We excel, in the research university, at preparing our students to do world-class research — everywhere except the classrooms in which they teach. From the beginning we insist that Ph.D. applicants explain their research plans. When they arrive we put them through their paces in methodology classes, carefully taking apart their ideas of what they want to accomplish and introducing them to the hard work of gathering data, performing analyses, testing and retesting hypotheses, and exploring all possible outcomes.

We want students to understand that what they think is true has to be questioned, repeatedly, and that their findings have to be defended. It is an iterative process, and we expect them to be rather poor at it when they begin — improving through honest critique and firm mentorship over time.

When it comes to teaching, however, the message they receive is very different. We don’t ask prospective students to address their teaching experience or philosophy in graduate-school applications, and we do not typically talk about teaching in coursework or qualifying examinations. Often it is not until graduate students enter the classroom, as teaching assistants responsible for their own sections, that they begin to think about what it might require to teach successfully.

In the midst of papers to grade and sections to prepare, conversations between even the best faculty instructors and assistants lean more toward the pragmatic. There is little room or incentive to see one’s time as a teaching assistant as an opportunity to simultaneously teach and analyze classroom success.

Some of this is because of the importance placed on graduate-student research. This makes a great deal of sense: Training the next generation of Ph.D.s to be world-class researchers in their chosen disciplines is a chief responsibility of modern universities. Time spent in the classroom is often seen as time spent away from one’s archive or laboratory, away from the process of inquiry and original analysis that leads to cutting-edge findings and future academic employment. This makes it all too easy to teach our graduate students that they must be skillful researchers, and only adequate teachers.

The fault line between teaching and research, however, is also created and maintained by our own misunderstanding, as largely 20th-century faculty, of the place of teaching in the 21st-century research university. With an increased national emphasis on graduation rates, student persistence, and student learning, rising undergraduate tuition costs, and the need to distinguish brick-and-mortar institutions from online offerings, teaching has become a much higher priority for all public institutions.

Merits and promotions are shifting to take teaching into greater account, new faculty are being given increased resources and encouragement to develop their pedagogy, and in some cases new positions are being created for tenure-track faculty who undertake what a recent National Research Council report has called “Discipline-Based Education Research.”

Whether current graduate students ultimately apply for traditional tenure-track research positions or in such new positions as pedagogy experts, they will be well served if their time in the classroom is time when they are encouraged to study how students learn in their field and adapt their practices for greatest success. Studying how undergraduates learn in a field actually also strengthens graduate students’ research processes in their own work. Breaking down the barrier between “discipline-based research” and “research into teaching” offers a win-win.

We are doing this, now, at the University of California at Davis with a group of teaching assistants assigned to the first course of our three-part Intro Biology series. With the support of the dean, department chair, and instructors, the teaching-assistant team was trained to implement active learning approaches in discussion sections, in partnership with the faculty who are teaching the main lecture course.

In this active-learning model, the undergraduate students complete online assignments even before the class begins, informing each teaching assistant, through data, as to where in the material they are struggling. Teaching assistants then use this data to adapt discussion content to address student strengths and weaknesses. In class, students are called upon randomly and asked to answer questions about core course material. If they do not know the answer, other peers are asked to assist, then the answers are refined by the teaching assistant, and finally the original student is brought back in to explain the correct answer.

Teaching assistants were also involved in determining whether the new method produced learning gains, why, and what improvements could be made for the next time the sections were taught.

For graduate students the approach enabled them to learn new teaching skills and better understand the role of distributed learning in biology students’ success. For the undergraduate students, their learning improved. By the course’s end we saw improvements in both exam pass rates and average total exam scores.

Think of how different it would be if we encouraged graduate students to think like this in the classroom and provided resources for them to do it. They might approach their first session nervous and unsure, but when they struggled to communicate concepts or students performed poorly on the material, teaching assistants would have at hand multiple approaches they might use and a means of asking which methods produced the best outcomes.

Rather than feeling that their job was to master knowledge and deliver lectures or overcome stage fright, they might see themselves as nodes where new ideas flow, where experimentation happens, and where data can help them continuously improve student learning. In the end, our Ph.D.s would emerge as skilled researchers in their professional fields and in their classrooms.

How do we get there? It will require a culture shift and a resource injection. Within our departments, we need to encourage students to see teaching as a process of experimentation, data collection, and refinement. Within our campuses we need to go beyond providing information-dumping single orientation sessions for all new teaching assistants. We should engage teaching assistants (and faculty) in experiments, such as the biosci project, as often as possible, with teams of pedagogy experts from our centers for excellence in teaching and learning.

On our campus this year, instead of the traditional single day of orientation for teaching assistants, we combined an introductory session in the fall with workshops and mentoring opportunities throughout winter and spring on topics ranging from active classroom techniques to adaptive learning tools to hybrid course construction. This allows assistants to gain experience, reflect on it, and when the time is right, get help from teaching experts and their own peers to improve their teaching impact and classroom dynamics. It is also important that teaching assistants learn to do assessment so that they are collecting data after courses (beyond student evaluations) and using that to think through their next approach to similar material.

For those of us dedicated to our students, this is an ideal moment to start mending the fault line between research in the discipline and research in the classroom, if not for ourselves, then certainly for our graduate students. We need to encourage our campuses to provide real teaching-assistant training and our teaching assistants to use such approaches in their sections, in true partnership with us. This way we can, at once, prepare our graduate students for rewarding teaching and secure employment, and provide our undergraduate students with active, engaged, and transformational learning.

Author Bio: Carolyn Thomas is a professor of American studies and vice provost and dean for undergraduate education at the University of California at Davis.