[R] Post-hoc pairwise Fisher’s exact tests

[AbouEl-Makarim Aboueissa]

Dear All:



Re: Post-hoc pairwise Fisher’s exact tests with confidence Intervals



I am trying to conduct a Post-hoc pairwise comparison for comparing the
proportions of the ones in the 7 groups using the fisher exact test. I have
a data with 7 variables, data = (0,1). Event = 1.



*Q1.* Is there a way to have the confidence intervals of Pi-Pj in the R
output of the R function “*pairwise_fisher_test()*”. Also, R yields only
the p values of the first 10 pairs (Pi,Pj).



*Q2.* I used the summary of the raw data, please see below. Is there a way
to use the raw data instead, please see the attached file?



*Q3.* I got an error message when I tried to use the R function “
*fisher_test()*”





#### R codes





Data1 <- as.table(rbind(

  c(24,9,28,37,18,10,28),                       # number of ones

  c(65,80,61,52,71,79,61)                       # number of zeros

))



dimnames(Data1) <- list(

  cases = c("yes", "no"),

  vars = c("x1", "x2", "x3", "x4", "x5", "x6", "x7")

)



Data1



install.packages("rstatix")

library(rstatix)





fisher_test(Data1, detailed = TRUE)



##### pairwise fisher test: Compare the proportion of ones between groups



pairwise_fisher_test(Data1)







With many thanks

abou


______________________


*AbouEl-Makarim Aboueissa, PhD*

*Professor, Statistics and Data Science*
*Graduate Coordinator*

*Department of Mathematics and Statistics*
*University of Southern Maine*

-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: Data_for_R_example.txt
URL: <https://stat.ethz.ch/pipermail/r-help/attachments/20211125/a6858e52/attachment.txt>