Table 3.

Summary of regression analyses for mental help effectiveness

Variables	A mental health professional can help the condition
	Equation 1		Equation 2		Equation 3		Equation 4
	b	s.e.	b	s.e.	b	s.e.	b	s.e.
Main effect
African American	.18*	.07	.15*	.08	−.16	.15	.79*	.37
Demographic controls
Age	–		−.004**	.00	−.004*	.00	−.004**	.00
Gender (Women = 1)	–		.16**	.05	.16**	.06	.17**	.05
Improve on its own	–		−.12***	.03	−.18***	.04
Seriousness	–		.13*	.05			.20***	.06
Interaction terms
Race × Improve on own	–				.16*	.06
Race × Seriousness	–						−.20	.11
Constant	3.44	.03	3.20	.22	3.70	.14	2.73	.22
R 2	.010	.081	.081	.061

^*p < .05

^**p < .01

^***p < .001.

The number of cases differs slightly for each dependent variable because we excluded respondents with missing data on those variables. Missing values for predictor variables were replaced using conditional mean imputation (Allison 2002)