Why raw positions are not comparable
Position 180 in a system that publishes 500 ranked institutions is not equivalent to position 180 in a system that publishes 2,000. The two numbers look alike but describe very different levels of relative standing. Publishers also switch from exact positions to bands in the lower parts of their tables, so a single number in the tail may represent a range of dozens of institutions.
Any comparison that treats these numbers as interchangeable will produce misleading conclusions. Normalisation is not a refinement here — it is a precondition for comparing at all.
The percentile approach
GURIndex converts each published result into a percentile within that system's ranked population for that edition. A percentile answers the question 'what share of ranked institutions did this institution outperform in this system?', which is meaningful regardless of how many positions the publisher chose to release or how it banded its tail.
Once every result is a percentile, the results can be averaged, compared and tracked over time. Ranking Achievement — 72% of the GURI Composite Score — is built directly from these normalised percentiles.
What to compare beyond the position
A well-formed comparison looks at four things, not one:
- Achievement — normalised standing in each system where the institution appears.
- Coverage — how many of the three systems rank it at all, since more independent sources means stronger evidence.
- Consistency — how closely the systems agree, measured by the Inter-Ranking Consistency Index.
- Momentum — whether normalised achievement is rising or falling across recent editions.
Comparing institutions, countries and systems
The same logic scales upward. Two institutions can be compared head to head; every ranked institution in a country can be compared against its national peers; and whole national higher-education systems can be compared against one another using aggregate measures of average performance, representation, legacy and excellence.
GURIndex provides all three levels, and each comparison can be exported as a formal performance report for use in planning documents, board papers and accreditation submissions.
Explore this on GURIndex
Frequently asked questions
Can you compare QS and THE rankings directly?
Not by comparing raw positions. The two systems publish different numbers of positions and weight different indicators. Converting each result into a percentile within its own ranked population makes the comparison valid, which is the approach GURIndex uses.
How do I compare two universities fairly?
Compare normalised achievement in each ranking system, then check how many systems cover each institution, how consistently those systems place it, and which direction its momentum points. GURIndex reports all four side by side in its comparison tool.
Why does a university rank higher in one system than another?
Because the systems measure different things. Reputation-weighted systems favour institutions with global brand strength; bibliometric systems favour research volume and citation impact. The gap between them describes the institution's profile.
Related guides
Use the GURIndex university comparison tool to compare two institutions side by side on composite score, ranking sources, legacy, momentum and consistency.
Compare QS, Times Higher Education and Shanghai (ARWU) rankings: what each measures, why they disagree, and how GURIndex combines all three into one index.
The Inter-Ranking Consistency Index measures how closely QS, THE and Shanghai agree about a university once results are normalised. Learn how IRCI works.