Graduate Students’ Effect Size Category Boundaries

Statisticians increasingly decry ritualistic categorizations of statistical measures. The interpretation of effect sizes is often guided by benchmarks, i.e., Cohen’s d = .2 represents a ‘small’ effect size; .5 represents a ‘medium’ effect size; and .8 (‘large’) represents a large effect size. We employed a cognitive science approach to investigate how researchers systematically categorize values between these benchmarks. We find effect size categories are typically separated by fuzzy boundaries, as predicted by psychological theories of categorization. Understanding the cognitive processes underlying statistical reasoning can help us consider how to move beyond ritualistic interpretation of statistical measures.