Re: st: Adjusting confidence intervals

[Ronán Conroy <rconroy@rcsi.ie>]

Ilesh Patel wrote:

> Hello.
>
> I have computed a difficulty score for a range of vision tasks and 
> want to see if there is a difference in the mean score by self-rated 
> overall health (scored from 1 – 4). Following is what I used:
>
> Sort overall_hlth
>
> by overall_hlth: ci diff_score
>
> This gives me the mean difficulty score and it’s CI for each of the 4 
> self-rated overall health groups. What command do I use in Stata to 
> adjust the CI for age given that I think age affects both the 
> difficulty score.
. xi:regress diff_score i.overall_hlth age

(this builds a model with overall health represented by dummy variables 
for all but the first level, and adjusts for age)

. adjust age, by(overall_hlth) xb ci label(Age-adjusted mean) 
cilabel(95% CI)

This will produce age adjusted mean scores with confidence intervals, 
evaluated at the overall mean age of the sample. I actually prefer 
taking a round figure for age that is somewhere towards the centre of 
the data, and doing

. adjust age=40, by(overall_hlth) xb ci label(Predicted mean at age 40) 
cilabel(95% CI)

--

Ronan M Conroy (rconroy@rcsi.ie) 
Senior Lecturer in Biostatistics 
Royal College of Surgeons 
Dublin 2, Ireland 
+353 1 402 2431 (fax 2764) 
-------------------- 
Just say no to drug reps 
http://www.nofreelunch.org/

*
*   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/