Alcohol Policy Support Among Mandated College Students

Abstract

Background

Alcohol consumption on college campuses is high, and often dangerous. College administrators have created policies to control alcohol consumption, but student body support or opposition of specific policies has been relatively unexplored.

Method

The current study examined the relations of alcohol policy support with gender and alcohol consumption. Mandated students (N = 229; 44% women) completed self-report assessments of alcohol policy support and alcohol consumption.

Results

Women supported policies to a greater extent than did men, as did lighter drinkers relative to heavier drinkers. Drinks per drinking day fully mediated the relation between gender and alcohol policy support.

Conclusion

While alcohol policy support differs by gender, this covariation is explained by differences in alcohol consumption. Findings have implications for addressing alcohol policy support among mandated college students.

1. Introduction

Some campus alcohol policies enjoy support among college students (DeJong, Towvim, & Schneider, 2007; Saltz, 2007; Wechsler, Lee, Kuo, Seibring, Nelson, & Lee, 2002; Wechsler, Lee, Nelson & Kuo, 2002). However, policy support appears to vary by policy content, such that approval rates for a single policy have ranged from 14.6% (i.e., “Eliminate low-price bar and liquor store promotions targeted to college students”) to 90.1% (i.e., “Use stricter disciplinary sanctions for students who engage in alcohol-related violence”) (DeJong et al., 2007). Policies vary by relevance to the average student and also in the underlying message frame (gain, loss, more or less punitive), which may influence support (Dorfman, Wallack, & Woodruff, 2005). However, little is known about the reasons for differential student support for campus alcohol policies.

Student characteristics may influence alcohol policy support. Heavier drinkers were less supportive of alcohol policies than lighter drinkers (Wechsler, Lee, Nelson, et al., 2002; Lavigne, Witt, Wood, Laforge, & DeJong, 2008), and women supported alcohol policies more than men (Lavigne et al., 2008). The relation between gender and drinking is well established (see Substance Abuse and Mental Health Services Administration, 2009), with men reporting more drinking than women. However, it is not clear if gender and alcohol consumption are independent or overlapping predictors of student alcohol policy support.

All prior research regarding alcohol consumption and alcohol policy support with college students has used volunteer samples. Students who volunteer for research studies may systematically differ from those ultimately affected by alcohol policy implementation. Therefore, it would be informative to assess predictors of alcohol policy support among students directly affected by them; specifically, students in the process of completing mandated sanctions associated with campus alcohol policies. Investigating mandated students’ attitudes toward alcohol policies provides unique information from students being reprimanded for policy violation. Thus, the goal of the present research is to examine the relations among gender, alcohol consumption, and alcohol policy support with a mandated sample. Specifically, we will test the hypothesis that alcohol consumption mediates the gender-alcohol policy support relation.

2. Method

2.1. Participants

Participants (N = 229; 44% women) were undergraduates (M_age = 18.56 years, SD = .72) attending a private university in the northeastern United States. All had violated campus alcohol policy and were required to participant in an alcohol educational program. Participants were predominately White (85%), first year (65%) students. Eligibility was based on (a) at least 18 years old, (b) the violation was alcohol-related, with no drug involvement, and (c) no previous disciplinary sanctions. Of the 454 sanctioned students referred, 229 (50%) consented to the study and completed the survey.

To characterize the non-consenters, limited descriptive data were collected from a subset of 116 non-consenting students (35% women; M_age = 18.48 years, SD = .68). Non-consenters were also predominately White (79%), first year (77%) students.

2.2. Measures

2.2.1. Alcohol Policy Support (APS)

Ten questions adapted from previous research (see DeJong et al., 2007; Lavigne et al., 2008; Wechsler, Lee, Kuo, et al., 2002) and themes found in alcohol policy literature assessed APS (see Table 1). Participants reported to what extent they supported or opposed the 10 items according to a 0 (Strongly Oppose) to 3 (Strongly Support) scale. Ratings for all 10 items were averaged to obtain a summary score for APS (Cronbach’s α = .83).

Table 1.

Comparison of APS between the current sample, DeJong and colleagues (2007), Lavigne and colleagues (2008), and Wechsler, Lee, Kuo, and colleagues (2002).

Policy	Mandated Sample		Volunteer Sample
	Current Samplea	DeJong et al. (2007)a	Lavigne et al. (2008)b	Wechsler, Lee, Kuo, et al. (2002)b
Prohibit kegs on campus	30%	56%	–	60%
Offer alcohol-free dorms	74%	–	–	89%
Require non-alcoholic beverages be available when alcohol is served at campus events and parties	89%	–	–	–
Ban advertisements of alcohol availability at campus events and partiesc	24%	52%	34%	55%
Provide more alcohol-free recreational and cultural opportunities such as movies, dances, sports, and lectures	86%	–	65%	89%
Make the alcohol rules more clear	82%	–	–	93%
Enforce the alcohol rules more strictly	15%	–	–	63%
Crack down on drinking at sororities and fraternities	13%	–	45%	56%
Hold hosts responsible for problems arising from alcohol use	34%	–	49%	55%
Crack down on under-age drinkingd	22%	46%	31%	–

Note. Policy wording slightly differed between studies; items that varied substantially are noted.

^aPercentage of students who reported support or strong support.

^bPercentage of students who reported support.

^cDeJong and colleagues’ (2007) item, “Restrict advertising that promotes alcohol consumption at on-campus parties or events;” Lavigne and colleagues’ (2008) item, “Restricting advertising that promotes excessive alcohol consumption at bar or events;” and Wechsler, Lee, Kuo, and colleagues’ (2002) item, “Ban alcohol advertisements on campus.”

^dDeJong and colleagues’ (2007) item, “Conduct undercover operations at bars, restaurants, and liquor stores to increase compliance with underage drinking laws;” and Lavigne and colleagues’ (2008) item, “Having undercover operations to increase enforcement of underage drinking laws.”

2.2.2. Alcohol Consumption

The Daily Drinking Questionnaire (Collins, Parks, & Marlatt, 1985) measured alcohol consumption. Participants completed a chart indicating the typical number of drinks they consumed each day of the week during the past month. Drinks per drinking day (DDD) was used as a representative measure of consumption.

2.3. Procedure

The Institutional Review Board approved all procedures, and a Certificate of Confidentiality was obtained for this study. Referred students met with a research assistant who gave them the option to participate in the current alcohol educational study or complete an online alcohol education program. Students who selected to participate in the study provided informed consent and then completed the online survey.

2.4. Analytic Strategy

Normality of DDD and APS was examined using the Kolmogorov-Smirnov Test. Measures of association were then conducted between APS and the two predictors, DDD and gender.

The test of mediation followed Baron and Kenny’s (1986) guidelines. Linear regressions were conducted with (a) gender as the predictor and APS as the criterion (Step 1); (b) gender as the predictor and DDD as the criterion (Step 2); and (c) gender and DDD as predictors and APS as the criterion (Step 3).

3. Results

3.1. Descriptive Statistics

To assess for recruitment bias, we compared data collected from 116 of the non-consenters to the 229 consenters on demographic variables. Consenters and non-consenters differed by freshman status, χ²(1) = 4.78, p = 0.03, with a greater percentage of non-freshmen (75%) than freshmen (62%) willing to participate in the study, and by “other” ethnicity, χ²(1) = 17.77, p < .01, with a greater percentage non-consenters identifying as “other” (7.8%) than consenters (0%); it should be noted that only nine participants identified as “other.” Age, gender, and alcohol consumption did not significantly differ between groups (ps > 0.05). The majority of non-consenters (61%) chose the alternative educational program because it could be completed on students’ own time from anywhere.

Participants averaged 13.92 (SD = 13.80) drinks per week and 4.51 (SD = 2.95) drinks per drinking day. The Kolmogorov-Smirnov Test revealed non-normality for DDD (Z = 2.41, p < .01). DDD was corrected to better approximate normality using a logarithm transformation; the transformed variable was used in all subsequent analyses. On average, men reported more DDD than women (t[225] = 4.13, p < .01). Raw DDD means for men and women were 5.18 (SD = 2.99) and 3.66 (SD = 2.68), respectively. The mean level of APS was 1.37 (SD = .47); men (M = 1.28, SD = .48) and women (M = 1.49, SD = .43) differed on APS, (t[219] = −3.28, p < .01). DDD moderately and negatively correlated with APS (r = −.43, p < .01).

Table 1 displays percentages of students who supported each policy in the current study, as well as percentages that supported similarly worded policy items in previous studies (see DeJong et al., 2007; Lavigne et al., 2008; Wechsler, Lee, Kuo, et al., 2002). In the current sample, the majority of participants supported “Offer alcohol-free dorms”, “Require non-alcoholic beverages be available when alcohol is served at campus events and parties”, “Provide more alcohol-free recreational and cultural opportunities such as movies, dances, sports, and lectures”, and “Make the alcohol rules more clear.” The policies least supported by this mandated sample pertained to stricter enforcement of campus alcohol policy and underage drinking laws.

3.2. Mediation Analysis

Gender predicted APS (R² = .05, F[1,219] = 10.76, p < .01), with women supporting alcohol policies more than men. Gender predicted DDD (R² = .07, F[1,225] = 17.03, p < .01), with men reporting more DDD. DDD predicted APS (R² = .19, F[2,217] = 50.19, p < .01), with more DDD associated with less APS. The regression coefficient for gender lost significance when DDD was included in the analysis (R² = .20, F[2,216] = 26.85, p < .01). Thus, DDD fully mediated the relation between gender and APS (see Figure 1).

Figure 1.

Mediation model for drinks per drinking day and the association between gender and alcohol policy support among mandated college drinkers. **p≤.01.

4. Discussion

The present study explored APS in a sample of college drinkers sanctioned for violating campus alcohol policy. Both gender and alcohol consumption predicted APS; however, alcohol consumption fully mediated the relation between gender and APS.

As displayed in Table 1, volunteer students support some alcohol policies (DeJong et al., 2007; Wechsler, Lee, Kuo, et al., 2002; Wechsler, Lee, Nelson, et al., 2002). The current study replicated these findings with a mandated sample. Many policies were supported by a lower percentage of students in this mandated sample relative to previous volunteer samples (see DeJong et al., 2007; Lavigne et al., 2008; Wechsler, Lee, Kuo, et al., 2002). In particular, mandated students expressed the least support for items that were more punitive (i.e., contained such words as “ban,” “prohibit,” or “crack down”), whereas items that were less punitive and actively promoted autonomy (i.e., provided options to participants) were supported more strongly. Mandated students may be less likely to support punitive policies because these students are in the process of satisfying sanctions for violating alcohol policies; therefore, their responses may be more influenced by their immediate context. On the other hand, mandated students may be more likely to support policies that promote autonomy and support free will. Messages framed to support core values, such as personal responsibility and free will, may be more appealing and may engender more support (Dorfman et al., 2005). Hence, students’ disciplinary status and message framing may play key roles in alcohol policy support among college students. To that end, policy makers should develop policy items that appear less punitive and promote autonomy, which may increase student support.

Volunteer female and male research participants differ on APS (Lavigne et al., 2008). The current study extended this finding showing that mandated male college students were less supportive of alcohol policies than mandated female college students. Thus, gender is an important predictor of APS among college students regardless of whether the sample consists of mandated or volunteer participants.

The current study revealed that alcohol consumption mediates the relation between gender and APS. We cannot rule out the influence of additional factors that may also be related to gender and drinking behaviors (e.g., conformity to social norms, compliance with authority), and encourage additional research on the role of these social-cognitive factors on APS. However, we have demonstrated that the frequently observed gender differences are largely explained by how much an individual drinks on a typical drinking day. It should be noted that Lavigne and colleagues (2008) observed a relation between gender and APS even after controlling for alcohol consumption. One difference between the current sample and Lavigne’s volunteer sample is that the volunteer sample was a lighter drinking sample and had a larger gender difference in reported alcohol consumption. It is possible that these differences may have influenced results. Nevertheless, the current finding indicates that a primary determinant of mandated college drinkers’ support or opposition of alcohol policies is alcohol consumption.

This study had several limitations. The students were sampled from only one university; therefore results may not be generalizable to all college students. Only 10 policies were assessed which did not cover all possible policy options. For example, policy options such as “Increasing enforcement of drinking and driving laws” or “Training bartenders to cut off intoxicated patrons” (Lavigne et al., 2008) were not assessed in the current study. Finally, we were not able to explicitly compare APS for mandated and volunteer students in this study, so we relied on comparisons to other samples. Thus, the relative levels of APS are suggestive and should be confirmed within a single study.

College drinkers are at high risk for experiencing alcohol related negative consequences (Wechsler, Davenport, Dowdall, Moeykens, & Castillo, 1994), especially the heavier drinking students (Wechsler et al., 1994). Notable, states with stricter alcohol policies have lower reported rates of heavy drinking among college students (Nelson, Naimi, Brewer, & Wechsler, 2005). By extension, alcohol policy enforcement on college campuses may attenuate the risk of experiencing alcohol related negative consequences. Thus, administrators should be mindful that heavier drinkers are less supportive of alcohol policies, and are the ones most affected by campus wide restrictions on alcohol as well as the ones who are sanctioned for violating the policies. With this in mind, clarifying benefits of policy enforcement to all students, including heavier drinkers, might garner more support; this might include student safety and maintaining attractive living spaces. Moreover, administrators should consider framing policies to promote autonomy, rather than hinder it. This may increase support among students and behavioral adherence to policies.