Psy 207B Introduction
to Statistics
Factorial
Analysis of Variance
Erwin Segal
return to syllabus
The study of Factorial Analyses of Variance leads us
into some of the most powerful methods in Statistics for analyzing data.
Most experimental research in psychology and other social sciences use
either variants of analysis of variance or other procedures which are derived
from them and from correlational analyses which we have only touched upon
in this course.
Factorial ANOVA has two independent variables which are
crossed with each other. That means each value of one variable is paired
with every value of the other variable. Two or more individuals are assigned
to each combination of values of the independent variables. For this kind
of analysis, there is only one dependent variable which is collected from
every individual in the study. Let's make sense out of this with an example.
A worked-out problem
using Excel
Click here for discussion
of linear analysis
Another
example
One can use Newman-Keuls probabilities to evaluate the q
statistic. It is more powerful than Tukey's HSD. Some authors feel, however,
that it increases the probability of a Type 1 error too much.