The κ statistic in rehabilitation research

An examination

Leigh R. Tooth, Kenneth Ottenbacher

Research output: Contribution to journalArticle

72 Citations (Scopus)

Abstract

Tooth LR, Ottenbacher KJ. The κ statistic in rehabilitation research: an examination. 2004;85:1371-6. The number and sophistication of statistical procedures reported in medical rehabilitation research is increasing. Application of the principles and methods associated with evidence-based practice has contributed to the need for rehabilitation practitioners to understand quantitative methods in published articles. Outcomes measurement and determination of reliability are areas that have experienced rapid change during the past decade. In this study, distinctions between reliability and agreement are examined. Information is presented on analytical approaches for addressing reliability and agreement with the focus on the application of the κ statistic. The following assumptions are discussed: (1) κ should be used with data measured on a categorical scale, (2) the patients or objects categorized should be independent, and (3) the observers or raters must make their measurement decisions and judgments independently. Several issues related to using κ in measurement studies are described, including use of weighted κ, methods of reporting κ, the effect of bias and prevalence on κ, and sample size and power requirements for κ. The κ statistic is useful for assessing agreement among raters, and it is being used more frequently in rehabilitation research. Correct interpretation of the κ statistic depends on meeting the required assumptions and accurate reporting.

Original languageEnglish (US)
Pages (from-to)1371-1376
Number of pages6
JournalArchives of Physical Medicine and Rehabilitation
Volume85
Issue number8
DOIs
StatePublished - Aug 2004

Fingerprint

Evidence-Based Practice
Sample Size
Biomedical Research
Tooth
Rehabilitation
Rehabilitation Research
Power (Psychology)

Keywords

  • Data analysis, statistical
  • Rehabilitation
  • Reliability and validity

ASJC Scopus subject areas

  • Rehabilitation

Cite this

The κ statistic in rehabilitation research : An examination. / Tooth, Leigh R.; Ottenbacher, Kenneth.

In: Archives of Physical Medicine and Rehabilitation, Vol. 85, No. 8, 08.2004, p. 1371-1376.

Research output: Contribution to journalArticle

@article{412a620aea654ed2b2d4894406cb3d9c,
title = "The κ statistic in rehabilitation research: An examination",
abstract = "Tooth LR, Ottenbacher KJ. The κ statistic in rehabilitation research: an examination. 2004;85:1371-6. The number and sophistication of statistical procedures reported in medical rehabilitation research is increasing. Application of the principles and methods associated with evidence-based practice has contributed to the need for rehabilitation practitioners to understand quantitative methods in published articles. Outcomes measurement and determination of reliability are areas that have experienced rapid change during the past decade. In this study, distinctions between reliability and agreement are examined. Information is presented on analytical approaches for addressing reliability and agreement with the focus on the application of the κ statistic. The following assumptions are discussed: (1) κ should be used with data measured on a categorical scale, (2) the patients or objects categorized should be independent, and (3) the observers or raters must make their measurement decisions and judgments independently. Several issues related to using κ in measurement studies are described, including use of weighted κ, methods of reporting κ, the effect of bias and prevalence on κ, and sample size and power requirements for κ. The κ statistic is useful for assessing agreement among raters, and it is being used more frequently in rehabilitation research. Correct interpretation of the κ statistic depends on meeting the required assumptions and accurate reporting.",
keywords = "Data analysis, statistical, Rehabilitation, Reliability and validity",
author = "Tooth, {Leigh R.} and Kenneth Ottenbacher",
year = "2004",
month = "8",
doi = "10.1016/j.apmr.2003.12.002",
language = "English (US)",
volume = "85",
pages = "1371--1376",
journal = "Archives of Physical Medicine and Rehabilitation",
issn = "0003-9993",
publisher = "W.B. Saunders Ltd",
number = "8",

}

TY - JOUR

T1 - The κ statistic in rehabilitation research

T2 - An examination

AU - Tooth, Leigh R.

AU - Ottenbacher, Kenneth

PY - 2004/8

Y1 - 2004/8

N2 - Tooth LR, Ottenbacher KJ. The κ statistic in rehabilitation research: an examination. 2004;85:1371-6. The number and sophistication of statistical procedures reported in medical rehabilitation research is increasing. Application of the principles and methods associated with evidence-based practice has contributed to the need for rehabilitation practitioners to understand quantitative methods in published articles. Outcomes measurement and determination of reliability are areas that have experienced rapid change during the past decade. In this study, distinctions between reliability and agreement are examined. Information is presented on analytical approaches for addressing reliability and agreement with the focus on the application of the κ statistic. The following assumptions are discussed: (1) κ should be used with data measured on a categorical scale, (2) the patients or objects categorized should be independent, and (3) the observers or raters must make their measurement decisions and judgments independently. Several issues related to using κ in measurement studies are described, including use of weighted κ, methods of reporting κ, the effect of bias and prevalence on κ, and sample size and power requirements for κ. The κ statistic is useful for assessing agreement among raters, and it is being used more frequently in rehabilitation research. Correct interpretation of the κ statistic depends on meeting the required assumptions and accurate reporting.

AB - Tooth LR, Ottenbacher KJ. The κ statistic in rehabilitation research: an examination. 2004;85:1371-6. The number and sophistication of statistical procedures reported in medical rehabilitation research is increasing. Application of the principles and methods associated with evidence-based practice has contributed to the need for rehabilitation practitioners to understand quantitative methods in published articles. Outcomes measurement and determination of reliability are areas that have experienced rapid change during the past decade. In this study, distinctions between reliability and agreement are examined. Information is presented on analytical approaches for addressing reliability and agreement with the focus on the application of the κ statistic. The following assumptions are discussed: (1) κ should be used with data measured on a categorical scale, (2) the patients or objects categorized should be independent, and (3) the observers or raters must make their measurement decisions and judgments independently. Several issues related to using κ in measurement studies are described, including use of weighted κ, methods of reporting κ, the effect of bias and prevalence on κ, and sample size and power requirements for κ. The κ statistic is useful for assessing agreement among raters, and it is being used more frequently in rehabilitation research. Correct interpretation of the κ statistic depends on meeting the required assumptions and accurate reporting.

KW - Data analysis, statistical

KW - Rehabilitation

KW - Reliability and validity

UR - http://www.scopus.com/inward/record.url?scp=3543085125&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=3543085125&partnerID=8YFLogxK

U2 - 10.1016/j.apmr.2003.12.002

DO - 10.1016/j.apmr.2003.12.002

M3 - Article

VL - 85

SP - 1371

EP - 1376

JO - Archives of Physical Medicine and Rehabilitation

JF - Archives of Physical Medicine and Rehabilitation

SN - 0003-9993

IS - 8

ER -