MultipleANOVAs: 'MultipleANOVAs'

In NumbersInternational/flipAnalysisOfVariance: Functions for computing ANOVA, MANOVA, and related methods

MultipleANOVAs

Description

Computes multiple ANOVAs.

Usage

MultipleANOVAs(
  dependents,
  independent,
  subset = NULL,
  weights = NULL,
  compare = "To mean",
  correction = "Table FDR",
  robust.se = FALSE,
  alternative = "Two-sided",
  missing = "Exclude cases with missing data",
  show.labels = FALSE,
  seed = 1223,
  p.cutoff = 0.05,
  data = NULL,
  ...
)

Arguments

dependents	The outcome variables.
independent	The factor representing the groups.
subset	An optional vector specifying a subset of observations to be used in the fitting process, or, the name of a variable in data. It may not be an expression. subset may not
weights	An optional vector of sampling weights, or, the name or, the name of a variable in data. It may not be an expression.
compare	One of "To mean", "Pairwise", "To first" (which implement's Dunnett's C, when combined with 'correction' == 'Tukey Range'), or "All"
correction	The multiple comparison adjustment method: "Table FDR", "Tukey Range", "None", "False Discovery Rate", "Benjamini & Yekutieli", "Bonferroni", "Free Combinations", "Hochberg", "Holm", "Hommel", "Single-step" "Shaffer", and "Westfall".
robust.se	Computes standard errors that are robust to violations of the assumption of constant variance. This parameter is ignored if weights are applied (as weights already employ a sandwich estimator).
alternative	The alternative hypothesis: "Two sided", "Greater", or "Less". The main application of this is when Compare us set 'To first' (e.g., if testing a new product, where the purpose is to work out of the new product is superior to an existing product, "Greater" would be chosen).
missing	How missing data is to be treated in the ANOVA. Options: "Error if missing data". "Exclude cases with missing data", and "Imputation (replace missing values with estimates)".
show.labels	Shows the variable labels, as opposed to the labels, in the outputs, where a variables label is an attribute (e.g., attr(foo, "label")).
seed	The random number seed used when evaluating the multivariate t-distribution.
p.cutoff	The alpha level to be used in testing.
data	An optional data.frame that can be used as alternative to specifying dependent, independent, and weights for the case when the independent variables are deferring lengths. Must have columns "dependent", "independent", "weights", and "dependent.names". The "dependent.names" column contains the character name of the dependent variable for each observation. The dependent, independent, weights, and subset arguments are all ignored if data is supplied.
...	Other parameters to be passed to OneWayANOVA.

Details

Conducts multiple OneWayANOVAs, and puts them in a list. If correction is "Table FDR", the false discovery rate correction is applied across the entire table. All other corrections are performed within rows. Additional detail about the other parameters can be found in OneWayANOVA.

NumbersInternational/flipAnalysisOfVariance documentation built on Feb. 26, 2024, 4:52 a.m.