📚 Key Points for Statistical Experiments/Investigations | 统计实验/实践考核要点
In your AQA GCSE Statistics practical assessment, you are expected to plan, carry out and evaluate a statistical investigation or experiment. Success depends on understanding the full enquiry cycle, choosing appropriate methods and being aware of potential pitfalls that can affect your conclusions. This guide highlights the key points you need to demonstrate in your practical work.
在 AQA GCSE 统计学的实践考核中，你需要规划、实施并评价一项统计调查或实验。成功取决于你是否理解完整的探究循环、选择了恰当的方法，并意识到可能影响结论的各种潜在陷阱。本指南将重点讲解你在实践作业中需要展现的关键要点。
1. The Statistical Enquiry Cycle | 统计探究循环
The practical assessment is structured around the statistical enquiry cycle (or PPDAC cycle: Problem, Plan, Data, Analysis, Conclusion). You need to show that you can move logically from a clear problem statement through to a well-supported conclusion, and then reflect critically on the process.
实践考核是围绕统计探究循环（即 PPDAC：问题、计划、数据、分析、结论）展开的。你需要展示自己能够从一个清晰的问题陈述出发，逻辑清晰地推进到有充分依据的结论，并能对整个流程进行批判性反思。
In your write-up, label each stage clearly. For example, state the hypothesis you intend to test, describe how you planned the data collection, present the raw data, carry out calculations and graphs, draw conclusions and finally evaluate the reliability of your findings.
在报告中，要清晰地标明每个阶段。例如，陈述你打算检验的假设，描述你是如何规划数据收集的，呈现原始数据，进行计算和绘制图表，得出结论，最后评价你研究结果的信度。
2. Formulating a Clear Hypothesis | 提出清晰的假设
A statistical investigation must begin with a well-defined hypothesis. This is a statement that predicts a relationship between two variables, such as ‘The length of a student’s foot is positively related to their height’ or ‘Reaction time decreases as temperature increases’. It should be testable using the data you plan to collect.
统计探究必须从一个清晰界定的假设开始。假设是对两个变量之间关系的一种预测性陈述，例如“学生脚的长度与身高正相关”或“反应时间随温度升高而缩短”。你所规划收集的数据必须能够检验这一假设。
Avoid vague questions like ‘Are people fit?’ Instead, turn the question into a measurable hypothesis: ‘The resting heart rate of students who exercise for more than 5 hours a week is lower than that of students who exercise less than 2 hours a week.’ This gives you a clear direction for data collection.
避免提出“人们是否健康？”这类模糊的问题。要把问题转化为可测量的假设：“每周锻炼时间超过 5 小时的学生的静息心率低于每周锻炼时间少于 2 小时的学生”。这样能为数据收集提供明确的方向。
3. Identifying Variables: Independent, Dependent and Control | 识别自变量、因变量与控制变量
In any experiment or comparative investigation, you must identify the independent variable (the one you change or compare), the dependent variable (the one you measure) and the control variables (the ones you keep constant). Clearly listing these variables is essential for achieving a valid comparison.
在任何实验或比较调查中，你都必须识别出自变量（你改变或比较的变量）、因变量（你测量的变量）和控制变量（你保持不变的变量）。清楚地列出这些变量对于实现有效比较至关重要。
For example, if you investigate how the length of a pendulum affects its period, the independent variable is the length of the string, the dependent variable is the time for 10 swings, and control variables include the mass of the bob, the angle of release and the type of string. Failing to control these would introduce confounding effects.
例如，假如你研究摆长如何影响周期，那么自变量是绳子长度，因变量是 10 次摆动的时间，控制变量包括摆锤质量、释放角度和绳子类型。若未能控制这些变量，就会引入混杂效应。
4. Designing the Experiment or Survey | 设计实验或调查
The design must be clear enough for someone else to replicate. Describe the apparatus, the number of trials or sample size, the levels of the independent variable you will use and how you will measure the dependent variable accurately. A pilot study can help refine your design.
实验设计必须足够清晰，以便他人可以重复。描述所用器材、试验次数或样本量、你将设置的自变量水平，以及如何精确测量因变量。预实验有助于完善你的设计。
When collecting primary data via a survey, consider how you will distribute the questionnaire, the format of questions (closed or open), and how you will avoid leading questions. The method of data collection should be appropriate for the target population and the hypothesis.
当通过问卷调查收集一手数据时，要考虑如何分发问卷、问题的格式（封闭式或开放式）以及如何避免诱导性问题。数据收集方法应适合目标总体和假设的要求。
5. Sampling and Randomisation | 抽样与随机化
Your practical work will often involve selecting a sample from a population. You should be able to justify the sampling method you choose – for instance, simple random sampling, stratified sampling or systematic sampling – and recognise the strengths and weaknesses of each in terms of bias and practicality.
你的实践作业常常需要从总体中抽取样本。你应该能够说明你选择某种抽样方法的理由——例如，简单随机抽样、分层抽样或系统抽样——并能从偏差和可行性角度认识每种方法的优缺点。
Randomisation is vital in comparative experiments. Allocating subjects randomly to treatment groups helps balance out uncontrolled variables. In your plan, state how you will randomise, e.g. using random number tables or a computer-generated list.
在比较实验中，随机化至关重要。将受试对象随机分配到处理组有助于平衡未控变量。在你的计划中，要说明如何实现随机化，例如，使用随机数表或计算机生成的列表。
6. Data Collection Sheets and Recording | 数据收集表与记录
Design a well-organised data collection sheet before you start. It should include columns for the independent variable, the dependent variable, any control checks and space for repeated trials. Clear headings, units and a layout that minimises recording errors are marks of good practice.
在开始之前，设计一份条理清晰的数据收集表。表中应包含自变量、因变量、任何控制检查以及重复试验所需的空间。清晰的表头、单位和有助于减少记录错误的布局，都是良好实践的标志。
Record data honestly and at the time of observation. Never tidy up results by removing data points simply because they do not fit a pattern, unless you have a clear experimental reason (such as a known equipment malfunction). Anomalies should be noted and discussed later.
要如实、即时地记录数据。切勿仅仅因为某些数据点不符合模式就将其清除，除非你有明确的实验性理由（例如已知的设备故障）。异常值应记录下来并在后面讨论。
7. Minimising Bias and Random Errors | 最小化偏差与随机误差
Bias can creep into an investigation through poor sampling, measurement methods or subjective judgement. For example, always reading a digital stopwatch in anticipation of a result can introduce expectation bias. Random errors occur due to unpredictable fluctuations, such as slight changes in environmental conditions or human reaction time.
偏差可能通过不良的抽样方法、测量方式或主观判断渗入调查过程。例如，总是带着预期读数去读取数字秒表就会引入期望偏差。随机误差则源于不可预测的波动，例如环境条件的轻微变化或人的反应时间。
Reduce bias by masking measurements where possible, using calibrated instruments and sticking to a clear protocol. Increase precision by taking repeated measurements and calculating a mean. The mean helps smooth out random errors, but systematic errors will remain and must be identified and corrected.
尽可能通过盲测、使用校准过的仪器并遵守清晰的操作规程来减少偏差。通过重复测量并计算平均值可以提高精密度。平均值有助于平滑随机误差，但系统误差仍然存在，必须予以识别和校正。
8. Reliability and Validity | 信度与效度
Reliability refers to the consistency of your results. If you repeated the experiment under the same conditions, would you obtain similar findings? You can assess reliability by comparing repeated readings or by calculating the range of a set of measurements. A small spread indicates good reliability.
信度指的是结果的一致性。如果在相同条件下重复实验，你是否会得到相似的结果？你可以通过比较重复读数或计算一组测量值的极差来评估信度。离散程度小说明信度好。
Validity asks whether you are actually measuring what you set out to measure. A valid experiment controls all the relevant variables and uses instruments that directly measure the intended quantity. A ruler may be a valid instrument for length, but a questionnaire about happiness may not be valid unless carefully designed.
效度问的是你是否真正测量了你打算测量的对象。一个有效的实验控制了所有相关变量，并使用能够直接测量目标量的仪器。尺子对于测量长度是一件有效的工具，但关于幸福感的问卷若不精心设计，就可能无效。
9. Ethical Considerations | 伦理考量
When your investigation involves human participants, you must respect their rights. Obtain informed consent, explain that participation is voluntary and that they can withdraw at any time. Keep personal data anonymous and store it securely. The AQA assessment expects you to mention how you addressed these ethical points.
如果你的调查涉及人类受试者，你必须尊重他们的权利。要获得知情同意，说明参与是自愿的，且他们可以随时退出。个人信息要匿名处理并安全保管。AQA 的评估期望你提及你是如何处理这些伦理要点的。
For experiments with physical activity or any risk of discomfort, carry out a risk assessment and minimise potential harm. Ethical conduct strengthens the integrity of your investigation and is a key aspect of good statistical practice.
对于涉及体力活动或任何可能导致不适的实验，要进行风险评估，并尽可能减少潜在伤害。符合伦理的操作能增强调查的诚信度，这是良好统计实践的一个关键方面。
10. Evaluation and Suggestions for Improvement | 评价与改进建议
An outstanding practical write-up includes an honest evaluation of limitations. Identify any sources of bias or uncontrolled variables, discuss how they might have affected the results and suggest specific improvements. For example, ‘The sample size was only 20, which may not represent the whole year group. Using a larger stratified sample would improve reliability.’
一份优秀的实践报告应包含对局限性的诚实评价。找出任何偏差来源或未受控的变量，讨论它们可能如何影响结果，并提出具体的改进建议。例如，“样本量仅为 20，可能无法代表整个年级。采用更大的分层样本可以提高信度”。
Comment on whether your conclusion is supported by the data, and suggest further investigations that could follow. Showing that you can reflect critically on your own work is exactly what examiners look for in the evaluation section.
要评论你的结论是否得到了数据的支持，并提出可以进一步开展的后续探究。展现出你能够批判性地反思自己的研究，这正是考官在评价部分所要寻找的。
Published by TutorHao | Statistics 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