QS排名 学术声誉 引文 雇主声誉 就业结果
每年六月新一版 QS 世界大学排名公布，家长群和留学群里都会出现同一张榜单截图，有人看前二十，有人只找自己想申的那一所。决定名次的并不是学校好坏这一个变量，而是一张公开的权重表加上几个数据来源。这篇文章把 2027 版的十项指标、权重分配、入榜门槛和结果的读法拆开讲，读完你至少能判断一张截图里的名次变动是被口碑拉动的，还是数据本身在变。
Every June a new edition of the QS World University Rankings lands, and the same screenshot travels through parent chats within hours; some readers scan the top twenty, others look only for the one university their child is applying to. A rank is not a single measure of quality, but the output of a published weighting table fed by a handful of data sources. The sections below take apart the ten indicators, the weights, the eligibility rules and the way those results should be read.
十项指标与五组 Lens 的权重分配
QS 把十项指标归成五组 Lens，权重加起来正好是 100%。研究与发现这一组最重，占 50%，由学术声誉 30% 和单位教师引文 20% 组成。就业与结果占 20%，里面雇主声誉 15%、就业结果 5%。全球参与占 15%，国际教师比例、国际研究网络、国际学生比例各占 5%。学习体验占 10%，全部落在师生比这一项上。可持续发展单独占 5%，是 2022 年之后加进来的新方向。记住这张权重表，榜单上任何一所学校的名次变动都能倒推回具体的某一项指标。
QS sorts ten indicators into five lenses whose weights add up to exactly 100%. Research and Discovery is the heaviest lens at 50%, built from Academic Reputation at 30% and Citations per Faculty at 20%, while Employability and Outcomes takes 20% through Employer Reputation at 15% and Employment Outcomes at 5%. Global Engagement accounts for 15%, split evenly between international faculty, international research network and international students; Learning Experience takes 10% and Sustainability closes with 5%, so almost every movement in the table traces back to one of these lines.
学术声誉 30% 与每年约 7000 所院校的同行提名
学术声誉占 30%，是单项里权重最高的一块。它的数据来自 QS 全球学术调查，请各学科专家提名自己领域里做得好的学校，QS 官网写明这一环节每年评估约 7,000 所院校。这份问卷的历史可以追到 2004 年，也是 QS 最早建立声誉数据的一年，如今它同时支撑着 QS 的学科排名与地区排名。它的长处是覆盖全球，短处同样明显：口碑变化很慢，一所学校十年前的名声可以在榜单上停留很久。很多学生以为名次变了就代表教学变了，实际上这一项常常是滞后指标，它记录的是同行记忆而不是当下课堂。
Academic Reputation carries the largest single weight at 30%, and it comes from the QS Global Academic Survey, where subject experts nominate the institutions they rate in their own field; QS 官网 states that around 7,000 institutions are evaluated at this stage each year. The survey dates back to 2004 and now also feeds the subject and regional rankings. Reputational data moves slowly, which is why a university's standing can outlast the reality on campus by several years; this indicator records what academics remember rather than what happens in a lecture room this term.
单位教师引文 20% 与 Scopus 五年被引数据
单位教师引文占 20%，数据来自 Elsevier 的 Scopus 数据库。QS 官网给出的算法是：把一所学校五年内的总被引次数除以教师人数，而不是除以论文数量。这个口径决定了规模小、研究密度高的理工类院校容易在这一项拿到高分，也让体量庞大但人均产出普通的学校吃亏。艺术与人文学科的引用本来就少，为了不让它们被自然科学压住，QS 对这项数据做了学科领域归一化，五个学科领域各占这一部分的 20%。常见的错误是拿这里的分数判断本科教学水平，它衡量的是研究产出与学术密度。
Citations per Faculty accounts for 20% and is drawn from Elsevier's Scopus database. QS 官网 describes the calculation as the total number of citations a university collects over five years divided by the number of academic staff, not by the number of papers; the ratio explains why small, research-intensive science institutions often score well here while very large universities with ordinary per-head output do not. Citation data is also normalised across five faculty areas so that medicine and natural sciences cannot crowd out the arts and humanities, which receive far fewer citations by nature.
雇主声誉 15% 与就业结果 5% 的区别
雇主声誉占 15%，配上就业结果 5%，合成就业方向的 20%。雇主声誉同样来自问卷，对象是全球范围内招聘毕业生的公司与人机构，请他们说出自己愿意招聘的学校和偏好招聘的专业。QS 官网指出，这是主流排名里唯一把雇主评价单列成指标的榜单。就业结果看的是毕业生去向，包括进入高技能岗位的比例与雇主构成，数据由学校通过 QS HUB 提交。这两项最容易被读混：雇主问卷问的是毕业生口碑，就业结果问的是实际去向，两者在总分里被压成了同一行数字。
Employer Reputation holds 15% and Employment Outcomes adds 5%, giving employability a fifth of the total score. Employer Reputation again comes from a survey, this time of graduate employers worldwide, who name the institutions they recruit from and the disciplines they prefer; QS 官网 notes that this is the only major global ranking to give employer opinion its own indicator. Employment Outcomes looks at what graduates actually do, including the share entering high-skill roles, with institutions submitting the underlying data through QS HUB rather than relying on survey perception.
全球参与 15% 里三项各 5% 的差别
全球参与占 15%，由三个各占 5% 的指标组成：国际教师比例、国际研究网络、国际学生比例。国际研究网络看的是学校能否把合作对象分散到不同国家，而不是长期只跟周边院校做项目，QS 官网在新一版分析里把这一项列为全球平均分最高的指标之一。国际学生比例最直白，就是国际学生占全校学生的比重。三项放在一起，本意是衡量一所学校对外连接的深度，可它们对政策的敏感度完全不同：前两项靠学校自己经营，最后一项会被签证与招生政策直接改写。
Global Engagement accounts for 15%, split evenly between the international faculty ratio, the international research network and the international student ratio. The research network indicator rewards universities that spread partnerships across several countries rather than working mainly with neighbours, and QS 官网 reports it as one of the highest-scoring indicators on average in the latest edition. The student ratio is the most exposed of the three: institutions manage the first two themselves, while the last one can be rewritten by visa rules and recruitment caps within a single cycle.
学习体验 10% 与权重为 0% 的多样性项
学习体验占 10%，全部给师生比，也就是教师人数除以学生人数。这一项对私立小班学校有利，对动辄几万人的公立大学不利，这也是同一梯队里不同类型学校名次忽上忽下的原因之一。可持续占 5%，评分包含环境与社会两部分，数据一部分来自学校提交，一部分来自外部机构。QS 的指标表里还列着一项国际学生多样性，权重写着 0%，它只作参考信息展示，不计入总分。看到有人拿这一项说事，先确认它在总分里占多少权重。
Learning Experience takes 10% and is spent entirely on the faculty student ratio, that is, academic staff divided by enrolled students. The indicator favours small private institutions and penalises large public universities with tens of thousands of students, which helps explain why institutions of the same standing swing in different directions from year to year. Sustainability adds a further 5%, covering environmental and social impact, and the indicator table also lists international student diversity at a weight of 0%, meaning it is displayed for reference only and never enters the final score.
入榜门槛 评估约 8000 所最终只排 1504 所
入榜不是报名就能进。QS 官网说明，一个常规周期里大约会评估 8,000 所院校，最终进榜的只有 1,504 所。资格线要求学校在本科与研究生两个层次都开设完整学位课程，至少覆盖两个学科领域，每个领域至少两个专业，每个专业至少有三届毕业生，课程还要有一部分在校内完成。过了资格线还有第二道线：学术声誉排进全球前 20%，五年内在 Scopus 至少被收录 100 篇论文，并且在所属区域排名中在榜，目前亚洲、阿拉伯地区、欧洲以及拉丁美洲与加勒比四个区域设有这项要求。名次表能被当作参考，前提就是后面这一整套筛选。
Being ranked is not automatic. QS 官网 explains that a typical cycle assesses around 8,000 institutions and ranks only 1,504 of them. To be eligible a university must offer full-degree programmes at both undergraduate and postgraduate level, cover at least two faculty areas with two subjects each, have at least three graduating cohorts in every subject, and deliver part of each programme on campus. A second filter then applies: top 20% globally for academic reputation, at least 100 papers indexed by Scopus over five years, and a place in the relevant regional ranking, which currently applies to Asia, the Arab Region, Europe and Latin America and the Caribbean.
少于 5000 人院校的例外条款
学生规模少于 5,000 人的学校走另一套规则，这是很多读者没注意到的技术细节。它只要满足以下任意一条就能进榜：学术声誉、雇主声誉、单位教师引文三项里有一项排进全球前 1,000，或者其中两项排进前 1,200，或者整体排进前 1,300。条款存在的意义，是让专业院校和规模很小的研究型学校不被体量直接淘汰。反过来说，你看到一所只有几千人的学校名次靠前，不必急着怀疑榜单出错，它的分数结构和大校本来就不一样。读榜时看一眼学校规模，比背下名次实用。
Institutions with fewer than 5,000 students follow a different route, a technical detail most readers never see. They qualify if any one of academic reputation, employer reputation or citations per faculty places them in the global top 1,000, if two of the three put them in the top 1,200, or if they reach the top 1,300 overall. The clause exists so that specialist and small research-led institutions are not eliminated purely by size, which is why a university of a few thousand students can legitimately sit high in the table with a very different score profile from a large comprehensive university.
2027 版结果 顶部稳定与中段位移
最新一版结果在 2026 年 6 月 18 日公布，共 1,504 所院校入榜，其中 98 所是新面孔，多数落在 1,000 到 1,500 名区间，新入榜院校里名次最高的是 Central European University，排在第 239 位。榜单顶部几乎没有变化：MIT 保持第一，Imperial College London 与 Stanford 并列第二，其后依次是 Oxford、Harvard 与 Cambridge。变化集中在中段：Yale、The Chinese University of Hong Kong 与 Johns Hopkins 进入前 20，香港中文大学比上一版上升 14 位。QS 官网把这种结构写成顶部稳定、中段竞争加剧，只看前二十会漏掉这份榜单真正在讲的事。
The latest edition was published on 18 June 2026, with 1,504 institutions ranked and 98 new entrants, most of them landing in the 1,000 to 1,500 band; the highest new entrant is Central European University at 239. The top of the table barely moved: MIT keeps first place, Imperial College London and Stanford share second, Oxford is fourth, followed by Harvard and Cambridge. Movement sits in the middle instead, where Yale, The Chinese University of Hong Kong and Johns Hopkins moved into the top 20 and CUHK climbed 14 places, a pattern QS 官网 describes as a stable top with intensifying competition underneath.
按指标选校而不是按名次选校
同一所学校在不同指标上的表现可以差得很远，这一点在榜单的指标标签页里看得最清楚。打算读研究型学位的人，重点看学术声誉与单位教师引文；打算本科毕业就工作的人，雇主声誉与就业结果更值得看；在意课堂规模与实验条件的人，师生比和可持续这两项比总名次更直接。很多学生以为名次高就等于适合自己，实际上总名次是把十项指标按固定权重压成一个数字，权重表里没有你的专业这一项。这也是专业排名与总排名经常给出不同答案的原因，申请前把两套结果放在一起看，比只记一个位次稳。
The same university can look strong in one indicator and weak in another, which is why the indicator view is more useful than the headline rank. Anyone heading for a research degree should weight academic reputation and citations per faculty; anyone planning to work straight after a first degree should look at employer reputation and employment outcomes; anyone who cares about class size and lab access gets more signal from the faculty student ratio and sustainability scores. A single overall number compresses ten indicators into one figure, and none of those weights refers to your subject, so the subject ranking and the overall ranking will often tell different stories.
国际生政策变化与排名读不出的部分
榜单读不出来的东西同样重要。QS 官网在新一版分析里提到，随着几个主要留学目的地收紧学生签证、设置招生上限或对国际学生加征费用，国际学生比例这一项的流向正在改变：香港、新加坡、马来西亚与韩国的平均国际学生比例在上升，英国与美国不少学校在这一项上同比下降。同一份分析给出的名次位移是，英国有 15 所、美国有 36 所院校下滑超过 20 位，加拿大 7 所、澳大利亚 2 所，而中国大陆有 29 所上升超过 20 位。这些数字说明的是生源与政策的相对变化，不等于教学质量在同步变化。选校时把排名当成参考坐标，把课程设置、录取要求、学费与毕业去向单独列出来对比，才是这份榜单该有的用法。
What the table cannot show matters just as much. QS 官网 notes in its latest analysis that as several major destinations tighten student visas, set recruitment caps or levy charges on international students, the international student ratio is shifting: Hong Kong, Singapore, Malaysia and South Korea improved their average, while many universities in the UK and the US recorded year-on-year declines. The same analysis counts 15 UK and 36 US institutions falling more than 20 places, against 29 institutions in mainland China rising by more than 20. Those movements describe where students and policies are going, not a synchronous change in teaching quality, so the ranking works best as a coordinate alongside course structure, entry requirements, tuition and graduate destinations.
牛津高阶英汉双解词典 第10版 商务印书馆¥168📱 去京东购买京东精选需要一对一辅导？Need One-on-One Tutoring?
A-Level / GCSE / IB / KS2-KS3 全科辅导
16621398022 同微信
Visit aleveler.com
或截图后用「扫一扫 → 相册」选择本图