Troubleshooting
Problem
How can I run a multivariate linear regression analysis (one with multiple dependent variables) in SPSS?
Resolving The Problem
You will need to have the SPSS Advanced Statistics module in order to run a linear regression with multiple dependent variables. The simplest way in the graphical interface is to click on Analyze->General Linear Model->Multivariate. Place the dependent variables in the Dependent Variables box and the predictors in the Covariate(s) box. To print the regression coefficients, you would click on the Options button, check the box for Parameter estimates, click Continue, then OK.
The output from this will include multivariate tests for each predictor, omnibus univariate tests, R^2, and Adjusted R^2 values for each dependent variable, as well as individual univariate tests for each predictor for each dependent. Checking the box for Parameter estimates in the Options dialog box produces the regression coefficients for each predictor for each dependent. In some cases people want a multivariate test for the entire regression. To do that, you would have to use syntax. Suppose you have predictors X1, X2, and X3, and dependents Y1 and Y2. The syntax to get the complete analysis at once, including the omnibus test for all predictors and dependents, would be:
GLM Y1 Y2 WITH X1 X2 X3
/PRINT PARAMETERS
/LMATRIX 'Multivariate test of entire model'
X1 1; X2 1; X3 1.
The string in quotes is an optional label for the output.
It is also possible to use the older MANOVA procedure to obtain a multivariate linear regression analysis. This requires using syntax. The basic form, which produces an omnibus test for the entire model, but no multivariate tests for each predictor, is:
MANOVA Y1 Y2 WITH X1 X2 X3.
In order to obtain the multivariate tests for each predictor, use:
MANOVA Y1 Y2 WITH X1 X2 X3
/PRINT PARAMETERS
/ANALYSIS X1 X2
/DESIGN CONSTANT, X1, X2, X3.
The output from this will include multivariate tests for each predictor, omnibus univariate tests, R^2, and Adjusted R^2 values for each dependent variable, as well as individual univariate tests for each predictor for each dependent. Checking the box for Parameter estimates in the Options dialog box produces the regression coefficients for each predictor for each dependent. In some cases people want a multivariate test for the entire regression. To do that, you would have to use syntax. Suppose you have predictors X1, X2, and X3, and dependents Y1 and Y2. The syntax to get the complete analysis at once, including the omnibus test for all predictors and dependents, would be:
GLM Y1 Y2 WITH X1 X2 X3
/PRINT PARAMETERS
/LMATRIX 'Multivariate test of entire model'
X1 1; X2 1; X3 1.
The string in quotes is an optional label for the output.
It is also possible to use the older MANOVA procedure to obtain a multivariate linear regression analysis. This requires using syntax. The basic form, which produces an omnibus test for the entire model, but no multivariate tests for each predictor, is:
MANOVA Y1 Y2 WITH X1 X2 X3.
In order to obtain the multivariate tests for each predictor, use:
MANOVA Y1 Y2 WITH X1 X2 X3
/PRINT PARAMETERS
/ANALYSIS X1 X2
/DESIGN CONSTANT, X1, X2, X3.
Related Information
[{"Product":{"code":"SSLVMB","label":"IBM SPSS Statistics"},"Business Unit":{"code":"BU059","label":"IBM Software w\/o TPS"},"Component":"Not Applicable","Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"Not Applicable","Edition":"","Line of Business":{"code":"LOB10","label":"Data and AI"}}]
Historical Number
20293
Was this topic helpful?
Document Information
Modified date:
01 December 2023
UID
swg21476743