st: Significance testing, grouped ordinal data

[Yerik Kaslow <yerik.kaslow@gmail.com>]

Hi Statalisters,

I am working with ordinal data (scale of 1-10, 1-5, etc) from patient
reported outcomes for health status.

Distribution is sometimes normal sometimes not normal.

I am trying to test for significant differences between 2 groups from
the same population:

Group1 are those who received the intervention therapy
Group2 are those in the control group

The data is clinical trial data where there are multiple responses for
the same patient. One patient might report their health status 4
times, another 21 times, another 1 time.

Thus far I have used ttests (assuming equal variance), as well as
ranksum tests for data that do not have a normal distribution.

I have thus far found some differences, but nothing that is
statistically significant and persistent. I am wondering if there are
other tests that I could be running that I am not running right now. I
want to be sure that their data are producing this result, as opposed
to me not running the right tests.

I would *really* appreciate some insight. I apologize if this sounds
like a very basic question. I have done reading online and want to be
sure there isn't something I am forgetting to do. I am not finding a
lot of advice online regarding how to deal with ordinal data. The data
are all survey responses, patients reporting their health status
subjectively.

I would very much appreciate insight.

Thank you,

Yerik Kaslow



--
Yerik Kaslow
yerik.kaslow@gmail.com
*
*   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/