Date:         Mon, 21 Nov 2005 21:12:35 +0100
Reply-To:     Henrik Lolle <lolle@socsci.aau.dk>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:         Henrik Lolle <lolle@socsci.aau.dk>
Subject:      Re: correlation question
Comments: To: Caroline McKenna <caroline0277@hotmail.com>
In-Reply-To:  <BAY103-F1075BF3D2990515444336EDC530@phx.gbl>
Content-Type: text/plain; charset="us-ascii"; format=flowed

Hi,

Depending on how many districts are in the analysis you should run
the analysis in the MIXED LINEAR procedure. And if you will say
anything statistical about the effect from size, you must have not
just a few districts. The effect and especialy the standard error
from the size variable will be estimated wrong if you just make a
correlation analysis on the individual level (pupil). If ordinary
analysis is used, it is better to analyse on the aggregate level (district).

Henrik

At 16:33 21-11-2005, Caroline McKenna wrote:

>I am in the middle of conducting a study where I am comparing the
>size of a school district with student test scores to identify if
>there is a correlation between district size and achievement levels.
>I just want to make sure I have the tests straight... I am planning
>on running an independent t test and also creating a scatter plot
>and using a correlation test to identify if there is a relationship.
>Am I on the right track? Any feedback is greatly appreciated.
>
>Caroline



***********************************************************
Henrik Lolle
Associate Professor
Department of Economics, Politics and Public Administration
Aalborg University
Fibigerstraede 1
DK 9220 Aalborg East
http://www.socsci.auc.dk/institut2/dansk/empl/lolle.htm
***********************************************************