Table 3.

Prediction of Physical Aggression in Adolescence (W-6) by Demographics and Family Functioning

	B (SE) Enter	B (SE) Final
Hispanic	−0.00 (0.02)	−0.00 (0.02)
White	−0.02 (0.02)	−0.02 (0.02)
Low SES	−0.02 (0.01)	−0.02 (0.01)
Adult Male Present	−0.01 (0.01)	−0.01 (0.01)
Family Cohesion	−0.04 (0.01)**	−0.02 (0.02)
Parental Monitoring	−0.01 (0.01)	−0.01 (0.02)
W-1 Physical Aggression	0.07 (0.01)***	0.07 (0.01)***
Cohesion × Hispanic		−0.08 (0.04)*
Cohesion × White		0.04 (0.04)
Monitoring × Hispanic		0.11 (0.04)**
Monitoring × White		0.02 (0.04)

Notes. Analyses controlled for study site using 3 dummy variables (4 study sites) and controlled for treatment condition using 3 dummy codes (4 treatment conditions). Hispanic and White ethnicity are dummy coded in comparison to African American ethnicity. Low SES is dummy coded such that (1) indicates participant’s family is below the poverty line. Adult male in the household is coded such that presence of an adult male = 1. Cohesion is measured at W-1 using combined parental and adolescent-report on the Family Relationships Scale (FRS; Tolan et al., 1997). Monitoring is measured at W-1 using combined parental and adolescent-report on the Pittsburgh Youth Survey (PYS; Thornberry et al., 1995). Aggression at W-1 and W-6 is measured using adolescent-report on the Problem Behavior Frequency Scale (PBFS; Farrell et al., 2000). W-1 = Wave 1; baseline; fall of sixth grade. W-6 = Wave 6; spring of eighth grade.

^*p < .05;

^**p < .01;

^***p < .001.