[R-sig-eco] question about adehabitatHS package

Luis Fernando García luysgarcia at gmail.com
Fri Aug 15 08:45:18 CEST 2014


Dear Friends!

I have started using the adehabitatHS for source selection. It is great,
but I have had some troubles interpreting the results and I wanted to know
if any of you knew this library well. for example I wanted to know if
 Pvalues indicate non-random (positive or negative) selection and
correspond to the chi square  value for the test. I wanted to know also for
 the plot:

> opar <- par(mfrow=c(2,2))
> plot(wiRatio)

How could I fix the last plot legend.


Thanks in advance for your help and you can find the script below!

> consumido=c(14,48,27,14,6,2,0,0,1)
> disponible=c(746,217,112,79,18,12,4,11,13)
>
names(consumido)=c("Rango1","Rango","Rango3","Rango4","Rango5","Rango6","Rango7","Rango8","Rango9")
> names(disponible)=names(consumido)
> (wi <- widesI(consumido, disponible, alpha = 0.05))



************** Manly's Selection ratios for design I ********

Significance of habitat selection:
   Khi2L       df   pvalue
126.5538   8.0000   0.0000


Table of ratios (p-values should be
 compared with Bonferroni level= 0.005555556 )
        used avail    Wi SE.Wi     P    Bi
Rango1 0.125 0.616 0.203 0.051 0.000 0.015
Rango  0.429 0.179 2.394 0.261 0.000 0.179
Rango3 0.241 0.092 2.609 0.437 0.000 0.195
Rango4 0.125 0.065 1.918 0.479 0.056 0.143
Rango5 0.054 0.015 3.607 1.433 0.069 0.270
Rango6 0.018 0.010 1.804 1.264 0.525 0.135
Rango7 0.000 0.003 0.000 0.000 0.000 0.000
Rango8 0.000 0.009 0.000 0.000 0.000 0.000
Rango9 0.009 0.011 0.832 0.829 0.840 0.062


Bonferroni classement
Based on 95 % confidence intervals on the differences of Wi :


habitat  Rango5  Rango3  Rango  Rango4  Rango6  Rango9  Rango1  Rango7
 Rango8
Rango5
 -----------------------------------------------------------------------
Rango3  -----------------------------------------------

Rango   -----------------------------------------------

Rango4  -----------------------------------------------

Rango6
 -----------------------------------------------------------------------
Rango9
 -----------------------------------------------------------------------
Rango1  --------                       ------------------------

Rango7  --------                       ----------------
 ----------------
Rango8  --------                       ----------------
 ----------------

	[[alternative HTML version deleted]]



More information about the R-sig-ecology mailing list