Hi,
I'm attempting to use the pmgram function of the ecodist package to
construct partial mantel correlograms and subsequent plots for analyzing
fish diversity measures in relation to site proximity at a number of set
distances, while controlling for environmental factors. My trouble is,
although the function is showing me the patterns I expect, I'm getting
"piecer" values that appear somewhat unusually high compared to what I'm
used to seeing for for a mantel r. In one case, I get a piecer value of > 1
(1.27). I belive I've read the relevant documents for ecodist (jstatsoft
article, ecodist pdf, etc), but I'm wondering if maybe I'm interpreting this
"piecer" value incorrectly, or if maybe I'm doing something else wrong. Any
help or suggestions are greatly appreciated.
A code snippet:
m.H <-
pmgram(dist(decostand(data[which(data$Group=="H"),c(2:12,14:39)],"hell")),as.dist(dist.H),dist(data[which(data$Group=="H"),41:48]),nperm=1000,breaks=c(0,5,10,20,30,40,50))
community data is a row (sites) by column (species) matrix. In this example
I've selected a specific group of sites and species. dist.H is a pairwise
distance matrix of (swim) distances between sites. The last bit for the
partial is columns of environmental variables.
And output:
$mgram
lag ngroup piecer pval
[1,] 5 21 1.26954133 0.001
[2,] 10 35 0.75870281 0.001
[3,] 20 57 0.37745070 0.011
[4,] 30 32 0.54731569 0.015
[5,] 40 48 0.17201527 0.408
[6,] 50 76 -0.08489383 0.584
Sincerely,
Roy Martin
USEPA Office of Research and Development
Cincinnati, OH
--
View this message in context:
http://r-sig-ecology.471788.n2.nabble.com/Problem-with-ecodist-pmgram-tp7578703.html
Sent from the r-sig-ecology mailing list archive at Nabble.com.
_______________________________________________
R-sig-ecology mailing list
R-sig-ecology@r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-ecology
