📚 GCSE CAIE Psychology: Key Points for Experimental / Practical Assessment | GCSE CAIE 心理学：实验/实践考核要点
Mastering the experimental component of CAIE IGCSE Psychology requires a clear understanding of research methodology, from hypothesis writing to ethical practice. This guide distils the essential points you need to demonstrate in your practical assessment, whether designing a study, analysing data, or evaluating findings. Focus on applying these principles to achieve high marks in Paper 2 and beyond.
掌握CAIE IGCSE心理学的实验部分需要你对研究方法有清晰的理解，从撰写假设到伦理实践无一不包。本指南浓缩了你在实践考核中必须展示的关键点，无论是设计研究、分析数据还是评估发现。专注于应用这些原则，助你在Paper 2及后续评估中取得高分。
1. The Scientific Method in Psychology | 心理学中的科学方法
Psychology is a science because it relies on empirical evidence gathered through systematic observation, measurement, and experimentation. The scientific method involves formulating a testable hypothesis, designing a controlled procedure to gather data, analysing the results objectively, and drawing conclusions that can be replicated. This approach minimises bias and ensures findings are valid and reliable.
心理学是一门科学，因为它依赖于通过系统观察、测量和实验获得的实证证据。科学方法包括提出可检验的假设、设计受控程序收集数据、客观分析结果，并得出可被重复验证的结论。这种方法能最大程度减少偏差，确保研究发现有效且可靠。
In your practical work, you must demonstrate an understanding of this cycle. Always start with a research question derived from a psychological theory or previous research. Then, operationalise your variables and choose an appropriate design. Remember that a single study rarely "proves" anything; rather, it supports or challenges existing explanations.
在你的实践操作中，你必须展示对这一循环的理解。始终从源于心理学理论或先前研究的研究问题出发。然后，将变量操作化并选择恰当的设计。请记住，单次研究很少能"证明"任何事物；相反，它支持或挑战已有的解释。
2. Formulating Hypotheses | 提出假设
A hypothesis is a precise, testable statement predicting the relationship between variables. In IGCSE Psychology, you must be able to write both experimental (alternative) and null hypotheses. The experimental hypothesis (H₁) predicts a significant difference or correlation, while the null hypothesis (H₀) states there will be no significant difference or relationship, and any observed effect is due to chance.
假设是一个精确、可检验的陈述，预测变量之间的关系。在IGCSE心理学中，你必须能够写出实验假设（备择假设）和零假设。实验假设(H₁)预测存在显著差异或相关，而零假设(H₀)则陈述不存在显著差异或关系，任何观察到的效应都由偶然因素造成。
Use directional (one-tailed) hypotheses when previous research allows you to predict the direction of the result, e.g. 'Participants who drink caffeine will recall more words than those who do not.' Use non-directional (two-tailed) hypotheses when the direction is uncertain, e.g. 'There will be a difference in word recall between caffeine and no-caffeine conditions.'
当先前研究允许你预测结果的方向时，使用定向（单尾）假设，例如'饮用咖啡因的参与者能比未饮用者回忆更多单词。'当方向不确定时，则使用非定向（双尾）假设，例如'在咖啡因和无咖啡因条件下，单词回忆量存在差异。'
H₁: μ₁ ≠ μ₂ (non-directional) or μ₁ > μ₂ (directional)
3. Variables and Operationalisation | 变量与操作化
Every experiment has an independent variable (IV) – the factor you manipulate, and a dependent variable (DV) – the factor you measure. To make your study replicable, you must operationalise these variables, meaning you define them in terms of exactly how they will be manipulated or measured. For example, 'memory' is too vague; operationalise it as 'the number of words correctly recalled from a list of 20 after a 2-minute delay.'
每个实验都有自变量(IV)——你操控的因素，以及因变量(DV)——你测量的因素。为使研究可复制，你必须将这些变量操作化，即准确定义它们将如何被操控或测量。例如，"记忆"过于模糊；应操作化为"在2分钟延迟后，从一份包含20个单词的列表中正确回忆出的单词数量"。
Control variables (CVs) are all other factors that must be kept constant to prevent them from affecting the DV. Failing to control extraneous variables can introduce confounding variables, which ruin the internal validity of your experiment. In your practical write-up, list at least three CVs (e.g., room temperature, time of day, instructions given) and explain how you kept them constant.
控制变量(CVs)是所有其他必须保持恒定的因素，以防止它们影响因变量。未能控制额外变量可能引入混淆变量，从而破坏实验的内部效度。在你的实践报告中，列出至少三个控制变量（例如，室温、一天中的时段、提供的指导语），并解释你是如何让它们保持恒定的。
4. Experimental Designs | 实验设计
Choosing the right experimental design is crucial. The three main designs are independent measures, repeated measures, and matched pairs. Each has strengths and limitations, and your ability to justify your choice and suggest improvements is frequently assessed.
选择合适的实验设计至关重要。三种主要设计是独立组设计、重复测量设计和配对组设计。每种设计都有其优势和局限，你能否为其选择提供理由并提出改进建议，通常是考核的重点。
| Design | 设计 | Description | 描述 | Strength | 优势 | Limitation | 局限 |
|---|---|---|---|
| Independent measures | 独立组 | Different participants in each condition. | 每种条件下使用不同参与者。 | No order effects. | 无顺序效应。 | Participant variables may differ between groups; more participants needed. | 组间可能存在参与者变量差异；需要更多参与者。 |
| Repeated measures | 重复测量 | The same participants take part in all conditions. | 同一批参与者接受所有条件。 | Controls participant variables; fewer participants needed. | 控制了参与者变量；需要较少参与者。 | Order effects (practice, fatigue) can occur. | 可能出现顺序效应（练习效应、疲劳效应）。 |
| Matched pairs | 配对组 | Participants are matched on a key trait, then split into conditions. | 根据关键特质将参与者配对，然后分入不同条件。 | Reduces participant variables without order effects. | 无需担心顺序效应即可减少参与者变量。 | Time-consuming and difficult to match perfectly. | 耗时且难以完美匹配。 |
To deal with order effects in a repeated measures design, you should use counterbalancing: half the participants do condition A then B, the other half do B then A. This spreads the order effects evenly across conditions.
为了应对重复测量设计中的顺序效应，你应使用抵消平衡法：一半参与者先接受条件A再接受B，另一半先B后A。这能将顺序效应均匀地分散到各条件中。
5. Sampling Methods | 抽样方法
The sample is the group of people who take part in your study, drawn from a target population. Your sampling method determines how representative your sample is, which in turn affects the generalisability of your findings. The main techniques include opportunity, random, systematic, stratified, and volunteer sampling.
样本是参与研究的人群，从目标总体中抽取。你的抽样方法决定了样本的代表性，进而影响研究结果的推广性。主要方法包括机会抽样、随机抽样、系统抽样、分层抽样和志愿者抽样。
Opportunity sampling uses whoever is available (e.g., classmates), which is quick but often biased. Random sampling gives every member of the population an equal chance, reducing bias, but it is difficult to achieve in practice. Stratified sampling ensures subgroups are proportionally represented, which is ideal but complex. Always name your sampling method and evaluate its impact on the study.
机会抽样选用任何可找到的人（例如同班同学），快捷但常有偏差。随机抽样让总体中每个成员都有同等被选中的机会，能减少偏差，但在实践中难以实现。分层抽样确保各子群按比例代表，是最理想但也最复杂的方法。始终要指明你所用的抽样方法，并评估其对研究的影响。
6. Ethical Guidelines | 伦理准则
All psychological research must follow ethical principles to protect participants from harm. For your IGCSE practical, you need to know the British Psychological Society (BPS) guidelines: informed consent, right to withdraw, protection from physical and psychological harm, confidentiality, and avoidance of deception (unless justified and followed by a thorough debrief).
所有心理学研究都必须遵循伦理原则，以保护参与者免受伤害。针对你的IGCSE实践，你需要了解英国心理学会(BPS)指南：知情同意、退出权、免受身心伤害、保密、以及避免欺骗（除非有充分理由并在事后进行彻底的情况说明）。
When planning an experiment, state how you will obtain consent (e.g., a signed form, and for under-16s, also parental consent). Explain that participants can leave at any time without penalty. Use numbers or codes instead of names to ensure confidentiality. If you use mild deception, you must debrief participants fully after the study, revealing the true purpose and allowing them to withdraw their data.
在策划实验时，要说明你将如何获取同意（例如签订同意书，对于16岁以下者还需获得家长同意）。解释参与者可以在任何时候退出而不受惩罚。使用编号或代码代替姓名以确保保密。如果你使用了轻微的欺骗，必须在研究结束后对参与者进行完整的情况说明，揭示真实目的并允许他们撤回数据。
7. Data Collection Techniques | 数据收集方法
While the experiment is the key method, you may also need to discuss other techniques such as questionnaires, interviews, and observations. Each generates different types of data: quantitative (numbers) or qualitative (words, descriptions). For your practical, be clear about what data you collected and why.
尽管实验是核心方法，你也可能需要讨论其他方法，如问卷、访谈和观察。每种方法产生不同类型的数据：定量（数字）或定性（文字、描述）。在你的实践报告中，要清楚地阐述你收集了什么数据及其原因。
Questionnaires can gather self-report data quickly but may suffer from social desirability bias. Standardised instructions and closed questions increase reliability. Observations, if used, must be clearly categorised (e.g., tallying specific behaviours). Always pilot your materials to check for clarity.
问卷能快速收集自我报告数据，但可能会受到社会赞许性偏差的影响。标准化的指导语和封闭式问题能提高信度。如果使用观察法，必须清晰地分类（例如，对特定行为进行计数）。始终要预试你的材料以检查其清晰度。
8. Descriptive Statistics | 描述性统计
Once data is collected, you must summarise it using descriptive statistics. Know how to calculate the mean, median, mode, and range. The mean is the arithmetic average, sensitive to outliers; the median is the middle value, more robust; the mode is the most frequent score. The range gives a simple measure of spread (highest minus lowest).
收集数据后，你必须使用描述性统计进行总结。掌握如何计算平均数、中位数、众数和极差。平均数是算术平均值，对异常值敏感；中位数是中间值，更稳健；众数是出现最频繁的分数。极差提供了一个简单的离散度测量（最大值减最小值）。
Mean (M) = Σx / N | Range = Xmax − Xmin
You do not need to perform inferential statistics at IGCSE, but you should state which average is most appropriate for your data and justify why. For skewed distributions, the median is often better. Also, present individual scores or a summary table to support your calculations.
在IGCSE阶段你无需进行推断统计，但你应该说明哪一种平均数最适合你的数据并给出理由。对于偏态分布，中位数通常更合适。此外，要呈现个体分数或汇总表来支持你的计算。
9. Graphical Representations | 图表呈现
Suitable graphs make data easier to interpret. For experiments comparing conditions, a bar chart is standard. Plot conditions on the x-axis (IV) and the mean DV on the y-axis. Include a clear title, labelled axes with units, and consistent scales. Do not use line graphs for categorical IVs.
恰当的图表能让数据更容易解读。对于比较不同条件的实验，条形图是标准选择。在x轴上绘制条件（自变量），在y轴上绘制因变量的平均值。图表需包含清晰标题、带单位的轴标签，以及一致的刻度。不要为分类自变量使用折线图。
If you investigated a correlation, a scatter graph is appropriate, and you should comment on the pattern (positive, negative, or zero). Always reference your graph in the results section and briefly describe what it shows, but save detailed interpretation for the discussion.
如果你研究的是相关关系，散点图是合适的，你应该对模式（正相关、负相关或零相关）进行说明。始终在结果部分引用你的图表并简要描述它所显示的内容，但要把详细的解释留到讨论部分。
10. Evaluating Experiments | 实验评估
Evaluation is a high-mark skill. You need to discuss reliability, validity, and ethical issues. Reliability refers to consistency; a study is reliable if it can be replicated with similar results. Validity asks whether the study truly measures what it intends to measure (internal validity) and whether findings apply to real life (ecological validity).
评估是一项高分技能。你需要讨论信度、效度和伦理问题。信度指一致性；如果一项研究能被重复并得到相似结果，就可以说它信度高。效度则考察研究是否真正测量了它想要测量的东西（内部效度），以及研究发现是否适用于现实生活（生态效度）。
To improve reliability, standardise procedures and instructions. To enhance validity, reduce demand characteristics, use a more natural setting, or avoid an artificial task. Always link your evaluation to the specific study you conducted, not just generic points. For example, if you used a memory test with nonsense syllables, point out that this lacks mundane realism.
为了提高信度，可以标准化流程和指导语。为了增强效度，应减少需求特征、使用更自然的环境或避免人为任务。始终要将评估与你所做的具体研究联系起来，而不是只提泛泛的要点。例如，如果你使用的记忆测试是无意义音节，应指出这缺乏日常真实性。
11. Writing a Practical Report | 撰写实践报告
A clear report structure helps the examiner follow your reasoning. The standard sections are: Title, Abstract (brief summary), Introduction (background and hypothesis), Method (design, participants, apparatus, procedure), Results (descriptive stats and graphs), Discussion (explanation, evaluation, improvements), and References. Follow this order strictly.
清晰的报告结构有助于考官跟随你的思路。标准部分包括：标题、摘要（简短概述）、引言（背景和假设）、方法（设计、参与者、器材、步骤）、结果（描述统计与图表）、讨论（解释、评估、改进）以及参考文献。请严格遵循这一顺序。
In the method section, provide enough detail for someone else to replicate your study exactly. In the results, do not interpret – just present your findings. Save interpretation for the discussion, where you relate results back to your hypothesis and the theory from the introduction. Suggest at least two realistic improvements, such as using a larger sample or a double-blind procedure.
在方法部分，要提供足够细节，让别人能精确复制你的研究。在结果部分不要解释——仅呈现你的发现。将解释留到讨论部分，在这里你要将结果与你的假设以及引言中的理论联系起来。提出至少两条切实可行的改进措施，比如使用更大的样本或双盲程序。
12. Common Pitfalls to Avoid | 常见错误避免
Many students lose marks by confusing hypotheses with aims, failing to operationalise variables clearly, or ignoring the null hypothesis. Another frequent mistake is mixing up 'results' and 'discussion' – results are what you found, discussion is what it means. Also, avoid overstating your conclusion; use cautious language like 'suggests' rather than 'proves'.
许多学生因为混淆了假设与目标、未能清晰操作化变量或忽略了零假设而丢分。另一个常见错误是将"结果"与"讨论"混为一谈——结果是你发现了什么，讨论是这些发现意味着什么。另外，避免夸大结论；要使用诸如"表明"这样的谨慎用语，而不是"证明"。
Ensure you have identified at least one control variable that is psychological rather than purely physical, such as 'participants had normal sleep the night before' or 'standardised encouragement was given'. Always check your ethical compliance: did you offer a debrief? Did you destroy data upon request? Tick off each BPS guideline.
确保你至少识别出一个心理层面的控制变量，而非仅仅物理变量，例如"参与者前一夜睡眠正常"或"给予了标准化的鼓励"。始终检查你的伦理合规性：你提供了情况说明吗？你有没有按照要求销毁数据？逐一对照BPS的每一条准则。
Published by TutorHao | Psychology 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