📚 Cambridge IGCSE Statistics: In-Depth Past Paper Analysis | 剑桥IGCSE统计学历年真题深度解析
Working through past papers is one of the most effective ways to prepare for the Cambridge IGCSE Statistics examination. By analyzing real questions from previous years, students can uncover recurring patterns, familiarise themselves with the exam format, and build the confidence needed to tackle both routine calculations and extended reasoning tasks. This article provides a detailed walkthrough of the key topics and question styles that have appeared in recent past papers, offering bilingual insights to help Year 10 learners master the syllabus and avoid common pitfalls.
刷历年真题是备考剑桥 IGCSE 统计学最有效的方法之一。通过分析以往的真题，学生可以发现反复出现的题型规律，熟悉考试形式，并建立起应对常规计算与拓展推理题所需的信心。本文细致梳理了近年来真题中出现的重点主题与提问方式，提供双语解析，帮助 Year 10 学生掌握考纲内容，避开常见的失分陷阱。
1. Exam Structure and Mark Allocation | 考试结构与分值分布
A typical Cambridge IGCSE Statistics paper consists of two components: Paper 1 (short-answer questions) and Paper 2 (long-answer questions). Past papers reveal that around 40% of marks are allocated to data handling and representation, 30% to probability and distributions, and the remainder to correlation, sampling, and inference. Understanding this weight helps you prioritise revision.
一份典型的剑桥 IGCSE 统计试卷包含两部分：卷 1（简答题）和卷 2（长答题）。历年真题显示，约 40% 的分数分配给数据处理与图表展示，30% 为概率与分布，其余为相关、抽样和推断。了解这一权重分配有助于你合理安排复习重点。
Examiners consistently test the same command words, such as ‘calculate’, ‘estimate’, ‘compare’ and ‘interpret’. Past papers demonstrate that simply obtaining a numerical answer is insufficient; you must often justify your steps or explain the significance of a result in context.
考官始终会使用相同的指令词，如 ‘calculate’ (计算)、’estimate’ (估计)、’compare’ (比较) 和 ‘interpret’ (解释)。真题表明，仅得出数字答案是不够的，你往往需要说明计算步骤，或结合具体情境解释结果的意义。
The mark schemes for past papers reward correct methods even if a final answer is inaccurate. This means you should always show clear working. Many common errors, such as misreading a cumulative frequency scale, cost numerous marks yearly, but method marks can still be salvaged.
真题的评分方案表明，即使最终答案有误，正确的解题方法仍可得分。因此务必展示清晰的解题过程。每年都有许多考生因读错累积频率图的刻度等常见错误而大量失分，但如果写出了正确方法，仍可拿到方法分。
2. Data Collection and Organization in Real Questions | 真题中的数据收集与整理
Past paper Section A questions frequently present raw data in tables or lists and ask you to construct grouped frequency tables. For instance, a 2019 paper required students to group the heights of 40 plants into equal intervals and decide on suitable class boundaries. Careless boundary choices (e.g., overlapping intervals) led to loss of accuracy marks.
在真题的 A 部分，常会以表格或列表形式给出原始数据，要求构建分组频率表。例如，2019 年的一道试题要求学生将 40 株植物的高度分成等距区间，并确定恰当的组界。草率的组界选择（如区间重叠）会导致准确度分数丢失。
Another classic task is to complete a two-way table from given totals. A 2021 question gave partial information about students’ favourite sports by gender; many candidates failed to check that all row and column sums matched the total. Practising these fill-in-the-blank tables sharpens your attention to detail.
另一经典题型是根据给出的总计完成双向表。2021 年的一道题给出了按性别划分的学生最喜爱运动的部分信息，许多考生未能检查所有行列合计是否与总人数吻合。多练习这类填表题可以提升你对细节的关注度。
Stem-and-leaf diagrams appear frequently in past papers because they test both ordering skills and the ability to identify key values. A 2020 paper asked candidates to draw an ordered stem-and-leaf plot and then find the median and range. Leaving the leaves unsorted was a common error that immediately lost the ordering mark.
茎叶图在真题中出现频率很高，因为它既考察排序能力，也考查识别关键值的能力。2020 年的一道题要求考生绘制有序茎叶图，然后找出中位数和极差。若叶片部分未排序，则直接丢掉排序分，这是常见错误。
3. Past Paper Breakdown: Measures of Central Tendency | 集中趋势量数真题详解
The mean, median and mode are tested in almost every paper. A 2018 question provided a frequency table of test scores and required the mean to be calculated. The correct formula is
Mean = Σfx ÷ Σf
, but many students mistakenly divided by the number of classes instead of total frequency. Always identify Σf before calculating the sum of fx.
平均数、中位数和众数几乎每套真题都会出现。2018 年的一道题给出了一份测验分数的频率表，要求计算平均数。正确公式为
平均数 = Σfx ÷ Σf
，但许多学生错误地除以组数而非总频数。务必在计算 fx 之和前先确定 Σf。
Questions on the median from a grouped table often include an extra step: finding the class interval containing the median using cumulative frequencies. A 2019 paper required interpolation inside the median class using the formula
Median ≈ L + (n/2 − F) × w ÷ fₘ
. Misplacing n/2 or forgetting to subtract the cumulative frequency before the class caused numerous miscalculations.
涉及分组表中位数的题目通常会多一步：利用累积频率找到中位数所在的组距。2019 年的一道题要求用公式在中位数区间内插值：
中位数 ≈ L + (n/2 − F) × w ÷ fₘ
。如果弄错 n/2，或者忘记减去该组之前的累积频数，会导致大量计算错误。
In problems where extreme values exist, the median is described as a better measure of central tendency than the mean. A typical exam question from 2022 asked, ‘Explain why the median might be a more appropriate measure here.’ The expected answer referred to skewness or outliers — simply stating ‘because the data is large’ earned no marks.
当数据存在极端值时，真题中会要求解释为什么中位数比平均数更适合作为集中趋势的度量。2022 年的一道典型题目问：“为什么在这里中位数可能更合适？” 期望答案要提及偏态分布或异常值——只写“因为数据很大”是不得分的。
4. Measures of Dispersion and Their Exam Applications | 离散程度量数及真题应用
Range, interquartile range (IQR) and standard deviation are the main dispersion measures tested. A 2017 question asked for the IQR from a cumulative frequency graph. Candidates had to read Q₁ at 25% and Q₃ at 75% of total frequency, then subtract. Many misread the graph’s horizontal axis scale, especially when each small square represented 2 units rather than 1.
极差、四分位距 (IQR) 和标准差是常考的离散程度量数。2017 年的一道题要求根据累积频率图求 IQR。考生需要在总频数的 25% 处读取 Q₁，75% 处读取 Q₃，然后相减。很多人读错了横轴刻度，尤其当每小格代表 2 个单位而非 1 时，读图错误屡见不鲜。
Standard deviation calculations from a frequency table appear in many Paper 2 questions. The formula
σ = √(Σfx²/Σf − (Σfx/Σf)²)
is often evaluated using a table with columns for x, f, fx, and fx². A 2020 past paper showed that many students omitted the square root step or incorrectly squared the fx column, revealing a lack of systematic approach.
许多卷 2 真题要求根据频率表计算标准差。公式为
σ = √(Σfx²/Σf − (Σfx/Σf)²)
。解题时通常借助包含 x、f、fx 和 fx² 的计算表。2020 年真题显示，不少学生忘记开方，或者直接错误地将 fx 列平方，暴露了缺乏系统性步骤的问题。
When asked to compare two datasets using the mean and standard deviation, a good exam answer always states both the measure of center and the measure of spread. Writing ‘Data A has a higher mean and is more spread out because its standard deviation is larger’ gains full marks, whereas a vague comparison like ‘Data A is bigger’ does not.
当题目要求使用平均数和标准差比较两组数据时，一份优秀的作答应同时说明集中趋势与离散程度。写出“数据 A 的平均数更高，且因为标准差更大，分布更分散”能拿到满分，而“数据 A 更大”这种模糊比较是无法得分的。
5. Probability: Step-by-step Progression in Past Papers | 概率题目的阶梯式难度
Probability questions typically start with simple enumeration from a sample space diagram. A 2019 question used a two-way table to present outcomes of rolling two dice; candidates had to find P(sum = 7). The solution required correctly identifying 6 favourable outcomes out of 36, reducing to 1/6. Overlooking the complete sample space led to invalid fractions.
概率题通常从样本空间图的简单计数开始。2019 年的一道题用双向表呈现掷两颗骰子的结果，要求计算 P(和为 7)。解题需要正确找出 36 种结果中的 6 种有利结果，化简为 1/6。忽略完整的样本空间会得出错误的分数。
Tree diagrams are indispensable for conditional probability questions. In a 2021 past paper, a bag contained 5 red and 3 blue counters; two were drawn without replacement. The tree required accurate second-branch probabilities (e.g., 4/7 and 3/7). A large number of students mistakenly kept the original 3/8 after one blue was removed, losing all marks for the ‘and’ rule calculation.
树状图是解决条件概率题不可或缺的工具。2021 年的一道真题中，袋子里有 5 个红色和 3 个蓝色筹码，抽取两次不放回。树状图的分支概率需要准确更新（如第二次为 4/7 与 3/7）。大量学生错误地在抽出一个蓝球后仍使用 3/8，导致依据“与”法则计算的分值全部丢失。
Venn diagram questions in past papers often combine set notation and probability. A 2022 question asked for P(A∪B) given the Venn diagram values. The correct approach was to sum all regions inside A or B and divide by the total. Mixing up intersection and union symbols was a notable error, as was forgetting to include the region outside both sets in the total.
真题中的韦恩图题常结合集合符号与概率。2022 年的一道题要求根据韦恩图求 P(A∪B)。正确的做法是将 A 或 B 内所有区域的值求和，再除以总数。混淆交集符号∪与并集符号，或者忘记在总数中包含两集以外的区域，都是显著错误。
6. Cumulative Frequency Graphs and Percentiles | 累积频率图与百分位数
Drawing an accurate cumulative frequency graph from a grouped frequency table is a perennial past-paper task. Marks are awarded for correct plotting of upper class boundaries against cumulative frequency and for a smooth curve. A 2018 exam question revealed that students often plotted midpoints instead of upper boundaries, distorting the entire graph.
根据分组频率表绘制准确的累积频率图是历年真题的常考任务。正确标出上组界对应的累积频率点，并用平滑曲线连接即可得分。2018 年的一道考题显示，许多考生误用了组中点而非上组界来描点，导致整个图形变形。
Estimating the median and quartiles from the graph involves drawing horizontal lines from the appropriate cumulative frequency positions. If total frequency is 200, the median is at 100, Q₁ at 50, and Q₃ at 150. Past papers emphasize that you must show construction lines on the graph, otherwise the reading marks are not awarded.
从图中估算中位数和四分位数需要从对应的累积频率位置画水平线。如果总频数为 200，中位数在 100 处，Q₁ 在 50 处，Q₃ 在 150 处。历年真题强调，必须在图上画出辅助线，否则读数分不予给出。
Percentile problems ask, for example, ‘Find the 90th percentile weight’. In a 2020 paper, candidates correctly found the position as 90% of total frequency but then failed to trace back to the horizontal axis using the curve. A common mistake was to report the cumulative frequency value as the percentile instead of the measurement on the x-axis.
百分位数问题比如“求第 90 百分位数的体重”。在 2020 年真题中，考生正确得出位置为总频数的 90%，但未能接着从曲线追溯回横轴。一个常见错误是将累积频率数值直接当作百分位数，而不是读取 x 轴上的测量值。
7. Scatter Diagrams, Correlation and Regression Analysis | 散点图与相关回归
Past papers often provide a scatter diagram and ask you to describe the correlation. The response must reference both direction and strength, e.g., ‘strong positive correlation’. Simply writing ‘positive’ is insufficient. A 2019 marking scheme also awarded a mark for identifying an outlier if one existed.
真题经常给出散点图，要求描述相关关系。答案必须同时提及方向与强度，如“强正相关”。只写“正相关”是不够的。2019 年的评分方案还指出，若存在异常点，识别出来也可得分。
Drawing the line of best fit by eye should pass through the ‘balance point’ and have roughly equal numbers of points above and below. A 2021 question penalised candidates whose line extended beyond the data range, as the exam required using it for interpolation, not extrapolation. Examiners look for a line that reflects the overall trend, not necessarily connecting any particular points.
肉眼绘制最佳拟合线时，应经过“平衡点”，并使线上下的点数大致相等。2021 年的一道题对超出数据范围的线条进行了扣分，因为题目要求内插而非外推使用。考官看重的是反映整体趋势的线，不一定非要通过某个特定点。
Calculating the equation of the regression line from given summaries often involves the formula
y = a + bx, b = Σ(x − x̄)(y − ȳ) ÷ Σ(x − x̄)²
. A 2020 past paper gave Σx, Σy, Σxy, and n, and asked for the line. Many students confused the numerator with Σxy − (Σx Σy)/n, which is correct, but miscomputed the denominator. Always write down the full expression and check your arithmetic.
从给定的汇总量计算回归线方程通常用到公式
y = a + bx, b = Σ(x − x̄)(y − ȳ) ÷ Σ(x − x̄)²
。2020 年真题提供了 Σx、Σy、Σxy 和 n，要求求回归线。许多学生虽然用了正确的分子 Σxy − (Σx Σy)/n，但算错了分母。务必写下完整表达式并检查计算。
8. Sampling Techniques in Past-Paper Contexts | 真题中的抽样方法
Questions on sampling methods test your ability to identify whether a scenario describes a simple random sample, stratified sample, systematic sample, or quota sample. In a 2019 paper, the description ‘every 10th person entering a shop is surveyed’ required the answer ‘systematic sampling’. Answers missed the mark when students wrote ‘random’ without specification.
关于抽样方法的题目考察识别简单随机抽样、分层抽样、系统抽样或配额抽样的能力。2019 年的一道题中，“每第 10 位进入商店的人接受调查”应回答“系统抽样”。如果只模糊地写“随机”而不说明具体类型，则不得分。
Stratified sampling problems often require you to calculate how many individuals to select from each group in proportion to the population. A 2021 paper gave a school with 400 boys and 600 girls, and a sample size of 80. The correct numbers were 32 boys and 48 girls. Where errors occurred, candidates forgot to multiply the total sample size by the stratum proportion.
分层抽样问题经常要求按比例计算从每一层中应抽取的个体数。2021 年的一道题给出某校男生 400 人、女生 600 人，样本量为 80。正确的人数为男生 32、女生 48。出错之处往往在于忘记用总样本量乘以该层比例。
A higher-order question might ask you to explain why a particular sampling method is appropriate. For instance, when different year groups have varied opinions, stratified sampling ensures representation. The expected answer must link the method’s advantage to the context — saying ‘it is fair’ is too vague.
高阶题目可能要求解释为什么某种抽样方法合适。例如，当不同年级的意见存在差异时，分层抽样能保证代表性。期望答案必须将方法的优势与题目情境联系起来，只说“它公平”过于笼统。
9. Common Mistakes and How to Avoid Them | 常见易错点与规避策略
One recurring mistake is confusing the formula for the sample standard deviation with the population standard deviation. Past papers often ask for the standard deviation of a dataset, but if the data is a sample, the denominator should be n−1. Analysing 2022 mark schemes shows that simply writing the larger formula without justification loses no marks if the correct version is used, but misapplication causes errors.
一个反复出现的错误是混淆样本标准差公式与总体标准差公式。真题常要求计算数据集的标准差，如果数据是样本，分母应为 n−1。分析 2022 年评分方案可知，只要使用了正确版本，不写理由不会丢分，但用错公式则必然导致错误。
Another frequent blunder involves misinterpreting ‘at least’ and ‘at most’ in probability. For example, ‘at least one head’ when tossing three coins means 1, 2, or 3 heads. Many students only considered exactly one head, ignoring the tail. Drawing a sample space or using 1 − P(no heads) helps avoid such oversights.
另一常见错误是误读概率中的“至少”与“至多”。例如抛三枚硬币时“至少一次正面”表示 1 次、2 次或 3 次正面。许多学生只考虑了恰好一次正面，忽略了其余情况。画出样本空间或使用 1 − P(无正面) 可以有效避免此类遗漏。
A subtle trap appears when reading scales on graphs. Past papers indicate that candidates frequently misread small divisions. If 10 small squares represent 2 units, each small square is 0.2 units. Taking a reading of 6.3 instead of 6.08 results from assuming a linear 1-to-1 scale. Always verify what one small division equals before plotting or reading.
读图时刻度是一个微妙的陷阱。真题表明，考生常常误读小格子。如果 10 小格代表 2 个单位，每小格就是 0.2 个单位。错误地假定 1 小格为 1 单位，会读到 6.3 而非 6.08。在描点或读数前，务必核实每小格代表的量值。
10. Effective Study Strategies Using Past Papers | 高效备考与真题活用策略
Start your revision by attempting a full past paper under timed conditions, then use the mark scheme to identify weak areas. Keep a log of errors: were they due to calculation, misreading, or concept gaps? For instance, if you consistently lose marks on cumulative frequency curve drawing, allocate two focused sessions to that topic.
开始复习时，先在计时条件下完成一套完整的真题，然后用评分方案找出薄弱环节。做好错误记录：是计算失误、审题不清还是概念欠缺？比如，如果累积频率曲线绘制总是丢分，就安排两个聚焦环节专门练习这一主题。
When you review a past paper, rewrite model answers in your own words, paying attention to the phrases the examiner expects, such as ‘positive skew because mean > median’. Over time, this builds a bank of precise statistical language that directly matches mark scheme terminology.
当回顾真题时，用自己的话重写高分范本，并留意考官期望的措辞，如“正偏态，因为平均数 > 中位数”。日积月累，这能帮助你建立一套与评分方案术语精准匹配的统计语言库。
In the final weeks before the exam, concentrate on the most frequently tested topics: cumulative frequency, tree diagrams, grouped mean and standard deviation, and scatter diagrams with a line of best fit. The past papers strongly indicate that mastering these four areas can secure a high proportion of the available marks.
考前最后几周，集中精力在最高频考点：累积频率、树状图、分组平均数和标准差，以及散点图与最佳拟合线。真题强烈表明，攻克这四个板块就能锁定卷面中很大比例的分数。
Published by TutorHao | Statistics 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