In: Statistics and Probability
In a clinical trial, 2222 out of 867867 patients taking a prescription drug daily complained of flulike symptoms. Suppose that it is known that 2.12.1% of patients taking competing drugs complain of flulike symptoms. Is there sufficient evidence to conclude that more than 2.12.1% of this drug's users experience flulike symptoms as a side effect at the alpha equals 0.1α=0.1 level of significance? Find the p-value
Answer:
Claim: To check whether the proportion of drug's users experience flulike symptoms as a side effect is more than 2.1% or not.
The Hypothesis is
H0: P = 0.021
H1: P > 0.021
Now,
n = number of patients = 867
x = number of patients taking a prescription drug daily complained of flulike symptoms=22
Now, we estimate the proportion p as
p̂ = x / n = 22 / 867 = 0.025
Test statistic:
= (0.025 – 0.021) / sqrt[0.021 * (1-0.021) / 867]
= 0.82
Now we find the P-value
= level of significance= 0.01
P-Value = P(Z > 0.82) this is right one tailed test
= 1- P( Z < 0.82)
= 0.4122
= 0.4122
Decision:
Here P-value > 0.01
That is here we fail to reject Ho (Null Hypothesis)
Conclusion:
That is there is not sufficient evidence that the proportion of drug's users experience flulike symptoms as a side effect is more than 2.1%