Dear all, I have a question regarding the 'y <- -loc[,2]' in ?cmdscale. Although I see that the plot is more sensible when using the '-loc' instead of just 'y <- loc[,2]', I don't understand if there is a statistical reason to do '-loc[,2]'. So is this just to make the graph look better, or should I always use -loc for the y-axis of a similar plot for a completely different data set. Kind regards, Paul Lemmens