Significant: Statistics Versus Everyday Speech

Read 'significant' correctly in research, where it does not simply mean important.

In casual speech, 'significant' means important or large. In statistics, 'statistically significant' means something narrower: the result would be unlikely to appear by random chance alone, given the study's assumptions. It does not automatically mean the effect is big or that it matters in real life. Worked example: 'A study of 50,000 students found that a new schedule produced a statistically significant improvement of 0. 3 points on a 100-point test. ' A careless reader thinks the schedule had a major effect.

A careful reader notices that with a huge sample, even a tiny difference can be statistically significant. The finding is probably real, but 0. 3 points is small. Researchers often separate 'statistical significance' from 'practical significance' or 'effect size' for exactly this reason. Whenever you see 'significant' in a report, ask which meaning the writer intends.

• 1Statistically significant means unlikely to be due to chance alone.
• 2It does not guarantee a large or important effect.
• 3Look for the effect size to judge practical importance.