📚 Year 12 CIE Psychology: Key Points for Experimental/Practical Assessment | CIE 心理学 Year 12 实验/实践考核要点
In the CIE AS Psychology (9990) syllabus, experimental and practical assessment skills form a central part of Paper 2: Research Methods. This component tests your ability to design, evaluate, and critique psychological experiments. Understanding the essential concepts such as variables, controls, ethics, and data analysis not only helps you score well on the exam but also equips you with the skills of a budding psychologist. This article walks you through the crucial aspects of experimental assessment, breaking down each area into clear, exam-relevant points.
在 CIE AS 心理学（9990）课程中，实验与实践评估技能是卷二《研究方法》的核心。这部分考查你设计、评价和批判心理学实验的能力。理解变量、控制、伦理和数据分析等关键概念不仅能帮助你在考试中取得好成绩，也能让你掌握一名初级心理学工作者必备的技能。本文带你逐一梳理实验评估的重点，将每个领域拆解为清晰与考试高度相关的内容。
Published by TutorHao | Psychology Revision Series | aleveler.com
1. The Role of Experiments in CIE Psychology | 实验在CIE心理学中的角色
Experiments are the only research method that allows psychologists to establish cause-and-effect relationships. By manipulating an independent variable (IV) and measuring its effect on a dependent variable (DV) while controlling extraneous variables, researchers can infer causation. In Paper 2, you are often asked to identify strengths and weaknesses of an experimental design, suggest improvements, or even design a simple experiment yourself. Practical assessment tasks, whether done in class or described in exam scenarios, are designed to evaluate your grasp of this logic.
实验是唯一能让心理学家确立因果关系的研究方法。通过操纵自变量（IV）并测量其对因变量（DV）的影响，同时控制额外变量，研究者可以推断出因果。在卷二里，你常常需要指出一个实验设计的优点和缺点，提出改进建议，甚至自行设计一个简单的实验。课堂实践评估或在考试情境中描述的实操任务，目的都在于考查你掌握这一逻辑的程度。
2. Identifying Independent and Dependent Variables | 识别自变量与因变量
nd one DV. The IV is the factor that the researcher deliberately changes or manipulates, while the DV is the factor that is measured to see if it was affected by the IV. In exam questions, you must be able to pick them out from a scenario. Look for the phrase 'whether … affects …' – the IV usually comes after 'whether', and the DV after 'affects'. For example, 'whether drinking caffeine affects reaction time' n time. Pay attention to the levels of the IV: it must have at least two conditions, such as experimental and control.
每个实验至少有一个自变量和一个因变量。自变量是研究者有意改变或操纵的因素，因变量则是被测量以观察其是否受自变量影响的因素。在考试题目中，你必须能从情境里把它们识别出来。可以留意“是否……影响……”这样的表述——自变量通常出现在“是否”之后，因变量在“影响”之后。例如，“饮用咖啡因是否影响反应时间”中，自变量为咖啡因摄入，因变量为反应时间。注意自变量的水平：它至少要有两个条件，如实验组和控制组。
3. Operationalising Variables Clearly | 明确操作化变量
Operationalisation means defining the IV and DV in a way that makes them precise and measurable. You cannot simply say 'memory' or 'aggression'; you must specify exactly how you will produce the conditions of the IV (e.g. list of 20 words vs. list of 5 words) and how you will record the DV (e.g. number of words correctly recalled). On Paper 2, marks are awarded for clear operationalisation. An ambiguous variable description will weaken the entire experiment because it becomes impossible to replicate or judge validity.
操作化是指用精确、可测量的方式定义自变量和因变量。你不能只是说“记忆”或“攻击性”，而必须具体说明如何形成自变量的各个条件（例如，呈现20个词的词表与5个词的词表），以及如何记录因变量（例如，正确回忆的词数）。在卷二里，清晰的操作化能得分。含糊的变量描述会削弱整个实验，因为它无法被复制，也难以判断其效度。
4. Formulating Testable Hypotheses | 构建可检验的假设
An experimental hypothesis can be directional (one-tailed) or non-directional (two-tailed). A directional hypothesis predicts the specific direction of the effect, e.g. 'Participants who drink caffeine will have faster reaction times than those who do not.' A non-directional hypothesis merely states there will be a difference, without specifying the direction. The null hypothesis always states there will be no significant difference (or relationship) between conditions. In your practical write-up or exam answer, you must state both the experimental and null hypotheses in fully operationalised form.
实验假设可以是方向性（单尾）或非方向性（双尾）的。方向性假设预测效应的具体方向，例如“饮用咖啡因的参与者反应时将快于未饮用者”。非方向性假设仅指出存在差异，而不指明方向。零假设总是说明条件之间没有显著差异（或关系）。在你的实践报告或考试答案中，必须以完全操作化的形式陈述实验假设和零假设。
5. Choosing an Experimental Design | 选择实验设计
Three main designs are used: independent measures, repeated measures, and matched pairs. In independent measures, different participants are allocated to each condition, which avoids order effects but may introduce participant variables. Repeated measures uses the same participants in all conditions, controlling for participant variables but risking order effects (practice, fatigue). Matched pairs design attempts to reduce participant variables by pairing similar individuals across conditions, but it is time-consuming and never perfect. Exam questions frequently ask you to justify the chosen design and discuss ways to overcome its weaknesses, such as counterbalancing for repeated measures or random allocation for independent measures.
主要使用三种设计：独立组设计、重复测量设计和匹配对设计。独立组设计将不同的参与者分配到各个条件，避免顺序效应但可能引入参与者变量。重复测量设计让同一批参与者经历所有条件，控制了参与者变量，但会带来顺序效应（练习效应、疲劳效应）。匹配对设计试图通过将相似的个体配对分配到不同条件来减少参与者变量，但耗时且难以完美。考试题目经常要求你论证所选设计的合理性，并讨论如何克服其弱点，例如对重复测量使用平衡法消除顺序效应，对独立组使用随机分配。
6. Controlling Confounding Variables | 控制混淆变量
Extraneous variables are any variables other than the IV that could affect the DV. When they systematically change alongside the IV, they become confounding variables and ruin internal validity. Common controls include standardisation of procedure, counterbalancing, random assignment, and using a control group. You must be able to identify potential confounds in a scenario and propose realistic controls. For example, if noise levels vary between conditions, suggest testing all participants in the same quiet room.
额外变量是除了自变量以外任何可能影响因变量的变量。当这些变量随自变量一起系统地变化时，它们就成了混淆变量，破坏了内部效度。常见的控制手段包括程序标准化、平衡法、随机分配以及使用对照组。你必须能够从情境中识别潜在的混淆变量，并提出切实可行的控制方法。例如，若不同条件下的噪音水平不一致，应建议让所有参与者在同一间安静的房间接受测试。
7. Sampling and Allocation of Participants | 参与者抽样与分配
The sample is the group of people who take part in the experiment. Sampling methods include opportunity, random, stratified, and volunteer sampling. Each has its strength and weakness in terms of representativeness and bias. After sampling, participants must be allocated to conditions. Random allocation is essential for independent measures to minimise pre-existing differences. For matched pairs, allocation is based on matching criteria. The exam may ask you to evaluate the sampling technique used and suggest how it affects generalisability (external validity).
样本是指参与实验的那群人。抽样方法包括机会抽样、随机抽样、分层抽样和志愿抽样。每种方法在代表性和偏差上各有优缺点。抽样之后，必须将参与者分配到各实验条件。对独立组设计而言，随机分配至关重要，以减少预先存在的差异。在匹配对设计中，分配依据匹配标准进行。考试可能会要求你评价所用的抽样技术，并说明它如何影响推广性（外部效度）。
8. Ethical Considerations in Practice | 实践中的伦理考虑
All psychological experiments must follow ethical guidelines: informed consent, right to withdraw, protection from harm, confidentiality, and debriefing. Even in classroom-based practicals, these principles apply. When evaluating an experiment on the exam, always consider whether any ethical issue is present. For instance, if participants were deceived about the purpose of the study, you must check whether the benefits outweigh the costs and whether they received a thorough debrief. Suggesting improvements like offering retrospective consent or ensuring anonymity shows higher-order evaluation skills.
所有心理学实验都必须遵循伦理准则：知情同意、随时退出权、免受伤害、保密和事后解释。即使在课堂实操中，这些原则也同样适用。在考试里评价一个实验时，永远要留意是否存在任何伦理问题。比如，若参与者被欺骗了关于研究目的，你必须权衡其益处是否大于代价，以及他们是否得到了充分的事后解释。提出改进建议，如取得事后同意或确保匿名，能展示出高阶的评价能力。
9. Standardised Procedures and Instructions | 标准化程序与指导语
To ensure reliability, the experiment must be conducted in the same way for all participants. This includes using identical instructions, materials, timings, and environments. Standardisation reduces experimenter bias and minimises situational variables. In your own practical work, write a step-by-step procedure so that someone else could replicate the experiment exactly. On Paper 2, you may be asked to identify procedural flaws, such as different wording of instructions for the two conditions, and recommend standardised scripts.
为确保信度，实验必须以完全一致的方式对所有参与者实施。这包括使用相同的指导语、材料、计时和环境。标准化能减少实验者偏差，并最小化情境变量。在你自己的实操中，应写出一步步的程序，以便他人可以精确复制该实验。在卷二里，你可能会被要求指出程序缺陷，例如两个条件使用了不同的指导语措辞，并建议采用标准化的脚本。
10. Collecting and Recording Data | 收集和记录数据
Quantitative data (numbers) is typical in experiments, though qualitative data can also be collected from post-experimental interviews. Record data in a clear raw data table with proper headings that include units. Avoid the common mistake of putting averages in the raw data table. The experimenter must also decide on the level of measurement: nominal, ordinal, or interval/ratio, as this determines which statistical test can be used. When evaluating, comment on whether the data collected truly reflects the DV as operationalised.
实验通常收集定量数据（数字），尽管也能通过实验后访谈收集定性数据。数据应记录在清晰的原始数据表中，表头要规范且包含单位。要避免把平均数放进原始数据表的常见错误。实验者还必须确定数据的测量层次：称名、顺序还是等距/比率，因为这决定了可以使用哪种统计检验。在进行评价时，要评论所收集的数据是否真正反映了已操作化的因变量。
11. Descriptive Statistics and Data Display | 描述统计与数据展示
For practical assessment, you need to calculate measures of central tendency (mean, median, mode) and measures of dispersion (range, standard deviation). Displaying data through bar charts, histograms, or scatterplots helps visualise patterns. Remember that a histogram is for continuous data and bars touch, whereas a bar chart is for categorical data and bars are separate. The exam may ask you to sketch or interpret a graph. Always provide clear labels, titles, and use an appropriate scale.
在实践评估中，你需要计算集中趋势量数（平均数、中位数、众数）和离中趋势量数（全距、标准差）。用条形图、直方图或散点图展示数据有助于可视化规律。记住，直方图适用于连续型数据，条形之间紧靠；条形图则适用于类别数据，条形之间分开。考试可能要求你绘制或解读图表。始终要提供清晰的标签、标题，并使用合适的刻度。
12. Evaluating and Improving an Experiment | 评估与改进实验
Evaluation is the most important high-order skill. You need to assess the validity (internal and external), reliability, ethicality, and generalisability of the experiment. Consider demand characteristics (cues that reveal the aim to participants) and social desirability bias. Use terms like 'mundane realism' to discuss whether the task resembles real life. To improve an experiment, suggest concrete changes: increase sample size, use a double-blind procedure, add a control group, or use counterbalancing. Link each suggestion back to a specific weakness you identified. This closes the loop of the experimental cycle and shows examiners your full command of research methodology.
评价是最重要的高阶技能。你需要评估实验的效度（内部与外部）、信度、伦理性和可推广性。要考虑需求特征（暴露实验目的的线索）和社会称许性偏差。使用像“日常真实感”这样的术语来讨论实验任务是否贴近现实生活。要改进一个实验，建议做出具体的改变：扩大样本量、采用双盲程序、增加对照组、或使用平衡法。每一条建议都应和你已识别的某个弱点挂钩。这样便闭合了实验循环的回路，向考官展示了你对研究方法论的全盘掌握。
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