A-Level Cambridge Statistics: Terminology Quick-Memorise Guide | A-Level 剑桥统计：词汇术语速记指南

📚 A-Level Cambridge Statistics: Terminology Quick-Memorise Guide | A-Level 剑桥统计：词汇术语速记指南

Mastering statistical vocabulary is half the battle in A-Level Cambridge Statistics. This guide organises key terms by topic, pairs them with clear definitions, and offers memory hooks to accelerate your recall. Both English and Chinese explanations are provided to bridge any language gap.

攻克 A-Level 剑桥统计，词汇关就是分水岭。本指南按专题整理核心术语，配以清晰释义和速记口诀，助你快速调用概念。中英双语对照讲解，消除理解死角。

---

1. Measures of Central Tendency | 集中趋势度量

Mean (x̄ for sample, μ for population) – the arithmetic average, calculated by summing all data values and dividing by the count. Memory aid: 'Mean' rhymes with 'between', it sits at the balancing point.

均值（样本 x̄，总体 μ） – 所有数据值之和除以数据个数。速记：均值如天平支点，总和归一点。

Median – the middle value when data are ordered. If the number of observations is even, it is the average of the two middle numbers. Think 'media n' as the 'middle one in the row'.

中位数 – 排序后居中的数值。偶数个时取中间两数的平均。记为“中位即中段，排序后截半”。

Mode – the most frequently occurring value. 'Mode' sounds like 'most', so it’s the most popular observation.

众数 – 出现频率最高的值。联想法：众数 = 最多人选的数。

---

2. Measures of Spread | 离散程度度量

Range = Maximum – Minimum. The simplest measure of spread. 'Range' reminds you of how far the data 'range' from end to end.

极差 = 最大值 − 最小值。最简单的分散度量。想成数据“极限之差”。

Interquartile Range (IQR) = Q₃ – Q₁. It covers the middle 50% of data. Think 'IQR' as 'InQuartile Reach' – the reach of the middle half.

四分位距 (IQR) = 第三四分位数 − 第一四分位数。涵盖中间50%的数据。口诀：IQR，中段跨度，去头去尾看中间。

Variance (σ² or s²) – average of squared deviations from the mean. Remember: Variance = 'variation squared', always in squared units.

方差（σ² 或 s²） – 各数据与均值离差平方的平均数。联想：方差就是“差”的平方和的平均，单位是原单位的平方。

Standard Deviation (σ or s) – square root of variance, bringing units back to original scale. 'Standard' means it’s the typical deviation from the mean.

标准差（σ 或 s） – 方差的平方根，恢复原单位。记法：标准差是方差的“降幂还原”。

---

3. Probability Foundations | 概率基础

Experiment – a repeatable process that gives rise to outcomes. E.g., rolling a die.

试验 – 可重复的、产生结果的过程。如掷骰子。

Outcome – a single possible result of an experiment. Memory trick: 'Out' + 'come' = what comes out.

结果 – 试验的一个可能出现的情况。联想：从试验中“出来”的单个结局。

Sample Space (S) – the set of all possible outcomes. Visualise it as the 'space' where all outcomes live.

样本空间 (S) – 所有可能结果的集合。想象成一张“全景底图”，囊括一切可能。

Event (A, B, …) – a subset of the sample space. An event 'happens' if the outcome belongs to that subset.

事件 (A, B, …) – 样本空间的子集。结果落入该子集，则事件发生。

P(A) – probability of event A, always between 0 and 1 inclusive. 0 = impossible, 1 = certain.

P(A) – 事件 A 的概率，值域 [0,1]。0 为不可能，1 为必然。

---

4. Probability Rules and Diagrams | 概率运算法则与图表

Addition Rule: P(A ∪ B) = P(A) + P(B) − P(A ∩ B). Useful for 'or' babilities. Remember the Venn diagram overlap is subtracted once to avoid double counting.

加法法则：P(A ∪ B) = P(A) + P(B) − P(A ∩ B)。用于“或”的情形。文氏图重叠部分被多算了一次，所以要减去。

Multiplication Rule for Independent Events: P(A ∩ B) = P(A) × P(B). Two events are independent if one does not affect the other. 'Multiply when independent' – think of multiplying probabilities along tree branches.

独立事件乘法法则：P(A ∩ B) = P(A) × P(B)。若两事件彼此不影响，则可以直接相乘。树图上顺着枝杈乘过去。

Conditional Probability: P(A|B) = P(A ∩ B) / P(B). 'Given that B occurred, what is the chance of A?' The symbol '|' looks like a gate – you condition on what is on the right.

条件概率：P(A|B) = P(A ∩ B) / P(B)。“在 B 已发生的条件下，A 的概率”。竖线“|”像一扇门，右侧是已知条件。

Tree Diagrams – multiplicative along branches, additive across branches for mutually exclusive outcomes. Label probabilities on branches, conditional probabilities at second stage.

树形图 – 沿分枝相乘，不同分枝的概率相加（互斥结果）。一级分枝标边际概率，二级分枝标条件概率。

---

5. Discrete Probability Distributions | 离散概率分布

Random Variable (X) – a variable whose value is a numerical outcome of a random phenomenon. Capital X denotes the variable; lowercase x denotes a specific value.

随机变量 (X) – 其数值由随机现象的结果决定。大写 X 代表变量，小写 x 代表具体取值。

Discrete Uniform Distribution – each of n outcomes has equal probability 1/n. Example: fair die roll. Mnemonic: 'Uniform' = all outcomes have the same probability 'uniform'.

离散均匀分布 – n 个结果中每个概率均为 1/n。如公平骰子。助记：均匀即概率“均一匀称”。

Binomial Distribution X ~ B(n, p) – xed number of trials, independent, same probability, two outcomes (success/failure). Think 'Bi'nomial = two outcomes, 'nom' = names for success/failure.

二项分布 X ~ B(n, p) – n 次独立试验中成功的次数，每次成功概率为 p。核心词：固定次数、独立、等概率、二结局。联想“二项”即两种可能项。

Poisson Distribution X ~ Po(λ) – models the number of events occurring in a fixed interval of time or space, with mean λ. Conditions: events occur singly, randomly, independently, and at a constant average rate. Memory hook: 'Poisson' sounds like 'fish', and fish arrive randomly in a net – events arrive at rate λ.

泊松分布 X ~ Po(λ) – 建模固定时间段或空间内事件发生的次数，均值为 λ。条件：事件单个出现、随机、独立、平均速率恒定。谐音助记：“泊松”似“泊来的鱼群”，以 λ 的速率随机入网。

---

6. Continuous Distributions and Normal | 连续分布与正态

Probability Density Function (pdf) f(x) – . Total area under the curve = 1. The height is not a probability; area gives probability.

概率密度函数 f(x) – 描述连续随机变量取某值时的相对似然性。曲线下总面积 = 1。高度不是概率，面积才是概率。

Cumulative Distribution Function (cdf) F(x) = P(X ≤ x) – the area under the pdf to the left of x. Always non-decreasing, from 0 to 1.

累积分布函数 F(x) = P(X ≤ x) – 密度函数左尾面积。单调不减，从 0 到 1。

Normal Distribution X ~ N(μ, σ²) – symmetric bell-shaped curve defined by mean μ and variance σ². 68–95–99.7 rule: ≈68% within 1σ of mean, 95% within 2σ, 99.7% within 3σ. Remember 'Normal bell – μ and σ tell'.

正态分布 X ~ N(μ, σ²) – 钟形对称曲线，由均值 μ 和方差 σ² 确定。68-95-99.7 法则：约 68% 在 μ±σ 内，95% 在 μ±2σ 内，99.7% 在 μ±3σ 内。记作“正态钟形，μ 定位，σ 定宽”。

Standard Normal Z ~ N(0, 1) – dardised value Z = (X − μ)/σ. Using Z-tables gives probabilities for any normal. Tip: Z is 'zero‑centered normal'.

标准正态 Z ~ N(0, 1) – 标准化 Z = (X − μ)/σ。查 Z 表可得任意正态概率。口诀：Z 是“归零归一”的正态。

---

7. Sampling and Estimation | 抽样与估计

Population vs Sample – population is the whole group; sample is a subset used to draw conclusions. Mnemonic: 'Pop' = everyone; 'Sample' = a sip.

总体 vs 样本 – 总体是全部对象，样本是用于推断的子集。联想：“总体”如满池水，“样本”如取一瓢饮。

Statistic – a number computed from a sample, e.g., sample mean x̄. Parameter – a number describing a population, e.g., μ. 'S' for statistic comes from Sample; 'P' for parameter comes from Population.

统计量 – 由样本计算出的数值，如 x̄。参数 – 描述总体的数值，如 μ。首字母记忆：Statistics from Samples, Parameters from Populations。

Unbiased Estimator – of the statistic equals the population parameter. Example: sample mean is unbiased for μ. 'Unbiased = aiming at the bullseye without systematic error.'

无偏估计量 – 统计量的期望值等于总体参数。例如样本均值是 μ 的无偏估计。联想：无偏如射箭不偏靶心。

Central Limit Theorem (CLT) – ample mean is approximately normal, regardless of the population’s shape. Key magic words: 'Sample mean → Normal when n large.'

中心极限定理 (CLT) – 大样本下，样本均值的分布近似正态，无论总体分布形状。口诀：“不管啥总体，大样均归一”。

---

8. Confidence Intervals | 置信区间

Confidence Interval – an interval estimate for a population parameter, with a stated confidence level (e.g., 95%). Formula: estimate ± margin of error. Margin of error = critical value × standard error.

置信区间 – 对总体参数的区间估计，附有置信水平（如 95%）。公式：点估计 ± 误差边际。误差边际 = 临界值 × 标准误。

Standard Error (SE) – standard deviation of a sampling distribution. For sample mean, SE = σ/√n when σ known, or s/√n when using sample standard deviation. 'Standard Error' tells you how much the estimate might 'err' on average.

标准误 (SE) – 抽样分布的标准差。样本均值的标准误：σ 已知时为 σ/√n，未知时用 s/√n。助记：标准误 = 估计值平均“误差”的尺度。

Width of Confidence Interval – wider with higher confidence level, smaller with larger sample size. Mnemonic: 'More confidence → wider net; more data → tighter aim.'

置信区间宽度 – 信心越高区间越宽，样本越大区间越窄。口诀：“越自信，网越宽；数据多，瞄得准”。

---

广告 / Sponsored

9. Hypothesis Testing | 假设检验

Null Hypothesis H₀ – the default assumption, often stating 'no effect' or 'no difference'. Think 'H₀ = H-nothing'.

原假设 H₀ – 默认假设，通常表述为“无效应”或“无差异”。记为“H₀ = 零假设”。

Alternative Hypothesis H₁ – the claim we seek evidence for. It can be one‑tailed (>, <) or two‑tailed (≠). 'H₁ is the one we hope to find.'

备择假设 H₁ – 我们希望寻找证据支持的断言。可分为单尾（>, <）或双尾（≠）。记作“H₁ 是想找到的那个”。

Significance Level α – the probability of rejecting H₀ when it is true (Type I error). Common values: 0.05, 0.01. 'α sets the bar for false alarm.'

显著性水平 α – 当 H₀ 为真时却拒绝 H₀ 的概率（第一类错误）。常用 0.05, 0.01。联想：α 是“误判的门槛”。

p-value – the probability, under H₀, of obtaining a test statistic as extreme as, or more extreme than, the observed one. 'p is low, reject H₀.'

p 值 – 在原假设下，观察到等于或比当前更极端的检验统计量的概率。判断口诀：“p 小就拒绝 H₀”。

Test Statistic – value calculated from sample data to decide about H₀. For a mean test with known σ: Z = (x̄ − μ₀)/(σ/√n). 'Test statistic transforms data to a standard scale.'

检验统计量 – 由样本数据算出的标准化值，用于判断 H₀。均值检验（σ 已知）：Z = (x̄ − μ₀)/(σ/√n)。用途：把数据映射到标准尺度上比较。

---

10. Errors and Power | 错误与功效

Type I Error – rejecting a true null. Probability = α. Recall: 'Type I = false positive, you see something that isn’t there.'

第一类错误 – 弃真错误，拒绝了一个真的原假设。概率为 α。速记：“一类错，假阳性，无而有”。

Type II Error – failing to reject a false null. Probability = β. 'Type II = false negative, you miss a real effect.'

第二类错误 – 存伪错误，未能拒绝一个假的原假设。概率为 β。速记：“二类错，假阴性，有却无”。

Power of a Test rrectly rejecting a false H₀. High power means the test is sensitive to detect the effect. 'Power = ability to detect truth.'

检验功效 = 1 − β，即正确拒绝假原假设的概率。功效高，试验能灵敏地发现效应。关键词：“功效即‘侦破真效果’的能力”。

---

11. Correlation and Regression | 相关与回归

Correlation Coefficient r – ationship between two variables. −1 ≤ r ≤ 1. r > 0: positive association; r < 0: negative. 'r is like a relationship score.'

相关系数 r – 衡量两变量间线性关系的强度和方向。−1 ≤ r ≤ 1。r > 0 为正相关，r < 0 为负相关。速记：r 就像“缘分指数”。

Product Moment Correlation Coefficient (PMCC) – same as r, the formal name for Pearson’s r. Suitable for linear relationships with bivariate normal data.

积矩相关系数 (PMCC) – 即 Pearson r，适用于二元正态分布的线性关系。记住“积矩”出自偏差乘积之和。

Regression Line (y = a + bx) – coefficient), a is the intercept. Least squares method minimises sum of squared residuals. 'Re' = again, 'gression' = stepping; regression steps back to the mean.

回归直线 (y = a + bx) – 用 x 预测 y。b 为斜率（回归系数），a 为截距。最小二乘法使残差平方和最小。词根：“回”归到均值的线。

Residual = observed y − predicted y. Residuals should show no pattern when plotted. 'Residual is the leftover after explaining.'

残差 = 观测值 − 预测值。残差图应呈现无模式随机分布。联想：残差就是回归“吃剩下”的部分。

---

12. Common Memory Aids and Quick-Fire Glossary | 通用速记法及术语快查

Acronyms and Rhymes: Use SOCS for describing distributions: Shape, Outliers, Centre, Spread. For Normal distribution bell: 'Mean fixes the centre, SD sets the width'. For hypothesis steps: PHANTOMS (Parameter, Hypotheses, Assumptions, Name the test, Test statistic, Obtain p‑value, Make decision, State conclusion).

首字母口诀与韵律： 描述分布用 SOCS：Shape（形状）, Outliers（异常值）, Centre（中心）, Spread（散度）。正态钟形：均值定中心，标准差定胖瘦。假设检验步骤：PHANTOMS（参数、假设、假定、检验名、检验统计量、得 p 值、决断、陈述结论）。

Word Origins: 'Binomial' from bi- (two) + nomial (names) → two-category trials. 'Poisson' named after mathematician Siméon Poisson; think 'Poisson process = random arrival process'. 'Regression' stems from 'regression to the mean'. Linking terms to their origins cements memory.

词源法： “Binomial” 由 bi-（二）+ nomial（项）→ 二项试验。“Poisson” 本为数学家泊松之名，可联想“泊松过程 = 随机抵达过程”。“Regression” 源于“回归到均值”。追根溯源，印象更牢。

The table below lists essential Cambridge Statistics terms with a one‑line memory hook. Keep it as a rapid revision card.

下表列出剑桥统计必背术语及一行速记口诀，可作快速复习卡片。

Term 术语	中文	Memory Hook 速记窍门
Mean	均值	Balance point, 'mean' between extremes
Median	中位数	Middle of the sorted row
Variance σ²	方差	Squared spread around the mean
Standard Deviation σ	标准差	Typical distance from mean; sqrt of variance
Sample Space S	样本空间	All possible outcomes under the sun
P(A ∪ B)	并事件概率	Add then subtract overlap (Venn diagram)
P(A|B)	条件概率	Gate '|' – given B, what's A?
Binomial B(n, p)	二项分布	Bi = two outcomes, n trials, p success prob.
Poisson Po(λ)	泊松分布	Random arrivals at constant rate λ
Normal N(μ, σ²)	正态分布	Bell shape, mean μ, variance σ²
Z-score	标准分	Z = (X − μ)/σ, zero-centred bridge to tables
CLT	中心极限定理	Sample mean ≈ Normal for large n
Confidence Interval	置信区间	Estimate ± critical × SE; wider = more confident
p-value	p 值	p low → reject H₀
Type I error	第一类错误	False positive – reject true H₀
Type II error	第二类错误	False negative – miss false H₀
Correlation r	相关系数	−1 to 1, strength and direction of line
Residual	残差	Observed − predicted, leftover scatter

Regularly reviewing these terms with their memory hooks will turn vocabulary into an automatic advantage in the exam. Pair them with past‑paper practice for best effect.

定期用这些速记口诀复习术语，能让你在考场上条件反射般地调用概念。搭配历年真题练习，效果最佳。

Published by TutorHao | Statistics Revision Series | aleveler.com

更多咨询请联系16621398022（同微信）