Table 3.
Alcohol Use as a Mediator of the Association of Belief in an AIAN Specific Biological Vulnerability (BV) with Substance Use Consequences
| Belief in BV → alcohol use models | Drinking days (R2 =.18) |
Drinks per drinking day (R2 =.10) |
||||
|---|---|---|---|---|---|---|
| B (SE) | β | f 2 | B (SE) | β | f 2 | |
| Belief in BV | 0.42 (0.17) | .25* | .05 | −0.02 (0.10) | −.02 | .00 |
| Substance use group | 0.59 (0.43) | .23 | .01 | 0.19 (0.25) | .13 | .00 |
| BV × Substance use group | −0.86 (0.27) | −.51** | .07 | −0.06 (0.16) | −.07 | .00 |
| Covariates | ||||||
| Age | 0.07 (0.02) | .30*** | .11 | −0.01 (0.01) | −.07 | .00 |
| Gender | −1.03 (0.42) | −.39* | .04 | −0.86 (0.24) | −.59*** | .09 |
| Alcohol use → consequence models | Consequences (R2 = .21) | Consequences (R2 = .14) | ||||
| B (SE) | β | f 2 | B (SE) | β | f 2 | |
| Drinking days | .23 (0.09) | .29* | .05 | — | — | — |
| Drinks/day | — | — | — | 0.24 (0.17) | .16 | .01 |
| Belief in BV | 0.15 (0.14) | .11 | .01 | 0.25 (0.14) | .18 | .02 |
| Substance use group | 1.12 (0.34) | .52** | .08 | 1.23 (0.36) | .57*** | .09 |
| Drinking days × Substance use group | 0.07 (0.13) | .08 | .00 | — | — | — |
| Drinks/day × Substance use group | — | — | — | −0.08 (0.24) | −.05 | .00 |
| BV × Substance use group | −0.05 (0.22) | −.04 | .00 | −0.26 (0.23) | −.19 | .01 |
| Covariates | ||||||
| Age | 0.02 (0.02) | .13 | .01 | 0.04 (0.01) | .22** | .05 |
| Gender | 0.22 (0.34) | .11 | .00 | 0.12 (0.37) | .06 | .00 |
Note. Substance use group was coded: alcohol use-only group=0 and alcohol and hard drug use group=1. Gender was coded: men=0 and women=1. Substance use consequences, drinking days, and drinks per drinking day were square-root transformed prior to analyses. An effect size (f 2) of .02 is conventionally interpreted as small, .15 as moderate, and .35 as large (Cohen, 1988).
*
p < .05.
**
p < .01.
***
p < .001.