Experiences in Close Relationships – Revised (ECR-R)

1. Abstract

The Experiences in Close Relationships – Revised (ECR-R) is a premier self-report psychometric instrument designed to assess individual differences with respect to adult romantic attachment. Developed by R. Chris Fraley, Niels G. Waller, and Kelly A. Brennan in 2000, the ECR-R represents an operational refinement of the original Experiences in Close Relationships (ECR) inventory through the application of sophisticated Item Response Theory (IRT) techniques. The instrument operationalizes romantic attachment across two continuous, orthogonal, and functionally independent dimensions: Attachment Anxiety and Attachment Avoidance. The inventory comprises 36 self-report items—18 per dimension—scored on a fully anchored 7-point Likert scale ranging from 1 (“Strongly disagree”) to 7 (“Strongly agree”).

Extensive psychometric investigations have established that the ECR-R demonstrates exemplary structural validity, exceptional internal consistency, and superior measurement precision across wide ranges of the underlying latent traits. Across diverse clinical and non-clinical populations, Cronbach’s alpha and McDonald’s omega coefficients regularly exceed .90 for both subscales (typically .91 to .95 for Anxiety and .90 to .94 for Avoidance). Confirmatory factor analytic investigations consistently support its stable two-factor structure across diverse demographic strata, cultures, and languages. By measuring attachment as continuous dimensional coordinates rather than discrete typological categories, the ECR-R prevents statistical artifactual attenuation, substantially mitigates measurement error, and enhances statistical power. As such, the ECR-R remains one of the most widely cited and psychometrically sound instruments in contemporary interpersonal, personality, social, and clinical psychology.

2. Keywords

Experiences in Close Relationships-Revised, ECR-R, adult attachment, attachment theory, attachment anxiety, attachment avoidance, item response theory, psychometrics, romantic relationships, internal working models

3. Authors

The Experiences in Close Relationships – Revised inventory was conceptualized, developed, and validated through the collaborative work of prominent psychometricians and personality psychologists:

• R. Chris Fraley, Ph.D. — Professor of Psychology at the University of Illinois at Urbana-Champaign. Dr. Fraley is an internationally recognized expert on attachment theory, personality development across the lifespan, and quantitative psychometric methodology, particularly the application of non-linear item response theory to psychological measurement.
• Niels G. Waller, Ph.D. — Professor of Quantitative Psychology and Psychometrics at the University of Minnesota. Dr. Waller has contributed fundamentally to factor analytic theory, computational statistics, latent variable modeling, and structural equation modeling.
• Kelly A. Brennan, Ph.D. — Research Psychologist whose foundational work with Phillip Shaver synthesized over sixty diverse adult attachment scales into the original Experiences in Close Relationships (ECR) instrument, establishing the definitive two-dimensional framework of adult romantic attachment.

4. Purpose

The primary purpose of the Experiences in Close Relationships – Revised (ECR-R) is to provide an empirical, highly reliable, and continuous dimensional assessment of adult romantic attachment orientations. Early operationalizations of attachment within adult relationships—such as the three-category forced-choice measure introduced by Hazan and Shaver (1987) or the four-category typological model advanced by Bartholomew and Horowitz (1991)—categorized individuals into discrete classes: secure, anxious-ambivalent (preoccupied), avoidant (dismissing), or fearful-avoidant. Although historically seminal, these categorical taxonomies suffered from severe psychometric limitations. Categorical typologies sacrifice massive amounts of individual variance, presuppose discontinuous boundaries where smooth gradients exist, exhibit suboptimal test-retest reliability near category thresholds, and arbitrarily reduce statistical power in predictive regression analyses.

To overcome these systemic issues, the ECR-R operationalizes attachment security not as a static category, but as a dynamic location within a two-dimensional continuous coordinate plane defined by Attachment Anxiety and Attachment Avoidance. Low scores on both dimensions define what is theoretically conceptualized as secure attachment. The theoretical rationale behind the ECR-R stems from John Bowlby’s proposition that attachment behavioral systems remain functional across the entire lifespan, continually guiding cognitive, emotional, and behavioral dynamics in close adult partnerships.

In research contexts, the ECR-R serves as the gold standard for testing hypotheses regarding dyadic conflict, affect regulation, relationship satisfaction, infidelity, dyadic communication, physiological stress reactivity (such as hypothalamic-pituitary-adrenal axis activation), and vulnerability to psychopathology. In applied clinical and counseling settings, the ECR-R aids practitioners in diagnosing interpersonal distress patterns, elucidating recurrent relationship dysfunctions, tracking therapeutic alliance rupture and repair dynamics, and systematically evaluating the outcomes of evidence-based interventions like Emotionally Focused Therapy (EFT) and attachment-informed cognitive behavioral therapies.

5. Psychological Construct

The ECR-R assesses two fundamental, conceptually distinct, yet interrelated latent constructs that govern the adult attachment behavioral system in romantic relationships: Attachment Anxiety and Attachment Avoidance.

Attachment Anxiety (Hyperactivation Strategies)

Attachment Anxiety captures individual differences in the extent to which an individual fears interpersonal abandonment, rejection, separation, or insufficient affection from their romantic partner. Individuals scoring high on this dimension operate under a persistent cognitive hypervigilance regarding partner availability, emotional accessibility, and responsiveness. Their internal working models of self are characterized by pervasive unworthiness, emotional vulnerability, and self-doubt, whereas their models of others tend to idealize partners as essential yet precarious sources of safety.

Behaviorally, high attachment anxiety triggers what attachment theorists classify as hyperactivating strategies. When confronted with perceived relationship stressors or perceived detachment, high-anxiety individuals amplify emotional distress, exhibit compulsive proximity-seeking, demand constant reassurance, and display intrusive monitoring behaviors. Sample items on the ECR-R capturing this construct include: “I often worry that my partner will not want to stay with me” and “When my partner is out of sight, I worry that he or she might become interested in someone else.”

Attachment Avoidance (Deactivation Strategies)

Attachment Avoidance measures individual differences in the degree to which a person experiences psychological discomfort with interpersonal intimacy, emotional dependency, closeness, and self-disclosure. Individuals scoring high on attachment avoidance maintain an excessively self-reliant stance and actively suppress vulnerability. Their internal working models of others are predominantly negative, viewing relationship partners as inherently unreliable, suffocating, untrustworthy, or intrusive, while their internal working models of self are defensively constructed around self-sufficiency and emotional invulnerability.

Behaviorally and affectively, high attachment avoidance manifests through deactivating strategies. Under relational stress or demands for intimacy, avoidant individuals deliberately divert attention away from attachment-related cues, suppress attachment-related memories, evade mutual self-disclosure, stonewall emotional communication, and withdraw physically or psychologically. Sample items on the ECR-R representing avoidance include: “I prefer not to show a partner how I feel deep down” and “I find it difficult to allow myself to depend on romantic partners.”

The Dimensional Matrix

Because these two dimensions are continuous and largely orthogonal, individuals can be located at any point within the bivariate space:

• Secure (Low Anxiety, Low Avoidance): Comfortable with emotional vulnerability, confident in partner support, capable of adaptive affect regulation and mutual interdependence.
• Preoccupied / Anxious (High Anxiety, Low Avoidance): Desiring extreme intimacy and validation, chronically plagued by fears of abandonment and unrequited commitment.
• Dismissing-Avoidant (Low Anxiety, High Avoidance): Compulsively self-reliant, dismissive of intimacy and vulnerability, emotionally detached.
• Fearful-Avoidant / Disorganized (High Anxiety, High Avoidance): Yearning for connection yet deeply terrified of intimacy and rejection, resulting in severe ambivalence and interpersonal paralysis.

6. Theoretical Framework

The ECR-R is firmly grounded in Attachment Theory, originally formulated by British psychoanalyst John Bowlby (1969, 1973, 1980) and empirically operationalized in infancy by developmental psychologist Mary Ainsworth (Ainsworth et al., 1978). Bowlby conceptualized attachment as an innate, evolutionarily conserved behavioral-motivational control system whose biological function is to maintain proximity between an infant and their primary caregiver (the attachment figure) to ensure protection from environmental hazards and physical predators.

Central to Bowlby’s formulation is the concept of Internal Working Models (IWMs)—cognitive representations composed of generalized expectations, declarative beliefs, and procedural rules concerning the self and primary attachment figures. These mental models crystallize through recurrent daily interactions during infancy and childhood. When caregivers are reliably available, warm, and responsive, the child internalizes a working model of the self as worthy of love and a model of others as benevolent and trustworthy. Conversely, inconsistent, rejecting, or neglectful caregiving fosters insecure internal working models characterized by self-doubt, hypervigilance, defensive independence, or profound mistrust.

In the late 1980s, Cindy Hazan and Phillip Shaver extended Bowlby’s developmental architecture into social and personality psychology by demonstrating that romantic partnerships among adults function as the primary attachment relationships of adult life. Romantic love represents an integration of three behavioral systems: attachment, caregiving, and sexuality. Adult partners become safe havens (sources of comfort and emotional soothing in times of threat) and secure bases (providing confidence and safety from which individuals can explore their external environments, careers, and personal growth).

In 1998, Kelly A. Brennan, Catherine L. Clark, and Phillip R. Shaver systematically pooled 482 individual items from 60 existing self-report adult attachment inventories, administered them to over 1,000 respondents, and factor-analyzed the pool. Their findings demonstrated that all self-report attachment items coalesce reliably into two broad, orthogonal dimensions: attachment-related anxiety and attachment-related avoidance. In 2000, Fraley, Waller, and Brennan refined this foundation through Nonparametric and Parametric Item Response Theory (IRT). They demonstrated that while the original ECR items were psychometrically robust, their measurement precision was unevenly distributed across the latent trait continuum. By employing IRT item information curves to select items with optimal discrimination and category threshold properties, Fraley et al. constructed the ECR-R, directly elevating adult attachment measurement to modern psychometric standards.

7. Validity

The ECR-R has undergone rigorous empirical validation across dozens of cultures, languages, and clinical populations, consistently demonstrating superior construct, convergent, discriminant, and predictive validity.

Construct and Factorial Validity

Item Response Theory analyses by Fraley, Waller, and Brennan (2000) demonstrated that the ECR-R items possess exceptionally high discrimination parameters (often $\alpha > 1.5$ and frequently reaching $> 2.0$), indicating that the items discriminate acutely between individuals exhibiting even slight differences on the underlying latent traits of anxiety and avoidance. Furthermore, the test information curves showed that the ECR-R provides high precision throughout the entire trait range (from $-3.0\sigma$ to $+3.0\sigma$), eliminating ceiling and floor effects that historically plagued older self-report attachment questionnaires.

Convergent Validity

Convergent validity is robustly established through strong correlations with complementary interpersonal and personality scales:

• Attachment Anxiety: Correlates strongly and positively with neuroticism ($r \approx .42$ to $.55$) on the NEO-PI-R, rejection sensitivity, trait anxiety on the STAI ($r > .50$), low self-esteem on the Rosenberg Self-Esteem Scale ($r \approx -.45$ to $-.60$), and emotional dependency.
• Attachment Avoidance: Correlates positively with alexithymia, fear of intimacy ($r \approx .60$), low agreeableness ($r \approx -.25$ to $-.35$), and negatively with interpersonal warmth, expressiveness, and empathy.

Discriminant Validity

Despite meaningful correlations with the Big Five personality traits (most notably neuroticism), multi-trait multi-method (MTMM) analyses and hierarchical structural equation modeling confirm that the ECR-R explains substantial unique variance in relationship outcomes after controlling for general personality dimensions. The intercorrelation between the two ECR-R subscales itself remains minimal to low ($r$ typically ranges from $.08$ to $.25$), demonstrating that anxiety and avoidance represent distinct, functionally uncoupled psychometric dimensions rather than opposite poles of a single continuum.

Predictive and Ecological Validity

The predictive utility of the ECR-R has been corroborated across longitudinal dyadic studies, observational lab paradigms, and experimental neurobiological investigations:

• Observational Interaction Paradigms: Simpson, Rholes, and Phillips (2001) demonstrated that baseline ECR-R avoidance scores predict observed behavioral withdrawal, emotional unresponsiveness, and lack of supportive physical contact when partners undergo acute laboratory-induced stress.
• Dyadic Satisfaction and Longitudinal Stability: In longitudinal studies of newlywed and cohabiting couples, high scores on either ECR-R subscale prospectively predict steep declines in relationship satisfaction, lower perceived partner responsiveness, increased conflict escalation, and significantly elevated rates of marital dissolution over 2- to 5-year follow-ups.
• Psychophysiological and Neurobiological Correlates: Functional magnetic resonance imaging (fMRI) studies show that high ECR-R anxiety scores correlate with heightened anterior cingulate cortex and amygdala reactivity to social rejection cues, whereas avoidant scores correlate with down-regulation of ventral striatal reward pathways during intimate partner interactions.

8. Reliability

The ECR-R demonstrates consistently outstanding reliability across both internal consistency and temporal stability metrics, making it one of the most reliable self-report inventories in social and personality science.

Internal Consistency

In the seminal validation study by Fraley, Waller, and Brennan (2000), internal consistency coefficients were exceptionally high:

• Attachment Anxiety: Cronbach’s $\alpha = .93$ to $.95$; McDonald’s $\omega_h > .88$.
• Attachment Avoidance: Cronbach’s $\alpha = .91$ to $.94$; McDonald’s $\omega_h > .87$.

Independent cross-cultural validation studies across diverse languages (including Spanish, French, Italian, German, Turkish, Chinese, and Japanese) confirm that internal consistency rarely drops below $.89$ for either subscale. These elevated alphas are not an artifact of mere item redundancy; IRT analyses confirm that the items span distinct severity thresholds across the latent continuum while sharing uniform factor saturations.

Test-Retest Stability

Temporal stability assessments have shown that while romantic attachment orientations can adaptively change in response to profound life events or sustained therapeutic interventions, individual ECR-R scores exhibit remarkable rank-order consistency over time:

• Short-term stability (3 weeks to 2 months): Test-retest correlation coefficients range between $r = .86$ and $r = .92$ for both anxiety and avoidance, demonstrating high stability against temporary mood fluctuations.
• Medium-term stability (6 months to 1 year): Stability coefficients remain high, consistently falling between $r = .70$ and $r = .80$.
• Long-term stability (2 to 4 years): Longitudinal studies track test-retest correlations between $r = .55$ and $r = .68$, indicating that while adult attachment functions as a stable trait-like disposition, it retains sufficient open-system malleability to register major relationship transitions, relationship loss, or deliberate psychological restructuring.

9. Factor Analysis

The structural integrity and latent architecture of the ECR-R have been extensively scrutinized through both Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA).

Item Response Theory and Exploratory Findings

During the design of the ECR-R, Fraley and colleagues simulated and empirically evaluated thousands of potential item configurations from Brennan et al.’s original 323-item pool using Samejima’s graded response model (a parametric IRT model for polytomous items). EFA with oblique and varimax rotations invariably yielded a dominant, clear two-factor solution accounting for the vast majority of common variance. The scree test, parallel analysis, and eigenvalues cleanly separated the 18 anxiety items and 18 avoidance items onto their designated target factors, with primary factor loadings typically ranging between $.60$ and $.85$, and cross-loadings universally remaining below $.20$.

Confirmatory Factor Analysis (CFA) and Model Fit

Subsequent independent validation studies (e.g., Sibley et al., 2005; Fairchild & Finney, 2006; Cameron et al., 2012) have confirmed that the two-factor oblique model offers superior fit compared to single-factor or higher-order hierarchical models. Typical goodness-of-fit indices obtained in large-sample CFAs satisfy strict psychometric standards:

• Root Mean Square Error of Approximation (RMSEA): Values typically range from $.045$ to $.062$ (indicating good to close fit).
• Comparative Fit Index (CFI): Values consistently range between $.92$ and $.97$.
• Tucker-Lewis Index (TLI): Values reliably exceed $.91$ to $.96$.
• Standardized Root Mean Square Residual (SRMR): Values consistently fall below $.055$.

Measurement Invariance

Extensive multi-group confirmatory factor analyses (MGCFA) have demonstrated configural, metric (weak), and scalar (strong) invariance across biological sexes (men vs. women), sexual orientations, and relationship status (dating vs. married). Scalar invariance establishes that differences in observed scores accurately reflect true latent mean differences rather than differential item functioning (DIF) or measurement bias.

10. Instrument / Measurement Tool

• Instrument Name: Experiences in Close Relationships – Revised (ECR-R)
• Authors: R. Chris Fraley, Niels G. Waller, and Kelly A. Brennan (2000)
• Test Type: Self-report psychometric rating scale / dimensional inventory
• Target Population: Adults and older adolescents (ages 16+) who have current or past romantic relationship experience
• Administration Time: Approximately 8 to 12 minutes
• Item Count: 36 items total
  • Attachment Anxiety Subscale: 18 items (Items 1 to 18)
  • Attachment Avoidance Subscale: 18 items (Items 19 to 36)
• Authentic Response Scale: 7-point Likert scale:
  • 1 = Strongly disagree
  • 2 = Disagree
  • 3 = Slightly disagree
  • 4 = Neutral / mixed
  • 5 = Slightly agree
  • 6 = Agree
  • 7 = Strongly agree
• Scoring and Computational Rules:
  • Reverse-Scored Items on Attachment Anxiety: Item 9 and Item 11 must be reverse-scored prior to calculating subscale metrics: $(8 – \text{Raw Score})$.
  • Reverse-Scored Items on Attachment Avoidance: Items 20, 22, 26, 27, 28, 31, 33, 34, 35, and 36 must be reverse-scored prior to calculating subscale metrics: $(8 – \text{Raw Score})$.
  • Subscale Scores: Computed as the unweighted arithmetic mean of all 18 items within each dimension after reverse-scoring. Anxiety Score = $\frac{\sum(\text{Items } 1-18)}{18}$; Avoidance Score = $\frac{\sum(\text{Items } 19-36)}{18}$. This preserves the original 1 to 7 interpretative metric.
  • Categorical Typology Mapping (Optional): While continuous analyses are strongly recommended to preserve variance, researchers can classify participants into Bartholomew’s four quadrants using sample medians or absolute scale midpoints (score of 4.0):
    • Secure: Low Anxiety (< 4.0) and Low Avoidance (< 4.0)
    • Preoccupied: High Anxiety (≥ 4.0) and Low Avoidance (< 4.0)
    • Dismissing: Low Anxiety (< 4.0) and High Avoidance (≥ 4.0)
    • Fearful: High Anxiety (≥ 4.0) and High Avoidance (≥ 4.0)

11. Permissions & Fee and Test Year

The Experiences in Close Relationships – Revised (ECR-R) was originally published in 2000 in the Journal of Personality and Social Psychology. In alignment with open scientific practices and the academic mission of its developers, the ECR-R is placed in the public domain for academic, non-commercial research, educational, and clinical evaluation purposes.

Fee Structure: There are no licensing fees, royalties, or costs required to administer the scale for academic research, doctoral dissertations, or routine clinical assessment. Researchers and clinicians may reproduce, digitize (e.g., via Qualtrics, REDCap, or Google Forms), and translate the instrument freely without seeking formal written permission, provided that proper scholarly attribution is given to the authors (Fraley, Waller, & Brennan, 2000). Commercial software vendors seeking to incorporate the ECR-R into proprietary commercial platforms or fee-for-service executive assessments should contact Dr. R. Chris Fraley directly for permission guidelines.

12. 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., & 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.
• Cameron, J. J., Finnegan, H., & Morry, M. M. (2012). Orthogonal dreams: Testing the factor structure and measurement invariance of the Experiences in Close Relationships-Revised scale. Interpersona, 6(1), 16–31. https://doi.org/10.5964/ijpr.v6i1.87
• Fairchild, A. J., & Finney, S. J. (2006). Investigating validity evidence for the Experiences in Close Relationships-Revised questionnaire. Educational and Psychological Measurement, 66(1), 116–135. https://doi.org/10.1177/0013164405278564
• Fraley, R. C., Waller, N. G., & Brennan, K. A. (2000). An item-response theory analysis of self-report measures of adult attachment. Journal of Personality and Social Psychology, 78(2), 350–365. https://doi.org/10.1037/0022-3514.78.2.350
• Hazan, C., & Shaver, P. R. (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
• Mikulincer, M., & Shaver, P. R. (2016). Attachment in adulthood: Structure, dynamics, and change (2nd ed.). Guilford Press.
• Sibley, C. G., Fischer, R., & Liu, J. H. (2005). Reliability and validity of the Experiences in Close Relationships-Revised (ECR-R) adult attachment questionnaire. Personality and Social Psychology Bulletin, 31(11), 1524–1536. https://doi.org/10.1177/0146167205276865
• Simpson, J. A., Rholes, W. S., & Phillips, D. (2001). Conflict in close relationships: An attachment perspective. In J. A. Simpson & W. S. Rholes (Eds.), Attachment theory and close relationships (pp. 200–225). Guilford Press.

13. Items of the Scale