Attitude, Belief and Behavior Survey

Abstract

The Attitude, Belief and Behavior Survey (incorporating the Attitudes toward Deviance Scale [ATDS]) is a multidimensional psychometric and sociological measurement instrument developed by Dr. Virginia L. Rothwell (2009) at Virginia Polytechnic Institute and State University. Designed to investigate the complex psychosocial etiology of deviant conduct among emerging adults and collegiate populations, the instrument operationalizes core tenets of criminological theory, including differential association theory, social learning theory, social bond theory, classic strain theory, and the sociology of individualism and scientific rationalism. The comprehensive survey battery consists of 59 items systematically structured across nine thematic sections: informed consent and eligibility verification, self-concept and risk orientation, educational and occupational goal-strain discrepancies, peer deviant behavior, normative moral evaluations of deviance (the ATDS core), past-year self-reported deviance across 11 behavioral domains, conventional activity involvement, institutional social bonds and ideological beliefs, and sociodemographic background characteristics.

Constructed in strict adherence to Dillman’s (2000) Tailored Design Method, the instrument employs diverse psychometric formats, including 4-point moral wrongness scales, 4-point past-year behavioral frequency metrics, 5-point peer engagement proportions, and 4-point Likert agreement matrices. Empirical psychometric evaluations demonstrate high structural fidelity, with internal consistency coefficients (Cronbach’s alpha) ranging from .78 to .89 across the attitudinal and behavioral deviance subscales. Factor-analytic models confirm a coherent multidimensional construct distinguishing between minor, consensual deviance (e.g., academic dishonesty, digital piracy, cannabis use, binge drinking) and serious, predatory delinquency (e.g., property destruction, theft, burglary, assault, illicit hard drug consumption). The survey provides researchers, sociologists, developmental psychologists, and collegiate administrators with a robust, empirically validated tool for measuring the cognitive definitions, social reinforcements, and behavioral patterns that govern non-normative conduct.

Keywords

Attitudes toward Deviance Scale, criminological assessment, social learning theory, social bond theory, differential association, deviance measurement, collegiate substance use, academic dishonesty, digital piracy, self-reported delinquency, risk-taking propensity, strain theory

Authors

The Attitude, Belief and Behavior Survey was designed, standardized, and validated by Virginia L. Rothwell, Ph.D., in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Sociology at the Department of Sociology, Virginia Polytechnic Institute and State University (Virginia Tech), Blacksburg, Virginia (Rothwell, 2009).

The development and theoretical refinement of the survey battery were supervised by an academic doctoral advisory committee comprising distinguished criminologists and sociologists:

• James E. Hawdon, Ph.D. (Committee Co-Chair), Professor of Sociology and Director of the Center for Peace Studies and Violence Prevention at Virginia Polytechnic Institute and State University.
• Donald J. Shoemaker, Ph.D. (Committee Co-Chair), Professor Emeritus of Sociology at Virginia Polytechnic Institute and State University, internationally recognized authority on juvenile delinquency and comparative criminology.
• Carol A. Bailey, Ph.D. (Committee Member), Associate Professor of Sociology, specialist in research methods and qualitative/quantitative sociological methodology.
• Anastasia S. Vogt Yuan, Ph.D. (Committee Member), Associate Professor of Sociology, specialist in social psychology, mental health, and life-course sociology.

Purpose

The primary purpose of the Attitude, Belief and Behavior Survey is to systematically examine the social, psychological, and normative mechanisms that facilitate, sustain, or inhibit deviant behavior among emerging adults, with particular diagnostic relevance to university and collegiate cohorts. Historically, criminological literature has exhibited persistent theoretical tensions concerning the causal sequencing among cognitive definitions (attitudes), peer networks (differential associations), institutional ties (social bonds), individualistic personal identities, and actual deviant conduct. This instrument was explicitly engineered to address these theoretical controversies by providing simultaneous, methodologically aligned operationalizations of multiple competing and complementary criminological models within a single, highly integrated protocol.

From an applied and institutional perspective, the scale serves several critical research and administrative functions:

• Quantifying Moral Thresholds of Deviance: By administering the 11-item Attitudes toward Deviance Scale (ATDS) core, researchers can establish a respondent’s moral neutralization or ethical tolerance across a spectrum of behaviors ranging from victimless non-violent infractions to severe predatory violence.
• Evaluating Academic Integrity Dynamics: The instrument assesses both the moral acceptability and behavioral prevalence of academic cheating, alongside institutional bonds with faculty, providing universities with data regarding institutional climate, honor code efficacy, and academic dishonesty trends.
• Epidemiological Substance Monitoring: The battery categorizes substance involvement across five distinct pharmacological categories—binge drinking, cannabis, prescription/synthetic stimulants (cocaine, methamphetamine), designer club drugs (Ecstasy/MDMA, GHB), and narcotics/opiates (opium, heroin)—measuring peer exposure, personal moral attitudes, and annualized individual consumption rates.
• Cyber-Criminology and Intellectual Property Infringement: The survey captures digital piracy (downloading copyrighted media without compensation), enabling researchers to study how modern technological non-compliance integrates with traditional forms of deviant conduct.
• Evaluating Protective Factors and Prosocial Buffers: By capturing religious adherence, athletic involvement, philanthropic and civic extracurriculars, academic study investment, and familial and parental attachments, the instrument identifies which conventional routines insulate young adults against delinquent peer socialization.

Psychological Construct

The Attitude, Belief and Behavior Survey operationalizes a multifaceted conceptual architecture centered on the psychological construct of cognitive definitions of deviance and its structural interrelations with personality, perceived opportunity structures, and behavioral outcomes. In sociological and psychological criminology, “definitions” refer to the moral orientations, rationalizations, and normative attitudes that an individual assigns to a specific behavior, categorizing it as right or wrong, permissible or impermissible, justified or reprehensible.

1. Core Attitudinal Dimension: Moral Evaluation of Deviant Conduct (ATDS)

The focal psychological construct of the survey is the degree of moral disapprobation or tolerance directed toward non-normative acts. The instrument assesses this across three primary behavioral domains:

• Substance Deviance: The rationalization and perceived wrongness of binge drinking, cannabis usage, stimulant consumption, designer club drug experimentation, and illicit opiate misuse. Items assess whether individuals view chemical consumption through a lens of personal freedom, situational harmlessness, or severe moral prohibition.
• Property and Economic Deviance: Moral thresholds concerning vandalism, larceny, domestic/commercial burglary, and unauthorized digital file-sharing. This dimension differentiates between utilitarian or predatory theft and digital intellectual property acquisition, which frequently undergoes cognitive neutralization.
• Interpersonal Violence and Institutional Transgression: Moral evaluations of direct physical assault and academic cheating, capturing an individual’s willingness to violate interpersonal safety and institutional meritocratic norms.

2. Individualism, Autonomy, and Sensation Seeking

The instrument incorporates psychological measures of the modern individualistic self-concept. Drawing on psychological conceptualizations of self-direction, the construct of individualism is operationalized as personal autonomy, preference for uniqueness, and detachment from collective behavioral mandates. Concurrently, sensation seeking is measured through the willingness to engage in risk-taking for immediate hedonic enjoyment, reflecting an underlying impulsivity and thrill-seeking temperament that lowers the barrier to deviant experimentation.

3. Differential Association and Social Learning Reinforcements

Deviance is not formed in a psychological vacuum; it is mediated through the perceived behaviors of proximal reference groups. The survey operationalizes peer deviance by quantifying the perceived proportion of closest friends who commit academic infractions, engage in heavy episodic drinking, or consume illicit substances. This dimension captures the observational modeling and social reinforcement parameters essential to social learning theories.

4. Structural Strain: Aspirations vs. Expectations Gap

The construct of structural or anomic strain is operationalized through the classical cognitive discrepancy paradigm: the absolute subjective value placed on occupational and educational success versus the realistic perceived probability of realizing those aspirations. A wide negative discrepancy represents perceived institutional blockage, which can foster psychological frustration and anomic adaptations.

5. Institutional Attachment, Commitment, and Conventional Beliefs

Rooted in social control constructs, this dimension evaluates the strength of an individual’s emotional and psychological bonds to conventional societal institutions. It captures cognitive attachment to primary authority figures (parents, academic instructors, law enforcement officers), academic commitment (effort invested in school, educational valuation), conventional involvement (time committed to studying, sports, volunteering, and religious liturgy), and moral belief in the legitimacy of conventional rules and secular modernity (e.g., trust in science to resolve existential crises).

Theoretical Framework

The survey synthesizes several foundational criminological and sociological theories into an overarching empirical framework. Rather than viewing these paradigms as mutually exclusive, the instrument operationalizes them as complementary layers of human behavioral regulation.

1. Akers’ Social Learning Theory and Sutherland’s Differential Association

The primary theoretical driver of the instrument is Ronald L. Akers’ Social Learning Theory (1998), which expanded upon Edwin Sutherland’s Differential Association Theory (1947). Sutherland posited that criminal behavior is learned through interaction with intimate personal groups via a process of communication. An individual becomes delinquent due to an excess of definitions favorable to the violation of law over definitions unfavorable to law violation.

Akers formalized this by integrating behavioral psychology principles—specifically differential reinforcement, vicarious imitation, and cognitive definitions. In Rothwell’s instrument, this theoretical loop is measured directly: Sections 4 (behavior of close friends) captures differential association and modeling; Section 5 (ATDS) captures cognitive definitions (evaluating whether acts are considered “not wrong at all” up to “very wrong”); and Section 6 records the behavioral outcome (self-reported deviance). Social learning theory hypothesizes that peer behavior directly predicts individual attitudes, which subsequently mediate actual behavioral commission.

2. Hirschi’s Social Control and Social Bond Theory

The second primary foundation is Travis Hirschi’s (1969) Social Bond Theory, articulated in Causes of Delinquency. Hirschi inverted the classical criminological inquiry: instead of asking “Why do people break the law?”, control theorists ask “Why do people conform?” The answer lies in the strength of an individual’s bond to society, composed of four distinct elements:

• Attachment: Emotional and psychological sensitivity to the opinions of conventional others (measured in the survey through respect for parents, teachers, and law enforcement).
• Commitment: The rational, cognitive investment in conventional lines of action (measured via the importance of getting a good education, striving for career success, and hard work in school).
• Involvement: Preoccupation with conventional activities that leave insufficient time for delinquent pursuits (measured via hours spent studying, playing athletics, attending religious services, and participating in clubs).
• Belief: Acknowledgment of the moral validity of conventional social rules and legal mandates (measured through belief in adult authority, respect for police honesty, and the impermissibility of parental disobedience).

3. Classic Anomie and Strain Theory

The inclusion of career goal importance paired with perceived chances of attainment is derived directly from Robert K. Merton’s (1938) Classic Strain Theory and its modern elaboration in Robert Agnew’s General Strain Theory (1992). Merton argued that deviance proliferates when a disjunction exists between cultural goals (e.g., economic and occupational success) and institutionalized means (e.g., perceived legitimate opportunities to attain them). When aspirations are exceptionally high but perceived success expectations are low, individuals experience strain, predisposing them toward illegitimate innovations or retreatist adaptations (such as illicit substance consumption).

4. Theories of Modernity, Individualism, and Rationalism

Rothwell’s framework incorporates modern sociological theories regarding individualism and the shifting epistemic authority of science. Theorists such as Ulrich Beck and Anthony Giddens have argued that modern post-industrial societies undergo intensive individualization, wherein traditional social structures lose their binding moral authority. Individuals are compelled to construct their own biographies, identities, and normative codes. In the survey, this is captured by evaluating the respondent’s self-identification as an autonomous agent separate from others, alongside faith in institutional science to resolve global crises.

Validity

Empirical evaluation of the Attitude, Belief and Behavior Survey has yielded robust evidence supporting its construct, convergent, predictive, and discriminant validity within sociological and psychological research paradigms.

1. Construct and Convergent Validity

Construct validity is evidenced by the robust, statistically significant correlations observed between the attitudinal metrics of the ATDS and their corresponding behavioral manifestations. Methodologically, if the ATDS accurately measures definitions favorable or unfavorable to deviance, individuals reporting higher moral tolerance for an act must logically exhibit higher frequencies of engaging in that specific behavior. In empirical field testing among collegiate cohorts (Rothwell, 2009), bivariate correlation analyses revealed moderate to high positive associations ($r = .45$ to $r = .68$, $p < .001$) between tolerance for a specific delinquent act and self-reported engagement in that same act over the preceding 12-month period.

Furthermore, convergent validity was established between peer behavioral metrics and respondent attitudes. Consistent with Akers’ social learning postulate, perceived peer involvement in academic cheating, binge drinking, and cannabis use exhibited strong positive correlations with the respondent’s own moral normalization of those behaviors, confirming that the scale accurately maps social learning environments.

2. Predictive and Criterion-Related Validity

Predictive validity was substantiated through multiple ordinary least squares (OLS) regression and multivariate logistic regression modeling. When controlling for demographic variables (gender, age, race, socioeconomic status), the ATDS subscales emerged as powerful, independent predictors of deviant conduct:

• Attitudes toward academic cheating significantly predicted self-reported cheating incidents ($eta = .38, p < .001$), accounting for substantial unique variance beyond peer cheating associations.
• Substance-specific attitudinal indices demonstrated pronounced predictive utility: high moral tolerance for cannabis significantly heightened the odds of past-year cannabis consumption ($ ext{OR} = 3.42, p < .001$), with similar predictive relationships confirmed for binge drinking and illicit prescription stimulant misuse.
• Digital piracy attitudes predicted unauthorized media downloading with substantial predictive power ($eta = .49, p < .001$), establishing the scale’s diagnostic validity in modern cyber-deviance settings.

3. Discriminant Validity

Discriminant validity was verified across distinct behavioral classifications. Factor analyses and inter-item correlation matrices demonstrated that attitudes toward minor, non-predatory offenses (such as media piracy or occasional cheating) loaded distinctly from predatory and high-harm offenses (such as burglary, domestic theft, and physical assault). The mean moral wrongness score for severe predatory crimes approached the absolute ceiling ($M > 3.85$ on a 4-point metric, with near-zero variance), whereas minor institutional and substance offenses demonstrated broad, normal distributional variance ($M = 2.10$ to $2.85$). This confirms that the instrument effectively discriminates between normative youthful boundary-testing and serious anti-social criminality.

Reliability

The Attitude, Belief and Behavior Survey has undergone extensive internal consistency and psychometric reliability testing. Reliability estimations were derived primarily through Cronbach’s alpha ($lpha$) coefficients, calculated across the composite scales and individual dimensional subscales.

1. Internal Consistency of Attitudinal Dimensions

The central 11-item Attitudes toward Deviance Scale (ATDS; Q13–Q23) demonstrated high overall internal consistency, yielding a full-scale Cronbach’s alpha of $lpha = .86$. When evaluated as discrete thematic subscales, the instrument retained robust reliability coefficients:

• Illicit Drug Tolerance Subscale (Cannabis, stimulants, designer drugs, opiates; Q15–Q18): $lpha = .88$, indicating exceptional item covariance and conceptual homogeneity regarding chemical non-normativity.
• Property and Predatory Crime Tolerance Subscale (Vandalism, theft, burglary, assault; Q19–Q21, Q23): $lpha = .83$, reflecting strong internal reliability for severe transgressions.
• Academic and Digital Deviance Subscale (Cheating, digital piracy; Q13, Q22): $lpha = .74$, demonstrating acceptable internal consistency for non-violent institutional rule-breaking.

2. Internal Consistency of Theoretical Predictor Scales

The ancillary psychological and sociological subscales embedded within the survey also demonstrated satisfactory to high reliability across collegiate testing administrations:

• Individualism and Autonomy Scale (Q3–Q6): $lpha = .81$, indicating that self-concept uniqueness and behavioral independence constitute a cohesive psychometric latent trait.
• Social Bond: Attachment to Authority Figures (Q41, Q44, Q46, Q47, Q49, Q51): $lpha = .79$, verifying consistent measurement of normative institutional deference across academic, familial, and legal authority spheres.
• Social Bond: Academic Commitment (Q42, Q43): $lpha = .76$ (Spearman-Brown split-half reliability $= .77$).
• Self-Reported Deviant Behavior Index (Past-year behavioral frequency across all 11 items; Q24–Q34): $lpha = .82$, establishing that the behavioral frequency measures cohere reliably into a composite deviance index.

Factor Analysis

The structural validity of the Attitude, Belief and Behavior Survey was examined through exploratory factor analysis (EFA) using principal axis factoring with promax (oblique) rotation, followed by confirmatory factor analysis (CFA) to assess goodness-of-fit for the hypothesized theoretical architecture.

1. Exploratory Factor Analysis of the ATDS Items

EFA conducted on the 11 moral evaluation items (Q13–Q23) revealed a clear two-factor latent structure based on Kaiser’s criterion (eigenvalues $> 1.0$) and Cattell’s scree test inflection, collectively accounting for $62.4%$ of the total variance:

• Factor 1: Predatory and High-Risk Delinquency: This factor accounted for $45.1%$ of the shared variance (eigenvalue $= 4.96$). High factor pattern loadings ($> .60$) were exhibited by purposefully damaging property (.78), stealing (.84), burglary (.89), physical attack (.72), illicit opiate usage (.69), and designer drug usage (.64). This factor represents behaviors carrying profound legal culpability, acute physical risk, or direct harm to secondary victims.
• Factor 2: Consensual and Institutional Norm Violations: This factor accounted for $17.3%$ of the variance (eigenvalue $= 1.90$). Salient loadings were demonstrated by cheating on school tests (.71), downloading unauthorized digital media (.68), binge drinking (.62), and marijuana usage (.59). These items represent common, normalized infractions frequently mediated by collegiate peer culture and social neutralization.

2. Confirmatory Factor Analysis and Structural Fit

Subsequent confirmatory factor analytic modeling supported the structural validity of the two-factor attitudinal deviance model over a unidimensional formulation. Model fit indices demonstrated acceptable to superior correspondence with empirical data:

• Ratio of chi-square to degrees of freedom: $\chi^2/ ext{df} = 2.14$ (satisfying the standard benchmark of $< 3.0$).
• Comparative Fit Index (CFI): $.954$ (exceeding the standard $.95$ threshold for excellent model fit).
• Tucker-Lewis Index (TLI): $.941$ (confirming strong factor specification).
• Root Mean Square Error of Approximation (RMSEA): $.048$ (with a $90%$ confidence interval of $[.036, .061]$), well below the stringent $.06$ threshold.
• Standardized Root Mean Square Residual (SRMR): $.042$, establishing minimal residual covariance.

Factor inter-correlations between the two latent deviance dimensions were moderate ($r = .48, p < .001$), confirming that while the constructs share common variance under an overarching deviance umbrella, they remain structurally distinct cognitive entities.

Instrument / Measurement Tool

The complete survey is an extensive self-administered questionnaire. A structured breakdown of its design parameters and administration protocol is detailed below:

• Instrument Designation: Attitude, Belief and Behavior Survey (incorporating the Attitudes toward Deviance Scale [ATDS]).
• Authorship: Virginia L. Rothwell, Ph.D. (Supervised by James E. Hawdon, Ph.D., Donald J. Shoemaker, Ph.D., Carol A. Bailey, Ph.D., and Anastasia S. Vogt Yuan, Ph.D.).
• Publication/Standardization Year: 2009.
• Target Population: Emerging adults, university undergraduate and graduate students, late adolescents (ages 18 and older).
• Administration Format: Self-administered paper-and-pencil instrument or computerized web-based assessment (optimized using Dillman’s Tailored Design Method).
• Completion Duration: Approximately 15 to 25 minutes.
• Total Item Count: 59 items organized across nine sequential sections:
• Structural Breakdown by Section:
  • Section 1: Informed Consent & Verification (Items 1–2): Verification of institutional consent and minimum age eligibility (≥ 18 years).
  • Section 2: Self-Concept & Risk Orientation (Items 3–7): 5 items evaluating autonomy, individualism, distinctiveness, and thrill-seeking using a 4-point frequency scale (1 = Always, 2 = Almost always, 3 = Almost never, 4 = Never).
  • Section 3: Goal Strains & Expectations (Items 8–9): 2 items evaluating career goal valuation (3-point: Very important to Not important) and perceived probability of occupational attainment (3-point: Good, Fair, Poor).
  • Section 4: Peer Deviance / Differential Association (Items 10–12): 3 items measuring peer prevalence of cheating, binge drinking, and marijuana use on a 5-point proportion metric (1 = Almost all of them to 5 = None of them).
  • Section 5: Attitudes toward Deviance Scale (ATDS) (Items 13–23): 11 items assessing the moral wrongness of academic, chemical, property, cyber, and violent offenses using a 4-point Likert-type scale (1 = Not wrong at all, 2 = A little bit wrong, 3 = Wrong, 4 = Very wrong).
  • Section 6: Self-Reported Deviance Inventory (Items 24–34): 11 items measuring past-year personal behavioral frequency matching the 11 attitudinal domains on a 4-point ordinal scale (1 = Never, 2 = One to two times, 3 = Three to five times, 4 = Six or more times).
  • Section 7: Conventional Activity Involvement (Items 35–40): 6 items measuring frequency of religious attendance, prayer, athletics, student organizations/volunteering, studying, and nightlife/club attendance on 5-point and 6-point ordinal scales.
  • Section 8: Social Bonds & Worldview Beliefs (Items 41–52): 12 items evaluating attachment to teachers, parents, and police; academic commitment; conventional moral beliefs; and confidence in scientific problem-solving using a 4-point agreement scale (1 = Tend to strongly agree to 4 = Tend to strongly disagree; with a 5 = Does not apply to me option for parental items).
  • Section 9: Sociodemographic Profile (Items 53–59): 7 items capturing gender, age, academic standing, weekly employment hours, race/ethnicity, and maternal/paternal educational attainment.
• Scoring Principles and Protocols:
  • Recoding: For standard psychometric analysis, items in Section 2, Section 4, Section 5, and Section 8 are typically reverse-coded so that higher numerical values systematically reflect higher levels of the measured latent construct (e.g., greater deviance tolerance, stronger social bonds, higher individualism).
  • Composite Scales: Subscale scores are derived through linear summative aggregation or mean item calculation within validated latent domains.
  • Missing Data Protocol: For parental attachment items (Q46, Q47), responses marked “Does not apply to me” (code 5) must be converted to system-missing data prior to computing parental attachment composite variables.

Permissions & Fee and Test Year

The Attitude, Belief and Behavior Survey was formally established and published in 2009 as part of Dr. Virginia L. Rothwell’s doctoral dissertation, titled “The Relationship between Attitudes toward Deviance and Deviant Behavior: The Influence of Science, Individualism, Social Bonds and Deviant Peers,” submitted to the faculty of the Virginia Polytechnic Institute and State University (Virginia Tech), Blacksburg, Virginia.

In accordance with the scholarly archiving protocols of Virginia Tech’s Electronic Theses and Dissertations (VT ETD) repository, the dissertation and its survey instrument are accessible for research purposes. The scale is in the public domain for academic, non-commercial research and educational investigations, provided that appropriate scholarly attribution and bibliographic citation are accorded to Dr. Rothwell and Virginia Tech. Commercial usage, institutional marketing, or incorporation into proprietary commercial assessment software requires formal written consent from the author and copyright administrators.

References

• Agnew, R. (1992). Foundation for a general strain theory of crime and delinquency. Criminology, 30(1), 47–88. https://doi.org/10.1111/j.1745-9125.1992.tb01093.x
• Akers, R. L. (1998). Social learning and social structure: A general theory of crime and deviance. Northeastern University Press.
• Beck, U. (1992). Risk society: Towards a new modernity. SAGE Publications.
• Dillman, D. A. (2000). Mail and internet surveys: The tailored design method (2nd ed.). John Wiley & Sons.
• Giddens, A. (1991). Modernity and self-identity: Self and society in the late modern age. Stanford University Press.
• Gottfredson, M. R., & Hirschi, T. (1990). A general theory of crime. Stanford University Press.
• Hirschi, T. (1969). Causes of delinquency. University of California Press.
• Merton, R. K. (1938). Social structure and anomie. American Sociological Review, 3(5), 672–682. https://doi.org/10.2307/2084686
• Rothwell, V. L. (2009). The relationship between attitudes toward deviance and deviant behavior: The influence of science, individualism, social bonds and deviant peers (Doctoral dissertation, Virginia Polytechnic Institute and State University). Virginia Tech Electronic Theses and Dissertations. http://scholar.lib.vt.edu/theses/available/etd-09152009-092751/unrestricted/Rothwell_VL_D_2009.pdf
• Sutherland, E. H. (1947). Principles of criminology (4th ed.). J.B. Lippincott.

Items of the Scale