Table 3. Regression Analysis Investigating Predictive Validity of SADQ, AAAQ Subscales, and SRC Task D Measures for PHDD at 2-, 4-, and 6-Month Follow-Up Assessments.

Variable	Cumulative		Simultaneous
	ΔR2	ΔFa	β	t	p	95% CI
Note. There was no evidence of multicollinearity in any of the regression models, as all variance inflation factors were <2.6 for AAAQ measures and <1.4 for SRC measures. SADQ = Severity of Alcohol Dependence Questionnaire; AAAQ = Approach and Avoidance of Alcohol Questionnaire; SRC = Stimulus-Response Compatibility Task; PHDD = percentage of heavy drinking days; df = degrees of freedom.
a Step 1, df = (1,118); Step 2, df = (5,114).
* p < .05.
2-month follow-up
SADQ	.001	.17	.05	.50	.621	[−.14, .24]
AAAQ Inclined-Indulgent	.05	1.20	.29	1.84	.069	[−.02, .61]
AAAQ Obsessed-Compelled			−.18	−1.09	.278	[−.52, .15]
AAAQ Resolved-Regulated			−.08	−.73	.465	[−.29, .13]
SRC alcohol approach bias (d)			.06	.54	.589	[−.16, .28]
SRC alcohol avoidance bias (d)			.13	1.19	.243	[−.09, .36]
4-month follow-up
SADQ	.001	.10	.03	.34	.733	[−1.56, .22]
AAAQ Inclined-Indulgent	.06	1.52	.18	.91	.377	[−.24, .60]
AAAQ Obsessed-Compelled			−.12	−.63	.533	[−.51, .27]
AAAQ Resolved-Regulated			.04	.32	.747	[−.19, .26]
SRC alcohol approach bias (d)			.12	1.23	.221	[−.07, .32]
SRC alcohol avoidance bias (d)			.22	2.06	.043*	[.07, .43]
6-month follow-up
SADQ	.001	.07	.01	.12	.902	[−.18, .21]
AAAQ Inclined-Indulgent	.06	1.50	.10	.44	.671	[−.43, .64]
AAAQ Obsessed-Compelled			−.02	−.10	.920	[−.47, .43]
AAAQ Resolved-Regulated			−.16	−.15	.884	[−.24, .20]
SRC alcohol approach bias (d)			.10	1.02	.309	[−.10, .31]
SRC alcohol avoidance bias (d)			.22	2.08	.039*	[.01, .42]