📚 AS AQA Statistics: Key Points for Experimental / Practical Assessment | AS AQA 统计：实验/实践考核要点
Practical work in AS Statistics goes beyond solving textbook problems; it develops the ability to design investigations, collect and analyse data, and draw valid conclusions. AQA's assessment expects you to understand the entire statistical enquiry process and apply critical thinking to real data scenarios.
AS统计中的实践工作不仅仅是解决课本问题，它培养你设计调查、收集和分析数据并得出有效结论的能力。AQA的评估要求你理解整个统计探究过程，并将批判性思维应用于实际数据情境。
1. The Statistical Enquiry Cycle | 统计探究循环
At the heart of any practical statistical investigation lies the statistical enquiry cycle (SEC). This iterative process includes: specifying a problem or question, planning how to collect data, collecting the data, processing and presenting the data, and interpreting the results to answer the original question. Mastering the SEC allows you to structure any data-based task systematically.
任何实践统计调查的核心在于统计探究循环（SEC）。这一循环过程包括：明确问题或疑问、计划如何收集数据、收集数据、处理和呈现数据，并解释结果以回答原始问题。掌握SEC能让你系统地构建任何基于数据的任务。
AQA examination questions often present a scenario and ask you to comment on aspects of the enquiry cycle—such as whether the sampling method is appropriate or if the conclusions are justified. You must be able to identify each stage and suggest improvements where necessary.
AQA考试题目通常会给出一个情景，并要求你就探究循环的各个方面进行评价——例如抽样方法是否恰当，或者结论是否有充分依据。你必须能够识别每个阶段，并在必要时提出改进建议。
2. Formulating a Clear Hypothesis or Question | 明确假设或问题的制定
A precise research question or hypothesis is the starting point of any investigation. In AS Statistics, you are expected to distinguish between a statistical hypothesis (a statement about a population parameter) and a non-statistical guess. A testable hypothesis often takes the form: 'The population mean of ... is ...' or 'There is a difference between the means of two populations'.
精确的研究问题或假设是任何调查的起点。在AS统计中，要求你区分统计假设（关于总体参数的陈述）和非统计猜测。一个可检验的假设通常采用以下形式："...的总体均值为..."或"两个总体的均值之间存在差异"。
When writing a hypothesis for a practical investigation, ensure it is specific, measurable and linked to the data you can collect. Avoid vague language such as 'higher' or 'better' without defining what you are comparing.
在为实践调查撰写假设时，要确保它具体、可度量，并与你能收集的数据相关联。避免使用模糊的词语，如"更高"或"更好"，而不明确你在比较什么。
3. Sampling and Data Collection Strategies | 抽样与数据收集策略
Choosing the right sampling method is crucial for obtaining representative data. Main methods include: simple random sampling, where every member of the population has an equal chance of being selected; stratified sampling, which divides the population into groups (strata) and samples proportionally; systematic sampling, selecting every nth individual; and opportunity (or convenience) sampling. In AS AQA, you need to know the advantages and drawbacks of each.
选择正确的抽样方法对于获取具有代表性的数据至关重要。主要方法包括：简单随机抽样，即总体中每个成员被选中的机会均等；分层抽样，将总体分成若干组（层）并按比例抽样；系统抽样，每隔n个个体选择一个；以及机会（便利）抽样。在AS AQA中，你需要了解每种方法的优点和缺点。
In practical assessments, you might be asked to design a sampling plan for a given scenario. Always consider the population definition, sample size, method of selection, and how to minimise bias. For example, stratified sampling is preferred when there are distinct subgroups you want to represent fairly.
在实践评估中，你可能需要针对给定情景设计一个抽样计划。始终要考虑总体的定义、样本量、选择方法以及如何尽量减少偏差。例如，当存在明显的子群体且你希望公平地代表它们时，分层抽样更受青睐。
4. Designing Experiments: Control, Randomisation and Replication | 实验设计：控制、随机化与重复
When planning an experiment to compare treatments, three fundamental principles apply: control, randomisation, and replication. Control involves keeping other variables constant to isolate the effect of the treatment. Randomisation ensures that subjects are allocated to treatment groups without systematic bias. Replication means using enough subjects or repeating the experiment to estimate experimental error and increase reliability.
在规划比较处理的实验时，适用三个基本原则：控制、随机化和重复。控制涉及保持其他变量不变，以隔离处理效果。随机化确保受试者被分配到处理组时没有系统偏差。重复意味着使用足够的受试者或重复实验，以估计实验误差并提高可靠性。
AQA often tests your understanding of why random allocation is necessary, how blocking can be used to reduce variation due to extraneous factors, and why a control group is essential in many experiments. You should be able to criticise a flawed experimental design and suggest practical improvements.
AQA经常考查你是否理解为什么需要随机分配，如何使用区组来减少外来因素造成的变异，以及为什么在许多实验中对照组必不可少。你应该能够批判一个有缺陷的实验设计，并提出实际可行的改进建议。
5. Dealing with Bias and Confounding Variables | 处理偏差与混杂变量
Bias occurs when a study systematically favours certain outcomes. Common types include selection bias (non-representative sample), measurement bias (faulty instrument or leading questions), and non-response bias. Confounding arises when the effect of the variable of interest is mixed with the effect of an extraneous variable, making it difficult to determine causation.
当一项研究系统性地偏向某些结果时，就会产生偏差。常见类型包括选择偏差（样本不具代表性）、测量偏差（有缺陷的工具或诱导性问题）和无应答偏差。混杂是指当兴趣变量的效应与外来变量的效应混在一起，使得难以确定因果关系。
In your practical work, demonstrate awareness by identifying potential sources of bias in a given scenario and proposing ways to reduce them—such as using blinding in experiments, ensuring anonymity in surveys, or controlling a confounding variable by holding it constant or randomising.
在你的实践工作中，要通过识别给定情景中的潜在偏差来源，并提出减少偏差的方法来展示你的意识——例如在实验中使用盲法、确保调查中的匿名性，或通过保持不变或随机化来控制混杂变量。
6. Data Types and Measurement Scales | 数据类型与测量尺度
Data can be classified as primary (collected by the investigator) or secondary (collected by others). It can also be categorical (qualitative) such as eye colour, or numerical (quantitative). Numerical data divides into discrete (countable) and continuous (measurable). Understanding the type of data influences the choice of statistical tests and presentation methods.
数据可以分为初级数据（由调查者自行收集）或次级数据（由他人收集）。也可以是分类数据（性质）如眼睛颜色，或数值数据（数量）。数值数据又分为离散型（可数）和连续型（可测）。理解数据类型会影响统计检验和呈现方法的选择。
| Data Type (English) | 数据类型（中文） | Example |
|---|---|---|
| Qualitative / Categorical | 定性 / 分类 | Favourite colour, gender |
| Quantitative Discrete | 定量离散 | Number of siblings, test score |
| Quantitative Continuous | 定量连续 | Height, time taken |
AQA may ask you to comment on the appropriateness of a bar chart, histogram, or stem-and-leaf diagram based on the data type. Make sure you can justify why a histogram is used for continuous data and why a pie chart is suitable for categorical data showing proportions.
AQA可能会要求你根据数据类型评价条形图、直方图或茎叶图的适用性。一定要能够论证为什么连续数据用直方图，为什么饼图适合展示分类数据的比例。
7. Organising and Presenting Data Effectively | 有效整理与呈现数据
Good data presentation makes patterns and anomalies visible. Common tools include frequency tables, bar charts, histograms (with equal class intervals or unequal widths where area represents frequency), cumulative frequency graphs, box plots, scatter graphs, and stem-and-leaf diagrams. For AS AQA, you must be able to construct and interpret these correctly.
良好的数据呈现可使模式和异常值清晰可见。常用工具包括频数表、条形图、直方图（等组距或不等组宽时面积代表频数）、累积频数图、箱线图、散点图和茎叶图。对于AS AQA，你必须能够正确绘制和解读这些图表。
frequency / class width. For box plots, identify outliers using the 1.5 × IQR rule. These are frequently examined practical skills.
在呈现数据时，一定要清晰标注坐标轴、给出标题并选择合适的尺度。对于直方图，要记住频率密度 = 频数 / 组距。对于箱线图，要用 1.5 × IQR 规则识别异常值。这些都是经常考查的实践技能。
8. Summary Statistics and Visual Interpretations | 汇总统计与可视化解读
Measures of central tendency (mean, median, mode) and measures of spread (range, interquartile range, standard deviation) summarise data sets. In practical work, you need to choose the most appropriate measures based on the data distribution and the presence of outliers. The mean and standard deviation are sensitive to outliers, whereas the median and IQR are robust.
集中趋势的度量（均值、中位数、众数）和离散程度的度量（极差、四分位距、标准差）用以汇总数据集。在实践工作中，你需要根据数据分布和是否存在异常值来选择最合适的度量。均值和标准差对异常值敏感，而中位数和四分位距具有稳健性。
Using box plots to compare two or more data sets is a key skill. Comment on the medians, the spread (IQR and range), skewness, and potential outliers. Relate your comments to the context—for example, 'the median waiting time at surgery A is lower, but its greater IQR suggests less consistency'.
使用箱线图比较两个或多个数据集是一项关键技能。要评论中位数、离散程度（IQR和极差）、偏态以及潜在的异常值。将你的评论与背景联系起来——例如，"诊所A的等待时间中位数较低，但其较大的IQR表明一致性较差"。
9. Probability and Simulation in Practice | 实践中的概率与模拟
Probability models underpin many statistical arguments. In practical investigations, you may need to calculate probabilities for binomial or normal distributions using given parameters. AQA also encourages the use of simulation (using random numbers or technology) to model real-world processes and estimate probabilities empirically.
概率模型是许多统计论证的基础。在实践调查中，你可能需要使用给定参数计算二项分布或正态分布的概率。AQA还鼓励使用模拟（利用随机数或技术）来模拟真实世界过程，并通过经验估算概率。
When using simulation, always state your assumptions, the number of trials, and how the simulation mimics the real situation. Recogn
Published by TutorHao | AS 统计 Revision Series | aleveler.com
更多咨询请联系16621398022（同微信）
或截图后用「扫一扫 → 相册」选择本图