Statistical significance refers to the scientific legitimacy of a research finding. Research findings are typically considered statistically significant only if the results would occur by chance less than five times out of one hundred. Statistical significance is based on a mathematical cut-off. It communicates very little about whether or not the finding is useful in real life (21). See practical significance for more information. In a study, a researcher finds a small relationship between happiness and time spent outdoors. In real life, however, the actual influence of time outdoors on happiness proves so small that being out all day has little affect. In this example, the research findings are statistically significant, but not practically significant.