Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Nick Cox <njcoxstata@gmail.com> |
To | statalist@hsphsun2.harvard.edu |
Subject | Re: st: Constructing socio-economic status scale using Principal Components Analysis |
Date | Wed, 28 Nov 2012 08:55:29 +0000 |
If factor analysis (or PCA) is apppropriate then the first factor (PC) _is_ the best single measure you can get. (Why call -0.005 very negative? It's nearly zero.) Nick On Wed, Nov 28, 2012 at 2:59 AM, Ameya Bondre <ameyabondre.jhsph@gmail.com> wrote: > I have a data-set with about 37 variables that can assess household > socio-economic status in a sample of about 6000 households. These > include variables measuring household wealth, access to water and > sanitation, rural households owning animals, etc. > > I used factor analysis (factor var1, var2, ...., pcf) and it gave me > 10 factors with eigen values more than 1. But the percentage of > variation in the total data was best explained by the first factor - > "factor1" had the highest eigen value and explained 18% of the > variation. > > Using the command "predict factor1, bar", I got the factor scores for > each of these 37 variables, for factor1. > > From the articles I read about principal components analysis, I got to > know that negative factor scores mean a lower SES and positive scores > mean a higher SES. Some scores are strongly defining the factor (0.30, > 0.17), some are very negative (-0.005) and most other scores are too > low and positive or negative but close to 0. > > The above steps also create a new variable called "factor1", which > gives "a score" to each data point (or each household). The scores for > the data points range from -2 to 1.8. > > I want to construct a socio-economic status scale using factor1. Is > that possible? * * 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/