evaulating multiple responses using Chi-square
Thanks Phillip That's helpful. I hadn't thought about the one-tailed aspect of it. The idea came from path analysis, wherein the goodness-of-fit test is a sum of the chi squares for the unspecified paths. I'll look into this further. Steve J. Stephen Brewer Professor Department of Biology PO Box 1848 University of Mississippi University, Mississippi 38677-1848 Brewer web page - http://home.olemiss.edu/~jbrewer/ FAX - 662-915-5144 Phone - 662-202-5877
On 6/11/15 6:32 AM, "Dixon, Philip M [STAT]" <pdixon at iastate.edu> wrote:
Yes, the idea has been considered, by no less than RA Fisher. If you want a name, it is sometimes known as Fisher's combined probability test. Usually applied to one-sided tests of the same null hypothesis. Yes, the p-values can come from different "sources". Before using it in the community context, I would want to think carefully about what it tells me when applied to two-sided tests of different species. Philip [[alternative HTML version deleted]]
_______________________________________________ R-sig-ecology mailing list R-sig-ecology at r-project.org https://stat.ethz.ch/mailman/listinfo/r-sig-ecology