The survival rate of patients with a tumor using an existing medication is known to be 40%. A pharmaceutical company claims that the survival rate of a new drug is higher. The new drug is given to 20 patients to test for this claim. Let be the number of cures out of the 20 patients. Suppose the rejection region is
Compute the type of error that can occur when the survival rate is 35% (round off to second decimal place).

Question

The survival rate of patients with a tumor using an existing medication is known to be 40%. A pharmaceutical company claims that the survival rate of a new drug is higher. The new drug is given to 20 patients to test for this claim. Let  be the number of cures out of the 20 patients. Suppose the rejection region is  
Compute the type of error that can occur when the survival rate is 35% (round off to second decimal place).

BlackTom AI · Accepted Answer

We are given a binomial setup. Let X be the number of cures out of 20 patients when the true survival rate is 0.35, so X ~ Binomial(n=20, p=0.35).

The rejection region is defined as X ≥ 12. To understand the error in the specified scenario (survival rate 35%), we examine the probability of observing X in the rejection region under p = 0.35:

- Compute P(X ≥ 12) for X ~ Binomial(20, 0.35).
  - A direct calculation uses the binomial tail: P(X ≥ 12) = sum_{k=12}^{20} C(20, k) (0.35)^k (0.65)^{20-k}.
  - This tail probability is small because 12 or more successes is far above the mean np = 7.

A practical approximation uses a normal approximation with continuity correction: for X ~ Binomial(20, 0.35), mean μ = np = 7 and variance σ^2 = np(1-p) = 20 × 0.35 × 0.65 ≈ 4.55, so σ ≈ 2.13. The standardized z-value for x = 11.5 is z ≈ (11.5 − 7) / 2.13 ≈ 2.11. The upper-tail probability beyond z ≈ 2.11 is about 0.0175, which rounds to approximately 0.02.

Therefore, P(X ≥ 12) under p = 0.35 is roughly 0.02. In plain terms, the chance of observing 12 or more cures when the true rate is 0.35 is about 2%. This is the likelihood of the test statistic landing in the rejection region given the true rate is 0.35.

Note: Depending on the framing, this probability is sometimes referred to as the test's power at p = 0.35 (the probability of rejecting H0 when the true p is 0.35). If one instead considers the Type II error, it would be 1 minus this value, i.e., approximately 0.98, representing the probability of failing to reject H0 when p = 0.35.

类似问题

A Type II error occurs when:

A Type II error occurs when:

What is the probability of making a Type II Error in a test?

更多留学生实用工具

智能学习助手

风格化写作助手

论文查重助手

文献引用助手

课堂转译助手

课堂笔记助手

Quiz搜索助手

学校历年真题

智能刷题助手

智能匹配练习

希望你的学习变得更简单