Richard Palmer-Jones <[email protected]>:
I think any suggestions would have to be conditional on what the
scores are to be used for. I would be inclined to condition only on
sex, since differences by race/ethnicity may be conflated with
differences by income. Also, are these age or cohort effects? Or
both?
webuse nhanes2
lpoly hei age if race==1 & sex==1
tw kdensity hei if race==1&sex==1&ageg==2||kdensity hei if
race==1&sex==1&ageg==5
Also note that you might want to use rank (on the interval 0,1)
directly, rather than a Z score, depending on your model...
On Mon, Apr 21, 2008 at 11:20 AM, Richard Palmer-Jones
<[email protected]> wrote:
> Dear Readers
>
> I want to calculate z-scores of adult male and female heights of
> different ages using reference population standards (e.g. healthy
> western populations). This is readily done for children, but I am not
> clear what distribution parameters to use for adults. I can see some
> debate about changing adult heights in western populations, and also
> differences by ethnicity and income, so population parameters (such as
> from NHANES) would not seem suitable.
>
> All suggestions welcome.
*
* 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/
