What each system measures
QS assigns substantial weight to academic and employer reputation surveys, alongside faculty-student ratio, citations per faculty and international staff and student mix. The effect is that recognised global brands, strong employer relationships and highly international campuses perform well, and that results are more sensitive to perception than in the other two systems.
Times Higher Education distributes weight across teaching environment, research environment, research quality, industry income and international outlook. It is the broadest of the three in scope and the only one that treats the teaching environment as a distinct pillar, using resourcing and volume proxies to do so.
ShanghaiRanking (ARWU) uses only objective, externally verifiable indicators: papers in leading journals, highly cited researchers, indexed publications, prize-winning alumni and staff, and per-capita academic performance. It runs no surveys, which makes it the most stable of the three and the narrowest in scope.
Different table sizes, different editions
The three systems publish tables of different sizes and on different annual cycles. The GURI 2026 edition therefore states the source edition for every figure it uses: QS 2026, THE 2026, ARWU 2025. ShanghaiRanking's cycle runs behind the other two, and GURI never relabels an ARWU edition to match the GURI edition year.
Because table sizes differ, raw positions from two systems cannot be averaged. GURI converts each published result into a percentile against that system's own published total first, which is what makes a cross-system comparison valid.
Where the three systems disagree most
Disagreement is structural, not accidental. A research-intensive institution with a modest international profile will look stronger in ShanghaiRanking than in QS; a highly international, teaching-oriented institution shows the opposite pattern. The table below lists the institutions inside the GURI top 300 with the widest gap between their strongest and weakest source percentile in the 2026 edition.
| University | Country | GURI rank | QS | THE | ARWU | Spread (pts) |
|---|---|---|---|---|---|---|
| King Fahd University of Petroleum & Minerals | Saudi Arabia | 186 | 67 | 184 | 701-800 | 70.6 |
| United Arab Emirates University | United Arab Emirates | 270 | 229 | 201–250 | 701-800 | 63.8 |
| Universiti Malaya (UM) | Malaysia | 234 | 58 | 201-250 | 601-700 | 61.2 |
| Hong Kong Baptist University | Hong Kong SAR, China | 271 | 244 | 201–250 | 601-700 | 53.8 |
| Khalifa University | United Arab Emirates | 273 | 177 | 201–250 | 601-700 | 53.8 |
| Soochow University | China | 163 | 801-850 | 51 | 101-150 | 52.5 |
| Polytechnic University of Turin | Italy | 299 | 242 | 401–500 | 601-700 | 48.9 |
| Qatar University | Qatar | 207 | 112 | 201–250 | 501-600 | 47.6 |
Spread is the gap in percentile points between an institution's strongest and weakest source result. Source editions: QS 2026, THE 2026, ARWU 2025.
Where they converge
When three methodologies built on different evidence reach the same judgement about an institution, that agreement is the strongest signal available in this sector. GURI measures it directly through the Inter-Ranking Consistency Index, which compares how far an institution's source percentiles deviate from one another.
| University | Country | GURI rank | QS | THE | ARWU | IRCI |
|---|---|---|---|---|---|---|
| Stanford University | United States | 16 | 3 | 5 | 2 | 99.9 |
| Massachusetts Institute of Technology | United States | 1 | 1 | 2 | 3 | 99.8 |
| University of Chicago | United States | 27 | 13 | 15 | 10 | 99.8 |
| University of Cambridge | United Kingdom | 10 | 6 | 3 | 4 | 99.7 |
| Cornell University | United States | 21 | 16 | 18 | 12 | 99.7 |
| Harvard University | United States | 7 | 5 | 5 | 1 | 99.7 |
A high Inter-Ranking Consistency Index means the three systems place the institution at a similar percentile despite using different indicators.
How much agreement is there overall?
Across the whole 2026 index, the average percentile spread between an institution's strongest and weakest source result is 21.9 points, with a median of 19.2. The distribution of consistency bands below shows how many institutions the three systems broadly agree on, and how many they read very differently.
Which system should you use?
The honest answer is that it depends on the question. For employability and international environment, QS carries the most relevant indicators. For a broad view including teaching resourcing, THE does. For pure research standing, ShanghaiRanking does. None of the three is a general-purpose verdict on institutional quality.
GURI is independent of QS, Times Higher Education and ShanghaiRanking, is not affiliated with or endorsed by any of them, and does not replace them. It reads their published results and reports performance across all three, together with how far they agree.
- Employability and international environment. QS indicators are the most relevant.
- Broad institutional profile including teaching resourcing. THE indicators are the most relevant.
- Research output and academic prestige. ShanghaiRanking indicators are the most relevant.
- Performance across all three at once, with agreement measured. the GURI Composite Score.
Explore this on GURI
Frequently asked questions
What is the difference between QS and THE rankings?
QS weights academic and employer reputation heavily; THE distributes weight across teaching environment, research environment, citation impact, industry income and international outlook. THE is broader in scope, QS more sensitive to reputation and employability.
Is ShanghaiRanking better than QS?
Neither is better in general. ShanghaiRanking is more objective and research-focused; QS covers reputation, employability and internationalisation that ShanghaiRanking does not measure. The appropriate system depends on the question.
Can QS, THE and ShanghaiRanking positions be compared directly?
Not as raw positions, because the systems publish different table sizes and band their tails differently. GURI converts each result into a percentile against that system's published total before comparing.
How often do the three systems disagree?
In the GURI 2026 edition, 620 institutions are ranked by all three systems, and the median spread between an institution's strongest and weakest source percentile is 19.2 points, so moderate disagreement is the norm rather than the exception.
Related insights
GURI is independent of QS, Times Higher Education and ShanghaiRanking and is neither affiliated with nor endorsed by them. All figures on this page are generated from the approved GURI 2026 production data.