Re: st: Adjusted Prevalence from Logistic models

[Richard Williams <richardwilliams.ndu@gmail.com>]

At 01:18 AM 10/24/2012, Dean Eurich wrote:
> Hi all,
>
> A simple question. If I run a logistic model:
>
> logistic BB age sex rural
>
> I get the OR for the effect of rural (which is a flag for rural=0 or 
> urban=1 dwelling) on the use of beta blockers (medication used to 
> treat cardiovascular disease).
>
> A reviewer has asked for us to present the adjusted prevalence for 
> the use of BB by rural and urban status. I am not to sure how to do this.
>
> I assume I could use the 'margin' command.
>
> So would it be as simple as
>
>  margins rural, atmeans
>
> (Although I realize I could do it as specific levels of sex and age as well)
>
> Thanks
>
> Dean

I'm not totally sure how you want to adjust. But if you want to use 
-margins- you should use factor variable notation, e.g.

logistic BB age i.sex i.rural
margins rural, atmeans

In general, I am not crazy about the atmeans option, and I especially 
don't like it when you have a variable like sex in the model. You'll 
be plugging in something like .5 for sex (the average person is half 
male, half female?) I'd rather use the default -asobserved- or plug 
in specific values or ranges of values for the other variables. For 
some highlights of what you can do with margins, see

http://www.nd.edu/~rwilliam/stats/Margins01.pdf

As a sidelight, when you say rural=0 or urban=1, that sort of sounds 
to me like heads I win, tails you lose. Aren't rural=0 and urban=1 
the same thing?

-------------------------------------------
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
HOME:   (574)289-5227
EMAIL:  Richard.A.Williams.5@ND.Edu
WWW:    http://www.nd.edu/~rwilliam

*
*   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/