27.03.2013 Views

SPSS® 12.0 Command Syntax Reference

SPSS® 12.0 Command Syntax Reference

SPSS® 12.0 Command Syntax Reference

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

560 FACTOR<br />

Matrix Input<br />

• FACTOR can read matrix materials written either by a previous FACTOR procedure or by<br />

a procedure that writes correlation or covariance matrices. For more information, see<br />

Universals on p. 3.<br />

• MATRIX=IN cannot be used unless a working data file has already been defined. To read an<br />

existing matrix data file at the beginning of a session, first use GET to retrieve the matrix<br />

file and then specify IN(COR=*), IN(COV=*) or IN(FAC=*) on MATRIX.<br />

• The VARIABLES subcommand cannot be used with matrix input.<br />

• For correlation and covariance matrix input, the ANALYSIS subcommand can specify a<br />

subset of the variables in the matrix. You cannot specify a subset of variables for factor<br />

loading matrix input. By default, the ANALYSIS subcommand uses all variables in the<br />

matrix.<br />

Format of the Matrix Data File<br />

Split Files<br />

• For correlation or covariance matrices, the matrix data file has two special variables created<br />

by the program: ROWTYPE_ and VARNAME_. Variable ROWTYPE_ is a short string variable<br />

with the value CORR (for Pearson correlation coefficient) or COV (for covariance) for<br />

each matrix row. Variable VARNAME_ is a short string variable whose values are the names<br />

of the variables used to form the correlation matrix.<br />

• For factor loading matrices, the program generates two special variables named<br />

ROWTYPE_ and FACTOR_. The value for ROWTYPE_ is always FACTOR. The values for<br />

FACTOR_ are the ordinal numbers of the factors.<br />

• For factor score coefficient matrices, the matrix data file has two special variables created:<br />

ROWTYPE_ and VARNAME_. If split-file processing is in effect, the split variables appear<br />

first in the matrix output file, followed by ROWTYPE_, VARNAME_, and the variables in the<br />

analysis. ROWTYPE_ is a short string with three possible values: MEAN, STDDEV, and<br />

FSCOEF. There is always one occurrence of the value MEAN. If /METHOD = CORRELA-<br />

TION then there is one occurrence of the value STDDEV. Otherwise, this value does not<br />

appear. There are as many occurrences of FSCOEF as the number of extracted factors.<br />

VARNAME_ is a short string who values are FACn where n is the sequence of the saved<br />

factor when ROWTYPE_ equals FSCOEF. Otherwise the value is empty.<br />

• The remaining variables are the variables used to form the matrix.<br />

• FACTOR can read or write split-file matrices.<br />

• When split-file processing is in effect, the first variables in the matrix data file are the split<br />

variables, followed by ROWTYPE_, VARNAME_ (or FACTOR_), and then the variables used<br />

to form the matrix.<br />

• A full set of matrix materials is written for each split-file group defined by the split variables.<br />

• A split variable cannot have the same variable name as any other variable written to the<br />

matrix data file.

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!