Chris was interested in comparing the effects of two sunscreens (i.e. Ointment A

Question

Chris was interested in comparing the effects of two sunscreens (i.e. Ointment A and B) on reducing the degree of sunburn for 12~15 year-old boys. He decided to ask each boy to apply Ointment A to his left arm and Ointment B to his right arm. After applying ointments, each boy was asked to stay under the sun for an hour and was then measured the degree of burn for both arms on a 10-point scale with 10 being the worst burn and 1 being no burn at all. He randomly recruited 25 boys.

Any potential flaws based on the way Chris designed this study?

a. Nothing is wrong

b. He didn’t randomly assign Ointment A to either left or right arm.

c. He shouldn’t randomly recruit boys.

part 2) Are the data paired?

a)True

b) False

part 3) Ignore the flaw you identify in Q9. What parametric test would you use to compare the effects of two ointments?

a)Two-sample t test with unequal variances

b)Wilcoxon signed rank test

c)Paired t test

d)Two-sample Z test

Explanation / Answer

The randomization condition in any experiment arises in many ways. The pairs may be random samples, order of the two treatments may be randomly assigned, or the treatments may be randomly assigned to one member of each pair. Therefore, Chris might have assigned Ointment A to either left or right arm, so the randomization condition has fulfilled. Ans>b

2) There are same number of observations. Samples were asked to put Ointment A in left hand and Ointment B in right hand, therefore, the data are paired. Ans>a

3) The paired data condition, randomization condition and nearly normal condition being fulfilled, use paired t test.

Ans>c

Related Questions

Navigate

• Browse (All)
• Browse C
• Subjects

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.