The second quiz (see dates page) will cover the following material:
- Complete an incomplete ANOVA table and answer questions from it. This is similar to the One-Way ANOVA HW #1.
- Answer questions for an analysis of 1-Way ANOVA data that will require you to reference R output that will be provided. This will be most similar to One-Way ANOVA HW #4, but will have some parts of One-Way ANOVA HW #3. Here is an example from a previous year’s exam. Note that this is an example only; the actual question will be different than this one.
- Fully answer short-answer questions from among the following list.
- Completely compare and contrast the meanings of MSWithin, MSTotal, and MSAmong. Consider all possible meanings for these terms.
- Completely compare and contrast the concepts of a “full” and a “simple” model.
- Describe when a Tukey’s HSD and when a Dunnet’s procedure should be used. Describe why the Dunnet’s method is “better” than Tukey’s method if Tukey’s method was applied to the situation where the Dunnet’s method should be used.
- Describe why performing multiple individual 2-Sample t-Tests to determine which specific pairs of means differ is problematic.
- Define experiment-wise and comparison-wise error rates. Identify the specific and relative sizes of each error rate and specifically comment on what happens to each as the number of comparisons made increases.
- List the assumptions of a one-way ANOVA and thoroughly describe how you would test each one (describe hypotheses and tests for each as appropriate).
- Mathematically show that the difference in two means of a log-transformed variable becomes a RATIO of two means on the original scale. You do not need complete sentences for this question but you should mathematically show each step in the proof.
The quiz is closed book and closed notes, you will not need to use R (but will need to be able to interpret results provided from R), you should bring a calculator and a pencil (exams written in pen will not be accepted), and be prepared to write neatly.
Please let me know if you have any questions. Thanks.