📚 Year 13 CIE Psychology: Experimental & Practical Exam Essentials | CIE 心理学实验与实践考核要点
In Year 13 CIE Psychology, mastering experimental methods and practical assessment is not just about running a study — it is about demonstrating deep understanding of design, control, ethics, statistics, and critical evaluation. Whether you are tackling a research methods question in Paper 3 or evaluating studies in Paper 4, you must think like a psychologist. This guide compiles the essential points to help you structure answers, avoid common pitfalls, and ace experiment-based questions.
在 CIE A-Level 心理学第二年中，掌握实验方法和实践考核要点不仅仅是“会做实验”，更要在答题中体现出对实验设计、变量控制、伦理规范、统计分析和批判性评价的透彻理解。无论你是在 Paper 3 中回答研究方法题，还是在 Paper 4 中评价研究，都必须从心理学家的视角思考。本文梳理了核心要点，帮助你组织答案、避开常见陷阱，轻松拿下实验相关题目。
1. Understanding the Core of Experimental Design | 理解实验设计的核心
Every experiment starts with the aim of establishing a cause-and-effect relationship between an independent variable (IV) and a dependent variable (DV). The researcher manipulates the IV and measures any resulting change in the DV while holding all other variables constant. A well-designed experiment eliminates alternative explanations, allowing the conclusion that the IV caused the observed effect.
任何实验都始于建立自变量（IV）与因变量（DV）之间因果关系的目标。研究者操纵自变量并测量因变量随之发生的变化，同时保持其他所有变量恒定。一个设计精良的实验能够排除替代解释，从而得出“自变量导致了所观察到的效应”这一结论。
Confounding variables are the enemy. If an extraneous variable systematically changes alongside the IV, you can no longer be certain what caused the DV to change. Key terms like random allocation, standardised procedures, and control conditions all serve to minimise confounds.
混淆变量是实验的大敌。如果某个额外变量与自变量一起系统性地变化，你就无法确定到底是什么导致了因变量的变化。像随机分配、标准化程序和控制条件这些关键术语，都是为了最大限度减少混淆。
2. Operationalising Variables | 操作化定义变量
Operationalisation means turning abstract concepts into measurable, concrete variables. For example, “aggression” could be operationalised as the number of times a participant delivers a loud noise blast to a confederate, or “memory” as the number of words correctly recalled from a list. Without clear operational definitions, the experiment lacks replicability and precision.
操作化是指将抽象概念转化为可测量、具体化的变量。例如，“攻击性”可以操作化为被试向同谋者发出高强度噪音的次数，而“记忆力”可以操作化为从列表中正确回忆出的单词数量。没有清晰的操作性定义，实验就缺乏可复制性和精确性。
Always state both the IV and DV in fully operationalised terms when answering an exam question. For instance: “The IV is whether participants receive 200 mg of caffeine or a placebo, and the DV is reaction time in milliseconds on a computerised task.” This shows the examiner you can translate a theory into a testable experiment.
在回答考试题目时，一定要用完全操作化的语言陈述自变量和因变量。例如：“自变量是被试究竟是摄入200毫克咖啡因还是安慰剂，因变量是在电脑任务中测得的反应时（毫秒）。” 这向考官展示了你能够将理论转化为可检验的实验。
3. Types of Experimental Design | 实验设计类型
The three main experimental designs are independent measures, repeated measures, and matched pairs. Each has distinct advantages, disadvantages, and control solutions.
三种主要的实验设计是独立组设计、重复测量设计和配对组设计。每一种都有独特的优点、缺点以及相应的控制方法。
| Design / 设计 | Key Feature / 关键特征 | Strength / 优势 | Weakness / 劣势 | How to Control / 控制方式 |
|---|---|---|---|---|
| Independent measures / 独立组 | Different participants in each condition / 每种条件使用不同被试 | No order effects / 无顺序效应 | Participant variables may differ / 被试变量可能不同 | Random allocation to conditions / 随机分配到不同条件 |
| Repeated measures / 重复测量 | Same participants in all conditions / 所有条件下使用相同被试 | Fewer participants needed; participant variables controlled / 所需被试少；控制被试变量 | Order effects (practice, fatigue) / 顺序效应（练习、疲劳） | Counterbalancing / 平衡法 |
| Matched pairs / 配对组 | Different but matched participants in each condition / 每种条件使用不同但经匹配的被试 | Reduces participant variables without order effects / 减少被试变量且无顺序效应 | Time-consuming; matching may be imperfect / 耗时；匹配可能不完美 | Pre-testing and careful matching on key traits / 前测并依据关键特质仔细匹配 |
When choosing a design in your answer, explicitly justify your choice and mention how you would control for its main weakness. For example, if you select repeated measures, explain how you will use counterbalancing to reduce order effects.
在答案中选择一种设计时，要明确说明你为何这样选择，并提及你将如何克服其主要弱点。例如，如果你选择重复测量设计，就要解释你将如何通过平衡法减少顺序效应。
4. Formulating Hypotheses | 假设的陈述
A hypothesis is a testable, directional or non-directional prediction of the outcome. An experimental (alternative) hypothesis states the expected effect of the IV on the DV, e.g., “Participants who drink 200 mg caffeine will have significantly faster reaction times than participants who drink a placebo.”
假设是一个可检验的、方向性或非方向性的结果预测。实验（备择）假设陈述自变量对因变量的预期效应，例如，“摄入200毫克咖啡因的被试，其反应时将显著快于摄入安慰剂的被试。”
You must also formulate a null hypothesis, which states that any difference or correlation is due to chance, e.g., “There will be no significant difference in reaction time between the caffeine and placebo conditions; any observed difference is due to chance.” Always use the word ‘significant’ in both hypotheses when inferential statistics will be used.
你还必须提出零假设，即任何差异或相关都是由偶然因素造成的，例如，“咖啡因组和安慰剂组在反应时上无显著差异；任何观察到的差异均源于偶然。” 当需要用到推断统计时，两种假设中都要使用“显著”一词。
A one-tailed (directional) hypothesis is justified when prior research strongly suggests the direction of the effect; otherwise, a two-tailed (non-directional) hypothesis is safer. Exam questions often ask you to justify the choice of tail.
当已有研究强烈提示效应的方向时，可以使用单尾（方向性）假设；否则使用双尾（非方向性）假设更为稳妥。考试题经常要求你解释为何选择单尾或双尾。
5. Controlling Extraneous and Confounding Variables | 控制额外变量与混淆变量
To isolate the effect of the IV, you must control extraneous variables (EV). If an EV is not controlled and varies with the IV, it becomes a confounding variable. Common techniques include standardised instructions, using the same room and materials for all participants, and keeping the time of day constant.
为了隔离自变量的效应，你必须控制额外变量（EV）。如果一个额外变量未被控制且随自变量一同变化，它就变成了混淆变量。常见的控制方法包括采用标准化指导语、对所有被试使用相同的房间和材料、保持一天中的时段不变等。
Random allocation is essential in independent measures to distribute participant variables evenly across conditions. In repeated measures, counterbalancing (e.g., ABBA design) helps manage order effects. Single-blind and double-blind procedures control for demand characteristics and experimenter bias respectively.
在独立组设计中，随机分配至关重要，可将被试变量均匀分布到各个条件中。在重复测量中，平衡法（例如ABBA设计）有助于处理顺序效应。单盲程序和双盲程序则分别控制需求特征和实验者偏差。
- Standardisation: keep everything except the IV identical across conditions.
- Randomisation: use random number tables or software for allocation and stimulus order.
- Blinding: in a single-blind trial, participants do not know which condition they are in; in a double-blind trial, neither participants nor the experimenter interacting with them knows.
- 标准化：保持除自变量外的所有条件完全一致。
- 随机化：使用随机数字表或软件进行分配和刺激顺序的安排。
- 盲法：在单盲实验中，被试不知道自己处于何种条件；在双盲实验中，被试和直接接触他们的实验者均不知情。
6. Sampling and Ethics | 抽样与伦理
The sample should represent the target population. Common sampling methods include opportunity, random, stratified, volunteer, and snowball sampling. Describe your sample clearly: how many participants, age range, gender, and how they were recruited. Acknowledge sampling bias and suggest how to improve representativeness.
样本应能代表目标人群。常见的抽样方法包括机会抽样、随机抽样、分层抽样、志愿者抽样和滚雪球抽样。要清楚描述你的样本：人数、年龄范围、性别，以及如何招募。要承认抽样偏差，并提出如何提高代表性的建议。
Ethical guidelines are paramount. You must mention informed consent, the right to withdraw, confidentiality, protection from harm, and debriefing. For example, “Participants will sign a consent form, be reminded they can withdraw at any time, and be fully debriefed after the study, including the true aim and how their data will be used.”
伦理规范至关重要。你必须提及知情同意、退出权、保密、免受伤害保护以及事后解释。例如，“被试将签署同意书，被告知可随时退出，实验后接受完整的事后解释，包括研究的真实目的及其数据用途。”
If deception is used, you must justify why it is necessary, explain how participants will be debriefed, and offer the right to withhold data. Always connect your ethical controls to the BPS or APA guidelines.
如果使用了欺骗手段，你必须说明其必要性，解释事后如何向被试解释，并提供撤回数据的权利。始终将你的伦理控制与BPS或APA准则联系起来。
7. Data Collection and Measurement | 数据收集与测量
Data can be quantitative (numbers) or qualitative (descriptive). Experiments usually produce quantitative data, but you may also use self-report measures, behavioural checklists, or physiological recordings. Specify the measurement scale: nominal, ordinal, interval, or ratio, as this determines which statistical test you can use.
数据可以是定量（数字型）或定性（描述型）的。实验通常产生定量数据，但你也可以使用自我报告量表、行为检核表或生理记录。要说明测量尺度：称名、顺序、等距或比率，因为这决定了你可以使用哪种统计检验。
Ensure your data collection tool is both reliable (consistent) and valid (measures what it claims to). For instance, you could improve reliability by using a standardised recording sheet and inter-rater reliability checks; validity can be enhanced by piloting the materials and removing ambiguous items.
确保你的数据收集工具既具有信度（一致性）又具有效度（测量了它声称要测量的东西）。例如，你可以通过使用标准化记录表和评分者信度检验来提高信度；通过预试材料和剔除模糊项目来提高效度。
Always describe exactly how the DV will be recorded: “Reaction time will be measured in milliseconds using SuperLab software and automatically saved.” Such detail shows practical methodology awareness.
要准确描述因变量将如何被记录：“反应时将使用SuperLab软件以毫秒为单位测量并自动保存。” 这样的细节体现了对实践方法的了解。
8. Descriptive and Inferential Statistics | 描述统计与推断统计
Descriptive statistics summarise data: mean, median, mode, range, and standard deviation. In an exam, you might be asked to sketch a bar chart or explain why a median is more appropriate than a mean when there are outliers.
描述统计用于概括数据：平均值、中位数、众数、全距和标准差。在考试中，你或许需要绘制条形图，或解释为什么当有异常值时中位数比平均值更合适。
Inferential statistics allow you to determine whether the results are significant. Choosing the correct test depends on the experimental design, level of measurement, and whether the data meets parametric assumptions. The following table summarises common tests for the CIE syllabus.
推断统计用来判断结果是否显著。选择正确的检验取决于实验设计、测量水平以及数据是否符合参数检验假设。下表总结了CIE课程范围内的常用检验方法。
| Design / 设计 | Level of Measurement / 测量水平 | Test / 检验 |
|---|---|---|
| Independent measures / 独立组 | Interval/ratio data, normally distributed / 等距/比率数据，正态分布 | Unrelated t-test / 独立样本 t 检验 |
| Independent measures / 独立组 | Ordinal data / 顺序数据 | Mann-Whitney U test / 曼-惠特尼 U 检验 |
| Repeated measures or matched pairs / 重复测量或配对组 | Interval/ratio, normal / 等距/比率，正态 | Related t-test / 相关样本 t 检验 |
| Repeated measures or matched pairs / 重复测量或配对组 | Ordinal / 顺序 | Wilcoxon signed-rank test / 威尔科克森符号秩检验 |
| Nominal data (any design) / 称名数据（任何设计） | Frequencies / 频次 | Chi-squared / 卡方检验 |
You must also interpret the calculated value against a critical value at a given significance level, typically p < 0.05. If the observed value is more extreme than the critical value (or in some tests, larger than the critical value), you reject the null hypothesis.
你还必须将计算值与给定显著性水平（通常为 p < 0.05）下的临界值进行比较。如果观测值比临界值更极端（或在某些检验中大于临界值），你就拒绝零假设。
9. Evaluating Experimental Studies | 评估实验研究
Evaluation is a higher-order skill. You need to discuss internal validity (did the IV really cause the change?), external validity (can results be generalised?), ecological validity (does the setting and task reflect real life?), and reliability (would the same results be obtained again?).
评价是一项高阶技能。你需要讨论内部效度（真的是自变量导致了变化吗？）、外部效度（结果能否推广？）、生态效度（情境和任务能否反映真实生活？）以及信度（能再次获得相同结果吗？）。
Common threats include demand characteristics (participants guess the aim and alter behaviour), social desirability bias, experimenter effects, and mundane realism. For each criticism, propose a practical improvement. For example, “Using a double-blind procedure would reduce experimenter bias and demand characteristics.”
常见的威胁包括需求特征（被试猜出目的并改变行为）、社会赞许性偏差、实验者效应以及世俗现实性。针对每个批评点，提出切实可行的改进方法。例如，“采用双盲程序可以减少实验者偏差和需求特征。”
Also evaluate ethical issues and cultural bias. Many classic studies used WEIRD (Western, Educated, Industrialised, Rich, Democratic) samples. Suggest using a more diverse sample or replicating the study in a different cultural context to enhance population validity.
还要评价伦理问题和文化偏差。许多经典研究使用了WEIRD（西方、受过教育、工业化、富裕、民主）样本。可建议使用更具多样性的样本，或在不同的文化背景下重复研究，以提高人群效度。
10. Common Pitfalls and Exam Tips | 常见错误与考试技巧
Pitfall 1: Vague or non-operationalised variables. Always specify exactly how you will manipulate the IV and measure the DV. Avoid statements like “memory will be tested”. Instead, say “participants will study a 20-word list for two minutes and then freely recall as many words as possible.”
错误一：变量模糊或未操作化。务必准确说明你如何操纵自变量及如何测量因变量。避免诸如“记忆力将受到测试”这样的语句。应该说“被试将学习一张包含20个单词的列表，时长两分钟，然后尽可能多地自由回忆。”
Pitfall 2: Ignoring the null hypothesis. Even when the question does not explicitly ask for it, mention the null hypothesis when outlining the analysis. It demonstrates statistical thinking.
错误二：忽略零假设。即便题目没有明确要求，在概述分析时也要提及零假设。这展现了你的统计思维。
Pitfall 3: Choosing a design without justification. Always link your design choice to the aim and suggest at least one control for its primary weakness. For example, “I will use independent measures to avoid order effects, and I will randomly allocate participants to ensure groups are comparable.”
错误三：选择了设计却不说明理由。始终将你的设计选择与研究目的联系起来，并至少提出一项应对其主要弱点的控制措施。例如，“我将使用独立组设计以避免顺序效应，并通过随机分配保证各组具有可比性。”
Exam tip: Use key terminology — random allocation, counterbalancing, standardisation, operationalisation, single/double blind — in every design question. Examiners reward precise use of methodological language.
考试技巧：在每一道设计题中都要使用关键术语——随机分配、平衡法、标准化、操作化、单/双盲。考官看重方法学术语的准确使用。
11. Applying These Skills to CIE Exam Questions | 将这些技能应用于CIE考题
Typical CIE Paper 3 research methods questions ask you to design an experiment, often worth 12–14 marks. You will need to address: aim, hypotheses (experimental and null), IV/DV with operationalisation, design and justification, sampling, procedure, controls, ethics, data analysis, and one or two evaluation points.
典型的 CIE Paper 3 研究方法题要求你设计一个实验，通常占12–14分。你需要涵盖以下内容：目的、假设（实验假设和零假设）、操作化自变量/因变量、实验设计及理由、抽样、程序、控制方法、伦理、数据分析，以及一两个评价要点。
A good answer follows a logical flow. Begin with the aim, then state both hypotheses. Next, describe the design and how participants will be allocated. Outline the procedure step by step, using a future tense as if you are about to run the study. End with how you will analyse the data (naming the inferential test and justification) and one strength and one limitation.
好的答案遵循逻辑顺序。从目的开始，然后陈述两种假设。接着描述设计以及如何分配被试。逐步概述程序，使用将来时态，仿佛你马上就要开展研究。最后说明你将如何分析数据（说出推断检验的名称并进行论证），并给出一个优点和一个局限。
In Paper 4 evaluation questions, you might be given a study to critique. Use the same framework: identify the design, note operationalised variables, suggest control improvements, highlight ethical strengths and weaknesses, and question the validity and reliability. Always back up each point with evidence from the stimulus material.
在 Paper 4 的评价题中，你可能会拿到一项研究并需要对其进行批判。运用相同的框架：识别设计，指出操作化变量，提出改进控制的方法，强调伦理方面的优缺点，并对其效度和信度提出质疑。每一点都要用题干材料中的证据来支撑。
Practice writing full experimental designs under timed conditions. The more you practise, the more automatically the structure and terminology will flow, boosting your confidence and your marks.
在限时条件下练习撰写完整的实验设计方案。练习得越多，结构和术语就越能自然而然地流露出来，从而提升你的信心和得分。
Published by TutorHao | Psychology Revision Series | aleveler.com
Find A Level Psychology Textbooks on eBay UK
New, used and second-hand copies of textbooks and revision guides are often much cheaper than retail — check current listings and prices before you buy.
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