A Proposal For Strength-Of-Agreement Criteria For Lin`s Concordance Correlation Coefficient

This equation was developed in a sample of Mexican and American adults. The body fat criterion was used for body fat proportionally derived from dual-energy X-ray absorptiometry (DXA). The reported correlation between bai and %gras (n-0.790) was greater than that of BMI (n-0.569), although the authors did not verify whether this difference in correlation coefficient was statistically significant. Similarly, the agreement between DXA`s BAI and %fett was rather bad when obesity levels were lower. Bergman et al. found similar results for cross-validation of FBCF in a sample of African Americans (4). What everyone agrees is that close to `pm;1, there is a perfect match (or perfect lag) and 0 is not a correlation; Everything in between should be interpreted with caution. The range from 0 to 1. Altman (1991) places the interpretation close to other correlation coefficients such as Pearson`s, with 80 as “excellent”. This interpretation is by no means engraved in stone; Other researchers have different interpretations. For example, McBride (2005) proposes the following guidelines for interpreting Lin`s correlation coefficient: The agreement between BAI, BAIFels, BMI and %fat was assessed on the basis of Lin`s correlation coefficient (C) (6). C assesses both the accuracy and accuracy of the relationship between two measurement methods and is the product of the correlation coefficient (-) between species-ratios and a bias correction factor (Cb) that measures the distance between the best adjustment line between them and the 45-degree line. If the correlation coefficient corresponds to one, there is a perfect match between two variables; in this specific study, this would mean that the FBCF provides a perfect estimate of %fat.

Hotelling`s t test for correlated correlations was used to test whether the correlation between BAI (baifels) and %fat is significantly different from the correlation between BMI and %fat. Sample T-tests were used to test differences in the average percentage of fat between BAI and BAIFels and %fat measured across the sample, in each sex and in certain %fat zones. In addition, general linear regressions of BAI, BAIFels and BMI were performed on %lipids measured to determine the amount of %-fat variance, which is explained by each anthropometric index. Journal: A Proposal for Strict Compliance Criteria for Lins Correlation Coefficient In this brief release, we test the FBCF`s ability to predict %fat in a sample of Euro-American adults. Our most important finding was that, although the FBCF is imprecise in cases of low obesity and generally has a poor match with %gras, the sample as a whole had a better match and a significantly stronger correlation with % fat than that between BMI and %gras. References: DG Altman (1991) Practical statistics for medical research. London: Chapman and Hall. Lin L.I-K (1989) A [CCC] for the assessment of reproducibility. Biometrics 45:255-268 McBride GB (2005) A proposal of resistance criteria for Lin`s match… NIWA Customer Report: HAM2005-062.