Jimjohn, I think this might be an example the ecological fallacy?
Regards,
Graeme.
-----Original Message-----
From: SPSSX(r) Discussion [mailto:
[hidden email]] On Behalf Of jimjohn
Sent: 17 September 2008 20:56
To:
[hidden email]
Subject: Statistics Question:
sorry guys, this isn't related to SPSS but it is related to statistics.
some of the variables in my data set include: Redemption Rate, Incentive,
Issue Date, Age, original balance, etc.
I have about 9 million cases, and I am analyzing the correlation between
redemption rate (between 0 and 100) and incentive (betwen -1 and 1). The
correlation comes out to .097
Now, I try grouping the variables by Issue Date and by Age. and then I
calculated the weighted redemption rate weighted by original balance, as
well as the weighted incentive weighted by original balance. Now, when I
analyze teh correlation between the new redemption rate and the new
incentive, I get a much higher correlation. It comes out to about .33.
Any ideas why the correlation woudl jump so much? I can't seem to understand
intuitively why ti would jump so much when I am just using the same
variables just aggregated. Thanks!
--
View this message in context:
http://www.nabble.com/Statistics-Question%3A-tp19539826p19539826.htmlSent from the SPSSX Discussion mailing list archive at Nabble.com.
=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD
The University of Aberdeen is a charity registered in Scotland, No SC013683.
=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD
Locked
