Factorial Designs Tyson S. Barrett, PhD PSY 3500 Experiments - - PowerPoint PPT Presentation

▶

Aug 10, 2023 962 likes •1.06k views

Factorial Designs Tyson S. Barrett, PhD PSY 3500 Experiments Manipulating a factor (some attribute of a participant) to discover a causal relationship How do we know our manipulation of the independent variable worked? If we were studying

SLIDE 1

Factorial Designs

Tyson S. Barrett, PhD PSY 3500

SLIDE 2

Experiments

Manipulating a factor (some attribute of a participant) to discover a causal relationship

How do we know our manipulation of the independent variable worked?

If we were studying how fear affects individuals ratings of various things, how could we do a manipulation check here?

SLIDE 3

Multiple Dependent Variables

Carryover Effects

Sometimes there’s one IV and many DVs There can be an affect of measuring dependent variables

Same Construct

Several DVs can measure different aspects of the same construct

Can combine multiple measures from the same construct into

ne (latent variable)

Cool statistical techniques can help with this

SLIDE 4

Factorial Designs

2 or more independent variables are tested simultaneously on one or more DVs

Treatment and Low Dose Treatment and High Dose Control and Low Dose Control and High Dose Treatment 1 and Morning Treatment 1 and Evening Treatment 2 and Morning Treatment 2 and Evening Control and Morning Control and Evening

2 x 2 Factorial Design 2 x 3 Factorial Design

SLIDE 5

Factorial Designs

2 or more independent variables are tested simultaneously on one or more DVs

The independent variable can be quasi-experimental (i.e., not manipulated) We are interested in interactions and main effects

When the effect of one IV depends on the level of another IV The effect of one IV does not depend on the level of another IV

SLIDE 6

Interactions

When the effect of one IV depends on the level of another IV

Interaction 1 1 0.0 −0.5 0.0 0.5 1.0 1.5

X1 X2

The effect of X1 depends on whether the person has a 0

r a 1 in X2

In this case, X1 has no effect when X2 is zero but has a big effect when X2 is one

SLIDE 7

Main Effects

The effect of one IV does not depend on the level of another IV

These are the effect of the variable regardless of other variables

Essentially the opposite of the interaction effects

There can be both an interaction effect and a main effect within the same factorial design

SLIDE 8

Which is which?

Main Effect X1 and X2 Interaction Main Effect X1 Main Effect X2 1 1 1 1 0.0 0.5 1.0 1.5 2.0 −0.5 0.0 0.5 1.0 1.5 0.0 0.5 1.0 1.5 2.0 1 2 3 4

X1 Outcome X2

Effect of X1?
Effect of X2?
Both have an effect?
Interaction effect?

Which is: