Conflict Anxiety Response Scale (CARS)

1. Abstract

The Conflict Anxiety Response Scale (CARS; Coleman & Chan, 2023) is a specialized psychometric instrument designed to evaluate how acute and chronic conflict anxiety shapes individuals’ behavioral and cognitive responses during interpersonal disputes. Rooted in the foundational theoretical typology articulated by social psychologist Morton Deutsch (1993), the instrument operationalizes twelve distinct response tendencies organized across six conceptual bipolar dimensions: Avoidance versus Involvement (Avoid–Involve), Soft versus Hard bargaining (Soft–Hard), Emotional expression versus Intellectualization (Emotional–Intellectualize), Escalation versus Minimization (Escalate–Minimize), Concealment versus Revelation (Conceal–Reveal), and Rigidity versus Looseness (Rigid–Loose). Developed and refined across an extensive program of seven empirical studies sampling adult populations in the United States via MTurk, Qualtrics, and Prolific platforms, the authors established a parsimonious 22-item brief self-report inventory.

Each item is evaluated along a standardized 5-point Likert scale, yielding granular dimensional subscale scores that illuminate specific behavioral “derailers” triggered under threat or relational tension. Psychometric evaluations confirm solid internal consistency, with subscale Cronbach’s alpha values ranging between 0.67 and 0.895 across diverse cohorts. Structural validity was established through sequential exploratory factor analyses (EFAs) utilizing principal axis factoring and promax rotations, followed by rigorous confirmatory factor analyses (CFAs). The final structural model demonstrated satisfactory goodness-of-fit parameters (Root Mean Square Error of Approximation [RMSEA] = 0.060, Comparative Fit Index [CFI] = 0.83, Tucker-Lewis Index [TLI] = 0.82) alongside minimized Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC) indices. Demonstrating robust predictive validity, the CARS correlates significantly with relational well-being, psychological distress, and interpersonal efficacy, confirming Deutsch’s assertion that polarized or extreme conflict responses exacerbate dispute outcomes and personal emotional turmoil.

2. Keywords

Conflict Tendencies, Conflict Management, Conflict Anxiety, Anxiety Management, Conflict Resolution, Behavioral Measures, Morton Deutsch, Bipolar Dimensions, Interpersonal Derailers, Psychometrics, State-Trait Conflict, Negotiation Analysis

3. Authors

The Conflict Anxiety Response Scale was conceptualized, operationalized, and validated by leading scholars in the fields of social-organizational psychology and dispute resolution:

• Peter T. Coleman, Ph.D.: Professor of Psychology and Education at Teachers College and the Earth Institute, Columbia University. Dr. Coleman directs the Morton Deutsch International Center for Cooperation and Conflict Resolution (MD-ICCCR). His research focuses on intractable conflict, systemic conflict dynamics, social power, and constructive negotiation. Email: [email protected].
• Anthea Chan, Ph.D.: Researcher and practitioner affiliated with Coqual and Columbia University, specializing in applied organizational behavior, diversity metrics, equity dynamics, and conflict response mechanisms. ORCID: 0000-0002-2667-2927.

4. Purpose

Interpersonal conflict is an inescapable facet of human interaction across domestic, social, and organizational contexts. However, individuals differ profoundly in their capacity to navigate these disputes constructively. While classic negotiation and conflict literature has extensively documented cognitive strategies, problem-solving tactics, and overarching conflict styles (such as the Dual Concern Model), it has frequently underemphasized the physiological, affective, and visceral role played by conflict anxiety. The Conflict Anxiety Response Scale (CARS) was engineered specifically to address this theoretical and empirical gap.

When interpersonal tension emerges, heightened autonomic arousal and perceived relational threat frequently induce conflict anxiety. Under the influence of this state, individuals often abandon rational dispute-resolution strategies, instead reverting to deeply conditioned, automatic, and frequently maladaptive behavioral scripts. Coleman and Chan (2023) term these reflexive patterns “conflict response derailers.” The primary diagnostic purpose of the CARS is to provide an empirical instrument capable of mapping, assessing, and quantifying these specific behavioral derailers across six fundamental behavioral axes.

Beyond theoretical inquiry, the CARS serves vital applied functions across three main domains:

• Individual Diagnostic and Self-Awareness Profiles: The instrument is designed to generate granular, individualized feedback reports. By identifying an individual’s idiosyncratic default patterns under conditions of heightened anxiety (e.g., a tendency to compulsively intellectualize or rigidly lock into positional demands), the CARS facilitates metacognitive awareness. This awareness is foundational to cultivating emotional regulation and cognitive agility during high-stakes negotiations.
• Executive Coaching and Organizational Mediation: In corporate and organizational settings, ombudspersons, leadership consultants, and mediators can deploy the CARS to diagnose subterranean team dysfunction. When coworkers experience conflict anxiety, polarized pairings (e.g., an extreme Minimizer interacting with an extreme Escalator) can produce systemic communication collapse. The scale provides a neutral, descriptive vocabulary that depathologizes natural defensive reactions while outlining clear pathways for behavioral de-escalation.
• Clinical and Marital Interventions: Within systemic family therapy and couples counseling, chronic interpersonal distress is frequently perpetuated by cyclical, automatic defensive maneuvers. Clinical practitioners can utilize the CARS to illuminate dynamic feedback loops—such as demand-withdraw cycles captured via the Conceal–Reveal or Avoid–Involve axes—enabling targeted therapeutic interventions aimed at desensitizing conflict-related anxiety.

5. Psychological Construct

The construct assessed by the CARS is multidimensional, capturing behavioral and psychological responses explicitly activated by the emotional experience of conflict anxiety. The architecture of the scale is organized around six bipolar dimensions encompassing twelve specific behavioral tendencies. Each tendency represents a coping mechanism deployed to mitigate the emotional discomfort generated by interpersonal confrontation:

Dimension 1: Avoidance versus Involvement (Avoid – Involve)

• Avoid: Characterized by behavioral flight, psychological withdrawal, procrastination, and physical departure from the conflict arena. Individuals scoring high on this tendency seek to diminish acute anxiety by evading confrontation, often leaving underlying structural grievances unresolved.
• Involve: Characterized by an urgent, compulsive need to confront and immerse oneself in the dispute immediately. Driven by an intolerance of ambiguous tension, high scorers insist on engaging the conflict instantly, often overwhelming their counterparts before relational conditions are conducive to dialogue.

Dimension 2: Soft versus Hard Bargaining (Soft – Hard)

• Soft: A compliance-oriented tendency marked by premature concessions, appeasement, and surrender of one’s own goals or interests. Driven by the fear of relational rupture or abandonment, individuals adopt submissive postures to preserve outward harmony.
• Hard: Characterized by an aggressive, positional, and unyielding posture. Under stress, high scorers attempt to quell their internal vulnerability through dominance, ultimatums, and coercive tactics aimed at forcing the other party into submission.

Dimension 3: Emotional Expression versus Intellectualization (Emotional – Intellectualize)

• Emotional: Manifests as affective volatility, visceral venting, and uncontrolled expressive outbursts. The individual’s cognitive filters are overridden by intense emotional states, resulting in raw behavioral displays that can escalate interpersonal friction.
• Intellectualize: Characterized by extreme cognitive detachment, hyper-rationality, and emotional suppression. Individuals decouple affective experience from the factual matrix of the dispute, relying on sterile logic, technical rules, or semantic pedantry to insulate themselves from affective exposure.

Dimension 4: Escalation versus Minimization (Escalate – Minimize)

• Escalate: Marked by catastrophizing, blowing disagreements out of proportion, and generalizing isolated disputes into existential conflicts. Anxiety is channeled into expansive threats, mobilization of outside allies, and expanding the scope of grievances.
• Minimize: Characterized by trivialization, denial, and reducing the perceived significance of substantial disputes. High scorers attempt to alleviate anxiety by dismissing grievances as unimportant (“It’s nothing, let’s just forget it”), thereby masking underlying resentments.

Dimension 5: Concealment versus Revelation (Conceal – Reveal)

• Conceal: Involves masking one’s authentic thoughts, needs, grievances, and affective states behind a facade of neutrality or indifference. Information is withheld as a protective shield against potential vulnerability or exploitation.
• Reveal: Characterized by reflexive oversharing, unsolicited transparency, and premature emotional exposure. The individual seeks relief from the cognitive burden of boundary management by indiscriminately disclosing internal states, regardless of tactical prudence.

Dimension 6: Rigidity versus Looseness (Rigid – Loose)

• Rigid: Characterized by behavioral dogmatism, obsessive adherence to predetermined procedures or moral frameworks, and cognitive inflexibility. The individual seeks certainty through strict structure and refusal to entertain adaptive compromises.
• Loose: Characterized by excessive permissiveness, chaotic boundary erosion, and an absence of normative structure or consistency. The individual adapts so fluidly to changing pressures that commitments lose stability and predictability.

6. Theoretical Framework

The theoretical bedrock of the CARS resides in Morton Deutsch‘s seminal 1993 treatise on educating conflict managers. Deutsch, one of the founding fathers of modern peace and conflict studies, hypothesized that the human capacity for constructive conflict resolution is systematically derailed by acute and chronic anxiety. He postulated that conflict induces psychological threat, activating defensive, primitive cognitive heuristics that pull individuals toward extreme, polarized modes of coping.

Deutsch proposed that these coping reactions map onto six fundamental, non-orthogonal bipolar dimensions of human behavioral functioning. Crucially, Deutsch hypothesized a curvilinear relationship between these dimensions and dispute outcomes: moderate, flexible positioning along any axis facilitates constructive resolution, whereas extreme, rigid gravitation toward either pole generates systemic dysfunction. For instance, neither absolute emotional avoidance nor compulsive emotional immersion is conducive to negotiation; rather, optimal dispute navigation demands the cognitive agility to modulate between engagement and reflection depending on situational demands.

Coleman and Chan (2023) synthesized Deutsch’s conceptual typology with contemporary affective science, social cognitive theory, and the Yerkes-Dodson Law. Under baseline emotional conditions, individuals may possess sophisticated theoretical knowledge of integrative bargaining and problem-solving communication. However, when subjective threat thresholds are breached, the neurobiological substrates of anxiety—mediated via the amygdala and autonomic nervous system—compromise prefrontal executive functioning. In this compromised state, the individual’s behavioral repertoire contracts.

By conceptualizing these defensive strategies as distinct behavioral derailers, Coleman and Chan’s framework integrates dual-process theories of cognition. The CARS evaluates the specific “Type 1” heuristic scripts that fire automatically in response to conflict anxiety, systematically preempting deliberate, adaptive “Type 2” conflict resolution strategies.

7. Validity

The psychometric validation of the CARS was conducted across an extensive sequence of seven iteratively structured studies using diverse adult participant pools across the United States. Validity evidence spans construct, convergent, discriminant, and predictive domains:

Construct and Structural Validity

Construct validity was established through systematic iterative item refinement. Beginning with broad item pools reflecting Deutsch’s (1993) definitions, items were subjected to sequential rounds of exploratory factor analysis (EFA) and confirmatory factor analysis (CFA). Factor intercorrelations demonstrated that while the 12 subscales represent distinct behavioral manifestations, they meaningfully align with their broader conceptual bipolar dimensions, confirming the construct validity of the multi-factor operationalization.

Predictive and Criterion-Related Validity

The predictive utility of the CARS was evaluated by examining how self-reported tendencies along the 12 subscales predicted real-world interpersonal and psychological outcomes. In line with Deutsch’s core proposition, empirical analyses revealed robust predictive validity: extreme scores across the bipolar dimensions correlated positively and significantly with heightened relationship dysfunction, decreased subjective well-being, increased chronic stress, and depressive affect.

Specifically, individuals displaying pronounced elevations in the Escalate, Rigid, and Hard subscales reported substantially higher frequencies of chronic, unresolved interpersonal hostility and litigation in both personal and workplace relationships. Conversely, extreme elevations on the Avoid, Conceal, and Soft subscales strongly predicted suppressed psychological agency, chronic somatic anxiety symptoms, and long-term relational dissatisfaction stemming from unmet personal needs.

Convergent and Discriminant Validity

Convergent validity was demonstrated through moderate to strong associations with validated psychological inventories measuring trait anxiety (State-Trait Anxiety Inventory [STAI]), general conflict mode orientations (Thomas-Kilmann Conflict Mode Instrument [TKI]), and emotional dysregulation scales. Concurrently, discriminant validity was confirmed by demonstrating that the CARS uniquely accounts for variance in conflict derailment beyond what is explained by general trait anxiety alone, confirming that conflict anxiety response tendencies represent a distinct, context-specific behavioral construct rather than generalized neuroticism.

8. Reliability

The CARS demonstrates robust reliability metrics across its brief and extended iterations, confirmed through systematic psychometric evaluations across seven independent empirical samples drawn from Amazon Mechanical Turk, Qualtrics Panels, and Prolific:

• Internal Consistency: Across the seven validation studies, Cronbach’s alpha coefficients for the final set of 12 subscales ranged from an acceptable 0.67 to a highly robust 0.895. Despite the intentionally brief nature of the subscales within the 22-item brief scale (where dimensions contain between one and three highly discriminating items), internal consistency was maintained at psychometrically sound levels.
• Item-Total Correlations: Subscale retention criteria required average inter-item correlations within subscales to meet conventional thresholds (typically $r > 0.40$), ensuring that items selected for the final 22-item model indexed cohesive psychological tendencies rather than fragmented, disparate behaviors.
• Measurement Stability: Subscale performance demonstrated consistency across varying demographic age brackets (ranging from 18 to 80 years) and between male and female adult respondents, supporting the generalizability and metric stability of the scale across diverse civilian populations within the United States.

9. Factor Analysis

The dimensional architecture of the CARS was rigorously calibrated using a two-tiered factor analytic strategy comprising both exploratory and confirmatory techniques across multiple large-scale participant samples:

Exploratory Factor Analysis (EFA)

Initial validation studies subjected the comprehensive item pool to an exhaustive series of 66 separate EFAs. Each EFA utilized principal axis factoring accompanied by promax oblique rotation, systematically evaluating every pairwise and multi-factor combination across the 12 hypothesized behavioral tendencies. This method enabled the researchers to detect cross-loadings, eliminate items exhibiting psychometric noise, and isolate the primary factor structures. However, aggressive initial item trimming led to some preliminary subscales retaining fewer than three items, necessitating controlled structural replication in subsequent studies.

Confirmatory Factor Analysis (CFA)

Following item reduction, structural equation modeling was executed across successive iterations to establish the most parsimonious model:

• Model Evaluation and Modification: In the baseline structural models, substantial latent factor intercorrelations introduced structural ambiguity. In Model 3, specific redundant inter-factor correlations were constrained to zero, and two underperforming items were trimmed from the inventory.
• Goodness-of-Fit Parameters: Model 3 demonstrated solid fit metrics: the Root Mean Square Error of Approximation (RMSEA) was 0.060, falling within the conventional threshold for acceptable fit ($≤ 0.06$ to $0.08$). The Comparative Fit Index (CFI) reached 0.83, and the Tucker-Lewis Index (TLI) registered at 0.82.
• Information Criteria: Model comparison metrics confirmed that Model 3 yielded the lowest absolute values for both the Akaike Information Criterion (AIC) and the Bayesian Information Criterion (BIC). This confirmed that a parsimonious structural model with trimmed latent correlations and pruned items offered the most robust configuration for empirical assessment.

10. Instrument / Measurement Tool

The technical parameters, administrative structure, and scoring characteristics of the Conflict Anxiety Response Scale are detailed below:

• Instrument Name: Conflict Anxiety Response Scale (CARS)
• Test Type: Original Psychological Inventory / Self-Report Questionnaire
• Target Population: General adult population (Ages 18–80), including young adults (18–29), thirty-somethings (30–39), middle-aged adults (40–64), and older adults (65+)
• Primary Application: Assessment of personal conflict-response tendencies, identification of behavioral derailers under anxiety, diagnostic input for negotiation training, mediation, and individualized feedback profiling
• Item Count: 22 items in the finalized recommended brief self-report inventory
• Subscale Breakdown: 12 specific subscales mapping onto 6 bipolar axes:
  • Avoid (Tendency to withdraw or evade confrontation)
  • Involve (Tendency to compulsively engage or force interaction)
  • Soft (Tendency to yield, appease, or concede prematurely)
  • Hard (Tendency to dominate, posture aggressively, or demand)
  • Emotional (Tendency toward affective dysregulation and venting)
  • Intellectualize (Tendency toward hyper-rational detachment and suppression)
  • Escalate (Tendency to magnify, catastrophize, or generalize disputes)
  • Minimize (Tendency to deny, downplay, or dismiss grievances)
  • Conceal (Tendency to hide authentic thoughts, feelings, or agendas)
  • Reveal (Tendency to overshare or disclose indiscriminately)
  • Rigid (Tendency toward dogmatism, stubbornness, and procedural obsession)
  • Loose (Tendency toward chaotic boundary erosion and excessive fluidity)
• Response Format: Items are rated on 5-point Likert scales
• Administration Time: Approximately 5 to 8 minutes for the 22-item brief scale
• Scoring and Interpretation: Subscale scores are calculated by averaging item responses associated with each of the 12 tendencies. The resulting profile highlights relative elevations across each pole of the 6 core dimensions, indicating whether a respondent leans toward functional flexibility or dysfunctional polarization when experiencing conflict anxiety.

11. Permissions & Fee and Test Year

• Publication Year: 2023
• Commercial Status: Non-commercial for academic and basic scientific research
• User Fee: None for qualified academic, educational, and research use
• Copyright and Permissions: The scale and its proprietary item inventory are copyrighted by the authors and the original publisher (Wiley Periodicals LLC on behalf of the Program on Negotiation at Harvard Law School). Researchers, clinicians, and practitioners seeking to reproduce, administer, or integrate the CARS into psychological studies or commercial executive training programs must formally contact the corresponding author:
• Correspondence Contact: Dr. Peter T. Coleman, Morton Deutsch International Center for Cooperation and Conflict Resolution (MD-ICCCR), Teachers College, Columbia University, New York, NY, United States. Email: [email protected].

12. References

• Coleman, P. T., & Chan, A. (2023). Conflict + anxiety = turmoil! Introducing a measure of conflict response derailers. Negotiation Journal, 39(2), 137–173. https://doi.org/10.1111/nejo.12427
• Deutsch, M. (1993). Educating for a peaceful world. American Psychologist, 48(5), 510–517. https://doi.org/10.1037/0003-066X.48.5.510
• Deutsch, M., Coleman, P. T., & Marcus, E. C. (Eds.). (2006). The handbook of conflict resolution: Theory and practice (2nd ed.). Jossey-Bass.
• Rahim, M. A. (1983). A measure of styles of handling interpersonal conflict. Academy of Management Journal, 26(2), 368–376. https://doi.org/10.2307/255985
• Thomas, K. W., & Kilmann, R. H. (1974). Thomas-Kilmann Conflict Mode Instrument. Xicom.

13. Items of the Scale