📚 Gender, Age, Ethnicity and Region as Factors in Voting Behaviour: An AS Mathematical Approach | 性别、年龄、种族和地区作为投票行为影响因素：AS数学视角
In Edexcel AS Mathematics, the statistics component equips students with tools to analyse real-world data. One compelling context is the study of voting behaviour, turnout and trends, where categorical variables like gender, age, ethnicity and region play a key role. By applying techniques such as two-way tables, percentage comparisons, stratified sampling and time series graphs, we can quantify and display how these factors relate to electoral outcomes. This article explores how AS Mathematical methods can be used to investigate voting patterns, without straying into political science, but instead focusing on data handling, interpretation and the limitations of statistical analysis.
在Edexcel AS数学中，统计部分为学生提供了分析真实世界数据的工具。投票行为、投票率和趋势是一个引人入胜的背景，其中性别、年龄、种族和地区等分类变量发挥着关键作用。通过应用双向表、百分比比较、分层抽样和时间序列图等技术，我们可以量化并展示这些因素与选举结果的关系。本文探讨如何运用AS数学方法研究投票模式，不涉足政治学本身，而是聚焦于数据处理、解读以及统计分析的局限性。
1. Understanding Variables: Categorical vs Numerical | 理解变量：分类变量与数值变量
In any statistical investigation we must first identify the types of variables involved. Gender, ethnicity and region are nominal categorical variables because they have no intrinsic order. Age, when grouped into brackets such as 18–24, 25–34, can be treated as an ordinal categorical variable. Voting behaviour could be recorded as 'voted'/'did not vote' (binary categorical) or by party choice (nominal). Recognising variable types determines which charts and summary statistics are appropriate.
在任何统计调查中，我们必须首先识别所涉及变量的类型。性别、种族和地区属于名义分类变量，因为它们没有内在顺序。年龄如果按18–24岁、25–34岁等分组，可视为有序分类变量。投票行为可记录为“已投票”/“未投票”（二分类）或按政党选择（名义）。认识到变量类型决定了哪些图表和汇总统计量是合适的。
For AS level, you are expected to distinguish between qualitative (categorical) and quantitative data. Since all our factors are categorical, we rely on frequency counts, proportions and percentages, not on measures like mean or standard deviation. For instance, reporting that 62% of women voted in an election compared to 58% of men is a meaningful comparison.
在AS水平，你应能区分定性数据（分类）和定量数据。由于我们所有因素都是分类的，我们依赖于频数、比例和百分比，而不是均值或标准差等度量。例如，报告一场选举中62%的女性投了票，而男性为58%，就是一个有意义的比较。
When collecting data on voting behaviour, we must define the population (e.g. all registered voters in a constituency) and the sampling frame. The response variable is often turnout (voted or not) or party preference, while the explanatory variables are the demographic factors. This framework underpins the rest of the analysis.
在收集投票行为数据时，我们必须定义总体（例如某选区的所有登记选民）和抽样框。响应变量通常是投票率（投票与否）或政党偏好，而解释变量是人口统计因素。这一框架构成了后续分析的基础。
2. Data Collection and Sampling Methods | 数据收集与抽样方法
Opinion polls and election studies rely on sampling because surveying the entire voting population is impractical. In AS Mathematics you learn about simple random sampling, systematic sampling, stratified sampling and quota sampling. For investigating how gender or region influences turnout, stratified sampling is especially useful: the population is divided into strata (e.g. male/female or by age group), and a random sample is taken from each stratum in proportion to its size.
民意调查和选举研究依赖于抽样，因为调查全部选民是不现实的。在AS数学中，你会学习简单随机抽样、系统抽样、分层抽样和配额抽样。在调查性别或地区如何影响投票率时，分层抽样尤为有用：先将总体分为层（如男/女或年龄组），然后按各层大小比例随机抽取样本。
When using a stratified sample by age, you ensure that the sample mirrors the age structure of the electorate. If 18–24-year-olds make up 12% of registered voters, then 12% of your sample should be from that group. This improves the accuracy of estimates for each subgroup and avoids sampling bias that could arise from a simple random sample accidentally underrepresenting minority ethnic groups or certain regions.
当使用按年龄分层抽样时，你确保样本反映了选民的年龄结构。如果18–24岁人群占登记选民的12%，则你样本的12%应来自该组。这提高了每个子群体估计的准确性，并避免了简单随机样本偶然低估少数族裔或某些地区而带来的抽样偏差。
Questionnaire design is another aspect: questions must be clear and unbiased. For instance, asking 'Did you vote in the last general election?' can be affected by response bias—people may over-report voting. Mathematically, we acknowledge that data may contain non-sampling errors which affect the validity of our conclusions.
问卷设计是另一个方面：问题必须清晰且无偏。例如，询问“您在上次大选中投票了吗？”可能受到回答偏差的影响——人们可能高报投票行为。从数学角度，我们承认数据可能含有非抽样误差，这会影响结论的有效性。
3. Organising Data: Two-Way Tables | 组织数据：双向表
A two-way table (contingency table) is a fundamental tool for showing the relationship between two categorical variables. Suppose we cross-tabulate gender against whether someone voted. The table might look like this hypothetical data from a survey of 1000 respondents:
双向表（列联表）是展示两个分类变量之间关系的基本工具。假设我们将性别与是否投票进行交叉制表，以下是一个假设的来自1000名受访者调查的数据表：
| Gender / 性别 | Voted / 已投票 | Did Not Vote / 未投票 | Total / 合计 |
|---|---|---|---|
| Male / 男 | 320 | 180 | 500 |
| Female / 女 | 370 | 130 | 500 |
| Total / 合计 | 690 | 310 | 1000 |
From this table we can calculate row percentages: 64% of men voted (320/500), whereas 74% of women voted (370/500). Similarly, we can compute column percentages to see that among voters, 46.4% were male and 53.6% female. Such percentage differences provide the numerical evidence for statements like 'women had a higher turnout rate than men in this survey.'
从该表我们可以计算行百分比：64%的男性投了票（320/500），而74%的女性投了票（370/500）。同样，我们可以计算列百分比，发现在投票者中，46.4%是男性，53.6%是女性。这样的百分比差异为“在该调查中女性投票率高于男性”等陈述提供了数字证据。
For AS Mathematics, you may be asked to complete missing entries in a two-way table or to use given percentages to work backwards to frequencies. Always check that totals are consistent and take care with rounding. Two-way tables can also be extended to include ethnicity or age groups, though large tables become harder to read.
对于AS数学，你可能需要填写双向表中的缺失项，或利用给定的百分比反向推算频数。始终检查总数是否一致，并注意四舍五入。双向表也可以扩展到包括种族或年龄组，但过大的表格会变得难以阅读。
4. Visualising Age-Based Turnout Trends | 可视化基于年龄的投票率趋势
Age is often grouped into bands: 18–24, 25–34, 35–44, 45–54, 55–64, 65+. Turnout rates typically increase with age, a pattern that can be displayed effectively using a bar chart or a line graph. Since age groups are ordered, a line graph connecting the midpoints of each age band is appropriate to show the trend.
年龄通常按区间分组：18–24岁、25–34岁、35–44岁、45–54岁、55–64岁、65岁以上。投票率通常随年龄增长而增加，这一模式可以用条形图或折线图有效展示。由于年龄组是有序的，连接每个年龄区间中点的折线图适合展示趋势。
Consider a dataset where the turnout percentages are: 18–24: 47%, 25–34: 54%, 35–44: 63%, 45–54: 72%, 55–64: 78%, 65+: 82% e. You could calculate the percentage point increase from the youngest to oldest group: 82% – 47% = 35 percentage points.
假设一组数据中投票率百分比为：18–24岁: 47%，25–34岁: 54%，35–44岁: 63%，45–54岁: 72%，55–64岁: 78%，65岁以上: 82%。以年龄组中点（如21, 30, 40, 50, 60, 75）为横轴绘制的折线图显示出明显的上升斜率。你可以计算从最年轻到最年长组的百分点增幅：82% – 47% = 35个百分点。
But a common mistake is to treat the sequence as a continuous numerical variable and calculate a mean age, which would be meaningless without the raw data. Instead, describe the trend: 'there is a positive association between age and turnout, with each successive age band showing a higher voting rate than the previous one.'
但一个常见错误是将序列视为连续数值变量并计算平均年龄，这在没有原始数据的情况下是无意义的。相反，应描述趋势：“年龄与投票率之间存在正相关，每个连续的年龄组都比前一组显示出更高的投票比例。”
Comparing turnout across age groups also requires caution: if a sample has very few respondents in the 18–24 band, the estimated turnout percentage will have a larger margin of error. You might discuss sample sizes alongside the percentages to give a complete picture.
比较不同年龄组的投票率也需要慎重：如果样本中18–24岁的受访者非常少，则估计的投票率百分比会有更大的误差范围。你可能需要在百分比旁边讨论样本量，以提供完整的图景。
5. Ethnicity and Voting Behaviour: Using Segmented Bar Charts | 种族与投票行为：使用分段条形图
Ethnicity, as a nominal variable, can be examined by calculating the percentage of each ethnic group voting for different parties or simply turning out to vote. Because the sizes of ethnic groups in the population differ widely, relative frequencies are essential. A segmented (stacked) bar chart allows comparison of distributions across groups.
种族作为一个名义变量，可以通过计算每个族裔群体投票给不同政党的百分比或简单投票率来进行研究。由于人口中各族裔群体的大小差异很大，相对频率是必不可少的。分段（堆叠）条形图允许比较不同群体之间的分布。
Create a 100% stacked bar chart where each bar represents an ethnic group, and the segments show the proportion voting for Party A, Party B, Party C, etc. This emphasises the composition of the vote within each group. Alternatively, side-by-side bar charts can compare turnout rates directly. For instance, if data shows that 65% of Group X voted compared to 58% of Group Y, a simple bar chart with two bars makes the difference immediately visible.
制作一个100%堆叠条形图，其中每个条形代表一个族裔群体，各段显示投票给政党A、政党B、政党C等的比例。这突出了每个群体内选票的构成。或者，并排条形图可以直接比较投票率。例如，如果数据显示X群体有65%的人投票，而Y群体为58%，一个有两个条形的简单条形图就能立即使差异一目了然。
Do not rely on raw frequencies alone. A statement like 'more votes from ethnic group X than Y' may simply be because group X has a larger population. By using percentages, you are making a like-for-like comparison. Mathematically, this is a key skill for interpreting contingency tables and graphical representations.
不要仅依赖原始频数。像“来自X族裔的选票多于Y族裔”这样的陈述可能仅仅是因为X群体人口更多。通过使用百分比，你进行的是同类比较。从数学上讲，这是解读列联表和图形表现的关键技能。
6. Regional Variations in Turnout and Voting Patterns | 投票率和投票模式的地区差异
Region is another nominal categorical factor. In the UK, turnout can vary significantly between London, the South East, the Midlands, the North, Scotland, Wales and Northern Ireland. A choropleth map (though not examinable in AS maths as a construction task) can be briefly described, but mathematically we use comparative bar charts or table summaries.
地区是另一个名义分类因素。在英国，伦敦、东南部、中部地区、北部、苏格兰、威尔士和北爱尔兰之间的投票率可能存在显著差异。虽然等值区域图在AS数学考试中不作为作图任务考察，但可简单描述；从数学角度我们使用比较条形图或表格摘要。
Suppose you sample three regions: Region A turnout 71%, Region B 66%, Region C 59%. The range (71% – 59% a measure of spread. You could also calculate the overall turnout and compare each region' average. This aligns with the AS skill of comparing distributions using simple statistics.
假设你抽样三个地区：A地区投票率71%，B地区66%，C地区59%。极差（71% – 59% = 12个百分点）提供了离散度的度量。你还可以计算总体投票率，并比较每个地区与平均值的偏差。这与AS技能中使用简单统计量比较分布是一致的。
A more detailed analysis might look at how region interacts with other variables. For example, does the relationship between age and turnout hold equally in all regions? This is called interaction, and while AS does not require formal testing, you can still comment on whether patterns appear consistent in two-way tables broken down by region.
更详细的分析可能考察地区如何与其他变量相互作用。例如，年龄与投票率的关系在所有地区是否都一致？这称为交互作用，虽然AS不要求正式检验，但你仍然可以根据按地区分组的双向表评述模式是否一致。
7. Time Series and Trends Over Elections | 时间序列与选举间的趋势
Voting behaviour trends over time are best shown with time series graphs. Plotting turnout percentages for successive general elections (e.g. 2001, 2005, 2010, 2015, 2017, 2019) reveals overall trends: perhaps a decline from 2001 to 2005, then a gradual rise. For AS Mathematics, you should be able to plot points, connect them with straight lines, and comment on the trend.
随时间变化的投票行为趋势最适合用时间序列图展示。绘制历次大选（如2001、2005、2010、2015、2017、2019）的投票率百分比，可以发现总体趋势：例如从2001年到2005年下降，随后逐渐上升。对于AS数学，你应该能够描点、用直线连接并评论趋势。
If we break down turnout by gender over several elections, we can plot two time series on the same axes. Use different line styles or colours and a key. Observing that the gap between male and female turnout has narrowed over time is a valid interpretation. Mathematically, you may be asked to calculate moving averages to smooth out irregular fluctuations and clarify the underlying trend.
如果我们将若干次选举的投票率按性别分解，我们可以在同一坐标轴上绘制两条时间序列。使用不同的线型或颜色并加图例。观察到男女投票率差距随着时间的推移而缩小，这是一个有效的解读。从数学上讲，你可能被要求计算移动平均数，以平滑不规则波动并厘清潜在趋势。
A moving average of order 3 for yearly data would be: (value₁ + value₂ + value₃)/3, then shift one time period. For voting behaviour, a moving average can help reveal whether a change is a blip or part of a steady trend. This technique is explicitly part of the Edexcel AS specification.
对于年度数据，3阶移动平均数为：(value₁ + value₂ + value₃)/3，然后向前移动一个时间段。对于投票行为，移动平均数有助于揭示变化是短暂波动还是稳定趋势的一部分。这一技术明确属于Edexcel AS的考纲内容。
8. Correlation Does Not Imply Causation | 相关性不意味着因果关系
When we find, for example, that older people vote more, or that a particular ethnic group has a lower turnout, it is tempting to say age or ethnicity causes the difference. However, from a purely statistical standpoint, all we have quantified is an association. There may be confounding variables such as income, education, or political engagement that are linked to both age/ethnicity and voting likelihood.
当我们发现，例如老年人投票更多，或某个族裔群体投票率偏低，很容易说年龄或种族导致了这种差异。然而，从纯统计的角度来看，我们所量化的只是一种关联。可能存在混杂变量，如收入、教育或政治参与度，它们同时与年龄/种族和投票可能性相关。
In AS Mathematics, you learn to distinguish between correlation and causation, especially when interpreting scatter diagrams and regression lines. Although categorical variables are not displayed on scatter plots, the logic applies: observed differences between groups may be due to other underlying factors. When writing conclusions, use cautious language: 'the data suggests that...' rather than 'this proves that...'
在AS数学中，你学习区分相关性和因果关系，特别是在解读散点图和回归线时。虽然分类变量不展示在散点图上，但逻辑是相通的：观察到的组间差异可能由其他潜在因素导致。撰写结论时，请使用谨慎的语言：“数据表明...”而非“这证明了...”。
9. Probability and Expected Frequencies | 概率与期望频率
d frequency of female voters = 0.5 × 0.69 × 1000 = 345. Actual observed was 370. Thus more women voted than expected under independence.
利用第3节的双向表，我们可以引入独立性的概念。如果性别与投票独立，那么一个人既是女性又是投票者的概率将是 P(女性) × P(投票者)。根据总计：P(女性)=500/1000=0.5，P(投票者)=690/1000=0.69，所以女性投票者的期望频率 = 0.5 × 0.69 × 1000 = 345。实际观察值为370。因此，女性投票人数多于独立条件下的期望值。
This difference between observed and expected helps quantify how strong the association is. In AS Mathematics you do not need to perform a chi-squared test, but you should be able to calculate expected frequencies and comment on the difference. Similar calculations can be done for age groups or ethnicity against turnout, providing a numerical basis for statements like 'the 65+ age group turned out to vote far more than expected if age and turnout were independent.'
观察值与期望值之间的这种差异有助于量化关联的强度。在AS数学中，你不需要进行卡方检验，但应能计算期望频率并评论差异。类似的计算也可以针对年龄组或种族与投票率进行，为诸如“如果年龄与投票率独立，那么65岁以上年龄组的投票人数远高于期望”等陈述提供数字依据。
10. Using Technology: Spreadsheets and Calculator | 使用技术：电子表格与计算器
Modern statistical work often involves spreadsheets. For this topic, you could use Excel or Google Sheets to input frequency data, generate bar charts and stacked percentage charts, and produce time series plots with trendlines. The 'Chart' function allows quick visualisation of variables like gender, age and region against turnout.
现代统计工作经常涉及电子表格。针对本主题，你可以使用Excel或Google表格输入频率数据，生成条形图和堆叠百分比图，并制作带趋势线的时间序列图。“图表”功能可以快速将性别、年龄和地区等变量与投票率进行可视化比较。
Your graphical calculator (e.g. Casio fx-CG50 or TI-84) can also handle two-way frequency tables and plot comparative bar charts. For the AS exam, you may be required to read values from given diagrams rather than draw them from scratch, but practising with technology reinforces understanding. Additionally, you can use calculator functions to sum frequencies and verify percentage calculations quickly.
你的图形计算器（如Casio fx-CG50或TI-84）也可以处理双向频数表并绘制比较条形图。在AS考试中，你可能需要从给定的图中读取数值，而不是从头绘制，但利用技术进行练习可以加深理解。此外，你可以使用计算器功能快速求和频率并验证百分比计算。
11. Interpreting Differences and Drawing Valid Conclusions | 解读差异并得出有效结论
After analysing the data, you must be able to communicate findings clearly. Use statements such as: 'In this sample, a higher proportion of women than men reported voting, a difference of 10 percentage points. However, this estimate is subject to sampling variability, so the true population difference may be slightly larger or smaller.'
在分析数据之后，你必须能够清晰地传达发现。使用如下的陈述：“在这个样本中，女性报告投票的比例高于男性，相差10个百分点。然而，该估计受到抽样变异的影响，因此真实的总体差异可能稍大或稍小。”
Avoid overgeneralising. A survey conducted in one region or with a small sample may not reflect national patterns. Recognising limitations, such as non-response bias (certain groups may be harder to reach) and the snapshot nature of polls, is part of rigorous statistical practice.
避免过度概括。在一个地区进行或以小样本进行的调查可能无法反映全国模式。认识到局限性，如无应答偏差（某些群体可能更难接触到）和民意调查的快照性质，是严谨统计实践的一部分。
Remember that in Edexcel AS exams, questions often ask you to 'comment on' or 'interpret' a given chart or table. Support your comments with numerical evidence: quote percentages, differences, and refer to sample sizes. This is more convincing than vague statements.
请记住，在Edexcel AS考试中，问题常要求你“评论”或“解读”给定的图表或表格。用数字证据支持你的评论：引用百分比、差异，并提及样本量。这比模糊的陈述更具说服力。
12. Summary and Exam Tips | 总结与考试技巧
The analysis of gender, age, ethnicity and region as factors in voting behaviour provides a rich context for practising AS statistical skills: identifying variable types, designing sampling methods, constructing and interpreting two-way tables, computing percentages and expected frequencies, drawing time series, and critically evaluating findings. Always link your mathematical steps back to the context.
分析性别、年龄、种族和地区作为投票行为的影响因素，为练习AS统计技能提供了丰富的背景：识别变量类型、设计抽样方法、构建和解读双向表、计算百分比和期望频率、绘制时间序列以及批评性地评估发现。在数学步骤中始终要联系实际背景。
Key exam advice: label your charts clearly, show percentage calculations step-by-step, mention sample size when comparing groups, and use cautious language about correlation and causation. With these tools, you'll be well-prepared to tackle any data-focused question in the Edexcel AS Mathematics paper.
关键考试建议：图表标注要清晰，逐步展示百分比计算，在比较群体时提及样本量，并对相关性和因果关系使用谨慎的语言。有了这些工具，你将为应对Edexcel AS数学试卷中任何以数据为中心的问题做好充分准备。
Published by TutorHao | Mathematics 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