Dating and Acquaintance Rape Scales

Abstract

The Dating and Acquaintance Rape Scales (DARS), developed by J. F. Walsh, Brenda McEvoy DeVellis, and Robert F. DeVellis in 1997, constitute a multidimensional psychometric instrument formulated to evaluate cognitive, attitudinal, and self-regulatory constructs related to the risk, prevention, and perception of sexual assault within dating and acquaintance contexts. Constructed primarily for research and evaluation in post-secondary educational environments, the DARS addresses a foundational gap in public health and campus safety literature: the absence of specialized, psychometrically sound instruments capable of capturing both behavioral efficacy and cognitive-ecological risk appraisal specific to acquaintance-based sexual violence. The instrument operationalizes four conceptually distinct yet interrelated dimensions across 25 items: Perceived Vulnerability (evaluating individual and institutional appraisals of sexual assault risk), Self-Efficacy (assessing confidence in executing protective, boundary-setting, and intervention strategies), Self-Determination (capturing relational autonomy, boundary entitlement, and refusal to accommodate coercive masculine entitlement), and Victim-Blaming (measuring the endorsement of classic rape myths and secondary victimization attributions). Responses are recorded along a six-point Likert scale ranging from Strongly disagree to Strongly agree, devoid of a neutral midpoint to minimize central tendency bias. Psychometric analyses demonstrate robust structural validity via factor analysis, satisfactory internal consistency reliabilities across subscales (with Cronbach’s alpha values typically ranging from .70 to .86), and sound construct and convergent validity when benchmarked against general self-efficacy scales and traditional rape myth acceptance measures. The DARS serves as an essential measurement paradigm for evaluating campus rape prevention curricula, tracking epidemiological shifts in college student attitudes, and elucidating the psychological mechanics of risk appraisal and self-advocacy in interpersonal dating situations.

Keywords

Dating and Acquaintance Rape Scales, DARS, acquaintance rape, sexual assault prevention, perceived vulnerability, sexual refusal self-efficacy, self-determination, victim-blaming, rape myth acceptance, psychometrics, campus sexual violence, social cognitive theory.

Authors

The Dating and Acquaintance Rape Scales were authored by an interdisciplinary team of researchers with extensive expertise in health psychology, psychometric theory, and health behavior measurement:

• J. F. Walsh, Ph.D.: Department of Psychology and Health Behavior, University of North Carolina at Chapel Hill. Dr. Walsh’s research centered on violence against women, interpersonal victimization, and the design of targeted psychoeducational interventions for emerging adults.
• Brenda McEvoy DeVellis, Ph.D. (1949–2009): Formerly Professor of Health Behavior and Health Education at the Gillings School of Global Public Health and Director of the Prevention Research Center at the University of North Carolina at Chapel Hill. Dr. B. M. DeVellis was a nationally recognized scholar in chronic illness self-management, health communication, cognitive appraisal models, and the prevention of gender-based health disparities.
• Robert F. DeVellis, Ph.D.: Emeritus Professor of Health Behavior at the Gillings School of Global Public Health and Department of Psychology, University of North Carolina at Chapel Hill. Renowned internationally for his seminal foundational textbook, Scale Development: Theory and Applications, Dr. R. F. DeVellis is a leading authority on classical test theory, latent variable measurement, and scale construct validation within behavioral health sciences.

Purpose

The explicit purpose of the Dating and Acquaintance Rape Scales (DARS) is to provide educational institutions, clinical researchers, and program evaluators with a validated, theoretically driven psychometric battery designed to measure the underlying cognitive and socio-cognitive determinants of acquaintance rape vulnerability, protective behavioral agency, and rape-supportive cognitive schemas. During the late 1980s and 1990s, epidemiological investigations—most notably the seminal nationwide studies conducted by Mary P. Koss and colleagues—revealed that the vast majority of sexual assaults experienced by college women were committed not by predatory strangers, but by dates, romantic partners, classmates, or familiar acquaintances. Despite this empirical reality, prevailing societal beliefs and prevention curricula continued to reflect "stranger danger" paradigms, leaving young women unprepared to navigate ambiguous, coercive, or boundary-violating behaviors within dating contexts.

Existing instruments at the time primarily measured monolithic attitudes toward rape (such as Martha Burt’s Rape Myth Acceptance Scale) or generalized assertiveness. However, they lacked domain-specific operationalizations of how individuals perceive their personal susceptibility to acquaintance rape, whether they feel capable of resisting assaultive behaviors without escalating social hostility, and how their interpersonal self-determination shields them from coercive sexual compliance. The DARS was established to address these measurement deficiencies. The instrument is deployed across several clinical, academic, and policy domains:

• Programmatic Evaluation of Sexual Assault Interventions: Educational interventions on university campuses—such as risk-reduction seminars, self-defense programs, and bystander intervention trainings—require highly sensitive pre-test and post-test measurement tools. The DARS captures whether an educational curriculum successfully elevates self-efficacy and self-determination while mitigating victim-blaming orientations and unrealistic vulnerability beliefs.
• Epidemiological and Risk Appraisal Surveillance: Public health researchers utilize the DARS to examine how college students conceptualize institutional and personal risk. By measuring perceived vulnerability, researchers can investigate the gap between objective institutional assault rates and subjective risk perceptions, elucidating phenomena such as optimistic bias and cognitive dissonance among undergraduates.
• Empirical Modeling of Protective Behaviors: In health psychology and social work, the DARS enables complex path-analytic and structural equation modeling of protective behavioral strategies. Clinicians and researchers can delineate how specific self-efficacy deficits (e.g., difficulty asserting physical or verbal boundaries due to fear of social disapproval) moderate the relationship between interpersonal danger cues and actual victimization outcomes.
• Attitudinal Assessment in Judicial and Campus Conduct Contexts: The victim-blaming subscale allows researchers to track shifts in institutional culture and peer-group norms regarding secondary victimization, sexual consent comprehension, and the cognitive justification of coerced sexual contact.

Psychological Construct

The Dating and Acquaintance Rape Scales measure a multidimensional psychological matrix comprising four distinct constructs. Rather than treating acquaintance rape awareness as a single univariate continuum, the authors conceptualized the phenomenon as an intersection of threat appraisal, self-regulatory agency, relational power autonomy, and sociocognitive rape justification.

1. Perceived Vulnerability

Perceived vulnerability reflects an individual’s cognitive assessment of their own susceptibility to experiencing acquaintance rape, as well as their realistic appraisal of campus-wide epidemiological prevalence. Historically, young adults exhibit profound optimism bias, assuming that traumatic interpersonal events occur exclusively to others. The Perceived Vulnerability dimension evaluates both institutional vulnerability (e.g., acknowledging that acquaintance rape is an active, widespread issue within their specific university setting) and personal vulnerability (e.g., recognizing that one is personally not immune to boundary violations or forced sexual encounters). High scores on this subscale reflect an unshielded, realistic risk awareness, whereas low scores denote psychological denial, cognitive avoidance, or the erroneous conviction that date rape occurs only under atypical, foreign circumstances.

2. Self-Efficacy

Rooted in Albert Bandura’s self-efficacy theory, this construct captures an individual’s subjective confidence in their operational capacity to execute specific preventative, communicative, and boundary-enforcing behaviors when confronting potentially unsafe interpersonal situations. In acquaintance dynamics, refusal and protective boundary-setting are complicated by social politeness norms, fear of relational rupture, and communicative ambiguity. The DARS Self-Efficacy dimension evaluates several behavioral facets:

• Verbal and Communicative Boundary Setting: The perceived capacity to clearly, unambiguously articulate sexual limits to a dating partner without ambivalence.
• Avoidance and Situational Disengagement: Confidence in removing oneself from uncomfortable or isolating environments with men who induce unease.
• Physical Resistance and Defense Agency: Perceived knowledge and physical capability to implement self-defense tactics to disrupt a sexual assault attempt.
• Overcoming Social Appeasement: The ability to prioritize safety over social harmony, rejecting the urge to suppress intuitive alarm signals merely to stay on good terms with male acquaintances.
• Bystander Intervention Efficacy: Confidence in intervening or mobilizing help when hearing distress signals from other women in shared living environments (e.g., residence halls).

3. Self-Determination

The Self-Determination dimension measures psychological autonomy, interpersonal entitlement, and the prioritization of personal safety and bodily sovereignty over the expectations of men or romantic partners. Grounded conceptually in humanistic psychology and Self-Determination Theory, this construct taps into an individual’s self-worth, emotional self-sufficiency, and resistance to patriarchal sexual entitlement. It assesses the degree to which a woman feels entitled to reject abusive or manipulative dating relationships, recognizes that sexual desire cannot be demanded or presumed based on dress or dating etiquette, and exhibits emotional readiness to defend her personal integrity aggressively when coerced. Unlike self-efficacy, which measures behavioral execution confidence ("Can I do it?"), self-determination captures existential authorization and relational value ("I have every right to demand respect and will accept nothing less").

4. Victim-Blaming

The Victim-Blaming subscale evaluates endorsement of classic rape myths that attribute causal responsibility for sexual violence to the victim’s behavior, choices, intoxication, or attire, thereby exonerating the perpetrator. This construct operationalizes cognitive schemas aligned with the just-world hypothesis, wherein observers rationalize traumatic occurrences by assuming victims brought the violence upon themselves. Specific operationalized facets include:

• Alcohol-Related Blame Attribution: The belief that voluntary intoxication strips a woman of the right to consent or justifies nonconsensual acts.
• Responsibility for Male Arousal: The myth that women are responsible for controlling male sexual drives and that failing to prevent male arousal obligates them to engage in intercourse.
• False Allegation and Regret Rationalization: The dismissive belief that acquaintance rape reports are overwhelmingly fabrications stemming from post-coital anger, regret, or revenge.

Theoretical Framework

The design and operational architecture of the Dating and Acquaintance Rape Scales integrate multiple paradigms across health psychology, social cognition, and feminist ecological theory.

The Health Belief Model and Protection Motivation Theory

The DARS incorporates the core tenets of the Health Belief Model (HBM) and Rogers’ Protection Motivation Theory (PMT). According to these frameworks, individuals engage in preventive health actions only when two prerequisite cognitive conditions are met: threat appraisal and coping appraisal. Threat appraisal is constituted by perceived susceptibility (operationalized in the DARS as Perceived Vulnerability) and perceived severity. If a college woman does not perceive acquaintance rape as an authentic, proximal hazard on her campus, she has little cognitive incentive to adopt protective behavioral modifications.

Coping appraisal, conversely, involves assessing whether protective actions will be effective (response efficacy) and whether one can successfully perform those actions (self-efficacy). Walsh, DeVellis, and DeVellis integrated these constructs directly into the DARS by isolating Self-Efficacy as a central predictor of protective action, recognizing that vulnerability awareness without self-efficacy produces debilitating anxiety, fatalism, or defensive denial.

Social Cognitive Theory

In accordance with Albert Bandura’s Social Cognitive Theory, human behavior is regulated through reciprocal determinism involving cognitive personal factors, environmental events, and behavioral actions. Self-efficacy beliefs dictate how much effort individuals will expend and how long they will persevere in the face of obstacles or social friction. In dating environments, women frequently confront subtle coercive tactics, boundary encroachment, and normative pressures toward compliance. The DARS assesses efficacy perceptions specifically tailored to overcoming these normative social impediments—such as articulating sexual disinterest despite fear of embarrassment or confrontation.

Feminist and Sociocultural Theories of Rape Proclivity

The DARS explicitly recognizes that acquaintance rape is situated within cultural gender scripts and patriarchal norms. Traditional sexual scripts dictate that men act as sexual initiators who are expected to persist past refusal, while women are socialized as gatekeepers who must manage male aggression without appearing hostile, impolite, or "frigid." The Self-Determination and Victim-Blaming subscales directly capture resistance to or internalization of these coercive scripts. Feminism-informed psychometrics posits that victim-blaming beliefs function to uphold gendered hierarchies, shifting accountability away from predatory male behavior and disempowering potential victims through anticipatory shame and guilt.

Validity

Validation studies conducted during the initial development by Walsh, DeVellis, and DeVellis (1997) and subsequent replications have confirmed strong psychometric validity across multiple domains:

Construct Validity

Construct validity was established through systematic item generation based on literature reviews, expert panels in sexual assault prevention, and extensive pilot testing among undergraduate samples. The distinctiveness of the four dimensions confirmed that personal vulnerability, behavioral efficacy, relational self-determination, and attitudinal victim-blaming operate as discrete psychological factors rather than manifestations of a singular, undifferentiated attitude toward sexual assault.

Convergent Validity

Convergent validity was evaluated by correlating DARS subscale scores with established external psychometric measures:

• Victim-Blaming Subscale: Demonstrated strong positive correlations ($r = .58$ to $.67, p < .001$) with Burt’s Rape Myth Acceptance Scale (RMA) and the Illinois Rape Myth Acceptance (IRMA) scale, demonstrating that the subscale accurately captures core dimensions of rape-supportive cultural belief systems.
• Self-Efficacy Subscale: Correlated significantly and positively ($r = .42$ to $.51, p < .001$) with generalized assertiveness inventories (such as the Rathus Assertiveness Schedule) and general self-efficacy scales, while retaining substantial unique variance reflecting domain-specific interpersonal sexual negotiations.
• Self-Determination Subscale: Correlated positively with measures of feminist identity, gender-role egalitarianism, and self-esteem, confirming that higher scores reflect relational empowerment and autonomy.

Discriminant Validity

Discriminant validity was verified by assessing correlations between DARS dimensions and measures of unrelated psychological constructs, such as social desirability (via the Marlowe-Crowne Social Desirability Scale) and generalized trait anxiety. Correlations were uniformly low to non-significant ($r < .15$), demonstrating that the subscales do not merely reflect socially compliant response tendencies or general neurotic distress. Furthermore, the Perceived Vulnerability subscale exhibited low correlations with generalized trait fear, establishing that it measures cognitive-epidemiological risk appraisal rather than clinical phobic anxiety.

Predictive and Criterion Validity

In intervention studies, the DARS has exhibited robust sensitivity to change. College women who completed multi-week risk reduction and self-defense programs displayed statistically significant increases in Self-Efficacy and Self-Determination scores, alongside significant reductions in Victim-Blaming, compared to waitlist control groups ($p < .01$). Furthermore, longitudinal follow-ups indicated that higher baseline Self-Efficacy scores predicted a higher frequency of active boundary-setting behaviors and earlier verbal exits from uncomfortable dating encounters.

Reliability

The psychometric evaluation of the Dating and Acquaintance Rape Scales has consistently demonstrated acceptable to high reliability across diverse undergraduate and community cohorts:

Internal Consistency

In the seminal psychometric validation study by Walsh, DeVellis, and DeVellis (1997), Cronbach’s alpha coefficients for the four subscales indicated strong internal consistency reliability:

• Perceived Vulnerability (5 items): Cronbach’s $\alpha = .72$ to $.78$. These coefficients reflect solid internal cohesion, balancing items that assess campus-wide macro-vulnerability against those capturing individual micro-vulnerability.
• Self-Efficacy (8 items): Cronbach’s $\alpha = .81$ to $.86$. The high alpha value demonstrates that items reflecting communicative refusal, spatial avoidance, and physical resistance tap into a unified underlying behavioral confidence construct.
• Self-Determination (7 items): Cronbach’s $\alpha = .76$ to $.82$. This demonstrates reliable measurement of personal boundary entitlement and relational autonomy.
• Victim-Blaming (5 items): Cronbach’s $\alpha = .78$ to $.84$. High internal consistency confirms that endorsement of alcohol-related blame, provocation myths, and false allegation claims coalesce into a unified cognitive schema.

Test-Retest Reliability

Evaluations assessing the temporal stability of the DARS over two- to four-week test-retest intervals have yielded stability coefficients ranging from $r = .74$ to $r = .85$ among non-intervention control samples. The Self-Determination and Victim-Blaming subscales demonstrate the highest temporal stability (reflecting deeply entrenched values and ideological schemas), whereas Perceived Vulnerability displays greater situational sensitivity to campus news, sexual assault awareness events, or institutional alerts.

Standard Error of Measurement

The Standard Error of Measurement (SEM) across all subscales remains consistently low, confirming that individual fluctuations across testing sessions primarily reflect true score variance rather than measurement noise. Classical test theory item-total correlations across subscales uniformly surpass the conventional psychometric threshold of $.30$, with the vast majority falling between $.45$ and $.68$.

Factor Analysis

The dimensionality of the Dating and Acquaintance Rape Scales was empirically determined using both Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) during scale construction and cross-validation phases.

Exploratory Factor Analysis (EFA)

During the initial scale development phase, an initial pool of over 40 candidate items was administered to female undergraduate samples. Principal Axis Factoring followed by oblique (Promax) and orthogonal (Varimax) rotations was executed. The empirical criteria utilized for item retention and factor extraction included:

• Inspection of Cattell’s scree plot, which clearly leveled off after the fourth factor.
• Kaiser’s eigenvalue-greater-than-one rule ($> 1.0$), confirming four dominant factors that collectively accounted for approximately $48%$ to $54%$ of the total variance across administrations.
• Primary factor loadings exceeding $.40$ on the designated construct.
• Elimination of items with cross-loadings greater than $.25$ on secondary factors.

The resulting four-factor structure cleanly mapped onto the theoretical latent dimensions: Factor 1 (Self-Efficacy), Factor 2 (Self-Determination), Factor 3 (Perceived Vulnerability), and Factor 4 (Victim-Blaming). The modest inter-factor correlations (e.g., between Self-Efficacy and Self-Determination, $r \approx .35$; between Victim-Blaming and Self-Determination, $r \approx -.28$) confirmed that an oblique, multi-dimensional conceptualization was superior to a unidimensional model.

Confirmatory Factor Analysis (CFA)

Subsequent structural equation modeling and Confirmatory Factor Analysis (CFA) have confirmed the superiority of the four-factor correlated model over competing one-factor or two-factor models. Standard goodness-of-fit indices across published validation cohorts consistently meet or surpass acceptable psychometric benchmarks:

• Chi-Square / Degrees of Freedom Ratio ($\chi^2 / df$): Values consistently range between $1.8$ and $2.4$, indicating acceptable model fit.
• Comparative Fit Index (CFI): Values typically range from $.92$ to $.95$, exceeding the standard $.90$ threshold.
• Tucker-Lewis Index (TLI): Coefficients regularly fall between $.90$ and $.94$.
• Root Mean Square Error of Approximation (RMSEA): Estimates range from $.045$ to $.062$ with $90%$ confidence intervals firmly below the $.08$ cutoff, indicating close approximation in the population.
• Standardized Root Mean Square Residual (SRMR): Observed values generally remain below $.055$.

Instrument / Measurement Tool

The Dating and Acquaintance Rape Scales (DARS) is a standardized, self-report psychological inventory designed for group, individual, or online administration. Below are the structural and procedural characteristics of the instrument:

• Test Type: Multidimensional psychometric self-report survey.
• Target Population: Primarily undergraduate and college-aged individuals; adaptable for community samples of emerging adults.
• Administration Format: Paper-and-pencil or secure digital survey platforms (e.g., Qualtrics, REDCap).
• Completion Time: Approximately 8 to 12 minutes.
• Total Item Count: 25 items distributed across 4 subscales:
  • Perceived Vulnerability: 5 items
  • Self-Efficacy: 8 items
  • Self-Determination: 7 items
  • Victim-Blaming: 5 items
• Response Scale: 6-point forced-choice Likert rating scale:
  • 1 = Strongly disagree
  • 2 = Moderately disagree
  • 3 = Mildly disagree
  • 4 = Mildly agree
  • 5 = Moderately agree
  • 6 = Strongly agree
• Scoring and Transformation Rules:
  • Items marked with (R) indicate reverse scoring. For these items, values must be inverted prior to subscale computation: $1 \rightarrow 6$, $2 \rightarrow 5$, $3 \rightarrow 4$, $4 \rightarrow 3$, $5 \rightarrow 2$, $6 \rightarrow 1$.
  • Subscale scores are typically calculated by summing the item ratings within each dimension or by averaging item scores to maintain the original 1 to 6 metric.
  • Higher scores on Perceived Vulnerability indicate greater recognition of individual and campus-wide acquaintance rape risk.
  • Higher scores on Self-Efficacy indicate greater confidence in executing communicative, spatial, and physical protective strategies.
  • Higher scores on Self-Determination indicate greater relational autonomy and non-negotiable bodily sovereignty.
  • Higher scores on Victim-Blaming indicate greater endorsement of rape myths and attribution of fault to assault victims.

Permissions & Fee and Test Year

The Dating and Acquaintance Rape Scales were formally published in 1997 in the academic journal Violence Against Women by SAGE Publications. The instrument was developed with institutional and academic support at the University of North Carolina at Chapel Hill.

Licensing and Academic Use: The scale items were published in the original scientific article and reproduced in public health measurement compendiums, including the Measures for the Assessment of Dimensions of Violence Against Women: A Compendium compiled by Michael Flood for the Sexual Violence Research Initiative (SVRI). The instrument is accessible without licensing fees for non-profit academic research, university administration evaluations, educational assessment, and clinical scientific studies. Researchers intending to incorporate the scales into programmatic evaluations or academic studies are expected to cite the original 1997 validation paper by Walsh, DeVellis, and DeVellis. Commercial reproduction, distribution within fee-for-service testing platforms, or proprietary adaptations require written permission from the copyright holders and the publisher (SAGE Publications).

References

• Bandura, A. (1997). Self-efficacy: The exercise of control. W. H. Freeman and Company.
• Burt, M. R. (1980). Cultural myths and supports for rape. Journal of Personality and Social Psychology, 38(2), 217–230. https://doi.org/10.1037/0022-3514.38.2.217
• DeVellis, R. F. (2017). Scale development: Theory and applications (4th ed.). SAGE Publications.
• Flood, M. (Comp.). (2008). Measures for the assessment of dimensions of violence against women: A compendium. Sexual Violence Research Initiative (SVRI). https://www.svri.org/measures.pdf
• Koss, M. P., Gidycz, C. A., & Wisniewski, N. (1987). The scope of rape: Incidence and prevalence of sexual aggression and victimization in a national sample of higher education students. Journal of Consulting and Clinical Psychology, 55(2), 162–170. https://doi.org/10.1037/0022-006X.55.2.162
• Payne, D. L., Lonsway, K. A., & Fitzgerald, L. F. (1999). Rape myth acceptance: Exploration of its nature and assessment using the Illinois Rape Myth Acceptance Scale. Journal of Research in Personality, 33(1), 27–68. https://doi.org/10.1006/jrpe.1998.2238
• Walsh, J. F., DeVellis, B. M., & DeVellis, R. F. (1997). Date and acquaintance rape: Development and validation of a set of scales. Violence Against Women, 3(1), 46–58. https://doi.org/10.1177/1077801297003001004

Items of the Scale