Help with ranking data from pairwise comparison


New Member
I am looking for a way to visualize pairwise comparisons from a non-parametric statistical test called Cochran's Q test. I am using the function below which returns the test statistic and p-value for each comparison.

nprod <- ncol(mydf)
	pd <- sort(colnames(mydf))
	cb <- combn(pd,2)
	nb <- ncol(cb) 
	out <- matrix(ncol=nb,nrow=2)
	rownames(out) <- c('Stat','Pvalue')
	colnames(out) <- apply(cb,2,paste,collapse='vs')
	for (i in 1:nb) {
		cqt <- CQT(na.omit(mydf[cb[,i]]))
		cqt.T <- cqt$T
		cqt.P <- cqt$p.value
		out[1,i] <- cqt.T
		out[2,i] <- cqt.P
From the object obtained, is there a way to visualize the comparisons with a letter-based ranking in which common letters identify levels that are not significantly different? E.g. something similar to the functions of the multcompView package?

I know nothing of programming, so please be understanding. I'll follow the thread and supply any other information, if needed.

Thank you in advance for your help!