In: Statistics and Probability
In studies for a medication, 6% of patients gained weight as a side effect. Suppose 703 patients are randomly selected. Use the normal approximation to the binomial to approximate the probability that:
A. Exactly 43 patients will gain weight as a side effect
B. No more than 43 patients will gain weight as a side effect
C. At least 57 patients will gain weight as a side effect. What does this result suggest?
P(gained weight as a side effect), p = 0.06
q = 1 - p = 0.94
Sample size, n = 703
P(X < A) = P(Z < (A - mean)/standard deviation)
Mean = np
= 703 x 0.06
= 42.18
Standard deviation =
=
= 6.2968
A. P(exactly 43 patients will gain weight as side effect) = P(X = 43)
= P(42.5 < X < 43.5)
= P(X < 43.5) - P(X < 42.5)
= P(Z < (43.5 - 42.18)/6.2968) - P(Z < (42.5 - 42.18)/6.2968)
= P(Z < 0.21) - P(Z < 0.05)
= 0.5832 - 0.5199
= 0.0633
B. P(no more than 43 patients will gain weight as a side effect) = P(X 43)
= P(X < 43.5) {with continuity correction)
= 0.5832
C. P(at least 57 patients will gain weight as a side effect) = P(X 57)
= 1 - P(X < 56.5)
= 1 - P(Z < (56.5 - 42.18)/6.2968)
= 1 - P(Z < 2.27)
= 1 - 0.9884
= 0.0116
Probability that at least 57 patients will gain weight as a side effect is less than 0.05. SO, the event is unusual. We are not likely to get at least 57 patients who will gain weight as a side effect.