Skip to content
CalcGospel 國際數學圖譜
返回

IAL 2021 Jan S3 Q4

A Level / Edexcel / S3

IAL 2021 Jan Paper · Question 4

题目

Problem

The scores in a national test of seven-year-old children are normally distributed with a standard deviation of 18 A random sample of 25 seven-year-old children from town A had a mean score of 52.4

(a) Calculate a 98% confidence interval for the mean score of the seven-year-old children from town A.

(4)

An independent random sample of 30 seven-year-old children from town B had a mean score of 57.8 A local newspaper claimed that the mean score of seven-year-old children from town B was greater than the mean score of seven-year-old children from town A.

(b) Stating your hypotheses clearly, use a 5% significance level to test the newspaper’s claim. You should show your working clearly.

(6)

The mean score for the national test of seven-year-old children is μ\mu.

Considering the two samples of seven-year-old children separately, at the 5% level of significance, there is insufficient evidence that the mean score for town A is less than μ\mu, and insufficient evidence that the mean score for town B is less than μ\mu.

(c) Find the largest possible value for μ\mu.

(4)
(Total 14 marks)
题目中文翻译

七岁儿童国家测试的分数服从正态分布,标准差为 18。 来自 A 镇的 25 名七岁儿童随机样本的平均分为 52.4。

(a) 求 A 镇七岁儿童平均分的 98% 置信区间。

来自 B 镇的 30 名七岁儿童独立随机样本的平均分为 57.8。 当地一家报纸声称,B 镇七岁儿童的平均分高于 A 镇七岁儿童的平均分。

(b) 清楚写出假设,在 5% 显著性水平下检验这家报纸的说法。你应清楚写出工作过程。

七岁儿童国家测试的总体均分为 μ\mu

单独考虑这两个样本,在 5% 显著性水平下,证据不足以表明 A 镇的平均分小于 μ\mu,也证据不足以表明 B 镇的平均分小于 μ\mu

(c) 求 μ\mu 的最大可能值。

解答

(a)

解法一

思路

展开

总体标准差已知,且分数服从正态分布,因此用标准正态分布构造均值的置信区间。98%98\% 双侧置信区间的两端各留 1%1\%,所以临界值为 z0.99=2.3263z_{0.99}=2.3263\ldots

答题过程

展开

For a 98%98\% confidence interval,

z0.99=2.3263z_{0.99}=2.3263\ldots

The standard error of the sample mean is

σn=1825=3.6.\frac{\sigma}{\sqrt{n}}=\frac{18}{\sqrt{25}}=3.6.

Therefore, the confidence interval is

xˉ±z0.99σn=52.4±2.3263(3.6)=(44.025,60.775).\begin{align*} &\,\bar{x}\pm z_{0.99}\frac{\sigma}{\sqrt{n}}\\ =&\,52.4\pm 2.3263\ldots(3.6)\\ =&\,(44.025\ldots,60.775\ldots). \end{align*}

Hence, to 3 significant figures, the 98%98\% confidence interval is

(44.0,60.8).\boxed{(44.0,60.8)}.

(b)

解法一

思路

展开

报纸声称 B 镇均分较高,因此进行右尾检验。两个样本相互独立且总体标准差同为 1818,所以样本均值之差的方差是两个样本均值方差之和。先算标准化检验统计量,再与 5%5\% 右尾临界值比较。

答题过程

展开

Let μA\mu_A and μB\mu_B be the population mean scores for towns A and B respectively.

H0: μB=μA,H1: μB>μA.\begin{align*} H_0:&\ \mu_B=\mu_A,\\ H_1:&\ \mu_B>\mu_A. \end{align*}

Under H0H_0, the standard error of XˉBXˉA\bar X_B-\bar X_A is

SE=18130+125=4.8744\begin{align*} \operatorname{SE}=&\,18\sqrt{\frac{1}{30}+\frac{1}{25}}\\ =&\,4.8744\ldots \end{align*}

The test statistic is therefore

z=57.852.44.8744=1.1078\begin{align*} z=&\,\frac{57.8-52.4}{4.8744\ldots}\\ =&\,1.1078\ldots \end{align*}

For a 5%5\% one-tailed test, the critical value is

z0.95=1.6449.z_{0.95}=1.6449.

Since

1.1078<1.6449,1.1078<1.6449,

the result is not significant, so H0H_0 is not rejected. There is insufficient evidence to support the newspaper’s claim that the mean score in town B is greater than the mean score in town A.

解法二

思路

展开

官方评分资料也接受用 p-value 作判断。沿用相同的右尾检验统计量 z=1.1078z=1.1078\ldots,求其右尾概率;若 p-value 大于显著性水平 0.050.05,便不拒绝原假设。

答题过程

展开

Using the same hypotheses and test statistic,

z=1.1078z=1.1078\ldots

The one-tailed p-value is

p=P(Z>1.1078)=0.13396\begin{align*} p=&\,P(Z>1.1078\ldots)\\ =&\,0.13396\ldots \end{align*}

Since

0.13396>0.05,0.13396>0.05,

H0H_0 is not rejected. There is insufficient evidence to support the newspaper’s claim that the mean score in town B is greater than the mean score in town A.

(c)

解法一

思路

展开

对每个城镇分别检验“该镇总体均分小于全国均分 μ\mu”。题目说两个检验都没有显著结果,因此各自的标准化统计量都不能小于 5%5\% 左尾临界值 1.6449-1.6449。这会分别给出 μ\mu 的上界;要让两项条件同时成立,必须取两个上界中较小的一个。

答题过程

展开

For each town, consider the lower-tailed test

H0: μtown=μ,H1: μtown<μ.\begin{align*} H_0:&\ \mu_{\text{town}}=\mu,\\ H_1:&\ \mu_{\text{town}}<\mu. \end{align*}

For the result to be not significant at the 5%5\% level, the test statistic must satisfy

xˉμ18/n1.6449.\frac{\bar{x}-\mu}{18/\sqrt{n}}\geqslant-1.6449.

For town A,

52.4μ18/251.644952.4μ1.6449(185)μ52.4+1.6449(185)μ58.3216\begin{align*} \frac{52.4-\mu}{18/\sqrt{25}}\geqslant&\,-1.6449\\ 52.4-\mu\geqslant&\,-1.6449\bigg(\frac{18}{5}\bigg)\\ \mu\leqslant&\,52.4+1.6449\bigg(\frac{18}{5}\bigg)\\ \mu\leqslant&\,58.3216\ldots \end{align*}

For town B,

57.8μ18/301.6449μ57.8+1.6449(1830)μ63.2056\begin{align*} \frac{57.8-\mu}{18/\sqrt{30}}\geqslant&\,-1.6449\\ \mu\leqslant&\,57.8+1.6449\bigg(\frac{18}{\sqrt{30}}\bigg)\\ \mu\leqslant&\,63.2056\ldots \end{align*}

Both conditions must hold, so the smaller upper bound is the limiting one. Therefore, the largest possible value of μ\mu is

μ=58.3\boxed{\mu=58.3}

to 3 significant figures.