Contextualizing NSSE Effect Sizes: Empirical Analysis and Interpretation of Benchmark Comparisons

NSSE staff are frequently asked to help interpret effect sizes. “Is .3 a small effect size?” “Is .5 a really large effect size?” An effect size (ES) is any measure of the strength of a relationship between two variables. In practice ES statistics are used to assess comparisons involving correlations, percentages, mean differences, probabilities, and so on. In cases where large sample sizes make it more likely that a difference – even a small one – will be statistically significant, ES statistics are often thought of as a measure of practical significance because they indicate the relative magnitude of the difference. Thus they are valuable in comparing abstract measurement indices such as the NSSE benchmarks which are computed on a 0 to 100 scale from sets of individual items using various response sets.