Professional Documents
Culture Documents
Basic Concepts
Two-Way ANOVA
• We have two grouping variables,
commonly referred to as:
– Factors
– Independent Variables
• best term if manipulated experimentally
– Predictors
– Grouping Variables
– Classification Variables
• We have one continuous variable,
commonly referred to as the
– Dependent variable
• best term if data collected experimentally
– Criterion variable
– Outcome variable
– Response variable
– Comparison variable
A 2 x 2 Design
• Has two levels of Factor A and two levels
of Factor B.
• This results in four combinations of level of
A and level of B.
• Each such combination is referred to as a
cell.
Your Party is Going Strong
• And then green aliens crash it
Data Collection
• You and colleagues monitor the aliens’
behavior.
• Half of them consume your ethanol-based
punch.
• Half of them sample your room-mate’s
barbiturate tablets.
• Post consumption, you blindly rate each
alien’s level of intoxication, and you
compute cell means.
Green Alien Party Crashers
Alcohol
Barbiturate none one marginal
none 00 10 05
one 20 30 25
marginal 10 20 15
30
20 no barb.
one barb.
10
0
no alch. one alch
Main Effects & Marginal Means
Alcohol
Barbiturate none one marginal
none 00 10 05
one 20 30 25
marginal 10 20 15
40
30
no barb.
20 one barb.
10
0
no alch. one alch.
Main Effects & Marginal Means
Alcohol
Barbiturate none one marginal
none 00 10 05
one 20 40 30
marginal 10 25 17.5
30
no barb.
20 one barb.
10
0
no alch. one alch.
30
20 no barb.
one barb.
10
0
no alch. one alch
Main Effects & Marginal Means
Alcohol
Barbiturate none one marginal
none 00 20 10
one 30 10 20
marginal 15 15 15
20 no barb.
one barb.
10
0
no alch. one alch
30 40
30
20 no barb. no barb.
one barb. 20 one barb.
10
10
0 0
no alch. one alch no alch. one alch.
Purple
30
20 no barb.
one barb.
10
0
no alch. one alch
Hypotheses Tested in Two-Way
ANOVA
1 = 2 = . . . = a
– That is, the mean of the criterion variable is
constant across the a levels of factor A.
1 = 2 = . . . = b
– That is, Factor B does not affect the mean of
the criterion variable.
• Factors A and B do not interact with one
another, A and B combine additively to
influence the criterion variable.