LISTSERV at the University of Georgia
Menubar Imagemap
Home Browse Manage Request Manuals Register
Previous messageNext messagePrevious in topicNext in topicPrevious by same authorNext by same authorPrevious page (September 1998)Back to main SPSSX-L pageJoin or leave SPSSX-L (or change settings)ReplyPost a new messageSearchProportional fontNon-proportional font
Date:   Thu, 10 Sep 1998 09:15:56 EST
Reply-To:   Tim Dunsworth <Tim_Dunsworth@METRO2.METRO.MSUS.EDU>
Sender:   "SPSSX(r) Discussion" <SPSSX-L@UGA.CC.UGA.EDU>
From:   Tim Dunsworth <Tim_Dunsworth@METRO2.METRO.MSUS.EDU>
Subject:   Forwarded: Reply to: z-scores
Comments:   To: "SPSSX-L~&VM.MARIST.EDU"

From: Tim Dunsworth:MET Date: ## 09/10/98 09:15 ##

Previous comments: From: Tim Dunsworth:MET Date: ## 09/10/98 09:14 ## Comparable in what sense? It may be that the higher states are band hotbeds so this is a real effect that needs to be considered in future analyses. And if you plan analyses of score vs predictors (eg. band budget, school size, etc) high states may be high precisely because they are high on these predictors, in which case your proposed normalization would undermine any hope opf seeing those predictor relationships. Think carefully about what you are trying to achieve before proceding. If you have some good reason to think that high states just have scoring inflation, you might proceed as you proposed, or you might do analyses as an ANOVA with states as the categorical factor and predictors as covariates, or ...? HTH.

From: mmhewitt@FLASH.NET:smtp Date: ## 09/10/98 04:49 ## I have 250 marching bands' scores from 10 different states. All scores are based on 100 points and are similar in format, but when I run an ANOVA to compare the means, they show statistical differences among them. Would it be valid and/or reliable to convert the scores to z-scores for each state to make them comparable with one another?

Thanks -- Mike Hewitt University of Arizona

Back to: Top of message | Previous page | Main SPSSX-L page