Hi I think that some of the confusion about this term does arise from its different uses in different areas. Although the idea of a manipulated variable, quantitative or qualitative, does arise often in the empirical sciences, I have come across a more general use of the term in statistics. At least one textbook by statisticians (perhaps Moore or McCabe?) at one time referred to any collection of information as an experiment. Hence, rolling a die was an experiment. I'm not sure what term other statisticians would use to refer to the more general notion of collection information or observations. Although perhaps confusing to a naive reader, experiment as any observation was clear in its context.
In at least some of the empirical sciences, experimental vs. non-experimental tends to depend on existance of one or more manipulated variables in experiments but not in non-experiments. Non-experimental is preferred to correlational/regression because method of analysis (anova, regression) is not tied to the experimental / non-experimental distinction in any strong way, although the two appear correlated in practice. That is, ANOVA can be used in non-experimental work when qualitative variables are involved and Regression can be used in experimental work. And as someone else properly noted both are realizations of the general linear model. Even in the empirical sciences, however, there are undoubtedly disciplines that use the term experiment more generically (i.e., equivalent to study or the collection of any set of observations). Your best bet would perhaps be to consider (1) the importance of having a term to denote any collection of observations, (2) the importance of having a term to distinguish between experiment and non-experiment, and (3) standard practice in your discipline with respect to these two criteria. Best wishes Jim ============================================================================ James M. Clark (204) 786-9757 Department of Psychology (204) 774-4134 Fax University of Winnipeg 4L05D Winnipeg, Manitoba R3B 2E9 [EMAIL PROTECTED] CANADA http://www.uwinnipeg.ca/~clark ============================================================================ . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
