boonlert wrote:
>
> Dear all
> I have seen articles in marketing and organization behavior journal
> using a likert scale to test a hypothesis by applying it as a metric scale
> (run regression as a dependent variable). In addition, some statistical
> books are presented a likert scale as an ordinal scale while a few books are
> represented it as an interval scale. Could anyone help me to make clear of
> it so we can use a likert scale with the right analysis tools?
> Any response will be highly appreciated.
> Regards,
> Boonlert W.
There is no hard and fast answer. In practice it depends how people use the
scale. If they treat the intervals as reasonably fixed, reasonably regular
intervals then the ratings can be considered for most purposes as if they are
interval measurements. The important property is not that of the scale, but
the relationship between the data you collect and the thing you are attempting
to measure.
Thom
===========================================================================
This list is open to everyone. Occasionally, people lacking respect
for other members of the list send messages that are inappropriate
or unrelated to the list's discussion topics. Please just delete the
offensive email.
For information concerning the list, please see the following web page:
http://jse.stat.ncsu.edu/
===========================================================================
