📚 Year 11 AQA Statistics: Core Topic Review | 核心知识点梳理
This article provides a concise yet comprehensive review of the key statistical concepts required for the AQA GCSE Statistics specification at Year 11. It covers data types, sampling, measures of location and spread, graphical representations, correlation, time series, index numbers, probability, the normal distribution, and quality assurance.
本文旨在为Year 11学生提供一份简洁而全面的AQA GCSE统计学核心知识点梳理，涵盖数据类型、抽样、集中趋势与离散度量、图表表示、相关分析、时间序列、指数、概率、正态分布以及质量控制等关键主题，助力高效备考。
1. Types of Data | 数据类型
Data can be classified as categorical (qualitative) or numerical (quantitative). Categorical data splits into nominal, where categories have no natural order (e.g., hair colour), and ordinal, where order matters (e.g., satisfaction rating: poor, good, excellent).
数据可分为分类数据（定性）和数值数据（定量）。分类数据又分为名义型（类别无自然顺序，如头发颜色）和有序型（顺序有意义，如满意度评分：差、良、优）。
Numerical data is either discrete (countable, often integers, e.g., number of cars in a household) or continuous (measurable, any value within a range, e.g., mass, time). Choosing the right graph or statistical measure depends critically on this classification.
数值数据可以是离散型（可数，常为整数，如家庭汽车数量）或连续型（可测，在区间内可取任意值，如质量、时间）。选择正确的图表或统计量在很大程度上取决于这一分类。
2. Sampling Methods | 抽样方法
A sample must be representative to allow valid conclusions about the population. The key methods are summarised below, with their strengths and weaknesses.
样本必须具有代表性，才能对总体做出有效推断。下面总结主要抽样方法及其优缺点。
Simple random sampling gives every member an equal chance of selection, reducing bias, but it may miss important sub-groups if the sample is small.
简单随机抽样使每个成员被选中的机会均等，减少了偏差，但如果样本较小，可能会遗漏重要的子群。
Stratified sampling divides the population into distinct strata (e.g., age bands) and samples each proportionally. This ensures key groups are represented, but it requires detailed population information.
分层抽样将总体划分为不同的层（如年龄段），并按比例从各层抽样。这保证了关键群体的代表性，但需要详细的总体信息。
Systematic sampling selects every kth individual from a list. It is quick and can be used when a sampling frame exists, but a hidden periodicity can introduce bias.
系统抽样从名单中每隔k个个体选取一个。它快捷且可用于存在抽样框的情形，但隐藏的周期性可能引入偏差。
Quota sampling allows interviewers to choose a fixed number of people in each quota category. It is cheap and fast but not random, so significant bias is possible.
配额抽样允许采访者在每个配额类别中选取固定数量的人。它成本低、速度快，但非随机，因此可能产生显著偏差。
Convenience sampling uses the most easily available subjects (e.g., friends), leading to poor generalisability and high bias. It should be avoided in rigorous statistical work.
便利抽样使用最容易获得的个体（如朋友），导致可推广性差、偏差大，在严谨的统计工作中应避免使用。
3. Measures of Central Tendency and Spread | 集中趋势与离散度量
The three main measures of central tendency are the mean, median, and mode. The mean for a sample of n values is given by:
x̄ = (∑ x) ÷ n
集中趋势的三个主要度量为均值、中位数和众数。n个值样本的均值由下述公式给出。
The median is the middle value when data are ordered; if n is even, it is the average of the two middle values. The mode is the most frequent value and is particularly useful for categorical data.
中位数是将数据排序后位于中间的值；如果n为偶数，则取中间两个值的平均数。众数是出现频率最高的值，对分类数据特别有用。
Spread is measured by range, interquartile range (IQR), and standard deviation. IQR = Q₃ – Q₁, where Q₁ and Q₃ are the lower and upper quartiles. It is resistant to outliers.
离散程度通过全距、四分位数间距（IQR）和标准差来度量。IQR = Q₃ – Q₁，其中Q₁和Q₃分别为下四分位数和上四分位数。IQR对异常值不敏感。
Standard deviation shows how widely data are dispersed around the mean. For a sample:
s = √[ Σ(x – x̄)² ÷ (n – 1) ]
标准差反映数据围绕均值的离散程度。样本标准差公式如上。
Variance is simply s². Always consider which measure of location and spread is most appropriate depending on the data shape and presence of outliers.
方差即s²。根据数据分布形状及是否存在异常值，务必考虑选择最合适的集中趋势和离散度量。
4. Cumulative Frequency and Box Plots | 累积频率和箱形图
Cumulative frequency is the running total of frequencies. A cumulative frequency graph plots cumulative frequency against the upper class boundary. The median, quartiles, and percentiles can be read from this smooth curve.
累积频率是频数的累加。累积频率图以累积频率对组上限作图。从该平滑曲线上可读出中位数、四分位数和百分位数。
A box plot (or box-and-whisker diagram) summarises the minimum, Q₁, median, Q₃, and maximum. It makes it easy to compare distributions side-by-side and to spot skewness or outliers.
箱形图（盒须图）概括了最小值、Q₁、中位数、Q₃和最大值，便于并列比较不同分布，并可看出偏态或异常值。
Outliers are often defined with the 1.5 × IQR rule: a point below Q₁ – flagged as an outlier. In such cases, the whisker extends to the most extreme non-outlier.
异常值通常用1.5 × IQR规则定义：低于Q₁ – 1.5 × IQR或高于Q₃ + 1.5 × IQR的点被标记为异常值。此时须线延伸至最远的非异常值。
5. Histograms with Unequal Class Widths | 不等组距的直方图
ncy density = frequency ÷ class width. The area of each bar then represents the frequency.
当组距宽度不等时，直接使用频数绘图会产生误导。应使用频率密度 = 频数 ÷ 组距。此时每个直条的面积代表频数。
Always label the vertical axis as 'Frequency density' ts. and carefully calculate heights. Remember that frequency = frequency density × class width for interpreting histograms.
务必将纵轴标记为“频率密度”并仔细计算条形高度。解读直方图时记住频数 = 频率密度 × 组距。
In an exam, you may be asked to complete a frequency density table and then draw or interpret the histogram. Check that the total area equals total frequency.
考试中可能会要求你完善频率密度表，然后绘制或解读直方图。检查总面积是否等于总频数。
6. Scatter Diagrams and Correlation | 散点图及相关性
A scatter diagram displays the relationship between two numerical variables. Correlation measures the strength and direction of a linear association. Positive correlation means as one variable rises, the other tends to rise; negative means the opposite.
散点图展示两个数值变量的关系。相关衡量线性关联的强度和方向。正相关意味着一变量升高，另一变量也趋于升高；负相关则相反。
Correlation is not the same as causation. The product-moment correlation coefficient, r, ranges from –1 (perfect negative) to +1 (perfect positive). A value near 0 indicates no linear correlation.
相关不等于因果。积矩相关系数r的取值范围从–1（完全负相关）到+1（完全正相关）。接近0的值表明没有线性相关。
A line of best fit should pass through the point (x̄, ȳ). Use it for interpolation (estimating within the data range), but avoid extrapolation (outside the range) as it is unreliable.
最佳拟合线应通过点(x̄, ȳ)。可用于内插（在数据范围内估计），但应避免外推（超出范围），因为不可靠。
7. Spearman's Rank Correlation Coefficient | 斯皮尔曼等级相关系数
When data are ordinal or the relationship is monotonic but not linear, Spearman's rank correlation coefficient, rₛ, is appropriate. It is based on the differences in ranks.
rₛ = 1 – (6 ∑ d²) ÷ [n(n² – 1)]
当数据为有序数据或关系单调但非线性时，宜使用斯皮尔曼等级相关系数rₛ，它基于排名的差异。计算公式如上。
Procedure: rank both sets of values separately; find the difference d = rank(x) – re these differences and substitute into the formula. Tied ranks receive the average rank.
步骤：分别对两组值进行排名；计算每对数据的差异 d = rank(x) – rank(y)；求d的平方并代入公式。存在并列排名时赋予平均排名。
Interpret rₛ similarly to r: +1 implies perfect positive association in ranks, –1 perfect negative, and 0 no association. A hypothesis test for significance often uses a critical value table.
解释rₛ类似于r：+1表示等级完全正相关，–1完全负相关，0无关联。显著性检验常使用临界值表。
8. Time Series Analysis and Moving Averages | 时间序列分析与移动平均
A time series is a sequence of observations recorded at regular intervals. Its components are the trend (long-term direction), seasonal variation (regular patterns within a year), and irregular (random) fluctuations.
时间序列是按固定间隔记录的一系列观测值。其组成包括趋势（长期方向）、季节变动（年内规律性模式）和不规则（随机）波动。
To isolate the trend, calculate a moving average. For quarterly data, a 4-point moving average is typical. Two steps: average four consecutive values, then centre by averaging pairs of these averages when needed.
为分解出趋势，需计算移动平均。对于季度数据，常用4项移动平均。分两步：平均连续四个值，必要时再对两个相邻移动平均取均值进行居中化。
Seasonal variation = actual value – typical seasonal effects, which are used to make forecasts or seasonally adjusted figures.
季节变动 = 实际值 – 趋势值。对各季节求平均，得出典型季节效应，用于预测或计算经季节调整的数据。
9. Index Numbers | 指数
period, which is set to 100. A simple price index is: (current price ÷ base price) × 100.
指数表示某个量相对于基期的变化，基期指数设为100。简单价格指数为：(现价 ÷ 基期价格) × 100。
Chain base indices link successive periods, each calculated as a percentage of the immediately preceding period. This can show growth rates more directly.
链基指数将连续各期链接起来，每期都以前一期的百分比计算，可更直接地反映增长率。
The formula for a weighted aggregate price index is ∑(price relative × weight) ÷ ∑ weight.
加权指数，如零售价格指数，利用权重综合多个项目以反映其重要性。加权综合价格指数的公式为∑(价比 × 权重) ÷ ∑ 权重。
10. Basic Probability and Risk | 基础概率与风险
For equally likely outcomes, P(A) = number of ways A can occur ÷ total number of outcomes.
概率量化事件发生的可能性。对于等可能结果，P(A) = 事件A发生的方式数 ÷ 总结果数。
The addition rule: P(A or B) = P(A) + P(B) – P(A and B). For mutually exclusive events, P(A and B) = 0, so the rule simplifies.
加法法则：P(A or B) = P(A) + P(B) – P(A and B)。若事件互斥，则P(A and B) = 0，法则可简化。
× P(B). Tree diagrams help visually with conditional probability and multi-stage problems.
独立事件乘法法则：P(A and B) = P(A) × P(B)。树形图有助于直观处理条件概率和多阶段问题。
risk in exposed group ÷ risk in unexposed group. Absolute risk reduction gives the actual difference in risk. Understanding these quantities is essential for interpreting statistical reports.
在医学情境中，相对风险 = 暴露组风险 ÷ 非暴露组风险。绝对风险降低表示风险的实际差值。理解这些量对解读统计报告至关重要。
11. Normal Distribution and Standardised Scores | 正态分布与标准分
The normal distribution is a continuous, symmetric, bell-shaped distribution described by its mean µ and standard deviation σ. The total area under the curve is 1.
正态分布是一种连续、对称的钟形分布，由其均值µ和标准差σ描述。曲线下总面积为1。
Key empirical rule: approximately 68% of values lie within 1σ of the mean, 95% within 2σ, and 99
Published by TutorHao | Year 11 统计 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