Effect of Scopolamine on Human Memory:
A Completely Randomized Three Treamtent Design (N = 28)
Scopolamine is a sedative used to induce sleep in patients. Behavioral Neuroscience (2004) researchers examined scopolamine's effects on memory for word-pair associates (see BB for a pdf of the study). For each word pair there was a second word pair with the same first word but a different second word (e.g., robber-jail and robber-police). Subjects were randomly divided into three treatment conditions. Group 1 were administered an injection of Scopolamine, Group 2 subjects were given an injection of Glycopyrrolate (an active Placebo), and Group 3 subjects were not given any drug (No Drug). Four hours latter subjects were shown 12 word pairs from the list and tested on how many of the associated word pairs they could recall. The number of words correctly recalled was recorded. The researchers predicted that Scopolamine treatment would result in poor performance and that the two control groups would not differ from each other.
Use SPSS to answer the following questions.
No inferential statistical test methods (No ANOVA or t-tests) are requested for this assignment. Warring, if any of these statistical tests are conducted they will be evaluated/graded carefully, and any errors or omissions from a complete presentation of the topic, will receive major point reductions. In short, I strongly discourage attempting t-tests, comparisons, or ANOVA at this stage of the course.
Please record the multiple choice answers on scantron form 882 and the other answers on 8 ½ x 11 paper. Please write or print clearly (typed strongly preferred), show all work, organize your written work so that it can be graded efficiently. Organize your paper so that it is easy to grade, e.g., highlight the most important or focal points. Do not staple, tape, etc. the scantron to the remainder of the test. Do not turn in a scantron with erasure marks, as they are normally miss-scored. Failure to follow the aforementioned may result in a loss of credit.
a. See the syllabusfor a description of the scoring method and see BB for examples of excellent work on previous tests. In general, the more you write and the more well organized your annotations and results-summary, the higher will be your grade.
1. Briefly explain why this is a completely randomized design.
2. Give three (general or conceptual) reasons for the unequal sample sizes (not the same reason with different particulars).
3. Identify the independent and the dependent/response variable.
4. Use SPSS Explore to describe the center, spread, and distribution shape for the recall measure separately for the three treatment conditions. Make sure to interpret the normal probability plot, box-plot, stem-and-leaf display, and descriptive statistics as they relate to the research questions of interest.
5. Calculate the 95% confidence interval for the three population means and generate the by-groups error-bar plot.
6. Use confidence intervals to address the researcher's two major hypotheses or research questions. Did the confidence interval contrast analysis support the researcher's theory?
7. Summarize the Results: Make sure to report all essential statistics and at least one graph that forms the basis for your answer to the substantive research questions. Reminder, this is a summary, so be selective. Make sure to justify the descriptive statistics that you choose to report.
xtra Credit: read the research article and comment on the statistical analysis, especially as they relate to topics that you would like to know more about. Comment on the structure of the results section and the information included in the results section.