Contextualizing student engagement effect sizes: An empirical analysis

Abstract

The concept of effect size--a measure of the strength of association between two variables--plays a crucial role in assessment, institutional research, and scholarly inquiry, where it is common with large sample sizes to find small or even trivial relationships or differences that are statistically significant. Using the distributions of effect sizes from the results of 984 institutions that participated in the National Survey of Student Engagement (NSSE) in 2013 and 2014, the authors empirically derived new recommendations for the interpretation of effect sizes which were grounded within the context of the survey. We argue for the adoption of new values for interpreting small, medium, and large effect sizes from statistical comparisons of NSSE Engagement Indicators, High-Impact Practices, and student engagement data more generally.