Skip to main content
v2026.11,986 entries · CC-BY 4.0

EDE-Q: Items, Scoring, Cutoffs and Validity

How the Eating Disorder Examination Questionnaire is scored, which published cutoffs apply to which sample, and what the validity evidence supports.

Written and maintained by CASRAI Editorial Board

Published · Last updated

On this page

This page describes a research and clinical-screening questionnaire for readers who administer, score or report it. It is not a self-assessment and not medical advice; a score cannot diagnose an eating disorder. Anyone worried about their own eating should speak with a qualified health professional.

What does the EDE-Q assess?

The Eating Disorder Examination Questionnaire (EDE-Q) is a 28-item self-report measure of eating-disorder attitudes and behaviours over the previous 28 days. It was adapted from the investigator-based Eating Disorder Examination (EDE) interview (Fairburn & Beglin, 1994). Twenty-two attitude items, each rated 0 to 6, feed four subscales: Restraint, Eating Concern, Shape Concern and Weight Concern. The global score is the mean of those four subscale scores. Six further free-response items ask how often behaviours such as binge eating and purging occurred; these are reported as frequencies and are not part of the global score (Schaefer et al., 2018). Higher scores mean more problematic attitudes. It is designed for ages 14 and over; for younger children the youth version (YEDE-Q) or the EDE for adolescents is recommended (CORC).

Scoring: formula with every symbol explained

Each subscale score is the mean of its items (each rated 0 to 6). The global score is then:

G = (R + E + S + W) / 4

  • G: the EDE-Q global score, on the same 0 to 6 range as the items.
  • R: Restraint subscale mean.
  • E: Eating Concern subscale mean.
  • S: Shape Concern subscale mean.
  • W: Weight Concern subscale mean.

Use the official scoring instructions that accompany the instrument to assign items to subscales; the behaviour-frequency items are excluded from G.

Worked example (illustrative numbers)

The values below are invented for demonstration and describe no real respondent. Suppose the four subscale means are R = 2.4, E = 1.8, S = 3.5 and W = 3.1.

  1. Sum the subscale means: 2.4 + 1.8 + 3.5 + 3.1 = 10.8.
  2. Divide by 4: 10.8 / 4 = 2.70.

The global score is 2.70. Whether that is “high” depends on the reference sample and threshold used, as the next sections explain.

How to interpret the EDE-Q: what is the clinical cutoff score?

There is no single, universal clinical cutoff. Published thresholds are sample-specific screening values chosen to balance sensitivity and specificity against an interview or diagnostic reference; they are not diagnostic. The CORC guidance page itself gives no cutoff and notes (citing Kelly et al., 2012) that the validity of the clinical cutoffs appears lower in samples of Black women than in samples of White women. The values below are those summarised by Schaefer et al. (2018):

Population Global cutoff Sensitivity / specificity Source
Women, community 2.30 .92 / .86 Mond et al. (2004)
Women, primary care 2.80 .80 / .80 Mond et al. (2008)
Women, clinical vs control 2.50 .86 / .86 Ro et al. (2015)
Men (245 patients, 205 undergraduates) 1.68 .77 / .77 (AUC .85, 95% CI .82 to .89) Schaefer et al. (2018)

In the male study, mean global scores were 3.06 in the clinical group and 1.04 in controls. Subgroup cutoffs were 1.81 (non-white) and 1.50 (white), and 1.71 (ages 18 to 20) versus 1.62 (ages 21 to 30); the authors described the differences as small. Their caution is important for interpretation: the EDE-Q may under-capture muscularity-oriented symptoms in men and does not fully assess all relevant domains for males with eating disorders. Their control group was self-identified, diagnoses came from non-structured interviews, and the clinical sample completed a 36-item version (EDE-Q 4) while controls completed the 28-item 6.0.

Reading guidance What it means
Score above a published cutoff The respondent resembles the clinical group in that study more than the comparison group; it flags a need for assessment, not a diagnosis.
Score below a cutoff Does not rule out an eating disorder, particularly where the cutoff came from a different sex, culture or setting.
Change over time Compare scores from the same respondent on the same version; see floor and ceiling effects for limits at the ends of the range.

Is the EDE-Q validated?

Yes, but with qualifications that matter for study design. Evidence summarised by CORC (read 2026-11-07):

  • Internal consistency: Cronbach’s alpha .74 to .89 in men and .75 to .93 in women (Rose et al., 2013). See Cronbach’s alpha.
  • Test-retest: r = .75 to .91 for subscales and .92 for the global score in a college sample (Rose et al., 2013); r = .66 to .77 in binge-eating-disorder patients (Reas et al., 2006). See test-retest reliability.
  • Factor structure: the original four-factor structure has limited support, and alternative structures have been proposed (Grilo et al., 2015; Carey et al., 2019; Rand-Giovannetti et al., 2020). Treat subscale scores with this in mind. See construct validity.
  • Discriminant and convergent validity: highly accurate at separating people with an eating disorder from those without, and moderate for binge-eating disorder versus obesity; convergent validity with the ED-15 is high. See convergent and discriminant validity.
  • Norms: published for UK adolescents aged 14 to 18 (White et al., 2013) and UK adult community samples (Carey et al., 2019), designed for pen-and-paper administration.

Assumptions and common mistakes

  • Applying a cutoff from one population (for example women in primary care) to another (for example male athletes).
  • Including the six behaviour-frequency items in the global mean.
  • Mixing versions (EDE-Q 4, 6.0) within one analysis without stating it.
  • Reporting only the global score when subscale structure is itself contested.
  • Treating a screening score as a diagnosis; the EDE interview is the reference standard.
  • Altering wording or response categories. CORC states the measure should be replicated exactly.

Access and permissions

The EDE-Q is copyrighted. According to CORC, it is free for non-commercial clinical and research use without asking permission, provided the copyright information is included and the measure is replicated exactly, including response categories. Digital administration is acceptable for non-commercial use, though the norms were developed for pen-and-paper. For commercial use or translation, CORC directs enquiries to [email protected]; CORC is not the copyright holder.

How to report it (APA 7 example)

Methods: “Eating-disorder psychopathology was measured with the 28-item Eating Disorder Examination Questionnaire (EDE-Q 6.0; Fairburn & Beglin, 1994). The global score is the mean of the Restraint, Eating Concern, Shape Concern and Weight Concern subscale scores (range 0 to 6; higher scores indicate greater severity). Cronbach’s alpha in this sample was .91.”

Results: “The mean global score was 2.70 (SD = 1.10).” (Replace with your own sample statistics; report alpha from your data rather than quoting published values.)

How to calculate the global score in Excel, R or Python

Excel (subscale means in B2:E2):

=AVERAGE(B2:E2)

R:

sub <- c(R = 2.4, E = 1.8, S = 3.5, W = 3.1)
global <- mean(sub)   # 2.7

Python:

sub = {"R": 2.4, "E": 1.8, "S": 3.5, "W": 3.1}
global_score = sum(sub.values()) / len(sub)   # 2.7

FAQ

What does the EDE-Q assess?

Eating-disorder attitudes (restraint, eating, shape and weight concern) and behaviour frequencies over the past 28 days, using 28 items.

How do I interpret the EDE-Q?

Compute subscale means and the global mean, then compare with norms or cutoffs from a sample resembling yours, treating the result as a screening signal rather than a diagnosis.

What is the clinical cutoff score for the EDE-Q?

No single value exists. Published global cutoffs include 2.30, 2.50 and 2.80 for women and 1.68 for men, each tied to its sample (see the table above).

Is the EDE-Q validated?

Reliability and discriminant validity are well supported; the four-subscale structure has limited support, and it may under-capture some male presentations.

Sources

  • Fairburn, C. G., & Beglin, S. J. (1994). Assessment of eating disorders: Interview or self-report questionnaire? International Journal of Eating Disorders, 16(4), 363-370.
  • Schaefer, L. M., et al. (2018). Identifying a male clinical cutoff on the Eating Disorder Examination-Questionnaire (EDE-Q). International Journal of Eating Disorders, 51(12), 1357-1360. pmc.ncbi.nlm.nih.gov/articles/PMC6310481 (read 2026-11-07).
  • CORC. Eating Disorder Examination Questionnaire (EDE-Q). corc.uk.net (read 2026-11-07).

Follow CASRAI

Research-administration guidance, standards updates and independent tool reviews.

Ask CASRAI · free to try

Ask about EDE-Q: Items, Scoring, Cutoffs and Validity

2 free questions a day, no account.

An AI assistant for research: funder rules, compliance, methods and publishing. It links the sources behind its answers and labels answers that come from a web search.

Answers draw on CASRAI's guides and dictionary plus the federal and funder documents we index: Federal Register, Grants.gov, Regulations.gov and UKRI.

Works on this site and inside Claude, Cursor and the AI tools you already use.

Everything CASRAI publishes — this page, the dictionary, the guides and the news — stays open to read, with no account and no card.

Referenced across the research world

University of Cambridge logoColumbia University logoCrossref logoUniversity of Edinburgh logoHarvard University logoUniversity of Oxford logoPrinceton University logoStanford School of Medicine logoUniversity College London logoORCID logoUniversity of Cambridge logoColumbia University logoCrossref logoUniversity of Edinburgh logoHarvard University logoUniversity of Oxford logoPrinceton University logoStanford School of Medicine logoUniversity College London logoORCID logo
  • University of Cambridge logo
  • Columbia University logo
  • Crossref logo
  • University of Edinburgh logo
  • Harvard University logo
  • University of Oxford logo
  • Princeton University logo
  • Stanford School of Medicine logo
  • University College London logo
  • ORCID logo

View CASRAI adoption →