In education research, statistical significance and effect size are 2 sides
of 1 coin; they complement each other but they do not substitute for each
other. Good research practice requires that, to make sound research decisio
ns, both sides should be considered. In a simulation study, the sampling va
riability of 2 popular effect-size measures (d and R-2) was examined. The v
ariability showed that what is statistically significant may not he practic
ally meaningful, and what appears to be practically meaningful could have b
een the result of sampling error, thus not trustworthy, Some practical guid
elines are suggested for combining the 2 sources of information in research
practice.