Department of Psychology
Master of Arts
The effect of intertest interval on test-retest reliability has received little attention in the research literature. In addition, no information has been gathered regarding the intertest intervals currently being used in test-retest reliability studies. This meta-analysis consists of 172 test-retest reliability studies. We hypothesized that the length of the intertest interval would be negatively associated with the test-retest reliability coefficient. The results of the study show that a median intertest interval of 14 days is typically usedin test-retest reliability studies. Data are also provided on typical intertest intervals and test-retest reliabilities categorized by the type of test and the type of sample used in the study. Of the 172 studies used to gather normative data, 69 studies used sample sizes large enough (minimum N = 100) to be used in the test of our hypothesis. The results show that there is a significant negative correlation, r = -.31, between intertest interval and test-retest reliability. This relationship is linear when used to describe intertest intervals shorter then 90 days, but shows a nonlinear relationship when studies using intertest intervals longer then 90 days are included.
Cromwell, Bryan, "A Meta-Analysis of Test-Retest Reliability Studies" (2005). Masters Theses & Specialist Projects. Paper 494.