Definitions for "Statistical significance"
a result is said to be significant when there is no more than a 5% chance that the same result could have been produced by random fluctuations. This is expressed as p = .05 .
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.
a difference found among groups after a comparative randomized investigation that is not likely to be caused by chance alone. The probability of it occurring by chance alone is often reported as P0.05.
A mathematical test that indicates that groups being compared are different
Results of a test to find out if a trend really is rising or falling, or whether any apparent rise or fall can be explained by random variation in the measurement.