Pre-U CCEA Statistics: Quick Terminology Memorisation Guide | Pre-U CCEA 统计：词汇术语速记指南

📚 Pre-U CCEA Statistics: Quick Terminology Memorisation Guide | Pre-U CCEA 统计：词汇术语速记指南

Mastering statistical terminology is half the battle in Pre-U CCEA Statistics. This guide breaks down the essential vocabulary into logical clusters, pairing each term with a visual or contextual memory hook to make revision faster and more effective. Whether you are preparing for a class test or the final examination, these mnemonics and structured explanations will help you recall definitions accurately and apply them confidently.

在 Pre-U CCEA 统计考试中，掌握术语词汇等于赢了一半。本指南将核心词汇划分成逻辑清晰的记忆群组，为每个术语配上视觉化或情境化的记忆钩子，让你的复习更快、更有效。无论是准备课堂测验还是最终大考，这些助记法和结构化解释都能帮助你准确回忆定义，自信地运用它们。

1. Populations, Samples and Sampling Frames | 总体、样本与抽样框

The population is the entire set of individuals or items that we want to draw conclusions about. Think of it as the ‘whole universe’ of your study. A sample is a subset of the population, selected to represent it. A sampling frame is the actual list of units from which the sample is drawn, like a phone book or a school register. The list may not match the population perfectly, which leads to undercoverage or overcoverage errors.

总体是我们想要得出结论的全部个体或项目的集合。可以把它想象成你研究的”整个宇宙”。样本是从总体中选出的一个子集，用来代表总体。抽样框是实际用于抽取样本的单元列表，比如电话簿或学校注册表。这份清单可能与总体不完全吻合，从而导致覆盖不足或覆盖过度的误差。

A census attempts to collect data from every member of the population. In contrast, a sample survey collects data from only a part of the population. The distinction is fundamental: a census is exhaustive but often impractical, while a sample is more manageable but introduces sampling error.

普查试图从总体中的每个成员收集数据。相比之下，抽样调查只从总体的一部分收集数据。这一区别是根本性的：普查很全面但往往不切实际，而样本更易于操作，却会引入抽样误差。

2. Variables: Quantitative vs. Qualitative | 变量：数值型与分类型

A variable is any characteristic that can take on different values. Quantitative variables are numerical and can be discrete (countable, like number of pets) or continuous (measurable, like height). Imagine discrete as ‘steps on a staircase’ and continuous as ‘points on a line’. Qualitative (categorical) variables describe qualities or groups, like eye colour or type of car.

变量是可以取不同值的任何特征。数值型变量是数字化的，可以是离散的（可数的，如宠物数量）或连续的（可测量的，如身高）。可以把离散型想象成”楼梯上的台阶”，连续型想象成”直线上的点”。分类型（定性）变量描述的是品质或类别，比如眼睛颜色或汽车型号。

Nominal variables are categories with no natural order (e.g., blood type A, B, AB, O). Ordinal variables have a meaningful order but no fixed intervals between ranks (e.g., satisfaction rating: very dissatisfied, dissatisfied, neutral, satisfied, very satisfied). The mnemonic ‘N-O Difference’: Nominal = No Order; Ordinal = Ordered.

名义变量是没有自然顺序的类别（例如血型 A、B、AB、O）。有序变量有有意义的顺序，但等级之间的间隔不固定（例如满意度评分：非常不满意、不满意、一般、满意、非常满意）。助记：”N-O差异”：Nominal = 无顺序；Ordinal = 有顺序。

3. Measures of Central Tendency: Mean, Median, Mode | 集中趋势量度：平均数、中位数、众数

The mean (x̄ for a sample, μ for a population) is the arithmetic average, calculated by summing all values and dividing by the count. It is sensitive to outliers – think of it as the ‘balancing point’ that shifts when a huge value appears. The median is the middle value when data are ordered; it resists outliers, like a sturdy centre pillar. The mode is the most frequently occurring value, useful for categorical data.

平均数（样本记为 x̄，总体记为 μ）是算术平均值，将所有数值加总再除以数据个数来计算。它对异常值敏感——可以把它看作”平衡点”，当出现一个极大值时就会移动。中位数是数据排序后的中间值；它能抵抗异常值的影响，就像一根坚固的中心柱。众数是出现频率最高的值，对分类数据很有用。

For a skewed distribution, the mean is pulled towards the tail. A quick memory aid: ‘The mean follows the skew’. In a positively skewed distribution (tail to the right), mean > median > mode. In a negatively skewed distribution, mean < median < mode. Symmetrical data have mean = median = mode.

对于偏态分布，平均数会被拉向尾部。一个快速记忆口诀：”平均数跟着偏斜走”。在正偏态分布（尾部在右）中，平均数 > 中位数 > 众数。在负偏态分布中，平均数 < 中位数 < 众数。对称分布的平均数、中位数和众数相等。

4. Measures of Dispersion: Range, IQR, Variance and Standard Deviation | 离散趋势量度：极差、四分位距、方差和标准差

The range (maximum minus minimum) is the simplest measure of spread, but it is heavily influenced by a single extreme value. The interquartile range (IQR = Q₃ – Q₁) measures the spread of the middle 50% of data, making it robust to outliers. Think of the range as looking at the whole fence, while IQR only cares about the house in the middle.

极差（最大值减最小值）是最简单的离散量度，但它极易受单个极端值的影响。四分位距（IQR = Q₃ – Q₁）度量的则是中间 50% 数据的散布程度，因此对异常值不敏感。可以把极差想象成观察整条篱笆，而 IQR 只关心中间的房子。

Variance is the average of the squared deviations from the mean. Its square root, the standard deviation (s or σ), restores the original units. A small standard deviation means data are tightly clustered around the mean; a large one indicates wide spread. The formula for sample variance, s² = Σ(x – x̄)² / (n – 1), uses n – 1 to get an unbiased estimate of the population variance – this is known as Bessel’s correction.

方差是离均差平方的平均值。其平方根，即标准差（s 或 σ），恢复了原单位。标准差小意味着数据紧密聚集在平均数周围；标准差大则表明数据分布很分散。样本方差的公式 s² = Σ(x – x̄)² / (n – 1) 使用 n – 1 以得到总体方差的无偏估计——这就是著名的贝塞尔校正。

5. Probability Basics and Venn Diagrams | 概率基础与韦恩图

Probability (P) quantifies the likelihood of an event, ranging from 0 (impossible) to 1 (certain). The sample space (S) is the set of all possible outcomes. For equally likely outcomes, P(A) = number of ways A can occur / total number of outcomes. Visualising with a Venn diagram helps: each circle represents an event, and the rectangle is the sample space.

概率 (P) 衡量事件发生的可能性，范围从 0（不可能）到 1（必然）。样本空间 (S) 是所有可能结果的集合。对于等可能的结果，P(A) = A 发生的方式数 / 总结果数。用韦恩图可视化很有帮助：每个圆圈代表一个事件，矩形则是样本空间。

The complement of A, denoted A’ or Aᶜ, is the event ‘A does not happen’, so P(A’) = 1 – P(A). Intersection (A ∩ B) means both A and B occur; union (A ∪ B) means A or B or both occur. Mutually exclusive events cannot happen at the same time: P(A ∩ B) = 0. Independent events have no influence on each other: P(A ∩ B) = P(A)P(B).

A 的补集，记作 A’ 或 Aᶜ，是” A 不发生”的事件，因此 P(A’) = 1 – P(A)。交集 (A ∩ B) 表示 A 和 B 同时发生；并集 (A ∪ B) 表示 A 或 B 或两者都发生。互斥事件不能同时发生：P(A ∩ B) = 0。独立事件彼此互不影响：P(A ∩ B) = P(A)P(B)。

6. Discrete Random Variables and Expectation | 离散随机变量与期望

A random variable (X) assigns a numerical value to each outcome in a sample space. For a discrete random variable, we can list all possible values and their probabilities in a probability distribution table. The sum of all probabilities must equal 1. The expected value E(X) is the long-run average, calculated as Σ [x · P(X=x)]. Think of it as the ‘balance point’ of the probability mass.

随机变量 (X) 为样本空间中的每个结果分配一个数值。对于离散随机变量，我们可以在概率分布表中列出所有可能的取值及其概率。所有概率之和必须等于 1。期望值 E(X) 是长期平均值，通过 Σ [x · P(X=x)] 计算得出。可以把它看作概率质量的 “平衡点”。

The variance of X, Var(X) = E(X²) – [E(X)]², measures the spread of the distribution. The standard deviation is the square root of Var(X). Pay attention to the units: if X is in cm, variance is in cm², so standard deviation brings you back to cm. For linear transformations, E(aX + b) = aE(X) + b, and Var(aX + b) = a²Var(X).

X 的方差 Var(X) = E(X²) – [E(X)]²，度量分布的离散程度。标准差是 Var(X) 的平方根。注意单位：如果 X 的单位是 cm，方差的单位就是 cm²，而标准差又回到了 cm。对于线性变换，E(aX + b) = aE(X) + b，且 Var(aX + b) = a²Var(X)。

7. Binomial and Poisson Distributions | 二项分布与泊松分布

The binomial distribution models the number of successes 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). A classic scenario: tossing a coin n times, counting heads.

二项分布用于描述在固定次数的独立试验中成功的次数，每次试验的成功概率 p 相同。如果 X ~ B(n, p)，则 P(X = r) = ⁿCᵣ pʳ (1–p)ⁿ⁻ʳ。均值为 np，方差为 np(1–p)。经典情景：抛硬币 n 次，数正面朝上的次数。

The Poisson distribution models the number of events occurring in a fixed interval of time or space, given a known average rate λ. If X ~ Po(λ), P(X = r) = (e⁻λ λʳ) / r!. Here the mean = variance = λ. Think of random, rare events: the number of calls to a call centre per hour, or the number of particles detected per second.

泊松分布用于描述在固定的时间或空间区间内，事件发生的次数，其已知平均发生率为 λ。如果 X ~ Po(λ)，P(X = r) = (e⁻λ λʳ) / r!。这里的均值等于方差，都等于 λ。可以联想随机、稀有的事件：每小时打入呼叫中心的电话数，或每秒检测到的粒子数。

A binomial distribution where n is large and p is small can be approximated by a Poisson with λ = np. The approximation works well if n > 50 and np < 5, or n > 20 and p < 0.1. Remember: ‘Large n, small p, let Poisson set you free.’

当 n 很大而 p 很小时，二项分布可以用泊松分布来近似，其中 λ = np。如果 n > 50 且 np < 5，或 n > 20 且 p < 0.1，这个近似效果很好。记住：”大 n 小 p，泊松解放你。”

8. Continuous Distributions: Normal and Standard Normal | 连续分布：正态与标准正态

The normal distribution is a symmetric, bell-shaped curve defined by two parameters: mean μ and standard deviation σ. About 68% of data lie within μ ± 1σ, 95% within μ ± 2σ, and 99.7% within μ ± 3σ – this is the empirical rule. Notation X ~ N(μ, σ²) highlights that the variance is σ², not σ. Many natural measurements like heights and weights are approximately normal.

正态分布是一种对称的钟形曲线，由两个参数定义：均值 μ 和标准差 σ。大约 68% 的数据落在 μ ± 1σ 内，95% 落在 μ ± 2σ 内，99.7% 落在 μ ± 3σ 内——这就是经验法则。记号 X ~ N(μ, σ²) 强调方差是 σ² 而不是 σ。许多自然的测量值如身高和体重都近似服从正态分布。

The standard normal distribution Z ~ N(0, 1) has mean 0 and standard deviation 1. Any normal variable X can be transformed to Z by standardising: Z = (X – μ) / σ. This lets us use standard normal tables to find probabilities. The trick ‘Z-score = distance from mean in standard deviations’ helps to remember the formula.

标准正态分布 Z ~ N(0, 1) 的均值为 0，标准差为 1。任何正态变量 X 都可以通过标准化转换为 Z：Z = (X – μ) / σ。这样我们就可以查标准正态分布表来求概率。”Z-分数 = 离均值有多少个标准差”这个口诀有助于记住这个公式。

9. Correlation and Regression | 相关与回归

Correlation measures the strength and direction of a linear relationship between two variables. The product moment correlation coefficient, r, ranges from –1 (perfect negative) to +1 (perfect positive). Note that r = 0 indicates no linear relationship, but there could be a strong non-linear pattern. Always plot the data first; correlation does not imply causation.

相关度量两个变量之间线性关系的强度和方向。积矩相关系数 r 的范围从 –1（完全负相关）到 +1（完全正相关）。注意，r = 0 表示没有线性关系，但可能存在很强的非线性关系。始终要先画散点图；相关不意味着因果。

Linear regression finds the line of best fit, y = a + bx, that minimises the sum of squared residuals. The constants are calculated using the least squares method: b = Sxy / Sxx, and a = ȳ – b x̄. The regression line passes through the point of means (x̄, ȳ). Use the line to predict y for a given x (interpolation is safer than extrapolation).

线性回归寻找最佳拟合直线 y = a + bx，使得残差平方和最小。常数通过最小二乘法计算：b = Sxy / Sxx，而 a = ȳ – b x̄。回归直线经过均值点 (x̄, ȳ)。可以用这条线根据给定的 x 预测 y（内插比外推更安全）。

10. Sampling Distributions and the Central Limit Theorem | 抽样分布与中心极限定理

A sampling distribution is the probability distribution of a statistic (like the sample mean) taken from repeated samples of the same size. It has its own center and spread. The standard error (SE) is the standard deviation of that sampling distribution. For the sample mean, SE = σ / √n. As sample size increases, the sample means become less variable – the precision improves.

抽样分布是指在相同样本容量下重复抽样，从样本中计算出的某个统计量（如样本均值）的概率分布。它有自身的中心和离散。标准误 (SE) 就是该抽样分布的标准差。对于样本均值，SE = σ / √n。随着样本容量增加，样本均值的变异性减小——精度得到改善。

The Central Limit Theorem (CLT) states that for a sufficiently large sample size (n ≥ 30 is a common rule of thumb), the sampling distribution of the mean will be approximately normal, regardless of the shape of the original population. This is the single most powerful tool in inference, because it allows us to use normal probability methods even when the underlying data are not normal.

中心极限定理 (CLT) 指出，对于足够大的样本容量（常用的经验法则是 n ≥ 30），样本均值的抽样分布将近似服从正态分布，无论原始总体的形状如何。这是推断统计中最强大的工具，因为它让我们即使面对非正态的原始数据也能使用正态概率方法。

11. Confidence Intervals and Hypothesis Testing | 置信区间与假设检验

A confidence interval gives a range of plausible values for an unknown population parameter. A 95% confidence interval for the mean μ is typically x̄ ± z* × σ/√n (if σ is known) or x̄ ± t* × s/√n (if σ is estimated). Interpretation: if we took many samples, 95% of the intervals constructed would capture the true μ. It does not say there is a 95% chance that μ lies in this particular interval.

置信区间给出了未知总体参数的一个合理取值范围。均值 μ 的 95% 置信区间通常为 x̄ ± z* × σ/√n（若 σ 已知）或 x̄ ± t* × s/√n（若 σ 需要估计）。解读方法：如果我们抽取许多样本，所构建的区间中有 95% 会包含真实的 μ。这并不是说 μ 落在这个特定区间内的概率是 95%。

Hypothesis testing starts with a null hypothesis (H₀) representing the status quo, and an alternative hypothesis (H₁). The p-value is the probability of observing a test statistic at least as extreme as the one obtained, assuming H₀ is true. A small p-value (typically < 0.05) leads us to reject H₀ in favour of H₁. Type I error is rejecting H₀ when it is true (false positive); Type II error is failing to reject H₀ when it is false (false negative).

假设检验从代表现状的零假设 (H₀) 和备择假设 (H₁) 开始。P 值是在假设 H₀ 为真的条件下，获得一个至少与当前观测值一样极端的检验统计量的概率。如果 P 值很小（通常 < 0.05），我们就拒绝 H₀ 而支持 H₁。第 I 类错误是 H₀ 为真却拒绝了它（假阳性）；第 II 类错误是 H₀ 为假却未能拒绝它（假阴性）。

12. Chi-Squared Tests for Association | 卡方独立性检验

The chi-squared (χ²) test for association is used with two categorical variables arranged in a contingency table. The hypotheses are: H₀: the variables are independent; H₁: they are associated. Expected frequencies are calculated under the assumption of independence: (row total × column total) / grand total. The test statistic is Σ (O – E)² / E, which follows a χ² distribution with (rows–1)(columns–1) degrees of freedom.

卡方 (χ²) 独立性检验用于列联表中的两个分类变量。假设为：H₀：变量相互独立；H₁：它们之间存在关联。在独立性假设下计算期望频数：(行合计 × 列合计) / 总计。检验统计量为 Σ (O – E)² / E，它服从自由度为 (行数–1)(列数–1) 的 χ² 分布。

A valid test requires that no expected frequency falls below 1, and no more than 20% of expected frequencies are below 5. If this is not met, you may need to combine categories. The chi-squared test is always one-tailed in the upper tail, because we are only concerned with large discrepancies between observed and expected values.

要进行有效的检验，要求没有期望频数低于 1，并且不超过 20% 的期望频数低于 5。如果不满足这些条件，可能需要合并一些类别。卡方检验始终是上尾单尾检验，因为我们只关心观测值与期望值之间存在较大偏差的情况。

Published by TutorHao | Statistics Revision Series | aleveler.com

更多咨询请联系16621398022（同微信）

Pre-U CCEA Statistics: Quick Terminology Memorisation Guide | Pre-U CCEA 统计：词汇术语速记指南