Date:         Mon, 21 Jan 2008 15:29:44 -0800
Reply-To:     Abdus Salam <mou_salam@yahoo.com>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:         Abdus Salam <mou_salam@yahoo.com>
Subject:      Which method would be more robust.
Content-Type: text/plain; charset=us-ascii

Dear expert listers,

I have a dataset where the variables are overall satisfaction(7 point scale),Value for money comparison(7 Point Scale) and also a variable with 13 attributes which contains Description ratings(7 point scale),One variable called attribute(7 point scale),Rating performance(7 Point Scale),Rating quality of Communication (7 point scale).


Now client needs a  robust Driver analysis based on these variable. What I think first

1) I have to make all the variables as Binary variable as I need to know only Top 2 Box performance.
2) To check the strength of relationship - I want to do correlation with overall satisfaction to all other variables.

Or:

To check which factor drives the overall satisfaction very well:

1) I have to make all the variables as Binary variable.

2) Then for removing the multicolinearity I have to do factor analysis on the base of 13 attributes.

3) After getting a good factor solution -taking those factor solution and other 4 variables as an independent variable; overall satisfaction as a dependent variable.

4) Finally wants to run a linear regression analysis with enter method.


Can anyone suggest me which one would be more robust method?


Thanks!
Mou.


      ____________________________________________________________________________________
Be a better friend, newshound, and
know-it-all with Yahoo! Mobile.  Try it now.  http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ

=====================
To manage your subscription to SPSSX-L, send a message to
LISTSERV@LISTSERV.UGA.EDU (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