📚 Graduate Retention Rates: The Student Retention Indicator in City Rankings | 毕业后留下比例：城市排名的学生留存指标
Every summer, hundreds of thousands of students graduate from universities in London, Manchester, Wuhan, Toronto and Berlin, and within twelve months most of them make one quiet but economically enormous decision: stay in the city where they studied, or leave. That decision, aggregated across a whole cohort, becomes a headline number in city rankings, university destination reports and government talent strategies. It is called the graduate retention rate, and it is one of the most quoted, most contested and most poorly understood indicators in the geography and economics of higher education.
每年夏天，伦敦、曼彻斯特、武汉、多伦多和柏林都有成千上万名学生毕业。在接下来的十二个月里，他们中的大多数人会做出一个安静却极具经济分量的决定：留在读书的城市，还是离开。把整个年级的这一决定加总起来，就变成了城市排名、大学去向报告和政府人才战略中的头条数字，这就是”毕业生留存率”。它是高等教育地理与经济学中最常被引用、最受争议，也最常被误解的指标之一。
1. What Graduate Retention Actually Means | 什么是毕业生留存率
Graduate retention is the proportion of graduates who received a degree from an institution located in a given city and who are still living, working, or both, in that same city at a defined moment after graduation. It is a flow indicator rather than a stock indicator: it does not tell you how many graduates live in a city in total, but how many of the graduates a city produces through its universities remain inside the local labour market.
毕业生留存率指的是：在某城市高校获得学位的毕业生中，在毕业后某个既定时间点仍在该城市居住、工作或两者兼有的人所占的比例。它是流量指标而不是存量指标：它并不告诉你这座城市总共有多少毕业生，而是告诉你这座城市通过大学”生产”出来的人才，有多少留在了本地劳动力市场。
Three distinctions are essential before quoting any figure.
在引用任何数字之前，必须先厘清三组区别。
- Student retention versus graduate retention. Student retention describes whether a student continues studying or stays at the same institution; graduate retention describes whether a graduate stays in the city after finishing. Rankings commentary routinely blurs the two.
学生留存与毕业生留存。学生留存描述学生是否继续学业或留在同一所学校；毕业生留存描述毕业生毕业后是否留在该城市。排名的评论文章经常把两者混为一谈。 - Residence-based versus workplace-based retention. A graduate may live in a commuter town and work in the city, or live in the city and work remotely for a firm abroad. Each definition produces a different rate.
按居住地界定与按工作地界定。毕业生可能住在卫星城而在市区上班，也可能住在市区却为海外公司远程工作。两种定义会得出不同的留存率。 - Intra-city retention versus intra-region retention. A graduate who moves from Manchester to Salford has left the city but not the region. Metropolitan-area definitions therefore inflate retention relative to municipal boundaries.
市域留存与区域留存。从曼彻斯特搬到索尔福德的毕业生离开了该市，却没有离开该区域。因此，按大都市区界定会比按市政边界界定得出更高的留存率。
2. Why Cities Compete for Graduates | 城市为何争夺毕业生
Human capital theory gives the simplest answer: productivity, innovation and wages rise where skilled workers cluster. Graduates are the cheapest skilled workers a city can acquire, because the city did not pay for their schooling but captures the tax revenue, consumption and knowledge spillovers that follow. Agglomeration economics adds a second layer: each additional graduate raises the returns to other skilled workers, so the benefit is cumulative rather than linear.
人力资本理论给出最简单的答案：技能型劳动者聚集的地方，生产率、创新与工资都会上升。毕业生是城市能获得的最便宜的技能型劳动力，因为城市没有为他们的教育付费，却能获得随之而来的税收、消费和知识溢出。集聚经济学则提供了第二层解释：每增加一名毕业生，都会提高其他技能型劳动者的回报，因此收益是累积的，而不是线性的。
The fiscal arithmetic reinforces the argument.
财政账本进一步强化了这一论点。
- A graduate typically pays more in income tax and social contributions across a working life than the local public services they consume, particularly in cities with a progressive tax system.
在一生的职业生涯中，毕业生缴纳的所得税和社会保险费通常高于其消耗的地方公共服务成本，在累进税制的城市尤为明显。 - Graduates delay retirement pressures: cities with shrinking native populations use graduate in-migration to keep their dependency ratio stable.
毕业生能延缓养老压力：本地人口萎缩的城市通过毕业生流入来维持抚养比的稳定。 - Graduate-intensive sectors, such as software, finance, life sciences and creative industries, are precisely the sectors that local industrial strategies target.
软件、金融、生命科学和创意产业等毕业生密集型行业，恰恰是地方产业战略重点争夺的行业。
The competition is therefore not symbolic. Cities design housing quotas, visa routes, wage subsidies and cultural amenities explicitly to shift the stay-or-leave decision.
因此，这种竞争并非象征性的。城市会通过设计住房配额、签证通道、工资补贴和文化设施，明确地影响毕业生的”留下或离开”决策。
3. The Core Formula and Its Components | 核心公式与构成
The standard measure is simple arithmetic, but every term is a definitional choice.
标准测量方式只是简单的算术，但每一个项都是一次定义上的选择。
Retention Rate (%) = (G_t ÷ C₀) × 100
Here G_t is the number of graduates from cohort C₀ who are living and/or working in the city at time t after graduation. Two derived measures are used constantly in exam answers and consultancy reports.
其中 G_t 是同期群 C₀ 中在毕业后第 t 时间点仍在该城市居住和/或工作的人数。在考试答案和咨询报告中，另有两个衍生指标被频繁使用。
Leakage Rate (%) = 100 − Retention Rate
Net Retention (%) = ((S_t + R_t − L_t) ÷ C₀) × 100
In the second formula S_t is stayers, R_t is returners who come back to the city after a period away, and L_t is leavers. Net retention is the honest version of the metric, because gross retention hides churn: a city can show a stable 45% retention rate while a third of its graduates leave and are replaced by returning or in-migrating graduates.
在第二个公式中，S_t 是留下者，R_t 是离开一段时间后又回到该城市的人，L_t 是离开者。净留存是更诚实的版本，因为毛留存掩盖了人员流动：一座城市可能显示稳定的 45% 留存率，但实际上有三分之一的毕业生离开，又被回流的毕业生补齐。
4. Cohort Definitions and the Denominator Problem | 同期群界定与分母难题
The denominator is where most retention statistics quietly break down. If the cohort is defined as all graduates of city institutions, it includes international students who were never eligible to stay, part-time and distance learners who never lived in the city, and students whose visa expires before the measurement date. Each of these groups mechanically depresses the rate.
分母是大多数留存统计悄然失灵的地方。如果把同期群定义为该市高校的全部毕业生，就会把本来就没有资格留下的国际学生、从未在该市居住的非全日制和远程学习者，以及签证在测量日期之前就到期的学生都算进去。这些群体中的每一类都会机械性地拉低留存率。
Sound studies therefore restrict the denominator, and the choice of restriction changes the result dramatically.
因此，严谨的研究会限定分母，而限定的方式会显著改变结果。
- Domicile filter: excluding graduates whose home country gives them no right to remain.
生源地过滤：排除那些祖国身份使其没有权利留下的毕业生。 - Eligibility filter: including only graduates with a post-study work route, such as the UK Graduate Route or Canada’s post-graduation work permit.
资格过滤：只纳入拥有毕业后工作通道的毕业生，例如英国毕业生签证或加拿大的毕业后工作许可。 - Survey filter: restricting the sample to respondents, which introduces non-response bias. Destination surveys often achieve response rates below 60%, and leavers are systematically less likely to answer.
调查过滤：只保留回答了问卷的样本，这会引入无应答偏差。去向调查的应答率常低于 60%，而离开的人在回答上系统性偏低。
A useful classroom rule: whenever a retention figure is quoted, ask who is in the denominator and who is not.
一条实用的课堂规则：只要有人引用留存率，先问清楚分母里包含了谁、又排除了谁。
5. Time Horizons: 6, 12 and 36 Months | 时间窗口：6、12、36 个月
Retention is not a single number but a curve. Most graduates leave within the first two years, and the steepest decline happens between six and eighteen months after graduation, when fixed-term graduate schemes end and the first genuine housing decision is made.
留存不是一个单一数字，而是一条曲线。大多数毕业生在前两年内离开，而下降最陡峭的区间是毕业后的第 6 至第 18 个月，此时固定期限的毕业生项目结束，第一次真正的住房决策也随之到来。
| Time horizon | What it captures | Main distortion |
| 6 months | Inertia and short graduate contracts | Far too early to represent a career choice |
| 12 to 15 months | First settled job; the UK Graduate Outcomes standard | Still captures temporary post-study visa periods |
| 36 months | Job mobility, housing formation, family ties | Reliable data are scarce and expensive |
| 5 to 10 years | True long-run attachment to a regional labour market | Sample attrition makes estimates fragile |
For data-response questions, the safest evaluative sentence is that a twelve-month rate measures early-career friction while a three-year rate measures labour-market attachment.
在数据题中，最稳妥的评述句是：12 个月的留存率衡量的初入职场的摩擦，而 3 年留存率衡量的是对劳动力市场的依附程度。
6. How Major City Rankings Treat Retention | 主流城市排名如何处理留存指标
Most global student-city rankings do not publish a dedicated graduate retention indicator. Retention enters indirectly, through employer activity, desirability and student voice, or it is estimated by the ranking body from national survey data. This is why two rankings can disagree about the same city without either being wrong.
大多数全球学生城市排名并不单独公布毕业生留存指标。留存是通过雇主活跃度、吸引力与学生评价间接进入的，或者由排名机构根据全国性调查数据推算。这就是为什么两份排名可能对同一座城市给出不同结论，而两者都不算错。
| Ranking or dataset | Treatment of retention | Caution |
| QS Best Student Cities | Indirect, via employer activity, desirability and student voice | Survey-weighted perceptions, not an administrative count |
| National graduate outcome surveys | Direct, based on location of work or study about 15 months after graduation | Response bias; city inferred from postcode or workplace |
| University destination reports | Direct but institution-specific | Universities have an incentive to present themselves favourably |
| Commercial talent competitiveness indices | Composite “talent retention” pillar | Methodology is proprietary and rarely reproducible |
The ranking lesson is that retention is usually a derived, modelled or surveyed quantity, never a census of where people actually sleep.
关于排名的结论是：留存率通常是一个推算、建模或调查得来的量，而不是对人们实际居住地的普查。
7. Adjusting for Salary and Cost of Living | 薪资与生活成本调整
Nominal salaries mislead. A graduate earning £32,000 in an expensive capital may be worse off than one earning £24,000 in a regional city, and the retention decision responds to real, not nominal, income.
名义工资会误导判断。在昂贵首都年薪 32,000 英镑的毕业生，处境可能不如在地方城市年薪 24,000 英镑的毕业生，而留存决策回应的是实际收入，不是名义收入。
Real Graduate Income = Nominal Salary ÷ (Cost-of-Living Index ÷ 100)
Housing is the dominant component of that index for recent graduates, because rent typically absorbs 30% to 45% of a young graduate’s net pay in high-demand cities. A practical extension is a housing-adjusted retention pressure index.
对刚毕业的学生而言，住房是这一指数中最重要的组成部分，因为在需求旺盛的城市，房租通常占年轻毕业生税后收入的 30% 到 45%。一个实用的延伸是”住房调整后的留存压力指数”。
Housing Pressure = (Median Monthly Rent ÷ Median Monthly Graduate Net Pay) × 100
When this ratio exceeds roughly 40, retention in the early years tends to fall even if salaries are high, because graduates cannot accumulate the deposit needed to settle. Cities that combine high nominal wages with severe housing constraints, such as London, Vancouver or San Francisco, therefore show strong graduate inflow and comparatively weak long-run retention.
当这一比率超过约 40 时，即使工资很高，早期留存率也往往下降，因为毕业生无法积累定居所需的存款。因此，伦敦、温哥华或旧金山这类”高名义工资 + 严重住房约束”的城市，会呈现出强劲的毕业生流入与相对疲软的长期留存并存的现象。
8. Push and Pull Factors Behind the Stay-or-Leave Decision | 留下或离开背后的推拉因素
Retention is the aggregate outcome of individual decisions, and those decisions can be modelled as a comparison between the value of staying and the value of leaving, minus the cost of moving.
留存是个体决策的加总结果，而这些决策可以被建模为”留下的价值”与”离开的价值”之间的比较，再减去迁移成本。
Stay if (V_stay − V_leave) > Switching Cost
The pull factors that raise V_stay and the push factors that lower it are not symmetrical, and this matters for policy.
提高 V_stay 的拉力与降低 V_stay 的推力并不对称，这对政策制定很重要。
- Pull: graduate job density, real wages, affordable rental supply, transport quality, social and romantic networks formed at university, visa security, cultural amenities.
拉力：毕业生岗位密度、实际工资、可负担的租赁住房供给、交通质量、大学期间形成的社会与情感网络、签证稳定性、文化设施。 - Push: rent inflation, congestion, limited internal promotion, family obligations elsewhere, weak local sectors for the graduate’s degree
Published by TutorHao | Exam Prep Revision Series | aleveler.com
更多咨询请联系16621398022（同微信）
屏轩国际教育cambridge primary/secondary checkpoint, cat4, ukiset,ukcat,igcse,alevel,PAT,STEP,MAT, ibdp,ap,ssat,sat,sat2课程辅导，国外大学本科硕士研究生博士课程论文辅导