On Wed, Jan 14, 2009 at 9:51 AM, Garry Anderson
<[email protected]> wrote:
> Hi Claudia,
> You only need to consider the exact p-value.
> Your conclusion that the two distributions are not equal to each other is consistent with the p-value.
The two-sided test tells you if there is a difference, and if your
threshold is the (arbitrary) p = 0.05 for declaring statistical
significance then you can say based on the Combined statistic that
there is a significant difference between male and females.
But with this output you can go further and look at a one-sided test,
and the results show you the direction of the difference along with
the associated p-value, in this instance the distribution in males is
lower than that in females.
Neil
--
"We should make things as simple as possible, but not simpler." -
Anon (not Albert Einstein)
Email - [email protected]
Website - http://slack.ser.man.ac.uk/
Photos - http://www.flickr.com/photos/slackline/
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
