concordance correlation coefficient using R
on 12/07/2008 09:23 AM paul murima wrote:
Hi. I have data which i would want to assess the degree of agreement between two assays, e.g., to evaluate reproducibility or for inter-rater reliability. I have used the Pearson product-moment correlation coefficient. It looks good ranginging between 0.90 to 0.998. Though this looks good. I am told the Concordance correlation coefficient will give a better picture of how reproducible the assay is. Does any one know how to run such a test using R? If so can you kindly provide the code, and bit of explanations to spruce it up? Thank you Paul
Look at the epi.ccc function in the epiR CRAN package. More information here: http://bm2.genes.nig.ac.jp/RGM2/R_current/library/epiR/man/epi.ccc.html You might also want to review Martin Bland's page pertaining to the design and analysis of measurement agreement studies: http://www-users.york.ac.uk/~mb55/meas/meas.htm HTH, Marc Schwartz