CCEA GCSE · 分析与预测

2023 CCEA GCSE Statistics 2260 历年真题分析与预测

The 2023 CCEA GCSE Statistics Higher Tier assessment comprises two distinct two-hour papers (Unit 1 and Unit 2), totaling 200 marks. The examination provides a balanced coverage of the entire syllabus, heavily emphasizing statistical distributions (Normal and Binomial models, Standardised z-scores), quality control processes (warning and action lines on control charts), bivariate analysis (Spearman's rank correlation and PMCC), and practical data representation (cumulative frequency diagrams, box plots, stem-and-leaf, and frequency trees). Success relies on precision in multi-step statistical calculations, proper contextual interpretation of summary statistics, and sound understanding of sampling methodologies.

更新: 2026年8月31日

分析模拟试题

难度 3.2/5

总分

200

时间

240 分钟

最高频考点

Cumulative frequency, box plots, and comparative statistical distributions

分卷

Unit 1 (With Calculator) Higher Tier [GST12]: 100 分 · 120 分钟Unit 2 (With Calculator) Higher Tier [GST22]: 100 分 · 120 分钟

重点章节

Standardised scores (z-scores) · 24 分Control charts and quality assurance · 25 分Cumulative frequency diagrams and box plots · 31 分Interpreting lines of best fit and correlation coefficients · 19 分Comparing data sets and drawing conclusions · 16 分

Executive Overview & Difficulty Verdict

The 2023 CCEA GCSE Statistics Higher Tier examination represents a standard, fair, yet technically rigorous assessment across Unit 1 and Unit 2. With an overall difficulty index of 3.2 / 5.0 (3 Stars), the assessment rewards students who possess solid algebraic manipulation skills, a clear conceptual understanding of statistical modeling, and the ability to articulate contextual conclusions rather than offering rote definitions.

Where the Marks Are concentrated

Marks across both papers are anchored heavily in four major domains:

Statistical Distributions & Standardised Scores: Unit 1 Question 10 and Unit 2 Question 10 test standard normal distributions, empirical rule intervals (\(\mu \pm \sigma\), \(\mu \pm 2\sigma\)), and standardised \(z\)-scores (\(z = \frac{x - \mu}{\sigma}\)) to determine likelihoods. Additionally, Unit 2 Question 8 features the Binomial distribution expansion \((p+q)^n\).
Data Representation & Spread: Box plots, stem-and-leaf diagrams, frequency polygons, and cumulative frequency curves feature prominently (e.g., Unit 1 Q4, Unit 2 Q7), accounting for substantial marks allocated to calculating medians, quartiles, and IQR.
Quality Assurance & Control Charts: Both papers test process control (Unit 1 Q5, Unit 2 Q3), requiring candidates to plot sample means, draw warning/action limits (at target \(\pm 2\sigma / 3\sigma\) or specified tolerances), and give actionable operational advice.
Correlation & Bivariate Data: Spearman’s rank correlation coefficient (\(r_s = 1 - \frac{6\sum d^2}{n(n^2-1)}\)) in Unit 1 Q9 and Product Moment Correlation Coefficient (PMCC) in Unit 2 Q4 test computational mastery alongside nuanced conceptual justifications (e.g., why linear regression is unsuitable when correlation is near zero).

Examiner Pitfalls & Lost Marks

Examiner analysis indicates regular mark loss in the following areas:

Contextual Comparisons: When comparing distributions (e.g., Unit 1 Q4g, Unit 2 Q7f), students often state numbers without statistical interpretation. Full marks require two distinct comments: one comparing location/average (median or mean) with context, and one comparing dispersion (IQR or standard deviation).
Control Chart Decisions: Failing to specify consecutive sample rules versus single sample breaches outside warning and action lines.
Sampling Assumptions: In Petersen capture-recapture questions (Unit 1 Q8), students often provide vague statements rather than formal assumptions such as a closed population (no births/deaths/migration), equal catchability, and non-loss of identification marks.
Interpreting Index Numbers: In chain-base index number problems (Unit 2 Q9), candidates must clearly state the percentage change relative to the preceding year rather than the base period.

Exam Strategy & Revision Advice

Master calculator operations for single-variable and bivariate statistics (sample mean, standard deviation \(\sigma\), PMCC \(r\)). Allocate roughly 1.1 minutes per mark, preserving at least 10 minutes at the end of each paper to audit arithmetic in multi-step tables (such as ranking and \(d^2\) columns).

准备好测试自己了吗？

将这些笔记转化为考试练习。获取此课题的无限AI题目，即时批改及详细解析。

练习此课题

图表

章节分数分布

快速找出分数最集中在哪些考点，优先分配复习时间。

题型分布

比较不同题型的占分，了解试卷结构和答题比重。

Calculation & Arithmetic78 分 · 22 题
Data Presentation & Plotting28 分 · 9 题
Interpretation & Contextual Comparison52 分 · 17 题
Methodology, Sampling & Explanation42 分 · 14 题

分数难度分布

估算哪些分数属于基础分、进阶分或高难度分。

78%属较易或中等难度

较易: 68 分中等: 88 分较难: 44 分

指令词频率

留意常见指令词，避免答题方向错配。

技能比重

显示本届试卷主要考核的能力比例。

复习回报

气泡越大代表越常重现；越高代表占分越多，适合用来排复习优先顺序。

Control Charts (Action & Warning limits, process control interpretation)

25 分 · 难度 2.4 · 重现率 100%

Box Plots & Cumulative Frequency (Plotting, IQR, Comparisons)

31 分 · 难度 2.6 · 重现率 100%

Standardised z-Scores & Normal Distribution Likelihood

20 分 · 难度 3.4 · 重现率 95%

Spearman's Rank Correlation & PMCC Calculation/Interpretation

19 分 · 难度 2.8 · 重现率 90%

Probability Trees, Two-Way Tables & Venn Diagrams

21 分 · 难度 2.5 · 重现率 90%

Chain Base & Simple Index Numbers

8 分 · 难度 3 · 重现率 80%

Petersen Capture-Recapture & Sampling Methodology

12 分 · 难度 2.2 · 重现率 85%

时间与分数

比较每部分的分数和建议作答时间，帮助安排考场节奏。

分分钟

下一年趋势预测

整理下一年较值得关注的考点，以及背后的原因。

Time Series, Trend Lines and Moving Averages (with Seasonal Variation)92%

Moving averages and seasonal variation calculations were minimal in this series and are core Higher tier syllabus requirements.

Stratified and Systematic Sampling Calculations88%

The 2023 papers emphasized cluster sampling and qualitative sampling frames; numeric stratified sampling allocations are heavily expected next cycle.

Histograms with Unequal Class Widths (Frequency Density)85%

Only basic frequency polygons were examined in 2023 Unit 1; frequency density histogram drawing and estimation of median are prime Higher tier targets.

Geometric Distribution & Tree Diagrams with Conditional Probability80%

Binomial was examined in Unit 2; geometric distribution models and multi-stage dependent tree diagrams alternate regularly.

累积分数阶梯

曲线显示逐题累积的分数，虚线为各等级的估算分界。看曲线在哪一题附近越过你的目标等级，就知道要稳答到第几题、哪些题目决定升降级。

考官重点

AI 推断

常见失分点

When comparing two distributions from box plots or summary statistics, candidates frequently give isolated numerical values without using explicit comparative adjectives (e.g. stating 'Paper 1 median is 45 and Paper 2 is 20' rather than 'Visitors to the museum are on average older because they have a higher median age').
Confusing warning and action thresholds on quality control charts, or failing to state the required corrective action (e.g., taking an immediate re-sample when a single point is between warning and action lines versus stopping the machine if two consecutive points or a single point beyond the action line occurs).
In chain-base index numbers, stating percentage changes relative to the initial base year rather than relative to the immediately preceding year.
In standardised score questions comparing two populations, failing to take the absolute value or clarify that a z-score closer to 0 indicates a value closer to that distribution's mean, hence more likely.

常见误区

Believing that a simple arithmetic mean can be used when component weights differ (e.g., mock exam weighting), neglecting the weighted mean formula.
Assuming linear regression is always valid regardless of the correlation coefficient value, overlooking that regression lines are inappropriate when correlation is approximately zero.
Assuming that capture-recapture produces an exact count rather than an estimate subject to biological and operational assumptions.

失分原因

Loss of method marks on grouped standard deviation due to manual entry errors instead of using calculator statistical list modes.
Incomplete Venn diagram labeling where probability proportions fail to sum to 1.0 or region intersections are counted twice.
Omitting clear units in final responses (e.g., seconds, cm, £, hectares).

等级分数线估算

更新: 2026年8月31日

来源: CCEA

等级	分数	百分比
A*	—	90%
A	—	80%
B	—	70%
C*	—	65%
C	—	60%
D	—	50%
E	—	40%
F	—	30%
G	—	20%

Official grade boundaries published by CCEA.

知道考官想要什么，就去拿分

thinka 是一款 AI 练习应用，超过 100,000 名 DSE 学生用它刷题、即时发现错处。你已经知道高频题型和失分点，现在就用 thinka 针对性练习，把分析变成实际分数。

免费开始练习查看方案

知道考什么了？用 thinka AI 针对性练习，即时批改。

免费开始练习

等级	分数	百分比
A*	—	90%
A	—	80%
B	—	70%
C*	—	65%
C	—	60%
D	—	50%
E	—	40%
F	—	30%
G	—	20%

等级	分数	百分比
A*	—	90%
A	—	80%
B	—	70%
C*	—	65%
C	—	60%
D	—	50%
E	—	40%
F	—	30%
G	—	20%

等级	分数	百分比
A*	—	90%
A	—	80%
B	—	70%
C*	—	65%
C	—	60%
D	—	50%
E	—	40%
F	—	30%
G	—	20%