Page 1 of 6 Exam 2 PS 306, Spring 2008 1. As you know, when designing a repeated measures experiment, or an experiment with a repeated measures factor, one must counterbalance. Why? In other words, if one did not counterbalance, why would the experiment be confounded? Be very explicit! An example might help. [5 pts.] If one did not counterbalance, then any position/order or carry-over effects would fall unequally on the conditions. Imagine, for example, a fatigue effect (decrease in performance over time). If all subjects were to first get Condition A, followed by Condition B, followed by Condition C, then it might appear that Condition C led to poorer performance than the other two conditions. At the very least, the fatigue effects would be another way to explain any results you obtain, which is a confound. Thus, complete counterbalancing (in this case) would ensure that the fatigue effects would fall equally on the three conditions (A, B, and C would occur in the first, second, and third positions equally often). 2. Well, of course you expect to tell me about the impact of various designs on the number of participants needed. For this problem, assume that we want to have a minimum of 25 pieces of data in each cell/condition. [10 pts] Design # of participants # of pieces of data A 4x7 completely between (independent groups) design 700 700 A 4x7 completely within (repeated measures) design 28 784 A 4x7 mixed design, with the first factor between (independent groups) and the second factor within (repeated measures) 112 784 A 4x7 mixed design, with the first factor within (repeated measures) and the second factor between (independent groups) 336 1344 A 5x6 mixed design, with the first factor within (repeated measures) and the second factor between (independent groups) 180 (Incomplete) 720 (Complete) 900 (Incomplete) 3600 (Complete) 3. In a study of early ability to detect a fear-relevant stimulus (a snake), LoBue and DeLoache (2008) presented 3-year-old children and adults (Age: 3-year old vs. adult) a series of 3x3 matrices of pictures. The subject’s task was to point out a target by touching one of the nine pictures on a touch-screen (Target: either a snake among eight non-snake distractors or a non-snake animal, such as a caterpillar, among eight snake distractors). Thus, we can think of this study as a 2x2 independent groups design. Below is a partially completed source table that is consistent with their results (Experiment 3). Complete the source table and interpret the results as completely as you can. Be sure to talk about the results as you might in a Discussion section (i.e., how would you make sense of these results). [15 pts]Page 2 of 6 Dependent Variable:Response Latency Source Type III Sum of Squares df Mean Square F Sig. Partial Eta Squared Observed Powerb Age 72.030 1 72.03 102.90 .000 .700 1.000 Target 10.830 1 10.83 15.47 .000 .259 .970 Age * Target 3.630 1 3.63 5.19 .028 .105 .604 Error 30.920 44 0.70 Corrected Total 117.410 47 € HSD = 3.76.7012= .91 There was a significant main effect for Age, F(1,44) = 102.9, MSE = .7, p < .001, η2 = .7. There was a significant main effect of target, F(1,44) = 15.47, p < .001, η2 = .259. There was also a significant interaction between Age and Target, F(1,44) = 5.19, p < .028, η2 = .105. Post hoc analyses using Tukey’s HSD indicate that the interaction was due to the fact that 3-year-old children were faster to pick the snake from among the nonsnake stimuli (M = 3.5) compared to picking the nonsnake stimuli from among snake stimuli (M = 5.0). On the other hand, adults responded to the snake (M = 1.6) and nonsnake stimuli (M = 2.0) equally quickly. It’s interesting that even very young children seem to treat snake stimuli differently from other animal stimuli. It’s not clear if their responses were slowed down by the number of distractor snakes present in the display (when picking out the nonsnake stimulus) or if they were quick to pick out a snake from among other stimuli. Nonetheless, these young children respond faster to the displays with a snake target among nonsnake distractors compared to displays with a nonsnake target among snake distractors. Adults have generally faster response times, but the difference in their speed to respond to the snake and nonsnake targets is not statistically significant.Page 3 of 6 4. One area of psychology looks at factors that influence decision-making. One factor that people have studied is how a decision is influenced by the way in which the information is delivered. Even though the information is identical, people’s decisions will differ when the information is placed in a different context (frame). Suppose that a researcher was interested in looking at the impact of four different frames on people’s willingness to engage in risky behavior (or to be more protective). One scenario involves the participant’s willingness to smoke cigarettes. The four frames are: NF = No Frame (so it just asks the participant to imagine that he or she has been smoking for a while and enjoys doing so), AF = Analytical Frame (with statistical information about the scenario, such as how many people die of lung cancer each year), EF = Experiential Frame (which attempts to make the scenario personally relevant by asking the participant to think about a family member dying from lung cancer), and AEF = Analytical + Experiential Frames (which puts the two types of information together). Participants read a series of scenarios and then gave a response that indicated their willingness to engage in risky behavior. The dependent variable is called Protect-Risk, where a positive score indicates a more protective response and a negative score represents a willingness to engage in riskier behavior. Suppose that the researcher is also interested in looking at the impact of age (Young 18-23, Middle 38-43, and Older 58-63). Complete the source table below and interpret the results as completely as you can. Finally, discuss the results as you might in a Discussion section. [15 pts] Tests of Between-Subjects Effects Dependent Variable:Protect-Risk Source Type III Sum of Squares df Mean Square F Sig. Partial Eta Squared Noncent. Parameter Observed Powerb Frame 46.017 3 15.339 57.449 .000 .782 172.346 1.000 Age 67.720 2 33.860 126.816 .000 .841 253.634 1.000 Frame * Age 2.352 6 .392 1.468 .210 .155 8.798 .515 Error 12.816 48 .267 Corrected