📚 AS Cambridge Statistics: In-Depth Analysis of Past Paper Questions | AS剑桥统计：历年真题深度解析
Mastering AS Cambridge Statistics requires more than memorising formulas; it demands a deep understanding of how to apply statistical concepts to real exam questions. This article provides an in-depth analysis of typical past paper questions, covering key topics such as data representation, probability, distributions, estimation, and hypothesis testing. We will dissect common pitfalls, highlight mark scheme requirements, and offer strategic advice to help you achieve top marks.
掌握AS剑桥统计不仅仅需要死记硬背公式，更需要深刻理解如何将统计概念应用到真实的考题中。本文对历年典型真题进行深度解析，涵盖数据表示、概率、分布、估计和假设检验等核心主题。我们将剖析常见错误，强调评分标准的要求，并提供策略性建议，助你拿下高分。
1. Understanding the AS Statistics Exam Structure | 理解AS统计考试结构
Cambridge AS Statistics is usually examined through Paper 5 (Statistics 1) and Paper 6 (Statistics 2). Paper 5 carries approximately 50 marks and lasts 1 hour 15 minutes, covering representation of data, probability, permutations and combinations, discrete random variables, and the normal distribution. Paper 6 is also worth about 50 marks and has the same time, focusing on the Poisson distribution, linear combinations of normal variables, sampling, estimation, and hypothesis tests. Both papers require clear working, correct notation, and final answers to at least three significant figures unless stated otherwise.
剑桥AS统计通常通过试卷5（统计1）和试卷6（统计2）进行考核。试卷5约50分，时长1小时15分钟，涵盖数据表示、概率、排列组合、离散随机变量和正态分布。试卷6同样约50分，时间相同，侧重泊松分布、正态变量的线性组合、抽样、估计和假设检验。两份试卷都要求解题步骤清晰、符号正确，且最终答案一般要保留三位有效数字，除非另有说明。
Past papers consistently mix straightforward calculation with interpretation. For instance, you may be asked to find a mean and then comment on its reliability or compare two data sets. Marks are often allocated for method (M), accuracy (A), and sometimes for a final statement (B). Always read the question carefully to see if units or specific rounding are required.
历年真题一贯将直接计算与解释说明相结合。例如，你可能被要求计算均值，然后评论其可靠性或比较两组数据。评分通常分为方法分（M）、准确度分（A），有时还有结论陈述分（B）。务必仔细审题，看是否需要标注单位或进行指定精度的舍入。
2. Data Representation and Descriptive Statistics | 数据表示与描述统计
A classic question gives a frequency table and asks for an estimate of the mean and standard deviation. Always use midpoints of intervals as representative values. For grouped data, the formula for mean, x̄ = Σfx / Σf, and variance, s² = (Σfx² / Σf) − (x̄)², must be applied carefully. The mark scheme often rewards correct substitution into the formula even if arithmetic slips occur.
典型的考题会给出频数表，要求估算均值和标准差。记得始终用组中值作为代表值。对于分组数据，均值公式为 x̄ = Σfx / Σf，方差公式为 s² = (Σfx² / Σf) − (x̄)²，计算时需细心代入。评分标准中，即使有算术错误，正确的公式代入通常也能拿到方法分。
When using cumulative frequency graphs to estimate the median and quartiles, draw smooth curves and show construction lines. The median is read at 50% cumulative frequency on the y-axis; state the x-axis value clearly. Interpolation from a grouped table is an alternative, using the formula: median = L + ((n/2 − F) / f) × w, where L is the lower class boundary of the median class, F is cumulative frequency before, f is frequency of the class, and w is the class width. This approach appears frequently in Paper 5.
在利用累积频数图估计中位数和四分位数时，须画出平滑曲线并展示作图线。在纵轴上对应50%累积频数处读取横轴数值即为中位数，并清晰标出。从分组表中进行插值是另一种方法，公式为：中位数 = L + ((n/2 − F) / f) × w，其中L是中位数组的下限，F为之前累积频数，f为该组频数，w为组距。这个方法在试卷5中很常见。
Always check for outliers and describe skewness. A quick formula: if mean > median, data is positively skewed. Comparing these measures helps in interpretation marks.
记得检查异常值并描述偏态。一个快速判断法是：如果均值大于中位数，则数据呈正偏态。对此进行比较有助于拿到解释分。
3. Probability and Tree Diagrams | 概率与树状图
Probability questions often involve conditional chance and independent events. A tree diagram is your best tool when multiple stages are present. Multiply along branches and add different paths. For conditional probability, P(A|B) = P(A ∩ B) / P(B). Many students lose marks by failing to adjust the probabilities for the second stage after a ‘without replacement’ scenario. Always reduce the denominator when items are not replaced.
概率题常涉及条件概率和独立事件。当存在多阶段时，树状图是最好的工具。沿分支相乘，不同路径相加。对于条件概率，P(A|B) = P(A ∩ B) / P(B)。许多学生因在“不放回”情境下未能调整第二阶段的概率而失分。若物品不被放回，要记得减少分母。
A typical exam question: ‘A bag contains 4 red and 6 blue balls. Two balls are drawn without replacement. Find the probability that the second ball is red given the first was blue.’ Here, after drawing a blue, 4 red and 5 blue remain, so the required probability is 4/9. Present your working clearly and sometimes it helps to write the reduced sample space.
典型考题：“一个袋中有4个红球和6个蓝球。不放回地抽取两球。已知第一个是蓝球，求第二个是红球的概率。”在这种情况下，取出一个蓝球后，剩下4个红球和5个蓝球，因此所求概率为4/9。清晰展示步骤，有时写出缩减后的样本空间会很有帮助。
Also, remember to use probability laws: P(A ∪ B) = P(A) + P(B) − P(A ∩ B). In ‘at least one’ problems, using the complement rule 1 − P(none) is often much quicker.
此外，别忘了使用概率运算法则：P(A ∪ B) = P(A) + P(B) − P(A ∩ B)。在“至少有一个”的问题中，利用补集法则 1 − P(一个也没有) 往往快捷得多。
4. Permutations and Combinations | 排列与组合
Permutation and combination problems frequently appear in Paper 5, often involving arrangements of letters or selections from a group. The number of distinct arrangements of a word with repeated letters is n! / (p! q! …), where p, q are frequencies of repeated letters. For example, the number of arrangements of ‘STATISTICS’ involves 10!/(3!3!2!) because S appears 3 times, T 3 times, and I 2 times.
排列与组合问题经常出现在试卷5中，常涉及字母排列或从群体中选取。含有重复字母的单词的不同排列数目为 n! / (p! q! …)，其中p, q是重复字母出现的次数。比如，单词“STATISTICS”的排列数为10!/(3!3!2!)，因为S出现3次、T 3次、I 2次。
When constraints are given, like ‘two particular letters must be together’, treat them as a single entity. Then multiply by the internal arrangements of that block. Alternatively, if they must not be together, subtract the ‘together’ count from the total. Another common type asks for committees with at least a certain number of men or women, where you sum combinations for each valid case.
当有约束条件时，如“两个特定字母必须相邻”，则把它们当作一个整体处理，然后再乘以该组内部的排列数。若要求它们不能相邻，则从总数中减去相邻的情况。另一种常见类型是要求委员会中至少包括特定数量的男性或女性，此时需对每种符合条件的情况的组合数求和。
Be meticulous with notation: use ⁿCᵣ or C(n, r). Candidates often lose marks by misinterpreting ‘and’ (multiply) and ‘or’ (add). Practice structured working to avoid muddling.
使用符号要严谨：用 ⁿCᵣ 或 C(n, r)。考生常因混淆“且”（相乘）和“或”（相加）而丢分。通过练习条理化步骤以避免混乱。
5. Discrete Random Variables and Expectation | 离散随机变量与期望
Questions on discrete random variables require constructing a probability distribution table. The sum of probabilities must equal 1, which often helps find an unknown k. Expectation E(X) = Σ x·P(X = x), and variance Var(X) = E(X²) − [E(X)]². The mark scheme awards marks for correct use of the formula, even if arithmetic mistakes creep in.
关于离散随机变量的题目需要构建概率分布表。所有概率之和必须等于1，这常常可以用来求出未知数k。期望值E(X) = Σ x·P(X = x)，方差Var(X) = E(X²) − [E(X)]²。评分标准会对正确使用公式给予分数，即使出现计算错误。
Be prepared for functions of X, such as E(3X + 2) = 3E(X) + 2, and Var(3X + 2) = 9Var(X). Many past papers ask for the expected profit or cost based on a given distribution; here linear transformations are essential. A typical trap: forgetting that adding a constant does not change variance.
要准备好应对X的函数，例如 E(3X + 2) = 3E(X) + 2，Var(3X + 2) = 9Var(X)。很多历年真题会根据给定的分布求期望利润或成本，此时线性变换很关键。一个典型陷阱是：忘记加常数项不改变方差。
Also, the concept of a fair game arises, where the expected gain is zero. Set E(winnings) = 0 to determine the entry fee or unknown prize. Always define your random variable clearly at the start of the solution.
此外，还会出现公平游戏的概念，即期望收益为零。设 E(奖金) = 0 来求入场费或未知奖金。解题开始时要清晰地定义随机变量。
6. The Binomial Distribution | 二项分布
The binomial distribution X ~ B(n, p) requires four conditions: fixed number of trials, two outcomes, constant probability p, and independent trials. You can use the formula P(X = r) = ⁿCᵣ pʳ (1 − p)ⁿ⁻ʳ or tables in the exam. In Paper 5, tables for n up to 20 are provided. A common mistake is reading cumulative probability wrongly: P(X ≤ r) is directly in the table, while P(X < r) = P(X ≤ r − 1).
二项分布 X ~ B(n, p) 需满足四个条件：试验次数固定、每次两种结果、概率p恒定、试验独立。你可以使用公式 P(X = r) = ⁿCᵣ pʳ (1 − p)ⁿ⁻ʳ 或考试中提供的表格。试卷5附有n最大到20的表格。常见错误是错误读取累积概率：P(X ≤ r) 可直接查表，而 P(X < r) = P(X ≤ r − 1)。
Past-paper questions often involve finding P(X ≥ k) using the complement, and then testing whether a claim is justified. For example, if 80% of items are acceptable, find the probability that at most 2 are faulty in a sample of 10. Define success carefully (faulty or acceptable). Another favourite: using a normal approximation when np and nq are both > 5, applying continuity correction. In Paper 6 this may be revisited with hypothesis testing.
真题中常要求用补集法求 P(X ≥ k)，然后检验某个说法是否合理。例如，若80%的产品合格，求样本量为10时最多2个次品的概率。要谨慎定义成功（次品还是合格品）。另一个常考内容是：当 np 和 n(1-p) 均大于5时使用正态近似，并应用连续性校正。试卷6可能会结合假设检验再次涉及。
Always show the parameters n and p explicitly. State the distribution, then the required probability, e.g. X ~ B(12, 0.35), P(X = 5) = 0.183. Present the answer to three significant figures.
要明确写出参数n和p。指明分布，然后写出所需概率，例如 X ~ B(12, 0.35), P(X = 5) = 0.183。答案保留三位有效数字。
7. The Normal Distribution | 正态分布
Normal distribution questions are highly structured. First standardise to Z ~ N(0, 1²) using Z = (X − μ) / σ. Then use normal tables for probabilities. In reverse questions, you are given a probability and must find the unknown parameter, often requiring the inverse normal table. Marks are given for the equation set-up, standardisation, and final answer.
正态分布题目结构性强。首先用 Z = (X − μ) / σ 标准化为 Z ~ N(0, 1²)，然后利用正态分布表查概率。在反向问题中，会给出概率要求求出未知参数，通常需要用到逆正态表。列方程、标准化过程和最终答案均设有分值。
A typical Paper 5 question: ‘The weights of apples are normally distributed with mean 150 g and standard deviation 20 g. Find the probability that a randomly chosen apple weighs between 140 g and 165 g.’ Step 1: convert each bound to Z scores: Z₁ = (140 − 150)/20 = −0.5, Z₂ = (165 − 150)/20 = 0.75. Step 2: P = Φ(0.75) − Φ(−0.5) = Φ(0.75) − (1 − Φ(0.5)) = 0.7734 − (1 − 0.6915) = 0.4649. Very common error: incorrectly handling negative Z-values; always sketch a diagram.
试卷5典型题：“苹果重量服从均值为150克、标准差为20克的正态分布。求随机选取一个苹果重量介于140克到165克之间的概率。” 步骤1：将上下限分别转换为Z分数：Z₁ = (140−150)/20 = −0.5，Z₂ = (165−150)/20 = 0.75。步骤2：P = Φ(0.75) − Φ(−0.5) = Φ(0.75) − (1 − Φ(0.5)) = 0.7734 − (1 − 0.6915) = 0.4649。非常常见的错误是错误处理负Z值；务必画简图。
For ‘find the value of k such that P(X > k) = 0.1’, use the inverse normal: find z such that Φ(z) = 0.9, then k = μ + zσ. If σ or μ is unknown but two probabilities are given, set up simultaneous equations. This tests deeper understanding.
对于“求k使得 P(X > k) = 0.1”这类题，使用逆正态：找到z使得 Φ(z) = 0.9，然后 k = μ + zσ。如果σ或μ未知但给出两个概率，则需建立方程组求解。这考查更深入的理解。
8. Sampling Distributions and the Central Limit Theorem | 抽样分布与中心极限定理
In Paper 6, the distribution of the sample mean X̄ is critical. If the population is normal N(μ, σ²), then X̄ ~ N(μ, σ²/n) exactly. Even if the population is not normal, by the central limit theorem (CLT), X̄ is approximately normal for large n (usually n ≥ 30). This allows us to use normal tables for probabilities about a sample mean.
在试卷6中，样本均值X̄的分布至关重要。如果总体服从正态分布 N(μ, σ²)，则 X̄ 精确服从 N(μ, σ²/n)。即使总体非正态，根据中心极限定理，当样本量n较大时（通常n≥30），X̄近似服从正态分布。这使我们能用正态表来求解关于样本均值的概率。
Questions often ask: ‘Find the probability that the mean of a random sample of 40 exceeds a certain value.’ Always begin by stating the distribution of X̄. If the population variance σ² is known, use the standard error σ/√n. Do not confuse the standard deviation of the population with the standard error of the mean. The mark scheme penalises use of σ instead of σ/√n.
题目常会问：“求容量为40的随机样本的均值超过某一特定值的概率。” 答题时始终应先写明X̄的分布。如果总体方差 σ² 已知，则使用标准误 σ/√n。不要混淆总体标准差与均值的标准误。评分标准会惩罚使用σ而非σ/√n的情况。
When σ is unknown, Paper 6 introduces the unbiased estimate s² of the population variance. Then the distribution of (X̄ − μ)/(s/√n) follows a t-distribution for normal populations, used in hypothesis testing and confidence intervals when n is small. You may be tested on identifying when to use t rather than Z.
当σ未知时，试卷6引入总体方差的无偏估计 s²。此时，对于正态总体，(X̄ − μ)/(s/√n) 服从t分布（当n较小时），用于假设检验和置信区间。可能会考到何时应使用t分布而非Z分布。
9. Confidence Intervals | 置信区间
Confidence intervals for the population mean μ are a staple of Paper 6. When σ is known, the 95% CI is x̄ ± 1.96 × (σ/√n). If σ is unknown, replace with sample standard deviation s and use the t-value with n−1 degrees of freedom: x̄ ± t × (s/√n). The table of t-values is provided, and you must choose the correct significance level for a two‑tailed interval.
总体均值μ的置信区间是试卷6的必考内容。当σ已知时，95%置信区间为 x̄ ± 1.96 × (σ/√n)。若σ未知，用样本标准差s替代，并使用自由度为n−1的t值：x̄ ± t × (s/√n)。考试提供t界值表，你需要选择正确的双尾显著性水平下的值。
A common exam question: ‘A sample of 10 gave x̄ = 52 and s = 4. Construct a 90% confidence interval for μ.’ First, note the small sample, so use t with 9 df. For 90% CI, two tails 5% each, t = 1.833. The interval is 52 ± 1.833 × (4/√10), giving (49.68, 54.32). Always state the interval with endpoints to three significant figures and interpret it in context: ‘We are 90% confident that the true mean lies in this interval.’
常见真题：“一个样本容量为10，得 x̄ = 52，s = 4。构建μ的90%置信区间。” 首先，注意是小样本，因此使用t分布，自由度9。90%置信区间对应双尾各5%，查表得 t = 1.833。区间为 52 ± 1.833 × (4/√10)，即 (49.68, 54.32)。始终给出端点并保留三位有效数字，并在文中解释：“我们有90%的把握认为真实均值落在该
Published by TutorHao | AS 统计 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