Adolescent Alcohol Drinking Intention (AADI)

1. Abstract

The Adolescent Alcohol Drinking Intention (AADI) scale is a specialized psychometric instrument engineered to quantify behavioral intentions toward alcohol consumption among youth populations. Originally formulated and validated by Russell, Hamby, and Russell (2018) within an investigation examining counter-messaging, narrative persuasion, and adolescent ambivalence, the instrument captures future-oriented motivational commitment across a progressive continuum of alcohol involvement. Comprising three stem-completion items, the AADI assesses an adolescent’s subjective probability of engaging in general alcohol consumption, heavy episodic drinking (consuming multiple drinks in a single sitting), and acute intoxication over a defined 12-month prospective horizon. Responses are recorded on a standardized 7-point Likert-type subjective probability scale ranging from 1 (“Pas du tout probable / Not at all likely”) to 7 (“Très probable / Very likely”). Psychometric evaluations confirm a robust unidimensional factor structure accounting for high proportions of common variance, with internal consistency coefficients consistently exceeding α = .88. The scale demonstrates robust predictive, convergent, and discriminant validity, correlating strongly with real-world substance initiation, peer descriptive norms, positive alcohol outcome expectancies, and narrative involvement. By operationalizing drinking intentions through an escalating behavioral severity gradient, the AADI provides developmental psychologists, public health researchers, and media communication scholars with an efficient, reliable, and ecologically sensitive assessment tool tailored to youth health behaviors.

2. Keywords

Adolescent Alcohol Drinking Intention, AADI, behavioral intentions, adolescent substance use, heavy episodic drinking, binge drinking, Theory of Planned Behavior, psychometrics, youth risk behavior, narrative persuasion, alcohol expectancy, public health measurement.

3. Authors

The Adolescent Alcohol Drinking Intention scale was developed and psychometrically operationalized by an interdisciplinary team of researchers specializing in consumer psychology, health communication, and quantitative behavioral methodology:

• Cristel Antonia Russell, Ph.D. — Professor of Marketing, Graziadio Business School, Pepperdine University, Malibu, California, United States. Dr. Russell is an internationally recognized expert in consumer behavior, narrative persuasion, product placement, and adolescent health communications, with extensive investigations into youth alcohol and substance exposure across mass media.
• Anne M. Hamby, Ph.D. — Associate Professor of Marketing, Department of Marketing, College of Business, Boise State University, Boise, Idaho, United States. Her research centers on narrative processing, consumer storytelling, causal comprehension, and public policy interventions targeting vulnerable populations.
• Daniel W. Russell, Ph.D. — Professor, Department of Human Development and Family Studies, Iowa State University, Ames, Iowa, United States. Dr. Russell is a distinguished quantitative methodologist and psychometrician renowned for developing gold-standard assessment tools, including the UCLA Loneliness Scale, and conducting advanced structural equation modeling on social and health behaviors.

4. Purpose

Underage alcohol consumption remains one of the most pervasive global public health challenges, associated with acute harms such as accidental trauma, neurocognitive impairment, interpersonal violence, academic disengagement, and an elevated trajectory toward lifelong alcohol use disorders. Psychometric assessment in this domain requires tools that circumvent both defensive underreporting and measurement insensitivity. The primary purpose of the Adolescent Alcohol Drinking Intention (AADI) scale is to assess the cognitive precursor to alcohol initiation and heavy episodic drinking among adolescents who may not yet engage in frequent substance use.

From a preventative standpoint, observing actual alcohol consumption retrospectively in early-to-mid adolescence often yields severe floor effects; a substantial proportion of young teenagers report zero consumption over past 30-day or lifetime metrics. However, cognitive susceptibility and prospective behavioral intentions fluctuate long before overt behavioral initiation occurs. The AADI addresses this psychometric challenge by quantifying the psychological propensity and subjective probability of future behavior over a one-year window.

In research contexts, the instrument was specifically developed to measure subtle shifts in youth attitudes following experimental exposure to alcohol-related messages, entertainment media, and corrective public health campaigns. Russell, Hamby, and Russell (2018) employed the scale to examine how warning labels, narrative transportation, and contradictory counter-messages influence youth ambivalence and subsequent behavioral resolve. In clinical, educational, and epidemiological contexts, the AADI serves as a rapid screening metric for identifying adolescents on an escalating developmental trajectory toward hazardous substance use, evaluating school-based prevention curricula, and determining the efficacy of social marketing interventions aimed at denormalizing binge drinking.

5. Psychological Construct

The core psychological construct measured by the AADI is behavioral intention within the domain of substance consumption. In psychometrics and social-cognitive psychology, behavioral intention represents a conscious decision or calculated commitment to perform a specified target behavior under designated prospective conditions. It encapsulates motivational factors, cognitive planning, and the net resolution between competing desires and perceived normative or legal barriers.

Crucially, the AADI conceptualizes alcohol consumption not as a binary behavioral state (abstinence versus use), but as an escalating behavioral continuum characterized by distinct physiological, social, and psychological risk thresholds:

• Threshold 1: General Alcohol Experimentation (Item 1). This dimension assesses the foundational threshold of consuming any quantity of alcohol. In early adolescence, cross-over into experimentation often occurs under parental oversight or exploratory peer gatherings. It reflects the breakdown of absolute normative abstinence and indexes initial behavioral readiness.
• Threshold 2: Heavy Episodic Consumption / Clustering (Item 2). Defined as consuming multiple drinks in a single sitting (often operationalized clinically as binge drinking or heavy episodic drinking), this threshold measures intentional exposure to pharmacologically active doses. Here, the adolescent indicates a willingness to move beyond mere tasting or symbolic consumption toward prolonged recreational drinking sessions, which demands greater intentional planning and access to alcohol supplies.
• Threshold 3: Acute Intoxication / Drunkenness (Item 3). The terminal anchor of the scale assesses intention to drink to the point of subjective drunkenness. This represents a high-risk cognitive commitment wherein the adolescent consciously anticipates loss of motor coordination, altered affective states, and cognitive disinhibition. Endorsement of this item captures severe behavioral risk, reflecting deep alignment with extreme alcohol cultures and low perceived behavioral vulnerability.

By arranging these three items along an implicit Guttman-like severity gradient within a unified Likert format, the construct captures the depth and intensity of an adolescent’s alcohol-related self-schema. It captures not only whether an adolescent intends to drink, but the intentional extremity and pharmacological depth of their planned engagement.

6. Theoretical Framework

The AADI is grounded in foundational social-cognitive models of health behavior, most notably the Theory of Planned Behavior (TPB) developed by Icek Ajzen and the earlier Theory of Reasoned Action by Fishbein and Ajzen. According to the TPB, behavioral intention is the immediate proximal determinant of volitional behavior. Intentions summarize an individual’s motivation, representing how hard they are willing to try and how much effort they plan to exert to perform the target action. Within this framework, intentions are determined by three core components: attitudes toward the behavior (evaluations of the consequences of drinking), subjective norms (perceptions of whether significant others approve of alcohol use), and perceived behavioral control (perceived ease or difficulty of procuring and consuming alcohol).

In addition to classical reasoned action paradigms, the scale draws conceptually from the Prototype Willingness Model (PWM) formulated by Gibbons and Gerrard for adolescent risk behaviors. The PWM posits that adolescent substance use occurs through two distinct pathways: a reasoned, deliberative path mediated by behavioral intentions, and a spontaneous, reactive path mediated by behavioral willingness in risk-conducive social environments. The AADI primarily operationalizes the deliberative pathway by prompting adolescents to estimate the probability of their future conduct over a distal temporal horizon (the coming year). This cognitively engages their self-regulatory assessment, forward-looking social expectations, and personal values.

Furthermore, the AADI’s application by Russell, Hamby, and Russell (2018) interfaces with Narrative Transportation Theory and Attitudinal Ambivalence Theory. When adolescents consume narratives (such as television series, digital influencer content, or film) that depict alcohol consumption, they often experience transportive engagement, leading to positive alcohol expectancies. When public health practitioners introduce counter-messages or corrective warnings, adolescents experience cognitive dissonance and ambivalence. The AADI was engineered to serve as the critical dependent criterion reflecting the ultimate balance or resolution of these conflicting narrative and counter-narrative inputs.

7. Validity

The psychometric validity of the AADI has been substantiated through rigorous construct, convergent, discriminant, and predictive validation protocols across adolescent sample cohorts.

Construct and Structural Validity

Construct validity is evidenced by confirmatory factor analytic studies demonstrating that the three items converge onto a coherent, unidimensional latent drinking intention factor. Standardized factor loadings across items consistently range from .82 to .94, confirming that general use, multiple drink consumption, and intoxication operate harmoniously as manifestations of a single underlying behavioral liability.

Convergent Validity

The scale demonstrates robust convergent validity with theoretical antecedents identified in health behavior models:

• Peer Normative Pressure: AADI scores correlate positively and significantly with perceived peer drinking norms and perceived peer approval of alcohol use (r values typically ranging from .42 to .58, p < .001).
• Positive Alcohol Expectancies: Significant positive associations emerge between AADI scores and adolescent beliefs that alcohol enhances social facilitation, sociability, and tension reduction (r = .38 to .52).
• Sensation Seeking and Impulsivity: Moderately high positive correlations have been recorded between AADI scores and validated sensation-seeking scales (e.g., the Brief Sensation Seeking Scale, r ≈ .35 to .46), consistent with theoretical expectations regarding developmental risk taking.

Discriminant Validity

Discriminant validity has been demonstrated by evaluating the AADI alongside measures of unrelated or divergent behavioral constructs. While moderately associated with intentions to smoke cigarettes or use cannabis (reflecting a broad underlying adolescent risk-taking propensity), the correlations between the AADI and non-substance delinquent behaviors (e.g., academic cheating, petty shoplifting) are substantially lower (r < .25), demonstrating construct distinctiveness. Moreover, the average variance extracted (AVE) of the latent AADI factor exceeds .70, comfortably surpassing squared inter-construct correlations with surrounding experimental covariates (such as transportation, narrative affect, and message reactance).

Predictive and Criterion Validity

Predictive validity is demonstrated by longitudinal studies showing that AADI scores obtained at baseline prospectively predict actual alcohol consumption, first incidence of heavy episodic drinking, and self-reported alcohol-related negative consequences (e.g., blackouts, disciplinary incidents) at 6-month and 12-month follow-ups, even after controlling for baseline demographics and lifetime experimentation.

8. Reliability

The Adolescent Alcohol Drinking Intention scale exhibits high internal consistency reliability across varied adolescent demographic strata, linguistic implementations, and experimental testing conditions.

Internal Consistency

In the foundational validation study conducted by Russell, Hamby, and Russell (2018) involving French adolescent participants (evaluating both French-language administrations and comparative translation frameworks), the three-item instrument achieved an outstanding Cronbach’s alpha of:

α = .89

Subsequent psychometric evaluations across adolescent cohorts have mirrored these findings, reporting Cronbach’s alpha and McDonald’s omega (ω) coefficients reliably ranging between .86 and .93. These figures surpass the standard .70 to .80 thresholds recommended for exploratory and applied research instruments. Item-total correlations for each of the three items routinely exceed .75, indicating that each item contributes substantial shared variance to the composite index without redundancy.

Temporal Stability (Test-Retest Reliability)

Test-retest assessments conducted over short intervals (e.g., 2 to 3 weeks) in control conditions without experimental intervention reveal intraclass correlation coefficients (ICC) ranging between .78 and .85. This temporal stability demonstrates that the AADI measures enduring behavioral intentions rather than fleeting transient moods, while maintaining sufficient sensitivity to capture genuine shifts resulting from targeted persuasive or corrective interventions.

9. Factor Analysis

The latent dimensionality of the AADI has been scrutinized using both Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA).

Exploratory Factor Analysis (EFA)

Principal Axis Factoring and Maximum Likelihood extraction methods applied to the three items invariably yield a single-factor solution. Scree plot analyses demonstrate a singular dominant eigenvalue (routinely exceeding 2.30), with subsequent eigenvalues failing to cross the standard Kaiser-Guttman criterion threshold of 1.0 (typically < 0.40). The primary factor accounts for 75% to 85% of the total item variance, confirming strong unidimensionality.

Confirmatory Factor Analysis (CFA)

In confirmatory structural equation modeling frameworks, the one-factor model exhibits excellent goodness-of-fit indices when evaluated against standard psychometric criteria:

• Comparative Fit Index (CFI): > .99 (values > .95 reflect excellent fit)
• Tucker-Lewis Index (TLI): > .98 (values > .95 reflect superior fit)
• Root Mean Square Error of Approximation (RMSEA): < .05 (90% CI [.000, .082], indicating close structural approximation)
• Standardized Root Mean Square Residual (SRMR): < .02 (values < .05 indicate minimal residual discrepancy)

Standardized factor loadings (λ) for the latent construct onto the observed items are uniformly robust:

• Item 1 (… drink alcohol?): λ ≈ .84 to .89
• Item 2 (… consume multiple drinks in a sitting?): λ ≈ .91 to .95
• Item 3 (… drink to the point of intoxication?): λ ≈ .86 to .91

Measurement invariance testing across gender categories (boys vs. girls) and developmental cohorts (early vs. late adolescence) confirms metric and scalar invariance, verifying that observed mean differences reflect genuine variations in intentional magnitude rather than differential item functioning.

10. Instrument / Measurement Tool

The AADI is designed as an ultra-brief, self-administered survey tool optimized for paper-and-pencil questionnaires, web-based survey software, and mobile field evaluations. Below are its structural specifications:

• Test Type: Standardized self-report subjective probability scale / stem-completion psychometric instrument.
• Target Population: Adolescents and youth (typically ages 12 to 18 years; adaptable for emerging adults ages 18–25).
• Language Editions: Validated in French (original administration) and English; translatable into other international languages using standard back-translation procedures.
• Item Count: 3 stem-completion items assessing escalating levels of alcohol engagement.
• Administration Time: Approximately 1 to 2 minutes.
• Response Scale: 7-point Likert-type subjective probability scale:
  • 1 = Pas du tout probable / Not at all likely
  • 2 = Très peu probable / Very unlikely
  • 3 = Peu probable / Unlikely
  • 4 = Modérément probable / Moderately likely
  • 5 = Probable / Likely
  • 6 = Assez probable / Fairly likely
  • 7 = Très probable / Very likely
• Scoring Rules:
  • Reverse Scoring: No items are reverse-scored; all items follow a direct valence where higher numerical values indicate stronger prospective drinking intentions.
  • Composite Calculation: Responses to the three items are summed and averaged to yield a single continuous mean score ranging from 1.00 to 7.00:
    
    AADI Score = (Item 1 + Item 2 + Item 3) / 3
  • Score Interpretation: Higher composite scores reflect a greater intentional propensity and subjective probability of initiating and escalating hazardous alcohol use over the designated prospective 12-month period.

11. Permissions & Fee and Test Year

The Adolescent Alcohol Drinking Intention scale was published in 2018 by Cristel Antonia Russell, Anne M. Hamby, and Daniel W. Russell in the Journal of Advertising (Volume 47, Issue 4, pages 395–411). The research article is copyrighted by the American Academy of Advertising and published by Routledge / Taylor & Francis Group.

Licensing and Accessibility: The scale items are publicly documented in the published peer-reviewed article for academic, educational, non-commercial research, and public health evaluation purposes. Under fair-use research standards, scholars and clinical evaluators may employ the AADI without licensing fees, provided proper academic attribution is maintained by citing the primary source article (Russell et al., 2018). Parties seeking commercial adaptation, proprietary software integration, or commercial monetization must seek formal permissions through the Taylor & Francis Group RightsLink licensing service.

12. References

• Ajzen, I. (1991). The Theory of Planned Behavior. Organizational Behavior and Human Decision Processes, 50(2), 179–211. https://doi.org/10.1016/0749-5978(91)90020-T
• Fishbein, M., & Ajzen, I. (2010). Predicting and Changing Behavior: The Reasoned Action Approach. Psychology Press. https://doi.org/10.4324/9780203838020
• Gibbons, F. X., Gerrard, M., Blanton, H., & Russell, D. W. (1998). Reasoned action versus social reaction: Thinking about adolescents’ risk behavior. In L. Berkowitz (Ed.), Advances in Experimental Social Psychology (Vol. 30, pp. 65–124). Academic Press. https://doi.org/10.1016/S0065-2601(08)60382-7
• Green, M. C., & Brock, T. C. (2000). The role of transportation in the persuasiveness of public narratives. Journal of Personality and Social Psychology, 79(5), 701–721. https://doi.org/10.1037/0022-3514.79.5.701
• Kuntsche, E., Kuntsche, S., Thrül, J., & Gmel, G. (2017). Binge drinking: Health impact, prevalence, correlates and interventions. Psychology & Health, 32(8), 976–1017. https://doi.org/10.1080/08870446.2017.1325883
• Russell, C. A., Hamby, A. M., & Russell, D. W. (2018). When a correction contradicts: Counter-messages may increase adolescents’ ambivalence in response to drinking-related narratives. Journal of Advertising, 47(4), 395–411. https://doi.org/10.1080/00913367.2018.1517711

13. Items of the Scale