Page 1 of 5 ID# Exam 2 PS 306, Spring 2006 As always, the Skidmore Honor Code is in effect, and at the end of the exam I’ll have a sheet on which you will indicate your adherence to the Honor Code. Read each question carefully and answer completely, showing all your work. Good Luck! 1. We’ve been talking about confounds, so first tell me how you would define a confound, including the kind of validity attacked by a confound. Then, as you know, I’ve indicated that a study in which one examines a non-manipulated characteristic of a participant is not a true experiment. One reason for that position is that such a study is potentially confounded. Imagine, if you will, a study that looks at the impact of intelligence on learning a list of non-words (e.g., the CVC stimuli used by Ebbinghaus). Participants fall into one of three intelligence groups (Average Intelligence, Above Average Intelligence, and High Intelligence), though you need not worry about the operational definition of those terms to answer the question. Let’s suppose that the results indicate that people in the High Intelligence group exhibited significantly better memory than those in the Average Intelligence group. Why might you describe such a study as confounded? Be very explicit! [10 pts.]Page 2 of 5 2. In an encoding specificity study, Suzuki and Takahashi (1997) were interested in whether pictures were better remembered if the test pictures were similar to the acquisition pictures in terms of color. Let’s treat their study as a completely independent groups design. That is, participants studied a set of pictures that were either all in color or were all in gray-scale. Then, at test half of the participants who got the pictures in color saw a test set of pictures in which the pictures were all in color (half old and some new). The other half of the participants who got the pictures in color saw a test set of pictures in which the pictures were all in gray-scale (half old and half new). The same procedure was used for participants who saw the original pictures in gray-scale (that is half got color test pictures and half got gray-scale test pictures). Thus, we can conceive of one factor in their study as the four types of Color Mode (CC, CG, GC, GG; C = Color and G = Gray, with the first letter acquisition and the second letter test). In addition, they were interested in the impact of delay. Let’s suppose that half of the participants were tested after a 5-min delay and half of the participants were tested after a 1-week delay, which defines the Delay factor. The DV is the percent of the old test items correctly identified as old. Thus, this is a 2x4 independent groups design as I’ve portrayed it here. Complete the source table below and analyze these data as completely as you can. How would you discuss your results (in a Discussion section)? [20 pts.] 12 83.417 4.078 1.17712 90.417 4.358 1.25812 74.500 2.611 .75412 87.000 3.814 1.10112 72.333 7.215 2.08312 88.167 2.980 .86012 83.667 3.420 .98712 87.250 3.388 .978Count Mean Std. Dev. Std. Err.CC, 1-weekCC, 5-minCG, 1-weekCG, 5-minGC, 1-weekGC, 5-minGG, 1-weekGG, 5-minMeans Table for Percent CorrectEffect: Color Mode * DelayPage 3 of 5 3. In a study of hyperactivity among elementary school boys, nine groups of participants were randomly selected from a school population of ADHD, 7-year-old boys. (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 on the activity levels of the participants. Both the drug Ritalin as well as a behavior modification program served as factors. The drug was varied from a Low dosage, to a Moderate dosage, to a High dosage of Ritalin. The behavior modification program consisted of giving the child ten 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 mod program was varied from using the program Every Day, to the program using the program Every Other Day, to using program Once a Week. After 4 weeks, all the children were evaluated for hyperactivity and were assigned scale scores ranging from a possible low of 0 (no indication of hyperactivity) to a high of 40 (extreme hyperactivity). Complete the source table below and interpret the data from this study as completely as you can. [20 points] 3 11.333 1.528 .8823 15.333 .577 .3333 18.667 .577 .3333 28.667 1.155 .6673 34.000 1.000 .5773 38.333 .577 .3333 20.333 .577 .3333 24.333 .577 .3333 27.667 .577 .333Count Mean Std. Dev. Std. Err.High, Every DayHigh, Every Other DayHigh, Once a WeekLow, Every DayLow, Every Other DayLow, Once a WeekMedium, Every DayMedium, Every Other DayMedium, Once a WeekMeans Table for HyperactivityEffect: Dosage * ProgramPage 4 of 5 4. In independent groups ANOVAs, such as the one in Problem #3: [10 pts.] a. If you were not provided the MSError (.741 in this case), you would still have been able to complete the source table, how could you have arrived at MSError? b. What is the function of MSError in the analysis? That is, what population parameter is it intended to estimate? c. Suppose that you re-computed the ANOVA as a single-factor analysis on the Dosage factor. What would that source table look like? SOURCE df SS MS FPage 5 of 5 5. 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 35 pieces of data in each cell/condition. [10 pts] Design # of participants # of pieces of data A 3x6 completely between (independent groups) design A 3x7 completely within (repeated measures) design A 3x7 mixed design, with the first factor between (independent groups) and the second factor within (repeated measures) A 3x7 mixed design, with the first factor within (repeated measures) and the second factor between (independent groups) A 4x8 mixed design, with the first factor between (independent groups) and the second factor within (repeated measures) 6. Suppose the data in Question 3 (Ritalin and Behavior Modification) had yielded the source table below. Tell me what you’d do next. [5 pts.] 2 27.185 13.593 1.266 .3060 2.531 .2322 19.185 9.593 .893