Thursday, October 14, 2004

Implications and Interpretations: lessons on data massaging

Significant vs. Important – pg. 226, Understanding Statistics in the Behavioral Science

The procedure we have been following in assessing the results of an experiment is first to evaluate directly the null hypothesis and then to conclude indirectly with regard to the alternative hypothesis. If we are able to reject the null hypothesis, we say the results are significant. What we mean by significant is that the results are probably not due to chance. Or that the independent variable has a real effect, which can be replicated by repeating the experiment. It might have been better to use the term reliable to convey this meaning. However the usage of significant is well established, so we will have to live with it. The point we wish to make is that we must not confuse significant with important. Significance does not imply importance.