📚 A-Level Maths: Example Responses MA04 Unit S2 – Common Mistakes | A-Level数学：MA04单元S2示例答案易错点总结
Working through example responses for Unit S2 often reveals a handful of recurring errors that trip up even well-prepared students. This article unpacks the most common pitfalls seen in past MA04-style scripts, covering hypothesis testing, approximations, continuous random variables, sampling distributions, confidence intervals, p-values, and discrete probability models. Each mistake is illustrated with typical student work, followed by a clear correction rooted in the S2 specification.
分析MA04单元S2的示例答案时，总能发现一些反复出现的错误，这些错误即使准备充分的学生也难以完全避免。本文剖析了过去试卷中最常见的易错点，涵盖假设检验、近似方法、连续型随机变量、抽样分布、置信区间、p 值解释以及离散概率模型。每个错误都配有典型的考生作答，并给出紧扣S2考纲的纠正方法。
1. Confusing One-Tailed and Two-Tailed Hypothesis Tests | 混淆单尾与双尾假设检验
Many students write down the alternative hypothesis H₁ without thinking about its direction and then blindly halve the significance level. For example, when testing whether a new teaching method increases the mean score, H₁ should be μ > μ₀ (right‑tailed). Mark schemes frequently penalise those who incorrectly set H₁: μ ≠ μ₀ and compare the test statistic with a two‑tail critical value, often leading to a non‑rejection when a significant result is actually present.
许多学生在写下备择假设 H₁ 时没有考虑方向，然后盲目地将显著性水平减半。比如，要检验一种新教学法是否提高平均分，H₁ 应为 μ > μ₀（右尾检验）。评分方案常会扣掉那些错误设置 H₁: μ ≠ μ₀ 并拿它跟双尾临界值比较的考生，这经常导致本该显著的结果变成不显著。
In S2 exam questions, the wording ‘has increased’ or ‘has decreased’ explicitly signals a one‑tailed test. If the problem says ‘has changed’, a two‑tailed test is required. Remember that in a one‑tailed test all α is placed in one tail, whereas in a two‑tailed test α is split equally. This also affects the p‑value: for a two‑tailed test you must double the one‑tail probability before comparing with α.
在S2考题中，文案中的“增加了”或“减少了”明确指向单尾检验。如果题目说“发生了变化”，则需要双尾检验。记住：单尾检验中全部的 α 放在一侧，而双尾检验要把 α 平分。这也会影响 p 值：对于双尾检验，必须把单尾概率翻倍后再与 α 比较。
H₀: μ = 50, H₁: μ > 50 → one‑tailed, p‑value = P(Z > z)
H₀: μ = 50, H₁: μ ≠ 50 → two‑tailed, p‑value = 2 × P(Z > |z|)
2. Applying Poisson Approximation Without Checking Conditions | 未检验条件就使用泊松近似
ghtly between exam boards, but for MA04 students are expected to verify that n is large, p is small, and usually np < 10 or np < 5. If the original binomial has p = 0.4 and n = 200, then np = 80, which is far too large for a Poisson approximation — here a normal approximation is appropriate instead.
一个典型的S2错误是，一看到二项分布的 n 较大就直接套用泊松分布。各考试局的经验准则略有差异，但在MA04中学生应当验证 n 大、p 小，并且通常是 np < 10 或 np < 5。如果原来的二项分布有 p = 0.4 且 n = 200，那么 np = 80，太大了，不适合用泊松近似——此时应当改用正态近似。
) under B(100, 0.02) are found from Poisson tables with λ = 2. A common mistake is to keep using binomial calculations alongside the Poisson, mixing up the two models.
一定要同时检查 n 的大小和 p 的值。泊松参数 λ 设为 np。如果近似合理，如 B(100, 0.02) 下的 P(X ≤ 3) 就通过 λ = 2 的泊松表得到。常见的错误是同时混用二项和泊松，把两个模型搅在一起。
| Binomial | Poisson suitable? |
| n=80, p=0.01 | Yes (n large, p small, np=0.8) |
| n=500, p=0.05 | np=25 → borderline, normal may be better |
| n=30, p=0.2 | No, n is not large enough for Poisson |
3. Mishandling Continuity Correction in Normal Approximations | 正态近似中连续性校正处理不当
. The classic mistake is to ignore the correction altogether, which leads to an inaccurate probability. Another widespread error is getting the direction of the ± 0.5 adjustment muddled: using ‘+0.5’ when ‘−0.5’ is required.
当用连续的正态分布去近似离散分布（二项或泊松）时，必须进行连续性校正。经典错误是完全忽略校正，导致概率不准确。另一个普遍错误是把 ±0.5 的调整方向搞反：该用 −0.5 的时候用了 +0.5。
ariable X, the probability P(X ≤ k) includes the value k. Under the approximating normal Y ~ students simply copy the uncorrected k and miss the subtle distinction between ≤ and <.
对于离散变量 X，概率 P(X ≤ k) 包含了 k 这个点。在近似的正态 Y ~ N(μ, σ²) 下，校正边界变为 Y ≤ k + 0.5。对于不包含 k 的 P(X < k)，则要用 Y ≤ k − 0.5。很多学生直接套用未校正的 k，忽略了 ≤ 和 < 之间的细微差别。
P(X ≤ 12) → use Y ≤ 12.5
P(X ≤ 12) → 用 Y ≤ 12.5
P(X < 12) → use Y ≤ 11.5
P(X < 12) → 用 Y ≤ 11.5
P(X ≥ 12) → use Y ≥ 11.5
P(X ≥ 12) → 用 Y ≥ 11.5
P(X > 12) → use Y ≥ 12.5
P(X > 12) → 用 Y ≥ 12.5
Always sketch a rough bar chart and shade the area you want. This visual check instantly reveals whether you need to extend the boundary by 0.5 to the left or right.
每次都画一张粗略的条形图并涂出你要的区域。这种直观检验能马上揭示你到底需要把边界向左还是向右延伸 0.5。
4. Mixing Up PDF and CDF in Continuous Random Variables | 连续型随机变量中混淆概率密度函数与累积分布函数
obability density function f(x) does not give a probability directly; it only provides the ‘ n and quartile calculations, where students set f(m) = 0.5 instead of solving ∫ₐₘ f(x) dx = 0.5.
概率密度函数 f(x) 并不直接给出概率，它只提供“密度”。一个严重错误是写出 P(1 < X < 2) = f(2) − f(1)，而不是对 f(x) 从 1 积分到 2。这种误用常出现在中位数和四分位数的计算中，学生令 f(m) = 0.5，而不是求解 ∫ₐₘ f(x) dx = 0.5。
formula for f(x), first find the constant by setting the total area to 1, then obtain F(x) by integration. Don’t forget to split the integration range when f(x) is defined piecewise.
永远使用累积分布函数 F(x) 或在区间上积分 f(x)。中位数 m 满足 F(m) = 0.5。如果题目给出了 f(x) 的表达式，先通过总面积等于 1 求出常数，再通过积分得到 F(x)。当 f(x) 是分段定义时，别忘了分段积分。
screte expectation formulas to continuous distributions. Variance also requires E(X²) − [E(X)]², and E(X²) = ∫ x² f(x) dx.
关于期望的错误，学生常常计算 E(X) = ∫ x f(x) dx 时忘记在积分内乘上 x，或者错误地把离散期望公式套用在连续分布上。方差也需要 E(X²) − [E(X)]²，而 E(X²) = ∫ x² f(x) dx。
5. Misapplying the Central Limit Theorem for Sample Means | 对样本均数的中心极限定理误用
The Central Limit Theorem states that for a sufficiently large sample size, the distribution of the sample mean x̄ is approximately Normal regardless of the population’s shape. A typical S2 error is to assume that a sample of size 5 or 10 from a highly skewed population will be normally distributed, or to forget that the standard deviation of x̄ is σ/√n, not σ.
中心极限定理指出，当样本容量足够大时，不管总体是什么形状，样本均值 x̄ 的分布都近似正态。典型的 S2 错误是，认为从一个高度偏态的总体中抽取容量 5 或 10 的样本，其样本均值也会服从正态分布；或者忘记了 x̄ 的标准差是 σ/√n，而不是 σ。
e CLT and state that n is large (usually n ≥ 30). When computing probabilities involving x̄, the test statistic is Z = (x̄ − μ) / (σ/√n). Forgetting the √n denominator is a mark‑losing mistake that can easily be avoided by always writing the standard error.
除非总体本身服从正态分布，S2 题目期望你援引中心极限定理并说明 n 足够大（通常 n ≥ 30）。在计算涉及 x̄ 的概率时，检验统计量为 Z = (x̄ − μ) / (σ/√n)。遗忘分母中的 √n 是一个丢分错误，养成每次写出标准误的习惯就能轻易避免。
Another common slip occurs when students are given the population variance σ² and mistakenly treat the sample mean’s variance as σ²/n only after a reminder. The distinction between σ² and σ²/n is fundamental in all sampling distribution work.
另一个常见的疏漏是，学生虽然知道总体方差 σ²，却总是在被提醒后才想起样本均值的方差是 σ²/n。在所有抽样分布的内容中，区分 σ² 和 σ²/n 都是基础。
6. Confidence Interval Calculation and Interpretation Errors | 置信区间的计算与解释错误
Confidence intervals in S2 are typically of the form x̄ ± z × σ/√n, where z is obtained from the Normal distribution for a given confidence level. Mistakes include using the wrong z‑value (e.g., 1.645 for 95% confidence instead of 1.96), forgetting to check whether σ is known or estimated, and failing to state the assumptions (random sample, normal population or large n).
S2 中的置信区间通常形如 x̄ ± z × σ/√n，其中 z 根据给定的置信水平由正态分布查得。常见错误包括用了错误的 z 值（如 95% 置信水平下用 1.645 而不用 1.96），忘记检查 σ 是已知的还是需要估计的，以及没有陈述假设条件（随机样本、正态总体或大样本）。
Essay‑style questions often ask for an interpretation of a 95% confidence interval. The statement ‘there is a 95% chance that μ lies in the interval’ is incorrect. The correct interpretation is that if many samples were taken and a 95% confidence interval constructed each time, about 95% of those intervals would contain μ. The probability applies to the process, not to the specific interval.
小论文式题目常要求解释 95% 置信区间。声称“μ 落在该区间内的概率为 95%”是不正确的。正确的解释是：如果重复抽取很多样本，并每次构造一个 95% 置信区间，那么大约 95% 的区间会包含 μ。概率针对的是整个流程，而不是某个具体的区间。
A practical check: if the calculated interval is absurdly wide or does not match the context, revisit the standard error calculation. Double‑check whether ‘population standard deviation’ or ‘sample standard deviation’ was intended.
一个实用的检查：如果算出的区间宽得离谱或与背景不符，就要重新审视标准误的计算。反复确认题目意图是“总体标准差”还是“样本标准差”。
7. Misunderstanding p-values and Significance | 对 p 值与显著性的误解
p‑value confusion is widespread. A low p‑value indicates that the observed data would be unlikely if the null hypothesis H₀ were true. It does not measure the magnitude of an effect or the probability that H₀ is false. In exam answers, many students write ‘p < 0.05, therefore H₀ is accepted’ — which is doubly wrong. The correct conclusion is ‘reject H₀’ when p < α, else ‘do not reject H₀’.
对 p 值的误解很普遍。一个很小的 p 值表示，如果原假设 H₀ 为真，观测到这样的数据是不太可能的。它并不衡量效应的大小，也不是 H₀ 为假的概率。在考试答案中，许多学生写道“p < 0.05，因此接受 H₀”——这犯了双重错误。正确的结论是：当 p < α 时“拒绝 H₀”，否则“不拒绝 H₀”。
Another error is stating that a p‑value of 0.03 means the result is ‘more significant’ than a p‑value of 0.001 because 0.03 is bigger. Significance is a binary decision at the chosen α; a smaller p‑value simply provides stronger evidence against H₀, but once the threshold is crossed, the decision is the same.
另一个错误是声称 p 值 0.03 的结果比 0.001 的结果“更显著”，因为 0.03 更大。在选定的 α 水平下，显著性是二元的判断；更小的 p 值只是提供了更强的反对 H₀ 的证据，但一旦越过阈值，决策都一样。
In two‑tailed tests, students sometimes compare the one‑tail probability with α without doubling it, thereby claiming significance when none exists. Always convert the test statistic to a p‑value properly and then state your conclusion in the context of the problem.
在双尾检验中，学生有时直接拿单尾概率与 α 比较而不翻倍，从而宣称有显著性，实则没有。务必正确将检验统计量转换为 p 值，然后在题目背景下给出结论。
8. Building Incomplete or Incorrect Discrete Probability Distributions | 构建不完整或错误的离散概率分布
When a question asks for the probability distribution of a discrete random variable, a frequent oversight is failing to list all possible values or miscalculating individual probabilities so that they do not sum to 1. For instance, in a ‘ball‑drawing without replacement’ problem, students might treat selections as independent events and use a constant probability, which leads to a distribution that doesn’t add up.
当题目要求一个离散随机变量的概率分布时，常见疏忽是未列出所有可能取值，或者算错各个概率导致总和不为 1。例如，在“无放回摸球”问题中，学生可能把各次抽取当作独立事件并用统一的概率，结果得到的分布加起来不等于 1。
ble, check that Σ P(X = x) = 1. If not, there is a missing outcome or an arithmetic error. When finding P(X = x) from a probability generating function or from given parameters, re‑eading the domain of X is crucial; probabilities for x values outside the domain are zero.
务必系统地列出样本空间，并用组合推理或树状图分配概率。构建好表格后，检验 Σ P(X = x) = 1。如果不为 1，说明有遗漏的取值或计算错误。在从概率母函数或给定参数求 P(X = x) 时，重读 X 的取值范围至关重要；超出该范围的 x 值概率为零。
tional probabilities often appear in S2 distribution problems. A common slip is to use P(A|B) = P(A ∩ B) / P(B) but then substitute P(A ∩ Unless explicitly stated, do not assume independence in conditional probability questions.
S2 分布问题中常出现条件概率。常见失误是使用 P(A|B) = P(A ∩ B) / P(B) 时，把 P(A ∩ B) 代换成 P(A) × P(B)，隐含了独立性假设。除非题目明示，不要在条件概率题中假设独立性。
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