onewayComp: Simultaneous Confidence Intervals and Hypothesis Tests for...

Description Usage Arguments Value Author(s) Examples

View source: R/onewayComp.R

Description

Generates Studentized confidence intervals and/or hypothesis tests for one-way designs. P-values can be adjusted ...

Usage

1
onewayComp(formula,data,alpha=.05,var.equal=TRUE,con=NA,nboot=0,adjust='one.step')

Arguments

formula

e.g. x~group, this would the same formula you would supply to lm for a one-way ANOVA

data

data frame containing x and group

alpha

significance level of the test, 1-alpha is the overall confidence level

var.equal

Defaults to TRUE, if FALSE, then Welch type corrections are used for the Studentized test statistics. FALSE along with adjust='one.step' gives the Games-Howell procedure.

con

optional contrast matrix. If con=NA, then all pairwise comparisons are made.

nboot

if nboot=0, then t-distributions are used for P-values and critical values (including Tukey max t), but if nboot>0 then resampling is used to approximate the critical values and P-values. For the bootstrap resampling separate samples are used and each sample is recentered so that the null hypothesis is true.

adjust

possible values are 'one.step' for Tukey Kramer or Games Howell procedures that give both hypothesis tests and confidence intervals (cannot be used with custom contrast matrix). 'bonferroni', 'none', 'holm', or any other adjustment that can be passed to p.adjust()

Value

call

Shows the details of the function call.

comp

A matrix, with names columns. 'diff' is the sample mean difference or sample mean contrast value. 'lwr' and 'upr' are the confidence interval limits (if available). 't' is the value of the Studentized test statistic. 'p' is the unadjusted p-value as would be computed by a single t-test. 'p adj' is the adjusted p-value as determined by the selected adjust method. 'Rej H_0' is 1 if p adj is less than alpha

pairw

Matrix showing pairwise comparisons and P-values.

Author(s)

Jeff Baggett, University of Wisconsin - La Crosse.

Examples

1
2
3
4
5
6
7
8
##---- Should be DIRECTLY executable !! ----
##-- ==>  Define data, use random,
##--	or do  help(data=index)  for the standard data sets.

## The function is currently defined as
function (x) 
{
  }

DataScienceUWL/DS705data documentation built on Jan. 15, 2022, 3:43 a.m.