heplots (version 1.2-0)

SocGrades: Grades in a Sociology Course

Description

The data set SocGrades contains four outcome measures on student performance in an introductory sociology course together with six potential predictors. These data were used by Marascuilo and Levin (1983) for an example of canonical correlation analysis, but are also suitable as examples of multivariate multiple regression, MANOVA, MANCOVA and step-down analysis in multivariate linear models.

Usage

data(SocGrades)

source

Marascuilo, L. A. and Levin, J. R. (1983). Multivariate Statistics in the Social Sciences Monterey, CA: Brooks/Cole, Table 5-1, p. 192.

Details

midterm1, midterm2, final, and possibly eval are the response variables. All other variables are potential predictors. The factors class, sex, and hssoc can be used with as.numeric in correlational analyses.

Examples

data(SocGrades)
# basic MLM
grades.mod <- lm(cbind(midterm1, midterm2, final, eval) ~ 
	class + sex + gpa + boards + hssoc + pretest, data=SocGrades)
	
Anova(grades.mod, test="Roy")

clr <- c("red", "blue", "darkgreen", "magenta", "brown", "black", "darkgray")
heplot(grades.mod, col=clr)
pairs(grades.mod, col=clr)

heplot3d(grades.mod, col=clr, wire=FALSE)

if (require(candisc)) {
	# calculate canonical results for all terms
	grades.can <- candiscList(grades.mod)
	# extract canonical R^2s
	unlist(lapply(grades.can, function(x) x$canrsq))
	# plot class effect in canonical space
	heplot(grades.can, term="class", scale=4)	

	# 1 df terms: show canonical scores and weights for responses
	plot(grades.can, term="sex")
	plot(grades.can, term="gpa")
	plot(grades.can, term="boards")
	}