Complex Experiments

Basic Experiment

• Simplest experimental design– Two levels of one independent variable

• Compares only two groups

Factorial Designs: Increasing the Number of Independent Variables

• Typically, two or three IVs are operating simultaneously (in the real world)

• Factorial designs include at least 2 IVs

• All levels of each IV are combined with all levels of the other IVs

• Simplest factorial is a 2 X 2 factorial design

A Tasty Example

2 X 2 Factorial Design

Factor A (IV1):

Type of Topping

Level 1 = Ketchup

Level 2 = Salsa

Factor B (IV2):

Type of Food

Level 1 = French Fries

Level 2 = Tortilla Chips

DV: Liking of taste

Other Factorial Designs

3 X 4 Factorial Design

2 X 3 Factorial Design

2 X 2 X 2 Factorial Design

Identify the number of experimental conditions

in each of these designs.

Main Effects and Interactions

• Interpretation of Factorial Designs:– A main effect tells us the effect each

variable has by itself.

– An interaction tells us that the effect of one independent variable depends on the particular level of the other.

For each example, determine:

• Is there a main effect for Type of Topping?

• Is there a main effect for Type of Food?

• Is there an interaction?

• Describe the graph in words (is there a moderator?).

-2.0-1.5-1.0-0.50.00.51.01.52.0

Ketchup Salsa

Type of Topping

Ta

ste

Pre

fere

nc

e

French Fries Tortilla Chips

1.

-2.0-1.5-1.0-0.50.00.51.01.52.0

Ketchup Salsa

Type of Topping

Ta

ste

Pre

fere

nc

e

French Fries Tortilla Chips

1.

-2.0-1.5-1.0-0.50.00.51.01.52.0

Ketchup Salsa

Type of Topping

Ta

ste

Pre

fere

nc

e

French Fries Tortilla Chips

2.

-2.0-1.5-1.0-0.50.00.51.01.52.0

Ketchup Salsa

Type of Topping

Ta

ste

Pre

fere

nc

e

French Fries Tortilla Chips

2.

-2.0-1.5-1.0-0.50.00.51.01.52.0

Ketchup Salsa

Type of Topping

Ta

ste

Pre

fere

nc

e

French Fries Tortilla Chips3.

-2.0-1.5-1.0-0.50.00.51.01.52.0

Ketchup Salsa

Type of Topping

Ta

ste

Pre

fere

nc

e

French Fries Tortilla Chips3.

-2.0-1.5-1.0-0.50.00.51.01.52.0

Ketchup Salsa

Type of Topping

Ta

ste

Pre

fere

nc

e

French Fries Tortilla Chips

4.

-2.0-1.5-1.0-0.50.00.51.01.52.0

Ketchup Salsa

Type of Topping

Ta

ste

Pre

fere

nc

e

French Fries Tortilla Chips

4.

-2.0-1.5-1.0-0.50.00.51.01.52.0

Ketchup Salsa

Type of Topping

Ta

ste

Pre

fere

nc

e

French Fries Tortilla Chips

5.

-2.0-1.5-1.0-0.50.00.51.01.52.0

Ketchup Salsa

Type of Topping

Ta

ste

Pre

fere

nc

e

French Fries Tortilla Chips

5.

Simple Main Effects

Remember, main effects are looking at an OVERALL effect of one IV across levels of the other IV (we examine the average of both levels).

We examine simple main effects when we isolate the effect of one IV at each level of the other IV.

Example: Berkowitz and LePage (1967)

2 X 2 Factorial Design

Factor A (IV1):

Presence of a Weapon

Level 1 = Gun

Level 2 = Badminton Racket

Factor B (IV2):

Amount of Provocation

Level 1 = High Provocation

Level 2 = Low Provocation

DV: Aggression: Number of shocks delivered

Berkowitz and LePage (1967)

3.07

2.40

4.67

5.87

1

2

3

4

5

6

No WeaponPresent

Weapon Present

Weapon Presence

Nu

mb

er

of

Sh

ocks D

eli

vere

dLow Provocation High Provocation

Berkowitz and LePage (1967)

3.07

2.40

4.67

5.87

1

2

3

4

5

6

No Weapon Present Weapon Present

Weapon Presence

Nu

mb

er

of

Sh

ocks D

elivere

dLow Provocation High Provocation

3.07

2.86

2.74

3.06

2.6

2.7

2.8

2.9

3.0

3.1

3.2

Control Shared Birthday

Se

lf-P

erc

ep

tio

ns

of

Att

rac

tiv

en

es

s

Other Person was Unattractive

Other Person was Attractive

Brown, Novick, Lord, & Richards (1992)

3.07

2.86

2.74

3.06

2.6

2.7

2.8

2.9

3.0

3.1

3.2

Control Shared Birthday

Se

lf-P

erc

ep

tio

ns

of

Att

rac

tiv

en

es

s

Other Person was Unattractive

Other Person was Attractive

Brown, Novick, Lord, & Richards (1992)

IV x PV Designs

• Factorial designs with manipulated and nonmanipulated variables (sometimes called IV x PV designs– Independent variable (IV) x participant variable

(PV)– Allows researchers to examine how different

individuals respond to the same manipulated IV

0

2

4

6

8

10

Men Women

He

alt

h a

nd

Lo

ng

ev

ity

Happily Married Unhappily Married Unmarried

High

Low

Orth-Gomer et al. (2000)

PV x PV design

• Forehand and Perkins (2005) studied consumer reactions to ads with celebrity voiceovers. They found that prior attitudes toward the celebrities influenced how much consumers liked or disliked the products, but this influence was greatest when consumers weren't sure which celebrity provided the voice-over.

• Participants liked the product more if:– they liked the celebrity doing the voiceover

AND– they did NOT recognize the celebrity’s voice in the

ad

Between and Within Group Designs

• Assignment procedures and factorial designs– Two basic ways of assigning participants to

conditions

1. Between (Independent) groups design

2. Within (Repeated measures) design

• Combination of the two basic ways is called a mixed factorial design

Example: Silverman et al (1978)

• Male’s ability at dart-throwing before and after subliminal presentation of a message “Beating Dad is OK”

• Poor design:

Before Message

After Message

X

X

X

X

X

X

X

X

X X