Comparison of two diverse measurement methods by graphical agreement

Abstract

Biological processes are complex, and several methods are often used to measure them. However, different methods could determine diverse parts of a single biological process. To date, there are no widely accepted and convenient methods for comparison between the results, so we consider graphical analysis with the ability to demonstrate the pattern of distribution of findings from one method across another. It appears that a two-series area plot is the most appropriate. After using normal values and a coding reference and examining the variables, unnecessary information is diminished and the graphics become more obvious. Three possibilities may be found: agreement or disagreement between variables or disagreement from normal values. Therefore, the graph may also be used to determine the corresponding normal values between variables. The association between variables may be tested using kappa coefficients, although graphical analysis remains more informative. Therefore, graphical analysis could compare two completely different variables that measure the same biological process or determine the range of normal values. © 2012 Informa Healthcare USA, Inc.