R help - Nov 2006 - interprete wilcox.test results

Dear All,

I am using wilcox.test to test two samples, data_a and data_b, earch sample has
3 replicates, suppose data_a and data_b are 20*3 matrix. Then I used the
following to test the null hypothesis (they are from same distribution.):

wilcox.test(x=data_a, y=data_b, alternative="g")
I got pvalue = 1.90806170863311e-09.

When I switched data_a and data_b by doing the following:
wilcox.test(x=data_b, y=data_a, alternative="g")
I got pvalue = 0.999999998111886.

I suppose because I am test them with same null hypothesis, I should get same
result (pvalue). Since I am not a savy statistician at all, maybe that is a
wrong assumption, could anyone please point out the right interpretion please?

Thanks!

Best,
 Baoqiang

On Wed, 8 Nov 2006, Baoqiang Cao wrote:
> Dear All,
>
> I am using wilcox.test to test two samples, data_a and data_b, earch sample
has 3 replicates, suppose data_a and data_b are 20*3 matrix. Then I used the
following to test the null hypothesis (they are from same distribution.):
>
> wilcox.test(x=data_a, y=data_b, alternative="g")
> I got pvalue = 1.90806170863311e-09.
>
> When I switched data_a and data_b by doing the following:
> wilcox.test(x=data_b, y=data_a, alternative="g")
> I got pvalue = 0.999999998111886.
>
> I suppose because I am test them with same null hypothesis,
no, you don't:

H_0: mean(data_a) <= mean(data_b) (alternative = "greater")

is different from

H_0: mean(data_b) <= mean(data_a)

check

wilcox.test(x=data_a, y=data_b, alternative="l")

Torsten
> I should get same result (pvalue). Since I am not a savy statistician at 
> all, maybe that is a wrong assumption, could anyone please point out the 
> right interpretion please?
>
> Thanks!
>
> Best,
> Baoqiang
>
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> 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.
>
>

Thanks Torsten! I got it!

Best,
 Baoqiang

---- Original message ----
>Date: Wed, 8 Nov 2006 17:26:18 +0100 (CET)
>From: Torsten Hothorn <Torsten.Hothorn at rzmail.uni-erlangen.de>  
>Subject: Re: [R] interprete wilcox.test results  
>To: Baoqiang Cao <caobg at email.uc.edu>
>Cc: r-help at stat.math.ethz.ch
>
>
>On Wed, 8 Nov 2006, Baoqiang Cao wrote:
>
>> Dear All,
>>
>> I am using wilcox.test to test two samples, data_a and data_b, earch
sample has 3 replicates, suppose data_a and data_b are 20*3 matrix. Then I used
the following to test the null hypothesis (they are from same distribution.):
>>
>> wilcox.test(x=data_a, y=data_b, alternative="g")
>> I got pvalue = 1.90806170863311e-09.
>>
>> When I switched data_a and data_b by doing the following:
>> wilcox.test(x=data_b, y=data_a, alternative="g")
>> I got pvalue = 0.999999998111886.
>>
>> I suppose because I am test them with same null hypothesis,
>
>no, you don't:
>
>H_0: mean(data_a) <= mean(data_b) (alternative = "greater")
>
>is different from
>
>H_0: mean(data_b) <= mean(data_a)
>
>check
>
>wilcox.test(x=data_a, y=data_b, alternative="l")
>
>Torsten
>
>> I should get same result (pvalue). Since I am not a savy statistician
at
>> all, maybe that is a wrong assumption, could anyone please point out
the
>> right interpretion please?
>>
>> Thanks!
>>
>> Best,
>> Baoqiang
>>
>> ______________________________________________
>> R-help at stat.math.ethz.ch mailing list
>> 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.
>>
>>