Year 11 Eduqas Statistics: In-depth Past Paper Analysis | 历年真题深度解析

📚 Year 11 Eduqas Statistics: In-depth Past Paper Analysis | 历年真题深度解析

Past papers are the most powerful revision tool for Eduqas GCSE Statistics. By analysing real exam questions, you can spot recurring themes, decode examiner expectations and practise applying statistical reasoning accurately. This article takes a deep dive into typical past-paper tasks across the entire specification, unpacking common question types, common mistakes and the step-by-step logic required to score full marks. Whether you are aiming for a Grade 5 or a Grade 9, you will learn how to approach each topic with the confidence that comes from truly understanding how the exam works.

历年真题是Eduqas GCSE统计学最有力的复习工具。通过分析真实考题，你可以发现反复出现的命题模式，理解考官的评分要求，并练习准确运用统计推理。本文将深度解读涵盖整个考纲的典型真题任务，剖析常见题型、常见错误以及获得满分的逐步逻辑。无论你的目标是5分还是9分，你都将学会如何以真正理解考试运作方式所带来的信心来应对每个主题。

1. Understanding the Exam Structure | 理解考试结构

The Eduqas GCSE Statistics assessment consists of two written papers, each worth 50 % of the final grade. Both papers cover the full specification, mixing straightforward data-handling tasks with extended problem-solving scenarios. Past papers reveal that questions are carefully scaffolded: early parts test basic recall and calculation, while later parts demand interpretation, evaluation and justification in context. Knowing this pattern allows you to manage your time – spend the first half securing marks on routine techniques before tackling the more demanding “explain” or “evaluate” style items.

Eduqas GCSE统计学考试由两份笔试组成，各占总成绩的50%。两份试卷均覆盖全部考纲内容，将直接的数据处理任务与拓展性问题解决场景相结合。历年真题显示，题目经过精心设计：前半部分考察基本记忆与计算，后半部分则要求在特定情境下进行解释、评估和论证。了解这一模式有助于你合理安排时间——先用前半段时间确保常规技术的分数，再攻克要求更高的“解释”或“评估”类题目。

One distinctive feature is the emphasis on mathematical reasoning rather than pure number crunching. A classic past-paper opening will ask you to “write down” a statistical term after reading a description, or to read a value directly from a diagram. These are quick wins. Later, you might be asked to compare two data sets and justify which one is more reliable – a skill that requires blending numerical evidence with real-world context. Train yourself to spot these question levels as you work through past papers so you never spend too long on a low-mark item.

一个显著特点是强调数学推理而非纯粹的数值计算。典型的真题开场要求你根据描述“写出”某个统计术语，或直接从图表中读取数值。这些都是能快速得分的环节。随后你可能会被要求比较两组数据，并论证哪组更可靠——这项技能需要将数字证据与现实情境相结合。在刷历年真题时，要训练自己识别题目层次，这样就不会在低分值题目上花费过多时间。

Marks allocated to parts of a question tell a story. In past papers, “calculate” steps often carry one or two marks, while “comment on” or “compare” steps can carry three or four. This signals that the examiner wants a structured, evidence-based answer, not a single word. When analysing mark schemes, notice that “both numbers stated” and “comparative statement” are separate marking points. Adopt this structure in your answers: state the relevant statistics, then make a comparison, and finally support it with a reason linked to the context.

题目各部分的分数分配包含重要信息。在历年真题中，“计算”步骤通常占1-2分，而“评论”或“比较”步骤可能占3-4分。这提示考官希望看到结构化、基于证据的答案，而非单个词语。分析评分标准时要留意，“说明两组数据”与“对比陈述”往往会是独立的得分点。在作答时采用这一结构：先写出相关统计量，再进行比较，最后结合情境给出理由加以支撑。

Time management is best practised by completing a full timed paper under exam conditions. Past papers from summer 2018 onwards reflect the current weighting of problem-solving – around 30 % of marks. Set a timer for 1 hour 45 minutes per paper. If you consistently struggle to finish, look at which sections you can accelerate: for example, rehearsing standard deviation computation until it becomes automatic saves precious minutes for the longer evaluative question at the end.

通过在全真考试条件下限时完成整套试卷是练习时间管理的最佳方法。2018年夏季及之后的历年真题反映了当前问题解决类题目约占总分30%的权重。为每份试卷设定1小时45分钟的计时器。如果你经常无法完成所有题目，查看可以加速的部分：例如，反复练习标准差计算直至条件反射，这能为末尾较长的评估类题目节省宝贵时间。

---

2. Data Collection and Sampling Methods | 数据收集与抽样方法

Past papers consistently test your ability to critique data collection methods. A typical question might describe a survey conducted by a company and ask you to identify the sampling method – simple random, stratified, systematic, quota or cluster – and then explain one advantage and one disadvantage in context. Merit comes from using the scenario details rather than giving a generic textbook answer. For instance, if the question says “the researcher stood at the entrance,” you can flag that this is opportunity sampling and that it may produce a biased sample because only people visiting at that time are included.

历年真题持续考查你对数据收集方法的批判能力。一道典型题目可能描述某公司开展的一项调查，要求你辨别抽样方法——简单随机、分层、系统、配额或整群抽样——然后结合情境解释一个优点和一个缺点。得分关键在于利用题目情境的细节，而非给出笼统的教科书答案。例如，若题目提及“研究人员站在入口处”，你可以指出这是便利抽样，可能产生有偏样本，因为只有当时到访的人被包括在内。

Stratified sampling frequently appears because it links directly to proportional reasoning. Past questions often provide a population table (e.g., by age group) and ask you to calculate how many individuals should be sampled from each stratum. The formula is (stratum size ÷ population size) × sample size, and marks are awarded for setting up the proportion and for rounding to a whole number. Examiners expect you to state clearly why you rounded up or down, so it is good practice to show the unrounded number first and then write “so we select 23 people from the stratum.”

分层抽样频频出现，因为它直接关联比例推理。历年真题常给出一个人口表格（如按年龄分组），要求你计算从每个层中应该抽取多少个个体。所用公式为（层大小 ÷ 总体大小）× 样本大小，建立比例式并舍入为整数即可得分。考官期望你清楚说明向上或向下舍入的理由，因此好的做法是先写出未舍入的数值，再陈述“由此我们从此层选取23人”。

Bias and reliability are favourite themes in evaluation parts. A past-paper scenario may state that a questionnaire was only emailed to existing customers. Here you must identify non-response bias and also recognise that the sample is not representative of all potential customers. Full-mark responses go further: they suggest an improvement, such as “send the questionnaire by post to a random sample of people living within 10 miles of the store to improve coverage.” Connecting criticism to a practical remedy is a pattern that runs through many higher-tier responses.

偏差与可靠性是评估部分最受欢迎的主题。一道真题情境可能描述问卷仅通过电子邮件发送给现有客户。此时你必须识别无应答偏差，并认识到样本不代表所有潜在客户。满分答案会进一步深入：提出改进建议，如“向居住在店铺10英里范围内随机抽取的人员邮寄问卷，以提高覆盖面”。将批评与实用补救措施联系起来，这一模式贯穿许多高分段答卷。

Sampling frame errors are easy to miss. If a question describes taking a random sample from the phone book, you need to point out that people without landlines are excluded, which under-represents younger age groups. Past mark schemes reward naming the group excluded and explaining the effect on the results. Practise scanning scenarios for hidden assumptions: “every 10th car entering a car park” sounds systematic, but is the car park used by employees and shoppers differently? Contextual thinking pays off.

抽样框错误容易被忽略。如果题目描述从电话簿中随机取样，你需要指出没有固定电话的人被排除在外，这会导致年轻年龄组代表性不足。历年评分标准奖励指出被排除的群体并解释其对结果的影响。练习扫描情境中的隐藏假设：“每隔10辆进入停车场的车”听起来是系统抽样，但该停车场的使用是否因雇员和购物者而不同？情境化思考必有回报。

---

3. Representing Data: Charts and Diagrams | 数据表示：图表与示意图

Eduqas past papers expect you both to read and to construct a variety of statistical diagrams. Bar charts, pie charts, population pyramids, cumulative frequency curves, histograms and box plots all appear regularly. A frequent task is to complete an unfinished histogram from a grouped frequency table when the class widths are unequal. The height of each bar must represent frequency density (frequency ÷ class width). Lost marks almost always come from forgetting to adjust for unequal class widths – a mistake examiners are acutely aware of and design questions to trap.

Eduqas历年真题要求你既会阅读也会构建多种统计图。条形图、饼图、人口金字塔、累积频率曲线、直方图和箱形图都经常出现。一项常见任务是根据不等组距的分组频数表补全未完成的直方图。每个直条的高度必须表示频率密度（频数 ÷ 组距）。失分几乎总因忘记针对不等组距进行调整——这是考官深知的易错点，专门设计题目来筛选。

For cumulative frequency graphs, questions typically ask you to find the median and interquartile range, and then to compare two distributions. You must clearly show your working on the graph – drawing lines from the quartile positions on the vertical axis to the curve and then down to the horizontal axis – and write both the value and its unit. Past mark schemes award a mark for the graph line markings, not just the final answer. Teach yourself to annotate your graph every time; during revision, check that your crossed lines are visible and that you have labelled them with “Q1″ or “median”.

对于累积频率图，题目通常要求你找到中位数和四分位距，然后比较两分布。你必须在图上清晰展示操作过程——从纵轴上的四分位数位置画线到曲线，再向下画至横轴——并写出数值连同单位。历年评分标准为图上标记线赋予分数，而不仅仅是最终答案。务必养成每次都在图上做标注的习惯；复习时应检查所作交叉线是否清晰可见，并是否标注了“Q1”或“median”。

Box plots drawn from summary data are a firm favourite. You might be given a list or stem-and-leaf diagram and asked to produce a box plot, labelling whiskers, quartiles and any outliers. Outliers are determined using the 1.5 × IQR rule. A past-paper examiner report noted that many candidates correctly identified an upper boundary but then incorrectly plotted the outlier as a cross within the whisker range. Outliers must be shown as separate symbols (usually crosses) beyond the whiskers; the whisker then extends to the largest value that is not an outlier.

根据摘要数据绘制箱形图是一个常考内容。你可能会得到一个列表或茎叶图，被要求绘制箱形图，并标注须线、四分位数以及任何异常值。异常值采用 1.5 × IQR 规则判定。一份真题考官报告指出，许多考生正确计算出上界，却错误地将异常值画在须线范围内。异常值必须用单独的符号（通常是叉号）绘制在须线之外；然后须线延伸至非异常值中的最大值。

Comparison of diagrams requires precise language. When asked to compare two box plots, past mark schemes reward identifying a difference in median, a difference in spread (range or IQR) and a difference in skew. Write in full sentences: “The median systolic blood pressure for the active group is 122 mmHg, whereas for the sedentary group it is 136 mmHg, which shows that sedentary adults tend to have higher blood pressure.” Avoid vague comments like “they are different” – always support with figures from the diagrams.

图的对比需要精确的语言。当被要求比较两个箱形图时，历年评分标准奖励指出中位数差异、离散程度（全距或四分位距）差异以及偏态差异。用完整句子表述：“运动组收缩压中位数为122 mmHg，而久坐组为136 mmHg，这表明久坐成年人往往血压更高。”避免“它们不同”之类的模糊表述——始终用图上的数字作为支撑。

---

4. Measures of Central Tendency and Spread | 集中趋势与离散程度的度量

Calculating and interpreting mean, median, mode and range form the bedrock of the GCSE, yet past papers reveal subtle ways these concepts are tested. You may be given a partially filled frequency table and asked to find a missing value using a known mean. Set up the equation (∑fx ÷ ∑f = given mean), solve for the unknown, and always check your answer by re-calculating the mean. This style of question appears regularly because it tests both procedural fluency and algebraic reasoning. Be careful with units: if the table values are in cm, give the mean in cm, not a mixture.

计算与解读平均值、中位数、众数和全距是GCSE的基础，然而历年真题揭示了这些概念被考查的精妙方式。你可能会得到一个部分填写的频数表，被要求利用已知平均数找出缺失值。建立方程（∑fx ÷ ∑f = 给定平均数），求解未知数，并始终通过重新计算平均数来检验答案。此类题目频繁出现，因为它既考查操作熟练度又考查代数推理。注意单位：如果表格数值以cm为单位，平均数的单位也应用cm，不能混用。

Choosing the appropriate measure is a higher-order skill frequently examined in context-rich scenarios. For a wages data set with a few extremely high salaries, the median is more representative than the mean because the mean is distorted by outliers. Past papers expect you to use the precise phrase “the median is not affected by extreme values” rather than simply “it’s better”. Likewise, when comparing two sets of exam marks with similar means but very different ranges, you must comment on the consistency of performance and link it to a recommendation or interpretation explicitly requested in the question.

选择恰当的度量指标是一项高阶技能，常在情境丰富的题目中加以考查。对于包含少数极高薪水的工资数据集，中位数比平均数更具代表性，因为平均数受异常值扭曲。历年真题期望你使用精确表述“中位数不受极端值影响”，而非简单地说“它更好”。类似地，当比较两组平均数相近但全距差异很大的考试成绩时，你必须对成绩的一致性加以评论，并将其与题目明确要求的建议或解读相联系。

Standard deviation appears in both the foundation and higher tiers, but depth of interpretation varies. You must be able to compute it using the formula: s = √[∑(x − x̅)² ÷ (n − 1)] for a sample, or the population version when instructed. Past papers often provide the sum of squared deviations to reduce arithmetic load, shifting the focus to interpretation. A smaller standard deviation indicates data are clustered closer to the mean. A typical comparison question: “Machine A produces nails with mean length 50.0 mm and standard deviation 0.7 mm; Machine B produces mean 50.1 mm and standard deviation 1.4 mm. Which machine is more reliable? Explain.” Your answer must refer to the lower variability of Machine A, not just the means.

标准差在基础和高阶试卷中均有出现，但解读的深度不同。你必须能使用公式计算样本标准差：s = √[∑(x − x̅)² ÷ (n − 1)]，或按题目要求使用总体版本。历年真题常提供离差平方和以减轻计算量，将重点转向解读。标准差越小，数据越集中在均值附近。一道典型的比较题：“机器A生产的钉子平均长度为50.0 mm，标准差为0.7 mm；机器B生产平均50.1 mm，标准差为1.4 mm。哪台机器更可靠？请解释。”你的答案必须提及机器A的变异较低，而不仅仅是讲均值。

Combining mean and range when comparing prompts careful thinking about advantages. The mean uses all data values, which is both a strength and a weakness: it is comprehensive but sensitive to outliers. The range is quick to compute but only considers extremes. A past-paper question might present a traffic survey comparing two junctions: one with a consistent flow and one with bursty traffic. Comment that the mean is similar but the range is much larger for the second junction, indicating unpredictable traffic, which is important for road safety decisions.

在比较时结合平均数与全距，促使学生审慎思考其优缺点。平均数使用所有数据值，这既是优点也是缺点：它信息全面，但对异常值敏感。全距易于计算，但只考虑极端值。一道真题可能呈现两个交叉口的交通流量调查：一个流量平稳，另一个呈爆发式。你应评论：平均数相似，但第二个交叉口全距大得多，表明交通流量不可预测，这对道路安全决策很重要。

---

5. Probability and Venn Diagrams | 概率与维恩图

Probability in Eduqas Statistics goes beyond simple fractions; it requires clear communication of likelihood and risk. Past paper questions often embed probability in a real-world setting, such as weather forecasts or effectiveness of a test. You must first identify the sample space and then express probabilities as fractions, decimals or percentages. When the question says “estimate the probability that…,” it usually expects you to use a relative frequency from the given data. Distinguish carefully between theoretical probability and experimental probability – past examiners have penalised candidates who treated a long-run relative frequency as an exact theoretical value without acknowledging uncertainty.

Eduqas统计学中的概率超越了简单的分数；它要求清晰传达可能性与风险。历年真题常将概率嵌入现实情境，如天气预报或检测有效性。你必须先确定样本空间，再将概率表示为分数、小数或百分比。当题目说“估计…的概率”时，通常希望你使用给定数据中的相对频率。务必仔细区分理论概率与实验概率——历年考官曾扣分那些将长期相对频率当作精确理论值而未承认不确定性的考生。

Venn diagrams with three or four events test setting up intersecting sets. A standard question provides a Venn diagram with numbers in regions and asks for probabilities of combined events, such as P(A ∪ B) or P(A′ ∩ B). The key is to total all numbers to find the total sample size, then apply the addition rule carefully. For P(A ∪ B) = P(A) + P(B) − P(A ∩ B). Past papers show that a common error is double-counting the intersection; show your subtraction explicitly. For conditional probability, use the formula P(A | B) = P(A ∩ B) ÷ P(B). Read the question wording: “given that” signals conditional probability.

包含三或四个事件的维恩图考查建立交集的能力。一道标准题目给出带有区域数字的维恩图，要求计算组合事件的概率，如 P(A ∪ B) 或 P(A′ ∩ B)。关键是将所有数字相加得出样本总数，然后仔细应用加法规则。对于 P(A ∪ B) = P(A) + P(B) − P(A ∩ B)。历年真题显示，常见错误是重复计算交集；要明确显示出减法步骤。对于条件概率，使用公式 P(A | B) = P(A ∩ B) ÷ P(B)。仔细读题：“已知…”即提示条件概率。

Tree diagrams are used for multi-stage events, and exams frequently provide a partially completed tree. Your first task is to fill in missing branch probabilities, remembering that probabilities on branches from the same node must sum to 1. Then calculate combined-event probabilities by multiplying along the branches, and finally add probabilities for mutually exclusive outcomes when the question asks for “at least one” or “either…or”. A past-paper trap: forgetting to multiply by the number of combinations when order matters. Write all routes explicitly and tick them off to avoid missing a branch.

树状图用于多阶段事件，考试常给出部分完成的树状图。你的首要任务是填上缺失的分支概率，并牢记同一节点出发的分支概率之和必须为1。然后沿分支相乘计算组合事件概率，最后在题目询问“至少一个”或“要么…要么…”时，将互斥结果的概率相加。真题陷阱：顺序重要时忘记乘以组合数量。明确写出所有路径并打勾核对，以避免遗漏某个分支。

Expected frequency questions link probability back to data. A past paper may give a probability of a component being faulty and ask how many faulty components you would expect in a delivery of 500. Simply multiply probability by 500. But be alert: if the probability is estimated from a sample, the expected value is an estimate, not a guarantee. Good answers acknowledge this by using language such as “we would expect approximately 23 faulty components.” This careful phrasing matches top-level mark scheme descriptors.

期望频数题目将概率与数据相联系。一道真题可能给出一个组件有缺陷的概率，并询问在500个交付件中预期有多少缺陷件。只需将概率乘以500。但要警觉：若概率是从样本估计而来，期望值仅为估计值，并非保证。优秀答案通过使用“我们预期大约23个缺陷件”这类措辞来承认这一点。这种谨慎用语符合最高等级评分标准的要求。

---

6. Bivariate Data and Scatter Graphs | 双变量数据与散点图

Scatter graphs are consistently examined through interpretation and line-of-best-fit tasks. You may be asked to plot additional points, describe the correlation (positive, negative or none) and discuss its strength. Past papers use real-life variables: temperature and ice cream sales, revision hours and test scores, engine size and fuel efficiency. When describing correlation, always mention direction, form (roughly linear?) and strength, using the data context. For example: “There is a strong, negative correlation between engine size and miles per gallon; as engine size increases, fuel efficiency tends to decrease.” Never confuse correlation with causation – a high-scoring response will add “but this does not necessarily prove that larger engines cause lower efficiency.”

散点图通过解读和最佳拟合线任务持续纳入考查。你可能被要求描点，描述相关性（正、负或无），并讨论其强弱。历年真题使用现实变量：温度与冰淇淋销量、复习时间与考试分数、发动机排量与燃油效率。描述相关性时，始终结合数据情境说明方向、形式（大致线性？）和强度。例如：“发动机排量与每加仑英里数之间存在强负相关；随着排量增大，燃油效率趋于降低。”切勿混淆相关与因果——高分答卷会加上“但这并不一定证明较大排量导致效率降低”。

Drawing a line of best fit by eye is a skill that requires practice. The line should pass through the middle of the points, with roughly equal numbers above and below, and it must follow the trend. Past examiners advise using a transparent ruler and ensuring the line extends across the full range of plotted data, not just through the cluster. After drawing the line, typical follow-up questions ask you to estimate one variable from the other, either by interpolation (within the data range) or extrapolation (beyond). Always indicate on the graph where you read off the estimate by drawing dashed construction lines, and state clearly if you are extrapolating, because extrapolation is less reliable and this recognition earns a mark.

目测绘制最佳拟合线是一项需练习的技能。该直线应穿过点的中心，上下各有大致相等数量的点，并且必须符合趋势。历年考官建议使用透明直尺，确保直线延伸至所有绘图数据的整个范围，而非仅穿过点簇。画线后，典型的后续问题是根据一个变量估计另一个变量，可以是内插（数据范围内）或外推（数据范围外）。始终在图上用虚辅助线标示取估计值的位置，并清楚说明是否为外推，因为外推较不可靠，承认这一点能得分。

Spearman’s rank correlation coefficient occasionally appears on higher-tier papers. The formula rₛ = 1 − (6∑d²) ÷ n(n² − 1) is provided, but you must rank data correctly, handling tied ranks by assigning the mean rank. A past-paper challenge: given two sets of ranks, interpret an rₛ value of 0.85. Your answer should state “strong positive correlation” and relate it to the context, e.g., “Judges A and B largely agree on the ranking of the gymnasts.” Remember that rₛ values close to 0 suggest no monotonic relationship; a perfect rₛ of 1 indicates complete agreement in ranking order.

斯皮尔曼等级相关系数偶尔出现在高阶试卷中。公式 rₛ = 1 − (6∑d²) ÷ n(n² − 1) 已提供，但你必须正确排定等级，处理并列等级时采用平均等级。一道真题挑战：给定两组等级，解读 rₛ 值为0.85。你的答案应说明“强正相关”，并联系情境，如“评委A和B对体操运动员的排名大体一致”。记住，rₛ 值接近0表示无单调关系；完美的 rₛ 值1代表排名顺序完全一致。

Common pitfalls include ignoring outliers when drawing the line of best fit, misreading the scale (especially when axes don’t start at zero) and confusing independent and dependent variables. The independent variable (explanatory) goes on the horizontal axis. Past papers often ask which variable is the response variable; correct identification is crucial because it affects later interpretations. Lay out your work clearly: label axes, title the graph and write a short sentence linking the evidence to the conclusion requested.

常见误区包括绘制最佳拟合线时忽略异常值、误读刻度（尤其当坐标轴不从零开始时）以及混淆自变量与因变量。自变量（解释变量）绘制在横轴上。历年真题常问哪个是响应变量；正确识别至关重要，因为它影响后续解读。清晰展示你的工作：标注坐标轴，为图表命名，并写一个简短句子将证据与所要求的结论联系起来。

---

7. Time Series Analysis | 时间序列分析

Time series questions present data recorded over time (e.g., quarterly sales, monthly temperatures) and assess your ability to identify trend and seasonal variation. Eduqas past papers frequently provide a time series graph and ask you to plot moving averages. A four-point moving average for quarterly data is common: average the first four values, then drop the first and add the next, and so on. Plot these moving averages on the same graph, ideally at the mid-point of the time intervals they represent. Smoothing reveals the underlying trend; you then explain that the moving average removes seasonal fluctuations to show the general direction.

时间序列题目呈现随时间记录的数据（如季度销售额、月气温），考查你识别趋势与季节变动的能力。Eduqas历年真题常提供时间序列图，要求你绘制移动平均线。季度数据的四点移动平均很常见：取前四个值的平均，然后去掉第一个加上下一个，依此类推。将这些移动平均值绘制在同一图上，理想位置是它们所代表的时间区间的中点。平滑处理揭示出潜在趋势；然后，你解释移动平均消除了季节波动，以显示总体方向。

Calculation of seasonal effects or variations is a higher-tier requirement. Once the trend line is established, you find the seasonal effect by subtracting the trend value from the actual value: seasonal variation = actual − trend. Positive variations indicate the data are above the trend, negative below. A past paper might ask you to calculate the mean seasonal variation for each quarter and use it to predict future sales. Always check that your seasonal variations sum to (approximately) zero for additive models; any significant difference suggests an irregular component.

季节效应或季节变差的计算是高阶要求。一旦确定了趋势线，你通过从实际值中减去趋势值来求得季节效应：季节变差 = 实际值 − 趋势值。正变差表示数据高于趋势，负变差表示低于趋势。一道真题可能要求你计算每个季度的平均季节变差，并用以预测未来销售额。始终检查可加模型中季节变差之和是否（近似）为零；任何显著差异均提示存在不规则成分。

Prediction tasks in time series combine the estimated trend with seasonal adjustments. For a future quarter, extend the trend (often using a line of best fit or the average step change) and then add the appropriate mean seasonal variation. Be cautious about extrapolation: state clearly that predictions assume the trend and seasonal pattern continue, which may not be valid if external factors change. Full-mark answers often include a qualifying statement such as “this forecast is only reliable if current economic conditions persist.” This shows critical evaluation, a skill highly rewarded in the evaluation strand.

时间序列中的预测任务结合了估计趋势与季节调整。对于未来某个季度，延伸趋势（通常使用最佳拟合线或平均步长变化），然后加上相应的平均季节变差。外推时需谨慎：清楚陈述预测假设趋势与季节模式持续不变，若外部因素变化则可能失效。满分答案常包含限定性陈述，如“此预测仅在当前经济状况保持不变时才可靠。”这展示了批判性评估，是评估环节高度奖励的能力。

Graphical interpretation of time series also includes commenting on unusual points. A past-paper graph might show a sudden spike in sales during December; identify this as a seasonal peak if it recurs annually. However, if a one-off spike appears, suggest a plausible cause such as a marketing campaign or a supply shortage and discuss its effect on the moving average. Interaction between context and statistical evidence is the hallmark of a top band response – always tie your observations back to the real-world scenario in the question stem.

时间序列的图形解读还包括对异常点的评论。一道真题图可能显示12月销售额突然飙升；若该情况每年重复出现，可将其识别为季节高峰。然而，若出现的是一次性飙升，则提出一个合理的诱因，例如营销活动或供应短缺，并讨论其对移动平均的影响。情境与统计证据之间的互动是高分段答卷的标志——始终将你的观察与题目给出的现实情境联系起来。

---

8. Index Numbers | 指数

Index numbers are used to compare changes in price, quantity or value over time relative to a base period. Eduqas past questions usually give a table of prices for a “basket” of goods in different years and ask to calculate a weighted price index such as the Retail Price Index (RPI). The first step is to compute the price relative for each item: (price in current year ÷ price in base year) × 100. Then, apply the weights – typically the expenditure in the base year – by multiplying each price relative by its weight, sum these weighted relatives and divide by total weight. The result is a single index number reflecting overall price change.

指数用于比较价格、数量或价值随时间相对于基期的变化。Eduqas历年真题通常给出不同年份“一篮子”商品价格表，要求计算加权价格指数，如零售物价指数(RPI)。第一步是计算每种商品的价比：(当年价格 ÷ 基年价格) × 100。然后，应用权数——通常是基年支出——将每个价比乘以对应权数，将这些加权价比求和，并除以总权数。结果即反映整体价格变动的单个指数。

Interpreting index values is key: an index of 120 means prices have increased by 20 % since the base year. Past papers ask you to compare two time points using indices, perhaps to calculate the percentage increase from a non-base year. If the index in 2018 is 140 and in 2022 is 158.2, the percentage change is (158.2 − 140) ÷ 140 × 100 = 13 %. Do not simply subtract indices – always use the percentage change formula. This area is rife with common mistakes; show your working clearly, writing the formula or at least the calculation steps.

解读指数值很关键：指数120表示价格自基年以来上涨了20%。历年真题要求使用指数比较两个时点，也许是计算从非基年开始的百分比增长率。若2018年指数为140，2022年为158.2，则百分比变化为 (158.2 − 140) ÷ 140 × 100 = 13%。不要简单地指数相减——始终使用百分比变化公式。此部分常见错误频发；清晰展示计算过程，写出公式或至少写出计算步骤。

Weighting is the most powerful concept in index numbers. Questions often ask why weights are used: to reflect the relative importance of items in the basket. For example, housing costs usually have a higher weight than theatre tickets. A past-paper evaluation asks, “Explain one limitation of a price index.” An excellent answer notes that the fixed basket becomes outdated as consumer preferences change, and that the index may not reflect the spending patterns of specific subgroups. Such critical awareness transforms a good answer into an outstanding one.

权数是指数概念中最有力的部分。题目常

Published by TutorHao | Year 11 统计 Revision Series | aleveler.com

更多咨询请联系16621398022（同微信）