Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: calculate alpha after polychoric factor analysis


From   <Seyi.Soremekun@lshtm.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   Re: st: calculate alpha after polychoric factor analysis
Date   Mon, 13 Feb 2012 15:18:18 +0000

Thanks (and for the polychoric command itself). The sample size is not an issue - all respondents answered all the questions, displaying -n- was just a check for me  (if that is what you meant in your note?)
 
I'm not sure about the appropriateness of using alpha either, however, I know that ordinal alphas are available in other types of software (Basto and Pereira 2012 J. Statistical Software 46(4)), and wondered if there was something I could use in Stata. 
I was hoping that I could ask stata to calculate the alpha on the saved matrix rather than a list of variables.
Seyi

>>> Stas Kolenikov <skolenik@gmail.com> 13/02/2012 14:34 >>>
I am not sure -alpha- is entirely appropriate in this situation. Tenko
Raykov published a number of papers in late 1990s conceptualizing
composite reliability extending the alpha to more general situations
of factor analysis. I don't know if any of that is implemented as a
third party package. Obviously, this is doable as -nlcom- following
-sem- or -confa-, although you would need to spend some time figuring
out how to best use either the estimated coefficients or the resulting
matrices.

Note also that -polychoric- computes correlations in pairwise fashion,
so there may be no single sample size that is suitable for factor
analysis (although this is probably just a scaling coefficient
somewhere, as the eigenvalues and eigenvectors produced in factor
analysis do not depend on -n-).

On Mon, Feb 13, 2012 at 8:54 AM,  <Seyi.Soremekun@lshtm.ac.uk> wrote:
> Dear Stata Listers,
>
> I would like to estimate the equivalent of  Cronbach's alpha after factor analysis conducted on a polychoric matrix of ordinal, Likert items. I started out in a pretty standard way thus:
>
> polychoric var1 - var29
> display r(sum_w) // check number of respondents answered all = 331
> matrix r = r(R)
> factormat r, n(331) factors(3) ipf
> alpha?
>
> Is there such functionality available, or might I have to re-do the whole analysis in R or SAS?
>
> thanks,
>
> Seyi
>
>
> Dr Seyi Soremekun
> Faculty of Epidemiology and Public Health
> London School of Hygiene and Tropical Medicine
> London WC1E 7HT
> +442079272464
>
>
> *
> *   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/ 



-- 
Stas Kolenikov, also found at http://stas.kolenikov.name 
Small print: I use this email account for mailing lists only.

*
*   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/


*
*   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/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index