Table 2.
| ||||||||
---|---|---|---|---|---|---|---|---|
k | Effect size |
p (ES) | 95% Confidence interval |
Standard error |
I2 | Q | p (Q) | |
| ||||||||
Treatment vs. control all studies | 55 | 0.211 | <.001 | 0.109, 0.242 | 0.052 | 61.568 | 140.508 | <.001 |
| ||||||||
Treatment vs. control continuous data studies | 28 | 0.243 | <.001 | 0.076, 0.410 | 0.085 | 72.307 | 97.497 | <.001 |
| ||||||||
Treatment vs. control dichotomous data studies | 27 | 0.167 | .004 | 0.054, 0.279 | 0.057 | 36.673 | 41.057 | .031 |
| ||||||||
Treatment subjects pre- vs. post-comparisons* | 47 | 0.418 | <.001 | 0.293, 0.542 | 0.064 | 78.296 | 211.943 | <.001 |
| ||||||||
Control subjects pre- vs. post-comparisons** | 24 | −0.022 | .697 | −0.135, 0.090 | 0.058 | 51.139 | 47.072 | .002 |
|
k denotes number of comparisons, Q is a conventional homogeneity statistic, I2 is the percentage of total variation among studies’ observed ES due to heterogeneity.
Four studies excluded as outliers.
Three studies excluded as outliers.