Date: Fri, 1 Sep 2006 11:11:16 +0100
Reply-To: Paul Mcgeoghan <mcgeoghan@Cardiff.ac.uk>
Sender: "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From: Paul Mcgeoghan <mcgeoghan@Cardiff.ac.uk>
Subject: repeated measures with varying covariates
Content-Type: text/plain; charset=US-ASCII
I have the following:
Dependent variable is TotalScore and this is recorded for 74 companies on 2
different websites and in 2 successive years so SITE and YEAR are 2 within
subject factors with 2 levels.
I have 3 between subject factors (each with 2 levels).
I have 3 constant covariates (LOGSIZE, GASHARE and FORSHARE) whose values do
not change over the 4 within subject levels.
I have 4 other covariates (LEVERAGE, RTA, RETURNEQUITY and FIXEDASSETS) whose
values are the same for the 2 websites but change for the 2 years.
I have syntax as follows:
TOTALSCORE04 TOTALSCORE05 TOTALSCORE04T TOTALSCORE05t BY BIG4 INDUSTRY
LOCATION WITH FIXEDASSETS04 TO FIXEDASSETS05t LEVERAGE04 TO LEVERAGE05t
RTA04 TO RTA05t RETURNEQUITY04 TO RETURNEQUITY05t
GASHARE FORSHARE LOGSIZE
/WSFACTOR = SITE 2 Polynomial TIME 2 Polynomial
/METHOD = SSTYPE(3)
/CRITERIA = ALPHA(.05)
/WSDESIGN = SITE TIME SITE*TIME.
The problem I have is in specifying the covarying covariates. An added problem
is that the covarying covariates do not vary over the 2 sites (only over the 2
years). Is the above the correct syntax?
Would it be better to use Linear Mixed Models Repeated Measures in this case?
Application support specialist (Statistics and Databases),
University Infrastructure Group (UIG),
Tel. 02920 (875035).