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 on April 23, and its replacement, statalist.org is already up and running.


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

Re: st: Generate one score for 2+ factors?


From   Nick Cox <njcoxstata@gmail.com>
To   "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu>
Subject   Re: st: Generate one score for 2+ factors?
Date   Wed, 26 Jun 2013 18:52:09 +0100

See http://www.stata.com/statalist/archive/2013-06/msg00415.html for a
previous reminder addressed to you directly to use full real names on
Statalist.

Predicted factor scores can be saved  using -predict-. Looking at the
help for -factor postestimation- shows that.

Then they are just new variables that you can combine as you wish.
Whether that's useful is another matter. For example, an idea I often
encounter is that if Factor1 and Factor 2 are good, then some
combination of the two will be even better. In terms of "proportion
explained", that is wrong. If you want a single construct to use you
need really good arguments not to use Factor 1. But plotting factor 1
and factor 2 against each other might give idea.

I can't remember if I ever knew what Thomson's method was, but I
suspect it was Thomson, not Thompson.

Nick
njcoxstata@gmail.com


On 26 June 2013 18:36, nola l <nola0908@gmail.com> wrote:
> I have a question on factor analysis. Please pardon me if it sounds
> too silly. I am new to factor analysis.
>
> When we run the factor analysis, normally we need to look at
> eigenvalues and see how many factors we need to keep.  Here is my
> results which suggested keep 2 factors.  Is there a program that can
> sort of combine two factor into one score? I read from literature that
> "the factors must first be generated using the Thompson's regression
> method and then each must be multiplied with its own proportion of
> variance explained" . Can we do that in Stata?
>
>  k  |  Eigenvalues  |  Proportion explained  |  Cum. explained
> ----+---------------+------------------------+------------------
>   1 |    1.763798   |    0.352760            |   0.352760
>   2 |    1.312006   |    0.262401            |   0.615161
>   3 |    0.941635   |    0.188327            |   0.803488
>   4 |    0.592744   |    0.118549            |   0.922037
>   5 |    0.389817   |    0.077963            |   1.000000
>
*
*   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/


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