Benchmarking

Our current approach

The Office for Students (OfS) uses benchmarking to make meaningful comparisons between higher education providers, different student groups and other groups or units of interest.

Because there are such differences between some of these groups and providers, average values for the whole of the higher education sector are not necessarily helpful when making these comparisons.

To account for some of the factors that contribute to these differences, we calculate a sector average that is then adjusted (‘weighted’) for each group based on the characteristics of the students in that group or the provision being offered.

The sector average, adjusted for these factors, is called the adjusted sector benchmark. It can be used in two ways: 

• to understand how well a provider has performed compared with performance for similar types of students, on similar types of courses in the higher education sector as a whole 
• to assess similarities between groups and providers. 

We use a set of guiding principles to select factors for each measure we are benchmarking. This involves taking account of the results of statistical modelling of the measure in question, as well as our policy objectives for benchmarking. As a result, the factors we use in calculating benchmarks vary according to the measure we are benchmarking.  

We use benchmarking in our regulation of student outcomes, the Teaching Excellence Framework (TEF) and the National Student Survey (NSS). If the same measure is used in more than one area of our regulation, we will normally use the same benchmarking factors. 

Methodology

Our current methodology to benchmarking is well established and has been used in the UK Performance Indicators (UKPIs) since 1997. 

Read more about our benchmarking approach and methodology in our description of student outcome and experience measures. 

Technical details of the methodology are also described in ‘Statistical analysis of performance indicators in UK higher education’ by D. Draper and others, Journal of the Royal Statistical Society, Series A, volume 167, part 3, 2004.