name: op-test-bartlett-sphericity
description: Bartlett's sphericity test on a vector's correlation matrix (is it the identity?); the factorability check, bit-equal to the PCA operator's Bartlett scalars.
kind: operator
category: TEST
operator: TEST_BARTLETT_SPHERICITY
type: reference
applies_to: process, compose, predict
examples_tags: [hypothesis-test, tier-1-test, parametric, multivariate, streaming-friendly]
Tests emit statistic / p-value; no Response.Components.
Use when
Checks whether a set of numeric fields are related at all, the first check before summarising them as a few underlying dimensions.
Questions it answers:
- Do the ten satisfaction items move together enough to be worth reducing to a few scores?
- Are these sensor readings related at all, or is each one independent noise?
Use something else:
MAT_CORRELATIONwhen you want how strongly each pair is related.
Params
alpha— float, default0.05.missing—listwiseonly.
Slot: exactly one of vector (a vectors entry) / fields, ≥ 2 members; no field, vector2 / fields2 or split_by.
weight— both kinds; N* = Σw or n_eff.
Inputs
Members: u4…u64, f32/f64.
Output
Statistic = χ² = −ln|R|·(N* − 1 − (2p + 5)/6), DF = p(p − 1)/2, PValue. Details: members, n, n_dropped, determinant (|R|).
Reading the output
statistic: A chi-square that grows as the fields’ correlations move away from all zero; a small p-value says the fields are related enough to summarise as fewer dimensions.p_value: a p-value. Below alpha (0.05 unless the request sets another) the result is called significant; that is not the same as important, so read the effect size for how big it is.details.determinant: The determinant of the correlation matrix, 1 when the fields are uncorrelated and near 0 when they are strongly related; the statistic grows as it shrinks.
Gotchas
- Streamable. Same rows ⇒ same bits as the PCA operator’s Bartlett scalars (
MAT_PCA). - Singular R:
PULSE_MATRIX_SINGULAR; N* ≤ 1 + (2p + 5)/6:PULSE_MATRIX_INSUFFICIENT_N. Both warn, null figures. - Large n rejects almost always.
See
pulse_examples_search tags=[multivariate]- Skills:
multivariate-design,op-mat-pca,statistical-testing