If they are reasonably normal, most people would treat them as continuous.
There is debate about this, but there are some studies that show that
for four items, it doesn't really matter.
Jeremy
On 23/03/07, Mann, Helen N <[log in to unmask]> wrote:
>
>
> Would you say that data from a Likert scale (answers on a scale from 1- 5)
> is parametric or non-parametric data? I've been running correlations on the
> default Pearson's and now am worried I may have needed Spearmans…..I'm sure
> I should know the answer but my mind is blank
--
Jeremy Miles
Learning statistics blog: www.jeremymiles.co.uk/learningstats
|