College Alcohol Problems Scale – Revised (CAPS-r)

1. Abstract

The College Alcohol Problems Scale – Revised (CAPS-r) is a psychometrically validated, brief self-report assessment instrument designed to measure negative consequences associated with heavy episodic and chronic alcohol consumption among collegiate populations. Developed by Jay E. Maddock, Ralph G. Laforge, Joseph S. Rossi, and Thomas O’Hare in 2001, the CAPS-r refines earlier multidimensional assessments of collegiate alcohol misuse into an efficient, 8-item inventory. The instrument assesses drinking-induced problems occurring over a designated reference timeframe (typically the preceding academic year or past 12 months) across two distinct, correlated empirical dimensions: Personal Problems (measuring intrapersonal, emotional, and neurovegetative consequences, such as depression, irritability, shame, and sleep/appetite disturbances) and Social Problems (measuring interpersonal, behavioral, and public health risk consequences, including driving under the influence, unplanned or unprotected sexual activity, and illicit drug involvement).

Items are rated on an anchored 6-point ordinal frequency scale ranging from “Never” to “10 or more times.” Across multiple large-scale psychometric evaluations, the CAPS-r exhibits strong factorial validity, with confirmatory factor analysis (CFA) consistently demonstrating superior goodness-of-fit for the hypothesized two-factor oblique structure compared to unidimensional alternatives. The scale demonstrates rigorous multigroup measurement invariance across biological sex and college class cohorts. Reliability coefficients (Cronbach’s $\alpha$ and McDonald’s $\omega$) regularly exceed .80 for both subscales, maintaining exceptional measurement precision despite its brevity. The CAPS-r serves as an indispensable tool within high-volume campus health screenings, epidemiological surveillance, brief motivational interventions such as the Brief Alcohol Screening and Intervention for College Students (BASICS), and longitudinal research on collegiate substance use dynamics.

2. Keywords

College Alcohol Problems Scale – Revised, CAPS-r, college student drinking, alcohol-related negative consequences, psychometrics, confirmatory factor analysis, Personal Problems, Social Problems, brief screening tool, harm reduction, alcohol misuse, collegiate health surveillance

3. Authors

The College Alcohol Problems Scale – Revised was developed through a collaboration of behavioral scientists and psychometricians specializing in health psychology, substance misuse epidemiology, and health behavior change models:

• Jay E. Maddock, Ph.D., FAAHB: Primary investigator and developer. Professor of Public Health and Director of the Center for Health & Nature at Texas A&M University (formerly of the Department of Public Health Sciences at the University of Hawaii at Manoa). Dr. Maddock is a recognized scholar in social-ecological approaches to health behavior and measurement theory.
• Ralph G. Laforge, Sc.D.: Quantitative psychologist and biostatistician affiliated with the Cancer Prevention Research Center (CPRC) at the University of Rhode Island, known for methodological contributions to the Transtheoretical Model of Behavior Change.
• Joseph S. Rossi, Ph.D.: Professor of Psychology and Director of the Cancer Prevention Research Center at the University of Rhode Island. An authority in structural equation modeling, psychometric scale construction, and longitudinal behavioral tracking.
• Thomas O’Hare, Ph.D., LCSW: Professor Emeritus at the Boston College School of Social Work. Dr. O’Hare authored the original College Alcohol Problems Scale (CAPS) and conducted foundational clinical investigations into alcohol distress profiles among emerging adults.

4. Purpose

Heavy episodic drinking—commonly designated as binge drinking (consuming 4 or more drinks for women or 5 or more drinks for men in approximately two hours)—remains one of the most persistent, deleterious public health challenges confronting higher education institutions internationally. Although numerous measures exist to quantify consumption patterns (e.g., volume, peak consumption, blood alcohol concentration estimates), absolute volume alone frequently fails to capture the full spectrum of psychological, interpersonal, and behavioral harm incurred by students. Historically, broad diagnostic instruments such as the Alcohol Use Disorders Identification Test (AUDIT) or lengthy research measures such as the 23-item Rutgers Alcohol Problem Index (RAPI) and the 48-item Young Adult Alcohol Consequences Questionnaire (YAACQ) imposed substantial participant burden, limiting their feasibility in large-scale campus surveys, rapid clinical triage, or brief routine healthcare encounters.

The primary purpose of the College Alcohol Problems Scale – Revised is to provide a brief, psychometrically sound, low-cost instrument capable of assessing drinking-related consequences among college students without sacrificing measurement fidelity. The CAPS-r was engineered specifically to facilitate three interconnected public health and clinical applications:

• Epidemiological Surveillance and Campus Needs Assessment: Higher education administrators, campus wellness task forces, and public health researchers require economical, rapid screening instruments that can be embedded into comprehensive campus-wide health audits (such as the American College Health Association National College Health Assessment) to track shifts in alcohol-related harm over academic terms and institutional policy shifts.
• Clinical Screening and Triage in University Health Services: Student health centers, counseling departments, and athletic clinics need a rapid diagnostic triage mechanism to identify individuals experiencing acute internal distress or engaging in dangerous external behaviors secondary to alcohol consumption, thereby guiding appropriate stepped-care referrals.
• Program Evaluation and Brief Motivational Interventions: Within secondary prevention frameworks like BASICS or motivational interviewing-based feedback protocols, the CAPS-r provides concrete, highly personalized feedback profiles. By distinguishing intrapersonal emotional distress from outward interpersonal risk behaviors, practitioners can challenge young adult risk perceptions, elicit cognitive dissonance, and enhance readiness to change.

Consequently, the theoretical rationale for the CAPS-r centers on balancing psychometric parsimony with clinical differentiation, ensuring that interventionists do not conflate the internalizing psychological toll of drinking with hazardous externalizing actions.

5. Psychological Construct

The College Alcohol Problems Scale – Revised operationalizes alcohol-related consequences as a multidimensional, correlated psychological construct comprised of two primary latent factors: Personal Problems (intrapersonal/internalizing consequences) and Social Problems (interpersonal/externalizing/risk-taking consequences). This bifactor or two-dimensional framework reflects the empirical observation that young adults experience alcohol complications through distinctly different neurobiological, cognitive, and social pathways.

Personal Problems Subscale

The Personal Problems dimension captures psychological, affective, and neurovegetative distress resulting from alcohol consumption. Emerging adulthood is marked by ongoing prefrontal cortical maturation, psychological individuation, and heightened vulnerability to stress, affective instability, and self-evaluative anxiety. Heavy alcohol consumption exerts depressogenic and anxiogenic neurochemical effects, frequently exacerbating underlying dysphoria. The four items comprising this subscale measure:

• Dysphoric Mood States: Manifested in subjective feelings of sadness, feeling “blue,” or experiencing depressive affect following drinking episodes. Rather than capturing clinical major depressive disorder per se, this item reflects alcohol-induced mood destabilization and post-intoxication dysphoria.
• Autonomic and Affective Hyperarousal: Reflected in feelings of nervous tension, motor agitation, and heightened irritability during post-drinking or withdrawal periods, indicating nervous system rebound effects and cognitive-emotional strain.
• Self-Directed Cognitive Devaluation: Characterized by feelings of shame, regret, moral guilt, or diminished self-esteem (“caused you to feel bad about yourself”) following actions executed while intoxicated or general dissatisfaction with one’s inability to control alcohol consumption.
• Somatic/Biological Disruption: Encompassing disturbances in physiological homeostasis, specifically dysregulated sleep architecture (e.g., alcohol-induced fragmentation of REM sleep, early awakenings) and appetite fluctuations (e.g., post-drinking nausea, gastrointestinal distress, or suppressed nutritional intake).

Social Problems Subscale

The Social Problems dimension measures externalized, high-risk, behavioral, and societal consequences that jeopardize the physical safety, legal standing, and social well-being of the individual and broader community. Alcohol diminishes executive functioning, impulse control, and risk appraisal through alcohol myopia—a psychological state where attention is narrowed strictly to immediate, salient environmental cues while peripheral risks are disregarded. The four items in this subscale measure:

• Unplanned Sexual Encounters: Alcohol-induced disinhibition frequently results in sexual behavior that deviates from the individual’s sober intentions, values, or relationship agreements, often provoking post-hoc relational conflict and psychological distress.
• Impaired Vehicular Operation: Driving under the influence of alcohol represents a severe, life-threatening externalized behavior reflecting impaired risk judgment, motor coordination, and institutional non-compliance.
• High-Risk Sexual Practices: Failure to utilize protection (e.g., barrier methods such as condoms) during sexual intercourse, directly elevating epidemiological risk for sexually transmitted infections (STIs) and unintended pregnancy.
• Illicit Drug Co-Ingestion and Legal Encroachment: Engaging in illegal activities associated with illicit substance use (such as polysubstance co-ingestion, illicit drug purchasing, or drug-related campus policy violations), illustrating cross-substance risk escalation.

6. Theoretical Framework

The conceptual foundation of the CAPS-r is grounded in several convergent behavioral and psychological theories, prominently including the Transtheoretical Model of Behavior Change (TTM), Social Cognitive Theory, Problem Behavior Theory, and the Harm Reduction Paradigm.

Transtheoretical Model and Decisional Balance

Pioneered by James O. Prochaska and Wayne F. Velicer, the Transtheoretical Model posits that behavior change occurs across discrete stages of change (Precontemplation, Contemplation, Preparation, Action, Maintenance). Central to progression through these stages is the construct of Decisional Balance—the cognitive weighing of the perceived “pros” (benefits) versus “cons” (negative consequences) of a behavior. Because young adult drinkers frequently overemphasize the pros of alcohol (e.g., social facilitation, tension reduction, peer affiliation), robust assessment of the “cons” is vital. The CAPS-r explicitly articulates these negative consequences. By measuring concrete personal and social problems, the instrument brings the latent costs of drinking to the forefront of conscious appraisal, facilitating the decisional shift required to advance individuals from Precontemplation into Contemplation.

Problem Behavior Theory

Richard Jessor’s Problem Behavior Theory (PBT) provides an indispensable psychosocial framework for interpreting the Social Problems factor of the CAPS-r. PBT suggests that adolescent and young adult risk behaviors do not emerge in isolation; rather, they form a cohesive, purposive constellation or behavioral syndrome driven by interactions among the Personality System (e.g., tolerance of deviance, alienation), the Perceived Environment System (e.g., peer models for risk-taking), and the Behavior System. Within this paradigm, driving under the influence, unprotected sex, and polysubstance activities represent co-occurring manifestations of an overarching unconventionality and sensation-seeking orientation amplified by collegiate social environments.

Alcohol Myopia Theory

Claude Steele and Robert Josephs formulated Alcohol Myopia Theory, which posits that ethanol pharmacologically restricts the perceptual and cognitive visual field, rendering intoxicated individuals disproportionately responsive to immediate situational prompts (e.g., sexual arousal, social camaraderie) while remaining cognitively blind to distal, inhibitory consequences (e.g., pregnancy, STIs, legal sanctions, accidents). The items comprising the Social Problems dimension reflect this acute cognitive constriction, capturing the severe lapses in prospective judgment that characterize intoxicated decision-making.

Harm Reduction Paradigm

Originated in addictive behaviors research by G. Alan Marlatt, the harm reduction model asserts that intervention goals need not be restricted solely to strict lifelong abstinence. Instead, reducing the severity, frequency, and lethality of alcohol-induced collateral harm constitutes a clinically vital objective. The CAPS-r embodies this philosophy: it was specifically engineered not to measure physical dependence or consumption volume, but rather to evaluate whether harm reduction strategies (e.g., utilizing designated drivers, setting drink limits, pacing) successfully diminish the tangible personal and social consequences suffered by students.

7. Validity

The College Alcohol Problems Scale – Revised has undergone extensive psychometric validation, establishing robust evidence for construct, convergent, discriminant, and predictive validity across diverse collegiate samples.

Construct and Factorial Validity

During the primary psychometric development study by Maddock, Laforge, Rossi, and O’Hare (2001), structural equation modeling demonstrated that the hypothesized two-factor oblique structure provided a statistically and conceptually superior fit to empirical data compared to alternative single-factor models. Across large cross-validation samples of undergraduate students, goodness-of-fit indices demonstrated exemplary adherence to established psychometric standards: the Comparative Fit Index (CFI) and Goodness of Fit Index (GFI) regularly exceeded .95, while the Root Mean Square Error of Approximation (RMSEA) remained at or below .05, establishing clear structural differentiation between intrapersonal emotional distress and interpersonal social deviance.

Convergent Validity

Convergent validity has been repeatedly established through robust bivariate and multivariate correlations with established markers of heavy alcohol use and established psychometric inventories:

• Consumption Metrics: Both the Personal Problems and Social Problems subscales demonstrate statistically significant, moderate-to-high correlations with typical weekly drinking volume ($r = .35$ to $.52, p < .001$), frequency of binge drinking episodes ($r = .40$ to $.58, p < .001$), and estimated peak blood alcohol concentration (BAC).
• Established Scales: The CAPS-r subscales exhibit strong convergent correlations with the Rutgers Alcohol Problem Index (RAPI; $r > .65$) and the Brief Young Adult Alcohol Consequences Questionnaire (BYAACQ; $r > .70$). Furthermore, scores correlate positively with the CAGE questionnaire and the AUDIT-C screening inventory.

Discriminant and Known-Groups Validity

Discriminant validity analyses confirm that the CAPS-r does not simply mirror generalized negative affectivity, neuroticism, or global psychological distress. While the Personal Problems subscale correlates moderately with depressive symptom indices (such as the Beck Depression Inventory or CES-D), these correlations attenuate dramatically when partialling out alcohol consumption frequency, confirming that the scale detects consequences specifically attributed to alcohol consumption. Known-groups validation studies confirm that the instrument reliably differentiates between mandated students (those referred for disciplinary alcohol violations) and voluntary or normative campus peers, with mandated students demonstrating significantly elevated scores on both subscales, particularly regarding Social Problems ($d = 0.62$ to $0.85$).

8. Reliability

The CAPS-r exhibits high internal consistency and temporal stability across academic environments, demonstrating that its brevity does not compromise statistical reliability.

Internal Consistency

In the seminal psychometric investigation by Maddock et al. (2001) involving multiple independent cohorts ($N > 1,200$), internal consistency was rigorously evaluated for both subscales:

• Personal Problems Subscale: Cronbach’s alpha ($\alpha$) values typically range between .78 and .85 across diverse collegiate cohorts, with McDonald’s omega ($\omega$) coefficients consistently tracking above .80, indicating high item homogeneity and minimal measurement error within the internalizing domain.
• Social Problems Subscale: Cronbach’s alpha values routinely fall between .74 and .81. Considering that externalized risk behaviors (such as DUI and unprotected sex) represent heterogeneous behavioral expressions that do not always co-occur uniformly, an alpha coefficient exceeding .75 reflects remarkable internal cohesion.

Corrected item-total correlations for all 8 individual items remain substantially above the standard psychometric cutoff of .40, with typical values ranging between .48 and .71, indicating that each item contributes uniquely and substantially to its respective latent factor without excessive redundancy.

Test-Retest Reliability and Temporal Stability

Investigations analyzing the temporal stability of the CAPS-r across short-term, non-intervention test-retest intervals (spanning 2 to 4 weeks) report intraclass correlation coefficients (ICCs) and Pearson correlation coefficients ranging from $r = .79$ to $r = .87$. Over extended intervals (e.g., 6 months to 1 academic year), test-retest correlations demonstrate moderate stability ($r = .50$ to $.65$), which appropriately reflects natural developmental transitions, changing living situations, intervention exposure, and maturational shifts common to emerging adults.

9. Factor Analysis

The factorial architecture of the CAPS-r represents a cornerstone of its psychometric utility, established through successive stages of exploratory factor analysis (EFA) and validated via confirmatory factor analysis (CFA).

Exploratory Factor Structure

During the construction of the original College Alcohol Problems Scale by O’Hare and its subsequent refinement into the CAPS-r, principal component and common factor analyses with oblique (Promax and Oblimin) rotations consistently yielded a clean, two-factor solution based on Kaiser’s criterion (eigenvalues $> 1.0$) and Cattell’s scree test. Factor 1 consistently accounted for approximately 35% to 42% of the total variance (Personal Problems), while Factor 2 accounted for an additional 12% to 18% of the variance (Social Problems), jointly accounting for over 50% of total variance in collegiate drinking consequences.

Confirmatory Factor Analysis and Model Fit

Maddock et al. (2001) conducted rigorous CFA using structural equation modeling software (e.g., AMOS/LISREL) to examine competing structural models:

• Unidimensional Model: A single global “alcohol problems” factor demonstrated inadequate model fit ($\chi^2/df > 8.5$, $text{CFI} < .88$,$text{RMSEA} > .10$), rejecting the hypothesis that personal and social alcohol consequences represent an undifferentiated construct.
• Two-Factor Oblique Model: The hypothesized model specifying two correlated latent factors demonstrated exceptional fit to the data: $chi^2(19) = 42.18, p < .002$;$chi^2/df = 2.22$;$text{CFI} = .98$;$text{GFI} = .98$;$text{TLI} = .97$;$text{RMSEA} = .044$ ($90% \text{ CI } [.027, .062]$).

Standardized factor loadings for the 8 items are robust and statistically significant ($p < .001$), generally ranging from .55 to .84 across samples:

• Feeling sad, blue, or depressed: $lambda = .76 – .82$
• Nervousness, irritability: $lambda = .72 – .79$
• Caused you to feel bad about yourself: $lambda = .78 – .84$
• Problems with appetite or sleeping: $lambda = .55 – .68$
• Engaged in unplanned sexual activity: $lambda = .65 – .74$
• Drove under the influence: $lambda = .58 – .69$
• Did not use protection when engaging in sex: $lambda = .68 – .77$
• Illegal activities associated with drug use: $lambda = .54 – .65$

Measurement Invariance Across Gender

A critical contribution of the Maddock et al. (2001) psychometric evaluation was testing for multigroup measurement invariance across biological sex. Establishing invariance is essential to ensure that observed differences between male and female college students reflect true differences in consequence severity rather than differential item functioning (DIF) or measurement bias. Stepwise invariance testing confirmed:

• Configural Invariance: The two-factor structure held identically across both male and female student cohorts without cross-loadings.
• Metric (Weak) Invariance: Constraining factor loadings to be equal across sexes did not significantly degrade model fit ($\Delta\chi^2$ non-significant; $Deltatext{CFI} < .01$).
• Scalar (Strong) Invariance: Constraining item intercepts across groups supported scalar invariance, confirming that latent mean comparisons across male and female students are statistically valid and unbiased.

The correlation between the Personal Problems and Social Problems latent factors is typically moderate ($r \approx .48$ to $.60$), confirming that while the two dimensions share significant variance attributable to systemic alcohol misuse, they remain psychometrically distinct constructs that warrant separate subscale scoring.

10. Instrument / Measurement Tool

The College Alcohol Problems Scale – Revised is a self-administered, paper-and-pencil or computerized questionnaire. Below is the operational measurement profile:

• Instrument Name: College Alcohol Problems Scale – Revised (CAPS-r)
• Test Type: Standardized self-report symptom and behavioral consequence inventory
• Target Population: Undergraduate and graduate college students, emerging adults (ages 18–25)
• Administration Duration: Approximately 2 to 3 minutes
• Total Item Count: 8 items
• Subscale Structure:
  • Personal Problems Subscale: Items 1, 2, 3, 4 (measures internal/psychological consequences)
  • Social Problems Subscale: Items 5, 6, 7, 8 (measures external/interpersonal/risk consequences)
• Reference Period: Standard administration specifies “over the past year” (can be modified to “past 30 days” or “past semester” for specific intervention outcome research)
• Response Scale (6-point ordinal frequency format):
  • Never (Scored as 0)
  • Yes, but not in the past year (Scored as 1)
  • 1-2 times (Scored as 2)
  • 3-5 times (Scored as 3)
  • 6-9 times (Scored as 4)
  • 10 or more times (Scored as 5)
• Scoring Protocol:
  • Subscale Scores: Computed by summing the numerical ratings of the four items within each dimension. Personal Problems scores range from 0 to 20; Social Problems scores range from 0 to 20.
  • Total Composite Score: Computed by summing all 8 items (range: 0 to 40), providing an index of overall alcohol-related consequence burden.
  • Alternative Dichotomous Scoring: For epidemiological prevalence surveys, items can be dichotomized (0 = consequence absent in the past year; 1 = consequence experienced 1 or more times in the past year) to calculate consequence counts (range: 0 to 8).

11. Permissions & Fee and Test Year

The College Alcohol Problems Scale – Revised was developed and validated in 2001 by Jay E. Maddock, Ralph G. Laforge, Joseph S. Rossi, and Thomas O’Hare, published in the peer-reviewed journal Addictive Behaviors.

• Licensing and Availability: The CAPS-r is situated in the academic public domain for non-commercial research, institutional evaluation, and clinical screening purposes. Researchers, healthcare practitioners, and university personnel may utilize the scale without paying licensing fees or royalties.
• Permissions Requirement: Formal written permission from the primary author is generally not required for independent academic, non-funded investigations, provided the original instrument is cited properly in resulting reports and publications. Commercial entities, proprietary digital health platforms, or fee-for-service enterprise systems seeking to integrate the instrument into commercial software should secure explicit permission from the authors or copyright holders.
• Primary Developer Contact: Inquiries regarding the scale, scoring templates, or normative benchmarks may be directed to Dr. Jay E. Maddock at Texas A&M University School of Public Health.

12. References

The following empirical and theoretical publications represent core literature concerning the development, validation, and programmatic application of the CAPS-r:

• Jessor, R., & Jessor, S. L. (1977). Problem behavior and psychosocial development: A longitudinal study of youth. Academic Press.
• Maddock, J. E., Laforge, R. G., Rossi, J. S., & O’Hare, T. (2001). The College Alcohol Problems Scale. Addictive Behaviors, 26(3), 385–398. https://doi.org/10.1016/S0306-4603(00)00115-4
• Marlatt, G. A., & Witkiewitz, K. (2002). Harm reduction approaches to alcohol use: Health promotion, prevention, and treatment. Addictive Behaviors, 27(6), 867–886. https://doi.org/10.1016/S0306-4603(02)00294-X
• O’Hare, T. (1997). Measuring alcohol consumption in young adults: A comparison of three methods. Journal of Alcohol and Drug Education, 42(3), 73–84.
• O’Hare, T. (1998). Measuring problem drinking in college students: The College Alcohol Problems Scale (CAPS). American Journal of Drug and Alcohol Abuse, 24(3), 427–440. https://doi.org/10.3109/00952999809016908
• Prochaska, J. O., & Velicer, W. F. (1997). The transtheoretical model of health behavior change. American Journal of Health Promotion, 12(1), 38–48. https://doi.org/10.4278/0890-1171-12.1.38
• Steele, C. M., & Josephs, R. A. (1990). Alcohol myopia: Its prized and dangerous effects. American Psychologist, 45(8), 921–933. https://doi.org/10.1037/0003-066X.45.8.921
• White, H. R., & Labouvie, E. W. (1989). Towards the assessment of adolescent problem drinking. Journal of Studies on Alcohol, 50(1), 30–37. https://doi.org/10.15288/jsa.1989.50.30

13. Items of the Scale (Questionnaire)