📚 SQA Sociology: Mastering Experiment Practice and Assessment Skills | SQA 社会学：实验与实践考核核心技能
In SQA Sociology at Year 10, experiments are a core research method that allow you to explore cause-and-effect relationships between social phenomena. Whether you are preparing for a written exam or a practical assignment, understanding how to design, conduct and evaluate an experiment is crucial for success. This comprehensive guide outlines the key assessment points for experimental methods, aligned with SQA criteria, and covers definitions, design, ethics, data handling, strengths and limitations, plus the exam and assignment techniques that separate a pass from a top-band answer.
在 SQA 十年級社会学中，实验是一种核心研究方法，能让你探究社会现象之间的因果关系。无论你是在准备书面考试还是实践作业，理解如何设计、开展和评估实验对于成功至关重要。本全面指南根据 SQA 标准概述了实验方法的关键评估要点，内容涵盖定义、设计、伦理、数据处理、优势与局限，以及区分及格与高分答案的考试与作业技巧。
1. Understanding the Role of Experiments in Sociology | 理解实验在社会学中的作用
Experiments are controlled investigations in which researchers manipulate one variable to observe its effect on another, while keeping all other factors constant. In sociology, experiments are used less frequently than in the natural sciences because of ethical concerns and the complexity of human behaviour, but they remain an essential tool for testing theories about social life.
实验是一种受控调查：研究者操纵一个变量，观察它对另一个变量的影响，同时保持所有其他因素不变。在社会学中，由于伦理顾虑和人类行为的复杂性，实验的使用频率低于自然科学，但它仍然是检验社会生活理论的重要工具。
For example, a sociologist might design an experiment to test whether teacher labelling affects pupil attainment: two matched groups of pupils could be given identical work, with only the teacher's reported expectations differing between groups. Any measurable difference in subsequent test scores would suggest a causal link between labelling and achievement.
例如，社会学家可以设计一个实验来检验教师标签是否影响学生成绩：给两个匹配的学生组布置相同的作业，仅让教师所表达的期望在两组之间存在差异。若随后的测验成绩出现可测量的差异，就说明标签与学业成就之间可能存在因果联系。
Experiments sit mainly within the positivist tradition, which argues that sociology should use the same logic of measurement and proof as the natural sciences. Interpretivists, by contrast, argue that experiments strip social action of meaning and cannot capture how people interpret their own situations.
实验主要属于实证主义传统，该传统主张社会学应使用与自然科学相同的测量与证明逻辑。相反，诠释主义者认为实验剥离了社会行动的意义，无法捕捉人们如何理解自身处境。
pulate the independent variable → measure the dependent variable → control everything else
实验逻辑：操纵自变量 → 测量因变量 → 控制其他一切因素
2. Core Concepts: Variables, Hypotheses and Operationalisation | 核心概念：变量、假设与操作化
A variable is any characteristic that can take different values across cases. The independent variable (IV) is the factor the researcher deliberately changes; the dependent variable (DV) is the outcome that is measured; extraneous variables are any other factors that could influence the DV and must therefore be held constant or controlled.
变量是指在不同的个案中可能取不同值的任何特征。自变量（IV）是研究者有意改变的因素；因变量（DV）是被测量的结果；无关变量是指任何其他可能影响因变量的因素，因此必须加以恒定或控制。
A hypothesis is a precise, testable statement predicting a relationship between two variables, for example 'pupils labelled as high ability will score higher on a standardised test than pupils labelled as low ability'. A null hypothesis states that there is no relationship, and researchers test against it to avoid confirming their own expectations.
假设是对两个变量之间关系的精确、可检验的陈述，例如"被标签为高能力的学生在标准化测验中会比被标签为低能力的学生得分更高"。虚无假设则认为两者之间没有关系，研究者以其为对照进行检验，以避免只证实自己的预期。
Operationalisation means converting abstract sociological concepts into concrete indicators that can be observed and measured in a real experiment. This step is where many candidates lose marks, because a hypothesis that cannot be operationalised cannot be tested.
操作化是指把抽象的社会学概念转化为可在真实实验中观察和测量的具体指标。这一步正是许多考生失分之处，因为无法操作化的假设也就无法检验。
| Concept | 概念 | Operational indicator | 操作指标 |
| Social class | 社会阶级 | Eligibility for free school meals | 免费校餐资格 |
| Educational attainment | 学业成就 | Score on a standardised test | 标准化测验分数 |
| Teacher expectation | 教师期望 | Grade predicted on a written report | 书面报告上的预测等级 |
| Conformity | 从众 | Number of times a participant changes an answer | 参与者改变答案的次数 |
3. Experimental Design: Laboratory, Field and Natural Experiments | 实验设计：实验室、田野与自然实验
A laboratory experiment takes place in a controlled, artificial setting, such as a psychology-style room or a simulated classroom. Its great strength is control: the researcher can isolate the IV and eliminate distractions, making a cause-and-effect claim much more defensible.
实验室实验在受控的人造环境中进行，例如心理实验式房间或模拟教室。其最大优势是控制：研究者可以分离自变量并排除干扰，使因果关系的论断更具说服力。
A field experiment takes place in the participants' natural setting, such as a real school, workplace or street, while the researcher still manipulates an IV. Ecological validity is higher because behaviour is more natural, but control over extraneous variables is weaker and ethical problems increase, since participants may not know they are in a study.
田野实验在参与者的自然环境中进行，例如真实的学校、工作场所或街道，研究者仍然操纵自变量。由于行为更自然，生态效度更高，但对无关变量的控制较弱，伦理问题也增加，因为参与者可能并不知道自己处于研究之中。
A natural experiment is not created by the researcher at all: the IV changes because of events outside the researcher's control, for example a change in school funding policy or the introduction of a new exam. The sociologist simply measures the before-and-after effect, so causal claims are weaker but the setting is completely realistic.
自然实验根本不由研究者创造：自变量的变化源于研究者无法控制的事件，例如学校拨款政策的变化或新考试的引入。社会学家只是测量前后效应的差异，因此因果论断较弱，但情境完全真实。
Control ↑ ecological validity ↓ (lab) · Control ↓ ecological validity ↑ (field) · No control, realistic setting (natural)
控制力 ↑ 生态效度 ↓（实验室）· 控制力 ↓ 生态效度 ↑（田野）· 无控制、情境真实（自然）
4. Sampling and Participant Selection | 抽样与参与者选择
Sampling is the process of selecting the participants who will take part in the experiment. The sampling frame is the list from which the sample is drawn, for example a school register or a list of employees. Careful sampling is essential because a biased sample undermines any generalisation you try to make.
抽样是选择参与实验者的过程。抽样框是抽取样本所依据的名单，例如学校名册或雇员名单。仔细抽样至关重要，因为带有偏差的样本会削弱你试图作出的任何推广。
Random sampling gives every member of the sampling frame an equal chance of selection and is the strongest technique for representativeness, but it is rarely possible in sociological fieldwork because schools and workplaces rarely grant such open access. Opportunity sampling uses whoever is available, which is quick and cheap but vulnerable to bias.
随机抽样使抽样框中的每位成员都有同等被选中的机会，是代表性最强的技术；但在社会学田野工作中很少可行，因为学校和工作场所很少给予如此开放的准入。机会抽样则使用任何可得之人，快捷省费，但容易产生偏差。
Volunteer sampling recruits people who respond to an advertisement, which raises ethical quality but produces a self-selecting group that is often unrepresentative. Matched-pairs sampling assigns participants to groups in pairs identical on key characteristics such as age, sex and prior attainment, which strengthens comparison between the experimental and control groups.
志愿者抽样招募响应广告的人，伦理质量较高，但会产生自我选择的群体，往往缺乏代表性。配对抽样依据年龄、性别、先前成绩等关键特征配对分配参与者，从而加强实验组与控制组之间的可比性。
5. Control and Experimental Groups | 控制组与实验组
The experimental group receives the treatment or manipulation of the IV, while the control group does not. Comparing the two groups allows the researcher to separate the effect of the IV from ordinary change such as maturation or practice, and this comparison is the heart of the experimental method.
实验组接受处理或自变量的操纵，而控制组不接受。比较两组使研究者能够把自变量的效应与成熟、练习等寻常变化区分开来，这种比较正是实验方法的核心。
Random allocation is the best way to create equivalent groups, because any differences between groups are then likely to be due to chance rather than to researcher bias. Where random allocation is impossible, matching participants on relevant variables is the next best option.
随机分配是形成等价组的最佳方式，因为组间差异此时更可能出于偶然，而非研究者的偏差。当随机分配不可行时，就相关变量对参与者进行配对，是次优选择。
Two classic distortions must be controlled. The Hawthorne effect occurs when participants change their behaviour simply because they know they are being studied. Demand characteristics occur when participants guess the hypothesis and behave in the way they think the researcher wants, which inflates apparent effects.
必须控制两种典型扭曲。霍桑效应是指参与者仅因知道自己正被研究而改变行为。需求特征是指参与者猜到假设后，按自己认为研究者希望的方式行动，从而夸大表面效应。
Practical controls include standardised instructions read out in identical words, blind or double-blind procedures where the participant or the researcher does not know the group allocation, and holding constant the room, time of day and materials used.
实际的控制手段包括：用完全相同的措辞宣读标准化指导语；采用单盲或双盲程序，使参与者或研究者不知道分组情况；以及保持房间、时段和所用材料恒定。
6. Ethical Issues in Sociological Experiments | 社会学实验中的伦理问题
Informed consent requires that participants understand the purpose of the study, what they will be asked to do and any risks involved, and that they agree freely to take part. Consent must be genuine rather than implied by silence, and for participants under sixteen it normally requires parental or guardian agreement as well.
知情同意要求参与者理解研究目的、他们将被要求做什么以及可能涉及的风险，并且自愿同意参与。同意必须是真正的，而不是以沉默表示默认；对于十六岁以下的参与者，通常还需要父母或监护人的同意。
Deception, such as misleading participants about the true aim, is sometimes used to prevent demand characteristics, but it must be minimal, justified and followed by a full debrief in which the real purpose is revealed. A debrief also gives participants the chance to ask questions and to withdraw their data.
欺骗，例如就真实目的误导参与者，有时被用来防止需求特征，但必须保持在最低限度、有正当理由，并在事后进行完整说明（debrief），揭示真实目的。事后说明也给参与者机会提问并撤回自己的数据。
Researchers must protect participants from physical and psychological harm, keep data confidential or anonymous, and guarantee the right to withdraw at any time without penalty. Extra care is required with children, prisoners and other vulnerable groups, and studies should be approved by an ethics committee before starting.
研究者必须保护参与者免受身体和心理伤害，对数据保密或匿名处理，并保证参与者有权在任何时候退出且不受任何惩罚。涉及儿童、囚犯及其他弱势群体时需格外谨慎，研究开始前应获得伦理委员会批准。
7. Validity, Reliability and Representativeness | 效度、信度与代表性
Internal validity asks whether the experiment really measured what it claimed to measure, and whether the change in the DV was genuinely caused by the IV rather than by an extraneous variable. Confounding variables, participant reactivity and faulty operationalisation are the main threats to internal validity.
内部效度追问的是：实验是否真的测量了它声称要测量的东西，因变量的变化是否真的由自变量引起，而非由无关变量引起。混淆变量、参与者的反应性以及操作化失误是内部效度的主要威胁。
External validity is the extent to which findings can be generalised beyond the experimental setting, and ecological validity is the specific extent to which the setting and behaviour resemble everyday life. Laboratory experiments often score high on internal validity but low on ecological validity.
外部效度是研究结果能推广到实验情境之外的程度，而生态效度则是情境与行为接近日常生活的具体程度。实验室实验通常内部效度高，但生态效度低。
Reliability means that repeating the experiment with the same procedures produces consistent results, which is why standardised instructions and a written procedure are so important. A study can be reliable but invalid: a broken thermometer gives the same reading every time, but the reading is wrong.
信度意味着以相同程序重复实验能得到一致的结果，这正是标准化指导语与书面程序如此重要的原因。一项研究可能信度高但效度低：坏掉的温度计每次读数都一样，但读数本身是错的。
results · Validity = accuracy of results · Representativeness = how typical the sample is
信度 = 结果的一致性 · 效度 = 结果的准确性 · 代表性 = 样本的典型程度
8. Conducting the Experiment: Practical Procedure | 实施实验：实践操作流程
Before the main study, a pilot study should be run with a small number of participants to test whether the instructions, materials and timings actually work. Piloting reveals ambiguous wording, tasks that are too easy or too hard, and practical problems that would otherwise ruin the main experiment.
在正式研究之前，应先用少量参与者进行试点研究，检验指导语、材料和时长是否可行。试点能暴露措辞歧义、任务过易或过难，以及本会毁掉主实验的实际问题。
s → obtain consent → allocate randomly to groups → read standardised instructions → run the task under identical conditions → collect and record data → debrief participants.
主程序应遵循固定流程，以便其他研究者可以重复。典型流程是：招募参与者 → 取得同意 → 随机分组 → 宣读标准化指导语 → 在相同条件下执行任务 → 收集并记录数据 → 事后说明。
During the run, the researcher should behave in exactly the same way with every participant, avoid extra explanations that differ between groups, and record any unexpected event such as a fire alarm or an interruption, since these must be reported in the write-up as possible sources of error.
在实验过程中，研究者应以完全相同的方式对待每位参与者，避免在不同组之间给出不同的额外解释，并记录任何意外事件，如火警或干扰，因为这些必须在报告中作为可能的误差来源加以说明。
9. Recording and Analysing Data | 记录与分析数据
Quantitative data from experiments is usually recorded in a coding sheet or table, with one row per participant and one column per variable. Clear tables make the data easy to check, and raw data should be kept so that another researcher can verify your analysis.
实验产生的定量数据通常记录在编码表或表格中，每位参与者一行，每个变量一列。清晰的表格便于核查数据，且应保留原始数据，以便其他研究者验证你的分析。
| Participant | 参与者 | Group | 组别 | Test score | 测验分数 | Notes | 备注 |
| P01 | 参01 | Experimental | 实验组 | 72 | - |
| P02 | 参02 | Control | 控制组 | 64 | Asked about the aim |
Analysis begins with descriptive statistics: frequencies, percentages, and a measure of central tendency such as the mean, median or mode, together with a measure of spread such as the range. Computing the mean difference between the experimental and control groups is the simplest way to express the size of the effect.
分析从描述性统计开始：频数、百分比，以及均值、中位数或众数等集中趋势指标，再加上极差等离散程度指标。计算实验组与控制组的均值差，是表达效应大小最简单的方式。
Data is then presented in a bar chart, histogram or scatter graph, and interpreted in relation to the original hypothesis. The key skill is to move beyond description: a top-band answer explains what the pattern means sociologically, links it back to the aim, and states whether the hypothesis was supported.
随后用条形图、直方图或散点图呈现数据，并结合最初的假设进行解释。关键技能是超越描述：高分答案会解释该模式在社会学上的含义，把它与研究目的相联系，并说明假设是否得到支持。
10. Strengths and Limitations of Experiments | 实验法的优势与局限
The main strengths are control, precision and replicability. Because the researcher isolates the IV and controls extraneous variables, experiments can support causal statements rather than mere correlations, and the standardised procedure means findings can be tested again by other sociologists.
主要优势是控制、精确与可重复。由于研究者分离自变量并控制无关变量，实验可以支持因果陈述，而不仅是相关关系；标准化程序也意味着其他社会学家可以再次检验结论。
The main limitations are artificiality, low ecological validity and reactivity. Laboratory settings do not resemble real social life, participants may behave unnaturally, and the Hawthorne effect or demand characteristics can distort the results, so findings may not travel beyond the laboratory door.
主要局限是人造性、生态效度低与反应性。实验室情境不像真实社会生活，参与者可能行为不自然，霍桑效应或需求特征会扭曲结果，因此结论可能走不出实验室的门。
Ethical and practical constraints are also severe. Sociologists cannot randomly allocate people to poverty, family breakdown or poor schooling, so many of the most important sociological variables can never be manipulated. Samples are usually small and unrepresentative, which weakens generalisation to the wider population.
伦理与实际限制同样严重。社会学家不可能把人随机分配到贫困、家庭破裂或劣质教育之中，因此许多最重要的社会学变量根本无法操纵。样本通常很小且不具代表性，从而削弱了向更大总体推广的能力。
Interpretivists add a theoretical objection: experiments treat people as objects responding to stimuli and ignore the meanings, motives and interpretations that make social action social. In their view a laboratory can measure behaviour but cannot explain it.
诠释主义者还提出理论上的反对意见：实验把人当作对刺激作出反应的客体，忽视了使社会行动之所以具有社会性的意义、动机与诠释。在他们看来，实验室可以测量行为，却无法解释行为。
11. Assessment Skills: Exam Questions and Practical Assignment | 考核技能：考试题目与实践作业
SQA command words determine what the marker wants. 'Describe' requires accurate knowledge of what an experiment is and how it is carried out; 'Explain' requires reasons, mechanisms and sociological concepts; 'Evaluate' requires a balanced judgement with strengths, limitations and a conclusion.
SQA 的指令词决定评分者想要什么。"描述"要求准确呈现实验是什么、如何开展；"解释"要求给出原因、机制与社会学概念；"评估"要求作出平衡判断，包含优势、局限与结论。
A typical exam question might ask you to evaluate the use of laboratory experiments in sociological research. A strong plan would define the method, give one detailed example, then weigh control and causal inference against artificiality, ethics, small samples and the interpretivist critique, finishing with a clear overall judgement.
典型考题可能要求你评估实验室实验在社会学研究中的使用。出色的提纲会先定义该方法，给出一个具体例子，然后在控制力与因果推断之间权衡人造性、伦理、小样本与诠释主义批评，最后给出明确的总体判断。
The practical assignment usually follows a fixed structure: aim and hypothesis, method and sampling, ethical considerations, findings, and evaluation of the method used. Marks are awarded separately for each section, so an unevaluated findings section is a common and avoidable loss.
实践作业通常遵循固定结构：目的与假设、方法与抽样、伦理考量、研究结果，以及对所用方法的评估。各部分独立给分，因此"结果部分缺少评估"是常见且可以避免的失分。
In every answer, link the method to the specific sociological topic being studied rather than writing about experiments in the abstract. Statements such as 'this experiment on labelling and attainment could not control parental support at home, which weakens the causal claim' show the applied understanding that reaches the top band.
在每一个答案中，都要把方法与所研究的具体社会学主题联系起来，而不是抽象地谈论实验。诸如"这项关于标签与学业成就的实验无法控制家庭中的父母支持，因而削弱了因果论断"这类表述，正体现了能达到高分段的应用性理解。
12. Common Errors and Top-Band Exam Technique | 常见错误与高分答题技巧
The most common error is confusing reliability with validity, or treating them as the same thing. Another frequent mistake is describing an experiment without ever naming an independent or dependent variable, which suggests the candidate has memorised a definition rather than understood the logic of the method.
最常见的错误是把信度与效度混为一谈，或把二者视为同一回事。另一个常见错误是描述实验时始终没有指出自变量或因变量，这说明考生只是背下了定义，并未理解该方法的逻辑。
Further errors include ignoring ethics until the final line, offering unbalanced evaluations that list only limitations, and failing to link the discussion to any sociological context such as education, crime or the family. Answers that stay generic rarely move beyond a middle grade.
其他错误包括直到最后一行才提及伦理、给出只列局限的不平衡评估，以及未能把讨论与教育、犯罪或家庭等社会学背景相联系。停留在泛泛而谈的答案很少能越过中等分数。
Top-band technique means using precise terminology, giving at least one fully developed example, structuring evaluation as strengths against limitations, and ending with an explicit judgement. Signposting words such as 'however', 'therefore' and 'this suggests that' make the argument easy for the marker to follow.
高分技巧意味着使用精确术语、给出至少一个充分展开的例子、把评估组织为"优势对局限"的结构，并以明确的判断收尾。诸如 however、therefore、this suggests that 之类的衔接词能让评分者轻松跟随你的论证。
| Revision checklist | 复习清单 | Can I do this? | 我做到了吗？ |
| Define IV, DV and extraneous variables | 定义自变量、因变量与无关变量 | ☐ |
| Operationalise two sociological concepts | 操作化两个社会学概念 | ☐ |
| Compare lab, field and natural experiments | 比较实验室、田野与自然实验 | ☐ |
| Explain four ethical requirements | 解释四项伦理要求 | ☐ |
| Write a balanced evaluation with a conclusion | 写出带结论的平衡评估 | ☐ |
Finally, revise experiments alongside the other research methods so that you can compare them directly. A question that asks why sociologists rarely use experiments is really asking you to weigh control and causal proof against ethics, artificiality and the interpretive critique of treating people as objects of study.
最后，要把实验法与其他研究方法放在一起复习，以便直接比较。若题目问"为什么社会学家很少使用实验"，实际上是在要求你权衡控制力与因果证明，与人造性、伦理以及把人当作研究客体的诠释主义批评之间的关系。
Published by TutorHao | SQA Sociology (Year 10) 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