Вы находитесь на странице: 1из 33

Introduction to Factorial

ANOVA Designs

Factorial Anova
With factorial Anova we have more than one
independent variable
The terms 2-way, 3-way etc. refer to how many
IVs there are in the analysis
The following will discuss 2-way design but may
extended to more complex designs
The analysis of interactions (i.e. moderating
situations) constitutes the focal point of factorial
design

Recall the one-way Anova


Total variability comes from:
Differences between groups
Differences within groups

SS total
SS Between groups

SSwithin groups

Factorial ANOVA
With factorial designs we can break down1 the
model variability into:
Main effects
Mean differences among the levels of a particular
factor

Interaction
Differences among cell means not attributable to
main effects
When the effect of one factor on the DV is
influenced by the levels of another

Partition of Variability
Total variability

Between-treatments var.

Factor A
variability

Within-treatments var.

Interaction
variability
Factor B
variability

Example:
Arousal, task difficulty and performance
Yerkes-Dodson

Example
SStotal = (X grand mean)2
SStotal = 360
Df = N 1 = 29

SSb/t =n(cell means grand mean)2


= 5(3-4)2 + 5(1-4)2
SSb/t =240
Df= K# of cells 1 = 5

SSw/in = (X respective cell means)2 or SStotal- SSb/t


SSw/in = 120
Df = N-K = 24

Sums of Squares Between


SSDifficulty = n(row means grand mean)2
= 15(6-4)2 + 15(2-4)2 = 120
Df = # of rows (levels) 1 = 1

SSArousal = n(col means grand mean)2


= 10(2-4)2 + 10(5-4)2 + 10(5-4)2 = 60
Df = # of columns (levels) 1 = 2

SSDxA1 = SSb/t - SSDifficulty - SSArousal

= 240 - 120 - 60 = 60
Df = dfb/t - dfDifficulty dfArousal = 5-1-2 = 2
x Or dfdiff X dfarous

Output
Mean squares and F-statistics are calculated as
before

Initial Interpretation
Significant main effects of task difficulty and
arousal level, as well as a significant interaction
Difficulty
Better performance for easy items

Arousal
Low worst

Interaction
Easy better in general but much more so with high arousal

Effect size
Eta-squared is given as the effect size for B/t
groups (SSeffect/SStotal)
Partial eta-squared is given for the remaining
factors: SSeffect/(SSeffect + SSerror)
End result: significance w/ large effect sizes

Graphical display of interactions


What are we looking for?
Do the lines behave similarly (are parallel) or
not?
Does the effect of one factor depend on the level
of the other factor?

No interaction

Interaction

Graphical display of interactions


Main effects and interactions

Comparison to regression

Data using deviation coding


ANOVA output top with bold correlates to the
regression output using an interaction product
term1
F1
1.00
1.00
1.00
-1.00
-1.00
-1.00
1.00
1.00
1.00
-1.00
-1.00
-1.00

F2
-1.00
-1.00
-1.00
-1.00
-1.00
-1.00
1.00
1.00
1.00
1.00
1.00
1.00

DV
1.00
2.00
1.00
2.00
3.00
2.00
4.00
3.00
4.00
1.00
3.00
2.00

Interpretation: sig main fx and


interaction
Note that with a significant interaction, the main
effects are understood only in terms of that
interaction
In other words, they cannot stand alone as an
explanation and must be qualified by the
interactions interpretation
Some take issue with even talking about the
main effects, but noting them initially may make
the interaction easier for others to understand
when you get to it

Interpretation: sig main fx and


interaction
However, interpretation depends on common sense, and
should adhere to theoretical considerations
Plot your results in different ways

If main effects are meaningful, then it makes sense to


talk about them, whether or not an interaction is
statistically significant or not

E.g. note that there is a gender effect but w/ interaction we now


see that it is only for level(s) X of Factor B

To help you interpret results, test simple effects

Is simple effect of A significant within specific levels of B?


Is simple effect of B significant within specific levels of A?

Simple effects
Simple effects are the analysis of the effects of
one factor at one level of the other factor
Some possibilities from previous example
Arousal for easy items (or hard items)
Difficulty for high arousal condition (or medium or
low)

Simple effects
SS
SS

arousal for easy items

= 5(3-6)2 + 5(6-6)2 + 5(9-6)2 = 90

arousal for difficult items

SS
SS
SS

difficulty at lo

= 5(3-2)2 + 5(1-2)2 = 10
= 5(6-5)2 + 5(4-5)2 = 10
= 5(9-5)2 + 5(1-5)2 = 160

difficulty at med

difficulty at hi

= 5(1-2)2 + 5(4-2)2 + 5(1-2)2 = 30

Simple effects
Note that the simple effect represents a partitioning of
SSmain effect and SSinteraction
NOT JUST THE INTERACTION!!

From Anova table:


SS + SS
= 60 + 60 = 120
arousal

arousal by difficulty

SS
SS

= 90
= 30
90 + 30 = 120
arousal for easy items

arousal for difficult items

SS

difficulty

+ SS

arousal by difficulty

= 120 + 60 = 180

SS
= 10
= 10
SS
= 160
SS
10 + 10 + 160 = 180
difficulty at lo

difficulty at med
difficulty at hi

Output

Multiple comparisons and contrasts


For main effects multiple comparisons and
contrasts can be conducted as would be
normally
One would have all the same considerations for
choosing a particular method of post hoc
analysis or weights for contrast analysis

Multiple comparisons and contrasts


With interactions post hocs can be run
comparing individual cell means
The problem is that it rarely makes theoretical
sense to compare many of the pairs of means
under consideration

Contrasts for interactions


We may have a specific result to
look for with regard to our interaction
For example, we may think based on
past research moderate arousal
should result in optimal performance
for difficult items
We would assign contrast weights to
reflect this hypothesis: Is moderate
different from the other levels for
difficult items?

Contrasts for interactions


For this 3 X 2 design the weights will order as follows:
A1B1 A1B2 A2B1 A2B2 A3B1 A3B2
Note for this example, SPSS is analyzing categories in alphabetical order
Arousal hi lo med
Task Diff Easy
In other words
Hi:Difficult Hi:Easy Lo:Difficult Med:Easy

Comparisons beyond an interaction


Interactions are what is interesting in a general
sense, and any statistical nitpicking beyond it is
likely to vary wildly in results with new data1
Furthermore with increasing design complexity,
getting bogged down in the details of e.g.
pairwise comparisons among a 3way or higher
order interaction is inclined to let an interested
reader lose site of the forest for the trees

And like our multiple comparisons, the focus tends


much toward statistical rather than practical
differences

As such, it is probably best to leave the


interpretation of an interaction to what you can
see graphically as we do in the typical multiple
regression scenario
A graph such as this encompasses main effects,
interactions and simple effects in an easy to
follow reference and allows one to focus on
trends rather than getting lost in a flurry of
statistics

A different model

If cognitive anxiety is low, then the performance


effects of physiological arousal will be low; but if it is
high, the effects will be large and sudden.

Appendix
Leftovers from old notes that some
may find useful

Pulling simple effects off in SPSS

Paste!

Pulling it off in SPSS


Add
/EMMEANS = tables(a*b)compare(a)
/EMMEANS = tables(a*b)compare(b)

Test for simple fx with no sig


interaction?
What if there was no significant interaction,
do I still test for simple effects?
Maybe, but more on that later
A significant simple effect suggests that at
2
0
least one of the slopes across levels is
0
0
0
significantly different than zero
R
A
V
However, one would not conclude that the nae
interaction is close enough just because M
there was a significant simple effect
The nonsig interaction suggests that the
slope seen is not statistically different from
the other(s) under consideration.

VAR00003

1.00
2.00

0
1.00

2.00

VAR00001

Pulling it off in SPSS


Analyze

General Linear Model

Univariate

Select Dependent Variable and Specify Fixed


and/or Random Factor(s) (Treatment Groups and
or Patient Characteristic(s), Treatment Sites, etc.)
Paste

Launches Syntax Window


Add /LMATRIX command lines

RUN

All

/LMATRIX Command
/LMATRIX <Title for 1st Contrast>
<Specify Weights for 1st Contrast>;
<Title for 2nd Contrast>
<Specify Weights for 2nd Contrast>;

<Title for Final Contrast>


<Specify Weights for Final Contrast>

As alluded to previously it is
possible to have:
Sig overall F
Sig contrast
Nonsig posthoc

With this figure ->


Nonsig omnibus F

4
03
0
0
0
R
A
V
n
a
e2
M

Nonsig contrast
x e.g. 1 & 3 VS. 2

Sig post hoc


x 1 vs. 2 sig

0
1.00

2.00

VAR00001

3.00

Вам также может понравиться