On Mon, 31 Mar 2003 07:58:12 +0100 ann e fitzmaurice
<[email protected]> wrote:
> hi
> I am using spss and stata
> I have experienced a problem with weighted analysis - the weights have been
> set in spss and stata
> I run either a crosstab in spss and include the line in the syntax (spss)
> count=asis ,or svytab in stata, this will produce identical tables and
> identical column percentages
>
> The problem that I have is that the test statistic are different, I can
> obtain a Pearson chi-square in spss, and also a p value based on the log
> likelihood, but when I ask stata to produce the same I get a slightly
> different value and different p value
>
> Can anyone explain what is happening and if I should be using the log
> likelihood values instead of Pearson in spss
A colleague has reported a similar problem to me, based a different
SPSS, Stata comparison, but also using weighted data. Sorry, I can't
remember the details, but I recall that the core of the problem was
the "sample size" that was used to calculate the test statistics. From
memory, SPSS was doing things incorrectly; it was using the weighted
sample size, rather than actual number of obs. (Stata got things
right.)
Stephen
----------------------
Professor Stephen P. Jenkins <[email protected]>
Institute for Social and Economic Research (ISER)
University of Essex, Colchester, CO4 3SQ, UK
Tel: +44 (0)1206 873374. Fax: +44 (0)1206 873151.
http://www.iser.essex.ac.uk
*
* 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/
