QS排名学术声誉 30%权重 学术问卷 提名制
每年六月，QS 发布新一轮世界大学排名，家长群里的第一个问题往往是名次，而不是名次背后的算法。QS 2027 版在 2026 年 6 月 18 日公布，MIT 连续第十五年排在第一位，Imperial College London 与 Stanford 并列第二，Oxford 第四，Cambridge 第六。这一版收录 106 个高等教育体系里的一千五百多所院校，美国 184 所入榜，英国 93 所，中国大陆 85 所。位次的变化常常只有一两位，支撑位次的九个指标权重却相对稳定。学术声誉一项占 30%，是所有单项里最高的。理解这 30% 是怎么来的，比记住学校掉了一位还是升了一位更有用，因为它决定你该在什么条件下参考这份榜单。
QS released the 2027 edition on 18 June 2026. MIT holds first place for a fifteenth consecutive year, Imperial College London and Stanford share second, Oxford is fourth and Cambridge sixth. The edition covers more than 1,500 universities from 106 higher education systems, with 184 entries from the United States, 93 from the UK and 85 from mainland China. Academic reputation alone carries 30% of the total score, more than any other single indicator, so knowing what that 30% measures tells you when the table is worth using.
学术声誉怎么收集：提名而不是打分 / How academic reputation is collected
QS 官网的指标页写得很清楚。学术声誉问的是哪所大学在学术上表现出色。它让学者按自己的专业领域提名院校，而不是给院校打分。这套做法 2004 年由 QS 引入。每年汇总的提名覆盖约七千所机构。学者在问卷开头先声明自己的学科方向与熟悉的地区。后面的题目范围由此收窄，一位经济学教授不会去评价医学院的临床教学。很多学生以为学术声誉是一个整体印象分，其实它是学科内的同行共识。同一所大学在不同学科里的声誉差别可以很大，只看总排名会漏掉这一点。
Academic reputation is not a score that universities earn by submitting their own figures. QS asks academics to nominate institutions within their own discipline and region, a method the company introduced in 2004 and now applied to roughly 7,000 institutions a year; the survey opens by asking respondents to state their subject and regional familiarity, which narrows the questions they can answer afterwards. That is why a strong economics department and a weak medical school can sit inside the same university, and why reading only the overall table hides how your own subject is regarded.
学术声誉也是入榜门槛 / Reputation is also an inclusion requirement
很多人把学术声誉当成九个指标里的一个分子，它还有第二重身份。QS 官网的入榜条件写明，一所学校要被纳入世界大学排名，必须处在其学术声誉的全球前 20%。这条门槛在总榜里不显示，却决定了一所学校能否出现在表格上。规模不足五千名学生的院校还有额外要求，它们的学术声誉、雇主声誉与每位教员引用三项里，至少要有一项进入前一千名。或者在其中两项进入前一千二百名。又或者整体进入前一千三百名。常见的错误是把入榜当成理所当然，看到某所学校没上榜就以为它办学不行。
Academic reputation does double duty in this method. It carries 30% of the score, and it also decides which universities are eligible to appear at all, because inclusion requires a place in the global top 20% for that indicator; institutions with fewer than 5,000 students face a further test and must reach the top 1,000 in one of reputation, employer reputation or citations, the top 1,200 in two of them, or the top 1,300 overall. A university missing from the table is therefore not the same as a university that failed to submit its data.
问卷怎么送到学者手里 / How the survey reaches respondents
学术声誉的数据来自 QS 的学术问卷，投递渠道由院校自己推动。QS 官网的问卷说明写明，每所院校每类问卷最多可以提交 400 个联系人。参与商学院排名的院校可以再提交 400 个。这额外的名额只用于商科榜单，不回流到世界大学排名。当某所院校的回复超过 350 份时，QS 会发出通知，请学校停止继续征集。提交上来的联系人名单会使用两年，也就是提交当年再加一年。学术与雇主问卷在 2027 版里安排在 2026 年 11 月至 2027 年 3 月初进行，联系人的提交窗口从 2026 年 9 月 15 日开到 2027 年 1 月 31 日。
The academic survey does not reach respondents by chance. Institutions submit contact lists, capped at 400 names per survey, while business schools may add another 400 that feed only their own tables; once a university passes 350 responses QS asks it to stop recruiting, and submitted contacts are used for two years. For the 2027 edition the academic and employer surveys ran between November 2026 and early March 2027, with contact lists accepted from 15 September 2026 to 31 January 2027.
十五万份学术问卷的规模 / The scale behind the scores
QS 官网的排名业绩页给出的数字是 151,000 份以上学术问卷回复。同一页写明雇主问卷回复超过 100,000 份。QS 每年分析约 8,000 所机构，其中约 2,000 所被排名。引用数据来自 Scopus 数据库，覆盖 1,640 万篇学术论文。样本量决定了声誉分数的随机波动有多大。名气大的学校落在多数受访者的熟悉范围内，样本变化对它们的影响较小。规模小或者在地区外知名度有限的学校，分数更容易受受访者名单影响。很多学生以为声誉分是一把绝对尺子，它更像是每年重做一次的抽样调查。看榜时留意学术声誉的具体分数，比只看位次更能判断差距。
QS reports more than 151,000 academic responses and over 100,000 employer responses, alongside analysis of about 8,000 institutions of which roughly 2,000 are ranked; the citation measures draw on 16.4 million papers indexed in Scopus. Sample size shapes how far a reputation score can move in a single year, and well-known universities sit inside the familiar range of most respondents, so their scores stay steady. Smaller or regionally focused institutions are more exposed to shifts in the contact lists, which is why reading the indicator score is safer than reading the position alone.
学术声誉与每位教员引用的分工 / Reputation versus citations
学术声誉与每位教员引用同属研究与发展这一组。两者合计占 50%。学术声誉占 30%，问的是同行怎么看。每位教员引用占 20%，看的是论文被引用的实际记录，数据来自 Scopus。一个讲印象，一个讲产出，两者的差距能说明问题。声誉高而引用低，可能意味着学校的历史声望还没有被近期研究支撑起来。引用高而声誉一般，常见于规模不大但研究方向集中的理工院校。很多学生以为两项说的是同一件事，看到引用分很高就以为学校名气也很大。
Academic reputation and citations per faculty sit in the same lens and together make up half the score, but they answer different questions. Reputation, at 30%, records what peers think, while citations, at 20%, measure the actual citation record drawn from Scopus. A university with strong reputation and weak citations may be living on history, whereas strong citations with modest reputation is common among smaller research-focused institutions, so the gap between the two figures is often more informative than either one.
五个维度的权重结构 / The five lenses and their weights
QS 官网把指标归成五组，每组带一个权重。研究与发展占 50%，里面是学术声誉 30% 与每位教员引用 20%。就业与成果占 20%，里面是雇主声誉 15% 与就业成果 5%。全球参与占 15%，里面是国际教员比例 5%、国际研究网络 5%、国际学生比例 5% 与国际学生多样性 0%。学习体验占 10%，只有师生比一项。可持续性占 5%。官网注明权重每年复核一次，隔几年回头看，同一项指标的份量可能已经变了。看榜前先确认自己看的是哪一版的方法论，这一步经常被跳过。
QS groups its indicators into five lenses. Research and Discovery takes 50%, Employability and Outcomes 20%, Global Engagement 15%, Learning Experience 10% and Sustainability 5%. Weightings are reviewed every year, which means an indicator that mattered in one edition may carry a different value in the next; checking the methodology page for the edition you are reading settles most arguments about the table.
声誉分为什么动得慢 / Why reputation moves slowly
学术声誉建立在受访学者的记忆之上，记忆更新得比数据慢。一位教授判断某所学校的水平，参考的是自己读过的论文、合作过的团队、带过的学生。这些印象来自多年的积累，不会因为某一年论文数量大增就立刻改变。QS 的权重每年复核，声誉这一项的历史惯性却很难被单次调整抵消。已经在前列的学校通常落在受访者的熟悉范围内。新校要挤进这个范围，需要几届毕业生的稳定输出。常见的错误是用一年的位次波动推断一所学校的长期走向，一两位的变化往往落在抽样误差之内。
Reputation rests on what academics remember, and memory updates more slowly than data. A professor judges a department through papers read, projects shared and students supervised over many years, so one strong year of output rarely changes the picture; universities already inside that circle keep their advantage, while newer names need several graduating cohorts to build recognition. Treating a one or two place move as a trend therefore over-reads what the survey can support.
国际学生比例 5% 与国际学生多样性 0% / Two international indicators, one at zero
全球参与这一组里有两条容易被混在一起的指标。国际学生比例占 5%，算的是国际学生在全校学生里的占比。占比越高得分越高。国际学生多样性列在官网的方法论表里，权重是 0%。这一项当年不参与总分计算，但仍然出现在指标列表里。很多学生以为国际学生越多，学校的国际化得分就越高，看到多样性这一栏没有分数就以为数据缺失。真正的读法是先确认权重那一列，再去比较具体分数。这两项只影响全球参与这一组，而这一组的总权重是 15%。
Two international indicators are easy to confuse. The international student ratio carries 5% and rewards a higher share of international students, while international student diversity appears in the same methodology table with a weighting of 0%, meaning it does not contribute to the total score in this edition. Reading the weighting column first tells you which indicator is actually doing work, and both measures feed only the Global Engagement lens, which is worth 15% overall.
学科榜单的权重与总榜不同 / Subject tables use different measures
想判断某所大学在自己专业上的位置，总榜不是最合适的工具。QS 单独发布学科排名，指标组合与总榜不同，权重也按学科调整。学科榜同样使用学术声誉，受访者只在相关学科里提名。样本集中在同一领域，比例上的误差影响更小。官网每年更新方法论页，写清当年用哪些指标与权重。很多学生以为总榜名次可以直接推到专业申请上，一所综合排名中游的学校可能在某个学科稳居前列。查专业时先看学科榜的位次，再用总榜了解学校的整体环境，这个顺序更贴合申请决策。
For a specific course, the subject tables are the better starting point. QS publishes them separately, with their own mix of indicators and subject-specific weightings, and the reputation element draws on nominations from academics working in that field, so the sample is concentrated in one discipline. A university sitting mid-table overall can hold a top position in a single subject, which is why checking the subject table before the overall ranking usually matches how admissions decisions are made.
雇主声誉 15% 与就业成果 5% / Employer reputation and employment outcomes
就业与成果这一组占 20%，由雇主声誉与就业成果两项组成。雇主声誉来自雇主的问卷，QS 官网给出的规模是 100,000 份以上回复。受访者按自己招聘毕业生的经验评价院校。这一项反映的是招聘端的认知，而不是毕业生的平均薪资。就业成果占 5%，看的是毕业生在就业市场的实际结果。两项放在一起看，可以判断一所学校的口碑与实际出口是否一致。常见的错误是用雇主声誉去推断某个专业的就业率，雇主声誉是全校层面的印象分。专业之间的差别要回到课程与实习安排上看。
Employability and Outcomes takes 20% of the score. Employer reputation accounts for 15% of the total and draws on more than 100,000 employer responses, which means it records how recruiters view a university rather than what its graduates earn; the remaining 5% looks at employment outcomes. The two figures together show whether a university's standing with employers is matched by results, and a high employer score says nothing about the placement support of one particular department.
这一版收录了哪些学校 / What the 2027 edition covers
2027 版在 2026 年 6 月 18 日公布，收录 106 个高等教育体系里的一千五百多所院校。美国有 184 所入榜，英国 93 所，中国大陆 85 所。前二十名里，MIT、Stanford、Harvard 与 Caltech 来自美国。Imperial College London、Oxford、Cambridge 与 UCL 来自英国，ETH Zurich 来自瑞士。官网的资格条件写明，参评院校要在本科与研究生两个层次都授予完整学位。院校还要在至少两个学科领域开设完整课程。看一版榜单时，先看清收录范围与资格条件，再讨论名次升降，比较的基础才成立。
The 2027 edition, published on 18 June 2026, ranks more than 1,500 universities across 106 higher education systems, with 184 entries from the United States, 93 from the UK and 85 from mainland China. Eligibility requires full degrees at both undergraduate and postgraduate level in at least two faculty areas, and inclusion requires a place in the global top 20% for academic reputation. Reading the scope and the entry rules before comparing positions keeps the comparison honest.
把位次当区间来读 / Reading ranks as bands, not exact places
总榜第 46 名与第 47 名之间的分差，往往小于两所学校在某个指标上的波动幅度。QS 公开的是总分与各项分数，位次只是分数排序的结果。同一所学校在不同年份换一两位，可能只是受访者名单变化带来的轻微偏移。判断差距要看分数而不是名次。分数相差不到一分时，两所学校的差别在统计上很难说清。很多学生以为并列名次意味着实力相同，并列通常说明总分存在四舍五入后的重合。跨国家比较时还要注意体系差异，一个国家的院校数量多，位次在中段的密度也更高。
Bands matter more than places. QS publishes overall and indicator scores, and the position is simply the order of those scores, so a university moving one or two places between editions may only reflect a slightly different respondent list; when two institutions are less than a point apart, the difference is hard to defend statistically. Reading tied positions as proof of equal strength also over-reads the data, because ties usually come from scores rounded to the same value.
给申请者的三条实操 / Three practical rules for applicants
把榜单用在申请上，可以先做三件事。第一件是查学科榜，确认目标专业所在学科的位次与分数，而不是从总榜推算。第二件是看指标分数。学术声誉、雇主声誉与师生比三项能反映学校的强项在哪一侧。第三件是回到课程页面，看必修模块、考核方式与实习安排，这些信息在排名里找不到。QS 官网每年发布方法论与数据表，申请季前花二十分钟读一遍，比刷榜单截图有用。家长常用的做法是拿排名当筛选门槛，先圈出学科榜前五十，再逐校核对入学要求，把榜单变成一张可执行的清单。
Three habits make the table useful for applications. Start with the subject table, then read the indicator scores for reputation, employer reputation and staff to student ratio instead of the headline position, and finally go back to the course pages for compulsory modules, assessment format and placement options, none of which appear in the ranking. Because QS publishes its methodology and underlying data each year, twenty minutes on the official pages in application season is worth more than a folder of screenshots.
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 全科辅导与升学规划
16621398022 同微信
关注公众号 tutorhao，获取更多学习资源
Visit aleveler.com
或截图后用「扫一扫 → 相册」选择本图