[R-sig-eco] split-split-plot design and adonis
Arnaud Foulquier
arnaud.foulquier at gmail.com
Sun Mar 4 18:28:00 CET 2012
I have run the two analysis. The only differences are the p-values that
appear in the anova tables. Other values (Df, SumsOfSqs, MeanSqs, F.Model)
are not modified because they are calculated on the "original" (not
permuted) data.
The p-value is calculated by comparing the value of F obtained with the
original data to the distribution obtained by permutations. If i understand
well, the strata argument is used to constrain the permutations of the
original data within the groups specified in the strata argument. The
distribution of the F-statistic that is generated by permutations is not the
same according to whether the strata argument is specified or not. This is
why the p-values are modified when the strata argument is specified.
However, I'm not sure that strata=Site_Hydro correctly specify the nested
structure of my data.
--
View this message in context: http://r-sig-ecology.471788.n2.nabble.com/split-split-plot-design-and-adonis-tp7332029p7342690.html
Sent from the r-sig-ecology mailing list archive at Nabble.com.
More information about the R-sig-ecology
mailing list