Most research in medical education, when examining the impact of an intervention, report findings based primarily on significance testing despite the controversy of its appropriate use. Moreover, the p-value used to determine rejection or acceptance of the null hypothesis tells nothing about the magnitude of the significance. Using a pre/post assessment of a Geriatric Interclerkship as a case study, this study examines the utility of "effect size" measures in augmenting significance testing results.
Presented at the AAMC (Association of American Colleges) Annual Meeting, RIME (Research in Medical Education) Program, November 2004.
Available at: http://works.bepress.com/mary_zanetti/22/