📚 AS CAIE Statistics: Exam Techniques and Marking Criteria | AS CAIE 统计：答题技巧与评分标准
Understanding how CAIE awards marks in the AS Statistics paper can dramatically improve your performance. This guide breaks down the marking principles, common pitfalls, and practical strategies for Probability & Statistics 1 (Paper 5). Whether you are tackling probability distributions, data representation, or calculations, applying these techniques will help you secure every possible method and accuracy mark.
理解 CAIE 在 AS 统计试卷中如何评分可以显著提高你的表现。本指南详细拆解了评分原则、常见失分点以及概率与统计 1（试卷 5）的实用策略。无论你面对的是概率分布、数据表示还是计算题，应用这些技巧都能帮你拿到每一个可能的方法分和准确性分。
1. Understanding the Exam Structure | 理解考试结构
The AS Statistics paper (Paper 5) lasts 1 hour 15 minutes and carries 50 marks. It typically contains 5 to 7 questions covering the full syllabus: representation of data, measures of central tendency and variation, probability, permutations and combinations, discrete random variables, the binomial distribution, and the normal distribution. Knowing the allocation of marks helps you decide how much time to spend on each question.
AS 统计试卷（试卷 5）时长为 1 小时 15 分钟，满分 50 分。试卷通常包含 5 至 7 题，覆盖全部考纲内容：数据表示、集中趋势和变异量数、概率、排列组合、离散型随机变量、二项分布以及正态分布。了解分值的分配能帮助你决定每道题该花多长时间。
Questions early in the paper are often shorter and more straightforward, while later questions integrate multiple concepts. Marks per question can range from 4 to 14. You must read the instructions on the front cover, which specify rounding rules (normally 3 significant figures) and the use of calculators. Misinterpreting these basics can cost you easy marks.
试卷前面的题目通常较短且直接，后面的题目则会综合多个概念。每题的分值从 4 分到 14 分不等。你必须仔细阅读封面上的说明，其中会指明四舍五入规则（通常为 3 位有效数字）以及计算器的使用要求。误解这些基本要求会让你丢掉容易拿到的分数。
2. Marking Criteria and Method Marks | 评分标准与方法分
CAIE uses a robust mark scheme with three main types of marks: M (method mark), A (accuracy mark), and B (independent mark). An M mark is awarded for using a correct method, even if your final answer is wrong, provided no error of principle is made. An A mark requires both the method and the final numerical result to be correct; it usually depends on the preceding M mark. A B mark is given for a correct statement, value, or diagram element that does not rely on a method path.
CAIE 使用一种严谨的评分方案，主要包括三种分数：M（方法分）、A（准确性分）和 B（独立分）。只要没有原则性错误，即使你的最终答案不正确，采用正确的方法也能获得 M 分。A 分则要求方法和最终数值结果都正确，它通常依赖于前面的 M 分。B 分授予不依赖于方法路径的正确的陈述、数值或图表元素。
| Mark Type | Meaning | Key Feature |
|---|---|---|
| M | Method mark | Awarded for a valid process, even if working contains a slip. |
| A | Accuracy mark | Requires both correct method and accurate answer; dependent on M. |
| B | Independent mark | Given for a standalone correct fact, plot, or statement. |
Always show your working clearly. Even if you cannot reach the final answer, a correct equation, formula, or substitution can earn M marks. Conversely, a correct answer with no working will not receive an M mark if the mark scheme demands method evidence. Write down the formula you are using before you substitute numbers.
始终要清楚地展示你的解题步骤。即便你无法算出最终答案，正确的方程、公式或代入也可能拿到 M 分。反之，如果评分标准要求展示方法，没有计算过程的正确答案将无法获得 M 分。在代入数字之前先把所使用的公式写下来。
3. Accuracy and Final Answers | 准确性与最终答案
Unless the question specifies otherwise, final answers should be given to 3 significant figures. This applies to probabilities, means, standard deviations, and z-values. However, when working with currency or data that naturally have two decimal places, follow that real-world convention. Answers that are not rounded suitably may lose the A mark.
除非题目另有规定，最终答案应保留 3 位有效数字。这适用于概率、均值、标准差以及 z 值。不过，在处理货币或本来就只有两位小数的数据时，应遵循其真实世界的惯例。未恰当四舍五入的答案可能会丢失 A 分。
Keep sufficient accuracy during intermediate steps. CAIE expects you to retain at least 4 significant figures in your working to avoid premature rounding errors. When using statistical tables, record the exact table value and then apply rounding only at the final answer. If you use a calculator's statistical functions, store results directly in memory rather than retyping truncated values.
在中间计算步骤中要保持足够的精度。CAIE 要求你在计算过程中至少保留 4 位有效数字，以避免过早舍入导致的错误。使用统计表时，记录精确的表值，只在最后一步才进行四舍五入。如果你使用计算器的统计功能，应直接将结果存入存储器，而不是重新键入被截断的数值。
For probabilities obtained from the normal distribution, standard practice is to give the answer to 4 decimal places. If a question says "find P(Z < 1.234)," your final answer should normally be 0.8910, or 0.891 as 3 s.f. unless told otherwise. The front cover of the exam paper will confirm the default accuracy rule.
对于由正态分布得到的概率，标准做法是给出 4 位小数。如果题目要求“求 P(Z < 1.234)”，你的最终答案通常应为 0.8910，或者按 3 位有效数字给出 0.891，除非另有说明。试卷封面会确认默认的精确度规则。
4. Working and Presentation | 解题步骤与呈现
chain of equalities that are not actually equal. Clearly label probabilities, for example 'P(X = 2) = ...' les you introduce. A well-structured solution helps the examiner assign marks efficiently.
你的书写应当符合逻辑且易于理解。正确使用等号，不要把实际并不相等的量用等号连起来。清楚地标注概率，例如 'P(X = 2) = ...'，并定义你引入的任何变量。一个结构清晰的解答能帮助考官有效评分。
When tackling a binomial probability, write the distribution statement: X ~ B(n, p). For a normal approximation, state the continuity correction and the standardized value. These explicit steps not only reveal your method but also prevent careless mistakes. Even if the final answer is incorrect, the distribution statement often carries a B mark.
处理二项分布概率时，要写出分布声明：X ~ B(n, p)。对于正态近似，要写明连续性校正项以及标准化后的值。这些明确的步骤不仅展现出你的方法，还能防止粗心错误。即使最终答案错了，分布声明本身往往也能拿到 B 分。
In graphical questions such as box plots or cumulative frequency diagrams, use a ruler and label the axes with clear scales. Plot points accurately and indicate the median, quartiles, and extreme values. A box plot with correct quartiles but a missing label on the axis may still earn the B marks if the shape is right, yet incomplete presentation can risk marks. Always check the mark scheme demands for scale and labels.
在箱线图或累积频率图等图形题中，要用尺子作图，并在坐标轴上标出清晰的刻度。准确描点并标出中位数、四分位数以及极值。若箱线图的四分位数正确但坐标轴缺少标签，如果形状正确仍可能获得 B 分，但不完整的呈现方式可能会丢分。务必核对评分标准对刻度和标签的要求。
5. Handling Probability Questions | 概率题处理技巧
Probability questions often involve tree diagrams, Venn diagrams, or conditional probabilities. Draw a clear diagram and label all branches or regions with probabilities. Write the multiplication and addition rules explicitly. For P(A ∩ B) or P(A | B), show the formula you are using, then substitute the numbers.
概率题通常会涉及树状图、韦恩图或条件概率。画出清晰的图表，并在所有分支或区域标注概率。明确写出乘法定理和加法定理。对于 P(A ∩ B) 或 P(A | B)，先写出你使用的公式，再代入数字。
When the question asks you to determine whether events are independent, do not just state 'yes' or 'no'. Show the check: P(A) × P(B) = P(A ∩ B) or P(A| mutual exclusivity; remember that mutually exclusive events cannot occur together, so P(A ∩ B) = 0. This difference is frequently tested.
当题目要求你判断事件是否独立时，不要只写“是”或“否”。要展示验证过程：P(A) × P(B) = P(A ∩ B) 或者 P(A|B) = P(A)，并结合题目情境得出结论。常见的失误是把独立性与互斥性混淆；要记住互斥事件不能同时发生，因此 P(A ∩ B) = 0。这个区别经常被考查。
6. Permutations and Combinations | 排列与组合
: e.g. ⁵P₂ or ⁵C₂. Setting out the terms of the multiplication principle step by step is often safer than only plugging into the calculator. For arrangements with repeated items, use the standard formula n!/(p!×q!×...).
在排列组合问题中，首先要判断顺序是否重要，以及是否允许重复。使用清晰的符号写出 nPr 或 nCr，例如 ⁵P₂ 或 ⁵C₂。将乘法原理的各项逐步列出往往比只在计算器里输入更安全。对于含有重复元素的排列，使用标准公式 n!/(p!×q!×...)。
If the problem involves selections with restrictions (e.g. 'at least one woman' or 'must include a particular item'), break the case into mutually exclusive scenarios. Calculate the number of ways for each scenario and add them together. Always check for overcounting: subtracting the unwanted arrangements from the total is often a clever alternative.
如果问题涉及有限制条件的选择（例如“至少一名女性”或“必须包含某一特定物品”），可以把情形拆分成互斥的情景。计算每种情景的排列数，再将它们相加。始终要检查是否重复计数：从总数中减去不符合要求的排列数通常是一个巧妙的替代方法。
7. Discrete Random Variables and Expectation | 离散型随机变量与期望
m to 1, which is a quick validation. Use the formulas: E(X) = Σ x p(x) and Var(X) = Σ x² p(x) − [E(X)]². Show the Σ columns in your working; this structure makes it easy to recover if you make an arithmetic slip.
对于离散型随机变量，要清晰地以表格呈现概率分布，列出每个取值及其概率。各概率之和必须等于 1，这是一个快速验证。使用公式：E(X) = Σ x p(x) 以及 Var(X) = Σ x² p(x) − [E(X)]²。在答题过程中列出 Σ 计算列；这样的结构能让你在出现算术失误时轻松找回分数。
When the distribution arises from a word problem, define your random variable first: Let X be the score/number of... This B or M mark is often the easiest to gain. Also show the working for E(X²) and variance step by step. Many candidates skip the E(X²) expansion and lose marks when they misapply the variance formula.
当分布来自文字题时，先定义你的随机变量：令 X 表示……的得分/数目。这个 B 或 M 分往往是最容易拿到的。同时要逐步展示 E(X²) 和方差的计算过程。很多考生会跳过 E(X²) 的展开，导致错误地使用方差公式而丢分。
8. Binomial Distribution | 二项分布
rials, independent trials, constant probability of success, and only two outcomes. Write X ~ B(n, p) immediately. For calculating P(X = r), use the formula: P(X = r) = ⁿCᵣ pʳ qⁿ⁻ʳ les, make sure you match the correct value of p and the block of cumulative probabilities.
要明确写出二项分布的条件：固定试验次数、各次试验相互独立、每次成功的概率恒定，以及只有两种可能的结果。立刻写出 X ~ B(n, p)。计算 P(X = r) 时，使用公式：P(X = r) = ⁿCᵣ pʳ qⁿ⁻ʳ，其中 q = 1-p。使用分布表时，务必确保匹配正确的 p 值以及累积概率的区间。
CAIE often expects you to use your calculator' how the candidate number you are evaluating: e.g. P(X ≤ 4) = ... The working line could be 'Using B(10, 0.35), P(X≤4) = 0.7515 (from calculator/table)' demonstrates method and avoids missing out on method marks if your key entry is mistaken.
CAIE 通常期望你使用计算器的二项分布概率密度函数或累积分布函数来加快计算。即便如此，仍要写出你要计算的量：例如 P(X ≤ 4) = ……。解题行可以是“利用 B(10, 0.35)，P(X ≤ 4) = 0.7515（来自计算器/表格）”。这样可以展示方法，避免在按键输入错误时丢失方法分。
9. Normal Distribution | 正态分布
The standard normal variable Z ~ N(0, 1) is central. For a general normal X ~ sketch of the bell curve and shade the required area. This helps you decide whether to use Φ(z), 1 − Φ(z), or the difference of two Φ values.
标准正态变量 Z ~ N(0, 1) 处于核心地位。对于一般正态 X ~ N(μ, σ²)，使用标准化公式 Z = (X − μ)/σ。始终画一张简单的钟形曲线草图并给所需区域涂上阴影。这能帮助你判断是使用 Φ(z)、1 − Φ(z) 还是两个 Φ 值的差。
When a question involves working backward (finding μ or σ or a percentile), set up the equation with the Z-value corresponding to the given probability. Write down the inverse normal statement clearly. Most board tables give values of z for specific probabilities; do not confuse the lower-tail and upper-tail areas.
当题目涉及反向求解（求 μ、σ 或百分位数）时，要用与给定概率相对应的 Z 值建立方程。清晰地写出逆正态声明。多数考试局的表格都会给出特定概率对应的 z 值，不要混淆下尾面积和上尾面积。
For continuity correction in normal approximations to binomial, use the half-unit adjustment: P(X < r) becomes P(X' < r − 0.5). Always state the approximation conditions, usually np > 5 and nq > 5. Showing this check can earn a B or M mark even before computing the probability.
对于二项分布的正态近似连续性校正，使用半单位修正：P(X < r) 变为 P(X' < r − 0.5)。始终说明近似条件，通常是 np > 5 且 nq > 5。展示这一检验即使在计算概率之前也可能获得 B 或 M 分。
10. Interpreting Results and Context | 结果解读与情境
Statistics is about making sense of data. In the exam, you must write a short conclusion in context when asked to interpret a probability, a mean, or a comparison. For example, "The probability that more than 3 items are defective is 0.033; this is very low, suggesting the production process is relatively reliable." A contextual sentence is often worth a B mark.
统计学的意义在于理解数据。考试中，当要求解释一个概率、一个均值或一项比较时，你必须结合背景写出简短的结论。例如，“超过 3 件产品有缺陷的概率是 0.033；这一概率很低，表明生产过程较为可靠。”结合背景的句子往往值一个 B 分。
In comparing two data sets using mean and standard deviation, go beyond just quoting the numbers. State which group has the higher average and which is more variable. Link your findings to the real-world scenario given in the question. A purely numerical description may not gain the full interpretation marks.
在利用均值和标准差比较两组数据时，不要仅仅列出数字。要说明哪一组的平均值更高、哪一组变异性更大。将你的发现与题目给出的实际情景联系起来。纯粹的数字描述可能拿不到全部的解读分。
11. Common Pitfalls to Avoid | 常见误区
One common mistake is confusing P(A∪B) with P(A∩B). The formula P(A∪B) = P(A) + P(B) − P(A∩ her is using nCr instead of nPr when arrangements matter. Always double-check the wording: 'chosen in a particular order' or 'arranged in a line' implies permutations.
一个常见错误是混淆 P(A∪B) 与 P(A∩B)。公式 P(A∪B) = P(A) + P(B) − P(A∩B) 至关重要；忘记减去交集就会导致重复计数。另一个错误是在顺序重要时使用 nCr 而非 nPr。务必反复检查题意：“按特定顺序选中”或“排成一列”意味着排列。
In grouped frequency calculations, use the midpoints for the mean but remember that the standard deviation formula requires the squared midpoints. Using class boundaries instead of midpoints will produce an inaccurate answer and lose accuracy marks. Also, when reading normal distribution tables, read the column and row headings carefully to avoid an offset error.
在分组频率计算中，使用组中值计算均值，但要记住标准差公式需要组中值的平方。用组限代替组中值会产生不准确的答案并丢失准确性分。此外，查阅正态分布表时，要仔细看清行列标题，避免发生偏移错误。
12. Time Management and Exam Strategy | 时间管理与考试策略
With 50 marks in 75 minutes, you have roughly 1.5 minutes per mark. Use this as a guide, but allow more time for the longer contextual questions at the end. Begin by scanning the whole paper; start with the question you feel most confident about to build momentum. Do not get stuck on a tough probability tree for more than 5 minutes.
75 分钟内完成 50 分的卷子，每分的作答时间大约为 1.5 分钟。以此作为参考，但要把更多时间留给最后较长的情境题。先浏览整份试卷，从你最有把握的题目开始，以建立信心。不要在一道很难的概率树状图上卡住超过 5 分钟。
If a question asks you to "show that" a particular value is obtained, set out your reasoning clearly, as the process is the core of the mark. Even if the final value is given, you must demonstrate the steps to earn the marks. When you are unable to complete a sub-part, move on, but write down any relevant formulas or distribution statements; you could still collect M and B marks.
若题目要求你“证明”某个值，要清晰地写出推理过程，因为过程才是得分的核心。即使最终值已经给出，你也必须展示步骤才能得到分数。当你无法完成某个小题时，先做后面的，但写下任何相关的公式或分布声明；你仍有可能拿到 M 和 B 分。
Published by TutorHao | Statistics 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