📚 Mastering Statistical Investigations: Core Practical Skills for Year 9 CIE | 掌握统计调查：Year 9 CIE 实践核心技能
Statistical investigations in Year 9 build the essential skills you need to plan, carry out, interpret and communicate data-based projects. The practical assessment focuses on your ability to turn a real-world question into a well-structured study, gather evidence responsibly and draw valid conclusions. Understanding the whole process – from formulating a hypothesis to reflecting on possible errors – is what sets a strong investigation apart from a simple number-crunching exercise.
Year 9 的统计调查旨在培养你规划、实施、解释和交流基于数据的项目所需的基本技能。实践考核重点关注你将现实问题转化为结构良好的研究、负责任地收集证据并得出有效结论的能力。理解从形成假设到反思可能误差的整个过程，正是一项优秀调查区别于简单数字运算的关键所在。
1. Understanding the Assessment Objectives | 理解考核目标
In a CIE statistical investigation, you are not just tested on calculations. Examiners look for clear evidence that you can pose a meaningful question, design a fair method to answer it, gather and present data accurately, analyse findings and evaluate the whole process. Every stage is linked to one or more assessment objectives: knowing, applying and reasoning.
在 CIE 统计调查中，考核的不仅仅是计算。考官会寻找明确的证据，表明你能提出有意义的问题、设计公平的方法来回答问题、准确地收集和呈现数据、分析结果并评估整个过程。每个阶段都与一个或多个考核目标相关：知识、应用和推理。
Practical tasks often begin with a vague real-life problem. Your job is to narrow it down into a clear statistical question that can be investigated with the time and resources available. For example, ‘Do students eat healthily?’ is too broad, whereas ‘What proportion of Year 9 students eat at least five portions of fruit and vegetables on a typical school day?’ is specific and measurable.
实践任务通常以一个模糊的现实生活问题开始。你的任务是将它缩小为一个清晰的统计问题，这个问题可以在现有的时间和资源下进行研究。例如，“学生吃得健康吗？”过于宽泛，而“在典型的上学日，Year 9 学生中有多大比例至少吃五份水果和蔬菜？”则是具体且可测量的。
Being aware of these objectives from the start helps you structure your report. Every decision you make – from the type of graph you draw to the way you handle outliers – should be linked back to the original question and the criteria on the mark scheme.
从一开始就意识到这些目标有助于你构建报告。你做出的每一个决定——从你绘制的图表类型到处理异常值的方式——都应该与原始问题和评分标准中的条目联系起来。
2. Planning a Statistical Investigation | 规划统计调查
Good planning prevents poor results. Start by writing a short hypothesis or a prediction that suggests a relationship between two variables, such as ‘Older students tend to spend more time on homework per evening.’ Make sure your hypothesis is testable using primary or secondary data.
良好的规划可以防止糟糕的结果。首先写一个简短的假设或预测，表明两个变量之间的关系，例如“年龄较大的学生每晚往往花更多时间做作业”。确保你的假设是可以使用一手或二手数据进行检验的。
Identify your population and decide whether you will collect data from every member (a census) or a representative sample. Think about practical constraints: how much time do you have? How will you contact participants? Will you need parental consent if the investigation involves younger learners? Writing a brief plan, even as a bulleted list, keeps your work focused.
确定你的总体，并决定你是从每个成员（普查）还是代表性样本中收集数据。考虑实际限制：你有多少时间？你将如何联系参与者？如果调查涉及低龄学习者，你是否需要家长同意？写一个简短的规划，哪怕只是分点列出，也能让你的工作保持专注。
Always include a pilot stage. Test your questionnaire or measurement tool on a few friends before the main data collection. A pilot can reveal ambiguous wording, missing options or practical difficulties that you can fix early.
始终包括试点阶段。在主要数据收集之前，先在几个朋友身上测试你的问卷或测量工具。试点可以揭示模棱两可的措辞、缺失的选项或实际的困难，你可以及早修正。
3. Sampling Methods and Their Importance | 抽样方法及其重要性
If a census is impossible, you must choose a sampling method. Common techniques for Year 9 include simple random sampling – where every member of the population has an equal chance of being selected – and stratified sampling, which ensures that key subgroups are represented in the correct proportions.
如果无法进行普查，你必须选择一种抽样方法。Year 9 常用的技术包括简单随机抽样——其中总体的每个成员被选中的机会均等——以及分层抽样，它确保关键子群体按正确比例被代表。
Convenience sampling, such as asking only the people on your bus, is quick but often leads to biased data. Be ready to explain why your chosen method reduces bias and how you could improve it if you repeated the investigation.
便利抽样，比如只询问你公交车上的乘客，速度很快但常常导致有偏差的数据。准备好解释为什么你选择的方法能减少偏差，以及如果你重复这项调查，你将如何改进它。
Sample size matters. A larger sample generally gives more reliable estimates of a population parameter. However, a large convenience sample is still biased. Discuss the trade-off between practicality and accuracy in your evaluation.
样本量很重要。更大的样本通常能给出更可靠的整体参数估计。然而，一个庞大的便利样本仍然是有偏差的。在你的评估中讨论实用性和准确性之间的权衡。
4. Questionnaire Design and Data Collection | 问卷设计与数据收集
A well-designed questionnaire produces clean, usable data. Questions must be neutral: avoid leading phrases like ‘Don’t you agree that homework is stressful?’ Instead, ask ‘How stressful do you find homework on a scale from 1 to 5?’
一份设计良好的问卷可以产生干净、可用的数据。问题必须中性：避免引导性语句，比如“你难道不认为作业压力很大吗？”，而应该问“你认为作业的压力有多大，从1到5打分？”
Include response options that cover all reasonable possibilities and do not overlap. For a question about daily water intake, use ranges like ‘0-0.5 litres’, ‘0.6-1.0 litres’, ‘1.1-1.5 litres’ and so on. A pilot will quickly show you if response boxes are missing.
提供涵盖所有合理可能性且不重叠的回答选项。对于每日饮水量的问题，可以使用诸如“0-0.5 升”、“0.6-1.0 升”、“1.1-1.5 升”等范围。试点将迅速显示你是否遗漏了选项。
When collecting data through observation or measurement, use the same instruments and protocols every time. This reduces measurement error. Record readings to a consistent degree of accuracy and always note the units clearly.
当通过观察或测量收集数据时，每次都使用相同的仪器和方案。这可以减少测量误差。以一致的准确度记录读数，并始终清楚地注明单位。
5. Organising and Cleaning Data | 整理和清洗数据
Raw data looks messy. Your first job after collection is to organise it into a table or spreadsheet. A frequency table is often the simplest way to group discrete or categorical data, while a grouped frequency table is needed for continuous data.
原始数据看起来是混乱的。收集数据后，你的首要任务就是将其整理到一个表格或电子表格中。对于离散或分类数据，频数表通常是最简单的分组方式，而对于连续数据，则需要分组频数表。
Check for anomalies. Did someone record a height of 25 metres instead of 1.25 metres? Such errors can be corrected if you spot them early, but only if you are confident the reading is a genuine mistake. Otherwise, flag the value as an outlier and consider analysing the data both with and without it.
检查异常情况。是否有人记录了 25 米而不是 1.25 米的身高？如果你及早发现，这类错误可以纠正，但前提是你确信该读数是真正的失误。否则，将该值标记为异常值，并考虑在保留和不保留该值的情况下分别分析数据。
Learn to tidy data. Each row should represent one observation and each column one variable. This consistent structure makes it much easier to calculate statistics and draw charts later.
学习整理数据。每一行应代表一个观测值，每一列代表一个变量。这种一致的结构使得后续计算统计量和绘制图表变得容易得多。
6. Choosing the Right Diagrams and Charts | 选择合适的图表
Diagrams tell the story of your data at a glance. The chart you pick depends on the type of data and the message you want to convey. Key types for Year 9 assessments include:
图表让你一眼就了解数据的故事。你选择的图表取决于数据的类型和你想要传达的信息。Year 9 考核中的关键类型包括：
- Bar charts – for comparing frequencies of discrete categories. Bars should be of equal width and separated by gaps.
- 柱状图 – 用于比较离散类别的频数。条形宽度相等，且条形之间有间隙。
- Pie charts – for showing proportions of a whole. Calculate each angle using the formula (frequency / total) x 360°.
- 饼图 – 用于展示整体中各部分的比例。使用公式 (频数 / 总数) × 360° 计算每个角度。
- Vertical line graphs – for displaying frequencies of numerical discrete data where there are only a few distinct values.
- 垂直折线图 – 用于展示只有少数几个不同数值的数值型离散数据的频数。
- Scatter graphs – for exploring the relationship between two continuous variables. Add a line of best fit if points show a clear linear trend.
- 散点图 – 用于探索两个连续变量之间的关系。如果点显示出清晰的线性趋势，则添加一条最佳拟合线。
- Histograms (for continuous data) – bars touch, and the area of each bar is proportional to the frequency.
- 直方图（用于连续数据）– 条形相互接触，且每个条形的面积与频数成正比。
Always label axes clearly, give your chart a title and, for scatter graphs, comment on the type of correlation: positive, negative or none.
始终清晰地标注坐标轴，给图表起一个标题，对于散点图，还要评论相关的类型：正相关、负相关或无相关。
7. Calculating Averages and Measures of Spread | 计算平均数与离散程度
Three measures of central tendency are commonly used in Year 9 investigations: the mean, the median and the mode. The mean is the sum of all values divided by the number of values (x̄ = Σx / n). The median is the middle value when data are ordered; for an even number of values, take the mean of the two middle numbers. The mode is the value that occurs most often.
Year 9 调查中通常使用三种集中趋势的度量：平均数、中位数和众数。平均数是所有数值的总和除以数值的个数（x̄ = Σx / n）。中位数是数据排序后位于中间的值；对于偶数个数值，取中间两个数的平均值。众数是出现次数最多的值。
Each measure has strengths. The mean uses all data but is sensitive to extreme values. The median is robust to outliers, making it a better choice for skewed distributions such as incomes or reaction times. The mode is useful for categorical data.
每种度量都有优点。平均数利用了所有数据，但对极值敏感。中位数对异常值具有稳健性，使其成为偏态分布（如收入或反应时间）的更好选择。众数对分类数据很有用。
To describe spread, use the range (highest value minus lowest value) or the interquartile range (IQR = Q₃ – Q₁). IQR ignores the top and bottom quarters and gives a reliable picture of the middle 50% of the data.
为了描述离散程度，使用极差（最大值减最小值）或四分位距（IQR = Q₃ – Q₁）。四分位距忽略了最高和最低的四分之一，给出了中间 50% 数据的可靠图景。
8. Conducting Simple Probability Experiments | 进行简单的概率实验
Many investigations involve chance. You might toss coins, roll dice or use spinners to collect experimental data on the relative frequency of an event. The experimental probability of an event A is estimated as: P(A) = (number of times A occurs) / (total number of trials).
许多调查涉及偶然性。你可能会抛硬币、掷骰子或使用转盘来收集关于事件相对频率的实验数据。事件 A 的实验概率估计为：P(A) = （A 发生的次数）/（试验总次数）。
Carrying out a large number of trials is essential because the law of large numbers says that the experimental probability will tend to stabilise around the theoretical probability as the number of trials increases. A Year 9 report should compare experimental results with expected values and discuss any discrepancies.
进行大量试验至关重要，因为大数定律指出，随着试验次数的增加，实验概率将趋于稳定在理论概率附近。Year 9 的报告应将实验结果与期望值进行比较，并讨论任何差异。
Randomisation devices must be fair. If you use a random number function on a calculator, make sure you understand how it works and state the seed if possible. If you design a spinner, check that the areas of the sectors are exactly proportional to the intended probabilities.
随机化装置必须是公平的。如果你使用计算器上的随机数功能，确保你了解它是如何工作的，并尽可能说明种子。如果你设计一个转盘，检查扇形的面积是否与预期的概率精确成比例。
9. Evaluating Reliability and Sources of Error | 评估可靠性与误差来源
No investigation is perfect. In your evaluation, you must identify at least two possible sources of error or bias and suggest realistic improvements. Common issues include a small or biased sample, poorly worded questions, measurement inaccuracies and recording mistakes.
没有哪个调查是完美的。在你的评估中，你必须至少找出两个可能的误差或偏差来源，并提出切实可行的改进建议。常见的问题包括样本量小或存在偏差、问题措辞不当、测量不准确以及记录错误。
Distinguish between random and systematic errors. A random error, such as a slight misreading of a stopwatch, can be reduced by taking repeat readings and averaging. A systematic error, like a metre ruler that has worn away at the end, will shift all measurements in one direction and requires a change of instrument.
区分随机误差和系统误差。随机误差，例如秒表的轻微误读，可以通过重复读数并取平均值来减少。系统误差，比如尺子末端磨损了，会使所有测量值向一个方向偏移，需要更换仪器。
Reliability can be checked by repeating parts of the experiment or by comparing results with published data. If you repeated the investigation with a different class, would you get similar results? Honest reflection on this question shows you are thinking like a statistician.
通过重复部分实验或与已发表的数据进行比较，可以检验可靠性。如果你用另一个班级重复调查，你会得到类似的结果吗？对这个问题进行诚实的反思表明你正在像统计学家一样思考。
10. Communicating Findings and Drawing Conclusions | 交流发现并得出结论
Your report should tell a clear story. Begin by restating your original hypothesis. Then summarise what the key statistics and charts reveal. Avoid simply listing every number; instead, highlight the most important patterns, trends or unexpected results.
你的报告应该讲述一个清晰的故事。首先重申你的最初假设。然后总结关键统计量和图表揭示了什么。避免简单地罗列每一个数字；相反，要突出最重要的模式、趋势或意外的结果。
Make a direct link between your evidence and your conclusion. For example, ‘The median time spent on devices was 4.2 hours for boys and 3.8 hours for girls, which supports the hypothesis that boys use screens for longer.’ If the data contradicts your prediction, say so clearly and offer a possible explanation.
在你的证据和结论之间建立直接联系。例如，“男孩使用电子设备的中位数时间是 4.2 小时，女孩是 3.8 小时，这支持了男孩看屏幕时间更长的假设。”如果数据与你的预测相矛盾，明确地说出来，并提供可能的解释。
Always include a brief discussion of limitations and next steps. What would you do differently next time? Could the findings apply to a wider population? This demonstrates a mature understanding of the investigative cycle and will earn higher marks in the reasoning strand.
始终包括对局限性和后续步骤的简要讨论。下次你会做哪些不同的尝试？这些发现能否适用于更广泛的总体？这展示了对调查循环的成熟理解，并将在推理方面获得更高的分数。
11. Ethical Considerations in Data Handling | 数据处理中的道德考量
Even at Year 9 level, it is important to treat participants with respect. Always obtain informed consent, even if it is just a friend agreeing to answer your questionnaire. Explain briefly what the data will be used for and keep responses anonymous unless you have permission to use names.
即使在 Year 9 级别，尊重参与者也很重要。始终获得知情同意，即使仅仅是一位朋友同意回答你的问卷。简要说明数据将用于什么目的，并对回答进行匿名处理，除非你获得了使用姓名的许可。
Avoid collecting sensitive personal information such as weight, income or health conditions that might embarrass someone. If your topic is sensitive, design questions so that participants can skip them without pressure. Store raw data securely and do not share it outside your project.
避免收集敏感的个人信息，如体重、收入或健康状况，这些可能会使人尴尬。如果你的话题敏感，设计问题时让参与者可以毫无压力地跳过。安全地储存原始数据，不要在你的项目之外分享它。
12. Common Mistakes and How to Avoid Them | 常见错误及如何避免
Many students lose marks by confusing correlation with causation. Just because taller students tend to have larger shoe sizes does not mean that being tall causes large feet – both are likely influenced by age and genetics. Always use phrases like ‘there is a trend’ or ‘the data suggests an association’, not ‘it causes’.
许多学生因为混淆了相关性和因果性而丢分。个子较高的学生往往鞋码较大，并不意味着身高导致了脚大——两者都可能受年龄和遗传的影响。始终使用诸如“存在一种趋势”或“数据表明存在关联”这样的说法，而不是“它导致”。
Other frequent errors include calculating means from grouped data without using midpoints, drawing bar charts with bars touching, forgetting to label axes, and using percentages without stating the base total. A quick checklist before submission will catch most of these.
其他常见错误包括：不使用组中值就从分组数据计算平均数；绘制的柱状图条形相互接触；忘记标注坐标轴；以及使用百分比时不注明基数总数。提交前快速核对一下就能发现大部分问题。
Practice by critiquing a flawed investigation. When you can spot errors in someone else’s work, you become more vigilant about your own. Swap drafts with a peer and give each other feedback focused on the assessment objectives.
通过批评一个有缺陷的调查来练习。当你能够发现别人工作中的错误时，你对自己的工作也会更加警惕。与同伴交换草稿，并围绕考核目标互相提供反馈。
Published by TutorHao | 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