📚 AS Mathematics Unit 5 Mark Scheme Jun19 Masterclass | AS 数学 Unit5 Jun19 评分方案知识点精讲
The June 2019 AS Mathematics Unit 5 mark scheme provides a clear window into the examiners’ expectations for statistical reasoning, probability calculations and data handling. This article distils the key topic areas, common pitfalls and mark-winning strategies that appeared in that session. By studying the mark scheme together with these focused explanations, you will sharpen your technique and improve your chances of scoring full marks in similar questions.
2019年6月 AS 数学第五单元评分方案清晰展示了考官对统计推理、概率计算及数据处理能力的考核重点。本文提炼了该次考试涉及的核心知识点、常见错误与得分技巧。将评分方案与这些精讲内容对照学习，能够帮助你打磨解题方法，在同类题目中力争满分。
1. Probability Trees and Conditional Probability | 概率树与条件概率
A probability tree diagram must be drawn with care: every branch at a given node should represent mutually exclusive outcomes whose probabilities sum to 1.
概率树图的绘制务必细心：同一节点上的每条分支代表互斥的结果，其概率之和必须等于 1。
When two events A and B are involved, conditional probability is often tested. Use the formula P(A|B) = P(A ∩ B) / P(B) directly, but only after you have identified the correct intersection from the tree.
当题目涉及两个事件 A 与 B 时，条件概率是常见考点。先根据树图准确找出交事件，再直接使用公式 P(A|B) = P(A ∩ B) / P(B)。
The mark scheme rewards clear labelling of probabilities and the use of fractions rather than rounded decimals in intermediate steps; premature rounding can lead to the final answer being outside the allowed tolerance.
评分方案提倡清晰地标注概率，并在中间步骤中使用分数而非四舍五入后的小数；过早舍入容易导致最终答案超出容差范围。
For independent events: P(A ∩ B) = P(A) × P(B)
独立事件时：P(A ∩ B) = P(A) × P(B)
2. Discrete Random Variables and Probability Distributions | 离散随机变量与概率分布
A discrete random variable X takes a finite set of values x₁, x₂, …, xₙ. The first check is that Σ P(X = x) = 1; any missing probability can be found by subtraction.
离散随机变量 X 取有限个值 x₁, x₂, …, xₙ。首要检验是全部概率之和 Σ P(X = x) = 1；任何缺失的概率都可通过减法求得。
The expected value E(X) = Σ x P(X = x) is a measure of central tendency. The mark scheme often awards method marks for showing the correct sum of products, even if arithmetic slips later.
期望值 E(X) = Σ x P(X = x) 衡量数据的集中趋势。评分方案通常会给正确的乘积和运算过程步骤分，即使后续算术出现小错。
Var(X) = E(X²) – [E(X)]²
方差 Var(X) = E(X²) – [E(X)]²
Many candidates lose marks by forgetting to square the expected value at the end. A good habit is to write E(X²) and [E(X)]² separately before subtracting.
很多考生因最后忘记减去期望值的平方而丢分。好习惯是先把 E(X²) 与 [E(X)]² 分别列出，再相减。
3. Binomial Distribution | 二项分布
The binomial model X ~ B(n, p) applies when there are a fixed number n of independent trials, each with the same probability p of success. The mark scheme expects you to check these conditions before applying the formula.
当存在固定次数 n 的独立试验、且每次成功概率 p 不变时，可使用二项分布模型 X ~ B(n, p)。评分方案期望你在用公式前先明确这些适用条件。
For a specific number of successes, use P(X = k) = ⁿCₖ pᵏ (1 – p)ⁿ⁻ᵏ. Cumulative probabilities are often read from tables, but always show the working that links the required probability to a table value.
计算特定成功次数时使用 P(X = k) = ⁿCₖ pᵏ (1 – p)ⁿ⁻ᵏ。累积概率通常查表，但务必写出如何将题目所求概率转化为查表值的推导过程。
P(X ≥ k) = 1 – P(X ≤ k – 1)
P(X ≥ k) = 1 – P(X ≤ k – 1)
A common mistake is to use the wrong tail, e.g. calculating P(X ≤ k) when P(X < k) is required. Read the inequality carefully.
常见错误是用错尾部，例如题目要求 P(X < k) 却计算了 P(X ≤ k)。务必仔细审读不等号方向。
4. The Normal Distribution | 正态分布
If X ~ N(μ, σ²), the standardised value is z = (x – μ) / σ. This converts any normal problem into a search on the standard normal table Φ(z).
若 X ~ N(μ, σ²)，标准化值为 z = (x – μ) / σ。这一步将任何正态问题转化为对标准正态表 Φ(z) 的查找。
When a question gives a probability and asks for an unknown mean or standard deviation, you need to work backwards: set up Φ⁻¹(p) = (x – μ)/σ and solve.
当题目给出概率值并要求反求未知的均值或标准差时，需要逆向操作：建立方程 Φ⁻¹(p) = (x – μ)/σ 并求解。
The mark scheme is strict about continuity corrections. They are not required for normal approximation to binomial in this AS unit, but do be careful if the question specifically mentions “use a suitable approximation”.
评分方案对连续性校正要求严格。本 AS 单元中正态近似二项时通常不涉及连续性校正，但若题目明确要求“使用适当近似”，则要小心处理。
Always sketch a bell curve and shade the required region; this simple visual can prevent many sign errors.
始终快速画一个钟形曲线并标出阴影区域；这个简单图示能避免大量符号错误。
5. Cumulative Frequency Diagrams and Box Plots | 累积频率图与箱线图
A cumulative frequency curve is plotted using the upper class boundaries. From the graph, you can estimate the median, lower quartile Q₁ and upper quartile Q₃.
累积频率曲线使用组上限绘制。通过此图可估算中位数、下四分位数 Q₁ 与上四分位数 Q₃。
The interquartile range IQR = Q₃ – Q₁ is used to identify outliers. Boundaries for mild outliers are Q₁ – 1.5 × IQR and Q₃ + 1.5 × IQR.
四分位距 IQR = Q₃ – Q₁ 用于识别异常值；轻度异常值的界限为 Q₁ – 1.5 × IQR 与 Q₃ + 1.5 × IQR。
When comparing box plots, refer to both location (median) and spread (IQR, range). The mark scheme often credits comparative phrases like “higher median” or “less spread”.
比较箱线图时需同时论述位置（中位数）和离散程度（IQR、全距）。评分方案常给予“中位数更大”或“分布更集中”等比较性表述的加分。
6. Histograms and Frequency Density | 直方图与频率密度
Histograms are used for grouped continuous data. The vertical axis is frequency density, not frequency: Frequency density = frequency ÷ class width.
直方图用于分组连续数据。纵轴代表频率密度而非频率：频率密度 = 频率 ÷ 组距。
To estimate the mean from a histogram, use the midpoint of each class and multiply by the frequency. The formula Σ(f × mid) / Σf appears frequently in mark schemes.
由直方图估计平均值时，使用每组组中值乘以频率。公式 Σ(f × 组中值) / Σf 在评分方案中反复出现。
Check whether the question asks for an exact width or proportional area; a common error is drawing the bars with incorrect density, causing the shape to misrepresent the data.
注意题目要求的是精确宽度还是按比例面积；常见错误是用错误的密度画条形，导致图形歪曲数据的实际分布。
7. Measures of Central Tendency and Spread | 集中趋势与离散度量
The mean, median and mode each summarise the centre of a data set in different ways. The mean is sensitive to outliers, while the median is resistant.
平均值、中位数和众数各以不同方式概括数据的中心。平均值对异常值敏感，中位数则较为稳健。
Variance is given by s² = Σ(x – x̄)² / (n – 1) for a sample, or σ² = Σ(x – μ)² / N for a population. The mark scheme penalises confusion between the two forms.
样本方差为 s² = Σ(x – x̄)²/(n – 1)，总体方差为 σ² = Σ(x – μ)²/N。评分方案对混淆这两种形式会扣分。
Coding of data (e.g. y = ax + b) changes the mean and standard deviation in predictable ways: ȳ = a x̄ + b, sᵧ = |a| sₓ. Use this to simplify calculations with large numbers.
数据编码（如 y = ax + b）会按一定规律改变均值与标准差：ȳ = a x̄ + b，sᵧ = |a| sₓ。利用编码可简化大数运算。
8. Scatter Graphs and Correlation | 散点图与相关性
A scatter graph reveals the relationship between two variables. Descriptions such as “strong positive correlation” or “weak negative correlation” must match the visual pattern.
散点图揭示两个变量间的关系。对相关性的描述如“强正相关”或“弱负相关”必须与图形形态一致。
The product moment correlation coefficient r ranges from –1 to +1. Values close to ±1 indicate strong linear correlation. In the mark scheme, you are usually given a formula book value, but you must interpret r in context.
积矩相关系数 r 的取值范围是 –1 至 +1。接近 ±1 表明强线性相关。评分方案中通常给出公式手册中的值，但必须结合情境解释 r 的含义。
r = Sxy / √(Sxx Syy)
r = Sxy / √(Sxx × Syy)
Never confuse correlation with causation — a frequent source of lost marks in interpretation questions.
切勿将相关关系与因果关系混为一谈——这是解释题中常见的丢分点。
9. Linear Regression | 线性回归
The least squares regression line of y on x is y = a + bx, where b = Sxy / Sxx and a = ȳ – b x̄. The mark scheme expects you to state the equation clearly after calculating these coefficients.
y 对 x 的最小二乘回归直线为 y = a + bx，其中 b = Sxy / Sxx，a = ȳ – b x̄。评分方案期望你计算出系数后清晰地写出回归方程。
Use the regression line for prediction only within the range of the data. Extrapolating beyond the observed x‑values is unreliable and may be penalised.
回归直线只适用于数据范围内的预测。超出观测 x 值范围的外推不可靠，可能被扣分。
If the question involves a change of units, remember that both a and b are affected; recalculating from coded sums is safer than converting the final line.
若题目涉及单位变化，a 和 b 都会受影响；从编码数据重新计算比直接转换最终直线更安全。
10. Hypothesis Testing for a Binomial Proportion | 二项比例假设检验
A hypothesis test begins by stating the null hypothesis H₀: p = p₀ and the alternative H₁: p < p₀, p > p₀ or p ≠ p₀, depending on the wording of the question.
假设检验的第一步是根据题意写出原假设 H₀: p = p₀ 与备择假设 H₁: p < p₀、p > p₀ 或 p ≠ p₀。
Assuming H₀, calculate the probability of obtaining the observed result or something more extreme. For a one‑tailed test, this is a single tail; for two‑tailed, double the appropriate tail probability.
在原假设成立的前提下，计算得到观测结果或更极端情况的概率。单尾检验只取一侧尾部概率；双尾检验则需将相应尾概率加倍。
Compare the calculated probability with the significance level, usually 5% or 1%. If p‑value < significance level, reject H₀. Always give a conclusion in the context of the problem and use non‑assertive language.
将计算所得概率与显著性水平（通常 5% 或 1%）比较。若 p 值 < 显著性水平，则拒绝 H₀。务必结合题目背景给出结论，并使用非断定性的措辞。
P(X ≥ observed | H₀) < 0.05 → reject H₀
P(X ≥ 观测值 | H₀) < 0.05 → 拒绝 H₀
Published by TutorHao | AS 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