Exam 1 - 1 of 7 ID# Exam 1 PS 306, Spring 2012 Read each question carefully and provide a clear and complete response, showing your work where appropriate. You must be as accurate as possible, but you need not respond in complete sentences. As always, the Skidmore Honor Code is in effect and you’ll be asked to write out the honor code statement at the end of the exam. The point value for each question is intended as a cue to the amount of time that I think you should need to complete your answer to that question (1 pt = 1 minute). Please limit yourself to correct answers (see below). J Good luck! 1. The repeated measure design has advantages over the independent groups design. First, articulate those advantages. (Examples would help.) In spite of those advantages, there are circumstances under which an independent groups design makes more sense. What are those circumstances? Finally, carryover effects and order effects are problematic for repeated measures designs, but not for independent groups designs. Articulate why that’s the case and then discuss the antidote to carryover and order effects. [10 pts]Exam 1 - 2 of 7 2. Nicholas Rule and his colleagues published an article (2011) titled “Mating interest improves women’s accuracy in judging male sexual orientation.” Their abstract reads as follows: People can accurately infer others’ traits and group memberships across several domains. We examined heterosexual women’s accuracy in judging male sexual orientation across the fertility cycle (Study 1) and found that women’s accuracy was significantly greater the nearer they were to peak ovulation. In contrast, women’s accuracy was not related to their fertility when they judged the sexual orientations of other women (Study 2). Increased sexual interest brought about by the increased likelihood of conception near ovulation may therefore influence women’s sensitivity to male sexual orientation. To test this hypothesis, we manipulated women’s interest in mating using an unobtrusive priming task (Study 3). Women primed with romantic thoughts showed significantly greater accuracy in their categorizations of male sexual orientation (but not female sexual orientation) compared with women who were not primed. The accuracy of judgments of male sexual orientation therefore appears to be influenced by both natural variations in female perceivers’ fertility and experimentally manipulated cognitive frames. a. In one study, they generated an accuracy measure from signal detection (A’, which is similar to d’, a measure that some of you know) for the women’s judgments of male faces (40 homosexual and 40 heterosexual). (That is, the better able they were to correctly label the men as heterosexual or homosexual, the higher their accuracy score.) They then correlated that accuracy judgment with there the women were in their menstrual cycles. The results from a linear analysis of the data are seen below. To the right is a figure from their paper. They report a significant curvilinear relationship between the two variables, but what would you report from a linear analysis of their data? Are their results consistent with their hypothesis that women who are ovulating will be more accurate in their assessment of a man’s sexual orientation? [5 pts]Exam 1 - 3 of 7 b. As we’ve discussed, one way to approach such a curvilinear relationship is to find a way to transform the data into a linear relationship. In this case, one might simply look at the days from ovulation (i.e., from 0, ovulating, to days away from ovulation—ignoring whether women were before or after ovulation). That is, I removed the sign from the days from ovulation. Doing so results in the output seen below: Interpret these results as completely as you can. If a woman were .4 from ovulation, what would you predict her accuracy score to be? If a woman were .9 from ovulation, what you predict her accuracy score to be? What proportion of the variance of accuracy scores is shared with absolute value of ovulation times? [10 pts]Exam 1 - 4 of 7 c. In one study from their paper, they attempted to manipulate the women’s romantic feeling by having half the subjects read a romantic story (the mating-prime condition) before judging the sexual orientation of the same male faces. (And, yes, I’m attempting to track down the story for purely scholarly reasons!) Thus, there was a control condition (no story) and the mating-prime condition. (These women were different from those in earlier studies). The dependent variable was again an accuracy score (A’). Complete the analyses below and interpret the results as completely as you can. [10 pts] ANOVA accuracy Sum of Squares df Mean Square F Sig. Between Groups .054 .009 Within Groups .275 Total .329Exam 1 - 5 of 7 3. Define what we mean by the term operational definition and then cite a specific example of an operational definition from both Milgram's shock study and Harlow's infant monkey study. [5 pts] 4. What is a ceiling effect? How would you know if you were dealing with a ceiling effect in a study? What would be the likely impact of a ceiling effect on MSBetween (MSTreatment) and MSWithin (MSError)? [5 pts] 5. Power is a very important consideration in any experimental design. Describe at least three specific ways in which one might increase the power of a study. [5 pts]Exam 1 - 6 of 7 6. External validity is important in some circumstances, but Mook tries to argue that it’s not particularly important in a lot of psychological research. First of all, define external validity. Then, using at least two of the studies cited in the article, explain circumstances in which Mook argues that external validity is not important (and why he thinks that way). Finally, using a single study that Mook discusses, indicate why a manipulation check would have been appropriate (or why a manipulation check would not have been needed). [10 pts]Exam 1 - 7 of 7 7. Dr. Richard Noggin is interested in the effect of different types of persuasive messages on a person’s willingness to engage in socially conscious behaviors. To that end, he asks his participants to listen to each of four different types of messages (Fear Invoking, Appeal to Conscience, Guilt, and Information Laden). After listening to each message, the participant rates how effective the message was on a scale of 1-7 (1 = very ineffective and 7