Date:         Thu, 30 May 2002 13:57:15 -0600
Reply-To:     William Dudley <william.dudley@nurs.utah.edu>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:         William Dudley <william.dudley@nurs.utah.edu>
Subject:      estimates for testing mediation
Content-Type: text/plain; charset=US-ASCII

I am using SPSS to test some mediation models and want to be clear on some terms.
Specifically, I am wondering if I am using the correct estimates from my SPSS analyses.

MacKinnon in his informative web site http://www.public.asu.edu/~davidpm/ripl/q&a.htm#q2 provides a test of the indirect (mediated effect) that includes terms  referred to as sea and seb (which I interpret as the standard error of regression coeffiencts b and a.  I assume that these are estimates listed as the std. error that are given in the coefficients table  in the "unstandardized Coefficients" section.

However in other places (MacKinnon and Dwyer, 1993), these terms seem to be reffered to as varainces (the authors use sigma squared).

whadyathink?
Thanks
Bill


***********************************************
*   William N. Dudley, PhD
*        University of Utah
*        College of Nursing
*        10 S 2000 E Front
*        Salt Lake City, UT 84122-5880
***********************************************