This is a test. Do your own work. If you have any questions, e-mail ME.

(7) 1. Eighteen spoofera minimax plants were potted individually and placed on a table in the greenhouse. Two growth treatments and a control were applied, six pots at each treatment level. The response was the plant's biomass after 5 months.

1. What design is this? b) Give Source and d.f. in ANOVA table, using specific names and numbers. c) Describe how you would change this design if you felt that light and temperature might change according to a pot's position on the table.

6. 2. Give an example from your field of a randomized block design. Identify the response, treatment, levels, block, block levels, and experimental units.

4. 3. Give an example from your field (identify response, factor) in which it would be difficult to do an observational study. (State why.) State an advantage and a disadvantage of experimental studies, compared to observational studies.

4. 4. Give two possible objections to transforming the response, e.g. by using a log transformation. Explain the difference between assumptions and hypotheses in statistics, and give an example.

6. 5. Suppose you have an RBD with four treatments and five blocks. a)Give the source and d.f. in the ANOVA table. b) Give two hypotheses you can test using Tukey’s procedure. Identify any symbols you use. c) How many experimental units do you need? d) How will you assign the experimental units to the treatments? Justify your answer.

7. 6. In a factorial, the following was printed out: Source SS d.f. MS

The response was the % of turtle eggs that hatched females.

State a hypothesis about the effect of temperature. Obtain the p-value for this test. Interpret the result using terms from this experiment. State a hypothesis about interaction. Obtain the p-value for this test. (It is significant.) Interpret the result using terms from this experiment. How might this second test change your thinking about the usefulness of the first one?

4. 7. Draw a normal plot (label axes) which suggests that the errors are normally distributed, except that one observation is unusually large. Draw a plot of residuals which suggests that the errors have more variance when the mean response is low. (label axes.)