📚 Psychology Experiments: Core Design and Variable Control Essentials | 心理学实验：核心设计与变量控制要点
A psychology experiment is a research method in which a researcher systematically manipulates one or more independent variables and measures their effect on a dependent variable, while controlling extraneous influences. The central goal is to establish cause-and-effect relationships through a controlled procedure.
心理学实验是一种研究方法：研究者系统地操纵一个或多个自变量，测量它们对因变量的影响，同时控制额外变量的干扰。其核心目标是通过受控程序确立因果关系。
1. What Is a Psychology Experiment? | 什么是心理学实验？
An experiment has three defining features: manipulation, measurement, and control. The researcher actively changes the independent variable (IV), records the dependent variable (DV), and attempts to hold all other conditions constant. For example, to test whether caffeine affects memory, the researcher might give one group a caffeinated drink and another group a placebo, then measure the number of words recalled.
实验具有三个核心特征：操纵、测量和控制。研究者主动改变自变量（IV），记录因变量（DV），并尽量保持其他条件恒定。例如，要检验咖啡因是否影响记忆，研究者可以让一组摄入含咖啡因的饮料、另一组摄入安慰剂，然后测量他们回忆单词的数量。
2. Independent and Dependent Variables | 自变量与因变量
The independent variable is the factor that the experimenter deliberately manipulates. It has at least two levels, such as “caffeine present” versus “caffeine absent”. The dependent variable is the outcome that is measured to see whether it changes as a result of the IV. A clear cause-effect conclusion requires that the IV is changed before the DV is measured, and that alternative explanations are eliminated.
自变量是实验者有意操纵的因素，它至少有“有咖啡因”和“无咖啡因”两个水平。因变量是为观察是否因自变量的变化而改变而测量的结果。要得出清晰的因果结论，必须先改变自变量、再测量因变量，并排除其他可能的解释。
Research question: Does amount of sleep affect test performance?
IV: hours of sleep (4 hours vs. 8 hours)
DV: score on a memory test
研究问题：睡眠时长是否影响测验成绩？
自变量：睡眠小时数（4小时 vs. 8小时）
因变量：记忆测验得分
3. Operational Definition | 操作性定义
An operational definition specifies exactly how a variable is measured or manipulated in a study. Instead of saying “memory”, a researcher might define memory operationally as “the number of correctly recalled words from a list of twenty presented once”. Operational definitions make research replicable, because other researchers know precisely what was done.
操作性定义是具体说明变量在研究中如何被测量或操纵。研究者不用模糊地说“记忆”，而可以把记忆操作性地定义为“在一次性呈现的20个单词中正确回忆的数量”。操作性定义使研究可以被重复验证，因为其他研究者能确切知道研究是如何进行的。
- Example: “Aggression” operationalized as the number of electric shocks a participant chooses to give in a supposed learning task.
示例：“攻击性”被操作性地定义为参与者在假定学习任务中选择给予电击的次数。 - Example: “Anxiety” operationalized as heart rate, self-reported rating, or level of cortisol in saliva.
示例：“焦虑”可以操作性地定义为心率、自我报告评分，或唾液中的皮质醇水平。
4. Control of Extraneous Variables | 额外变量的控制
Extraneous variables are variables other than the IV that could affect the DV. They include participant variables such as age and intelligence, situational variables such as noise and time of day, and experimenter variables such as tone of voice. If left uncontrolled, they threaten the internal validity of an experiment.
额外变量是除自变量之外可能影响因变量的变量，包括参与者的年龄、智力等个体变量，噪音、时间段等情境变量，以及实验者的语气等主试变量。如果不对它们加以控制，实验的内部效度就会受到威胁。
Common control techniques include:
- Standardisation: give identical instructions and procedures to all participants.
标准化：向所有参与者提供相同的指导语和程序。 - Random allocation: assign participants to conditions using chance, so participant variables are roughly evenly distributed.
随机分配：用随机方式将参与者分配到各条件组，使个体变量在各组大致均等。 - Holding conditions constant: use the same room, time, lighting, and equipment for every participant.
保持条件恒定：让每位参与者在同一房间、同一时间、相同光线和设备下进行。 - Double-blind procedure: neither the participant nor the experimenter knows the condition, reducing experimenter effects.
双盲程序：参与者和主试都不知道参与者属于哪个条件组，以减少主试效应。
5. Confounding Variables | 混淆变量
A confounding variable is an extraneous variable that changes systematically along with the IV, so the researcher cannot tell whether the effect on the DV is caused by the IV or by the confounding variable. For example, if all caffeinated participants are tested in the morning and all placebo participants are tested in the afternoon, time of day is confounded with caffeine. Any difference in memory could be due to caffeine or to circadian rhythm.
混淆变量是随自变量共同发生系统性变化的额外变量，使研究者无法判断因变量的变化究竟是由自变量还是混淆变量引起的。例如，如果所有摄入咖啡因的参与者都在上午测试，而安慰剂组都在下午测试，那么时间就与咖啡因产生了混淆。记忆成绩的差异可能来自咖啡因，也可能来自昼夜节律。
Confounding is avoided by ensuring that conditions differ only in the IV, using random allocation, counterbalancing, and carefully controlling the procedure. A confounded experiment lacks internal validity, because the causal conclusion is ambiguous.
避免混淆的方法是确保各条件组除自变量外没有任何不同，并采用随机分配、平衡设计和严格控制程序。混淆实验缺乏内部效度，因为因果结论是模糊的。
6. Random Allocation and Counterbalancing | 随机分配与平衡设计
Random allocation involves placing participants into groups by chance, such as flipping a coin or using a random number generator. It does not eliminate individual differences, but it distributes them across conditions, reducing the likelihood that pre-existing differences systematically bias the results.
随机分配是采用抛硬币或随机数生成器等随机方法将参与者分入各组。它不能消除个体差异，但能把差异分散到各条件组中，从而降低既有差异系统性影响结果的可能性。
Counterbalancing is used in repeated-measures designs to control order effects. If all participants do condition A first and condition B second, fatigue or practice could influence their scores. Counterbalancing ensures that half of the participants complete A then B, while the other half complete B then A.
平衡设计用于重复测量设计，以控制顺序效应。如果所有参与者都先完成条件A再完成条件B，疲劳或练习效应就可能影响成绩。平衡设计让一半参与者先A后B，另一半先B后A。
- ABBA design: A then B, followed by B then A, giving a symmetrical sequence.
ABBA设计：A-B然后B-A，形成对称顺序。 - Latin square: each condition appears equally often in each position.
拉丁方设计：每个条件在每个位置出现的次数相同。 - Random order: the sequence of conditions is randomised for each participant.
随机顺序：每个参与者的条件顺序随机排列。
7. Experimental and Control Conditions | 实验条件与控制条件
The experimental condition receives the active treatment, while the control condition provides a baseline against which the effect can be compared. In drug research, the control group may receive a placebo that looks identical to the real drug but has no active ingredient. The placebo effect is a real psychological response; it must be controlled for by using an inactive substance or a sham procedure.
实验条件接收积极处理，控制条件则提供比较基线。在药物研究中，控制组可能服用与真药外观相同但不含活性成分的安慰剂。安慰剂效应是一种真实的心理反应，必须通过使用非活性物质或模拟程序来控制。
| Condition | Treatment | Purpose |
| Experimental group 实验组 |
Receives the IV manipulation 接受自变量处理 |
To observe the effect of the IV 观察自变量的效应 |
| Control group 控制组 |
Receives no treatment or a placebo 不接受处理或接受安慰剂 |
To provide a baseline comparison 提供基线比较 |
8. Demand Characteristics and Investigator Effects | 需求特征与主试效应
Demand characteristics are cues in a study that lead participants to guess the hypothesis and change their behaviour accordingly. For example, if a participant realises the study is about obedience, they may behave in a way they think the researcher wants. Techniques to reduce demand characteristics include using cover stories, keeping the hypothesis hidden, and conducting single-blind studies where participants do not know their condition.
需求特征是研究中能让参与者猜测假设并据此改变行为的线索。例如，如果参与者意识到研究是关于服从的，他们可能会按自己认为研究者希望的方式行事。减少需求特征的方法包括使用掩护性说明、隐藏研究假设，以及采用参与者不知晓自己所在条件的单盲研究。
Investigator effects occur when the researcher’s own expectations, tone, or body language influence the results. The most powerful safeguard is the double-blind procedure, in which participants and experimenters are both unaware of the condition. Automated instructions and computer-delivered tests also reduce investigator influence.
主试效应发生在研究者自身的期望、语气或肢体语言影响结果时。最有效的防范措施是双盲程序，即参与者和主试都不知道条件归属。自动指导语和电脑化测验也能减少主试的影响。
9. Reliability and Validity | 信度与效度
Reliability refers to the consistency of a measurement. A reliable test produces similar results on repeated measurements. Inter-rater reliability asks whether different observers produce the same coding; test-retest reliability asks whether scores are stable over time. Validity refers to whether the research actually measures what it claims to measure.
信度指测量结果的一致性。一个信度高的测验在重复测量时应得到相似结果。评分者间信度考察不同观察者的编码是否一致；重测信度则考察分数随时间是否稳定。效度指研究是否真正测量了它声称要测量的内容。
Internal validity is present when changes in the DV can confidently be attributed to the IV. External validity refers to how well findings generalise to other people, settings, and times. A highly controlled laboratory experiment may have strong internal validity but weaker ecological validity because the artificial setting differs from real life.
内部效度指因变量的变化能被可靠地归因于自变量。外部效度指研究结果能在多大程度上推广到其他人群、情境和时间。高度受控的实验室实验可能内部效度较强，但生态效度较弱，因为人工情境与真实生活不同。
10. Common Designs: Repeated Measures, Independent Groups, Matched Pairs | 常见设计：重复测量、独立组、匹配组
Three research designs are common in psychology experiments. Each has strengths and limitations that influence variable control.
心理学实验常用三种研究设计。每种设计都有其优点和局限，并影响变量控制的方式。
| Design 设计类型 |
Description 描述 |
Advantages 优点 |
Disadvantages 缺点 |
| Repeated measures 重复测量 |
The same participants take part in all conditions. 同一组参与者参加所有条件。 |
Eliminates participant variables; needs fewer participants. 消除个体变量；所需参与者较少。 |
Order effects such as practice or fatigue. 存在练习、疲劳等顺序效应。 |
| Independent groups 独立组 |
Different participants are assigned to each condition. 不同参与者被分配到不同条件。 |
No order effects; each participant is tested once. 没有顺序效应；每位参与者只测试一次。 |
Participant variables may create differences between groups. 个体变量可能在组间造成差异。 |
| Matched pairs 匹配组 |
Participants are matched on a key variable, then split into conditions. 参与者在关键变量上匹配后分入不同条件。 |
Controls for the matched variable while avoiding order effects. 控制匹配变量，同时避免顺序效应。 |
Matching is difficult and time-consuming. 匹配过程困难且耗时。 |
11. Ethical Considerations in Experiments | 实验中的伦理考量
Psychological research must protect the welfare and dignity of participants. Ethical principles include informed consent, the right to withdraw at any time without penalty, confidentiality of data, and protection from physical or psychological harm. In some experiments, deception may be necessary to avoid demand characteristics, but it must be justified and followed by full debriefing.
心理学研究必须保护参与者的福祉和尊严。伦理原则包括知情同意、随时退出且不受处罚的权利、数据保密，以及免受身心伤害。在某些实验中，为避免需求特征可能需要欺骗，但这种做法必须有充分理由，并在研究结束后进行完整的事后说明。
- Informed consent: participants agree in advance based on accurate information.
知情同意：参与者在获得准确信息后预先同意参加。 - Right to withdraw: participants may leave at any time and may demand data removal.
退出权：参与者可随时退出，并可要求删除其数据。 - Debriefing: after the study, participants are fully informed about the true purpose and any deception used.
事后说明：研究结束后，研究者应向参与者完整说明真实目的和使用过的欺骗手段。 - Protection from harm: any risk must be no greater than in ordinary life, and distress must be monitored.
免受伤害：风险不得高于日常生活水平，并须监测参与者是否感到痛苦。
12. Checklist for Evaluating an Experiment | 评估实验的检查清单
When analysing any psychology experiment, a structured evaluation helps identify strengths and weaknesses in design and variable control.
在分析任何心理学实验时，结构化评估有助于识别其设计和变量控制上的优点与不足。
- Was the IV clearly manipulated with quantifiable levels?
自变量是否被清晰操纵，并有可量化的水平？ - Was the DV objectively measured with good reliability and validity?
因变量是否被客观测量，并具有良好的信度和效度？ - Were extraneous variables controlled through standardisation, random allocation, or matching?
额外变量是否通过标准化、随机分配或匹配得到控制？ - Could any confounding variable account for the results?
是否有任何混淆变量可以解释研究结果？ - Are demand characteristics or investigator effects likely to have influenced behaviour?
需求特征或主试效应是否可能影响了行为？ - Does the design have strong internal validity and appropriate external validity?
该设计是否具备较强的内部效度和适当的外部效度？ - Were ethical guidelines followed throughout?
整个研究过程是否遵守了伦理准则？
A well-designed psychology experiment is not simply a matter of manipulating variables; it is a disciplined process of reducing ambiguity so that the observed change in behaviour can be traced back to a single cause. Mastery of these design principles is essential for both conducting research and critically evaluating the work of others.
一个设计良好的心理学实验不只是简单操纵变量，而是一个减少不确定性的规范过程，使观察到的行为变化能够追溯到单一原因。掌握这些设计原则，对开展研究和批判性地评价他人研究都至关重要。
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Cancel reply