📚 Year 11 WJEC Statistics: Practical Investigation Essentials | WJEC 统计：实践调查考核要点
The practical investigation is a core component of the WJEC GCSE Statistics course. It assesses your ability to plan, carry out, analyse and evaluate a real-world statistical enquiry. Mastering these skills not only boosts your controlled assessment performance but also deepens your understanding of how statistics is applied beyond the classroom.
实践调查是 WJEC GCSE 统计课程的核心组成部分。它考查你规划、开展、分析和评估一项真实世界统计探究的能力。掌握这些技能不仅能够提升你受控评估的成绩，还能加深你对统计在课堂之外如何应用的理解。
1. Planning Your Investigation | 规划调查
Every successful investigation begins with a clear plan. State your aim precisely and identify the target population. Specify whether you are exploring a relationship, comparing groups or estimating a population parameter. A well-defined question drives every subsequent step and keeps your project focused.
每一项成功的调查都始于清晰的规划。准确陈述你的目标，并明确目标总体。说明你是在探索关系、比较组别还是估计总体参数。一个定义清晰的问题会驱动后续每一步，并使你的项目保持聚焦。
Outline the type of data you need: discrete, continuous, categorical or ordinal. Decide whether you will collect primary data firsthand or use secondary data from reliable sources. A timeline and a checklist of tasks help you manage the controlled assessment window efficiently.
概述你需要的数据类型：离散型、连续型、分类型或有序型。决定是亲自收集一手数据，还是使用来源可靠的二手数据。一个时间表和任务清单有助于你高效管理受控评估时段。
2. Formulating Hypotheses and Predictions | 提出假设与预测
Translate your broad aim into a testable null hypothesis (H₀) and an alternative hypothesis (H₁). For example, H₀: There is no association between hours of sleep and test scores. H₁: There is a positive association between hours of sleep and test scores. Stating hypotheses sharpens your statistical thinking.
将你的宽泛目标转化为可检验的零假设 (H₀) 和备择假设 (H₁)。例如，H₀：睡眠时间与考试成绩之间没有关联。H₁：睡眠时间与考试成绩之间存在正相关。提出假设会强化你的统计思维。
Where appropriate, make a quantitative prediction, such as ‘The median revision time for boys will be at least 20% higher than for girls.’ Predictions based on preliminary research or reasoning add depth to your investigation and make your conclusions more meaningful.
在适当情况下，做出定量预测，例如“男孩复习时间的中位数将比女孩至少高出 20%”。基于初步研究或推理的预测能够增加调查的深度，并使你的结论更有意义。
3. Sampling Techniques | 抽样方法
Selecting an appropriate sampling method is critical. Random sampling methods, such as simple random sampling, stratified sampling and systematic sampling, help to reduce bias and allow you to generalise findings to the population. Stratified sampling is particularly useful when the population contains distinct subgroups.
选择合适的抽样方法至关重要。随机抽样方法，如简单随机抽样、分层抽样和系统抽样，有助于减少偏差，并使你的发现能够推广到总体。当总体包含不同子群时，分层抽样特别有用。
Non-random methods like quota sampling and opportunity sampling are easier to implement but introduce selection bias. If you must use such methods, acknowledge the limitations and discuss how they might affect the validity of your conclusions. Always justify your choice and state your sampling frame clearly.
非随机方法，如定额抽样和便利抽样，实施起来更容易，但会引入选择偏差。如果必须使用这些方法，应承认其局限性，并讨论它们可能如何影响结论的有效性。务必证明你选择的合理性，并清晰陈述你的抽样框。
Sample size matters: a larger sample increases precision and makes your results more representative. Use a pilot study to test your instruments and to estimate variability, which can help you determine a suitable sample size.
样本量很重要：更大的样本能提高精确度，并使结果更具代表性。通过试点研究来测试你的工具并估计变异性，这有助于确定合适的样本量。
4. Designing Data Collection Instruments | 设计数据收集工具
Questionnaires and observation sheets must be designed to minimise bias and maximise clarity. Write questions that are unambiguous, neutral, and easy to answer. Avoid leading questions such as ‘Don’t you agree that homework is useful?’ which push respondents toward a particular answer.
问卷和观察表的设计必须最小化偏差，最大化清晰度。编写的问题应当明确、中立且易于回答。避免诱导性问题，例如“你不觉得家庭作业很有用吗？”，这类问题会把受访者推向特定答案。
Use closed questions for quantitative data (e.g. multiple-choice, rating scales) and pre-code responses where possible. If you include open questions, plan how you will categorise the responses later. Pilot your questionnaire on a small group to identify ambiguous wording or missing options.
对定量数据使用封闭式问题（例如选择题、评分量表），并尽可能对回答进行预编码。如果包含开放式问题，要预先规划稍后如何对回答进行分类。在小范围群体中试测你的问卷，以识别措辞含混或选项缺失之处。
For experiments or measurements, specify the equipment, units and procedures in detail. Standardising instructions for participants and recording conditions consistently reduces experimenter effects and ensures reliability.
对于实验或测量，详细说明设备、单位和步骤。向参与者提供标准化指导语，并一致地记录实验条件，可以减少实验者效应，确保信度。
5. Collecting Primary and Secondary Data | 收集一手与二手数据
Primary data is gathered directly from your own investigation. It gives you full control over methodology and relevance. Whether you are measuring reaction times or conducting a traffic survey, always follow ethical guidelines: obtain informed consent, ensure anonymity, and do not pressure participants.
一手数据通过你自己的调查直接收集。它使你能完全控制方法和相关性。无论你是在测量反应时间还是进行交通流量调查，始终遵循道德准则：获取知情同意、确保匿名，并不施加压力给参与者。
Secondary data comes from existing sources such as government databases, academic papers or reputable websites. Always verify the credibility of the source and record the date of access. Secondary data can save time and expand your sample size, but you must check for any hidden biases or outdated information.
二手数据来自现有来源，如政府数据库、学术论文或信誉良好的网站。始终核实来源的可信度，并记录访问日期。二手数据可以节省时间并扩大样本量，但你必须检查是否有任何隐性偏差或过时信息。
Present both types of data clearly, and never alter or omit data points that do not fit your expectation. Data integrity is fundamental to a trustworthy investigation.
清晰呈现两类数据，绝不要更改或删除不符合你预期的数据点。数据完整性是可信调查的基石。
6. Organising and Representing Data | 数据整理与呈现
Raw data must be organised before analysis. Tally charts and frequency tables help you count occurrences for discrete and categorical variables. For continuous data, group the values into equal-width intervals and construct a grouped frequency table. Include columns for cumulative frequency if needed.
原始数据必须在分析前进行整理。划记表和频数表有助于你统计离散变量和分类变量出现的次数。对于连续数据，将数值归入等宽区间并构建分组频数表。如有需要，可增加累积频数列。
Choose the most appropriate diagram for your data. Bar charts compare categories, pie charts show proportions, and histograms display the distribution of continuous data with area proportional to frequency. Remember that in a histogram, frequency density = frequency ÷ class width.
为数据选择最适当的图表。条形图用于比较类别，饼图用于展示比例，直方图则用面积表示频率，呈现连续数据的分布。记住，在直方图中，频率密度 = 频率 ÷ 组距。
For bivariate data, construct scatter graphs to visualise correlation. A line of best fit can be drawn to model a trend. Box plots (box-and-whisker diagrams) are excellent for comparing distributions, as they clearly show the median, quartiles and range.
对于双变量数据，构建散点图来观察相关性。可以绘制最佳拟合线来模拟趋势。箱线图（盒须图）非常适合比较分布，因为它能清晰显示中位数、四分位数和极差。
7. Calculating Summary Statistics | 计算汇总统计量
Summarising data numerically allows you to compare datasets objectively. Measures of central tendency include the mean, median and mode. For a sample of n values, the mean is calculated as:
用数字概括数据使你能客观地比较数据集。集中趋势的度量包括平均数、中位数和众数。对于包含 n 个值的样本，平均数计算如下：
mean x̄ = Σx / n
The median is the middle value when data are ordered; for an even number of observations, it is the average of the two middle values. The median is resistant to outliers, making it preferable for skewed distributions.
中位数是数据排序后位于中间的值；对于偶数个观测值，它是中间两个值的平均数。中位数不易受异常值影响，因此更适合偏态分布。
Measures of spread include the range (maximum – minimum), the interquartile range (IQR = Q₃ – Q₁) and the standard deviation. The sample standard deviation s measures how data points deviate from the mean:
离散程度的度量包括极差（最大值 – 最小值）、四分位距 (IQR = Q₃ – Q₁) 和标准差。样本标准差 s 衡量数据点偏离平均数的程度：
s = √[ Σ(x – x̄)² / (n – 1) ]
Always discuss why you have chosen particular statistics. For instance, if you identify outliers, report both the mean and median, or use the IQR alongside the standard deviation to give a complete picture.
始终讨论你为何选择特定统计量。例如，如果你识别出异常值，应同时报告平均数和中位数，或同时使用四分位距和标准差，以给出完整图景。
8. Probability and Simulation | 概率与模拟
Probability can underpin predictions in your investigation. You may calculate experimental probability from observed frequencies or compare it with theoretical probability. For independent events, the probability of both occurring is P(A and B) = P(A) × P(B).
概率可以为你的调查中的预测提供基础。你可以从观测频率中计算实验概率，或将其与理论概率进行比较。对于独立事件，两者同时发生的概率为 P(A 与 B) = P(A) × P(B)。
Simulation is a powerful technique when real data collection is impractical. You can model random processes using random number tables, dice, coins or spreadsheet functions such as RAND() and RANDBETWEEN(). A simulation of 100 coin tosses, for example, helps illustrate sampling variability.
当实际数据收集不可行时，模拟是一项强大的技术。你可以使用随机数表、骰子、硬币或电子表格函数（如 RAND() 和 RANDBETWEEN()）来模拟随机过程。例如，模拟 100 次掷硬币有助于说明抽样变异性。
Design your simulation carefully, state the assumptions, and run it enough times to obtain stable estimates. Relate the relative frequency back to the true probability as the number of trials increases, demonstrating the concept of the Law of Large Numbers.
仔细设计你的模拟，陈述假设条件，并运行足够多次以获得稳定估计。随着试验次数的增加，将相对频率与真实概率联系起来，以此来展示大数定律的概念。
9. Analysing and Interpreting Results | 分析与解释结果
Analysis moves beyond description to find patterns and answer your original question. Calculate percentage changes, compare summary statistics across groups, and comment on the shape of distributions using terms like symmetric, positively skewed or negatively skewed.
分析超越描述，去发现模式并回答你最初的问题。计算百分比变化，比较各组的汇总统计量，并使用对称、正偏态或负偏态等术语对分布形状进行评论。
For bivariate data, quantify the strength of correlation with Spearman’s rank correlation coefficient or simply interpret the scatter. Always write a clear sentence that links your statistical findings back to the context: ‘The data suggests that as daily screen time increases, concentration levels tend to decrease, with a moderate negative correlation.’
对于双变量数据，使用斯皮尔曼等级相关系数量化相关强度，或直接解读散点图。始终写一个清晰的句子，将统计发现与背景联系起来：“数据表明，随着每天屏幕使用时间的增加，注意力水平趋于下降，呈现中度负相关。”
Use confidence intervals if appropriate: a 95% confidence interval for a mean gives a range of plausible values for the population mean. This shows the precision of your estimate and helps you assess whether differences are statistically meaningful.
如果适当，使用置信区间：均数的 95% 置信区间给出了总体均数的一个合理取值范围。这显示了你的估计的精确度，并有助于评估差异是否具有统计学意义。
10. Evaluating the Investigation | 调查评估
A high-quality evaluation critically reflects on the entire process. Identify any sources of bias, such as non-response bias in questionnaires or measurement error in equipment. Discuss how these may have influenced your data and whether they would likely overestimate or underestimate the true value.
高质量的评估能对整个过程进行批判性反思。识别任何偏差来源，如问卷中的无回答偏差或设备测量误差。讨论这些偏差可能如何影响了你的数据，以及它们可能导致高估还是低估真实值。
Consider limitations in your sampling method and sample size. Was your sample representative? Could you generalise the findings? If you used opportunity sampling, explain why this might limit the scope of your conclusions and how a stratified approach could have strengthened the study.
思考抽样方法和样本量的局限性。你的样本具有代表性吗？你能将发现推广吗？如果使用了便利抽样，解释为何这可能限制结论的范围，以及分层方法如何能增强研究。
Suggest realistic improvements: a larger sample, better calibration of instruments, more varied data sources, or a longer observation period. Showing that you can learn from the investigation’s shortcomings demonstrates higher-order statistical skills.
提出切实可行的改进建议：更大的样本、更好的仪器校准、更多样化的数据来源，或更长的观察期。表明你能从调查的不足中汲取教训，这展示出高阶的统计技能。
11. Writing a Clear Report | 撰写清晰报告
Structure your write-up logically: title, introduction, methodology, data presentation, analysis, evaluation and conclusion. Use headings and subheadings to guide the reader. Every table and graph must be clearly labelled with a title, axis labels and a key if needed.
按逻辑组织你的报告：标题、引言、方法、数据呈现、分析、评估和结论。使用标题和副标题引导读者。每个表格和图表必须清晰标注标题、轴标签，必要时附上图例。
Write in the third person and use formal, technical language: ‘The sample of 50 students was selected using a stratified random sampling technique…’ Avoid conversational phrases. Embed statistical terminology naturally to show your fluency.
使用第三人称和正式、专业的语言：“采用分层随机抽样技术选取了 50 名学生作为样本……”避免口语化的表达。自然融入统计术语，以展示你的流利程度。
Summarise your key findings concisely in the conclusion and directly address your hypotheses. State whether the evidence supports H₁ or if you fail to reject H₀, and explain what this means in the given context without overclaiming.
在结论中简要总结关键发现，并直接回应你的假设。说明证据是支持 H₁ 还是你未能拒绝 H₀，并在不夸大其词的前提下解释这在给定背景中意味着什么。
12. Common Mistakes to Avoid | 常见错误
Watch out for these frequent pitfalls. Confusing correlation with causation: a scatter graph showing a strong link does not prove that one variable causes the other. Always use cautious language like ‘associated with’ rather than ’causes’.
留意这些常见陷阱。混淆相关与因果：散点图显示强关联并不证明一个变量导致另一个变量。始终使用谨慎的语言，比如“与……相关联”，而不是“导致”。
Misusing percentages: do not average percentages from groups of very different sizes without weighting them appropriately. Also, avoid creating misleading graphs by truncating axes or using inconsistent scales; your histogram must use frequency density not raw frequency if class widths differ.
误用百分比：不要在没有恰当加权的情况下，对来自规模差异很大的组的百分比取平均数。此外，避免通过截断坐标轴或使用不一致的刻度来制作误导性图表；如果组距不同，直方图必须使用频率密度而非原始频率。
Overcomplicating the analysis: stick to techniques you understand well. It is better to apply a simpler method rigorously than to attempt an advanced test and misinterpret the result. Finally, never ignore ethical considerations or data protection when handling personal information.
过度复杂化分析：坚持使用你透彻理解的技术。严谨地运用更简单的方法，要好过尝试复杂检验却误读结果。最后，在处理个人信息时，绝不要忽视伦理考量和数据保护。
Published by TutorHao | WJEC Statistics 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