This is not so much an S-Plus question as it is a statistics question, but
I'm wondering what tools (if any) S-Plus has to help me investigate this.
Effect size is, apparently, used fair bit in the social sciences. It
appears to be considered an adjunct (in some cases, an alternative) to more
classical significance testing. I'm looking into this because I often
(though not always) use large data sets and am uncomfortable with the fact
that very small, physically insignificant, differences in, say, the mean
heights of the analyzed and 24 hr forecast 500 hPa pressure surfaces result
in *statistically* significant differences. I'm concerned about the power
of classical significance tests for my applications, which is one of the
things Effect Size is intended to address. I have never heard Effect Size
mentioned in the physical sciences or in meteorology (my
speciality). Because of this, the nomenclature surrounding Effect Size
seems a bit opaque to me. If anyone knows of some writing on Effect Size
that uses more standard statistical jargon, I'd be be happy to know of
it. Also, I'm very happy to entertain a dialog with anyone to help me
better understand how I might approach my statistical problems differently.
Kim Elmore
Kim Elmore, Ph.D.
"All of weather is divided into three parts: Yes, No, and Maybe. The
greatest of these is Maybe" The original Latin appears to be garbled.
|