Take a Scientific Approach to Your Teaching, Part 2

Take a Scientific Approach to Your Teaching, Part 2

In Part 1, we looked at why and how you might want to study your own teaching, and how to choose a topic to focus on. Today we’ll examine different approaches you can take and the ethical guidelines for your research. Part 2 of a two-part series.

How Do You Study Your Teaching?

Once you have a question in mind, the next step is to determine an appropriate way to study it. Here, you have many options:

  • Correlational studies look at whether some variable correlates with one or more other variables. For example, do students’ perceptions of their instructor correlate with how motivated they are to learn? You could test whether students’ ratings of the professor (positive or negative) correlate with their attendance or with their performance on exams and assignments.

  • Between-group comparison: Divide your students into two or more groups and give each group a different experience to see the effects. Sometimes an instructor teaches two sections of the same course, making it relatively easy to give each section a different “treatment.” For example, does an active learning activity embedded in lecture boost how well students do on a quiz that week in class, compared to a group of students who didn’t do the learning activity? Even if you don’t teach two sections of a course, you can sometimes creatively divide students into two groups. Perhaps you divide the class and bring in each half for a different part of a class period, or you send each group different brief assignments to do before class one day.

  • Within-group comparison: Give all of your students different “treatments” at different times and see how their performance or some other outcome varies with the treatment. For example, you might make lecture slides available before the start of class for a randomly-selected subset of lectures, and after class for another randomly-selected subset. Does student attendance or performance on a short quiz depend on whether they received lecture slides before or after class?

Ethical Considerations

When doing SoTL, it is critical to keep our students’ best interests at heart. If you intend to publish any findings from your research, then you should seek approval from the Institutional Review Board (IRB): http://humansubjects.stanford.edu/ They can help you be sure that your research approach is ethical.

One important ethical consideration is whether and how to get your students’ informed consent to participate in a research study. Cluing your students to the exact nature of your study risks changing their behavior in ways that can bias your results, but you (and the IRB) will likely want to find some way to let students know that research is going on.  Another concern is whether students are even free to consent or not consent to the research: If research involves standard classroom practices, then students will participate by default. One solutions to this problem is to give students the option to withdraw their data from any published analyses.

Of course, you shouldn’t manipulate aspects of the classroom that you suspect will harm your students. Given existing research on the costs of multi-tasking, for example, I wouldn’t encourage half of my students to engage in online shopping during lecture to test its effects on learning.

You should also take steps to ensure that “control” groups or conditions are receiving the typical best practices you have to offer and aren’t being disadvantaged for the sake of the study. Our research questions should be aimed at whether we can be more effective than we already are. I would also make sure that if your class is graded on a curve, that students receiving different treatments are graded on separate curves.

This last point brings us to an important consideration. Many teachers worry that giving students different treatments in their course is unfair. What if one group of students ends up with a higher grade in the course than others as a result of something you manipulated? This concern can prevent many scholars from taking a more rigorous scientific approach to testing their teaching methods. As teachers, we want to make sure all of our students get the “best,” right? But consider: could taking a more rigorous, scientific approach to testing your teaching methods actually be the more ethical approach to teaching?

Teachers unintentionally manipulate their students’ experience all the time, giving some students potential advantages over others, without having any idea what the outcomes are. From quarter to quarter, from lecture to lecture, from office hours to office hours, we make changes to what we are doing. We may be making things better. We may be making things worse. Without the tools of science, we can’t tell which is which. Science allows us to bring our manipulations into focus and examine their effects. Science give us the potential to improve the experience for ALL of our students.

Have you done research on your own teaching? Want to try? Share your thoughts in a comment below.


Bridgette Hard, Ph.D. headshotBridgette Hard, Ph.D. is a cognitive psychologist and the Coordinator of the Psychology One Program at Stanford.



See Also

Garung, R.A.R. & Wilson, J.H. (2014). Doing the scholarship of teaching and learning: Measuring systematic changes to teaching and improvements in learning. New Directions in Teaching and Learning, No. 136. San Francisco: Jossey-Bass.

What is the Scholarship of Teaching and Learning? with Mary Huber

A three-part series by expert Mary Huber on SoTL and its life at Stanford.

Assessing learning in informal science programs

A survey of methods to assess learning when students are doing science outdoors.

Everyday Practice and Enduring Mystery

Coverage of a 2014 Stanford forum on teaching and learning as complex systems.