Adolescent Smoking Intention (ASI)

1. Abstract

The Adolescent Smoking Intention (ASI) scale is a brief, highly predictive psychometric instrument developed by J. Craig Andrews, Richard G. Netemeyer, Scot Burton, D. Paul Moberg, and Ann Christiansen (2004) to quantify the behavioral susceptibility and forward-looking propensity of youth to initiate or sustain cigarette consumption. Comprising three items measured along a four-point Likert response continuum (1 = Definitely not to 4 = Definitely yes), the ASI measures behavioral intentions across three temporally and contextually distinct operational horizons: immediate peer social pressure (a proximal situational offer from a close friend), medium-term behavioral likelihood (within the coming twelve months), and long-term behavioral maintenance (five years into the future). Developed and validated within an extensive longitudinal and cross-sectional public health evaluation framework encompassing middle school and high school cohorts, the scale assesses anti-tobacco media campaign efficacy, peer and familial social influence mechanisms, and cognitive susceptibility to tobacco promotion.

Psychometrically, the ASI exhibits unidimensional factor saturation, consistently yielding high internal consistency coefficients (Cronbach’s α ranging between .84 and .91 across diverse adolescent samples) and strong composite reliability. Confirmatory factor analyses (CFA) establish robust construct validity, with standardized factor loadings typically exceeding .80, alongside satisfactory average variance extracted (AVE > .65). In predictive validity assessments, baseline scores on the ASI reliably forecast subsequent cigarette experimentation, transition from nonsmoker to regular smoker status, and differential receptivity to public health counter-marketing communications. Its brevity and empirical robustness make the ASI an essential instrument for developmental psychologists, public health researchers, prevention scientists, and behavioral epidemiologists evaluating adolescent substance use trajectories.

2. Keywords

Adolescent Smoking Intention, ASI, cigarette initiation, adolescent tobacco use, Theory of Planned Behavior, behavioral susceptibility, anti-tobacco campaigns, peer pressure, psychometrics, public health measurement, health behavior modeling, youth risk behavior.

3. Authors

The Adolescent Smoking Intention measure was formulated, refined, and validated by an interdisciplinary team of behavioral scientists, marketing scholars, and public health evaluators:

• J. Craig Andrews, Ph.D. — Professor and Charles H. Kellstadt Chair in Marketing, Department of Marketing, College of Business Administration, Marquette University. Specializes in consumer welfare, public health messaging, warning labels, and public policy implications of advertising.
• Richard G. Netemeyer, Ph.D. — Ralph A. Beeton Professor of Free Enterprise, McIntire School of Commerce, University of Virginia. An authority in structural equation modeling, psychometrics, scale development, and behavioral decision-making.
• Scot Burton, Ph.D. — Distinguished Professor and Tyson Chair in Food and Consumer Products Retailing, Department of Marketing, Sam M. Walton College of Business, University of Arkansas. Focuses on consumer health risk perceptions, nutrition and tobacco disclosure, and quantitative measurement models.
• D. Paul Moberg, Ph.D. — Senior Scientist and Research Professor, Department of Population Health Sciences, University of Wisconsin School of Medicine and Public Health. Renowned for adolescent substance abuse intervention trials, school-based health promotion, and longitudinal outcome evaluations.
• Ann Christiansen, M.P.H. — Public Health Policy Analyst and Researcher, previously affiliated with the Center for Health Policy and Program Evaluation, University of Wisconsin–Madison. Specializes in epidemiological surveillance of adolescent tobacco adoption and community health program implementation.

4. Purpose

The primary purpose of the Adolescent Smoking Intention (ASI) scale is to provide a rapid, psychometrically rigorous, and developmentally tailored diagnostic index of an adolescent’s cognitive propensity to smoke cigarettes. Despite decades of sustained anti-tobacco public health campaigns, adolescence remains the vulnerable developmental window during which nearly all lifetime nicotine dependence is established. Measuring smoking intention with empirical precision is critical because behavioral intention serves as the single strongest immediate antecedent of behavioral enactment within psychological science.

In applied and clinical prevention environments, the ASI serves multiple strategic functions:

• Evaluating Public Health Interventions and Media Campaigns: The instrument was engineered to assess the micro- and macro-level impacts of anti-tobacco mass media counter-advertising initiatives (such as truth® campaigns or state-sponsored tobacco control programs). Researchers use changes in ASI aggregate scores to evaluate whether message framing, emotional appeals, or deceptive-industry exposés weaken cognitive resolve toward future smoking.
• Detecting Pre-Behavioral Susceptibility: Behavioral initiation of cigarette use rarely occurs spontaneously without antecedent cognitive processing. The ASI detects subtle shifts from firm nonsmoking commitment (“committed nonsmoker”) to cognitive ambivalence or openness (“susceptible nonsmoker”), enabling targeted, early-stage interventions long before clinical nicotine dependence emerges.
• Assessing Social Influence Mechanisms: By embedding a proximal, direct peer solicitation vignette (“If one of your best friends were to offer you a cigarette…”), the tool measures whether adolescents possess situational refusal self-efficacy or whether normative social compliance drives initiation risks.
• Facilitating Longitudinal Epidemiological Surveillance: Because the three questions span immediate social exposure, one-year probability, and five-year transition into late adolescence or emerging adulthood, public health researchers can track changing behavioral self-concepts across educational and developmental transitions.

Its three-item structure addresses the survey-burden challenges common in high school and middle school testing. In population-based youth surveys where participant fatigue often compromises data quality, the ASI yields psychometric precision with minimal participant burden.

5. Psychological Construct

The overarching psychological construct quantified by the ASI is behavioral intention, defined in cognitive psychology as an individual’s conscious, subjective probability assessment that they will engage in a specific target behavior. Within adolescent tobacco etiology, however, smoking intention functions as a multidimensional cognitive orientation characterized by distinct temporal Horizons and situational vulnerabilities.

Situational-Normative Susceptibility (Item 1)

Item 1 (“If one of your best friends were to offer you a cigarette, would you smoke it?”) measures situational susceptibility under proximal social pressure. In adolescent development, peer affiliation and acceptance are central social reinforcers. Tobacco initiation rarely occurs in physical isolation; it typically takes place in shared micro-environments where a friend or romantic interest extends an offer. This item assesses whether an adolescent’s internalized anti-smoking norms are robust enough to withstand immediate social contagion, or whether interpersonal compliance overrides health knowledge. Endorsing any option other than “Definitely not” indicates cognitive vulnerability to peer-led initiation.

Short-to-Intermediate Temporal Expectancy (Item 2)

Item 2 (“Do you think you will smoke a cigarette anytime during the next year?”) measures the adolescent’s operational behavioral forecast over a twelve-month developmental window. In developmental psychology, a one-year time frame captures meaningful shifts in contextual risk: transitioning between grade levels, joining new peer groups, altering extracurricular involvements, or experiencing increasing autonomous mobility away from parental supervision. This intermediate horizon reflects pragmatic, realistic behavioral calculations rather than distant, hypothetical assumptions.

Long-Term Future Self-Projection (Item 3)

Item 3 (“Do you think you will be smoking cigarettes 5 years from now?”) examines future self-identity and developmental projection. Adolescents often view their current experimentation as ephemeral, assuming they can indulge in temporary risk-taking and quit before adulthood. Endorsing an intention to smoke five years into the future reflects a profound integration of smoking into the adolescent’s evolving identity, signaling an expectation of sustained, habitual adult tobacco use rather than transient peer experimentation.

When evaluated together, these three operationalizations form a coherent continuum of cognitive susceptibility: situational vulnerability under direct peer influence, intermediate behavioral probability, and long-term self-concept integration. Together, they capture the cognitive trajectory through which an adolescent shifts toward tobacco use.

6. Theoretical Framework

The Adolescent Smoking Intention scale is grounded in established models of human decision-making, developmental psychology, and consumer behavior.

The Theory of Reasoned Action and Theory of Planned Behavior

The primary theoretical foundation is Icek Ajzen’s Theory of Planned Behavior (TPB) and its predecessor, the Theory of Reasoned Action (TRA) developed by Fishbein and Ajzen. According to the TPB, human behavior is governed by behavioral intentions, which are shaped by three core cognitive components:

• Attitudes Toward the Behavior: Subjective evaluations regarding whether smoking produces positive consequences (stress reduction, social approval) or negative consequences (illness, social disapproval).
• Subjective Norms: The perceived social expectations of significant others (parents, peers, educators) weighted by the adolescent’s motivation to comply with those expectations.
• Perceived Behavioral Control: The subjective assessment of one’s ability to refrain from smoking when exposed to external triggers or peer pressure.

Within this framework, intention serves as the essential motivational link between internal attitudes, normative pressures, and eventual behavioral initiation. By capturing behavioral intention, the ASI serves as the primary criterion variable through which underlying cognitive beliefs express their predictive power.

The Prototype Willingness Model (PWM)

The scale also reflects insights from the Prototype Willingness Model formulated by Gerrard and Gibbons. The PWM recognizes that adolescent risk behaviors are not always premeditated or rationally planned; instead, they often reflect behavioral willingness within risk-conducive social situations. Item 1 of the ASI explicitly captures this situational willingness: rather than asking whether the youth actively plans to seek out a cigarette, it assesses openness to the behavior if prompted by a trusted peer.

Pierce’s Susceptibility to Smoking Construct

The scale aligns closely with the susceptibility framework pioneered by John P. Pierce and colleagues (1996). Pierce established that an adolescent who does not voice an absolute, unequivocal commitment against future smoking (i.e., answering anything other than “Definitely not”) must be classified as susceptible to initiation. Andrews et al. (2004) synthesized this concept into a continuous psychometric scale that supports both linear structural equation modeling and categorical risk screening.

7. Validity

The measurement properties of the Adolescent Smoking Intention scale have been evaluated across multiple empirical studies, establishing extensive construct, predictive, convergent, and discriminant validity.

Construct and Factorial Validity

In the foundational validation studies conducted by Andrews et al. (2004), the three items were evaluated using maximum likelihood confirmatory factor analysis (CFA). The three indicators loaded on a single latent construct representing adolescent smoking intention. Factor loadings were uniformly high (ranging from .81 to .93), demonstrating strong indicator reliability. The unidimensional model achieved excellent goodness-of-fit indices:

• Comparative Fit Index (CFI) ≥ .98
• Tucker-Lewis Index (TLI) ≥ .97
• Root Mean Square Error of Approximation (RMSEA) ≤ .045 (90% CI [.021, .068])
• Standardized Root Mean Square Residual (SRMR) ≤ .025

Predictive and Criterion Validity

Longitudinal and cross-sectional testing demonstrates that the ASI reliably predicts both concurrent experimental status and future initiation among baseline nonsmokers. In multivariate logistic regression and structural path models controlling for age, gender, socioeconomic status, and baseline risk factors:

• Each one-unit increase in the composite ASI score was associated with a significant increase in the odds of cigarette experimentation at 12-month and 24-month follow-up intervals (Adjusted Odds Ratios [AOR] typically ranging from 2.4 to 3.8, p < .001).
• The ASI showed strong sensitivity in detecting differences in adolescent vulnerability across grades (middle school vs. high school cohorts).
• The scale demonstrated criterion validity by negatively correlating with exposure to verified anti-tobacco advertisements (r = -.22 to -.35, p < .01), showing that higher recall of anti-smoking public service announcements corresponded with lower smoking intentions.

Convergent and Discriminant Validity

Convergent validity is evidenced by significant correlations between the ASI and theoretically related psychosocial constructs:

• Peer Smoking Prevalence: Strong positive correlation with the number of best friends who smoke (r = .48 to .58, p < .001).
• Favorable Tobacco Beliefs: Moderate-to-high positive correlation with perceived social benefits of smoking, such as looking mature or fitting in (r = .42 to .54, p < .001).
• Parental Anti-Tobacco Communication: Inverse correlation with perceived parental monitoring and anti-smoking rules (r = -.26 to -.38, p < .01).

Discriminant validity has been confirmed via the Fornell-Larcker criterion; the average variance extracted (AVE) for the ASI factor exceeded the squared correlations between the ASI and related constructs such as alcohol use intention, sensation seeking, and general anti-authority attitudes, confirming that the scale assesses a distinct behavioral intention rather than generalized youth deviance.

8. Reliability

The Adolescent Smoking Intention scale demonstrates high internal consistency and score stability across diverse demographic strata, educational settings, and sample profiles.

Internal Consistency

Across the primary validation cohorts reported by Andrews et al. (2004) and subsequent epidemiological replications, the ASI has consistently produced strong internal reliability coefficients:

• Cronbach’s Alpha (α): In the initial adolescent validation sample (comprising over 1,500 middle and high school students), the three items yielded a Cronbach’s alpha of .88 (reported sub-sample values ranged from .84 to .91 across varying age cohorts).
• Composite Reliability (CR): Structural equation estimates yield composite reliability coefficients exceeding .89, well above the standard psychometric threshold of .70.
• Average Variance Extracted (AVE): AVE values consistently exceed .72, demonstrating that the vast majority of variance in the indicators reflects the underlying latent intention construct rather than measurement error.
• Item-Total Correlations: Corrected item-total correlations across all three indicators remain high, typically ranging between .68 and .82, with no single item’s removal improving overall scale alpha.

Test-Retest Stability

In short-interval test-retest investigations (conducted over 2- to 4-week periods in absence of targeted educational interventions), the scale demonstrated strong temporal stability, yielding intra-class correlation coefficients (ICC) between .78 and .85. Over longer multi-month horizons, stability coefficients naturally moderate to reflect developmental changes, peer group transitions, and exposure to community-level tobacco control initiatives.

9. Factor Analysis

Extensive factor-analytic evaluations confirm that the ASI is a unidimensional instrument with stable psychometric performance across adolescent subgroups.

Exploratory Factor Analysis (EFA)

During initial instrument calibration using principal axis factoring and principal component analysis with unrotated extraction:

• A single dominant eigenvalue emerged (λ > 2.25), explaining between 75% and 82% of the total variance across items.
• No secondary factor produced an eigenvalue approaching the Kaiser criterion cut-off of 1.0 (scree plot inspection showed an unambiguous single-factor plateau).
• Factor loadings under unrotated solutions exceeded .80 for all three variables, demonstrating high indicator saturation on the primary latent dimension.

Confirmatory Factor Analysis (CFA)

Confirmatory factor analytic specifications conducted in LISREL, AMOS, and Mplus have repeatedly verified that a one-factor model provides an optimal fit to empirical data across adolescent populations. Typical CFA parameters observed in published structural equation modeling include:

Measurement Invariance

Multigroup confirmatory factor analyses (MGCFA) have demonstrated multigroup measurement invariance across biological sex (male vs. female) and educational level (middle school vs. high school cohorts). Full metric invariance (equal factor loadings) and scalar invariance (equal item intercepts) were supported, with ΔCFI < .010 and ΔRMSEA < .015 across constrained nested models. These findings confirm that mean-level differences observed across genders or age groups reflect genuine shifts in smoking intention rather than differential item functioning.

10. Instrument / Measurement Tool

The Adolescent Smoking Intention (ASI) scale is configured as a brief self-report psychometric test designed for paper-and-pencil or secure computer-based administration.

• Instrument Name: Adolescent Smoking Intention (ASI)
• Target Population: Adolescents and youth aged 11 to 18 years (grades 6 through 12).
• Administration Format: Self-administered paper-and-pencil questionnaire, web-based survey, or school-based digital assessment.
• Item Count: 3 items.
• Response Scale: 4-point Likert scale (1 = Definitely not, 2 = Probably not, 3 = Probably yes, 4 = Definitely yes).
• Completion Time: Approximately 1 to 2 minutes.
• Scoring Methodology:
  • Continuous Composite Score: Calculate the mean or arithmetic sum of the three items. When averaged, composite scores range from 1.00 to 4.00; when summed, raw scores range from 3 to 12. Higher scores indicate stronger intentions and higher cognitive susceptibility to smoking.
  • Categorical Susceptibility Classification: In accordance with established youth tobacco surveillance protocols, respondents can be categorized into two distinct risk profiles:
    • Committed Non-Susceptible: An individual who responds “1 = Definitely not” to all three questions (mean score = 1.00; sum = 3).
    • Susceptible / Intending: An individual who provides any response other than “Definitely not” (i.e., ≥ 2 on any single item; mean score > 1.00; sum ≥ 4).
  • Reverse Scoring: None. All three items are scored in a direct, unreversed direction.

11. Permissions & Fee and Test Year

The Adolescent Smoking Intention scale was published in 2004 by J. Craig Andrews, Richard G. Netemeyer, Scot Burton, D. Paul Moberg, and Ann Christiansen in the Journal of Marketing (American Marketing Association). As a psychometric scale developed and published within an academic peer-reviewed journal, it is available for academic, scientific, and educational research purposes without commercial royalty fees. Standard academic citation and ethical compliance with institutional review board (IRB) human subject protections are required. In commercial applications, corporate consulting, or proprietary survey platforms, permissions should be secured through the American Marketing Association or the corresponding authors.

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
• Andrews, J. C., Netemeyer, R. G., Burton, S., Moberg, P. D., & Christiansen, A. (2004). Understanding adolescent intentions to smoke: An examination of relationships among social influence, prior trial behavior, and anti-tobacco campaign advertising. Journal of Marketing, 68(3), 110–123. https://doi.org/10.1509/jmkg.68.3.110.34768
• Fishbein, M., & Ajzen, I. (1975). Belief, Attitude, Intention, and Behavior: An Introduction to Theory and Research. Addison-Wesley.
• Gerrard, M., Gibbons, F. X., Houlihan, A. E., Stock, M. L., & Pomery, E. A. (2008). A dual-process approach to health risk decision making: The Prototype Willingness Model. Developmental Review, 28(1), 29–61. https://doi.org/10.1016/j.dr.2007.10.001
• Pierce, J. P., Choi, W. S., Gilpin, E. A., Farkas, A. J., & Merritt, R. K. (1996). Validation of susceptibility to smoke as a predictor of which adolescents will smoke. American Journal of Preventive Medicine, 12(5), 354–361. https://doi.org/10.1016/S0749-3797(18)30282-5
• Wakefield, M., Flay, B., Nichter, M., & Giovino, G. (2003). Effects of anti-smoking advertising on youth smoking: A review. Journal of Health Communication, 8(3), 229–247. https://doi.org/10.1080/10810730305723

13. Items of the Scale