What does it mean for an experimental result to show significance?

Answer 1

For data to show significance, the resulting percentage must be less than the chosen level of significance.

The term "significance" links to statistics. A scientist sets forth two hypotheses: the null hypothesis that no change or effect is present, and the alternative hypothesis that a change or effect has occurred. When the data is quantitative, the conclusion is not black-and-white. Probability can push the results one way or the other and the results will vary each time the same exact experiment is carried out.One way is to take the average results of many, many repetitions of the same experiment. Another is to use a statistical test like the chi-squared. The scientist chooses a "good enough" level, the level of significance, say 5% or (if he's very picky and demanding) 1%. The lower the level of significance, the more precision and accuracy is being demanded. For example, picking 100% is ridiculous because while the scientist can be 100% statistically confident, this demonstrates absolutely no precision or accuracy. The test then calculates a percentage from the data. If the percentage is lower than the level of significance, say 2.8% compared to the chosen 5%, this indicates a high certainty and the null hypothesis is rejected. (If the percentage is higher, say 13% compared to the 5%, neither the null nor the alternative can be rejected nor accepted- simply put, no conclusion can be drawn.)