The Situation Temptation Scales- Smoking Version<\/strong> (STS-S), developed by Wayne F. Velicer<\/a>, Carlo C. DiClemente<\/a>, Joseph S. Rossi, and James O. Prochaska<\/a> (1990), is an empirically validated psychometric instrument designed to assess an individual’s perceived temptation to smoke across a diverse array of situational contexts. Grounded within the Transtheoretical Model<\/a> (TTM) of behavior change and integrated with Albert Bandura<\/a>‘s self-efficacy<\/a> theory, the STS-S measures the situational intensity of urges that challenge an individual\u2019s resolve to maintain tobacco abstinence or achieve cessation. The instrument exists primarily in two structurally parallel formats: a full 20-item version and an abbreviated 9-item version. Both configurations evaluate three correlated, structurally invariant primary factors: Negative Affect<\/em>, Positive\/Social<\/em>, and Habitual\/Craving<\/em> (or Habit\/Addictive). Respondents rate each situation on a standardized 5-point Likert-type scale ranging from 1 (Not at all tempted<\/em>) to 5 (Extremely tempted<\/em>). Extensive psychometric investigations across clinical, community, and population-based samples demonstrate exceptional internal consistency reliability, with Cronbach’s alpha coefficients routinely exceeding .90 for Negative Affect, .85 for Positive\/Social, and .80 for Habitual\/Craving dimensions. Structural equation modeling and confirmatory factor analysis confirm a robust hierarchical or three-factor oblique architecture that demonstrates strong measurement invariance across biological sex, age, and stages of change (Precontemplation, Contemplation, Preparation, Action, and Maintenance). Furthermore, the scale demonstrates rigorous predictive validity for smoking cessation outcomes, situational lapses, and clinical relapse trajectories, establishing it as an indispensable assessment battery in addiction medicine, behavioral health psychology, and clinical trials targeting tobacco dependence.<\/p>\n Situation Temptation Scales, STS-S, Transtheoretical Model, Smoking Cessation, Nicotine Dependence, Self-Efficacy, Relapse Prevention, Negative Affect, Addictive Behaviors, Psychometrics<\/p>\n The Situation Temptation Scales for smoking were conceptualized, developed, and psychometrically validated by leading behavioral scientists affiliated with the Cancer Prevention Research Center (CPRC) at the University of Rhode Island<\/a> and the Department of Psychology at the University of Maryland, Baltimore County<\/a>:<\/p>\n The primary purpose of the Situation Temptation Scales- Smoking Version is to provide a standardized, theoretically rigorous, and psychometrically sound measurement of the magnitude of temptation an individual experiences to smoke cigarettes when confronting specific high-risk internal and external cues. In the study of addictive behaviors, relapse cannot be conceptualized merely as a biological reflex or an unpredictable event; rather, it is a complex behavioral process precipitated by dynamic environmental, cognitive, and affective stimuli. The STS-S operationalizes situational temptation as a critical mediator of behavior change, capturing the subjective vulnerability that threatens an individual’s commitment to cessation.<\/p>\n In clinical environments, the STS-S serves as a vital diagnostic and treatment-planning instrument. By administering the inventory during baseline evaluations, clinicians can profile a client\u2019s idiosyncratic high-risk triggers. For instance, identifying whether a patient experiences peak temptation primarily in response to emotional distress (Negative Affect) versus social gatherings and alcohol consumption (Positive\/Social) allows cognitive-behavioral therapists to tailor relapse prevention strategies. Tailored interventions may include emotional regulation skills training, cognitive restructuring, assertive refusal training, or environmental stimulus control. Longitudinal administration of the scale enables clinicians to track shifts in temptation across therapeutic phases, providing quantitative verification of whether behavioral coping mechanisms successfully attenuate subjective cravings in response to conditioned cues.<\/p>\n In epidemiological, public health, and clinical trial research, the STS-S provides an indispensable outcome measure. Researchers utilize the scale to evaluate the efficacy of pharmacological agents (such as nicotine replacement therapy, bupropion, or varenicline) and behavioral modalities (such as motivational interviewing, acceptance and commitment therapy, or digital health interventions) in mitigating situational cravings. Moreover, the STS-S is employed to test structural hypotheses within health behavior models, such as examining how self-regulatory capacity and situational temptation interact to predict longitudinal transitions across the stages of change, smoking cessation maintenance, and prospective relapse risk over 6- to 24-month intervals.<\/p>\n The Situation Temptation Scales- Smoking Version operationalizes the construct of situational temptation<\/em>, which represents an individual’s conscious perception of the urge or desire to engage in tobacco use when challenged by specific intrapersonal or interpersonal circumstances. As conceptualized by Velicer et al. (1990), temptation functions as the functional inverse or dynamic counterweight to situational self-efficacy<\/a>\u2014an individual’s confidence in their ability to abstain from smoking across identical stressful, social, or conditioned situations. Rather than treating craving as an undifferentiated, static biological state, the STS-S delineates temptation as a multifaceted psychological construct characterized by three distinct, replicable domains:<\/p>\n The Negative Affect<\/em> dimension measures the degree to which distressing emotional states and internal psychological conflicts evoke an impulse to smoke. This construct reflects the pervasive tendency of tobacco users to employ nicotine as an artificial pharmacological coping mechanism for stress reduction, tension attenuation, and negative mood management. Individuals exhibiting elevated scores on this subscale experience profound temptation when confronting acute interpersonal friction, profound sadness, anxiety, frustration, or existential crises. Key behavioral manifestations and representative scenarios include:<\/p>\n The Positive\/Social<\/em> dimension captures temptation triggered by pleasant, socially affirming, celebratory, or peer-reinforced environments. This facet reflects the sociocultural conditioning of smoking, wherein cigarette consumption has become deeply paired with recreational relaxation, social bonding, dining, and celebratory rituals. Rather than smoking to escape an aversive state, individuals scoring high on this dimension are motivated by reward enhancement, sensory indulgence, and peer affiliation. Situations embodying this construct include:<\/p>\n The Habitual\/Craving<\/em> (also termed Habit\/Addictive) dimension assesses temptation arising from neurobiological dependence, automatic behavioral routines, visceral craving, and depleted physical energy. This construct taps into conditioned stimulus-response patterns established over thousands of smoking repetitions, wherein specific times of day, physiological transitions, or decreases in systemic nicotine concentration automatically trigger an impulse to smoke in the absence of explicit emotional turbulence. Characteristic manifestations include:<\/p>\n The theoretical foundations of the Situation Temptation Scales are deeply anchored within two seminal frameworks in psychological science: the Transtheoretical Model<\/a> (TTM) of behavior change formulated by Prochaska and DiClemente (1983; Prochaska, DiClemente, & Norcross, 1992) and the Cognitive Social Learning Theory of relapse pioneered by G. Alan Marlatt and Judith R. Gordon (1985), integrated with Albert Bandura\u2019s (1977, 1986) self-efficacy paradigm.<\/p>\n Within the TTM architecture, intentional behavior change does not occur as a single, discrete event, but rather unfolds across a temporal continuum comprising five distinct stages: Precontemplation<\/em> (no intention to quit within the foreseeable future), Contemplation<\/em> (recognizing the problem and intending to quit within six months), Preparation<\/em> (actively planning to take action within the next 30 days and attempting preliminary modifications), Action<\/em> (overt modification of behavior and total tobacco abstinence for less than six months), and Maintenance<\/em> (sustained abstinence lasting beyond six months up to several years). As individuals progress through these stages, cognitive, emotional, and behavioral processes of change alter the psychological valuation of the addictive behavior.<\/p>\n Velicer, DiClemente, Rossi, and Prochaska (1990) integrated Bandura’s self-efficacy construct into the TTM by conceptualizing situational confidence and situational temptation as complementary, mathematically inverse, yet phenomenologically distinct cognitive dimensions. Bandura defined self-efficacy as the subjective conviction that one can successfully execute the behavior required to produce a desired outcome in the face of obstacles. In the context of tobacco cessation, Bandura’s concept translates into an individual’s confidence that they can abstain from smoking across specific challenging situations. In parallel, Marlatt and Gordon\u2019s (1985) relapse model posits that exposure to high-risk situations (such as interpersonal conflict, social pressure, and negative emotional states) presents an immediate threat to perceived control. If an individual lacks adequate behavioral or cognitive coping strategies, the immediate temptation to smoke surges, precipitating a lapse (an initial breach of abstinence), which may subsequently escalate into a full-blown relapse via the Abstinence Violation Effect<\/a> (AVE).<\/p>\n Velicer and colleagues demonstrated that while confidence and temptation represent opposing psychological forces, measuring temptation directly yields superior predictive sensitivity during the earliest stages of change (Precontemplation and Contemplation), where overt confidence is nearly non-existent but temptation varies markedly according to cue exposure. Across the stages of change, empirical research demonstrates a profound, systematic crossover: in Precontemplation, situational temptation is at its absolute zenith while confidence is minimal; as individuals advance into Action and Maintenance, temptation systematically declines in an inverse linear trajectory, reaching its nadir in long-term Maintenance. Thus, the STS-S captures the dynamic decay of cue-elicited craving across the behavioral change spectrum.<\/p>\n The Situation Temptation Scales- Smoking Version has undergone rigorous empirical validation across diverse populations, establishing strong construct, convergent, discriminant, and predictive validity.<\/p>\n Initial structural validation by Velicer et al. (1990) evaluated large, representative cohorts of smokers and quitters ($N > 1,000$<\/span>). Through exploratory and subsequent confirmatory factor analyses, the hypothesized three-factor architecture\u2014Negative Affect, Positive\/Social, and Habitual\/Craving\u2014was confirmed. Confirmatory factor analytic (CFA) models demonstrated excellent fit indices, establishing that the 20-item configuration and its streamlined 9-item counterpart successfully reflect three distinct, correlated first-order constructs under a higher-order overarching temptation dimension. Furthermore, factorial invariance testing has established metric and scalar invariance across biological sex, diverse age strata, and across each of the five TTM stages of change, confirming that the instrument measures identical psychological constructs regardless of a participant’s demographic background or cessation status.<\/p>\n Convergent validity has been repeatedly corroborated via statistically significant correlations between the STS-S subscales and established indicators of nicotine dependence and tobacco consumption. The Habitual\/Craving subscale consistently correlates positively with the Fagerstr\u00f6m Test for Nicotine Dependence<\/a> (FTND), biological biomarkers such as expired carbon monoxide (CO) and plasma cotinine levels, baseline cigarettes-per-day consumption, and the brevity of time to the first morning cigarette ($r = .40$<\/span> to $.65, p < .001$<\/span>). The Negative Affect subscale correlates substantially with validated measures of depressive symptoms (e.g., the Beck Depression Inventory<\/a>) and perceived stress scales (e.g., the Perceived Stress Scale, PSS), confirming that the subscale specifically captures affect-driven impulses to smoke.<\/p>\n Discriminant validity is supported by the near-perfect inverse relationship observed between the STS-S subscales and their counterpart Situational Self-Efficacy\/Confidence subscales. Intercorrelations between temptation and confidence in identical situations generally range from $r = -.70$<\/span> to $r = -.88$<\/span>, indicating that while they are closely linked polar constructs, temptation reliably explains unique variance in affective vulnerability and lapse dynamics that general self-efficacy scales fail to capture entirely.<\/p>\n The predictive validity of the STS-S has been established across prospective clinical cessation trials. In longitudinal studies tracking smokers attempting cessation, elevated aggregate scores on the STS-S at baseline significantly predict early treatment attrition, acute lapse occurrences during the first 14 days of an attempt, and long-term relapse at 6- and 12-month follow-up evaluations. Specifically, elevated Negative Affect temptation scores reliably emerge as the single strongest psychological predictor of post-cessation relapse, especially among individuals exhibiting high emotional distress or underlying affective vulnerability. Conversely, reductions in STS-S scores during active behavioral intervention reliably differentiate successful long-term maintainers from relapsers, underscoring the scale’s sensitivity to therapeutic change.<\/p>\n The STS-S exhibits exceptional internal consistency and temporal stability across both the 20-item full instrument and the 9-item abbreviated form. In the foundational psychometric evaluation conducted by Velicer et al. (1990), internal consistency coefficients (Cronbach’s alpha<\/a>) for the 20-item version demonstrated high measurement precision across all three subscales:<\/p>\n In subsequent large-scale clinical trials and cross-sectional evaluations across diverse international cohorts, Cronbach’s alpha values have shown remarkable replication, consistently ranging between .88 and .96 for Negative Affect, .82 and .90 for Positive\/Social, and .78 and .85 for Habitual\/Craving. The full 20-item total composite score typically yields an internal consistency coefficient exceeding $\\alpha = .93$<\/span>.<\/p>\n For the streamlined 9-item short-form version\u2014which selects the three highest-loading, most representative items per factor\u2014internal consistency remains robust despite reduced scale length:<\/p>\n Test-retest reliability evaluations over intervals of two to three weeks among stable smokers in the Precontemplation stage (where behavioral changes have not yet been initiated) demonstrate excellent stability coefficients ranging from $r_{tt} = .81$<\/span> to $.89$<\/span>, confirming that the scale is not confounded by random transient daily fluctuations, but rather reflects enduring cognitive-affective behavioral patterns.<\/p>\n The structural development and refinement of the STS-S utilized rigorous exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) protocols designed to isolate distinct situational domains while minimizing cross-loadings and measurement error.<\/p>\n During the initial item generation phase, Velicer and colleagues compiled a comprehensive pool of candidate items reflecting potential relapse situations drawn from Marlatt’s relapse taxonomy, clinical interviews with smokers, and prior behavioral inventories. Principal component analysis (PCA) followed by oblique rotation (promax and oblimin) revealed a distinct three-component structure. Items demonstrated high primary factor loadings ($> .60$<\/span>) onto their designated latent constructs, with minimal secondary cross-loadings ($< .25$<\/span>). The three extracted factors accounted for more than 60% of the total item variance. The inter-factor correlations revealed that while the three dimensions are moderately correlated ($r \\approx .45$<\/span> to $.65$<\/span>), confirming an underlying global construct of temptation, they maintain sufficient unique variance to warrant independent subscale scoring.<\/p>\n Confirmatory factor analysis was conducted to compare alternative theoretical models: (a) a unidimensional single-factor model where all items load onto general temptation, (b) an orthogonal three-factor model assuming complete independence of the subscales, (c) an oblique three-factor model allowing latent factors to covary, and (d) a second-order hierarchical model where three first-order factors load onto a higher-order general Temptation dimension. Multiple independent structural equation modeling evaluations have consistently demonstrated that the oblique three-factor model and the hierarchical second-order model exhibit superior fit across multiple standard indices:<\/p>\n In CFA evaluations, standardized factor loadings for the 20-item instrument remain uniformly strong: items loading on the Negative Affect latent factor exhibit standardized path coefficients ranging from $.78$<\/span> to $.92$<\/span>; items on the Positive\/Social latent factor load between $.62$<\/span> and $.84$<\/span>; and items on the Habitual\/Craving factor load between $.58$<\/span> and $.80$<\/span>. The 9-item short-form version yields even tighter factor loadings (all standardized $lambda > .70$<\/span>), with CFA model fit metrics frequently surpassing CFI and TLI values of $.97$<\/span>, validating its utility in extensive clinical batteries and large population-based surveys where participant burden must be minimized.<\/p>\n The Situation Temptation Scales- Smoking Version was formally published in 1990 following foundational validation studies conducted across the late 1980s by Velicer, DiClemente, Rossi, and Prochaska. The primary seminal citation is: Velicer, W. F., DiClemente, C. C., Rossi, J. S., & Prochaska, J. O. (1990). Relapse situations and self-efficacy: An integrative model. Addictive Behaviors, 15(3), 271\u2013283.<\/em><\/p>\n Licensing and Accessibility<\/strong>: In alignment with the longstanding public health mission of the late authors and the Cancer Prevention Research Center (CPRC) at the University of Rhode Island, as well as the Habits Lab at the University of Maryland, Baltimore County (UMBC), the Situation Temptation Scales are considered accessible for academic, clinical, educational, and non-commercial research endeavors without the requirement of licensing fees or royalty payments. Researchers and clinicians are granted permission to reproduce and administer the scale provided that the original authors are appropriately cited and acknowledged in all resulting manuscripts, reports, and clinical platforms. For commercial deployments, electronic medical record integrations, or digital commercial health applications, inquiries regarding formal licensing agreements should be addressed to the respective institutional technology transfer offices at the University of Rhode Island or UMBC.<\/p>\nKeywords<\/h2>\n
Authors<\/h2>\n
\n
Purpose<\/h2>\n
Psychological Construct<\/h2>\n
1. Negative Affect (Emotional Distress)<\/h3>\n
\n
2. Positive\/Social (Social Facilitation and Celebration)<\/h3>\n
\n
3. Habitual\/Craving (Automaticity and Physiological Conditioning)<\/h3>\n
\n
Theoretical Framework<\/h2>\n
Validity<\/h2>\n
Construct and Structural Validity<\/h3>\n
Convergent and Discriminant Validity<\/h3>\n
Predictive and Criterion Validity<\/h3>\n
Reliability<\/h2>\n
\n
\n
Factor Analysis<\/h2>\n
Exploratory Factor Analysis (EFA)<\/h3>\n
Confirmatory Factor Analysis (CFA) and Model Fit<\/h3>\n
\n
Instrument \/ Measurement Tool<\/h2>\n
\n
\n
\n
\n
\n
\n
Permissions & Fee and Test Year<\/h2>\n
References<\/h2>\n
\n
13. Items of the Scale (Questionnaire)<\/h2>\n