📚 Statistical Distributions | 统计分布
A statistical distribution describes how the values of a random variable are spread or arranged. In A‑Level Statistics, understanding the shape, centre, and spread of distributions is fundamental to making inferences, calculating probabilities, and modelling real‑world phenomena. This article covers the key discrete and continuous distributions you need for the Edexcel specification: binomial, Poisson, and normal, along with the concepts of probability mass functions, cumulative distribution functions, expectation, and variance. We also explore approximations and the use of statistical tables, ensuring you have a solid grasp for both problem‑solving and examination success.
统计分布描述了一个随机变量取值的散布或排列方式。在A‑Level统计中，理解分布的形状、中心和散布特征是进行推断、计算概率以及建立真实世界模型的基础。本文涵盖了Edexcel考试大纲所需的核心离散与连续分布：二项分布、泊松分布和正态分布，同时包括概率质量函数、累积分布函数、期望与方差等概念。我们还会探讨分布之间的近似以及统计表的使用，帮助你扎实掌握问题求解技巧，从容应对考试。
1. Discrete and Continuous Distributions | 离散分布与连续分布
A discrete distribution is used when the random variable can only take a finite or countably infinite set of distinct values, such as the number of heads in ten coin tosses or the number of customers arriving per hour. Each possible value has a calculable probability, and these probabilities sum to 1.
当随机变量只能取有限个或可数无限个不同值（例如抛十次硬币中正面的次数或每小时到达的顾客数）时，我们使用离散分布。每个可能取值都有一个可以计算的概率，且所有概率之和等于1。
In contrast, a continuous distribution describes a random variable that can take any value within an interval, such as height, weight, or time. Because there are infinitely many possible values, the probability of the variable taking any single exact value is zero; we instead work with the probability of falling within an interval, represented by the area under the probability density function (PDF) curve.
相反，连续分布描述的是可以在一个区间内取任意值的随机变量，如身高、体重或时间。由于可能取值有无穷多个，变量取任意单一精确值的概率为零；我们转而计算落在区间内的概率，这由概率密度函数（PDF）曲线下的面积表示。
The distinction is crucial because the mathematical tools differ: discrete distributions use the probability mass function (PMF) and cumulative sums, while continuous distributions rely on the PDF and integration. For Edexcel Statistics, you will encounter both types, with the normal distribution being the primary continuous model.
这种区分至关重要，因为数学工具不同：离散分布使用概率质量函数（PMF）和累加求和，连续分布则依赖概率密度函数和积分。在Edexcel统计中，你会遇到这两种类型，其中正态分布是主要的连续模型。
2. Probability Mass Function (PMF) and Cumulative Distribution Function (CDF) | 概率质量函数与累积分布函数
x. The PMF must satisfy two conditions: 0 ≤ P(X = x) ≤ 1 for all x, and the sum over all possible x equals 1.
对于离散随机变量X，概率质量函数写成P(X = x)，表示X等于某个特定值x的精确概率。PMF必须满足两个条件：对所有x有0 ≤ P(X = x) ≤ 1，且所有可能x的概率之和等于1。
at is particularly useful for finding probabilities of ranges, such as P(3 ≤ X ≤ 7) = F(7) – F(2). CDFs are the foundation for using statistical tables efficiently.
累积分布函数F(x) = P(X ≤ x)是直到x（含）的所有概率之和。它提供累加汇总，在求区间概率时尤其有用，如P(3 ≤ X ≤ 7) = F(7) – F(2)。CDF是高效使用统计表的基础。
For continuous variables, the CDF is given by the integral of the PDF from the lower bound up to x. In the Edexcel course, you will mainly work with tabulated values of the CDF for the binomial, Poisson, and standard normal distributions, rather than carrying out many summation calculations from scratch.
对于连续变量，CDF由概率密度函数从下界到x的积分给出。在Edexcel课程中，你主要接触的是二项、泊松和标准正态分布的CDF表格值，而不是从头进行大量求和计算。
3. Expected Value and Variance | 期望值与方差
The expected value, E(X) or μ, is the long‑ the centre of the distribution. Variance, Var(X) or σ², measures the spread: Var(X) = E[(X – μ)²] = E(X²) – [E(X)]².
期望值E(X)或μ是随机变量的长期加权平均值，权重为概率。对于离散分布，E(X) = Σ x · P(X = x)。它代表分布的中心。方差Var(X)或σ²衡量散布程度：Var(X) = E[(X – μ)²] = E(X²) – [E(X)]²。
Crucially, for any distribution, the standard deviation σ is the square root of the variance. These summary statistics are essential for comparing data sets and for constructing confidence intervals. You will be required to know and use the formulas for the binomial and Poisson distributions, as well as interpret μ and σ for the normal distribution.
关键是，对于任何分布，标准差σ是方差的平方根。这些概括性统计量对于比较数据集和构建置信区间至关重要。你需要熟知并运用二项和泊松分布的公式，并能解释正态分布的μ和σ。
For the binomial distribution B(n, p), E(X) = np and Var(X) = np(1 – ibution N(μ, σ²), μ is the mean and σ² is the variance directly defined by the parameters.
对于二项分布B(n, p)，E(X) = np，Var(X) = np(1 – p)。对于泊松分布Po(λ)，均值与方差均等于λ。对于正态分布N(μ, σ²)，μ是均值，σ²即由参数直接定义的方差。
4. Binomial Distribution | 二项分布
The binomial distribution models the number of successes in a fixed number n of independent trials, where each trial has only two possible outcomes (success or failure) and the probability of success p remains constant. It is denoted X ~ B(n, p).
二项分布用于描述在固定次数的独立试验n中成功的次数，每次试验只有两种可能结果（成功或失败），且成功概率p保持不变。记作X ~ B(n, p)。
The PMF is given by P(X = r) = ⁿCᵣ pʳ (1 – p)ⁿ⁻ʳ, where ⁿCᵣ = n! / [r!(n – r)! the exam. Remember that the tables typically give P(X ≤ x) for selected values of n and p.
其概率质量函数为P(X = r) = ⁿCᵣ pʳ (1 – p)ⁿ⁻ʳ，其中ⁿCᵣ = n! / [r!(n – r)!]。当n较小时可直接计算，但n较大时你会使用考试中提供的二项累积分布表。请注意，表格通常给出选定n和p下P(X ≤ x)的值。
es. A typical exam question might ask you to find P(X ≥ k) by using the complement P(X ≤ k – s check that the situation meets the binomial conditions before applying the distribution.
二项模型的适用条件包括独立性、固定试验次数、恒定p，以及只有两种结果。典型考题可能要求你通过补集P(X ≤ k – 1) 求P(X ≥ k)。在应用分布之前，务必检查实际情况是否满足二项条件。
5. Poisson Distribution | 泊松分布
The Poisson distribution models the number of events occurring in a fixed interval of time or space, provided the events happen independently and at a constant average rate λ. It is written X ~ Po(λ). Common examples include the number of phone calls per hour at a call centre or the number of flaws per metre of fabric.
泊松分布用于描述在固定时间或空间区间内事件发生的次数，条件是事件相互独立且以恒定平均速率λ发生。记作X ~ Po(λ)。常见例子包括呼叫中心每小时接到的电话数量，或每米织物上的瑕疵数。
The PMF is P(X = r) = (e⁻ᵞ × λʳ) / r! for r = 0, 1, 2, … ve Poisson distribution table gives P(X ≤ x) for various λ, and you may need to read values carefully when λ is not an integer.
其概率质量函数为P(X = r) = (e⁻ᵞ × λʳ) / r!，r = 0, 1, 2, … 。对于泊松分布，E(X) = λ，Var(X) = λ。这个均值等于方差的独特性质是泊松模型的诊断性检查。泊松累积分布表给出各λ下P(X ≤ x)的值，当λ不是整数时需要仔细读取。
ences. In exam problems, you often need to scale λ appropriately when the interval changes – for example, if λ = 3 per hour, then for a 2‑hour period λ = 6. Adjusting the rate parameter proportionally is a key skill.
泊松过程要求事件单个、随机地发生，且不受先前事件的影响。考试题目中，当区间改变时你经常需要按比例调整λ——例如，若每小时λ = 3，则对于2小时时段λ = 6。按比例调整速率参数是一项关键技能。
6. Normal Distribution | 正态分布
The normal distribution is the most important continuous distribution in statistics. It is symmetric and bell‑shaped, defined entirely by its mean μ and variance σ². We write X ~ N(μ, σ²). Many natural measurements, such as IQ scores and measurement errors, approximately follow a normal distribution.
正态分布是统计学中最重要的连续分布。它呈对称的钟形，完全由其均值μ和方差σ²定义。我们记作X ~ N(μ, σ²)。许多自然测量数据，如智商分数和测量误差，都近似服从正态分布。
The probability density function is f(x) = (1 / √2πσ²) × e⁻ᐟ⁽ˣ⁻ᵘ⁾²/²σ²ᐟ to evaluate this directly. Instead, you standardise to the standard normal distribution Z ~ N(0, 1²) using the transformation Z = (X – allows you to use the standard normal table to find probabilities for any normal variable.
其概率密度函数为f(x) = (1 / √2πσ²) × e⁻ᐟ⁽ˣ⁻ᵘ⁾²/²σ²ᐟ，但你无需直接计算该式。相反，你通过变换Z = (X – μ) / σ 将其标准化为标准正态分布Z ~ N(0, 1²)。这使你能够利用标准正态表计算任何正态变量的概率。
The normal distribution is also the foundation for many inferential techniques, such as confidence intervals and hypothesis tests for means. The famous 68–95–99.7 rule tells us that about 68% of values lie within 1σ of the mean, 95% within 2σ, and 99.7% within 3σ. This empirical rule is extremely useful for quick checks.
正态分布也是许多推断技术的基础，例如均值的置信区间和假设检验。著名的68–95–99.7法则告诉我们，大约68%的值落在均值±1σ内，95%落在±2σ内，99.7%落在±3σ内。这一经验法则在快速检查时极其有用。
7. Standard Normal and Z‑Scores | 标准正态与Z分数
The standard normal distribution Z ~ N(0, 1) has mean 0 and standard deviation 1. The Z‑score tells you how many standard deviations an observation is away from the mean. If X ~ N(μ, σ²), then the Z‑score is z = (x – , you find the corresponding z, then look up Φ(z) = P(Z ≤ z) in the standard normal table.
标准正态分布Z ~ N(0, 1)的均值为0，标准差为1。Z分数表示一个观测值距离均值有多少个标准差。如果X ~ N(μ, σ²)，则Z分数为z = (x – μ) / σ。对于给定的x，先找到对应的z，然后在标准正态表中查找Φ(z) = P(Z ≤ z)。
pically giving the area to the left of z. For probabilities to the right, use P(Z > z) = 1 – Φ(z). For symmetric intervals, such as P(–a < Z < a), use 2Φ(a) – r the table gives lower or upper tail probabilities and whether z is positive or negative.
标准正态表是累积的，通常给出z左侧的面积。对于右侧的概率，使用P(Z > z) = 1 – Φ(z)。对于对称区间，如P(–a < Z < a)，使用2Φ(a) – 1。熟练掌握表格至关重要：注意表格给出的是下尾还是上尾概率，以及z的正负。
Working backwards, you may be given a probability and asked to find the corresponding z‑ uch that P(Z > z) = 0.025, you look up the 0.975 quantile, which gives approximately 1.96.
反向求解时，可能会给出概率让你求相应的z值，这是逆正态问题。在Edexcel考试中，你将使用百分位数表来解决。例如，要找使P(Z > z) = 0.025的z值，你查找0.975分位数，约为1.96。
8. Approximations between Distributions | 分布之间的近似
When certain conditions are met, one distribution can be used to approximate another, which simplifies calculations. The two key approximations in the Edexcel Statistics course are: binomial approximated by Poisson, and binomial or Poisson approximated by normal.
在满足特定条件时，可以用一种分布近似另一种，从而简化计算。Edexcel统计课程中的两个关键近似是：用泊松近似二项，以及用正态近似二项或泊松。
typically n > 50 and np < 5, or n > 20 and p < 0.1 (guideline varies). Let λ = np, then X ~ ause the Poisson table is easier to use for large n and small p than the binomial formula.
用泊松近似二项时，条件是n大且p小，通常n > 50且np < 5，或n > 20且p < 0.1（准则有所不同）。令λ = np，则X ~ B(n, p) ≈ Po(λ)。当n很大p很小时，泊松表比二项公式更容易使用。
l, apply when n is large and p is not too close to 0 or 1; a common rule is np > 5 and n(1 – p) > 5. Then X ~ B(n, p) ≈ N(np, np(1 – the normal approximation to the Poisson is appropriate when λ is large (usually λ > 10): X ~ Po(λ) ≈ N(λ, λ).
用正态近似二项时，条件是n大且p不太接近0或1；常用法则是np > 5且n(1 – p) > 5。此时X ~ B(n, p) ≈ N(np, np(1 – p))。类似地，当λ较大时（通常λ > 10），可用正态近似泊松：X ~ Po(λ) ≈ N(λ, λ)。
r x, we replace P(X ≤ x) with P(X < x + 0.5) under the normal curve, and P(X = x) with P(x – + 0.5). Always check the question for guidance on whether to use a continuity correction.
用连续分布近似离散分布时，连续性校正可提高精度。对于整数x，在正态曲线下将P(X ≤ x)替换为P(X < x + 0.5)，将P(X = x)替换为P(x – 0.5 < X < x + 0.5)。务必查看题目指示是否需要使用连续性校正。
9. Using Statistical Tables | 使用统计表
Your Edexcel examinations provide formula booklets that include tables for the binomial cumulative distribution, Poisson cumulative distribution, and standard normal distribution (both cumulative and percentage points). It is vital to become efficient at locating the correct row and column under time pressure.
Edexcel考试提供公式手册，包含二项累积分布表、泊松累积分布表以及标准正态分布表（累积表和百分位数表）。在时间压力下，能够迅速找到正确的行和列至关重要。
he correct block for n and p. The table gives P(X ≤ x), so for P(X = x) calculate P(X ≤ x) – P(X ≤ x – 1). For P(X < x) use P(X ≤ x – 1). For P(X ≥ x) use 1 – P(X ≤ x – 1). Similar logic applies to the Poisson table.
对于二项表，确保使用正确的n和p分块。表格给出P(X ≤ x)，因此P(X = x)用P(X ≤ x) – P(X ≤ x – 1)计算。P(X < x)用P(X ≤ x – 1)。P(X ≥ x)用1 – P(X ≤ x – 1)。类似的逻辑也适用于泊松表。
mal cumulative table often gives Φ(z) for positive z to 2 or 3 decimal places. Remember Φ(–z) = 1 – Φ(z) due to symmetry. The percentage points table gives z‑ ese tables with correct standardisation is the core of many normal distribution questions.
正累积态表通常给出正z对应的Φ(z)，保留2或3位小数。根据对称性，记住Φ(–z) = 1 – Φ(z)。百分位数表给出常用尾部概率（如0.05、0.025、0.01）对应的z值。将这些表格与正确的标准化结合是许多正态分布问题的核心。
10. Common Problems and Exam Techniques | 常见问题与应试技巧
A frequent exam task is to determine which distribution to use. Read the context carefully: if there is a fixed number of trials with constant p, think binomial. If events are occurring randomly over time/space at a constant rate, think Poisson. If the variable is continuous and symmetric, or if you are given μ and σ, think normal.
考试中一个常见任务是判断应使用哪种分布。仔细阅读情境：如果有固定试验次数且p恒定，考虑二项；如果事件在时间/空间上随机发生且速率恒定，考虑泊松；如果变量是连续的且对称，或给出了μ和σ，考虑正态。
Always state the distribution and its parameters clearly, e.g., X ~ B(20, 0.3) or Y ~ Po(4.5). This simple step often earns marks and structures your solution. When finding probabilities for worded problems, define your random variable explicitly – this clarifies your reasoning and avoids common mistakes.
始终清晰地写出分布及其参数，例如X ~ B(20, 0.3)或Y ~ Po(4.5)。这一简单步骤通常能得分，并搭建你的解题框架。在应用题中求概率时，明确定义随机变量——这会使你的推理更清晰，并避免常见错误。
Watch out for ‘at least’, ‘more than’, ‘between’ phrasing. Translate ‘ te variable that equals P(X ≥ 6). Drawing a quick sketch or number line can help avoid off‑by‑one errors. Also, double‑check whether a continuity correction is appropriate in approximation problems.
留意“至少”、“超过”、“介于”等措辞。将“超过5”翻译为X > 5，对于离散变量这等于P(X ≥ 6)。快速画出草图或数轴有助于避免缺一错误。此外，在近似问题中请反复检查是否适合使用连续性校正。
11. Applications of Statistical Distributions | 统计分布的应用
Statistical distributions are not just theoretical; they underpin quality control, finance, biological studies, and engineering. For example, a manufacturer may use the binomial distribution to model the number of defective items in a batch, or the Poisson distribution to predict the number of machine breakdowns in a month. The normal distribution is used extensively in setting confidence intervals for population parameters and in hypothesis testing.
统计分布并非纯理论，它们是质量控制、金融、生物研究和工程等领域的基础。例如，制造商可能使用二项分布模拟一批产品中的次品数量，或使用泊松分布预测一个月内的机器故障次数。正态分布被广泛用于设定总体参数的置信区间以及假设检验。
Understanding distributional assumptions helps you assess whether a model is appropriate. If data show that the variance is much larger than the mean, a Poisson model may be unsuitable; a binomial model requires independence and a fixed p. These checks are vital for valid conclusions in real‑life statistical projects and exam scenarios.
理解分布的假设有助于评估模型是否合适。如果数据显示方差远大于均值，泊松模型可能不合适；二项模型要求独立性和固定的p。这些检查对于实际统计项目和考试情境中得出有效的结论至关重要。
Many questions ask you to find the mode, median, or make predictions based on the distribution. For symmetric normal distributions, the mean, median, and mode coincide. For skewed distributions like the Poisson with small λ, the mean and median differ, and the distribution is positively skewed, which has implications for predictions.
许多题目要求你找出众数、中位数或基于分布进行预测。对于对称的正态分布，均值、中位数和众数重合。对于小λ的偏态分布如泊松，均值与中位数不同，分布呈正偏态，这对预测有一定影响。
12. Linking Distributions to Statistical Inference | 分布与统计推断的联系
The distributions you study form the basis of statistical inference. The normal distribution is used to construct z‑based confidence intervals and perform z‑tests. The binomial and Poisson distributions underlie hypothesis tests for proportions and rates. Understanding their properties – particularly the expected values and variances – enables you to derive standard errors and test statistics.
你所学习的分布构成了统计推断的基础。正态分布用于构建基于z的置信区间和进行z检验。二项和泊松分布支撑着关于比例和速率的假设检验。理解它们的性质——尤其是期望值和方差——使你能够推导标准误和检验统计量。
For example, when testing a population proportion, the sample proportion Ƞ̂ is approximately normally distributed under the null hypothesis: Ƞ̂ ~ N(p, p(1 – p)/n) for large n, by the central limit theorem. This links the binomial distribution back to the normal. Similarly, the sum or mean of independent Poisson variables tend to normality as λ increases, underpinning many tests.
例如，检验总体比例时，根据中心极限定理，对于大样本，样本比例Ƞ̂近似服从正态分布：Ƞ̂ ~ N(p, p(1 – p)/n)。这将二项分布与正态分布联系起来。类似地，随着λ增大，独立泊松变量的总和或均值趋向正态分布，为许多检验提供了理论依据。
In the Edexcel specification, you are expected to interpret parameters, choose the correct distribution for a scenario, calculate probabilities, and understand the assumptions underlying each model. Strengthening your fluency with distributions will pay dividends in later topics such as sampling distributions, confidence intervals, and hypothesis testing.
在Edexcel大纲中，你需要解释参数、为场景选择正确的分布、计算概率，并理解每个模型背后的假设。扎实掌握分布知识将使你在后续的学习（如抽样分布、置信区间和假设检验）中受益匪浅。
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