Indices of Deprivation 2015 released. What does it measure?

IMD 2015 Infographic as JPGThe Indices of Deprivation 2015 is today released by the Department for Communities and Local Government, having been updated by the Brighton-based Oxford Consultants for Social Inclusion (OCSI). The data and supporting guidance are available here:

It is the fourth incarnation of the Indices, which has previously been issued in 2004, 2007 and 2010. It is commonly used by Government, public-sector bodies and funding to assess levels of deprivation across England and Wales, often with a view to identifying and comparing deprivation between geographical areas. This is often used to target funding, resources and policy initiatives.

The Index brings together multiple datasets/indicators across 7 domains for each small scale geographical area (Lower Super Output Areas each containing on average between 1,000 and 1,500 people). These are then scored based on their attributes, with each area ranked nationally based on its score, with 1 being the most deprived and 32,844 being the least deprived.

The Index represents a useful strategic tool, providing a universal and standardised approach to exploring the multi-disciplinary nature of deprivation in a form which is available with complete coverage of England for free. It particularly appeals to those seeking to delineate one areas attributes from those of another in a definitive and evidence-based way. This has become increasingly prevalent in the light of political and financial imperatives to target those most in need.

However, before using it, it is important to consider what limitations or methodological obstacles may sit in the way of those seeking to use it. Taking it at face value can often lead to assumptions being made that cannot be substantiated. Ultimately, it is designed to identify the average levels of deprivation within a given geographical area at a particular point in time and comparing it in relation to that occurring in other similarly sized areas. There is a need to ensure that it is not employed for tasks to which it is substantially less suited or inferring meaning where it cannot be substantiated.

Factors to consider:

It is area rather than people orientated, which is an obvious distinction but one which is commonly blurred. Many deprived people do not live in deprived areas, nor is everyone in a deprived area deprived. It is easy to forget when using area-based classifications that the score or ranking for the area reflects its average and there will be people in that location whose characteristics do not align with these.

Analysis has previously shown that in Sussex approximately 60% of deprived people live outside of the 20% most deprived areas. A focus on the most ‘deprived’ areas may therefore lose sight of the greater proportion of people experiencing equivalent forms of deprivation but who simply live in areas rated by the Index as less deprived. The ultimate risk is that comprehensive targeting of only the ‘most deprived’ area by the majority of funders and service providers may lead a majority of those affected to receive less support purely based on where they live in relation to others with comparable needs rather than their level of need.

Areas classified as ‘deprived’ are not homogenous. The characteristics which may lead them to be classified in this way differ from place to place, as do the reasons which lead to these features being exhibited. It is crucial that the underpinning data of an area of interest is interrogated to determine the features that give rise to its ranking and to use this with broader contextual understanding to assess the causative factors which give rise to them. The Index does not provide an insight into the numbers of people classified in a specific way, for example, the numbers of unemployed. This can only be assessed by using actual measures of the numbers affected.

Relationships also exist between locations and it may be difficult to identify these causative factors from the ‘effects’ highlighted by the rankings set out in the Index. For example, adjacent geographical areas may have very similar or very different rankings and it is important to understand how such patterns have developed and why the differences occur.

As the nature of deprivation may differ from area to area, so too might the impacts experienced by people on the ground. As with all statistical models, there are certain attributes which cannot be measured and are therefore not reflected within the Index data. For example, issues may affect individuals or societal groups within the same area in different ways. Areas that are geographically isolated, with more dispersed populations or in possession of fewer assets and services may have additional barriers for those seeking to overcome their disadvantage, as well restricting the responses which support organisations and service providers may make.

Given that the Index brings together multiple pieces of information and translates this into a single average score and ranking, it is important to note that different areas may be classified in the same way for fundamentally different reasons. The very nature of an Index (bringing together multiple datasets) can mean that the average score and therefore the ranking obtained may result from one dataset/indicator being very high and one very low. To better understand the characteristics of each area, interrogation of each of the domains and the underlying datasets is required.

It is likely that there will be some movement in the classification of areas since the last incarnation of the Index of Deprivation in 2010. When exploring this aspect it is important to consider the ‘relative’ nature of the Index and the impact that this has when seeking to compare any change in an areas ‘ranking’ between 2010 and 2015. For example, if an area (X) experiences a beneficial rise in incomes, services and employment, one would expect it to reduce its level of deprivation and therefore see a decline in its deprivation ranking. However, if those areas with similar rankings (to X) in the 2010 Index have experienced a greater increase in their attributes, then X may remain in its original position in the rankings and possibly be classified as more deprived in the rankings even if on an indicator by indicator basis it shows improvement.

As outlined above, the Index does not provide a direct measure of deprivation but a relative one. Each area is positioned based on its relationship with other areas. The ranking attributed to each area are therefore not directly linked to characteristics that it may exhibit. For example, an area with a ranking of 100 is not twice as deprived as one ranked at 200. Additionally, on interrogation of the underpinning data an area rated as less deprived (based on rakings) may also have more people identified by some indicators, for example, being unemployed.

Fundamentally, it is important to recognise that some elements of deprivation exist in pretty much all geographical areas and there is a need to distinguish between deprivation and poverty. Deprivation is generally a broader and more encompassing term related to an inability to access key services, housing and other life essentials, whilst poverty places a greater emphasis on the limitations imposed through a lack of income.

However, the Index weights each of the 7 domains, with those associated with ‘poverty’ such as Income and Employment given more importance than those associated with factors such as Access to Services. This may lead to a focus on those areas where concentrations of this type are most evident (urban areas), rather than those experiencing access issues (rural areas). Therefore, by its very nature the IoD has a tendency to highlight these urban concentrations.


It is crucial for anyone seeking to use the 2015 Indices of Deprivation to be clear on what they wish to use it for and to ascertain whether this will provide them with the answers they require. Like all Indices and data models it possesses limitations and attempts to extrapolate from it further meaning than it can provide is potentially misleading.

Ultimately how far down the path of understanding the pattern of difference that exists between physical locations and the people that reside in them does the IoD 2015 take us? Surely it is merely the start point in a longer, more detailed and forensic process for exploring and responding to differences.

“Supporting rural communities in Sussex to be vibrant and diverse places in which to live and work.”