Factor analysis on dichotomously scored items

#1
Hi
have a set of 30 dichotomously scored items (scored 0 and 1) and want to do exploratory factor analysis on them. I have heard that performing EFA on such items is problematic and the EFA should be carried out on the tetrachoric correlation matrix. How can do this? Does SPSS automatically take care of this? Please advise.

Thanks
Anthony
 

spunky

King of all Drama
#2
I have heard that performing EFA on such items is problematic and the EFA should be carried out on the tetrachoric correlation matrix.
That is correct. When one is conducting Factor Analysis (whether Exploratory or Confirmatory) it is important to keep in mind the scaling of your variables. Like, in your case, binary.


How can do this? Does SPSS automatically take care of this? Please advise.
SPSS will not do this for you and it cannot calculate a tetrachoric/polychoric correlation matrix as it stands. You'll either need to use an external SPSS macro or use a more powerful software like R.
 

noetsi

Fortran must die
#3
Actually SPSS has a link to R that will generate a polychoric correlation. Since I use SAS rather than SPSS I don't know the details.

One approach to adapting factor analysis for ordinal variables is to use polychoric correlations, rather than the Pearson correlations that are used by SPSS Factor. SPSS does not have a built-in procedure for computing polychoric correlations, but there is an extension command (SPSSINC HETCOR) to print polychoric and polysrial correlations available in the SPSS Community for SPSS Statistics versions from 17.0 upwards. (Click the "Downloads for IBM SPSS Statistics" link, then the "Extension Commands" link under "Tools and Utilities". Look for the downloadable file SPSSINC_HETCOR.zip in the list of extension commands there. This command calculates polyserial, polychoric, and Pearson correlations between variables with the type determined by the variable measurement levels. The package includes a dialog box interface for the procedure. Beginning with Version 19, this file is installed with R Essentials.. The Programmability Plug-ins and Essentials for SPSS Statistics versions 19 and above are available on the "Downloads for IBM SPSS Statistics" page. There may also be SPSS macros available on the internet to do this. See Technote 1479694 for instructions to read a correlation matrix in text format into SPSS and then analyze that matrix with the Factor procedure
http://www-01.ibm.com/support/docview.wss?uid=swg21477550

When I have run polychoric versus spearman's (with likert variables that have five levels) I have not found huge differences. But that could be luck.