University and subject league tables methodology

CONTENTS

1. Where does the data come from?

2. What measures do we use?

3. 2025 league table scores

4. Arts, Drama and Music institutions

5. How we ensure accuracy

6. Notes

Where does the data come from?

All the data in our university league tables comes from sources in the public domain. 

The National Student Survey and the 2021 Research Excellence Framework are overseen by the regulatory bodies for UK universities: 

• Office for Students (OfS) for England
• Scottish Funding Council (SFC)
• Higher Education Funding Council for Wales (HEFCW)
• The Department for the Economy in Northern Ireland

The following data are collected from UK universities by Higher Education Statistics Agency (HESA), and supplied by Jisc, the UK digital agency that supports further and higher education:

• Entry standards
• Student-staff ratios
• Spending on academic services
• Facilities spending
• Graduate prospects
• Continuation
• International student enrolments

In a few cases, the source data wasn't available and was obtained directly from the individual universities. We have a protocol for queries and corrections to the league tables.

What measures do we use?

Entry Standards (maximum score n/a)

The average UCAS tariff score of new undergraduate students.

Where does it come from?

HESA data for 2021–22.

How does it work?

Each student's exam results were converted to a numerical score (UK A Level A*=56, A=48 ... E=16, etc.; Scottish Highers A=33, B=27, etc.) and added up to give a score total.

For the ranking, points derived from Scottish qualifications were discounted by 25% to allow for the fact that the Scottish education system gives more opportunities for students to accumulate points, and so provide a fairer comparison with providers with most of their students from other parts of the UK. However, in the published table it is the raw score that is displayed, i.e. before the discount is applied.

An average for all students at the university was calculated and the results were then adjusted to take account of the subject mix at the university. Students on a foundation year were excluded. 

What should you look out for?

A high average score (over 200, or A*AAA, at some universities) doesn't mean all students score so highly or that you need to take lots of A levels to get in. The actual grades needed will vary by subject and few, if any, courses will ask for grades in more than three subjects. Universities that have a specific policy of accepting students with low grades as part of an access policy will tend to have their average score depressed. 

The UCAS tariff does not include some international qualifications taken by international students. This means that the average tariff score for providers with international students can be lower than it would have been if these qualifications were assigned tariff points by UCAS. 

Changes to the way A level and other grades were awarded in 2021 because of the pandemic resulted in a general increase in UCAS tariff scores in 2021–22 HESA data. 

Student satisfaction (maximum score 4.00)

A measure of student views of the teaching quality at the university. 

Where does it come from?

The National Student Survey (NSS), a survey of final-year undergraduate students in 2023.

How does it work?

The NSS asks questions about a variety of aspects of teaching. The average satisfaction score for the first 24 questions in the survey was calculated and then adjusted for the subject mix at the university. Due to the distribution of the data, and to avoid this measure having an undue influence on the overall ranking, the z-score was divided by three. 

What should you look out for?

The survey is a measure of student opinion, not a direct measure of quality, so it may be influenced by a variety of biases such as the effect of prior expectations. A high-ranked university expected to deliver excellent teaching could score lower than a lower-ranked university that, while offering lower-quality teaching, does better than students expect from it. 

Research quality (maximum score 4.00)

A measure of the quality of the research undertaken in the university.

Where does it come from?

The 2021 Research Excellence Framework (REF) undertaken by the regulatory bodies.

How does it work?

Each university department entered in the assessment exercise achieved a quality profile that gave the proportion of research in each of four categories from 4* to 1* (with any remaining activity being unclassified). For the research assessment measure, the categories 4* to 1* were given a numerical value of 4 to 1, which allowed a grade point average to be calculated. An overall average was then calculated weighted according to the number of staff in each unit of assessment. If no submission to the REF was made, then the minimum score was used. 

What should you look out for?

All staff with responsibility for research were submitted in the REF but universities vary in the proportion of staff who are designated teaching only. This issue is addressed by the research intensity measure (below). 

Research intensity (maximum score 1.00)

A measure of the proportion of teaching staff involved in research.

Where does it come from?

The 2021 Research Excellence Framework (REF) HESA data from 2019–20.

How does it work?

The number of teaching staff eligible to be submitted to the REF was multiplied by the proportion actually submitted and divided by the total number of teaching staff. If no submission to the REF was made, then the minimum score was used. 

What should you look out for?

Universities can decide who is designated with responsibility for research. In some cases, researchers might be switched to teaching only as a way of getting the best possible quality profile and so the research intensity measure can be an underestimate of the actual research intensity. The REF also allowed for some staff who were eligible to be submitted to nonetheless be omitted. 

Graduate prospects – outcomes (maximum score 100.0)

A measure of the success in employability or further study of graduates completing their first degree. 

Where does it come from?

The Graduate Outcomes survey which took place fifteen months after the graduation of students in HESA data for 2020–21. 

How does it work?

The number of graduates who take up employment or further study is divided by the total number of graduates with a known destination, before being expressed as a percentage. Only highly skilled employment (i.e., where graduates are normally recruited) was included. The results are then adjusted to take account of the subject mix at the university. 

What should you look out for?

A relatively low score on this measure doesn't mean many graduates were unemployed. It may be that some had low-level jobs that don’t normally recruit graduates. Some universities recruit a high proportion of local students and so if they're located in an area where graduate jobs are hard to come by this can depress the outcome. 

Graduate prospects – on track (maximum score 100.0)

A measure of the proportion of graduates who agree that their activity is on track with their future plans.

Where does it come from?

The Graduate Outcomes survey which took place fifteen months after the graduation of students in HESA data for 2020–21.

How does it work?

The proportion of graduates who agreed or strongly agreed with the statement ‘My [activity] fits with my future plans’.

What should you look out for?

Graduate's plans vary and this is a measure of how happy they are with their current situation in relation to those plans. The graduate prospects score provides a measure of absolute success while this provides a measure of how happy the student is with their situation regardless of how successful it might appear to be. 

Student-staff ratio (maximum score n/a)

A measure of the average staffing level in the university.

Where does it come from?

Calculated using HESA data for 2021–22.

How does it work?

A student-staff ratio (the total number of undergraduate and postgraduate students divided by the number of academic staff) was calculated, with the results adjusted for subject mix.

What should you look out for?

A low student-staff ratio (a small number of students for each member of staff) doesn't guarantee good quality of teaching or good access to staff.

Academic services spend (maximum score n/a)

The expenditure per student on all academic services.

Where does it come from?

HESA data for 2019–20, 2020–21 and 2021-22. 

How does it work?

A university's expenditure on library and computing facilities (staff, books, journals, computer hardware and software, but not buildings) plus museums, galleries and observatories. This figure was divided by the number of full-time equivalent students in the latest year. Expenditure over three years was averaged to allow for uneven expenditure.

What should you look out for?

Some universities are the location for major national facilities, such as the Bodleian Library in Oxford and the national computing facilities in Bath and Manchester. The local and national expenditure is very difficult to separate and so these universities will tend to score more highly on this measure.

Facilities spend (maximum score n/a)

The expenditure per student on staff and student facilities.

Where does it come from?

HESA data for 2019–20, 2020–21 and 2021-22. 

How does it work?

A university's expenditure on student facilities (e.g. sports, careers services, health, counselling) was divided by the number of full-time equivalent students in the latest year. Expenditure over three years was averaged to allow for uneven expenditure.

What should you look out for?

This measure tends to disadvantage some collegiate universities as it mostly includes central university expenditure. In Oxford and Cambridge, for example, a significant amount of facilities expenditure is by the colleges, but this does not appear in the university’s accounts.

Continuation (maximum score 100.0)

Continuation measures the proportion of students who continue in their studies after their first year. 

Where does it come from?

HESA data for 2020-21 and 2021-22. 

How does it work?

The number of first-year students who continued to the next year, left with a qualification or transferred to another university was divided by the total number of first-year students. 

What should you look out for?

The way universities dealt with the pandemic had a positive impact on progression and so this increased continuation rates. 

Summary of measures

How we compile the tables

• A statistical technique called the z-transformation is applied to each measure to create a score – this ensures each measure contributes the same amount to the overall score and so reduces any need for scaling 
• Six of the measures are then adjusted to take account of the subject mix at a university (a side-effect of this is that it’s impossible to recalculate the total score in the tables using the published data as you'd need full access to all the raw data) 
• The z-scores (adjusted z-scores where the subject mix was taken into account) on each measure are then weighted (see table above) and totalled to give an overall score for the university 
• These total z-scores are then transformed to a scale where the top score was set at 1000, with the remainder being a proportion of the top score – this scaling doesn't affect the overall ranking, but it avoids giving any university a negative overall score 

2025 league table scores

The measures for the subject tables are the same as for the main university league table, except only six measures are used: entry standards, student satisfaction, research quality, continuation, graduate prospects – outcomes and graduate prospects – on track. 

• A research intensity measure as used in the main table was not possible for the subject tables. However, there was data on the proportion of staff eligible for REF 2021 who were actually submitted. This was used as an additional measure in the background but is not displayed, partly to avoid confusion with the intensity measure in the main table and partly because two-thirds of the outcomes are the same. 
• If subject level NSS data were not available, aggregated data across similar subjects were used.
• For continuation and graduate prospects, two years of data were aggregated; scores were withheld where the number of students was too small to calculate a reliable percentage. 
• The calculation of the overall score is also the same, except there's no need for any subject mix adjustment and weights are 1.0 for student satisfaction, entry standards, and continuation, 0.8 for research assessment, 0.67 for graduate prospects – outcomes, and 0.33 for graduate prospects – on track (with % staff submitted at 0.2). 
• In tables where the range of data for continuation or graduate prospects is very small, the weight has been reduced to avoid the measure having a disproportionate impact on the overall ranking. 
• To qualify for inclusion in a subject table, a university normally must have data for student satisfaction and at least two from entry standards, continuation and graduate prospects – outcomes. If student satisfaction is missing, then all the other measures must be present.
• A blank (n/a) in one of the columns isn't a zero score but indicates that no valid data was available, and the final score was calculated using the data that was available. 

Arts, Drama and Music institutions

• Our Arts, Drama & Music league table includes several specialist colleges that don't meet the full criteria for inclusion in the main table – some of these institutions will also be listed in their relevant subject tables 
• Note that other institutions also offer courses in Arts, Drama and Music – you can find them in the relevant subject tables and on the main universities ranking table 
• The methodology used is the same as for the main table 

How we ensure accuracy

Each university is provided with complete sets of its own HESA data well in advance of publication. In addition, where we find anomalous figures in the HESA data, universities are given a further opportunity to check for and notify any errors. 

Notes

Main university table

• Birkbeck, University of London: declined to release its data for use in all UK league tables. 
• Highlands & Islands: given its unique collegiate structure with 12 academic partners, it would be inappropriate for the university to appear in the table. 
• The Open University does not appear in the table because its students are distant learners and so data are mostly unavailable. 
• Ravensbourne & University College Birmingham:  asked not to be included in the table. 
• NSS data not available: Cambridge and Oxford. 
• Student-staff ratio (SSR) data provided by the university: Buckingham and Exeter.  
• Staff and student facilities spend provided by the university: Buckingham (2019–20 and 2020–21), East London (2019–20), London Met (2019–20) and Queen Mary (2019–20). At the time of initial publication 2021–22 data were not available for Northampton. This is now available and has been incorporated.
• Academic services spend provided by the university: Buckingham (2019–20 and 2020–21), East London (2019–20), London Met (2019–20), Loughborough (2019–20) and Queen Mary (2019–20). At the time of publication 2021–22 data were not available for Northampton. This is now available and has been incorporated.
• Continuation data not available: Falmouth (average of other z scores used). 
• Research GPA data not available: Buckingham (minimum used). 
• Research intensity data not available: Buckingham (minimum used). 

Arts, Drama & Music league table

• Research GPA data not available: Liverpool Institute for Performing Arts and Arts University Plymouth (minimum used). 
• Research Intensity data not available: Liverpool Institute for Performing Arts and Arts University Plymouth (minimum used). 

Advisory board

We consult the universities on methodology. Once a year, an Advisory Board of university and other relevant experts meets to discuss the methodology and how it can be improved. 

We have taken every effort to ensure accuracy but cannot take responsibility for errors or omissions. The data providers don't necessarily agree with the data aggregations or manipulations we use and are also not responsible for any inferences or conclusions derived from them. 

• READ MORE
• League tables background and inclusion criteria