There is a considerable literature on establishing equivalence between methods on the basis of their ability to enumerate samples. Various measures have been used therein, all with some shortcomings. The "concordance correlation coefficient" was first proposed by Lin (1989) for assessment of concordance in continuous data. It represents a breakthrough in assessing agreement between alternative methods for continuous data in that it appears to avoid all the shortcomings associated with the panoply of usual procedures (Pearson correlation coefficient r, paired t-tests, least squares analysis for slope and intercept, coefficient of variation, intraclass correlation coefficient). It is robust on as few as 10 pairs of data (Lin 1989). It appears, with a worked example, in the popular biostatistical text by Zar (1996). It is important to note that there are typographical errors in Lin’s original paper (and therefore repeated in Zar’s book). Corrections have since been published (Lin 2000).
A precautionary testing procedure analogous to that for Cohen’s kappa has been implemented in this calculator. However there is as yet no literature giving a descriptive scale for the degree of agreement. Accordingly it is tentatively suggested that the following scales be used for values of the coefficient in given ranges.
|Strength-of-agreement||Continuous variables||QuantiTray methods|
Feedback from users of this calculator on the appropriateness of this scale is sought.
Replace the data in this input box with your data. They must be comma-, space-, or tab-delimited. They can be copied from an Excel spreadsheet if you wish.
For further details
See the technical report written for the Ministry of Health. Further details are given in the recent book: McBride, G.B. (2005). Using Statistical Methods for Water Quality Management: Issues, Problems and Solutions. Wiley, New York.