Controlling Behaviors Scale (CBS-R)

Abstract

The Controlling Behaviors Scale – Revised (CBS-R) is an established, multidimensional psychometric self-report and collateral-report instrument designed to quantify non-physical, non-fatal coercive and domineering tactics within intimate partner relationships. Originating from the foundational work of John Archer and Nicola Graham-Kevan at the University of Central Lancashire, the instrument operationalizes the behavioral dimensions conceptualized within the influential Duluth Power and Control Wheel and Michael P. Johnson’s typology of intimate partner violence (IPV). The CBS-R measures controlling behaviors through 32 items distributed across seven distinct clinical and behavioral subscales: Economic Control, Threatening Control, Intimidating Control, Emotional Control, Isolating Control, Using Children, and Minimizing. Each item is evaluated along a 5-point Likert-type frequency metric ranging from 0 (Never) to 4 (Always). Across forensic, clinical, student, and community samples worldwide, the CBS-R demonstrates robust psychometric properties, featuring high internal consistency (with subscale Cronbach’s alpha coefficients routinely exceeding .75 to .92 and full-scale alpha exceeding .90), excellent construct and convergent validity alongside instruments such as the Conflict Tactics Scales (CTS2), and established predictive validity regarding physical aggression, psychological trauma, and relationship dissolution. This paper provides an exhaustive psychometric, theoretical, and clinical review of the CBS-R, explicating its factor structure, clinical scoring parameters, measurement invariance, and empirical administration.

Keywords

Controlling Behaviors Scale, CBS-R, coercive control, intimate partner violence, psychological abuse, domestic violence typology, intimate terrorism, psychometrics, relational aggression, behavioral control

Authors

The Controlling Behaviors Scale was originated and systematically validated by:

• Nicola Graham-Kevan, Ph.D. – Professor of Criminal Justice Psychology, School of Psychology and Humanities, University of Central Lancashire (UCLan), Preston, Lancashire, United Kingdom. Dr. Graham-Kevan has served as a preeminent international expert in aggressive behavior, intimate partner violence typologies, forensic assessment, and criminal risk management.
• John Archer, Ph.D. – Emeritus Professor of Psychology, School of Psychology and Humanities, University of Central Lancashire, Preston, United Kingdom. Professor Archer is globally recognized for his pioneering meta-analyses on sex differences and aggression, evolutionary models of human violence, and domestic violence epidemiology.

Purpose

The primary clinical and empirical objective of the Controlling Behaviors Scale – Revised (CBS-R) is to provide an objective, reliable, and standardized assessment of interpersonal coercive tactics operating within marital, cohabiting, or dating dyads. For decades, the study of domestic abuse was predominantly anchored to measures of overt physical assaults—such as the physical assault subscales of Murray A. Straus’s Conflict Tactics Scales (CTS). However, contemporary forensic psychology and domestic abuse research demonstrate that non-physical controlling tactics constitute the structural bedrock of relationship pathology, severe psychological trauma, and physical violence escalation.

The CBS-R was expressly formulated to test and quantify Michael P. Johnson‘s dynamic typology of intimate partner violence. Johnson posited a categorical distinction between situational couple violence (conflict-instigated, common couple violence devoid of a broader pattern of pervasive systemic control) and intimate terrorism (a unilateral, systematic campaign of terror, intimidation, and multifaceted subjugation). By measuring behaviors that constrain personal autonomy, regulate interpersonal interactions, control material survival, and manipulate emotional well-being, the CBS-R enables researchers and clinicians to:

• Distinguish between isolated acts of reactive physical conflict and systemic patterns of coercive domination.
• Assess the bidirectional dynamics of control, as the instrument can be administered to measure both perpetration (self-report of behaviors enacted toward partner) and victimization (dyadic report of behaviors experienced from partner).
• Formulate clinical risk assessments in forensic settings, domestic abuse shelters, child custody and welfare evaluations, and perpetrator intervention programs.
• Investigate evolutionary psychological and sociological theories regarding sexual proprietory behavior, mate guarding, and the structural dynamics of relational power imbalances.

Psychological Construct

The psychological construct assessed by the CBS-R is interpersonal coercive control, defined as an enduring, multidimensional pattern of interpersonally directed acts used to establish, enforce, and maintain dominance and micro-regulation over an intimate partner’s thoughts, emotions, behaviors, social networks, and practical liberties. Rather than viewing controlling actions as discrete or disconnected behavioral infractions, the construct represents a systemic matrix of micro-regulations. The CBS-R operationalizes this overarching construct through seven distinct behavioral dimensions:

1. Economic Control (Items 1–4)

Economic control involves the deliberate sabotage, subjugation, or deprivation of a partner’s financial agency and material resources. By restricting educational pursuits, sabotaging workplace employment, monopolizing household capital, concealing assets, or refusing to contribute equitably to shared household expenditures, the controlling agent renders the targeted partner financially dependent, thereby removing their material capacity for autonomous decision-making or relationship exit.

2. Threatening Control (Items 5–8)

Threatening control encompasses overt and covert conditional ultimatums aimed at inducing paralyzing fear and compliance. Tactics include explicit declarations of physical harm directed toward the partner, instrumental threats of immediate relationship abandonment, coercive self-harm threats (e.g., threatened suicide to prevent separation), and threats of reputational destruction via the malicious disclosure of sensitive or humiliating personal information.

3. Intimidating Control (Items 9–13)

Intimidating control captures overt non-verbal and environmental expressions of dominance designed to maintain a climate of psychological terror. It manifests through menacing facial expressions or non-verbal posturing, coercive enforcement of unwanted actions, explosive destruction of the partner’s personal belongings, hostility toward the partner’s extended family, and displaced aggressive outbursts directed at domestic pets.

4. Emotional Control (Items 14–18)

Emotional control (frequently discussed in contemporary clinical discourse as severe psychological denigration or gaslighting) targets the internal cognitive and affective integrity of the partner. Behaviors include systemic verbal insults, degrading the partner’s self-concept whenever they demonstrate autonomy, intentional public humiliation, and epistemic invalidation—specifically telling the partner that they are mentally unhinged, delirious, or chronic liars.

5. Isolating Control (Items 19–24)

Isolating control operationalizes obsessive mate-guarding and social alienation. The perpetrator deliberately severs the partner’s access to external systems of support, including family, peers, and community networks. Items measure the active restriction of contact, obsessive surveillance (interrogating whereabouts and communications), spatial tracking, suspiciousness, and the intentional provocation of relational jealousy to manipulate the partner’s emotional state.

6. Using Children (Items 25–29)

This contextual dimension, applicable to parents, captures the instrumentalization of dependent children as tactical weapons in interpersonal coercion. Behaviors include inducing parental guilt, turning children into coerced messengers, threatening custodial severance, orchestrating explosive confrontations in the children’s presence, and deploying physical aggression directly before the children to amplify trauma and enforce subjugation.

7. Minimizing (Items 30–32)

The minimizing dimension reflects defensive epistemic distortion, externalization of blame, and DARVO-type mechanisms (Deny, Attack, and Reverse Victim and Offender). Items assess the systematic weaponization of false accusations of violence, cognitive minimization of the severity of physical strikes, and coercive victim-blaming whereby the injured party is held psychologically responsible for provoking the assault.

Theoretical Framework

The conceptual foundation of the CBS-R is situated at the intersection of three major psychological frameworks: the Duluth Power and Control Model, Johnson’s Dyadic IPV Typology, and Evolutionary Mate-Retention Theory.

The Duluth Framework

Developed in the early 1980s by the Domestic Abuse Intervention Project (DAIP) in Duluth, Minnesota, the Power and Control Wheel organized clinical observations from female victims of battering. It postulated that physical violence does not occur in a vacuum; rather, physical strikes serve as the reinforcing enforcement mechanism for an overarching wheel of non-physical tactics: economic exploitation, coercion and threats, intimidation, emotional degradation, isolation, and the manipulation of children. Graham-Kevan and Archer (2003, 2004) translated these descriptive, sociological categories into psychometrically quantifiable variables suitable for scientific, hypothesis-driven psychological inquiry.

Johnson’s Typology of Intimate Partner Violence

Sociologist Michael P. Johnson (1995, 2008) addressed the long-standing “paradox” in domestic violence research: why family conflict survey research consistently reported gender symmetry in relationship violence, whereas clinical, shelter, and criminal justice samples revealed near-exclusive male perpetration of catastrophic terror. Johnson demonstrated that this contradiction stemmed from sampling different phenomena. He distinguished:

1. Situational Couple Violence (SCV): Symmetric, low-level physical conflicts triggered by acute disagreements, characterized by an absence of a general, pervasive motive to control one’s partner.
2. Intimate Terrorism (IT): Asymmetric, chronic, and severe physical violence nested within a systematic matrix of coercive, controlling tactics designed to subjugate the other partner entirely.
3. Violent Resistance (VR): Acts of physical defense or counter-violence enacted by victims against intimate terrorists.
4. Mutual Violent Control (MVC): Rare dyads wherein both partners actively compete for total tyrannical dominance using both violence and pervasive coercive control.

The CBS-R serves as the gold-standard instrument for operationalizing the control axis of Johnson’s matrix, allowing researchers to categorize dyads based on statistical cut-offs of controlling behavior scores combined with physical violence metrics.

Evolutionary Mate-Retention Theory

From an evolutionary psychology perspective (Archer & Graham-Kevan, 2009; Buss, 1988), controlling behaviors represent extreme, maladaptive manifestations of mate-retention tactics and sexual proprietariness. Isolating a mate, monitoring their movements, depleting their autonomous resources, and using threats can be viewed as coercive strategies evolved to prevent infidelity, paternity uncertainty, or mate defection. The CBS-R captures both direct-guarding tactics (isolation, surveillance) and negative-inducement tactics (threats, emotional degradation).

Validity

The psychometric validity of the CBS-R has been extensively substantiated across diverse samples, including university students, community dyads, forensic psychiatric patients, battered women’s shelter residents, and male criminal justice batterer intervention cohorts.

Construct and Factorial Validity

Graham-Kevan and Archer (2003, 2005) demonstrated that the CBS-R possesses robust factorial construct validity. Confirmatory factor analyses across community and clinical populations have consistently confirmed that non-physical controlling tactics resolve into the postulated multi-factor domain rather than a single undifferentiated negative affect factor. Intercorrelations among the subscales range from moderate (.35) to high (.78), confirming that while each subscale captures an autonomous behavioral domain, they converge coherently upon the broader higher-order construct of interpersonal coercive control.

Convergent and Discriminant Validity

The CBS-R demonstrates high convergent validity with established measures of relationship dynamics and interpersonal abuse:

• Conflict Tactics Scales (CTS & CTS2): Controlling behavior scores on the CBS-R correlate strongly with CTS2 Psychological Aggression subscale scores (r = .62 to .78, p < .001) and moderately to strongly with CTS2 Physical Assault scores (r = .40 to .65, p < .001).
• Psychological Maltreatment of Women Inventory (PMWI): High convergence (r > .70) has been established between the isolating, emotional, and threatening subscales of the CBS-R and the Dominance-Isolation and Emotional-Verbal subscales of the PMWI.
• Discriminant Validity: The scale discriminates effectively between coercive relationship dynamics and general relationship dissatisfaction. While global relationship satisfaction (e.g., measured via the Dyadic Adjustment Scale) correlates negatively with CBS-R scores (typically r = -.25 to -.40), factor analyses demonstrate that controlling behavior items do not cross-load onto general relationship distress or neuroticism inventories, establishing that coercive control is distinct from common marital unhappiness.

Predictive and Criterion-Related Validity

The CBS-R possesses exceptional criterion-related and predictive validity. In pivotal studies testing Johnson’s typology, Graham-Kevan and Archer (2003, 2008) demonstrated that high scores on the CBS-R reliably differentiate intimate terrorists from situational couple violence perpetrators across shelter and community samples. Furthermore, longitudinal studies confirm that high CBS-R scores predict escalation into severe physical violence, partner injury, and clinical post-traumatic stress disorder (PTSD) symptomatology in victimized partners over time.

Reliability

The Controlling Behaviors Scale – Revised exhibits consistently high reliability indices across multiple cultural and demographic contexts.

Internal Consistency

Across validation studies conducted in the United Kingdom, North America, and international translations (e.g., the Chinese C-CBS-R by Tiwari, 2015), the CBS-R has shown robust internal consistency:

• Total Scale: Full-scale Cronbach’s alpha coefficients typically range between α = .88 and .94 for both self-reported perpetration and partner victimization reports.
• Economic Control: Cronbach’s α ranges between .70 and .82.
• Threatening Control: Cronbach’s α ranges between .75 and .86.
• Intimidating Control: Cronbach’s α ranges between .74 and .85.
• Emotional Control: Cronbach’s α ranges between .78 and .89.
• Isolating Control: Cronbach’s α ranges between .80 and .90.
• Using Children: Cronbach’s α ranges between .76 and .88 among subsamples with dependent offspring.
• Minimizing: Cronbach’s α ranges between .71 and .84.

Test-Retest Stability

In stability studies over intervals of 4 to 8 weeks in non-intervened community couples, the CBS-R demonstrated high test-retest reliability, with intraclass correlation coefficients (ICCs) ranging from .76 to .88 across subscales, confirming that coercive controlling behaviors represent an enduring behavioral pattern rather than fleeting, mood-congruent state fluctuations.

Factor Analysis

The structural composition of the Controlling Behaviors Scale has been scrutinized through both Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA).

Exploratory Factor Structure

In the original developmental iterations (Graham-Kevan, 2004; Graham-Kevan & Archer, 2003), principal axis factoring with oblimin rotation was conducted on the core behavioral items. The scree plot and eigenvalue criteria (λ > 1.0) consistently extracted five core primary factors representing Economic, Threatening, Intimidating, Emotional, and Isolating control, with children-specific items loading cleanly onto an autonomous factor when parents were evaluated independently. Factor loadings for designated primary items routinely exceeded .45, with primary loadings commonly falling between .55 and .82, and minimal problematic cross-loadings (rarely exceeding .25 on secondary dimensions).

Confirmatory Factor Analysis and Model Fit

Confirmatory factor analytic investigations (e.g., Graham-Kevan & Archer, 2008, 2009; Tiwari, 2015) have evaluated alternative structural representations, including: (a) a single-factor unidimensional model, (b) a series of uncorrelated orthogonal factors, and (c) a correlated multi-factor model (or hierarchical second-order model where subscales load onto a single higher-order Coercive Control construct).

The multi-factor and hierarchical models demonstrate superior fit to empirical data compared to unidimensional alternatives. Standard structural equation modeling fit indices in published literature yield:

• Comparative Fit Index (CFI): .92 to .96
• Tucker-Lewis Index (TLI): .91 to .95
• Root Mean Square Error of Approximation (RMSEA): .042 to .058 (with 90% confidence intervals within acceptable psychometric bounds)
• Standardized Root Mean Square Residual (SRMR): .045 to .061

These findings substantiate the scale’s theoretical design: intimate partner control is both clinically nuanced into specific functional domains (e.g., economic deprivation vs. social isolation) and unified under a coordinated higher-order syndrome of interpersonal subjugation.

Instrument / Measurement Tool

The CBS-R is an assessment tool designed for self-administration, research evaluation, or clinical interview formatting. Below is the technical breakdown of the instrument:

• Measurement Tool Name: Controlling Behaviors Scale – Revised (CBS-R)
• Target Population: Adults (18 years and older) who are currently involved in or have previously been involved in an intimate partner, dating, cohabiting, or marital relationship.
• Administration Modality: Paper-and-pencil questionnaire, digital psychometric platform, or structured clinical assessment. Can be administered in self-report format (“How often did you enact this behavior?”) or collateral / victim report format (“How often did your partner enact this behavior toward you?”).
• Item Count: 32 items total.
  • Core items for all respondents: 27 items (Items 1–24 and Items 30–32).
  • Parental subscale: 5 items (Items 25–29, administered specifically to individuals with dependent children).
• Response Format: 5-point Likert-type frequency scale:
  • 0 = Never
  • 1 = Rarely
  • 2 = Sometimes
  • 3 = Often
  • 4 = Always
• Subscale Structural Division:
  • Economic Control: Items 1, 2, 3, 4 (Score range: 0–16)
  • Threatening Control: Items 5, 6, 7, 8 (Score range: 0–16)
  • Intimidating Control: Items 9, 10, 11, 12, 13 (Score range: 0–20)
  • Emotional Control: Items 14, 15, 16, 17, 18 (Score range: 0–20)
  • Isolating Control: Items 19, 20, 21, 22, 23, 24 (Score range: 0–24)
  • Using Children: Items 25, 26, 27, 28, 29 (Score range: 0–20; conditional)
  • Minimizing: Items 30, 31, 32 (Score range: 0–12)
• Scoring and Interpretation Procedures:
  • Continuous Dimensional Scoring: Subscale scores are obtained by calculating the arithmetic sum or mean of the respective items. A total global control score is calculated by summing all administered items (or using mean item scores to account for parental item non-applicability). Higher scores reflect greater frequency and breadth of coercive controlling behavior.
  • Dichotomous Categorical Scoring (Typology Allocation): In line with Johnson’s typological methodology, items can be dichotomized (e.g., 0 = Never, 1 = Ever enacted / score ≥ 1) to calculate the cumulative breadth of controlling tactics. Dyads scoring above established population median or clinical cut-offs on control breadth alongside physical violence measures are categorized as exhibiting Intimate Terrorism or Violent Resistance, whereas those scoring below the control threshold with active physical aggression are categorized as displaying Situational Couple Violence.

Permissions & Fee and Test Year

The Controlling Behaviors Scale was initially operationalized and published by Nicola Graham-Kevan and John Archer in 2003 (with further revisions and doctoral work established in Graham-Kevan, 2004, 2005, and 2008–2010). The instrument is widely considered an open-access academic and clinical research measure available for non-commercial scientific, educational, and clinical research purposes, provided appropriate scholarly attribution is cited. No commercial licensing fee is required for academic investigations or non-profit clinical settings. Researchers seeking to implement large-scale institutional testing, automated diagnostic commercial software, or translated adaptations are advised to contact the primary developer, Dr. Nicola Graham-Kevan at the University of Central Lancashire.

References

• Archer, J., & Graham-Kevan, N. (2009). Control tactics and partner violence in heterosexual relationships. Evolution and Human Behavior, 30(6), 445–452. https://doi.org/10.1016/j.evolhumbehav.2009.06.007
• Buss, D. M. (1988). From vigilance to violence: Tactics of mate retention in American undergraduates. Ethology and Sociobiology, 9(5), 291–317. https://doi.org/10.1016/0162-3095(88)90024-3
• Graham-Kevan, N. (2004). Physical aggression and controlling behavior within relationships: The effect of sampling (Doctoral dissertation). University of Central Lancashire, Preston, United Kingdom.
• Graham-Kevan, N., & Archer, J. (2003). Physical aggression and control in heterosexual relationships: The effect of sampling. Violence and Victims, 18(2), 181–196. https://doi.org/10.1891/vivi.2003.18.2.181
• Graham-Kevan, N., & Archer, J. (2005). Investigating three explanations of women’s relationship aggression. Psychology of Women Quarterly, 29(3), 270–277. https://doi.org/10.1111/j.1471-6402.2005.00221.x
• Graham-Kevan, N., & Archer, J. (2008). Does controlling behavior predict physical aggression and violence to partners? Journal of Family Violence, 23(7), 539–548. https://doi.org/10.1007/s10896-008-9177-8
• Johnson, M. P. (1995). Patriarchal terrorism and common couple violence: Two forms of violence against women. Journal of Marriage and the Family, 57(2), 283–294. https://doi.org/10.2307/353683
• Johnson, M. P. (2008). A typology of domestic violence: Intimate terrorism, violent resistance, and situational couple violence. Northeastern University Press.
• Straus, M. A., Hamby, S. L., Boney-McCoy, S., & Sugarman, D. B. (1996). The Revised Conflict Tactics Scales (CTS2): Development and preliminary psychometric data. Journal of Family Issues, 17(3), 283–316. https://doi.org/10.1177/019251396017003001
• Tiwari, A. (2015). Evaluating the psychometric properties of the Chinese Controlling Behaviors Scale – Revised (C-CBS-R). School of Nursing, Li Ka Shing Faculty of Medicine, The University of Hong Kong.

Items of the Scale