Page 1 of 6 Exam 2 PS 306, Fall 2005 1. [Based on Huck & Sandler, Rival Hypotheses] Sailors have long had a reputation for drinking spirited beverages in prodigious amounts. Apparently some members of the U.S. Navy have decided to continue this tradition as evidenced by a recent upsurge on interest in the development of alcohol treatment centers for naval personnel. It is refreshing to find that the Navy is also interested in evaluating the effectiveness of these new programs. One such evaluation is described below. Three researchers used a battery of personality inventories and a measure of anxiety to examine personality changes resulting from entrance into one of the alcohol treatment centers. The analysis of pretest/post-test differences on the 404 alcoholics for whom complete data were available revealed significant positive changes on level of trust, emotional stability, and extroversion. These positive changes were accompanied by significant decreases in both pathology (depression, hysteria) and anxiety. Ratings by the (former) alcoholics’ commanding officers indicated the short-term success rate to be over 80 percent. This result stands in marked contrast to the 45 percent rate of success reported in the decade before the development of special treatment centers. Based on this study, do you now have reason to believe that the U.S. Navy is spending our tax money effectively? [10 pts] There are a number of ways to criticize this study: 1. The two groups are not equivalent. That is, there may have been societal changes, etc., that made excessive drinking less of a societal taboo in the prior decade (which might explain the lower success rate). To create an appropriate control group, you’d need to have a group of equivalent Navy alcoholics (e.g., half of the group of 404) and not treat them in one of the alcohol treatment centers. 2. The key numbers reported were generated in a highly subjective fashion, with individual commanding officers making the assessments. Those officers, of course, were aware of the sailors’ participation in the treatment program, and may have made their assessments more positive because of that fact. 3. For the more objective data (numbers not reported here), none of them actually address the issue of the sailors’ alcoholism. That is, the sailors may become more trusting, more emotionally stable, more extroverted, less anxious, and less pathological—but they may still be alcoholics. 2. Dr. Julie Ard was interested in the effects of music on studying using an encoding specificity paradigm. That is, she was interested in the extent to which the similarity of the study and test situations affected performance. To test her hypotheses, she randomly assigned participants to five acquisition conditions (heavy metal, rock, classical, jazz, and blues). People in these groups studied material while listening to a particular type of music. After a brief delay, half of the people in each condition were tested under identical music (same) and half of the people were tested with no music (different). The dependent variable was the percentage score on the test (100 = perfect performance). Complete the analysis and interpret the results below as completely as possible, paying particular attention to the extent to which the results support the hypothesis. [10 pts.] 4 1874.140 468.535 6.539 .0001 26.156 .9931 295.840 295.840 4.129 .0451 4.129 .5084 39.260 9.815 .137 .9682 .548 .07790 6448.800 71.653DF Sum of Squares Mean Square F-Value P-Value Lambda PowerMusicTest SettingMusic * Test SettingResidualANOVA Table for ScorePage 2 of 6 10 80.700 5.889 1.86210 78.300 9.546 3.01910 82.100 9.122 2.88510 80.500 9.168 2.89910 72.400 7.905 2.50010 67.500 8.168 2.58310 82.800 5.692 1.80010 78.300 7.602 2.40410 76.300 10.078 3.18710 72.500 10.124 3.202Count Mean Std. Dev. Std. Err.Blues, DifferentBlues, SameClassical, DifferentClassical, SameHeavy Metal, DifferentHeavy Metal, SameJazz, DifferentJazz, SameRock, DifferentRock, SameMeans Table for ScoreEffect: Music * Test Setting There is no significant interaction between Music and Test Setting, F(4,90) = .137, MSE = 71.653, p = .968. There are, however, main effects for both Music and Test Setting. Performance in the Different (i.e., No Music at Test) condition was significantly better (M = 78.86) than performance in the Music at Test conditions (M = 75.42), F(1,90) = 4.129, MSE = 71.653, p = .045. The type of music had an effect on performance, F(4,90) = 6.539, MSE = 71.653, p < .001. To determine which specific types of music differed, you would need to compute a post hoc test. ! HSD = 3.9571.65320= 7.48 Thus, performance doesn’t differ under the Blues (M = 79.5), Classical (M = 81.3), and Jazz (M = 80.6) conditions, but all are better than Heavy Metal (M = 70.0). You should note that the results are inconsistent with the encoding specificity hypothesis. That is, Same conditions at acquisition and test should be higher than Different conditions at acquisition and test, but they are not. However, note that the different test conditions involve being tested with no music playing, rather than different music. Thus, you might also note that the study would be strengthened by a control condition where the participants study with no music. It may well be the case that people generally perform better if they are tested with no music distracting them. 3. For the following designs, tell me the number of participants needed and the total number of pieces of data that one would have in the study. [10 pts] Design Total Participants Total Pieces of Data A 3x5 completely independent groups (between) design, so that you achieve a minimum of 40 scores per cell. 600 600 A 3x5 completely repeated measures (within) design, so that you achieve a minimum of 40 scores per cell. 60 900Page 3 of 6 A 4x6 mixed design, with the first factor repeated and the second factor independent groups, so that you achieve a minimum of 40 scores per cell. 288 1152 A 4x6 mixed design, with the first factor independent groups and the second factor repeated measures, so that you achieve a minimum of 40 scores per cell. 168 1008 A 5x6 mixed design, with the first factor independent groups and the second factor repeated measures, so that you achieve a minimum of 40 scores per cell. 210 1260 4. Slaughter, Stone, and Reed (2004) were interested in studying infant perception of bodies and faces. In their abstract, they