Factorial Design

factorial designs

designs with more than one IV (or factor) where all levels of each IV are combined with all levels of the other IVs

main effects

the effect each variable has by itself, irrespective of the effect of the other variable

interaction effects

the effect of one IV depends on the particular level of the other IV

simple effects

the effect of one IV on the particular level of the other IV, only relevant when interaction effects are present

2×2 factorial design examples

• Top right: main effect between A, but not B

  • Main effect between A - difference between A1 and A2

  • No main effect between B - the levels of the IV are the same

• Bottom left

  • Main effect for B 

    • Level for B1 is much higher than B2

  • No interaction effect  

• Bottom right

  • B has a main effect (can see that B1 is higher than B2)

  • Main effect of A (A2 is higher than A1)

experimental research vs non-experimental or quasi-experimental research

experimental research: manipulate IV

non-experimental research: non-manipulated IV

manipulated vs non-manipulated IVs

IVs in factorial designs can be manipulated or non-manipulated

• If all IVs are non-manipulated → non-experimental or quasi-experimental

• If any IV is manipulated → experimental

between-subject IV

each participant exposed to one level of IV

within-subject IV

each participant exposed to all levels of IV

between-subject or within-subject IVs

• All IVs in factorial design are between subject → between-subject factorial design

• All IVs in factorial design are within subject → within-subject factorial design

mixed factorial design

some IVs in factorial design are between-subject and some are within-subject

Is there a main effect(s)?

Is there an interaction effect?

Is there a simple effect?

Main Effect
- There is a main effect of workload.
Students have a higher level of stress
when the workload is high as compared
to when the workload is low.
- There is a main effect of the roommate
situation. Students have a higher level
of stress when they do not have a
roommate as compared to when they do.
- There are main effects of workload and
roommate situation.

Interaction Effect
- There is an interaction between
workload and roommate situation.
- The effect of workload on stress depends
on the roommate situation.
- The effect of roommate situation on
stress depends on the workload

Simple Effect
Only relevant when there is an
interaction.
- When students have a roommate, having
a high workload is associated with more
stress than having a low workload.
- When students do not have a roommate,
having a high workload is associated
with more stress than having a low
workload.
- The difference in stress level between
high versus low workload is larger when
students have a roommate than when
they do not.