Hi, Since folks are taking the time to point out some subtle issues here, taking an example from the UCLA Stats web site:
https://stats.idre.ucla.edu/other/mult-pkg/faq/general/faq-why-is-the-mann-whitney-significant-when-the-medians-are-equal/ Grp1 <- rep(c(-2, 0, 5), each = 20) Grp2 <- rep(c(-1, 0, 10), each = 20) > Grp1 [1] -2 -2 -2 -2 -2 -2 -2 -2 -2 -2 -2 -2 -2 -2 -2 -2 -2 -2 -2 -2 0 0 [23] 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 5 5 5 5 [45] 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 > Grp2 [1] -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 -1 0 0 [23] 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 10 10 10 10 [45] 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 > median(Grp1) [1] 0 > median(Grp2) [1] 0 > wilcox.test(Grp1, Grp2) Wilcoxon rank sum test with continuity correction data: Grp1 and Grp2 W = 1400, p-value = 0.03096 alternative hypothesis: true location shift is not equal to 0 So, in contrast to the original problem, here is an example where you have equal medians, but a significant test result. The key concept is that the Wilcoxon Rank Sum test is not strictly a test of differences in medians. That is, the null hypothesis for the test is not that the medians are equal, and you are either accepting or rejecting that null. Javed, I would suggest spending some time with a good tutorial on non-parametric statistics. Regards, Marc Schwartz > On Mar 19, 2019, at 6:25 PM, Jim Lemon <drjimle...@gmail.com> wrote: > > Hi Javed, > Easy. > > A<-c(2000,2100,2300,2400,6900,7000,7040,7050,7060) > median(A) > [1] 6900 > B<-c(3300,3350,3400,3450,3500,7000,7100,7200,7300) > median(B) > [1] 3500 > wilcox.test(A,B,paired=FALSE) > > Wilcoxon rank sum test with continuity correction > > data: A and B > W = 26.5, p-value = 0.233 > alternative hypothesis: true location shift is not equal to 0 > > Jim > > On Wed, Mar 20, 2019 at 3:48 AM javed khan <javedbtk...@gmail.com> wrote: >> >> Hi >> >> This is my function: >> >> wilcox.test(A,B, data = data, paired = FALSE) >> >> It gives me high p value, though the median of A column is 6900 and B >> column is 3500. >> >> Why it gives p value high if there is a difference in the median? >> >> Regards ______________________________________________ R-help@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.
