[R-sig-eco] adonis with continuous variables

[alfredo Tello]

Hi Jari (and everyone else who might read this),

Just one last thing (I hope I'm not being annoying! ). I believe this can be
of interest to other people as well.

In a recent paper in the ISME (International Society for Microbial Ecology)
Journal, Thomas Bell used adonis to test for evidence of a distance-decay
relationship in bacteria throughout a woodland. He shows the distance-decay
plot (distance vs. bray curtis dissimilarity) and then tests for it using
adonis and presents p-values and r^2 from adonis. 

This would suggest to me that using adonis on a regression type approach
could, in distance-decay studies, replace the traditional mantel test or
multiple regression approach. However, I don't understand how you would
input this into adonis (on the right hand side) so that you would in fact be
testing for "distance" and not for "location". Basically, to input the
pairwise distant differences(?¿).

Any comments are welcome!

Cheers,

A

--
View this message in context: http://r-sig-ecology.471788.n2.nabble.com/adonis-with-continuous-variables-tp6155914p6157342.html
Sent from the r-sig-ecology mailing list archive at Nabble.com.