Charlotte opined: "That's not quite true. You shouldn't use a t test for
ordinal data. I generally distinguish between nominal ("categorical"),
ordinal ("ranked") and interval/ratio ("continuous") data."
I'm with Jim on this one. Distributional assumptions are important
for t tests, but the ordinal versus interval distinction not. Even if that
distinction were important, how would you ever know whether a scale is
really interval rather than just ordinal. If ordinal, the function relating
data to true scores must be positive monotonic, if interval then that
function must be positive linear. How could we ever know the nature of that
function? It is largely a matter of faith. The transcendental way out of
this is to generalize one's results to the population of true scores which
could be considered to be a positive linear function of ones data.
+++++++++++++++++++++++++++++++++++++++++++++++++
Karl L. Wuensch, Department of Psychology,
East Carolina University, Greenville NC 27858-4353
Voice: 252-328-4102 Fax: 252-328-6283
[EMAIL PROTECTED]
http://core.ecu.edu/psyc/wuenschk/klw.htm
---
You are currently subscribed to tips as: [EMAIL PROTECTED]
To unsubscribe send a blank email to [EMAIL PROTECTED]
