Forum: help

Monitor Forum | Start New Thread

Nested Flat Threaded Ultimate

Show 25 Show 50 Show 75 Show 100

Doubt about pairwise function of RRPP package [ Reply ] By: Jayme Marques on 2022-02-26 13:19

[forum:49336] Rplot.png (1) downloads

Hello R forge comunnit, I have a specific doubt about RRPP and GEOMORPH package functions. I am trying to compare the trajectory allometry of skull shape in four species of snakes ( my data of shape are geometric morphometric coordinates). I am using the function Procd.lm to run the following MANCOVA: alometria.all<-procD.lm(shape.sym~log(ttlength)*spp,iter=999) Using the summary function I obteined this: summary(alometria.all) ## RESULT Df SS MS Rsq F Z Pr(>F) log(ttlength) 1 0.14032 0.140319 0.34286 93.479 5.7990 0.001 ** spp 3 0.06072 0.020239 0.14836 13.483 8.9833 0.001 ** log(ttlength):spp 3 0.01008 0.003359 0.02463 2.238 3.8872 0.001 ** Residuals 132 0.19814 0.001501 0.48415 Total 139 0.40926 # My first question is: Is this result saying that at least one species differs from the others in the slope trajectory? correctly ? Using the pairwise function in the MANCOVA, to pairwise compare the slopes between species the result is this: allo_ALL<-pairwise(alometria.all, groups=spp) summary(allo_ALL) #Pairwise distances between means, plus statistics d UCL (95%) Z Pr > d BOTALT:BOTFON 0.04171967 0.04842692 -0.1259108 0.536 BOTALT:BOTJAR 0.03888497 0.04319270 0.1569560 0.438 BOTALT:BOTPUB 0.02516634 0.03021651 -0.8676365 0.817 BOTFON:BOTJAR 0.02331394 0.03105392 -1.0254375 0.847 BOTFON:BOTPUB 0.04259982 0.05003583 -0.4108532 0.654 BOTJAR:BOTPUB 0.03828906 0.04364420 -0.3659096 0.638 Saying that all my species do not differ in the distance means Correctly? I want to know how can I dyscover what is the different pointed between the species in the first summary. I noted that using pairwise function with the TOTAL LENGTH (ttlength) as a covariable: allo_ALL<-pairwise(alometria.all, groups=spp, covariete= ttlength) the result is: Pairwise distances between slope vector (end-points), plus statistics d UCL (95%) Z Pr > d BOTALT:BOTFON 0.0006912669 0.0007153177 1.3756084 0.083 BOTALT:BOTJAR 0.0007486346 0.0004477359 4.7925516 0.001** BOTALT:BOTPUB 0.0006960572 0.0005804760 2.8198602 0.002** BOTFON:BOTJAR 0.0004313468 0.0006309879 -0.4105356 0.650 BOTFON:BOTPUB 0.0004007854 0.0006455627 -0.9518647 0.821 BOTJAR:BOTPUB 0.0002903250 0.0004558577 -0.7147862 0.754 Showing significant differents between species slopes. But I dont know if its making sense in real.. Could some one help me please? Thanks for the attention!