Hello again,
I hope that you (listmembers) don't mind me posting questions like this to
the list and I hope that this is an appropriate place to ask such questions.
I've tried to find the answer to this question in Tabachnick & Fidell, but
maybe my question is too simplistic for this book!
I have a dataset with 3 IV's and 1 DV. All of this data has come from Likert-
type scales (i.e. strongly disagree = 1 to 5 = strongly agree). Much research
in this area treats the data from Likert scales as interval, but many would
argue that this data is really ordinal.
I have drawn scaterplots of may data (hence the previous post) and am
content that linear relationships exist between my IV's and my DV.
Is multiple linear regression an appropriate test to use on ordinal data, or is
there a non-parametric equivalent?
Thanks in advance,
Brian
P.S. If anyone knows a good "multiple regression for beginners book" please
let me know!
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Dr. Brian McMillan
Mother and Infant Research Unit
22 Hyde Terrace,
Leeds LS2 9LN.
Tel: 0113 233 1575
Fax: 0113 244 9730
http://www.leeds.ac.uk/miru
|