Table 4.

Model specifications and outcomes for a priori multiple-group power analysis using Monte Carlo simulations (n repetitions = 10,000)

Path	DRD4-7r n = 143			DRD4-no7 n = 262
	Estimate	Coverage	Power	Estimate	Coverage	Power
Autoregressive paths positive social preference	0.60	0.95	1.00	0.60	0.95	1.00
Autoregressive paths prosocial behavior	0.60	0.95	1.00	0.60	0.95	1.00
Autoregressive paths conduct problems	0.60	0.94	1.00	0.60	0.94	1.00
Positive social preference predicting prosocial behavior	0.12	0.94	0.80	0.00	0.95	0.05
Positive social preference predicting conduct problems	−0.12	0.94	0.81	0.00	0.95	0.06
Prosocial behavior predicting positive social preference	0.05	0.95	0.51	0.05	0.95	0.51
Conduct problems predicting positive social preference	−0.05	0.94	0.51	−0.05	0.94	0.51
Correlations positive social preference and prosocial behavior	0.10	0.95	0.22	0.10	0.95	0.36
Correlations positive social preference and conduct problems	−0.10	0.95	0.23	−0.10	0.95	0.38
Correlations prosocial behavior and conduct problems	0.10	0.95	0.23	0.10	0.95	0.37

Estimates of paths reflect standardized regression coefficients. Correlations between constructs reflect residual error correlations. Means of all constructs were estimated to be 0 and variances of all constructs were estimated to be 1. Recurring paths were constrained to be similar over time, hence estimates hold for all recurring paths. Estimates < 0.05 are considered too small to interpret, estimates ≥0.05 are small but meaningful, estimates ≥0.10 are moderate, estimates ≥0.25 are large (Keith 2006)