(No Mod) Factorial Designs

Question 1

Define factorial design

Answer 1

Containing more than 2 predictor variables

Question 2

Why is 4 or more predictors difficult to manage in an experiment?

Answer 2

Difficult to analyze and may not be able to get enough participants for each group if it is a between subjects design: may mean less power

Question 3

2 x 2 design (LEFT TO RIGHT)

Answer 3

NUMBER OF PREDICTORS X NUMBER OF LEVELS

2 predictors with 2 levels
EXAMPLE: no therapy vs wilderness camp therapy, nonviolent vs violent criminals

Question 4

2 x 3 design

Answer 4

2 predictors with 3 levels

Question 5

Why can we not conduct a 2x2 factorial design separately?

Answer 5

It is more efficient to do it together
Can look for interaction effects

Question 6

Main Effect

Answer 6

isolating one predictor to see its effect on the outcome

Question 7

Interaction Effect

Answer 7

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

Question 8

Real life interaction applications

Answer 8

usually when the statement “this is true, ESPECIALLY IN THIS CASE”
“especially among teenagers”
“especially among elderly”

Question 9

Mixed Factorial Design

Answer 9

Contains a between subjects variable and a within subjects variable

Question 10

Quasi experimental

Answer 10

does not use random assignment because the feature is already preexisting (can’t randomly assign people to be alcoholics or not)

Question 11

How to tell whether there is an interaction effect on a bar graph

Answer 11

To identify an interaction in a bar graph, look for a “difference in differences.” First, look for the difference between high- and normal expectations for 7-year- olds: a difference of about 15 points. Now look at the difference between high and normal expectations for 15-year-olds: a difference of 0 points. The difference of 15 is different from the difference of 0. The differences are different. This is the mark of an interaction: the effect of one IV (teacher expectations) is different across different levels of another IV (age).

Question 12

How to tell a main effect on a bar graph

Answer 12

Use the same logic as for the line graphs: are the black bars, in general, higher or lower than the lighter bars? Yes, the black bars are higher. Are the two bars for 7- year-olds higher or lower than the two bars for 15-year-olds? It looks like the bars for 15-year- olds might be slightly higher.