linear.comparison: Linear comparisons

In mattsigal/psyc6130: Package for PSYC6130

Description

Computes an F test for a linear comparison among group means. Uses procedures described in Maxwell & Delaney (2004; Designing Experiments and Analyzing Data), pages 1157-1162. This routine is suitable for one-way designs.

Usage

  linear.comparison(y, group, c.weights, alpha = 0.05,
    var.equal = TRUE)

  ## S3 method for class 'linear.comparison'
 print(x)

Arguments

y	Dependent variable
group	Grouping variable
c.weights	A list containing weights for multiple linear contrasts; or a numeric vector containing weights for 1 contrast
alpha	compute 100*(1-alpha)% confidence interval; default is 0.05
var.equal	logical. Are the variances equal? Default is TRUE.

Author(s)

Rob Cribbie and Phil Chalmers

Examples

## Not run: 
data(contrasts)
attach(contrasts)

linear.comparison(achiev,major,c.weights=list(c(1,0,-.5,-.5,0),
 c(1,-2,0,0,1),c(1,0,-1,-1,1),c(0,1,1,-1,-1)),var.equal=F)
 
## End(Not run)

mattsigal/psyc6130 documentation built on May 21, 2019, 1:26 p.m.