AS CAIE Statistics: Key Points for Experimental and Practical Assessment | AS CAIE统计：实验/实践考核要点

📚 AS CAIE Statistics: Key Points for Experimental and Practical Assessment | AS CAIE统计：实验/实践考核要点

In AS level CAIE Statistics, questions on designing experiments, planning investigations, and evaluating data collection methods often appear in written papers. Even though there is no separate practical exam, you are expected to understand key experimental principles, sampling techniques, and how to minimise bias. This article covers essential points for such ‘practical’ assessment preparation.

在AS CAIE统计学中，尽管没有独立的实验操作考试，但笔试常涉及实验设计、调查规划以及数据收集方法评估等题型。你需要掌握实验基本原则、抽样技术以及如何减少偏倚。本文汇总了这些实验/实践考核的核心要点，助你高效备考。

1. Understanding Experiments and Observational Studies | 理解实验与观察研究

An experiment deliberately imposes a treatment on subjects to observe a response, while an observational study simply collects data without intervention. In CAIE questions, you must be able to distinguish between them.

实验是主动对受试者施加某种处理并测量响应，而观察研究仅在不干预的情况下收集数据。CAIE考题常要求你区分二者。

For example, measuring plant heights after applying different fertilisers is an experiment; recording the grades of students who already take extra tuition is an observational study.

例如，施用不同肥料后测量植株高度是实验；仅记录已参加补习的学生成绩则属观察研究。

---

2. Principles of Experimental Design | 实验设计原则

Good experimental design follows key principles: control (keeping other variables constant), randomisation (allocating subjects to treatments randomly), and replication (repeating the experiment on multiple subjects). These principles allow cause-and-effect conclusions.

良好的实验设计遵循以下原则：控制（保持其他变量不变）、随机化（将受试者随机分配到处理组）和重复（在多个受试者上重复实验）。这些原则使得因果推断成为可能。

In AS Statistics, you may be asked to explain why randomisation is necessary – to avoid bias and balance out confounding variables.

在AS统计中，可能要求解释随机化的必要性——即避免偏倚并平衡混杂变量。

---

3. Randomisation and Replication | 随机化与重复

Randomisation ensures that each experimental unit has an equal chance of receiving any treatment. It helps eliminate systematic differences between groups, making the comparison fair.

随机化确保每个实验单元有同等机会接受任意处理，消除组间系统差异，使比较更加公平。

Replication means applying each treatment to several independent units. It allows estimation of experimental error and increases the reliability of conclusions.

重复是指将每种处理施加于多个独立单元，从而估计实验误差，提高结论的可靠性。

---

4. Control Groups and Blinding | 对照组与盲法

A control group receives no treatment or a standard treatment, providing a baseline for comparison. In many experiments, a placebo is used to isolate the psychological effect.

对照组不接受处理或使用标准处理，为比较提供基线。许多实验中会使用安慰剂，以分离心理效应。

Blinding (single-blind or double-blind) prevents bias. In single-blind studies, subjects do not know which treatment they receive; in double-blind, neither the subject nor the assessor knows.

盲法（单盲或双盲）可防止偏倚。单盲研究中，受试者不知自己接受何种处理；双盲中，评估者和受试者均不知情。

---

5. Sampling Methods in Data Collection | 数据收集中的抽样方法

Simple random sampling gives every member of the population an equal chance of selection, minimising bias. Stratified sampling divides the population into groups (strata) and samples randomly from each, ensuring representation.

简单随机抽样使总体中每个成员被选中的概率相等，最大限度地减少偏倚。分层抽样将总体分为多个层，然后从各层随机抽取，确保代表性。

Systematic sampling selects members at regular intervals from a list, while quota sampling selects a fixed number from each category but is non-random and prone to bias.

系统抽样按固定间隔从名单中抽取；配额抽样按类别固定数量选取，但非随机且易产生偏倚。

Cluster sampling divides the population into clusters, then randomly selects entire clusters for study; it is cost-effective when the population is widely spread.

整群抽样将总体分成群，随机抽取若干整群进行研究，当总体分布广泛时成本效益高。

In exam, you should justify why a simple random sample may be impractical and suggest alternatives like stratified. Always mention advantages and disadvantages of the chosen method.

考试中需解释为何简单随机抽样不可行，并提出如分层抽样的替代方案。务必说明所选方法的优缺点。

---

6. Designing Questionnaires and Surveys | 设计问卷与调查

A well-designed questionnaire should contain clear, unambiguous, and neutral questions. Avoid leading questions that suggest a particular answer. Use closed questions for easy analysis and open questions for richer detail.

设计良好的问卷应包含清晰、无歧义且中性的问题。避免引导性问题暗示特定答案。封闭式问题便于分析，开放式问题可获取更详尽的信息。

Pilot the survey on a small group to identify any problems before the main data collection. This helps check question clarity and estimate required sample size.

在正式收集数据前进行小范围预调查，以发现潜在问题。这有助于检验问题清晰度并预估所需样本量。

---

7. Identifying Sources of Bias | 识别偏倚来源

Bias can arise from non-random sampling, poor questionnaire design, non-response, or measurement error. Selection bias occurs when the sample is not representative of the population.

偏倚可能源于非随机抽样、问卷设计不佳、无应答或测量误差。当样本不代表总体时出现选择偏倚。

To reduce bias, use random sampling, improve response rates with follow-ups, and standardise measurement procedures. Confounding variables should be controlled through proper experimental design.

为减少偏倚，应采用随机抽样，通过追踪提高应答率，并标准化测量程序。混杂变量需通过恰当的实验设计加以控制。

---

8. Ethical Considerations in Experiments | 实验中的伦理考量

When designing an experiment involving human subjects, you must consider informed consent, confidentiality, and the right to withdraw. The experiment should not cause harm.

设计涉及人类受试者的实验时，必须考虑知情同意、隐私保密以及中途退出的权利。实验不应造成伤害。

In AS Statistics, ethical issues are often evaluated in the context of medical trials, such as using a placebo when effective treatment exists. You may be asked to comment on the appropriateness of a design.

在AS统计中，常在医学试验背景下评估伦理问题，例如当已有有效疗法时是否使用安慰剂。可能要求你评论实验设计的适当性。

---

9. Planning and Describing a Statistical Investigation | 规划与描述统计调查

CAIE questions frequently ask you to outline the steps of a statistical enquiry: define the problem, plan data collection, collect data, analyse and interpret data, draw conclusions.

CAIE试题经常要求你概述统计调查的步骤：界定问题、规划数据收集、收集数据、分析解释数据、得出结论。

When describing an experiment, mention how to allocate groups, what to measure (response variable), how to control variables, and how many replications. Mention blocking if there is a known source of variation.

描述实验时，要说明如何分配组别、测量什么（响应变量）、如何控制变量以及重复次数。若存在已知变异来源，应提及区组化。

A clear plan should state the treatment levels and the number of replicates per treatment. Estimating sample size in advance ensures sufficient power to detect a meaningful effect.

清晰的计划应陈述处理水平及每种处理的重复数。提前估计样本量可确保有足够的能力检测出有意义的效应。

The sample variance is often used to measure spread:

s² = Σ(x − x̄)² / (n − 1)

样本方差常用于度量离散程度：

s² = Σ(x − x̄)² / (n − 1)

---

10. Common Mistakes and Exam Tips | 常见错误与考试技巧

Common mistakes: confusing observational studies with experiments, forgetting to mention randomisation, and suggesting convenience sampling without acknowledging bias.

常见错误：混淆观察研究与实验、忘记提及随机化、建议便利抽样却不指出偏倚。

Always justify your choice of sampling or experimental design with reference to the context, and use appropriate terminology like ‘replication’, ‘control group’, ‘blinding’. Practice past paper questions on planning investigations—they carry significant weight in AS Statistics.

务必结合情景来论证所选的抽样或实验设计，并使用‘重复’、‘对照组’、‘盲法’等术语。多练习历年试题中的调查规划题，它们在AS统计学中占有重要分值。

Published by TutorHao | Statistics Revision Series | aleveler.com

更多咨询请联系16621398022（同微信）