Hi all,

I was to test if individual's dimensional scores derived from principal component analysis (PCA) predicts an individual's spherical equivalence measure (gold standard clinical measurement for refractive error). To make clearer, i did a PCA using five variables and obtained individual's scores on PC1 and PC2 (where both of these components explained 74% of the variance). Categorical methods of agreement analysis will be possible after categorizing both the PCs and the clinical measurement, but I wonder if there is an alternative method which can fit with these numerical measurements having different scales. I thought Bland-Altman, Intraclass correlation coefficient , Lin's concordance correlation coefficient etc are applicable to same measuring scales.

Simply, my question is similar to "Can we perform an agreement analysis how weight (kg) can predict height (cm)?"

Thanks

Similar questions and discussions