IBM Support

Will Amos read an SPSS correlation/covariance matrix file?

Troubleshooting


Problem

I have a covariance data matrix to analyze using Amos. Will Amos automatically distinguish an SPSS Statistics raw data file from a covariance or correlation matrix file? If not, how do I specify the type of data file being read?

Resolving The Problem

Provided that you have saved your matrix as an SPSS Statistics matrix file, Amos will read it as such. That is, of two files "raw file.sav" and "covfile.sav", SPSS Statistics and Amos will expect the former to be laid out as a flat file of rows and columns where the rows are the individual units to be analyzed (cases) and the columns are the observations on those units (variables). The latter file is expected to be a square matrix of covariances or correlations along with their associated vectors of standard deviations, Ns, means, etc. For further information on matrix file formats in SPSS, please see the SPSS Command Syntax Reference Guide in the section on the MATRIX DATA command syntax.

SET PRINTBACK = NONE.
* Open the 'Hamilton.sav' data in the '..\Amos\Tutorial' directory *.
GET FILE='C:\Program Files\IBM\SPSS\Amos\Tutorial\Hamilton.sav'.
TITLE 'This is a regular SPSS raw data file (*.sav)'.
LIST.
CORRELATIONS
/VARIABLES=sat income educatn
/PRINT=TWOTAIL NOSIG
/MISSING=LISTWISE
/MATRIX = OUT(*).
MCONVERT.
TITLE 'This is a covariance matrix file created from the above data (*.sav)'.
LIST.

Save this file and use it as input for Amos.

Dr. James Arbuckle has provided two further examples on his web site:
http://amosdevelopment.com/support/faq/enter_sample_correlations.htm
http://amosdevelopment.com/support/faq/enter_sample_covariances.htm

[{"Product":{"code":"SSLVC7","label":"SPSS Amos"},"Business Unit":{"code":"BU001","label":"Analytics Private Cloud"},"Component":"Not Applicable","Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"Not Applicable","Edition":""}]

Historical Number

19374

Document Information

Modified date:
16 June 2018

UID

swg21476687