Don Allen wrote:
Now I have a hard time accepting that "number of chairs" is interval data.
Interval? It is clearly ratio data -- zero means *no* chairs, two chairs is half of four chairs, etc. I think what you mean is that you don't believe that number of chairs measures desired interpresonal closeness with interval-level accuracy (e.g., two chairs' distance means I don't want you near exactly twice as much as one chair's distance).away.
A non parametric analysis would have been far more appropriate. Editorial rigour just ain't what it used to be.
There is a long and involved debate about the actual importance of levels of measurement. Stanley Stevens invented them primarily to support his claim that psychology was a science even though it didn't have measurement accuracy on the same level as the physical sciences. His analysis/rhetoric was eagerly picked up by psychological methods books of the time in order to advance the same scientific claim (though to students who didn't know better, rather than to other professional scientists and statisticians). The actual statistical importance of such "levels" was seriously called into question by Lord (1953), however, and there has been a contingent of statisticians ever since who think that the strong distinctions between different levels of measurement that we typically make in experimental psychology is, to a first approximation, a bunch of hooey.
Lord, F.M. (1953). On the statistical treatment of football numbers. American Psychologist, 8, 750-751.
Best, Christopher Green
--- You are currently subscribed to tips as: [EMAIL PROTECTED] To unsubscribe send a blank email to [EMAIL PROTECTED]
