This study aimed to highlight the relationship between moderate-to-vigorous physical activity (MVPA) as assessed by accelerometer devices and the Physical Activity Questionnaire for Children (PAQ-C) to estimate the convergent validity of the questionnaire. A systematic review and a meta-analysis were applied by collecting pertinent studies (PubMed, Web of Science, PsycINFO, and SCOPUS) from 1997 until November 2020. The relationship between PAQ-C and MVPA scores was estimated considering correlation coefficients such as the effect size. Fisher's transformation was used to convert each correlation coefficient into an approximately normal distribution. The pooled correlations between PAQ-C and MVPA scores were measured by r values after converting the Fisher's z values back into correlation coefficients for presentation. A total of 13 studies were included in the meta-analysis, and a random effects model was adopted. The pooled correlation between PAQ-C and MVPA scores was significant but with a moderate effect size (r = 0.34 [0.29, 0.39], Z = 15.00, p < 0.001). No heterogeneity among the studies was observed (I2 < 25%). In conclusion, the results highlighted a moderate relationship (around 0.30-0.40) between PAQ-C and accelerometer measurements. These results suggested to concurrently administer both tools to reach a more comprehensive description of children's PA, in terms of quality and quantity.
Keywords: MVPA; PA; accelerometer; children; questionnaire.