Date:         Wed, 18 Jul 2007 16:14:51 -0500
Reply-To:     "Swank, Paul R" <Paul.R.Swank@uth.tmc.edu>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:         "Swank, Paul R" <Paul.R.Swank@uth.tmc.edu>
Subject:      Re: standardized vs. unstandardized coefficients
Comments: To: roxana dragan <dragan.roxana@GMAIL.COM>
In-Reply-To:  <200707181700.l6IGSn80029666@mailgw.cc.uga.edu>
Content-Type: text/plain; charset="us-ascii"

The unstandardized parameters tell you the difference in Y per unit
change in X whereas the standardized tells you the difference in Y in
standard deviations per standard deviation difference in X. Many people
think the latter is better because you can directly compare the effcts
across other predictors but this is not good practice. In a perfect
world where all the predictors are measured with the same degree of
reliability and there are no outliers or overly influential data points,
this might be true but it rarely is the case. In addition, the
standardized estimates are more subject to sampling variablity. I would
stick to unstandardized coefficients, or if you must include the
standardized ones, make sure you include the unstandardized ones as
well.

Paul R. Swank, Ph.D. Professor
Director of Reseach
Children's Learning Institute
University of Texas Health Science Center-Houston

-----Original Message-----
From: SPSSX(r) Discussion [mailto:SPSSX-L@LISTSERV.UGA.EDU] On Behalf Of
roxana dragan
Sent: Wednesday, July 18, 2007 12:00 PM
To: SPSSX-L@LISTSERV.UGA.EDU
Subject: standardized vs. unstandardized coefficients

Hi,

Can anybody tell me what is the difference between standardized and
unstandardized coefficients in a linear regression? I am rusty on
regressions, did not use them for about 4 years.

Thanks,
roxana