Year 12 WJEC Statistics: High-Frequency Topics and Common Mistakes | Year 12 WJEC 统计：高频考点与易错题分析

📚 Year 12 WJEC Statistics: High-Frequency Topics and Common Mistakes | Year 12 WJEC 统计：高频考点与易错题分析

Mastering AS Statistics requires a clear understanding of the most frequently examined topics and the ability to avoid repeating typical errors. This article identifies ten core areas that dominate the WJEC Year 12 specification and highlights the common pitfalls students encounter in each. Whether you are revising probability, distributions, or hypothesis testing, this guide will sharpen your exam technique and deepen your understanding.

掌握 AS 统计需要透彻理解最高频的考点，并有能力避开反复出现的典型错误。本文梳理了 WJEC Year 12 考纲中占主导地位的十个核心专题，并指出了学生在每个专题中最常遇到的陷阱。无论你正在复习概率、分布还是假设检验，这篇指南都将助你打磨答题技巧、加深理解。

1. Data Representation & Summary Statistics | 数据表示与概括统计量

Histograms, stem-and-leaf diagrams, and box plots appear in almost every paper. When class widths are unequal, frequency density must be calculated as frequency ÷ class width. Candidates often plot frequency itself on the y-axis, leading to a distorted picture of the distribution.

直方图、茎叶图和箱线图几乎出现在每一份试卷中。当组距不相等时，必须用频数除以组距来计算频率密度。考生经常直接在纵轴上标绘频数，导致分布图失真。

Outlier identification uses the rule Q₁ − 1.5 × IQR or Q₃ + 1.5 × IQR. A frequent mistake is to treat any data point outside the whiskers as an anomaly without checking whether it is a legitimate measurement. Also, when computing the mean from a grouped frequency table, the midpoint of each class must be used; assuming the endpoints are midpoints is a classic error.

异常值的识别规则是 Q₁ − 1.5 × IQR 或 Q₃ + 1.5 × IQR。常见的错误是不加核验就认为须线之外的点都是异常，而忽略了它们可能是合理的测量值。此外，用分组频数表计算均值时必须使用组中点；将区间端点当作中点是一个经典错误。

For sample standard deviation, the denominator is n−1, not n. In questions that ask for an estimate of the population variance, failing to use Bessel’s correction costs marks.

计算样本标准差时分母是 n−1 而非 n。如果题目要求估计总体方差，忘记使用贝塞尔校正就会失分。

2. Probability | 概率

The addition rule P(A ∪ B) = P(A) + P(B) − P(A ∩ B) and the multiplication rule for independent events P(A ∩ B) = P(A) × P(B) are high-frequency tools. A very common error is to apply the multiplication rule to mutually exclusive events, forgetting that for mutually exclusive events P(A ∩ B) = 0, so P(A ∪ B) = P(A) + P(B).

加法公式 P(A ∪ B) = P(A) + P(B) − P(A ∩ B) 和独立事件的乘法公式 P(A ∩ B) = P(A) × P(B) 是高频工具。一个极常见的错误是将乘法公式用于互斥事件，而忘记互斥事件满足 P(A ∩ B) = 0，因此 P(A ∪ B) = P(A) + P(B)。

Conditional probability P(A|B) = P(A ∩ B)/P(B) is often misinterpreted. Students sometimes reverse the condition or treat it as equivalent to P(B|A). Tree diagrams help visualise the structure, but marks are lost when probabilities on the second set of branches are not multiplied by the preceding probabilities.

条件概率 P(A|B) = P(A ∩ B)/P(B) 经常被误解。学生有时会颠倒条件，或将其等同于 P(B|A)。树状图有助于直观理解结构，但当第二层分支上的概率没有乘以前一层概率时就会失分。

Another pitfall is failing to recognise when events are independent from a worded context. If the question states ‘the events are independent’, the candidate must explicitly use P(A ∩ B) = P(A) × P(B) to find missing probabilities.

另一个陷阱是无法从文字情境中识别出事件相互独立。如果题目指明“事件相互独立”，考生必须明确使用 P(A ∩ B) = P(A) × P(B) 来求缺失的概率。

3. Discrete Random Variables | 离散随机变量

A probability distribution table must show each outcome and its associated probability, with the probabilities summing to exactly 1. The expectation E(X) = Σ x·P(X = x) and variance Var(X) = E(X²) − [E(X)]² are tested regularly. A mistake that occurs repeatedly is miscalculating E(X²) by squaring the probabilities instead of the values of x.

概率分布表必须列出每个结果及其对应的概率，且概率之和恰好为 1。期望 E(X) = Σ x·P(X = x) 和方差 Var(X) = E(X²) − [E(X)]² 经常被考到。反复出现的一个错误是在计算 E(X²) 时把概率平方而不是将 x 值平方。

Linear transformations feature strongly: E(aX + b) = aE(X) + b, Var(aX + b) = a²Var(X). Many candidates incorrectly write Var(aX + b) = aVar(X) + b, forgetting that the variance of a constant is zero and that the multiplier is squared.

线性变换是重点：E(aX + b) = aE(X) + b，Var(aX + b) = a²Var(X)。许多考生误写为 Var(aX + b) = aVar(X) + b，忘记了常数的方差为零且乘数需平方。

When a distribution table contains an unknown probability p, setting the sum of probabilities equal to 1 is the first step. After finding p, always verify that all probabilities lie between 0 and 1; negative probabilities indicate a calculation error earlier in the question.

当分布表中含有未知概率 p 时，首要步骤是令概率之和等于 1。求出 p 之后，务必验证所有概率都在 0 与 1 之间；负概率表明前面计算有误。

4. Binomial Distribution | 二项分布

The binomial distribution X ~ B(n, p) requires a fixed number of independent trials, each with the same probability of success p. A common misstep is using the binomial model when samples are taken without replacement from a small population, which violates the independence condition unless the population is large enough for the sampling fraction to be below 10%.

二项分布 X ~ B(n, p) 要求固定次数的独立试验，且每次试验的成功概率 p 相同。一个常见误区是在从小总体中无放回抽样时使用二项模型，这破坏了独立性条件，除非总体足够大使得抽样比例低于 10%。

Calculating individual probabilities uses the formula P(X = k) = nCk × pᵏ × (1−p)ⁿ⁻ᵏ. When using statistical tables, candidates often confuse ‘less than’ with ‘less than or equal to’. For example, P(X < 5) = P(X ≤ 4), not P(X ≤ 5). This simple slip can change the entire tail probability.

计算个别概率使用公式 P(X = k) = nCk × pᵏ × (1−p)ⁿ⁻ᵏ。使用统计表时，考生经常混淆“小于”与“小于等于”。例如 P(X < 5) = P(X ≤ 4)，而非 P(X ≤ 5)。这个简单失误可能彻底改变尾部概率。

When approximating a binomial distribution with a normal distribution, the continuity correction is essential: P(X ≤ k) ≈ P(Z ≤ (k + 0.5 − μ)/σ). Omitting the ±0.5 adjustment is one of the most widespread errors. Also ensure that np and n(1−p) are both greater than 5 for the approximation to be valid.

用正态分布近似二项分布时，连续性校正至关重要：P(X ≤ k) ≈ P(Z ≤ (k + 0.5 − μ)/σ)。忘记 ±0.5 的调整是最普遍的失误之一。同时要确保 np 和 n(1−p) 都大于 5，以保证近似的有效性。

5. Poisson Distribution | 泊松分布

The Poisson distribution X ~ Po(λ) models the number of events occurring in a fixed interval of time or space, provided events are independent and occur at a constant average rate λ. The mean and variance both equal λ. A typical mistake is using the Poisson model when events are not independent, such as in contagious phenomena where one occurrence raises the probability of another.

泊松分布 X ~ Po(λ) 用于对固定时间或空间区间内发生的事件数进行建模，前提是事件彼此独立且以恒定平均率 λ 发生。其均值与方差都等于 λ。一个典型错误是在事件不独立时使用泊松模型，例如在传染性现象中，一次发生会提高另一次发生的概率。

When using the formula P(X = r) = (e⁻λ × λʳ)/r!, the zero factorial 0! = 1 is sometimes forgotten. For cumulative probabilities P(X > k), candidates often attempt to read the table directly instead of calculating 1 − P(X ≤ k).

使用公式 P(X = r) = (e⁻λ × λʳ)/r! 时，零阶乘 0! = 1 有时会被遗忘。对于累积概率 P(X > k)，考生常常试图直接从表格中读取，而不是计算 1 − P(X ≤ k)。

Poisson approximation to the binomial requires large n and small p, with λ = np. A frequent error is applying the approximation when p is close to 0.5 or n is small. Also, if λ is large (λ > 10), a normal approximation with continuity correction may be more suitable, but the question will usually guide the chosen method.

泊松近似二项需要 n 大且 p 小，令 λ = np。常见的错误是在 p 接近 0.5 或 n 较小的情况下使用该近似。另外，若 λ 较大（λ > 10），采用带连续性校正的正态近似可能更合适，不过题目通常会引导选择的方法。

6. Normal Distribution | 正态分布

Standardising with Z = (X − μ)/σ is the backbone of normal distribution questions. When finding P(X > a), many students fail to convert to the complementary tail correctly. Tables give P(Z < z), so P(Z > z) = 1 − P(Z < z). Drawing a sketch of the standard normal curve reduces sign errors.

用 Z = (X − μ)/σ 进行标准化是正态分布问题的核心。在求 P(X > a) 时，许多学生未能正确转换为互补尾部。表格给出的是 P(Z < z)，因此 P(Z > z) = 1 − P(Z < z)。绘制标准正态曲线草图可减少符号错误。

Inverse normal problems ask: given a probability, find the value of k such that P(X < k) = p. The most frequent blunder is failing to check whether the probability refers to the lower tail or the upper tail. If P(X > k) = 0.1, then P(X < k) = 0.9, and the table value for 0.9 must be used, not 0.1.

逆正态问题要求：给定概率，求 k 使得 P(X < k) = p。最常见的疏忽是没有判断概率指的是左尾还是右尾。如果 P(X > k) = 0.1，那么 P(X < k) = 0.9，必须使用 0.9 的表格值，而非 0.1。

Linear combinations of independent normal variables are frequently examined: if X ~ N(μ₁, σ₁²) and Y ~ N(μ₂, σ₂²), then aX + bY ~ N(aμ₁ + bμ₂, a²σ₁² + b²σ₂²). A critical error is adding standard deviations instead of variances.

独立正态变量的线性组合经常被考查：若 X ~ N(μ₁, σ₁²) 且 Y ~ N(μ₂, σ₂²)，则 aX + bY ~ N(aμ₁ + bμ₂, a²σ₁² + b²σ₂²)。一个关键错误是将标准差相加，而不是将方差相加。

7. Correlation and Regression | 相关与回归

The product moment correlation coefficient (PMCC) r measures the strength of linear association. A value close to 0 does not necessarily mean there is no relationship; the relationship could be non-linear. Candidates often overstate the implications of a small r.

积矩相关系数 (PMCC) r 衡量线性关联的强度。接近 0 的 r 值并不一定表示没有关系；这种关系可能是非线性的。考生常会过度解读一个较小的 r 值。

The regression line of y on x, y = a + bx, is calculated from b = Sxy/Sxx and a = ȳ − bx̄. It should only be used to predict values of y within the range of observed x-values. Extrapolation beyond this range is unreliable, yet many candidates ignore this restriction when answering interpretation questions.

y 对 x 的回归直线 y = a + bx 由 b = Sxy/Sxx 和 a = ȳ − bx̄ 算出。该直线只能用于在观测 x 值范围内预测 y 值。在此范围之外的外推是不可靠的，然而许多考生在回答解释性问题时忽略这一限制。

Spearman’s rank correlation coefficient is used when data are not normally distributed or when the relationship is non-linear but monotonic. Errors arise when ranking: equal values must be assigned the mean rank, and failure to adjust changes the coefficient appreciably.

斯皮尔曼秩相关系数适用于数据非正态分布或关系呈非线性但单调的情形。排序时容易出错：等值必须赋予平均秩次，若未作调整，系数会明显改变。

8. Sampling | 抽样

WJEC papers regularly test knowledge of simple random, stratified, systematic, and quota sampling. A stratified sample allocates numbers in proportion to stratum sizes: nᵢ = (Nᵢ/N) × n. Miscalculating these proportions or rounding inconsistently is a common mistake.

WJEC 试题定期考查对简单随机、分层、系统和配额抽样的理解。分层抽样按层的大小比例分配样本量：nᵢ = (Nᵢ/N) × n。计算比例错误或不一致的舍入是常见错误。

Quota sampling is non-random and can introduce interviewer bias because the selection within strata is left to the sampler’s judgement. Students often describe it as quicker and cheaper but fail to mention this source of bias, losing marks for incomplete evaluation.

配额抽样是非随机的，可能引入调查员偏差，因为层内的选择取决于抽样者的判断。学生通常描述它更快、更便宜，却未提及这一偏差来源，导致评价不完整而失分。

When a sampling frame is not available, simple random sampling is impossible. Candidates must then consider other methods like systematic sampling or quota sampling and justify their choice with reference to the specific context of the question.

如果没有抽样框，简单随机抽样便无法实施。考生此时必须考虑系统抽样或配额抽样等其他方法，并参照题目特定情境说明理由。

9. Hypothesis Testing – Discrete Distributions | 假设检验 – 离散分布

Setting the null hypothesis H₀ and the alternative hypothesis H₁ correctly is the first crucial step. For a binomial test with H₀: p = 0.3, a one-tailed test uses H₁: p > 0.3 or p < 0.3, while a two-tailed test uses H₁: p ≠ 0.3. Choosing the wrong tail is a frequent error.

正确设定原假设 H₀ 与备择假设 H₁ 是首要关键步骤。对于 H₀: p = 0.3 的二项检验，单尾检验使用 H₁: p > 0.3 或 p < 0.3，而双尾检验使用 H₁: p ≠ 0.3。选错尾部是常见错误。

When using the critical region method, the calculated test statistic must be compared with the critical value(s). Many candidates fail to state whether the test statistic lies inside or outside the critical region and then draw an incorrect conclusion. The conclusion should be phrased as ‘reject H₀’ or ‘do not reject H₀’, never as ‘accept H₀’.

使用临界区域法时，必须将计算的检验统计量与临界值进行比较。许多考生未说明检验统计量是否落入临界区域，从而得出错误结论。结论应表述为“拒绝 H₀”或“不拒绝 H₀”，绝不可说“接受 H₀”。

For p-value method, a p-value less than the significance level α leads to rejecting H₀. In two-tailed tests, the p-value must be doubled before comparing to α. Forgetting to double the one-tailed probability is a classic oversight that results in an incorrect decision.

在 p 值法中，p 值小于显著性水平 α 则拒绝 H₀。在双尾检验中，p 值必须翻倍后再与 α 比较。忘记将单尾概率翻倍是一个经典的疏忽，会导致错误决策。

10. Hypothesis Testing – Normal Mean | 假设检验 – 正态均值

Testing a population mean when the variance is known uses the z-statistic Z = (x̄ − μ₀)/(σ/√n). If the population variance is unknown and the sample is small, the t-distribution with n−1 degrees of freedom should be used. Confusing z and t distributions, or using n instead of n−1 for degrees of freedom, is a routine mistake.

当方差已知时，检验总体均值使用 z 统计量 Z =

Published by TutorHao | Year 12 统计 Revision Series | aleveler.com

更多咨询请联系16621398022（同微信）

Year 12 WJEC Statistics: High-Frequency Topics and Common Mistakes | Year 12 WJEC 统计：高频考点与易错题分析