If a University Rises Ten Places in the Rankings, Has Its Education Improved?

Geoffrey Chen

When Numbers Start Making Decisions · Season Two, “When the Measure Becomes the Target” · Article 6

1. What changed between 451–500 and 401–450?

Western Sydney University’s Sustaining Success 2021–2026 Strategic Plan placed world rankings directly among its performance targets. It sought to move from the 251–300 band to 200–250 in the Times Higher Education ranking, from 451–500 to 401–450 in QS, and to increase the number of subjects within the world’s top 200. Western Sydney University: “Sustaining Success 2021–2026”

This is neither secret nor peculiar to one university. Australian universities promote global positions in websites, annual reports, international recruitment and strategy. Adelaide University set an ambition to remain consistently in the global top 100 and stressed the value of ranking when it debuted at 82 in QS. Adelaide University: “Adelaide University debuts in world’s top 100”

When a ranking is merely a table produced by an external publisher, a university can agree with it or ignore it. When the rank enters strategy, it changes internal decisions: whom to hire, which disciplines to support, what kinds of publication to encourage and how much to invest in international reputation and data submission. Description becomes a target.

If a university rises ten places, it may genuinely have improved. Yet the object improving can be “the university as defined by this ranking”, not necessarily the education received by each student, the service provided to every local community or the public value of every research program.

2. How is one place calculated?

In the methodology for the QS World University Rankings 2026, research and discovery account for 50 per cent of the total: 30 per cent from academic reputation and 20 per cent from citations per faculty. Employability and outcomes receive 20 per cent, global engagement 15 per cent, learning experience 10 per cent, and sustainability 5 per cent. The learning-experience component relies principally on the faculty-student ratio rather than direct observation of teaching. QS: “World University Rankings methodology”

Each variable has a rationale. Citations can indicate research influence. Employer views provide information about graduate reputation. International networks show cross-border engagement. Faculty-student ratio may relate to the availability of academic time. Ranking compresses thousands of universities with different systems, scales and missions onto one comparable axis, offering international students and governments a low-cost form of navigation.

But an overall position performs three acts of compression.

First, different values become indicators. Teaching, basic research, employment, internationalisation and sustainability have no natural exchange rate. The methodology’s author must choose weights.

Second, different disciplines become one institution. Medicine and natural science tend to have higher citation density than the humanities, local-language scholarship and community-based research. Field normalisation can reduce but not remove the influence of disciplinary composition.

Third, continuous scores become ordinal positions. A university ranked 39th and one ranked 49th may have very similar underlying scores. Movement by other institutions can shift a university whose own performance did not change.

A place is therefore not a measurement of institutional essence. It is jointly generated by a set of values, data and relative competition.

3. Why do rankings have practical power?

Rankings have no legislative force, yet can influence international students, partners, donors, media reputation and the recruitment of prominent researchers. The University of Queensland has explicitly noted that international students commonly consider global standing when choosing a university, making rankings important. UQ: “UQ ranked in the world’s top 50 universities”

This influence creates a feedback loop. A high rank attracts students and researchers, generating revenue, networks and citations that support the next ranking cycle. A lower-ranked regional university may make indispensable contributions to first-in-family students, remote communities and local industries, but have less capacity to accumulate global reputation.

The ranking begins both to predict opportunity and distribute it. Unlike ATAR, it does not draw a formal eligibility boundary. Instead, many dispersed choices produce a de facto threshold. No single decision-maker declares that institutions below 100 are unworthy of partnership. Families, agents, collaborators and managers use the number, and their combined actions create structural consequences.

No observer understands hundreds of universities in full. A ranking stores a quickly accessible judgement for a society with distributed knowledge. Its convenience makes it easy to forget who defined that judgement and the purpose for which it was generated.

4. How do universities learn to adapt?

A ranking target can stimulate real improvement. It can support high-quality research, international collaboration, more sustainable faculty-student ratios and better tracking of graduate outcomes. Those changes improve the indicator and institutional capacity together.

It can also direct optimisation towards proxies. Universities can recruit highly cited researchers, move resources to citation-rich fields, encourage publication in English-language international journals, expand international-student ratios and invest in reputation campaigns. Each action may have academic value. In combination, the methodology becomes an invisible curriculum telling universities which kinds of knowledge are most valuable to produce.

Other work may add little to global position: intensive support for underprepared students, long partnerships with local schools, preservation of regional histories and languages, teaching-heavy professional programs, and practical research for local businesses and public bodies. If every strategy meeting asks whether such work will lift the ranking, those missions must justify themselves in a language poorly fitted to them.

The deepest change does not require fraudulent submission. The organisation can sincerely learn to understand itself through ranking vocabulary. Research success becomes citations, internationalisation becomes a population ratio, and learning experience becomes faculty-student ratio. A proxy passes from management instrument into the language of value.

5. Why a higher rank may not mean better teaching

The Australian Universities Accord Final Report provided a direct warning. Eleven Australian universities were in the top 200 of the overall 2024 Times Higher Education ranking. Only six were in the top 200 on its teaching dimension, down from seven in 2015, and the teaching rank had fallen for five of the six leading Australian universities considered. Australian Universities Accord: Final Report

This does not prove that a highly ranked university teaches poorly. It shows that the aggregate position cannot substitute for examination of a particular dimension. Strong research influence can lift an institution while student support, curriculum experience or teaching resources fail to improve in parallel.

Nor should a student equate the university-wide position with the quality of their degree. Strength in engineering does not make every discipline equally strong. General employer reputation does not determine outcomes in a particular region, occupation or course. For a student balancing work, care and travel, scheduling and support can matter more than global citations.

Rankings are therefore better used as an entry point for investigation than as the final decision about fees, migration, career and several years of a person’s life.

6. How can universities avoid being governed by one table?

First, a strategy may report rankings without treating a position in one league table as the institution’s overall line of success. Movement should be decomposed into raw indicators, distinguishing changes in the university from changes in methodology and movement by competitors.

Second, different missions require different evidence. Australian mission-based compacts already recognise autonomous universities with distinct missions. Department of Education: “Mission based Compacts” Research-intensive institutions, regional universities, Indigenous education and professionally oriented institutions cannot be exhaustively evaluated on one global axis.

Third, teaching and student outcomes need measures closer to their purpose: course-level experience, learning gain, completion, equity gaps, long-term employment and professional competence. Their limitations should be published so that employment itself does not become a new target that encourages exclusion of higher-risk students.

Fourth, decision-makers should record the resources shifted by ranking targets. If investment in a citation-rich field rises, which area receives less? If international-reputation spending grows, how does it compare with student support? The opportunity cost of a metric cannot remain outside the score.

Fifth, public work resistant to counting should receive protection. It still requires accountability, but that accountability may combine peer review, community feedback, case outcomes and long observation rather than forcing every contribution into a global place.

Conclusion: rankings may compare universities, but should not decide what a university is

World rankings make an extremely complex sector visible. They preserve partial facts about research, reputation, employability and international networks. Abolishing every ranking would not abolish comparison; it could return comparison to less transparent impressions of brand.

My judgement is that universities may use rankings as external reference points and diagnostic indicators, but should not elevate movement by a number of places above their mission. Governments, councils and students should not use the overall position instead of specific evidence about teaching, access, disciplines and local contribution. Any ranking KPI should sit beside teaching, equity and public-value conditions that cannot be offset by a higher total score.

A university that rises ten places has changed its position relative to others under one set of weights. To prove that education improved, we must return to particular students, courses, research and social relationships. A number can compress a university. It should not compress what the university permits itself to become.

Principal sources

Discover more from Geoffrey Chen

Subscribe to get the latest posts sent to your email.