eelbrain.test.pairwise

eelbrain.test.pairwise(y, x, match=None, sub=None, cells=None, data=None, par=True, corr='Hochberg', trend=False, title=True, mirror=False, labels=None)[source]

Pairwise comparison table

Parameters:

  • y (Var | str) – Dependent measure.

  • x (Factor | Interaction | NestedEffect | str) – Categories to compare.

  • match (Factor | Interaction | NestedEffect | str) – Repeated measures factor.

  • sub (Var | ndarray | str) – Perform tests with a subset of the data.

  • cells (Sequence[str | tuple[str, ...]]) – Cells to include. All entries have to be cells of model. Can be used to change the order of cells in the table.

  • data (Dataset) – If a Dataset is given, all data-objects can be specified as names of Dataset variables.

  • par (bool) – Use parametric test for pairwise comparisons (use non-parametric tests if False).

  • corr (bool | Literal['hochberg', 'bonferroni', 'holm']) – Method for multiple comparison correction.

  • trend (bool | str) – Marker for a trend in pairwise comparisons.

  • title (bool | str) – Title for the table.

  • mirror (bool) – Redundant table including all row/column combinations.

  • labels (dict[str | tuple[str, ...], str]) – Alternative labels for x as {cell: label} dictionary.

Returns:

table – Table with results.

Return type:

fmtxt.Table