regards to teaching, then he would have known about this rule and could have made at least a more informed decision.
Another challenge many of our graduate students in teaching roles have come across, and one that I myself came across as a new instructor, is familiarity with the slew of active learning techniques that can be employed in a classroom of several hundred. While I discuss the importance of active learning in 691G, and demonstrate some techniques, many instructors, both new and experienced, are more reluctant to attempt such techniques in a large classroom format due to logistical concerns. Moreover, novice instructors who do attempt active learning techniques in the large lecture setting often give up after negative experiences. My own experience as an instructor shows that these negative experiences are often the result of issues with either: implementation, choice of activity, or trying to “shoehorn” an activity or technique with inadequate consideration of the instructional context. By observing class, GTAs learn the techniques of each technique. Through our two-hour after-class meetings, they are encouraged to ask questions about what they have observed to be better prepared to implement them in their own classrooms in the future.
To add further complexity, many, though certainly not all, active learning techniques in a large classroom environment require some application of technology. Flipped classrooms require an online homework system to help students be prepared and some sort of classroom response system for quizzing. Moreover, many aspects of running a large-enrollment class are much smoother with technology: effective LMS usage helps students navigate the course, asynchronous communication platforms help build community, and tools like 3-D printing can help create an equitable learning environment.
The apprenticeship model I have developed over the past year provides an excellent opportunity for GTAs to gain a deeper familiarity with these tools. GTAs in this new approach complete tasks such as writing quizzes in the Edfinity online homework system, managing LMS content, creating and facilitating teams in CATME, using the VEVOX audience response system, using mail merges to provide feedback to students, as well as using Excel to facilitate communication between these tools. While I do not expect my GTAs to ultimately use all these technologies in their future classes, at least not at first, I do hope that exposure to them all will give them confidence to address pedagogical challenges in their future courses with technology.
In addition to these technical details of active learning implementation and technological familiarity, I strive to teach the GTAs in this apprenticeship model the fundamentals of instructional design: both at the micro class level and at the macro course level. With regards to individual classes, my role as a member of the executive board of the MSP (our faculty union), requires me to miss a few classes a semester. After observing several weeks of class, the apprentice GTAs are in a great position to gain experience in front of a class covering for me. I work very closely with the GTAs in helping them prepare for their sessions. Not only do I work with them to prepare their slides and notes, but we also do a “dress rehearsal” in the room to ensure that they are comfortable both operating the room’s technology, and demonstrations, and to ensure that they are comfortable with the performance aspect. After their lecture, I ask the GTA to review the recording of class and use a protocol from Harvard (https://cepr.harvard.edu/files/cepr/files/l1a_teacher_video_selfie.pdf) to reflect on their performance and how they will improve in the future. I also provide feedback on their instruction based on the video recording.
At the macro course level, we discuss the basic principles of course design such as backward design and essential questions. The goal of the backward design discussion is to help GTAs see the utility of this approach and to convince them that one of the first tasks is to write exams. The literature is very clear on the effectiveness of this approach. However, many novice instructors, and even some experienced ones, approach course development from slides: the first thing they do after writing a syllabus is begin writing slides. While this is an understandable approach, going into the semester with a few weeks ready to go definitely provides a sense of relief, almost any instructor will, at some point in the semester, catch up to what they had developed prior to the semester’s start. The subsequent writing as one goes is generally sustainable unless there are other significant instructional tasks to complete, such as writing and administering exams. This argument provides a practical reason to backward design supplementing the best-practices argument.
Ultimately, I have been very pleased with this change in GTA usage from student-resource to apprentice. Not only does the apprentice model help our graduate students become better instructors in the future, but it also is more engaging. The student-resource model, by contrast, has always yielded mixed results: some GTAs are highly committed to the role, while others feel pulled towards other obligations. This model also has a benefit for the department of requiring fewer graduate TAs, a benefit particularly salient in the current funding situation. If GTAs are the primary source of office hours etc., then many GTAs are required. The apprenticeship model, by contrast, uses a single GTA per section with the student-resource aspect of office hours etc. served, often more effectively, by cheaper undergraduate TAs. In short, I definitely plan to continue using this approach I the future.