CoChat
Sign Up
Cited by 665Sleep and related disordersPhysical Activity and HealthCircadian rhythm and melatonin

Aerobic exercise improves self-reported sleep and quality of life in older adults with insomnia

Open access
Reid, Kathryn J., Baron, Kelly Glazer, Lu, Brandon, Naylor, Erik, Wolfe, Lisa, Zee, Phyllis C.·2010·Sleep Medicine
DOI 10.1016/j.sleep.2010.04.014PMC PMC2992829
Open this paper in CoChat — free account

Abstract

Objective To assess the efficacy of moderate aerobic physical activity with sleep hygiene education to improve sleep, mood and quality of life in older adults with chronic insomnia.Methods Seventeen sedentary adults aged >or=55 years with insomnia (mean age 61.6 [SD±4.3] years; 16 female) participated in a randomized controlled trial comparing 16 weeks of aerobic physical activity plus sleep hygiene to non-physical activity plus sleep hygiene. Eligibility included primary insomnia for at least 3 months, habitual sleep duration <6.5h and a Pittsburgh Sleep Quality Index (PSQI) score >5. Outcomes included sleep quality, mood and quality of life questionnaires (PSQI, Epworth Sleepiness Scale [ESS], Short-form 36 [SF-36], Center for Epidemiological Studies Depression Scale [CES-D]).Results The physical activity group improved in sleep quality on the global PSQI (p<.0001), sleep latency (p=.049), sleep duration (p=.04), daytime dysfunction (p=.027), and sleep efficiency (p=.036) PSQI sub-scores compared to the control group. The physical activity group also had reductions in depressive symptoms (p=.044), daytime sleepiness (p=.02) and improvements in vitality (p=.017) compared to baseline scores.Conclusion Aerobic physical activity with sleep hygiene education is an effective treatment approach to improve sleep quality, mood and quality of life in older adults with chronic insomnia.

Integrity signals

Registry

Crossref · 10.1016/j.sleep.2010.04.014

Cites

53 of 53 references verified0 citation errors · 0 not checkable

Statistics

Claims the paper makes

0 of 14 cited sources checked · 0 supported · 0 flagged

  • Findingown data

    There was a significant time by condition interaction for the PSQI sleep onset latency sub-score.

    “a significant time by condition interaction for sleep onset latency (F(1,15)=5.39, p= .035)”
    Quote in the paperSubstantial claim
  • Findingown data

    A reduction in global PSQI scores was associated with a greater reduction in depressive symptoms on the CES-D.

    “A reduction in global PSQI scores was also associated with a greater reduction in depressive symptoms on the CES-D (r=0.487, p=0.048).”
    Quote in the paperNot a substantial claim
  • Findingown data

    There were trends for time by condition effects in the SF-36 role emotional and mental health domains.

    “a trend for time by condition effects in role emotional (F (1,15)= 4.214, p= .058) and mental health (F (1,15) 3.44, p = .08).”
    Quote in the paperNot a substantial claim
  • Findingown data

    There were no significant time or time by condition effects for the SF-36 role physical, general health, or social functioning domains.

    “There were no significant time or time by condition effects for role physical, general health, or social functioning.”
    Quote in the paperNot a substantial claim
  • Findingown data

    Reductions in global PSQI scores were not associated with improvements in maximum heart rate or VO2max.

    “Reductions in global PSQI scores were not associated with greater improvements in maximum heart rate or VO2max.”
    Quote in the paperNot a substantial claim
  • Classified as probably not a substantial claim. 11
  • May combine two claims. 25

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

    Findingown data

    There was a trend toward an effect for time on maximum heart rate, but no time by condition effect.

    “There was a trend toward an effect for time on maximum heart rate (F (1,15)= 3.51, p= .08) but no time by condition effect.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    There was no significant time or time by condition effect for sleep latency measured as a continuous variable.

    “There was not a significant time or time by condition effect for sleep latency measured as a continuous variable.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    The current study achieved an average 4.8 point reduction in PSQI score, compared with a 1.8 point decrease reported in a 25-week Tai Chi study.

    “Compared to a 25 week Tai Chi study that reported a significant decrease in PSQI score of 1.8 points [39], the current study achieved an average 4.8 point reduction in PSQI score.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [39]

      Not checked in the free run.

  • May combine two claims.

    Findingown data

    There was an effect for time on SF-36 vitality and a trend toward improvement over time in physical function.

    “There was an effect for time on vitality (F(1,15), 7.207, p= .02) and a trend toward improvement over time in physical function (F (1,15)= 4.309, p=.056).”
    Quote in the paperSubstantial claim

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

    Findingown data

    The exercise group showed a trend toward improvement in V02max, while the non-physical activity group did not.

    “Pre-post comparisons for each group demonstrated a trend toward improvements in V02max for the exercise group (t(9) -1.93, p= .08) but not for the non-physical activity group.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    In older adults with chronic insomnia, the exercise group showed a significant improvement in global PSQI scores from baseline to post-treatment, while the non-physical activity group did not.

    “participants in the exercise group demonstrated significant improvement in global PSQI scores (t(15)= -5.62, p< 0.0001) while participants in the non-physical activity group did not demonstrate improvement.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The exercise group showed significant pre-post improvements in CES-D scores, while the non-physical activity group did not.

    “pre-post comparisons for each group demonstrated significant improvements in CES-D scores for the exercise group (t (9)= 2.34, p=.04) but not for the non-physical activity group.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    There were no effects of time or time by condition on V02max or maximum heart rate.

    “There were no effects of time or time by condition on V02max or maximum heart rate.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    There was a significant time effect and a significant time by condition interaction for the PSQI daytime dysfunction sub-score.

    “for daytime dysfunction (time F (1,15)= 12.32, p= .003, time × condition F(1,15)= 5.98, p= .03)”
    Quote in the paperSubstantial claim

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

    Findingown data

    In the exercise group alone, there was a trend for a correlation between change in maximum heart rate and PSQI scores.

    “when the physical activity group was analyzed alone, there was a trend for a correlation between change in maximum HR and PSQI scores (r=0.671, p=0.058).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    There was a significant time effect and a significant time by condition interaction for the PSQI sleep duration sub-score.

    “a significant time and time by condition interaction for sleep duration (time F(1,15)=4.59 p=.049, time × condition F(1,15)=4.59, p= .049)”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    When sleep duration was analyzed as a continuous variable, there was a significant time effect and a trend toward a time by condition interaction.

    “there was a significant time effect for sleep duration and a trend toward a time by condition interaction for sleep duration (time F(1,15) = 6.44, p=.02, time by condition F(1,15) = 3.22, p=.09).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    For the CES-D, there was no significant time or time by condition effect.

    “For the CES-D, there was not a significant time or time by condition effect (Figure 1b).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    The exercise group had a significant reduction in ESS scores from baseline to post-treatment, while the non-physical activity group did not.

    “there was a significant reduction in the ESS score for the exercise group (t(9)= -2.75, p=0.02) but not in the non-physical activity group”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingcitation

    The exercise plus sleep hygiene group increased self-reported sleep duration by 1.25 hours, which is higher than reported for other non-pharmacological interventions for insomnia.

    “The increase in self reported sleep duration by 1.25 hours in the exercise plus sleep hygiene education group is higher than what has been reported for other non-pharmacological interventions for insomnia [30, 40]”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [30] source

      Not checked in the free run.

    • cites [40] source

      Not checked in the free run.

  • May combine two claims.

    Findingown data

    For the ESS, there was a trend for a time effect but no time by condition interaction.

    “For ESS there was a trend of an effect for time (F (1, 15)= 3.18, p= .09), but there was not a time by condition interaction (F (1, 15)= 2.8, p= .12).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingreasoning

    There was a significant reduction in the PSQI sleep latency sub-score but not in sleep latency measured as a continuous variable.

    “There was also a significant reduction in the sleep latency sub-score of the PSQI but not in sleep latency measured as a continuous variable.”
    Quote in the paperNot a substantial claim

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

    Findingcitation

    At baseline, SF-36 scores in this study were similar to those previously reported for people with mild-to-severe insomnia.

    “At baseline, SF-36 scores in this study were similar to those previously reported for people with mild-severe insomnia [42-44].”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites [42]

      Not checked in the free run.

    • cites [43] source

      Not checked in the free run.

    • cites [44]

      Not checked in the free run.

  • May combine two claims.

    Interpretationcitation

    The effect sizes for improvements in sleep duration and subjective sleep quality in this study were similar to or greater than those reported for cognitive behavioral intervention and other physical activity studies.

    “The effect size for the improvements in sleep duration and subjective sleep quality in this study were similar or even greater than that reported for cognitive behavioral intervention and other physical activity studies [18, 30, 36-38].”
    Quote in the paperSubstantial claim0 of 5 sources checked
    • cites [18] source

      Not checked in the free run.

    • cites [30] source

      Not checked in the free run.

    • cites [36] source

      Not checked in the free run.

    • cites [37] source

      Not checked in the free run.

    • cites [38] 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 conductown data

    Five participants withdrew or were excluded after randomization; four from the non-physical activity group withdrew because they wanted the exercise condition, and one from the exercise group was withdrawn for non-compliance.

    “Five participants withdrew or were excluded following randomization. The reason given for withdrawal by the four in the non-physical activity group was that they wanted to be in the exercise condition and the one participant from the activity group was withdrawn due to non-compliance.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Study design and conductown data

    Participants were randomly assigned to either the aerobic physical activity or non-physical activity group.

    “Participants were randomly assigned to either an aerobic physical activity or non-physical activity intervention group and measurements were collected at baseline and post treatment (16 weeks).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationown data

    A 16-week program of moderate-intensity aerobic physical activity plus sleep hygiene education is effective in improving self-reported sleep quality, mood, and quality of life in older adults with chronic insomnia.

    “Results from this study indicate that a sixteen week program of moderate intensity aerobic physical activity plus sleep hygiene education is effective in improving self reported sleep quality, mood and quality of life in older adults with chronic insomnia.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The non-physical activity condition (recreational or educational activities) adequately controls for non-specific effects of attention, social contact, and sleep hygiene education, so that differences in outcomes can be attributed to aerobic exercise.

  • May combine two claims.

    Interpretationown data

    The elevated baseline WASO indicates sleep maintenance insomnia, so improved sleep continuity may have contributed to the self-reported increase in sleep duration.

    “In addition, the elevated baseline WASO as measured by actigraphy and PSG indicate that this group of patients had sleep maintenance insomnia type. Thus, improved sleep continuity may also have contributed to the self reported increase in sleep duration.”
    Quote in the paperSubstantial claim

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

    Study design and conductown data

    The non-physical activity group averaged 4.2 sessions per week for 139 minutes per session.

    “On average participants completed 4.2 (±0.73) sessions a week for an average of 139 (±64.4) minutes each session.”
    Quote in the paperNot a substantial claim

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

    Study design and conductown data

    The exercise group averaged 3.27 sessions per week for 32.3 minutes per session.

    “On average participants exercised 3.27 (±0.52) times a week for an average of 32.3(±12.5) minutes each session.”
    Quote in the paperNot a substantial claim

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

    Study design and conductown data

    Data were analyzed for 17 of the 18 participants who completed the study; one participant's data were removed because of an acute stressful event at the end of the study.

    “Data were analyzed for 17 of 18 participants that completed the study. Data from one participant was removed due to an acute stressful event (de ath in the family) at the end of the study.”
    Quote in the paperNot a substantial claim

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

    Study design and conductown data

    The study was a randomized parallel pre-post trial comparing 16 weeks of aerobic physical activity plus sleep hygiene education with non-physical activity plus sleep hygiene education in sedentary adults aged 55 and older with insomnia.

    “This study used a randomized parallel pre-post design. Participants were randomly assigned to either an aerobic physical activity or non-physical activity intervention group and measurements were collected at baseline and post treatment (16 weeks).”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Study design and conductown data

    The study had limitations including reliance on self-reported sleep quality measures, a relatively small sample, and a female preponderance.

    “There are several potential limitations in this study, including the reliance on self- reported sleep quality measures, the relatively small number of participants and the female preponderance of the sample.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationown data

    The improvements in subjective sleep quality associated with exercise remained after adjusting for depressive symptoms, indicating that the effect of exercise on sleep quality is independent of depressive symptoms.

    “However, the improvements in subjective sleep quality (PSQI score) associated with exercise remained after adjusting for depressive symptoms (CES-D score), indicating that the positive effect of exercise on sleep quality is also independent of depressive symptoms.”
    Quote in the paperSubstantial claim

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

    Study design and conductreasoning

    The higher dropout rate in the non-physical activity group may have affected internal validity if participants with low expectations for that intervention selectively dropped out after randomization.

    “Another limitation was the higher dropout rate in the non-physical activity group. This may have been due in part to the desire of some participants to be randomized to the exercise condition. The internal validity of the study may be affected if those with low expectations for the non-physical activity intervention selectively dropped out after randomization.”
    Quote in the paperNot a substantial claim
30 of 30

  • The paper relies on this without support.

    Self-reported measures of overall sleep quality, sleep latency, sleep duration, alertness, and quality of life are the most widely used measures of insomnia treatment effectiveness in clinical practice.

    “self-reported measures of overall sleep quality, sleep latency, sleep duration, alertness and quality of life are the most widely used measure of insomnia treatment effectiveness in clinical practice.”

  • The paper relies on this without support.

    The non-physical activity condition (recreational or educational activities) adequately controls for non-specific effects of attention, social contact, and sleep hygiene education, so that differences in outcomes can be attributed to aerobic exercise.

    “Participants in this condition selected recreational or education activities (e.g., parks and recreation, YMCA, museum lectures, church activities) which met for approximately 45 minutes, 3 to 5 times per week, for 16 weeks.”
2 of 2

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?

  • Several mechanisms provide a rationale for the improvement in insomnia symptoms from exercise, including reducing core body temperature for sleep onset, increasing melatonin and serotonin secretion, reducing depression and anxiety, and improving autonomic regulation which may reduce hyperarousal associated with insomnia.

    “Several mechanisms that provide a rationale for the improvement in insomnia symptoms from exercise have been suggested [133,136,137,138,139]; these include but are not limited to reducing core body temperature for sleep onset [140,141,142,143], increasing melatonin and serotonin secretion [131,144], reducing depression and anxiety, and improving autonomic regulation which may reduce hyperarousal”

    From The Cyclical Battle of Insomnia and Mental Health Impairment in Firefighters: A Narrative Review (2024) · cites this paper as [137]

    The Discussion suggests that exercise’s effects on sleep may involve mechanisms beyond cardiopulmonary fitness, and the Results link improved sleep quality with reduced depressive symptoms. The paper does not identify reduced core body temperature, increased melatonin or serotonin secretion, or improved autonomic regulation as mechanisms; it also does not establish that reduced depression mediates the sleep improvement. (LLM, conf=0.84)

    Experimental — accuracy not yet measured

  • In a review of the literature, Kline concluded that the three studies exclusively including patients diagnosed with chronic insomnia had the most robust results regarding insomnia.

    “Kline [7], in a review of the literature, concluded that the three studies that exclusively included patients diagnosed with chronic insomnia had the most robust results regarding insomnia [8–10].”

    From Is sleep disturbance in patients with chronic pain affected by physical exercise or ACT-based stress management? - A randomized controlled study (2018) · cites this paper as [10]

  • Three of the four trials measuring adverse events found no adverse events in any participants.

    “Three trials found no adverse events in any of the participants (Reid et al., 2010; Afonso et al., 2012; Chan et al., 2016).”

    From Exercise can improve sleep quality: A systematic review and meta-analysis (2018) · cites this paper as Reid et al., 2010

  • Recent randomized controlled trials have confirmed that exercise has positive effects on sleep quality, sleep onset latency, total sleep time, sleep efficiency, and insomnia severity.

    “Recent randomized controlled trials (RCTs) have confirmed that exercise has positive effects on sleep quality, sleep onset latency, total sleep time, sleep efficiency, and insomnia severity (Passos et al., 2010; Reid et al., 2010; Hartescu, Morgan & Stevinson, 2015).”

    From Exercise can improve sleep quality: A systematic review and meta-analysis (2018) · cites this paper as Reid et al., 2010

  • Only four of the nine included trials examined adverse effects.

    “First, only four of the nine included trials examined adverse effects (Reid et al., 2010; Afonso et al., 2012; Hartescu, Morgan & Stevinson, 2015).”

    From Exercise can improve sleep quality: A systematic review and meta-analysis (2018) · cites this paper as Reid et al., 2010

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.

  • Not confirmedF(1, 15) = 5.39reported p = .04 · recomputed p = .035

    “F (1,15) = 5.39, p = .04”

  • ConsistentF(1, 15) = 9.96reported p = .007 · recomputed p = .007

    “F (1,15)= 9.96, p = .007”

  • Consistentt(15) = -5.62reported p < .0001 · recomputed p < .001

    “t(15)= -5.62, p < .0001”

  • ConsistentF(1, 15) = 26.09reported p < .001 · recomputed p < .001

    “F(1,15)= 26.09, p < .001”

  • ConsistentF(1, 15) = 5.39reported p = .035 · recomputed p = .035

    “F(1,15)=5.39, p = .035”

  • ConsistentF(1, 15) = 4.59reported p = .049 · recomputed p = .049

    “F(1,15)=4.59, p = .049”

  • ConsistentF(1, 15) = 12.32reported p = .003 · recomputed p = .003

    “F (1,15)= 12.32, p = .003”

  • ConsistentF(1, 15) = 5.98reported p = .03 · recomputed p = .027

    “F(1,15)= 5.98, p = .03”

  • ConsistentF(1, 15) = 6.44reported p = .02 · recomputed p = .023

    “F(1,15) = 6.44, p = .02”

  • ConsistentF(1, 15) = 3.22reported p = .09 · recomputed p = .093

    “F(1,15) = 3.22, p = .09”

  • Consistentt(9) = 2.34reported p = .04 · recomputed p = .044

    “t (9)= 2.34, p = .04”

  • ConsistentF(1, 15) = 4.309reported p = .056 · recomputed p = .056

    “F (1,15)= 4.309, p = .056”

  • ConsistentF(1, 15) = 7.207reported p = .017 · recomputed p = .017

    “F (1,15)= 7.207, p = .017”

  • ConsistentF(1, 15) = 4.214reported p = .058 · recomputed p = .058

    “F (1,15)= 4.214, p = .058”

  • ConsistentF(1, 15) = 3.18reported p = .09 · recomputed p = .095

    “F (1, 15)= 3.18, p = .09”

  • ConsistentF(1, 15) = 2.8reported p = .12 · recomputed p = .115

    “F (1, 15)= 2.8, p = .12”

  • Consistentt(9) = -2.75reported p = .02 · recomputed p = .022

    “t(9)= -2.75, p = .02”

  • ConsistentF(1, 15) = 3.51reported p = .08 · recomputed p = .081

    “F (1,15)= 3.51, p = .08”

18 of 18

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.

  • Exercise-group t test degrees of freedom imply 16 subjects, not 10

    Minor

    The Results section reports an exercise-group paired t test as t(15)= -5.62, which implies 16 paired observations. But the same paper elsewhere reports the exercise-group pre-post comparison as t(9)= 2.34 / t(9)= -2.75 / t(9) -1.93, implying only 10 subjects in that group.

  • Sleep-latency p-value differs between the abstract and results

    Minor

    The abstract gives p=0.049 for the sleep-latency comparison, while the results give p=.035 for the sleep-onset-latency time-by-condition interaction. These are different values for the reported between-group effect.

  • Sleep-duration p-value differs between the abstract and results

    Minor

    The abstract reports p=0.04 for the sleep-duration comparison, but the results report p=.049 for its time-by-condition interaction. The difference is not explained by rounding to the precision shown.

3 of 3

Cites

Flagged first, then by citations

4 of the 57 references couldn’t be matched to a known paper.

Cited by

Sorted by citations

Related

Authors
Reid, K. J., Baron, K. G., Lu, B., Naylor, E., Wolfe, L., & Zee, P. C.
Affiliations
Northwestern University (US)
Corresponding author
Zee, Phyllis C.
ORCID
Venue
Sleep Medicine
Publisher
Elsevier BV
Published
2010-10
Volume
11
Issue
9
Pages
934-940
ISSN
1389-9457 (ISSN-L)
DOI
10.1016/j.sleep.2010.04.014
PMC
PMC2992829
OpenAlex
W2061105741
Cited by
665
References
57
Citations by year
2026: 39 · 2025: 40 · 2024: 57 · 2023: 36 · 2022: 71 · 2021: 45 · 2020: 39 · 2019: 43 · 2018: 54 · 2017: 43 · 2016: 38 · 2015: 43 · 2014: 44 · 2013: 31 · 2012: 26
FWCI
30.56
Type
journal article
Language
en
Open access
Licence
Locations
  • Sleep Medicine (publishedVersion)
  • PubMed (publishedVersion)
  • PubMed Central (acceptedVersion)
Funders
National Institute on Aging (K23 HL091508, M01 RR00048, P01 AG11412, T32AG020506)
Topics
  • Social Sciences › Psychology › Experimental and Cognitive Psychology › Sleep and related disorders
  • Health Sciences › Medicine › Physiology › Physical Activity and Health
  • Life Sciences › Neuroscience › Endocrine and Autonomic Systems › Circadian rhythm and melatonin
Keywords
aerobic exercise, chronic insomnia, older adults, sleep hygiene education, sleep quality, mood, quality of life, depressive symptoms, daytime sleepiness, sleep latency, physical activity interventions
MeSH terms
Affect, Depression, Exercise Therapy, Female, Humans, Sleep Initiation and Maintenance Disorders, Male, Middle Aged, Patient Compliance, Psychiatric Status Rating Scales, Quality of Life, Sleep, Exercise, Polysomnography, Sedentary Behavior

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 under definition version 3 (extract-v3) on 2026-09-28.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-09-28.

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