Chapter 10 - Complex Designs

Question 1

complex (factorial) designs

Answer 1

• 2 or more factors (IVs)

- can be 2 manipulated factors, 1 manipulated factor and 1 non-manipulated factor, etc.

Question 2

interaction

Answer 2

• When effects of the first IV/factor on the DV change depending on the level of the second IV/factor
• Second IV = moderator or moderator variable
• Ex. Training dogs to sit: verbal command condition: tells half to sit, tells half nothing. Food condition: holds treat for half, nothing for other half. The effect of command on proportion of dogs sitting and is different depending on whether food is present or not.

Question 3

3 questions to ask when interpreting a graph

Answer 3

• Are the lines parallel?
• Are the midpoints of the lines different?
• Are the average values of the DV at each level of the IV on the x-axis different?

Question 4

Interpreting graphs: Are the lines parallel?

Answer 4

• if lines are parallel, there’s NO interaction
• WHY? Because the impact of one IV is consistent across all levels of another IV (simple main effects are the same)
• If the two lines intersect or look like they could if they were extended, there’s an interaction

Question 5

Interpreting graphs: Are the midpoints of the lines different?

Answer 5

• If so, there is a main effect of the moderator variable (variable B/2nd IV)
• WHY? The average of one condition is different from the average of the other condition -> marginal means of moderator variable.
• If they’re the same/overlap – no main effect of moderator variable

Question 6

Interpreting graphs: Are the values of the DV at each level of the IV on the x-axis different?

Answer 6

If they’re different values, there is a main effect of variable A

Question 7

Simple main effect

Answer 7

• effect of one IV on the DV within a SINGLE LEVEL of another IV
• ex. More dogs will sit when you tell them to sit rather than not telling them to sit – If you hold food in your hand. If you don’t hold food in your hand, telling dogs to sit vs. Not telling them to sit makes little difference
• you have to decide on one way to split the data

Question 8

participant variable

Answer 8

things you can’t manipulate (ex. ethnicity, gender, sexual orientation, age, intelligence, relationship status, etc.)

Question 9

2-way factorial design

Answer 9

• when there are two factors and each factor has two sub-factors
• aka: 2-by-2 factorial design
• ex. Study hypothesized violent video games resulted in more aggression than playing non-violent video games. Half played violent game, half played non-violent games, Half were men, half were women
• 2 factors: video game & gender. Each factor has 2 more factors = violent vs. Nonviolent; male vs. Female

Question 10

cross-factorial designs

Answer 10

• researchers study/cross all possible conditions
• Ex. Studying women and men playing violent and non-violent video game -> you can get both men and women to play violent and non-violent games

Question 11

nested factorial designs

Answer 11

• alternative to crossed factorial designs
• 1 IV is nested within the other IV, preventing full crossing
• Ex. School district 1 and school district 2 -> then look at specific schools within. District = overarching IV, schools are nested -> you can’t put one school from District 1 into District 2

Question 12

IV x PV design

Answer 12

• PV = participant variable
• allows researchers to investigate how different types of people respond to the same manipulated variable (ex. how psychopaths and non-psychopaths respond when asked to tell a lie vs. the truth)

Question 13

Main effect

Answer 13

• effect of 1 variable on the DV -> when one variable influences the DV more than the other
• difference between 2 marginal means

Question 14

Marginal means

Answer 14

• the average of main effect values

- if main effects are the same, there’s no marginal means

Question 15

mixed factorial design

Answer 15

combines both repeated measures and independent groups designs