📚 Year 12 Edexcel Psychology: Experimental and Practical Assessment Key Points | Edexcel 心理学：实验与实践考核要点
Experimental and practical skills are not a separate topic in Edexcel Year 12 Psychology — they are woven into every question that asks you to design, evaluate, or interpret research. Mastering the key points of experimental design, data handling, and ethical practice will sharpen both your AS exam performance and the skills required for any internal practical assessment. This guide pulls together the essential knowledge you need in a straightforward, exam-focused way.
实验与实践技能并非 Edexcel 12 年级心理学中一个孤立的主题 —— 它渗透在每一道要求你设计、评估或解释研究的考题中。掌握实验设计、数据处理和伦理实践的关键要点，既能提升你的 AS 考试成绩，也能帮助你完成校内实践评估。本文以考试为导向，系统梳理你需要掌握的核心知识。
1. The Scientific Approach to Psychology | 心理学的科学方法
Psychology adopts the scientific method by formulating a testable hypothesis and collecting objective data under controlled conditions. This distinguishes it from everyday assumptions and allows researchers to establish cause-and-effect relationships.
心理学采用科学方法，通过提出可检验的假设并在控制条件下收集客观数据，从而区别于日常推测，使研究者得以确立因果关系。
An aim states the purpose of the study, while a hypothesis is a precise, testable prediction. Hypotheses can be directional (one-tailed), predicting the direction of an effect, or non-directional (two-tailed), simply predicting a difference or relationship.
研究目的陈述了研究的目标，假设则是精确、可检验的预测。假设可分为定向的（单尾），预测效应的方向，或非定向的（双尾），仅预测存在差异或关系。
Replicability, objectivity, and falsifiability are hallmarks of the scientific approach. This means procedures must be standardised so others can repeat the study, and the hypothesis must be capable of being disproven.
可重复性、客观性和可证伪性是科学方法的标志。这意味着程序必须标准化，以便他人能够重复研究，且假设必须能被证伪。
2. Variables and Operationalisation | 变量与操作化定义
The independent variable (IV) is the factor manipulated by the researcher to create two or more conditions. The dependent variable (DV) is the factor measured to assess the effect of the IV.
自变量是研究者操纵以创设两个或以上条件的因素，因变量则是测量的因素，用以评估自变量的效应。
Operationalisation means defining variables in clear, measurable terms. For example, 'memory' could be operationalised as the number of words correctly recalled from a list after a 2-minute delay, rather than a vague concept.
操作化定义指用清晰、可测量的术语界定变量。例如，“记忆力”可被操作化为延迟 2 分钟后从列表中正确回忆的单词数，而非一个模糊的概念。
Without careful operationalisation, replication becomes impossible, and internal validity is threatened. Every IV and DV in your practical work must be operationalised precisely.
缺乏细致的操作化定义，研究便无法重复，内部效度也会受损。在你的实践作业中，每一个自变量和因变量都必须精确地进行操作化定义。
3. Experimental Designs: Strengths and Weaknesses | 实验设计：优缺点对比
Independent groups design recruits different participants for each condition. This eliminates order effects but may suffer from participant variables (individual differences) that confound the results.
独立组设计为每个条件招募不同的被试，这消除了顺序效应，但可能因被试变量（个体差异）而混淆结果。
Repeated measures design uses the same participants in all conditions, which controls participant variables perfectly. However, order effects such as practice, boredom, or fatigue must be managed through counterbalancing or randomisation of conditions.
重复测量设计在所有条件下使用同一批被试，这完美地控制了被试变量。然而，必须通过平衡顺序或条件随机化来管理练习、厌烦或疲劳等顺序效应。
Matched pairs design attempts to gain the best of both worlds: different but similar participants are matched on key traits (e.g., IQ, age) and then assigned to different conditions. This reduces participant variables while avoiding order effects, though matching is time-consuming and never perfect.
匹配对设计试图兼得两者之长：不同的但相似被试在关键特质（如智商、年龄）上进行匹配，然后分配到不同条件。这减少了被试变量，又避免了顺序效应，尽管匹配过程耗时且无法达到完美。
When evaluating an experimental design, always link its strengths and weaknesses to the specific scenario and suggest a practical improvement.
在评估实验设计时，始终将其优缺点与具体情景相联系，并提出切实可行的改进方法。
4. Controlling Extraneous and Confounding Variables | 控制额外变量与混淆变量
An extraneous variable is any variable, other than the IV, that could affect the DV if not controlled. When an extraneous variable systematically varies with the IV, it becomes a confounding variable and ruins internal validity.
额外变量指除自变量外任何可能影响因变量的变量，若未加以控制，当额外变量与自变量共同系统变化时，就会变成混淆变量，破坏内部效度。
Standardisation of procedures, instructions, and timings keeps the experience consistent for all participants. Random allocation distributes participant variables evenly across conditions and is a key control in independent groups design.
程序、指导语和时间安排的标准化使所有被试的体验保持一致。随机分配将参与者变量均匀地分布到各个条件，是独立组设计中关键的控制手段。
Single-blind and double-blind procedures help control demand characteristics and experimenter bias. In a double-blind trial, neither participants nor the researcher administering the conditions know who is in which group.
单盲和双盲程序有助于控制需求特征和实验者偏见。在双盲试验中，被试和实施条件的研究者均不知道分组情况。
5. Sampling in Psychological Research | 心理学研究中的抽样方法
Random sampling gives every member of the target population an equal chance of being selected, which should yield a representative sample. In practice, it is difficult to achieve a truly random sample due to access and non-response.
随机抽样使目标总体中的每个成员都有同等被选中的机会，理论上能产生代表性样本。然而在实际操作中，由于接触途径和无应答问题，很难获得真正随机的样本。
Opportunity sampling selects participants who are readily available, such as students in a school. It is quick and convenient but often leads to a biased, unrepresentative sample.
机会抽样选取那些方便获得的人，例如中学里的学生。此法快捷方便，但往往导致样本有偏差且不具代表性。
Stratified sampling divides the target population into key subgroups (strata) and then draws a random sample from each in proportion to their occurrence in the population. This improves representativeness but requires detailed population data.
分层抽样将目标总体划分为主要子群（层），然后按各层在总体中的比例从中随机抽取样本。这提升了代表性，但需要详尽的总体数据。
Volunteer (self-selected) sampling relies on participants responding to advertisements. Such samples may be motivated and cooperative, yet they are prone to volunteer bias — typically more motivated, with particular personality traits.
志愿者（自我选择）抽样依赖被试回应广告。这种样本积极性高、配合度好，但易产生志愿者偏差——他们通常动机更强，具有某种人格特质。
6. Ethical Guidelines and Risk Management | 伦理准则与风险管理
The British Psychological Society (BPS) Code of Ethics requires researchers to obtain informed consent, avoid deception where possible, protect participants from physical and psychological harm, debrief after the study, and ensure confidentiality.
英国心理学会伦理准则要求研究者获取知情同意、尽可能避免欺骗、保护被试免受身心伤害、研究后进行事后说明，并确保保密性。
If deception is necessary, it must be justified and followed by a thorough debrief. Participants should leave a study in the same state as they entered it, with no lasting distress.
若必须有欺骗，必须给出正当理由，并在之后进行详尽的解释。被试离开研究时应与进入时状态无异，不留持久困扰。
The right to withdraw enables participants to stop at any time without penalty. In practical assessments, you must state how this and all other ethical safeguards were implemented.
退出权允许被试在任何时候终止参与且不受惩罚。在实践考核中，你必须陈述如何落实这一权利及所有其他伦理保护措施。
7. Types of Experiments and Their Applications | 实验类型及其应用
Laboratory experiments are conducted in a controlled, artificial environment. They offer high internal validity and precise control, but low ecological validity, as behaviour may not mirror real life.
实验室实验在受控的人工环境中进行。它们内部效度高、控制精准，但生态效度低，因为行为可能无法反映真实生活。
Field experiments manipulate the IV in a natural setting (e.g., a train station or playground). They have higher ecological validity but less control over confounding variables, and ethical issues around consent often arise.
现场实验在自然情境中操纵自变量（如火车站或操场）。它们生态效度较高，但对混淆变量的控制较弱，且常涉及同意相关的伦理问题。
Natural experiments occur when the IV varies naturally (e.g., comparing groups before and after a law change). The researcher does not manipulate the IV, so causal conclusions must be drawn cautiously.
自然实验中，自变量自然变化（如比较法律变更前后的群体）。研究者不操纵自变量，因此推断因果关系时须谨慎。
Quasi-experiments examine a naturally occurring IV that is a characteristic of the person, such as gender or personality type. No random allocation is possible, limiting internal validity.
准实验研究人的某种特征作为自然发生的自变量，如性别或人格类型。无法随机分配，内部效度受限。
8. Data Collection and Descriptive Statistics | 数据收集与描述性统计
Quantitative data numerates behaviour, allowing statistical analysis and comparisons. Qualitative data captures the richness of experience through words or images, offering depth but being harder to analyse objectively.
定量数据对行为进行量化，便于统计分析和比较。定性数据通过文字或图像捕捉体验的丰富性，有深度但难以客观分析。
Measures of central tendency summarise the data. The mean uses all values but is sensitive to outliers. The median is the middle score, unaffected by extreme scores. The mode is the most frequent value, useful for nominal data.
集中趋势量数概括数据。均值使用所有数值但对异常值敏感，中位数是中间值，不受极端分数影响，众数是最常见值，适用于名义数据。
Measures of dispersion indicate spread. The range (highest minus lowest) is quick to compute but heavily influenced by a single extreme. The standard deviation (SD) measures the average deviation from the mean; a larger SD indicates greater variability.
离散量数表示数据的离散程度。极差（最大值减最小值）计算简单，但受单一极端值影响大。标准差测量各数据偏离均值的平均幅度；标准差越大，变异性越大。
Visual displays include bar charts for discrete categories, histograms for continuous grouped data, and scatterplots for correlations. Always label axes clearly and include a title.
视觉呈现包括用于离散类别的条形图、用于连续分组数据的直方图，以及用于相关性的散点图。坐标轴必须清晰标注，并包含标题。
9. Inferential Testing: Probability and Significance | 推论检验：概率与显著性
Inferential statistics help us decide whether the pattern we see in our sample reflects a real effect in the population or is due to chance. The null hypothesis (H₀) states there is no effect or relationship, while the alternative hypothesis states there is an effect.
推论统计帮助判断样本中观察到的模式究竟反映了总体的真实效应，还是偶然导致。零假设陈述没有效应或关系，备择假设则表明存在效应。
is the probability threshold for rejecting the null hypothesis. At p ≤ 0.05, there is a 5% probability that the result occurred by chance, so we accept the experimental hypothesis.
显著性水平（通常为 p ≤ 0.05）是拒绝零假设的概率阈值。当 p ≤ 0.05 时，结果出于偶然的概率为 5% 或更低，因此我们接受实验假设。
to reject a false null hypothesis. Choosing a stricter significance level (e.g., p ≤ 0.01) reduces Type I errors but increases Type II risk.
第一类错误（假阳性）指零假设为真时却拒绝了它。第二类错误（假阴性）指零假设为假时却未能拒绝它。选择更严格的显著性水平（如 p ≤ 0.01）可减少第一类错误，但会增加第二类错误的风险。
Choosing the correct test depends on the research design and level of measurement. Nominal data (categories) may use a Chi-squared or Sign test; ordinal data (ranked) often use Wilcoxon, Mann-Whitney U, or Spearman’s rho; interval/ratio data can use the related or unrelated t-test.
选择正确的检验取决于研究设计和测量层次。名义数据（类别）可用卡方或符号检验；顺序数据（排序）常使用 Wilcoxon、Mann-Whitney U 或 Spearman 秩相关；等距/等比数据可使用相关或无关 t 检验。
The calculated (observed) value is compared with a critical value from statistical tables. If the observed value is larger than (or smaller than, as directed) the critical value at the chosen significance level and degrees of freedom or N, the result is significant.
计算（观测）值与统计表中的临界值进行比较。若观测值在所选显著性水平及自由度或 N 下大于（或根据规则小于）临界值，则结果显著。
10. Ensuring Reliability and Validity | 确保信度与效度
Reliability refers to consistency. Internal reliability measures whether items within a test produce similar results; external reliability checks whether the same test gives consistent scores over time (test-retest) or between observers (inter-rater reliability).
信度指一致性。内部信度衡量测试内部各项目是否得到相近结果；外部信度检查相同测试在不同时间（重测信度）或不同观察者之间（评分者信度）是否得分一致。
Validity concerns accuracy — are we measuring what we intend to measure? Internal validity questions whether the IV really caused the change in the DV, free from confounding. External validity (including ecological and population validity) asks if findings can be generalised.
效度关乎准确性——我们是否在测量想测量的东西？内部效度质疑因变量的变化是否确实由自变量引起，且无混淆。外部效度（包括生态效度和总体效度）询问研究结果能否推广。
To improve reliability, standardise all procedures, provide clear instructions, and use pilot studies to refine measures. To improve validity, ensure careful operationalisation, use a representative sample, and control confounding variables through random allocation and counterbalancing.
为提高信度，应标准化全部程序、提供清晰指导语，并通过试点研究改进测量工具。为提高效度，应确保精细的操作化定义、使用代表性样本，并通过随机分配和平衡顺序控制混淆变量。
11. Writing the Lab Report and Communication | 实验报告撰写与沟通
A standard practical report follows the structure: Abstract (summary of aims, method, results, conclusion), Introduction (literature review and hypotheses), Method (design, participants, materials, procedure), Results (descriptive and inferential statistics), Discussion (explanation, relation to theory, limitations), References, and Appendices.
标准的实践报告遵循如下结构：摘要（目的、方法、结果、结论概述）、引言（文献回顾与假设）、方法（设计、被试、材料、程序）、结果（描述统计和推论统计）、讨论（解释、与理论关联、局限）、参考文献和附录。
Always write in the past tense and third person, avoiding personal pronouns. Tables and graphs must be numbered and titled, with clear reference to them in the text.
始终使用过去时与第三人称撰写，避免使用人称代词。图表必须有编号和标题，并在正文中明确引用。
In the discussion, explicitly link your findings back to the hypothesis and the background literature. Identify two or three specific weaknesses and suggest realistic modifications for future research.
在讨论中，明确将发现与假设及背景文献联系起来。指出两至三个具体的不足，并对未来研究提出切实的改进建议。
12. Practical Tips for Exam and Assessment Success | 考试与评估的实战技巧
When faced with a 'Design a study' question, work systematically: state an operationalised IV and DV, choose an appropriate design, describe the sample and sampling method, detail the procedure step by step, specify controls, and address ethics explicitly.
面对“设计一项研究”类题目时，系统地作答：陈述操作化定义的自变量和因变量，选择合适的实验设计，描述样本及抽样方法，逐步详述程序，明确控制措施，并清晰阐述伦理考虑。
In evaluation questions, use the P.E.E.L. structure (Point, Evidence, Explain, Link) to show why a methodological choice is a strength or a weakness in context. Always offer a concrete suggestion for improvement rather than simply criticising.
在评价类题目中，采用 P.E.E.L. 结构（观点、证据、解释、联系）说明为何某种方法选择在此情境中是优点或缺点。务必提出具体的改进建议，而非仅仅批评。
For data interpretation tasks, read the table or graph carefully, note means and SDs, and describe patterns clearly. If asked whether a difference is significant, refer to both the descriptive pattern and the reported inferential test result.
对于数据解读任务，仔细阅读表格或图形，注意均值和标准差，并清晰描述模式。若被问及差异是否显著，应同时提及描述性模式和所报告的推论检验结果。
Regular hands-on practice with designing mini-experiments, writing risk assessments, and running simple statistical tests on class data will build the fluency needed for exam questions. Use the key terms accurately — they are the vocabulary that earns high marks.
经常动手练习设计小实验、撰写风险评估、并对课堂数据进行简单的统计检验，能培养应对考题的流利度。准确使用关键术语——这些是赢得高分的词汇。
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