Drinking Refusal Self-Efficacy Questionnaire-Revised (DRSEQ-R)

Abstract

The Drinking Refusal Self-Efficacy Questionnaire-Revised (DRSEQ-R) is an established psychometric instrument designed to evaluate an individual’s belief in their ability to resist or refuse alcohol across diverse high-risk contexts and situational prompts. Developed by Tian P. S. Oei, Penelope A. Hasking, and Ross McD. Young in 2005, the DRSEQ-R is a psychometrically refined 19-item revision of the original 31-item Drinking Refusal Self-Efficacy Questionnaire (DRSEQ; Young, Oei, & Crook, 1991). Grounded in Albert Bandura‘s social cognitive theory and the cognitive-behavioral relapse prevention model formulated by G. Alan Marlatt, the DRSEQ-R conceptualizes drinking refusal self-efficacy as a multidimensional construct rather than a global trait. The instrument comprises three robust, correlated subscales: Social Pressure Refusal Self-Efficacy (5 items), Negative Emotional Relief Refusal Self-Efficacy (7 items), and Opportunistic Refusal Self-Efficacy (7 items). Respondents rate their subjective confidence on an authentic 6-point Likert-type continuum ranging from 1 (“I would drink”) to 6 (“I am sure I would NOT drink”), yielding subscale scores as well as an aggregate global index where elevated scores reflect superior refusal self-efficacy.

Extensive psychometric investigations among clinical cohorts diagnosed with alcohol use disorder (AUD) as well as non-clinical adolescent, university, and community samples indicate that the DRSEQ-R possesses exemplary internal consistency (Cronbach’s alpha and composite reliability values typically span .83 to .94 across subscales and surpass .92 for the total scale), robust test-retest reliability, and well-demonstrated factorial validity across diverse demographic and cultural groups. Moreover, the questionnaire exhibits strong convergent, discriminant, and predictive validity, functioning as a sensitive predictor of alcohol consumption volume, binge drinking episodes, treatment completion, and post-intervention relapse latency. This article provides a comprehensive academic analysis of the DRSEQ-R, delineating its historical development, theoretical architecture, psychometric properties, clinical administration, scoring protocols, and complete authentic scale items.

Keywords

Drinking Refusal Self-Efficacy, DRSEQ-R, Alcohol Use Disorder, Social Cognitive Theory, Cognitive Behavioral Therapy, Psychometrics, Relapse Prevention, Addiction Assessment, Social Pressure, Opportunistic Drinking

Authors

The Drinking Refusal Self-Efficacy Questionnaire-Revised was developed and psychometrically standardized by a team of clinical psychologists and behavioral researchers specializing in addictive behaviors:

• Tian P. S. Oei, Ph.D., Dip.Clin.Psych., FAPS: Emeritus Professor of Clinical Psychology at the School of Psychology, The University of Queensland, Brisbane, Australia. A prominent authority in cognitive behavioral therapy and substance abuse, Professor Oei has authored seminal theoretical models integrating cognitive expectancies and self-efficacy mechanisms in cross-cultural and clinical cohorts.
• Penelope A. Hasking, Ph.D.: Professor of Psychology, previously affiliated with the School of Psychology at The University of Queensland and currently a leading researcher in mental health, substance use, and non-suicidal self-injury within university and adolescent populations at Curtin University and Monash University, Australia.
• Ross McD. Young, Ph.D., MAPS: Professor and behavioral health scientist, previously Dean of the Faculty of Health at Queensland University of Technology (QUT), Brisbane, Australia. Professor Young’s research focuses extensively on the interaction between cognitive vulnerability factors, genetic predispositions, and clinical outcomes in addictive and psychiatric disorders.

Purpose

The primary purpose of the DRSEQ-R is the quantitative assessment of situational drinking refusal self-efficacy—defined as an individual’s perceived competence and cognitive mastery to decline alcohol consumption when confronted with environmental, interpersonal, emotional, or opportunistic drinking cues. While global self-efficacy inventories evaluate generalized optimism or coping agency across life domains, addiction research demonstrates that self-efficacy is fundamentally context- and behavior-specific. Generalized measures frequently fail to predict substance use initiation, escalation, or relapse; conversely, domain-specific instruments such as the DRSEQ-R directly target the discrete psychological decision-points that precede alcohol consumption.

In clinical settings, the DRSEQ-R serves vital diagnostic, prescriptive, and evaluative functions. Prior to initiating evidence-based psychotherapeutic protocols—such as Cognitive Behavioral Therapy (CBT), Motivational Interviewing (MI), or Twelve-Step Facilitation—clinicians administer the DRSEQ-R to establish a comprehensive baseline profile of a patient’s cognitive vulnerabilities. For instance, a patient may demonstrate relatively high confidence in resisting alcohol when alone or experiencing negative affect, but display critically impaired self-efficacy when facing overt social facilitation or peer encouragement in hospitality environments. Such differentiated assessment allows clinicians to tailor treatment plans, targeting specific cognitive distortions, behavioral role-play, and cue-exposure exercises to the specific domains in which the client is most vulnerable.

In longitudinal research and clinical trials, the DRSEQ-R functions as an essential mediator and outcome metric. Tracking changes in DRSEQ-R subscale scores throughout the therapeutic continuum enables researchers and clinicians to assess whether therapeutic gains are mediated by increases in cognitive refusal confidence. Furthermore, post-discharge DRSEQ-R scores serve as a reliable prognostic marker for relapse vulnerability: individuals who exit detoxification or residential rehabilitation programs with persistently suppressed Negative Emotional or Opportunistic refusal self-efficacy exhibit significantly accelerated latencies to first lapse and full relapse. In non-clinical and prevention paradigms, such as university health services or secondary schools, the DRSEQ-R identifies non-dependent individuals who possess hazardous drinking patterns or an elevated risk of developing alcohol dependence, thereby informing early psychoeducational and harm-reduction interventions.

Psychological Construct

The psychological construct evaluated by the DRSEQ-R is drinking refusal self-efficacy (DRSE), conceptualized as a situational, state-sensitive cognitive appraisal of personal agency. DRSE reflects an individual’s subjective confidence that they can intentionally abstain from drinking alcohol in specific circumstances that typically invite, facilitate, or provoke consumption. The DRSEQ-R captures three distinct yet interrelated psychological dimensions identified through structural equation modeling and factor analysis:

1. Social Pressure Refusal Self-Efficacy

This dimension reflects an individual’s cognitive mastery and perceived capacity to decline alcohol in contexts characterized by direct or implicit interpersonal and social influences. Social settings frequently present intense normative expectations, overt offers, modeling behaviors, and peer pressure to drink. Individuals with suppressed social pressure refusal self-efficacy experience elevated anxiety regarding interpersonal evaluation, fear social alienation, or possess an entrenched behavioral habit of consuming alcohol as an affiliative lubricant. In the DRSEQ-R, this subscale encompasses scenarios such as being out to dinner, attending a pub or club, receiving an explicit drink offer from someone, or observing friends, partners, or spouses consuming alcohol. Clinically, deficits in this domain underscore assertiveness deficits, social anxiety, or high susceptibility to social contagion.

2. Negative Emotional Relief Refusal Self-Efficacy

Negative emotional relief refusal self-efficacy taps an individual’s belief that they can successfully withhold alcohol consumption when encountering negative affective states, internal emotional distress, or physiological tension. Rooted in the tension reduction hypothesis and self-medication paradigms, alcohol is widely utilized by vulnerable individuals as a chemical coping mechanism to attenuate negative affect. This dimension evaluates internal triggers including anger, frustration, worry, feeling upset, depressive moods, nervousness, and sadness. Individuals scoring low on this subscale perceive alcohol as indispensable for emotional regulation and psychological distress tolerance; consequently, affective dysregulation becomes a direct catalyst for compulsive drinking and high-risk relapse episodes.

3. Opportunistic Refusal Self-Efficacy

Opportunistic refusal self-efficacy quantifies the individual’s confidence to resist drinking in environments or situations where alcohol is easily accessible, habitual, or paired with routine solitary or unmonitored activities. Unlike the explicit interpersonal dynamics of the social factor or the overt psychological distress of the negative emotional factor, opportunistic situations typically involve benign, low-arousal, or habitual contexts. These include situations such as being home alone, watching television, listening to music or reading, taking a lunch break, commuting home from work, or relaxing immediately after completing a sporting activity. Deficits in opportunistic refusal self-efficacy reveal conditioned automated stimulus-response patterns where alcohol consumption has become intertwined with relaxation rituals, boredom management, or conditioned environmental cues.

Theoretical Framework

The conceptual architecture of the DRSEQ-R rests at the intersection of two foundational psychological paradigms: Albert Bandura‘s Social Cognitive Theory (1977, 1986, 1997) and G. Alan Marlatt‘s Cognitive-Behavioral Model of Relapse (1985; Marlatt & Gordon, 1985; Witkiewitz & Marlatt, 2004).

Bandura’s Triadic Reciprocal Causation and Self-Efficacy Expectancies

In social cognitive theory, human functioning is governed by reciprocal determinism, a dynamic interplay among personal cognitive factors, environmental influences, and behavioral patterns. Bandura differentiated between two primary cognitive expectancies: outcome expectancies (the belief that a given behavior will lead to specific outcomes, such as “alcohol will alleviate my social anxiety”) and efficacy expectancies or self-efficacy (the personal conviction that one can successfully execute the behavior required to produce the desired outcome, such as “I am capable of refusing a drink even if my peers are urging me to join them”).

Subsequent psychometric modeling by Oei and colleagues demonstrated that while outcome expectancies (such as those measured by the Alcohol Expectancy Questionnaire) dictate the initial motivation or desire to drink, self-efficacy operates as the proximal, decisive cognitive gateway determining behavioral execution. Even when alcohol expectancies are elevated, robust drinking refusal self-efficacy can buffer against consumption. Conversely, when refusal self-efficacy collapses, positive outcome expectancies dominate, precipitating alcohol use. Crucially, self-efficacy is not a static personality trait; it is situation-dependent, fluctuating based on physiological state, affective arousal, environmental mastery experiences, and social modeling.

Marlatt’s Relapse Prevention Framework

Marlatt and Gordon’s relapse taxonomy posits that the primary obstacle in maintaining behavioral change or sobriety is confronting high-risk situations. High-risk contexts typically bifurcate into intrapersonal triggers (e.g., negative emotional states, physical pain, high-arousal positive states) and interpersonal triggers (e.g., interpersonal conflict, direct social pressure). When an individual encounters a high-risk scenario, their immediate cognitive appraisal determines the behavioral trajectory:

• If an adequate, confident coping response is generated—anchored in elevated refusal self-efficacy—the individual avoids alcohol, precipitating an increased sense of mastery and lowering the probability of future relapse.
• If an individual lacks the belief in their capacity to execute refusal behaviors (low refusal self-efficacy), combined with anticipatory positive expectancies regarding the pharmacological effects of alcohol, an initial lapse (the first drink) is highly probable.
• This initial lapse often triggers the Abstinence Violation Effect (AVE), characterized by cognitive dissonance, internal self-blame, and feelings of failure, which culminates in a catastrophic escalation from a lapse into an uncontrolled relapse.

The DRSEQ-R operationalizes this framework by directly quantifying an individual’s subjective resilience across the exact high-risk categories delineated by Marlatt, providing empirical measurement of the proximal cognitive mechanism underlying relapse vulnerability.

Validity

The DRSEQ-R has undergone extensive psychometric evaluation across diverse populations, establishing solid evidence for construct, convergent, discriminant, and predictive validity.

Construct and Factorial Validity

Construct validity was initially established by Oei, Hasking, and Young (2005) through large-scale structural modeling involving both non-clinical community/university samples ($N = 654$) and clinical populations undergoing treatment for alcohol dependence ($N = 373$). Confirmatory Factor Analysis (CFA) systematically demonstrated that the revised 19-item, 3-factor oblique model exhibited superior fit relative to alternative unidimensional or orthogonal configurations. The factor structure remained stable across independent clinical and non-clinical cohorts, confirming that the tri-dimensional conceptualization (Social Pressure, Negative Emotional Relief, Opportunistic) is an accurate structural representation of the construct across varied degrees of drinking severity.

Convergent and Discriminant Validity

Convergent validity has been repeatedly corroborated via statistically significant correlations with validated indices of alcohol involvement, drinking severity, and cognitive expectancies:

• Measures of Hazardous Consumption: DRSEQ-R scores demonstrate moderate to strong negative correlations with the Alcohol Use Disorders Identification Test (AUDIT) (Pearson $r$ values typically ranging from $-.45$ to $-.68$), showing that lower refusal self-efficacy tracks elevated hazardous and harmful drinking.
• Timeline Followback (TLFB): Longitudinal studies utilizing the TLFB method have established inverse relationships between baseline DRSEQ-R scores and total standard drinks consumed per week, frequency of heavy episodic drinking, and peak blood alcohol concentration.
• Alcohol Expectancies: DRSEQ-R subscales correlate inversely with positive alcohol expectancies measured by the Alcohol Expectancy Questionnaire (AEQ), specifically expectancies regarding tension reduction, social facilitation, and liquid courage ($r = -.35$ to $-.55$).
• Discriminant Validity: The DRSEQ-R successfully discriminates between clinical populations (individuals meeting DSM criteria for Alcohol Use Disorder) and non-clinical social drinkers. Clinical cohorts routinely register significantly lower refusal self-efficacy scores across all three subscales compared to non-clinical cohorts ($p < .001$). Furthermore, the instrument demonstrates discriminant validity against general self-efficacy measures (e.g., the Generalized Self-Efficacy Scale), confirming that the DRSEQ-R measures domain-specific alcohol refusal rather than non-specific generalized personal agency.

Predictive and Ecological Validity

The predictive utility of the DRSEQ-R is well-documented in clinical longitudinal trials. Studies tracking patients following residential detoxification or cognitive-behavioral outpatient treatment have shown that post-treatment DRSEQ-R scores significantly predict latency to lapse and full relapse over 3-, 6-, and 12-month follow-up periods. Negative Emotional Relief Refusal Self-Efficacy has repeatedly emerged as the single strongest unique predictor of relapse when controlling for baseline drinking severity, age, and depression indices. In university cohorts, the Opportunistic and Social Pressure subscales significantly predict frequency of weekend binge drinking and alcohol-related interpersonal problems.

Reliability

The DRSEQ-R consistently exhibits strong reliability across clinical and non-clinical samples.

Internal Consistency

In the seminal validation study by Oei, Hasking, and Young (2005), Cronbach’s alpha ($lpha$) coefficients confirmed strong internal consistency for the composite score and all three subscales in both clinical and community populations:

• Total DRSEQ-R Scale: Clinical sample $\alpha = .94$; Non-clinical sample $\alpha = .93$.
• Negative Emotional Relief Subscale (7 items): Clinical sample $\alpha = .93$; Non-clinical sample $\alpha = .91$.
• Social Pressure Subscale (5 items): Clinical sample $\alpha = .89$; Non-clinical sample $\alpha = .86$.
• Opportunistic Subscale (7 items): Clinical sample $\alpha = .88$; Non-clinical sample $\alpha = .83$.

Subsequent psychometric investigations evaluating cross-cultural translations (e.g., Chinese, Spanish, Japanese, and Turkish adaptations) and adolescent extensions (DRSEQ-A) have supported these parameters, with composite reliability (Raykov’s $rho$) and McDonald’s omega ($\omega$) routinely exceeding .85 across subscales.

Test-Retest Stability

Evaluation of temporal stability across non-clinical college and community cohorts unexposed to therapeutic intervention has demonstrated high test-retest reliability over 2-week and 4-week intervals. Pearson correlation coefficients ($r$) for test-retest reliability generally range between .78 and .87 for subscales and .85 to .91 for the aggregate scale, indicating temporal stability in the absence of targeted cognitive-behavioral intervention while retaining sensitivity to genuine therapeutic change.

Factor Analysis

The transition from the original 31-item DRSEQ (Young, Oei, & Crook, 1991) to the 19-item DRSEQ-R (Oei, Hasking, & Young, 2005) was guided by systematic factor-analytic modeling designed to address psychometric limitations, redundant item content, and unstable cross-loadings present in the earlier version.

Exploratory and Confirmatory Modeling

The original DRSEQ contained 31 items organized into three factors (Social Pressure, Emotional Relief, and Opportunistic). However, clinical applications revealed that several items exhibited weak primary loadings, complex secondary cross-loadings, or colloquial phrasing that reduced cross-cultural applicability. Oei, Hasking, and Young (2005) conducted exploratory factor analyses (EFA) with oblimin rotation on split halves, sequentially pruning items that loaded below .40 on their primary factor or demonstrated cross-loadings exceeding .30 on secondary dimensions. This reduced the inventory from 31 to 19 items.

Model Fit and Confirmatory Factor Analysis

The revised 19-item structure was formally evaluated via Confirmatory Factor Analysis using Maximum Likelihood estimation. The authors tested multiple competitive structural models, including:

1. A 1-Factor Unidimensional Model, positing that all items reflect a single, undifferentiated drinking refusal capacity.
2. A 3-Factor Orthogonal Model, presuming three independent, uncorrelated situational refusal dimensions.
3. A 3-Factor Oblique Model, positing three distinct but correlated situational factors.
4. A Higher-Order Hierarchical Model, featuring three first-order situational dimensions loading onto a single higher-order global refusal self-efficacy construct.

The 3-factor oblique model exhibited superior fit indices across both clinical and non-clinical cohorts, outperforming the unidimensional and orthogonal alternatives. Typical fit indices obtained in structural equation investigations of the 19-item model meet modern standards:

• Comparative Fit Index (CFI): $ge .95$ (clinical = .96, non-clinical = .95).
• Tucker-Lewis Index (TLI): $ge .94$.
• Root Mean Square Error of Approximation (RMSEA): $.048$ to $.056$ (with 90% confidence intervals bounded within $.040$ and $.065$).
• Standardized Root Mean Square Residual (SRMR): $le .045$.

Standardized factor loadings ($lambda$) across all 19 items are strong and uniform, ranging from $.64$ to $.89$ ($p < .001$). Inter-factor correlations ($r$) between the three dimensions are moderate to high (typically $.50$ to $.72$), indicating that while the three subscales share common variance associated with generic inhibitory control, they capture distinct situational coping domains.

Instrument / Measurement Tool

The Drinking Refusal Self-Efficacy Questionnaire-Revised (DRSEQ-R) is a brief, self-administered psychometric instrument designed for routine clinical screening, research, and longitudinal outcome evaluation.

Instrument Specifications

• Construct Measured: Situational self-efficacy to refuse or resist alcohol consumption.
• Target Population: Adolescents and adults (ages 14 and older); applicable across non-clinical, at-risk, and clinical substance-dependent populations.
• Format: 19 self-report items completed via paper-and-pencil, computer, or digital survey software.
• Administration Time: Approximately 5 to 10 minutes.
• Stem Question: “How sure are you that you would have an alcoholic drink when …”
• Authentic Response Scale: 6-point Likert-type scale:
  • 1 = I would drink
  • 2 = I would probably drink
  • 3 = I might drink
  • 4 = I might NOT drink
  • 5 = I would probably NOT drink
  • 6 = I am sure I would NOT drink

Subscale Allocation

• Social Pressure Refusal Self-Efficacy (5 items): Items 1, 2, 3, 4, 5.
• Negative Emotional Relief Refusal Self-Efficacy (7 items): Items 6, 7, 8, 9, 10, 11, 12.
• Opportunistic Refusal Self-Efficacy (7 items): Items 13, 14, 15, 16, 17, 18, 19.

Scoring and Interpretation Procedures

• Item Directionality: The DRSEQ-R utilizes a direct positive scoring gradient. A score of 1 reflects certainty of drinking (extremely low refusal self-efficacy), whereas a score of 6 reflects certainty of non-drinking (maximal refusal self-efficacy). No reverse-scoring is necessary.
• Subscale Sum Scores:
  • Social Pressure: Sum of items 1–5 (Range: 5 to 30).
  • Negative Emotional Relief: Sum of items 6–12 (Range: 7 to 42).
  • Opportunistic: Sum of items 13–19 (Range: 7 to 42).
• Mean Subscale Scores: Subscale sums can be divided by their respective item count to provide a comparable metric on the original 1–6 metric (Mean = Sum / Item Count).
• Total Score: Sum of all 19 items (Range: 19 to 114). Higher aggregate and subscale scores denote greater confidence in resisting alcohol across challenging situations.
• Clinical Interpretation Guidelines:
  • Mean Scores 1.00 – 2.99: Markedly impaired refusal self-efficacy. High vulnerability to alcohol consumption and relapse in the specified context. Direct cognitive-behavioral intervention and skills training indicated.
  • Mean Scores 3.00 – 4.49: Ambivalent or moderate refusal self-efficacy. Individual possesses unstable refusal beliefs, vulnerable to erosion under elevated stress, intoxication, or persistent peer pressure.
  • Mean Scores 4.50 – 6.00: Robust refusal self-efficacy. The individual possesses strong confidence in their capacity to maintain abstinence or deliberate moderation within the specified domain.

Permissions & Fee and Test Year

The revised instrument was formally published in 2005:

• Publication Year: 2005 (DRSEQ-R revision; original DRSEQ published in 1991).
• Copyright & Ownership: The DRSEQ-R was developed by Tian P. S. Oei, Penelope A. Hasking, and Ross McD. Young, and published in Drug and Alcohol Dependence (Elsevier).
• Permissions & Research Access: The scale is widely accessible for non-commercial research, academic, and clinical assessment purposes under fair-use scientific conventions, provided proper scholarly attribution is cited. Researchers and clinicians wishing to utilize the instrument in published empirical investigations do not typically require paid licenses for non-commercial academic use, but must reference the foundational validation paper (Oei, Hasking, & Young, 2005). Commercial organizations, proprietary digital health applications, and pharmaceutical clinical trial deployments should contact the corresponding author or copyright clearance holders to obtain formal commercial licensing permissions.

References

• Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84(2), 191–215. https://doi.org/10.1037/0033-295X.84.2.191
• Bandura, A. (1997). Self-efficacy: The exercise of control. W. H. Freeman and Company.
• Marlatt, G. A., & Gordon, J. R. (Eds.). (1985). Relapse prevention: Maintenance strategies in the treatment of addictive behaviors. Guilford Press.
• Oei, T. P. S., & Baldwin, A. R. (1994). Expectancy theory and alcohol consumption. Drug and Alcohol Dependence, 34(3), 249–253. https://doi.org/10.1016/0376-8716(94)90165-1
• Oei, T. P. S., Hasking, P. A., & Young, R. M. (2005). Drinking refusal self-efficacy questionnaire-revised (DRSEQ-R): A new factor structure with confirmatory factor analysis. Drug and Alcohol Dependence, 78(3), 297–307. https://doi.org/10.1016/j.drugalcdep.2004.11.010
• Oei, T. P. S., & Jardim, C. L. (2007). Alcohol expectancies, drinking refusal self-efficacy and drinking behavior in Asian and Australian students. Drug and Alcohol Dependence, 87(2-3), 281–287. https://doi.org/10.1016/j.drugalcdep.2006.09.005
• Scherer, M. (2010). Forgiveness and the Bottle: Promoting Self-forgiveness with Alcohol Misuse (Doctoral dissertation, Virginia Commonwealth University). VCU Theses and Dissertations, Paper 2109. http://scholarscompass.vcu.edu/cgi/viewcontent.cgi?article=3108&context=etd
• Witkiewitz, K., & Marlatt, G. A. (2004). Relapse prevention for alcohol and drug problems: That was Zen, this is Tao. American Psychologist, 59(4), 224–235. https://doi.org/10.1037/0003-066X.59.4.224
• Young, R. M., Oei, T. P. S., & Crook, G. M. (1991). Development of a drinking self-efficacy questionnaire. Journal of Psychopathology and Behavioral Assessment, 13(1), 1–15. https://doi.org/10.1007/BF00960734

Items of the Scale