Statistical significance

In inferential statistics, a result is statistically significant if it is unlikely to have occurred by chance. Statistical software packages generally report statistical significance using a test statistic and a p (probability) value (ranging between 0 and 1). If the p-value is less than a pre-selected critical value (critical $\alpha$) then it is considered to be "statistically significant".

The greater the statistical power, the greater the likelihood of a test being statistically significance. Power increases as each of these components increases:

• Effect size
• Sample size
• Critical value (critical $\alpha$)

Null hypothesis significance testing

During the 1900's, social sciences relied heavily on Null Hypothesis Significance Testing (NHST), however over-reliance on this technique was increasingly criticised during the 1980's and 1990's, leading to widespread recognition of the importance of also considering effect sizes, confidence intervals and statistical power as adjuncts or alternatives to tests of statistical significance during the 2000's.

See also

• Inferential statistics decision-making tree
• Statistical significance (Wikipedia)