Statistical Significance

Also known as: Statistical Significance, p-value, Hypothesis Testing, Significance Test

Probability that an observed result is due to chance; a result is significant when that probability is less than 5% (p < 0.05).

Statistical Significance is a fundamental concept indicating whether an observed result—a difference between groups, a correlation, a change over time—is large enough that it is unlikely to be due simply to sampling chance.

Conventionally, a result is statistically significant when the probability that it is due to chance (p-value) is less than 0.05 (5%), corresponding to a confidence level of 95%. In high-precision research, p < 0.01 is used.

Important: statistical significance ≠ practical importance. A result can be statistically significant but trivial in magnitude (especially with very large samples), or it can be practically important but not reach statistical significance (especially with small samples).

Atlantia applies standard significance tests in all its quantitative analyses and clearly communicates when differences are and are not statistically significant.