📚 Pre-U CAIE Statistics: Vocabulary Mnemonics Guide | Pre-U CAIE 统计：词汇术语速记指南
Mastering the precise technical vocabulary of Pre-U CAIE Statistics is half the battle. This guide breaks down essential terms into logical groups and supplies memorable mnemonics and clear explanations to help you internalise the language of data, chance, and inference. Whether you are grappling with distribution names, hypothesis testing jargon, or the subtleties of sampling, leaning on structured word associations will accelerate your revision and boost your confidence in exam answers.
掌握 Pre-U CAIE 统计中精确的专业词汇是成功的一半。本指南将核心术语按逻辑分组，并提供好记的口诀和清晰的解释，帮助你真正内化数据、概率和推断的语言。无论你是在为分布名称、假设检验术语还是抽样的微妙之处而苦恼，借助结构化的词语联想都能加速你的复习进度，并让你在考场上更加自信。
1. Data Types and Measurement Scales | 数据类型与度量尺度
Qualitative data describe qualities or categories that cannot be measured numerically, often called categorical data. Quantitative data come from measurements or counts and are numerical. A key split within quantitative data is discrete versus continuous: discrete data can only take certain isolated values (usually produced by counting), while continuous data can take any value within a range (typically obtained by measuring).
定性数据描述的是无法用数字衡量的性质或类别，通常称作分类数据。定量数据来自测量或计数，是数值型的数据。定量数据内部一个关键区分是离散与连续：离散数据只能取某些孤立的数值（通常由计数产生），而连续数据可以取一个范围内的任意值（通常通过测量获得）。
Mnemonic: ‘Quantitative’ contains ‘quant-‘ hinting at quantity. ‘Discrete’ sounds like ‘discreet’ – think separate, countable items. ‘Continuous’ starts with ‘continue’ – values flow without gaps, like time.
记忆窍门：’Quantitative’ 含有 ‘quant-‘, 暗示数量；’Discrete’ 与 discreet 发音相似，想象成分开的可数个体；’Continuous’ 以 continue 开头，数值像时间一样无间断流动。
Measurement scales add another layer: nominal data are named categories with no order (e.g. eye colour), ordinal data have a clear order but unknown gaps between ranks, interval data have equal gaps but no true zero (e.g. temperature in °C), and ratio data have a meaningful zero, allowing ratios (e.g. height, mass).
度量尺度又增添了一个层次：名义数据是仅有命名而无顺序的类别（如眼睛颜色）；顺序数据有明确排序但等级之间差距未知；间隔数据间距相等但没有真正的零点（如摄氏温度）；比率数据有有意义的零点，可以计算倍数（如身高、质量）。
2. Measures of Central Tendency | 集中趋势的度量
The mean (x̄ for a sample, μ for a population) is the arithmetic average, calculated as Σxᵢ/n. It uses every observation but is sensitive to outliers. The median is the middle value after ordering data; it is resistant to outliers. The mode is the most frequently occurring value or class.
均值（样本用 x̄，总体用 μ）是算术平均数，计算公式为 Σxᵢ/n。它使用所有观测值但对异常值敏感。中位数是排序后位于中间的值，能抵抗异常值。众数是出现频率最高的值或组。
Quick recall: ‘Mean is the average, median is the middle, mode is the most.’ For symmetrical unimodal data, mean = median = mode. Skewness pulls the mean away from the median.
快速回忆口诀：’Mean 是平均数，Median 是正中间，Mode 是出现最多的。’ 对于对称单峰数据，均值=中位数=众数。偏态会将均值拉离中位数。
When data are grouped, we estimate the mean using midpoints and the modal class is the class with highest frequency. The median is found by linear interpolation: median = L + ( (n/2 − F) / f ) × w, where L is the lower boundary, F is cumulative frequency before the median class, f is the frequency of the median class, and w is the class width.
数据分组时，我们使用组中值估计均值，众数组是频数最高的组。中位数通过线性插值求得：中位数 = L + ( (n/2 − F) / f ) × w，其中 L 为下边界，F 为中位数组之前的累积频数，f 为中位数组的频数，w 为组距。
3. Measures of Dispersion | 离散程度的度量
Range = maximum − minimum; it is simple but affected by extremes. Interquartile range (IQR) = Q₃ − Q₁, the spread of the middle 50% of data, robust to outliers. Variance and standard deviation quantify average squared deviation from the mean. For a sample, variance s² = Σ(xᵢ − x̄)²/(n−1); the denominator n−1, called Bessel’s correction, ensures an unbiased estimate of the population variance.
极差 = 最大值 − 最小值，虽然简单但受极端值影响。四分位距 IQR = Q₃ − Q₁，是中间 50% 数据的散布范围，能抵抗异常值。方差和标准差衡量的是各数据与均值偏差平方的平均水平。样本方差 s² = Σ(xᵢ − x̄)²/(n−1)；分母使用 n−1（贝塞尔校正）以保证对总体方差的无偏估计。
Mnemonic: IQR gives a ‘robust middle spread’. Variance is ‘variability squared’; standard deviation is the square root, returning to original units. Remember ‘standard deviation = √variance’.
记忆点：IQR 提供稳健的中间散布。方差是 ‘差异的平方’，标准差是方差的平方根，使单位恢复原状。记住 ‘标准差 = √方差’。
Standard deviation is the most widely used dispersion measure for symmetric distributions. The formula σ = √( Σ(xᵢ − μ)² / N ) for a population and s = √( Σ(xᵢ − x̄)²/(n−1) ) for a sample. In exam answers, always state which measure and why, e.g. ‘use IQR because of outliers’.
对于对称分布，标准差是最常用的离散指标。总体标准差 σ = √( Σ(xᵢ − μ)² / N )，样本标准差 s = √( Σ(xᵢ − x̄)²/(n−1) )。答题时务必说明选用的度量及理由，例如 ‘因为有异常值，故用 IQR’。
4. Foundational Probability Terms | 概率基础术语
A sample space S is the set of all possible outcomes. An event A is a subset of S. The complement of A, denoted A’ or Aᶜ, is everything not in A: P(A’) = 1 − P(A). The union A ∪ B occurs if either A or B or both happen. The intersection A ∩ B occurs only if both A and B happen simultaneously.
样本空间 S 是所有可能结果的集合。事件 A 是 S 的子集。A 的补集 A’ 或 Aᶜ 是 S 中所有不在 A 中的结果：P(A’) = 1 − P(A)。并集 A ∪ B 表示 A 或 B 至少有一个发生。交集 A ∩ B 表示 A 与 B 同时发生。
Mutually exclusive events cannot happen together: A ∩ B = ∅, so P(A ∪ B) = P(A) + P(B). Independent events satisfy P(A ∩ B) = P(A) × P(B), or equivalently P(A|B) = P(A). Do not confuse ‘mutually exclusive’ with ‘independent’ – the former forbids simultaneous occurrence, the latter means one event’s occurrence does not affect the probability of the other.
互斥事件不可能同时发生：A ∩ B = ∅，因此 P(A ∪ B) = P(A) + P(B)。独立事件满足 P(A ∩ B) = P(A) × P(B)，或等价地 P(A|B) = P(A)。切勿混淆 ‘互斥’ 与 ‘独立’——前者禁止同时发生，后者表示一个事件的发生不影响另一个事件的概率。
Conditional probability is written P(A|B) and read ‘probability of A given B’. Formula: P(A|B) = P(A ∩ B) / P(B), provided P(B) > 0. It underpins tree diagrams, Bayes’ theorem, and contingency table analysis.
条件概率写作 P(A|B)，读作 ‘在 B 发生的条件下 A 的概率’。公式为 P(A|B) = P(A ∩ B) / P(B)，前提是 P(B) > 0。它是树状图、贝叶斯定理和列联表分析的基础。
5. Discrete Probability Distributions | 离散概率分布
A probability distribution lists all possible outcomes of a discrete random variable X and their probabilities, with Σ P(X = x) = 1. Expectation E(X) = Σ x · P(X = x) gives the long-run average. Variance Var(X) = E(X²) − [E(X)]².
概率分布列出了离散随机变量 X 的所有可能取值及其概率，且满足 Σ P(X = x) = 1。期望 E(X) = Σ x · P(X = x) 代表长期平均值。方差 Var(X) = E(X²) − [E(X)]²。
Binomial distribution B(n, p): models the number of successes in n independent trials, each with success probability p. Conditions summarised by the mnemonic BINS: Binary outcome, Independent trials, Number of trials fixed, Same probability p for each trial. If X ~ B(n, p), then P(X = r) = ⁿCᵣ pʳ (1 − p)ⁿ⁻ʳ, E(X) = np, Var(X) = np(1−p).
二项分布 B(n, p)：模拟 n 次独立试验中成功的次数，每次成功概率为 p。条件用 BINS 助记：二元结果、试验独立、试验次数固定、每次概率相同 p。若 X ~ B(n, p)，则 P(X = r) = ⁿCᵣ pʳ (1 − p)ⁿ⁻ʳ，E(X) = np，Var(X) = np(1−p)。
Poisson distribution Po(λ): models the number of randomly occurring events in a fixed interval of time or space, with mean rate λ. Conditions: events occur singly, randomly, independently, and at a constant average rate. If X ~ Po(λ), then P(X = r) = e⁻λ λʳ / r!, E(X) = Var(X) = λ. Poisson can approximate Binomial when n is large and p is small, setting λ = np.
泊松分布 Po(λ)：模拟固定时间或空间内随机发生的事件个数，平均发生率为 λ。条件：事件单个出现、随机、独立、且平均率恒定。若 X ~ Po(λ)，则 P(X = r) = e⁻λ λʳ / r!，E(X) = Var(X) = λ。当 n 很大、p 很小时，可用 λ = np 的泊松分布近似二项分布。
6. Continuous Probability Distributions | 连续概率分布
For a continuous random variable, probability is represented by area under a probability density function (pdf). The total area equals 1, and P(X = a) = 0 for any single point. We compute probabilities over intervals: P(a < X < b) = ∫ₐᵇ f(x) dx.
对于连续随机变量，概率由概率密度函数 (pdf) 下的面积表示。总面积为 1，任何单点的概率 P(X = a) = 0。我们计算区间概率：P(a < X < b) = ∫ₐᵇ f(x) dx。
The Normal distribution N(μ, σ²) is the cornerstone of continuous models. It is bell-shaped, symmetric about μ, with points of inflection at μ ± σ. Approximately 68% of data lie within μ ± σ, 95% within μ ± 2σ, and 99.7% within μ ± 3σ. These are the empirical rule figures.
正态分布 N(μ, σ²) 是连续模型的基石。图形呈钟形，关于 μ 对称，拐点位于 μ ± σ。约 68% 的数据落在 μ ± σ 内，95% 落在 μ ± 2σ 内，99.7% 落在 μ ± 3σ 内。这就是经验法则的数据。
Standardisation converts any Normal variable X ~ N(μ, σ²) to Z ~ N(0, 1) using Z = (X − μ) / σ. The standard Normal table then gives cumulative probabilities. Key mnemonic: Z tells you ‘how many standard deviations X is away from the mean’. In applications, check whether you need to apply a continuity correction when approximating a discrete distribution with a Normal distribution.
标准化将任意正态变量 X ~ N(μ, σ²) 转化为 Z ~ N(0, 1)，公式为 Z = (X − μ) / σ。借助标准正态表可得累积概率。核心口诀：Z 值告诉 ‘X 距离均值有多少个标准差’。应用中，若用正态分布近似离散分布，需检查是否需要连续性校正。
The Normal approximation to binomial requires np > 5 and nq > 5, setting μ = np, σ = √(npq). The continuity correction adjusts a discrete value by ±0.5 to improve accuracy.
正态近似二项分布需满足 np > 5 和 nq > 5，设 μ = np, σ = √(npq)。连续性校正将离散值 ±0.5 进行调整以提高精度。
7. Sampling and the Central Limit Theorem | 抽样与中心极限定理
A population consists of all items of interest; a sample is a subset selected for study. A parameter is a numerical summary of a population (e.g. μ, σ), while a statistic is a numerical summary of a sample (e.g. x̄, s). The sampling distribution of a statistic describes how the statistic varies from sample to sample.
总体由所有关注对象构成，而样本是其中选出用于研究的一部分。参数是总体的数值概括（如 μ, σ），统计量是样本的数值概括（如 x̄, s）。统计量的抽样分布描述该统计量在不同样本间如何变化。
An estimator is unbiased if its expected value equals the true parameter, e.g. E(x̄) = μ and E(s²) = σ². The Central Limit Theorem (CLT) states that for a sufficiently large sample size (typically n ≥ 30), the sampling distribution of the sample mean x̄ is approximately Normal, regardless of the shape of the population distribution, with mean μ and standard deviation σ/√n (the standard error).
若估计量的期望值等于参数真值，则称该估计量无偏，如 E(x̄) = μ，E(s²) = σ²。中心极限定理 (CLT) 指出，当样本量足够大（通常 n ≥ 30），无论总体分布形状如何，样本均值 x̄ 的抽样分布近似正态，均值为 μ，标准差为 σ/√n（称为标准误差）。
Standard error is a key term: the standard deviation of a sampling distribution. For the sample mean, SE = σ/√n. If σ is unknown, we estimate it using s, giving estimated standard error s/√n. This concept feeds into confidence intervals and hypothesis tests.
标准误差是一个关键术语：指抽样分布的标准差。对于样本均值，SE = σ/√n。若 σ 未知，则用 s 估计，得估计标准误差 s/√n。这概念延伸至置信区间与假设检验。
8. Hypothesis Testing Lexicon | 假设检验词汇
The null hypothesis H₀ is a statement of no effect or no difference, assumed true until evidence suggests otherwise. The alternative hypothesis H₁ or Hₐ is what we hope to prove, often one-tailed (directional) or two-tailed. The significance level α is the probability of rejecting H₀ when it is actually true (a Type I error).
原假设 H₀ 是关于无效应或无差别的陈述，在证据表明确实之前被假定为真。备择假设 H₁ 或 Hₐ 是我们希望证实的陈述，常为单侧（有方向）或双侧。显著性水平 α 是当 H₀ 为真时却拒绝 H₀ 的概率，即犯第一类错误的概率。
Type I error: rejecting a true H₀ (probability α). Type II error: failing to reject a false H₀ (probability β). The power of a test is 1 − β, the probability of correctly rejecting a false H₀. Simple mnemonic: Type I is a ‘false alarm’, Type II is a ‘missed detection’.
第一类错误：拒绝正确的 H₀（概率 α）。第二类错误：未能拒绝错误的 H₀（概率 β）。检验的功效为 1 − β，即正确拒绝错误 H₀ 的概率。简明记忆法：第一类错误是’误报警’，第二类错误是’漏报’。
The p-value is the probability, assuming H₀ is true, of obtaining test results at least as extreme as the observed. If p-value ≤ α, we reject H₀; otherwise, we do not reject H₀. Do not say ‘accept H₀’, only ‘do not reject’.
p 值是在 H₀ 为真的情况下，获得至少与观测结果一样极端的检验结果的概率。若 p 值 ≤ α，拒绝 H₀；否则不拒绝 H₀。回答时不要说 ‘接受 H₀’，只能说 ‘不拒绝 H₀’。
Test statistic: a standardised value computed from sample data (e.g. z = (x̄ − μ₀)/(σ/√n)). The critical region (rejection region) is the set of test statistic values that lead to rejecting H₀. The critical value separates the critical region from the acceptance region.
检验统计量：由样本数据计算出的标准化值，例如 z = (x̄ − μ₀)/(σ/√n)。拒绝域是导致拒绝 H₀ 的检验统计量取值的集合。临界值将拒绝域与接受域分开。
9. Correlation and Regression | 相关与回归
Correlation measures the strength and direction of a linear relationship between two variables. Pearson’s product-moment correlation coefficient r ranges from −1 to +1. r = 1 indicates perfect positive linear correlation; r = −1 perfect negative linear correlation; r = 0 suggests no linear correlation. Spearman’s rank correlation coefficient ρ (or rₛ) assesses monotonic relationships using ranks, useful when data are ordinal or non-normal.
相关衡量两个变量之间线性关系的强度与方向。皮尔逊积矩相关系数 r 的取值范围从 −1 到 +1。r = 1 表示完全正线性相关，r = −1 表示完全负线性相关，r = 0 提示无线性相关。斯皮尔曼等级相关系数 ρ（或 rₛ）利用秩次评估单调关系，适合顺序型数据或数据不正态的情况。
Mnemonic: ‘r is fiercely linear; Spearman cares about order, not shape.’ Also remember that correlation does not imply causation.
记忆点：’r 只在乎线性；斯皮尔曼在意秩次而非图形形状。’ 还需牢记相关不意味着因果。
Regression analysis models the relationship by fitting a line, often using least squares regression. The regression equation is y = a + bx, where b = Sxy / Sxx and a = ȳ − b x̄. The coefficient b gives the change in y for a unit increase in x. Residuals are the differences between observed y-values and those predicted by the line: e = y − ŷ. The residual sum of squares is minimised by the least squares line.
回归分析通过拟合一条线来建模关系，常用最小二乘法回归。回归方程为 y = a + bx，其中 b = Sxy / Sxx，a = ȳ − b x̄。系数 b 表示 x 每增加一个单位时 y 的变化量。残差是观测 y 值与直线预测值之间的差值：e = y − ŷ。最小二乘直线使残差平方和最小。
Extrapolation means using the regression line outside the range of the original data; it is unreliable. Interpolation within the data range is safer. Always comment on the validity of predictions.
外推指在原始数据范围之外使用回归线进行预测，这并不可靠。内推在数据范围内进行预测则更安全。作答时一定要对预测的有效性加以评论。
10. Summary of Mnemonic Devices | 速记技巧汇总
Consolidate the vocabulary with compact anchors. For Binomial conditions, recall ‘BINS’. For measures of location, ‘Mean, Median, Mode: 3Ms’. For dispersion: ‘IQR cuts off extremes; standard deviation measures typical distance from the mean’. For probability: ‘Mutually Exclusive cannot coexist (empty intersection); Independent means no influence’. For distributions: ‘Binomial is discrete counts with n and p; Poisson models rare events with constant rate; Normal is the bell-shaped continuous workhorse’. For testing: ‘If p is low, H₀ must go (p ≤ α → reject)’. ‘Type I is a false positive, Type II a false negative’. For regression: ‘LSRL minimises the vertical squared gaps’. Linking terms to vivid mental images helps recall under exam pressure.
用简明的线索把词汇串联起来。二项分布条件记 ‘BINS’。位置度量记 ‘三个 M：Mean, Median, Mode’。离散程度：’IQR 切除极端值；标准差度量到均值的典型距离’。概率：’互斥不能共存（交集为空）；独立则表示互不影响’。分布：’二项是离散计数，带 n 和 p；泊松模拟恒定速率的稀有事件；正态是钟形连续的主力’。检验：’若 p 值低，H₀ 就弃（p ≤ α → 拒绝）’。’第一类错误是假阳性，第二类错误是假阴性’。回归：’LSRL 使纵向平方差最小’。将术语与生动的心理图像结合，有助于在考试压力下回忆。
Another effective trick is to create contrasting pairs: parameter vs statistic, sample vs population, discrete vs continuous, null vs alternative, Type I vs Type II. Writing your own glossary with definitions, symbols, and conditions is a proven revision method. Regularly test yourself by describing a term in plain English before checking the formal definition.
另一个有效策略是制作对照对子：参数与统计量，样本与总体，离散与连续，原假设与备择假设，第一类错误与第二类错误。编写带有定义、符号和条件的个人术语表是经过验证的复习方法。定期自测，先用通俗语言描述一个术语，再对照正式定义检查。
Keep this guide close as you practise past papers, and soon the specialised vocabulary will become second nature.
在练习历年真题时随身携带本指南，这些专业词汇很快就会变得得心应手。
Published by TutorHao | Statistics Revision Series | aleveler.com
更多咨询请联系16621398022（同微信）
屏轩国际教育cambridge primary/secondary checkpoint, cat4, ukiset,ukcat,igcse,alevel,PAT,STEP,MAT, ibdp,ap,ssat,sat,sat2课程辅导，国外大学本科硕士研究生博士课程论文辅导