Comparing injunctive marijuana use norms of salient reference groups among college student marijuana users and nonusers

Abstract

Marijuana is the most commonly used illicit drug among college students and has the potential for various negative outcomes. Perceptions of what constitutes typical approval/acceptability of a reference group (i.e. injunctive social norms) have been shown to have strong utility as predictors of health-risk behaviors in the college context, yet this construct remains significantly understudied for marijuana use despite its potential for use in social norms-based interventions. The current research evaluated individuals’ marijuana approval level and their perceptions of others’ marijuana approval level (i.e. injunctive norms) for various reference groups (typical student on campus, one’s close friends, one’s parents) as a function of individual user status (abstainers, experimenters, occasional users, regular users). A diverse sample of 3553 college students from two universities completed an online survey. Among all user status groups, individual approval yielded mean scores paralleling that of perceived close friends’ approval and all groups were relatively uniform in their perception of typical students’ approval. Higher levels of marijuana use tended to produce higher endorsements of individual approval, perceived close friends’ approval, and perceived parental approval. Among occasional and regular users, there were no differences between one’s own approval level for use and the perceptions of close friends’ approval, nor did they think the typical student was more approving than themselves. Abstainers and experimenters, however, perceived typical students and close friends to have more permissive attitudes than themselves. Implications and future directions for research regarding the role of injunctive marijuana use norms in the development of social norms intervention are discussed.

1. Social norms and marijuana

Social norms theory (see Berkowitz, 2004; Perkins, 2003) asserts that indirect peer influence, in the form of perceptions, acts on an individual’s own behavior regardless of the accuracy of the perceived norm. The construct of perceived approval or attitude of others (i.e. injunctive norm; Cialdini, Reno, & Kallgren, 1990) has been used in frameworks for understanding a wide range of behaviors (see review by Ajzen, 1991) including substance use (e.g., Conner, Warren, Close, & Sparks, 1999; Norman & Conner, 2006). Yet the functionality and influence of injunctive norms have been understudied in marijuana users. In one study assessing a sample of high school graduates during the summer prior to attending college, the authors found a unique and positive relationship between both descriptive and injunctive norms on marijuana use (Neighbors, Geisner, & Lee, 2008). Given the paucity of extant research and potential applicability of findings for use in prevention and intervention efforts, including how to best focus such efforts to specific user types, the present study focused exclusively on injunctive marijuana norms.

1.1 Specificity of normative referents

Recent evidence (Borsari & Carey, 2003; Larimer et al., 2009; Neighbors et al., 2010; Neighbors, Lee, Lewis, Fossos, & Walter, 2009) supports several theoretical perspectives (i.e., Social Comparison, Festinger, 1954; Social Impact Theory, Latané, 1981; Social Identity Theory, Hogg, Abrams, Otten, & Hinkle, 2003, Terry & Hogg, 1996) suggesting that the reference groups to which individuals are closely connected by proximity or identification are more relevant and have greater influence on individual behavior and attitudes than reference groups to which individuals are remotely connected. Because injunctive norms define the social approval by important others, the endurance and influence of these norms on marijuana use may differ according to the level of proximity of the normative referent. Therefore, a necessary step to extend normative research on injunctive norms and marijuana use requires first identifying what types of perceptions different users have of different reference groups, relative to their own approval level. Secondly, it is important to compare the perceptions of the different reference groups by individual user status, to garner a sense of the strength of similarity or dissimilarity between perceptions, as a function of how frequently the individual actually uses marijuana. These two components provide data that can then help to answer an important question in social norms research: what types of normative education are appropriate and effective, for which health-risk behaviors, and for whom (Mattern & Neighbors, 2004).

1.2 Study aims and hypotheses

The current study evaluated individuals’ marijuana approval level and their perceptions of others’ marijuana approval level for various reference groups (typical student on campus, one’s close friends, one’s parents) as a function of individual user status (abstainers, experimenters, occasional users, regular users). It was generally anticipated that higher levels of individual marijuana use would also reflect progressively higher levels of individual approval and perceptions of others’ approval. We also hypothesized that individual approval levels would be most closely aligned to perceived close friend approval and that regardless of user status, students would perceive the typical student to hold more permissive attitudes than themselves. Finally, we expected perceived parental approval to obtain the lowest approval score regardless of individual user status.

2. Materials and methods

2.1 Participants

Participants were 3553 (61% female) students at two west-coast campuses, one a large public university and the other a private mid-sized university. They had a mean age of 19.88 years (SD = 1.36) and were 54.7% Caucasian, 18.5% Asian, 12.7% Hispanic/Latino, 6.8% Multiracial, 3.0% African American, and 4.3% other.

2.2 Design and procedure

During the first two weeks of the fall semester, 7,000 randomly selected students received invitations to participate in an online survey, Upon indicating their consent on an IRB-approved form and entering their unique pin number, they were routed to the confidential survey. Participants received $20 for completing this 30 minute baseline survey. .

2.3 Measures

Measures relevant to the current study included demographics, perceived injunctive marijuana norms, and individual marijuana use.

2.3.1 Individual attitudes and injunctive norms

Participants were first asked about their own approval of four items regarding the frequency of marijuana use: (1) abstaining from marijuana use, (2) trying marijuana once or twice, (3) smoking marijuana occasionally, and (4) smoking marijuana regularly. Response options were on a Likert scale ranging from 1 (strongly disapprove) to 7 (strongly approve).

Injunctive norms were assessed by asking participants about their perceptions of how much a typical student on their campus, their close friends, and their parents approved of the same four items, using the same Likert-type response scale. Item 1 (referring to abstaining from use) was reversed scored. Composites were then computed for the actual/self approval (α = .77), perceived typical student approval (α = .69), perceived close friend approval (α = .80), and perceived parental approval (α = .69).

2.3.2 Marijuana use

The frequency of marijuana use was assessed via the following question: “In the past year, how many occasions did you use marijuana?” Response options were on a Likert scale ranging from 0 (never) to 6 (40 or more times).

3. Results

3.1 Classification of marijuana user status

For the purpose of the investigation, the following categorization was applied to define the four user status groups, based on the prevalence of use in the past year: abstainers (0 times; n = 2,134), experimenters (1–2 times; n = 470), occasional users (3–19 times; n = 588), and regular users (20+ times; n = 361). Thus, 40.0% of the sample used marijuana at least once in the past year.

3.2 Repeated-measures analytic model

A 4 × 4 between- and within-subjects ANOVA model was conducted to examine differences in self/actual marijuana approval as well as perceived approval of the various referent groups as a function of respondent user status. Respondent marijuana user status (abstainers, experimenters, occasional users, and regular users) was specified as the between-subjects factor. Marijuana approval measures (self/actual approval, perceived typical student approval, perceived close friends’ approval, and perceived parental approval) served as the within-subjects factor.

Results revealed highly statistically significant main effects for marijuana user status, F(3, 3549) = 488.71, p < .001, and for marijuana approval, F(3, 10647) = 1999.91, p < .001. A marijuana user status × marijuana approval interaction also emerged, F(9, 10647), p < .001. Mean scores for each approval measure, as a function of marijuana user status, are graphed in Figure 1.

Figure 1.

Self/actual marijuana approval and perceived marijuana approval of referent groups as a function of user status.

3.3 Decomposition of the interaction via pair-wise differences

The statistically significant interaction was decomposed in a series of follow-up between-subjects and within-subjects t-tests. Results are presented in Table 1. Specific between-subjects differences were initially examined. The four status groupings of marijuana usage differed statistically, in all possible comparisons, on self/actual approval, perceived close friends’ approval, and perceived parental approval, ps < .001. Experimenters and occasional users each reported significantly higher means on perceptions of typical student approval than abstainers (ps < .001), but no other differences were discovered on typical student approval.

Table 1.

Marijuana User Status Differences (Between-Subjects) and Marijuana Approval Differences (Within-Subjects)

	Marijuana User Status
	Abstainers (A)		Experimenters (E)		Occasional Users (O)		Regular Users (R)		Between- Subjects F test	Between- Subjects Contrasts (p < .001)
Marijuana Approval	M	(SD)	M	(SD)	M	(SD)	M	(SD)
Self/actual approval	2.45a	(1.14)	3.61a	(0.83)	4.11a	(0.84)	4.79a	(0.97)	816.86*	A < E < O < R
Perceived typical student appoval	3.85b	(1.10)	4.13b	(0.86)	4.07a	(0.93)	3.96b	(1.00)	13.70*	A < E; A < O
Perceived close friends approval	2.95c	(1.35)	3.92c	(1.06)	4.21a	(1.04)	4.94a	(1.15)	380.73*	A < E < O < R
Perceived parental approval	1.71d	(0.97)	2.24d	(1.14)	2.49b	(1.14)	2.79c	(1.20)	173.25*	A < E < O < R
Within-Subjects F test	2325.91*		475.30*		537.20*		408.82*

Note. In the same column, means with different subscripts are statistically significant in within-subjects contrasts, p < .001.

Sample sizes for Abstainers (n = 2134), Experimenters (n = 470), Occasional Users (n = 588), and Regular users (n = 361).

^*p < .001.

Next, specific within-subjects effects were examined. In both the abstainer and experimenter groups, mean differences were evidenced on all four of the marijuana approval composites in all comparisons. The highest score emerged on perceived typical student approval, followed by perceived close friends’ approval, self/actual approval, and perceived parental approval, ps < .001. Occasional users estimated that parental approval was significantly lower than self/actual approval, perceived approval of typical student, and perceived approval of close friends, ps < .001, which were each not significantly different. Among regular users, both self/actual approval and perceived close friends’ approval were reported to be significantly higher than perceived typical student approval, which in turn was significantly higher than perceived parental approval, ps < .001.

4. Discussion

Consistent with hypotheses, a general pattern across user statuses revealed that higher levels of individual marijuana use tended to equate to progressively higher levels of individual approval, perceived close friend approval, and perceived parental approval of marijuana use. However, contrary to hypotheses, the perceptions of typical student approval of marijuana were relatively stable across all groups. Within user status group findings revealed that for abstainers and experimenters, their own approval of marijuana use was significantly lower than both perceived close friend and perceived typical student approval. For occasional and regular users, there was no significant difference between their own and perceived close friends levels of marijuana approval. Among regular users, their own approval and the perceived approval of their close friends was significantly higher than their perceptions of typical student approval.

Utilizing injunctive norms may nuance the potential applicability of normative feedback interventions targeting college student marijuana users. For abstainers and experimenters, the focus should be on correcting their normative beliefs and reinforcing their moderate attitudes and behaviors as recent evidence suggests that college students overestimate the actual prevailing approval levels about marijuana use held by fellow students (LaBrie, Hummer, Lac, & Lee, 2010). While this may be a reasonable platform for prevention efforts among abstainers and experimenters, it may not be a potentially beneficial intervention approach for more regular marijuana users—the students most in need of harm-reducing interventions.

Social norms interventions operate on the premise that normative feedback is effective because users are able to compare their attitudes and behaviors to those of a salient referent group. It is expected that persons receiving the feedback will want to adjust their attitudes and behaviors to more closely reflect those of their peers. However, given that the heaviest users in this sample appear to know they are more approving of marijuana use than other students, and consistent with problem behavior theory and deviant behavior theory (Jessor & Jessor, 1977), a social norms intervention may not be a particularly effective form of treatment. Regular users may see themselves as distinct and different from other students (Pedersen, 2009; Suchman, 1968) and, thus, may choose to associate with heavier-using and more approving peers. From this perspective, there are two particularly salient hypothetical scenarios with implications resulting from the current data. The first is that providing feedback about the approval level of one’s close friends may be an important mechanism for positive behavioral change in the heavy using group. Yet this strategy is highly dependent on whether misperceptions exist between one’s own approval level and the approval of one’s friends, particularly among heavier users. If these students overestimate how approving their friends really are about marijuana use, highlighting the discrepancy between the student’s own approval level, his/her perceptions of friends, and the actual approval level of his/her peers may be an effective intervention strategy.

The second possible scenario is that heavier users accurately know their close friends’ attitudes and usage levels, in which case, normative education about even that reference group would not be appropriate as it may simply reinforce existing permissive attitudes or heavier use. Thus, the documentation of misperceptions regarding the injunctive norms of close friend and other social friend networks is a very necessary avenue for future research. Such information would help to differentiate the role of perceived close friend injunctive norms in the development of personal attitudes about marijuana use and provide important insight into their potential clinical utility for use in prevention and intervention programs.

Finally, a positive linear relationship was present between individual marijuana use levels and perceived parental approval. A recent meta-analytic review examining the relationship between parental monitoring and marijuana use disclosed a robust link between parental monitoring and lower marijuana consumption in adolescents (Lac & Crano, 2009). It may therefore be worth exploring the value of strengthening the communication and appropriate monitoring between parents and college students about expectations and attitudes regarding marijuana use, particularly if parents are not as approving of use as occasional and regular marijuana users might believe.

4.1 Conclusion

It is important for researchers and other health professionals to continue exploring how drugs other than alcohol can be incorporated into normative reeducation programs and in what ways these efforts and their functionality are congruent or discrepant when compared to the social norms approach to college student drinking.

Research Highlights.

• This study evaluated 3553 college students’ marijuana approval level and their perceptions of others’ marijuana approval level (i.e. injunctive norms) for various reference groups (typical student on campus, one’s close friends, one’s parents) as a function of individual user status (abstainers, experimenters, occasional users, regular users).

• Among all user status groups, individual approval paralleled that of perceived close friends’ approval and all groups were relatively uniform in their perception of typical students’ approval.

• Higher levels of marijuana use tended to produce higher endorsements of individual approval, perceived close friends’ approval, and perceived parental approval.

• Among occasional and regular users, there were no differences between one’s own approval level for use and the perceptions of close friends’ approval, nor did they think the typical student was more approving than themselves. Abstainers and experimenters, however, perceived typical students and close friends to have more permissive attitudes than themselves.