Data science is in demand, producing pressure for institutions of higher education to add data science majors, minors, or certificates to their catalog. However, new data science programs miss opportunities that already exist within current curricula. A data scientist is someone who uses computational stills to work with, understand, and explain the meaning of large datasets. These skills involve the fundamental ability to understand research methodology, statistical analysis and interpretation, and ethical behavior—areas in which traditional undergraduate psychology curricula excel. Given that psychology is one of the most popular and diverse undergraduate majors1,2, psychology programs are in the perfect position to impact the future workforce by enhancing data science skill instruction at the undergraduate level.
Data scientists employ their skills across numerous contexts, so which data science skills should psychology add to the curricula? Consider creating modular activities that will push your students toward data science tools and skills. Start with the basics. One critical first step is data management. Many students overestimate their understanding of file structure, permissions, and storage on local machines and in the cloud. Consider students working on an assignment for their psychological statistics course. A basic lesson in data management involves teaching students how to properly name and store files within a local working directory. Starting with this fundamental lesson ensures that all of your students are ready for more advanced data science activities such as accessing and transforming publicly available datasets (e.g., from Open Science Frameworkopens in new window). As an added bonus, perhaps you won’t have to grade 50 submissions named “research paper.”
How about data visualization? Most psychology instructors discuss data in class because it’s connected to their content delivery. Some of that data might be publicly available. Instead of assigning a paper for background reading or presenting results in a slideshow, you could teach students how to recreate figures from the paper or create simpler visualizations from the data. Your students might know how to create graphs in spreadsheet software, but you could teach them how to create the visualizations with popular data science toolkits based on R or Python programming languages. In our classes, we train students to create customizable visualizations using the seaborn libraryopens in new window within Jupyter Notebookopens in new window, a free, web-based platform that empowers users to rapidly write, test, and revise code. Creating informative and elegant visualizations with code is an essential data science skill, and it presents an excellent opportunity to introduce psychology students to computer programming. As your students learn to code their data visualizations, you can demonstrate how small changes in code can dramatically affect the resulting visualization. This kind of active learning naturally leads to conversations about the qualities of a good figure, just as it increases students’ confidence in their use of data science tools—and it opens the doors to exploration in other areas of data science.
Once your students are comfortable visualizing data within a programming environment, you can teach them how to code statistical analyses of the data in their visualizations. In our courses, we teach students how to define, compute, and transform variables while they conduct descriptive analyses with the pandas libraryopens in new window. The combination of publicly available psychological data with freely available data visualization and analysis tools presents a transformative opportunity for classroom conversations that can bolster students’ understanding of the science before, during, and after they read the authors’ interpretation of those data.
There are many other places to incorporate data science skills within undergraduate psychology classes. Do you usually clean lab data for your students before they analyze it? Teach them this essential data science skill instead. Such experiences will naturally lead you and your students to creating powerful data narratives that can transparently communicate the entire data processing pathway. There are a multitude of ways to add these skills to our courses. Yes, you will have to give up some of your course content to make room for data science skills. But our primary job as educators is to prepare our students for the future. The professional and economic demands for data science skills will continue to grow. We owe it to our students to prepare them for this data science future.
Aileen M. Bailey, PhD, received a BA in psychology and mathematics from Beloit College (Wisconsin) and an MS and PhD in biopsychology from the University of Georgia. As a professor at St. Mary’s College of Maryland, she teaches courses in research methods, statistics, biological psychology/behavior neuroscience, and the psychology of learning. At St. Mary’s College, she has served as the chair of the psychology program, coordinator of the neuroscience program, acting chair for the Department of Biology, and as the Aldom-Plansoen Honors College Professor. Bailey’s scholarship includes publications in Behavioral Brain Research, the Journal of Neuroscience, and Nature Neuroscience.
James Mantell, PhD, is associate professor, chair of the psychology department, and current Aldom-Plansoen Honors College Professor at St. Mary’s College of Maryland. Mantell received a BA in psychology and philosophy from Millersville University (Pennsylvania) and an MA and PhD in cognitive psychology from the University at Buffalo. His current research addresses data science pedagogy, auditory perception, and music cognition.
