=========================================================================
Date:         Thu, 13 Jul 2006 15:06:52 -0300
Reply-To:     "Della Mora, Marcelo" <marcelo.dellamora@citigroup.com>
Sender:       "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:         "Della Mora, Marcelo" <marcelo.dellamora@citigroup.com>
Subject:      Likert scale
Comments: cc: Dominic Fernandes <dominic@spssme.com>
Content-Type: text/plain; charset="iso-8859-1"

Hi Dominic

Try reliability analysis that tests internal consistency
and relationship of each items to total score: correlation item-total, co-variance...

hope this helps you


Marcelo Della Mora



-----Original Message-----
From: SPSSX(r) Discussion [mailto:SPSSX-L@LISTSERV.UGA.EDU]On Behalf Of
Dominic Fernandes
Sent: Thursday, July 13, 2006 5:45 AM
To: SPSSX-L@LISTSERV.UGA.EDU
Subject: Likert scale


Hi All,

I have a question. How do we analyze a Likert scale (consisting of 7 responses) course evaluation questionnaire given to 25 teachers who attended an in-service course? Shall we treat the responses as scale variables and use parametric tests or shall we treat the responses as ordinal variables and use the nonparametric test.

Thank you in advance for your assistance.