Cited by 81Psychotherapy Techniques and ApplicationsDigital Mental Health InterventionsTelemedicine and Telehealth Implementation

Working Alliance Inventory for Online Interventions-Short Form (WAI-TECH-SF): The Role of the Therapeutic Alliance between Patient and Online Program in Therapeutic Outcomes

Open access
Herrero, Rocío, Vara, Mª Dolores, Miragall, Marta, Botella, Cristina, García-Palacios, Azucena, Riper, Heleen, Kleiboer, Annet, Baños, Rosa Mª·2020·International Journal of Environmental Research and Public Health

Abstract

Background: Therapeutic alliance (TA) between the patient and therapist has been related to positive therapeutic outcomes. Because Internet-based interventions are increasingly being implemented, a tool is needed to measure the TA with Internet-based self-guided programs. The Working Alliance Inventory for online interventions (WAI-TECH-SF) was adapted based on the WAI Short Form (Hatcher & Gillaspy, 2006). The objectives of this study were: (1) to analyse the psychometric properties of the WAI-TECH-SF; (2) to explore the differences in the WAI-TECH-SF scores according to different categories of the sample; and (3) to analyse whether the WAI-TECH-SF can predict therapeutic outcomes and satisfaction with the treatment. Methods: 193 patients diagnosed with depression were included and received blended Cognitive-Behavioural Therapy. Measures of preferences, satisfaction, and credibility about the treatment, TA with the online program, depressive symptoms, and satisfaction with the treatment were administered. Results: An exploratory factor analysis revealed a one-dimensional structure with adequate internal consistency. Linear regression analyses showed that the WAI-TECH-SF predicted changes in depressive symptoms and satisfaction with the treatment. Conclusions: WAI-TECH-SF is a reliable questionnaire to assess the TA between the patient and the online program, which is associated with positive therapeutic outcomes and satisfaction with the treatment.

AI snapshot

Written from the paper’s full text.

TL;DR

The study adapted the Working Alliance Inventory for online interventions (WAI-TECH-SF) to measure therapeutic alliance between patients and a self-guided online program in blended CBT for depression. It found a unidimensional, internally consistent scale whose scores predicted greater depressive symptom reduction and higher treatment satisfaction.

Key findings

  • Exploratory factor analysis supported a one-dimensional WAI-TECH-SF structure explaining 73.49% of total variance, with Cronbach’s α = 0.97; the retained factor had a raw eigenvalue of 9.08 versus 1.53 at the 95th percentile from parallel analysis.
  • WAI-TECH-SF scores predicted changes in depressive symptoms, F(1,188) = 14.42, p < 0.001, explaining 6.7% of the variance, and predicted treatment satisfaction, F(1,187) = 185.53, p < 0.001, explaining 49.7% of the variance.
  • There were no significant WAI-TECH-SF differences by sex, age-range, initial depression severity, treatment preference, expectations, or credibility; patients with high versus low education scored higher, p = 0.042.
  • In exploratory regression, level of education predicted WAI-TECH-SF scores (β = 0.180, p = 0.015; 2.7% variance), and a second model added age (education β = 0.205, p = 0.006; age β = 0.167, p = 0.024; 4.9% variance).

Methods

  • 193 patients diagnosed with depression (ages 19–69; M = 40.44, SD = 12.79; 64.2% women) received blended CBT in the European e-Compared project across primary or specialized care.
  • Measures included the WAI-TECH-SF at post-assessment, PHQ-9 at pre- and post-assessment, MINI diagnosis, treatment preference, CEQ expectancy/credibility, and CSQ-8 satisfaction.
  • Analyses used SPSS v.26: missing data imputation, EFA with Maximum Likelihood extraction and parallel analysis, Cronbach’s alpha, t-tests/ANOVAs, and simple/stepwise linear regressions; power analysis was also reported.

Limitations

  • The paper states the WAI-TECH-SF was administered only at the end of treatment, so it could not establish whether alliance preceded symptom and satisfaction changes or support causal conclusions.
  • The paper states the sample was only depressive patients, so replication in other diagnoses is needed; adherence or number of sessions was not registered; and analyses included only patients who completed the questionnaire after the self-guided IBI.
  • Evident from the text: the study used a clinical-trial sample with exclusion criteria and no control condition for the alliance analyses, which may limit generalizability and causal interpretation.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.3390/ijerph17176169

Cites

56 of 56 references verified0 citation errors · 0 not checkable

Statistics

Claims the paper makes

0 of 45 cited sources checked · 0 supported · 0 flagged

  • Findingown data

    Higher WAI-TECH-SF scores predicted a greater decrease in depression symptoms.

    “Higher scores on the WAI-TECH-SF predicted a greater decrease in depression symptoms.”
    Quote in the paperSubstantial claim
  • Findingown data

    Patients with high education levels achieved higher scores on the WAI-TECH-SF than patients with low education levels, p = 0.042.

    “However, there were significant differences in the WAI-TECH-SF scores based on the level of education. Patients with high (vs. low) education levels achieved higher scores on the WAI-TECH-SF, p = 0.042.”
    Quote in the paperNot a substantial claim
  • Findingown data

    Exclusion of any item increased the alpha value for the overall scale.

    “We analysed the item–total correlation, and the exclusion of any item increased the alpha value for the overall scale.”
    Quote in the paperNot a substantial claim
  • Findingown data

    Cronbach's alpha coefficient for the WAI-TECH-SF was high for the overall scale (α = 0.97).

    “Cronbach’s alpha coefficient for the WAI-TECH-SF was high for the overall scale (α = 0.97).”
    Quote in the paperSubstantial claim
  • Findingown data

    One dimension explained 73.49% of the total variance.

    “Factorial rotation with one dimension was performed using the Maximum Likelihood extraction method, which showed that one dimension explained 73.49% of the total variance.”
    Quote in the paperSubstantial claim
  • May combine two claims. 16
  • Classified as probably not a substantial claim. 14
  • May combine two claims.

    Findingown data

    There were no significant differences in WAI-TECH-SF scores based on sex, age-range, initial severity of depression, preferences for any of the treatments offered, expectations about the treatment, and credibility of the treatment.

    “Independent-sample t-tests and one-way ANOVAs showed that there were no significant differences in the WAI-TECH-SF scores based on sex, age-range, initial severity of depression, preferences for any of the treatments offered, expectations about the treatment, and credibility of the treatment.”
    Quote in the paperNot a substantial claim
  • May combine two claims.

    Findingown data

    The WAI-TECH-SF has a one-dimensional structure with adequate internal consistency.

    “An exploratory factor analysis revealed a one-dimensional structure with adequate internal consistency.”
    Quote in the paperSubstantial claim
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    The average WAI-TECH-SF score was around 58 (on a scale ranging from 12 to 84).

    “The average score on the WAI-TECH-SF was around 58 (on a scale ranging from 12 to 84).”
    Quote in the paperNot a substantial claim
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Clarke et al. did not find a relationship between TA and therapeutic outcomes.

    “In this study, they did not find a relationship between TA and therapeutic outcomes.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [22] Clarke, 2016, Therapeutic alliance with a fully automated mobile phone and web-based intervention: Secondary analysis of a randomized controlled trial, JMIR Ment. Health source

      Not checked in the free run.

  • Classified as probably not a substantial claim.

    Study design and conductnone given

    All patients were diagnosed with depression using the MINI.

    “All the patients were diagnosed with depression using the MINI.”
    Quote in the paperSubstantial claim
  • May combine two claims.

    Premisecitation

    Individuals have the ability to form a bond and be open with an online application.

    “Previous studies have found that individuals have the ability to form a bond and be open with an online application [28].”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [28] Berry, 2018, Assessing therapeutic alliance in the context of mHealth interventions for mental health problems: Development of the mobile Agnew relationship measure (mARM) questionnaire, J. Med. Internet Res. source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Clarke et al. found that TA is high even when the intervention does not involve human support.

    “they found that TA is high even when the intervention does not involve human support.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [22] Clarke, 2016, Therapeutic alliance with a fully automated mobile phone and web-based intervention: Secondary analysis of a randomized controlled trial, JMIR Ment. Health source

      Not checked in the free run.

  • May combine two claims.

    Interpretationown data

    The WAI-TECH-SF is associated with positive therapeutic outcomes and satisfaction with treatment.

    “which is associated with positive therapeutic outcomes and satisfaction with the treatment.”
    Quote in the paperSubstantial claim
  • Classified as probably not a substantial claim.

    Premisecitation

    Therapeutic alliance (TA) between the patient and therapist has a relevant influence on therapeutic outcomes.

    “Evidence shows that the therapeutic alliance (TA) (also called the working alliance) has a relevant influence on therapeutic outcomes [1].”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [1] Cameron, 2018, The relationship between the therapeutic alliance and clinical outcomes in cognitive behaviour therapy for adults with depression: A meta-analytic review, Clin. Psychol. Psychother. source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductcitation

    The adaptation of the patient version of the WAI-SF was carried out following the recommendations of Hambleton and Patsula.

    “An adaptation of the patient version of the WAI-SF [16] was carried out following the recommendations of Hambleton and Patsula [39].”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [16] Hatcher, 2007, Development and validation of a revised short version of the working alliance inventory, Psychother. Res. source

      Not checked in the free run.

    • cites [39] Hambleton, 1999, Increasing the validity of adapted tests: Myths to be avoided and guidelines for improving test adaptation practices, J. Appl. Test. Technol.

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    The study was approved by the corresponding ethical committee in each country.

    “The study was approved by the corresponding ethical committee in each country:”
    Quote in the paperNot a substantial claim
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    WAI-Tech total scores were not associated with change in therapeutic outcomes.

    “Results showed lower scores on the bond subscale, and the total scores on WAI-Tech were not associated with the change in therapeutic outcomes.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [29] Kiluk, 2014, Only connect: The working alliance in computer-based cognitive behavioral therapy, Behav. Res. source

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    A systematic review showed an effect of TA on anxiety and depressive outcomes, with higher levels of TA related to better clinical outcomes.

    “a systematic review showed an effect of TA on anxiety and depressive outcomes, pointing out that higher levels of TA were related to better clinical outcomes [23].”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [23] Pihlaja, 2018, Therapeutic alliance in guided internet therapy programs for depression and anxiety disorders–a systematic review, Internet Interv. source

      Not checked in the free run.

  • May combine two claims.

    Study design and conductown data

    Missing values in the WAI-TECH-SF, PHQ-9, and CEQ ranged from 0% to 1.6%, were missing at random according to Little's MCAR test (p > 0.05), and were imputed using the Expectation-Maximization Algorithm method.

    “The percentage of missing values in the WAI-TECH-SF, PHQ-9, and CEQ scores ranged from 0% to 1.6%. After testing that the values were missing at random using Little’s MCAR test (p > 0.05), they were imputed using the Expectation–Maximization Algorithm method [40].”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [40] Schafer, J.L. (1997). Analysis of Incomplete Multivariate Data. Monographs on Statistics and Applied Probability, Chapman & Hall. source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductnone given

    All participants were informed about the study and gave informed consent before the trial, in accordance with the Declaration of Helsinki.

    “All participants were informed about the study and gave their informed consent before the beginning of the trial, in accordance with the Declaration of Helsinki.”
    Quote in the paperNot a substantial claim
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Patients with higher scores on TA were more satisfied with the intervention after treatment.

    “Moreover, patients with higher scores on TA were more satisfied with the intervention after the treatment.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [24] Babl, 2020, The association of therapeutic alliance with long-term outcome in a guided Internet Intervention for depression: Secondary analysis from a randomized control trial, J. Med. Internet Res. source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    TA is moderately associated with better treatment outcomes in face-to-face therapy, regardless of the therapeutic framework or patient characteristics.

    “Several meta-analyses have found that TA is moderately associated with better treatment outcomes in face-to-face therapy [1,2], regardless of the therapeutic framework or patient characteristics, among other aspects [3].”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites [1] Cameron, 2018, The relationship between the therapeutic alliance and clinical outcomes in cognitive behaviour therapy for adults with depression: A meta-analytic review, Clin. Psychol. Psychother. source

      Not checked in the free run.

    • cites [2] Horvath, 2011, Alliance in individual psychotherapy, Psychotherapy source

      Not checked in the free run.

    • cites [3] Wampold, 2018, The alliance in adult psychotherapy: A meta-analytic synthesis, Psychotherapy source

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    Gómez-Penedo et al. found that the WAI for guided Internet interventions had a two-factor solution ("tasks-goals" and "bonds") with adequate internal consistency and external validity in a sample of patients with mild to moderate depression.

    “Results of the WAI for guided Internet interventions (i.e., WAI-I) showed a two-factor solution (“tasks-goals” and “bonds”) with adequate internal consistency and external validity in a sample of patients with mild to moderate depression.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [24] Babl, 2020, The association of therapeutic alliance with long-term outcome in a guided Internet Intervention for depression: Secondary analysis from a randomized control trial, J. Med. Internet Res. source

      Not checked in the free run.

  • May combine two claims.

    Study design and conductown data

    The current study had 69.38% and 62.68% power for one-way ANOVAs with three and four groups, respectively, 75.49% for the t-test, and 99.77% for the regression analyses with one predictor to detect a medium effect size at p < 0.05.

    “Results indicated that the current study had 69.38% and 62.68% power for one-way ANOVAs with three and four groups, respectively, 75.49% for the t-test, and 99.77% for the regression analyses with one predictor to detect a medium effect size at p < 0.05.”
    Quote in the paperNot a substantial claim
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Kiluk et al.'s WAI-Tech showed similar psychometric properties to the original WAI scale.

    “Kiluk et al. [29] conducted an adaptation of the 36-item original version of the WAI called WAI-Tech, which was designed to measure the TA between cocaine-dependent patients and the online program, and it showed similar psychometric properties to the original scale (i.e., the WAI).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [29] Kiluk, 2014, Only connect: The working alliance in computer-based cognitive behavioral therapy, Behav. Res. source

      Not checked in the free run.

  • May combine two claims.

    Study design and conductown data

    The KMO value was 0.96 and Bartlett's Test of Sphericity was significant, showing that it was appropriate to perform a factor analysis.

    “The KMO value was (0.96), and the Barlett’s Test of Sphericity value, χ2(66) = 2587.26, p < 0.001, showed that it was appropriate to perform a factor analysis.”
    Quote in the paperSubstantial claim
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductown data

    The therapists were trained in how to deliver the blended treatment.

    “All of them were trained in how to deliver the blended treatment.”
    Quote in the paperNot a substantial claim
  • May combine two claims.

    Premisecitation

    In a WAI adapted to measure TA between patients and the virtual environment, the three dimensions proposed by Bordin did not arise in exploratory factor analysis, and only one general dimension was found.

    “in the WAI adapted to measure the TA between the patients and the virtual environment, where the word “therapist” was replaced by “virtual environment”, the three dimensions proposed by Bordin [4] did not arise in the exploratory factor analysis, and only one general dimension was found [25].”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [25] Miragall, 2015, Working alliance inventory applied to virtual and augmented reality (WAI-VAR): Psychometrics and therapeutic outcomes, Front Psychol. source

      Not checked in the free run.

  • May combine two claims.

    Study design and conductown data

    The WAI-TECH-SF was only administered at the end of the treatment, which did not allow exploration of whether patient-online program TA preceded symptoms and satisfaction throughout therapeutic sessions.

    “First, the WAI-TECH-SF was only administered at the end of the treatment, which did not allow us to explore whether the “patient and online program TA” preceded the symptoms and satisfaction throughout the therapeutic sessions.”
    Quote in the paperSubstantial claim
  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    The adherence or number of sessions performed by the patients was not registered.

    “Third, the adherence or number of sessions performed by the patients was not registered.”
    Quote in the paperNot a substantial claim
  • May combine two claims.

    Interpretationown data

    The study sample was only composed of depressive patients, limiting generalizability.

    “Second, the study sample was only composed of depressive patients.”
    Quote in the paperNot a substantial claim
  • May combine two claims.

    Study design and conductown data

    The statistical analyses of TA were only conducted with patients who accepted to fill in the questionnaire after the self-guided IBI was finished, but the normal distribution and wide range of variability of WAI-TECH-SF scores allowed reliable conclusions.

    “Fourth, the statistical analyses of TA were only conducted with the patients that accepted to fill in the questionnaire after the self-guided IBI was finished. However, the normal distribution (e.g., skewness = −0.56; kurtosis = −0.83) and the wide range of variability of the WAI-TECH-SF scores (i.e., from 12 to 84) allowed us to draw reliable conclusions.”
    Quote in the paperSubstantial claim
  • May combine two claims.

    Interpretationown data

    In the case of TA with an online program during a self-guided IBI, the theoretical distinction between task, goals, and bond with the online program was not psychometrically significant, and a single factor could explain the majority of the explained variance of the TA between the patient and the online program.

    “More specifically, in the case of the TA with an online program during a self-guided IBI, the theoretical distinction between task, goals, and bond with the online program was not psychometrically significant, and a single factor could explain the majority of the explained variance of the TA between the patient and the online program.”
    Quote in the paperSubstantial claim
  • May combine two claims.

    Interpretationreasoning

    Patients can feel that the program is "taking care" of them, in terms of allowing them to achieve therapeutic goals, proposing appropriate tasks to achieve these goals, and making them feel "embraced" and "cared for" by the program.

    “Thus, patients can feel that the program is “taking care” of them, in terms of allowing them to achieve therapeutic goals, proposing appropriate tasks to achieve these goals, and making them feel “embraced” and “cared for” by the program.”
    Quote in the paperSubstantial claim
29 of 29

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

Statistics recomputed

Every reported test result, recomputed from its test statistic and degrees of freedom. A result shows as not adding up only once an independent check has confirmed it.

  • Consistentχ²(33) = 14.27reported p = .998 · recomputed p = .998

    “χ2(33) = 14.27, p = .998”

  • Consistentχ²(66) = 2587.26reported p < .001 · recomputed p < .001

    “χ2(66) = 2587.26, p < .001”

  • ConsistentF(1, 181) = 6.059reported p = .015 · recomputed p = .015

    “F(1,181) = 6.059, p = .015”

  • ConsistentF(2, 181) = 5.696reported p = .004 · recomputed p = .004

    “F(2,181) = 5.696, p = .004”

  • ConsistentF(1, 188) = 14.42reported p < .001 · recomputed p < .001

    “F(1,188) = 14.42, p < .001”

  • ConsistentF(1, 187) = 185.53reported p < .001 · recomputed p < .001

    “F(1,187) = 185.53, p < .001”

6 of 6

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

  • Nested regression models imply different sample sizes

    Minor

    The first model's F(1,181) implies 183 observations, whereas the second model's F(2,181) implies 184. Adding age to the first model cannot increase the number of observations while retaining its education predictor.

  • A scale described as 10-point ranges only from 1 to 9

    Minor

    A scale with scores from 1 through 9 has nine points, not ten.

  • Twelve-item scale lists item 10 twice and omits item 12

    Minor

    The three four-item dimension lists name only 11 distinct items: item 10 appears under both goals and tasks, while item 12 appears under none.

3 of 3

Cites

Flagged first, then by citations

Cited by

Sorted by citations

Authors
Herrero, R., Vara, M. D., Miragall, M., Botella, C., García-Palacios, A., Riper, H., Kleiboer, A., & Baños, R. M.
Affiliations
  • Universitat de València (ES)
  • Polibienestar Research Institute, University of Valencia, 46022 Valencia, Spain
  • Universitat Jaume I (ES)
  • CIBERObn Physiopathology of Obesity and Nutrition, Instituto de Salud Carlos III, 28029 Madrid, Spain
  • GGZ inGeest (NL)
  • Vrije Universiteit Amsterdam (NL)
Corresponding author
Miragall, Marta
Venue
International Journal of Environmental Research and Public Health
Publisher
MDPI AG
Published
2020-08-25
Volume
17
Issue
17
Pages
6169
ISSN
1660-4601 (ISSN-L)
OpenAlex
W3080571288
Cited by
81
References
56
Citations by year
2026: 8 · 2025: 9 · 2024: 12 · 2023: 5 · 2022: 9 · 2021: 32 · 2020: 6
FWCI
5.20
Type
journal article
Language
en
Open access
Locations
  • International Journal of Environmental Research and Public Health (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • LA Referencia (Red Federada de Repositorios Institucionales de Publicaciones Científicas) (publishedVersion, other-oa)
  • Pure Amsterdam UMC (publishedVersion, cc-by)
  • Repositori UJI (Universitat Jaume I) (submittedVersion, cc-by-sa)
  • RePEc: Research Papers in Economics (submittedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion)
  • Amsterdam UMC (VU Amsterdam) - Institutional Repository (submittedVersion, other-oa)
  • VU Research Portal (submittedVersion, other-oa)
  • Data Archiving and Networked Services (DANS) (submittedVersion, other-oa)
Funders
European Commission (FP7-Health-2013-Innovation-N603098)
Topics
  • Social Sciences › Psychology › Clinical Psychology › Psychotherapy Techniques and Applications
  • Social Sciences › Psychology › Applied Psychology › Digital Mental Health Interventions
  • Health Sciences › Medicine › Public Health, Environmental and Occupational Health › Telemedicine and Telehealth Implementation
Keywords
therapeutic relationship, online intervention, psychometric properties, depression, treatment satisfaction, treatment credibility, depressive symptoms, questionnaire adaptation
MeSH terms
Therapeutic Alliance, Internet-Based Intervention, Adolescent, Adult, Aged, Behavior Therapy, Depression, Female, Humans, Male, Middle Aged, Professional-Patient Relations, Psychometrics, Surveys and Questionnaires, Reproducibility of Results, Treatment Outcome, Patient Satisfaction, Young Adult

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-09.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-09.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review