Chapter 10: Correlated-Groups Designs & Inferential Statistics

Question 1

What are the two types of correlated-groups designs?

Answer 1

The within-subjects design and The matched-subjects design and they are analyzed in exactly the same way

Question 2

What is the difference of a between-subjects and within-subjects design?

Answer 2

Between-subjects design requires more participants, within-subjects is vulnerable to order (practice, fatigue) effects, and within-subjects design is more likely to result in a significant test statistic

Question 3

What are the different ways to order the treatment conditions and why?

Answer 3

Such that any practice/fatigue effects will be evenly distributed (1)randomization or (2)counterbalancing

Question 4

What are the different types of counterbalancing?

Answer 4

Complete counterbalancing and Partial counterbalance

Question 5

What is a complete counterbalance?

Answer 5

all possible orders of treatment conditions are represented equally often

Question 6

What are the two types of partial counterbalance?

Answer 6

(1) a Latin square: having treatment condition appear in each position equally often
(2) a balanced Latin square: having treatment condition appear in each position equally often as well as having each treatment condition precede and follow every other treatment condition equally often

Question 7

What is the statistical test for correlated groups?

Answer 7

Within/matched subjects t test: can be used to determine if the numerical difference between the means of two treatment conditions is “statistically significant”