Re: st: Standardized coefficients using nestreg: svy

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

At 06:17 PM 11/6/2011, Patricia Logan-Greene wrote:
> Hello All,
>
> We are hoping to get standardized betas when using nested 
> regressions and the svy commands. The usual command (,beta) doesn't 
> work with the svy commands. Adding listcoef after the nestreg 
> command, unfortunately, only provides the betas for the last step. 
> One can repeat each step of the nested regression separately, adding 
> listcoef after each step, however the sample n in earlier steps does 
> not always match the n included in the nested regression, as cases 
> with missing values on later steps are excluded from all steps in nestreg.
>
> Is there another way to get the betas? I suppose one could limit the 
> sample to those with full data on all variables, OR could 
> standardize all the variables before running the regression. Would 
> love to hear if there's another option.

It isn't that hard to use the same sample throughout. Do something like

gen touse = !missing(y, x1, x2, x3, x4, x5, x6)
svy, subpop(touse): reg y x1 x2
listcoef
svy, subpop(touse): reg y x1 x2 x3 x4
listcoef

Whether you should do this is another matter. The fact that the 
-beta- option doesn't work is one warning sign. I could swear there 
were threads explaining why standardized coefficients with svy: were 
a bad idea, but I can't find them, so maybe I am just imagining things.


-------------------------------------------
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/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/