RT Journal
A1 Singh AS, Uijtdewilligen L, Twisk JR, van Mechelen W, Chinapaw MM
T1 PHysical activity is not related to performance at school—reply
JF Archives of Pediatrics & Adolescent Medicine
JO Archives of Pediatrics & Adolescent Medicine
YR 2012
FD July 1
VO 166
IS 7
SP 678
OP 679
DO 10.1001/archpediatrics.2012.337
UL http://dx.doi.org/10.1001/archpediatrics.2012.337
AB 
In their letter, Hattie and Clinton report that they have performed a meta-analysis averaging all outcomes from all included studies.1 We, however, believe that the heterogeneity of the included studies is too large to combine them in a meta-analysis. The included studies differ largely regarding study sample, intervention contents, measurement instruments, outcome variables, follow-up duration, and study design. Therefore, we believe that the data are not appropriate to perform a meta-analysis. Instead, we chose to perform a methodological quality assessment combined with a best-evidence synthesis2 to summarize the findings of all studies. This method is commonly used in systematic literature reviews and takes into account the number, methodological quality, and consistency of outcomes of the studies. Based on this method, we conclude that there is strong evidence for a positive prospective relationship between physical activity and school performance.1 Importantly, this conclusion is very different from the statement that we found “evidence for a strong positive relationship.”