题目
Bill owns a restaurant.
Over the next four weeks Bill decides to carry out a sample survey to obtain the customers’ opinions.
(a) Suggest a suitable sampling frame for the sample survey.
(b) Identify the sampling units.
(c) Give one advantage and one disadvantage of taking a census rather than a sample survey.
Bill believes that only 30% of customers would like a greater choice on the menu.
He takes a random sample of 50 customers and finds that 20 of them would like a greater choice on the menu.
(d) Test, at the 5% significance level, whether or not the percentage of customers who would like a greater choice on the menu is more than Bill believes. State your hypotheses clearly.
(Total for Question 2 is 9 marks)
题目中文翻译
Bill 拥有一家餐厅。
在接下来的四周里,Bill 决定进行抽样调查以获取顾客的意见。
(a) 建议一个适合此抽样调查的抽样框架。
(b) 识别抽样单位。
(c) 给出普查而非抽样调查的一个优点和一个缺点。
Bill 认为只有 30% 的顾客希望菜单有更多选择。
他随机抽取了 50 名顾客样本,发现其中 20 人希望菜单有更多选择。
(d) 在 5% 的显著性水平下,检验希望菜单有更多选择的顾客比例是否大于 Bill 的预期。清楚地陈述你的假设。
(第 2 题共 9 分)
解答
(a)
解法一
思路
展开
抽样框应是能够识别目标总体中每个抽样单位的完整名单。这里的总体是在这四周内到餐厅用餐的所有顾客,因此应使用这些顾客的完整记录,而不能只列部分顾客或顾客意见。
答题过程
展开
A suitable sampling frame is
(b)
解法一
思路
展开
抽样单位是实际被选择和调查的基本对象,而不是顾客给出的意见。因此,本调查中的抽样单位是每一名到餐厅用餐的顾客。
答题过程
展开
The sampling units are
(c)
解法一
思路
展开
普查涵盖总体中的每一名顾客,因此通常能减少抽样误差,结果更准确;但需要收集并分析更多数据,所以更加耗时。题目各要求一个优点和缺点,清楚标明即可。
答题过程
展开
An advantage of taking a census is that
A disadvantage is that
(d)
解法一
思路
展开
令 为希望菜单有更多选择的顾客比例。题目检验该比例是否大于 0.3,所以使用右尾二项检验。在原假设下 ;观察到 20 人,计算 并与 0.05 比较,最后写出结合餐厅情境的结论。
答题过程
展开
Let be the proportion of customers who would like a greater choice on the menu. The hypotheses are
Let be the number of such customers in the sample. Under ,
The observed value is , so the upper-tail probability is
Since
is not rejected. There is insufficient evidence at the significance level to suggest that more than of the restaurant’s customers would like a greater choice on the menu.
解法二
思路
展开
官方评分资料也接受临界区域法。从较大的整数开始寻找右尾概率不超过 0.05 的最小边界;比较 与 后得到临界区域,再检查观察值 20 是否落入其中。
答题过程
展开
Under , . The adjacent upper-tail probabilities are
and
Therefore, the critical region is
The observed value is not in the critical region, so is not rejected. There is insufficient evidence at the significance level to suggest that more than of the restaurant’s customers would like a greater choice on the menu.