Rev 68997 | Blame | Compare with Previous | Last modification | View Log | Download | RSS feed
% File src/library/stats/man/model.tables.Rd% Part of the R package, https://www.R-project.org% Copyright 1995-2015 R Core Team% Distributed under GPL 2 or later\name{model.tables}\alias{model.tables}\alias{model.tables.aov}\alias{model.tables.aovlist}\title{Compute Tables of Results from an Aov Model Fit}\description{Computes summary tables for model fits, especially complex \code{aov}fits.}\usage{model.tables(x, \dots)\method{model.tables}{aov}(x, type = "effects", se = FALSE, cterms, \dots)\method{model.tables}{aovlist}(x, type = "effects", se = FALSE, \dots)}\arguments{\item{x}{a model object, usually produced by \code{aov}}\item{type}{type of table: currently only \code{"effects"} and\code{"means"} are implemented. Can be abbreviated.}\item{se}{should standard errors be computed?}\item{cterms}{A character vector giving the names of the terms forwhich tables should be computed. The default is all tables.}\item{\dots}{further arguments passed to or from other methods.}}\details{For \code{type = "effects"} give tables of the coefficients for eachterm, optionally with standard errors.For \code{type = "means"} give tables of the mean response for eachcombinations of levels of the factors in a term.The \code{"aov"} method cannot be applied to components of a\code{"aovlist"} fit.}\value{An object of class \code{"tables.aov"}, as list which may contain components\item{tables}{A list of tables for each requested term.}\item{n}{The replication information for each term.}\item{se}{Standard error information.}}\section{Warning}{The implementation is incomplete, and only the simpler cases have beentested thoroughly.Weighted \code{aov} fits are not supported.}\seealso{\code{\link{aov}}, \code{\link{proj}},\code{\link{replications}}, \code{\link{TukeyHSD}},\code{\link{se.contrast}}}\examples{\donttest{options(contrasts = c("contr.helmert", "contr.treatment"))npk.aov <- aov(yield ~ block + N*P*K, npk)model.tables(npk.aov, "means", se = TRUE)## as a test, not particularly sensible statisticallynpk.aovE <- aov(yield ~ N*P*K + Error(block), npk)model.tables(npk.aovE, se = TRUE)model.tables(npk.aovE, "means")}}\keyword{models}