📚 Decoding the A-Level Further Maths Unit 4 June 2022 Mark Scheme: Key Question Types and Scoring Techniques | 解读A-Level进阶数学第四单元2022年6月评分方案：关键题型与得分技巧
Understanding the mark scheme for A-Level Further Mathematics Unit 4, often focused on Statistics 2 or an equivalent applied module, is just as important as mastering the content itself. The June 2022 mark scheme offers a transparent look at how examiners allocate method marks, accuracy marks, and independent marks across common question types such as hypothesis testing, normal approximations, confidence intervals, and chi-squared tests. This article breaks down the recurring patterns and scoring techniques, helping you to maximise your marks by writing solutions that match the mark scheme expectations.
对A-Level进阶数学第四单元（通常对应统计学2或同等应用模块）而言，理解评分方案与掌握知识本身同样重要。2022年6月的评分方案清晰展示了考官如何在假设检验、正态近似、置信区间和卡方检验等常见题型中分配方法分、准确性分和独立分。本文剖析其中的规律和得分技巧，助你写出符合评分方案预期的解答，稳稳拿下分数。
1. Structure of the Mark Scheme | 评分方案的结构
The June 2022 mark scheme follows the standard format: marks are labelled as M1, M2, A1, A2, B1, etc. An M mark is awarded for a correct method, even if the arithmetic is flawed; an A mark is for an answer that follows from previous work, and a B mark is independent of method. In multi-step statistics problems, you will often see marks split into a method for calculating a test statistic, an accuracy mark for the numerical result, and a further mark for a correct conclusion in context.
2022年6月评分方案遵循标准格式：分数标记为M1、M2、A1、A2、B1等。M分为方法分，即使算错也可能得到；A分是基于前面步骤得出的答案的正确性；B分独立于方法。在多步统计问题中，常会看到：计算检验统计量的方法分、数值结果的准确性分、以及根据上下文得出正确结论的另一个分数。
Paying attention to the exact wording required for conclusions is vital. For a hypothesis test, writing simply "reject H₀" without comparing to a significance level or critical value often loses a mark. The mark scheme expects a clear comparison: "Since 2.34 > 1.96, there is sufficient evidence to reject H₀ at the 5% level."
关注结论所需的准确措辞非常重要。假设检验中，仅写“拒绝原假设”而不与显著性水平或临界值进行比较，通常会丢分。评分方案期望清晰的比较：“因为 2.34 > 1.96，所以在 5% 显著性水平下有充分证据拒绝 H₀。”
2. Method Marks vs Accuracy Marks | 方法分与准确性分的区别
M marks are earned by showing a correct approach, such as writing the formula for a z-test statistic or the expression for expected frequencies in a chi-squared test. A marks are then given for the correct numerical result, provided it follows from the method. If the method is incorrect, no A marks are available. However, if a small slip like a calculator error leads to a wrong final answer, the M mark can still be awarded.
M分通过展示正确方法获得，例如写出z检验统计量公式或卡方检验中预期频数的表达式。然后A分根据方法得出的正确数值结果给予。如果方法错误，A分就无法拿到。但如果只是计算器按错之类的小失误导致最终答案错误，M分仍然可能得到。
A common trap is failing to state the exact test statistic with an appropriate label, e.g. "Z = -1.645". The mark scheme sometimes gives a B1 for stating the correct distribution, like "X ~ N(50, 4/100)", before any calculation. So always include distribution information explicitly.
常见的陷阱是没有清晰地写出带合适标签的检验统计量，例如“Z = -1.645”。评分方案有时会单独给出B1分，用于声明正确的分布，例如在计算前写“X ~ N(50, 4/100)”。因此始终要明确写出分布信息。
3. Hypothesis Testing – Mark Allocation Patterns | 假设检验 – 分数分配规律
In the June 2022 paper, hypothesis testing questions typically allotted 6–8 marks. A typical mark distribution was: B1 for stating H₀ and H₁ correctly, M1 for finding the test statistic using the correct formula, A1 for the correct value, M1 for comparing with a critical value or using probability, A1 for the correct comparison, and a final B1 or M1 for the conclusion in context. Do not forget to define the parameter (μ or p) early on.
在2022年6月的试卷中，假设检验题目通常占6–8分。典型的分数分配为：正确写出H₀与H₁得B1，用正确公式求出检验统计量得M1，正确数值得A1，与临界值比较或使用概率得M1，正确比较得A1，最后结合上下文得出结论得B1或M1。不要忘记尽早定义参数（μ 或 p）。
For a one-tailed test on the mean with known variance, the step-by-step mark scheme requires: definition of μ, hypotheses, calculation of the standard error σ/√n, substitution into the z-formula, numerical value of z, comparison with 1.645 (or a p-value), and a concluding statement. Omitting the standard error step may cause loss of the M mark.
对于已知方差下均值的单尾检验，分步评分方案要求：定义μ、假设、计算标准误σ/√n、代入z公式求出z值、与1.645（或p值）比较、以及结论陈述。遗漏标准误步骤可能导致丢失M分。
4. Normal Approximation to Poisson – Continuity Correction | 泊松分布的正态近似 – 连续性校正
re allocated for stating the approximating distribution, writing the continuity-corrected bound, standardising correctly, and obtaining the probability.
当泊松分布均值为λ时用正态分布近似，评分方案希望你使用连续性校正。例如，在λ = 10时求 P(X ≤ 12)，要用 N(10, 10) 并计算 P(X ≤ 12.5)。分数分配包括：写出近似的正态分布、写出连续性校正后的边界、正确标准化、以及得到概率值。
Many candidates forget to write the corrected boundary as 12.5 for "<" or "≤" and 11.5 for "≥" or ">" ing normal is λ, not λ/n, so be careful not to confuse with the sample mean approximation.
不少考生忘记将“<”或“≤”的校正边界写成12.5，或将“≥”或“>”写成11.5，从而丢失校正步骤的M分。此外，近似正态的方差是λ，而不是λ/n，注意不要与样本均值的近似相混淆。
5. Confidence Intervals – Formula and Interpretation | 置信区间 – 公式与解释
numerical substitution. A typical 6-mark question would give M1 for the correct formula x̄ ± z × σ/√n, A1 for the correct critical z-value (e.g., 1.96 for 95%), A1 for the correct interval endpoints, and then a further A1 or B1 for interpreting the interval in context. Simply giving the interval without stating "We are 95% confident that the true mean lies between …" may miss the interpretation mark.
2022年试卷中的置信区间题目注重精确的数值代入。一道典型的6分题会给出：正确公式 x̄ ± z × σ/√n 得M1，正确临界z值（如95%时1.96）得A1，端点计算正确得A1，然后结合上下文解释区间再得A1或B1。只给出区间而不说“我们有95%信心真实均值在…之间”可能会丢掉解释分。
When the population variance is unknown and a t-distribution is used, the mark scheme allocates a B1 for stating the degrees of freedom and a mark for the correct t critical value. Make sure to look up the correct t-value from tables and avoid mistakenly using z for small samples.
当总体方差未知而使用t分布时，评分方案会给出B1用于陈述自由度，并为正确的t临界值给分。确保从表中查找正确的t值，避免小样本下误用z值。
6. Chi-Squared Tests – Expected Frequencies and Degrees of Freedom | 卡方检验 – 预期频数与自由度
otal / grand total) and for stating the null and alternative hypotheses. The test statistic formula Σ (O – degrees of freedom must be correctly determined. A key source of lost marks is using (rows – 1) × (columns – 1) carelessly when some cells are combined.
卡方拟合优度或独立性检验的题目在评分方案中结构清晰。计算每个预期频数（行合计×列合计/总合计）得到M分，写出原假设和备择假设也能得分。检验统计量公式 Σ (O – E)² / E 必须正确使用，自由度必须正确确定。失分的常见原因是当某些格子合并时随意使用 (行数–1)×(列数–1) 计算自由度。
The June 2022 mark scheme also included a B1 for stating the critical value from the chi-squared table and a B1 for the correct conclusion about independence or goodness-of-fit, together with context. Always reference the significance level and degrees of freedom in the conclusion.
2022年6月的评分方案还包括给出卡方表临界值的B1分，以及关于独立性或拟合优度结合上下文得出正确结论的B1分。记得在结论中提及显著性水平和自由度。
7. Continuous Random Variables – PDFs and CDFs | 连续随机变量 – 概率密度函数与累积分布函数
s are given for the correct piecewise definition. A common error is forgetting to state that F(x) = 0 below the lower bound and F(x) = 1 above the upper bound.
涉及概率密度函数的题目需要在区间上积分pdf。评分方案给分方式：正确建立积分得M1，正确积分得A1，正确代入上下限再得A1。若要求累积分布函数F(x)，则分数给在正确的分段定义上。常见错误是忘记在定义域下方写明F(x)=0、上方写明F(x)=1。
ility) and solve. Marks are given for forming the equation and for the final numerical value. Always check that the solution lies within the support of the distribution.
求中位数或百分位数时，评分方案期望你设F(m)=0.5（或相应概率）并求解。建立方程可得M分，正确数值结果可得A分。务必检查解是否落在分布的支撑集内。
8. Central Limit Theorem – Wording Requirements | 中心极限定理 – 措辞要求
Many mark schemes for CLT-related problems explicitly list a mark for stating that "since n is large, the sample mean is approximately normally distributed" or that "the distribution of X̄ is approximately N(μ, σ²/n)". This must appear before standardisation. The June 2022 paper awarded this B1 only if the statement was clearly linked to the Central Limit Theorem.
许多涉及中心极限定理的评分方案都明确列出：必须写出“由于n足够大，样本均值近似服从正态分布”或“X̄的分布近似为 N(μ, σ²/n)”，才能得到这一分数，且必须在标准化之前出现。2022年6月试卷只在明确关联到中心极限定理时才给出此B1分。
Even if σ is unknown and estimated by s, the CLT still applies, and you must mention that the approximation is valid because of the large sample size. Losing this mark often costs 1 or 2 marks overall, so always include the justification.
即使未知σ而用s估计，中心极限定理依然适用，你必须提及因样本量大所以近似有效。丢失这一分往往导致整体掉1-2分，所以要始终包含该理由。
9. Avoiding Common Loss-Marks Pitfalls | 避免常见失分陷阱
From the June 2022 mark scheme, the most frequent avoidable mistakes included: using an incorrect critical value (e.g., 1.645 for a two-tailed test instead of 1.96), forgetting the continuity correction in normal approximations, not showing the substitution step explicitly, and omitting units or context in the final statement.
从2022年6月的评分方案来看，最频繁的可避免错误有：使用错误的临界值（例如双尾检验用了1.645而非1.96），正态近似中忘记连续性校正，没有明确写出代入步骤，以及在最后陈述中遗漏单位或上下文。
Another trap is giving too many decimal places in final answers. While A marks can tolerate slight rounding, if an answer is expected to 3 significant figures, providing 5 or 6 can sometimes be penalised if it contradicts the required accuracy. Stick to 3 s.f. unless otherwise stated.
另一个陷阱是最终答案给出过多小数位。虽然A分容许轻微的舍入差异，但若答案要求3位有效数字而你给出5或6位，有时会因不符合要求精度而罚分。除非另有说明，坚持使用3位有效数字。
10. High-Scoring Techniques and Time Management | 高分解题技巧与时间管理
To score highly on the Unit 4 exam, read the mark scheme clues within the question: when a part carries 3 marks, expect to demonstrate a method, a calculation, and a final answer. Show all working, even for simple steps, because the M mark is often awarded for the first correct operation. Use exact values (fractions or surds) during intermediate steps to avoid rounding errors later.
想在第四单元考试中斩获高分，应抓住题目中隐藏的评分线索：若某小题为3分，就预计要展示方法、计算过程和最终答案。所有步骤都要展示，哪怕是简单的步骤，因为M分常奖励第一个正确的运算。中间步骤使用精确值（分数或根式）以避免后面的舍入误差。
Timing is crucial: allocate roughly 1.5 minutes per mark. For longer hypothesis testing or chi-squared questions, leave time for the conclusion. A quick scan of the mark scheme after attempting past papers will reveal exactly where marks are dropped and how to adjust your presentation to secure them.
时间分配至关重要：大约每1.5分钟对应1分。对于较长的假设检验或卡方检验题目，留出时间写结论。做完历年真题后快速浏览评分方案，能准确揭示何处容易丢分，以及如何调整答题呈现方式以锁定这些分数。
Published by TutorHao | Mathematics Revision Series | aleveler.com
更多咨询请联系16621398022（同微信）
或截图后用「扫一扫 → 相册」选择本图