The differences between
and misunderstanding of statistical effect size, significance and
power have been one of the greatest issues in modern statistics. It
is critical that the researcher learn these concepts. Remembering
that the key thing is the size of the key statistical effects (parameters
or coefficients) is foremost. Knowing that confidence in the statistic
is crucial supplements this. To be confident in a statistic we would
hope that it would be statistically significant, accurate, and powerful.
Once we have these in place simultaneously, we have a statistic with
the potential to substantially improve our business knowledge and
hopefully strategies.