Assignment 2: Power

Due Feb. 14(A2&B2) & Feb. 16(C2)

 TO BE COMPLETED INDIVIDUALLY

 

For this assignment, you will use the simulation site (http://www.ruf.rice.edu/~lane/stat_sim/index.html) you used for Assignment 1.  For each of the following questions, use the Simulate 5000 button. 

 

1)      You are conducting a between-subjects test with an n of 18, Population Mean A = 175 and Population Mean B = 151.  The standard deviation is 48. 

 

a.       Calculate the effect size of these data (show your work). (2 marks)

 

b.      What is your Type II error rate (show your work)? (2 marks)

 

2)      You are conducting a between-subjects test with an n of 18, Population Mean A = 175 and Population Mean B = 163.  The standard deviation is 48. 

 

a.       Calculate the effect size of these data (show your work). (2 marks)

 

b.      What is your Type II error rate (show your work)? (2 marks)

 

3)      Compare the results that you obtained for questions 1 and 2.  What is the relationship between effect size and power? (2 marks)

 

4)      You are conducting a between-subjects test with an n of 18, Population Mean A = 6 and Population Mean B = 5.  The standard deviation is 4. 

 

a.       Calculate the effect size of these data (show your work). (2 marks)

 

b.      What is your Type II error rate (show your work)? (2 marks)

 

5)      Now assume that you have an effect size of 1.  How many participants (approx.) would you need to have power of .95 in a between-subjects experiment? (1 mark)

 

6)      Assume that you have an effect size of 0.6.  How many participants (approx.) would you need to have power of .95 in a between-subjects experiment? (1 mark)

 

7)      Assume that you have an effect size of 0.2.  How many participants (approx.) would you need to have power of .95 in a between-subjects experiment? (1 mark)

 

8)      Compare the number of participants you needed in questions 5, 6 and 7.  What does this tell you about the relationship between effect size and the number of participants required in an experiment to have the desired power? (2 marks)

 

9)      What is the relationship between sample size and power? (2 marks)

 

10)  Would you expect the Type II error to be greater in a study with 20 participants or 75 participants (assuming all other things are equal)?  Explain your reasoning. (2 marks)

 

11)  Assume that you have an effect size of 0.75.  How many participants would you need to have power of .95 in a within-subjects experiment with:

 

a.       rho = 0? (1 mark)

b.      rho = 0.4? (1 mark)

c.       rho = 0.8? (1 mark)

 

12)  What is the relationship between rho and power? (2 marks) 

 

13)  In general, would you expect a within-subjects design or a between-subjects design to be more powerful?  Explain your reasoning. (2 marks)

 

14)  Enter the following information into the appropriate spaces on the simulation page: N = 40, Pop Mean A = 915, Pop Mean B = 879, rho = 0, sd = 54.  What is the effect size (show your calculation)?  Now conduct both a between-subject simulation and a within-subject simulation.  Which test is more powerful?  Why? (3 marks)

 

15)   Now imagine that you only have access to a total of 12 participants.  You can do one of two things, use a between-subjects design and randomly assign them to one of two conditions (n= 6) or you can use a within-subjects design.  Enter the following information into the appropriate spaces on the simulation page: Pop Mean A = 915, Pop Mean B = 879, rho = 0, sd = 54.  Conduct both a between-subject simulation and a within-subject simulation.  Which design would be more powerful under these circumstances of limited participation and why?  (3 marks)

 

16)  Your father has high blood pressure and the doctor said that he needs to lose weight because he is a great risk for a heart attack.  You decide to look into diet programs that will help him lose the weight.  You request information from two different weight loss programs.  One program, Julie Craig, sends you information regarding the effectiveness of their program.  You also receive information regarding the effectiveness of a new diet program on the market, the South Bend diet.  Both programs conducted stats comparing the pre-program and post-program weight in the form of a t-test examining weight loss in pounds; they have provided the p values from their respective tests. The information you obtained is listed below. 

 

South Bend Diet                                 Julie Craig

n = 25                                                  n = 4523

sd = 40                                                            sd = 40

effect size = 0.75                                  effect size = 0.10

p = .04                                                 p = .001

 

If your father’s weight was 260 pounds before the program started, what would your father’s final weight be after the implementation of each program, assuming his improvements were typical or average (show your calculations)?  (4 marks)

 

Which program would you select for your father (assume the programs cost the same amount)?  Explain your reasoning. (2 marks)