📚 GCSE AQA Statistics: Vocabulary & Terminology Quick Reference Guide | GCSE AQA 统计：词汇术语速记指南
Vocabulary is where marks are quietly lost in GCSE AQA Statistics. A candidate who can calculate a standard deviation but cannot explain what an 'outlier' or a 'stratified sample' actually is will still drop marks on the written questions. This guide is a compact reference of the key terms you must be able to define, use in context and spell correctly in the exam.
在 GCSE AQA 统计考试中，术语往往是悄悄丢分的地方。一个会算标准差却讲不清什么是"离群值"或"分层抽样"的考生，仍然会在文字说明题上失分。本指南是一份精炼的核心术语速查表，涵盖你必须能够定义、会用在情境中、并在考卷上拼写正确的关键词。
Each section defines one family of terms, gives the meaning in plain English and the matching Chinese translation, and adds the exam-style wording that markers expect to see. Use it as a checklist: tick the terms you can define without looking, then revisit only the ones you cannot.
每一节聚焦一类术语，给出简明英文释义与对应的中文翻译，并补充考官期待看到的考试用语。把它当作清单使用：能不看资料就说清定义的打勾，只回头复习没掌握的那些。
1. Data Types and Collection | 数据类型与收集
Qualitative data (categorical data): non-numerical data that describes a quality or category, such as eye colour, type of transport or favourite subject.
定性数据（分类数据）：描述性质或类别的非数值数据，例如眼睛颜色、交通方式或最喜欢的科目。
Quantitative data: numerical data that can be measured or counted, such as height in cm or number of siblings.
定量数据：可以测量或计数的数值数据，例如以厘米计的身高或兄弟姐妹的人数。
Discrete data: quantitative data that can only take specific values, usually whole numbers obtained by counting, such as the number of students in a class.
离散数据：只能取特定值（通常是通过计数得到的整数）的定量数据，例如一个班的学生人数。
Continuous data: quantitative data that can take any value within a range, limited only by the accuracy of the measuring instrument, such as height, mass or time.
连续数据：在一定范围内可以取任意值的定量数据，仅受测量仪器精度限制，例如身高、质量或时间。
Primary data: data collected first-hand by the researcher for the specific purpose of the current investigation.
一手数据：研究者为当前调查的特定目的亲自收集的数据。
Secondary data: data that has already been collected by someone else, for another purpose, and is reused by you.
二手数据：由他人为其他目的已经收集、被你再次使用的数据。
Raw data: data that has not yet been processed, ordered or organised into a table.
原始数据：尚未经过处理、排序或整理成表格的数据。
Population: the entire group of individuals or items that you are interested in studying.
总体：你所要研究的所有个体或项目的整个集合。
Sample: a subset of the population, selected so that conclusions can be drawn about the whole population.
样本：从总体中选出的子集，用于对整体总体作出结论。
Census: a survey that collects data from every single member of the population.
普查：从总体中每一个成员收集数据的调查。
Univariate and bivariate data: univariate data involves one variable only; bivariate data involves two variables measured on the same individual, allowing a relationship to be investigated.
单变量与双变量数据：单变量数据只涉及一个变量；双变量数据对同一个个体的两个变量进行测量，从而可以研究两者之间的关系。
| Type | Can be counted or measured? | Example |
| Qualitative | Neither - describes a category | Colour, make of car |
| Discrete | Counted only | Number of goals scored |
| Continuous | Measured, any value in a range | Mass, temperature, time |
2. Sampling Techniques | 抽样技术
Sampling frame: a complete list of all the members of the population from which a sample is taken, for example a school register.
抽样框：包含总体全部成员、用于抽取样本的完整名册，例如学校的学生名册。
Simple random sampling: every member of the population has an equal chance of being selected, for example names drawn from a hat or random numbers generated by a calculator.
简单随机抽样：总体中每个成员被选中的机会均等，例如从帽子中抽签或用计算器生成随机数抽取。
Systematic sampling every kth item on the sampling frame is selected, where k = population size ÷ sample size.
系统抽样：确定随机起点后，从抽样框中每隔 k 个选取一个项目，其中 k = 总体规模 ÷ 样本规模。
Stratified sampling: the population is divided into groups called strata that share a characteristic, and a random sample is taken from each stratum in proportion to its size.
分层抽样：将总体按某一共同特征分成若干"层"，并从每一层按该层规模的比例随机抽取样本。
Quota sampling: the interviewer selects a fixed number of people with specified characteristics; the choice of individuals is not random.
配额抽样：访问员按规定的特征选取固定数量的人；具体选谁不是随机的。
Convenience (opportunity) sampling: selecting whoever is easiest to reach, such as the first twenty people in the canteen.
便利抽样（偶遇抽样）：选取最容易接触到的人，例如食堂里前二十个人。
Cluster sampling: the population is divided into clusters (often geographical), some clusters are chosen at random and every member of the chosen clusters is used.
整群抽样：将总体分成若干群（常按地理划分），随机抽取若干群，并使用被选中群内的全部成员。
The key formula to memorise is the number taken from each stratum:
记住这个关键公式，用来计算每层应抽取的人数：
Stratum sample size = (stratum size ÷ population size) × sample size
每层样本数 = (该层人数 ÷ 总体人数) × 样本总量
3. The Statistical Enquiry Cycle and Data Quality | 统计探究循环与数据质量
Statistical enquiry cycle: the five-stage process of Problem, Plan, Data, Analysis and Conclusion that structures every statistical investigation.
统计探究循环：由"提出问题、制定计划、收集数据、分析数据、得出结论"五个阶段构成的流程，用于组织每一次统计调查。
Hypothesis: a testable statement about the population that the investigation sets out to support or reject.
假设：关于总体、可被检验的陈述，调查的目的即在于支持或否定它。
Bias: a systematic error that pushes results consistently in one direction, so the sample is not representative of the population.
偏差：使结果持续偏向某一方向的系统性误差，导致样本不能代表总体。
Representative sample: a sample whose characteristics closely match those of the population.
有代表性的样本：特征与总体高度吻合的样本。
Reliability: the extent to which repeating the investigation would produce similar results; larger samples generally increase reliability.
可靠性：重复该调查能得到相似结果的程度；样本越大，可靠性通常越高。
Validity: the extent to which the data actually measures what it claims to measure.
有效性：所收集的数据确实测到了它声称要测的东西的程度。
Leading question: a question worded in a way that suggests a particular answer, such as "Do you agree that the canteen should stop serving junk food?"
诱导性问题：措辞上暗示某一答案的问题，例如"你同意食堂应该停止供应垃圾食品吗？"
Response bias: when people give inaccurate answers, for example by exaggerating or by trying to please the interviewer.
作答偏差：被调查者给出不准确答案的情况，例如夸大事实或迎合访问员。
Non-response: when selected individuals do not reply or refuse to take part, which can make the sample unrepresentative if their views differ.
无应答：被选中的人没有回复或拒绝参与；如果这些人观点不同，就会使样本失去代表性。
Pilot survey: a small trial run of a questionnaire used to identify unclear questions before the main survey.
预调查：正式调查前对问卷进行的小规模试测，用于发现表述不清的问题。
4. Organising Data: Tables and Frequency Distributions | 数据整理：表格与频数分布
Frequency: the number of times a value or category occurs in a data set.
频数：某个数值或类别在数据集中出现的次数。
Tally chart: a table using tally marks to record frequencies as data is collected.
计数表：用划记符号在收集数据时记录频数的表格。
Grouped frequency table: a table in which data is placed into class intervals rather than listed individually, used when data is continuous or has many values.
分组频数表：把数据归入组距而不是逐一列出的表格，用于连续数据或数值很多的数据。
Class interval (group): a range of values in a grouped frequency table, such as 10 ≤ x < 20.
组距（组）：分组频数表中的一段数值范围，例如 10 ≤ x < 20。
Class width e upper and lower class boundaries of an interval; 10 ≤ x < 20 has a class width of 10.
组宽：一个组的上、下组界之差；10 ≤ x < 20 的组宽为 10。
Midpoint (mid-interval value): the value halfway through a class interval, calculated as (lower boundary + upper boundary) ÷ 2, and used to estimate the mean of grouped data.
组中值：位于组距中点的数值，等于（下界 + 上界）÷ 2，用于估计分组数据的平均数。
Cumulative frequency: a running total of the frequencies, showing how many values are less than or equal to a given value.
累计频数：频数的逐步累加，表示小于或等于某一数值的数据有多少个。
Two-way table: a table that cross-classifies data by two variables, allowing conditional statements to be read off.
双向表：按两个变量交叉分类数据的表格，可用于读出条件性的结论。
Outlier ule is any value more than 1.5 × IQR below the lower quartile or above the upper quartile.
离群值：与其余数据相比异常偏离的数值；常用判据是低于下四分位数或高于上四分位数超过 1.5 × 四分位距。
5. Charts and Diagrams | 图表与图形
Bar chart: a diagram for qualitative or discrete data in which frequencies are shown by bars of equal width with gaps between them.
条形图：用于定性或离散数据的图形，用等宽且有间隔的条形表示频数。
Bar-line chart: a bar chart in which bars are replaced by vertical lines, often used for discrete data.
线条图：用竖线代替条形的图，常用于离散数据。
Pie chart ngles are proportional to the frequencies; each sector angle = (frequency ÷ total) × 360°.
饼图：把圆分成与频数成比例的扇形；每个扇形的圆心角 = (频数 ÷ 总数) × 360°。
Pictogram: a diagram using pictures or symbols to represent frequencies, with a key stating what each symbol is worth.
象形图：用图形或符号表示频数的图，需配图例说明每个符号代表多少。
Histogram: a diagram for continuous data with no gaps between bars, in which the area of each bar (not its height) is proportional to the frequency.
直方图：用于连续数据的图形，条形之间没有间隔，每个条形面积（而非高度）与频数成正比。
Frequency density: the height of a bar in a histogram, calculated as frequency ÷ class width; it must be used whenever class widths are unequal.
频数密度：直方图中条形的高度，等于频数 ÷ 组宽；当各组组宽不相等时必须使用它。
Frequency polygon: a line graph joining the midpoints of the tops of the bars of a histogram or the midpoints of class intervals.
频数多边形：连接直方图各条形顶端中点（或各组组中值）的折线图。
Stem-and-leaf diagram: a diagram that splits each value into a stem and a leaf, keeping the original data values while showing the shape of the distribution.
茎叶图：把每个数值分成"茎"与"叶"的图，既保留了原始数据值，又能显示分布的形状。
Cumulative frequency graph (ogive): a graph of cumulative frequency against the upper class boundary, used to estimate the median, quartiles and percentiles.
累计频数图（累积曲线）：以累计频数对组上界作图得到的曲线，用于估计中位数、四分位数和百分位数。
6. Measures of Central Tendency (Averages) | 集中趋势的度量（平均数）
Mean: the sum of all values divided by the number of values; it uses every piece of data but is affected by extreme values.
算术平均数：全部数值之和除以数值个数；它用到了所有数据，但会受极端值影响。
Mean = Σx ÷ n
Median: the middle value when the data is arranged in order; if there are two middle values, the median is their mean.
中位数：把数据按顺序排列后位于中间的数值；若中间有两个值，中位数取二者的平均。
Mode: the value or category that occurs most often; a data set may have no mode or more than one mode.
众数：出现次数最多的数值或类别；一组数据可能没有众数，也可能有多个众数。
Modal class: the class interval with the highest frequency in a grouped frequency table.
众数组：分组频数表中频数最大的那个组。
Estimated mean: the mean calculated from grouped data using the midpoint of each class, so it is an estimate rather than an exact value.
估计平均数：用各组的组中值对分组数据算出的平均数，因此是估计值而非精确值。
Estimated mean = Σ(f × x) ÷ Σf
Weighted mean: an average in which each value is multiplied by a weight reflecting its importance, then divided by the total of the weights.
加权平均数：每个数值先乘以反映其重要性的权重，再除以权重之和得到的平均数。
Choosing an average: the mean is best when the data is fairly symmetrical, the median is best when there are outliers or the data is skewed, and the mode is best for categorical data.
如何选择平均数：数据大致对称时用平均数最好；存在离群值或数据偏斜时用中位数最好；分类数据则用众数最好。
7. Measures of Spread (Dispersion) | 离散程度的度量
Range: the largest value minus the smallest value; simple to calculate but based on only two values.
极差：最大值减去最小值；计算简单，但只依赖两个数值。
Lower quartile (LQ): the value one quarter of the way through the ordered data; 25% of the data lies below it.
下四分位数：有序数据中位于四分之一处的数值；有 25% 的数据小于它。
Upper quartile (UQ): the value three quarters of the way through the ordered data; 75% of the data lies below it.
上四分位数：有序数据中位于四分之三处的数值；有 75% 的数据小于它。
Interquartile range (IQR): a measure of spread containing the middle 50% of the data, and unaffected by outliers.
四分位距：包含数据中间 50% 的离散程度度量，不受离群值影响。
IQR = Upper quartile − Lower quartile
Percentile: the value below which a given percentage of the data lies; the 90th percentile has 90% of values below it.
百分位数：有给定百分比的数据小于它的那个数值；第 90 百分位数之下有 90% 的数据。
Box plot (box-and-whisker diagram): a diagram showing the minimum value, LQ, median, UQ and maximum value, used to compare two distributions visually.
箱线图（箱须图）：显示最小值、下四分位数、中位数、上四分位数与最大值的图形，用于直观比较两个分布。
Standard deviation: a measure of spread that describes the typical distance of values from the mean; it uses every value and has the same units as the data.
标准差：描述各数值距平均数典型距离的离散程度度量；它用到了每个数值，且单位与原始数据相同。
σ = √(Σ(x − μ)² ÷ n)
Variance: the square of the standard deviation; it is measured in squared units, so the standard deviation is usually easier to interpret.
方差：标准差的平方；其单位为原单位的平方，因此标准差通常更易解释。
Comparing distributions: always quote a measure of central tendency and a measure of spread, and comment on the context in a sentence such as "the mean score is higher but the results are more variable".
比较分布：务必同时给出集中趋势和离散程度的度量，并结合情境说明，例如"平均分更高，但成绩波动更大"。
8. Probability Vocabulary | 概率词汇
Trial: a single performance of an experiment, such as one roll of a dice, and outcome is a single possible result of that trial.
试验：进行一次实验，例如掷一次骰子；而结果指该次试验的某一个可能取值。
Event: a set of one or more outcomes, for example "rolling an even number".
事件：一个或多个结果的集合，例如"掷出偶数"。
Sample space: the set of all possible outcomes of an experiment, often shown in a list, table or Venn diagram.
样本空间：实验中所有可能结果的集合，常用列表、表格或韦恩图表示。
Mutually exclusive events: events that cannot happen at the same time, so P(A ∩ B) = 0 and P(A ∪ B) = P(A) + P(B).
互斥事件：不能同时发生的事件，因此 P(A ∩ B) = 0，且 P(A ∪ B) = P(A) + P(B)。
Independent events: events where the outcome of one does not affect the probability of the other, so P(A ∩ B) = P(A) × P(B).
独立事件：一个事件的结果不影响另一个事件的概率，因此 P(A ∩ B) = P(A) × P(B)。
Complement: the event that A does not happen, written A', with P(A') = 1 − P(A).
对立事件（补事件）：A 不发生的那个事件，记作 A'，且 P(A') = 1 − P(A)。
Relative frequency: the number of times an event occurs divided by the total number of trials, used to estimate probability from experimental data.
相对频数：事件发生的次数除以试验总次数，用于根据实验数据估计概率。
Relative frequency = number of successes ÷ number of trials
Expected frequency number of times an event is predicted to occur, calculated as P(event) × number of trials.
期望频数：预测某事件将发生的次数，等于 P(事件) × 试验次数。
Tree diagram: a diagram showing the outcomes of two or more successive events, on which probabilities are multiplied along branches and added between them.
树状图：显示两个或多个相继事件结果的图形；沿分支相乘，不同分支之间相加。
Venn diagram: a diagram of overlapping circles used to show the relationship between two or three events and their probabilities.
<
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