Non-random Sampling | 非随机抽样

📚 Non-random Sampling | 非随机抽样

In A-Level Mathematics and Statistics, understanding how data is collected is as important as analysing the data itself. Sampling methods are divided into random and non-random techniques, each with distinct features, advantages and drawbacks that can significantly affect the reliability of conclusions.

在 A-Level 数学与统计中，理解数据如何收集与数据分析本身同样重要。抽样方法分为随机和非随机两大类，每一类都有鲜明的特点、优势和局限，会显著影响结论的可靠性。

1. Sampling Methods Overview | 抽样方法概述

Sampling is the process of selecting a subset of individuals from a population to estimate characteristics of the whole population. A population can be all students in a school, all manufactured items in a factory, or any defined group.

抽样是从总体中选择一部分个体以估计整个总体特征的过程。总体可以是学校所有学生、工厂所有产品，或任何明确界定的群体。

Sampling methods fall into two broad categories: probability (random) sampling, where every member has a known, non-zero chance of being selected, and non-probability (non-random) sampling, where selection relies on human judgment or convenience rather than chance.

抽样方法分为两大类：概率（随机）抽样，每个成员有已知、非零的入选机会；非概率（非随机）抽样，选择依赖于人的判断或便利性，而非概率。

In Edexcel A-Level Statistics, you are expected to identify, describe and critique both random and non-random sampling techniques, and to understand how the choice of method affects bias and generalisability.

在 Edexcel A-Level 统计课程中，你需要识别、描述和评价随机与非随机抽样技术，并理解方法的选择如何影响偏差和推广性。

2. What is Non-random Sampling? | 什么是非随机抽样？

Non-random sampling refers to any sampling technique where the selection of individuals does not involve a known probability mechanism. The sample is chosen based on subjective criteria, ease of access, or self-selection.

非随机抽样指任何不涉及已知概率机制来选择个体的抽样技术。样本根据主观标准、易得性或自我选择而产生。

Because there is no random element, it is impossible to calculate the sampling error or to apply probability theory to make valid inferences about the population. This does not mean non-random samples are always bad; they can be useful for exploratory research or when a sampling frame is unavailable.

由于没有随机因素，不可能计算抽样误差，也无法应用概率理论对总体作出有效推断。但这不意味着非随机样本总是不好；在没有抽样框或进行探索性研究时它们可能很有用。

Common types of non-random sampling tested in your exams include quota sampling, opportunity sampling, judgment sampling, voluntary sampling and snowball sampling.

考试中常见的非随机抽样类型包括配额抽样、便利抽样、判断抽样、自愿抽样和滚雪球抽样。

3. Quota Sampling | 配额抽样

Quota sampling is a non-random method where the researcher sets quotas for different subgroups (e.g. gender, age, ethnicity) to ensure the sample reflects the structure of the population. Within each quota, interviewers are free to choose any members who fit the category, often the most accessible ones.

配额抽样是一种非随机方法，研究者为不同子群体（如性别、年龄、种族）设定配额，以确保样本反映总体结构。在每个配额内，采访者可自由选择符合该类别的任何成员，通常是那些最容易接触到的人。

For example, if a population is 48% male and 52% female, the researcher might instruct interviewers to collect data from exactly 48 males and 52 females. The selection within each gender group is left to the interviewer’s convenience and is non-random.

例如，若总体中 48% 为男性、52% 为女性，研究者可能要求采访者恰好收集 48 名男性和 52 名女性的数据。每个性别组内的选择由采访者的便利决定，是非随机的。

Quota sampling is relatively quick and cheap because no sampling frame is needed. However, it introduces selection bias: interviewers may avoid people who look unfriendly, live in hard-to-reach areas, or are simply different from themselves, leading to an unrepresentative sample.

配额抽样相对快捷便宜，因为不需要抽样框。然而，它会引入选择偏差：采访者可能会避开看起来不友善、住在难以到达的区域或与自己不同的人，从而导致样本缺乏代表性。

4. Opportunity Sampling (Convenience Sampling) | 便利抽样

Opportunity sampling, also called convenience sampling, involves selecting individuals who are easily available at the time of data collection. A classic example is a researcher standing in a shopping centre and interviewing passers-by.

便利抽样，也称方便抽样，指选择在数据收集时最容易找到的个体。典型的例子是研究者站在购物中心采访路过的行人。

This method requires no sampling frame and is extremely quick and low-cost. However, it often results in a highly biased sample that over-represents certain groups – for instance, people who shop at a particular location, are available during working hours, or are willing to stop and talk.

这种方法不需要抽样框，极为快捷且成本低廉。但通常会导致样本高度偏差，过度代表某些群体——例如，在特定地点购物的人、工作时间有空的人或愿意停下来交谈的人。

In A-Level exam questions, you may be asked to explain why an opportunity sample might not be representative and how this limits the validity of any conclusions drawn.

在 A-Level 考试中，你可能会被要求解释为什么便利样本可能不具代表性，以及这如何限制所得结论的有效性。

5. Judgement Sampling (Purposive Sampling) | 判断抽样

Judgement sampling relies on the expertise of the researcher to hand-pick individuals who are considered typical, knowledgeable or informative for the study. This is a deliberately subjective choice rather than a probability-based selection.

判断抽样依赖研究者的专业知识来挑选被认为具有典型性、知识性或信息量丰富的个体。这是一种刻意主观的选择，而非基于概率。

This technique is common in small-scale or qualitative research, such as selecting a specific group of experts for a focus group. The quality of the sample depends entirely on the researcher’s judgment; poor judgment can lead to omission of important subgroups and produce misleading results.

这种技术常见于小规模或定性研究，例如为一组焦点小组选择特定的专家群体。样本质量完全取决于研究者的判断；判断不当可能导致重要子群体被遗漏，产生误导性结果。

From a mathematical statistics perspective, judgement sampling does not allow for standard error calculation and cannot be used to construct confidence intervals or hypothesis tests that require random selection.

从数理统计的角度看，判断抽样无法进行标准误差计算，也不能用于构建需要随机选择的置信区间或假设检验。

6. Voluntary Sampling (Self-selection Sampling) | 自愿抽样

In voluntary sampling, individuals choose to take part in the study on their own initiative. Common examples include online polls, phone-in surveys, and questionnaires returned by only those who feel strongly about the topic.

在自愿抽样中，个体主动选择参与研究。常见例子包括在线投票、电话打入调查和仅由对话题有强烈感受的人返回的问卷。

This method is very inexpensive and can reach a wide geographic audience quickly. However, voluntary response bias is a major problem: people with extreme opinions are much more likely to respond, while the silent majority remains unheard. The sample often over-represents strong positive or negative views.

这种方法成本极低，能迅速接触到广泛的受众。但自愿响应偏差是主要问题：持有极端观点的人更有可能回应，而沉默的大多数则未被听到。样本往往过度代表强烈的正面或负面观点。

In your exam, you should be able to identify when a sample is voluntary and explain why any conclusions from such a sample cannot be safely generalised to the wider population.

在考试中，你应该能够识别样本何时是自愿性的，并解释为何由此得出的任何结论不能安全地推广到更广泛的总体。

7. Snowball Sampling | 滚雪球抽样

Snowball sampling is a technique where initial participants are recruited and then asked to refer other people they know who fit the study criteria. The sample size grows like a rolling snowball.

滚雪球抽样是一种先招募初始参与者，然后请他们推荐认识的其他符合研究标准的人的技巧。样本量像滚雪球一样增长。

This method is particularly useful when studying hidden or hard-to-reach populations, such as people with rare medical conditions, members of underground communities, or experts in a niche field. A sampling frame is often impossible to construct for these groups.

这种方法在研究隐蔽或难以接触的群体时特别有用，例如罕见病患者、地下社群成员或小众领域的专家。对这些群体通常无法构建抽样框。

The main drawback is that the sample is highly dependent on the social networks of the initial seeds. People who are more isolated or who do not wish to be identified will still be excluded, and the resulting sample can be far from representative.

主要缺点是样本高度依赖初始种子成员的社交网络。更孤立或不愿被识别的人仍会被排除在外，由此得到的样本可能远不具有代表性。

8. Advantages and Disadvantages Summary | 优缺点总结

Non-random sampling methods share several practical advantages: they usually require no list of the population, are faster and cheaper to implement, and can be the only feasible approach when randomisation is impossible.

非随机抽样方法有几个共同的实用优势：通常不需要总体名单，实施起来更快更便宜，而且在无法随机化时可能是唯一可行的方法。

However, the core disadvantage is the inability to objectively measure sampling error. Because the selection probability is unknown, you cannot apply the central limit theorem or calculate margins of error, standard errors or confidence intervals that are mathematically justified.

然而，核心劣势是无法客观衡量抽样误差。由于选择概率未知，你无法应用中心极限定理，也不能计算具有数学依据的误差范围、标准误差或置信区间。

The table below summarises key characteristics of the main non-random sampling methods for quick revision.

下表总结了主要非随机抽样方法的关键特性，以便快速复习。

Method	Method (中文)	Key Feature	Main Bias Risk
Quota	配额抽样	Fixed number per subgroup	Interviewer selection bias
Opportunity	便利抽样	Ease of access	Over-representation of accessible groups
Judgement	判断抽样	Expert choice	Researcher subjectivity
Voluntary	自愿抽样	Self-selection	Voluntary response bias
Snowball	滚雪球抽样	Referral chains	Network homogeneity bias

9. Bias and Limitations in Non-random Sampling | 非随机抽样中的偏差与局限性

Bias in non-random sampling arises because the sample is not a miniature replica of the population. Selection bias is the most common: the way participants are chosen systematically favours certain outcomes or characteristics.

非随机抽样中的偏差源于样本并非总体的微缩复制品。选择偏差是最常见的：参与者被选择的方式系统性地偏向某些结果或特征。

Interviewer bias, prevalent in quota sampling, occurs when fieldworkers consciously or unconsciously select respondents who are similar to themselves or easier to approach. This can distort the demographic and attitudinal profile of the sample.

采访者偏差在配额抽样中普遍存在，当实地工作人员有意或无意选择与自己相似或更易接触的受访者时，就会发生这种情况。这会扭曲样本的人口结构和态度画像。

Voluntary response bias is typical of online surveys and phone-in polls. Those with strong opinions are far more likely to respond, while indifferent individuals remain silent, making the sample appear more polarised than the true population.

自愿响应偏差常见于在线调查和电话投票。持有强烈观点的人更可能回应，漠不关心的人保持沉默，使得样本比真实总体更显两极化。

Non-sampling errors, such as non-response and measurement errors, also affect non-random samples. While these errors exist in random sampling too, the lack of randomisation makes it impossible to assess their impact using probability models.

非抽样误差，如无响应和测量误差，也会影响非随机样本。虽然这些误差在随机抽样中也存在，但缺乏随机化使得无法用概率模型评估其影响。

10. Comparison with Random Sampling Methods | 与随机抽样方法的比较

Random sampling techniques, such as simple random sampling, stratified sampling, systematic sampling and cluster sampling, all use a known probability mechanism. This allows statistics like the sample mean to be unbiased estimators of population parameters, and permits calculation of standard errors.

随机抽样技术，如简单随机抽样、分层抽样、系统抽样和整群抽样，都使用已知的概率机制。这使得样本均值等统计量成为总体参数的无偏估计量，并允许计算标准误差。

In contrast, non-random samples cannot guarantee unbiasedness. Any parameter estimate from a non-random sample is suspect because we do not know how the selection probabilities differ across the population. Statistical inference is technically invalid.

相比之下，非随机样本无法保证无偏性。从非随机样本得到的任何参数估计都值得怀疑，因为我们不知道总体的选择概率如何变化。统计推断在技术上是无效的。

That said, non-random methods are widely used in business, social research and pilot studies where resources are limited or a complete sampling frame does not exist. The key exam skill is to evaluate fitness for purpose: a non-random sample may be perfectly adequate for generating hypotheses, but not for testing them.

话虽如此，非随机方法在商业、社会研究和试点研究中广泛应用，因为资源有限或根本不存在完整的抽样框。关键的考试技能是评估目的的适用性：非随机样本可能完全足够用于生成假设，但不适合用于检验假设。

11. Exam Tips and Common Misconceptions | 考试技巧与常见误解

Edexcel questions often present a real-world scenario and ask you to critique the sampling method used. Always identify the method by name, explain how it is implemented, and then discuss at least one advantage and one disadvantage related to representativeness and bias.

Edexcel 考题经常给出一个真实世界的情景，要求你评价所使用的抽样方法。始终按名称识别方法，解释其实施方式，然后至少讨论一个与代表性和偏差相关的优点和缺点。

A common misconception is that a larger sample automatically makes a non-random sample acceptable. While a larger sample size may reduce chance fluctuations, it does nothing to correct the systematic bias built into the selection process. A large biased sample remains biased.

一个常见的误解是，较大的样本量会自动使非随机样本可接受。虽然较大的样本量可能减少随机波动，但对纠正选择过程中内置的系统偏差毫无作用。有偏差的大样本仍然有偏差。

Another pitfall is confusing quota sampling with stratified sampling. Stratified sampling selects members randomly from each stratum, while quota sampling leaves the selection within each quota to interviewer convenience. Only stratified sampling allows probabilistic inference.

另一个陷阱是将配额抽样与分层抽样混淆。分层抽样从每个层中随机选择成员，而配额抽样将每个配额内的选择权交给采访者的便利性。只有分层抽样允许概率推断。

When justifying your answer, use phrases like ‘introduces selection bias’, ‘cannot calculate standard error’, ‘not scientifically generalisable’, and ‘over-represents accessible/enthusiastic individuals’ to show precise statistical thinking.

在论证答案时，使用诸如“引入选择偏差”“无法计算标准误差”“不能科学推广”“过度代表易接触/热情的个体”等表述，以展示精确的统计思维。

12. Conclusion and Key Takeaways | 结论与关键要点

Non-random sampling is a collection of pragmatic data-collection techniques that sacrifice mathematical rigour for speed, cost and feasibility. They are an essential part of the A-Level Statistics syllabus because they illustrate the real-world tension between ideal statistical design and practical constraints.

非随机抽样是一系列实用的数据收集技术，牺牲了数学严谨性以换取速度、成本和可行性。它们是 A-Level 统计大纲的重要组成部分，因为它们展示了理想统计设计与实际约束之间的现实拉锯。

Remember that no statistical test, however sophisticated, can rescue a sample that is fundamentally biased in its composition. Always evaluate the sampling method before interpreting any numerical summary.

请记住，无论统计检验多么复杂，都无法拯救一个构成上根本有偏差的样本。在解释任何数字总结之前，务必先评价抽样方法。

By mastering the characteristics, strengths and weaknesses of quota, opportunity, judgement, voluntary and snowball sampling, you will be well prepared to tackle any Edexcel examination question on non-random sampling with confidence and precision.

通过掌握配额抽样、便利抽样、判断抽样、自愿抽样和滚雪球抽样的特征、优势和弱点，你将为自信、精准地回答 Edexcel 考试中关于非随机抽样的任何问题做好充分准备。

Published by TutorHao | Mathematics Revision Series | aleveler.com

更多咨询请联系16621398022（同微信）

Non-random Sampling | 非随机抽样