Table 4.

Papers giving paternally-reported ECBI Intensity scores.

Author	Significantbenefit (P <0.05)?	n1	mean1	sd1	n2	mean2	sd2
Bodenmann et al. [32]a	No	50	109.2	18.5	50	110.1	25.2
Connell et al. [50]	Yes	11	154.55	17.44	12	111.0	12.41
Hahlweg et al. [51]*	No	16	10.7	7.0	18	7.7	5.1
Hahlweg et al. [13]*	No	57	9.3	7.2	141	10.2	6.9
Markie-Dadds & Sanders [56]	No	NOT REPORTED
Morawska & Sanders [61]	No	24	111.57	20.41	49	106.07	24.37
Sanders et al. [65]	Yes	71	127.34	22.39	184	113.13	27.34

^aECBI subscale data reported in [32] were assumed to have been transposed, and are corrected here. In this paper attrition rates at the post-treatment assessment are unknown and we assumed they remained constant. n1, mean1 and sd1 are, respectively, group size, mean and standard deviation for the control groups, and n2, mean2 and sd2 are the corresponding figures for the Triple-P intervention groups. Means and standard deviations are for ECBI-I subscale data, apart from the two papers marked with an asterisk, where the CBCL-E was reported. No paternally-reported data are tabulated in Markie-Dadds and Sanders [47], but there is a statement that 'Analyses of father-reported measures of child behavior failed to produce any significant effects.' CBCL-E, Child Behavior Checklist - Externalizing scale; ECBI-I, Eyberg Child Behavior Inventory - Intensity scale.