Background and purpose
1. HEFCE’s Participation of Local Areas (POLAR) classification is a UK-wide measure of educational disadvantage based on young participation rates in higher education (HE). It estimates how likely young people are to go into HE according to where they live, and assigns local areas into quintiles. The population is split into five groups and assigned equally across five quintiles, where quintile 1 areas have the lowest rates of young participation and quintile 5 areas have the highest rates. This means that the quintile used in the classification applies to the local area and not to each individual.
2. This report gives details of the latest release of the classification, POLAR4. It is an update to a series of reports that HEFCE has published since the first release of POLAR in 2005.
3. This report explains the methodology of how young participation rates are calculated for local areas. It provides analysis of the POLAR4 classification and discusses how it differs from the classification it has superseded, POLAR3.
4. Each area in the UK is classified in a POLAR quintile according to the level of young participation in that area. The young participation rate is calculated by dividing the number of young people from each area who enter HE aged 18 or 19 by the cohort population in that area.
5. POLAR4 has been calculated using administrative data for five cohorts of young people. The five cohorts are those who were aged 15 at the start of the school years that began from autumn 2006 to autumn 2011, and hence who could have entered HE between the academic years 2009-10 and 2013-14 aged 18 or between 2010-11 and 2014-15 aged 19.
6. The methodology for POLAR4 differs from previous releases in the choice of geographical location, and because of changes in the availability of data. ‘Local areas’ have been defined as middle layer super output areas (MSOAs) in England and their equivalents in the devolved nations.
7. Local young participation rates span almost the full range of possibilities, with some areas having 100 per cent young participation. This is the first POLAR classification to have no areas with a participation rate of 0 per cent. Just over half of the young population live in areas where the participation rate is between 20 and 40 per cent. The median participation rate across all areas in the POLAR4 classification is 37 per cent.
8. Large differences exist between the median participation rates from quintile 1 areas and quintile 5 areas. On average young people from quintile 5 areas are around three times more likely to enter higher education than those who live in quintile 1 areas.
9. Across the four nations in the UK, Northern Ireland and Scotland have the highest median rates of young participation. Young participation is lowest in Wales. Comparing English regions, the median level of young participation in London is at least eight percentage points higher than all other regions.
10. These differences in young participation rates mean that there is variation in the distribution of POLAR quintile across nations and regions. At a national level, the lowest proportion of quintile 1 areas is in Scotland, while the lowest proportion of quintile 5 areas is in Wales.
11. At a regional level, London looks very different from the other regions in England because it has much higher levels of participation in higher education. Approximately 45 per cent of local areas in London are classified as quintile 5 compared with 1.3 per cent classified as quintile 1, so that for every POLAR4 quintile 1 area in London there are around 34 quintile 5 areas. The North East has the greatest proportion of quintile 1 areas.
12. Participation rates have also been calculated at a lower geographical level (lower layer super output areas in England). Local areas are likely to be more homogeneous at this level, but smaller population sizes mean that the participation rates are less well estimated. This trade-off between the robustness of the participation rates and the homogeneity in local areas means that MSOA level is the preferred geographical area for the calculation of participation rates for POLAR4, but comparison between the two is useful to investigate the sensitivity of the POLAR4 classification to choice of geographical area. Overall, around 90 per cent of these smaller areas would be assigned to the same quintile or move up or down by just one quintile if these smaller area participation rates were used for POLAR4. This indicates that POLAR4 is consistent across different geographical aggregations.
Comparison between POLAR3 and POLAR4
13. While the main difference between POLAR classifications is the set of cohorts used to form them, POLAR4 has some further differences. First, the methodology used to calculate population estimates for POLAR4 is different from those used in previous iterations of the classification. This change was necessitated by a change in government policy that meant that a data source previously used will not cover the required level of information in the future. The second difference is the change in geography level used for local areas to align with population geography levels published by the statistical bodies across the UK.
14. The median young participation rate is higher for all five quintiles in POLAR4 than POLAR3. The largest percentage point increase is for quintile 5, but the greatest proportional increase is for quintile 1. Comparing classifications across 2,560,000 postcodes, we found that just over 55 per cent of them are in the same quintile in both POLAR3 and POLAR4. A further 35 per cent of postcodes have moved up or down by one quintile. Only a very small proportion of postcodes (1.3 per cent) moved from quintile 5 in POLAR3 to quintile 1 in POLAR4 or vice versa, and these large, but rare, changes at postcode level are associated with the change of geographical area used for the POLAR classification.
15. This document is for information only.