Eating Disorder Belief Questionnaire (EDBQ)

Abstract

The Eating Disorder Belief Questionnaire (EDBQ) is an established 32-item psychometric instrument designed to assess the underlying cognitive architecture, core beliefs, and conditional assumptions characteristic of eating disorders, including anorexia nervosa, bulimia nervosa, and other specified feeding or eating disorders. Developed by Myra Cooper, Gillian Todd, and Adrian Wells in 1998, the instrument operationalizes cognitive-behavioral formulations that distinguish between surface-level automatic thoughts and deeper schema structures. Factor-analytic investigations reveal a robust four-factor latent structure: (1) Negative Self-Beliefs, (2) Weight and Shape as Means to Self-Acceptance, (3) Weight and Shape as Means to Acceptance by Others, and (4) Control Over Eating. Unlike conventional Likert-type scales, the EDBQ utilizes a continuous visual analogue rating format ranging from 0 (“I don’t usually believe this at all”) to 100 (“I am usually completely convinced that this is true”), specifically instructing respondents to rate their visceral, “emotional” conviction rather than rational or intellectual endorsement. The EDBQ demonstrates exceptional internal consistency (Cronbach’s α typically ranging from .87 to .94 across clinical and non-clinical cohorts), strong test-retest stability (r = .80 to .92), and high convergent, discriminant, and construct validity. By capturing unconditional negative self-evaluations alongside the conditional rules linking body shape and dietary control to self-worth, the EDBQ serves as an indispensable tool in clinical diagnosis, cognitive-behavioral case formulation, schema therapy, and empirical research.

Keywords

Eating Disorder Belief Questionnaire, EDBQ, cognitive schemas, core beliefs, conditional assumptions, anorexia nervosa, bulimia nervosa, cognitive behavioral therapy, psychometrics, body image, dietary restraint, schema therapy.

Authors

The Eating Disorder Belief Questionnaire was developed by a team of clinical psychologists and cognitive-behavioral researchers specializing in eating pathology and cognitive theory:

• Myra Cooper, DPhil, DipClinPsych — Department of Clinical Psychology, The Isis Education Centre, Warneford Hospital, University of Oxford, Oxford, United Kingdom. Dr. Cooper is renowned for her pioneering work modeling core beliefs, cognitive biases, and schema-level processing across the spectrum of eating disorders.
• Gillian Todd, MPhil, MSc — Department of Clinical Psychology, Addenbrooke’s NHS Trust, Cambridge, and University of East Anglia, Norwich, United Kingdom. An expert clinician and clinical trialist focused on specialized cognitive-behavioral protocols for eating disorders.
• Adrian Wells, PhD — Professor of Clinical and Experimental Psychopathology, Division of Psychology and Mental Health, University of Manchester, Manchester, United Kingdom. Renowned as the originator of Metacognitive Therapy (MCT) and a leading authority on the role of cognitive and metacognitive beliefs in emotional disorders.

Purpose

The primary clinical and empirical purpose of the Eating Disorder Belief Questionnaire (EDBQ) is to capture, quantify, and delineate the deeper cognitive schemas—specifically unconditional core beliefs and conditional assumptions—that maintain eating pathology. Traditional screening and assessment instruments, such as the Eating Attitudes Test (EAT-26) or the Eating Disorder Examination Questionnaire (EDE-Q), focus predominantly on surface-level cognitive phenomena (e.g., conscious preoccupation with calories, explicit weight dissatisfaction) and behavioral symptoms (e.g., purging, dietary restriction, objective binge eating episodes). While tracking behavioral frequencies is essential for diagnostic classification, such instruments often fail to illuminate the cognitive vulnerability factors that trigger and perpetuate these maladaptive behaviors.

The EDBQ was specifically engineered to bridge this diagnostic and therapeutic gap. Drawing upon Aaron Beck’s cognitive therapy paradigm and modern schema theories, the scale evaluates the cognitive mechanisms through which individuals link physical morphology and appetite control to fundamental existential worth. Specifically, the EDBQ operationalizes two primary cognitive levels: general unconditional negative self-beliefs (such as feelings of defectiveness, failure, and unlovability) and intermediate conditional assumptions (such as the belief that achieving a thin physique is a prerequisite for interpersonal acceptance or internal self-worth).

In clinical practice, the EDBQ serves several distinct functions:

• Individualized Cognitive Case Formulation: It allows clinicians conducting Cognitive Behavioral Therapy (CBT) or Schema-Focused Therapy to rapidly identify idiosyncratic intermediate rules and unconditional self-schemas, mapping the precise links between patients’ negative self-evaluations and their restrictive or compensatory eating behaviors.
• Targeting “Hot” Emotional Cognitions: By explicitly instructing patients to record their “gut-level” emotional conviction (how they feel deep down) rather than their rational, intellectual perspective, the EDBQ accesses emotionally charged cognitive schemas that drive behavior under distress, bypassing defensive intellectualization.
• Treatment Monitoring and Relapse Prevention: Clinical trials demonstrate that while surface symptoms may remit early in treatment, underlying negative self-beliefs and conditional rules can persist as residual vulnerabilities. Monitoring EDBQ trajectory enables therapists to track schema modification throughout therapy, serving as a reliable predictor of long-term recovery versus relapse.

Psychological Construct

The psychological construct assessed by the EDBQ is the hierarchical cognitive belief system that constitutes the cognitive vulnerability architecture of eating disorders. Cognitive theory posits that human thinking is organized on multiple levels: voluntary automatic thoughts at the periphery, intermediate conditional assumptions and rules in the middle, and deep, unconditional core beliefs at the schema core. The EDBQ operationalizes these constructs into four distinct, empirically verified dimensions:

1. Negative Self-Beliefs (Items 1–10)

This subscale captures pervasive, unconditional, global negative core beliefs regarding the self. These beliefs reflect deep-seated feelings of defectiveness, incompetence, social unworthiness, and fundamental inadequacy. Items such as “I’m stupid,” “I’m a failure,” “I’m useless,” “I don’t like myself very much,” and “I’m unlovable” do not mention food, weight, or body shape. Instead, they capture the primary schema deficits that characterize individuals vulnerable to severe psychopathology. In individuals with eating disorders, these negative core beliefs represent the psychological baseline of distress from which compensatory mechanisms arise.

2. Weight and Shape as Means to Self-Acceptance / Self-Esteem (Items 11–15, 21–23)

This dimension operationalizes conditional, instrumental assumptions connecting body shape, firmness, and weight directly to self-worth, internal competence, and moral value. Examples include: “If my thighs are firm it means I’m a better person,” “If my hips are narrow it means I’m successful,” “If I gain weight it means I’m a bad person,” and “If my body is lean I can feel good about myself.” Within this construct, thinness and physical firmness are not merely aesthetic goals; they serve as moralized metrics of personal virtue and self-efficacy. Weight gain is catastrophized as total moral degradation, confirming the primary negative core belief that one is worthless or “nothing.”

3. Weight and Shape as Means to Acceptance by Others (Items 16–20)

This subscale assesses external, interpersonal conditional assumptions regarding relational security, social inclusion, and affection. The construct reflects beliefs that interpersonal warmth, validation, and protection from rejection are strictly conditional upon physical thinness. Exemplified by items such as “If my body shape is in proportion people will love me,” “If my hips are thin people will approve of me,” “If I lose weight people will care about me,” and “If I lose weight I’ll count more in the world,” this dimension represents the patient’s coping strategy to mitigate intense fears of abandonment, ostracism, or interpersonal unacceptability.

4. Control Over Eating (Items 24–32)

This construct assesses rigid, dichotomous beliefs, rules, and catastrophic predictions centered on food intake, appetite regulation, and purging behaviors. Items measure cognitive distortions regarding metabolic processes, moralized categorizations of food, and the functional utility of compensatory behaviors (e.g., “If I eat desserts or puddings I’ll get fat,” “Body fat/flabbiness is disgusting,” “If I binge and vomit I can stay in control,” “If I stay hungry I can guard against losing control and getting fat,” and “If I eat a forbidden food I won’t be able to stop”). Here, eating restraint and compensatory rituals (vomiting, dietary elimination) are cognitively framed as non-negotiable defensive strategies necessary to prevent catastrophic loss of physical and psychological self-regulation.

Theoretical Framework

The EDBQ is grounded in the cognitive-behavioral conceptualization of emotional disorders established by Aaron T. Beck (1976), further elaborated for eating pathology by Myra Cooper (1997, 2005) and integrated with Christopher Fairburn’s transdiagnostic cognitive-behavioral model (Fairburn et al., 2003).

Beck’s Hierarchy of Cognition

Beck’s foundational paradigm articulates a tripartite cognitive hierarchy:

1. Core Beliefs / Schemas: Deeply ingrained, rigid, global cognitive structures developed during early development that filter and process environmental stimuli (e.g., “I am unlovable”).
2. Intermediate Beliefs (Assumptions, Rules, and Attitudes): Conditional propositions, often structured as “If… then…” statements, that serve as compensatory strategies designed to navigate the threat posed by active negative core beliefs.
3. Automatic Thoughts: Involuntary, situation-specific cognitions that flash across awareness in response to immediate environmental triggers.

Cooper’s Specific Cognitive Model of Eating Disorders

Building upon Beck’s architecture, Cooper (1997) proposed a comprehensive cognitive model explaining how general schemas interact with disorder-specific schemas. Cooper asserted that individuals who develop eating disorders possess underlying, unconditional negative self-beliefs (defectiveness, failure, unlovability) rooted in developmental experiences of invalidation, trauma, or adverse attachment. Because experiencing unconditional worthlessness generates unbearable psychological pain, the individual develops secondary, conditional assumptions that bridge physical appearance and self-worth:

“If I can control my body, shape, and weight, then I will be acceptable to myself and others; if I fail to maintain this control, my worthlessness is confirmed.”

Under this theoretical framework, dietary restriction, compulsive exercise, weighing rituals, and body checking become cognitive and behavioral safety maneuvers designed to sustain the conditional assumption. Binge eating, purging, and subjective weight fluctuations trigger the primary schema of defectiveness, reinforcing the negative self-beliefs and closing the perpetuating cycle. The EDBQ directly measures both the fundamental core tier (Factor 1) and the intermediate compensatory tier (Factors 2, 3, and 4), rendering the cognitive etiology transparent.

Emotional vs. Rational Cognitive Beliefs

A critical theoretical premise of the EDBQ is the distinction between rational (intellectual) and emotional (visceral) belief ratings. Patients with eating disorders frequently state in clinical sessions: “Rationally, I know eating a piece of bread won’t make me gain five pounds instantly, but emotionally, I feel absolutely convinced that it will.” Classical psychometric measures that ask for general agreement capture the patient’s rational awareness, diluting clinical sensitivity. The EDBQ explicitly instructs patients to evaluate their conviction based on how they feel “deep down” (hot cognitions), successfully bypassing intellectualized defense mechanisms and isolating the active schemas driving behavioral impairment.

Validity

The psychometric validity of the Eating Disorder Belief Questionnaire has been comprehensively evaluated across clinical cohorts (including inpatients and outpatients with anorexia nervosa, bulimia nervosa, and binge eating disorder) and non-clinical comparison groups across multiple international investigations.

Construct and Factorial Validity

In the seminal validation study by Cooper, Todd, and Wells (1998), the four-factor solution was derived through exploratory factor analysis and accounted for 54.8% of the total variance. Subsequent confirmatory factor analyses across diverse linguistic and clinical samples (e.g., European, North American, and Australian studies) have repeatedly confirmed the structural integrity of this four-dimensional model, yielding acceptable to excellent fit indices (e.g., Comparative Fit Index [CFI] > .90; Root Mean Square Error of Approximation [RMSEA] ≤ .06; Standardized Root Mean Square Residual [SRMR] ≤ .07).

Convergent Validity

Convergent validity is supported by robust positive correlations with concurrent measures of eating pathology, body image distress, and general psychological dysfunction:

• The EDBQ total score and subscales demonstrate large, statistically significant correlations with the Eating Disorder Examination Questionnaire (EDE-Q) global score (r = .65 to .78, p < .001) and specific subscales such as Shape Concern and Weight Concern (r > .70).
• Significant correlations have been documented with the Eating Disorder Inventory (EDI-2 / EDI-3) subscales, particularly Drive for Thinness (r = .68 to .81) and Body Dissatisfaction (r = .60 to .75).
• The Negative Self-Beliefs subscale exhibits strong convergent associations with the Beck Depression Inventory (BDI-II) (r = .62 to .74) and the Rosenberg Self-Esteem Scale (RSES) (inverting at r = -.65 to -.76), verifying that it accurately assesses global dysphoria and pervasive self-depreciation.

Discriminant and Known-Groups Validity

The EDBQ displays remarkable discriminant validity, successfully separating clinical from non-clinical cohorts. Cooper, Todd, and Wells (1998) demonstrated that clinical patients with anorexia nervosa and bulimia nervosa scored significantly higher across all four EDBQ subscales compared to non-clinical dieters and non-clinical non-dieters (F-values exceeding 40.0, p < .0001). Importantly, the instrument successfully differentiated dieters from clinical eating disorder patients: while non-clinical dieters endorsed moderate levels of control over eating, they scored dramatically lower than clinical patients on Negative Self-Beliefs and Weight and Shape as Means to Self-Acceptance. This confirms that unconditional self-defectiveness and the moralization of shape are unique hallmarks of clinical eating pathology rather than normative dietary restraint.

Reliability

The EDBQ exhibits exceptional reliability across multiple empirical parameters, demonstrating internal consistency, item-total homogeneity, and temporal stability over repeated administrations.

Internal Consistency

Across validation studies involving clinical populations (inpatients and outpatients with eating disorders) and non-clinical female samples, the EDBQ has demonstrated high internal consistency coefficients (Cronbach’s alpha):

• Full Scale Total Score (32 items): α = .94 to .97
• Negative Self-Beliefs (10 items): α = .89 to .94
• Weight/Shape as Means to Self-Acceptance (8 items): α = .88 to .93
• Weight/Shape as Means to Acceptance by Others (5 items): α = .85 to .91
• Control Over Eating (9 items): α = .83 to .89

Item-to-subscale correlations generally range from .52 to .84, with no single item removal resulting in an increase in the composite alpha coefficient, confirming exceptional internal homogeneity.

Test-Retest Stability

Temporal stability assessments over intervals ranging from 14 days to 4 weeks in stable clinical and non-clinical populations demonstrate high test-retest reliability:

• Negative Self-Beliefs: r = .86 to .91 (p < .001)
• Weight/Shape as Means to Self-Acceptance: r = .84 to .89 (p < .001)
• Weight/Shape as Means to Acceptance by Others: r = .80 to .87 (p < .001)
• Control Over Eating: r = .82 to .88 (p < .001)
• EDBQ Global Score: r = .88 to .93 (p < .001)

These findings substantiate that the EDBQ measures enduring cognitive schemas and trait-like belief systems rather than transient affective states or day-to-day fluctuations in body satisfaction.

Factor Analysis

The factorial validity of the EDBQ was established through rigorous exploratory and confirmatory factor analytic methodologies.

Original Exploratory Factor Analysis (EFA)

In the scale’s derivation phase (Cooper, Todd, & Wells, 1998), an initial pool of potential belief statements derived from clinical transcripts of cognitive therapy sessions with eating disorder patients was subjected to Principal Components Analysis (PCA) followed by orthogonal (Varimax) and oblique (Oblimin) rotations. Kaiser-Meyer-Olkin (KMO) measures of sampling adequacy (> .90) and Bartlett’s Test of Sphericity confirmed data factorability.

Inspection of the scree plot and eigenvalues greater than 1.5 revealed a clear four-factor latent structure accounting for 54.8% of the total variance:

• Factor 1: Negative Self-Beliefs: Eigenvalue = 9.87; accounted for 24.3% of the variance. Items 1 through 10 loaded heavily on this factor (factor loadings ranging from .61 to .84), showing negligible cross-loadings (< .25) on eating- or weight-specific factors.
• Factor 2: Weight and Shape as Means to Self-Acceptance: Eigenvalue = 4.12; accounted for 14.2% of the variance. Items 11, 12, 13, 14, 15, 21, 22, and 23 loaded between .55 and .81.
• Factor 3: Control Over Eating: Eigenvalue = 2.45; accounted for 9.1% of the variance. Items 24, 25, 26, 27, 28, 29, 30, 31, and 32 exhibited strong loadings ranging from .48 to .76.
• Factor 4: Weight and Shape as Means to Acceptance by Others: Eigenvalue = 1.98; accounted for 7.2% of the variance. Items 16, 17, 18, 19, and 20 loaded between .59 and .82.

Subsequent Confirmatory Factor Analyses (CFA)

Subsequent psychometric investigations have tested this four-factor model against alternative single-factor or two-factor models (e.g., a general negative affect factor vs. an eating belief factor). Structural equation modeling has repeatedly verified that the four-factor correlated model provides a superior fit to the empirical data:

• Satorra-Bentler χ² / df ratio: < 2.2
• Comparative Fit Index (CFI): .91 – .94
• Tucker-Lewis Index (TLI): .90 – .93
• Root Mean Square Error of Approximation (RMSEA): .052 – .064 (90% CI [.047, .071])
• Standardized Root Mean Square Residual (SRMR): .055 – .068

Inter-factor correlations indicate that while Factors 2, 3, and 4 correlate moderately to strongly with one another (r = .50 to .72), Factor 1 (Negative Self-Beliefs) maintains moderate correlations (r = .40 to .55), affirming that general negative self-schemas represent a distinct cognitive layer from the disorder-specific eating and shape assumptions.

Instrument / Measurement Tool

The operational features and structural parameters of the Eating Disorder Belief Questionnaire are summarized below:

• Instrument Name: Eating Disorder Belief Questionnaire (EDBQ)
• Construct Assessed: Core cognitive schemas, unconditional negative self-beliefs, conditional shape/weight assumptions, and rigid eating control beliefs.
• Administration Format: Paper-and-pencil self-report inventory or secure digital/interactive online assessment.
• Item Count: 32 items.
• Response Scale: 0 to 100 continuous Visual Analogue Scale (VAS). The anchors are explicitly defined as: 0 = “I don’t usually believe this at all” to 100 = “I am usually completely convinced that this is true”.
• Administration Time: Approximately 8 to 12 minutes.
• Target Population: Adolescents and adults (ages 14 and older) evaluated for or diagnosed with eating disorders, clinical populations undergoing cognitive-behavioral psychotherapy, and research participants in body image and psychopathology investigations.
• Scoring Protocol:
  • Each item receives a continuous score from 0 to 100 based on the respondent’s mark along the visual line (or numerical input in digital versions).
  • Subscale 1: Negative Self-Beliefs: Mean of items 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 (Total sum divided by 10).
  • Subscale 2: Weight and Shape as Means to Self-Acceptance: Mean of items 11, 12, 13, 14, 15, 21, 22, 23 (Total sum divided by 8).
  • Subscale 3: Weight and Shape as Means to Acceptance by Others: Mean of items 16, 17, 18, 19, 20 (Total sum divided by 5).
  • Subscale 4: Control Over Eating: Mean of items 24, 25, 26, 27, 28, 29, 30, 31, 32 (Total sum divided by 9).
  • Global EDBQ Index: Calculated either as the mean across all 32 items or as the composite mean of the four subscale scores. Higher scores indicate greater cognitive schema conviction and elevated psychopathology.
• Special Administration Instructions: Respondents are explicitly instructed to base their ratings on what they emotionally feel or believe in their everyday lives rather than what they rationally understand to be true.

Permissions & Fee and Test Year

The Eating Disorder Belief Questionnaire was first published in 1998 in the peer-reviewed journal Behaviour Research and Therapy by Pergamon / Elsevier Science Ltd. Under standard academic publishing conventions, the questionnaire items and scoring framework were placed into the scientific domain for research, academic, and non-commercial clinical diagnostic purposes by authors Myra Cooper, Gillian Todd, and Adrian Wells. No royalty fees or purchase costs are charged for using the measure in independent clinical practice or academic scientific research. Clinicians and research teams utilizing the instrument are required to cite the primary seminal paper (Cooper, Todd, & Wells, 1998) and ensure that scale integrity is maintained without altering the standardized instructions, items, or response anchors. Commercial applications, automated diagnostic platforms, or distribution within copyrighted software systems require appropriate publisher and author permissions.

References

The following seminal peer-reviewed literature provides the empirical, conceptual, and psychometric foundation for the Eating Disorder Belief Questionnaire:

• Beck, A. T. (1976). Cognitive therapy and the emotional disorders. International Universities Press.
• Cooper, M. J. (1997). Cognitive theory in anorexia nervosa and bulimia nervosa: Progress, problems and future directions. Clinical Psychology Review, 17(4), 399–417. https://doi.org/10.1016/S0272-7358(97)00015-X
• Cooper, M. J. (2005). Cognitive research in anorexia nervosa: A review. Behavioral and Cognitive Psychotherapy, 33(4), 389–406. https://doi.org/10.1017/S135246580500234X
• Cooper, M. J., Todd, G., & Wells, A. (1998). Content, construct and structural validity of the Eating Disorder Belief Questionnaire. Behaviour Research and Therapy, 36(3), 355–362. https://doi.org/10.1016/S0005-7967(97)10014-9
• Cooper, M. J., & Turner, H. (2000). Underlying assumptions and core beliefs in anorexia nervosa and dieting. British Journal of Clinical Psychology, 39(2), 215–218. https://doi.org/10.1348/014466500163239
• Fairburn, C. G., Cooper, Z., & Shafran, R. (2003). Cognitive behaviour therapy for eating disorders: A “transdiagnostic” theory and treatment strategy. Behaviour Research and Therapy, 41(5), 509–528. https://doi.org/10.1016/S0005-7967(02)00088-8
• Swinbourne, J. M. (2009). The nature and treatment of comorbid eating and anxiety disorders (Doctoral dissertation, University of Sydney). Faculty of Science, School of Psychology.
• Tovée, M. J., Benson, P. J., Emery, J. L., Mason, D. G., & Cohen-Tovée, E. M. (1999). Measurement of body shape and extreme attractiveness judgements. Nature, 399(6733), 211–212. https://doi.org/10.1038/20329
• Young, J. E., Klosko, J. S., & Weishaar, M. E. (2003). Schema therapy: A practitioner’s guide. Guilford Press.

13. Items of the Scale (Questionnaire)