For each analysis, we calculated Cohen's d as an estimate of effect size.
The estimate of effect did not differ much between the different study designs.
High quality: further research is very unlikely to change our confidence in the estimate of effect.
The between-group differences were significant, however, the eta-squared values (estimate of effect size) were small at 2.2%.
The imprecision of reporting lowers the quality of evidence, which means that further research is likely to have an impact on our confidence in the estimate of effect.
There is a propensity for small negative studies not to reach full publication, and this might lead to an exaggerated estimate of effect [19].
The outcome is rated as moderate quality of evidence by GRADE software, which suggests further research is likely to impact on the confidence in the estimate of effect.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
As an estimate of effect size, we calculated Cramér's V, which is a measure of the strength of association for contingency tests (Cramér 1946; Siegel and Castellan 1988).
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Of the 54 updated reviews that showed quantitative changes, the median change in the estimate of effect size was 18.2% (95% CI 13.2% 23.1%).
