On 22 Nov 2002 22:01:59 -0800, [EMAIL PROTECTED] (Jay Warner) wrote: >So how do I do the same thing, when the measurement in each class is >not a count? I have a mass for each class, an amount on a continuous >scale, which I can also report as a percentage of the total.
I don't think we can say without knowing where these masses came from. If they were just conversions of counts of independent observations, then the chi-square method is probably what you want. If they arose in some other way, it might not be. The general idea is to think of some discrepancy measure between what you observed and what you expected to observe, and then somehow work out the distribution of that measure if the null hypothesis is true. The chi-square is one such statistic, but there are lots of them. The result that the null distribution is chi-square on a certain number of degrees of freedom is based on the assumption that the values arose as counts of independent observations. If they didn't, the reference distribution is likely different. Duncan Murdoch . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
