Lecture 13 Psyc 300A

Lecture 13Psyc 300A

Review• Confounding, extraneous variables • Operational definitions • Random sampling vs random

assignment• Internal validity• Null hypothesis • Type I and type II errors

Review: Confounding and extraneous variables

• Extraneous variables can be confounds, but can also add variability (noise). For each, provide extraneous variable and confound:

• Study 1: Effect of distraction on pain perception using cold immersion.

• Study 2: Do girls benefit from sixth grade middle school?

Review: Operational definitions• For each of the previous studies,

operationalize the IV and DV

Review: Random sampling vs random assignment

• What is the difference between the two?

• Random assignment is a way to prevent confounding

Review: Internal validity• What is internal validity?• Internal validity: Ability to make

valid inferences concerning the relationship between the IV and DV in an experiment. (effect on the DV is caused only by the IV)

Type I and Type II ErrorsAccept the Null Hypothesis

Reject the Null Hypothesis

Null is really True(chance is responsible)

Correct Decision

Type I Error

Null is really False(chance is not responsible)

Type II Error

Correct Decision

Power• Power is the probability of avoiding

a Type II error.• Power is related to:

– Alpha level– Effect size (mean and sd)– Number of participants

Using More Than Two Levels of an IV• What is a level of an IV?

• In an experiment with an experimental and control group, how many levels?

• Can we have more than two levels?• Example:

– Golf club study– Anxiety management techniques for

speech-giving• Graphing the relationship

Advantages of Multi-level Designs• Efficiency (fewer participants

needed and less time)• Ability to see relationships better

– Ex: Caffeine and Performance (0, 2, 4 cups of coffee)

Graphing Relationships: One IV

0

1

2

3

4

5

6

7

8

9

10

Romantic Action

Movie type

Ratin

g

0

2

4

6

8

10

Romantic Action

Movie TypeRating

Multifactor Designs• Factorial design: A design in which

all levels of each IV are combined with all levels of the other IVs.

• Advantages of factorial designs:– More efficient (fewer participants and

less experimenter time)– Allows us to see how variables

interact

Example: Movie Preferences

0

2

4

6

8

10

Romantic Action

Movie Type

Rat

ing

menwomen

Men Women Mean

Romantic

3 6 4.5

Action 7 4 5.5

Mean 5 5

What a Factorial Design Tells You• Main effect: The effect of an IV on

the DV, ignoring all other factors in the study

• Interaction effect: When the effect of one IV on a DV differs depending on the level of a second IV.

• Graphing a factorial design• Interpreting the interaction first

Examples of Main Effects and Interactions

• A1= morning• A2= late

afternoon

• B1= high fat diet• B2= low fat diet

• DV: 0-50 rating of energy level

More Main Effects and Interactions• A1= morning

• A2= late afternoon

• B1= high fat diet• B2= low fat diet

• DV: 0-50 rating of energy level

More Main Effects and Interactions• A1= morning

• A2= late afternoon

• B1= high fat diet• B2= low fat diet

• DV: 0-50 rating of energy level

Group Activity: Main Effects and InteractionsMake graphs of the

following situations:

Var A Var B AxB interaction

p < .05 n.s. p < .05

p < .05 p < .05 p < .05

n.s. p < .05 n.s.

n.s. n.s. p < .05

Factorial Designs: Naming Conventions

• The first number is the number of levels in first IV, second number is number of levels in second IV, etc.

• 2 x 2• 2 x 3• 2 x 2 x 3• Between-subjects,

repeated measures (within), mixed

A 2 x 3 Interaction

0

2

4

6

8

10

Relaxation Focus Affirmation

Coping Technique

Rat

ing hi stress

lo stress