📚 A-Level Further Mathematics Unit 4 Mark Scheme Jun19: Common Pitfalls | A-Level 进阶数学 Unit 4 评分标准 Jun19: 易错点总结
This article summarises the most frequent mistakes made by candidates in the June 2019 sitting of the A-Level Further Mathematics Unit 4 examination (CIE 9231 Paper 4 – Further Probability & Statistics). Each common error is explained with reference to mark scheme expectations, followed by a Chinese translation for bilingual clarity.
本文总结了 2019 年 6 月 A-Level 进阶数学 Unit 4 考试（CIE 9231 试卷 4——进阶概率与统计）中考生最常犯的错误。每个易错点均结合评分标准的要求进行说明，并附有中文翻译，便于双语对照理解。
1. Confidence Intervals for the Mean Using the t-distribution | 使用 t 分布进行均值的置信区间估计
Many candidates incorrectly used the normal distribution (z-values) instead of the t-distribution when the population variance was unknown and the sample size was small. The mark scheme required the critical value from tn−1 at the specified confidence level. Failing to state the correct degrees of freedom (n−1) or rounding the t-value too early led to loss of accuracy marks.
许多考生在总体方差未知且样本量较小时，错误地使用了正态分布（z 值）而非 t 分布。评分标准要求在指定的置信水平下使用 tn−1 的临界值。未能正确注明自由度（n−1）或过早舍入 t 值会导致准确性扣分。
The correct formula for the confidence interval is x̄ ± tn−1 × s/√n, and candidates must clearly state the value of the standard error before substitution.
置信区间的正确公式为 x̄ ± tn−1 × s/√n，考生在代入前必须明确写出标准误差的值。
2. Hypothesis Testing: Direction of the Alternative Hypothesis | 假设检验：备择假设的方向
A common mistake was setting up the alternative hypothesis H1 incorrectly. For a two-tailed test, some wrote H1: μ ≠ … but then used a one-tailed critical region, or vice versa. The mark scheme penalised mismatch between the stated H1 and the rejection region. Candidates must clearly define the parameter and use the correct inequality symbols (≠, >, or <).
一个常见错误是错误设定备择假设 H1。对于双尾检验，有些考生写了 H1: μ ≠ … 却使用了单尾拒绝域，反之亦然。评分标准对陈述的 H1 和拒绝域不匹配的情况会予以扣分。考生必须明确定义参数，并使用正确的不等号（≠、> 或 <）。
3. Poisson Distribution with Scaled Rates | 涉及比例调整的泊松分布
When the Poisson rate λ was given per unit of length, time, or area, many candidates forgot to scale λ to the new interval correctly. For example, if λ = 2 per metre, then over 3.5 metres the parameter becomes 7. Errors in this scaling step made the entire subsequent probability calculation incorrect, even if the Poisson formula was applied correctly.
当泊松分布的参数 λ 是基于单位长度、时间或面积给出时，许多考生忘记按比例将 λ 调整到新的区间。例如，λ = 2 每米，则在 3.5 米内参数变为 7。这一缩放步骤的错误会使后续的概率计算完全错误，即使泊松公式运用正确也不例外。
4. Unbiased Estimators: Confusing Sample Variance Formulas | 无偏估计量：混淆样本方差公式
The mark scheme required the use of s² = Σ(x − x̄)²/(n−1) as the unbiased estimator for population variance. A frequent error was dividing by n instead of (n−1), or using the formula Σx²/n − x̄² with the wrong denominator. Candidates were expected to explicitly state “unbiased estimate of variance” and show the divisor n−1.
评分标准要求使用 s² = Σ(x − x̄)²/(n−1) 作为总体方差的无偏估计。常见错误是用 n 而不是 (n−1) 作除数，或者在使用 Σx²/n − x̄² 时分母错误。考生应明确写出“方差的无偏估计”，并显示除数为 n−1。
5. Continuous Random Variables: Integration Limits | 连续随机变量：积分上下限
In problems involving probability density functions (pdf), several candidates lost marks by using wrong integration limits, especially when the pdf was defined piecewise. The mark scheme insisted on displaying the correct limits and ensuring the total area equalled 1. Using indefinite integrals without evaluating limits was not accepted.
在涉及概率密度函数的问题中，多名考生因使用错误的积分上下限而失分，特别当概率密度函数分段定义时。评分标准要求明确写出正确的上下限，并验证总面积为 1。不代入上下限而只给出不定积分不予接受。
6. Type I and Type II Errors: Definitions and Interpretation | 第 I 类错误和第 II 类错误：定义与解释
The difference between a Type I error (rejecting H0 when it is true) and a Type II error (not rejecting H0 when it is false) was frequently confused. The mark scheme awarded marks only for precise wording. Moreover, candidates needed to interpret these errors in the context of the given scenario, not just give generic definitions.
第 I 类错误（当 H0 成立时拒绝 H0）与第 II 类错误（当 H0 不成立时未拒绝 H0）经常被混淆。评分标准仅对准确的措辞给分。此外，考生需要结合给定情景解释这些错误，而不仅仅是给出笼统的定义。
7. Chi-squared Test: Degrees of Freedom and Expected Frequencies | 卡方检验：自由度与期望频数
Many candidates incorrectly calculated the degrees of freedom for a chi-squared test of independence, using (rows × columns) instead of (rows − 1)×(columns − 1). Another mistake was rounding expected frequencies before calculating the test statistic, causing a loss of precision. The mark scheme required expected values to be used to at least one decimal place or more before rounding.
许多考生错误地计算了独立性卡方检验的自由度，使用了（行 × 列）而不是（行 − 1）×（列 − 1）。另一个错误是在计算检验统计量之前便对期望频数进行了舍入，导致精度下降。评分标准要求期望值至少保留一位小数后再进行后续计算。
8. Probability Generating Functions: Derivatives at 1 | 概率生成函数：在 1 处的导数
When using the probability generating function G(t), candidates often differentiated incorrectly or forgot to differentiate before substituting t = 1. The mark scheme clearly showed G'(t) and G”(t) with t→1 to find E(X) and Var(X). Algebraic errors in solving G'(1) and G”(1) were penalised, especially if the final mean or variance was given without working.
在使用概率生成函数 G(t) 时，考生常常求导错误或忘记在代入 t = 1 之前先求导。评分标准明确显示使用 G'(1) 和 G”(1) 来求 E(X) 和 Var(X)。求解 G'(1) 和 G”(1) 时的代数错误将被扣分，尤其在最终均值和方差未展示计算过程时。
9. Exponential Distribution and the Memoryless Property | 指数分布与无记忆性
The memoryless property P(X > s + t | X > s) = P(X > t) was often misapplied. Some candidates attempted to calculate the conditional probability using long-winded integration, overlooking the simple relation. The mark scheme rewarded recognition of the memoryless property, which shortened working considerably. However, stating the property without appropriate justification was insufficient.
无记忆性 P(X > s + t | X > s) = P(X > t) 经常被误用。一些考生试图通过冗长的积分计算条件概率，却忽视了这一简单关系。评分标准认可对无记忆性的识别，这能大幅简化计算过程。然而，仅陈述性质而无适当说明是不够的。
10. Sampling Distributions of the Sample Mean | 样本均值的抽样分布
When the population was not normal, candidates were expected to invoke the Central Limit Theorem to justify a normal approximation for the sample mean, provided the sample size was sufficiently large. Many omitted the statement “by the Central Limit Theorem” and lost communication marks. Additionally, the variance of the sample mean is σ²/n; using σ/√n as the variance was a frequent slip.
当总体非正态时，考生需借助中心极限定理来论证样本均值近似正态，前提是样本量足够大。许多考生遗漏了“根据中心极限定理”这一表述，从而丢失了表述分的。此外，样本均值的方差是 σ²/n；将 σ/√n 误作方差是常见的疏忽。
11. Combining Random Variables: Means and Variances | 随机变量的组合：均值与方差
In questions involving sums or differences of independent normal variables, candidates frequently misapplied the rules E(aX + bY) = aE(X) + bE(Y) and Var(aX + bY) = a²Var(X) + b²Var(Y). Carelessness with signs (particularly when subtracting variables) led to incorrect variances. The mark scheme demanded clear separation of the mean and variance calculations with correct algebraic signs.
在涉及独立正态变量求和或求差的问题中，考生经常错误运用规则 E(aX + bY) = aE(X) + bE(Y) 和 Var(aX + bY) = a²Var(X) + b²Var(Y)。对符号的疏忽（特别是在变量相减时）导致方差计算错误。评分标准要求均值和方差的计算清晰分开，且代数符号使用正确。
12. Interpretation of p-values and Conclusions | p 值的解释与结论
A significant number of candidates failed to correctly interpret the p-value in context. Merely stating “p < 0.05 so reject H0” without relating the conclusion to the problem context lost marks. The mark scheme expected a contextual sentence such as “there is sufficient evidence at the 5% level to suggest that …”. Vague or non-contextual conclusions were penalised.
大量考生未能正确结合背景解释 p 值。仅仅陈述“p < 0.05 故拒绝 H0”而未将结论与问题背景联系起来会扣分。评分标准期望一句结合背景的陈述，例如“在 5% 显著性水平下有充分证据表明……”。模糊或脱离背景的结论将被扣分。
Published by TutorHao | Further Mathematics 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