[R] About wilcox test

[David Winsemius]

On 3/12/19 9:23 AM, Thomas Subia via R-help wrote:
> Javid wrote:
>
> "I have two set of data in excel:
> A column( 16.38, -31, -16.77, 127, -57, 23.44 and so on)
> B column ( -12, -59.23, -44, 34.23, 55.5, -12.12 and so on)


If those are in an excel spreadsheet than you need to indicate exactly 
how they  became `mydata`. And then you need to read the help page. The 
Usage section indicates that there are two forms. The first which you 
are in effect choosing has the first two arguments as data objects. The 
second version which appear to desire has a formula object as it first 
argument. Try:

This would be a way to use the first form:


  with( mydata, wilcox.test(A,  B,  paired = FALSE))

And this would succeed for the second form under the assumption that it 
is a dataframe with only two columns of numeric data:


  wilcox.test(values~ ind, data=stack(mydata), paired = FALSE)

Please read the Posting guide and in the future post in plain text.

David

>
> I run the wilcox test as :
>
> wilcox.test(A , B, data = mydata, paired = FALSE)
>
> I got always the p value very high, like 0.60
>
> Even I make changes in the data, it gives me 0.7, 0.4 etc which is too high
> than 0.05 and can not thus reject the null hypothesis.
>
> What could be the problem as I know there is difference in the data?"
> How did you conclude there is a difference in the data? 2-sample t-test?Some details and a data set would be helpful for someone to investigate this further.
> Thomas SubiaStatisticianIMG Precision
>
>
>
> 	[[alternative HTML version deleted]]
>
> ______________________________________________
> R-help using 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.