students | R Documentation |
Students were administered two parallel forms of a test after a random assignment to three different treatments.
data("students")
A data frame with 35 observations on the following 3 variables.
treatment
a factor with levels AA
, C
, and
NC
.
low
the result of the first test.
high
the result of the second test.
The data arise from a large study of risk taking (Timm, 2002).
Students were randomly assigned to three different
treatments labelled AA, C and NC. Students were administered two
parallel forms of a test called low
and high
. The aim is to
carry out a test of the equality of the bivariate means of each treatment
population.
N. H. Timm (2002), Applied Multivariate Analysis. Springer, New York.
data("students", package = "HSAUR") layout(matrix(1:2, ncol = 2)) boxplot(low ~ treatment, data = students, ylab = "low") boxplot(high ~ treatment, data = students, ylab = "high")