Psychology Research Methods & Experimental Design Essentials | 心理学研究方法与实验设计要点

📚 Psychology Research Methods & Experimental Design Essentials | 心理学研究方法与实验设计要点

Research methods form the backbone of psychology as a scientific discipline. Understanding how psychologists design experiments, collect data, and draw conclusions is essential for any student preparing for examinations in this subject.

研究方法构成了心理学作为一门科学学科的基石。理解心理学家如何设计实验、收集数据并得出结论，对于任何准备心理学考试的学生来说都是至关重要的。

---

1. The Scientific Method in Psychology | 心理学中的科学方法

The scientific method is a systematic approach to acquiring knowledge through observation, hypothesis testing, and empirical evidence. In psychology, this involves formulating a research question, developing a testable hypothesis, designing a study, collecting data, analysing results, and drawing conclusions that can be replicated.

科学方法是一种通过观察、假设检验和实证证据来获取知识的系统性途径。在心理学中，这包括提出研究问题、建立可检验的假设、设计研究、收集数据、分析结果并得出可复制的结论。

A good hypothesis must be falsifiable, meaning it can be proven wrong through experimentation. It should also be precise and operationalised — that is, variables must be defined in terms of how they will be measured.

一个好的假设必须是可证伪的，即可以通过实验证明其错误。它还应该是精确且可操作化的——也就是说，变量必须根据其测量方式来定义。

Hypothesis → Design → Data Collection → Analysis → Conclusion → Replication

假设 → 设计 → 数据收集 → 分析 → 结论 → 复制

---

2. Independent and Dependent Variables | 自变量与因变量

The independent variable (IV) is the factor that the researcher manipulates or changes systematically. The dependent variable (DV) is the outcome that is measured to see the effect of the IV. For example, if studying the effect of caffeine on reaction time, caffeine dosage is the IV and reaction time is the DV.

自变量（IV）是研究者系统地操纵或改变的因素。因变量（DV）是为了观察自变量效应而测量的结果。例如，研究咖啡因对反应时间的影响时，咖啡因剂量是自变量，反应时间是因变量。

Control variables are factors that are kept constant across all conditions to ensure that any observed effect is due to the IV alone. In a caffeine experiment, time of day, participant age, and prior sleep could all serve as control variables.

控制变量是在所有条件下保持恒定的因素，以确保观察到的任何效应仅由自变量引起。在咖啡因实验中，一天中的时间、参与者的年龄和先前睡眠情况都可以作为控制变量。

Extraneous variables are unwanted factors that could influence the DV if not controlled. Confounding variables go further — they systematically vary alongside the IV, making it impossible to determine which variable caused the effect.

额外变量是如果不加以控制就可能影响因变量的不想要的因子。混淆变量更为严重——它们与自变量一起系统性变化，使得无法确定是哪个变量导致了效应。

---

3. Experimental Designs | 实验设计类型

There are three main experimental designs in psychological research. In independent groups design, different participants are assigned to each condition of the IV. In repeated measures design, the same participants take part in every condition. In matched pairs design, participants are paired based on relevant characteristics, with each member of the pair assigned to a different condition.

心理学研究中有三种主要的实验设计。在独立组设计中，不同的参与者被分配到自变量的每个条件中。在重复测量设计中，相同的参与者参加所有条件。在匹配组设计中，参与者根据相关特征进行配对，每对中的一个成员被分配到不同条件。

• Independent groups: No order effects, but participant variables may differ between groups.
• 独立组：没有顺序效应，但参与者变量可能在组间不同。
• Repeated measures: Controls participant variables, but order effects and demand characteristics are risks.
• 重复测量：控制参与者变量，但存在顺序效应和要求特征的风险。
• Matched pairs: Balances participant variables and avoids order effects, but matching is difficult and time-consuming.
• 匹配组：平衡参与者变量并避免顺序效应，但匹配困难且耗时。

Counterbalancing is used in repeated measures designs to address order effects. Half the participants experience condition A then B, while the other half experience B then A, thereby distributing practice and fatigue effects evenly.

平衡法用于重复测量设计中以解决顺序效应。一半参与者先经历条件A再经历条件B，而另一半先经历B再经历A，从而均匀分布练习效应和疲劳效应。

---

4. Laboratory, Field, and Natural Experiments | 实验室、现场与自然实验

Laboratory experiments take place in a controlled environment where the researcher has high levels of control over extraneous variables. They offer high internal validity but may suffer from low ecological validity because the setting is artificial.

实验室实验在受控环境中进行，研究者对额外变量有高度控制。它们提供高内部效度，但由于环境人为，可能生态效度较低。

Field experiments are conducted in real-world settings such as schools or workplaces. The researcher still manipulates the IV but participants are in natural surroundings, increasing ecological validity. However, control over extraneous variables is reduced.

现场实验在真实世界环境中进行，如学校或工作场所。研究者仍然操纵自变量，但参与者处于自然环境中，从而提高了生态效度。然而，对额外变量的控制会降低。

Natural experiments involve the researcher taking advantage of a naturally occurring event or change. The IV is not manipulated by the researcher — for example, comparing children adopted at different ages. These allow study of variables that cannot be ethically manipulated, but causality is harder to establish.

自然实验是指研究者利用自然发生的事件或变化。自变量并非由研究者操纵——例如，比较不同年龄被收养的儿童。这些实验允许研究无法在伦理上操纵的变量，但因果关系更难确立。

---

5. Sampling Techniques | 抽样方法

Participants in psychological research are selected using various sampling methods. Random sampling gives every member of the target population an equal chance of being selected, producing the most representative sample. Opportunity sampling selects whoever is available at the time, which is convenient but often biased.

心理学研究中的参与者通过多种抽样方法选取。随机抽样赋予目标总体中每个成员同等的被选中机会，产生最具代表性的样本。机会抽样选择当时可获得的任何人，方便但常有偏差。

Method	Advantage	Disadvantage
Random | 随机	Representative | 代表性高	Difficult to obtain | 难以获得
Opportunity | 机会	Easy and quick | 简便快捷	Biased sample | 样本偏差
Systematic | 系统	Unbiased selection | 无偏选择	May miss patterns | 可能遗漏模式
Stratified | 分层	Reflects population proportions | 反映总体比例	Time-consuming | 耗时
Snowball | 滚雪球	Access to hidden populations | 接触隐蔽群体	Lacks generalisability | 缺乏普遍性

Volunteer sampling relies on participants who actively respond to advertisements. While ethically sound because participation is voluntary, this method may attract a particular type of person, creating a volunteer bias that threatens representativeness.

志愿者抽样依赖于主动响应广告的参与者。虽然因为参与是自愿的在伦理上合理，但这种方法可能吸引特定类型的人，产生志愿者偏差，威胁代表性。

---

6. Correlational Research | 相关研究

Correlational research examines the relationship between two or more variables without manipulating any of them. The correlation coefficient ranges from −1 to +1, where the sign indicates the direction of the relationship and the absolute value indicates its strength.

相关研究在不操纵任何变量的情况下检视两个或多个变量之间的关系。相关系数的范围从−1到+1，其中符号表示关系的方向，绝对值表示关系的强度。

r = −1 (perfect negative) → r = 0 (no relationship) → r = +1 (perfect positive)

r = −1（完全负相关）→ r = 0（无关系）→ r = +1（完全正相关）

It is crucial to remember that correlation does not imply causation. Two variables may be correlated because a third variable influences both, or because of a bidirectional relationship. For example, ice cream sales and drowning incidents are positively correlated, but eating ice cream does not cause drowning — warm weather drives both.

务必记住，相关不代表因果。两个变量可能相关是因为第三个变量同时影响两者，或因为存在双向关系。例如，冰淇淋销量与溺水事件正相关，但吃冰淇淋不会导致溺水——炎热天气同时驱动了两者。

---

7. Qualitative vs Quantitative Methods | 定性研究与定量研究

Quantitative methods involve collecting numerical data that can be statistically analysed. Questionnaires with closed questions, structured observations, and experiments are typical examples. These methods emphasise objectivity, reliability, and generalisability.

定量方法涉及收集可以通过统计分析的数字数据。带有封闭式问题的问卷、结构化观察和实验是典型例子。这些方法强调客观性、可靠性和普遍性。

Qualitative methods gather rich, in-depth information in the form of words, images, or behaviours. Unstructured interviews, case studies, and naturalistic observations fall into this category. They prioritise validity and the subjective experience of participants.

定性方法以文字、图像或行为的形式收集丰富、深入的信息。非结构化访谈、案例研究和自然观察属于此类别。它们优先考虑效度和参与者的主观体验。

Mixed-methods research combines both approaches within a single study, allowing the strengths of one to compensate for the weaknesses of the other. For example, a researcher might use a questionnaire to measure anxiety levels quantitatively, then interview a subset of respondents to understand the meaning behind their scores.

混合方法研究在单一研究中结合两种方法，使一种方法的优点弥补另一种方法的缺点。例如，研究者可能使用问卷定量测量焦虑水平，然后对部分受访者进行访谈以理解其分数背后的意义。

---

8. Reliability and Validity | 信度与效度

Reliability refers to the consistency of a measurement. A reliable psychological test produces the same results when repeated under identical conditions. Test-retest reliability checks consistency over time, while inter-rater reliability examines whether different observers produce similar recordings.

信度指的是测量的一致性。一个可靠的心理学测试在相同条件下重复时会产出一致的结果。重测信度检查跨时间的一致性，而评分者间信度检视不同观察者是否产生相似的记录。

Validity concerns whether a study actually measures what it claims to measure. Internal validity asks whether the IV genuinely caused changes in the DV. External validity concerns the generalisability of findings to other people, settings, and times. Ecological validity is a specific form of external validity referring to whether findings apply to real-life settings.

效度关注研究是否真正测量了其所声称要测量的内容。内部效度询问自变量是否真正导致了因变量的变化。外部效度关注研究结果能否推广到其他人、其他情境和其他时间。生态效度是外部效度的一种特定形式，指研究结果是否适用于真实生活情境。

To improve validity, researchers can use standardised instructions, double-blind procedures, and carefully operationalised definitions. Demand characteristics — where participants guess the research aims and alter their behaviour — can be reduced by using deception where ethically permissible or by using naturalistic observations.

为提高效度，研究者可以使用标准化指令、双盲程序和仔细操作化的定义。要求特征——即参与者猜测研究目的并改变其行为——可以通过在伦理允许的情况下使用欺骗或使用自然观察来减少。

---

9. Research Ethics | 研究伦理

Ethical considerations are paramount in psychological research. The British Psychological Society and the APA provide guidelines protecting participants from harm. Informed consent requires that participants be given sufficient information to make a voluntary decision about participation. Deception must be justified, and participants must be debriefed afterwards.

伦理考量在心理学研究中至关重要。英国心理学会和美国心理协会提供了保护参与者免受伤害的指导方针。知情同意要求参与者获得充分信息后自愿决定是否参与。欺骗必须有正当理由，且事后必须向参与者进行事后说明。

• Confidentiality: Participant data must be kept private and anonymous where possible.
• 保密性：参与者数据必须保密，尽可能匿名。
• Right to withdraw: Participants may leave the study at any time without penalty.
• 退出权：参与者可以随时退出研究而不受处罚。
• Protection from harm: Researchers must avoid physical or psychological distress.
• 免受伤害：研究者必须避免造成身体或心理困扰。
• Debriefing: Full disclosure of the research purpose after participation.
• 事后说明：参与者完成后充分告知研究目的。

Ethical guidelines also apply to animals, which may only be used when necessary and when alternatives are unavailable. Cost-benefit analysis weighs the potential value of research against the distress it may cause.

伦理指南同样适用于动物，只有在必要且没有替代方案时才可使用。成本效益分析权衡研究的潜在价值与其可能造成的困扰。

---

10. Descriptive Statistics | 描述性统计

Descriptive statistics summarise and organise data. Measures of central tendency include the mean (arithmetic average), median (middle score when data are ordered), and mode (most frequent score). The mean is sensitive to outliers, so the median is often preferred for skewed distributions.

描述性统计用来汇总和组织数据。集中趋势的度量包括平均数（算术平均值）、中位数（数据排序后的中间分数）和众数（出现频率最高的分数）。平均数对极端值敏感，因此偏态分布中常优先使用中位数。

Measures of dispersion describe the spread of data. The range is the difference between the highest and lowest scores. The standard deviation (SD) indicates the average distance of scores from the mean. A small SD means scores cluster tightly around the mean; a large SD means they are widely spread. Standard deviation is a more informative measure than the range because it uses all data points.

离散程度的度量描述了数据的分布广度。全距是最高分与最低分之间的差值。标准差（SD）表示各分数与平均数的平均距离。SD小意味着分数紧密聚集在平均数周围；SD大意味着分数分布广泛。标准差比全距更具信息量，因为它使用了所有数据点。

SD = √[Σ(x − x̄)² ⁄ n]

标准差 = √[Σ(x − x̄)² ⁄ n]

---

11. Inferential Statistics and Significance | 推断统计与显著性

Inferential statistics allow researchers to determine whether their results are likely to have occurred by chance. A significance level of p < 0.05 means there is less than a 5% probability that the observed result is due to random variation. If this threshold is met, the null hypothesis — which states that the IV had no effect — is rejected.

推断统计使研究者能够确定其结果是否可能由偶然因素造成。p < 0.05 的显著性水平意味着观察到的结果由随机变异导致的概率小于5%。如果达到此阈值，则拒绝零假设——即自变量没有效应的假设。

Type 1 errors occur when a researcher incorrectly rejects a true null hypothesis (a false positive). Type 2 errors occur when a researcher fails to reject a false null hypothesis (a false negative). Choosing an appropriate significance level and using adequate sample sizes help balance these risks.

第一类错误是指研究者错误地拒绝了真实的零假设（假阳性）。第二类错误是指研究者未能拒绝错误的零假设（假阴性）。选择适当的显著性水平和使用足够的样本量有助于平衡这些风险。

Statistical tests must be chosen based on the research design, level of measurement, and whether the data are normally distributed. Parametric tests like the t-test assume normal distribution, while non-parametric tests like the Mann-Whitney U test do not. A significant result does not necessarily mean the effect is practically important — effect sizes should also be reported to indicate the magnitude of the effect.

统计检验必须根据研究设计、测量水平和数据是否正态分布来选择。像t检验这样的参数检验假设正态分布，而像Mann-Whitney U检验这样的非参数检验则不作此假设。显著结果不一定意味着效应在实际中重要——还应报告效应量以指示效应的大小。

---

12. Evaluating Research — Common Exam Questions | 评估研究——常见考点

Examination questions frequently ask students to evaluate a research study by identifying its strengths and limitations. A structured evaluation should address sampling bias, demand characteristics, investigator effects, ethical issues, and the balance between reliability and validity. Use the GRAVE mnemonic — Generalisability, Reliability, Application, Validity, Ethics — to systematically critique any study.

考试题目常常要求学生通过识别研究的优点和局限性来评估一项研究。结构化的评估应涉及抽样偏差、要求特征、研究者效应、伦理问题以及信度与效度之间的平衡。使用GRAVE记忆法——普遍性、信度、应用、效度、伦理——来系统性地评析任何研究。

Investigator effects occur when the researcher unintentionally influences participant behaviour through subtle cues. A double-blind procedure, where neither the participant nor the researcher knows the condition assignment, can eliminate this problem. Standardised procedures also reduce variability between different data collectors.

研究者效应是指研究者通过微妙的线索无意中影响参与者行为。双盲程序——即参与者和研究者都不知道条件分配——可以消除这个问题。标准化程序也减少了不同数据收集者之间的变异性。

Ultimately, the best research design depends on the research question. Understanding the trade-offs between different methods — control versus realism, breadth versus depth, and objectivity versus subjectivity — is the mark of a sophisticated psychology student. Mastering these evaluation skills is essential for achieving top marks in any psychology examination.

归根结底，最佳的研究设计取决于研究问题。理解不同方法之间的权衡——控制与现实性、广度与深度、客观性与主观性——是成熟心理学学生的标志。掌握这些评估技能对于在任何心理学考试中取得高分至关重要。

---

Published by TutorHao | Psychology Revision Series | aleveler.com

更多咨询请联系16621398022（同微信）