Determinants of Underage College Student Drinking: Implications for Four Major Alcohol Reduction Strategies

Abstract

Guided by the assumptions of the social ecological model and the social marketing approach, this study provides a simultaneous and comprehensive assessment of 4 major alcohol reduction strategies for college campuses: school education programs, social norms campaigns, alcohol counter-marketing, and alcohol control policies. Analysis of nationally representative secondary survey data among 5,472 underage students reveals that alcohol marketing seems to be the most formidable risk factor for underage drinking, followed by perceived drinking norms (injunctive norm) and lax policy enforcement. This analysis suggests that, to make social norms campaigns and alcohol control policies more effective, alcohol reduction strategies should be developed to counter the powerful influence of alcohol marketing and promotions.

Notes

¹Social norms marketing has adopted a variety of approaches to target heavy-drinking college students, including providing them with personalized normative feedback (Neighbors, Larimer, & Lewis, 2004; Berkowitz, 2004). With a slightly narrower focus, the present study instead closely follows the original conception of social norms marketing.

^a Dummy variables with male = 1, female = 0; White = 1, other races = 0; fraternity/sorority membership and residence = 1, no = 0.

^b The question wordings were as follows: “Describe your father's/mother's (or the person who served as your father/mother in raising you) use of alcohol during most of the time that you were growing up,” reported on a 5-point scale ranging from 1 (abstainer or in recovery) to 5 (problem drinker).

^c This variable was measured by asking, “How important is it for you to participate in religious activities at college?” reported on a 4-point scale ranging from 1 (not at all important) to 4 (very important).

^d This variable was measured by averaging the following question items reported on a 4-point scale ranging from 1 (not at all important) to 4 (very important): (1) to get away from troubles; (2) to relax or relieve tension; (3) to get drunk; (4) to have a good time with my friends; (5) to celebrate; (6) as a reward for working hard.

^e This variable was measured by averaging the following question items: (1) to fit in with my friends; (2) to feel more comfortable when I'm with the opposite sex; and (3) everyone else is drinking.

²Worth mentioning are descriptive statistics on underage students’ drinking patterns. About 62% of underage students drank on at least one or two occasions in the past 30 days. Also, when they drank alcohol, they reported that they usually had more than four drinks. Approximately 48% of underage students reported that they had more than five drinks in a row at least once in the past 2 weeks, an amount that is often regarded as binge drinking (or heavy episodic drinking). Only less than 26% described themselves as abstainers, whereas the other students described themselves as infrequent drinkers (29.7%), light drinkers (18.8%), moderate drinkers (22.2%), and heavy or problematic drinkers (2.7%). Independent samples t test also shows that men tended to drink more alcohol and more frequently than did women.

³It should be noted that our social reward drinking motive includes the three motives from Cooper's four-factor drinking motive scales (2004): coping motive (“get away from troubles”), social motive (“have a good time with my friends” “celebrate”), and enhancement motive (e.g., “get drunk”). The reason for this inconsistency may be that the CAS data do not employ all the 20-item motive questions from Cooper's scale. Regardless, we used the drinking motive variables for control purposes, not for validation purposes. Hence, the inconsistent scales should not substantially affect our findings. Future research could further explore the validity of the drinking motive scales.

^a Pearson correlation coefficient.

^b Betas are standardized regression coefficients, taken from the final equation with all the predictors entered.

*p < .05. **p < .01. ***p < .001.

⁴We tested moderation effects to explore the extent to which education, social norms, and policy strategies can interact with one another to counter the powerful influence of alcohol marketing. A total of ten sets of interaction terms were created: (1) School Education × Descriptive Norm, (2) School Education × Injunctive Norm, (3) School Education × School Policy, (4) School Education × Community Policy, (5) Descriptive Norm × Injunctive Norm, (6) Descriptive Norm × School Policy, (7) Descriptive Norm × Community Policy, (8) Injunctive Norm × School Policy, (9) Injunctive Norm × Community Policy, and (10) School Policy × Community Policy. Bivariate correlations indicate that, except for (1) School Education × Descriptive Norm and (3) School Education × School Policy, all the remaining interaction terms were significantly negatively related to drinking behavior. However, when these interaction terms were entered in the regression model, none of them remained significant.