FAQ/wmixed - CBU statistics Wiki
Self: FAQ/wmixed

Reproduction of SPSS Resolution 22273 which may also be accessed by clicking on Knowledgebase and entering 22273 in the Number of the resolution you want to review box after clicking here.

Resolution number: 22273 Created on: Dec 17 2001 Last Reviewed on: Jul 8 2008

  • = Problem Subject: Repeated measures ANCOVA with a constant covariate in MIXED in SPSS =

Problem Description: How do you run a repeated measures Analysis of Covariance (ANCOVA) with a constant covariate in SPSS MIXED?

Resolution Subject: Running this analysis in MIXED is very straightforward.

Resolution Description: Running this analysis in MIXED is very straightforward. Because time is the most common dimension in which repeated measures are taken, repeated measures ANCOVA with a constant covariate is sometimes known as repeated measures with a non-time-varying covariate. This solution will provide the command syntax for estimating this model in SPSS MIXED using sample data from Winer (1971), p.803.

Here is command syntax for reading in the sample data from Winer:

DATA LIST FREE / person a b x y . 
BEGIN DATA 
1 1 1 3 10 
1 1 2 3 8 
2 1 1 5 15 
2 1 2 5 12 
3 1 1 8 20 
3 1 2 8 14 
4 1 1 2 12 
4 1 2 2 6 
5 2 1 1 15 
5 2 2 1 10 
6 2 1 8 25 
6 2 2 8 20 
7 2 1 10 20 
7 2 2 10 15 
8 2 1 2 15 
8 2 2 2 10 
END DATA. 

*Here is the command syntax for running the ANCOVA model. 

MIXED Y BY a b WITH x 
/FIXED = a b a*b x 
/REPEATED = b | SUBJECT(person) COVTYPE(CS). 

References:

Winer, B. J. (1971). Statistical principles in experimental design (2nd Ed.). New York: McGraw-Hill.

Winer, B. J., Brown, D. R. & Michels, K. M. (1991). Statistical principles in experimental design (3rd Ed.). New York: McGraw-Hill.

None: FAQ/wmixed (last edited 2013-03-08 10:17:32 by localhost)