Similarities and differences between the Indices of Deprivation across the UK
Please note this information is correct as of August 2015. The detail has not been updated to reflect the most recent English Indices of Deprivation 2015 , but the principles regarding using Indices of Deprivation in the United Kingdom still apply
Indices of deprivation identify areas of multiple deprivation at the small area level.
Based on a methodology developed by the Social Disadvantage Research Centre at the University of Oxford, separate indices have been constructed for England, Northern Ireland, Scotland and Wales.
Though not directly comparable, each index is based on the concept that distinct dimensions of deprivation such as income, employment, education and health can be identified and measured separately. These dimensions, sometimes referred to as 'domains' are then aggregated to provide an overall measure of multiple deprivation and each individual area is allocated a deprivation rank and score.
The indices are used to help target policies and funding, and reinforce a common goal to improve the quality of life in disadvantaged communities. However, the indices may not be used together to create a single UK index.
The purpose of this section is to provide a broad overview of the methodology used to construct the four separate Indices of Deprivation. Those wishing to access more detailed information should follow the links to the Indices' technical reports shown in the 'Useful links' section.
The methodologies used to derive the separate indices of multiple deprivation are broadly similar. In each case, the index measures the level of multiple deprivation experienced by individuals in small areas. Firstly, levels of deprivation are measured for a number of separate dimensions or 'domains'. The number and title of domains varies between each index but each measures deprivation across key themes of income, employment, education and health.
For each domain, deprivation is calculated according to a series of summary statistics or 'indicators'. These indicators are designed to measure key features of a particular deprivation theme. The choice of indicators is agreed through consultation and varies between each index. Typically there are between two and six indicators per domain. (For more details see the 'Indicator comparison table' link in the 'Useful tables' section below).
For each index, the indicators are combined in broadly the same way to provide a domain level measure of deprivation. Where possible and most notably in the income and employment domains, indicators are summed and divided by the 'at risk population' (for example, the number of income support claimants as a proportion of the total population) to give an overall area rate of deprivation. Where rates are not possible, appropriate weights for combining indicators into a single deprivation score are selected using a range of techniques including Maximum Likelihood factor analysis.
The domain level scores which represent specific dimensions of deprivation are then ranked and transformed to an exponential distribution. Weights are then applied to provide the overall Index of Multiple Deprivation for each country in the UK. The weights are selected according to a number of criteria which are used to assess the level of importance attached to each domain theme in the overall measurement of deprivation. As each index uses different indicators and domains, the weights applied to aggregate domain scores are different. Table 1 provides a summary of the domains and weights used in each index.
Table 1: Domains and percentage weights applied for each Index
In each index, the areas which have the highest rank of 1 are those areas which are considered most deprived. The deprived can then be grouped together by the user within various bands, for example, the 10%, 15% and 20% most deprived areas. Figure 1 highlights the patterns of deprivation according to each Index.
Figure 1: Indices of Multiple Deprivation (rank) for each constituent country in the United Kingdom
Download a pdf version of this map (pdf 419 Kb)
For reasons outlined in the next section, areas found in the 20% of most deprived areas in England do not necessarily experience the same level of deprivation as found in the 20% of most deprived areas in Wales. Equally, the areas with rank 1 in England, Wales, Scotland and Northern Ireland are not necessarily the four most deprived areas within the UK.
Whilst the methodologies used to construct the four separate indices are conceptually similar, separate indices of deprivation have been developed in acknowledgement of the differences between each country. Aside from different policy objectives, the countries themselves are different. For example, Wales and Scotland have vastly different landscapes and a greater proportion of rural areas than England. The 'key' factors which influence the levels of deprivation within each country are therefore likely to be different.
The following sections provide further detail on the differences between the indices and by implication, why they cannot be used together as a single UK Index of Multiple Deprivation.
1. Different underlying indicators used to construct each Index
A key difference between the indices constructed for the different administrations is the individual indicators on which they are based. The first difference is in the number of indicators used. Scotland uses only 38 indicators whereas Northern Ireland uses 52. The differences are particularly apparent in the number of indicators used to represent accessibility of services: in the English index only four indicators are used whereas Northern Ireland uses 12. The relative level of influence of each indicator on the overall level of deprivation will therefore be different in each index.
As well as different numbers of indicators, there is much variation in the way indicators are defined. Again using the accessibility of services theme, whilst the key services considered are largely compatible across the four nations, the way in which accessibility is measured is not. England focuses on road distance to key services, Scotland measures accessibility by drive time and public transport travel time, Wales considers the average time to access key services by public and private transport, and Northern Ireland uses travel time by car.
In addition, the choice and definition of indicators reflects the different influences which exist in each country as well as the data available. For example, the education system in Scotland is different to the other nations and the choice of education indicators for the Scottish Index reflects this.
Finally, in the Scottish Index, the income and employment domains are allocated higher weights and as a result may have more influence over the overall level of multiple deprivation than the same domains in the English, Welsh and Northern Irish indices. However, it should be noted that research completed for the Scottish Index showed that changes in the domain weights made little difference to the overall deprivation score. The extent to which this is true for other indices is unclear.
The 'Indicator comparison table' link in the 'Useful tables' section (below) provides a complete summary of the indicators and domains used by each Index.
2. Currency of the data used and timing of updates
The current indices relate to different time periods. Therefore, even where definitions of the indicators are compatible across the indices, deprivation is not being measured for the same time period and we are not comparing 'like with like'. This is perhaps most acute in the income and employment domains which have the most commonality in indicator definitions across the Indices.
Each administration is subject to different controls on the timing and frequency at which the indices should be updated. Reasons include: policy cycles, user input, access to new administrative or statistical data and the availability of resources, which are likely to differ across the four countries.
As each new index seeks to make use of the best information available at the time, new indicators may be adopted. This has the potential to further exacerbate the differences between the indices. Nevertheless, differences in timing and frequency of updates can also be seen as an opportunity for continuous improvement of the indices. Each nation is committed to learning lessons from others and will seek to harmonise underlying indicator definitions where appropriate.
3. Different underlying geographies
Even if the indicators used for each index were compatible and from the same period in time, the biggest obstacle to comparability between the indices is the geography on which deprivation is measured.
Each index is based on measuring deprivation at the small area level. Different interpretations of these 'small areas' exist between England and Wales, Scotland, and Northern Ireland. The key differences are the population sizes of the small areas and the criteria used to aggregate census output areas. Table 2 summarises the population sizes for each country and target population design thresholds.
Only the Welsh IMD relates to small area geographies that were created following the 2011 Census, the other IMDs relate to small area geographies created following the 2001 Census.
Table 2: Population sizes and thresholds for small areas used
1 Based on 2001 Census geographies
2 Based on 2011 Census geographies
The Indices for both England and Wales are based on data collected at the lower layer super output area or 'LSOA' level of geography. LSOAs were generated by aggregating census output areas (OAs), taking into account population size, mutual proximity and social homogeneity.
Northern Ireland uses super output areas (SOAs) but they have been generated using physical and electoral ward boundaries. The population for a SOA in Northern Ireland averages around 2,000 people. Multiple deprivation scores for SOAs in Northern Ireland therefore relate to slightly more populous areas than that for England and Wales.
Scotland's small areas are known as 'data zones'. There are 6,505 in total and they are smaller than the areas used by other nations, averaging a population size of around 800 people. Data zones were created by aggregating 2001 Census output areas. The data zones group together areas with similar social characteristics and in contrast to the LSOAs, they take account of physical boundaries.
The size of the small areas on which the different indices are based is by far the biggest difference to note. At half the target population size of Northern Ireland's SOAs, the Scottish Index has the potential to pick up on smaller pockets of deprivation. The same is true for England and Wales albeit to a lesser extent than for Scotland.
Indices of Deprivation are designed to identify small area deprivation and so assist in it the development of more targeted policies and more informed funding allocation. As each country has its own responsibility for tackling deprivation, and the combination of factors which influence the levels of deprivation in each country may differ, separate indices have been constructed to help address the issue more effectively.
For these reasons, the indices cannot be used as a single UK Index of Multiple Deprivation. The indices use different underlying indicators and domains, are updated at different times and frequencies and most critically, use different small area geographies to record levels of deprivation. It is for these reasons that areas ranked in the top ten most deprived areas in England do not necessarily experience the same levels of deprivation as the top ten most deprived areas in Scotland, Northern Ireland or Wales.
Opportunities for making the indices more comparable, where appropriate, will be sought through future updates and it is likely that through each development, lessons will be learned. These lessons will be shared across the four nations for the continual improvement to the measurement of deprivation.
Using Indices of Deprivation in the United Kingdom - 2014 Guidance Paper (PDF 296 Kb)
Indicator comparison table - a downloadable table showing a complete listing of indicators for each index and how they compare (45 Kb xls).
Key components and differences - a downloadable table highlighting key components of the four indices of deprivation and how they compare (25 Kb xls).