inst/partials/functions-short.md

In factorH: Multifactor Nonparametric Rank-Based ANOVA with Post Hoc Tests

Function reference (short)

srh.kway.full()

One-call pipeline: ANOVA on ranks, descriptives, post hocs, simple effects.

Syntax:

srh.kway.full(y ~ A + B (+ C ...), data, max_levels = 30)

Returns list: anova, summary, posthoc_cells, posthoc_simple, meta.

write.srh.kway.full.tsv()

Export full result to a single TSV.

Syntax:

write.srh.kway.full.tsv(obj, file = "srh_kway_full.tsv", sep = "\t", na = "", dec = ".")

dec can be "." or ","; numeric output has no scientific notation; character tables are normalized for dec=",".

srh.kway()

K-way SRH-style ANOVA on ranks (Type II SS), tie-corrected p-values. Reports Effect, Df, Sum Sq, H, Hadj, p.chisq, k, n, eta2H, eps2H.

srh.effsize()

Two-factor SRH with effect sizes (eta2H, eps2H) from H.

nonpar.datatable()

Compact descriptives (global rank means, medians, quartiles, IQR) for all main effects and interactions.

srh.posthoc()

Dunn-Bonferroni pairwise matrices for a specified effect. Options: method, digits, triangular (lower/upper/full), numeric, sep.

srh.posthocs()

Dunn-Bonferroni pairwise matrices for all effects. Returns a named list with P.adj tables for A, B, A:B, etc.

srh.simple.posthoc()

Simple-effects post hocs within levels of conditioning factors. compare selects target factor; scope = "within" or "global".

srh.simple.posthocs()

Enumerate all simple-effect configurations for a design. Returns a named list like "COMPARE(gender) | BY(condition x age_cat)".

factorH documentation built on Sept. 11, 2025, 9:09 a.m.