Usage Note 22555: How are the estimated factor scores in the OUT= data set computed? How can I compute them myself or on a second data set?
 |  |  |  |
For every observation used in the analysis, PROC FACTOR will multiply the standardized data vector by the standardized scoring coefficients. If you want to score a new data set based on the original factors, then just use PROC SCORE DATA= the new data set, SCORE= the OUTSTAT data set from PROC FACTOR (be sure to specify the options SCORE OUTSTAT= when you run PROC FACTOR), type='SCORE', and OUT= a data set to contain the estimated factor scores. Use a VAR statement if appropriate.
*
For software releases that are not yet generally available, the Fixed
Release is the software release in which the problem is planned to be
fixed.
An explanation of how the standardized scoring coefficients are computed is given in Usage Note 22554.
If you want to compute the factor scores manually, first standardize the data using the same mean and/or standard deviation used by PROC FACTOR when the scoring coefficients were extracted (this information is in the OUTSTAT= data set above). Then multiply the standardized data by the standardized scoring coefficients. For an example of the calculations you need to perform, see SAS/STAT User's Guide, The SCORE Procedure. The Getting Started section has an example that shows the proper calculations in detail.
Operating System and Release Information
| Product Family | Product | System | SAS Release | |
| Reported | Fixed* | |||
| SAS System | SAS/STAT | All | n/a | |
| Type: | Usage Note |
| Priority: | low |
| Topic: | Analytics ==> Spatial Analysis SAS Reference ==> Procedures ==> FACTOR Analytics ==> Multivariate Analysis Analytics ==> Psychometrics |
| Date Modified: | 2008-01-25 08:06:54 |
| Date Created: | 2002-12-16 10:56:39 |