Monitoring the Future Survey/30 Day Use

Abstract

The Monitoring the Future Survey/30 Day Use (often designated in psychometric and epidemiologic research as the MTF 30-Day Substance Use Battery) is an internationally recognized epidemiological and psychometric measurement instrument designed to capture past-month frequency, prevalence, and severity of licit and illicit substance consumption. Developed at the Institute for Social Research at the University of Michigan under sustained sponsorship from the National Institute on Drug Abuse (NIDA), the 30-day assessment serves as the gold-standard index of active, ongoing substance involvement among adolescents and emerging adults. The battery evaluates twelve core pharmacological classes: combustible tobacco cigarettes, smokeless tobacco, alcohol use frequency, subjective alcohol intoxication (drunkenness), cannabis/hashish, powder cocaine, crack cocaine, volatile inhalants, non-prescribed amphetamines, non-prescribed sedatives, non-prescribed tranquilizers, and non-prescribed prescription opioids/narcotics. Administered via structured self-report utilizing graded categorical frequency response frameworks (ranging from zero occasions to daily/sub-daily quantitative consumption), the instrument possesses well-established psychometric parameters. Confirmatory factor analytic investigations consistently demonstrate high construct validity, characterized by a bifactor architecture featuring a robust general substance use liability factor (ω_h > .80) alongside substance-specific residual clusters (e.g., licit/normative substances, illicit street narcotics, and non-medical prescription drugs). Test-retest reliability coefficients over brief intervals range from .78 to .94 across diverse substance categories, while longitudinal cross-wave stability demonstrates predictive fidelity for subsequent substance use disorders classified under DSM-IV and DSM-5 criteria. This article provides an exhaustive psychometric, theoretical, and empirical analysis of the MTF 30-day use battery, detailing its measurement model, behavioral epidemiology utility, and psychometric validation across five decades of continuous national surveillance.

Keywords

Monitoring the Future, MTF-30, adolescent substance use, 30-day prevalence, epidemiological psychometrics, drug surveillance, psychostimulants, prescription drug misuse, binge drinking, measurement invariance.

Authors

The Monitoring the Future instrument battery was conceptualized, developed, and standardized by principal investigators at the Survey Research Center within the Institute for Social Research (ISR), University of Michigan, Ann Arbor, Michigan, United States:

• Lloyd D. Johnston, Ph.D. — Distinguished Research Scientist Emeritus, Survey Research Center, Institute for Social Research, University of Michigan. Contact: [email protected].
• Patrick M. O’Malley, Ph.D. — Research Professor Emeritus, Survey Research Center, Institute for Social Research, University of Michigan.
• Jerald G. Bachman, Ph.D. — Distinguished Research Scientist Emeritus, Survey Research Center, Institute for Social Research, University of Michigan.
• John E. Schulenberg, Ph.D. — Research Professor, Survey Research Center, Institute for Social Research, and Professor of Psychology, University of Michigan.
• Richard A. Miech, Ph.D. — Research Professor, Survey Research Center, Institute for Social Research, University of Michigan (Current Principal Investigator).

Purpose

The primary diagnostic and epidemiological purpose of the Monitoring the Future 30-Day Use scale is to quantify proximate, current psychoactive substance involvement among secondary school students, college populations, and non-college young adults. While lifetime prevalence measures capture historical experimentation or exploratory behavior that may have ceased years prior, and 12-month prevalence encompasses seasonal or episodic experimentation, 30-day use isolates active, ongoing consumption. In both psychometric theory and behavioral pharmacology, past-month consumption provides the most sensitive empirical indicator of active habituation, continuous pharmacological reinforcement, and proximal clinical risk for chemical dependence, adverse neurodevelopmental sequelae, academic dysfunction, and acute behavioral impairment.

From a public health and behavioral surveillance standpoint, the MTF-30 addresses several distinct scientific needs:

• Temporal Sensitivity to Epidemiological Inflections: Because the 30-day window is temporally proximate to the testing moment, recall bias and cognitive decay are markedly reduced compared to annual or lifetime recall frameworks. This temporal precision enables public health officials, prevention scientists, and epidemiologists to identify emerging drug epidemics, regulatory policy impacts (such as changes in minimum legal purchasing ages or clean indoor air laws), and secular shifts in national consumption patterns.
• Distinction Between Frequency and Dosage: The instrument assesses not merely binary consumption (use versus non-use), but categorical gradations of exposure. For high-volume substances such as tobacco, alcohol, and cannabis, the measure differentiates light or experimental past-month use from habitual, daily, or multiple-daily use, providing a nuanced behavioral gradient that maps cleanly onto subclinical and clinical diagnostic criteria.
• Clinical and Preventive Stratification: In clinical triage and school-based intervention contexts, the MTF-30 serves as a standardized screener to stratify youth across developmental trajectories of problem behavior. An affirmative response to any illicit substance or frequent combustible/alcohol exposure in the preceding 30 days triggers targeted secondary prevention or formal clinical assessment for substance use disorder (SUD) as codified in the American Psychiatric Association’s DSM-5-TR.
• Longitudinal Etiological Research: Within developmental psychopathology, the MTF-30 provides a standardized metric that can be tracked across longitudinal panel cohorts from age 14 into middle adulthood, permitting precise trajectory modeling (e.g., latent growth curve modeling and growth mixture modeling) to map the escalation, maturation out, or chronic persistence of chemical dependency.

Psychological Construct

The overarching psychological construct measured by the MTF 30-day battery is proximal substance involvement, defined as the frequency, behavioral density, and diversity of psychoactive xenobiotic consumption within an immediate developmental and environmental window. Rather than treating substance use as a unitary or monolithic construct, the battery evaluates multiple distinct, interrelated behavioral and pharmacological dimensions.

1. Combustible and Smokeless Nicotine Consumption

This subscale captures compulsive nicotine delivery via combustible leaf tobacco (cigarettes) and oral smokeless formulations (chewing tobacco, snuff). The construct reflects physiological nicotine dependence, daily autonomic regulation, and sensory-motor behavioral routines. Items assess consumption along a fine-grained continuous-categorical spectrum ranging from zero exposure to intensive multiple-pack-per-day intake. Psychometrically, daily combustible smoking within the past 30 days is historically one of the strongest indicators of established physiological addiction and is linked to the developmental internalization of externalizing tendencies.

2. Alcohol Ingestion Frequency and Subjective Intoxication

The battery operationalizes alcohol involvement through two distinct yet highly collinear dimensions: behavioral frequency of consumption (drinking more than a few sips) and acute behavioral impairment (frequency of being drunk or “very high” from alcohol). This structural division is critical: while frequency reflects normative social ingestion or social drinking patterns in emerging adult cohorts, the drunkenness indicator assesses acute neuropharmacological intoxication, loss of behavioral control, and high-risk sensation seeking. The dissociation between mere use and heavy episodic intoxication provides critical predictive validity for alcohol-related consequences such as blackouts, motor vehicle collisions, and interpersonal aggression.

3. Cannabinoid Involvement

Cannabis involvement is operationalized via past-month occasions of consuming marijuana (grass, weed, pot) or concentrated hashish. This dimension captures engagement with exogenous tetrahydrocannabinol (THC) and related cannabinoids. In the expanded epidemiological administration, this is supplemented with quantitative estimates of daily or sub-daily joint/unit equivalents. The construct taps into both recreational sensation seeking and coping-motivated self-medication for affective distress, social anxiety, or sleep disturbances.

4. Illicit Psychostimulant, Dissociative, and Inhalant Consumption

This dimension encompasses low-prevalence, high-severity illicit substances including cocaine powder, crack cocaine, and volatile hydrocarbon inhalants (glues, aerosols, solvents). Engaging in the non-medical inhalation or insufflation of these compounds during a 30-day period signifies a severe departure from developmental norms, indicating marked deviance proneness, neurobehavioral disinhibition, elevated psychiatric comorbidity, and immersion in deviant peer networks.

5. Non-Medical Prescription Drug Misuse (Iatrogenic and Illicit Sourcing)

A central psychometric contribution of the MTF architecture is the explicit differentiation of non-prescribed pharmaceutical classes: amphetamines/prescription stimulants (e.g., Adderall, Ritalin, Dexedrine), sedatives/barbiturates, benzodiazepine tranquilizers, and non-heroin prescription narcotics/opioids (e.g., oxycodone, hydrocodone, codeine). The construct measures instrumental, self-directed consumption without medical authorization (“on your own — without a doctor’s orders”). This taps into modern paradigms of functional chemical coping, neuroenhancement (in academic and occupational environments), and secondary physiological physical dependence emerging outside of illicit street-drug supply chains.

Theoretical Framework

The conceptual framework of the Monitoring the Future 30-day measurement architecture is grounded in several landmark paradigms within developmental psychology, sociological criminology, and behavioral pharmacology.

Problem Behavior Theory (PBT)

Formulated by Richard Jessor and Shirley L. Jessor (1977), Problem Behavior Theory posits that adolescent substance use does not occur in isolation but represents one facet of a broader, systematically organized constellation of transition-marking or unconventional behaviors. According to PBT, adolescent deviance is governed by three intersecting systems: the Personality System (values, expectations, beliefs, and personal controls), the Perceived Environment System (parental models, peer pressures, and perceived approvals), and the Behavior System (conventional vs. unconventional behavioral enactments). The MTF 30-day scale serves as the operational metric for the unconventional Behavior System; frequent past-month engagement in licit and illicit drug use reflects the behavioral manifestation of low conventionality, high peer-model approval for deviance, and weakened structural controls.

The Social Development Model

The Social Development Model (SDM), conceptualized by J. David Hawkins and Richard F. Catalano, synthesizes social control theory, social learning theory, and differential association theory. SDM asserts that substance use emerges when a youth’s bonding to prosocial institutions (family, school, conventional community) is disrupted or when the youth forms strong social and psychological attachments to antisocial peers who reinforce drug-taking behaviors. Within this model, the 30-day use battery functions as an index of active antisocial reinforcement. Because the items capture behavior within the last month, affirmative reports signal current, active socialization within substance-using peer networks, indicating that prosocial bonding mechanisms have been superseded by deviant socialization processes.

The Gateway Hypothesis and Developmental Staging

The classical sequential model of drug involvement, pioneered by Denise Kandel, outlines an orderly, progressive developmental hierarchy of substance experimentation: beginning with legal, socially accessible substances (beer, wine, cigarettes), progressing to distilled spirits and cannabis, and culminating in illicit street drugs and the non-medical misuse of prescription pharmaceuticals. The MTF 30-day scale reflects this developmental ordering. By assessing identical 30-day temporal windows across the full spectrum of pharmacological substances, the scale enables Guttman-like simplex scaling, revealing where an individual resides along the continuum of developmental drug involvement.

Behavioral Economics and Reinforcement Pathology

From a behavioral economic perspective, active substance use reflects a dynamic interaction between the commodity’s reinforcing efficacy, behavioral acquisition costs (price, accessibility, legal sanctions), and the availability of alternative prosocial reinforcers (academic success, athletics, family relationships). The 30-day window captures current reinforcement allocation: individuals reporting multiple occasions across diverse drug categories are exhibiting high behavioral economic demand elasticity for psychoactive reinforcement, characterized by excessive delay discounting where immediate pharmacological reward systematically devalues distal, long-term health and academic goals.

Validity

The validity of the Monitoring the Future 30-day self-report methodology has been subjected to extensive empirical verification across five decades of psychometric evaluation, cross-validation studies, and methodological experimentation conducted by the ISR team and external research bodies.

Construct and Factorial Validity

Construct validity is substantiated through structural equation modeling (SEM) and latent variable measurement models across thousands of adolescent cohorts. Factor analytic investigations reveal that MTF 30-day items demonstrate strong factor loadings on an overarching latent dimension of general substance use (λ = .55 to .88), with secondary residual factors representing specific chemical classes (e.g., CNS depressants, stimulants, cannabinoids). The observed covariance patterns align with theoretical predictions: substances sharing pharmacological classifications (e.g., sedatives and tranquilizers; powder cocaine and crack) exhibit high specific correlations, whereas disparate substances (e.g., inhalants and smokeless tobacco) demonstrate lower shared specific variance, validating the multi-dimensional structure of the instrument.

Convergent and Criterion Validity

Extensive validity studies have examined the concordance between self-reported 30-day drug consumption on the MTF questionnaire and objective external benchmarks:

• Biochemical Validation: Sub-sample validation studies utilizing carbon monoxide (CO) expired-air analysis, salivary cotinine, and urinary drug screens have demonstrated high concordance with MTF self-report. Specifically, Johnston and colleagues established that adolescent self-reports of 30-day cigarette and cannabis use correlate at r = .82 to .89 with biological markers when anonymity and confidentiality are rigorously protected.
• Collateral and Peer Reports: Studies employing peer-nomination methods and multi-informant designs show significant correlations (r = .60 to .75) between self-reported past-month substance use on the MTF battery and independent reports provided by close friends and siblings regarding the target individual’s drug consumption.
• Cross-Validation with Emergency Department and Overdose Data: Secular trends observed in MTF 30-day prevalence rates for opioids, cocaine, and amphetamines closely mirror population-level emergency department presentations captured by the Drug Abuse Warning Network (DAWN) and mortality statistics compiled by the National Center for Health Statistics (NCHS), establishing macroscopic criterion-related validity.

Predictive and Longitudinal Validity

The predictive utility of the MTF 30-day scale is documented across longitudinal follow-up panels that track respondents from age 18 to ages 35, 50, and 65. Bachman et al. demonstrated that affirmative 30-day use of cannabis or illicit prescription drugs at age 18 predicts an increased odds ratio (OR = 3.8 to 6.2) for meeting DSM-IV and DSM-5 criteria for Substance Use Disorder at age 28, even after controlling for baseline demographic variables, socioeconomic status, academic achievement, and parental psychopathology. Furthermore, past-month heavy drinking (drunkenness) at high school senior baseline strongly predicts subsequent young-adult motor vehicle collisions, physical injuries, employment instability, and marital dissolution.

Discriminant Validity

The MTF-30 effectively discriminates between normative adolescent exploration and clinically severe chemical dependence. In comparative analyses involving clinical adolescent populations in residential substance abuse treatment versus general school samples, the MTF 30-day severity index differentiated the two groups with extreme precision (area under the receiver operating characteristic curve [AUC] > .94). Within the normative school population, the measure demonstrates clear discriminant validity against measures of non-externalizing emotional distress, such as pure internalizing depressive and anxious states, showing substantially higher correlations with externalizing conduct problems and antisocial behavior (r = .52 to .68) than with unipolar depression (r = .14 to .22).

Reliability

The psychometric reliability of the MTF 30-day measurement battery has been confirmed through classical test theory metrics, longitudinal panel stability analysis, and item response theory (IRT) precision estimations.

Internal Consistency

When evaluated as an aggregated composite score of poly-substance involvement or illicit drug severity, the internal consistency of the 30-day battery is remarkably high. Across nationally representative samples of 8th, 10th, and 12th graders, Cronbach’s alpha coefficients for the overall substance involvement composite consistently range between α = .82 and α = .91. Subscale internal consistency estimates are similarly robust:

• Alcohol Involvement Index (Frequency of drinking + Frequency of drunkenness): Spearman-Brown split-half coefficient = .86; Cronbach’s α = .87.
• Illicit Drug Composite (Excluding alcohol and tobacco): Cronbach’s α = .84 to .89 across survey waves.
• Prescription Drug Misuse Index (Amphetamines, sedatives, tranquilizers, narcotics): Cronbach’s α = .81 to .86.

Test-Retest Stability

Direct short-term test-retest reliability studies conducted over 2- to 4-week intervals demonstrate exceptional response stability. Pearson product-moment correlations and intraclass correlation coefficients (ICC) across specific substances confirm high temporal reproducibility:

• Past-30-day cigarette frequency: r = .91, ICC = .90.
• Past-30-day alcohol occasions: r = .84, ICC = .83.
• Past-30-day alcohol drunkenness: r = .82, ICC = .81.
• Past-30-day cannabis occasions: r = .88, ICC = .87.
• Past-30-day illicit drug use (composite): r = .79 to .85.

Longitudinal stability coefficients over 1-year and 2-year spans reveal substantial autoregressive stability in late adolescence (paths from Grade 10 to Grade 12 ranging from β = .54 to .71 for cannabis and tobacco), reflecting the consolidation of behavioral habits, while still capturing developmental change.

Consistency Across Parallel Forms and Item Contexts

Because the Monitoring the Future study employs multiple rotating questionnaire forms (six distinct forms among 12th-grade cohorts) to maximize content coverage, core 30-day items are embedded across varying form positions and surrounding questionnaire contexts. Methodological analyses by Johnston and colleagues demonstrate that variation in surrounding item context produces negligible variance shifts (< 1.5% absolute prevalence deviation), confirming the structural robustness and contextual invariance of the 30-day operational indicators.

Factor Analysis

Extensive exploratory factor analyses (EFA) and confirmatory factor analyses (CFA) conducted on the MTF 30-day substance use battery support a hierarchical or bifactor latent structure rather than a strictly unidimensional construct.

Exploratory Factor Analysis (EFA)

Early psychometric evaluations of the correlation matrix of 30-day substance use indicators among secondary school students regularly yielded a four-factor solution using principal axis factoring with promax or varimax rotation:

• Factor 1: Socially Permitted / Licit Substances — Dominated by cigarette smoking frequency, smokeless tobacco use, alcohol drinking frequency, and alcohol intoxication (eigenvalues > 3.5, accounting for 35–42% of common variance; factor loadings: .62 to .84).
• Factor 2: Cannabis Involvement — Characterized by past-month marijuana frequency and quantity of joints smoked per day (loadings > .80). In many developmental phases, cannabis cross-loads moderately onto the licit factor due to its normative prevalence among older adolescents.
• Factor 3: Severe Illicit Street Drugs — Comprising powder cocaine, crack cocaine, inhalants, and hallucinogens/LSD (loadings: .58 to .79).
• Factor 4: Non-Medical Prescription Drug Misuse — Marked by high loadings on non-prescribed amphetamines, sedatives, tranquilizers, and other narcotics (loadings: .61 to .83).

Confirmatory Factor Analysis (CFA) and Model Fit

Contemporary psychometric structural evaluations test competing measurement models: a strictly unidimensional model, a correlated four-factor model, and a bifactor model. Across large-scale representative samples (e.g., N > 15,000 per cohort), the bifactor structural model demonstrates superior empirical fit to the data:

In the bifactor model, the general substance liability factor (General G) accounts for over 72% of the common variance (Explained Common Variance [ECV] = .72), with hierarchical omega (ω_h) exceeding .84. Standardized factor loadings on the general factor are robust across all items: combustible cigarettes (λ = .58), alcohol frequency (λ = .64), drunkenness (λ = .72), cannabis (λ = .78), cocaine (λ = .75), inhalants (λ = .52), amphetamines (λ = .70), and narcotics (λ = .76). These results demonstrate that while specific substance groupings capture pharmacologically distinct variance, a single generalized behavioral liability toward substance engagement accounts for the predominant variance in the 30-day battery.

Measurement Invariance

Multi-group CFA testing confirms that the MTF-30 measurement model achieves full metric (weak) and scalar (strong) invariance across biological sex (male vs. female), race/ethnicity (White, Black, Hispanic), and high school grade level (8th, 10th, and 12th grades). Changes in fit indices between unconstrained and constrained multi-group models remain well within standard psychometric thresholds (ΔCFI < .010, ΔRMSEA < .005), confirming that observed differences in 30-day prevalence rates across demographic subgroups represent true behavioral differences rather than differential item functioning (DIF) or measurement artifact.

Instrument / Measurement Tool

• Instrument Name: Monitoring the Future Survey — 30-Day Substance Use Scale (MTF-30)
• Target Population: Adolescents (Grades 8, 10, 12), college students, and emerging/middle adults (ages 14–65)
• Administration Format: Standardized self-administered pencil-and-paper questionnaire (scannable optical forms) or secure web-based computer-assisted self-interviewing (CASI)
• Completion Time: Approximately 3 to 6 minutes for the dedicated 30-day substance use module (embedded within the broader 45-minute omnibus MTF survey)
• Item Count: 12 core standardized primary items assessing past-month consumption across distinct drug categories
• Response Scale Structure: Item-specific categorical frequency scales
  • Item 1 (Cigarettes): 7-point categorical scale (1 = Not at all, 2 = Less than one cigarette per day, 3 = One to five cigarettes per day, 4 = About one-half pack per day, 5 = About one pack per day, 6 = About one and one-half packs per day, 7 = Two packs or more per day)
  • Item 2 (Smokeless Tobacco): 6-point categorical scale (1 = Not at all, 2 = Once or twice, 3 = Once to twice per week, 4 = Three to five times per week, 5 = About once a day, 6 = More than once a day)
  • Items 3 through 12 (Alcohol, Drunkenness, Cannabis, Inhalants, Illicit Drugs, Prescription Misuse): 7-point ordinal occasion scale (1 = 0 Occasions, 2 = 1-2 Occasions, 3 = 3-5 Occasions, 4 = 6-9 Occasions, 5 = 10-19 Occasions, 6 = 20-39 Occasions, 7 = 40 or more Occasions)
• Scoring Rules:
  • 30-Day Prevalence (Dichotomous): Items are individually dichotomized into non-use (Category 1 = 0) versus any affirmative use (Categories 2 through 6/7 = 1) to derive population-level prevalence estimates for each substance class.
  • Substance Severity / Frequency Indices: Ordinal item scores (1 to 7) may be examined directly as continuous or pseudo-continuous indicators of consumption density.
  • Polysubstance Involvement Index: Calculated by summing the number of distinct pharmacological classes for which the respondent endorsed past-30-day use (> 0 occasions), producing a total polysubstance diversity score ranging from 0 to 12.
  • Composite Substance Use Severity Score: Standardized z-score conversion followed by summation across items, yielding an overarching latent severity metric.
• Reverse Scoring: None. Higher numerical values correspond directly to higher frequency, higher volume, or greater behavioral severity of substance use.

Permissions & Fee and Test Year

The Monitoring the Future study was established in 1975 by Lloyd D. Johnston, Jerald G. Bachman, and Patrick M. O’Malley at the University of Michigan. The research program has been supported continuously by major research grant funding from the National Institute on Drug Abuse (NIDA), a branch of the United States National Institutes of Health (NIH) (Grant Numbers R01DA001411 and R01DA016518).

Because the survey was developed under United States federal research sponsorship, the core question stems and item formulations reside in the public domain for non-commercial academic, scientific, and clinical research purposes. Researchers, school administrators, and behavioral health professionals may utilize the questions without royalty fees or licensing costs. However, formal citation and attribution to the University of Michigan Institute for Social Research and the Monitoring the Future project are required in all publications and presentations. Commercial repackaging, incorporation into proprietary digital testing software, or large-scale redistribution requires express institutional authorization from:

Monitoring the Future Project
Institute for Social Research, University of Michigan
426 Thompson Street, Ann Arbor, MI 48104-2321
Email: [email protected] | Phone: (734) 764-8354

References

Bachman, J. G., O’Malley, P. M., Schulenberg, J. E., Johnston, L. D., Freedman-Doan, P., & Messersmith, E. E. (2008). The education-drug use connection: How successes and failures in school relate to adolescent smoking, drinking, drug use, and delinquency. Lawrence Erlbaum Associates / Psychology Press. https://doi.org/10.4324/9780203809839

Hawkins, J. D., Catalano, R. F., & Miller, J. Y. (1992). Risk and protective factors for alcohol and other drug problems in adolescence and early adulthood: Implications for substance abuse prevention. Psychological Bulletin, 112(1), 64–105. https://doi.org/10.1037/0033-2909.112.1.64

Jessor, R., & Jessor, S. L. (1977). Problem behavior and psychosocial development: A longitudinal study of youth. Academic Press. https://doi.org/10.1016/C2013-0-10892-0

Johnston, L. D., O’Malley, P. M., & Bachman, J. G. (2001). Monitoring the Future national survey results on drug use, 1975–2000: Volume I, Secondary school students (NIH Publication No. 01-4924). National Institute on Drug Abuse. https://eric.ed.gov/?id=ED460144

Johnston, L. D., O’Malley, P. M., Bachman, J. G., & Schulenberg, J. E. (2012). Monitoring the Future national survey results on drug use, 1975–2011: Volume II, College students and adults ages 19–50. Institute for Social Research, The University of Michigan. https://files.eric.ed.gov/fulltext/ED536294.pdf

Kandel, D. B. (2002). Stages and pathways of drug involvement: Examining the gateway hypothesis. Cambridge University Press. https://doi.org/10.1017/CBO9780511499777

Miech, R. A., Johnston, L. D., O’Malley, P. M., Bachman, J. G., Schulenberg, J. E., & Patrick, M. E. (2023). Monitoring the Future national survey results on drug use, 1975–2022: Secondary school students. Institute for Social Research, The University of Michigan. https://doi.org/10.3998/2027.42/179261

Oetting, E. R., & Donnermeyer, J. F. (1998). Primary socialization theory: The etiology of drug use and deviance. I. Substance Use & Misuse, 33(4), 995–1026. https://doi.org/10.3109/10826089809056252

Patrick, M. E., Schulenberg, J. E., Martz, M. E., Maggs, J. L., O’Malley, P. M., & Johnston, L. D. (2017). Extreme binge drinking among 12th-grade students in the United States: Prevalence and predictors. JAMA Pediatrics, 171(1), 37–43. https://doi.org/10.1001/jamapediatrics.2016.3278

Schulenberg, J. E., Merline, A. C., Johnston, L. D., O’Malley, P. M., Bachman, J. G., & Laetz, V. B. (2005). Trajectories of marijuana use during the transition to adulthood: The big picture based on national longitudinal data. Journal of Drug Issues, 35(2), 255–280. https://doi.org/10.1177/002204260503500203

13. Items of the Scale (Questionnaire)