1 / 12

WARM – UP #38

Learn how to conduct a completely randomized experiment to test the effectiveness of a new ointment for treating shingles. Explore the role of double-blind testing and consider gender differences in the experiment design.

mjuan
Download Presentation

WARM – UP #38

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. WARM – UP #38 • 38. A research doctor has discovered a new ointment that she believes will be more effective than the current medication in the treatment of shingles (a painful skin rash). Eight patients have volunteered to participate in the initial trials of the hits ointment. You are the statistician hired as a consultant to help design a completely randomized experiment. • Describe how you will conduct this experiment.. • Suppose the last names of the patients start with the letters A to H. Using the random numbers listed below, show which patients you will assign to each treatment. Explain you randomization procedure clearly. • 41098 18329 78458 31685 55259 • Can this experiment be double-blind? How? • Its is suspected that males react differently to the ointment than females. Redesign your experiment with this extra info in mind.

  2. Describe how you will conduct this experiment..

  3. b) Suppose the last names of the patients start with the letters A to H. Using the random numbers listed below, show which patients you will assign to each treatment. Explain you randomization procedure clearly.

  4. Can this experiment be double-blind? How? • Assuming that the ointments looked alike, it would be possible to blind the experiment for the patient and the evaluating doctor. If both the subject and the evaluator are blinded, the experiment is double-blind. • Its is suspected that males react differently to the ointment than females. Redesign your experiment with this extra info in mind. • Before randomly assigning patients to treatments, identify them as male or female. Having blocks for males and females will eliminate variation in improvement due to gender.

  5. HW: Page 316: #34, 35, 41, 42

  6. HW: Page 316: #34, 35, 41, 42 34. This double-blind experiment has 1 factor (vitamin E), at 2 levels (vitamin E and no vitamin E), resulting in 2 treatments. The response variable measured is the time it takes the patient to recover from the surgery. Randomly select half of the patients who agree to the study to get large doses of vitamin E after surgery. Give the other patients in the study a similar looking placebo pill.

  7. HW: Page 316: #34, 35, 41, 42 35. This experiment has 1 factor (reading program), at 2 levels (phonics and whole language), resulting in 2 treatments. The response variable is reading score on an appropriate reading test after a year in the program. After randomly assigning students to teachers, randomly assign half the reading teachers in the district to use each method. There may be variation in reading score based on school within the district, as well as by grade. Blocking by both school and grade will reduce this variation.

  8. HW: Page 316: #34, 35, 41, 42 This Matched-Pairs experiment has 1 factor (hand), at 2 levels (right, left), resulting in 2 treatments. The response variable is the difference in deactivation time between left and right hand. Find a group of volunteers. Using a matched design, we will require each volunteer to deactivate the machine with his or her left hand, as well as with his or her right hand. Randomly assign the left or right hand to be used first. Randomization will equalize any variability in time that may result from experience gained after deactivating the machine the first time.

  9. HW: Page 316: #34, 35, 41, 42

  10. 42. This experiment has two factors (water temperature, wash cycle). The factor water temperature has 2 levels (cold, hot), and the factor wash cycle has 2 levels (regular, delicates). 2 factors, at 2 levels each, results in 4 treatments (hot regular, hot-delicates, cold-regular, cold-delicates). The response variable is the level of cleaning of the grass stains.

  11. Confounding Variable A factor or (3rd External Variable) that affects the Response variable but is NOT the experimental factor (Explanatory Variable) we are studying is called a Confounding Variable. • These variables exist in Experiments.

  12. Test Review 13 men and 7 women are waiting to be interviewed for jobs. If only 3 candidates will be hired, what is the probability that 2 will be men and 1 will be a women Describe how you would assign digits and perform a 5 run simulation for this event. Explain what a run is and when you stop. Perform the simulation by using 4→rand and then randInt(1, 20, 3) FIVE times. Using the digits 01-20. Assign 01-13 as the selection of a man and 14-20 as that of a woman. A run is the selection of 3 random digits. Perform 5 runs and stop. Note the number of runs that contain 2 men/1women are counted.

More Related