Hello,
I have a problem with conflicting advice concerning some
statistics I need to run for my project and thought it would be a good idea to
just openly ask some people not involved in the actual project. I am looking at
gene flow (based on mtDNA) between several populations of frogs. Basically, I am interested in how (1)
geographic distance and (2) the number of roads between study ponds influence
genetic variation.
Now my statistics question:
(1) Do
I simply run one partial Mantel test looking at the influence of one variable
while controlling for the effect of the other, or, do I run two separate Mantel
tests, plus a partial Mantel?
(2) Regardless
of which test I use, do I test one-sided or two-sided? I obviously expect to
find less variation with increasing distance and more roads, but a lot of
people cautioned me and said it is always better to test two-sided to be more
conservative and to account for patterns different from what I would expect to
find.
I would highly appreciate if someone could give me some
advice on these issues.
Many thanks in advance and regards,
Herbert
_________________________________________________________________
Drag n’ drop—Get easy photo sharing with Windows Live™ Photos.
http://www.microsoft.com/windows/windowslive/products/photos.aspx
|