📚 AS Cambridge Psychology: Experimental & Practical Assessment Essentials | AS剑桥心理学：实验与实践考核要点
AS Cambridge Psychology Paper 2 (Research Methods) tests your ability to think like a scientific researcher. You must design investigations, handle variables, analyse data, and critique studies. This article synthesises the essential experimental and practical knowledge you need to score highly, from formulating hypotheses to evaluating reliability and validity. Every section pairs an English explanation with its Chinese equivalent, ensuring dual-language clarity.
AS剑桥心理学试卷二（研究方法）考查你是否具备科学研究者般的思维。你需要设计研究、处理变量、分析数据并评论实验。本文综合了从提出假设到评估信度与效度的核心实验与实践知识，助你夺取高分。每个部分均提供中英文对照解释，确保双语清晰。
1. Types of Experiments | 实验类型
Cambridge AS candidates must distinguish between laboratory, field, natural, and quasi-experiments. A laboratory experiment takes place in a highly controlled environment, allowing the researcher to manipulate the independent variable (IV) and control confounding variables strictly. Field experiments occur in a natural setting, where the IV is still manipulated, but extraneous variables are harder to control. Natural experiments exploit a naturally occurring change in the IV, so the researcher does not manipulate it directly—quasi-experiments likewise use a pre-existing IV (such as gender or age group) without random allocation.
剑桥AS考生必须区分实验室实验、现场实验、自然实验和准实验。实验室实验在高度受控的环境中进行，研究者可以操纵自变量并严格控制混淆变量。现场实验发生在自然环境中，研究者依然操纵自变量，但额外变量更难控制。自然实验利用自变量的自然变化，研究者并不直接操纵；准实验同样基于先定的自变量（如性别或年龄组）且没有随机分配。
Each type carries distinct strengths, such as high internal validity in lab experiments and high ecological validity in field or natural experiments. However, lab experiments may lack mundane realism, while natural and quasi-experiments make causal inference difficult due to reduced control. Examiners frequently ask you to propose a suitable experiment type for a given scenario and justify your choice.
每种类型各有优缺点，例如实验室实验内部效度高，而现场实验和自然实验生态效度高。然而，实验室实验可能缺乏现实性，自然实验和准实验则因控制不足难以推断因果关系。考官常要求你为给定情境提出合适的实验类型并说明理由。
2. Hypotheses and Operationalisation | 假设与操作化
Every investigation begins with a clearly stated hypothesis. An experimental or alternative hypothesis predicts the effect of the IV on the DV. It can be directional (one-tailed), specifying the expected direction of the difference, or non-directional (two-tailed), simply stating that a difference exists. You must also formulate a null hypothesis, which asserts that any observed difference is due to chance and there is no true effect.
每项研究都始于清晰陈述的假设。实验假设或备择假设预测自变量对因变量的影响，它可以是方向性（单尾）的，明确差异的预期方向；也可以是非方向性（双尾）的，仅陈述存在差异。你还必须提出零假设，主张任何观察到的差异都源于偶然且没有真实效应。
Operationalisation is critical: to make variables measurable, you need to define exactly how the IV will be manipulated and how the DV will be recorded. For instance, if you are testing whether caffeine improves memory, ‘caffeine’ could be operationalised as a 200 mg caffeine tablet versus a placebo, and ‘memory’ as the number of words correctly recalled from a 20-word list. Without operational precision, replication becomes impossible and marks are lost.
操作化至关重要：要使变量可测量，你需要准确定义如何操纵自变量以及如何记录因变量。例如，若测试咖啡因是否提升记忆力，“咖啡因”操作化为200毫克咖啡因片与安慰剂对比，“记忆力”操作化为从20词列表中正确回忆的单词数。缺乏操作精确性，研究无法复制，考试中也会失分。
3. Research Designs and Controls | 研究设计与控制
Three core experimental designs appear repeatedly in AS assessments: independent groups design, repeated measures design, and matched pairs design. Independent groups use different participants in each condition, which avoids order effects but may suffer from participant variables. Repeated measures expose the same participants to all conditions, controlling individual differences but risking order effects such as practice and fatigue.
AS考试中反复出现三种核心实验设计：独立组设计、重复测量设计和匹配对设计。独立组设计在不同条件下使用不同被试，避免了顺序效应，但可能受到被试变量的影响。重复测量让同一批被试接受所有条件，控制了个体差异，但可能出现练习效应和疲劳效应等顺序效应。
Matched pairs design attempts to combine the best of both worlds: participants are pre-matched on a relevant characteristic (e.g., IQ) and then each member of the pair is assigned to a different condition. This reduces participant variables and removes order effects, though it is time-consuming and difficult to match perfectly. Control of confounding variables is achieved through random allocation (in independent groups), counterbalancing (in repeated measures), standardised procedures, and controlled environmental settings.
匹配对设计试图结合两者优点：被试按相关特征（如智商）进行预先匹配，然后将每对成员分别分配到不同条件。这减少了被试变量并消除了顺序效应，但耗时且难以完美匹配。通过随机分配（独立组）、平衡法（重复测量）、标准化程序以及受控环境来控制混淆变量。
4. Sampling Methods | 抽样方法
The way you select participants influences the generalisability of findings. Random sampling gives every individual in the target population an equal chance of selection, minimising bias, but it is often impractical. Stratified sampling divides the population into strata and selects proportionally, ensuring representation of subgroups. Opportunity sampling uses readily available individuals—it is convenient but prone to researcher bias.
被试的选择方式影响研究结果的普适性。随机抽样让目标群体中每个个体都有均等中选机会，最大程度减少偏差，但实际操作中常难以实现。分层抽样将总体分为不同层次并按比例选择，保证亚群体的代表性。机会抽样使用易得个体，方便快捷但易受研究者偏见影响。
Volunteer sampling, often through advertisements, may attract a particular profile of participant (e.g., highly motivated), leading to volunteer bias. In the exam, you must be ready to evaluate a study’s sample and suggest how a more representative sample could be obtained, linking it to the target population.
志愿者抽样常通过广告招募，可能吸引特定类型的被试（如高度积极者），导致志愿者偏差。考试中你必须能评估研究的样本，并建议如何获取更具代表性的样本，且要与目标群体相关联。
5. Data Collection Tools | 数据收集工具
AS Psychology expects you to understand both quantitative and qualitative data collection methods. Questionnaires can contain closed questions (generating numerical data) or open questions (producing qualitative insights). Structured interviews follow a fixed schedule, whereas unstructured interviews are more flexible but harder to analyse. Observations can be structured (using coding schemes) or unstructured, and the researcher may be overt or covert, participant or non-participant.
AS心理学要求你理解定量与定性数据收集方法。问卷可以包含封闭式问题（产生数字数据）或开放式问题（产生定性洞见）。结构化访谈遵循固定提纲，而非结构化访谈更为灵活但分析难度大。观察可以是结构化（使用编码方案）或非结构化，研究者可以是公开或隐蔽、参与或非参与。
For each method, consider standardisation, reliability, and ethical implications. Covert observations raise issues of deception and informed consent, while overt observations risk demand characteristics. Be prepared to craft a simple data collection tool in a design question, specifying question type, format, and how responses will be recorded.
对于每种方法，都要考虑标准化、信度和伦理问题。隐蔽观察引发欺骗和知情同意的问题，公开观察则可能存在需求特征。准备好在实验设计题中设计一个简单的数据收集工具，说明问题类型、格式以及如何记录反应。
6. Descriptive Statistics and Data Display | 描述统计与数据展示
You are expected to calculate and interpret measures of central tendency (mean, median, mode) and dispersion (range, standard deviation). The mean is the arithmetic average, sensitive to outliers; the median is the middle value in an ordered list; the mode is the most frequent score. The standard deviation quantifies the spread of scores around the mean – a larger SD indicates greater variability.
你需要计算并解释集中趋势（平均值、中位数、众数）和离散程度（全距、标准差）。平均值是算术平均数，对极端值敏感；中位数是有序数据的中间值；众数是出现频率最高的数值。标准差量化了分值在平均值周围的分散程度——标准差越大，变异越大。
Graphical displays must be chosen appropriately: bar charts are ideal for displaying frequencies or means from discrete categories; histograms display continuous frequency data; scattergrams illustrate correlations. When describing data, identify patterns, peak scores, and outliers before linking them to the hypothesis.
图表必须恰当选择：条形图适合显示离散类别的频数或均值；直方图呈现连续频数数据；散点图展示相关关系。描述数据时，先识别模式、峰值和异常值，再将其与假设联系起来。
σ = √( Σ (x − μ)² ÷ N )
This formula for standard deviation calculates the root mean square deviation from the mean, where σ = standard deviation, x = each value, μ = mean, N = number of data points. Use it accurately when required.
该标准差公式计算各项与均值之间离差的平方平均数的平方根，σ 表示标准差，x 为每个数值，μ 为均值，N 为数据点个数。必要时请准确运用。
7. Ethical Guidelines in Practice | 实践中的伦理准则
Ethics permeate every stage of research. The British Psychological Society (BPS) guidelines require informed consent, avoidance of deception (or debriefing after its use), right to withdraw, protection from psychological and physical harm, confidentiality, and privacy. In exam scenarios, you must identify ethical problems in a described study and suggest practical improvements.
伦理贯穿研究的每个阶段。英国心理学会（BPS）准则要求知情同意、避免欺骗（或事后说明）、可随时退出、保护免受身心伤害、保密以及隐私。在考试情境中，你必须识别所述研究中的伦理问题并提出实际改进建议。
For example, if a study uses a simulated prison environment, you might note that participants could experience distress, and suggest constant monitoring, prescreening for vulnerability, and full debriefing. When designing your own study, build in ethical safeguards from the outset—handling of personal data, anonymity of responses, and the right to omit answers.
例如，若一项研究使用模拟监狱环境，你可以指出被试可能遭受痛苦，并建议持续监测、预先筛选脆弱个体以及全面事后说明。在设计自己的研究时，从一开始就要建立伦理保障——个人数据的处理、匿名作答以及跳过问题的权利。
8. Evaluating Research: Validity and Reliability | 评估研究：效度与信度
Validity refers to whether a study measures what it intends to measure. Internal validity considers whether the IV truly caused the change in the DV, free from confounding variables. External validity includes ecological validity (generalising to real-life settings) and population validity (generalising across people). Standardisation and control improve internal validity, whereas naturalistic settings enhance ecological validity.
效度指研究是否测量了它想要测量的内容。内部效度考量因变量的变化是否确实由自变量引起，不受混淆变量干扰。外部效度包含生态效度（推广到现实生活环境）和人群效度（在不同人群中推广）。标准化和控制能提升内部效度，而自然情境则增强生态效度。
Reliability deals with consistency. If the same procedure is repeated under similar conditions, it should yield consistent results. Internal reliability can be checked through split-half methods. External reliability refers to replicability across time (test-retest) and across observers (inter-rater reliability). For observations, clear coding schemes and observer training raise inter-rater reliability.
信度涉及一致性。如果在相似条件下重复相同程序，应能得到一致结果。内部信度可通过分半法检验。外部信度指在时间推移（重测信度）和不同观察者之间（评分者信度）的可复制性。对于观察法，清晰的编码方案和观察者培训可提升评分者信度。
9. Common Pitfalls and Examiner Tips | 常见失分点与考官建议
Many candidates lose marks by confusing research designs or failing to operationalise variables adequately. Another frequent error is neglecting the null hypothesis: even in design questions, stating the null hypothesis demonstrates understanding of statistical testing. In 6- to 12-mark evaluation questions, avoid simply listing strengths and weaknesses—always link each point to the specific study, using terms like ‘internal validity’ and ‘participant variables’.
许多考生因混淆研究设计或未能充分操作化变量而失分。另一个常见错误是忽略零假设：即使在设计类题目中，陈述零假设也能证明你对统计检验的理解。在6至12分的评估题中，切忌简单罗列优缺点——始终将每一点与该具体研究联系起来，使用“内部效度”“被试变量”等术语。
When asked to draw or interpret a graph, label axes fully and include units. For data analysis, always give the appropriate measure of central tendency—if the data contains an outlier, the median may be more representative than the mean. Time management is crucial: read design scenarios carefully, underline key terms, and plan your answer structure before writing.
当需要绘制或解读图表时，要充分标注坐标轴并包含单位。分析数据时，务必选择恰当的集中趋势指标——若数据含有异常值，中位数可能比平均值更具代表性。时间管理至关重要：仔细阅读设计情境，标记关键词，写之前在脑中规划答案结构。
10. Applying Skills to Exam Questions | 考试题目中的技能应用
AS Paper 2 typically includes a mandatory design question where you create your own study from a prompt. Start by identifying the IV and DV, then state directional and null hypotheses. Next, describe the design type, sampling method, apparatus, and detailed procedure. Address ethical considerations and propose how data will be analysed—descriptive statistics first, perhaps a bar chart, and if required, a statement about which inferential test might be appropriate (though not calculated).
AS试卷二通常会有一道必做的设计题，让你根据提示自行创建研究。首先要确定自变量和因变量，再陈述方向性假设和零假设。然后描述设计类型、抽样方法、工具和详细程序。处理伦理考量，并提出如何分析数据——先描述统计，可能绘制条形图，如有需要，还可以指出哪种推断性检验合适（但不需要计算）。
You may also be asked to evaluate an existing study. Use the ‘VALID’ approach: think about Variables, Analysis, Limits, Internal/External validity, Demand characteristics, Debriefing/ethics. For each identified issue, provide a concrete improvement. Practice with past papers to internalise the structure, and always frame your answers in psychological terminology.
你也可能需要评估一项已有研究。采用“VALID”思路：思考变量、分析、局限性、内/外部效度、需求特征、事后说明/伦理。对每个识别出的问题，给出具体的改进措施。通过历年真题练习来内化结构，并始终用心理学专业术语组织答案。
Published by TutorHao | Psychology Revision Series | aleveler.com
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