Year 12 AQA Psychology: Experimental and Practical Assessment Key Points | AQA 心理学：实验与实践考核要点

📚 Year 12 AQA Psychology: Experimental and Practical Assessment Key Points | AQA 心理学：实验与实践考核要点

In Year 12 AQA Psychology, mastering the practical application of research methods is not only essential for the written exams but also for any internal experimental assessments. This guide breaks down every key concept you need to design, conduct, and evaluate psychological experiments with confidence, from defining variables to interpreting data ethically.

在 AQA 心理学第一年课程中，掌握研究方法的实际应用不仅对笔试至关重要，对任何校内实验考核也同样关键。本指南将逐一解析从定义变量到合乎伦理地解释数据所需的每一个核心概念，帮助你自信地设计、实施和评估心理学实验。

1. Understanding Variables: IV, DV, and Controls | 理解变量：自变量、因变量与控制

The independent variable (IV) is the factor you manipulate or compare. The dependent variable (DV) is what you measure to see if the IV had an effect. All other variables that might influence the DV are extraneous variables and must be controlled; if an extraneous variable varies systematically with the IV, it becomes a confounding variable, ruining the internal validity of your experiment.

自变量（IV）是你操纵或比较的那个因素。因变量（DV）是你测量的结果，用来判断 IV 是否有效。所有其他可能影响 DV 的变量都是额外变量，必须加以控制；如果某个额外变量与 IV 系统地一起变化，它就成了混淆变量，会破坏实验的内部效度。

Operationalisation is about turning these abstract concepts into measurable operations. For example, 'memory' could be operationalised as the number of words correctly recalled from a list of 20. Clear operationalisation of both IV and DV allows for accurate replication and reduces measurement ambiguity.

操作化就是把抽象概念转化为可测量的操作。例如，“记忆”可以操作化为从包含 20 个单词的列表中正确回忆出的单词数量。清楚地操作化 IV 和 DV，既便于准确复制，又能减少测量上的模糊性。

2. Formulating Hypotheses: Directional, Non-directional, and Null | 提出假设：方向性、非方向性与零假设

A hypothesis is a precise, testable statement predicting the relationship between the IV and the DV. A directional (one‑tailed) hypothesis states the expected direction of the effect, e.g., 'Participants who drink 200 mg of caffeine will recall more words than those who drink none.' A non‑directional (two‑tailed) hypothesis simply predicts a difference, e.g., 'There will be a difference in word recall between participants who consume caffeine and those who do not.'

假设是一个精确、可检验的陈述，用来预测 IV 与 DV 之间的关系。方向性（单尾）假设会说明预期的效应方向，例如“摄入 200 毫克咖啡因的参与者比不摄入者回忆出更多单词”。非方向性（双尾）假设则只预测存在差异，比如“摄入咖啡因与不摄入咖啡因的参与者在单词回忆上存在差异”。

In practical assessments, you will often also be required to write a null hypothesis, which states that any observed difference or correlation is due to chance and that there is no true effect in the population. Directional hypotheses are used when previous research strongly suggests a particular outcome; otherwise, a non‑directional hypothesis is safer.

在实践考核中，你还常常需要写出零假设，它声明任何观察到的差异或相关性只是由偶然因素造成的，在总体中并不存在真实效应。当先前研究强烈暗示某一特定结果时，使用方向性假设；否则，用非方向性假设更稳妥。

3. Experimental Designs: Independent Groups, Repeated Measures, Matched Pairs | 实验设计：独立组、重复测量与匹配对

Independent groups design allocates different participants to each condition of the IV. Its main strength is the absence of order effects; however, it requires more participants and can be compromised by individual differences between groups. Random allocation to conditions helps minimise this problem.

独立组设计将不同参与者分配到 IV 的各个水平中。其主要优点是没有顺序效应；但它需要更多参与者，而且不同组之间的个体差异可能会干扰结果。随机分配至各条件有助于减轻这一问题。

Repeated measures design uses the same participants in all conditions. This eliminates participant variables and uses fewer participants, but introduces order effects such as practice and fatigue. Counterbalancing (ABBA design) is essential to distribute these effects evenly across conditions.

重复测量设计在全部条件下使用同一批参与者。这消除了参与者变量，所需参与者也较少，但会引入练习效应和疲劳效应等顺序效应。采用平衡设计（ABBA 设计）把这些效应均匀地分配到各个条件是必不可少的。

Matched pairs design attempts to combine the strengths of both: participants are paired on key characteristics (e.g., age, IQ) and then each member of the pair is allocated to a different condition. This controls for participant variables without order effects, but it is time‑consuming and matching is never perfect.

匹配对设计试图结合两者的优点：根据关键特征（如年龄、智商）将参与者配对，然后把每对中的成员分别分配到不同条件。这样既控制了参与者变量，又避免了顺序效应，但过程耗时，而且匹配永远不可能完美。

4. Types of Experiments: Lab, Field, Natural, and Quasi | 实验类型：实验室、现场、自然与准实验

A laboratory experiment is conducted in a highly controlled, artificial environment. It allows precise control of extraneous variables, making replication easy, but it may lack ecological validity due to the artificial setting and task. Demand characteristics are also more likely.

实验室实验在高度受控的人为环境中进行。它能精确控制额外变量，易于复制，但由于环境和任务的人为性，可能缺乏生态效度。需求特征也更容易出现。

A field experiment takes place in a natural setting where the IV is still deliberately manipulated. It yields higher ecological validity, but control over extraneous variables is limited and ethical issues around informed consent often arise. A natural experiment simply records the effect of a naturally occurring IV (e.g., comparing children before and after the introduction of television in a remote area); the researcher does not manipulate the IV, so causal conclusions are weak.

现场实验在自然环境中进行，IV 仍然被有意操纵。它具有较高的生态效度，但对额外变量的控制有限，并且经常涉及知情同意的伦理问题。自然实验只是记录一个自然发生的 IV 的影响（例如，比较偏远地区电视引入前后儿童的变化）；研究者不操纵 IV，因此因果推断较弱。

Quasi‑experiments also have an IV that is an existing characteristic of the participants, such as gender or personality type. There is no random allocation, making it difficult to establish cause and effect, but they are valuable when ethical or practical constraints prevent manipulation.

准实验的 IV 也是参与者的既有特征，例如性别或人格类型。由于没有随机分配，很难确定因果关系，但当伦理或实际限制不允许操纵时，这种方法很有价值。

5. Controlling Extraneous Variables: Standardisation and Randomisation | 控制额外变量：标准化与随机化

Standardisation means keeping the procedure exactly the same for all participants, apart from the manipulation of the IV. Standardised instructions, timings, and environments ensure that any difference in DV is not due to variations in procedure.

标准化意味着除了操纵 IV 之外，整个过程对每位参与者完全一致。标准化的指导语、计时和环境能确保 DV 的任何差异不是由程序变化引起的。

Randomisation refers to using chance methods to determine the order of conditions or the allocation of participants, thereby minimising the impact of uncontrolled variables. For example, using a random number generator to assign participants to groups turns systematic bias into random error, which is less likely to confound results.

随机化指的是用机遇方法决定条件的顺序或参与者的分配，从而将未受控变量的影响降到最低。例如，用随机数字生成器把参与者分到各组，能把系统偏差转化为随机误差，减小混淆结果的可能性。

6. Demand Characteristics and Investigator Effects: Single-blind and Double-blind Procedures | 需求特征与研究者效应：单盲与双盲程序

Demand characteristics are cues in the research situation that lead participants to guess the purpose of the study and alter their behaviour accordingly. This threatens internal validity because participants may try to 'help' the researcher or deliberately act against the hypothesis.

需求特征是研究情境中那些让参与者猜出研究目的并据此改变行为的线索。这会威胁内部效度，因为参与者可能试图“帮助”研究者，或故意做出与假设相反的行为。

Investigator effects occur when a researcher's expectations or behaviour unintentionally influence participants' responses. In a single‑blind procedure, participants do not know which condition they are in, reducing demand characteristics. In a double‑blind procedure, neither the participants nor the researcher administering the experiment knows the condition, controlling for both demand characteristics and investigator effects.

广告 / Sponsored

当研究者的期望或行为无意中影响了参与者的反应时，就发生了研究者效应。单盲程序中，参与者不知道自己处于哪个条件，从而减少了需求特征。双盲程序中，参与者和主试都不知道分组情况，这能同时控制需求特征和研究者效应。

7. Sampling Techniques: Random, Stratified, Volunteer, and Opportunity | 抽样方法：随机、分层、自愿与机会抽样

Random sampling gives every member of the target population an equal chance of being selected, making the sample highly representative in theory. However, it is often impractical and may still produce a non‑representative sample by chance. Stratified sampling involves dividing the population into subgroups (strata) and then randomly sampling from each in proportion to their occurrence, which ensures good representation of key characteristics.

随机抽样让目标人群中的每个成员都有相等的被选机会，理论上样本代表性很高。但它常常难以实际操作，而且仍有可能碰巧产生不具有代表性的样本。分层抽样先将总体划分为子群体（层），再按比例从每层随机抽取，这样能确保关键特征的良好代表性。

Volunteer sampling relies on individuals who choose to join the study, often by responding to an advert. It is convenient but tends to attract a certain type of person, limiting generalisability. Opportunity sampling simply uses whoever is available and willing at the time; it is the most common in student practicals but is prone to bias and offers weak population validity.

自愿抽样依赖人们主动选择参与研究，通常是通过回应广告。这种方法方便，但容易吸引某一类人，限制了结论的推广性。机会抽样只是使用当时方便且愿意参与的人；这在学生实验中最为常见，但容易有偏差，总体效度较弱。

8. Ethical Considerations and BPS Guidelines | 伦理考量与 BPS 准则

Every psychological study must adhere to ethical principles such as informed consent, the right to withdraw, protection from harm, confidentiality, and deception avoidance. The British Psychological Society (BPS) Code of Ethics and Conduct provides a framework that must be followed in any practical assessment.

每一项心理学研究都必须遵守知情同意、退出权、免受伤害、保密和避免欺骗等伦理原则。英国心理学会（BPS）的伦理行为准则为所有实践评估提供了必须遵守的框架。

If deception is unavoidable, participants must be fully debriefed afterwards, and the deception must be justified by the scientific value of the study. For participants under 16, parental consent is required. In a practical exam situation, you must be able to explain how these ethical issues would be addressed, even if the experiment is only hypothetical.

如果欺骗不可避免，必须在事后对参与者进行彻底的事后解说，并且欺骗必须因研究的科学价值而具有正当理由。对于未满 16 岁的参与者，需获得家长同意。在实际考核中，即使只是假设性实验，你也必须能够说明如何处理这些伦理问题。

9. Pilot Studies and Refining Procedures | 试点研究及程序优化

A pilot study is a small‑scale trial run of the actual experiment. It allows the researcher to test standardised procedures, check that the instructions are clear, identify ambiguous DV measures, and see whether timing is realistic. Any problems can then be fixed before the main study, saving time and resources.

试点研究是对真实实验进行的一次小规模试运行。它能让研究者检验标准化程序，检查指导语是否清晰，识别 DV 测量中模糊的地方，并判断时间安排是否合理。在主研究开始前解决这些问题，可以节省时间和资源。

In a practical assessment, describing a pilot study shows that you understand the importance of refining your design. You might mention how you would adjust the number of trials, reword the consent form, or alter the difficulty of the task. Even saying 'a pilot study would be needed to estimate the standard deviation for a power calculation' demonstrates deeper methodological awareness.

在实践考核中，描述试点研究表示你理解优化设计的重要性。你可以提到如何调整试次数量、修改同意书的措辞或改变任务难度。哪怕只是说“需要试点研究来估算标准差以便进行效能计算”，也能展现出更深的方法论意识。

10. Data Analysis: Descriptive Statistics and Graphs | 数据分析：描述统计与图表

Descriptive statistics summarise data. Measures of central tendency include the mean (the arithmetic average), median (the middle score), and mode (the most frequent score). The mean is the most sensitive as it uses all data points, but it is easily distorted by outliers; the median is better for skewed distributions.

描述统计用来概括数据。集中趋势的度量包括平均数（算术平均值）、中位数（中间值）和众数（出现最频繁的值）。平均数最灵敏，它使用了所有数据点，但容易被异常值扭曲；中位数更适合偏态分布。

Measures of dispersion tell us about spread: the range (highest minus lowest) is quick but crude; standard deviation is a more precise measure of how much scores deviate from the mean. When presenting data, bar charts are used for discrete categories, while histograms and frequency polygons suit continuous data; scattergrams show correlations.

离散趋势的度量告诉我们数据的分布情况：全距（最大值减最小值）计算快却很粗糙；标准差能更精确地衡量各分数偏离平均数的程度。在呈现数据时，条形图用于离散类别，而直方图和频数多边形适合连续数据；散点图则用来显示相关性。

Mean (M) = Σx ÷ N

Standard Deviation (s) = √[ Σ(x - M)² ÷ (N - 1) ]

11. Evaluating Experiments: Validity and Reliability | 评估实验：效度与信度

Internal validity asks whether the study truly measures what it set out to measure, without confounding. External validity concerns whether findings generalise to other settings (ecological validity), populations (population validity), and times (historical validity). A key threat to internal validity in experiments is the presence of confounding variables.

内部效度关心的是研究是否真正测量了它所想要测量的东西，并且没有混淆。外部效度则涉及研究结果能否推广到其他环境（生态效度）、人群（总体效度）和时代（历史效度）。实验的一个主要内部效度威胁就是混淆变量的存在。

Reliability refers to consistency. If the same procedure produces similar results on a second occasion (test‑retest) or across different observers (inter‑rater reliability), the measurement is reliable. In practical assessments, you might be asked to suggest ways to improve reliability, such as clearer operational definitions or more standardised protocols.

信度指的是结果的一致性。如果同一程序在第二次实施时（重测信度）或不同观察者之间（评分者信度）产生相似的结果，那么测量就是可信的。在实践考核中，你可能需要提出提高信度的方法，比如更清晰的操作化定义或更标准的程序。

Balancing internal and external validity is often a trade‑off. A highly controlled lab study may be strong on internal validity but weak on ecological validity. You should always mention this trade‑off when evaluating experimental designs.

内部效度和外部效度往往需要权衡。一个严格控制变量的实验室研究可能内部效度很高，但生态效度较低。在评估实验设计时，你应当始终提到这种权衡。

12. Common Pitfalls in Practical Assessments | 实践考核中的常见误区

One classic mistake is confusing the DV with how it is operationalised. For instance, 'memory' is not the DV – recall score is. Another is writing a hypothesis that is not testable, such as 'Music makes people happy.' Always ensure your hypothesis refers to measurable variables and predicted outcomes.

一个典型错误是把 DV 和它的操作化方式混为一谈。例如，“记忆”不是因变量，回忆得分才是。另一个是写出的假设无法检验，比如“音乐让人快乐”。务必确保你的假设提及了可测量的变量和预测的结果。

Students often overlook the importance of counterbalancing in repeated measures designs, or forget to specify which ethical safeguards are relevant. In data analysis, choosing the wrong graph type or using mean for nominal data is common. Finally, remember that in an exam, you are not expected to memorise experiments, but to apply your knowledge of methods to novel scenarios – so focus on the reasoning behind each decision.

学生常常忽略在重复测量设计中平衡顺序的重要性，或者忘记指出哪些伦理保护措施是相关的。在数据分析中，选错图表类型，或者对名义数据使用平均数，都是常见问题。最后记住，考试并不要求你背下具体实验，而是要求你将方法学知识应用到新情境中——所以把重点放在每个决策背后的推理上。

Published by TutorHao | Psychology Revision Series | aleveler.com

更多咨询请联系16621398022（同微信）