Using the Spearman-Brown Prophecy Formula to Improve Medical School Examination Quality

Abstract

Most measurement experts suggest a minimum reliability estimate of .60 to .70 is desirable for routine medical education assessments with low to moderate stakes. The most common recommendation for improving examination reliability is to increase examination length. Unfortunately, adjustments in examination length may result in unpredictable reliability estimation and involve a great deal of trial and error for educators. The Spearman-Brown prophecy formula may help medical educators accurately predict the effects of adding or removing items on reliability estimation, and help educators make immediate improvements to the psychometric quality and functioning of their examinations.