1. Imagine you are about to conduct an experiment investigating reaction times as a function of stimulus complexity. You are concerned that your participants may show great differences in reaction times even when the participants are treated identically. These differences may obscure any effect of stimulus complexity. You have reason to believe that participants of the same sex and approximate age will be more similar to one another than randomly selected participants, so you decide to match on these two variables.
Following is a list of participants, together with their sexes and ages. Form matched pairs of participants based on these variables. Then randomly assign the first member of each pair to either Group A or Group B. Assign the remaining member of each pair to the opposite group. Show each step of your work (matching and group assignment). (4 points)
Participant Sex Age
1 Male 21
2 Male 19
3 Female 31
4 Female 25
5 Male 34
6 Female 22
7 Male 25
8 Female 35
9 Male 27
10 Male 22
11 Female 20
12 Male 42
13 Female 23
14 Female 18
15 Male 24
16 Female 46
17 Male 18
18 Male 29
19 Female 24
20 Female 32

2. Counterbalancing
You want to conduct an experiment investigating the detection rate for stray aircraft on a radar screen as a function of the number of aircraft already present. There will be four levels of aircraft present (4, 8, 16, and 32), and you suspect that participants will differ widely in their basic abilities to detect stimuli in the complex environment represented by the display. Show how you would arrange the order of treatments for each of 12 participants. How many participants would be required for an equivalent between-subjects design? In addition to the economy of participants, why would the within-subjects design be preferable? (4 points)

3. The literature has shown that children diagnosed with Fetal Alcohol Syndrome (FAS) are more active and impulsive than children not receiving this diagnosis. They also seem to have a more difficult time controlling themselves (i.e., self-restraint). These problems typically become less severe as the child ages.
An experiment is designed to study the effect of maternal alcohol consumption on a measure of impulsivity in children and adults. The mean values (of a dimensionless measure of impulsivity that is larger when there is more impulsivity) in the following table resulted from analysis of this 2X2 factorial experiment (It’s a fake experiment I made up. Don’t worry – no one gave pregnant women vodka for the sake of research!)
Age (Factor B)
Child Adult
Maternal diet (Factor A) Chocolate milk 5 20
Vodka 35 21

Plot these results and interpret them. (2 points).