Sampling Methods: Types and Applications | 抽样方法的类型与适用场景

📚 Sampling Methods: Types and Applications | 抽样方法的类型与适用场景

In A-Level Mathematics (Statistics), understanding sampling methods is essential for conducting valid statistical investigations. Sampling is the process of selecting a subset of individuals from a population in order to estimate characteristics of the whole population. Each method has distinct advantages and limitations, and the choice of method depends on the research context, available resources, and the need for representativeness.

在 A-Level 数学（统计学）中，理解抽样方法对于开展有效的统计调查至关重要。抽样是从总体中选择一部分个体以估计整体特征的过程。每种方法都有其独特的优势和局限，方法的选择取决于研究背景、可用资源以及对代表性的需求。

---

1. Why Sampling Matters | 为什么抽样很重要

Sampling is fundamental to statistics because studying an entire population is often impractical. Imagine trying to measure the height of every student in a country — the time, cost, and effort would be enormous. A carefully chosen sample allows statisticians to make reliable inferences about the population with far fewer resources.

抽样是统计学的基石，因为研究整个总体往往不切实际。设想测量一个国家所有学生的身高——所需的时间、成本和精力将是巨大的。精心选取的样本能够让统计学家以少得多的资源对总体做出可靠的推断。

Key reasons for sampling include:

• Sampling saves time, money, and labour. 抽样节省时间、金钱和人力。

• Sampling can produce accurate estimates when the sample is representative. 当样本具有代表性时，抽样可以产生准确的估计。

• Sampling is the only feasible option when testing is destructive, such as quality-checking light bulbs. 当检验具有破坏性时（例如灯泡的质量检查），抽样是唯一可行的选择。

• Sampling reduces the volume of data, making analysis simpler and more manageable. 抽样减少了数据量，使分析更简单、更易于管理。

---

2. Census vs Sample | 普查与抽样

A census is a survey conducted on every member of the population. A sample, by contrast, is a subset of the population that is actually observed. Both approaches have their place in statistical investigation, and it is important to understand their trade-offs.

普查是对总体中每一个成员进行的调查。相比之下，样本是实际观测到的总体的一部分。这两种方法在统计调查中都有其适用场景，理解它们的取舍非常重要。

Advantages of a census | 普查的优点

• It provides completely accurate results because every individual is measured. 它提供完全准确的结果，因为每个个体都被测量。

• There is no sampling error or bias arising from selection. 不存在由选择引起的抽样误差或偏倚。

• Detailed information about rare subgroups can be obtained. 可以获得关于稀有子群的详细信息。

Disadvantages of a census | 普查的缺点

• It is time-consuming and expensive. 它耗时且昂贵。

• It may be impractical for large or inaccessible populations. 对于庞大或难以接触的总体可能不切实际。

• It may be destructive if testing destroys the item. 如果检验会破坏物品，则可能具有破坏性。

Advantages of a sample | 抽样的优点

• It is quicker and cheaper than a census. 比普查更快、更便宜。

• It is feasible for populations that are infinite or constantly changing. 对于无限或不断变化的总体是可行的。

• It causes less disruption and can be more manageable. 造成的干扰更小，且更易于管理。

Disadvantages of a sample | 抽样的缺点

• Results are estimates, not exact values. 结果是估计值而非精确值。

• There is always the possibility of sampling error and bias. 始终存在抽样误差和偏倚的可能性。

• Small or rare subgroups may be missed entirely. 较小或稀有的子群可能完全被遗漏。

---

3. Simple Random Sampling | 简单随机抽样

Simple random sampling is the most basic and theoretically ideal probability sampling method. Under this method, every member of the population has an equal and independent chance of being selected. A sampling frame — a complete list of all individuals in the population — is required.

简单随机抽样是最基础且在理论上最理想的概率抽样方法。在该方法下，总体中的每个成员都有相等且独立的机会被选中。需要一个抽样框——即总体中所有个体的完整名单。

How to carry it out | 如何实施

• Assign a unique number to every member of the sampling frame. 为抽样框中的每个成员分配一个唯一编号。

• Use a random number generator, a random number table, or a lottery method to select the required number of individuals. 使用随机数生成器、随机数表或抽签法选出所需数量的个体。

Advantages | 优点

• It is unbiased — every member has an equal chance of selection. 无偏——每个成员被选中的机会均等。

• It is easy to understand and the theory is simple. 易于理解，理论简单。

• It allows for straightforward calculation of sampling error. 便于直接计算抽样误差。

Disadvantages | 缺点

• A complete and accurate sampling frame is required, which may not always be available. 需要完整且准确的抽样框，但这并非总可获得。

• Selected individuals may be geographically spread out, making data collection expensive. 被选中的个体可能在地理上分散，使数据收集成本高昂。

• It may under-represent small subgroups purely by chance. 可能纯粹因随机性而使较小的子群代表性不足。

---

4. Systematic Sampling | 系统抽样

Systematic sampling involves selecting every k-th member from a sampling frame after a random starting point. The sampling interval k is calculated as:

k = N ÷ n

where N is the total population size and n is the desired sample size. For example, if N = 1000 and n = 100, then k = 10, meaning every 10th person is selected after a random start between 1 and 10.

系统抽样是从抽样框中随机起点开始，每隔 k 个成员选取一个。抽样间隔 k 的计算公式为：

k = N ÷ n

其中 N 是总体总量，n 是所需样本量。例如，若 N = 1000，n = 100，则 k = 10，意味着在 1 到 10 之间的随机起点之后，每第 10 个人被选中。

Advantages | 优点

• It is simpler and faster than simple random sampling when the sampling frame is long. 当抽样框很长时，它比简单随机抽样更简单、更快。

• The sample is spread evenly across the population list, giving good coverage. 样本在总体名单中分布均匀，覆盖良好。

• It is easier to manage in the field than fully random selection. 在实际操作中比完全随机选择更容易管理。

Disadvantages | 缺点

• If there is a periodic pattern in the sampling frame, the sample may be biased. 如果抽样框存在周期性模式，样本可能产生偏倚。

• It may still require a complete sampling frame. 它仍然可能需要完整的抽样框。

• It is not truly random unless the starting point is chosen randomly. 除非起点是随机选择的，否则它不是真正随机的。

---

5. Stratified Sampling | 分层抽样

Stratified sampling divides the population into distinct groups called strata (e.g., by age, gender, or income level). A random sample is then drawn from each stratum in proportion to the stratum’s size in the population. This guarantees representation of every subgroup.

分层抽样将总体划分为不同的组，称为层（例如按年龄、性别或收入水平）。然后从每一层中按该层在总体中所占比例抽取随机样本。这保证了每个子群都有代表。

The sample size for each stratum is calculated using the proportion formula:

每层的样本量使用比例公式计算：

nᵢ = (Nᵢ ÷ N) × n

where nᵢ is the sample size from stratum i, Nᵢ is the population size of stratum i, N is the total population size, and n is the total sample size.

其中 nᵢ 是来自第 i 层的样本量，Nᵢ 是第 i 层的总体大小，N 是总体总量，n 是总样本量。

Example | 示例

A school has 600 students: 300 males and 300 females. A stratified sample of 60 students is required. The number of males selected = (300 ÷ 600) × 60 = 30; the number of females selected = (300 ÷ 600) × 60 = 30. Simple random sampling should then be used within each stratum.

一所学校有 600 名学生：300 名男生和 300 名女生。需要抽取 60 名学生的分层样本。男生抽取人数 = (300 ÷ 600) × 60 = 30；女生抽取人数 = (300 ÷ 600) × 60 = 30。随后应在每一层内使用简单随机抽样。

Advantages | 优点

• All subgroups are represented proportionally, reducing sampling error. 所有子群都按比例得到代表，减少了抽样误差。

• It allows comparisons between different strata. 它允许在不同层之间进行比较。

• It often gives more precise estimates than simple random sampling for the same total sample size. 在相同总样本量下，它通常比简单随机抽样给出更精确的估计。

Disadvantages | 缺点

• It requires prior knowledge of the population structure. 需要事先了解总体结构。

• It is more complicated to administer than some other methods. 比其他一些方法更复杂。

• If strata are poorly defined, the method loses its advantage. 如果分层定义不当，该方法将失去其优势。

---

6. Quota Sampling | 配额抽样

Quota sampling is a non-random sampling method. The population is divided into groups (similar to strata), and an interviewer selects a pre-determined quota of individuals from each group. Crucially, the selection within each group is not random — it is left to the convenience or judgment of the interviewer.

配额抽样是一种非随机抽样方法。总体被划分为若干组（类似于分层），访问员从每组中选择预先确定的配额个体。关键在于，每组内的选择不是随机的——而是由访问员的便利或判断决定。

Advantages | 优点

• It is quick, cheap, and easy to administer. 快捷、便宜且易于实施。

• It does not require a sampling frame. 它不需要抽样框。

• It ensures that each predefined group is represented. 它确保每个预设的组都有代表。

Disadvantages | 缺点

• It is biased because the interviewer can choose who to include. 由于访问员可以选择纳入对象，因此存在偏倚。

• It is not possible to calculate sampling error meaningfully. 无法有意义地计算抽样误差。

• The results are generally not fully representative of the population. 结果通常不能完全代表总体。

---

7. Cluster Sampling | 整群抽样

Cluster sampling involves dividing the population into naturally occurring groups called clusters (e.g., schools, neighbourhoods, or city blocks). A random selection of clusters is made, and then every member of the selected clusters is included in the sample.

整群抽样是将总体划分为自然形成的组，称为群（例如学校、社区或街区）。先随机选择若干群，然后所选群中的每一个成员都纳入样本。

Advantages | 优点

• It is highly cost-effective, especially for geographically dispersed populations. 成本效益高，尤其适用于地理上分散的总体。

• It requires a listing of clusters rather than a full sampling frame of individuals. 只需要群的清单，而不是个体的完整抽样框。

• It is practical when individual members are hard to identify but clusters are easy to define. 当个体成员难以识别但群易于定义时，非常实用。

Disadvantages | 缺点

• It may be less precise than simple random sampling with the same sample size. 在相同样本量下，它可能不如简单随机抽样精确。

• Clusters may be very similar internally but differ from each other, increasing sampling error. 群内部可能非常相似但彼此差异大，从而增加抽样误差。

• If clusters are large, the sample may still be substantial. 如果群很大，样本规模可能仍然可观。

---

8. Convenience (Opportunity) Sampling | 便利（机会）抽样

Convenience sampling, also known as opportunity sampling, involves selecting individuals who are easiest to reach. For example, a researcher might stand outside a shopping centre and interview whoever passes by. This method is frequently used in pilot studies and exploratory research.

便利抽样，也称为机会抽样，是选择最容易接触到的个体。例如，研究者可能站在购物中心外，采访路过的任何人。这种方法常用于先导研究和探索性研究。

Advantages | 优点

• It is the fastest and cheapest method of all. 它是所有方法中最快、最便宜的。

• No sampling frame is required. 不需要抽样框。

• It is useful for testing surveys before a major study. 在大规模研究之前测试问卷时很有用。

Disadvantages | 缺点

• The sample is highly unlikely to be representative of the population. 样本极不可能代表总体。

• There is severe bias because only accessible people are included. 由于只包括可接触到的人，存在严重偏倚。

• It is impossible to generalise the results with any confidence. 无法以任何置信度推广结果。

---

9. Choosing the Right Method | 如何选择正确的抽样方法

Choosing an appropriate sampling method requires balancing cost, time, accuracy, and the nature of the population. The table below summarises the key features of each method:

选择合适的抽样方法需要在成本、时间、准确性和总体性质之间取得平衡。下表总结了每种方法的主要特征：

Method | 方法	Random? | 随机吗？	Sampling Frame? | 需要抽样框吗？	Bias Level | 偏倚水平	Cost & Time | 成本与时间
Simple Random | 简单随机	Yes | 是	Yes | 是	Low | 低	Medium | 中等
Systematic | 系统	Partly | 部分	Yes | 是	Low-Medium | 低至中	Medium | 中等
Stratified | 分层	Yes | 是	Yes (with strata info) | 是（需分层信息）	Very Low | 极低	High | 高
Quota | 配额	No | 否	No | 否	Medium-High | 中至高	Low | 低
Cluster | 整群	Partly | 部分	Cluster list only | 仅需群清单	Medium | 中等	Low-Medium | 低至中
Convenience | 便利	No | 否	No | 否	Very High | 极高	Very Low | 极低

When deciding, consider the following questions: Is a complete list of the population available? Do you need to compare subgroups? Is the population widely dispersed geographically? How large is your budget and time allowance? The answers will guide you to the most appropriate method.

在决策时，请考虑以下问题：能否获得完整的总体名单？是否需要比较子群？总体在地理上是否分散？预算和时间允许的范围有多大？这些答案将引导你选择最合适的方法。

---

10. Common Pitfalls in Exams | 考试中的常见误区

Examiners frequently report that students confuse the different sampling methods or fail to state their limitations. Here are the most common mistakes and how to avoid them:

考官经常反映学生混淆不同的抽样方法或未能说明其局限性。以下是最常见的错误以及如何避免它们：

• Confusing stratified and quota sampling: In stratified sampling, the selection within each stratum is random; in quota sampling, it is not. Remember this key difference. 混淆分层抽样和配额抽样：分层抽样中每层内的选择是随机的；配额抽样中则不是。记住这一关键区别。

• Forgetting the sampling frame: Simple random and systematic sampling cannot work without a complete and accurate sampling frame. 忘记抽样框：没有完整准确的抽样框，简单随机抽样和系统抽样无法实施。

• Ignoring practical limitations: Always state both advantages and disadvantages when evaluating a method in exam questions. 忽略实际局限性：在考试题目中评估一种方法时，务必同时陈述优点和缺点。

• Misusing the systematic interval formula: Remember k = N ÷ n, and ensure the start point is chosen randomly. 误用系统抽样间隔公式：记住 k = N ÷ n，并确保起点是随机选择的。

• Claiming convenience sampling is representative: It is quick and cheap, but heavily biased. Never describe a convenience sample as representative. 声称便利抽样具有代表性：它确实快捷便宜，但偏倚严重。切勿将便利样本描述为具有代表性。

• Not calculating stratified sample sizes correctly: Use the proportion nᵢ = (Nᵢ ÷ N) × n, and check that the stratum sample sizes add up to the total sample size. 未正确计算分层样本量：使用比例公式 nᵢ = (Nᵢ ÷ N) × n，并检查各层样本量之和是否等于总样本量。

By mastering the distinctions between these methods and practising calculation questions, you will be well prepared for both multiple-choice and structured exam questions on sampling. Always read the context carefully — the practical situation described in the question often provides strong clues about the most suitable method.

通过掌握这些方法之间的区别并练习计算题，你将能够从容应对关于抽样的选择题和结构性问题。务必仔细阅读题目背景——题目中描述的实际情境往往为选择最合适的方法提供了强有力的线索。

---

Published by TutorHao | Mathematics Revision Series | aleveler.com

更多咨询请联系16621398022（同微信）