Decoding the College Rankings

DataRes at UCLA
10 min readMar 12, 2024

Daniel Wang (Project Lead), Alvin Huang, Armaan Bassi, Aadi Malaviya

Introduction

College rankings play a pivotal role in determining where millions of students go to school each year, but which factors truly shape them? Our investigation will dive into the QS World University Rankings to find out. In this article, we’ll explore why some schools rank higher than others, how schools in different regions compare, and the differences between public and private universities. We’ll also see which schools rose and fell the most in the rankings since 2017 and examine UCLA’s strengths and weaknesses compared to its peers. We decided to use QS because their rankings provide a comprehensive look at universities across the world, not just in the U.S., and they utilize a robust and transparent methodology to score each school. For this article, we will limit our analysis to only the top 100 universities globally in 2024.

Which Factors Are Most Important?

To begin, let’s take a closer look at the distribution of overall scores among the top 100 universities.

The histogram appears right-skewed, as only a select group of prestigious universities scored higher than 90. Among the top 100, the 60–65 score range had the most schools, followed by schools around the 85 mark.

Here is a qualitative look at the distribution of the top 100 universities by size, focus, age, and country:

Among the top 100 universities, almost all are either large (≥ 12,000 students) or extra large (> 30,000 students) in size; over 80 are full comprehensive schools (all 5 faculty areas + medical school); and more than 70 are considered “historic” (100+ years old). Geographically speaking, the United States has the most top 100 schools, followed by the United Kingdom and Australia.

Remarkably, over 20 countries are represented by at least one top 100 school. We will dive deeper into the analysis by geographic region later in the article.

To explore the relationships between quantitative variables in our data, we will utilize a correlation matrix.

Looking at the matrix, we observe that Academic Reputation, Employment Outcomes, Employer Reputation, and Citations per Faculty are most strongly correlated with a school’s overall score. This aligns with the QS scoring methodology for 2024, which is as follows:

Employment Outcomes

Employment Outcomes is a new variable introduced in the 2024 edition of the QS World Rankings, and it only holds a 5% weight in the scoring, but arguably the biggest reason for people to attend a more-highly ranked school is to obtain better job opportunities.

This graph illustrates that going to a higher-ranked school generally leads to better employment outcomes. All but one of the schools that received an overall score of 95 or above scored a perfect 100 for employment outcomes. For more information about how QS measures employment outcomes, see this link.

Sustainability

Another newly introduced variable in the 2024 edition of the QS rankings was sustainability, which is becoming a point of emphasis for many colleges. However, is a focus on sustainability actually important when it comes to a school’s overall score?

There does not seem to be much of a relationship between a school’s overall score and sustainability, although all of the schools that received an overall score of 90 or higher scored higher than 80 for sustainability. The R-squared value between overall score and sustainability is just 0.10, which may suggest that sustainability does not need to be a main focus for schools to improve their rankings.

Comparing Schools by Geographic Region

The map above visualizes the impressive global geographic distribution of the top 100 schools. North America seems to have more prestigious universities compared to other regions, while Europe and the rest of the world seem to have a higher concentration of lower-scoring schools among the top 100 universities. As mentioned earlier, over 20 countries have at least one top 100 school!

At a broader regional level, Europe actually has the most top 100 schools, followed by North America and Asia.

However, as we observed from the map, it seems like North America has a higher proportion of top-ranked schools compared to other regions. Defining “top-ranked” as ranks 1–20, “average-ranked” as 21–50, and “low-ranked” as 51–100, we obtain the following pie charts:

While Europe has the most top 100 schools, 62.2% are classified as low-ranked. On the other hand, 32.3% of the top 100 schools in North America were top-ranked, the most of any region. Unexpectedly, 30% of Oceania’s top 100 schools are considered top-ranked.

Unsurprisingly, schools in North America have the highest median score compared to the other regions, followed by Oceania, Asia, Europe, and Latin America. North America is known to house many of the world’s most prestigious universities, which is reflected by the high median and Q3 scores for North American universities. Europe has a notably lower median score but also has exceptional schools like the University of Cambridge and the University of Oxford, as represented by the high upper tail in Europe’s boxplot.

As for employment outcomes, North American schools dominate the other regions due to North America’s strong and robust economy. Colleges in North America are known to offer and encourage internship/co-op programs that give students work experience while they are still in school, and the overall entrepreneurial culture and economic freedom in North America allow for better employment opportunities. Conversely, schools in Europe have a much lower median employment outcome score. This discrepancy could be attributed to the fact that Europe relies more heavily on industries like manufacturing and agriculture and also has stricter labor market regulations compared to North America.

Here is a stacked bar chart to compare schools by region in several other variables:

One of our observations was that Latin America actually has the highest average academic reputation and employer reputation scores among the regions. While Latin America does have the fewest number of top 100 schools, their universities, like the University of São Paulo and Universidad Nacional Autónoma de México, tend to be ranked highly in specific fields or disciplines and have strong academic and employer reputations in certain areas. Among the other variables, Asia had the highest average faculty-to-student ratio, Oceania had the most citations per faculty and international student ratio on average, and Europe had the highest average score for international research network. North American universities seem to be well-rounded in all areas when taking the average scores.

Public vs. Private Schools

Of the top 100 universities in the world, 80% are public universities, and 20% are private universities. The main differences between public and private universities are centered around funding, governance, and size. Public schools receive funding mainly from state governments, while private colleges rely on private sources of funding like tuition fees and donations, typically receiving little to no government aid. Along similar lines, public universities are regulated by state governments, while private schools usually have their own independent governing bodies. In terms of size and scope, public institutions tend to be larger and contribute to a broader range of academic programs and majors, whereas private schools are more focused.

Among the top 100 universities, private universities outperform public ones in most areas. While private schools generally scored higher, it’s important to remember that only 20 out of the top 100 fall into the private category—the rest are public universities. When we compare public and private schools, they are similar, with noticeable differences mainly in the faculty-to-student ratio and employment outcomes categories. Private universities often boast smaller overall enrollments, creating a more intimate learning environment, which translates to a higher faculty-to-student ratio score. Private universities also have higher employment outcomes as they offer specialized programs that closely align with industry needs, providing graduates with a competitive edge in the job market. The strong alumni networks associated with private schools play a pivotal role by offering valuable connections, mentorship opportunities, and potential job placements.

Looking at the kernel density estimate, there is a much wider range of overall scores and employment outcomes for public universities, and private institutions tend to score higher for employment outcomes across various levels of overall score.

Risers and Fallers

The two graphs above show the five biggest risers and fallers in rank from 2017 to 2024 among U.S. universities. The biggest riser was UC Berkeley, gaining 18 spots, and the biggest faller was Ohio State University. The 2024 rankings in particular were influenced by the new methodology that QS implemented:

First of all, 10% of the overall weight was deducted from both Academic Reputation and Faculty Student Ratio, and 5% weightage was added to Employer Reputation. On top of that, QS has decided to add three new criteria to their ranking system, each carrying a 5% weight: International Research Network, Employment Outcomes, and Sustainability. These changes explain why there is a trend of universities having a consistent ranking before significantly rising or dropping in 2024. One school that unfortunately faced the consequences of these changes was Rice University. Rice was ranked 90th in 2017, ascending to its peak at rank 85 in 2020. However, it then experienced a steady decline, plummeting to rank 145 in 2024.

Another intriguing observation is the rise in rankings among Australian schools from 2017 to 2024. When looking at the top 100 colleges worldwide (without restricting it to the U.S.), many of the biggest risers were schools from Australia. The University of Sydney, the University of Melbourne, the University of New South Wales, and the University of Western Australia all rose more than 25 places in the world rankings, heavily benefiting from the new QS weighting system. These schools often struggled in the Academic Reputation and Faculty Student Ratio categories, which both saw notable weight drops, and they tend to excel in other areas, like International Research Network, which was a newly introduced metric in 2024.

How about UCLA?

UCLA excels in academic and employer reputation, achieving a perfect score in both categories, and the university sets its students up with excellent employment opportunities. UCLA also performs well with sustainability, international research network, and citations per faculty. However, UCLA struggles when it comes to international faculty and student ratios, as well as faculty-to-student ratio. UCLA is known for its high enrollment numbers, resulting in larger class sizes and reduced faculty-to-student ratios.

The QS ranking system certainly favors UC Berkeley, as the university climbed to the #10 rank in 2024 and achieves maximum scores in employment outcomes, employer reputation, academic reputation, and sustainability, with a near perfect score in international research network as well. Similar to UCLA, faculty-to-student ratio and international student ratio are two of UC Berkeley’s biggest challenges. UCSD was the only other UC school to rank in the top 100 globally. UCSD’s biggest strength is its international research network, and UCSD also scores well in academic reputation, but it struggles in employment outcomes compared to UC Berkeley and UCLA.

In case you were wondering, USC ranked outside of the top 100 (116th), so we decided not to bother with them.

Rank Over Time

While UCLA’s rank methodically dropped from 31st in 2017 to 44th in 2023, its ranking soared back up to #29 in the world in this year’s edition of the QS rankings.

Conclusion

While we all see the signs around campus that say UCLA is the #1 public university in the U.S., we hope that this article helps shed light on the factors that actually go into determining the college rankings. Each school ultimately has its own individual strengths and weaknesses that a singular ranking will never be able to fully encapsulate. For future analysis, we could compare different ranking systems (QS, CWUR, Times Higher Education, U.S. News & World Report, etc.) against each other to observe areas of consensus and disagreement. But for now, we hope you enjoyed reading our article, and we would like to wish Gene Block a happy retirement!

Sources

--

--