| How reliable are chance-corrected measures of agreement? | |
| | |
MedLine Citation:
|
PMID: 8310189 Owner: NLM Status: MEDLINE |
Abstract/OtherAbstract:
|
Chance-corrected measures of agreement are prone to exhibit paradoxical and counter-intuitive results when used as measures of reliability. It is demonstrated that these problems arise with Cohen's kappa as well as with Aickin's alpha. They are the consequence of an analogy to Simpson's paradox in mixed populations. It is further shown that chance-corrected measures of agreement may yield misleading values for binary ratings. It is concluded that improvements in the design and the analysis of reliability studies are a prerequisite for valid and pertinent results. |
| | |
Authors:
|
I Guggenmoos-Holzmann |
Related Documents
:
|
11809359 - The work limitations questionnaire's validity and reliability among patients with osteo... 15587489 - The validity and reliability of the global index of safety (gis). 10424679 - Psychophysical assessment of the affective components of non-painful touch. 19237509 - Comparison of cephalometric measurements with digital versus conventional cephalometric... 18851739 - Production and quality of clinical practice guidelines in argentina (1994-2004): a cros... 17540249 - Confronting models with data. 20400369 - Hiv-1 subtype c transmission network: the phylogenetic reconstruction strongly supports... 22115179 - Irefr: an r package to manipulate the irefindex consolidated protein interaction database. 3986299 - Comparison of nonlinear pharmacokinetic parameters estimated from the sinusoidal perfus... |
Publication Detail:
|
Type: Journal Article; Research Support, Non-U.S. Gov't |
Journal Detail:
|
Title: Statistics in medicine Volume: 12 ISSN: 0277-6715 ISO Abbreviation: Stat Med Publication Date: 1993 Dec |
Date Detail:
|
Created Date: 1994-03-14 Completed Date: 1994-03-14 Revised Date: 2006-11-15 |
Medline Journal Info:
|
Nlm Unique ID: 8215016 Medline TA: Stat Med Country: ENGLAND |
Other Details:
|
Languages: eng Pagination: 2191-205 Citation Subset: IM |
Affiliation:
|
Department of Biostatistics and Medical Informatics, Free University of Berlin, Germany. |
Export Citation:
|
APA/MLA Format Download EndNote Download BibTex |
| MeSH Terms | |
Descriptor/Qualifier:
|
Classification Data Interpretation, Statistical* Diagnosis* Epidemiology* Humans Likelihood Functions Linear Models Models, Statistical* Observer Variation* Reproducibility of Results Sensitivity and Specificity* |
From MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine
Previous Document: Bayesian analysis of binary data from an audit of cervical smears.
Next Document: Estimating kappa from binocular data and comparing marginal probabilities.