Cognitive Science 14
Homework Set 5 Questions

1. An experimenter was interested in dieting and weight losses among men and women. It was believed that in the first 2 weeks of a standard dieting program, women would tend to lose more weight than men. As a check on this notion, a random sample of 15 husband-wife pairs were put on the same strenuous diet. Their weight losses after 2 weeks showed the following:
   
   
   

   Did wives lose significantly more () than husbands? What are we assuming here?

2. An experiment was concerned with the possible effect of a small lesion in a particular area of a hamster's brain on the activity level of the animal. Thirty hamsters were selected at random and randomly divided into two groups of 15 animals each. Group I was used as a control group and operated upon but without making any lesion, whereas Group II was given a small lesion. Because three animals died in Group II, the actual number turned out to be 12 in that group. After full recovery by the remaining hamsters, each was given access to a running wheel, and the number of revolutions per fixed unit of time was recorded. The following data show activity in hundreds of revolutions per unit time. Did the lesion significantly reduce the activity of the hamsters ()?
   
   
3. Find the 95 % confidence limits for the mean difference in wheel revolutions for the two populations of hamsters in the previous exercise. Use the pooled estimate of variance.

4. A study tested whether a new drug had an effect (positive or negative) on short-term memory in Alzheimer patients. They gathered a sample of 6 male and 6 female Alzheimer patients in their seventies from the SD area. Subjects were randomly assigned to either placebo or drug conditions with 3 males and 3 females in each condition. All subjects took a pill early in the morning for one month. The pill contained 150mg of the drug in the experimental group and a neutral substance for the control group. After 1 month of treatment subjects were tested using a standard memory test (the larger the score, the better). The lab assistant tested all male subjects in the morning and female subjects in the afternoon. Table 1 shows the results obtained with the 12 subjects.
   
   

   We can think of the experiment as consisting of four groups: Female Drug as Group 1, Female Placebo as Group 2, Male Drug as Group 3, and Male Placebo as Group 4.

   1. For each possible intervening variable, state how it was controlled or if it were not controlled:
      1. Differences between males and females in memory?
      2. Differences of different individuals in memory?
      3. Differences due to how much drug was taken?
      4. Differences due to the time of day the test was taken?
   2. Compute the sample mean and variance for each group.
   3. Compute the pooled estimate of variance. How many degrees of freedom does it have?
   4. Do we have enough evidence to say the drug has an effect on women?
   5. Do we have enough evidence to say the drug has an effect on men?
   6. Find the 95 % confidence intervals of each of the 4 group means.
   7. Let represent the effect of the drug in females, and let represent the effect in males. Do you have enough evidence to say the drug has an interaction, that is that it has a different effect in males than it does in females?
5. Is there a potential confound that could explain differences between how the males and females performed?