You are in :

POLAR FAQs

Here are some frequently asked questions relating to the POLAR maps, grouped by topic.

If you cannot find the answer you need, we may be able to help if you e-mail m.corver@hefce.ac.uk.

Participation rates

How is the young cohort size estimated?

We developed our own small area annual cohort estimate method which uses 1991 Census and child benefit data. This method is described in full in Annex A of the 'Young participation in higher education' report.

How are the entrants counted?

The number of entrants is determined by linking together records of HE students studying in UK higher education institutions and GB further education institutions. The definition used for entrants in the POLAR measure is very similar to that used for the entrants for the YPR(A) statistic which is described in Annex C of the 'Young participation in higher education' report. There are some slight technical differences in the two measures that are a consequence of refinements of the methods in the two years between the development of POLAR and the publication of the report.

What about part-time study or undergraduate study other than degree, HND or HNC?

The restriction to full-time degree, HND and HNC programmes is required to ensure compatibility between different data sources and years. Fortunately, this definition captures nearly all of the participation for this age group. Annex C in the 'Young participation in higher education' report looks at the effect of changing the entrant definitions. It finds that, for example, including part-time study modes and HE diplomas and certificates increases the number of entrants from England by only a few percent. It also shows that the participation rates for just these 'excluded' entrants are highest for high participation areas and lowest for low participation areas. Together the small number and similar distribution of these 'excluded' entrants means that if it were possible to robustly include them then the local participation patterns would be very similar to those shown in POLAR.

What are the differences between the POLAR participation quintiles and those used in the 'Young participation in higher education' report?

The POLAR participation ward quintiles are very similar to the quintiles used in the report that are formed by ranking wards using the YPR(A) measure. The quintiles are not exactly the same since

  • POLAR uses quintiles of the GB cohort, the report considers England and Scotland separately
  • POLAR uses three cohorts; an extra cohort (the 2000 cohort) was available for the report
  • POLAR uses the 1999 version of the All Fields Postcode Directory; the work in the report uses the 2003 version
  • the POLAR quintiles are set using the integer participation bands given in the map keys so that they are not exactly quintiles of the young cohort
  • as the report is concerned with change through time it excludes a small number of micro-areas that are thought to have changed in nature since 1991. POLAR includes these areas in the participation rates.

However, the correspondence is good between the POLAR quintiles and the YPR(A) [and the HEI only YPR(H)] based quintiles in the report. In particular, the findings in the report about trends in quintile participation and the different nature of high and low participation areas can be taken to apply in general terms to the POLAR ward quintiles.

Where can I get area participation rates that use the exact definitions from the 'Young participation in higher education' report?

Data sets and tables of YPR(A) rates for regions, LSC areas, local education authorities and parliamentary constituencies are available on this site. These are based on the data sets used in the 'Young participation in higher education report' and give participation rates for each of the cohorts reaching 18 between 1997 and 2000.

Areas

Do institutions such as boarding schools or halls of residence distort the local participation rates?

There is careful processing to handle cases where an HE entrant returns, for example, a boarding school or university term time address on the student records. Accordingly there should not be any significant distortion from these cases but let us know if you suspect there might be a problem of this kind with a particular small area.

I know this area - the participation rate does not seem possible.

First check that the area you know is the same physical area shown on the map. For example, sometimes a ward might include several localities but the ward name may refer to only one. If you still think the rate shown is very different from what you would expect then let us know.

This small area has changed a lot over the period shown - how should I interpret the results?

The participation rate shown is an average for the 1997 to 1999 18 year-old cohorts. If the type of children living in an area has changed through new building or redevelopment then the rate shown is, of course, harder to interpret.

Are the areas used too mixed to be useful for looking at participation inequalities?

It depends. One of the purposes of having the POLAR maps for a hierarchy of geographies is to show that the larger geographies can be useful for a participation overview but mask the real area inequality in participation that is apparent at the level of neighbourhoods. Our work has suggested that wards are small enough geographical units to approximate these participation neighbourhoods while retaining sufficient numbers of young people so that meaningful rates can be shown for each area.

Annex F in the 'Young participation in higher education' report examines these issues in more depth. It concludes that, as far as can be determined, wards generally show acceptable internal participation homogeneity and that low participation micro-areas are overwhelmingly found in low participation wards. Note that this finding applies only to young participation. It is possible for a ward to have rich families and poor pensioners and yet correctly appear as homogenous using young participation: this is because we are only considering a section of the residents who might well be more homogenous than the ward residents as a whole.

Why are the ward young participation rates reported as quintile bands in the tables?

Although three cohorts are used for the POLAR rates the typical combined cohort size for wards remains small at around 150 (see Table 28 in 'Young participation in higher education'). With these small cohort sizes the proportional variation in the observed participation rate resulting from random fluctuations is quite large (see Annex F in 'Young participation in higher education'). Because of this the ward rates are reported as quintile bands as this fulfils POLAR's purpose of providing a robust discrimination between high and low participation areas and avoids focusing on small differences between wards that are indistinguishable from random fluctuations. For the larger units, such as parliamentary constituencies, the proportional random variation is much less and the actual participation rates are given.

Using POLAR

What do I need to view the maps?

The maps are in portable document format (PDF) and can be used with a pdf viewer such as Adobe Acrobat Reader.

Why do the maps appear blank or give error messages?

This is usually caused by using older versions of Acrobat Reader. Try using the latest version.

How do I use the maps with Acrobat Reader?

First make sure that you have the 'basic tools', 'file' and 'viewing' toolbars enabled (Windows>Toolbars). The +/- magnifying glasses will zoom the view in or out. You can also hold down the left mouse button to drag out a rectangle to enlarge or reduce it. Select the 'hand tool' and hold down the left mouse button to move the map. To find a particular ward, zoom in to the desired scale and then use the find button (binoculars) and enter the ward name. With Acrobat Reader 6 the 'snapshot' tool copies the selected area to the clipboard as an image.

Why are some of the ward and school names overprinted?

This is most likely to happen if there is a wide range of population densities in the region, two schools share the same site or a school happens to be in the middle of a ward. It is a limitation of our automated production of the maps which we have tried to minimise by making the maps large.

Why are some schools not on the maps / Why are some schools shown which have closed?

The POLAR maps show the young participation rates for the cohorts that were 18 between 1997 and 1999. These cohorts would have left their final year of compulsory education between 1995 and 1997: we have used the results from these years so that the school GCSE results relate to the participation cohorts. This has the consequence that new schools are not shown and that some schools are shown that may have since closed or merged.

Why are schools missing from the Scottish and Welsh maps?

The school data are based on the Government's School and College Achievement and Attainment tables (www.dcsf.gov.uk/performancetables) which only cover English schools.

Why are the 1998 ward boundaries jagged and sometimes have detached areas?

These boundaries are aggregated for 1991 Census enumeration districts (see the POLAR guide). We advise using the 1991 Census wards unless you particularly need the 1998 geography - both maps show participation for the same cohorts, only the boundaries used are different.

How do I print the maps?

The portable document format (PDF) files used in POLAR are designed and optimised for viewing on screen. In particular they are able to show large areas with fine detail by exploiting the ability of most PDF viewers to allow functions such as zooming and searching.

Most of the POLAR maps are very large (the ward maps are based on A0 sheets) and are not designed to be printed. They can be printed using large format colour printers (though the colours, designed for screen viewing, are often too dark on paper) but will be illegible on standard A4 or A3 printers.

Can I use the maps in my regional strategy (etc) report?

It is permissible to reproduce extracts from the POLAR maps in reports, etc (on a non-commercial basis) provided that the following acknowledgment is given with the map.

POLAR series (www.hefce.ac.uk/polar). Copyright Higher Education Funding Council for England. Some boundaries derived from data provided with the support of ESRC/JISC copyright ED-LINE consortium.

There are a number of ways to incorporate the PDF maps into a word processed document, none of them straightforward. We are not able to help you with this but can suggest some methods that have worked for some users. For small extracts using a screen capture (such as the snapshot tool in Adobe Acrobat Reader 6) can work. More technical users might want to consider using utilities based around Ghostscript (www.cs.wisc.edu/~ghost/) to convert the PDF files to image files or encapsulated postscript files (which, after adjusting the bounding box to the desired extract, can be inserted into some word processing software directly).

Will POLAR be updated?

The purpose of the ward-level POLAR maps is to show the location of areas of high and low participation. The small size of wards (with a typical annual cohort of 50) means that any realistic year-on-year change in participation would be swamped by random fluctuations associated with small counts. One solution to this is to group small areas into larger groups. This is the approach of the 'Young participation in higher education' report and it finds that, for the 1994 to 2000 cohorts, the local pattern of young participation inequality is not changing much.

Nevertheless, the reliability of the ward participation rates would be further improved if they were based on more than three cohorts. It is our intention to maintain POLAR and, in due course, update it so that the participation rates are formed from more cohorts and the local maps are based on the 2001 Census geography. As an intermediate step, some extra tables showing annual participation rates based on the YPR(A) measure from the 'Young participation in higher education' report have been added to POLAR.

The text on the maps is too small to read.

Use the zoom function in Acrobat reader to enlarge the selected area. See also "How do I use the maps with Acrobat Reader?".

Last updated 12 July 2007