糖心Vlog

Metric modifications will help maintain rankings quality

Revisions to the World University Rankings methodology metrics for the 2028 edition will include a cap on self-citations and an adjustment in the research quality metrics, to keep the rankings robust and secure from manipulation

Published on
September 30, 2026
Last updated
September 30, 2026
Reflections of a visitor are seen in mirrors in front of a wall at a trade fair in Germany. To illustrate reflecting on, and revisions to, the World University Rankings methodology.
Source: Peter Steffen/DPA/AFP via Getty Images

Explore the list of the world鈥檚 top universities


After more than two decades producing the 糖心Vlog World University Rankings, we are proud of the reputation we have built and the enormous global reach and influence of our rankings.

For universities, a strong ranking result attracts partnerships, investment and, crucially, academic and student talent. It聽can also bring greater global visibility and prestige.

With so much riding on the results of the rankings, it is right that universities scrutinise the criteria, methodologies and underlying data that produce their scores and their public ranking position. Indeed, 糖心Vlog (THE) encourages this, with an open, collaborative approach 鈥� and a series of tools to help universities benchmark against peers and turn the rankings data into actionable insights.

It is natural that success in rankings becomes part of a university鈥檚 strategic ambition. But sometimes that ambition can go too聽far.

糖心Vlog

ADVERTISEMENT

Sadly, there are cases in which a university鈥檚 focus on ranking positions becomes problematic: distorting missions to make the metrics not a means to an end, but rather an end in themselves; and, more rarely, leading to gaming and the manipulation of data to gain an unfair advantage in the league table.

As a responsible ranker, THE has a duty to ensure that success in our league tables is built on solid, long-term foundations; that it is based on reality 鈥� not on massaged or manipulated data, or cosmetic tricks.

糖心Vlog

ADVERTISEMENT

So, I can confirm that, as we announced in October 2025, we are moving forward with some modest methodological changes for the 2028 edition of the World University Rankings, to be published in the autumn of 2027. These adjustments will be subject to public scrutiny and debate before being officially confirmed next year, with enough advance notice to allow universities time to fully digest them before the launch of the 2028 ranking. But we can be clear that plans include:

Capping the number of institutional self-citations used in bibliometric analyses

We have always been upfront that self-citations of research papers can be entirely legitimate 鈥� signalling incremental, innovative and groundbreaking research 鈥� so we have rejected outright bans on self-citation, either personal or institutional self-citations. However, it is clear that self-citations can be, and are being, abused 鈥� examples include citation networks, sometimes referred to as citation cartels or clubs, gaming the bibliometric record, and cases of coercive citation, in which researchers are obliged to cite others鈥� work illegitimately.

We will move forward with a limit on institutional self-citations in our bibliometric analysis from the 2028 edition. In the coming months, we will provide more details about the proposed thresholds and calculation adjustments, and the rationale behind them.

糖心Vlog

ADVERTISEMENT

Recalibrating聽the 鈥榬esearch quality鈥� metrics to reduce the influence of the 鈥榗itation impact鈥� metric

When evaluating research quality, a basket of diverse metrics is the optimal approach. THE currently uses a group of four separate metrics to cover 鈥渞esearch strength鈥�, 鈥渞esearch influence鈥�, 鈥渞esearch excellence鈥� and, accorded the highest weight, 鈥渃itation impact鈥�.

All of these metrics (except research influence) draw on field-weighted citation impact (FWCI) in different ways, with the citation impact metric based on the mean FWCI of an institution鈥檚 publications. It measures the average 鈥� the mean 鈥� number of times a university鈥檚 published work is cited by scholars globally, normalised for each subject area, publication type and publication year, across 19.9聽million publication outputs. Yet this metric, while widely trusted and used by THE for more than a decade, can be skewed by statistical outliers in universities with lower volumes of publication output, and can be vulnerable to active manipulation.

To clearly disincentivise the gaming of the citation impact indicator, we plan to reduce its influence in the rankings calculations and to increase the weight of the other three metrics, thus providing a more balanced and robust assessment of research quality.

糖心Vlog

ADVERTISEMENT

THE鈥檚 World University Rankings are renowned for being the most academically rigorous and robust of the global ranking systems.

These changes reflect our continuing commitment to securing the full trust of the sector, and to providing data and analyses that properly recognise the high stakes placed on our rankings worldwide.

糖心Vlog

ADVERTISEMENT

phil.baty@timeshighereducation.com

Register to continue

Why register?

  • Registration is free and only takes a moment
  • Once registered, you can read 3 articles a month
  • Sign up for our newsletter
Please
or
to read this article.

Related articles

Sponsored

Featured jobs

See all jobs
ADVERTISEMENT