The Applicability of the Spearman-Brown Prophecy Formula When Applied to Physical Performance Tests

Data from four physical performance tests were collected from approximately 260 subjects and were used to determine (a) whether the split-half method is a valid procedure for estimating the reliability of multiple trial physical performance tests, and (b) whether the Spearman-Brown prophecy formula will be an accurate predictor of the reliability of a physical performance test when its length is increased. A significant day-to-day variation in test scores was found at the .05 level. The assumptions underlying the use of the Spearman-Brown prophecy formula were significantly violated at the .05 level. Seventy-nine percent of the reliability coefficients computed by the split-half method were higher than the test-retest reliability coefficients. Seventy-two percent of the reliability coefficients predicted by the Spearman-Brown prophecy formula were higher than the obtained test-retest reliability coefficients.