The FACTOR Procedure
-
Overview
- Getting Started
-
Syntax
-
Details
Input Data Set Output Data Sets Confidence Intervals and the Salience of Factor Loadings Simplicity Functions for Rotations Missing Values Cautions Factor Scores Variable Weights and Variance Explained Heywood Cases and Other Anomalies about Communality Estimates Time Requirements Displayed Output ODS Table Names ODS Graphics
-
Examples
- References
| Output Data Sets |
The OUT= Data Set
The OUT= data set contains all the data in the DATA= data set plus new variables called Factor1, Factor2, and so on, containing estimated factor scores. Each estimated factor score is computed as a linear combination of the standardized values of the variables that are factored. The coefficients are always displayed if the OUT= option is specified, and they are labeled "Standardized Scoring Coefficients."
If partial variables are specified in the PARTIAL statement, the factor analysis is on the residuals of the variables, which are regressed on the partial variables. In this case, the OUT= data set also contains the (unstandardized) residuals, which are prefixed by R_ by default. For example, the residual of variable X is named R_X in the OUT= data set. You might also assign the prefix by the PARPREFIX= option. Because the residuals are factor-analyzed, the estimated factor scores are computed as linear combinations of the standardized values of the residuals, but not the original variables.
The OUTSTAT= Data Set
The OUTSTAT= data set is similar to the TYPE=CORR or TYPE=UCORR data set produced by the CORR procedure, but it is a TYPE=FACTOR data set and it contains many results in addition to those produced by PROC CORR. The OUTSTAT= data set contains observations with _TYPE_=’UCORR’ and _TYPE_=’USTD’ if you specify the NOINT option.
The output data set contains the following variables:
the BY variables, if any
two new character variables, _TYPE_ and _NAME_
the variables analyzed—those in the VAR statement, or, if there is no VAR statement, all numeric variables not listed in any other statement. If partial variables are specified in the PARTIAL statement, the residuals are included instead. By default, the residual variable names are prefixed by R_, unless you specify something different in the PARPREFIX= option.
Each observation in the output data set contains some type of statistic as indicated by the _TYPE_ variable. The _NAME_ variable is blank except where otherwise indicated. The values of the _TYPE_ variable are as follows:
- MEAN
means
- STD
standard deviations
- USTD
uncorrected standard deviations
- N
sample size
- CORR
correlations. The _NAME_ variable contains the name of the variable corresponding to each row of the correlation matrix.
- UCORR
uncorrected correlations. The _NAME_ variable contains the name of the variable corresponding to each row of the uncorrected correlation matrix.
- IMAGE
image coefficients. The _NAME_ variable contains the name of the variable corresponding to each row of the image coefficient matrix.
- IMAGECOV
image covariance matrix. The _NAME_ variable contains the name of the variable corresponding to each row of the image covariance matrix.
- COMMUNAL
final communality estimates
- PRIORS
prior communality estimates, or estimates from the last iteration for iterative methods
- WEIGHT
variable weights
- SUMWGT
sum of the variable weights
- EIGENVAL
eigenvalues
- UNROTATE
unrotated factor pattern. The _NAME_ variable contains the name of the factor.
- SE_UNROT
standard error estimates for the unrotated loadings. The _NAME_ variable contains the name of the factor.
- RESIDUAL
residual correlations. The _NAME_ variable contains the name of the variable corresponding to each row of the residual correlation matrix.
- PRETRANS
transformation matrix from prerotation. The _NAME_ variable contains the name of the factor.
- PREFCORR
prerotated interfactor correlations. The _NAME_ variable contains the name of the factor.
- SE_PREFC
standard error estimates for prerotated interfactor correlations. The _NAME_ variable contains the name of the factor.
- PREROTAT
prerotated factor pattern. The _NAME_ variable contains the name of the factor.
- SE_PREPA
standard error estimates for the prerotated loadings. The _NAME_ variable contains the name of the factor.
- PRERCORR
prerotated reference axis correlations. The _NAME_ variable contains the name of the factor.
- PREREFER
prerotated reference structure. The _NAME_ variable contains the name of the factor.
- PRESTRUC
prerotated factor structure. The _NAME_ variable contains the name of the factor.
- SE_PREST
standard error estimates for prerotated structure loadings. The _NAME_ variable contains the name of the factor.
- PRESCORE
prerotated scoring coefficients. The _NAME_ variable contains the name of the factor.
- TRANSFOR
transformation matrix from rotation. The _NAME_ variable contains the name of the factor.
- FCORR
interfactor correlations. The _NAME_ variable contains the name of the factor.
- SE_FCORR
standard error estimates for interfactor correlations. The _NAME_ variable contains the name of the factor.
- PATTERN
factor pattern. The _NAME_ variable contains the name of the factor.
- SE_PAT
standard error estimates for the rotated loadings. The _NAME_ variable contains the name of the factor.
- RCORR
reference axis correlations. The _NAME_ variable contains the name of the factor.
- REFERENC
reference structure. The _NAME_ variable contains the name of the factor.
- STRUCTUR
factor structure. The _NAME_ variable contains the name of the factor.
- SE_STRUC
standard error estimates for structure loadings. The _NAME_ variable contains the name of the factor.
- SCORE
scoring coefficients to be applied to standardized variables if the SCORE option is specified on the PROC FACTOR statement. The _NAME_ variable contains the name of the factor.
- USCORE
scoring coefficients to be applied without subtracting the mean from the raw variables if the SCORE option is specified on the PROC FACTOR statement. The _NAME_ variable contains the name of the factor.