Psychology 372 Final Exam.

 

Due: 12:00  Wednesday

 

  1. Examine the procedures for selecting and evaluating published and unpublished studies in meta-analytic studies such as Del Vecchio & O'Leary (2004) and McLeod & Weisz (2004) and present a list of qualities that are considered elements of strong research design. (1- 2 pages)
  2. Imagine that you were designing a study to look at the difference that a history of alcohol use makes in the development of hostility toward family members in old age.  You want to study a group of elderly heavy drinkers and elderly non-drinkers.  Provide a list of variables that you believe you should match your groups on.  Provide a list of variables you believe should be statistically controlled for. (0.5 -1 page)
  3. Why is reporting "effect size" considered superior to just reporting a statistically significant difference between groups? (0.5-1 page)
  4. In a recent article advocating using antidepressant medication for children even though randomized clinical trials have shown no difference between placebo and control groups, Thomas Kramer, MD wrote the following: 

There are a number of reasons why placebo-controlled trials of antidepressants for children often have trouble separating the responses of the drug group from the placebo group. Subjects participating in clinical drug trials get a lot of attention. They come in for frequent visits and talk about depression often. This talking about depression can get them thinking about depression and can be, in effect, de facto cognitive therapy. When you consider the fact that children are considerably more impressionable than adults, it may explain why medications that are in common clinical use in the treatment of depression in children may not look so great in a clinical study. If you have concerns about the efficacy of newer-generation antidepressants in the treatment of depression in children, talk to child psychiatrists who use them. The enthusiasm for these medications among the practitioners who pharmacologically treat depression in children is quite strong. If they didn't work that well, these are the people who would know.

 

Do you agree or disagree with Dr. Kramer's reasoning. Why or why not? (1-2 pages)

 

 

 

 

 

  1. Using the data available at http://www.wooster.edu/psychology/method/examdat.sav  answer the following questions.

 

Do men drink more than women?

Do fraternity/sorority members drink more than non-members?

Is the fraternity/sorority "effect" greater for men than it is for women?

 

Are the differences significant?