📚 IGCSE OCR Business: A Guide to Conducting Experiments | IGCSE OCR 商务：实验操作指南
Experiments play a crucial role in business decision-making, allowing firms to test ideas, measure cause-and-effect relationships, and make evidence-based choices. This guide walks you through the entire experimental process, from hypothesis formation to data analysis, as required by the IGCSE OCR Business specification.
实验在商业决策中扮演着至关重要的角色，它使企业能够测试想法、衡量因果关系，并做出基于证据的选择。本指南将带你走完整套实验流程，从假设提出到数据分析，完全满足 IGCSE OCR 商务课程的要求。
1. Understanding Experiments in Business | 理解商业中的实验
An experiment in business is a research method where one variable is deliberately changed to see how it affects another variable, while other factors are kept constant. This helps managers understand what drives consumer behaviour, sales, or productivity.
商业中的实验是一种研究方法，通过刻意改变一个变量来观察它如何影响另一个变量，同时保持其他因素不变。这有助于管理者了解是什么推动了消费者行为、销售额或生产率。
For example, a supermarket might change the layout of a product display (the independent variable) and measure the resulting change in sales (the dependent variable). The store would keep factors like price, promotions, and location constant to isolate the effect.
例如，一家超市可能改变产品陈列的布局（自变量），并测量由此带来的销售额变化（因变量）。商店会保持价格、促销和位置等因素不变，以分离出这种效果。
2. Types of Business Experiments | 商业实验的类型
Business experiments fall into two main categories: laboratory experiments and field experiments. Laboratory experiments take place in an artificial setting where the researcher has high control over variables. Field experiments occur in a natural business environment, such as a real shop floor or online store.
商业实验主要分为两类：实验室实验和现场实验。实验室实验在研究人员对变量有高度控制的人为环境中进行。现场实验则发生在自然的商业环境中，比如真实的商店或在线店铺。
Laboratory experiments offer high internal validity because extraneous factors can be minimised, but they may lack external validity since behaviour in a lab may not reflect real-world buying decisions. Field experiments provide more realistic results but are harder to control and replicate.
实验室实验具有较高的内部效度，因为外部因素可以被最小化，但它们可能缺乏外部效度，因为实验室中的行为可能无法反映真实的购买决策。现场实验能提供更真实的结果，但更难控制和复制。
3. Formulating a Hypothesis | 提出假设
A hypothesis is a clear, testable prediction about the relationship between two variables. In business, it often takes the form: ‘If we change X, then Y will increase/decrease.’ A good hypothesis must be specific, measurable, and linked to the experiment’s aim.
假设是关于两个变量之间关系的一个清晰、可测试的预测。在商业中，它通常采取这样的形式：“如果我们改变 X，那么 Y 将会增加/减少。”一个好的假设必须具体、可衡量，并与实验目标相关联。
For instance, ‘Offering free delivery on orders over £50 will increase average order value by 10% compared to a control group that receives no free delivery.’ This hypothesis sets a clear metric and allows for statistical testing.
例如，“对超过 50 英镑的订单提供免费送货服务，将使平均订单价值比不提供免费送货的对照组增加 10%。”这个假设设定了一个清晰的指标，并允许进行统计检验。
4. Identifying Variables | 识别变量
Every experiment involves three types of variables: independent variable (what you change), dependent variable (what you measure), and controlled variables (what you keep the same). Defining these precisely is critical for a valid experiment.
每个实验都涉及三种变量：自变量（你改变的是什么）、因变量（你测量的是什么）和控制变量（你保持不变的是什么）。精确地定义这些变量对于实验的有效性至关重要。
In a marketing campaign test, the independent variable might be the type of advertisement (video vs. image), the dependent variable could be click-through rate, and controlled variables might include audience demographics, time of day, and budget.
在一次营销活动测试中，自变量可能是广告类型（视频与图片），因变量可能是点击率，而控制变量可能包括受众人口统计特征、时段和预算。
5. Controlling Extraneous Variables | 控制无关变量
Extraneous variables are any factors other than the independent variable that could affect the dependent variable. If not controlled, they can lead to a false conclusion about cause and effect, a phenomenon known as confounding.
无关变量是指除自变量外任何可能影响因变量的因素。如果不加以控制，它们可能导致对因果关系的错误结论，即所谓的混杂现象。
Common methods to control extraneous variables include randomisation (assigning subjects randomly to experimental and control groups), matching (pairing subjects with similar characteristics), and keeping the setting constant. In online A/B testing, splitting web traffic randomly is a typical control technique.
控制无关变量的常用方法包括随机化（将受试者随机分配到实验组和对照组）、匹配（将特征相似的受试者配对）以及保持环境不变。在在线 A/B 测试中，随机分配网站流量是一种典型的控制技术。
6. Sampling and Sample Size | 抽样与样本量
Choosing the right sample is essential for generalising results to a broader population. Sampling methods in business experiments include random sampling, stratified sampling, and quota sampling. The sample must be representative of the target market.
选择合适的样本对于将结果推广到更广泛的群体至关重要。商业实验中的抽样方法包括随机抽样、分层抽样和配额抽样。样本必须能够代表目标市场。
A larger sample size generally increases the reliability of results and reduces the margin of error. However, practical constraints such as cost and time must be balanced. A minimum sample size can be calculated using statistical formulas to detect a meaningful effect.
较大的样本量通常会提高结果的可靠性并降低误差范围。然而，必须平衡成本和时间等实际限制。可以使用统计公式计算出最小样本量，以检测有意义的效应。
7. Designing the Experiment | 实验设计
The classic experimental design involves two groups: an experimental group that receives the treatment (e.g., a new pricing strategy) and a control group that does not. Comparing outcomes between these groups reveals the treatment effect.
经典的实验设计包含两个组：接受处理（例如新的定价策略）的实验组和不接受处理的对照组。比较这两组的结果可以揭示处理效应。
Repeated measures designs expose the same subjects to both treatment and control conditions at different times, reducing individual differences. Independent groups designs use different participants for each condition, avoiding order effects but requiring larger samples.
重复测量设计让相同的受试者在不同时间分别经历处理条件和对照条件，从而减少个体差异。独立组设计则为每种条件使用不同的参与者，避免了顺序效应，但需要更大的样本量。
8. Data Collection and Measurement | 数据收集与测量
Accurate data collection is the backbone of any experiment. Businesses use quantitative data (sales figures, website clicks, production output) and qualitative data (customer feedback, observation notes). Tools must be calibrated and procedures standardised.
准确的数据收集是所有实验的支柱。企业使用定量数据（销售额、网站点击量、生产产出）和定性数据（顾客反馈、观察笔记）。工具必须经过校准，程序必须标准化。
For example, if measuring employee productivity after a training programme, ensure the same measurement criteria are used before and after the intervention. Digital analytics platforms can automatically record consumer behaviour in real-time, reducing human error.
例如，如果在培训项目后测量员工生产率，需确保干预前后使用相同的测量标准。数字分析平台可以自动实时记录消费者行为，从而减少人为错误。
9. Ethical Considerations | 伦理考量
Conducting business experiments raises ethical issues, especially when human participants are involved. Informed consent, confidentiality, and the right to withdraw are fundamental principles. Deception, even for boosting sales, must be avoided.
进行商业实验会引发伦理问题，特别是当涉及人类参与者时。知情同意、保密和退出权是基本原则。即使是为了提高销售额，也必须避免欺骗。
Companies should also consider the wider impact of experiments. For instance, price discrimination experiments may disadvantage certain consumer groups and damage a brand’s reputation. Ethical review processes similar to academic research are increasingly adopted in industry.
公司还应考虑实验的更广泛影响。例如，价格歧视实验可能会使某些消费者群体处于不利地位，并损害品牌声誉。类似于学术研究的伦理审查流程在工业界正被越来越多地采用。
10. Analysing and Interpreting Results | 分析与解释结果
Once data is collected, it must be analysed to determine whether the hypothesis is supported. Statistical tests, such as the t-test or chi-square test, help assess whether differences between groups are significant or due to chance.
收集数据后，必须进行分析以确定假设是否得到支持。统计检验，如 t 检验或卡方检验，有助于评估组间差异是显著的还是偶然的。
Interpretation goes beyond statistics: managers must consider practical significance. A statistically significant 1% rise in sales might not justify the cost of implementing the change. Presenting findings using graphs and summary tables aids decision-making.
解释远不止统计数字本身：管理者必须考虑实际意义。统计上显著的 1% 销售增长可能无法证明实施变革的成本是合理的。用图表和汇总表展示发现有助于决策。
11. Advantages and Limitations of Experiments | 实验的优势与局限
Experiments provide strong evidence of causality, allow for precise replication, and enable managers to make data-driven decisions. They are particularly useful for testing new products, pricing strategies, and advertising messages before full-scale launch.
实验提供了有力的因果关系证据，允许精确复制，并使管理者能够做出数据驱动的决策。它们在全面推出新产品、定价策略和广告信息之前进行测试时特别有用。
However, experiments can be time-consuming and expensive. The artificial nature of some settings may limit real-world applicability, and ethical constraints sometimes prevent a true experiment from being conducted. In fast-moving markets, conditions change quickly, making long-term experiments impractical.
然而，实验可能耗时且昂贵。某些环境的人为性质可能限制了现实世界的适用性，并且伦理限制有时会阻碍进行真正的实验。在快速变化的市场中，条件变化很快，使得长期实验不切实际。
12. Real-World Application: A/B Testing in E-commerce | 现实应用：电子商务中的A/B测试
A/B testing is a practical form of field experiment widely used by online businesses. Two versions of a webpage (version A and version B) are shown to randomly selected visitors, and their behaviours, such as conversion rate or time spent on page, are compared.
A/B 测试是一种被在线企业广泛使用的现场实验形式。两个版本的网页（版本 A 和版本 B）被展示给随机选择的访问者，并比较他们的行为，例如转化率或在页面上停留的时间。
For example, a fashion retailer might test two different call-to-action button colours. By analysing click data from thousands of visitors, it can determine which version performs better and deploy the winning design across the entire site, directly boosting sales with minimal risk.
例如，一家时装零售商可能会测试两种不同的行动号召按钮颜色。通过分析数千名访问者的点击数据，它可以确定哪个版本效果更好，并将获胜的设计部署到整个网站，以最低的风险直接提升销售额。
Published by TutorHao | Business 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