This preview shows pages 1–2. Sign up to view the full content.
10/19/10
Class Notes: ANOVA (Analysis of Variance)
When to use: When you want to compare the means of more than two groups.
Imagine an example where we know information about three samples. Maybe a
researcher is studying three ways of teaching (group discussion, lecture, and a
combination). The researcher gets the following means of class grades for students in
each class format:
Group Discussion:
80
Lecture:
85
Combination:
95
The logic of ANOVA
The new question we ask is: Are the sample means different from one another by MORE
than we would expect if H
o
is true?
Or, is the variance
between groups
greater than
would be expected if the null hypothesis was true?
Why not just use a bunch of ttests?
Because there’s a large chance you’ll make a type
one error. Inflated Alpha. Also who would want to conduct 7 different ttests for ex.
ANOVA permits the control of alpha at a predetermined value when simultaneously
testing the equality of any number of means. In ANOVA, all differences for all pairs are
examined simultaneously to see if one more of the means deviates significantly from one
or more of the other means.
ANOVA is an
omnibus
test
Omnibus means overall. It doesn’t tell us detail. It tells us somewhere in these means
there is a difference.
One way ANOVAin terms of number of independent variables with multiple
conditions. Can have 2, 3, 4, etc.
ANOVA uses a new test: The F test
F
obt
= variance among sample means
Variance within means
**Variance within is the variance we would expect by chance.
The F test allows us to look at why variance exists between the group means
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
This is the end of the preview. Sign up
to
access the rest of the document.
This note was uploaded on 01/17/2011 for the course PSYC 3990 taught by Professor Vandellen during the Fall '10 term at University of Georgia Athens.
 Fall '10
 VANDELLEN

Click to edit the document details