Abstract

The need to quantify agreement between two raters or two methods of measuring a response often arises in research. Kappa statistics (unweighted and weighted) are appropriate when the data are nominal or ordinal, whereas the concordance correlation coefficient is more appropriate when the data are measured on a continuous scale. We develop weighted product-moment and concordance correlation coefficients which are applicable for repeated measurements study designs. We consider two distinct situations in which the repeated measurements are paired or unpaired over time. We illustrate the methodology with examples comparing (1) two assays for measuring serum cholesterol, (2) two estimates of dietary intake, from a food frequency questionnaire and dietary recalls, and (3) two measurements of percentage body fat, from skinfold calipers and dual energy x-ray absorptiometry.

Original languageEnglish (US)
Pages (from-to)341-353
Number of pages13
JournalBiometrics
Volume52
Issue number1
DOIs
StatePublished - Mar 1 1996

Fingerprint

Repeated Measurement Designs
Concordance
Correlation coefficient
Adipose Tissue
Eating
Cholesterol
X-Rays
Product Moments
Repeated Measurements
Serum
Research
Questionnaire
Categorical or nominal
Percentage
calipers
Quantify
diet recall
food frequency questionnaires
Statistics
Oils and fats

All Science Journal Classification (ASJC) codes

  • Statistics and Probability
  • Biochemistry, Genetics and Molecular Biology(all)
  • Immunology and Microbiology(all)
  • Agricultural and Biological Sciences(all)
  • Applied Mathematics

Cite this

Chinchilli, Vernon M. ; Martel, Juliann K. ; Kumanyika, Shiriki ; Lloyd, Tom. / A weighted concordance correlation coefficient for repeated measurement designs. In: Biometrics. 1996 ; Vol. 52, No. 1. pp. 341-353.
@article{bc668eed346043d6a0de2aa3852d29a9,
title = "A weighted concordance correlation coefficient for repeated measurement designs",
abstract = "The need to quantify agreement between two raters or two methods of measuring a response often arises in research. Kappa statistics (unweighted and weighted) are appropriate when the data are nominal or ordinal, whereas the concordance correlation coefficient is more appropriate when the data are measured on a continuous scale. We develop weighted product-moment and concordance correlation coefficients which are applicable for repeated measurements study designs. We consider two distinct situations in which the repeated measurements are paired or unpaired over time. We illustrate the methodology with examples comparing (1) two assays for measuring serum cholesterol, (2) two estimates of dietary intake, from a food frequency questionnaire and dietary recalls, and (3) two measurements of percentage body fat, from skinfold calipers and dual energy x-ray absorptiometry.",
author = "Chinchilli, {Vernon M.} and Martel, {Juliann K.} and Shiriki Kumanyika and Tom Lloyd",
year = "1996",
month = "3",
day = "1",
doi = "10.2307/2533172",
language = "English (US)",
volume = "52",
pages = "341--353",
journal = "Biometrics",
issn = "0006-341X",
publisher = "Wiley-Blackwell",
number = "1",

}

A weighted concordance correlation coefficient for repeated measurement designs. / Chinchilli, Vernon M.; Martel, Juliann K.; Kumanyika, Shiriki; Lloyd, Tom.

In: Biometrics, Vol. 52, No. 1, 01.03.1996, p. 341-353.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A weighted concordance correlation coefficient for repeated measurement designs

AU - Chinchilli, Vernon M.

AU - Martel, Juliann K.

AU - Kumanyika, Shiriki

AU - Lloyd, Tom

PY - 1996/3/1

Y1 - 1996/3/1

N2 - The need to quantify agreement between two raters or two methods of measuring a response often arises in research. Kappa statistics (unweighted and weighted) are appropriate when the data are nominal or ordinal, whereas the concordance correlation coefficient is more appropriate when the data are measured on a continuous scale. We develop weighted product-moment and concordance correlation coefficients which are applicable for repeated measurements study designs. We consider two distinct situations in which the repeated measurements are paired or unpaired over time. We illustrate the methodology with examples comparing (1) two assays for measuring serum cholesterol, (2) two estimates of dietary intake, from a food frequency questionnaire and dietary recalls, and (3) two measurements of percentage body fat, from skinfold calipers and dual energy x-ray absorptiometry.

AB - The need to quantify agreement between two raters or two methods of measuring a response often arises in research. Kappa statistics (unweighted and weighted) are appropriate when the data are nominal or ordinal, whereas the concordance correlation coefficient is more appropriate when the data are measured on a continuous scale. We develop weighted product-moment and concordance correlation coefficients which are applicable for repeated measurements study designs. We consider two distinct situations in which the repeated measurements are paired or unpaired over time. We illustrate the methodology with examples comparing (1) two assays for measuring serum cholesterol, (2) two estimates of dietary intake, from a food frequency questionnaire and dietary recalls, and (3) two measurements of percentage body fat, from skinfold calipers and dual energy x-ray absorptiometry.

UR - http://www.scopus.com/inward/record.url?scp=0029927989&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0029927989&partnerID=8YFLogxK

U2 - 10.2307/2533172

DO - 10.2307/2533172

M3 - Article

C2 - 8934601

AN - SCOPUS:0029927989

VL - 52

SP - 341

EP - 353

JO - Biometrics

JF - Biometrics

SN - 0006-341X

IS - 1

ER -