Are social norms the best predictor of outcomes among heavy drinking college students? Clayton Neighbors, Ph.D. Associate Professor Psychiatry & Behavioral.

Are social norms the best predictor of outcomes

among heavy drinking college students?

Clayton Neighbors, Ph.D.

Associate ProfessorPsychiatry & Behavioral Sciences

Center for the Study Health and Risk BehaviorsUniversity of Washington

This research was supported by National Institute on Alcohol Abuse and Alcoholism Grant R01AA014576

Neighbors, C., Lee, C. M., Lewis, M. A., Fossos, N., & Larimer, M. E. (2007). Are social norms the best predictor of outcomes among heavy drinking college students? Journal of Studies on Alcohol and Drugs, 68, 556-565.

College Student DrinkingCollege Student Drinking

• 80% of US college students drink.

• 40% engage in heavy episodic drinking (5/4 drinks for men/women on at least one occasion in the previous 2 weeks).

• Consequences include:

– c. 1700 alcohol-related deaths per year

– 500,000 alcohol-related injuries

– Criminal behavior

– Academic problems

– Unwanted sexual activity

– Hangovers, fights, embarrassment, etc.

Hingson et al., 2002, 2005; Johnson et al, 2000, 2005; Wechsler et al., 2002

Predictors of DrinkingPredictors of Drinking

• Social influences/Perceived norms are among the strongest and most consistent predictors of heavy drinking among college students (Borsari & Carey, 2003; Perkins, 2002; Wood et al., 2001).

• Really?

– What about sex?

– What about Greek status?

– What about drinking motives?

– What about drinking expectancies?

• What about problems?

• Who cares?

Social NormsSocial Norms

• Descriptive norms: Perception of others’ behavior.

– Typically operationalized as perception of typical students drinking.

• Injunctive Norms: Perceptions of others’ approval.

– Typically operationalized as perceived approval of close others (e.g., friends and parents)

DemographicsDemographicsSex: • Men drink more than women. Differences have been relatively

stable among college students over the past 25 years (Johnston, O’Malley, Bachman, & Schulenberg, 2005).

• Men also report more problems than women, although the problems tend to be more public (Perkins, 2002).

Greek membership: • Greek students drink more and experience more problems than

non-Greek students (Borsari and Carey, 1999; Cashin et al., 1998; Kahler et al., 2003; Larimer et al., 2000; Lo & Globetti, 1995; Meilman et al., 1999) .

• The difference between Fraternity and non-Fraternity men is larger than the difference between Sorority and non-Sorority women (Larimer et al. 2000; 2004).

• Research suggests that both selection and socialization effects contribute to heavier drinking by Greek affiliated students (Baer, Kivlahan, & Marlatt, 1995; Read, Wood, Davidoff, McLacken, & Campbell, 2002; Sher, Bartholow, & Nanda, 2001; Kahler et al., 2003; McCabe et al., 2005).

Drinking MotivesDrinking Motives (Cooper, 1994)

• Types of drinking motives– Social: “because it helps you enjoy a party”– Enhancement: “because you like the feeling”– Conformity: “so you won’t feel left out”– Coping: “to forget about your problems”

• Associated with drinking and problems.

• Social/enhancement are often combined.

• Coping has been more strongly linked to problems.

• Associated with different drinking contexts and patterns.

• Conformity is the weak link.

(Carey & Correia, 1997; Cooper, 1994; Cooper et al., 1995; Hussong, 2003; Martens et al., 2003; MacLean & Lecci, 2000; Mccabe, 2002; Mohr et al., 2005; Neighbors et al., 2004; Park & Levenson, 2002; Read et al., 2003; Stewart and Devine, 2000; Wild & Cunningham, 2001; Wood et al., 1992).

Expectancies Expectancies (Fromme, Stroot, & Kaplan, 1993)

• Positive expectancies: “I would be outgoing”• Evaluation of positive effects: “ Being outgoing is...”• Negative expectancies: “I would neglect my obligations” • Evaluation of negative effects: “Neglecting my obligations is...”

• Complex literature. Diversity of assessment methods.• Associated with drinking and problems.• Positive expectancies more strongly associated with

drinking.

(Burden & Maisto, 2000; DelBoca et al., 2002; Darkes et al., 2004; Fromme, Stroot, & Kaplan, 1993; Jones et al., 2001; Leigh & Stacy, 2004; McNally & Palfai; 2001; Neighbors et al., 2003; Palfai & Ostafin, 2003; Read et al., 2003, 2004; Sher et al., 1996; Wiers et al., 2002; Wood et al., 2001).

MethodMethodBaseline participants for a 5-year indicated prevention trial utilizing web-based

normative feedbackSocial Norms and Alcohol Prevention (SNAP) R01AA014576

Invited to ScreenFirst Year Students

N = 4103

Completed ScreenN = 2105(51.3%)

Met Screening CriteriaN = 896(42.6%)

Completed BaselineN = 818(91.3%)

Participants• 818 (57.6% women) first-year undergraduate students

enrolled at a large west coast university. • Mean age of participants was 18.14 (SD = .46). • Ethnicity: 65.2% white, 24.3% Asian, 10.5% other Procedures • All assessments were web-based.• 20 minute screening survey assessed eligibility for the

intervention study.• Inclusion criteria was reporting consuming 4/5 drinks

for women/men on one occasion during the previous month.

• Eligible students were invited to complete a 45 minute baseline assessment immediately following screening.

MeasuresMeasuresAlcohol consumption: Typical number of drinks per week over the past three months

from the Daily Drinking Questionnaire (Collins et al., 1985)

Alcohol related consequences: Rutgers Alcohol Problems Index (RAPI; White & Labouvie, 1989)

Social Norms:

Descriptive Norm: Perceived number of drinks per week consumed by the typical student on campus. Drinking Norms Rating Form (Baer et. al., 1991)

Injunctive Norms (Friends): How would your friends respond if they knew: (1) you drank alcohol every weekend, (2) you drank alcohol daily, (3) you drove a car after drinking, and (4) you drank enough alcohol to pass out” (Baer, 1994).

Injunctive Norms (Parents): How would your parents respond if they knew: (1) you drank alcohol every weekend, (2) you drank alcohol daily, (3) you drove a car after drinking, and (4) you drank enough alcohol to pass out” (Baer, 1994).

Drinking Motives: Social Motives, Coping Motives, Enhancement Motives, Conformity Motives (Cooper, 1994).

Expectancies: Positive Expectancies, Subjective Evaluation of Positive Effects, Negative Expectancies, Subjective Evaluation of Negative Effects (Fromme et al., 1993).

Primary Analyses Alcohol Consumption

Simultaneous regression evaluating unique contribution of all predictors.

Stepwise regression to evaluate the strongest predictors of alcohol consumption in decreasing order. At each step, the best unique predictor is added until no additional predictors account for a significant unique proportion of variance.

Alcohol Related Consequences Simultaneous regression evaluating unique contribution of all predictors.

Stepwise regression to evaluate the strongest predictors of alcohol consumption in decreasing order.

Consumption as a mediator of the relationship between predictors and consequences. Mediation evident for predictors that are 1) associated with consumption,

2) associated with problems, and 3) no longer significantly associated with problems when controlling for consumption.

Hierarchical regression. Step 1 = alcohol consumption. Step 2 = all predictors.

Correlations

VariablesDrinks Per Week

AlcoholRelatedConsequences

Sex .20*** .09*

Greek Membership .27*** .10**

Descriptive Norm .41*** .17***

Injunctive Norm (Friends) .39*** .27***

Injunctive Norm (Parents) .20*** .16***

Social Motives .17*** .19***

Coping Motives .13*** .35***

Enhancement Motives .28*** .22***

Conformity Motives .03 .24***

Positive Expectancies .15*** .17***

Evaluation of Positive Effects .08* .04

Negative Expectancies .04 .25***

Evaluation of Negative Effects .18*** .30***

Note. *p < .05; **p < .01; ***p < .001

Simultaneous Regression: DV = Drinks Per Week

Note. *p < .05; **p < .01; ***p < .001. R2 = .37.

Predictor B β t d

Sex 3.71 0.17 5.76*** 0.41

Greek Membership 5.03 0.19 6.67*** 0.47

Descriptive Norm 0.32 0.34 11.88*** 0.84

Injunctive Norm (Friends) 2.04 0.18 5.49*** 0.39

Injunctive Norm (Parents) 1.20 0.07 2.16* 0.15

Social Motives -0.13 -0.01 -0.29 0.02

Coping Motives 0.82 0.07 1.93 0.14

Enhancement Motives 1.78 0.16 4.14*** 0.29

Conformity Motives -1.35 -0.11 -2.97** 0.21

Positive Expectancies 0.67 0.03 0.70 0.05

Evaluation of Positive Effects -0.25 -0.01 -0.39 0.03

Negative Expectancies -0.34 -0.01 -0.41 0.03

Evaluation of Negative Effects 1.33 0.06 1.89 0.13

Stepwise Regression: DV = Drinks Per Week

Note. *p < .05; ***p < .001.

Step Predictor R2 R2∆

1 Descriptive Norm 0.169 0.169***

2 Injunctive Norm (Friends) 0.277 0.108***

3 Greek Membership 0.311 0.034***

4 Sex 0.334 0.022***

5 Enhancement Motives 0.356 0.022***

6 Injunctive Norm (Parents) 0.361 0.004*

7 Conformity Motives 0.364 0.003*

8 Coping Motives 0.368 0.004*

Simultaneous Regression: DV = Alcohol Related Consequences

Note. *p < .05; **p<.01; ***p < .001. R2 = .24.

Predictor B β t d

Sex 0.96 0.06 1.89 0.13

Greek Membership 1.30 0.07 2.19* 0.15

Descriptive Norm 0.07 0.10 3.07** 0.22

Injunctive Norm (Friends) 0.75 0.09 2.56* 0.18

Injunctive Norm (Parents) 1.17 0.09 2.66** 0.19

Social Motives 0.25 0.03 0.69 0.05

Coping Motives 1.85 0.21 5.50*** 0.39

Enhancement Motives 0.53 0.07 1.57 0.11

Conformity Motives 0.04 0.00 0.12 0.01

Positive Expectancies -0.42 -0.02 -0.56 -0.04

Evaluation of Positive Effects -0.52 -0.04 -1.02 -0.07

Negative Expectancies 3.35 0.18 5.05*** 0.36

Evaluation of Negative Effects 1.99 0.13 3.56*** 0.25

Stepwise Regression: DV = Alcohol Related Consequences

Note. *p < .05; **p<.01; ***p < .001.

Step Predictor R2 R2∆

1 Coping Motives 0.121 0.121***

2 Injunctive Norm (Friends) 0.162 0.041***

3 Negative Expectancies 0.195 0.033***

4 Evaluation of Negative Effects 0.210 0.015***

5 Descriptive Norm 0.221 0.011**

6 Injunctive Norm (Parents) 0.228 0.006*

7 Greek Membership 0.368 0.004*

Hierarchical Regression: DV = Alcohol Related Consequences

Note. *p < .05; **p<.01; ***p < .001.

Predictor Step 1 R2 = .19. B β t d

Drinks Per Week 0.31 0.44 13.76*** 0.97

Predictors Step 2 R2 = .33.

Sex -0.03 0.00 -0.06 0.00

Greek Membership -0.04 0.00 -0.06 0.00

Descriptive Norm -0.02 -0.03 -0.92 -0.06

Injunctive Norm (Friends) 0.21 0.03 0.74 0.05

Injunctive Norm (Parents) 0.85 0.07 2.05* 0.14

Social Motives 0.29 0.03 0.83 0.06

Coping Motives 1.63 0.19 5.13*** 0.36

Enhancement Motives 0.06 0.01 0.18 0.01

Conformity Motives 0.40 0.04 1.19 0.08

Positive Expectancies -0.59 -0.03 -0.84 -0.06

Evaluation of Positive Effects -0.46 -0.03 -0.94 -0.07

Negative Expectancies 3.44 0.18 5.51*** 0.39

Evaluation of Negative Effects 1.63 0.11 3.10** 0.22

Discussion Among heavy drinking students social norms are

the best predictors of alcohol consumption.

Problems are most strongly associated with affect regulation (coping, negative expectancies).

Alcohol consumption mediates the relationship between most predictors and alcohol problems. Coping and negative expectancies are independently related to problems.

Discussion Limitations

Restricted sample—only those who reported at least one heavy episode in the previous month.

Entering first year students.

Cross-sectional – self report.

Disconnection in reference groups between descriptive and injunctive norms.

Assessment of problems (RAPI).

Social influence versus affect regulation

Consumption versus problems

Intervention implications

Your thoughts?

Correlations