📚 IGCSE WJEC Statistics: In-Depth Analysis of Past Papers | IGCSE WJEC 统计：历年真题深度解析
This revision guide examines the WJEC IGCSE Statistics past papers in depth. It identifies the most common question types, the recurring pitfalls that quietly drain marks, and the high-impact strategies that help candidates move from a correct calculation to a full-mark written conclusion.
本复习指南深入解析 WJEC IGCSE 统计历年真题。它梳理了最常见的题型、反复出现却在不知不觉中吞掉分数的失分点，以及帮助考生从正确计算走向满分文字结论的高效策略。
1. Why In-Depth Past Paper Analysis Works | 为什么真题深度解析有效
Past papers are the syllabus in disguise. WJEC examiners reuse a small number of question archetypes across series, so a candidate who has genuinely analysed ten papers has effectively seen the assessment before it happens. The goal is not to memorise answers but to recognise patterns.
真题其实是换了外衣的考纲。WJEC 考官在不同考季反复使用少数几种题目原型，因此真正分析过十套真题的考生，等于提前见过考题。目标不是背答案，而是识别模式。
A three-pass method works best. First sit a paper untimed and unaided. Then mark it strictly against the official mark scheme, writing the examiner's own wording next to every lost mark. Finally, rewrite only the weak answers a week later and compare the two versions side by side.
最有效的做法是三轮法。第一轮不限时、不查资料地完整做一遍。第二轮对照官方评分标准严格批改，把考官的原话抄在每一处失分旁边。第三轮在一周后只重写薄弱题目，并把前后两版答案并排对比。
Keep a log with three columns: topic, marks lost, and the reason for the loss. Most candidates discover that losses cluster around three or four repeating causes rather than spreading evenly, which makes the final weeks of revision far more targeted.
建议做一个三栏记录表：考点、失分、失分原因。多数考生会发现失分集中在三四个重复原因上，而不是均匀分布，这样最后几周的复习就能精准得多。
2. Past Paper Structure and Mark Allocation | 真题结构与分值分布
WJEC IGCSE Statistics past papers usually combine short data-handling questions with longer scenario-based problems. Marks are typically split between statistical calculations, diagram interpretation and written conclusions, so numerical accuracy alone is not enough.
WJEC IGCSE 统计历年真题通常把短数据处理题和较长情境题结合起来。分值一般分布在统计计算、图表解读和文字结论之间，因此仅有数值准确是不够的。
| Question type | 题型 | Typical marks | 常见分值 | What earns the marks | 得分要点 |
|---|---|---|
| Short data handling | 1 - 4 | Correct value, correct unit, rounding as requested |
| Table or diagram interpretation | 2 - 5 | Accurate reading, comparing two groups, quoting figures |
| Statistical enquiry / scenario | 6 - 9 | Method, calculation, conclusion linked to context |
| Probability and diagrams | 3 - 6 | Correct method, simplified fraction or decimal, clear notation |
Command words such as 'compare', 'justify' and 'evaluate' appear frequently. Examiners reward clear comparative language, relevant units and an explicit reference to the context, rather than just a final number.
指令词如 'compare'、'justify' 和 'evaluate' 出现频率很高。阅卷人看重清晰的比较性语言、相关单位以及对情境的明确引用，而不只是一个最终数值。
Timing should mirror the tariff. A useful rule is roughly one minute per mark plus ten minutes of checking, so a 4-mark interpretation question deserves four minutes of thinking and writing, not forty seconds of guessing.
时间分配应与分值匹配。一条实用的规则是大约每分一分钟，再加十分钟检查，因此一道 4 分的解读题值得花四分钟思考与书写，而不是四十秒草率作答。
3. Assessment Objectives and Mark Scheme Logic | 考核目标与评分标准逻辑
WJEC IGCSE Statistics questions are written against three assessment objectives. AO1 covers recalling and using statistical techniques. AO2 covers selecting the right technique, applying it in context and interpreting results. AO3 covers analysing, evaluating and communicating findings clearly.
WJEC IGCSE 统计的题目围绕三项考核目标设计。AO1 考查对统计方法的记忆与运用；AO2 考查选择正确方法、在情境中应用并解读结果；AO3 考查分析、评价并清晰表达结论。
Mark schemes use short codes that tell you exactly where credit sits. M marks are for a valid method, A marks are for an accurate answer, B marks are for a stated fact or value, and ft means follow through, so an early arithmetic slip does not necessarily cost the later method marks.
评分标准使用简短代号，明确指出分数落在哪里。M 分给正确方法，A 分给准确答案，B 分给陈述性事实或数值，ft 表示"后续跟分"，因此前面算错一步并不必然丢掉后面的方法分。
Codes such as 'oe' (or equivalent) and 'cao' (correct answer only) also matter. 'Oe' signals that alternative wording or forms are accepted, while 'cao' means the final value must be exact with no follow-through.
'oe'（或等价表达）与 'cao'（仅正确答案）这类代号同样重要。'Oe' 表示接受其他说法或形式，而 'cao' 意味着最终数值必须完全正确，不提供后续跟分。
Always finish a calculation with the unit and, where relevant, a sentence of interpretation. A bare number on a 3-mark enquiry question typically earns only the M and A marks, leaving the communication mark unclaimed.
计算完成后一定要写单位，并在需要时补上一句解读。一道 3 分的情境探究题，只写一个光秃秃的数字通常只能拿到 M 分和 A 分，表达分白白丢掉。
4. Command Words Decoded | 指令词解码
Command words determine the shape of a full-mark answer. Many candidates lose marks not because they cannot do the statistics, but because they answer a different question from the one printed on the paper.
指令词决定满分答案的形状。很多考生丢分并不是因为不会统计，而是因为回答的不是试卷上印的那道题。
| Command word | 指令词 | What it demands | 作答要求 |
|---|---|
| Calculate / Work out | Show method, give value with unit |
| Compare | Two groups, two figures, explicit comparative wording |
| Describe | State what the data or diagram shows, no explanation needed |
| Explain | Give a statistical cause or reason, using 'because' |
| Justify | Support a decision with a calculated figure or clear evidence |
| Evaluate | Weigh strengths and weaknesses, then reach a judgement |
| State / Write down | No working required, but the value must be clear |
| Suggest | A sensible contextual reason, not a calculation |
A reliable habit is to underline the command word and the number of marks before starting. If the question says 'compare' and carries 2 marks, the answer needs two distinct comparative statements, each referring to the context.
一个可靠的习惯是动笔前先划出指令词和分值。如果题目要求 'compare' 且为 2 分，答案就需要两条互不重复的比较性陈述，且都要联系情境。
5. Data Collection, Sampling and Critiquing Sources | 数据收集、抽样与数据来源评价
Sampling questions ask you to name a method, describe how to carry it out in the given context, and then state one advantage and one disadvantage. Naming the method alone rarely earns full marks.
抽样题要求你写出方法名称、结合给定情境说明操作步骤，然后指出一个优点和一个缺点。只写出方法名称很少能拿满分。
- Simple random sampling | 简单随机抽样: every member of the population has an equal chance.
- Systematic sampling | 系统抽样: choose every k-th item from an ordered list.
- Stratified sampling | 分层抽样: sample proportionally from each subgroup.
- Quota sampling | 配额抽样: fixed numbers from each group, chosen by the interviewer.
- Opportunity (convenience) sampling | 偶遇抽样: whoever is available, quick but biased.
The stratified sampling formula is worth memorising, because examiners often ask for the number taken from one stratum.
分层抽样的公式值得背下来，因为考官常要求计算某一层应抽取的人数。
Number from a stratum = (stratum size ÷ population size) × sample size
A census measures every member of the population: accurate but slow and expensive. A sample is faster and cheaper but introduces sampling error. Always link the choice back to the scenario, such as cost, time or the need for precision.
普查测量总体中的每一个个体：准确但耗时且昂贵。抽样更快更便宜，但会带来抽样误差。选择时一定要回到情境中说明理由，例如成本、时间或对精确度的要求。
When criticising a questionnaire, look for leading questions, overlapping response boxes, missing options, personal or sensitive questions placed early, and unclear units. These reappear in almost every series.
评价问卷时，重点找诱导性问题、选项区间重叠、缺少选项、把私人或敏感问题放在开头，以及单位不明确。这些几乎每个考季都会出现。
6. Averages, Spread and Comparing Distributions | 平均数、离散程度与分布比较
Averages and measures of spread form the backbone of the paper. Knowing how to calculate them is AO1; knowing which one best represents a data set, and why, is AO2 and carries the higher marks.
平均数与离散程度的度量是整份试卷的骨架。会算只是 AO1；知道哪一个最能代表该数据集以及原因，才是 AO2，也对应更高的分值。
÷ n Range = maximum − minimum IQR = Q₃ − Q₁
s = √( Σ(x − x̄)² ÷ (n − 1) )
ich is a favourite exam question because it combines a calculation with an interpretation.
离群值通常用 1.5 × IQR 规则来界定，这是考官偏爱的题型，因为它把计算与解读结合在一起。
Outlier if value < Q₁ − 1.5 × IQR or value > Q₃ + 1.5 × IQR
When comparing two distributions, always make two separate points: one about a typical value (mean or median) and one about consistency (range, IQR or standard deviation). Finish with a sentence that answers the actual question, such as which shop is better value.
比较两个分布时，一定要分成两点：一点讲典型值（平均数或中位数），一点讲一致性（极差、四分位距或标准差）。最后用一句话回答真正的提问，例如哪家店更划算。
Useful phrasing includes 'on average, group A has a higher value' and 'group B is more consistent because its interquartile range is smaller'. Both statements must quote figures drawn from the data.
可用的表述包括"平均而言，A 组的数值更高"以及"B 组更稳定，因为其四分位距更小"。两条陈述都必须引用数据中的具体数值。
7. Charts, Tables and Diagram Interpretation | 图表、表格与图示解读
Diagram questions reward precision. Reading a value off a cumulative frequency curve or a histogram demands attention to the scale, and examiners routinely award a mark for the correct reading and a further mark for its use.
图表题奖励精确。从累积频率曲线或直方图上读数时要注意刻度，考官通常会给"读对数值"一分，再给"正确使用"一分。
- Bar chart | 条形图: compares categories; bars must be equal width.
- Pie chart | 饼图: shows proportions; each sector is 360° × (frequency ÷ total).
- Histogram | 直方图: for continuous data; the vertical axis is frequency density.
- Cumulative frequency curve | 累积频率曲线: read quartiles, median and percentiles.
- Box plot | 箱线图: shows median, quartiles and extremes in one picture.
For a histogram, the key relationship is straightforward, and forgetting it is one of the most common errors in the whole paper.
对于直方图，关键关系很直接，而忘记它正是整份试卷中最常见的错误之一。
Frequency density = frequency ÷ class width
Pie chart questions often require the reverse step: given a sector angle, find the frequency, or given the frequency, find the angle.
饼图题常常需要反向计算：已知扇形角度求频数，或已知频数求角度。
Sector angle = (frequency ÷ total frequency) × 360°
When a question asks you to criticise a diagram, check the axis labels, whether the vertical axis starts at zero, whether the classes are equal width, and whether the chart is drawn to scale rather than designed to exaggerate.
当题目要求你评价一幅图时，检查坐标轴标签、纵轴是否从零开始、各组组距是否相等，以及图形是否按比例绘制而非刻意夸大。
8. Probability, Venn Diagrams and Tree Diagrams | 概率、韦恩图与树状图
Probability is one of the most predictable areas of the paper. The two rules that appear again and again are the addition rule for combined events and the multiplication rule for independent events.
概率是试卷中最可预测的部分之一。反复出现的两条规则是复合事件的加法规则和独立事件的乘法规则。
P(A ∪ B) = P(A) + P(B) − P(A ∩ B)
P(A ∩ B) = P(A) × P(B) for independent events
Expected frequency is a frequent final part of a probability question, and it is easy to secure provided you multiply by the correct total.
期望频数常作为概率题的最后一小问，只要乘以正确的总数就能稳稳拿分。
Expected frequency = P(event) × number of trials
Tree diagrams require care with the second set of branches. When the item is not replaced, the denominator and numerator both change, and examiners specifically test this condition.
树状图要特别小心第二层分支。当抽取后不放回时，分母与分子都会改变，而考官恰恰专门考查这一条件。
Venn diagram questions often give the intersection first and expect you to work outwards. A quick sketch with totals written in each region prevents the double counting that costs marks year after year.
韦恩图题通常先给出交集，然后要求你向外推算。随手画一张草图并在每个区域标出总数，能避免年年都在发生的重复计数错误。
9. Scatter Diagrams, Correlation and Lines of Best Fit | 散点图、相关与最佳拟合线
Scatter graph questions test three things: describing correlation, using the line of best fit, and understanding the limits of prediction.
散点图题考查三件事：描述相关性、使用最佳拟合线，以及理解预测的局限。
Describe correlation with two descriptors at once, such as 'strong positive correlation'. Then, if asked to use the line, read across carefully and state the predicted value with its unit.
描述相关性时要用两个描述词，例如"强正相关"。如果需要使用拟合线，要仔细横向读取，并带单位写出预测值。
Interpolation means predicting inside the range of the data and is generally reliable. Extrapolation means predicting outside it and is risky, so answers should mention that the trend may not continue.
内插指在数据范围内进行预测，通常可靠。外插指在数据范围之外预测，风险较大，因此答案中应提到趋势未必延续。
Correlation does not prove causation. A high correlation between two variables may be explained by a third factor, and a one-line comment about this earns marks in evaluation questions.
相关不等于因果。两个变量高度相关可能由第三个因素造成，在评价题中写一句相关说明就能得分。
Higher-tier questions sometimes ask for Spearman's rank correlation coefficient, which measures the strength of agreement between two rankings.
较高难度的题目有时会要求计算斯皮尔曼等级相关系数，它衡量两组排名之间的一致程度。
rₛ = 1 − 6Σd² ÷ (n(n² − 1))
10. Time Series, Index Numbers and Spreadsheets | 时间序列、指数与电子表格
Time series questions ask you to separate a trend from seasonal variation. Moving averages smooth out the seasonal peaks and dips so the underlying direction becomes visible.
时间序列题要求你把趋势与季节性波动分开。移动平均可以抹平季节性的高峰与低谷，让潜在走势显现出来。
For quarterly data, use a four-point moving average, centred so that each value aligns with a time period. For monthly data, use twelve points. Seasonal variation is then the difference between the actual value and the trend.
季度数据
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