By Janie H. Wilson, Ph.D., Georgia Southern University

I will begin by admitting that I started teaching 25 years ago as part of my graduate-school assistantship. At that time, I asked the department chair to avoid assigning me to teach statistics because I had seen many years of student struggles, including my own. I agreed to teach research methods, where I could share my passion for experimental and correlational designs. A few weeks into teaching the course, I realized my mistake. I explained to students that the two-group design we were using could be analyzed with a t-test. They stared blankly. I briefly explained that a t-test analyzed two groups when the dependent variable represented interval or ratio data. They glanced around the room at other students, clearly wondering if anyone knew what the heck I was talking about. One student raised her hand and assured me that she “kind of” remembered it.

Based on the curriculum, I knew research methods had a prerequisite: statistics. How could they not remember a t-test? That term, I had to reteach the t-test, ANOVA, Pearson’s r, simple regression, and chi square. I did not do a good job of teaching the topics – I simply was not prepared to tackle detailed statistics in research methods.

After the term ended, I gave a lot of thought to teaching statistics. Clearly I would have to teach analyses in research methods, so why not tackle the prerequisite course? To prepare, I looked back on the way I learned statistics. Mainly the focus had been on hand-calculations. Thinking about it now, I believe the approach made sense at the time. After all, when I began college, we typed term papers on a typewriter, not a PC! Computer labs popped up on campus pretty quickly, but even then, undergraduates did not learn statistical software as part of a statistics course. Later when I attended graduate school, hand-calculations remained the focus, including exams covering matrix algebra.

Based on my undergraduate and graduate training, I prepped my statistics course with a heavy emphasis on hand-calculations. When I taught statistics for the first time, I spent a week helping students work through their math anxiety as best I could. On exams, I graded based on the process rather than the final answer because students usually made minor math errors along the way. Throughout the term, after students struggled through hand-calculations, I showed them the magic of statistical software. When the answer appeared in a matter of seconds, students often asked me why they had learned all of the math. My answer was always the same: If you know the math, you will understand the analysis better.

As the years passed, I stood by my decision to focus on hand-calculations. Even when my (younger) colleagues urged me to consider focusing on computer software so I could spend class time on theory and more examples, I gave the same response: If they know the math, they will understand the analysis better.

It turned out that my noble intentions had no substance. When my colleagues taught research methods, students who had taken my statistics course did not remember how to analyze data using – you guessed it – a t-test. The simplest analysis was lost in the fog of a summer or holiday break. I had done nothing to solve the problem of students forgetting statistics. In fact, I had to be honest with myself that I never had any evidence that my students understood analyses better after going through hand-calculations.

I wish I could say my course immediately changed, but that would be untrue. I can say that my eyes had been opened, and I started watching what was really happening in my classroom. I would go through how to calculate standard deviation by hand, appreciating the “aha!” moment when students understood that we were obtaining an average spread of values. But rather than feel the elation of a job well done, I wondered what the point was. They were never going to work in a lab where they would calculate values by hand. In today’s world, most students have a powerful computer in their pockets.

As I continued to take students through hand-calculations, I noticed that about 50% of my class time was used when students worked through examples. Sure, the activity kept them awake, but they often produced the wrong answer and became so bogged down in the math that the big picture was never clear to them. I began to ask them to put down their pencils and talk through an example with me. I explained that whether or not they remembered to take the square root of the final number was not the point; they needed to understand what the number meant for research. No matter what I said, they grabbed their pencils as soon as they could and dove into the problem again, determined to conquer the math.

Although I have been slow to change the way I teach, the process has begun. And with so much class time free from hand-calculations, I can work with students on research design to provide context for each analysis. We have time to work through more examples, and I have even started incorporating APA style. I remain determined to help students build a solid foundation in our discipline with knowledge of statistics and research methods, the backbone of psychology as a science.

I am open to change. My next goal is to fully integrate research methods and statistics. Even if the curriculum continues to offer statistics and methods as separate courses, I can integrate methods into statistics for context, and I can integrate statistics into the methods course for repetition and more complete examples. Integration enhances student retention of the information, and I am delighted that Psychology Departments are beginning to rethink the curriculum and abandon sequenced courses in favor of integration. By letting go of hand-calculations, we make room for the important context offered by research methods.

Recommended Readings

Barron, K. E., & Apple, K. J. (2016). Debating curricular strategies for teaching statistics and research methods: What does the current evidence suggest? Teaching of Psychology, 41(3), 187-194. DOI: 10.1177/0098628314537967

Pliske, R. M., Caldwell, T. L., Calin-Jageman, R. J., & Taylor-Ritzler, T. (2016). Demonstrating the effectiveness of an integrated and intensive research methods and statistics course sequence. Teaching of Psychology, 42(2), 153-156. DOI: 10.1177/0098628315573139

Stranahan, S. D. (1995). Sequence of research and statistics courses and student outcomes. Western Journal of Nursing Research, 17(6), 695-699.

Wilson, J. H. (2017). Teaching challenging courses: Focus on statistics and research methods. In Obeid, R., Schwartz, A. M., Shane-Simpson, C., & Brooks, P. J. (Eds.), How we teach now: The GSTA guide to student-centered teaching. Society for the Teaching of Psychology e-book http://teachpsych.org/ebooks/howweteachnow

Wilson, J. H., & Joye, S. W. (2017). Demonstrating interobserver reliability in naturalistic settings. In Stowell, J. R. & Addison, W. E. (Eds.), Activities for teaching research methods and statistics in psychology: A guide for instructors. Washington, DC: American Psychological Association.

Wilson, J.H., & Joye, S.W. (2017). Research methods and statistics: An integrated approach. Thousand Oaks, CA: Sage Publications.