A-Level CAIE Psychology: Experimental and Practical Assessment Essentials | A-Level CAIE 心理学：实验/实践考核要点

📚 A-Level CAIE Psychology: Experimental and Practical Assessment Essentials | A-Level CAIE 心理学：实验/实践考核要点

In CAIE A Level Psychology (9990), experimental and practical skills are assessed through written examination papers rather than a separately marked laboratory portfolio or coursework folder. The main home for these skills is Paper 2 Research Methods, but all specialist options also expect you to use experimental evidence, evaluate methodology and apply practical reasoning to novel scenarios. This article sets out the key assessment demands for experimental and practical work, with paired English-Chinese explanations to help you revise efficiently and answer with precision.

在 CAIE A Level 心理学（9990）中，实验与实践技能通过书面考试卷评估，而不是通过单独评分实验报告或课程作业。主要考查载体是 Paper 2 Research Methods，同时所有专题选项也要求运用实验证据、评估研究方法，并将实践推理应用于新情境。本文列出实验与实践考核的关键要求，采用中英对照讲解，帮助你高效复习、精准作答。

1. Where Practical Skills Are Assessed | 实践技能的考查位置

CAIE Psychology 9990 does not require a separate practical submission: your experimental understanding is tested inside written papers, especially Paper 2 Research Methods. You will meet data-based questions, scenario-based design questions, and questions that ask you to evaluate a study as if you were the researcher planning or improving it.

CAIE 心理学 9990 不需要单独提交实践报告：实验理解在书面试卷中考查，尤其是 Paper 2 Research Methods。你会遇到数据题、情境设计题，以及要求你像研究者一样规划或改进研究的评估题。

Paper 2 is therefore not just about memorising terms; it is a practical reasoning paper. You must be able to identify variables, propose controls, choose a design, handle ethics, interpret descriptive statistics, and judge whether a claim is supported by the data.

因此 Paper 2 不只是背诵术语，而是一份实践推理卷。你必须能识别变量、提出控制措施、选择实验设计、处理伦理问题、解释描述统计，并判断数据是否支持某一结论。

2. Hypotheses and Variables | 假设与变量

A hypothesis must state the expected relationship or difference between the independent variable (IV) and the dependent variable (DV). You should be able to write an experimental/alternative hypothesis and a null hypothesis, and you must operationalise each variable so that it can be manipulated or measured in a replicable way.

假设必须说明自变量（IV）与因变量（DV）之间预期的关系或差异。你要会写实验假设/备择假设和零假设，并必须对每个变量进行操作化定义，使其能够以可复制的方式被操纵或测量。

For example, “Students who sleep for eight hours will recall more words from a 20-word list than students who sleep for four hours” is a directional experimental hypothesis. The IV is hours of sleep, operationalised as two conditions, and the DV is the number of words correctly recalled.

例如，“睡眠 8 小时的学生比睡眠 4 小时的学生能从 20 个单词表中回忆出更多单词”是一个方向性实验假设。自变量是睡眠时长，操作化为两个条件；因变量是正确回忆的单词数量。

H₁: Sleep condition will affect the number of words recalled.
H₀: There will be no significant difference in word recall between the two sleep conditions.

H₁：睡眠条件会影响回忆单词的数量。
H₀：两种睡眠条件在单词回忆数量上无显著差异。

Poor operationalisation is a common weakness. “Memory is better” is too vague, whereas “number of words correctly recalled after a five-minute delay” is precise enough to be tested and replicated.

操作化不当是常见弱点。“记忆力更好”过于模糊，而“五分钟后正确回忆的单词数量”则足够精确，可以被检验和复制。

3. Experimental Designs | 实验设计

You need to know independent measures, repeated measures and matched pairs designs, including their strengths, limitations and suitable remedies. The choice of design directly affects participant variables, order effects and the statistical test you later use.

你需要掌握独立组设计、重复测量设计和配对组设计，包括它们的优点、局限与合适的补救措施。设计的选择直接关系到被试变量、顺序效应以及之后使用的统计检验。

Design	Key Feature	Strength	Limitation
Independent measures	Different participants in each condition	No order effects	Participant variables may confound results
Repeated measures	Same participants in all conditions	Fewer participant variables	Order effects, fatigue, practice
Matched pairs	Different but matched participants	Controls key participant variables	Matching is time-consuming and imperfect

For repeated measures, counterbalancing or randomisation of condition order is essential to reduce order effects. For independent measures, random allocation to conditions helps distribute participant variables evenly across groups.

对于重复测量设计，平衡顺序或随机安排条件顺序对减少顺序效应至关重要。对于独立组设计，将参与者随机分配到各条件有助于均匀分散被试变量。

4. Controls, Standardisation and Demand Characteristics | 控制、标准化与需求特征

A well-controlled experiment uses standardised procedures, standardised instructions and, where relevant, a control group or placebo condition. Standardisation means that every participant experiences the same environment, materials, timings and instructions, apart from the IV manipulation.

控制良好的实验使用标准化程序、标准化指导语，并在相关时设置对照组或安慰剂条件。标准化意味着除自变量操作外，每个参与者都经历相同的环境、材料、时间和指导语。

Watch out for demand characteristics: participants may change their behaviour because they guess the aim of the study. Investigator effects can also influence results through expectations. Single-blind and double-blind procedures reduce these risks, as does using a pilot study to refine instructions and materials.

要注意需求特征：参与者可能因为猜到研究目的而改变行为。研究者效应也会通过期望影响结果。单盲和双盲程序可以降低这些风险，使用试点研究完善指导语和材料也有帮助。

Standardised instructions | 标准化指导语
Constant environmental conditions | 恒定的环境条件
Random allocation or counterbalancing | 随机分配或顺序平衡
Control group, placebo or baseline condition | 对照组、安慰剂或基线条件
Single-blind or double-blind procedure | 单盲或双盲程序
Pilot study to identify confusion | 通过试点研究发现歧义

When evaluating a study, do not simply say “it lacked control”. Specify exactly what should have been controlled, how it might have threatened validity, and what practical control would solve the problem.

评估研究时，不要只说“缺乏控制”。要具体指出本应控制什么、它如何威胁效度，以及哪种实际控制措施可以解决这一问题。

5. Sampling Methods and Representativeness | 取样方法与代表性

You may be asked to recommend a sampling method for a planned experiment or to evaluate the sample used in a published study. The main methods in the CAIE syllabus are random, stratified, opportunity and volunteer/self-selected sampling.

你可能会被要求为计划中的实验推荐取样方法，或评估已发表研究中使用的样本。CAIE 大纲中的主要方法有随机取样、分层取样、机会取样和志愿者/自选取样。

Sampling Method	How It Works	Main Issue
Random	Every member has an equal chance of selection	Can be difficult in practice; may still be unrepresentative by chance
Stratified	Population is divided into strata, then sampled proportionally	More representative but time-consuming
Opportunity	Uses whoever is available	Convenient but often biased
Volunteer/self-selected	Participants choose to take part	May attract a particular personality type

A representative sample improves generalisability, but practical constraints often force researchers to use opportunity sampling. In your answers, link the sampling method to the target population and to the conclusions that can be drawn.

有代表性的样本能提高研究结论的可推广性，但实际限制常迫使研究者使用机会取样。作答时，要将取样方法与目标总体以及可推断的结论联系起来。

6. Ethical Considerations | 伦理考量

Ethics are assessed in almost every practical design question. You must know the core principles: informed consent, protection from harm, privacy and confidentiality, right to withdraw, avoidance of deception where possible, and full debriefing.

几乎所有实践设计题都会考查伦理。你必须掌握核心原则：知情同意、免受伤害、隐私与保密、随时退出权、尽可能避免欺骗，以及完整的事后解释。

Informed consent: participants should know the aims and procedures before agreeing. | 知情同意：参与者在同意前应了解目的与程序。
Protection from harm: psychological and physical safety must be prioritised. | 免受伤害：必须优先保障心理与身体安全。
Privacy and confidentiality: personal data must be anonymised and protected. | 隐私与保密：个人数据必须匿名并受到保护。
Right to withdraw: participants can leave at any time without penalty. | 退出权：参与者可随时退出且不受到惩罚。
Deception: avoid unless necessary; if used, debrief fully afterwards. | 欺骗：除非必要否则避免；若使用，事后必须充分解释。
Debriefing: explain the true aim and offer support or withdrawal of data. | 事后解释：说明真实目的，提供支持或允许撤回数据。

When proposing an experiment, you should say how you would obtain consent, protect data, minimise stress and debrief participants. When evaluating a classic study, apply the same ethical standards but also acknowledge the historical context and current guidelines.

在提出实验方案时，应说明如何获得知情同意、保护数据、减少压力并进行事后解释。评估经典研究时，既要应用相同的伦理标准，也要承认历史背景与现行指南的差异。

7. Data Collection Techniques | 数据收集技术

Practical questions may ask you to choose or justify a data collection technique. Common techniques include laboratory experiments, field experiments, natural experiments, questionnaires, interviews, observations and psychometric tests. Each method has different strengths for validity, reliability and ethics.

实践题可能会要求你选择或论证一种数据收集技术。常见技术包括实验室实验、现场实验、自然实验、问卷、访谈、观察和心理测量。每种方法在效度、信度和伦理方面都有不同的优势。

For example, a laboratory experiment offers high control over extraneous variables and can be replicated easily, but may lack ecological validity. A natural experiment has high external validity but low control and cannot establish cause and effect as strongly.

例如，实验室实验对额外变量控制力强、易于复制，但可能缺乏生态效度。自然实验外部效度高，但控制力弱，难以强有力地建立因果关系。

You should also know self-report techniques, behavioural observations and physiological measures. When designing your own investigation, choose a technique that directly measures your operationalised DV and minimises bias.

你还需要了解自我报告法、行为观察和生理测量。设计自己的研究时，要选择一种直接测量已操作化因变量、并能最大限度减少偏差的技术。

8. Descriptive Statistics and Data Presentation | 描述统计与数据呈现

You must be able to calculate and interpret measures of central tendency (mean, median, mode) and measures of dispersion (range, standard deviation). You should also know which measure is most appropriate for different levels of measurement and for skewed data.

你必须会计算和解释集中趋势量数（平均数、中位数、众数）和离散量数（全距、标准差）。还要知道对于不同测量水平和偏态数据，哪种量数最合适。

Mean: x̄ = Σx ÷ n

Standard deviation: s = √[Σ(x − x̄)² ÷ (n − 1)]

The mean uses all scores but is sensitive to outliers. The median is more robust for skewed distributions, while the mode is the only measure suitable for nominal data. The standard deviation shows how spread out scores are around the mean; a larger value indicates greater variability.

平均数使用所有分数但对异常值敏感。中位数对偏态分布更稳健，而众数是唯一适合称名数据的量数。标准差反映分数围绕平均数的离散程度；数值越大，变异性越大。

Data can be presented in tables, bar charts, histograms, or scattergraphs. A bar chart is used for discrete categories, a histogram for continuous interval data, and a scattergraph for correlational analysis. Always label axes and include units where relevant.

数据可以用表格、条形图、直方图或散点图呈现。条形图用于离散类别，直方图用于连续等距数据，散点图用于相关分析。务必标注坐标轴并在相关处注明单位。

9. Inferential Statistics and Significance | 推断统计与显著性

Inferential statistics help you decide whether the results are likely to reflect a real effect or to have occurred by chance. A null hypothesis is tested, and a probability value (p value) is calculated. In psychology the conventional significance level is p < 0.05.

推断统计帮助你判断结果是否可能反映真实效应，还是偶然发生。检验零假设并计算概率值（p 值）。心理学通常采用 p < 0.05 的显著性水平。

Type I error: rejecting a true null hypothesis; false positive. | I 类错误：拒绝真实零假设；假阳性。
Type II error: failing to reject a false null hypothesis; false negative. | II 类错误：未拒绝错误零假设；假阴性。

The choice of statistical test depends on the level of measurement, the experimental design, and whether the study tests a difference or a correlation. Below is a simplified guide for commonly used CAIE tests.

统计检验的选择取决于测量水平、实验设计以及研究检验的是差异还是相关。下面是 CAIE 常用检验的简化指南。

Data Type	Design	Test
Nominal	Independent measures	Chi-square
Ordinal	Independent measures	Mann-Whitney U
Ordinal	Repeated/matched	Wilcoxon signed-rank
Ordinal/interval correlation	Related pairs	Spearman rₛ

When writing about significance, do not only state “p < 0.05". Explain what that means: there is less than a 5% probability that the observed result occurred by chance, assuming the null hypothesis is true, so we reject the null hypothesis.

写显著性时，不要只写“p < 0.05”。要解释其含义：假设零假设为真，观察到的结果由偶然因素产生的概率小于 5%，因此我们拒绝零假设。

10. Validity, Reliability and Generalisability | 效度、信度与可推广性

Practical evaluation questions reward precise use of validity and reliability. Internal validity asks whether the IV really caused the change in the DV. External validity asks whether findings can be generalised beyond the study setting, sample and time.

实践评估题奖励对效度和信度的准确使用。内部效度关注自变量是否真的引起了因变量的变化。外部效度关注研究结果能否推广到研究情境、样本和时间之外。

Internal validity | 内部效度：control, cause and effect, confounding variables.
Ecological validity | 生态效度：task and setting reflect real life.
Population validity | 总体效度：sample represents the target population.
Temporal validity | 时间效度：findings hold over time.

Reliability means consistency. A measure is reliable if it gives similar results under consistent conditions. Inter-rater reliability is relevant for observations, while test-retest reliability is relevant for questionnaires or psychometric tests.

信度指一致性。如果一种测量在一致条件下给出相似结果，就具有信度。评分者间信度适用于观察研究，重测信度适用于问卷或心理测验。

To improve validity, standardise procedures, use a representative sample and ensure tasks reflect the real-world behaviour being studied. To improve reliability, use clear operational definitions, train observers and pilot the materials.

提高效度的方法是标准化程序、使用代表性样本并确保任务反映所研究的现实行为。提高信度的方法是使用清晰的操作定义、培训观察者并对材料进行试点。

11. Common Pitfalls in Practical Responses | 实践答题常见失分点

Many students lose marks because their answers are generic rather than context-specific. Avoid these common mistakes when completing design and evaluation questions.

很多学生因为答案笼统、脱离情境而失分。在完成设计题和评估题时，要避免以下常见错误。

Writing a hypothesis without operationalising the IV and DV | 写假设时未操作化自变量和因变量
Confusing extraneous variables with confounding variables | 混淆额外变量与混淆变量
Describing a control without explaining what it controls or why | 只描述控制措施却不说明控制什么或为何控制
Ignoring ethics in a design answer | 在设计答案中忽略伦理
Giving a statistical test without justifying its choice | 给出统计检验却不论证选择理由
Evaluating a study with vague phrases such as “low validity” | 用“效度低”等模糊表述评估研究

For each limitation, offer a concrete improvement. Instead of “the sample was biased”, say “the sample was drawn from one school using opportunity sampling, so it may not represent the wider population; using stratified sampling by age and school type would improve population validity”.

每一个局限都要提出具体改进。不要只说“样本有偏差”，而应说“样本通过机会取样来自一所学校，可能无法代表更广泛总体；按年龄和学校类型进行分层取样会提高总体效度”。

12. Exam Strategy for Experimental and Practical Questions | 实验与实践题考试策略

In the exam, read the practical scenario carefully and identify the IV, DV, design, sample and procedure before writing. Use the command word to shape your response: describe, explain, design, or evaluate.

考试

Published by TutorHao | A-Level Psychology Revision Series | aleveler.com

更多咨询请联系16621398022（同微信）

A-Level CAIE Psychology: Experimental and Practical Assessment Essentials | A-Level CAIE 心理学：实验/实践考核要点