Background/Purpose
The international physical activity questionnaire (IPAQ) is one of the most widely used measurement tools in a large scale studies for its practicality. The purpose of this study was to investigate the convergent validity of IPAQ with meta-analytic approach and to estimate the differences in the degree of convergence of IPAQ by moderator variables.
Method
The relevant studies were surveyed from the five electronic databases. A random effect model was applied to calculate the weighted mean effect sizes (ESzr) 95% CI. In addition, the weighted mean ESzr were reversed to mean correlation coefficient (r) for ease interpretation.
Analysis/Results
A total of 134 ESzr from 14 studies were calculated. Overall mean correlation of IPAQ was .45(95% CI=.43, .46) and heterogeneous (Q=2243.07, df=133, p<.001). Examinations of moderators showed higher correlation in short form (r=.53) than long form (.27), and Asia (.60) had highest correlation among five regions (N. America=.38, S. America=.41, Europe=.26, and Africa=.43). Interviewer administration (.55) had higher correlation than others (self = .31, and telephone = .32). Also, studies in which questionnaires were used had greatest correlation (.57) among four instruments (Accelerometer=.33, Pedometer=.20, and PA-log=.45). There were fairly similar correlations among intensity categories of PA (low=.53, moderate=.36, vigorous=.45, and total PA=.43).
Conclusions
The IPAQ have secured moderate convergent validity. However, observed differences in the degree of convergence by moderator variables indicated necessity for further studies to find out the causes of such differences.