Relationship Scales Questionnaire (RSQ)

Abstract

The Relationship Scales Questionnaire (RSQ), developed by Don W. Griffin and Kim Bartholomew (1994), is an established self-report psychometric instrument designed to assess adult attachment styles within interpersonal and romantic contexts. Synthesizing item content from foundational adult attachment instruments—including Hazan and Shaver’s (1987) romantic attachment measure, Bartholomew and Horowitz’s (1991) Relationship Questionnaire (RQ), and Collins and Read’s (1990) Adult Attachment Scale (AAS)—the RSQ comprises 30 items evaluated on a 5-point Likert scale ranging from 1 (“Not at all like me”) to 5 (“Very much like me”). The instrument operationalizes attachment across Bartholomew’s four-category typology: Secure, Fearful-Avoidant, Preoccupied, and Dismissing-Avoidant, derived from orthogonal intersections of internal working models of the self (attachment anxiety) and internal working models of others (attachment avoidance). Although originally scored to yield four continuous prototype ratings, psychometric investigations, most notably by Kurdek (2002), strongly advocate for dimensional scoring reflecting the underlying higher-order axes of Model of Self (attachment anxiety) and Model of Other (attachment avoidance). Across clinical, developmental, and social psychological research, the RSQ demonstrates robust construct, convergent, and predictive validity, correlating systematically with interpersonal functioning, dyadic adjustment, emotion regulation strategies, and personality traits. While individual four-prototype subscales often demonstrate modest internal consistency estimates (α = .41 to .70) due to item brevity and semantic breadth, dimensional composite scores achieve satisfactory reliability (α = .75 to .86) and temporal stability over multi-month test-retest intervals. This paper provides a comprehensive psychometric review of the RSQ, detailing its theoretical foundation, structural validity, scoring methodologies, and clinical applicability.

Keywords

Relationship Scales Questionnaire, RSQ, adult attachment, attachment styles, internal working models, attachment anxiety, attachment avoidance, Bartholomew four-category model, interpersonal relationships, psychometrics

Authors

The Relationship Scales Questionnaire was formulated and psychometrically standardized by:

• Don W. Griffin, Ph.D. — Department of Psychology, University of British Columbia, Vancouver, British Columbia, Canada. Specialist in social cognition, interpersonal judgment, and statistical modeling in relationship science.
• Kim Bartholomew, Ph.D. — Department of Psychology, Simon Fraser University, Burnaby, British Columbia, Canada. Renowned developmental and social psychologist who pioneered the four-category adult attachment model and observational interview coding systems for adult relationship orientations.

Correspondence regarding the instrument’s original theoretical and developmental framework is historically affiliated with the Department of Psychology at Simon Fraser University.

Purpose

The primary purpose of the Relationship Scales Questionnaire is to measure adult attachment orientations continuously along empirically and theoretically validated dimensions. The RSQ addresses key methodological limitations of early categorical forced-choice paradigms in relationship science, which compelled respondents to self-assign into single attachment categories without accommodating continuous variability or blended behavioral orientations.

In research contexts, the RSQ is widely implemented to examine the influence of cognitive-affective representations of attachment on:

• Dyadic Relationship Quality: Quantifying how mutual attachment orientations predict relational satisfaction, stability, conflict resolution strategies, and communication dynamics in dating, cohabiting, and married couples.
• Affective and Stress Regulation: Evaluating how internal working models mediate physiological reactivity to interpersonal stress, distress disclosure, emotional suppression, and help-seeking behaviors.
• Psychopathology and Interpersonal Trauma: Exploring attachment-related vulnerability factors for major depressive disorder, generalized anxiety, borderline personality organization, and complex post-traumatic stress disorder (C-PTSD).
• Transitions Across the Lifespan: Investigating the stability and plasticity of attachment schemas during developmental milestones, such as the transition to university, marriage, parenthood, or divorce.

In clinical assessment, the RSQ serves as an efficient diagnostic adjunct. It illuminates a client’s implicit expectations regarding dependency, emotional intimacy, trust, and autonomy. Understanding whether a client predominantly manifests fearful avoidance versus dismissive detachment empowers psychotherapists to tailor the therapeutic alliance, anticipate alliance ruptures, navigate transference and countertransference dynamics, and formulate targeted interventions grounded in attachment-informed psychotherapy.

Psychological Construct

The RSQ operationalizes adult attachment through the structural synthesis of two fundamental cognitive-affective axes: the Model of Self and the Model of Other. Originally articulated by John Bowlby and conceptualized dimensionally by Bartholomew and Horowitz (1991), these representations correspond directly to Attachment Anxiety (the degree to which an individual worries about rejection, abandonment, or unworthiness of love) and Attachment Avoidance (the degree to which an individual fears emotional intimacy, suppresses dependency, and insists upon behavioral self-reliance).

The intersection of these two orthogonal dimensions yields four distinct attachment prototypes:

1. Secure Attachment (Positive Self, Positive Other)

Individuals scoring high on the Secure prototype possess a baseline sense of personal self-worth coupled with an expectation that other people are fundamentally trustworthy, reliable, and supportive. They comfortably engage in mutual emotional disclosure, establish adaptive intimacy without fear of engulfment, and accept interdependent reliance without anxiety over abandonment. Behaviorally, secure adults resolve interpersonal disputes constructively, assert boundaries without hostility, and solicit social support during acute distress.

2. Preoccupied Attachment (Negative Self, Positive Other)

Characterized by high attachment anxiety and low attachment avoidance, individuals high in the Preoccupied prototype maintain an impoverished evaluation of their own self-worth alongside an idealized, hyper-vigilant valuation of attachment figures. They experience pervasive apprehension regarding potential abandonment, emotional rejection, or lack of reciprocity. Consequently, they engage in hyperactivating strategies, including demanding reassurances, attempting complete emotional merger, and reacting with intense affective distress when proximity is perceived to be threatened.

3. Fearful-Avoidant Attachment (Negative Self, Negative Other)

Representing a confluence of high attachment anxiety and high attachment avoidance, the Fearful prototype embodies an acute approach-avoidance conflict. Such individuals deeply crave emotional intimacy, validation, and closeness, but simultaneously anticipate that interpersonal exposure will inevitably result in rejection, exploitation, or emotional devastation. Consequently, they defend against anticipated hurt through preemptive withdrawal, chronic distrust, and self-protective social inhibition, remaining persistently distressed and isolated.

4. Dismissing-Avoidant Attachment (Positive Self, Negative Other)

Defined by low attachment anxiety and high attachment avoidance, individuals exhibiting the Dismissing prototype construct an overt self-concept of self-sufficiency, invulnerability, and autonomy, while actively devaluing the necessity of close emotional relationships. They employ deactivating regulatory strategies, which involve the active cognitive suppression of attachment needs, denial of vulnerability, avoidance of intimacy, and a tendency to reject emotional dependence from romantic partners or peers.

Theoretical Framework

The theoretical bedrock of the RSQ originates in John Bowlby’s (1969, 1973, 1980) evolutionary-ethological Attachment Theory. Bowlby posited that infants possess an innate psychobiological motivational system designed to maintain proximity to primary caregivers for survival. Through repeated daily interactions, children internalize cognitive representations of these early caregiving dynamics, termed Internal Working Models (IWMs). These models encompass expectations regarding whether the self is worthy of care and whether caregivers are responsive and benevolent.

While Mary Ainsworth’s (1978) Strange Situation protocol classified infant attachment into three broad categories (Secure, Anxious-Ambivalent, and Anxious-Avoidant), early adult attachment researchers like Cindy Hazan and Phillip Shaver (1987) translated these categories into adult romantic orientations. However, Bartholomew (1990) and Bartholomew and Horowitz (1991) observed an essential theoretical limitation in Hazan and Shaver’s tripartite framework: it conflated two clinically and functionally distinct forms of adult avoidance.

Drawing directly upon Bowlby’s specification of dual internal working models, Bartholomew articulated that avoidance could stem either from a conscious defensive devaluation of intimacy (Dismissing style, rooted in an overt positive model of self and negative model of others) or from intense fear of rejection (Fearful style, rooted in a negative model of self and negative model of others). By framing these internal working models as continuous, bipolar axes (Self: Positive vs. Negative; Other: Positive vs. Negative), Griffin and Bartholomew (1994) established the 2 × 2 matrix that forms the conceptual architecture of the RSQ.

Validity

The psychometric validity of the RSQ has been evaluated across diverse demographic, cultural, and clinical samples over three decades:

Construct and Structural Validity

Construct validity was initially confirmed by Griffin and Bartholomew (1994) by evaluating the correspondence between self-reported RSQ scores and external criteria, including peer ratings, romantic partner ratings, and semi-structured Adult Attachment Interviews (AAI). Multitrait-multimethod (MTMM) matrix analyses indicated that latent variables for the self-model and other-model extracted from the RSQ converged significantly across informant perspectives (coefficients typically ranging from .35 to .58, p < .001).

Convergent Validity

The RSQ exhibits substantial convergent associations with alternative attachment inventories. The RSQ Model of Self (Anxiety) correlates strongly with the Anxiety subscale of the Experiences in Close Relationships (ECR; Brennan, Clark, & Shaver, 1998) inventory (r = .65 to .78). Similarly, the RSQ Model of Other (Avoidance) demonstrates moderate-to-high correlations with the ECR Avoidance subscale (r = .58 to .72). When compared against the Big Five personality dimensions, RSQ Anxiety positively tracks Neuroticism, whereas RSQ Avoidance correlates negatively with Extraversion and Agreeableness.

Discriminant and Predictive Validity

Discriminant validity is supported by the divergence between the Fearful and Dismissing subscales. Although both subscales reflect behavioral avoidance of closeness, Fearful attachment correlates significantly with measures of social anxiety, interpersonal dependency, self-criticism, and depressive symptomatology, whereas Dismissing attachment is unrelated or negatively related to subjective distress, instead predicting interpersonal coldness, narcissistic tendencies, and reduced empathy in dyadic interactions.

Reliability

The internal consistency of the RSQ subscales varies noticeably depending on whether researchers utilize the raw four-prototype subscales or the composite dimensional indices:

Internal Consistency (Cronbach’s Alpha)

Empirical studies typically report the following Cronbach’s alpha ranges for the four 4-to-5 item prototype subscales:

• Secure: α = .41 to .66
• Fearful: α = .64 to .76
• Preoccupied: α = .53 to .69
• Dismissing: α = .55 to .71

These relatively lower alpha values reflect the conscious construction of the RSQ: Griffin and Bartholomew designed the prototype scales using a small number of broad, heterogeneous items sampled from divergent historical measures (Hazan & Shaver, 1987; Collins & Read, 1990) to capture complex clinical phenotypes rather than narrow, homogenous traits.

Conversely, when items are aggregated into the higher-order continuous dimensions of Model of Self (Anxiety) and Model of Other (Avoidance), internal consistency increases substantially, with Cronbach’s α values consistently falling between .75 and .86 (Kurdek, 2002; Scharfe & Bartholomew, 1994).

Test-Retest Reliability

Longitudinal stability studies across non-clinical adult samples demonstrate stability over time. Scharfe and Bartholomew (1994) documented 8-month test-retest correlations ranging from .51 to .68 for both young adults and their romantic partners. Over shorter intervals (2 to 8 weeks), stability coefficients consistently range between .70 and .82, confirming that while attachment representations possess trait-like stability, they remain sensitive to significant relational transitions or therapeutic change.

Factor Analysis

Exploratory factor analyses (EFA) and confirmatory factor analyses (CFA) conducted across international samples consistently substantiate the presence of two robust, higher-order latent dimensions rather than four isolated, orthogonal factors:

Dimensional Extraction

Kurdek (2002) conducted rigorous factor analytic comparisons across large cohorts of heterosexual and homosexual cohabiting couples. EFA using principal axis factoring with oblimin and varimax rotations invariably extracts two primary factors accounting for the majority of variance:

• Factor 1: Attachment Avoidance (Model of Other) — Marked by strong positive loadings from items highlighting discomfort with closeness, distrust, and preference for self-reliance (e.g., Items 1, 12, 13, 20, 24, 26).
• Factor 2: Attachment Anxiety (Model of Self) — Marked by strong positive loadings from items tapping anxiety regarding unworthiness, abandonment, and abandonment hypervigilance (e.g., Items 9, 11, 16, 21, 23, 28).

Confirmatory Factor Analytic Fit

CFA evaluations modeling a four-factor structure corresponding to the four discrete prototypes frequently demonstrate suboptimal fit (e.g., CFI < .85, RMSEA > .08) due to shared variance and cross-loadings across prototypes (particularly between Fearful and Dismissing items regarding discomfort with closeness). In contrast, two-factor models (Anxiety vs. Avoidance) consistently demonstrate acceptable to superior fit indices (CFI ≥ .91, TLI ≥ .90, RMSEA ≤ .055), leading contemporary psychometricians to strongly endorse dimensional scoring over prototype isolation.

Instrument / Measurement Tool

• Test Type: Self-report psychological scale / continuous attachment questionnaire.
• Format: Paper-and-pencil or secure computer-administered digital assessment.
• Number of Items: 30 declarative statements.
• Administration Time: Approximately 5 to 10 minutes.
• Response Scale: 5-point Likert scale (1 = Not at all like me, 2 = Slightly like me, 3 = Somewhat like me, 4 = Moderately like me, 5 = Very much like me).
• Target Population: Adults and adolescents aged 16 years and older; applicable to general community, university, and clinical populations.
• Scoring Protocols:
  • Prototype Subscale Scoring:
    • Secure (5 items): Mean of items 3, 9 (reverse scored), 10, 15, and 28 (reverse scored).
    • Fearful (4 items): Mean of items 1, 5, 12, and 24.
    • Preoccupied (4 items): Mean of items 6 (or reversed in dimensional scoring), 8, 16, and 25.
    • Dismissing (5 items): Mean of items 2, 6 (reverse scored), 19, 22, and 26.
  • Dimensional Scoring (Recommended by Griffin & Bartholomew, 1994; Kurdek, 2002):
    • Model of Self (Anxiety): Calculated as (Fearful + Preoccupied) − (Secure + Dismissing). High scores reflect high attachment anxiety and self-doubt.
    • Model of Other (Avoidance): Calculated as (Dismissing + Fearful) − (Secure + Preoccupied). High scores reflect high attachment avoidance and interpersonal detachment.
  • Categorical Assignment: Categorical classification is methodologically discouraged. If strictly required, individual prototype means should be transformed into standardized z-scores, assigning participants to the prototype associated with their highest standardized value.

Permissions & Fee and Test Year

The Relationship Scales Questionnaire was published in 1994 by Don W. Griffin and Kim Bartholomew. The scale is in the public domain for non-commercial academic research and clinical assessment. No royalties, licensing fees, or formal permissions are required for non-commercial institutional use, provided proper scholarly citation is credited to Griffin and Bartholomew (1994). Diagnostic software distributors, commercial assessment publishers, or corporate practitioners seeking fee-based deployment must contact the copyright holders or representative institutional technology transfer offices.

References

• Ainsworth, M. D. S., Blehar, M. C., Waters, E., & Wall, S. (1978). Patterns of attachment: A psychological study of the strange situation. Lawrence Erlbaum Associates.
• Bartholomew, K. (1990). Avoidance of intimacy: An attachment perspective. Journal of Social and Personal Relationships, 7(2), 147–178. https://doi.org/10.1177/0265407590072001
• Bartholomew, K., & Horowitz, L. M. (1991). Attachment styles among young adults: A test of a four-category model. Journal of Personality and Social Psychology, 61(2), 226–244. https://doi.org/10.1037/0022-3514.61.2.226
• Bowlby, J. (1969). Attachment and loss: Vol. 1. Attachment. Basic Books.
• Bowlby, J. (1973). Attachment and loss: Vol. 2. Separation: Anxiety and anger. Basic Books.
• Bowlby, J. (1980). Attachment and loss: Vol. 3. Loss: Sadness and depression. Basic Books.
• Brennan, K. A., Clark, C. L., & Shaver, P. R. (1998). Self-report measurement of adult attachment: An integrative overview. In J. A. Simpson & W. S. Rholes (Eds.), Attachment theory and close relationships (pp. 46–76). Guilford Press.
• Collins, N. L., & Read, S. J. (1990). Adult attachment, working models, and relationship quality in dating couples. Journal of Personality and Social Psychology, 58(4), 644–663. https://doi.org/10.1037/0022-3514.58.4.644
• Griffin, D. W., & Bartholomew, K. (1994). Models of the self and other: Fundamental dimensions underlying measures of adult attachment. Journal of Personality and Social Psychology, 67(3), 430–445. https://doi.org/10.1037/0022-3514.67.3.430
• Hazan, C., & Shaver, P. (1987). Romantic love conceptualized as an attachment process. Journal of Personality and Social Psychology, 52(3), 511–524. https://doi.org/10.1037/0022-3514.52.3.511
• Kurdek, L. A. (2002). On being insecure about the assessment of adult attachment styles. Journal of Social and Personal Relationships, 19(6), 811–834. https://doi.org/10.1177/0265407502196005
• Scharfe, E., & Bartholomew, K. (1994). Reliability and stability of adult attachment representations. Personal Relationships, 1(1), 23–43. https://doi.org/10.1111/j.1475-6811.1994.tb00053.x

13. Items of the Scale (Questionnaire)