📚 Practical Investigation Essentials for OCR Statistics | OCR统计实践考核要点
In Year 11 OCR Statistics, the practical investigation is a crucial component that tests your ability to design, conduct, and evaluate a statistical enquiry. It requires you to demonstrate skills from formulating a hypothesis to interpreting data and drawing conclusions. This guide covers the essential points to help you excel in your practical assessment.
在11年级OCR统计课程中，实践考核是关键组成部分，考察你设计、实施和评估统计探究的能力。你需要展示从提出假设到解释数据并得出结论的各项技能。本指南涵盖帮助你取得优异成绩的基本要点。
1. Understanding the Marking Criteria | 理解评分标准
Before you start, read the exam board’s marking criteria carefully. Marks are typically allocated for planning, data collection, processing and presentation, analysis, and evaluation. Knowing where the marks come from helps you focus your effort on each stage of the investigation.
开始之前，请仔细阅读考试局的评分标准。分数通常分配到计划、数据收集、处理与展示、分析和评估等多个环节。了解分数来源能帮你把精力集中到调查的每个阶段上。
Your teacher may provide a specific mark scheme or checklist. Ensure you cover all required elements: a clear hypothesis, a description of the method, raw data tables, appropriate graphs, calculations of averages and measures of spread, a written interpretation, and an evaluation of limitations.
老师可能提供具体的评分方案或检查清单。确保你覆盖所有必要元素：清晰的假设、方法描述、原始数据表、恰当的图表、平均数和离散程度的计算、书面解释以及局限性的评估。
2. Formulating a Clear Hypothesis | 提出清晰的假设
A hypothesis is a testable statement that predicts an outcome or a relationship between variables. It should be specific and based on preliminary research or reasoning. Avoid vague statements like ‘there will be a difference’ without stating what kind of difference you expect.
假设是一个可检验的陈述，它预测结果或变量之间的关系。假设应当具体，并基于初步研究或推理。避免模糊的说法，例如“会有差异”，而不说明你预期的差异类型。
For example, a strong hypothesis might be: ‘Students who sleep more than 8 hours per night will have a higher median test score than those who sleep fewer than 7 hours.’ This directly identifies the groups and the predicted direction, making it suitable for statistical testing.
例如，一个有力的假设可以是：“每晚睡眠超过8小时的学生，其考试成绩的中位数将高于每晚睡眠不足7小时的学生。”这直接明确了组别和预测方向，使其适合进行统计检验。
3. Choosing the Right Data Collection Method | 选择正确的数据收集方法
Decide whether to conduct an experiment, an observational study, or a questionnaire. An experiment involves manipulating one variable (e.g., adding fertiliser) and measuring another (plant height), whereas a questionnaire collects self-reported data like opinions or habits. Your choice must align with your hypothesis and practical feasibility.
决定是开展实验、观察性研究还是问卷调查。实验涉及操控一个变量（如施肥）并测量另一个变量（植株高度），而问卷收集自我报告的数据如意见或习惯。你的选择必须与假设和实际可行性一致。
If you use a questionnaire, design questions that are unbiased, clear, and easy to answer. Avoid leading questions and double-barrelled questions. Provide closed-response options where possible to simplify data processing, but also consider allowing an ‘other’ category for completeness.
如果使用问卷，设计的问题要无偏、清晰且易于回答。避免引导性问题和双重问题。尽可能提供封闭式选项以简化数据处理，但也要考虑设置“其他”类别以保证完整性。
4. Sampling Techniques and Bias | 抽样技术与偏差控制
The quality of your conclusions depends heavily on the sample. A random sample gives every member of the population an equal chance of selection, reducing selection bias. Stratified sampling ensures subgroups (strata) are proportionally represented, which is useful when the population has clear divisions like year groups or gender.
结论的质量在很大程度上取决于样本。随机样本使总体中每个成员都有相同的被选机会，减少选择偏差。分层抽样确保子群（层）按比例被代表，当总体有明确划分如年级或性别时，这种方法很有用。
Beware of convenience sampling, where you choose the most easily available subjects — it often leads to unrepresentative data. Always describe your sampling frame and justify your chosen method. Including a brief evaluation of potential sampling bias earns high marks.
谨防便利抽样，即选择最容易获得的个体——这往往会导致数据不具代表性。务必描述抽样框并说明所选方法的理由。简要评估潜在抽样偏差将获得高分。
| Sampling Method | Key Advantage | Key Disadvantage |
|---|---|---|
| Simple Random | Minimises bias, easy to understand | May not represent small subgroups well |
| Stratified | Proportional representation of strata | Requires knowledge of population structure |
| Systematic | Quick and simple to implement | Can introduce periodicity bias |
5. Designing Data Recording Sheets | 设计数据记录表
A well-structured data recording sheet saves time and reduces errors during data collection. Create columns for all variables you plan to measure, with clear headings and units. Leave space for repeated trials or multiple observations if your investigation involves replication.
结构良好的数据记录表可以在数据收集过程中节省时间并减少错误。为你计划测量的所有变量创建列，使用清晰的标题和单位。如果调查涉及重复，留出空间记录多次试验或观察结果。
Use pre-drawn tally charts for categorical data to keep count easily. For numerical data, include a column for raw readings and, if needed, another for calculated values. The raw data table should appear in your final report under the ‘Data Collection’ section, not just in an appendix.
对于分类数据使用预设的划记表以便轻松计数。对于数值型数据，设置一列用于原始读数，必要时另加一列用于计算值。原始数据表应出现在最终报告的“数据收集”部分，而非仅在附录中。
6. Conducting the Experiment or Survey | 开展实验或调查
Follow your plan carefully while remaining flexible enough to note any unexpected occurrences. If you are conducting an experiment, control all other variables (e.g., light, temperature) so that any observed change can reasonably be attributed to the variable you are testing. Document any difficulties faced during data collection.
仔细按照计划执行，同时足够灵活地记录任何意外情况。如果你在开展实验，要控制所有其他变量（如光照、温度），使观察到的变化能合理地归因于你测试的变量。记录数据收集过程中遇到的任何困难。
For surveys, ensure anonymity and confidentiality to encourage honest responses. Pilot your questionnaire on a few people first to check for ambiguous questions. Record the date, time, and context of data collection — these details add credibility and can explain anomalies later.
对于调查，确保匿名性和保密性以鼓励真实回答。先在少数人身上测试问卷，以发现含糊的问题。记录数据收集的日期、时间和背景——这些细节可增加可信度，并有助日后解释异常。
7. Organising and Representing Data | 数据整理与展示
Once you have collected your data, organise it into a clear format. Use frequency tables for discrete or grouped data, and calculate cumulative frequencies if needed. Presenting data visually is vital — choose graphs appropriate for the data type: bar charts or pictograms for categorical data, histograms for continuous grouped data, and scatter graphs for bivariate numerical data.
收集数据后，将其整理成清晰的格式。对离散或分组数据使用频数表，必要时计算累积频数。用可视化方式呈现数据至关重要——根据数据类型选择合适的图表：分类数据用条形图或象形图，连续分组数据用直方图，双变量数值数据用散点图。
Label every graph with a title, axis labels with units, and use sensible scales. Avoid distorting the scale or using 3D effects that mislead the reader. In a scatter graph, you should also draw a line of best fit if the relationship appears linear, and comment on outliers.
给每个图表加上标题、带单位的坐标轴标签，并使用合理的刻度。避免扭曲刻度或使用误导读者的三维效果。在散点图中，如果关系呈线性，还应画出最佳拟合线，并评述异常值。
8. Calculating Summary Statistics | 计算汇总统计量
Summary statistics condense a large dataset into a few meaningful numbers. Calculate measures of central tendency — the mean (average), median, and mode — and explain which one best represents your data. For symmetrical data, the mean is appropriate; for skewed data, the median is often more robust.
汇总统计量将大量数据浓缩为几个有意义的值。计算集中趋势的度量——平均值（平均数）、中位数和众数，并解释哪一个最能代表你的数据。对于对称数据，平均值合适；对于偏态数据，中位数通常更稳健。
x̄ = Σx / n
Also calculate measures of spread, such as the range, interquartile range (IQR), and standard deviation. The IQR is less affected by outliers than the range, while standard deviation takes every data point into account. Present these in a summary table for clarity.
还要计算离散程度的度量，如极差、四分位距和标准差。四分位距受异常值的影响小于极差，而标准差考虑了每个数据点。将这些结果呈现在汇总表中以便清晰。
IQR = Q₃ − Q₁
s = √[ Σ(x − x̄)² / (n − 1) ]
9. Analysing Results and Drawing Inferences | 分析结果与推断
In this section, go beyond just stating numbers — explain what they tell you in the context of your hypothesis. If you compared two groups, comment on the difference in medians or means and whether the spread of data overlaps. Use the calculated statistics to support whether the evidence supports or refutes your hypothesis.
在这一部分，不要只陈述数字——要结合你的假设解释它们说明了什么。如果你比较了两个组，评述中位数或平均值的差异以及数据分布是否有重叠。用计算出的统计量来支持证据是支持还是否定了你的假设。
Where appropriate, discuss correlation versus causation. A strong correlation between two variables does not necessarily mean that one causes the other — there may be a confounding factor. A level of statistical thinking is to acknowledge this limitation and suggest further investigations that could establish causation.
适当时，讨论相关性与因果关系的区别。两个变量之间的强相关并不一定意味着一方导致另一方——可能存在混杂因素。统计思维的一个层面是承认这一局限，并提出可确定因果关系的进一步调查。
10. Evaluating the Investigation and Suggesting Improvements | 评估调查并提出改进
No investigation is perfect, and examiners expect you to reflect honestly on weaknesses. Identify potential sources of bias, measurement errors, or sample size limitations. Discuss how these might have affected your results — for example, a small sample size could make an apparent pattern unreliable.
没有完美的调查，考官希望你诚实地反思缺点。识别潜在的偏差来源、测量误差或样本量限制。讨论这些因素如何影响你的结果——例如，较小的样本量可能使明显的模式不可靠。
Suggest specific improvements: if you had a small sample, propose collecting more data; if measurement tools were imprecise, suggest using digital instruments. Linking your evaluation back to the validity of your conclusion shows higher-order thinking and lifts marks.
提出具体的改进措施：如果样本量小，建议收集更多数据；如果测量工具不精确，建议使用数字仪器。将评估与结论的有效性联系起来，展示高阶思维，能提高分数。
11. Writing a Clear and Structured Report | 撰写清晰有条理的报告
Your final report should follow a logical structure: introduction, hypothesis, method, data collection (raw data), processed data and diagrams, analysis, evaluation, and conclusion. Use headings and subheadings to guide the reader. Keep language precise and avoid unnecessary jargon, unless it is correctly defined.
最终报告应遵循逻辑结构：引言、假设、方法、数据收集（原始数据）、处理后数据和图表、分析、评估和结论。使用标题和子标题引导读者。语言要精确，避免不必要的术语，除非定义清楚。
Include all original data and evidence, even if some results seem uninteresting. Integrity is key. If you encountered problems, describe how you dealt with them. A well-presented report with correct spelling, grammar, and neat presentation creates a positive impression and makes it easier to allocate marks.
纳入所有原始数据和证据，即使某些结果看似无趣。诚信是关键。如果遇到问题，描述你是如何应对的。一份拼写正确、语法规范、呈现整洁的报告会产生好印象，也便于评卷人分配分数。
12. Common Pitfalls to Avoid | 需要避免的常见陷阱
One frequent error is making the hypothesis too vague or impossible to test. Another is ignoring outliers without explanation — you should flag them in your analysis and comment on their possible cause. Also, avoid using the wrong type of average or graph for your data, as this misinterprets the distribution.
常见的错误是假设过于模糊或无法检验。另一个错误是不作解释就忽略异常值——你应在分析中标记它们并评述可能的原因。还要避免为数据选用错误的平均数类型或图表，因为这会曲解分布。
Finally, don’t leave the evaluation to the last minute. A rushed evaluation that simply says ‘everything went well’ earns few marks. Instead, think critically about what could be done differently and how that would strengthen the trustworthiness of your findings.
最后，不要将评估留到最后一刻。草率的评估仅仅说“一切顺利”，几乎得不到分数。相反，要批判性地思考可以做出哪些不同操作，以及这如何增强结果的可靠性。
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Cancel reply