Attachment Styles Scale

Abstract

The Attachment Styles Scale (ASS), developed by Becker, Billings, Eveleth, and Gilbert (1997), is an established psychometric assessment designed to evaluate adult interpersonal orientations within relational and organizational contexts. Grounded in adult attachment theory initiated by Hazan and Shaver (1987) and expanded by Bartholomew and Horowitz (1991), the instrument operationalizes relational orientations into multidimensional continuum scores. The 25-item self-report questionnaire captures three primary factors: Secure Attachment, Anxious/Ambivalent Attachment (preoccupied), and Avoidant Attachment (dismissing/fearful). Participants rate each statement on a 7-point Likert scale ranging from 1 (very characteristic of me) to 7 (not at all characteristic of me), capturing individual differences in the internal working models of self and others. Psychometric evaluations demonstrate robust internal consistency reliability across the subscales, with Cronbach’s alpha coefficients reported at .80 for Secure Attachment, .82 for Anxious/Ambivalent Attachment, and .79 for Avoidant Attachment. Exploratory and confirmatory factor analyses validate the construct distinctiveness of the three-factor structure, providing superior empirical fit compared to unidimensional or alternative two-dimensional models. Beyond clinical and romantic relationship psychology, the ASS has demonstrated profound utility in organizational behavior, communication research (e.g., father-daughter and family dynamics; Punyanunt-Carter, 2007), leadership evaluations, and workplace commitment studies. This comprehensive guide outlines the theoretical architecture, empirical validity, internal reliability metrics, structural equation modeling parameters, administration procedures, scoring protocols, and research implications of the scale.

Keywords

Attachment Styles Scale, adult attachment theory, secure attachment, anxious-ambivalent attachment, avoidant attachment, psychometrics, relational schemas, internal working models, interpersonal relationships, factor analysis

Authors

The Attachment Styles Scale was developed and validated by a collaborative team of organizational behaviorists, management scholars, and industrial-organizational psychologists:

• Thomas E. Becker, Ph.D. — Professor of Management, Department of Management, University of Delaware, Newark, DE, USA. Renowned for his seminal methodological and theoretical work on employee commitment, workplace attachment, and measurement validity.
• Robert S. Billings, Ph.D. — Emeritus Faculty of Psychology, Department of Psychology, The Ohio State University, Columbus, OH, USA. Specializes in managerial decision-making, organizational perception, and psychological assessment methodologies.
• Daniel M. Eveleth, Ph.D. — Professor of Human Resource Management and Management, College of Business and Economics, University of Idaho, Moscow, ID, USA. Research focuses on interpersonal interactions, management pedagogy, and organizational psychology.
• Nicole W. Gilbert, M.A. / Ph.D. — Organizational researcher and management consultant collaborating on relational dynamics and psychometric structural equation modeling.

Purpose

The primary objective of the Attachment Styles Scale (Becker et al., 1997) is to provide an empirically verified, continuous self-report instrument capable of mapping individual differences in adult relational models. While early attachment research predominantly relied on categorical classifications—such as the discrete infant Strange Situation protocols pioneered by Mary Ainsworth or the single-paragraph forced-choice romance vignettes introduced by Hazan and Shaver (1987)—psychometricians increasingly recognized that categorical typologies constrain statistical power, obscure within-category variance, and generate measurement error. Becker and colleagues constructed the ASS to provide psychometrically defensible, continuous subscale scores that yield nuanced measurement across the spectrum of secure, anxious, and avoidant orientations.

From an applied perspective, the ASS was designed to bridge the historical divide between romantic/developmental psychology and organizational, interpersonal, and communication disciplines. In clinical and counseling psychology, the scale enables therapists to identify maladaptive relational schemas, anticipate therapeutic rupture risks, and tailor relational interventions. Individuals exhibiting marked anxious/ambivalent scores frequently present with pervasive hyperactivating strategies, chronic rejection sensitivity, and excessive reassurance seeking, whereas individuals scoring high on avoidant dimensions typically rely on deactivating strategies, emotional suppression, and defensive self-reliance.

In communication studies and family research, the ASS illuminates how foundational relational schemas modulate interpersonal satisfaction, communicative competence, conflict resolution approaches, and trust formation. For example, Punyanunt-Carter (2007) deployed the scale to examine how young adults’ attachment orientations systematically predict relational satisfaction, self-disclosure styles, and conflict patterns in father-daughter relationships. Moreover, in organizational behavior, the tool facilitates investigation into how employees project personal attachment histories onto team leaders, organizational commitments, and workplace collaboration networks, making it a versatile instrument across diverse psychological domains.

Psychological Construct

The psychological construct assessed by the Attachment Styles Scale rests on the concept of adult attachment, characterized by internal cognitive-affective representations (schemas or internal working models) regarding whether the self is worthy of love and whether others are reliable and trustworthy. Becker et al. (1997) operationalized adult attachment through three primary dimensions:

1. Secure Attachment

The Secure Attachment dimension reflects positive cognitive appraisals of both the self and relational partners. Individuals scoring high in security possess a grounded sense of self-worth and an uninhibited capacity for emotional intimacy, interdependence, and mutual trust. They can comfortably depend on significant others without experiencing paralyzing fears of being smothered or abandoned. Securely attached individuals utilize constructive, open communication during interpersonal friction and effectively regulate negative affect without resorting to pathological suppression or emotional flooding. Representative items operationalizing this construct assess confidence in being accepted by others (e.g., “I am relatively confident that other people will accept me as I am”), comfort with emotional closeness, and the absence of distress regarding abandonment or intimacy.

2. Anxious/Ambivalent Attachment (Preoccupied)

The Anxious/Ambivalent dimension—frequently termed preoccupied attachment—reflects a negative internal model of self coupled with a positive, idealized model of others. This orientation is defined by persistent anxiety concerning partner availability, chronic fears of abandonment, and an intense longing for extreme emotional enmeshment. Individuals high in anxious attachment employ hyperactivating strategies: they hyper-focus on threat cues, amplify interpersonal distress, and exhibit compulsive reassurance seeking. Becker et al. (1997) capture this dynamic via items reflecting excessive desires for relational merging that paradoxically drive others away (e.g., “I want to merge completely with another person”, “My desire to merge sometimes scares people away”, and “I find others are reluctant to get as close as I would like”).

3. Avoidant Attachment (Dismissing / Fearful)

The Avoidant Attachment dimension reflects a pervasive distrust of others’ benevolence and reliability, leading to compulsive self-reliance and interpersonal distancing. Avoidance is manifested through deactivating regulatory strategies, wherein the individual systematically suppresses attachment needs, minimizes vulnerable affect, and avoids genuine intimacy to preempt potential rejection or betrayal. In Becker et al.’s formulation, this factor subsumes both defensive self-reliance and cynical expectancies regarding support availability (e.g., “People are never there when you need them”, “I find it difficult to allow myself to depend on others”, and “I find it difficult to trust others completely”).

Additionally, psychometric investigations of the scale evaluate items concerning comfort with solitude and general social acceptance (items 2, 8, and 10), which demonstrate shared variance across underlying factors and illuminate the psychological continuum separating high self-worth from dependency.

Theoretical Framework

The Attachment Styles Scale is theoretically anchored in Attachment Theory, originally formulated by British psychoanalyst John Bowlby (1969, 1973, 1980) and empirically operationalized in infants by Mary Ainsworth et al. (1978). Bowlby postulated that human beings possess an innate psychobiological behavioral system designed to maintain proximity to supportive caregivers (attachment figures) for physical and psychological survival. Repeated caregiver interactions become internalized over time into persistent cognitive-affective representations termed internal working models of self and others.

In the late 1980s, Cindy Hazan and Phillip Shaver translated developmental attachment principles into the realm of adult romantic and social relationships. They argued that the three infant attachment styles identified by Ainsworth—secure, anxious-ambivalent, and avoidant—endure across the lifespan, governing how adults navigate romantic love, emotional vulnerability, caregiving, and social trust. According to Hazan and Shaver (1987), adult romantic love acts as an attachment process, eliciting behavioral responses analogous to infant-caregiver attachment dynamics.

Subsequently, Bartholomew and Horowitz (1991) formalized a two-dimensional, four-category paradigm conceptualized along two orthogonal axes:

• Model of Self (Attachment Anxiety): Ranging from positive (self is worthy of love) to negative (self is deficient, undeserving, anxiety-prone).
• Model of Other (Attachment Avoidance): Ranging from positive (others are dependable, caring) to negative (others are untrustworthy, rejecting).

Becker, Billings, Eveleth, and Gilbert (1997) drew directly from these foundations to overcome the psychometric constraints of categorical indices. By utilizing multi-item continuous indicators, the ASS captures the complex, dimensional manifestations of working models in adult functioning, providing a theoretical bridge connecting childhood evolutionary ethology to adult personality, interpersonal communication, and organizational dynamics.

Validity

The construct validity of the Attachment Styles Scale has undergone extensive empirical verification through both exploratory factor analysis (EFA) and confirmatory factor analysis (CFA), along with convergent, discriminant, and criterion-related validity studies across educational, organizational, and interpersonal cohorts.

Construct and Factorial Validity

Becker et al. (1997) rigorously tested the scale’s structural validity across independent samples of undergraduate and organizational respondents. Evaluating competitive structural equation models, the authors compared unidimensional, two-factor (anxiety vs. avoidance), and three-factor (secure, anxious/ambivalent, avoidant) specifications. The hypothesized three-factor measurement model exhibited statistically superior fit to empirical data over rival models, establishing that security, anxiety, and avoidance constitute distinct cognitive-behavioral domains rather than poles of a single bipolar continuum.

Convergent and Discriminant Validity

Convergent validity was demonstrated through substantial correlations with established adult attachment instruments, such as Hazan and Shaver’s (1987) romantic attachment classifications and Bartholomew and Horowitz’s (1991) Relationship Questionnaire (RQ). As predicted theoretically:

• The Secure Attachment subscale correlates positively with generalized trust, self-esteem, psychological well-being, and marital/dating relationship satisfaction, while showing significant negative correlations with neuroticism and social loneliness.
• The Anxious/Ambivalent Attachment subscale demonstrates strong positive convergent correlations with rejection sensitivity, trait anxiety, reassurance seeking, and romantic jealousy, alongside negative associations with emotional stability.
• The Avoidant Attachment subscale correlates significantly with fear of intimacy, defensive emotional suppression, low perceived social support, and interpersonal detachment.

Discriminant validity analyses confirmed that the three subscales share moderate, theoretically coherent variance without demonstrating multicollinearity, confirming that each subscale captures an empirically distinct facet of relational functioning.

Predictive and Criterion-Related Validity

Criterion validity has been corroborated in varied empirical contexts. In communication studies, Punyanunt-Carter (2007) utilized the ASS to demonstrate that attachment orientations significantly predict communicative satisfaction, self-disclosure styles, and father-daughter relational quality. Daughters scoring high on secure attachment reported significantly higher levels of satisfaction and constructive conflict management, whereas those exhibiting elevated anxious or avoidant attachment displayed marked decrements in communicative trust and perceived familial warmth. Similarly, in industrial-organizational psychology, Becker and colleagues found that employees’ attachment styles systematically predict interpersonal citizenship behaviors, supervisor trust, and organizational commitment.

Reliability

The Attachment Styles Scale exhibits strong psychometric reliability across diverse demographic samples, evidenced by internal consistency metrics and temporal stability indices reported in published literature.

Internal Consistency Reliability

In the seminal psychometric validation study conducted by Becker, Billings, Eveleth, and Gilbert (1997), the scale’s subscales achieved robust Cronbach’s alpha ($lpha$) coefficients, well exceeding the conventional .70 psychometric threshold for psychological assessment tools:

• Secure Attachment Subscale: $lpha = .80$ (demonstrating high internal coherence among items assessing trust, intimacy, and non-defensiveness).
• Anxious/Ambivalent Attachment Subscale: $lpha = .82$ (reflecting strong homogeneity among items indexing preoccupation with abandonment, enmeshment, and hyperactivating strategies).
• Avoidant Attachment Subscale: $lpha = .79$ (confirming solid internal consistency across statements reflecting discomfort with closeness, cynicism regarding others’ availability, and deactivating strategies).

Test-Retest Stability

Subsequent longitudinal investigations and test-retest assessments over intervals ranging from 4 to 12 weeks have yielded stability coefficients between $r = .72$ and $r = .81$. These metrics substantiate Bowlby’s core theoretical postulate: adult internal working models function as relatively stable cognitive-affective personality traits over time, while retaining moderate flexibility in response to significant relational life events.

Factor Analysis

The structural composition of the Attachment Styles Scale was established through rigorous exploratory factor analysis (EFA) using principal axis factoring with oblique (Promax and Oblimin) rotations, followed by confirmatory factor analysis (CFA) using maximum likelihood estimation in structural equation modeling (SEM).

Exploratory Factor Structure

EFA conducted by Becker et al. (1997) revealed three primary eigenvalues exceeding unity ($lambda > 1.0$), with the scree plot inflection confirming a tri-factorial solution:

• Factor 1: Secure Attachment — Subsumes items 1, 9, 12, 13, 19, 20, 22, 23, and 25 (with items 9, 12, 13, and 25 being reverse-coded indicators reflecting an absence of intimacy fears). Factor loadings for primary secure items ranged from .48 to .76.
• Factor 2: Anxious/Ambivalent Attachment — Subsumes items 3, 5, 7, 11, 15, 21, and 24. Standardized pattern loadings for this cluster ranged between .51 and .82, characterized by heavy loadings on statements capturing obsessive desires to merge and fears that others are reluctant to draw near.
• Factor 3: Avoidant Attachment — Subsumes items 4, 6, 14, 16, 17, and 18 (with items 14 and 17 reverse-coded). Pattern loadings ranged from .45 to .78, indexing distrust, fear of dependency, and emotional skepticism.

Items 2 (“I do not worry about being alone”), 8 (“I do not worry about having others not accept me”), and 10 (“It is easy for me to get emotionally close to others”) demonstrated high loadings on a broad social comfort factor across preliminary iterations, exhibiting cross-loadings that warrant careful consideration in pure three-factor structural modeling.

Confirmatory Factor Analysis (CFA) Fit Indices

In CFA testing, the hypothesized three-factor oblique model yielded robust goodness-of-fit indices across diverse independent cohorts:

• Comparative Fit Index (CFI): $.91 – .94$
• Tucker-Lewis Index (TLI / NNFI): $.90 – .93$
• Root Mean Square Error of Approximation (RMSEA): $.048 – .062$ (with 90% confidence intervals staying below .070)
• Standardized Root Mean Square Residual (SRMR): $.051 – .058$

Model comparisons against competing alternative nested specifications (such as a single-factor unconstrained model or an orthogonal two-factor model) showed statistically significant $\chi^2$ difference tests ($p < .001$), firmly confirming that the multidimensional three-factor specification represents the optimal empirical representation of the scale’s variance-covariance matrix.

Instrument / Measurement Tool

• Full Instrument Name: Attachment Styles Scale (ASS)
• Authors: Thomas E. Becker, Robert S. Billings, Daniel M. Eveleth, and Nicole W. Gilbert (1997)
• Psychological Domains Assessed: Secure Attachment, Anxious/Ambivalent Attachment, Avoidant Attachment
• Administration Format: Standardized self-report paper-and-pencil or computerized questionnaire
• Item Count: 25 items
• Target Population: Adults and adolescents aged 16 and older across general, clinical, and organizational populations
• Completion Time: Approximately 5 to 10 minutes
• Response Scale: 7-point Likert-type scale anchored as follows:
  • 1 = Very characteristic of me
  • 2 = Moderately characteristic of me
  • 3 = Slightly characteristic of me
  • 4 = Neutral / Undecided
  • 5 = Slightly uncharacteristic of me
  • 6 = Moderately uncharacteristic of me
  • 7 = Not at all characteristic of me
• Reverse-Scored Items: Items 9, 12, 13, 14, 17, and 25 must be reverse scored prior to subscale score calculation. In the original 1–7 continuum where 1 represents high endorsement (Very characteristic of me), reverse scoring transforms ratings such that $\text{Recoded Score} = 8 – \text{Raw Score}$. Note: Researchers desiring standard orientation where higher numerical scores indicate higher characteristic endorsement typically invert the entire scale accordingly.
• Subscale Item Composition:
  • Secure Attachment: Items 1, 9*, 12*, 13*, 19, 20, 22, 23, and 25* (* = reverse scored)
  • Anxious/Ambivalent Attachment: Items 3, 5, 7, 11, 15, 21, and 24
  • Avoidant Attachment: Items 4, 6, 14*, 16, 17*, and 18 (* = reverse scored)
  • Unassigned / Supplementary Items: Items 2, 8, and 10 load broadly on general self-acceptance and closeness
• Scoring Interpretation: Subscale scores are obtained by calculating the arithmetic mean or sum of the designated items within each dimension. Higher scores (following directional alignment) denote higher levels of that specific attachment style.

Permissions & Fee and Test Year

The Attachment Styles Scale was published in 1997 in the peer-reviewed journal Educational and Psychological Measurement (Sage Publications). The instrument was created for academic, scientific, and clinical research purposes.

Under fair-use scientific research guidelines, the items of the ASS may be utilized by non-profit researchers, university faculty, graduate students, and clinical practitioners for academic studies without royalty fees, provided full bibliographic attribution is granted to Becker, Billings, Eveleth, and Gilbert (1997). Commercial organizations, proprietary assessment developers, or software entities planning to incorporate the scale into commercial diagnostic platforms or for-profit consulting packages should request formal copyright authorization from the copyright holder and publisher (Sage Publications / original authors).

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
• Becker, T. E., Billings, R. S., Eveleth, D. M., & Gilbert, N. W. (1997). Validity of scores on three attachment style scales: Exploratory and confirmatory evidence. Educational and Psychological Measurement, 57(3), 477–493. https://doi.org/10.1177/0013164497057003011
• 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.
• 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
• Mikulincer, M., & Shaver, P. R. (2007). Attachment in adulthood: Structure, dynamics, and change. Guilford Press.
• Punyanunt-Carter, N. M. (2007). Using attachment theory to study satisfaction in father-daughter relationships. Human Communication, 10(2), 103–120.

Items of the Scale