Cover Image
QR Code

How to Be a Statistical Detective.

Statistical errors are surprisingly common in the biomedical literature and contribute to the reproducibility crisis.[1] Some errors are impossible to spot without access to the underlying dataset, but many are detectable from information available in the paper. Statcheck automatically extracts thes...

Full description

Published in:PM & R: Journal of Injury, Function & Rehabilitation Vol. 12; no. 2; pp. 211 - 216
Main Author: Sainani, Kristin L.
Format: Journal Article
Published: Wiley-Blackwell Feb2020
Online Access:View this record in EBSCOhost
fields @attributes:
  recordID: 1
pdfLink:
plink: https://search.ebscohost.com/login.aspx?direct=true&db=ccm&AN=141526723&site=ehost-live
header:
  @attributes:
    shortDbName: ccm
    uiTerm: 141526723
    longDbName: CINAHL Complete
    uiTag: AN
  controlInfo:
    bkinfo:
    dissinfo:
    jinfo:
      jid:
        19341482
        8NET
      jtl: PM & R: Journal of Injury, Function & Rehabilitation
      issn: 19341482
      maglogo: N
    pubinfo:
      dt: Feb2020
      vid: 12
      iid: 2
      pid: 480
      pub: Wiley-Blackwell
      place: Malden, Massachusetts
    artinfo:
      ui:
        141526723
        141526723
        NLM31850680
        10.1002/pmrj.12305
        NLM31850680
        141526723
      ppf: 211
      ppct: 5
      formats:
      tig:
        atl: How to Be a Statistical Detective.
      aug:
        au: Sainani, Kristin L.
        affil: Department of Epidemiology and Population Health, Stanford University, Stanford CA
      sug:
        subj:
          Statistics
          Exercise of Self-Care Agency Scale
      ab: Statistical errors are surprisingly common in the biomedical literature and contribute to the reproducibility crisis.[1] Some errors are impossible to spot without access to the underlying dataset, but many are detectable from information available in the paper. Statcheck automatically extracts these numbers from the text and then checks whether an F statistic of 2.791 with numerator degrees of freedom of 1 and denominator degrees of freedom of 50 actually yields a I P i value less than.05. When I uploaded the fun vs exercise paper, statcheck extracted 11 test statistics, 3 of which it flagged as inconsistent, including the F(1,50) = 2.791, I P i < .05 statistic. It can only check statistics that are reported in the text (not tables) and in which the test statistic, degrees of freedom, and I P i value are all available.
      pubtype: Academic Journal
      doctype: Journal Article
      ougenre: Article
    language: English
    refInfo:
    holdings:
      @attributes:
        islocal: N