Table 4.

Factors Predicting the Incidence of Substance Abuse by Age 26

	Total sample (n=1,208)				Male group (n=585)
	Marginal effect	Robust Std.Error	p-value		Marginal effect	Robust Std. Error	p-value
Gender (female)	−0.350	0.025	0.000	***
Race/ethnicity (African-American)	0.060	0.033	0.151		0.080	0.071	0.322
CPC preschool participation (ages 3-4)	−0.010	0.023	0.509		−0.030	0.035	0.303
Child protection services (ages 4-9)	0.170	0.057	0.000	***	0.220	0.067	0.002	**
Family conflict (ages 5-10)	−0.010	0.041	0.943		−0.020	0.076	0.873
Parent sub abuse experience (ages 5-10)	−0.030	0.059	0.687		−0.060	0.137	0.689
School mobility (ages 6-9)	−0.030	0.017	0.098		−0.040	0.028	0.211
CPC school-age participation (ages 6-9)	0.040	0.032	0.219		0.050	0.042	0.283
Social maturity (ages 7-9)	0.000	0.004	0.896		0.000	0.006	0.573
Family risk (age 8)	0.000	0.006	0.752		0.000	0.014	0.874
Parent expectations (ages 8-10)	−0.050	0.021	0.012	*	−0.060	0.035	0.075
Intrinsic motivation (ages 9-10)	−0.020	0.018	0.181		−0.040	0.030	0.270
No trouble-making behavior (ages 9-12)	−0.030	0.006	0.000	***	−0.040	0.011	0.000	***
Reading achievement (age 10)	0.000	0.001	0.345		0.000	0.002	0.390
School mobility (ages 10-14)	0.030	0.016	0.060		0.060	0.025	0.016	*
School quality (ages 10-14)	−0.060	0.044	0.237		−0.080	0.081	0.285
Parent substance abuse experience (ages 10-15)	0.100	0.055	0.037		0.140	0.094	0.150
Personal substance use experience (ages 10-15)	0.460	0.115	0.000	***	0.400	0.081	0.000	***
School mobility (ages 14-18)	0.040	0.020	0.018	*	0.110	0.045	0.005	**
Deviant peer affiliation (age 16)	0.080	0.025	0.002	**	0.130	0.048	0.006	**
School dropout (by age 16)	0.030	0.032	0.218		0.020	0.050	0.575
Wald Chi-squared	2,968.73				2,207.66
Prob. > Chi-squared	0.00				0.00
Pseudo R-squared	0.33				0.16

Note:

^***p<0.001

^**p<0.01

^*p<0.05.

Probit coefficients are transformed to marginal effects (dy/dx), which are the change in substance abuse (in percentage points) per 1-unit change in the explanatory variables.