r - Shuffle data before doing glm, then repeating x times -

i have data contains group variable (0/1), , score pr individual, 2000 individuals. data set looks this:

id group score   a1 1 3.5   a2 1 3.2   a3 0 2.8   a4 0 2.5   

i want test if group variable can predicted through score, , have used following in r:

glm(group~score,family=binomial) 

now test p-value shuffling group variable, doing glm again. @ least 10,000 times, , possibly more, each time printing p-value score in file, there 1 row per permutation. i've looked @ sample(), struggle combine glm() , how output p-value only. in script/formula change number of permutations, , change glm formula if choose add covariates.

thank help!

you're on right track.

example (i added 1 more value suppress warnings "fitted probabilities numerically 0 or 1")

ex <- read.table(textconnection( "id group score   a1 1 3.5   a2 1 3.2   a3 0 2.8   a4 0 2.5 a5 1 2.4"),header=true)  g0 <- glm(group~score,data=ex,family=binomial) 

now need function compute summary p-value (you can on fly in replicate, way cleaner).

pvalfun <- function() {    g <- update(g0,data=transform(ex,group=sample(group)))    coef(summary(g))["score","pr(>|z|)"] } res <- replicate(1000,pvalfun()) 

or

library(plyr) res <- raply(1000,pvalfun(),.progress="text") 

or

library(glmperm) ptest2 <- prr.test(group~score,"score",data=ex,family=binomial) summary(ptest2)