Elsevier

Drug and Alcohol Dependence

Volume 180, 1 November 2017, Pages 349-355
Drug and Alcohol Dependence

Full length article
Cross-cultural examination of college drinking culture in Spain, Argentina, and USA: Measurement invariance testing of the College Life Alcohol Salience Scale

https://doi.org/10.1016/j.drugalcdep.2017.08.016Get rights and content

Highlights

  • Examined internalized norms of college student drinking culture in three countries.

  • Validated a Spanish version of the College Life Alcohol Salience Scale (CLASS).

  • Metric invariance across countries was found for a 12-item version of the CLASS.

  • A 12-item version demonstrates scalar invariance across sex and drinker status.

  • The CLASS has a similar nomological network across the U.S., Argentina, and Spain.

Abstract

Objective

Perceptions about what is “normal” drinking in college, measured by the College Life Alcohol Salience Scale (CLASS; 15 items), have been robustly associated with elevated levels of problematic alcohol use, yet the role of these beliefs has not been studied outside the U.S. The present work examined measurement invariance of the CLASS across sex, drinker status, and in individuals from three different countries (i.e., U.S., Argentina, Spain). Additional goals were to evaluate differences on the CLASS (i.e., latent mean differences) as a function of sex, drinker status and country and to compare construct validity (i.e., correlations with alcohol variables) across sex and different countries.

Method

A large sample of 1841 college students enrolled in universities from the U.S., Spain and Argentina completed, via an online survey, a battery of instruments that assess college alcohol beliefs, drinking motives, alcohol consumption and alcohol-related negative consequences.

Results

We found that a shortened 12-item version of the CLASS to be invariant across sex and drinker status, but only metric invariance was found across countries. As expected, men and drinkers showed significantly higher scores on the CLASS than women and non-drinkers, respectively. Bivariate correlations between CLASS scores and drinking outcomes strongly supported criterion-related validity of this measure across multiple countries and sex with differing strengths in relationships with alcohol-related constructs.

Conclusions

Taken together, perceptions of the centrality of alcohol to the college experience appear to be an important target for college student alcohol interventions across various cultures and countries, especially for male college student drinkers.

Introduction

Globally, extensive research has been conducted identifying risk/protective factors (e.g., personality traits, negative affect, alcohol expectancies) that may contribute to problematic alcohol use among college students, with an eye toward prevention and early intervention (Mallett et al., 2013, Stone et al., 2012, White and Hingson, 2014, Wicki et al., 2010). Perceptions about normative drinking in relevant peer groups (drinking norms) have been found to be a robust risk factor associated with alcohol-related outcomes among college students (Borsari and Carey, 2003, Mallett et al., 2013, Wicki et al., 2010). Yet, most research has focused on two specific types of normative perceptions, descriptive norms (i.e., beliefs about the drinking behaviors of others, Neighbors et al., 2007, Lewis and Neighbors, 2006) or injunctive norms (i.e., beliefs about the degree to which others approve/disapprove of drinking, Neighbors et al., 2008, LaBrie et al., 2010).

Recent research has identified another type of normative perception that may also be an important influence on drinking behavior, perceptions about alcohol and the college ethos. In 2010, Osberg and colleagues developed and validated the College Life Alcohol Salience Scale (CLASS) to assess the internalization of college student drinking culture, or beliefs regarding the degree to which alcohol use is considered an integral part of the college experience. Since its development, the CLASS has been shown to be robustly associated with elevated levels of alcohol use and consequences among college students (Bravo et al., 2017, Osberg et al., 2010, Osberg et al., 2011, Osberg et al., 2012). Moreover, the CLASS has been shown to have stronger associations with negative alcohol-related consequences than many other established predictors of alcohol outcomes (e.g., descriptive norms, injunctive norms, alcohol expectancies; Hustad et al., 2014, Osberg and Boyer, 2016, Pearson and Hustad, 2014, Ward et al., 2015). Together, these findings suggest that this type of normative perception may be an important target for intervention.

Despite the increasing research conducted on the college alcohol beliefs assessed by the CLASS, no published study to date has examined the CLASS outside of the U.S. As such, how these beliefs differ across countries is unknown. Given that drinking on college campuses and college life greatly differs around the world, we believe cross-cultural research examining these beliefs across countries is needed to determine the extent to which these beliefs are an important intervention target in distinct cultural contexts. For instance, living on campus, which has been positively associated with alcohol use (Lorant et al., 2013), is prevalent among U.S. college students, but not among Argentinean or Spanish students. Further, Argentina and Spain are two Spanish-speaking countries in which regular use of alcohol is deeply entrenched in society and associated with meals, family celebrations, and social events. As already noted, it is yet unknown if results from the U.S. extends to other countries with different cultural milieu and language. In that sense, the present study represents an extraordinary opportunity to further explore social norms regarding the role of alcohol within the college context across three countries with similarities and unique cultural and college-related components.

Beyond cultural differences, sex differences in perceptions about alcohol and the college experience have been observed. For example, using the CLASS, researchers have found that men endorse higher college alcohol beliefs compared to women (Bravo et al., 2017; Hustad et al., 2014; Pearson and Hustad, 2014). To date, it is unclear whether these findings regarding sex and college perceptions reflect actual differences in the degree to which men and women hold these perceptions, or whether they instead reflect measurement bias in the measure itself. That is, differences that have been reported may be attributed to sex differences in item responses, rather than to sex-based differences in the latent trait of college alcohol beliefs (Millsap, 2012). Further, the CLASS has been administered among both drinkers and non-drinkers (Osberg et al., 2010, Osberg et al., 2011, Osberg et al., 2012), yet how drinking status may influence such perceptions and whether the CLASS measures these perceptions similarly across drinkers and non-drinkers is unknown.

The goals of the present study were to: a) test the extent to which college alcohol beliefs are captured using the same items (i.e., measurement invariance) across different countries, sex (men vs women), and drinker status (i.e., non-drinkers vs drinkers), b) examine how men vs. women, drinkers vs. non-drinkers, and individuals in different countries compare on the degree to which college students view drinking to be an integral part of the college experience (i.e., latent mean differences), and c) characterize how this construct relates to drinking motives and alcohol-related outcomes across sex and different countries (i.e., comparing construct validity).

Section snippets

Participants and procedures

Participants were college students recruited from four universities across three countries to participate in an online survey regarding personal mental health, personality traits, and alcohol use behaviors (see Bravo et al., in press for more information on recruitment procedures). Although 1864 students were recruited across sites, for the present study only data from students that completed the CLASS (n = 1841) were included in the final analysis from each sample (two U.S. sites combined, n = 

CFAs

The 15-item CLASS provided poor fit to the data based on most fit indices in the total sample [CFI = 0.888, TLI = 0.869, RMSEA = 0.071 (90% CI [0.067, 0.075]), SRMR = 0.048] and in the Argentinean and Spanish subsamples (model fit was acceptable in the U.S. subsample; see Table 1). Given the poor fit of the 15-item version, we conducted post-hoc modifications to produce a better fitting version of the measure in the total sample. Based on model modification indices, item 10 (“Drinking alcohol is a

Discussion

The present study sought to adapt a Spanish version of the CLASS, examine measurement invariance across sex, drinker status, and nationality, and to examine the construct validity of college alcohol beliefs among college students in the U.S., Spain, and Argentina. We found a 12-item version of the CLASS to be scalar invariant across drinker status and sex, but only metric invariant across countries. Further, we found that drinkers reported significantly higher scores on the CLASS than

Conclusions

Consistent with the goals of the present study, we found that a 12-item version of the CLASS demonstrates scalar invariance across sex and drinker status and metric invariance across three countries (U.S., Argentina, and Spain). Supporting previous research, we found that drinkers (compared to non-drinkers) and men (compared to women) reported higher CLASS scores. Extending previous research, the CLASS was robustly associated with alcohol-related outcomes for men and women and across all three

Conflict of interest

No conflict declared.

Role of funding sources

There was no direct funding for this study in the US. However, Dr. Pearson is supported by a career development grant (K01-AA023233) from the National Institute of Alcohol Abuse and Alcoholism (NIAAA) and Dr. Bravo is supported by a training grant (T32-AA018108) from the NIAAA. NIAAA had no role in the study design, collection, analysis or interpretation of the data, writing the manuscript, or the decision to submit the paper for publication. Data collection in Spain was supported, in part, by

Contributors

Dr. Bravo coordinated the efforts of the research team, conceptualized the research questions, conducted the analyses, drafted the introduction, method, statistical analyses, and results sections (including tables). Dr. Pearson assisted with the statistical analyses and edited the introduction, method, and results sections of the first draft. Dr. Pilatti wrote the abstract and part of the discussion section. Dr. Read wrote the clinical implications section of the discussion section and edited

Acknowledgments

No acknowledgments declared.

References (40)

  • F.F. Chen

    Sensitivity of goodness of fit indexes to lack of measurement invariance

    Struct. Equ. Model.

    (2007)
  • G.W. Cheung et al.

    Evaluating goodness-of-fit indexes for testing measurement invariance

    Struct. Equ. Model.

    (2002)
  • R.L. Collins et al.

    Social determinants of alcohol consumption: the effects of social interaction and model status on the self-administration of alcohol

    J. Consult. Clin. Psychol.

    (1985)
  • P.J. Conrod et al.

    Long-term effects of a personality-targeted intervention to reduce alcohol use in adolescents

    J. Consult. Clin. Psychol.

    (2011)
  • M. Cugnon

    A Hop Across the Pond: Comparing European Vs. American College Experiences

    (2015)
  • R.A. Fisher

    Frequency distribution of the values of the correlation coefficient in samples from an indefinitely large population

    Biometrika

    (1915)
  • L. Hu et al.

    Cutoff criteria for fit indexes in covariance structure analysis: conventional criteria versus new alternatives

    Struct. Equ. Model.

    (1999)
  • J.T.P. Hustad et al.

    The role of alcohol perceptions as mediators between personality and alcohol-related outcomes among incoming college student drinkers

    Psychol. Addict. Behav.

    (2014)
  • International Alliance for Responsible Drinking

    Drinking Guidelines: General Population

    (2016)
  • K.G. Jöreskog et al.

    LISREL 8: User’s Guide

    (1993)
  • Cited by (35)

    • A multidimensional view of intellectual capital: The impact on innovation performance

      2021, Journal of Open Innovation: Technology, Market, and Complexity
    • College alcohol beliefs and negative alcohol-related consequences: A moderated mediation model of enhancement drinking motives and restricted eating

      2020, Addictive Behaviors
      Citation Excerpt :

      For the purposes of the present study, we focus on enhancement motives’ potential role as a moderator between CABs and restricted eating. Research indicates enhancement motives are positively associated with CABs (Bravo, Pearson, et al., 2017). Moreover, studies consistently demonstrate enhancement motives are positively associated with positive alcohol expectancies, supporting the notion that beliefs activate motives aimed at attaining certain mood states (Cooper, Kuntsche, Levitt, Barber, & Wolf, 2016; Read, Wood, Kahler, Maddock, & Palfai, 2003).

    • Contexts of alcohol use: A latent class analysis among Argentinean college students

      2020, Drug and Alcohol Dependence
      Citation Excerpt :

      Because of the dichotomous response format (0= no; 1= yes), the total score reflects the number of negative consequences experienced in the previous month. We used the 12-item Spanish version (Bravo et al., 2017) of the College Life Alcohol Salience Scale (CLASS; Osberg et al., 2010). Although the original CLASS is a 15-item measure, this 12-item version was found to be invariant across three countries (i.e., the U.S., Spain and Argentina), sex, and drinker status (Bravo et al., 2017).

    View all citing articles on Scopus
    View full text