Modeling Run Test Validity: A Meta-Analytic Approach

摘要

Previous summaries of the research evidence have shown that run tests are valid indicators of VO2max, but analysis also indicated that validity differed from one test situation to another. This study utilized data from 166 samples (N = 5,757) to test the general hypothesis that differences in testing methods could account for the cross-situational variation in validity. Only runs >2 km or >12 min were included. These criteria restricted attention to tests with maximal validity. The estimated average validity (r= .75). Validity was higher for fixed-time runs than for fixed-distance runs and in samples with greater variability in VO2max. This difference must be interpreted cautiously because studies that directly compared these 2 types of run test have found little or no difference. Validity was not related to the age, gender, fitness, or running experience of the population tested or to the method used to measure VO2max. A random-effects model estimated the 95% credibility interval for the validity of run tests at r = .52 to r = .84. The evidence was consistent with the view that some methods factors affect run test validity, but tests are equally valid for different types of people. This summary provides a point of. departure for the design and interpretation of future run test validation studies.