UK Graduate Jobs and Earnings Data Update: How to Use It When Choosing a Degree | 英国毕业生就业与收入数据更新：选专业时如何参考

2026-10-15 13:09 · 本文部分内容由AI辅助创作

📚 UK Graduate Jobs and Earnings Data Update: How to Use It When Choosing a Degree | 英国毕业生就业与收入数据更新：选专业时如何参考

Every year, thousands of students choose a UK degree on the basis of reputation, guesswork, or what a relative happened to say at a family dinner. The next refresh of the UK's official graduate employment and earnings statistics offers something far better: large-scale, tax-record-based evidence about what graduates of each subject actually earn and actually do, several years after they leave university.

每年都有成千上万的学生依据声誉、猜测，或者亲戚在家庭聚餐上随口的一句话来选择英国学位。即将更新的英国官方毕业生就业与收入统计提供了更好的东西：基于税务记录的大规模证据，说明各专业的毕业生在离校数年之后，实际挣多少钱、实际在做什么工作。

This article explains what the updated data will contain, how to read it correctly, and how to turn it into a practical shortlist for your UCAS application. It is written for A-Level, IB and foundation students, and for the parents and counsellors advising them.

本文解释这次更新会包含哪些内容、如何正确解读，以及如何把它转化为 UCAS 申请中的实际选校选专业清单。文章面向 A-Level、IB 和预科学生，也适合为他们提供建议的家长与升学顾问阅读。

1. What Is Actually Being Updated | 这次更新的究竟是什么

The release is a refresh of official longitudinal statistics that link university student records to tax, benefits and student-loan data. In practice this means the government can follow an entire cohort of graduates, year by year, and report their earnings and employment status at fixed points after graduation.

这次发布的是官方纵向统计的更新版，它把大学学籍记录与税务、福利和学生贷款数据关联起来。实际效果是：政府可以逐届追踪整批毕业生，逐年报告他们在毕业后固定时间点上的收入与就业状况。

The key advantage of this approach is scale and objectivity. It is not a survey of a few thousand volunteers; it is administrative data covering hundreds of thousands of individuals.

这种方法的核心优势在于规模与客观性。它不是对几千名志愿者的问卷调查，而是覆盖数十万人的行政数据。

Subject-level earnings, by degree level — 按学历层次划分的专业层面收入数据
Institution-level comparisons, including value-added measures — 院校层面比较，包括"增值"指标
Breakdowns by gender, region and background — 按性别、地区与背景的细分数据
Employment rates and broad occupation groups — 就业率与职业大类分布

Because the dataset is longitudinal, the update does not simply show one snapshot. It shows trajectories: what a graduate earned one year out, three years out, five years out and ten years out.

由于该数据集是纵向的，更新内容并不只是某一时点的快照，而是轨迹：毕业生在离校一年、三年、五年和十年后分别挣多少。

2. Who Publishes What: LEO, Graduate Outcomes and Discover Uni | 谁发布什么：LEO、Graduate Outcomes 与 Discover Uni

Students often confuse three different sources. They measure different things, and mixing them up leads to bad decisions.

学生常常混淆三个不同的数据来源。它们测量的东西不同，混用会导致错误决策。

Source	What it measures	Coverage
LEO (Longitudinal Education Outcomes), Department for Education	Actual earnings from tax records, 1, 3, 5 and 10 years after graduation	England, mainly graduates who took student loans
Graduate Outcomes survey, HESA	Employment, study and job type 15 months after graduation, self-reported	UK-wide, all graduates including international
Discover Uni, Office for Students	Course-level summary combining both, plus student satisfaction	UK, searchable by course and university

For earnings, LEO is the strongest evidence base because it uses tax records rather than memory. For destination and job type, Graduate Outcomes is richer. Discover Uni is the most accessible starting point for a school student exploring options.

就收入而言，LEO 是最有力的证据基础，因为它使用税务记录而非个人回忆。就去向与岗位类型而言，Graduate Outcomes 信息更丰富。对正在探索选项的中学生而言，Discover Uni 是最容易上手的入口。

Check the official statistics release calendar before you rely on any figures. Data cycles are usually annual, and each new release revises previous estimates.

在依赖任何数字之前，请核对官方统计发布日历。数据周期通常为一年，每次新发布都会修订此前的估计值。

3. The Degree Premium by Level: Bachelor's, Master's, Doctoral | 学历溢价：本科、硕士与博士

A common assumption is that each additional degree level automatically adds a fixed amount to earnings. The data do not support that simplification.

一种常见假设是：每多读一个学历层次，收入就会自动增加一个固定数额。数据并不支持这种简化。

At the aggregate level, graduates earn substantially more than non-graduates over a working lifetime, and postgraduate qualifications add a further average premium. But that average conceals enormous variation. A master's in a subject with weak labour-market demand can pay less than a bachelor's in a high-demand subject.

在总体层面，毕业生一生收入明显高于非毕业生，研究生学历平均还会带来额外溢价。但这个平均值掩盖了巨大的差异。需求疲软专业的硕士，收入可能低于高需求专业的本科。

Level	Typical pattern	Main caveat
Bachelor's	Solid premium in most subjects; extremely wide spread between subjects	Subject choice matters more than institution
Master's	Positive average premium, but heavily concentrated in business, computing, economics, engineering	Some taught master's show little or no earnings gain
Doctoral	High overall, but with a slow start if entering academia	Huge difference between PhD in finance or AI and PhD in low-demand humanities fields

Note that the figures above describe orders of magnitude, not exact values. Always open the most recent official release for precise numbers, because medians shift with each cohort and each methodology revision.

请注意，上表描述的是量级而非精确数值。务必查阅最新官方发布以获得准确数字，因为中位数会随每届学生和方法修订而变化。

4. Subject Is the Biggest Single Lever | 专业是最大的单一杠杆

If you look at only one variable when planning your degree, look at subject. Across the official data, the subject you study explains far more of the variation in later earnings than the university you attend.

如果在规划学位时只能看一个变量，那就看专业。在官方数据中，你所学专业对日后收入差异的解释力，远大于你所就读的大学。

Certain subjects sit consistently at the top: medicine, dentistry, veterinary science, economics, mathematics, computing and most engineering disciplines. Others sit consistently lower on median earnings five years after graduation: creative arts, design, media, and some humanities and social science combinations.

某些专业稳定地处于高位：医学、牙医学、兽医学、经济学、数学、计算机以及大多数学科门类的工程专业。另一些专业在毕业后五年的收入中位数上则稳定偏低：创意艺术、设计、传媒，以及部分人文与社会科学组合。

Subject group (indicative)	Median earnings, approx. 5 years out
Medicine and dentistry	Highest band, roughly £45,000-£60,000 and rising steeply with seniority
Economics, mathematics, computing, engineering	Upper band, roughly £32,000-£48,000
Law, business and management, physical sciences	Mid band, roughly £27,000-£38,000
Biological sciences, social sciences, humanities	Lower-mid band, roughly £24,000-£32,000
Creative arts, design, media, performing arts	Lowest band, roughly £20,000-£26,000

These are indicative bands for England-domiciled graduates and are intended to show relative position, not precise predictions. Your own outcome depends on university, region, industry, and above all on what you do with the degree.

以上为学生贷款覆盖范围内的英格兰毕业生指示性区间，意在显示相对位置，而非精确预测。你个人的结果取决于院校、地区、行业，最重要的是你如何利用这个学位。

One crucial point: a lower median does not mean a subject is worthless. It means the typical path for that subject is different, and you must plan deliberately if you want a higher-earning route.

关键一点是：中位数较低并不意味着某专业没有价值。它意味着该专业的典型路径不同，如果你想走收入更高的路线，就必须有意识地规划。

5. Institution Effects: Real, but Smaller Than Subject | 院校效应：真实存在，但小于专业效应

Graduates of highly selective universities do earn more on average. However, once subject and prior attainment are controlled for, the institution premium shrinks considerably, and in some subject areas it nearly disappears.

高度择优录取大学的毕业生平均收入确实更高。但在控制专业和入学前成绩之后，院校溢价会显著缩小，在某些学科领域甚至几乎消失。

The official data now include value-added or "adjusted" measures, which compare a university's graduate earnings against what would be expected given its intake. This is a much fairer comparison than raw league tables.

官方数据现在包含增值或"调整后"指标，即把某校毕业生收入与根据其生源预期应有的水平进行比较。这比原始排行榜公平得多。

Raw median earnings favour universities with already-advantaged intakes — 原始收入中位数有利于生源本已占优的大学
Adjusted measures show which universities genuinely lift outcomes — 调整后指标显示哪些大学真正提升了学生结果
For most students, subject and course structure matter more than a few places in a ranking — 对多数学生而言，专业与课程结构比排名上的几个位次更重要

A strong student in a mid-ranked university studying computing will typically out-earn a strong student at a prestigious university studying a low-demand subject.

在中等排名大学读计算机的优秀学生，其收入通常高于在名校读低需求专业的优秀学生。

6. The Time Dimension: 1, 3, 5 and 10 Years Out | 时间维度：毕业后 1、3、5、10 年

剑桥AS&A Level完整力学进口原版图书教材国际考试教材牛津大学出版社

¥235.0

📱 去京东购买

京东精选

Earnings rankings change shape over time, and this is where the longitudinal data becomes genuinely useful.

收入排名会随时间改变形态，而这正是纵向数据真正有用的地方。

Point after graduation	What tends to happen
1 year	Clinical subjects and vocational courses lead immediately; arts and humanities often show low earnings partly because of further study and portfolio careers
3 years	Professional training routes kick in: law, accountancy, teaching, engineering chartership
5 years	Gap between subject groups widens; finance, technology and management routes accelerate
10 years	Highest earners are concentrated in finance, law, senior management, medicine and technology; some creative graduates also reach high earnings, but with far greater variance

The practical lesson is simple: do not judge a subject by its one-year figures alone. Many degrees with modest early earnings, such as law or architecture, require several years of training before the earnings curve rises.

实用结论很简单：不要仅凭毕业后一年的数字评判一个专业。许多早期收入平平的学位，例如法律或建筑，需要数年培训后收入曲线才会抬升。

Equally, do not assume that a slow start guarantees a strong finish. That is true for some subjects and false for others.

同样，也不要假设起步慢就必然终点好。对某些专业而言如此，对另一些则并不成立。

7. Gaps by Gender, Region and Background | 性别、地区与背景差距

The updated data will again show that earnings gaps within the same subject are substantial. These gaps are not a reason to avoid a subject, but they are a reason to be informed.

更新后的数据将再次显示，同一专业内部的收入差距相当大。这些差距不是回避某专业的理由，但却是了解情况的理由。

Within the same subject and institution, male graduates often show higher median earnings than female graduates, particularly ten years out. Part of this reflects different working patterns and career breaks, and part reflects differing sector and negotiation patterns.

在同一专业、同一院校内，男性毕业生收入中位数常高于女性毕业生，十年后尤为明显。其中一部分反映了工作模式与职业中断的差异，另一部分反映了行业分布与薪酬谈判方式的差异。

Region matters too. Graduates working in London and the South East typically earn more in nominal terms, but housing and living costs absorb much of that difference. A £32,000 salary in Manchester and a £40,000 salary in London are not as far apart in real terms as they appear.

地区同样重要。在伦敦和东南部工作的毕业生名义收入通常更高，但住房与生活成本会吃掉其中很大一部分差异。曼彻斯特的 £32,000 与伦敦的 £40,000，在实际购买力上的差距远没有看上去那么大。

Background also shapes outcomes: graduates who received free school meals or were the first in their family to attend university often show lower average earnings even within the same subject. This is a signal about networks and access, not about ability.

家庭背景也影响结果：曾领取免费校餐或家中第一代大学生的毕业生，即便在同一专业内，平均收入也往往更低。这反映的是人脉与机会获取，而非能力。

8. What These Numbers Cannot Tell You | 这些数字无法告诉你的事

Earnings data is powerful but narrow. Treating it as the only criterion for choosing a degree is a mistake in both directions: it over-weights salary and under-weights everything that determines whether you finish the course and thrive afterwards.

收入数据很有力，但视角狭窄。把它当作选专业的唯一标准，在两个方向上都出错：它高估了薪水的重要性，也低估了决定你是否能读完并发展良好的其他因素。

Job satisfaction, autonomy and meaning — 工作满意度、自主性与意义感
Unemployment risk and job security within a subject — 专业内部的失业风险与职业稳定性
Self-employment and freelance income, which tax records capture unevenly — 自雇与自由职业收入，税务记录对其捕捉并不完整
Career paths that require a postgraduate qualification before the payoff appears — 需要先取得研究生学历才有回报的职业路径
Small cohort sizes, where medians can be based on very few people — 小规模群体，其中位数可能仅基于极少数人

There is also a structural point: these statistics describe the past. By the time you graduate, the labour market for your subject may have changed. Use the data to understand the shape of a field, not to forecast your own salary.

还有一个结构性要点：这些统计描述的是过去。等你毕业时，你所在专业的劳动力市场可能已经改变。用这些数据理解一个领域的形态，而不是预测你个人的薪水。

9. International Students: Read the Caveats Carefully | 国际学生：务必谨慎解读

International students should be especially careful. The earnings data most widely quoted is drawn largely from England-domiciled graduates who took student loans, which means it does not represent international graduates.

国际学生需要格外谨慎。被引用最广的收入数据主要来自英格兰本地、曾申请学生贷款的毕业生，因此并不代表国际毕业生群体。

If you are an international student, prioritise three other sources of evidence.

如果你是国际学生，请优先关注另外三类证据。

Graduate Outcomes destination data, which includes international graduates — Graduate Outcomes 去向数据，其中包含国际毕业生
Visa and post-study work rules, which affect which jobs you can realistically take — 签证与毕业后工作规定，直接影响你实际能从事哪些工作
Employer sponsorship lists and sector hiring patterns in your target country — 目标国家的雇主担保名单与行业招聘规律

Also remember that many international graduates return home or move to a third country. Their earnings then reflect a different labour market entirely, and no UK dataset will capture that.

同时要记住，许多国际毕业生会回国或前往第三国。他们的收入反映的是完全不同的劳动力市场，任何英国数据集都无法涵盖。

10. A Practical Framework for Choosing a Course | 选专业的实操框架

Here is a five-step method that converts official data into a decision you can defend at interview and live with for three years.

以下是一个五步方法，把官方数据转化为一个你可以在面试中自圆其说、并愿意与之共处三年的决定。

Step 1: List your hard constraints. Budget, visa conditions, location, subject prerequisites, and any genuine non-negotiables.

第一步：列出硬性约束。预算、签证条件、地理位置、专业先修要求，以及任何真正不可妥协的事项。

Step 2: Check subject-level earnings, not just institution. Look at the median and, more importantly, the 25th and 75th percentiles of your intended subject.

第二步：查专业层面的收入，而不只是院校。看你目标专业的中位数，更重要的是第 25 百分位和第 75 百分位。

Step 3: Check employment and further-study rates. A subject with modest early earnings but a high rate of progression to professional training may be a strong long-term choice.

第三步：查看就业与升学率。早期收入平平但进入专业培训比例很高的专业，长期看可能是优选。

Step 4: Look at the dispersion, not just the average. A subject where the middle half earns between £22,000 and £55,000 offers very different risk to one where the middle half earns £28,000 to £34,000.

第四步：看分布离散度，而不只是平均值。中间一半人收入在 £22,000 到 £55,000 的专业，其风险特征与中间一半在 £28,000 到 £34,000 的专业完全不同。

Step 5: Test the reality. Speak to two recent graduates and one employer. Twenty minutes of conversation often changes more than twenty hours of spreadsheet reading.

第五步：检验现实。与两位近期毕业生和一位雇主交谈。二十分钟的对话，往往比二十小时看表格更能改变你的判断。

11. Common Misreadings to Avoid | 需要避免的常见误读

Most bad decisions based on this data come from a handful of repeated errors.

基于这些数据的糟糕决定，大多源于少数几个反复出现的错误。

Confusing median with average — 把中位数与平均数混为一谈。极端高收入者会拉高平均数，中位数才反映"典型"毕业生
Ignoring selection effects — 忽视选择效应。进入某专业的本来就更可能是某类学生，收入差异不全是专业造成的
Comparing a single year across different release years — 用不同发布年份的单年数据做横向比较
Reading correlation as causation — 把相关当作因果。数据说明"读这个专业的人挣多少"，而非"读这个专业会让你挣多少"
Forgetting cost of living — 忘记生活成本。名义收入必须放在地区物价下解读
Overlooking the 25th percentile — 忽视第 25 百分位。它展示了这条路走得不顺时的下限

One final caution: no dataset can tell you whether you will enjoy a subject for three years, or whether you will still be curious about it at thirty-five. That judgement remains yours.

最后一个提醒：没有任何数据集能告诉你，你是否会喜欢一个专业三年，或者到三十五岁时是否仍对它保持好奇。这个判断仍然属于你自己。