AS Cambridge Statistics: Essay Writing Framework and Model Essay | AS剑桥统计学：论文写作框架与范文

📚 AS Cambridge Statistics: Essay Writing Framework and Model Essay | AS剑桥统计学：论文写作框架与范文

Success in AS Cambridge Statistics goes far beyond calculating probabilities or plugging numbers into formulas. Examiners want to see clear, logical communication: how you set up a problem, justify a choice of test, interpret a p‑value, and reach a conclusion in context. This article provides a step‑by‑step writing framework and model essays for the most common extended‑response questions, so you can present your statistical reasoning with confidence.

AS剑桥统计学的成功远不止于计算概率或将数字代入公式。考官希望看到清晰、逻辑严谨的交流：你如何建立问题、证明检验方法的选择、解释p值并得出符合上下文的结论。本文提供针对常见长篇简答题的逐步写作框架与范文，让你从容展示统计推理过程。

1. Why Essay Structure Matters in Statistics | 为什么统计答题需要结构化

Many students treat statistics questions as pure calculations. In reality, more than 40% of marks in Paper 5 often come from interpretation, justification, and conclusion statements. A clear structure ensures you don’t miss these communication marks. It also helps you stay focused when a question asks you to ‘comment on’, ‘interpret’, or ‘justify’ your findings.

许多学生把统计题当作纯计算处理。实际上，Paper 5 中超过 40% 的分数常常来自解释、论证和结论陈述。清晰的结构能确保你不会丢失这些表达分，还能在题目要求“评论”“解释”或“论证”你的发现时让你保持思路明确。

Think of every extended statistical response as a mini‑essay with five core parts: Setup and Assumptions, Method and Calculations, Interpretation, Validation, and Conclusion. Once this framework is internalised, you can apply it to any topic.

把每一道统计长篇回答看作包含五个核心部分的微型论文：设定与假设、方法与计算、解释、验证和结论。一旦内化这个框架，你就能将其应用到任何主题中。

2. The 5‑Part Writing Framework | 五部分写作框架

The framework below works for hypothesis tests, confidence intervals, correlation analysis, and even questions on probability distributions that require interpretation. Memorise it and use it as a checklist during the exam.

下面的框架适用于假设检验、置信区间、相关性分析，甚至是需要解释的概率分布问题。请记下它，并在考试中当作清单来使用。

Part	English Prompt	中文要点
1. Setup	Define parameters, state hypotheses, check assumptions	定义参数，陈述假设，检验前提
2. Method	Name the test or interval, give formula, substitute values	指明检验或区间名称，给出公式，代入数值
3. Calculation	Compute test statistic, p‑value or critical value	计算检验统计量、p 值或临界值
4. Interpret	Compare to significance level, make a decision in context	与显著性水平比较，在上下文中做出决策
5. Conclude	Write a non‑technical conclusion that answers the original question	写出通俗结论，回答原始问题

Note: In AS Statistics, you may not need to check every assumption formally unless the question demands it, but mentioning them shows statistical maturity and can earn a mark.

注意：在AS统计学中，除非题目明确要求，否则你可能不需要正式检验每一项假设，但提及它们能体现统计成熟度，有时会因此得分。

3. Mastering the Introduction: Parameters and Hypotheses | 引言精讲：参数与假设

Always define the population parameter of interest clearly. For a hypothesis test, write the null and alternative hypotheses in mathematical notation and also in words. Use symbols like μ for mean, p for proportion, ρ for correlation coefficient.

务必清晰定义感兴趣的总体参数。进行假设检验时，把原假设和备择假设用数学符号和文字同时写出。使用 μ 表示均值，p 表示比例，ρ 表示相关系数。

Example for a two‑tailed test on a mean:

均值双侧检验示例：

Let μ be the population mean height of 18‑year‑old boys. H₀: μ = 175 cm; H₁: μ ≠ 175 cm.

设 μ 为 18 岁男孩的总体平均身高。H₀: μ = 175 cm; H₁: μ ≠ 175 cm。

Avoid vague language such as “the null hypothesis is that there is no difference.” Instead, be specific: “The null hypothesis states that the population mean weight is 500 g.”

避免使用模糊语言，如“原假设是没有差异”。应当具体：“原假设陈述总体平均重量为 500 克。”

4. Choosing and Describing the Test or Interval | 选择并描述检验或区间方法

Tell the examiner exactly which statistical procedure you are using. For a one‑sample t‑test, write: “Since the population variance is unknown and the sample size is small, a one‑sample t‑test is appropriate.” Provide the formula and explain every symbol.

准确告诉考官你使用的统计方法。对于单样本 t 检验，写出：“由于总体方差未知且样本量较小，应采用单样本 t 检验。”提供公式并解释每个符号。

For example: t = (x̄ − μ₀) / (s / √n), where x̄ is the sample mean, μ₀ the hypothesised mean, s the sample standard deviation, and n the sample size.

例如：t = (x̄ − μ₀) / (s / √n)，其中 x̄ 为样本均值，μ₀ 为假设的总均值，s为样本标准差，n为样本量。

If using a normal approximation to a binomial, say: “Since np and nq are both greater than 5, the normal approximation to the binomial is valid.” Always justify your choices.

如果采用二项分布的正态近似，写道：“由于 np 和 nq 均大于 5，可以用正态近似二项分布。”一定要为选择提供依据。

5. Presenting Calculations Clearly | 清晰展示计算过程

Calculations must be laid out step‑by‑step, not squeezed into a single line. Show intermediate values: standard error, test statistic, degrees of freedom (if applicable). Use standard notation and align your working neatly.

计算必须逐步展开，不要挤在一行里。展示中间值：标准误、检验统计量、自由度（如适用）。使用标准符号并保持书写整齐。

s/√n = 4.2 / √25 = 0.84

t = (83.6 − 80) / 0.84 = 4.2857 ≈ 4.29

For a p‑value, state: “Using the t‑distribution with 24 degrees of freedom, the two‑tailed p‑value is approximately 0.0002.” Where exact p‑values are not obtainable from tables, give the range: “p < 0.01" or "0.02 < p < 0.05".

对于 p 值，表述：“利用自由度为 24 的 t 分布，双侧 p 值约为 0.0002。”如果无法从表格获得精确 p 值，则给出范围：“p < 0.01”或“0.02 < p < 0.05”。

6. Interpretation: Comparing with Significance Level | 解释：与显著性水平的比较

This is the step where many candidates lose marks. You must state whether the result is significant, and what that means in context. Do not just say “Reject H₀”. Explain clearly:

这一步是许多考生失分的环节。必须说明结果是否显著以及它在上下文中的含义，不要只说“拒绝 H₀”。要清楚解释：

“Since the p‑value (0.0002) is less than the 5% significance level, we reject the null hypothesis. There is sufficient evidence to suggest that the population mean height is not 175 cm.”

“因为 p 值 (0.0002) 小于 5% 显著性水平，我们拒绝原假设。有足够证据表明总体平均身高不是 175 厘米。”

If the result is not significant: “The p‑value (0.12) exceeds the 5% significance level, so we do not reject H₀. There is insufficient evidence to conclude that there has been a change in satisfaction rating.”

如果结果不显著：“p 值 (0.12) 超过 5% 显著性水平，因此我们不拒绝 H₀。没有足够证据推断满意度评分发生了变化。”

7. Writing a Conclusion in Plain English | 用简洁英语写出结论

Your final sentence should answer the original practical question without using technical jargon. Imagine you are explaining the result to someone who has never studied statistics.

最后一句应回答原始实际问题，不使用专业术语。设想你正向从未学过统计学的人解释结果。

For a charity campaign question: “The data provide strong evidence that the campaign increased average donations.”

对于慈善募捐问题：“数据提供了有力证据，表明此次募捐活动增加了平均捐款额。”

For a machine calibration problem: “We can be confident that the machine is now filling bottles to the correct amount on average.”

对于机器校准问题：“我们可以确信该机器现在平均填充量正确无误。”

This contextual conclusion often carries the final mark and separates A grades from B grades.

这种结合上下文的结论往往带有最后的关键分，是区分 A 与 B 等级的关键。

8. Model Essay 1: Hypothesis Test for a Population Mean | 范文 1：总体均值的假设检验

Question: A company claims the mean lifetime of its light bulbs is 1500 hours. A consumer group tests 36 bulbs and finds a mean of 1470 hours with a standard deviation of 120 hours. Test, at the 5% significance level, whether the company’s claim is justified. Write your answer using the 5‑part framework.

问题：某公司声称其灯泡的平均使用寿命为 1500 小时。一个消费者组织测试了 36 只灯泡，发现均值为 1470 小时，标准差为 120 小时。在 5% 显著性水平下，检验该公司的说法是否合理。用五部分框架写出答案。

Setup: Let μ be the true mean lifetime of bulbs produced by the company. H₀: μ = 1500; H₁: μ < 1500 (one‑tailed, because we suspect the true mean is lower than claimed). As n = 36 is moderately large, by the Central Limit Theorem the sample mean follows an approximate normal distribution even if the population is not normal.

设定：设 μ 为公司所产灯泡的真实平均使用寿命。H₀: μ = 1500; H₁: μ < 1500（单侧，因为我们怀疑真实均值低于声称值）。由于 n = 36 足够大，根据中心极限定理，即便总体不服从正态分布，样本均值也近似服从正态分布。

Method: One‑sample z‑test for a mean (population variance estimated by s²). Test statistic: z = (x̄ − μ₀) / (s / √n).

方法：单样本均值 z 检验（总体方差由 s² 估计）。检验统计量：z = (x̄ − μ₀) / (s / √n)。

Calculation: x̄ = 1470, μ₀ = 1500, s = 120, n = 36. Standard error = 120/√36 = 20. z = (1470 − 1500)/20 = −1.5. The critical value for a one‑tailed test at 5% is −1.645. Alternatively, p‑value = P(Z < −1.5) = 0.0668.

计算：x̄ = 1470, μ₀ = 1500, s = 120, n = 36. 标准误 = 120/√36 = 20. z = (1470 − 1500)/20 = −1.5. 5% 单侧临界值为 −1.645. 或者，p 值 = P(Z < −1.5) = 0.0668。

Interpretation: Since z = −1.5 > −1.645 (or p‑value = 0.0668 > 0.05), we do not reject H₀ at the 5% significance level.

解释：因为 z = −1.5 > −1.645（或 p 值 = 0.0668 > 0.05），我们在 5% 显著性水平下不拒绝 H₀。

Conclusion: There is insufficient evidence to challenge the company’s claim that the mean bulb lifetime is 1500 hours. The observed sample mean is lower, but the difference could reasonably be due to sampling variability.

结论：没有充分证据质疑公司声称的灯泡平均寿命为 1500 小时的说法。观察到的样本均值较低，但这一差异可能合理归因于抽样变异。

9. Model Essay 2: Correlation and Regression | 范文 2：相关与回归

Question: Eight students’ scores in Mathematics (x) and Physics (y) are recorded. The product moment correlation coefficient is r = 0.782. Test, at the 1% significance level, whether there is positive correlation in the population. Find the equation of the regression line of y on x, and interpret the slope.

问题：记录了八名学生的数学 (x) 和物理 (y) 分数。积矩相关系数为 r = 0.782。在 1% 显著性水平下，检验总体中是否存在正相关。求 y 对 x 的回归直线方程，并解释斜率。

Setup: Let ρ be the population correlation coefficient between Mathematics and Physics scores. H₀: ρ = 0; H₁: ρ > 0 (one‑tailed). Sample size n = 8.

设定：设 ρ 为数学与物理成绩的总体相关系数。H₀: ρ = 0; H₁: ρ > 0（单侧）。样本量 n = 8。

Method: Use a t‑test for correlation coefficient. Test statistic: t = r √(n − 2) / √(1 − r²), with ν = n − 2 degrees of freedom.

方法：使用相关系数的 t 检验。检验统计量：t = r √(n − 2) / √(1 − r²)，自由度 ν = n − 2。

Calculation: r = 0.782, n = 8. t = 0.782 × √(6) / √(1 − 0.782²) = 0.782 × 2.4495 / √(1 − 0.6115) = 1.9155 / √0.3885 = 1.9155 / 0.6233 ≈ 3.073. Critical value from t‑table, one‑tailed at 1% with 6 df is 3.143. p‑value ≈ 0.011.

计算：r = 0.782, n = 8. t = 0.782 × √(6) / √(1 − 0.782²) = 0.782 × 2.4495 / √(1 − 0.6115) = 1.9155 / √0.3885 = 1.9155 / 0.6233 ≈ 3.073。查 t 分布表，自由度 6 时单侧 1% 临界值为 3.143。p 值 ≈ 0.011。

Interpretation: Since t = 3.073 < 3.143 and p‑value > 0.01, we do not reject H₀ at the 1% level. There is not enough evidence to claim positive correlation at this strict significance level.

解释：因为 t = 3.073 < 3.143 且 p 值 > 0.01，我们在 1% 水平下不拒绝 H₀。在如此严格的显著性水平下，没有足够证据声称存在正相关。

For the regression equation y = a + bx, suppose summary statistics give b = 0.65 and a = 30.2. The equation is y = 30.2 + 0.65x. Interpretation: “For each additional mark in Mathematics, the Physics mark increases on average by 0.65 marks.”

对于回归方程 y = a + bx，假设汇总统计量给出 b = 0.65，a = 30.2。方程为 y = 30.2 + 0.65x。解释：“数学成绩每增加 1 分，物理成绩平均提高 0.65 分。”

10. Common Pitfalls and How to Avoid Them | 常见陷阱及避免方法

Pitfall 1: Forgetting to state the parameter. Always define μ, p, or ρ explicitly. A hypothesis test without a defined parameter is meaningless to the examiner.

陷阱 1：忘记陈述参数。务必明确定义 μ、p 或 ρ。对考官而言，没有定义参数的假设检验毫无意义。

Pitfall 2: Using “accept H₀”. In Neyman‑Pearson testing, we never ‘accept’ the null; we ‘do not reject’ it. This subtlety matters in mark schemes.

陷阱 2：使用“接受 H₀”。在 Neyman‑Pearson 检验框架中，我们从不“接受”原假设，而是“不拒绝”它。这个细微差别在评分方案中很重要。

Pitfall 3: Confusing significance level and p‑value. A common mistake: “p‑value is less than the null hypothesis”. Be precise.

陷阱 3：混淆显著性水平和 p 值。常见错误：“p 值小于原假设”。要表达精确。

Pitfall 4: Writing a conclusion that is purely statistical. “Reject H₀” is not a conclusion. The conclusion must address the real‑world scenario: “There is evidence that the new drug reduces blood pressure.”

陷阱 4：结论纯统计化。“拒绝 H₀”不是结论。结论必须联系现实场景：“有证据表明新药可降低血压。”

11. Adapting the Framework to Exam Time Pressure | 在考试时间压力下调整框架

You don’t need to write lengthy paragraphs for every step. Use bullet points or short sentences where appropriate, especially for the setup and calculation sections. However, keep the interpretation and conclusion in full sentences to demonstrate communication skills.

不必每一步都写冗长段落。在适当时使用要点或短句，尤其对于设定和计算部分。然而，解释和结论要保持完整句子，以展示交流能力。

Example of succinct presentation:

简洁表述示例：

H₀: μ = 50; H₁: μ > 50
z = (52.3 − 50)/(4/√25) = 2.875
p‑value = 0.0020 < 0.05
Reject H₀. Strong evidence that the population mean has increased.

Even in this compressed form, the logical flow is clear and all marking points are visible.

即便在这种压缩形式下，逻辑流仍然清晰，所有得分点都显而易见。

12. Practice Exercise and Final Advice | 练习与最终建议

Pick any past‑paper question requiring a written statistical conclusion. Apply the 5‑part framework on a blank sheet. Check the mark scheme afterwards to see how many marks were for communication. You will likely be surprised by how heavily those marks are weighted.

选一道要求写出统计结论的历年真题。在白纸上套用五部分框架。之后对照评分方案，看看多少分是沟通表达分。你或许会惊讶于它们所占的分量。

Remember: statistical writing is a skill that improves with deliberate practice. Every time you solve a question, take two extra minutes to polish the language of your interpretation and conclusion. The clarity of your reasoning will make your answers stand out.

记住：统计写作是一项通过刻意练习得以提升的技能。每次解题时，多花两分钟润色你的解释和结论语言。清晰的推理会让你的答案脱颖而出。

Published by TutorHao | Statistics Revision Series | aleveler.com

更多咨询请联系16621398022（同微信）

AS Cambridge Statistics: Essay Writing Framework and Model Essay | AS剑桥统计学：论文写作框架与范文