South Oaks Gambling Screen – Revised for Adolescents (SOGS-RA)

1. Abstract

The South Oaks Gambling Screen – Revised for Adolescents (SOGS-RA; Winters, Stinchfield, & Fulkerson, 1993) is one of the most widely utilized psychometric screening instruments designed specifically to assess problem gambling behaviors and associated negative psychosocial consequences among youth populations aged 12 to 18 years. Adapted from the classic adult South Oaks Gambling Screen (SOGS; Lesieur & Blume, 1987), the SOGS-RA reformulates adult-oriented indicators—such as lines of credit, spousal conflicts, and check kiting—into developmentally congruent behaviors applicable to adolescents, including academic absenteeism, borrowing lunch money or allowances without repayment, interpersonal friction with parents and peers, chasing losses, and covert gambling activities. The core instrument consists of a preliminary behavioral filter question (“Do you gamble?”) followed by 12 substantive items referencing the past 12-month timeframe. Item responses are recorded either through standardized frequency gradations (Every time, Most of the time, Some of the time, Never) or dichotomous endorsements, which are then recoded into a cumulative problem index ranging from 0 to 12 points. Extant empirical investigations demonstrate robust psychometric properties across diverse high school, clinical, and community cohorts: internal consistency estimates typically yield Cronbach’s alpha coefficients between .74 and .84, with test-retest reliability across brief intervals exceeding r = .70. Factor-analytic investigations predominantly support an essential unidimensional construct representing adolescent gambling severity, alongside minor secondary facets capturing intra-personal loss of control and inter-personal consequences. Clinical cutoffs categorize youth into non-problem gamblers (scores 0–1), at-risk or transitionally problematic gamblers (scores 2–3), and problem/probable pathological gamblers (scores 4 or higher). This comprehensive review delineates the historical development, psychometric properties, theoretical underpinnings, diagnostic utility, and practical administration standards of the SOGS-RA.

2. Keywords

South Oaks Gambling Screen – Revised for Adolescents, SOGS-RA, adolescent gambling, problem gambling, behavioral addiction, psychometrics, screening instruments, chasing losses, adolescent risk behavior, scale validation

3. Authors

The South Oaks Gambling Screen – Revised for Adolescents was developed by Ken C. Winters, Ph.D., Randy D. Stinchfield, Ph.D., and Jayne Fulkerson, Ph.D. at the Department of Psychiatry, Division of Child and Adolescent Psychiatry, University of Minnesota Medical School (Minneapolis, Minnesota, USA).

• Ken C. Winters, Ph.D.: Senior Scientist at the Oregon Research Institute (Minnesota branch) and Adjunct Professor in the Department of Psychiatry at the University of Minnesota. Dr. Winters is an internationally recognized authority on adolescent substance use disorders and youth behavioral addictions.
• Randy D. Stinchfield, Ph.D.: Associate Professor in the Department of Psychiatry at the University of Minnesota Medical School. Dr. Stinchfield specializes in psychometrics, diagnostic nosology, and gambling assessment instruments.
• Jayne Fulkerson, Ph.D.: Professor at the University of Minnesota School of Nursing, specializing in quantitative methodology, family systems, and adolescent health behaviors.

4. Purpose

The fundamental purpose of the SOGS-RA is to provide a brief, psychometrically sound, developmentally sensitive screening tool for detecting early manifestations of maladaptive betting patterns, loss of cognitive-behavioral control, and the functional sequelae of gambling among adolescents. During the late 1980s and early 1990s, epidemiological surveys increasingly reported elevated rates of informal and commercial gambling engagement among youth, spanning activities such as card games, sports pools, lottery scratch tickets, and dice wagering. However, standard instruments designed for adults—chiefly the original 20-item SOGS (Lesieur & Blume, 1987)—exhibited substantial construct non-equivalence when administered to adolescents. Adult indices relied heavily on financial disruptions that youth rarely encounter, including mortgaging real estate, obtaining illegal loan-shark credits, liquidating retirement portfolios, or experiencing workplace discipline.

Winters and colleagues (1993) sought to resolve this psychometric dissonance by recalibrating the diagnostic thresholds and operational indicators to reflect the developmental milieu of teenagers. The SOGS-RA targets the behavioral, affective, and relational disruptions specific to secondary school students and emerging young adults. Its primary applications span:

• School-Based and Community Epidemiological Surveys: Estimating population-level base rates of gambling participation, identifying shifts from casual recreational wagering to disordered engagement, and evaluating public health prevention curricula (Macdonald, Turner, & Somerset, 2008).
• Clinical Triage and Intake Screening: Serving as a first-line triage instrument in child and adolescent mental health clinics, youth substance use rehabilitation facilities, and juvenile diversion programs to identify co-occurring behavioral addictions.
• Intervention and Prevention Outcome Tracking: Evaluating changes in adolescent gambling severity over time following psychoeducational, cognitive-behavioral, or family-systems interventions.

By establishing a standardized 12-month recall frame, the SOGS-RA differentiates transient recreational betting experimentation from entrenched problem patterns characterized by cognitive preoccupation, loss chasing, and interpersonal deceit.

5. Psychological Construct

The psychological construct measured by the SOGS-RA is adolescent problem gambling severity, conceptualized as a continuum ranging from subclinical recreational involvement to severe behavioral dysregulation and functional impairment. Grounded in criteria adapted from both the diagnostic framework for pathological gambling in the Diagnostic and Statistical Manual of Mental Disorders (DSM-III-R and DSM-IV) and social-ecological formulations of youth delinquency, the SOGS-RA captures multiple empirical facets:

1. Chasing Losses and Cognitive Preoccupation

Chasing represents a hallmark cognitive distortion and behavioral trap of disordered gambling (Item 1). It reflects an adolescent’s compulsion to return to betting arenas on subsequent days with the explicit, maladaptive intent to recoup accumulated financial losses. This behavior demonstrates the progression from recreational leisure to an escalating, instrumental cycle driven by irrational expectations of imminent restitution.

2. Dishonesty, Dissimulation, and Impression Management

The scale measures the erosion of interpersonal veracity through items assessing false claims regarding gambling outcomes (Item 2: claiming to have won money when one actually lost) and the covert concealment of physical paraphernalia (Item 8: hiding lottery tickets, betting slips, or IOUs from parents and close companions). These behaviors reflect profound impression management motivated by shame, guilt, or fear of parental sanctions.

3. Impaired Self-Regulation and Loss of Volitional Control

Self-regulatory failure is tapped via indicators of gambling longer or wagering larger amounts of money than originally intended (Item 4), as well as conscious, unsuccessful attempts to cease or cut back betting (Item 7). In adolescents whose prefrontal cortical executive controls are still maturing, these items mark heightened vulnerability to immediate reinforcement schedules over long-term negative consequences.

4. Interpersonal, Social, and Academic Disruption

Disordered betting consistently damages primary relational networks. The SOGS-RA evaluates social conflict through reports of arguments with parents, siblings, or peers (Items 3 and 9), explicit interpersonal criticism from external observers who detect problem behaviors (Item 5), and systemic institutional failure, specifically truancy from school or absenteeism from part-time employment (Item 11).

5. Financial Strain, Delinquency, and Illegal Acts

Because adolescents typically lack independent credit lines or substantial financial capital, economic distress manifests through developmentally contextualized fiscal behaviors. The SOGS-RA evaluates borrowing funds without subsequent repayment (Item 10) and committing illegal acts such as petty theft, shoplifting, or taking money from family members’ wallets to bankroll wagers or satisfy gambling-related debts (Item 12).

6. Affective Distress

Subjective distress is indexed by dysphoric emotional responses following gambling episodes (Item 6), wherein the young person experiences acute remorse, depressive affect, anxiety, or self-reproach concerning the magnitude of losses or the uncontrollable nature of the betting event.

6. Theoretical Framework

The SOGS-RA is anchored in integrated biopsychosocial models of adolescent risk behavior and behavioral addiction, drawing upon three prominent theoretical frameworks:

1. The Pathways Model of Problem Gambling

Formulated by Blaszczynski and Nower (2002), the Pathways Model posits three distinct etiologic trajectories toward problem gambling: (a) behaviorally conditioned gamblers, who drift into dysfunction through early big wins and habituation; (b) emotionally vulnerable gamblers, who utilize wagering as a maladaptive coping mechanism to escape negative affective states, childhood trauma, or mood disorders; and (c) antisocial impulsivist gamblers, characterized by profound neurobiological disinhibition, sensation-seeking, and comorbid conduct problems. The SOGS-RA items systematically tap the operational footprints of all three trajectories—conditioning mechanisms appear in chasing behavior (Item 1), emotional vulnerability manifests in subjective dysphoria (Item 6), and impulsive/antisocial features surface in academic truancy (Item 11) and illicit acquisition of funds (Item 12).

2. Problem Behavior Theory (PBT)

Jessor and Jessor’s (1977) Problem Behavior Theory posits that adolescent risk activities—including substance abuse, early sexual activity, delinquency, and disordered gambling—do not occur in isolation. Instead, they constitute an organized syndrome of non-conventionality driven by reciprocal interactions between the perceived social environment (e.g., peer modeling, low parental monitoring), personality dispositions (e.g., high value on independence vs. low value on academic achievement), and behavioral practices. The SOGS-RA operationalizes problem gambling as an integrated component of this adolescent risk syndrome, specifically linking betting to broader delinquency (theft, truancy, and deception).

3. Developmental Neurobehavioral Vulnerability Models

Contemporary developmental neuroscience highlights the asynchronous maturation of the adolescent brain: the socioemotional subcortical dopamine reward system (nucleus accumbens, ventral striatum) develops precociously during puberty, whereas cognitive control systems within the prefrontal cortex mature well into the third decade of life (Steinberg, 2008). Consequently, youth exhibit heightened sensitivity to immediate rewards and variable-ratio reinforcement schedules coupled with compromised response inhibition. SOGS-RA items assessing an inability to stop betting (Item 7) or wagering beyond intended thresholds (Item 4) capture this fundamental neurodevelopmental vulnerability.

7. Validity

Extensive empirical studies have evaluated the construct, convergent, discriminant, and criterion-related validity of the SOGS-RA across community, educational, and clinical samples:

Construct and Criterion Validity

In their seminal validation study across a stratified sample of Minnesota high school students (N = 702), Winters, Stinchfield, and Fulkerson (1993) demonstrated that adolescents categorized as problem gamblers by the SOGS-RA exhibited statistically significant elevations across multiple objective behavioral markers compared to non-problem peers. Problem gamblers demonstrated significantly higher weekly gambling frequencies (p < .001), larger median financial sums wagered per week, and a substantially greater diversity of gambling formats engaged in simultaneously. Re-evaluating these parameters in a broader statewide adolescent cohort, Winters, Stinchfield, and Kim (1995) replicated these findings, verifying that SOGS-RA severity classifications accurately reflect real-world gambling involvement.

Convergent Validity

Convergent validity has been established through strong correlations between SOGS-RA scores and alternative adolescent behavioral addiction and psychopathology measures:

• DSM-IV Adolescent Criteria (DSM-IV-MR-J): SOGS-RA scores demonstrate strong positive correlations (ranging from r = .68 to .78) with the Diagnostic and Statistical Manual Fourth Edition Revised for Juveniles adapted by Fisher (2000), confirming that both instruments measure the same underlying pathological construct despite differing diagnostic thresholds.
• Substance Use and Delinquency: Significant positive associations are consistently documented between elevated SOGS-RA scores and concurrent adolescent alcohol abuse, tobacco consumption, illicit drug experimentation, and conduct disordered behaviors (Stinchfield, 2000; Derevensky & Gupta, 2000).
• Cognitive Distortions: SOGS-RA scores correlate positively with validated measures of adolescent gambling-related cognitive fallacies, such as the Gamblers’ Belief Questionnaire (GBQ) and perceived illusions of control.

Discriminant Validity

The scale effectively discriminates between adolescent non-gamblers, purely recreational/social gamblers, and individuals experiencing functional impairment. Discriminant function analyses show that non-problem recreational gamblers do not endorse items involving structural distress (theft, chronic truancy, or intense familial conflict), whereas individuals scoring in the problem category (scores ≥ 4) report these severe functional sequelae at high rates (Stinchfield, 2002).

Social Desirability Considerations

Kuentzel and Henderson (2002) examined social desirability response biases on SOGS-derived metrics, reporting that youth who score high on impression management scales may modestly underreport items referencing theft (Item 12) or borrowing without repayment (Item 10). Nevertheless, construct validity remains robust even after partialling out social desirability variance.

8. Reliability

The psychometric reliability of the SOGS-RA has been corroborated through diverse evaluations of internal consistency and temporal stability:

Internal Consistency

In the original derivation study by Winters et al. (1993), the SOGS-RA demonstrated an overall Cronbach’s alpha of α = .80 among active adolescent gamblers. Subsequent independent investigations across varied geographic and cultural contexts have affirmed this reliability level:

• Winters, Stinchfield, and Kim (1995) reported an internal consistency coefficient of α = .81 in an expanded multi-school sample.
• Derevensky and Gupta (2000) obtained an alpha of α = .78 in a large Canadian cohort of junior high and high school students.
• Stinchfield (2002) observed Cronbach’s alpha values ranging from .74 to .84 across diverse demographic subsamples, indicating that the 12 items possess strong internal coherence without excessive item redundancy.
• In international adaptations (e.g., European and Latin American translations), Cronbach’s alpha has consistently ranged between .72 and .82, underscoring cross-linguistic reliability (Edgren et al., 2016).

Test-Retest Stability

Temporal stability assessments conducted across 2- to 4-week test-retest intervals have yielded stability coefficients between r = .71 and .82, demonstrating adequate longitudinal consistency for a behavioral screening instrument. Shorter 7-day test-retest trials within structured classroom settings have demonstrated intraclass correlation coefficients (ICC) exceeding .85.

9. Factor Analysis

Structural evaluations of the SOGS-RA employing both Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) have yielded valuable insights into its latent organization.

Unidimensional Model

The original authors (Winters et al., 1993) conducted principal components analysis and found that a dominant single-factor solution accounted for a substantial proportion of the total variance (eigenvalue > 3.5, accounting for 35% to 42% of common variance across studies). All 12 items loaded significantly onto this general “Gambling Severity” factor, with standardized factor loadings ranging from .42 to .74.

Alternative Multidimensional Findings

While the unidimensional conceptualization supports calculating a single summary score for clinical triage, subsequent CFA investigations (e.g., Stinchfield, 2002; Edgren et al., 2016) have occasionally highlighted a plausible two-factor oblique model, decomposing the items into:

1. Intrapersonal Loss of Control and Affective Preoccupation: Items 1, 4, 6, 7, and 8.
2. Interpersonal Conflicts and Maladaptive Consequences: Items 2, 3, 5, 9, 10, 11, and 12.

In CFA evaluations, both the unidimensional model with correlated error residuals between closely phrased interpersonal items (such as Item 3 and Item 9) and the two-factor oblique model demonstrate adequate goodness-of-fit metrics: Comparative Fit Index (CFI) ≥ .94, Tucker-Lewis Index (TLI) ≥ .92, and Root Mean Square Error of Approximation (RMSEA) ≤ .055. Because the correlation between the two oblique factors routinely exceeds r = .80, researchers and clinicians overwhelmingly retain the unidimensional operational scoring method.

10. Instrument / Measurement Tool

The structural characteristics, administration protocols, and scoring algorithms for the SOGS-RA are detailed below:

• Instrument Type: Standardized self-report psychometric screening questionnaire / paper-and-pencil or digital behavioral assessment.
• Target Population: Adolescents and youth aged approximately 12 to 18 years (grades 7 through 12).
• Administration Time: Approximately 3 to 5 minutes.
• Initial Gating Question: “Do you gamble? YES / NO (If no, STOP here)”. Respondents endorsing “NO” are categorized as non-gamblers (Score = 0) and bypass the remaining items.
• Recall Period: Past 12 months (“In the past 12 months…”).
• Item Count: 12 substantive items.
• Response Scale: Respondents evaluate the frequency of behaviors using a 4-point response continuum:
  • Every time
  • Most of the time
  • Some of the time
  • Never
• Scoring and Recoding Rules:
  • The SOGS-RA uses a dichotomous scoring approach to convert frequency responses into clinical screening points.
  • Standard Dichotomization Rule: For each item, an endorsement of “Some of the time”, “Most of the time”, or “Every time” receives 1 point. A response of “Never” receives 0 points. (Note: In the original Winters et al., 1993 coding manual, certain items assessing severe antisocial acts or chasing can be alternatively weighted based on stricter threshold frequencies, but the standard universal convention across epidemiologic studies assigns 1 point for any affirmative non-zero endorsement).
  • Total Score Range: 0 to 12 points.
• Diagnostic Classification and Cutoff Thresholds:
  • 0 to 1 point: No Problem / Non-Problem Gambler (Recreational betting behavior displaying no clinically significant dysregulation or psychosocial distress).
  • 2 to 3 points: At-Risk Gambler (Transitional or subclinical gambling involvement exhibiting early warning signs of loss of control or interpersonal conflict; targeted psychoeducational intervention recommended).
  • 4 or more points (4–12): Problem Gambler / Probable Pathological Gambler (Severe behavioral addiction pattern characterized by functional impairment, chasing losses, interpersonal strife, and academic or legal consequences; clinical comprehensive assessment and specialized intervention indicated).

11. Permissions & Fee and Test Year

• Initial Publication Year: 1993 (Winters, Stinchfield, & Fulkerson, 1993). Expanded statewide psychometric validation published in 1995 (Winters, Stinchfield, & Kim, 1995).
• Copyright and Intellectual Property: The SOGS-RA was developed with funding from state and academic research grants and published in peer-reviewed scientific literature. The instrument is considered an open-access public domain research instrument for non-commercial educational, clinical, and scientific investigations.
• Usage Fees: Free of charge for research, educational, and public health assessment applications.
• Permission Requirements: Formal commercial deployment, integration into proprietary clinical software, or reproduction in commercial test batteries requires written permission and citation of the original authors (Ken C. Winters and Randy D. Stinchfield). In standard academic and screening contexts, appropriate citation of the 1993 validation article satisfies ethical and scholarly attribution requirements.

12. References

• Blaszczynski, A., & Nower, L. (2002). A pathways model of problem and pathological gambling. Addiction, 97(5), 487–499. https://doi.org/10.1046/j.1360-0443.2002.00015.x
• Derevensky, J. L., & Gupta, R. (2000). Youth gambling problems: A review of the empirical literature and a conceptual model. Journal of Gambling Studies, 16(2–3), 315–349. https://doi.org/10.1023/A:1009425316977
• Edgren, R., Castrén, S., Mäkelä, M., Pallesen, S., & Salonen, A. H. (2016). Reliability and validity of the South Oaks Gambling Screen-Revised for Adolescents (SOGS-RA) among Finnish high school students. Journal of Gambling Issues, 33, 107–123. https://doi.org/10.4309/jgi.2016.33.7
• Fisher, S. (2000). Developing the DSM-IV-DSM-IV-MR-J criteria to identify adolescent problem gambling in non-clinical populations. Journal of Gambling Studies, 16(2–3), 253–273. https://doi.org/10.1023/A:1009473232002
• Jessor, R., & Jessor, S. L. (1977). Problem behavior and psychosocial development: A longitudinal study of youth. Academic Press.
• Kuentzel, J. G., & Henderson, J. M. (2002). Preliminary evidence of social desirability effects on South Oaks Gambling Screen (SOGS) scores. Poster presented at the 2002 meeting of the Center for Responsible Gaming, Las Vegas, NV.
• Lesieur, H. R., & Blume, S. B. (1987). The South Oaks Gambling Screen (SOGS): A new instrument for the identification of pathological gamblers. American Journal of Psychiatry, 144(9), 1184–1188. https://doi.org/10.1176/ajp.144.9.1184
• Macdonald, J., Turner, N. E., & Somerset, M. (2008). Life skills, mathematical reasoning and critical thinking: Curriculum for the prevention of problem gambling. Final Report to the Ontario Problem Gambling Research Centre. Centre for Addiction and Mental Health. https://pubmed.ncbi.nlm.nih.gov/18095146/
• Steinberg, L. (2008). A social neuroscience perspective on adolescent risk-taking. Developmental Review, 28(1), 78–106. https://doi.org/10.1016/j.dr.2007.08.002
• Stinchfield, R. (2000). Gambling and correlates of gambling among Minnesota public school students. Journal of Gambling Studies, 16(2–3), 153–173. https://doi.org/10.1023/A:1009439818818
• Stinchfield, R. (2002). Youth gambling: How prevalent is it and what are the risk factors? The Prevention Researcher, 9(2), 3–7.
• Winters, K. C., Stinchfield, R. D., & Fulkerson, J. (1993). Toward the development of an adolescent gambling severity scale. Journal of Gambling Studies, 9(1), 63–84. https://doi.org/10.1007/BF01019925
• Winters, K. C., Stinchfield, R. D., & Kim, L. G. (1995). Monitoring adolescent gambling in Minnesota. Journal of Gambling Studies, 11(2), 165–182. https://doi.org/10.1007/BF02107112

13. Items of the Scale