I do not claim to know the answer but have a suggestion to make and would
welcome comments on it. Could the cut off not be made in a post hoc
analysis, say at the median response so that half the respondents were in
the bottom group and the other half in the top group?
--
Kev Hopayian, Seahills, Leiston Rd, Aldeburgh, Suffolk IP15 5PL, England
Web site: http://www.suffolk-maag.ac.uk/kevhop
----------
>From: "E. Karen Rosenbloom" <[log in to unmask]>
>To: [log in to unmask]
>Subject: likert to dichotomous variable....
>Date: Thu, Nov 16, 2000, 9:37 am
> help needed
> I wondered if there is an accepted rule about at what stage can a 5
> point Likert scaled be response be considered a dichotomous response??
> I have looked and looked!
>
>
>
> E. Karen Rosenbloom Pharmacist
> Senior lecturer Derby University.
> 21 Brook Road Loughton Essex IG10 1BW
> telephone 020 8508 3011
>
>
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|