Reminder:

DO NOT hand in unedited SAS output/code.

ONLY include the results required

Answer questions IN ORDER; Include question number; Label graphs

Include SAS input for questions at the end of the entire homework as Appendix

SAVE your SAS code, since you may be asked to continue problems on successive homework

*Reading*: Montgomery Chapter 3

1. You want to compare 5 treatments using a one-way fixed effects model. You want enough

replication so that the F test has at least 80% power (at the 5% significance level) when

the largest difference between means is 4 units. Suppose that 2 = 9. How large must

n be (assuming balanced experiment, i.e., ni = n)? Make sure you show your software

inputs or hand calculations to receive full credits.

2. Recall Exercise 5 in Lecture 9. There were a = 6 randomly chosen schools each with

n = 5 teachers.

a. How much power does this study have if the true variances were such that 22=2? Make sure you show your software inputs or hand calculations to receive full

credits.

b. In a random effects situation you can either increase a and/or n to increase the power.

You investigate different combinations and find the following:

a = 11, n = 5, 81:6% power

a = 9, n = 6, 82:2% power

a = 8, n = 7, 83:7% power

a = 7, n = 8, 83:6% power

If it costs $20 in time and resources to evaluate each teacher once at a school and $100 in

time and resources to access a school, which of these options would you choose? Explain your

answer.

### Like this:

Like Loading...

*Related*