CoChat
Sign Up
Cited by 715Physical Activity and HealthSports and Physical Education StudiesRecreation, Leisure, Wilderness Management

Domain-Specific Physical Activity and Mental Health: A Meta-analysis

Open access
White, Rhiannon Lee, Babic, Mark J., Parker, Philip D., Lubans, David R., Astell-Burt, Thomas, Lonsdale, Chris·2017·American Journal of Preventive Medicine
DOI 10.1016/j.amepre.2016.12.008
Open this paper in CoChat — free accountOpen PDF

Abstract

Context The mental health benefits of physical activity are well established. However, less is known about whether the relationship between physical activity and mental health is consistent across different life domains. It is important to understand how context may influence the relationship between physical activity and mental health so that interventions and policy guidelines can be tailored to maximize positive effects. Evidence acquisition In 2015, systematic searches of four databases identified 13,435 records, of which 98 studies met the inclusion criteria. Evidence synthesis Included studies were published between 1988 and 2015 and had a combined sample size of 648,726. Of the 98 included studies, 93 examined leisure-time physical activity, 14 examined work-related physical activity, 15 examined transport physical activity, 16 examined household physical activity, three examined school sport, and three examined physical education. Multi-level meta-analyses showed that leisure-time physical activity (r =0.13) and transport physical activity (r =0.13) both had a positive association with mental health. Leisure-time physical activity (r = -0.11) and school sport (r = -0.09) both had an inverse association with mental ill-health. However, physical activity was not consistently associated with lower mental ill-health across domains, as work-related physical activity was positively associated with mental ill-health (r =0.09). Household physical activity and participation in…

Integrity signals

Registry

Crossref · 10.1016/j.amepre.2016.12.008

Cites

182 of 182 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

13 of 31 cited sources checked · 3 supported · 0 flagged · accuracy not yet measured

  • Premisecitation

    Evidence shows that exercise involving choice is associated with increased positive affect and enjoyment compared to exercise with no choice.

    “Further, evidence also shows that exercise involving choice is associated with increased positive affect and enjoyment, compared to exercise with no choice.[62][63][64]”
    Quote in the paperSubstantial claim1 of 3 sources checked
    • cites [62] Parfitt, 2006, The psychological and physiological responses of sedentary individuals to prescribed and preferred intensity exercise, Br J Health Psychol source
    • cites [63] Vazou-Ekkekakis, 2009, Affective consequences of imposing the intensity of physical activity: does the loss of perceived autonomy matter, Hellenic Journal of Psychology
    • cites [64] Lubans, 2016, Mediators of psychological well-being in adolescent boys, J Adolesc Health source

      The Discussion reports that opportunities for choice in school sport mediated the intervention’s effect on psychological well-being, supporting a link between choice and well-being. But the Outcome measures section identifies flourishing—not positive affect or enjoyment—as the outcome, and the Discussion says effects on hedonic well-being were not captured. The paper does not establish the claimed increase in positive affect and enjoyment compared with exercise offering no choice. (LLM, conf=0.87)

      Experimental — accuracy not yet measured
  • Premisecitation

    Evidence supports psychosocial mechanisms that explain the effect of physical activity on mental health.

    “Evidence supports a number of psychosocial mechanisms that explain the effect of PA on mental health; however, there is little understanding as to whether these mechanisms play the same role in different PA domains.22”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [22] Lubans, 2016, Physical activity for cognitive and mental health in youth: A systematic review of mechanisms, Pediatrics source

      The Psychosocial mechanisms section (p. 17) reports evidence linking improved physical self-perceptions with well-being, including one study in which changes in self-perceptions partially mediated reduced depressive symptoms. But the Conclusion (p. 21) says evidence for the specific mechanisms remains limited, so the claim overstates how firmly a number of psychosocial mechanisms have been shown to explain the effect. (LLM, conf=0.87)

      Experimental — accuracy not yet measured
  • Premisecitation

    Mental health and mental ill-health are not always negatively correlated.

    “While this result is plausible given that mental health and mental ill-health are not always negatively correlated, 65 it demonstrates that workrelated PA is associated with mental health for some people and mental ill-health for others.”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [65] Keyes, 2005, Mental illness and/or mental health? Investigating axioms of the complete state model of health, J Consult Clin Psychol source

      The Discussion (p. 8) says mental health and mental illness are distinct, correlated axes, so they are not invariably opposites for individuals. But the Results (p. 6) report a negative correlation of −.53; the paper does not show that their correlation is sometimes nonnegative. (LLM, conf=0.84)

      Experimental — accuracy not yet measured
  • Findingown data

    School sport has no significant association with mental health.

    “Household PA, school sport, and physical education are not significantly associated with mental health.”
    Quote in the paperSubstantial claim
  • Findingown data

    Leisure-time physical activity has a positive association with mental health (r = .13).

    “leisure-time physical activity (r = .13)and transport physical activity (r = .13)both have a positive association with mental health.”
    • Also stated in: § Effect Sizes, Heterogeneity, and Moderator Analyses · plain
      “Leisure-time PA has a weak positive association with mental health and the effect size is not significantly moderated by the mental health outcome, sex, age, study design, PA measure, or risk of bias.”
    Quote in the paperSubstantial claim
  • May combine two claims. 21
  • Classified as probably not a substantial claim. 15

  • May combine two claims.

    Premisecitation

    Work-related physical activity is likely to be viewed as a compulsory task and may be enforced by another person or only carried out for the purpose of receiving an external reward such as pay.

    “Alternatively, work-related PA is likely to be viewed as a compulsory task and may be enforced by another person or only carried out for the purpose of receiving an external reward (i.e., pay).13”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [13] Asztalos, 2009, Specific associations between types of physical activity and components of mental health, J Sci Med Sport source

      The Discussion (page unavailable) says active transportation can involve some compulsion: getting to work is “a must,” and economic reasons may motivate biking to work. It does not say that work-related physical activity is enforced by another person or undertaken for pay; its example concerns commuting rather than physical activity performed at work. (LLM, conf=0.86)

      Experimental — accuracy not yet measured

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

    Study design and conductown data

    Initial agreement between reviewers on risk of bias was 90% and kappa was .81, indicating strong agreement; after discussion 100% agreement was achieved.

    “Initial agreement between the two reviewers was 90% and the kappa coefficient (κ = .81)indicated strong agreement.48 Upon discussion 100% agreement was achieved.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [48] McHugh, 2012, Interrater reliability: The kappa statistic, Biochemia Medica source

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

    Study design and conductown data

    Risk of bias of individual studies was assessed using six criteria adapted from STROBE and CONSORT, with two reviewers independently assigning scores.

    “Six criterion were adapted from the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement 42 and the Consolidated Standards of Reporting Trials (CONSORT) statement to assess risk of bias of individual studies.43”
    • Also stated in: § Risk of Bias of Individual Studies · plain
      “Two reviewers independently assigned a 1 (present and explicitly explained) or a 0 (absent or inadequately described) to each included study on each criterion”
    Quote in the paperNot a substantial claim1 of 2 sources checked
    • cites [42] Von Elm, 2007, The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies, Prev Med source
    • cites [43] Moher, 2001, The CONSORT statement: revised recommendations for improving the quality of reports of parallel group randomized trials, BMC Med Res Methodol source

  • May combine two claims.

    Premisecitation

    Physical activity enhances self-efficacy and exposes individuals to challenges that offer opportunities to develop confidence and a sense of mastery.

    “For example, PA enhances self-efficacy and exposes individuals to challenges that offer opportunities to develop confidence and a sense of mastery.51,52”
    Quote in the paperNot a substantial claim1 of 2 sources checked
    • cites [51] Craft, 2004, The benefits of exercise for the clinically depressed, Prim Care Companion J Clin Psychiatry source

      The Self-Efficacy Hypothesis section suggests that exercise may enhance self-efficacy by providing a meaningful mastery experience, which supports the confidence-and-mastery idea. But it describes findings on generalized self-efficacy in clinically depressed people as equivocal, so the citing sentence states the enhancement more definitively than the source does. Page numbers were not provided in our copy. (LLM, conf=0.85)

      Experimental — accuracy not yet measured
    • cites [52] Paluska, 2000, Physical activity and mental health, Sports Med source

  • May combine two claims.

    Interpretationreasoning

    Work-related physical activity is associated with mental health for some people and mental ill-health for others.

    “While this result is plausible given that mental health and mental ill-health are not always negatively correlated, 65 it demonstrates that workrelated PA is associated with mental health for some people and mental ill-health for others.”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [65] Keyes, 2005, Mental illness and/or mental health? Investigating axioms of the complete state model of health, J Consult Clin Psychol source

      The citation is attached to the rationale about mental health and mental ill-health, not to the citing authors’ finding about work-related physical activity. The Discussion (p. 8) says mental health and mental illness are distinct rather than opposite ends of one continuum, but the Results (p. 6) report a negative correlation of −.53 between them. The source therefore supports their not being exact opposites, but not a claim that they are uncorrelated or non-negatively correlated. (LLM, conf=0.82)

      Experimental — accuracy not yet measured

  • Classified as probably not a substantial claim.

    Premisecitation

    Individual studies have shown that leisure-time physical activity has a stronger relationship with mental health and mental ill-health compared to work-related, transport, and household physical activity.

    “Further, a number of individual studies have shown that leisure-time PA has a stronger relationship with mental health and mental illhealth, compared to work-related PA, transport PA, and household PA. [13][14][15]”
    Quote in the paperSubstantial claim2 of 3 sources checked
    • cites [13] Asztalos, 2009, Specific associations between types of physical activity and components of mental health, J Sci Med Sport source

      The Discussion reports that sports participation, unlike the other activity types studied, was consistently associated with less stress and psychological distress. But the Methods distinguish sports from leisure active transportation, and the study measured commuting activity rather than physical activity performed at work. The finding therefore supports a narrower comparison than the claim's broad leisure-time versus work-related, transport, and household comparison. (LLM, conf=0.87)

      Experimental — accuracy not yet measured
    • cites [14] Kull, 2012, Relationship between low depressiveness and domain specific physical activity in women, Health Care Woman Int source
    • cites [15] Ohta, 2007, Effect of the physical activities in leisure time and commuting to work on mental health, J Occup Health source

      The Results (p. 3) report that greater leisure-time exercise was associated with lower GHQ scores in men, supporting a relationship with mental health. They also report an association for walking or cycling to work in men; the paper does not establish that the leisure-time relationship was stronger, and its stated scope (p. 2) does not include household or work-related physical activity. (LLM, conf=0.88)

      Experimental — accuracy not yet measured

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

    Premisecitation

    Asztalos reported that transport physical activity was associated with increased stress among blue collar workers but not among white collar workers.

    “Furthermore, Asztalos reported that transport PA was associated with increased stress among blue collar workers but not among white collar workers.13”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [13] Asztalos, 2009, Specific associations between types of physical activity and components of mental health, J Sci Med Sport source

      The Results report an association between biking to and from work and more stress specifically in men with blue-collar jobs, not blue-collar workers generally or transport physical activity as a whole. The Discussion says the only significant association among people with white-collar jobs concerned housework and distress, which supports the absence of a reported transport–stress association in that group. (LLM, conf=0.88)

      Experimental — accuracy not yet measured

  • Classified as probably not a substantial claim. May combine two claims.

    Premisecitation

    Systematic reviews have shown that physical activity is associated with greater mental health and a reduced risk of mental ill-health, specifically depression and anxiety.

    “Systematic reviews have shown that physical activity (PA) is associated with greater mental health, 6 and a reduced risk of mental ill-health, specifically depression and anxiety.7,8”
    Quote in the paperSubstantial claim2 of 3 sources checked
    • cites [6] Bize, 2007, Physical activity level and health-related quality of life in the general adult population: a systematic review, Prev Med source
    • cites [7] Biddle, 2011, Physical activity and mental health in children and adolescents: a review of reviews, Br J Sports Med source

      The Abstract reports potentially beneficial effects on depression and a small beneficial effect on anxiety, and the Overall Conclusions say physically active young people appear less likely to have mental health problems. But the review concerns children and adolescents, not people generally, and describes the evidence for depression as weak and for anxiety as limited. (LLM, conf=0.83)

      Experimental — accuracy not yet measured
    • cites [8] Janssen, 2010, Systematic review of the health benefits of physical activity and fitness in school-aged children and youth, Int J Behav Nutr Phys Act source

  • May combine two claims.

    Findingown data

    Physical activity domain significantly moderates the strength of the relationship between physical activity and mental ill-health (p < .001) and explains 46% of the variance.

    “Overall, PA domain significantly moderates the strength of the relationship between PA and mental ill-health (p < .001)and explains 46% of the variance.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Age is a significant predictor of the association between work-related physical activity and mental ill-health (p = .03), with a larger association among adults (r = .10) than adolescents (r = .01), explaining 13% of the variance.

    “In contrast, work-related PA has a weak positive association with mental ill-health and age is a significant predictor (p = .03)as the association is larger among adults (r = .10)than adolescents (r = .01),but this only explains 13% of the variance.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Nineteen percent of included studies had high risk of bias, 59% moderate, and 18% low.

    “Nineteen percent of studies have a high risk of bias and 59% have a moderate risk of bias, while only 18% have a low risk of bias.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Work-related physical activity was positively associated with mental health as well as mental ill-health.

    “Interestingly, work-related PA was positively associated with mental health as well as mental ill-health.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Age is a significant moderator of the association between work-related physical activity and mental health (p = .002) and explains 89% of the variance, with work-related physical activity positively associated with mental health only in the one study examining adults >65 years (r = .37).

    “Further, age is a significant moderator (p = .002)and explains 89% of the variance, as work-related PA is only positively associated with mental health in the one study examining adults >65 years (r = .37).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingnone given

    Transport physical activity is positively associated with mental health among adults, but no such relationship was found for adolescents.

    “While transport PA is positively associated with mental health among adults, no such relationship was found for adolescents.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The purpose of the trip is the only significant moderator of the association between transport physical activity and mental health (p = .01) and explains the majority of variation in effect sizes (R2 = 76%).

    “The purpose of the trip (e.g., walking to work, walking during leisure-time) is the only significant moderator (p = .01)and explains majority of the variation in effect sizes (R² = 76%).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Study design and conductown data

    Ninety-eight percent of included studies were observational, the majority cross-sectional.

    “Firstly, 98% of the included studies were observational, the majority of which were cross-sectional.”
    Quote in the paperSubstantial claim

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

    Study design and conductcitation

    The methods were conducted in accordance with the PRISMA statement.

    “The methods detailed below were conducted in accordance with the 'Preferred Reporting Items for Systematic Reviews and Meta-Analysis' (PRISMA) statement.16”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [16] Moher, 2009, Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement, Ann Intern Med source

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

    Study design and conductown data

    Two reviewers independently assessed titles/abstracts and full texts, and discrepancies were resolved by discussion with a third reviewer until consensus was reached.

    “Two reviewers independently assessed the title and abstract of each study and records recommended for exclusion by both reviewers were excluded.”
    • Also stated in: § Study Selection · plain
      “Discrepancies were resolved by discussion between the two reviewers, along with a third reviewer, until consensus was reached.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Premisecitation

    Choosing to participate for enjoyment is central to autonomous motivation.

    “Choosing to participate for enjoyment is central to autonomous motivation.58”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [58] Ryan, 2000, Intrinsic and extrinsic motivations: Classic definitions and new directions, Contemp Educ Psychol source

  • May combine two claims.

    Interpretationreasoning

    Physical activity is not automatically associated with greater mental health and reduced mental ill-health, and contextual factors are crucial to such relationships.

    “This finding suggests that PA is not automatically associated with greater mental health and reduced mental ill-health, and that contextual factors are crucial to such relationships.”
    Quote in the paperSubstantial claim

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

    Premisecitation

    The distraction hypothesis suggests that physical activity diverts individuals' attention from stressful life events, leading to improved mental wellbeing.

    “The distraction hypothesis suggests that PA diverts individuals' attention from stressful life events which leads to improved mental wellbeing.55”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [55] Leith, 2010

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Classified as probably not a substantial claim. May combine two claims.

    Premisecitation

    Self-determination theory posits that behaviours undertaken due to autonomous motivation are more likely to be associated with the satisfaction of psychological needs and that wellbeing is enhanced when psychological needs are satisfied.

    “Self-determination theory posits that behaviours which are undertaken due to autonomous motivation are more likely to be associated with the satisfaction of psychological needs (i.e., autonomy, competence, and relatedness), and that wellbeing is enhanced when an individual's psychological needs are satisfied.59”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [59] Weinstein, 2010, When helping helps: autonomous motivation for prosocial behavior and its influence on well-being for the helper and recipient, J Pers Soc Psychol source

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

    Study design and conductown data

    Data extraction was performed by one researcher and cross-checked by a second reviewer, with discrepancies discussed and resolved.

    “A second reviewer cross-checked the extracted data for accuracy and all discrepancies were discussed and resolved.”
    Quote in the paperNot a substantial claim

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

    Premisecitation

    The social interaction hypothesis suggests that participating in physical activity with other people may influence mental health outcomes.

    “Alternatively, the social interaction hypothesis suggests that participating in PA with other people may influence mental health outcomes.54”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [54] Bailey, 2005, Physical activity alone and with others as predictors of sense of belonging and mental health in retirees, Aging Ment Health source

  • May combine two claims.

    Interpretationcitation

    Lifestyle physical activity outside leisure-time may improve physical health but may not benefit mental health.

    “While lifestyle PA outside leisure-time may improve people's physical health, such behaviours may not benefit mental health.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Study design and conductown data

    Egger's test results for the mental health and mental ill-health meta-analyses revealed low asymmetry, suggesting low risk of bias across studies.

    “Non-significant Egger's test results for the mental health (t = 1.6121 , p = .11)and mental ill-health meta-analyses (t = 0.0925, p = .93)revealed low asymmetry, suggesting low risk of bias across studies.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Study design and conductnone given

    Studies included in this review measured the relationship between habitual physical activity and mental health.

    “studies included in this review measured the relationship between habitual PA and mental health.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Premisecitation

    Sallis' guideline for interpreting systematic review results states that evidence is inconclusive if less than four studies are included.

    “Sallis' guideline for interpreting systematic review results states that evidence is inconclusive if less than four studies are included.73”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [73] Sallis, 2000, A review of correlates of physical activity of children and adolescents, Med Sci Sports Exerc source

  • Classified as probably not a substantial claim.

    Study design and conductreasoning

    Not identifying the source of this variation is a limitation of the current study.

    “Not identifying the source of this variation is a limitation of the current study.”
    Quote in the paperSubstantial claim
29 of 29

  • The paper relies on this without support.

    Less is known about whether the relationship between physical activity and mental health is consistent across different life domains.

    “However, less is known about whether the relationship between physical activity and mental health is consistent across different life domains.”
1 of 1

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.

  • Reported estimate falls outside its confidence interval

    Minor

    Table 2 gives a household-activity mental health correlation of .00 alongside a 95% confidence interval from .09 to .17. The point estimate is outside its own interval.

  • Self-report claim conflicts with the study listing

    Minor

    The paper says all 98 studies used self-report activity measures, but its study summary lists the activity measure for Armstrong (2004) as n/a.

2 of 2

Cites

Flagged first, then by citations

10 of the 192 references couldn’t be matched to a known paper.

Cited by

Sorted by citations

Related

Authors
White, R. L., Babic, M. J., Parker, P. D., Lubans, D. R., Astell-Burt, T., & Lonsdale, C.
Affiliations
  • Australian Catholic University (AU)
  • University of Newcastle Australia (AU)
Corresponding author
Lonsdale, Chris
ORCID
Venue
American Journal of Preventive Medicine
Publisher
Elsevier BV
Published
2017-05
Volume
52
Issue
5
Pages
653-666
ISSN
0749-3797 (ISSN-L)
DOI
10.1016/j.amepre.2016.12.008
OpenAlex
W2582924421
Cited by
715
References
192
Citations by year
2026: 65 · 2025: 90 · 2024: 119 · 2023: 120 · 2022: 95 · 2021: 95 · 2020: 72 · 2019: 33 · 2018: 22 · 2017: 3
FWCI
4.36
Type
journal article
Language
en
Open access
Licence
www.elsevier.com/tdm/userlicense/1.0/ (tdm)
Locations
  • American Journal of Preventive Medicine (publishedVersion)
  • PubMed (publishedVersion)
  • Research Bank (Australian Catholic University) (submittedVersion)
  • Research Online (University of Wollongong) (submittedVersion)
  • Figshare (submittedVersion)
  • NOVA (University of Newcastle Australia) (submittedVersion)
Funders
  • Australian Research Council (DP130104659)
  • Australian Postgraduate Award
  • Australian Catholic University
  • Australian Research Council Future Fellowship
  • National Heart Foundation of Australia
Topics
  • Health Sciences › Medicine › Physiology › Physical Activity and Health
  • Social Sciences › Life-span and Life-course Studies › Sports and Physical Education Studies
  • Social Sciences › Psychology › Social Psychology › Recreation, Leisure, Wilderness Management
Keywords
physical activity, mental health, mental illness, leisure-time physical activity, work-related physical activity, household physical activity, school sports, physical education
MeSH terms
Community Participation, Environment, Female, Health Planning, Health Policy, Health Promotion, Humans, Health Knowledge, Attitudes, Practice, Male, Mental Health, United States, Health Behavior, Exercise, Needs Assessment

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 2026-10-11.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-11.

Cited sources are checked by gpt-6-sol.

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