[Page 737]
Symptom class/level |
k |
N |
ru |
95% CI |
Chi2 |
runw |
|
Emotional problems |
|
||||||
| All | 14 | .17 | |||||
| All (w/o KFW) | 13 | 18,146 | .13 | .12 - .15 | 74.40* | .14 | |
| Male | 4 | 4,029 | .09 | .06 - .12 | 19.32* | .13 | |
| Female | 6 | 7,517 | .15 | .12 - .17 | 35.74* | .16 | |
Behavioral problems |
|||||||
| All | 20 | .17 | |||||
| All (w/o KWF) | 19 | 151,820 | .11 | .11 - .12 | 81.41* | .14 | |
| Male | 7 | 70,859 | .11 | .11 - .12 | 37.81* | .15 | |
| Female | 9 | 74,261 | .11 | .10 - .11 | 40.35* | .13 | |
CSA = child sexual abuse;
k = number of effect sizes (samples);
N = total number of participants in the k samples;
ru = unbiased effect size estimate (positive values indicate
better adjustment for control participants);
CI = confidence interval for ru;
Chi2 = within-group homogeneity statistic (based on dl = k -1);
runw = unweighted mean effect size. "AIl' indicates all
samples were used, including Kendall-Tackett et al. (1993), in which some data
were missing and statistics could not be computed, as indicated by blank spaces;
"AIl (w/o KWF)" indicates all samples were used except for
KendaIl-Tackett et al. (1993).
* p < .05, indicating non-homogeneity of effect sizes.