Skip to content
CalcGospel 國際數學圖譜
返回

CIE 9231 2026 June Paper 41 Q1

A Level / CIE / FS

CIE 9231 2026 June Paper 41 Paper · Question 1

题目

Problem

The manager of a village store is interested in the shopping habits of his customers. From previous records, the median number of items bought per customer was found to be 88. The manager believes the median number of items bought per customer has increased. He selects a random sample of 1515 customers and records the number of items each customer buys at his store. The results are shown in the stem-and-leaf diagram.

(a) Suggest a reason why using a Wilcoxon signed-rank test to examine the manager’s belief might not be appropriate in this situation.

(1)

(b) Use a sign test at the 5%5\% significance level to investigate the manager’s belief.

(6)
题目中文翻译

一家乡村商店的经理对顾客的购物习惯感兴趣。根据以往记录,每位顾客所购物品数量的中位数为 88。经理认为,每位顾客所购物品数量的中位数已经增加。他随机抽取 1515 位顾客,记录每位顾客在店内购买的物品数量。结果如茎叶图所示。

(a) 提出一个理由,说明在这种情况下,使用 Wilcoxon 符号秩检验来考察经理的看法可能不合适。

(1)

(b) 在 5%5\% 的显著性水平下,使用符号检验考察经理的看法。

(6)
原卷通用作答要求

Answer all questions. You should use a calculator where appropriate.

You must show all necessary working clearly; no marks will be given for unsupported answers from a calculator.

Give non-exact numerical answers correct to 3 significant figures, or 1 decimal place for angles in degrees, unless a different level of accuracy is specified in the question.

请回答所有题目。适当时应使用计算器。必须清楚展示所有必要的计算过程;仅给出计算器答案而没有过程,不予评分。除非题目另有精度要求,非精确数值答案保留 3 位有效数字,以度为单位的角度保留 1 位小数。

解答

(a)

解法一

思路

展开

查看茎叶图的尾部。符号秩检验要求总体关于其中位数对称。

答题过程

展开

The observations have a long upper tail, with large values such as 5656 and 7474. This suggests a right-skewed population, so the symmetry assumption of the Wilcoxon signed-rank test may be inappropriate.

(b)

解法一

思路

展开

从茎叶图数出高于 88 的人数。用单侧精确二项概率检验是否有超过一半的顾客购买多于 88 件。

答题过程

展开

Let mm be the population median. Test H0 ⁣:m=8H_0\colon m=8 against H1 ⁣:m>8H_1\colon m>8. From the diagram, 1212 observations exceed 88 and 33 are below it, with no observations equal to 88. Under the null hypothesis, the probability of exceeding 88 is at most 1/21/2. The upper-tail probability is largest at the boundary 1/21/2, so we use S∼Binomial⁡(15,1/2)S\sim\operatorname{Binomial}(15,1/2) for the test. The tail probability is

 Pr⁡(S≥12)=Pr⁡(S≤3)= (150)+(151)+(152)+(153)215= 0.0175781….\begin{align*} &\,\Pr(S\ge12)=\Pr(S\le3)\\ =&\,\frac{\binom{15}{0}+\binom{15}{1}+\binom{15}{2}+\binom{15}{3}}{2^{15}}\\ =&\,0.0175781\ldots. \end{align*}

Since this is below 0.050.05, we reject H0H_0. There is sufficient evidence at the 5%5\% level that the population median number of items bought per customer has increased.

解法二

思路

展开

把高于基准的个数近似为正态变量,作连续性修正后进行单侧检验。

答题过程

展开

Test H0 ⁣:m=8H_0\colon m=8 against H1 ⁣:m>8H_1\colon m>8, where mm is the population median. There are 1212 observations above 88 out of 1515, with no ties. At the null boundary, S∼Binomial⁡(15,1/2)S\sim\operatorname{Binomial}(15,1/2), so its normal approximation has mean 7.57.5 and variance 3.753.75.

For the upper tail S≥12S\ge12, the continuity-corrected statistic is

z=11.5−7.53.75=2.06559….z=\frac{11.5-7.5}{\sqrt{3.75}}=2.06559\ldots.

Since z>1.645z>1.645, we reject H0H_0 at the 5%5\% level. There is sufficient evidence that the population median number of items bought per customer has increased.