Page 1 of 8 Exam 3 PS 217, Spring 2005 1. In a classic study, Tulving and Gold (1963) studied word identification under conditions of amounts of relevant and irrelevant context. Let’s conceive of their study as a 2x5 independent groups design, with Context (Relevant vs. Irrelevant) and Number of Words (0, 1, 2, 4, and 8) as the two factors. For the very briefly presented target word performer, for instance, some participants would see no preceding words (0) or the first 1, 2, 4, or 8 words from one of two sentences. The Relevant sentence was “The actress received praise for being an outstanding...” The Irrelevant sentence was “The dog retrieved the burrito from the neighbor’s...” Obviously, the issue was the extent to which the preceding context would help the participant to identify the target word that was presented very briefly. The dependent variable was the time it took the participant to identify the target word. Complete the source table below and analyze the data as completely as you can, providing an interpretation for the obtained results. [15 pts] 4 132.467 33.117 3.375 .0120 13.500 .8411 14344.533 14344.533 1461.920 <.0001 1461.920 1.0004 10560.467 2640.117 269.067 <.0001 1076.267 1.000110 1079.333 9.812DF Sum of Squares Mean Square F-Value P-Value Lambda PowerNumber of WordsContextNumber of Words * ContextResidualANOVA Table for ID Time 12 68.333 3.200 .92412 70.500 3.989 1.15112 73.083 2.314 .66812 67.917 2.778 .80212 80.917 3.147 .90812 55.917 3.679 1.06212 84.333 2.964 .85612 53.333 3.025 .87312 92.500 3.425 .98912 42.167 2.406 .694Count Mean Std. Dev. Std. Err.0, Irrelevant0, Relevant1, Irrelevant1, Relevant2, Irrelevant2, Relevant4, Irrelevant4, Relevant8, Irrelevant8, RelevantMeans Table for ID TimeEffect: Number of Words * Context 405060708090100Cell Mean0 1 2 4 8CellRelevantIrrelevantInteraction Line Plot for ID TimeEffect: Number of Words * Context First, I would test for the homogeneity of the variances in the groups. Using Hartley’s FMax Crit = 8.0, with FMax = 2.97 I would presume that there is no concern about heterogeneity of variance and use α = .05 for the ANOVA. ! FMax=15.915.35= 2.97 There is a significant main effect for number of words, F(4,110) = 3.375, MSE = 9.812, p = .012. There is also a significant main effect for context, F(1,110) = 1461.92, MSE = 9.812, p < .001. There is also a significant interaction, F(4,110) = 269.067, MSE = 9.812, p < .001. To interpret the interaction, I would first compute Tukey’s HSD: ! HSD = 4.589.81212= 4.14 Thus, when there are no words present (so that there is not really any context), people don’t differ in their speed of identifying the target word. However, with one or more wordsPage 2 of 8 of context present, people take significantly longer to identify the target word when the word(s) provided an irrelevant context compared to those shown the word(s) in a relevant context. As seen in the figure, it appears that the more words of context provided, the greater the impact, so that the more irrelevant context you have, the more difficult it is to identify the target word and the more relevant context you have, the easier it is to identify the target word. 2. In a study of hyperactivity among elementary school boys, 63 students were randomly selected from a school population of ADHD, 7-year-old boys are randomly assigned to one of 9 groups (n = 7). (ADHD is Attention Deficits with Hyperactivity, and left untreated, it can prevent a child from attending to incoming learning stimuli and may also create major disruptions in the classroom.) The researcher wanted to study the classroom effects of both the drug Ritalin as well as a behavior modification program on the activity levels of the students. The drug administered at three levels: 5 mg Ritalin, 10 mg Ritalin, and 20 mg Ritalin. The behavior modification program consisted of giving each student 10 tokens to start the day and then taking away a token for each hyperactive infraction. The tokens that were saved could then be exchanged for some valued prize. The behavior modification program was varied across three levels: no program, program every other day, and program every day. After four weeks, all the children were evaluated for hyperactivity and were assigned scaled scores ranging from a possible low of 0 (no indication of hyperactivity) to a high of 40 (extreme hyperactivity). A partially completed StatView output of the data is seen below. Complete the source table and then analyze the data as completely as possible, providing a complete interpretation. Behavior Mod (None, Every other day, Every day). Ritalin (5 mg, 10 mg, 20 mg). [10 pts] 2 1283.143 641.571 26.292 <.0001 52.583 1.0002 657.238 328.619 13.467 <.0001 26.934 .9994 128.762 32.190 1.319 .2746 5.277 .37654 1317.714 24.402DF Sum of Squares Mean Square F-Value P-Value Lambda PowerBehavior ModRitalinBehavior Mod * RitalinResidualANOVA Table for Hyperactivity Score 21 19.524 6.274 1.36921 11.810 4.131 .90121 22.524 6.983 1.524Count Mean Std. Dev. Std. Err.BM Ev Oth DayBM Every DayNo Beh ModMeans Table for Hyperactivity ScoreEffect: Behavior Mod 21 17.762 5.864 1.28021 14.095 5.787 1.26321 22.000 8.283 1.807Count Mean Std. Dev. Std. Err.10 mg20 mg5 mgMeans Table for Hyperactivity ScoreEffect: Ritalin First, I would test for the homogeneity of the variances in the groups. Using Hartley’s FMax Crit = 17.5, with FMax = 4.89 I would presume that there is no concern about heterogeneity of variance and use α = .05 for the ANOVA. ! FMax=54.9511.24= 4.89 There is a significant main effect for behavior modification, F(2,54) = 26.292, MSE = 24.402, p < .001. There is also a significant main effect for Ritalin, F(2,54) = 13.467, MSE = 24.402, p < .001. There was no significant interaction, F(4,54) = 1.319, MSE = 24.402, p = .275. Thus, I would focus on the main effects, which would use the same Tukey’s HSD: ! HSD = 3.4224.40221= 3.69Page 3 of 8 The means for Behavior Modification would be 19.5, 11.8, and 22.5 for Every Other Day, Every Day, and No Beh Mod, respectively. The means for Ritalin would be 17.8, 14.1, and 22 for 10mg, 20 mg, and Placebo, respectively. For Behavior Mod: Administering the Behavior Mod every day leads to less hyperactivity than administering Beh Mod every other day or not using Beh Mod (and those two levels produce statistically similar results). For Ritalin: Dosage 20 leads to significantly less hyperactivity than