|
|  |
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
Re: st: Adjusted odds ratio
On 12 Noll 2006, at 09:51, Maarten buis wrote:
Easiest way is to divide X by it's standard deviation, so a unit
change
in the new X variable is a standard deviation change. I find that
spending some extra time to make sure that a unit change in each
variable is meaningfull to me, is usually time well spent. In your
case
you apperently prefer a standard deviation, I ususally like the range
of the variable, for instance I am often comparing persons with the
lowest socioeconomic status with people with the highest socioeconomic
status, which I find easier to talk about then standard deviations.
Can I add another option here? – Quantiles.
I often deal with measures that are on arbitrary scales, such as
stigma, depression, worry and assumptive world. Even scales with the
same apparent numeric range have very different distributions. For
this reason, I often use anything from quartiles to deciles. This
makes it possible to compare regression coefficients (relative risks,
odds ratios etc) between different measures.
I don't like using the standard deviation because my readers, like
everyone else who passed through elementary statistics classes on the
way to something more interesting, don't understand it.
=========
Ronán Conroy
Royal College of Surgeons in Ireland
[email protected]
+353 (0) 1 402 2431
+353 (0) 87 799 97 95
http://www.flickr.com/photos/ronanconroy
*
* 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/
