I have been using R to calculate the significance of Procrustes correlations. With one series of data, where there are five cases, the value returned for the correlation coefficient is one although there are differences as shown by the procrustes error graph. Is there a statistical reason for this? Similarly, when we look at the cas scores there is not a systematic relationship between the two series of data. Thanks Tara -- Tara Dirilgen School of Biology & Environmental Science Science Center West University College Dublin Belfield, Dublin 4 IRELAND tara.dirilgen at ucdconnect.ie [[alternative HTML version deleted]]
Hi Tara, Providing a simple example script that reproduces your case and using it to support your question would increase your chances to obtain an answer. Best, Pierrick On Thu, Feb 19, 2015 at 5:43 PM, Tara Dirilgen <tara.dirilgen at ucdconnect.ie> wrote:> I have been using R to calculate the significance of Procrustes > correlations. With one series of data, where there are five cases, the > value returned for the correlation coefficient is one although there are > differences as shown by the procrustes error graph. Is there a statistical > reason for this? Similarly, when we look at the cas scores there is not a > systematic relationship between the two series of data. > > Thanks > > Tara > > -- > Tara Dirilgen > School of Biology & Environmental Science > Science Center West > University College Dublin > Belfield, Dublin 4 > IRELAND > tara.dirilgen at ucdconnect.ie > > [[alternative HTML version deleted]] > > ______________________________________________ > R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code.
Hi again, Just to clarify my question from previous email... Example script: A <- data[, c(1,2,3,4)] B <- data[, c(5,6,7,8)] library(vegan) vare.proc <- procrustes(A,B, scale=FALSE, symmetric=FALSE) vare.proc summary(vare.proc) plot(vare.proc) #plot(vare.proc, kind=2) residuals(vare.proc) protest(A,B, scores = "sites", permutations = 999) Data A and B are case scores from PCA's. *Question* Protest output- 'Correlation in a symmetric Procrustes rotation' comes out as 1 for each different data I input. This does not seem correct as the graphical output suggests otherwise as well the raw case scores show that there is not a systematic relationship between the two series of data. I am wondering what the reason for this is. Thanks & kind regards, Tara On 19 February 2015 at 16:43, Tara Dirilgen <tara.dirilgen at ucdconnect.ie> wrote:> I have been using R to calculate the significance of Procrustes > correlations. With one series of data, where there are five cases, the > value returned for the correlation coefficient is one although there are > differences as shown by the procrustes error graph. Is there a statistical > reason for this? Similarly, when we look at the cas scores there is not a > systematic relationship between the two series of data. > > Thanks > > Tara > > -- > Tara Dirilgen > School of Biology & Environmental Science > Science Center West > University College Dublin > Belfield, Dublin 4 > IRELAND > tara.dirilgen at ucdconnect.ie >-- Tara Dirilgen School of Biology & Environmental Science Science Center West University College Dublin Belfield, Dublin 4 IRELAND tara.dirilgen at ucdconnect.ie [[alternative HTML version deleted]]