Experimental Economics and Data Analysis: A Practical Guide for IB & CIE | 经济实验与数据分析：IB与CIE操作指南

📚 Experimental Economics and Data Analysis: A Practical Guide for IB & CIE | 经济实验与数据分析：IB与CIE操作指南

Economics is often seen as a purely theoretical subject, yet experimental methods and hands‑on data analysis are now central to understanding real‑world behaviour, especially in IB and CIE syllabuses. Whether you are conducting a classroom market simulation, testing prospect theory, or analysing unemployment figures, this guide walks you through the key steps of designing, executing, and interpreting economic experiments and empirical exercises.

经济学常被视为纯理论学科，但实验方法和动手数据分析如今已成为理解现实行为的关键，尤其在IB和CIE课程中。无论你是在课堂上模拟市场、检验前景理论，还是分析失业数据，本指南将带你逐步完成经济实验和实证练习的设计、执行与解读全过程。

1. The Nature of Economic Experiments | 经济实验的本质

Unlike natural sciences where controlled laboratory conditions are routine, economics studies human choices that are shaped by incentives, information and social context. An economic experiment recreates a simplified decision‑making environment to isolate cause‑and‑effect relationships, such as how a change in tax rate affects labour supply or how framing influences risk preferences. Both IB Internal Assessments and CIE Paper 4 data‑response tasks reward students who treat data investigation as a form of structured experiment.

不同于可以常规使用受控实验室条件的自然科学，经济学研究的是受激励、信息和社会环境塑造的人类选择。经济实验通过再现简化的决策环境来分离因果关系，例如税率变化如何影响劳动供给，或框架效应如何影响风险偏好。无论是IB内部评估还是CIE卷四数据回应题，那些把数据探究视为结构化实验的学生往往能获得高分。

Experimental economics usually falls into three broad categories: laboratory experiments with real monetary incentives, field experiments that test policies in natural settings, and natural experiments that exploit exogenous shocks. For IB/CIE students, the practical focus is on classroom‑based market games, online behavioural tasks, and econometric exercises using secondary data from sources such as World Bank Open Data or national statistics offices.

实验经济学通常分为三大类：有真实金钱激励的实验室实验、在自然环境中检验政策的实地实验，以及利用外生冲击的自然实验。对于IB和CIE学生而言，实践重点是基于课堂的市场游戏、在线行为任务，以及利用世界银行公开数据或各国统计局二手数据的计量练习。

2. From Hypothesis to Variables: Designing Your Experiment | 从假设到变量：实验设计

Start with a clear economic question, for example: ‘Does providing recycling bins for free increase household recycling rates?’ Formulate a testable hypothesis using standard A‑level terminology — null hypothesis H₀ and alternative hypothesis H₁. Define independent, dependent and control variables meticulously. In the recycling case, the independent variable is access to free bins, the dependent variable is kilogrammes of recyclables collected per week, and controls might include household income, education level and prior environmental attitudes.

从清晰的经济问题开始，比如：“免费提供回收箱会提高家庭回收率吗？”用标准的A‑level术语构造可检验的假设——零假设H₀和备择假设H₁。细致定义自变量、因变量和控制变量。在回收案例中，自变量是免费回收箱的可及性，因变量是每周收集的可回收物千克数，控制变量可能包括家庭收入、教育水平和先前的环保态度。

All IB Economics IAs and CIE investigative tasks must be grounded in an explicit theoretical framework. Use a diagram or a model — such as a demand–supply graph, a production possibility frontier, or a game‑theory payoff matrix — to predict the direction of change before you collect any data. This prediction becomes your theoretical benchmark against which real outcomes are compared.

所有IB经济内部评估和CIE探究任务都必须建立在明确的理论框架之上。在收集任何数据之前，使用图表或模型（如供求图、生产可能性边界或博弈论收益矩阵）预测变化的方向。该预测将成为与现实结果进行比较的理论基准。

3. Selecting Participants and Ethical Considerations | 选择参与者和伦理考量

In a classroom market experiment, participants are usually classmates. Avoid selection bias by randomly assigning roles — buyers and sellers, or proposers and responders in an ultimatum game. For survey‑based research, reach a sample size that is both manageable and statistically meaningful: aim for at least 30 observations per condition to satisfy the central limit theorem. Always obtain informed consent and anonymise responses; never fabricate data.

在课堂市场实验中，参与者通常是同班同学。通过随机分配角色（买家与卖家，或最后通牒博弈中的提议者与回应者）来避免选择偏差。对于基于问卷的研究，要达到既可行又有统计意义的样本量：每个处理至少30个观测值以满足中心极限定理。务必获得知情同意并对回答匿名处理；绝不捏造数据。

Ethics in economic experiments go beyond privacy. Deception, though sometimes used in psychology, is discouraged in economics because it can contaminate participants’ future behaviour in similar tasks. If you must withhold certain details, debrief participants fully at the end of the session and offer them the chance to withdraw their data.

经济实验中的伦理不止于隐私。尽管欺骗在心理学中偶有使用，但在经济学中不被鼓励，因为它会污染参与者在类似任务中的未来行为。如果必须隐瞒某些细节，应在实验环节结束时充分向参与者说明，并允许他们撤回数据。

4. Data Collection Methods: Surveys, Simulations and Observations | 数据收集方法：问卷、模拟与观察

Choose a method that best captures the economic variable of interest. Controlled lab simulations — for example, a double‑oral auction on a whiteboard — generate high‑frequency price‑quantity data that can be plotted in real time. Online platforms such as VeconLab or classroom‑friendly tools like Google Sheets with live‑editing enable collaborative market experiments even in remote settings.

选择最能捕捉感兴趣经济变量的方法。受控实验室模拟——例如白板上的双向口头拍卖——生成可实时绘制的高频价量数据。VeconLab等在线平台或Google Sheets等支持实时编辑的课堂友好工具，即使在远程环境下也能开展协作式市场实验。

Structured questionnaires are ideal for investigating consumer confidence, inflation expectations or willingness to pay. Use Likert scales or multiple‑choice questions to make responses quantifiable. For natural‑experiment‑style analysis, download time‑series data on variables such as exchange rates, inflation and GDP growth from reliable databases. Always record the source, date of access and any transformation applied (e.g., deseasonalising or taking logs).

结构化问卷是调查消费者信心、通胀预期或支付意愿的理想工具。使用李克特量表或选择题使回答可量化。对于自然实验风格的分析，从可靠数据库下载汇率、通胀和GDP增长等变量的时间序列数据。务必记录来源、访问日期以及所应用的任何变换（如去季节化或取对数）。

5. Classic Behavioural Experiment: The Ultimatum Game | 经典行为实验：最后通牒博弈

The ultimatum game is a simple yet powerful tool for exploring fairness, rationality and strategic behaviour. Two players split a fixed sum, say £10. The proposer suggests a division; the responder can accept (both get the proposed amounts) or reject (both get zero). Standard theory predicts the proposer offers the smallest possible positive amount and the responder accepts. Real‑world results, however, show that offers below 30% are frequently rejected, challenging the assumption of pure self‑interest.

最后通牒博弈是探究公平、理性和策略行为的简单而有力的工具。两名玩家分配一笔固定金额，比如10英镑。提议者提出分配方案；回应者可以接受（双方获得提议的金额）或拒绝（双方均得零）。标准理论预测提议者提供最小的可能正金额而回应者接受。然而，现实结果显示低于30%的提议经常被拒绝，挑战了纯粹自利的假设。

To run this experiment in class, prepare numbered envelopes containing different splits. Pair students randomly; ensure the negotiations are anonymous to reduce social‑desirability bias. After the rounds, calculate the average accepted offer and the rejection rate. Discuss how these metrics deviate from the Nash equilibrium and link them to behavioural concepts like inequity aversion and bounded rationality.

要在课堂上开展此实验，准备装有不同分配方案的编号信封。随机配对学生；确保谈判匿名以减少社会期许偏差。多轮之后，计算平均被接受的提议和拒绝率。讨论这些指标如何偏离纳什均衡，并将其与不平等厌恶和有限理性等行为概念联系起来。

6. Simulating a Market: Price Discovery in a Double Auction | 模拟市场：双向拍卖中的价格发现

Assign half the class as buyers (each given a maximum willingness‑to‑pay) and the other half as sellers (each given a minimum willingness‑to‑accept). Allow open outcry trading for several periods. Buyers aim to buy below their valuation, sellers to sell above cost. The equilibrium price and quantity predicted by the intersection of demand and supply usually emerge within a few rounds, even without central coordination.

将全班一半分配为买家（每人给定最高支付意愿），另一半为卖家（每人给定最低接受意愿）。允许多轮次的公开叫价交易。买家力求以低于估值买入，卖家力求以高于成本卖出。即使在无中央协调的情况下，供求交点预测的均衡价格和数量通常也会在几轮内出现。

Record every transaction price and quantity. Plot the convergence path: in early periods prices may be dispersed, but they tend to narrow toward the theoretical equilibrium. This exercise vividly demonstrates Adam Smith’s ‘invisible hand’ and is directly applicable to understanding market efficiency — a topic examined in both IB Paper 2 and CIE Paper 4. For analysis, calculate consumer surplus and producer surplus from the recorded trades.

记录每笔成交价格和数量。绘制收敛路径：初期价格可能分散，但趋于向理论均衡靠拢。此练习生动展示了亚当·斯密的“看不见的手”，并且直接适用于理解市场效率——这是IB卷二和CIE卷四都考查的主题。在分析中，根据记录的交易计算消费者剩余和生产者剩余。

7. Descriptive Statistics and Graphical Analysis | 描述统计和图形分析

Once data are collected, begin with descriptive statistics. Report mean, median, standard deviation and range for each key variable. For example, if you measured the price elasticity of demand (PED) for cinema tickets across different income groups, compute group means and visualise them with bar charts and box‑whisker plots. Always label axes clearly: Quantity demanded (units per week) on the x‑axis and Price ($) on the y‑axis, using appropriate scales.

收集数据后，从描述统计入手。报告每个关键变量的均值、中位数、标准差和极差。例如，如果测量了不同收入群体对电影票的需求价格弹性（PED），计算各组均值并用条形图和箱线图可视化。务必清晰标注坐标轴：x轴为需求量（每周单位数），y轴为价格（美元），并使用合适的刻度。

PED = %ΔQ_d ÷ %ΔP

A common mistake is to ignore outliers. Identify anomalies — such as a student who reported a willingness to pay of $0.01 for a luxury good — and decide on a consistent rule for handling them, whether exclusion or winsorising, with full transparency in the methodology paragraph.

一个常见错误是忽略异常值。识别异常——例如一名学生报告的奢侈品支付意愿为0.01美元——并制定一致的处理规则，无论是剔除还是缩尾处理，并在方法段落中完全透明地说明。

8. Inferential Statistics: Testing Your Hypothesis | 推断统计：检验你的假设

To determine whether observed differences are statistically significant, use a t‑test for comparing two groups (e.g., treatment vs. control) or a chi‑squared test for categorical data (e.g., whether gender is associated with choice of fair trade products). Most school‑level experiments can be analysed with a simple two‑sample t‑test assuming unequal variances. Calculate the t‑statistic and compare it with the critical value at a 5% significance level.

要判断观察到的差异是否统计显著，可使用t检验比较两组（如处理组与对照组），或使用卡方检验处理分类数据（如性别是否与公平贸易产品选择相关）。大多数学校级别的实验可用假设方差不等的简单双样本t检验进行分析。计算t统计量并与5%显著性水平下的临界值比较。

t = (x̄₁ – x̄₂) / √(s₁²/n₁ + s₂²/n₂)

Interpret the p‑value correctly: a p‑value of 0.03 means there is only a 3% probability of obtaining a result at least as extreme as the observed, assuming the null hypothesis is true. If p < 0.05, reject H₀. But also discuss economic significance — a statistically significant increase of 0.2% in saving rate may be too small to matter for policy.

正确解释p值：p值为0.03意味着，在零假设为真的情况下，只有3%的概率得到至少与观察结果一样极端的结果。若p < 0.05，拒绝H₀。但同时讨论经济显著性——储蓄率统计显著的0.2%提升可能小到对政策毫无意义。

9. Calculating and Interpreting Elasticities | 弹性的计算和解读

Elasticity measurement is a core experimental skill. For price elasticity of demand, use the midpoint formula to avoid inconsistency when direction of change is reversed. Collect data on two price‑quantity points and apply:

弹性测度是核心实验技能。对于需求价格弹性，使用中点公式以避免方向反转时的不一致。收集两个价量点的数据并应用：

PED = [(Q₂ – Q₁) / ((Q₂ + Q₁)/2)] ÷ [(P₂ – P₁) / ((P₂ + P₁)/2)]

You can design a mini‑experiment by manipulating the price of a classroom snack and recording purchase intentions. For income elasticity (YED), survey monthly pocket money and demand for a normal good. For cross elasticity (XED), see how demand for bus rides changes when train fares rise. All three elasticities are examinable and strengthen the internal coherence of an IA or CIE data‑response answer.

你可以通过操纵课堂零食的价格并记录购买意向来设计一个迷你实验。对于收入弹性（YED），调查每月零花钱和对正常商品的需求。对于交叉弹性（XED），观察火车票价上涨时公交车乘坐需求如何变化。这三种弹性都可考查，并能增强IA或CIE数据回应题答案的内部一致性。

10. Writing the Experimental Report: Structure and Style | 撰写实验报告：结构与风格

Follow the standard IB Economics IA structure — introduction, methodology, data and diagram, analysis, evaluation — even if you are completing a CIE investigation. Start with a concise context paragraph that links your experiment to a real‑world economic issue, such as sugar taxes or congestion charging. Present your raw data in a clearly labelled table and processed data in well‑designed charts.

遵循标准的IB经济IA结构——引言、方法、数据和图表、分析、评估——即便你是在完成CIE探究。以一个将实验与现实经济问题（如糖税或交通拥堵费）联系起来的简明背景段落开头。用一个标注清晰的表格呈现原始数据，用设计良好的图表呈现处理后的数据。

The evaluation must go beyond ‘sample size was small’. Critically assess whether the experimental design accurately captured ceteris paribus, whether demand or supply curve shifts were controlled, and whether external validity (generalisability) was compromised. Compare your findings with existing economic studies; this demonstrates top‑band research skills.

评估必须超越“样本量小”的层面。批判性评估实验设计是否准确捕捉了其他条件不变，是否控制了需求或供给曲线的移动，以及外部有效性（可推广性）是否受损。将你的发现与现有经济研究进行比较；这展示了最高分段的探究技能。

11. Common Pitfalls and How to Avoid Them | 常见误区及规避方法

Treating correlation as causation. Always use the phrase ‘is associated with’ rather than ’causes’ unless you have a well‑identified natural experiment with a proper control group.

将相关性当作因果关系。除非你有一个识别清晰且带有适当对照组的自然实验，否则始终使用“与……相关”而非“导致”。
Ignoring the time dimension. Macroeconomics experiments often deal with lags. When analysing the effect of interest rate changes on investment, allow a minimum of two quarters for the transmission mechanism to work.

忽略时间维度。宏观经济学实验常涉及时滞。分析利率变化对投资的影响时，至少让传导机制运行两个季度。
Over‑reliance on averages. Distributions matter. A mean inflation rate of 3% could hide extremes of deflation and hyperinflation in different product categories; report measures of dispersion.

过度依赖平均值。分布很重要。3%的平均通胀率可能掩盖不同产品类别中通缩和恶性通胀的极端情形；应报告离散指标。
Neglecting behavioural biases. Students often assume rational decision‑making; remember that anchoring, loss aversion and overconfidence can systematically skew results.

忽视行为偏差。学生们常假设理性决策；要记住锚定效应、损失厌恶和过度自信会系统性地扭曲结果。

12. Conclusion: From Experiment to Economic Insight | 结语：从实验到经济洞见

Conducting an economic experiment or rigorous data analysis transforms abstract models into tangible learning. It sharpens your ability to think like an economist — to question assumptions, interrogate data, and appreciate the gap between theoretical predictions and observed behaviour. For IB and CIE candidates, this hands‑on mindset not only boosts grades but also builds the analytical foundation required for university‑level economics.

开展经济实验或严谨的数据分析能将抽象模型转化为具体的学习体验。它磨砺了你像经济学家一样思考的能力——质疑假设、审视数据，并体会理论预测与实际行为之间的差距。对IB和CIE考生而言，这种动手思维不仅能提高分数，还将构建大学经济学所需的分析基础。

Always remember that economics is a social science: its experiments involve people, not just numbers. Treat your subjects ethically, design your investigation carefully, and let the data tell a story. The skills outlined in this guide — from hypothesis formulation to statistical inference — are transferable to any empirical discipline, making them invaluable for your academic journey.

始终牢记经济学是一门社会科学：其实验涉及的是人，而不仅仅是数字。合乎伦理地对待你的受试者，精心设计你的探究，让数据讲述一个故事。从假设构造到统计推断，本指南所概述的技能可迁移至任何实证学科，使它们在你的学术旅程中弥足珍贵。

Published by TutorHao | Economics Revision Series | aleveler.com

更多咨询请联系16621398022（同微信）

Experimental Economics and Data Analysis: A Practical Guide for IB & CIE | 经济实验与数据分析：IB与CIE操作指南