Quantcast

kruskal wallis post hoc?

classic Classic list List threaded Threaded
7 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

kruskal wallis post hoc?

Iasonas Lamprianou
Dear all,

I run a kruskal wallis test and found significant results. Then, I conducted all pairwise comparisons and found no significant results. Could anyone please give me a hint as to why this happens or redirect me towards a specific web page where I can find more info? (I used alpha=5% and made no bonferroni or other correction for the pairwise comparisons)
Thank you

 
Dr. Iasonas Lamprianou
Department of Social and Political Sciences
University of Cyprus
        [[alternative HTML version deleted]]


______________________________________________
[hidden email] 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.
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: kruskal wallis post hoc?

gaiarrido
CONTENTS DELETED
The author has deleted this message.
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: kruskal wallis post hoc?

Tal Galili
In reply to this post by Iasonas Lamprianou
Hi Iasonas ,
This is a stat question and not an R question.
But the general answer is that it could happen :)

The R question would have been if their is a Tukey HSD for kruskel.test,
the answer is yes:
http://stats.stackexchange.com/questions/17342/is-there-a-nonparametric-equivalent-of-tukey-hsd
But if you didn't get any significant result from the pairwise comparison,
I would say that the post hoc correction wouldn't help you (it could be
that the reason for this significance is based on some weird contrast...)

Best,
Tal





----------------Contact
Details:-------------------------------------------------------
Contact me: [hidden email] |  972-52-7275845
Read me: www.talgalili.com (Hebrew) | www.biostatistics.co.il (Hebrew) |
www.r-statistics.com (English)
----------------------------------------------------------------------------------------------




On Thu, Jan 12, 2012 at 8:35 AM, Iasonas Lamprianou <[hidden email]>wrote:

> Dear all,
>
> I run a kruskal wallis test and found significant results. Then, I
> conducted all pairwise comparisons and found no significant results. Could
> anyone please give me a hint as to why this happens or redirect me towards
> a specific web page where I can find more info? (I used alpha=5% and made
> no bonferroni or other correction for the pairwise comparisons)
> Thank you
>
>
> Dr. Iasonas Lamprianou
> Department of Social and Political Sciences
> University of Cyprus
>        [[alternative HTML version deleted]]
>
>
> ______________________________________________
> [hidden email] 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.
>
>

        [[alternative HTML version deleted]]

______________________________________________
[hidden email] 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.
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: kruskal wallis post hoc?

Iasonas Lamprianou
Thank you for the result, I will have a look at the link.

 
Dr. Iasonas Lamprianou
Department of Social and Political Sciences
University of Cyprus


>________________________________
> From: Tal Galili <[hidden email]>
>To: Iasonas Lamprianou <[hidden email]>
>Cc: "[hidden email]" <[hidden email]>
>Sent: Thursday, 12 January 2012, 10:48
>Subject: Re: [R] kruskal wallis post hoc?
>
>
>Hi Iasonas ,
>This is a stat question and not an R question.
>But the general answer is that it could happen :)
>
>
>The R question would have been if their is a Tukey HSD for kruskel.test, the answer is yes:
>http://stats.stackexchange.com/questions/17342/is-there-a-nonparametric-equivalent-of-tukey-hsd 
>But if you didn't get any significant result from the pairwise comparison, I would say that the post hoc correction wouldn't help you (it could be that the reason for this significance is based on some weird contrast...)
>
>
>Best,
>Tal
>
>
>
>
>
>
>
>
>----------------Contact Details:-------------------------------------------------------
>Contact me: [hidden email] |  972-52-7275845
>Read me: www.talgalili.com (Hebrew) | www.biostatistics.co.il (Hebrew) | www.r-statistics.com (English)
>----------------------------------------------------------------------------------------------
>
>
>
>
>
>On Thu, Jan 12, 2012 at 8:35 AM, Iasonas Lamprianou <[hidden email]> wrote:
>
>Dear all,
>>
>>I run a kruskal wallis test and found significant results. Then, I conducted all pairwise comparisons and found no significant results. Could anyone please give me a hint as to why this happens or redirect me towards a specific web page where I can find more info? (I used alpha=5% and made no bonferroni or other correction for the pairwise comparisons)
>>Thank you
>>
>> 
>>Dr. Iasonas Lamprianou
>>Department of Social and Political Sciences
>>University of Cyprus
>>       [[alternative HTML version deleted]]
>>
>>
>>______________________________________________
>>[hidden email] 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.
>>
>>
>
>
>
        [[alternative HTML version deleted]]


______________________________________________
[hidden email] 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.
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: kruskal wallis post hoc?

Frank Harrell
The Kruskal-Wallis test is a special case of the proportional odds ordinal logistic model.  You can get any contrast you want by testing regression coefficients.   In a couple of weeks the rms package's contrast function will allow for individual confidence intervals of effects that together have a 0.05 type I error, by using the multcomp package (called automatically from contrast.rms).
Frank
Iasonas Lamprianou wrote
Thank you for the result, I will have a look at the link.

 
Dr. Iasonas Lamprianou
Department of Social and Political Sciences
University of Cyprus


>________________________________
> From: Tal Galili <[hidden email]>
>To: Iasonas Lamprianou <[hidden email]>
>Cc: "[hidden email]" <[hidden email]>
>Sent: Thursday, 12 January 2012, 10:48
>Subject: Re: [R] kruskal wallis post hoc?
>
>
>Hi Iasonas ,
>This is a stat question and not an R question.
>But the general answer is that it could happen :)
>
>
>The R question would have been if their is a Tukey HSD for kruskel.test, the answer is yes:
>http://stats.stackexchange.com/questions/17342/is-there-a-nonparametric-equivalent-of-tukey-hsd 
>But if you didn't get any significant result from the pairwise comparison, I would say that the post hoc correction wouldn't help you (it could be that the reason for this significance is based on some weird contrast...)
>
>
>Best,
>Tal
>
>
>
>
>
>
>
>
>----------------Contact Details:-------------------------------------------------------
>Contact me: [hidden email] |  972-52-7275845
>Read me: www.talgalili.com (Hebrew) | www.biostatistics.co.il (Hebrew) | www.r-statistics.com (English)
>----------------------------------------------------------------------------------------------
>
>
>
>
>
>On Thu, Jan 12, 2012 at 8:35 AM, Iasonas Lamprianou <[hidden email]> wrote:
>
>Dear all,
>>
>>I run a kruskal wallis test and found significant results. Then, I conducted all pairwise comparisons and found no significant results. Could anyone please give me a hint as to why this happens or redirect me towards a specific web page where I can find more info? (I used alpha=5% and made no bonferroni or other correction for the pairwise comparisons)
>>Thank you
>>
>> 
>>Dr. Iasonas Lamprianou
>>Department of Social and Political Sciences
>>University of Cyprus
>>       [[alternative HTML version deleted]]
>>
>>
>>______________________________________________
>>[hidden email] 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.
>>
>>
>
>
>
        [[alternative HTML version deleted]]


______________________________________________
[hidden email] 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.
Frank Harrell
Department of Biostatistics, Vanderbilt University
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: kruskal wallis post hoc?

Peter Dalgaard-2

On Jan 12, 2012, at 14:11 , Frank Harrell wrote:

> The Kruskal-Wallis test is a special case of the proportional odds ordinal
> logistic model.

Eh? Can you elaborate on that?

I would expect that at best it is equivalent to some _test_ in a polr-type model. It is never really clear what the model is when some groups are different and others not.

>  You can get any contrast you want by testing regression
> coefficients.   In a couple of weeks the rms package's contrast function
> will allow for individual confidence intervals of effects that together have
> a 0.05 type I error, by using the multcomp package (called automatically
> from contrast.rms).
> Frank
>
> Iasonas Lamprianou wrote
>>
>> Thank you for the result, I will have a look at the link.
>>
>>  
>> Dr. Iasonas Lamprianou
>> Department of Social and Political Sciences
>> University of Cyprus
>>
>>
>>> ________________________________
>>> From: Tal Galili &lt;tal.galili@&gt;
>>> To: Iasonas Lamprianou &lt;lamprianou@&gt;
>>> Cc: "r-help@" &lt;r-help@&gt;
>>> Sent: Thursday, 12 January 2012, 10:48
>>> Subject: Re: [R] kruskal wallis post hoc?
>>>
>>>
>>> Hi Iasonas ,
>>> This is a stat question and not an R question.
>>> But the general answer is that it could happen :)
>>>
>>>
>>> The R question would have been if their is a Tukey HSD for kruskel.test,
> the answer is yes:
>>> http://stats.stackexchange.com/questions/17342/is-there-a-nonparametric-equivalent-of-tukey-hsd 
>>> But if you didn't get any significant result from the pairwise comparison,
> I would say that the post hoc correction wouldn't help you (it could be that
> the reason for this significance is based on some weird contrast...)
>>>
>>>
>>> Best,
>>> Tal
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>> ----------------Contact
> Details:-------------------------------------------------------
>>> Contact me: Tal.Galili@ |  972-52-7275845
>>> Read me: www.talgalili.com (Hebrew) | www.biostatistics.co.il (Hebrew) |
> www.r-statistics.com (English)
>>> ----------------------------------------------------------------------------------------------
>>>
>>>
>>>
>>>
>>>
>>> On Thu, Jan 12, 2012 at 8:35 AM, Iasonas Lamprianou &lt;lamprianou@&gt;
> wrote:
>>>
>>> Dear all,
>>>>
>>>> I run a kruskal wallis test and found significant results. Then, I
> conducted all pairwise comparisons and found no significant results. Could
> anyone please give me a hint as to why this happens or redirect me towards a
> specific web page where I can find more info? (I used alpha=5% and made no
> bonferroni or other correction for the pairwise comparisons)
>>>> Thank you
>>>>
>>>>  
>>>> Dr. Iasonas Lamprianou
>>>> Department of Social and Political Sciences
>>>> University of Cyprus
>>>>        [[alternative HTML version deleted]]
>>>>
>>>>
>>>> ______________________________________________
>>>> R-help@ 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.
>>>>
>>>>
>>>
>>>
>>>
>> [[alternative HTML version deleted]]
>>
>>
>> ______________________________________________
>> R-help@ 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.
>>
>
>
> -----
> Frank Harrell
> Department of Biostatistics, Vanderbilt University
> --
> View this message in context: http://r.789695.n4.nabble.com/kruskal-wallis-post-hoc-tp4288008p4288894.html
> Sent from the R help mailing list archive at Nabble.com.
>
> ______________________________________________
> [hidden email] 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.

--
Peter Dalgaard, Professor,
Center for Statistics, Copenhagen Business School
Solbjerg Plads 3, 2000 Frederiksberg, Denmark
Phone: (+45)38153501
Email: [hidden email]  Priv: [hidden email]

______________________________________________
[hidden email] 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.
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: kruskal wallis post hoc?

Frank Harrell
Peter,

The score test from the P.O. model for the global null hypothesis (k-1 degrees of freedom for comparing k groups) is almost exactly the Kruskal-Wallis test statistic.  For the case where k=2 (Wilcoxon test) the numerator of the score test is exactly the numerator of the Wilcoxon-Mann-Whitney U test.

Frank
Peter Dalgaard-2 wrote
On Jan 12, 2012, at 14:11 , Frank Harrell wrote:

> The Kruskal-Wallis test is a special case of the proportional odds ordinal
> logistic model.

Eh? Can you elaborate on that?

I would expect that at best it is equivalent to some _test_ in a polr-type model. It is never really clear what the model is when some groups are different and others not.

>  You can get any contrast you want by testing regression
> coefficients.   In a couple of weeks the rms package's contrast function
> will allow for individual confidence intervals of effects that together have
> a 0.05 type I error, by using the multcomp package (called automatically
> from contrast.rms).
> Frank
>
> Iasonas Lamprianou wrote
>>
>> Thank you for the result, I will have a look at the link.
>>
>>  
>> Dr. Iasonas Lamprianou
>> Department of Social and Political Sciences
>> University of Cyprus
>>
>>
>>> ________________________________
>>> From: Tal Galili <tal.galili@>
>>> To: Iasonas Lamprianou <lamprianou@>
>>> Cc: "r-help@" <r-help@>
>>> Sent: Thursday, 12 January 2012, 10:48
>>> Subject: Re: [R] kruskal wallis post hoc?
>>>
>>>
>>> Hi Iasonas ,
>>> This is a stat question and not an R question.
>>> But the general answer is that it could happen :)
>>>
>>>
>>> The R question would have been if their is a Tukey HSD for kruskel.test,
> the answer is yes:
>>> http://stats.stackexchange.com/questions/17342/is-there-a-nonparametric-equivalent-of-tukey-hsd 
>>> But if you didn't get any significant result from the pairwise comparison,
> I would say that the post hoc correction wouldn't help you (it could be that
> the reason for this significance is based on some weird contrast...)
>>>
>>>
>>> Best,
>>> Tal
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>> ----------------Contact
> Details:-------------------------------------------------------
>>> Contact me: Tal.Galili@ |  972-52-7275845
>>> Read me: www.talgalili.com (Hebrew) | www.biostatistics.co.il (Hebrew) |
> www.r-statistics.com (English)
>>> ----------------------------------------------------------------------------------------------
>>>
>>>
>>>
>>>
>>>
>>> On Thu, Jan 12, 2012 at 8:35 AM, Iasonas Lamprianou <lamprianou@>
> wrote:
>>>
>>> Dear all,
>>>>
>>>> I run a kruskal wallis test and found significant results. Then, I
> conducted all pairwise comparisons and found no significant results. Could
> anyone please give me a hint as to why this happens or redirect me towards a
> specific web page where I can find more info? (I used alpha=5% and made no
> bonferroni or other correction for the pairwise comparisons)
>>>> Thank you
>>>>
>>>>  
>>>> Dr. Iasonas Lamprianou
>>>> Department of Social and Political Sciences
>>>> University of Cyprus
>>>>        [[alternative HTML version deleted]]
>>>>
>>>>
>>>> ______________________________________________
>>>> R-help@ 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.
>>>>
>>>>
>>>
>>>
>>>
>> [[alternative HTML version deleted]]
>>
>>
>> ______________________________________________
>> R-help@ 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.
>>
>
>
> -----
> Frank Harrell
> Department of Biostatistics, Vanderbilt University
> --
> View this message in context: http://r.789695.n4.nabble.com/kruskal-wallis-post-hoc-tp4288008p4288894.html
> Sent from the R help mailing list archive at Nabble.com.
>
> ______________________________________________
> [hidden email] 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.

--
Peter Dalgaard, Professor,
Center for Statistics, Copenhagen Business School
Solbjerg Plads 3, 2000 Frederiksberg, Denmark
Phone: (+45)38153501
Email: [hidden email]  Priv: [hidden email]

______________________________________________
[hidden email] 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.
Frank Harrell
Department of Biostatistics, Vanderbilt University
Loading...