Standard statistical approaches use p values to compare treatments suggesting only that differences exist or do not. By themselves, p values do not indicate the size of an effect in a study. Comparing only p values across studies risks giving undo emphasis to very large studies with very small effects. Strict reliance on p values misses biologically important patterns and small sample size compounds this risk. Calculating unitless standardized effect sizes facilitates comparisons of response variables measured in different units and among published studies. We combined null hypothesis testing using t tests with calculation of Cohen's d and its confidence intervals to look past p values in two study systems. First, we measured benthic metrics under stream bridges and upstream of bridges. Secondly, we removed the dominant vegetation in experimental ponds and compared responses of organisms to this perturbation. Bridges reduced macroinvertebrate richness and abundance, but statistically significant reductions in corelated benthic metrics under bridges were undetectable. Large standardized effect sizes of reductions in EPT (Ephemeroptera, Plecoptera, and Trichoptera) richness and Ephemeroptera supported our hypothesis that bridges would reduce benthic diversity. No significant differences were found between pond treatments despite robust effect sizes supporting hypothesized reductions in most organisms. We conclude standardized effect sizes provide a more reasoned and nuanced approach to evaluating differences between means in biologically important variables.