selecting-statistical-test

To choose the right statistical test, you evaluate your study design, measurement scale, sample size, and assumption-check results like normality and equal variance to select a valid parametric or nonparametric test.

Use a nonparametric test when your data fails assumption checks for normality or equal variance. The selection process applies explicit gating checks to justify switching from parametric to nonparametric methods.

For paired before-after data, the recommended statistical test depends on the measurement scale and normality checks. The decision tree evaluates these explicit gating conditions to recommend the correct paired test.

For sparse 2x2 contingency tables with categorical data, the test selection uses expected cell counts as a gating check to recommend the appropriate exact test, ensuring valid conclusions for binary data.

Yes, this test selection method handles one-sample and repeated-measures scenarios by mapping the study design and measurement scale to appropriate statistical tests, including continuous, ordinal, and binary data.

The recommended statistical test includes an explanation by naming the specific gating check, such as normality or expected cell counts, that determined the choice. This makes the result defensible and easy to audit.