📚 A-Level CIE Economics: A Guide to Experimental Methods | A-Level CIE 经济：实验方法指南
Economics is a social science, yet it increasingly relies on experimental methods to test hypotheses about human behaviour, market outcomes, and policy effectiveness. For A-Level CIE Economics students, understanding how experiments work — and their limitations — is essential for evaluating evidence and writing high-quality essays. This guide covers the key types of experiments used in economics, design principles, validity concerns, and how to apply this knowledge in exam contexts.
经济学是一门社会科学，但它越来越依赖实验方法来检验有关人类行为、市场结果和政策有效性的假设。对于 A-Level CIE 经济学学生来说，理解实验如何运作——以及它们的局限性——对于评估证据和写出高质量的论文至关重要。本指南涵盖了经济学中使用的关键实验类型、设计原则、有效性问题，以及如何在考试中应用这些知识。
1. Why Experiments Matter in Economics | 为什么实验在经济学中很重要
Traditionally, economics relied on observational data and theoretical models. However, causal inference from observational data is often confounded by endogeneity and omitted variable bias. Experiments allow economists to isolate causal effects by manipulating a treatment variable while controlling for extraneous factors. This shift toward experimental methods has been recognised by several Nobel Prizes, notably in 2019 to Abhijit Banerjee, Esther Duflo, and Michael Kremer for their experimental approach to alleviating global poverty. For CIE students, experiments demonstrate how economists build knowledge and test theories, a core part of the syllabus.
传统上，经济学依赖观察数据和理论模型。然而，从观察数据进行因果推断常常受到内生性和遗漏变量偏差的困扰。实验允许经济学家通过操纵处理变量同时控制外来因素来分离因果效应。这种转向实验方法的趋势已获得多项诺贝尔奖的认可，特别是 2019 年阿比吉特·班纳吉、埃斯特·迪弗洛和迈克尔·克雷默因其缓解全球贫困的实验方法而获奖。对于 CIE 学生来说，实验展示了经济学家如何构建知识并检验理论，这是教学大纲的核心部分。
2. Laboratory Experiments in Economics | 经济学中的实验室实验
Laboratory experiments take place in a controlled environment where participants make decisions, often using computers. A classic example is the double auction market experiment, where buyers and sellers trade under known conditions to test whether prices converge to equilibrium. The experimenter can vary parameters such as information, money endowments, or the number of traders. Key advantages are high internal validity (clear causal identification) and replicability. However, laboratory experiments often lack external validity because the environment is artificial and participants may behave differently knowing they are being observed (Hawthorne effect). In A-Level answers, you should discuss how lab experiments help test core economic assumptions like rationality or supply-demand dynamics, but also acknowledge their limitations in mimicking real-world markets.
实验室实验在受控环境中进行，参与者通常使用计算机做出决策。一个经典的例子是双向拍卖市场实验，买家和卖家在已知条件下交易，以检验价格是否收敛到均衡。实验者可以改变参数，如信息、货币禀赋或交易者数量。关键优势是内部有效性高（明确的因果识别）和可重复性。然而，实验室实验通常缺乏外部有效性，因为环境是人为的，参与者知道被观察时可能表现出不同的行为（霍桑效应）。在 A-Level 答案中，你应该讨论实验室实验如何帮助检验核心经济假设，如理性或供需动态，但也承认它们在模仿真实市场方面的局限性。
3. Field Experiments | 田野实验
Field experiments are conducted in naturally occurring settings — firms, schools, villages — where subjects often do not know they are participating in an experiment. A famous instance is the RAND Health Insurance Experiment, which randomly assigned families to different health insurance plans to measure the effect of cost-sharing on healthcare utilisation. Another example would be sending fake CVs with randomly assigned characteristics to test for discrimination in hiring. Field experiments offer greater external validity than lab experiments because the context is realistic, and they can still provide a clean causal estimate if randomisation is properly implemented. However, they are costlier, harder to replicate, and may face ethical challenges around informed consent. In your exam, highlight how field experiments bridge the gap between controlled settings and real-world complexity.
田野实验在自然发生的环境中进行——公司、学校、村庄——受试者通常不知道他们正在参与实验。一个著名的例子是 RAND 医疗保险实验，该实验将家庭随机分配到不同的医疗保险计划中，以衡量成本分摊对医疗保健利用的影响。另一个例子是发送随机分配特征的虚假简历，以测试招聘中的歧视。田野实验比实验室实验具有更高的外部有效性，因为情境是现实的，如果随机化得到适当实施，它们仍然可以提供清晰的因果估计。然而，它们成本更高，更难复制，并且可能在知情同意方面面临伦理挑战。在考试中，强调田野实验如何在受控环境和现实复杂性之间架起桥梁。
4. Natural Experiments | 自然实验
Natural experiments arise when an external event or policy change creates a situation resembling a randomised experiment. The economist does not control the treatment assignment; instead, they exploit variation that is ‘as good as random’. A classic case is Card and Krueger’s study of the minimum wage in New Jersey and Pennsylvania, where the treatment (minimum wage increase) was assigned by geography. Another is the use of twin births as an instrument to study the effect of family size on child outcomes. Natural experiments can have high external validity because they use real-world policy changes, but careful analysis is needed to demonstrate that the assignment mechanism is indeed exogenous. A-Level questions often ask you to evaluate the strengths and weaknesses of natural experiments compared to other methods.
自然实验发生在外部事件或政策变化创造出类似随机实验的情境时。经济学家不控制处理分配；相反，他们利用“如同随机”的变异。一个经典案例是卡德和克鲁格对新泽西州和宾夕法尼亚州最低工资的研究，处理（最低工资提高）是按地理位置分配的。另一个例子是使用双胞胎出生作为工具变量来研究家庭规模对子女结果的影响。自然实验可能具有很高的外部有效性，因为它们利用现实世界的政策变化，但需要仔细分析以证明分配机制确实是外生的。A-Level 问题经常要求你评价自然实验相对于其他方法的优势和劣势。
5. Key Design Elements: Randomisation and Control Groups | 关键设计要素：随机化和对照组
Any credible experiment requires a control group that does not receive the treatment, and random assignment to treatment and control groups. Randomisation ensures that, on average, observable and unobservable characteristics are balanced across groups, so any difference in outcomes can be attributed to the treatment. For example, in a school feeding programme experiment, children are randomly assigned to receive meals or not. The difference in test scores between the two groups yields the causal effect. Without randomisation, selection bias can confound results: perhaps the children who choose to participate are already more motivated. In CIE answers, you need to explain why randomisation is the ‘gold standard’ for causal inference and what happens when it is not feasible (leading to quasi-experiments).
任何可信的实验都需要一个不接受处理的对照组，以及随机分配到处理组和对照组。随机化确保平均而言，可观察和不可观察的特征在组间是平衡的，因此任何结果差异都可以归因于处理。例如，在学校供餐计划实验中，儿童被随机分配是否获得餐食。两组之间考试成绩的差异得出因果效应。如果没有随机化，选择偏差会混淆结果：也许选择参与的儿童本来就更有动力。在 CIE 答案中，你需要解释为什么随机化是因果推断的“黄金标准”，以及当它不可行时会发生什么（导致准实验）。
6. Internal Validity | 内部有效性
Internal validity refers to how well an experiment establishes a causal relationship between the treatment and the outcome within the study setting. Threats to internal validity include attrition (participants dropping out), spillover effects (control group being indirectly affected), experimenter bias, and non-compliance with treatment. A study with high internal validity can convincingly argue that the treatment caused the observed effect. Lab experiments tend to have higher internal validity, while field and natural experiments face more potential confounds. In exam answers, you can enhance your evaluation by discussing whether a study’s internal validity is strong enough to support its conclusions — and what could weaken it, such as small sample sizes or poor compliance.
内部有效性是指实验在研究环境内确立处理与结果之间因果关系的能力。内部有效性的威胁包括流失（参与者退出）、溢出效应（对照组被间接影响）、实验者偏见以及不遵守处理。具有高内部有效性的研究可以令人信服地论证处理导致了观察到的效应。实验室实验往往具有较高的内部有效性，而田野实验和自然实验面临更多潜在的混杂因素。在考试答案中，你可以通过讨论一项研究的内部有效性是否足够强以支持其结论——以及什么可能削弱它，例如小样本量或依从性差，来增强你的评价。
7. External Validity | 外部有效性
External validity concerns the extent to which experimental findings can be generalised to other populations, settings, and time periods. A study run on US college students may not apply to rural farmers in India; an experiment conducted in a booming economy may not hold during a recession. Economists worry about ‘travelling’ results across contexts. Replicating experiments in different settings and using meta-analyses help establish external validity. An A-Level essay might contrast a randomised control trial (RCT) in a specific region with a global policy recommendation — you should note that unless the sample is representative and the context similar, the external validity remains questionable. Policy-makers often demand multiple experiments before scaling up.
外部有效性涉及实验发现可以推广到其他人群、环境和时间段的程度。一项在美国大学生中进行的研究可能不适用于印度的农村农民；在经济繁荣时期进行的实验在衰退期间可能不成立。经济学家担心结果在不同背景下的“迁移”。在不同环境中复制实验和使用荟萃分析有助于建立外部有效性。A-Level 论文可能会对比特定地区的随机对照试验（RCT）与全球政策建议——你应该指出，除非样本具有代表性且情境相似，否则外部有效性仍然存疑。政策制定者通常要求在推广之前进行多次实验。
8. Ethical Considerations in Economic Experiments | 经济实验中的伦理考量
Ethical issues in economic experiments include informed consent, deception, and potential harm to participants. While deception is common in psychology, many economists avoid it to maintain trust and replicability. In field experiments, withholding a beneficial treatment from the control group can be ethically contentious, especially when dealing with poverty or health. For instance, a clean water experiment that provides filters only to the treatment group raises questions about fairness. Some designs use phase-in or encouragement designs to ensure all participants eventually receive the treatment. CIE students should be prepared to discuss these ethical dimensions, showing an understanding that economic research must balance scientific rigour with respect for individuals’ rights.
经济实验中的伦理问题包括知情同意、欺骗和对参与者的潜在伤害。虽然欺骗在心理学中很常见，但许多经济学家避免使用它，以保持信任和可重复性。在田野实验中，对对照组拒绝给予有益的处理可能在伦理上存在争议，尤其是在涉及贫困或健康问题时。例如，一个仅向处理组提供过滤器的清洁水实验引发了公平性问题。一些设计使用分阶段引入或鼓励设计，以确保所有参与者最终都接受处理。CIE 学生应准备好讨论这些伦理维度，表现出对经济研究必须在科学严谨性与尊重个人权利之间取得平衡的理解。
9. Quasi-Experiments and Other Approaches | 准实验和其他方法
When random assignment is impossible, economists turn to quasi-experimental methods. These include difference-in-differences (comparing trends before and after a policy change between treated and untreated groups), instrumental variables (using a variable that affects treatment but not the outcome directly), and regression discontinuity designs (exploiting a cut-off for assignment). For example, a study might use a test score threshold for scholarship eligibility to estimate the causal effect of financial aid on university completion. Quasi-experiments can produce credible causal estimates, but they rely on additional assumptions that must be defended. Knowing these methods helps you evaluate real-world empirical evidence and enriches your exam responses with methodological sophistication.
当随机分配不可能时，经济学家转向准实验方法。这些包括双重差分法（比较政策变化前后处理组和未处理组的趋势）、工具变量法（使用一个影响处理但不直接影响结果的变量）和断点回归设计（利用分配的分界点）。例如，一项研究可能使用奖学金资格的成绩阈值来估计财务援助对大学完成率的因果效应。准实验可以产生可靠的因果估计，但它们依赖于必须被辩护的额外假设。了解这些方法有助于你评估现实世界的实证证据，并用方法论上的深度丰富你的考试答案。
10. Applying Experimental Concepts in A-Level Exams | 在 A-Level 考试中应用实验概念
In CIE Economics papers, you may be asked to evaluate a statement like ‘A tax on sugary drinks reduces obesity’ using experimental evidence, or to assess how a government might use a trial programme to test a policy before nationwide rollout. The key is to demonstrate your understanding of causation, control, and validity. Structure your answer by defining the type of experiment, discussing design issues (randomisation, sample size, context), internal and external validity, and ethical aspects. Use real-world examples such as the PROGRESA conditional cash transfer programme in Mexico, which used random assignment and led to policy changes. Remember to balance strengths and weaknesses, and always link back to the question.
在 CIE 经济学试卷中，你可能被要求使用实验证据评价诸如“对含糖饮料征税可以减少肥胖”的陈述，或评估政府如何利用试点项目在全面推广前测试政策。关键是要展示你对因果关系、控制和有效性的理解。通过定义实验类型、讨论设计问题（随机化、样本量、情境）、内部和外部有效性以及伦理方面来构建你的答案。使用现实世界的例子，如墨西哥的 PROGRESA 有条件现金转移支付项目，该项目使用了随机分配并导致了政策变化。记住要平衡优势和劣势，并始终回扣问题。
11. Common Pitfalls and How to Avoid Them | 常见陷阱及如何避免
Students often confuse correlation with causation: just because two variables move together does not mean one causes the other. Experiments help, but even experimental results can be misinterpreted if internal validity is weak. Another pitfall is assuming that an experiment’s findings are universally applicable without checking external validity. Additionally, overlooking ethical constraints or practical limitations weakens an essay. To avoid these, always critically assess the methodology: ask how the treatment was assigned, whether the sample is representative, and what confounding factors might remain. In your revision, practise past paper questions that require evaluation of evidence, and use the ‘IDEA’ structure: Identify, Define, Explain, and Assess.
学生经常混淆相关性与因果关系：仅仅因为两个变量一起变动并不意味着一个导致另一个。实验有所帮助，但如果内部有效性弱，即便实验结果也可能被误解。另一个陷阱是假设实验发现普遍适用，而不检查外部有效性。此外，忽视伦理约束或实际限制会削弱论文。为了避免这些，始终批判性地评估方法论：询问处理是如何分配的，样本是否具有代表性，以及可能仍然存在哪些混杂因素。在复习中，练习需要评估证据的历年真题，并使用“IDEA”结构：识别、定义、解释和评估。
12. Summary and Final Tips | 总结与最后提示
Experimental methods have revolutionised economics, providing powerful tools to separate causation from correlation. Mastering these concepts will not only boost your exam performance but also deepen your appreciation of how economic knowledge is constructed. Focus on the three main types (lab, field, natural), design elements (randomisation, control), validity (internal vs external), and ethical considerations. When writing, be precise with terminology and always support your points with concrete examples. Keep this guide handy during revision, and you’ll be well prepared to tackle any experimental method question that appears on your CIE A-Level Economics paper.
实验方法已经彻底改变了经济学，提供了将因果关系从相关性中分离出来的强大工具。掌握这些概念不仅能提高你的考试成绩，还能加深你对经济知识如何构建的理解。重点关注三种主要类型（实验室、田野、自然）、设计要素（随机化、对照）、有效性（内部与外部）以及伦理考量。写作时，术语要精确，并始终用具体例子支持你的观点。在复习期间随身携带本指南，你将准备好应对 CIE A-Level 经济学试卷上出现的任何实验方法问题。
Published by TutorHao | Economics Revision Series | aleveler.com
更多咨询请联系16621398022（同微信）
屏轩国际教育cambridge primary/secondary checkpoint, cat4, ukiset,ukcat,igcse,alevel,PAT,STEP,MAT, ibdp,ap,ssat,sat,sat2课程辅导，国外大学本科硕士研究生博士课程论文辅导