ShapiroTest

StatisticalTest

DashAI.back.statistical_tests.helper_tests.ShapiroTest

---

Test for normality of a single set of scores.

This helper test assesses whether a sample is plausibly drawn from a normal distribution. It is frequently used as a diagnostic step before applying parametric tests that assume normality, such as ANOVA or paired t-tests.

References

Shapiro, S. S., & Wilk, M. B. (1965). An analysis of variance test for normality (complete samples). Biometrika, 52(3/4), 591-611.

Methods

get_metadata(cls) -> dict

Defined on ShapiroTest

Metadata for Shapiro-Wilk Test.

run(self, scores: dict[str, list[float]], alpha: float = 0.05, **kwargs) -> DashAI.back.statistical_tests.statistical_test_result.StatisticalTestResult

Defined on ShapiroTest

Run the Shapiro-Wilk test for normality of a single score sample.

Parameters

scores : dict[str, list[float]]

Mapping containing exactly one score vector to evaluate.

alpha : float, optional

Significance level used to judge the null hypothesis of normality, by default 0.05.

Returns

StatisticalTestResult

A result object containing the Shapiro statistic, p-value, and the significance outcome.