Date:         Mon, 9 Jun 1997 13:45:17 EDT
Reply-To:     brenda shapiro <SHAPIRO@UCONNVM.UCONN.EDU>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@UGA.CC.UGA.EDU>
From:         brenda shapiro <SHAPIRO@UCONNVM.UCONN.EDU>
Subject:      standard error of estimate
Content-Type: TEXT/PLAIN; charset=US-ASCII

I have a question about data analysis.  I am analyzing data on two groups of
subjects (autistics and language disordered), and originally conducted
partial correlations between variables of memory and social functioning,
controlling for IQ.  I have been told that I also need to take the range
of the variables into account, and should do some kind of analysis such
as a standard error of estimate.  I have not been able to find a book
that talks about this issue.  (the person who told me this said it is
because restricted range can affect correlations).  ANy ideas on what
kind of analysis to conduct and how to do it in SPSS?I am mainly interested
in looking at the relationship between memory and social within groups and
then comparing this relationship between groups (different pattern?).