📚 Year 11 OCR Statistics: Comprehensive Syllabus Breakdown | Year 11 OCR 统计：课程大纲全面解析
In Year 11, students following the OCR GCSE Statistics specification consolidate and extend the statistical skills required to collect, process, analyse and interpret data in a wide range of contexts. This comprehensive guide breaks down the entire syllabus into manageable sections, explaining key concepts, common pitfalls and examination tips. Whether you are aiming to master sampling techniques, probability distributions or hypothesis testing, this resource will support your revision and deepen your understanding of the subject.
在 Year 11 阶段，学习 OCR GCSE 统计课程的学生将进一步巩固和拓展收集、处理、分析和解释数据的统计技能。本全面解析将整个课程大纲拆分为易于掌握的小节，讲解关键概念、常见错误和应试技巧。无论你的目标是掌握抽样方法、概率分布还是假设检验，这份资料都能助力你的复习并加深对学科的理解。
1. Data Collection and Sampling Methods | 数据收集与抽样方法
Primary data is information you collect yourself through experiments, surveys or observations, while secondary data has already been gathered by someone else, such as government statistics or newspaper articles. The choice between these two depends on cost, time and reliability. A census collects data from every member of a population, giving completely accurate results but often being impractical. Sampling involves selecting a subset of the population, and the method must minimise bias to produce a representative sample.
一手数据是你通过实验、调查或观察亲自收集的信息，而二手数据是他人已收集好的资料，例如政府统计数据或报刊文章。两者之间的选择取决于成本、时间和可靠性。普查从总体的每一个成员收集数据，结果完全准确但往往不切实际。抽样则选取总体的一个子集，其方法必须尽量减少偏差以得到代表性样本。
Simple random sampling gives every individual an equal chance of being chosen, for example using a random number generator to select names from a list. It is unbiased but may miss certain subgroups.
简单随机抽样使每个个体有均等的机会被选中，例如用随机数生成器从名单中选择姓名。这一方法无偏差但可能遗漏某些子群体。
Stratified sampling divides the population into distinct strata (e.g. age groups, genders) and takes a random sample from each in proportion to its size. This guarantees representation of all key groups but requires knowledge of the population structure beforehand.
分层抽样将总体划分为明显的层（如年龄组、性别），然后按比例从每一层随机抽取样本。这保证了所有关键群体的代表性，但需要事先了解总体结构。
Systematic sampling selects individuals at regular intervals from an ordered list. It is quick to use but can introduce bias if the list has an underlying pattern.
系统抽样从有序列表中每隔固定间隔选取个体。使用快捷，但如果列表存在潜在模式则会引入偏差。
Cluster sampling divides the population into clusters, randomly selects some clusters and then surveys all members of those clusters. It is cost-effective for geographically spread populations but may be less precise than stratified sampling.
整群抽样将总体分成群组，随机选取一些群组，然后调查这些群组的所有成员。对地理分布广的总体经济有效，但可能不如分层抽样精确。
Quota sampling is a non-random method where interviewers select people to meet fixed quotas of characteristics. It is cheap and fast but highly susceptible to researcher bias and cannot be used to calculate sampling error.
配额抽样是一种非随机方法，访调员选择人员以满足设定的特征配额。成本低、速度快，但极易受研究者偏见影响，并且无法计算抽样误差。
When designing a questionnaire, questions must be clear, unbiased and avoid leading the respondent. Pilot surveys help identify ambiguous wording. Sampling frames (a list of all sampling units) must be up-to-date and inclusive, otherwise coverage error occurs.
设计问卷时，问题必须清晰、无偏见并避免引导受访者。试调查有助于发现含糊的措辞。抽样框（所有抽样单位的列表）必须最新且完备，否则会出现涵盖误差。
2. Types of Data and Data Handling | 数据类型与数据处理
Data is classified into categorical (qualitative) and numerical (quantitative) types. Categorical data can be nominal, where categories have no natural order (e.g. eye colour, car make), or ordinal, where categories can be ranked (e.g. satisfaction ratings from 'poor' to 'excellent'). Numerical data is either discrete, arising from counting that produces whole numbers (e.g. number of students in a class), or continuous, measured on a scale and able to take any value within a range (e.g. height, time).
数据分为分类（定性）数据和数值（定量）数据。分类数据可以是名义数据，即类别没有自然顺序（如眼睛颜色、汽车品牌）；或有序数据，即可排序（如满意度评级从“差”到“优秀”）。数值数据可以是离散数据，源于计数产生整数（如班级学生人数）；或连续数据，在量尺上测量并可在范围内取任意值（如身高、时间）。
Boundaries must be defined so that no data value falls into two groups; for example, intervals like 0 ≤ h < 10, 10 ≤ h < 20 avoid overlap. The midpoint of an interval is used when calculating estimates of the mean from grouped data.
将连续数据分组成组距可简化大数据集，但组距宽度的选择会影响分布的形状。必须明确定义边界以保证数据值不会落入两组；例如，组距 0 ≤ h < 10, 10 ≤ h < 20 避免了重叠。在根据分组数据估算平均值时，使用组距的中点值。
l distance from other values and can be detected using the interquartile range rule (lower fence = Q₁ – 1.5 × IQR, upper fence = Q₃ + 1.5 × IQR).
数据清理包括识别并处理异常值、缺失值和不一致之处。异常值是指与其他数值距离异常远的观测值，可以使用四分位距法则检测（下限 = Q₁ – 1.5 × IQR，上限 = Q₃ + 1.5 × IQR）。
3. Charts, Graphs and Diagrams | 图表与图示
e story you want to tell. For categorical data, bar charts (with gaps between bars) and pie charts are standard. A pie chart shows proportions of a whole, with each slice angle calculated as (frequency / total) × 360°.
数据的可视化呈现对于揭示模式、趋势和比较至关重要。图表的选择取决于数据类型以及你想传达的信息。对于分类数据，条形图（柱形之间有间隙）和饼图是标准选择。饼图显示整体的比例，每块扇形角度计算为（频数 / 总数）× 360°。
For discrete and continuous numerical data, histograms with no gaps between bars are used, where the area of each bar is proportional to frequency. If class intervals are unequal, the frequency density (frequency ÷ class width) must be plotted on the vertical axis. Cumulative frequency diagrams and box plots (box-and-whisker plots) are used to display the spread and quartiles of a dataset. A cumulative frequency curve allows estimation of the median and other percentiles.
对于离散和连续数值数据，使用柱形之间无间隙的直方图，每根柱子的面积与频数成正比。若组距不相等，则垂直轴须绘制频率密度（频数 ÷ 组距宽度）。累积频数图和箱线图（盒须图）用于展示数据集的离散程度和四分位数。累积频数曲线可用来估算中位数和其他百分位数。
Scatter diagrams plot bivariate data to investigate relationships between two variables. A line of best fit drawn by eye (or the least squares regression line) can model the association. Stem-and-leaf diagrams retain the original data values while showing shape, and back-to-back versions allow comparison of two datasets.
散点图绘制双变量数据以探究两个变量之间的关系。目测绘制的最佳拟合线（或最小二乘回归线）可以对关联关系进行建模。茎叶图在展示分布形态的同时保留了原始数据值，背靠背茎叶图则可以比较两个数据集。
4. Measures of Central Tendency | 集中趋势的度量
The mean, median and mode are the three main averages. The mean (x̄) is the sum of all data values divided by the number of values: x̄ itive to outliers. For grouped data, the estimated mean is Σ(fx) / Σf, using class midpoints x and frequencies f.
平均数、中位数和众数是三种主要的集中趋势度量。平均数（x̄）是所有数据值之和除以数据个数： x̄ = Σx / n。它利用了所有数据，但对异常值敏感。对于分组数据，估算平均数为 Σ(fx) / Σf，使用组中点值 x 和频数 f。
The median is the middle value when data are ordered. If there are n values and n is odd, the median is the (n+1)/2 th value; if n is even, it is the average of the n/2 th and (n/2 + 1) th values. The median is robust to outliers and is the preferred average for skewed distributions. The mode is the most frequently occurring value and is useful for categorical data where the mean cannot be calculated.
中位数是数据按大小排序后位于中间的值。如果有 n 个值且 n 为奇数，中位数是第 (n+1)/2 个值；若 n 为偶数，则为第 n/2 个和第 (n/2 + 1) 个值的平均值。中位数不受异常值影响，是偏态分布中首选的集中趋势度量。众数是出现频率最高的值，对于无法计算平均数的分类数据有用。
From a cumulative frequency graph, the median is estimated by reading the value at half the total frequency. The shape of a distribution influences the choice of average: in a positively skewed distribution, mean > median > mode; in a negatively skewed distribution, mean < median < mode; in a symmetrical distribution, all three coincide.
从累积频数图中，通过读取总频数一半处的值可估算中位数。分布的形状影响平均数的选择：在正偏态分布中，平均数 > 中位数 > 众数；在负偏态分布中，平均数 < 中位数 < 众数；在对称分布中，三者重合。
5. Measures of Dispersion and Spread | 离散程度与分布度量
ures of spread describe how tightly the data cluster around the centre. The range (maximum – the simplest but is heavily affected by extreme values. The interquartile range (IQR = Q₃ – Q₁) measures the middle 50% s can be found by ordering the data and splitting into four equal parts; Q₂ is the median.
离散度量描述数据围绕中心聚集的紧密程度。极差（最大值 – 最小值）最简单，但极易受极端值影响。四分位距（IQR = Q₃ – Q₁）衡量中间 50% 的数据，不受异常值影响。四分位数可通过将数据排序并分成四个等份求得；Q₂ 即中位数。
more sophisticated measures that use every data value. For a population, variance σ² = Σ(x – μ)² / N; for a sample, s² = Σ(x – x̄)² / (n – n indicates data are tightly clustered around the mean; a large one indicates wide spread.
方差和标准差是更为精细的度量，利用了每一个数据值。对于总体，方差 σ² = Σ(x – μ)² / N；对于样本，s² = Σ(x – x̄)² / (n – 1)。标准差 σ（或 s）是方差的平方根，回到原始单位。标准差小表示数据紧密围绕平均数聚集；大则表示离散程度高。
ally higher values. The coefficient of variation (CV = standard deviation / mean) allows comparison of relative dispersion across datasets with different units or widely different means.
比较数据集时，应同时使用平均数与标准差：平均数更高且标准差相近，则说明数值普遍较高。变异系数（CV = 标准差 / 平均数）可用于比较单位不同或平均数差异悬殊的数据集的相对离散程度。
6. Introduction to Probability | 概率入门
y likely outcomes is P(event) = number of favourable outcomes / total number of possible outcomes. Experimental probability is based on repeated trials: relative frequency = number of successes / number of trials. As the number of trials increases, experimental probability tends to approach theoretical probability (the law of large numbers).
概率是事件可能性的度量，在 0（不可能）到 1（确定）的标度上表示。对于等可能结果，理论概率为 P(事件) = 有利结果数 / 可能结果总数。实验概率基于重复试验：相对频率 = 成功次数 / 试验次数。随着试验次数增加，实验概率趋于理论概率（大数定律）。
or B) = P(A) + P(B). For non-mutually exclusive events, we subtract the intersection: P(A ∪ B) = P(A) + P(B) – P(A ∩ rule applies to independent events: P(A and B) = P(A) × P(B). Conditional probability, P(A|B), is the probability of A occurring given that B has occurred: P(A|B) = P(A ∩ B) / P(B).
事件可使用加法法则组合：对于互斥事件 A 和 B，P(A 或 B) = P(A) + P(B)。对于非互斥事件，需减去交集：P(A ∪ B) = P(A) + P(B) – P(A ∩ B)。乘法法则适用于独立事件：P(A 与 B) = P(A) × P(B)。条件概率 P(A|B) 是在 B 已发生的条件下 A 发生的概率：P(A|B) = P(A ∩ B) / P(B)。
Tree diagrams systematically list outcomes of multistage experiments. Branch probabilities multiply along a path, and probabilities of different paths leading to the same outcome are added. Venn diagrams illustrate sets, intersections and unions, useful for solving conditional probability problems.
树状图系统地列出多阶段试验的结果。路径上的分支概率相乘，导向同一结果的不同路径概率相加。韦恩图说明集合、交集和并集，对求解条件概率问题很有用。
7. Discrete Probability Distributions | 离散概率分布
ties must equal 1. The expected value E(X), or mean of the distribution, is calculated as E(X) = Σ [x · P(X = x)]. It represents the long-run average outcome if the experiment is repeated many times.
概率分布列出离散随机变量的所有可能结果及其概率。所有概率之和必须等于 1。期望值 E(X)，即分布的均值，计算公式为 E(X) = Σ [x · P(X = x)]。它表示若实验重复多次，长期平均的结果。
The variance of a random variable is Var(X) = E(X²) – the linear transformation rules apply: E(aX + b) = aE(X) + b and Var(aX + b) = a² Var(X).
随机变量的方差为 Var(X) = E(X²) – [E(X)]²，其中 E(X²) = Σ [x² · P(X = x)]。标准差是方差的平方根。对于 X 的函数，如 aX + b，适用线性变换规则：E(aX + b) = aE(X) + b，Var(aX + b) = a² Var(X)。
in a fixed number of independent trials, each with the same probability of success p. If X ~ B(n, p), then P(X = r) = ⁿCᵣ pʳ (1 – p)ⁿ⁻ʳ. The mean is np and the variance is np(1 – p). Binomial probabilities can be calculated using the formula, tables or calculators.
二项分布模拟在固定次数的独立试验中，每次试验成功概率 p 相同的成功次数。若 X ~ B(n, p)，则 P(X = r) = ⁿCᵣ pʳ (1 – p)ⁿ⁻ʳ。均值为 np，方差为 np(1 – p)。二项概率可用公式、表格或计算器计算。
8. Bivariate Data and Correlation | 双变量数据与相关性
Bivariate data involves two related variables, such as height and weight. A scatter diagram reveals the direction, form and strength of the relationship. Correlation measures the strength of linear association, ranging from –1 (perfect negative) through 0 (no linear correlation) to +1 (perfect positive).
双变量数据涉及两个相关变量，如身高和体重。散点图揭示关系的方向、形式和强度。相关系数度量线性关联的强度，范围从 –1（完全负相关）到 0（无线性相关）再到 +1（完全正相关）。
nt correlation coefficient (PMCC), denoted by r, is calculated using the formula r = Sxy / √Sxx Syy, where Sxx = Σ(x – x̄)², Syy = Σ(y – Ȳ)² and Sxy = Σ(x – x̄)(y – Ȳ). A value close to +1 indicates strong positive linear correlation; a value close to –1 indicates strong negative linear correlation. Spearman' cient can be used when data is non-linear or contains outliers; it works with ranked data.
积矩相关系数（PMCC），记为 r，使用公式 r = Sxy / √Sxx Syy 计算，其中 Sxx = Σ(x – x̄)²，Syy = Σ(y – Ȳ)²，Sxy = Σ(x – x̄)(y – Ȳ)。接近 +1 的值表示强正线性相关；接近 –1 表示强负线性相关。当数据非线性或含有异常值时，可使用斯皮尔曼等级相关系数；它基于排序数据。
of the least squares regression line of y on x is y = a + bx, where b = Sxy / Sxx and a = ȳ – b x̄ generally reliable, while extrapolation (predicting outside the range) can be misleading.
回归分析基于相关性，用来自一个变量预测另一个变量。y 对 x 的最小二乘回归线方程为 y = a + bx，其中 b = Sxy / Sxx，a = ȳ – b x̄。该直线使得各点到直线的垂直距离平方和最小。内插法（在数据范围内预测）通常可靠，而外推法（在范围外预测）可能产生误导。
9. Time Series Analysis | 时间序列分析
A time series is a sequence of data points recorded at regular time intervals (e.g. monthly sales, daily temperatures). The aim is to identify patterns and forecast future values. A time series can be decomposed into four key components: trend (long-term movement), seasonal variation (regular fluctuations within a year), cyclical variation (longer-term waves often linked to economic cycles) and random (irregular) variation.
时间序列是按固定时间间隔记录的一系列数据点（如月销售额、日温度）。其目的是识别模式并预测未来值。时间序列可分解为四个关键成分：趋势（长期变动）、季节变动（一年内的规则波动）、循环变动（通常与商业周期相关的更长周期波动）和随机（不规则）变动。
ed by averaging each set of four consecutive values and placing the result centrally. The cycle must then be centred when the period is even. Estimated trend values allow calculation of seasonal effects: additive model (actual = trend + seasonal + random) or multiplicative model (actual = trend × seasonal factor × random).
移动平均法可以平滑短期波动并揭示潜在趋势。对于季度数据，四点移动平均的计算方式是依次取四期数值求平均，并将结果置于中心位置。当周期为偶数时，还需进行中心化处理。估算的趋势值可用于计算季节效应：加法模型（实际值 = 趋势 + 季节 + 随机）或乘法模型（实际值 = 趋势 × 季节因子 × 随机）。
Forecasting involves projecting the trend line forward and adding or multiplying the appropriate seasonal component. The reliability of forecasts decreases the further ahead you predict. Plotting original data with moving averages and trend lines aids visual interpretation.
预测包括向前延伸趋势线并加减或乘上相应的季节成分。预测越远其可靠性越低。将原始数据与移动平均和趋势线一起绘制有助于直观解读。
10. Statistical Inference and Hypothesis Testing | 统计推断与假设检验
ere is no effect or no difference, and an alternative hypothesis (H₁) that contradicts it. For a binomial test of a proportion, H₀ might be p = 0.5 and H₁: p > 0.5. The test is based on the probability of obtaining the observed result, or more extreme, assuming H₀ is true.
统计推断利用样本数据对总体得出结论。假设检验从零假设（H₀）开始，它声称无效应或无差异；备择假设（H₁）与之相反。对于比例的二项检验，H₀ 可以是 p = 0.5，H₁: p > 0.5。检验基于假设 H₀ 为真时，得到观测结果或更极端结果的概率。
ed to the significance level α (often 0.05). If p-value ≤ α, we reject H₀ in favour of H₁; the result is statistically significant. If p-value > α, we fail to reject H₀. This does not prove H₀ is true, only that there is insufficient evidence against it.
p 值是零假设下观测检验统计量（如成功次数）或更极端结果出现的概率。将其与显著性水平 α（常用 0.05）相比较。若 p 值 ≤ α，则拒绝 H₀ 支持 H₁；结果具有统计显著性。若 p 值 > α，则未能拒绝 H₀。这并不证明 H₀ 为真，只是没有足够的证据反驳它。
= α). Type II error occurs when H₀ is not rejected when H₁ is true. The power of a test (1 – probability of Type II error) increases with larger sample sizes and larger effect sizes.
第一类错误发生在 H₀ 实际为真却被拒绝时（概率 = α）。第二类错误发生在 H₁ 为真却未能拒绝 H₀ 时。检验的功效（1 – 第二类错误概率）随样本量和效应量的增大而提高。
11. Critical Evaluation of Statistics | 统计的批判性评价
Not all statistics are trustworthy. Critical thinking involves questioning the source, methodology and presentation of data. Common misuses include cherry-picking a favourable subset of data, using a misleading scale on a graph (e.g. truncated axis to exaggerate differences), and confusing correlation with causation. Simpson's paradox illustrates how trends can reverse when data is aggregated across subgroups.
并非所有统计数据都可信。批判性思维要求质疑数据的来源、方法及呈现方式。常见的误用包括：挑选有利的数据子集（摘樱桃），在图表中使用误导性的标尺（如截断坐标轴以夸大差异），以及混淆相关性与因果关系。辛普森悖论说明，当子组的汇总数据合并时，趋势可能发生逆转。
Bias can enter at every stage: selection bias (non-representative sample), measurement bias (faulty instrument or leading question), non-response bias (those who do not respond differ systematically from those who do) and confirmation bias (interpreting results to support pre-existing beliefs). Recognising these pitfalls helps evaluate survey reports, media claims and scientific studies.
偏差可出现在每个阶段：选择偏差（样本不具代表性）、测量偏差（工具缺陷或引导性问题）、无应答偏差（未回应者与回应者系统性地不同）以及确认偏差（解读结果以支持原有信念）。认清这些陷阱有助于评价调查报告、媒体报道和科学研究。
The UK Statistics Authority and similar bodies enforce a Code of Practice to ensure official statistics are trustworthy, high quality and valuable. Context, limitations and margin of error should always accompany published data.
英国统计管理局及类似机构强制执行实务守则，以确保官方统计可信、高质量、有价值。已发布的数据应始终附有背景、局限性和误差范围。
12. Exam Preparation and Assessment Tips | 备考与评估技巧
The OCR GCSE Statistics exam consists of two equally weighted papers. Each paper includes a mixture of short-structured and longer questions, some requiring interpretation of data provided in tables or graphs. Questions assess AO1 (knowledge and understanding), AO2 (application) and AO3 (analysis and evaluation).
OCR GCSE 统计考试由两份权重相同的试卷组成。每份试卷包含短结构题和较长的题目，部分题目要求解读表格或图表中的数据。试题评估 AO1（知识与理解）、AO2（应用）和 AO3（分析与评价）。
When tackling longer questions, read the scenario carefully, identify the statistical concepts involved and show all working clearly. For calculation questions, write down the formula first, substitute numbers, and give an answer to an appropriate degree of accuracy (often 3 significant figures unless stated otherwise). Diagrams such as Venn diagrams, tree diagrams and cumulative frequency graphs carry marks for correct scales and labelling.
解答较长题目时，仔细阅读情景，识别所涉及的统计概念，并清晰地展示所有步骤。对于计算题，先写出公式，代入数值，再给出适当精度的答案（通常为 3 位有效数字，除非另有说明）。韦恩图、树状图和累积频数图等图示因正确的刻度和标注而得分。
Use past papers and mark schemes to familiarise yourself with command words like 'interpret', 'compare' and 'evaluate'. 'Compare' requires a quantitative statement of similarity or difference, such as 'the median of Sample A is 5 units higher than that of Sample B'. 'Evaluate' demands a judgement supported by statistical reasoning. Manage your time: spend roughly one minute per mark, leaving time to check calculations and written conclusions.
利用历年真题和评分标准熟悉指令词，如“解读”、“比较”和“评价”。“比较”要求给出相似或差异的定量陈述，如“样本 A 的中位数比样本 B 高 5 个单位”。“评价”要求以统计推理支持判断。管理好时间：大致每分花费一分钟，留出时间检查计算和书面结论。
Published by TutorHao | Statistics Revision Series | aleveler.com
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