Nana,
I forward to you some feedback I got from posing a similar question a while ago! Hope they are helpful.
Andreas
References:
Take a look at Cohen et al.'s (1999) POMP scores, which put things on a 0 to 100 scale, like a percentage.
Cohen, P., Cohen, J., Aiken, L. S., & West, S. G. (1999). The problem of units and the circumstance for POMP. Multivariate Behavioral Research, 34, 315-346.
----- Original Message -----
From: Plinio Monteiro
To:
[hidden email]
Sent: Thursday, January 17, 2008 8:21 PM
Subject: Re: Likert-Scale and Normalization
Hello,
Sorry about my poor english.
You can transform any scale in a 10 point scale by applying the following procedure:
Yi=([Xi-mín(Xi)]/[máx[Xi]-mín(Xi)])x10.
Where máx[Xi] and mín(Xi) are the maximum and minimum possible scores on a particular scale.
So if you have a value of 4 in a 1-7 scale, you can convert it as the example: ((4-1)/(7-1))x10=(3/6)x10=5.
Of course the number of points in a scale and the nature of the labels can affect the results, so it may be a problem compare scales of
different numbers of points even if you take a common scale for that.
Best regards
Plínio Rafael Reis Monteiro
Consultor em pesquisa de marketing e estatística empresarial
Mestre em Marketing pela UFMG
Fone: 98115773
-------------------------------------------------------------- To unsubscribe from SEMNET, send email to
[hidden email] with the body of the message as: SIGNOFF SEMNET Search the archives at
http://bama.ua.edu/archives/semnet.html====================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
