Elsevier

Addictive Behaviors

Volume 26, Issue 3, May–June 2001, Pages 385-398
Addictive Behaviors

The College Alcohol Problems Scale

https://doi.org/10.1016/S0306-4603(00)00116-7Get rights and content

Abstract

A short, reliable two-factor instrument measuring drinking-related negative consequences was developed from a previous measure using two samples of college students. In Study I, data on alcohol use and problems associated with alcohol use were collected on 382 introductory psychology students. The original College Alcohol Problems Scale (CAPS) was tested and found to fit the data poorly. Sequential methods were used to develop a revised instrument. Principal components analyses (PCA) on half of the sample were conducted on 20 items written to measure negative consequences related to college student drinking. Results indicated a two-factor solution measuring social and emotional problems. Confirmatory factor analyses (CFA) on the other half of the sample confirmed the two-factor structure. Further refinement of the instrument resulted in the revised CAPS (CAPS-r), an eight-item two-factor scale. In Study II, the response format was altered to coincide with the Young Adult Problem Screening Test. A total of 726 students completed the instrument as part of a university-wide random sample. CFA showed that the hypothesized model fit well across all measures of model fit and the factor structure was invariant across gender. Additional analyses revealed that the scale was internally consistent and externally valid. A short reliable and valid measure of alcohol-related problems is needed to enable low-cost data collection on college campuses across the nation, as well as to facilitate program evaluation and routine epidemiological surveillance and monitoring.

Introduction

Problems with alcohol are a major concern on college campuses today. Efforts to influence the rate of student alcohol problems are increasingly seen to be urgent priorities. Two recent nationwide surveys of college campuses have revealed that heavy drinking is the norm on many campuses and that alcohol-related problems are widespread Wechsler et al., 1994, Wechsler et al., 1998. Psychosocial interventions to reduce heavy, episodic drinking have traditionally not been highly effective (Moskowitz, 1989). However, more recent efforts have shown some reduction in harmful drinking (i.e., Darkes & Goldman, 1993, Darkes & Goldman, 1998, Haines & Spear, 1996, Marlatt et al., 1998).

Good measures of negative consequences related to alcohol consumption are needed to measure primary outcomes for these studies. College-age drinkers, who drink less often than adults but in greater quantities (Harford & Mills, 1978), are at high risk for alcohol problems. These problems are more often related to acute consumption rather than alcoholism or alcohol dependence. Alcoholism or alcohol dependence measures, such as the AUDIT (Babor & Grant, 1989), CAGE (Mayfield, McLeod, & Hall, 1974), and MAST (Selzer, 1971) do a good job of assessing individuals with chronic alcohol problems, but do not detect the heavy, episodic drinking patterns of many college students. Problems such as blackouts, hangovers, drunk driving, and unprotected sex are prevalent for college drinkers. Traditional measures have done a poor job assessing these problems Harrell et al., 1986, Mayer & Filstead, 1980, O'Gorman et al., 1976, White, 1989. Construction of a sound psychometric problem index is essential for intervention research. Recently, interventions to reduce college student drinking have become a national priority (US Department of Health and Human Services, 1997). Negative consequences are a central outcome measure of the effectiveness of this type of research. Highly reliable and valid measures are needed for the precise measurement of these outcomes.

In recent years, several instruments have been designed to measure negative related consequences among college students both using both ad hoc and psychometrically sound methods. Typically, these instruments have examined negative consequences as a global index, ignoring possible subfactors. Typical approaches to the development of an alcohol problem index involve asking an exhaustive list of possible problems that college students face without assessing the underlying factor structure. This decreases the sensitivity of the instrument and treats all problems equally. An extensive measurement study of alcohol problems resulted in the development of the Rutgers Alcohol Problem Index (White & Labouvie, 1989), which greatly improved the traditional ad hoc method of assessing alcohol-related problems. However, the researchers ignored the three-factor structure that resulted from a factor analysis on their original problems scale and opted for a single-factor instrument. The results of their measurement development lead to a fairly long, reliable and valid single-factor measure of alcohol problems. However, this scale does not provide a short psychometrically valid way to assess problems of drinking or examine possible subfactors that may exist in the data. Another measure that employed excellent measurement techniques was the Young Adult Alcohol Problems Screening Test (YAAPST; Hurlbut & Sher, 1992). While, this measure is well developed, it consists of only one factor and is fairly long (23 items) with many of the items being very infrequently endorsed reducing the variance to assess change in intervention studies.

One measure of alcohol problems developed for college students was constructed with careful attention to the underlying factor structure (O'Hare, 1997). The College Alcohol Problems Scale (CAPS) contains two subscales measuring community and socioemotional problems. This scale was constructed using principal components analysis (PCA) on a group of college students who had been cited for alcohol violations.

The goal of this study was to further refine the CAPS for use on the general population of college drinkers. This was accomplished in two consecutive studies. In Study I, the original version of the 10-item CAPS was examined using confirmatory factor analysis (CFA) and found to fit the data poorly. A revised version of the CAPS (CAPS-r) was then constructed using the 10 items from the original CAPS with 10 additional items used in the construction of the original scale. Sequential methods were used to find a two-factor eight-item solution. In Study II, a large, random sample of the entire university was collected. The response format was altered to match the YAAPST and analyses indicated that the scale fit well across a representative sample and was invariant across gender. Examination of external validity indicated that the scale was highly related to alcohol use and attitudes among college students.

Section snippets

Participants

Participants were 382 introductory psychology students who completed a questionnaire on alcohol behavior and attitudes in the fall of 1996 or the spring of 1997 for class credit. This study was conducted to develop measures and provide pilot data for a randomized trial of a standardized, individualized intervention for college drinkers based on the Transtheoretical Model. Informed consent was obtained from all participants before they completed the survey. Of this sample, 348 (91%) reported

Participants

A representative sample of undergraduates answered an anonymous survey of alcohol use and attitudes as part of the Campus Climate Checkup in the spring of 1999. The survey was conducted to assess drinking rates on campus, which would be used to direct policy and services. It was completed by 726 respondents using the Harvard Survey Method (DeJong & Wechsler, 1998). Informed consent was obtained from all participants before they completed the survey. Over 90% (91.3%; n=663) of the sample

Discussion

The objective of this study was to replicate and refine a short, psychometrically sound instrument to measure collegiate problem drinking. The development of the eight-item CAPS-r fulfills these goals. The CAPS-r is shown to have a stable factor structure, which is evident in both the exploratory and confirmatory samples. The personal and social problems subscales are only moderately correlated indicating a need to differentiate between the two subscales. While most studies show that males

References (41)

  • J. Darkes et al.

    Expectancy challenge and drinking reduction: process and structure in the alcohol expectancy network

    Experimental and Clinical Psychopharmacology

    (1998)
  • W. DeJong et al.

    Preventing alcohol related problems on campus: methods for assessing student use of alcohol and other drugs

    (1998)
  • R.C. Engs

    Drinking patterns and drinking problems of college students

    Journal of Studies on Alcohol

    (1977)
  • M. Haines et al.

    Changing the perception of the norm: a strategy to decrease binge drinking among college students

    Journal of American College Health

    (1996)
  • T.C. Harford et al.

    Age-related trends in alcohol consumption

    Journal of the Studies on Alcohol

    (1978)
  • Harrell, A., Kapsak, K., & Cook, R. (1986). Screening for adolescent drinking problems: Phase I development. Final...
  • J.I. Horn

    A rationale and test for the number of factors in factor analysis

    Psychometrika

    (1965)
  • S.C. Hurlbut et al.

    Assessing alcohol problems in college students

    Journal of American College Health

    (1992)
  • K.G. Joreskog et al.

    LISREL 7: a guide to the program and applications

    (1989)
  • G.J. Lautenschlager

    A comparison of alternatives to conducted Monte Carlo analyses for determining parallel analysis criteria

    Multivariate Behavioral Research

    (1989)
  • Cited by (75)

    • Sleep in college students and young adults

      2023, Encyclopedia of Sleep and Circadian Rhythms: Volume 1-6, Second Edition
    • Is Pornography Consumption Related to Risky Behaviors During Friends With Benefits Relationships?

      2020, Journal of Sexual Medicine
      Citation Excerpt :

      To examine whether these results are robust to plausible alternate explanations, we included a measure of trait self-control42 (α = 0.84 in study 1, α = 0.84 in study 2), a single item assessing binge drinking (“How often in the last 30 days did you have 5 or more drinks on one occasion?”),43 a measure of broader problematic patterns of alcohol use among students in college44 (α = 0.84 in study 1, α = 0.85 in study 2), and a measure of attitudes toward casual sex from the sociosexuality inventory45 (α = 0.86 in study 1, α = 0.88 in study 2). We also included a 2-item measure of openness to experience from a brief personality measure that is particularly used when scale brevity is important for the study and trait is used as a control variable46 (α = 0.40 in study 1, α = 0.38 in study 2).

    • Measuring alcohol-related consequences and motives among students attending historically black colleges and universities

      2015, Addictive Behaviors
      Citation Excerpt :

      Each of the eight items is answered on a 6-point Likert-type scale ranging from 0 (never/almost never) to 5 (10 or more times) and the total scores range from 0 to 40. The CAPS-r has displayed adequate psychometric properties in previous studies using college student samples (Maddock et al., 2001; West & Graham, 2001). Young Adult Alcohol Consequences Questionnaire (YAACQ; Read et al., 2006): The YAACQ is a 48 item instrument specifically designed to assess a broad range of alcohol-related consequences experienced by college students (e.g., I often have found it difficult to limit how much I drink, I have passed out from drinking).

    View all citing articles on Scopus
    View full text