Year 12 CAIE Statistics: In-depth Analysis of Past Paper Questions | Year 12 CAIE 统计：历年真题深度解析

📚 Year 12 CAIE Statistics: In-depth Analysis of Past Paper Questions | Year 12 CAIE 统计：历年真题深度解析

Past paper questions are the most valuable resource for mastering CAIE AS Level Statistics. By analysing recurring question types, mark schemes and common pitfalls, students can refine their problem-solving techniques and boost exam confidence. This article dissects key topic areas, offers step-by-step breakdowns of typical exam items, and highlights the subtle reasoning examiners expect.

历年真题是掌握 CAIE AS 统计的最宝贵资源。通过分析反复出现的题型、评分标准和常见陷阱，学生可以改进解题技巧并提升考试信心。本文剖析核心知识点，提供典型考题的分步拆解，并突出考官期望的细微推理。

1. Representation of Data: Stem-and-Leaf and Box Plots | 数据表示：茎叶图与箱线图

Stem-and-leaf diagrams frequently appear in Paper 5, often combined with back-to-back comparisons for two data sets. A common task is to find the median, quartiles and interquartile range directly from the stem plot. Remember that when the number of data values n is odd, the median is the middle value; when n minimum, Q₁, median, Q₃ and maximum, and any outliers identified using the 1.5 × IQR rule.

茎叶图在试卷 5 中频繁出现，通常结合背靠背比较两组数据。常见任务是从茎叶图中直接找出中位数、四分位数和四分位距。记住数据个数 n 为奇数时中位数是中间值；n 为偶数时中位数是中间两个数的平均数。箱线图必须清晰显示最小值、Q₁、中位数、Q₃ 和最大值，以及使用 1.5 × IQR 规则识别出的任何异常值。

Examiners often ask for a comparison of central tendency and spread using these diagrams. Key phrases include "median is higher", "IQR is smaller indicating less variability", and "the data is positively skewed because the median is closer to the lower quartile". Always support statements with numerical values from the diagram.

考官经常要求使用这些图比较集中趋势和离散程度。关键表达包括“中位数更高”、“IQR 更小表明变异性更低”、“由于中位数更靠近下四分位数，数据呈正偏态”。始终用图中的数值支持陈述。

---

2. Measures of Central Tendency and Variation with Coded Data | 编码数据的集中趋势与变异度量

Many past papers provide data in coded form, e.g. y = (x − a)/b. Candidates must calculate the mean and standard deviation of the coded values, then reverse the coding to obtain those for the original data. If ȳ is the coded mean, then the original mean is x̄ = a + bȳ. For standard deviation, the original is sₓ = b × s_y (since adding a constant does not affect spread). A frequent mistake is forgetting to reverse the coding on variance or standard deviation correctly.

许多历年真题以编码形式提供数据，例如 y = (x − a)/b。考生需计算编码值的均值和标准差，然后反向编码以获得原始数据的值。若 ȳ 为编码均值，则原始均值为 x̄ = a + bȳ。对于标准差，原始为 sₓ = b × s_y（因为加常数不影响离散度）。一个常见错误是忘记正确反向计算方差或标准差。

Questions often move on to test confidence intervals or hypothesis testing for the original mean. Reversing coding correctly is essential for accurate results. Always show the steps: Σy, Σy², coded variance, then original standard deviation.

题目通常紧接着检验原始均值的置信区间或假设检验。正确反向编码对结果准确性至关重要。始终展示步骤：Σy、Σy²、编码方差，然后是原始标准差。

---

3. Probability: Venn Diagrams and Conditional Statements | 概率：维恩图与条件语句

ents, intersections and unions. A typical exam format gives P(A) = 0.6, P(B) = 0.5 and P(A ∪ B) = 0.8, then asks for P(A ∩ B), P(A' ∩ B) and the conditional probability P(A | B). The addition rule P(A ∪ B) = P(A) + P(B) − P(A ∩ B) is fundamental. Conditional probability is evaluated as P(A | B) = P(A ∩ B)/P(B).

试卷 5 中的概率题通常涉及补集、交集和并集。一个典型的考题形式给出 P(A) = 0.6，P(B) = 0.5 和 P(A ∪ B) = 0.8，然后要求求 P(A ∩ B)、P(A' ∩ B) 和条件概率 P(A | B)。加法公式 P(A ∪ B) = P(A) + P(B) − P(A ∩ B) 是基础。条件概率按 P(A | B) = P(A ∩ B)/P(B) 计算。

Fill in the intersection first, then the exclusive parts. When events are independent, P(A ∩ B) = P(A) × P(B); for mutually exclusive events, P(A ∩ B) = 0. Be careful to distinguish independence from mutual exclusivity — they are not the same.

维恩图是可视化重叠概率的强大工具。先填入交集，再填入互斥部分。当事件独立时，P(A ∩ B) = P(A) × P(B)；对于互斥事件，P(A ∩ B) = 0。注意区分独立性与互斥性——它们并不相同。

---

4. Permutations and Combinations: Arrangements with Restrictions | 排列与组合：带限制条件的排列

CAIE Statistics 1 tests arrangement and selection problems under specific constraints. A classic question: “How many ways can the letters of the word PARALLEL be arranged if all A's are together?” The standard approach is to treat the identical A's as a single block, count arrangements of the block with remaining letters, then multiply by arrangements within the block. Always account for repeated letters by dividing by factorial of identical items.

CAIE 统计 1 考查带特定限制条件的排列和选择问题。经典题目：“如果所有 A 都在一起，单词 PARALLEL 的字母有多少种排列方式？”标准方法是先把相同的 A 视为一个块，计算块与其余字母的排列数，再乘以块内部的排列。始终通过除以相同物品的阶乘来计算重复字母。

When questions involve arrangements where certain items must not be together, use the total unrestricted arrangements minus arrangements where the items are together. For combinations, the key formula is ⁿCᵣ = n!/(r!(n − r)!). Exam questions frequently combine selections from different groups, requiring multiplication of combination counts. Be meticulous about whether order matters.

当题目涉及特定物品不能相邻时，用无限制的总排列数减去这些物品在一起的排列数。对于组合，关键公式是 ⁿCᵣ = n!/(r!(n − r)!)。考试题常结合不同组别的选择，需要将组合数相乘。务必仔细判断顺序是否重要。

---

5. Discrete Random Variables: Probability Distribution Table and Expectation | 离散随机变量：概率分布表与期望

A typical exam problem provides a probability distribution of a discrete random variable X with unknown probabilities, often expressed in terms of a constant k. Using the fact that Σp(x) = 1 allows solving for k a function like Y = 2X + 3, the linear rule E(Y) = 2E(X) + 3 and Var(Y) = 4Var(X) applies.

一个典型考题给出离散随机变量 X 的概率分布，其中含未知概率，通常表示为常数 k 的式子。利用 Σp(x) = 1 可解出 k。然后考生计算 E(X) = Σ[x·p(x)] 和 Var(X) = Σ[x²·p(x)] − [E(X)]²。对于 Y = 2X + 3 这样的函数，适用线性规则 E(Y) = 2E(X) + 3 和 Var(Y) = 4Var(X)。

Mark schemes heavily award the correct method for variance: Σx²p(x) minus square of mean. A common mistake is forgetting to square the mean or misapplying the formula. Also, ensure that probabilities are left as fractions rather than decimals, unless instructed otherwise, to maintain exactness.

评分标准非常重视方差计算的正确方法：Σx²p(x) 减去均值的平方。常见错误是忘记对均值平方或误用公式。此外，除非另有说明，确保概率保留为分数而非小数，以保持精确性。

---

广告 / Sponsored

6. Binomial Distribution: Conditions and Probability Calculations | 二项分布：条件与概率计算

The binomial distribution is appropriate when there are a fixed number of independent trials, each with two possible outcomes and constant probability of success p ing candidates to state these conditions. Then probability calculations follow: P(X = r) = ⁿCᵣ pʳ (1 − p)ⁿ⁻ʳ booklet, and students must know how to use them effectively for P(X ≥ r) or P(r ≤ X ≤ s).

当存在固定次数的独立试验、每次试验有两种可能结果且成功概率 p 不变时，适合使用二项分布。考试题通常先要求考生陈述这些条件。然后是概率计算：P(X = r) = ⁿCᵣ pʳ (1 − p)ⁿ⁻ʳ。累积二项概率表包含在公式手册中，学生必须知道如何有效使用它们来计算 P(X ≥ r) 或 P(r ≤ X ≤ s)。

s P(X ≤ 3), read directly from tables. For P(X > 3) use 1 − P(X ≤ 3). Questions might also involve finding the most likely number of successes (the mode) by evaluating P(X = r)/P(X = r − 1) or simply comparing probabilities.

许多历年真题设置一个场景，例如一个转盘落在红色上的概率 p = 0.3，问在 10 次转动中，少于 4 次落在红色的概率。这即是 P(X ≤ 3)，可直接从表格读取。对于 P(X > 3) 则用 1 − P(X ≤ 3)。题目也可能涉及通过评估 P(X = r)/P(X = r − 1) 或简单比较概率来求最可能成功次数（众数）。

---

7. Normal Distribution: Standardisation and Inverse Problems | 正态分布：标准化与逆问题

Working with the normal distribution requires converting X ~ N(μ, σ²) to the standard normal Z ~ N(0,1) using Z = (X − μ)/σ nd require finding either μ or σ with other values known. This involves inverse normal tables: locate the z-value corresponding to Φ⁻¹(0.8) = 0.8416, then form an equation (a − μ)/σ = 0.8416 and solve.

处理正态分布需要将 X ~ N(μ, σ²) 转化为标准正态 Z ~ N(0,1)，使用 Z = (X − μ)/σ。历年真题常给出 P(X < a) = 0.8，要求在其他值已知时求 μ 或 σ。这涉及逆正态表：找到 Φ⁻¹(0.8) = 0.8416 对应的 z 值，然后建立等式 (a − μ)/σ = 0.8416 并求解。

A common question type is finding probabilities for a sample total or mean, e.g., the total weight of 6 apples where individual weights are normally distributed. Here use ΣX ~ N(6μ, 6σ²). Continuity corrections are not required for the normal distribution when the original data is continuous, but they do appear when a normal approximation to the binomial is used — a crossover topic.

一个常见的题型是求样本总和或均值的概率，例如 6 个苹果的总重量，其中个体重量服从正态分布。这里使用 ΣX ~ N(6μ, 6σ²)。当原始数据为连续时，正态分布不需要连续性校正；但当使用正态近似二项分布时，连续性校正就会出现——这是一个交叉知识点。

---

8. Hypothesis Testing for Mean Using Normal Distribution | 使用正态分布对均值进行假设检验

ull hypothesis H₀: μ = μ₀ and alternative H₁: μ < μ₀ or μ > μ₀ or μ ≠ μ₀. A test statistic Z = (X̄ − μ₀)/(σ/√n) is calculated and compared to critical values or the exact probability (p-value) is found. The conclusion must explicitly "reject H₀" or "do not reject H₀" in context.

在试卷 5 中，已知方差时对总体均值的假设检验很常见。结构包括陈述原假设 H₀: μ = μ₀ 和备择假设 H₁: μ < μ₀ 或 μ > μ₀ 或 μ ≠ μ₀。计算检验统计量 Z = (X̄ − μ₀)/(σ/√n) 并与临界值比较，或求得确切概率（p 值）。结论必须在上下文中明确“拒绝 H₀”或“不拒绝 H₀”。

omparison and a contextualised conclusion. For one-tailed tests, the critical z-value at 5% r -1.645 (lower). For two-tailed, ±1.96. Always specify the significance level used. A non‑ficant result does not prove H₀ true; rather, there is insufficient evidence to reject it.

评分标准为正确假设、检验统计量计算、比较和具体背景结论分配分值。对于单尾检验，5% 显著性水平下的临界 z 值为 1.645（上尾）或 -1.645（下尾）。双尾为 ±1.96。始终指明使用的显著性水平。不显著的结果并不证明 H₀ 为真；而是没有足够证据拒绝它。

---

9. Correlation and Linear Regression: Interpretation and Prediction | 相关与线性回归：解释与预测

Scatter diagrams and the product‑moment correlation coefficient r feature regularly. Candidates may be given sums such as Σx, Σy, Σx², Σy², Σxy and asked to compute r. The formula is provided, but careful substitution is essential. Interpretation of r must describe the direction and strength of linear correlation, e.g., “r = 0.92 indicates strong positive linear correlation”.

散点图和积矩相关系数 r 经常出现。考生可能被给定诸如 Σx、Σy、Σx²、Σy²、Σxy 的求和项，并要求计算 r。公式已提供，但仔细代入至关重要。对 r 的解释必须描述线性相关的方向和强度，例如“r = 0.92 表明强正线性相关”。

Linear regression follows: the equation of the least squares regression line y = a + bx where b = S_xy/S_xx and a = ȳ − bഥ x̄. The line can be used for prediction only within the range of the data (interpolation), not beyond (extrapolation). A common question tests the reliability of predictions, and the answer should warn about extrapolation if the x‑value lies outside the observed range.

接下来是线性回归：最小二乘回归线方程 y = a + bx，其中 b = S_xy/S_xx，a = ȳ − bഥ x̄。该直线仅可用于数据范围内的预测（内插），不可用于范围外（外推）。一个常见考题检验预测的可靠性，若 x 值超出观察范围，答案应警示外推风险。

---

10. Exam Technique: Working with Mark Schemes and Time Management | 应试技巧：利用评分标准与时间管理

Past paper analysis reveals that many marks are lost through incomplete working, omitted units or poor labelling on diagrams. Show all steps clearly, even when using a calculator, because method marks are awarded for correct intermediate expressions. When calculating probabilities, draw a standard normal curve annotated with the mean and the area of interest to reduce errors.

真题分析揭示，许多失分是由于解题过程不完整、遗漏单位或图表标注不佳。即使使用计算器，也要清晰展示所有步骤，因为正确的中途表达式可获得方法分。计算概率时，绘制标注有均值和关心区域的标准正态曲线，可减少错误。

Time pressure is a real challenge in Statistics 1. Allocate roughly one minute per mark. If stuck on a part, move on and return later. Always attempt the last sub‑question, which often involves interpretation and carries relatively easy marks. Practise with past papers under timed conditions, and review the mark schemes to understand what constitutes a complete answer.

时间压力是统计 1 的真实挑战。大致按每分钟一分的比例分配时间。若在某个部分卡住，继续前进稍后再回。务必尝试最后一个子问题，它通常涉及解释且分值相对容易。在计时条件下练习历年真题，并回顾评分标准以理解怎样的答案算完整。

Published by TutorHao | Statistics Revision Series | aleveler.com

更多咨询请联系16621398022（同微信）