Methodology
Population
23 . This analysis uses linked National Pupil Database ( NPD ) 19 to UCAS data . The initial population includes 10 cohorts of learners who obtained their Key Stage 4 qualifications ( most commonly GCSEs ) between summer 2010 and 2019 , while they were in year 11 of secondary education . We start with this relatively long time series to consider recent trends within the broader context of how participation rates were changing before the Uni Connect programme was introduced .
24 . Learners in this population , which we refer to as the ‘ Key Stage 4 population ’ in this report , will have :
a . Attended a state-funded mainstream school or college in England . b . Lived in England . c . Been 16 years old by the end of their Key Stage 4 academic year .
25 . Table B1 in Annex B provides population counts of these learners , for those living in areas targeted by Uni Connect and those living in other areas .
26 . NPD data for this population of Key Stage 4 learners was linked to UCAS applicant data using personal characteristics such as name and postcode . The linking approach is ‘ fuzzy ’, in that it takes account of differences in how personal characteristics are recorded between the datasets , by allowing for typos and misspelling of names , for example . Ultimately this enables us to track how many of the original population of 16-year-old learners in England applied through UCAS and were then accepted to start a higher education course by age 18 . 20
27 . It should be noted that if a learner applies to higher education more than two years after their final Key Stage 4 year , this would not be captured . Additionally , some applicants may be accepted to start a higher education course without using the UCAS Undergraduate scheme . For example , they may apply to a conservatoire , or to a part-time course . Nonetheless , most applications made by 18-year-olds will be through UCAS . 21
19
The Department for Education ( DfE ) does not accept responsibility for any inferences or conclusions derived from the NPD data by third parties .
20
The linking process does not use information from Key Stage 5 . This is done to avoid introducing a time series bias , whereby more recent cohorts of learners are less likely to match than earlier cohorts because their Key Stage 5 information is not yet available . This means that the proportion of learners identified as having applied through UCAS ( the application rate ) is likely to be slightly underestimated in this report .
21
A small number of applicants are placed through UCAS routes outside of the main scheme , including Direct Clearing and Records of Prior Acceptance ( RPA ), which are both included in this analysis . Between 1,105 and 2,430 applicants were placed through Direct Clearing in each cycle between 2016 and 2021 , compared with between 855 and 2,320 placed through RPA .
12