Factorial design (skip analysis and variance) and t-test for paired observation

Question 1

Again. What is the difference between t-test between unpaired observations and a t-test between paired observations?

Answer 1

1. for unpaired observations
   data of different participants in different conditions (groups)
   example: between-subjects design
2. for paired observations
   data of the same participants in different conditions
   example: within-subjects design

Question 2

What are the three different variants of factorial design?

Answer 2

Complete between-subjects design
Complete within-subjects design
Mixed design

Question 3

What are the advantages of factorial design?

Answer 3

1. Efficiency: one experiment with two factors instead of two experiments with one factor each.
2. Generalization: the effect of alcohol dose applies to cars with and without power steering
3. Interaction: does the effect of alcohol dose depend on power steering?

Question 4

Define the “main effect” in factorial designs!

Answer 4

Effect of a factor after averaging across the levels of all other factors.

e.g. Main effect alcohol dose?
Average across the levels of power steering
Main effect power steering?
Average across levels of alcohol dose.

Question 5

Definition interaction effect A × B

Answer 5

Interaction A x B: the effect of A depends on the level of B.
Otherwise: no interaction effect A × B.
symmetrical: interaction A × B = interaction B × A