POLAR4 for Russell Group


We have recently completed a dashboard for use which takes the POLAR4 methodology and re-maps participation into HE for Russell Group institutions.  It has been noted that using POLAR for high tariff institutions was becoming increasingly problematic as a metric as the likelihood of learners coming from low participation neighbourhoods into these universities was small. This project was commissioned and funded by UCL.

While there are methodological problems with POLAR4 (particularly in terms of the use of ONS forecast data to establish the base populations) any new measure of participation needs to align with POLAR4 for the sake of accurate comparison.


POLAR4 differs from POLAR3 in two main ways. Firstly, the unit of geography that was used in the calculations of POLAR3 (and indeed POLAR and POLAR2) were parliamentary wards. For POLAR4 MSOAs are used (middle super output areas) which are units of geography that are based on population numbers, rather than parliamentary constituencies. This is preferable as the MSOAs have a more even distribution of population, whereby the minimum population is 5,000 and the mean average is 7,200. POLAR4 covers participation for the academic years 20010/11 through to 2014/15, thereby updating the participation rates of young learners (aged 18 and 19).

Alongside this, POLAR4 is the first iteration of POLAR to include all countries in the UK which has enables us to comment on participation rates in Wales, Scotland and Northern Ireland for the first time.

We followed the POLAR4 methodology to calculate the participation rates of young learners by mission group – in this instance for Russell Group. We purchased the participation data from HESA for the five cohorts of young people who were aged 15 at the start of the school years that began from 2006 to 2011 (aged 18 or 19 when entering HE) between the years of 2009/10 and 2013/14 (aged 18) or between 2010/11 and 2014/15 (aged 19).

The baseline populations will be the same as POLAR4 and these data from the POLAR4 workbooks were used.

The geographies used were:

  • MSOA 2011 (England and Wales)
  • SOA 2011 (Northern Ireland)
  • Intermediate Zones 2001 (Scotland)

For the full POLAR4 methodology including the underlying population estimates, please see POLAR4 Classification.


The dashboard was developed in R and shiny.

It can be found by clicking here (and will open in a new window).

Due to the large file sizes, only the England maps were put online, although of course the calculations were done for all countries.

The map takes approximately 10 seconds to load (dependent on local network speeds) but once loaded there is no lag time at all in navigating the map.

dashboard snapshot

Despite the problems with the underlying population figures with POLAR4, the dashboard we have developed for the Russell Group institutions is rigorous in its comparison with the overall POLAR4 maps as we used the same methodology and the same underlying populations. Therefore the quintile shifts deal with comparative proportion of learners attending any versus Russell Group institutions, thereby sidestepping the issue of faulty population figures.

I am sure that many will find this version of POLAR4 useful and we hope to create dashboards for other mission groups.

Please feel free to contact us regarding this, or any other project.


This entry was posted in Higher Education, progression, widening participation. Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s