📚 GCSE Edexcel Statistics: Vocabulary & Terminology Quick Memorisation Guide | GCSE Edexcel 统计：词汇术语速记指南
Mastering the precise terminology of GCSE Edexcel Statistics is your key to better performance in both coursework and exams. This guide provides a structured, bilingual walkthrough of core vocabulary, pairing each term with a concise English definition and its Chinese equivalent to accelerate memorisation. Where helpful, memory hooks, formulas and visual cues are included to reinforce your learning, and every section is matched to the wording that examiners actually use in Edexcel papers.
掌握 GCSE Edexcel 统计的准确术语是你在课程作业和考试中取得更好成绩的关键。本指南通过结构化的双语培训，将核心词汇配以简洁的英文定义和中文对应，加速记忆。在适当之处，还加入了记忆辅助、公式和视觉联想，以巩固你的学习；每一个小节都对应 Edexcel 试卷中命题人真正使用的表述方式。
1. Why Terminology Matters | 为什么术语如此重要
In Edexcel GCSE Statistics (specification 1ST0), marks are awarded for using the correct statistical word at the correct moment. A question that asks you to 'describe the distribution' wants the vocabulary of shape and spread, not a story about the data. A question that says 'state' wants a value with no working, while 'explain' wants a reason linked to context. Learning the word list is therefore not extra work - it is the fastest way to convert knowledge into marks.
在 Edexcel GCSE 统计（考纲代码 1ST0）中，分数是给"在正确的时刻用了正确的统计词汇"的。要求你 describe the distribution（描述分布）的题目，考的是描述形状与离散程度的词汇，而不是关于数据的故事。要求 state（说出）的题目只要数值、不要过程，而 explain（解释）则要求结合具体情境给出理由。因此背术语不是额外负担，而是把知识变成分数的最快路径。
The command words below appear again and again. Memorise them as a pair with their Chinese meaning, and read them twice before you start writing.
下面这些指令词会反复出现。请把它们和中文含义成对记忆，并在动笔前读两遍。
- State / Give - write the value only, no working | 说出／给出——只写数值，不需要过程
- Write down - the answer should be almost immediate from the data | 写出——答案应能直接从数据中读出
- Describe - say what you see: shape, centre, spread, outliers | 描述——说明你看到什么：形状、中心、离散程度、异常值
- Explain - give a reason that mentions the context of the question | 解释——给出与题目情境挂钩的理由
- Compare - you must mention both data sets, using a measure of average and a measure of spread | 比较——必须同时提到两组数据，并用一个平均指标和一个离散指标
- Comment - give a short justified judgement, often about reliability or bias | 评论——给出简短且有依据的判断，常涉及可靠性或偏差
- Criticise - find the fault, then say how to improve it | 批评——找出问题，再说明如何改进
- Estimate - an approximate value obtained by calculation or from a graph | 估计——通过计算或从图象读出的近似值
A useful memory hook: D-E-C-C stands for Describe, Explain, Compare, Criticise - the four command words that carry the most marks on the longer questions.
一个实用的记忆口诀：D-E-C-C，即 Describe、Explain、Compare、Criticise，这四个指令词在长题中占分最多。
2. Types of Data | 数据类型
Qualitative data describes qualities or categories that cannot be measured with numbers, such as eye colour, brand names or types of transport. It is often non-numerical and is used when we classify items into groups. If you can put the answer into a named box rather than a measuring instrument, it is qualitative.
定性数据描述无法用数字衡量的性质或类别，如眼睛颜色、品牌名称或交通方式。它通常是非数值型的，用于将事物分组归类。如果答案只能放进"分类框"而不是"测量仪"，它就是定性数据。
Quantitative data consists of numerical measurements or counts. It can be further split into discrete data (countable values like number of siblings, shoe size) and continuous data (measurable values that can take any number within a range, such as height, weight, time). The quick test is simple: ask yourself whether a value of 3.7 would make sense. If yes, the data is continuous; if no, it is discrete.
定量数据由数值测量或计数组成。可进一步分为离散数据（可数的值，如兄弟姐妹数量、鞋码）和连续数据（可在一定范围内取任意值的测量值，如身高、体重、时间）。快速判断法很简单：问自己"3.7 这个值是否合理"。合理就是连续数据，不合理就是离散数据。
Related terms include categorical data (the same thing as qualitative data) and ordinal data, which can be ordered but not measured precisely, for example satisfaction ratings such as poor, good, excellent. Primary data is collected first-hand by you; secondary data is gathered from existing sources.
相关术语包括分类数据（等同于定性数据），以及有序数据——它可以排序但无法精确测量，例如"差、好、优秀"这类满意度评分。一手数据由你自己第一手收集；二手数据则来自已有的资料来源。
A compact way to hold all of this in mind is the table below. Cover the right-hand column and test yourself until every row is automatic.
下面这张表可以用最紧凑的方式把以上内容装进脑子。遮住右侧一列自我测试，直到每一行都能脱口而出。
| Term 术语 | Definition 定义 | Example 例子 |
| Qualitative 定性 | Non-numerical category 非数值类别 | Eye colour 眼睛颜色 |
| Quantitative 定量 | Numerical measurement or count 数值测量或计数 | Mass in kg 质量（千克） |
| Discrete 离散 | Countable, whole values only 只能取可数的整值 | Number of goals 进球数 |
| Continuous 连续 | Any value in a range 区间内任意值 | Time in seconds 时间（秒） |
| Ordinal 有序 | Ordered categories, unequal gaps 可排序但间距不均 | Star ratings 星级评分 |
3. Categorical and Ordinal Data | 分类数据与有序数据
Categorical data simply means data placed into named groups. It is the preferred term in Edexcel mark schemes when the answer is a label rather than a number, and it splits into two families: nominal data, where the groups have no natural order (favourite colour, type of pet), and ordinal data, where the groups do have an order but the gaps between them are not guaranteed to be equal.
分类数据就是指被放入"有名称的分组"里的数据。当答案是标签而非数字时，Edexcel 评分标准更偏好这个说法。它又分为两类：名义数据，各组之间没有天然顺序（最喜欢的颜色、宠物种类）；以及有序数据，各组之间有顺序，但组与组之间的间距不保证相等。
Ordinal data is the classic trap. A response of 'excellent' is better than 'good', but we cannot claim the difference between excellent and good equals the difference between good and poor. So you may compare ordinal data by median or mode, but quoting a mean of a satisfaction score is statistically unsafe.
有序数据是最经典的陷阱。回答"优秀"比"好"更好，但我们不能声称"优秀与好"的差距等于"好与差"的差距。因此，有序数据可以用中位数或众数比较，但拿满意度分数去算平均数在统计上是不严谨的。
Bivariate data involves two variables measured on the same individual (height and shoe size), while multivariate data involves more than two. When one variable is used to explain or predict the other, we call the explanatory variable the independent variable and the outcome the dependent variable.
二元数据指对同一个个体测量两个变量（身高与鞋码），多元数据则涉及两个以上变量。当一个变量用来解释或预测另一个变量时，我们把用于解释的一方称为自变量，被预测的一方称为因变量。
4. Primary and Secondary Data | 一手数据与二手数据
Primary data is collected first-hand by you or your team for the specific purpose of your investigation - a questionnaire you design, an experiment you run, a tally you keep. Its great advantage is control: you know exactly how the data was obtained, so you can judge its reliability.
一手数据是你或你的团队为本次调查的特定目的而亲自收集的数据——你设计的问卷、你做的实验、你记录的计数。它最大的优势是可控：你确切知道数据是如何获得的，因此可以判断其可靠性。
Secondary data is gathered from existing sources such as government databases, published reports, company records or the internet. It is often cheaper, faster and available in far larger quantities, but you have no control over how it was collected, and it may be out of date or collected for a different purpose.
二手数据来自已有资料，例如政府数据库、已发布的报告、公司记录或互联网。它通常更便宜、更快捷，数量也大得多，但你无法控制它的收集方式，而且它可能已经过时，或原本是为别的目的而收集的。
An exam answer that earns full marks usually names both a strength and a weakness, then links them to context: 'Secondary data on national rainfall is reliable because it comes from the Met Office, but it may not reflect the microclimate of one school field.'
能拿满分的答案通常会同时指出一个优点和一个缺点，再与情境挂钩："全国降雨量的二手数据可靠，因为它来自英国气象局，但它未必能反映某一所学校操场的微气候。"
5. Populations, Samples and Census | 总体、样本与普查
The population is the whole group you want to draw conclusions about, and every member of it is a sampling unit. A census measures every single member of the population. It gives the most accurate picture, but it is expensive, slow and sometimes impossible - you cannot put every can of drink through a taste test and still sell it.
总体是你想要得出结论的整个群体，其中的每一个成员称为一个抽样单元。普查则测量总体中的每一个成员。它能给出最准确的图景，但费用高、耗时长，有时根本不可行——你不可能把每一罐饮料都开罐品尝后还拿去售卖。
A sample is a subset of the population chosen for study, and a sampling frame is the list from which the sample is drawn - a register of names, a list of postcodes, a database of customer numbers. Without a sampling frame you cannot take a genuine random sample, and this is why a sampling frame matters when you are asked to criticise a survey design.
样本是从总体中选出来接受研究的一部分，而抽样框是抽取样本所依据的名册——如姓名登记表、邮编列表、客户编号数据库。没有抽样框就无法进行真正意义上的随机抽样，这也是为什么当题目要求你批评一项调查设计时，抽样框往往是答题要点。
A sample is representative if its structure mirrors the population, and biased if some parts of the population are systematically more likely to be included. A biased sample does not simply mean an unlucky one - it means the method itself favours one group, for example surveying only people leaving a gym about exercise habits.
如果样本的结构与总体一致，它就是有代表性的；如果总体中某些部分被系统性地更容易选中，它就是有偏的。有偏样本不仅指"运气不好"的样本，而是方法本身偏向某一群体，例如只在健身房门口调查锻炼习惯。
6. Sampling Methods | 抽样方法
A simple random sample gives every member of the population an equal chance of selection, usually by numbering the sampling frame and using a random number generator. It is unbiased and easy to justify, but it may by chance miss small subgroups, and it needs a complete list.
简单随机抽样使总体中每个成员被选中的机会相等，通常做法是给抽样框编号并使用随机数生成器。它无偏且容易论证，但可能碰巧漏掉小型子群体，而且需要一份完整的名单。
A systematic sample selects every k-th item from an ordered list after a random start. It is quick and spreads across the list, but if the list has a repeating pattern the sample can become biased - choosing every 10th house on a street where houses alternate in size is a classic failure.
系统抽样在有序列表中随机确定起点后，每隔 k 个抽取一个。它速度快，且分布贯穿整个列表，但如果列表本身存在周期性规律，样本就可能产生偏差——在同一条街上每隔 10 户抽一户，而这条街的房屋大小交替变化，就是典型失败案例。
A stratified sample divides the population into strata with a shared characteristic (year group, gender, region) and then samples each stratum in proportion. The number taken from each stratum is calculated as follows.
分层抽样把总体按共同特征划分成若干层（年级、性别、地区），再按比例从每层抽取。每层抽取的数量按下式计算。
Stratum sample size = (stratum size ÷ population size) × sample size
每层样本量 = (该层人数 ÷ 总体人数) × 样本量
A quota sample also uses groups, but the interviewer fills each quota with whoever is available, so it is quick and cheap yet not random. An opportunity sample (also called a convenience sample) simply takes whoever is on hand. Both are common in real life and both are easy targets for a 'criticise' question.
配额抽样同样按群体划分，但访问员用"碰到谁就填谁"的方式填满配额，因此快捷、省钱，却不是随机的。机会抽样（又称便利抽样）则直接取手边可得的人。两者在现实生活中都很常见，也都是"批评类"题目的常见靶子。
7. Questionnaires and Data Collection | 问卷与数据收集
A closed question offers fixed response boxes and produces data that is easy to tabulate; an open question invites free text and produces richer answers that are harder to analyse. A well-designed questionnaire is short, avoids leading questions ('Do you agree that the canteen is too expensive?'), avoids overlapping response boxes, and includes a pilot study so that problems are found before the real survey begins.
封闭式问题提供固定选项框，产生的数据易于汇总；开放式问题允许自由作答，答案更丰富但更难分析。设计良好的问卷应当简短，避免诱导性问题（"你是否同意食堂太贵了？"）、避免选项区间重叠，并进行预调查（试点研究），在正式调查前发现问题。
Response bias occurs when the people who answer differ systematically from those who do not - for example, an online survey about internet use. Related vocabularies you should know are reliability (would the same method give consistent results if repeated?), validity (does the data actually measure what it claims to measure?) and accuracy (how close a value is to the true value).
回答偏差出现在"作答的人"与"未作答的人"存在系统性差异时——例如一项有关上网情况的在线调查。相关词汇你还需要掌握：可靠性（同一方法重复使用是否给出一致结果？）、有效性（数据是否真的测量了它声称测量的东西？）以及准确性（数值与真值有多接近）。
When a population is hard to count, Edexcel uses the capture-recapture method. A sample is captured, marked and released; later a second sample is taken and the number of marked individuals in it is recorded.
当总体难以清点时，Edexcel 会使用标记重捕法。先捕捉一个样本，做标记后放回；之后再取第二个样本，记录其中被标记个体的数量。
Population estimate N ≈ (M × C) ÷ R
总体估计值 N ≈ (M × C) ÷ R
Here M is the number marked in the first sample, C is the size of the second sample, and R is the number of marked individuals found in the second sample. A typical exam question then asks you to comment on why the estimate may be unreliable - for example, marks may wear off or marked animals may become easier to spot.
其中 M 是第一次样本中被标记的数量，C 是第二个样本的规模，R 是第二个样本中被标记个体的数量。典型考题随后会要求你评论该估计值为何可能不可靠——例如标记可能脱落，或被标记的动物更容易被再次发现。
8. Tables, Charts and Histograms | 表格、图表与直方图
In a frequency table, the frequency is how many times a value occurs; in grouped data the values are bundled into class intervals, and the class width is the difference between the upper and lower class boundaries. A two-way table cross-tabulates two categorical variables, and a tally chart is simply a running count before the totals are written up.
在频数表中，频数指某个数值出现的次数；在分组数据中，数值被归入组区间，组距则是上组界与下组界之差。双向表交叉列出两个分类变量，画记表则是在汇总前进行的连续计数记录。
For a histogram the vertical axis is not frequency but frequency density, because the bars may have unequal widths. The area of each bar represents frequency, which is why you must always calculate before you draw.
绘制直方图时，纵轴不是频数，而是频数密度，因为各柱宽度可能不相等。每一根柱的面积才代表频数，因此动笔之前必须先计算。
Frequency density = frequency ÷ class width
频数密度 = 频数 ÷ 组距
A bar chart is used for discrete or categorical data with gaps between bars; a pie chart shows proportion, where the angle for each sector is (frequency ÷ total) × 360°. A stem and leaf diagram keeps the original values while showing shape, and it needs a key. A cumulative frequency curve plots the running total against the upper class boundary, and it is read with the help of a scale drawing tool: median at 50%, quartiles at 25% and 75%.
条形图用于离散数据或分类数据，柱与柱之间留有空隙；饼图展示比例，每个扇形的角度为（频数 ÷ 总数）× 360°。茎叶图在保留原始数值的同时展现分布形状，并且必须加图例（键）。累积频数曲线把累计频数与上组界对应作图，并借助比例尺作图读取数值：中位数
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