On Tue, Dec 11, 2012 at 3:45 PM, MIGUEL ANGEL GRUPO DE INVESTIGACIO
ODONT SIMANCAS PALLARES <msimancasp@unicartagena.edu.co> wrote:
>
> Hi Statalisters.
>
> I have some continous data to work with. Some of the observations are
> negative values. How to make appropiate statistics in Stata to deal
> with this negative values? I want to be sure that the statistics are
> correctly performed. I want to perform descriptive statistics and then
> non-parametric statistics such as t-test's and Kruskal-Wallis.
The t-test isn't a nonparametric test, but Kruskal-Wallis is. In neither
case would the presence of negative values matter. A rank-based procedure
like Kruskal-Wallis blows away the original scale of the data anyhow
because of the rank transformation. It will have issues in the presence of
ties.
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/faqs/resources/statalist-faq/
* http://www.ats.ucla.edu/stat/stata/
