Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: About Rank Tests with more than 2 ranked lists.


From   Roger Newson <roger.newson@kcl.ac.uk>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: About Rank Tests with more than 2 ranked lists.
Date   Mon, 10 Oct 2005 13:52:02 +0100

As I understand it (correct me if I'm wrong), you want to test the hypothesis that none of the 4 efficiency measures has a tendency to be higher than any of the rest. I do not know why you should want to do this. However, if that is what you want to do, then I would proceed as follows.

First, reshape the data (probably using -reshape-) to have one observation per individual per score (and therefore 4 observations per individual). The dataset created might have variables -id- identifying the individual assessed, -scoreseq- with values equal to 1, 2, 3 or 4 denoting which of the 4 scores the observation represents, and -score- containing the value of that score for that individual. Then, use -tabulate- to generate 4 new dummy variables indicating the 4 possible values of -scoreseq-. Then, calculate Somers' D of these 4 dummy variables with respect to -score- (clustering by -id-), and use -test- to test the hypothesis that all 4 of these Somers' D parameters are zero (as will be the case if none of the 4 scores tends to be greater than, or less than, the other 3). After reshaping the data, the code might be as follows:

tabulate scoreseq, gene(scoreseq_)
somersd score scoreseq_*, tr(z) cluster(id)
test scoreseq_1 scoreseq_2 scoreseq_3 scoreseq_4

This should perform a chi-squared test for the hypothesis that all 4 Somers' D parameters are zero.

I hope this helps.

Roger


At 13:06 10/10/2005, Alvaro wrote:

Thank you very much for your response, Roger. I'm studying these pdf files that you recommended.

My intention is to compare 4 distributions to each other. My concern, then, is on how to do it with only one application of a test, if possible. If not, which tests would be available?

best,
alvaro

--On 10 October 2005 12:27 +0100 Roger Newson <roger.newson@kcl.ac.uk> wrote:


At 11:07 10/10/2005, Alvaro wrote:
Dear colleagues,
I have four lists of efficiency scores, which allow to order several
individuals according to their performance in different sectors.
I need to test if these rank-distributions are similar.

I wonder, is there any test that would help me to do this in just one
test ?
Alvaro has not specified which rank distributions are to be compared with
which other rank distributions, or what question exactly is being asked.
However, the -somersd- package, downloadable from SSC, produces
confidence limits for a wide range of rank statistics, and probably can
do whatever Alvaro actually wants to do. It comes with a .pdf manual
which explains the principles behind these statistics.

I hope this helps.

Roger


--
Roger Newson
Lecturer in Medical Statistics
Department of Public Health Sciences
Division of Asthma, Allergy and Lung Biology
King's College London

5th Floor, Capital House
42 Weston Street
London SE1 3QD
United Kingdom

Tel: 020 7848 6648 International +44 20 7848 6648
Fax: 020 7848 6620 International +44 20 7848 6620
   or 020 7848 6605 International +44 20 7848 6605
Email: roger.newson@kcl.ac.uk
Website: http://phs.kcl.ac.uk/rogernewson/

Opinions expressed are those of the author, not the institution.

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/

--
Roger Newson
Lecturer in Medical Statistics
Department of Public Health Sciences
Division of Asthma, Allergy and Lung Biology
King's College London

5th Floor, Capital House
42 Weston Street
London SE1 3QD
United Kingdom

Tel: 020 7848 6648 International +44 20 7848 6648
Fax: 020 7848 6620 International +44 20 7848 6620
  or 020 7848 6605 International +44 20 7848 6605
Email: roger.newson@kcl.ac.uk
Website: http://phs.kcl.ac.uk/rogernewson/

Opinions expressed are those of the author, not the institution.

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index