whether the significant differences found are large enough to matter.

the number of subjects needed to correctly reject the null hypothesis.

the power of the statistical analysis.

where significant differences exist among groups.