When looking at the biplots generated from PCA analyses, the software program I'm using (PAST) is showing slightly different values for the loadings on the variables than it lists in the loadings spreadsheet. For instance, one variable will be given a loading of (0.2456, 0.7101) on axes 1 & 2, respectively. But in the biplot the line for that variable will end at around (0.13, 0.1). The relative magnitude and direction of the lines will be consistent between variables, but the specific end points are off. Is this because the software is graphing the variable loadings on a separate scale from the data points, and just not showing both scales? Sorry for the rather specific question, but the help manual with the program was of no help.