03 · What You Need to Know
Research Output Is an Input to Ranking Performance, Not a Ranking Formula
Start by Distinguishing Output, Score, and Rank
Three concepts are easily collapsed in university strategy discussions.
Research output
The research products attributed to an institution under a defined measurement system, such as eligible publications within a specified period.
Indicator score
The institution's performance after the ranking organization applies the formula, normalization, denominator, and standardization used for a particular indicator.
Ranking position
The institution's relative position after weighted indicators are combined and compared with other eligible universities.
Increasing the first does not mechanically increase the third. Additional output must pass through the ranking methodology before it affects the overall result.
QS Does Not Simply Count Publications
In the current QS World University Rankings methodology, Research and Discovery contributes 50% of the overall score. Academic Reputation carries 30%, while Citations per Faculty contributes 20%. QS states that this lens reflects research quality, volume, and reputation.
The flagship ranking does not currently contain a standalone publication-count indicator equivalent to Times Higher Education's Research Productivity measure. Publication volume nevertheless matters because publications create the research corpus from which citations arise and contribute to the institution's wider scholarly presence.
But the citation indicator is Citations per Faculty, not total publications. QS divides citation counts by faculty numbers to account for institutional size. Consequently, adding publications that generate little citation attention does not translate directly into a proportional improvement in the citation score.
THE Does Measure Research Productivity Directly
Times Higher Education's World University Rankings 2026 includes Research Productivity at 5.5% within its 29% Research Environment pillar. THE counts relevant publications in Scopus-indexed academic journals per scholar, scales the measure for institutional size, and normalizes it for subject.
Greater publication output can therefore contribute more directly to this indicator, but the calculation is still not simply “more papers equals more points.” Staff numbers and disciplinary profile are part of the methodology.
THE Gives Much More Weight to Research Dimensions Beyond Productivity
Research Productivity is only one THE research indicator. Research Reputation contributes 18% and Research Income 5.5% within Research Environment. The separate Research Quality pillar contributes another 30% through Citation Impact at 15%, Research Strength at 5%, Research Excellence at 5%, and Research Influence at 5%.
THE therefore gives substantial weight to how research is received and positioned within scholarly literature, not merely how many publications an institution produces.
| Change in research activity |
Possible ranking effect |
Why the effect is not automatic |
| More publications |
Can strengthen output or productivity measures |
Indicators may normalize by staff, subject, or other factors |
| More publications receiving citations |
Can strengthen citation-related performance |
Citation formulas and normalization differ among rankings |
| More highly cited publications |
Can affect measures of research excellence or influence |
Effects depend on distribution, field, time window, and methodology |
| More international co-authored publications |
Can contribute to international research indicators |
Rankings operationalize international collaboration differently |
| Greater research visibility |
May contribute to reputation over time |
Reputation is survey-based and may change slowly |
More Output Can Matter for Ranking Eligibility
Research volume sometimes matters before an institution receives any indicator score. THE's World University Rankings 2026 can exclude institutions whose relevant research output is below 1,000 publications during 2020 to 2024, with a minimum of 100 publications in each year. THE also applies other eligibility criteria.
For an institution below such a threshold, increasing eligible research output can be strategically consequential because it may determine whether the university can receive a rank.
Once the threshold is comfortably exceeded, however, producing another thousand papers does not necessarily create an equivalent ranking benefit. Eligibility and scoring are different functions.
More Publications Create More Opportunities for Citations
At the institutional level, a larger research portfolio creates more opportunities for individual papers to be discovered, used, and cited. This can matter because citation-related indicators carry substantial weight in both QS and THE.
Yet opportunity is not outcome. If additional papers attract little scholarly use, citation performance may not rise in proportion to publication volume.
This is why citations matter to university rankings differently from publication counts.
The Distribution of Citation Performance Matters
THE illustrates why a simple total can be misleading. Its Citation Impact measure examines average normalized citation performance, while Research Strength uses the 75th percentile of field-weighted citation impact. Research Excellence considers publications in the world's top 10% by field-weighted citation impact, and Research Influence considers the importance of citing publications.
Two universities could therefore publish the same number of papers and receive very different research-quality scores because the scholarly reception of those papers differs.
Research Reputation Does Not Rise Mechanically With Annual Output
QS Academic Reputation contributes 30% to its overall ranking, while THE Research Reputation contributes 18%. Both rely on surveys of academics, although their survey methodologies and indicators differ.
Reputation may reflect accumulated scholarly visibility, disciplinary strengths, influential researchers, historical standing, networks, and previous research achievements. Increasing publication output this year does not guarantee that academics around the world will immediately change their perceptions of the institution.
Research reputation may therefore lag substantially behind changes in current output.
Disciplinary Composition Changes the Meaning of More Research
Universities do not all have the same subject mix. Publication and citation practices differ substantially across disciplines, which is why ranking systems normalize several research indicators.
THE's subject rankings make this variation explicit. In 2026, Citation Impact carries 7.5% in Arts and Humanities but 17.5% in Life Sciences, Medical and Health, Physical Sciences, and Psychology. Research-productivity weights also vary across subjects.
A strategy of simply maximizing paper counts can therefore favor disciplines whose publication cultures naturally produce large numbers of indexed journal articles.
Institutional Size Can Change the Result
Many ranking indicators attempt to distinguish scale from intensity. QS divides citations by faculty. THE scales research productivity by scholar numbers and normalizes research income and several other indicators.
A university that expands both its faculty and its publication output may therefore experience a different ranking effect from one that increases output while faculty size remains stable.
This is another reason raw publication totals are inadequate for predicting rank.
Non-Research Indicators Still Affect the Overall Ranking
Even dramatic research improvement cannot be interpreted without the rest of the ranking methodology.
QS currently assigns 20% to Employability and Outcomes, 10% to Learning Experience, 15% to Global Engagement, and 5% to Sustainability alongside its 50% Research and Discovery lens. THE assigns 29.5% to Teaching, 7.5% to International Outlook, and 4% to Industry alongside Research Environment and Research Quality.
A university can therefore improve research performance while other indicators remain weak or decline.
Rank Is Relative to Other Universities
Perhaps the most important reason more research does not guarantee a higher rank is that rankings are comparative.
Suppose a university increases publications by 15%, citations by 20%, and research income by 10%. Those are genuine improvements. But if comparable universities improve more rapidly, the institution may remain in the same ranking band or move downward.
A falling rank therefore does not necessarily mean the university's research deteriorated. It can mean that its relative position changed.
Absolute improvement
The institution performs better than it did previously on an underlying research measure.
Relative improvement
The institution improves sufficiently to outperform competing universities under the ranking methodology.
Methodological Changes Can Move Rankings Without Changes in Research
Ranking organizations can revise weights, definitions, data windows, normalization procedures, eligibility rules, and indicators. QS states that indicator weights are reviewed annually, while THE publishes methodology for each ranking edition.
An institution's rank can therefore change partly because the measurement system changed, even if its underlying research activity did not change dramatically.
This is why year-to-year rank movement should be interpreted alongside methodological changes rather than treated as a pure measure of institutional progress.
Increasing Output Can Produce Unintended Incentives
Publication targets can encourage useful behavior when they prompt researchers to complete and disseminate worthwhile work. They can also create pressure to prioritize easily publishable questions, divide work into unnecessarily small outputs, or devote attention to whatever activities the metric rewards most strongly.
The problem is not measurement itself. It is treating the measured output as though maximizing it were identical to improving the underlying research system.
DORA's guidance on quantitative research indicators emphasizes that no single indicator captures the complexity of research quality and recommends using quantitative measures clearly, transparently, specifically, contextually, and fairly.
Watch Out
Do not convert a ranking objective directly into individual publication quotas without examining the actual ranking formula. An institutional indicator based on normalized citations, reputation, staff-adjusted productivity, or research excellence may respond very differently from a simple count of faculty publications.
More Research Helps Most When It Is Part of a Stronger Research System
A sustainable increase in output may result from more capable researchers, stronger doctoral programs, better infrastructure, effective collaboration, improved grant support, appropriate workload, and functioning research administration.
In that situation, greater publication volume is evidence of a broader improvement in institutional research capacity.
Increasing output without strengthening the system that produces it is a different strategy. It may improve a short-term metric while creating workload, quality, or integrity problems that eventually undermine the research environment.
The Underlying Research Portfolio Matters More Than One Annual Count
Rankings generally use multi-year bibliometric windows and accumulated reputation data rather than evaluating only the most recent calendar year's output. Consequently, research strategy operates over longer periods than an annual publication campaign.
Universities seeking durable ranking improvement need research portfolios capable of producing credible, visible work repeatedly rather than a temporary surge timed around a particular ranking cycle.