Northern Ireland Assembly Flax Flower Logo

NORTHERN IRELAND ASSEMBLY

COMMITTEE FOR
FINANCE AND PERSONNEL

Multiple Deprivation Measures

4 November 2009

Members present for all or part of the proceedings:

Ms Jennifer McCann (Chairperson)
Mr Peter Weir (Deputy Chairperson)
Dr Stephen Farry
Mr Simon Hamilton
Mr Fra McCann
Mr Mitchel McLaughlin
Mr David McNarry
Mr Adrian McQuillan
Mr Declan O’Loan
Mr Ian Paisley Jnr
Ms Dawn Purvis

Witnesses:

Dr Norman Caven )
Dr David Marshall ) Northern Ireland Statistics and Research Agency
Miss Cathryn McBurney )

The Chairperson (Ms J McCann):

Appearing before the Committee today are Dr Norman Caven, the chief executive of the Northern Ireland Statistics and Research Agency (NISRA), and Dr David Marshall and Miss Cathryn McBurney who are also from that agency. You are all very welcome. I remind witnesses, and those in the Public Gallery, that the session is being recorded by Hansard. Therefore, all mobile phones must be switched off completely.

I invite the witnesses to make their presentations.

Dr Norman Caven ( Northern Ireland Statistics and Research Agency):

Thank you very much.

We are before the Committee today because in 2008 Departments decided that there should be an update of the measures of multiple deprivation. That decision was taken at an official level, and the Minister of Finance and Personnel subsequently wrote to Executive colleagues informing them that the decision had been taken. There were no objections. Following that, a consultation document was published and circulated in 2009, and a copy was also submitted to the Committee. NISRA is now in the process of obtaining the views of those consultees

The multiple deprivation measures are normally updated approximately every five years, usually quite soon after a census has been completed and again in the middle of any given decade. Essentially, those measures look at the concentrations of deprivation at a certain geographical level. What we are trying to do is measure something which is relative. It is not absolute deprivation that we are measuring; we are measuring it relative to the Northern Ireland average, and we are looking at concentrations of deprivation. That does not mean that everyone in a particular geographical area is deprived, because there will be deprived people who live in wards that are not included in the measures of multiple deprivation and non-deprived people who live in deprived areas. We are not just looking at poverty; we are looking at a whole range of domains, such as education and access to services.

We have been doing this for several decades. The information that came out of the 1971 census informed the Belfast areas of need; the 1981 work informed the Making Belfast Work initiative; and the 1991 material informed the targeting social need objectives. Therefore, there is a long history in Northern Ireland of using those measures as an adjunct to the social policies that attempt to tackle deprivation at an individual level.

Through the decades we have moved away from an initial reliance solely on census variables, which tie us to the time when the census of population is carried out. As rich sources of administrative data have become available we have tried to capture those in order to make the measure of deprivation more gainful, more refined and more developed. That process has been going on, and we have employed outside experts to help us in that — Professor Brian Robson in the 1990s and Professor Mike Noble since 2000.

We as statisticians will produce the measure, but Departments have the latitude to use that in their policies as they see fit. Some will use the measure of multiple deprivation, which combines all of the domains of deprivation into one overall measure. Others will use single domains because their policies are focused, for example, on education, and they will use the results of the education domain alone. The material that we have provided to the Committee includes examples of how the Department for Social Development and the Department of Education use the measures. Again, that is a matter for those Departments after the information has been produced, although we do provide guidance on how it could be used.

Mike Noble carried out some work in 2001, and in 2005 using the 2001 census material that was becoming available then; those are our current measures of multiple deprivation. In 2001 and 2005 when we were combining those measures into an overall measure, we gave a weighting to each of them according to their importance. Income is given a weighting of 25%, employment 25%, and so on down to crime and disorder with 5%. That weighting has been in place since 2001.

The material that we have provided to the Committee also outlines how we compare with the other countries of the United Kingdom and the Republic of Ireland. Professor Noble has also produced measures for England, Wales and Scotland. The most recent updates to those measures have been made at different times; in England in 2007, in Wales in 2008 and in Scotland in 2009. Those areas all hold to the common Noble methodology that we are going to employ for the update in Northern Ireland.

Measures in the Republic of Ireland have been based more on their census of population, which, as you may know, takes place every five years rather than every 10 years, as in the UK, although the Combat Poverty Agency is looking at trying to produce something in addition to the census approach, utilising a Noble-type approach.

Dr David Marshall ( Northern Ireland Statistics and Research Agency):

Departments have agreed to an update of the 2005 measures, which we are now doing, but that is strictly an update; we are not talking about a review or a fundamental change of methodology. The Noble methodology will continue to be in use, but we are consulting on how to measure deprivation most effectively. Any significant issues that are raised through the consultation will be taken on board for a fuller review after the 2011 census.

Statisticians in NISRA are working with colleagues in all Government Departments in Northern Ireland. It is a collaborative effort across government because of the way we measure deprivation in different domains. There is a steering group consisting of senior officials from the Departments, with representation from the director of research at the Equality Commission, the chief executive of Omagh District Council and various representations from external parties who are users of the measures of deprivation.

We have run a consultation exercise, which closes on Friday 6 November and has included a number of public meetings. We plan to run a peer review process. Once the results of the consultation come back, we as a steering group will look at those and an external peer review will also look at them. We will see how those match together.

The 2005 measures have 43 indicators, and, this time, our consultation document proposes a temporal update of those statistics — moving the statistics from 2003-04 up to the current day. We are retaining 35 of those 43 indicators. We are only revising where Noble has explicitly recommended that in the 2005 report, where data is no longer available — I will talk about that later — or where we are moving away from the use of the 2001 census towards the use of administrative data.

Noble sets criteria for indicators, which must be domain specific. For example, a number of people asked that receipt of free school meals be included in the education domain as an indicator of educational disadvantage. Receipt of free school meals is an indicator of income disadvantage; it is a passport from income-related benefits. Educational disadvantage is measured by GCSE scores and educational attainment, rather than through receipt of free school meals. That is the approach that is taken; indicators must be specific to their domain.

The indicators must measure major features of deprivation. They can be classed as major because of the numbers of people or because they are deemed to be important by the community. The indicators must be applicable across all of Northern Ireland. They must be measured in a consistent way across Northern Ireland, and they must reflect rural and urban issues. I will come on to that when I talk about the health domain. The figures have to be up to date and robust.

We must have data that is available for the whole of Northern Ireland at small-area level in a consistent form. Many people want specific indicators brought in that are just not available for the small areas that we require. We effectively created, for the 2005 review, a geography called super output areas. There are around 900 super output areas in Northern Ireland, each with a population of around 2,000 people. The key is that they are consistent in population size. We produced data on the seven domains and on the overall measure at super output area level.

We have also produced statistics at output area level. There are over 5,000 output areas in Northern Ireland, and they are much smaller, with around 350 people in each one. This time, we hope to produce more statistics at output area level. We have been asked to do that, and statistics for that level are quite important for understanding targeted need. However, there are issues with obtaining statistics at that level. For example, when assessing health deprivation in areas where there are only 300 or 400 people, there can be issues to do with confidentiality.

We also produce summary measures by Assembly constituency, local government district and electoral ward. We have provided the Committee with maps of the 26 district council areas, the 890 super output areas, the 5,022 output areas in Northern Ireland and some detailed maps of the Belfast area. I do not propose to talk about that in detail, but detailed maps are available on the NISRA website for those who are interested in specific output areas and super output areas. That is the geographical framework to which we work. The maps are fascinating, and I could talk about them all morning.

To give you a flavour of the proposals to update the deprivation measures for 2009-2010, I will concentrate on two domains. We have similar information across all seven of the domains. I will concentrate on the income deprivation domain and the health deprivation and disability domain. How does Noble define income deprivation? The domain is conceptualised as:

“the proportion of the population experiencing income deprivation at the small area level.”

Typically, we have taken account of the number of people who are on income-related benefits as a proportion of the population. We also create statistics for children and for old people.

There is a great deal of similarity between what we did in 2005 and the consultation document’s proposals in 2009. In 2005, we used income support households. We continue to use those households in 2009-2010, but the benefits system has changed since 2005. We are now using state pension credit households and employment and support allowance households as well. That is a change in the way that the information is captured in the social security system. Similarly, jobseeker’s allowance, which was included in 2005, is also included in 2009-2010.

Tax credit has arisen as an issue in the consultation exercise. In 2005, working families’ tax credit and disabled person’s tax credit were the tax credits that were available. Those have changed to child tax credit and working tax credit. When putting together the consultation exercise, we talked to colleagues in the Revenue Commission, but, unfortunately, the information on tax credits was not available to us. Therefore, we proposed to go forward without using the tax credit information. That has come forward in the consultation as a significant issue, and we have been in discussions with the Revenue Commission and the Department for Social Development about getting access to that information. I cannot say that we will be able to do that, but we hope to be able to include that information; that is one of the themes that is coming through the consultation. We also hope to be able to add information on housing benefit, and that will include rate rebates. Our proposal for 2009-2010 is effectively a count of the people who are receiving any of those benefits, as a proportion of the population.

The health deprivation and disability domain identifies areas where there are high rates of premature death or areas where relatively high proportions of the population’s quality of life is poor, because they have poor health or are disabled. In 2005, we had four indicator areas: years of potential life lost — effectively, mortality statistics that identified areas where people were dying younger than the 75-year threshold; comparative illness and disability ratio, which covered areas where there were high levels of disability as measured through social security benefits; a combination of people who had poor mental health or anxiety disorders, based on prescriptions, and suicides; and areas in which large numbers of people were registered as having cancer.

We took account of the age structures of areas. That is one of the things that we do within the health domain to ensure that we are comparing like with like — that and the number of people.

We propose to continue with those four indicators, with slight differences in the mental health indicator, in 2010. We want to include people who are staying in mental health hospitals, along with three further indicators. Noble referred to the health of children in his 2005 report; it is seen as a key development. The percentage of babies with low birth weight is seen as a marker for poor health later in life, as is poor dental health.

The other indicator is where the urban/rural divide comes in — emergency hospital admissions. In 2005, it was thought that there might be some urban/rural bias in that area; people in urban areas might find it easier to go to hospital, because they lived closer to them. Our colleagues in the Department of Health have looked at that, and they say that there is no urban or rural influence in cases where patients stay in hospital for four or more days after emergency admission.

There are another five domains, but I will not go into those this morning.

Dr Caven:

The consultation period will finish on 6 November. We will collate the consultation responses and publish them on the NISRA website. There will then be a peer review of the final proposals, and they will be mediated through the interdepartmental steering group, which will also involve the stakeholders. We will then be ready to decide what measures we should use in the exercise. We will advise the Committee of that before we crank any numbers through. We will then produce the actual results and, if the Committee so wishes, we will come back and present those. Finally, we will hold some dissemination events for potential users in spring 2010.

The Chairperson:

It seems a bit unfair that a large proportion of the working families on low incomes will be left out of this because of the lack of information about tax credits.

Dr Caven:

As Dr Marshall said, we are trying to ensure that we get that information. The people who are in receipt of any of the benefits mentioned will be included. A lot of people who receive a tax credit will actually be included because they are in receipt of some other benefit as well. There will be some people at the margins who are in receipt of only a tax credit and who will be excluded if we do not get that variable. We are trying to rectify that.

Mr Weir:

I understand what is being said; however, tax credits are one of the bluntest instruments. I appreciate that people on low incomes can receive tax credits, but they can apply to high incomes, too. People who are reasonably well off can also get tax credits.

The Chairperson:

What you have suggested, and what I have heard through different consultations with the community sector, is that people on working families’ tax credit will no longer be considered. That gap needs to be sorted out. The information must be collated and put into that income domain. A number of people who belong to low-income families or who are on the borderline will not to be included in that domain.

Dr Marshall:

We will include only those people on tax credits whose income is less than 60% of the median. We will not include everybody who receives tax credits; we will include only those people who are at the lower end of the income spectrum as identified through the tax credit process. We are working towards trying to get that. It all goes back to the fact that the Revenue Commission stopped supplying information to people after it lost some information in late 2007. Effectively, that information would have been available to us but for the fact that some information was lost in 2007.

The Chairperson:

Are you working to remedy that?

Dr Marshall:

Yes.

Mr Hamilton:

You never know, you might find the information on a train.

I am a bit of a nerd, and I am quite fascinated by this sort of thing, and by the maps in particular. Perhaps, I should just put my anorak on now and be done with it.

Mr Weir:

That is Committee uniform.

Mr Hamilton:

A lot of the changes that you are suggesting are wise, sensible, right and proper. I want to talk about geography and the scale at which we can measure deprivation in different areas. If we continue to do that at super output level or higher, we are not going to learn a terrible lot about anything, because no matter which measures are changed, whether it be one benefit or another, the same areas will still be in the top 10, 20 or 100 areas of deprivation. The same areas will constantly come up.

I encourage you to do all that you can to get more and better information at output level — I am glad to hear that you are talking about that. My constituency of Strangford is perceived to be quite affluent and well-off and not having any deprivation or disadvantage issues. However — and David McNarry can back me up on this point — small areas in the constituency are missed because they are surrounded by extremely affluent areas. That happens in every constituency. Sometimes, we have difficulty in securing interventions to solve problems in certain areas because the data at super output level does not support what we know to be the situation on the ground. This is more a point of encouragement, but the more information you can obtain at the lower output level, the more that will help us to get the interventions that we sometimes find difficult to secure. Anything you can do to enhance the information at that level will be very useful.

Dr Caven:

In 2005, we applied the income and employment domains to the smaller geographical area that you mentioned. As David said, we will try to extend that this time round. There are some difficulties with using very small numerations; for instance, confidentiality regarding the information that is used, and the robustness of that information can also be a problem. Contingent on those two problems being solved, we will try to extend the information gathered at that output area level.

Mr McNarry:

Simon and I represent the same constituency, and I support his comments. However, sometimes the problem goes deeper; sometimes, the point that Simon made is not recognised as there is no data to support it. For example, with respect to targeting social need, it was very difficult to prove that there was need in small areas without prying in people’s private affairs and almost physically pointing out those who were affected.

You talked about services, and moving away from using the population figure of 10,000; is that necessary, given the current circumstances? Pockets of deprivation will be particularly affected by the closure of post offices and schools and by the removal of accident and emergency services. Many of the services that you identified are being removed, reduced, or are likely to be reduced. Will you take into account the effects that those cuts could have on the pockets of deprivation that, as Simon said, are hard to identify?

Dr Marshall:

We did not talk in detail about proximity to services deprivation domain. There will be some changes to that domain, such as the service centre indicator, which you mentioned. One aspect of the domain is the measurement of how close people are to a large service centre. We have moved away from considering this from the perspective of populations of 10,000 or more. Cathryn will detail how we have done that.

Mr McNarry:

I understand that, and I am not criticising you. I am saying that we are witnessing the closure of post offices and other service centres. In other areas, service centres are under threat. That may only be temporary; I do not know.

Dr Marshall:

We will use the information available in 2009, but there will be some slight differences. For example, Tyrone County Hospital was raised as a key issue during one of the consultation meetings: it no longer provides accident and emergency services. Therefore, the hospital will be removed from the list of hospitals with accident and emergency services.

Miss Cathryn McBurney ( Northern Ireland Statistics and Research Agency):

Instead of looking at a population of 10,000, we now compare towns and cities on the range and types of services that they have. Mr McNarry mentioned post offices and hospitals with accident and emergency services, which are two separate indicators. Areas with a low number of post offices will show up as being more deprived than those with a high number. Similarly, an area that is a long distance from a hospital with accident and emergency services will be shown as being more deprived.

Mr McNarry:

I understand that. However, your figures are based on the situation as it was yesterday; I am talking about likely impacts on people tomorrow. I am asking you to have a look at the information and determine whether the people whom you consulted might be saying something different now. For instance, they may have said that they were not thinking of closing some A&E department, but they might be doing so now; or they may have said that a post office was not going to be closed, but it may be closed now. The same goes for shops and other services. What you are saying is important, but, unfortunately, it might need some adjustment.

Dr Caven:

We will look at that.

Mr McNarry:

Thank you.

Mr Weir:

Logically, we should then talk about the day after tomorrow; however, that sounds like the title of a disaster movie. I hope that we are not going into disaster territory.

Mr McNarry:

Do not get ahead of yourself.[Laughter.]

Mr Weir:

I agree with what Simon and David have said. I suspect that the situation is not unique to the North Down and Ards areas, but examining small areas is vital. I suspect that settlement patterns are small throughout Northern Ireland and that consequently, unlike England or Scotland where there is deprivation in vast estates coupled with a high level of social segregation, many deprived areas here are literally across the road from quite affluent areas. The need to drill down is vital, because it goes to the core of ensuring that areas are not left out.

When NISRA is drilling down and decides to use wards as the building blocks, are the divisions within wards determined on a purely arbitrary basis or are they based on geographical linkage? One could examine a couple of wards and come up with different results depending on where the geographical lines are drawn.

Secondly, you said that to ensure that you had a reasonably common methodology, the Noble methodology was commonplace in England, Wales and Scotland, and may not be that different from the one used in the Republic. Are the methodologies identical or just roughly similar? Of course, none of these things can be perfect; however, it is vital that the methodologies are as good as possible because of the impact that decisions based on them might have on communities. Is it possible to have an external review to verify that we have a robust and universally accepted methodology?

Dr Caven:

I will answer the first question. There are large and small areas within the super output areas that form the main units of analysis. They are based on the 2001 census of population, in which output areas with a degree of socio-economic similarity were created — in other words, houses with similar socio-economic characteristics were grouped together. In a super output area, the output areas are built up from similar types of dwellings and socio-economic characteristics.

Mr Weir:

Are those output areas created by examining individual cases?

Dr Caven:

Yes; they are built up from the individual census returns.

Dr Marshall:

With regard to similarities across jurisdictions; in essence, the situation in England, Scotland and Wales is not identical, but there are domains in all three countries in which mathematical mechanisms are identical.

The differences relate to variations in administration systems. For example, the education system in Scotland is different, so the indicators there are slightly different. Furthermore, the weightings for income and employment are not identical but are very similar. There are slight differences in the South of Ireland, but the concept of domains of deprivation is the same. Instead of having the seven domains that we have in Northern Ireland, the South has three. Nevertheless, in essence, the concept is identical. Cathryn will say a few words about the peer review.

Miss McBurney:

We hope to have an external peer review of our final decisions. At the moment, the issue is out for consultation, and when the responses come in, we, as statisticians, will determine which indicators go in. Our decisions will then be assessed by external academic experts, who will give their views on our blueprint and then pass it on to our steering group.

Mr Weir:

How are the external academic experts selected?

Dr Marshall:

We hope that Professor Noble will look at it for us. That is the plan; but he has not yet given a formal commitment to do so.

Mr Weir:

Do not take this the wrong way; but, given that this concept is based largely on Professor Noble’s analysis, can his assessment be considered to be external? Is it not like getting Dr Frankenstein to look at his monster?

Dr Marshall:

The domains and the fabric are his, but we are not consulting on those; we are consulting on the indicators. Professor Noble could have a significant impact on the changes that we have since 2005. We are updating what already exists: we are not looking at it in a totally new way.

Mr Weir:

I am a little bit concerned about how external and independent such scrutiny will be.

Ms Purvis:

I welcome the review and the consultation, particularly the efforts to obtain greater detail on the smaller output areas. We know how important those measures are in determining Government policies and services to tackle deprivation.

I want to return to the information available on people who claim tax credits. I concur with the Chairperson’s assertion that the working poor are not being taken into account in the income domain. Many families who receive tax credits do not claim, or are not entitled to claim, any other benefits, and are missing from the income domain.

With respect to the information that is not available to you from HM Revenue and Customs (HMRC), I have heard in another Committee that the Department for Employment and Learning has submitted a business case to HMRC for income tax data. In addition, the Department for Social Development has submitted a business case to receive tax credit information. It seems to me that there is a hole with respect to information held by Westminster Departments. Are the Executive making a business case to HMRC for that information, or are Departments applying for it on an individual basis? It seems to me that only one business case should be made for that information.

Dr Caven:

New needs can arise in Departments. However, you are right: to date, individual Departments have been making representations. I am also making representations, at my level, to the senior statistician in HMRC.

Ms Purvis:

Does it not make sense for all Departments to collectively make a single application to HMRC for all the relevant information?

Dr Caven:

As David said, we cannot promise that we will get the information. We will consider the issue to be significant if we are not able to get the information. In the first instance, there may be merit in seeing how the separate representations go.

Ms Purvis:

Is tax credit information used in multiple deprivation measures in England and Wales?

Miss McBurney:

Yes.

Ms Purvis:

So, there is no reason why they should not give the information to you?

Dr Caven:

That is what we would say.

Ms Purvis:

Employment support allowance includes the former incapacity benefit, but there is no reference to, for example, disability living allowance or carer’s allowance. There is evidence that not only do people with disabilities suffer levels of deprivation, their carers also suffer. Why is there no specific measure in that domain?

Dr Marshall:

With respect to the health domain, the comparative illness and disability ratio includes people who are not only in receipt of the incapacity element of employment support allowance and those held back on incapacity benefit as it has changed, it also includes those on disability living allowance and attendance allowance. It includes all the relevant disability and health benefits. As to your specific question about carer’s allowance, I am not sure if it is included in the —

Miss McBurney:

It is included in the employment domain. Employment, and whether someone can work, were measured, and the two benefits that you mentioned, DLA and carer’s allowance, are included as separate indicators, but not specifically in the income domain.

Ms Purvis:

Thank you.

Dr Farry:

I welcome the witnesses. Returning to the “nerd theme” mentioned earlier, how does your consultation exercise sit with the current review of boundaries by the Local Government Boundaries Commissioner? Potentially, your ward boundaries and district council boundaries will change, and the final recommendations on that will be laid before the Assembly in the next few weeks or months at the same time as this exercise is being conducted.

Mr McLaughlin:

Stephen is interested in the possible deprivation facing the Alliance Party.

Dr Marshall:

The wards, output areas and super output areas are linked to the current 26 council districts with respect to the framework. However, we have developed this exercise in such a way that we can re-work the data. The report will be published next spring under the current framework of super output areas and output areas, but we believe that we can rework the data relatively easily if there are to be 11 council districts and 460 wards.

The problem is that, beneath those 460 wards, we will have to create a new set of output areas and super output areas. As Norman has said, we created the output areas specifically to examine the problem of disadvantaged households within each area, using the 2001 census. Creating a new set of output areas will have to wait until after the 2011 census, because we will have to go through the process again. So, the small-area detail will have to remain with the output areas that exist, but we will reconfigure a new set of output areas after the next census.

Dr Farry:

You are running a risk. The political likelihood is that there will be a new set of boundaries by the time your figures are published. Your figures will be based on old ward boundaries and old district council boundaries, and decision-makers will have moved on to a new set of units.

Dr Marshall:

We can aggregate the statistics for wards and district councils to accommodate the new boundaries, but we will not be able to create a new set of output areas. The very detailed geography will have to wait until after the 2011 census. Most of those statistics, within council areas, will be sufficient.

Dr Farry:

Have you examined the new ward boundaries to see how many of the super output areas or output areas will cut across the new wards and how many will be unaffected?

Dr Marshall:

There will be significant changes to the ward boundaries, so we cannot continue to use the existing ones. They will not add up. We had hoped that that would not be so, but the changes are significant and we will have to reconfigure the geography after the 2011 census.

Dr Farry:

Will everything be reconciled after the 2011 census?

Dr Marshall:

Yes.

Mr F McCann:

Some of the questions that I had intended to ask have been asked by Dawn Purvis.

When weighting the deprivation measures, the crime and disorder domain was given 5%. Crime and disorder has a great impact on people from a psychological and mental perspective. Why has it been given such a low weighting?

Dr Caven:

That weighting has gone through consultation processes in 2001 and 2005, and we are proposing to leave it as it is in this consultation process. If issues arise about the weightings during consultation we will examine them.

Mr F McCann:

Therefore, the weighting was based on figures from 2001.

Dr Caven:

Yes, and 2005.

The weightings that are before the Committee today are the same as those that went into the previous consultations. No difficulties were raised about them then. If any issues are raised about them during the current consultation, we will examine them.

Mr F McCann:

That leads me on to my second question. If, during the consultation period, large numbers of people focus on crime and disorder, or other weightings, will those weightings be changed?

Dr Caven:

We will certainly consider that.

As I have said, once the consultation responses are received we will examine them and take a professional view. We will them mediate the data through the process of peer review, and the weightings will be debated again before we crank any numbers through.

Mr F McCann:

One thing that concerns me about consultations is that we go to the expense of asking people for their opinions, yet those opinions are not taken on board when it comes to the final results. In my view, when a consultation is carried out and a large number of people request change, that change should be made.

Dr Marshall:

We are not going to throw away the responses. In 2005, recommendations were made during the consultation exercise, and though we may not have identified or addressed those recommendations at that time, we have tried to address them in the current update. We may also be unable to address all of the issues that are raised in this consultation, at this time, but we will definitely take them on board. Sometimes it takes time to put systems in place or obtain the necessary data to support the changes that people suggest.

I reiterate that we will not throw the responses away. It is very important for us to capture that information, and it may be used this time or the next time we examine the deprivation measures.

Mr F McCann:

It is right that small pockets of deprivation are removed. However, in many of the wards the deprivation is generational, which is very difficult to deal with.

Are you concerned that when the new data come out, it will be used to create a smaller pot of money? If that is the case, it will make deprivation worse in the areas that the multiple deprivation measures initially sought to deal with?

Dr Caven:

Our job is to carry out the statistical exercise. As I said earlier, whenever Departments choose to use the results as the basis of their policy proposals, they may consider those issues. All of the money could be spent on the top 2% of the deprived areas or it could be layered very thinly over the top 30%. Those decisions are best taken by the Departments using the statistics available. We are not prescriptive in how the data is used.

Mr F McCann:

I understand that NISRA deals solely with statistics, but the statistics that you produce have a knock-on effect on areas that have suffered from social deprivation for ever and a day. Do you draw the Government’s attention to the fact that to effectively deal with deprivation, additional resources are required?

Dr Caven:

As I said at the beginning, there is a range of social policies dealing with social need, and they are sometimes mediated through the individual with respect to a lot of the indicators. For a long time, social policy has also examined areal units of deprivation, and programmes have been introduced to focus resources, as an adjunct to mainstream social policy.

Departments are best placed to decide where to focus those resources in their own programmes. Sometime programmes, such as Making Belfast Work or Belfast Areas of Need, have been interdepartmental. That is fairly well known to Departments’ senior officials.

Mr McLaughlin:

Welcome, and thank you for your presentation. I welcome the review and will not repeat the concerns that other members have raised, because you have dealt with them already. I have a concern, and it relates to the issue of free school meals. Given the difficulties that you are having in gathering all the inputs and data you need, I feel that free school meals is also a key indicator of deprivation. It has implications not only for the health and well-being of schoolchildren but for their employability, and so on, in later life. What are your thoughts on free school meals as an indicator of deprivation?

Dr Marshall:

Children who receive free school meals are, obviously, from low-income families, and we measure that through the income domain. There have been a number of debates on the education domain during the consultation exercise; and, in the past, we did not have information on primary school children because it was not captured at the required spatial level of detail. We are now able to gather that information, and we propose to include more detailed information on primary school children, particularly their performance in Key Stage 2 examinations.

Mr McLaughlin:

Will that information be based on the fact that a family would be in receipt of one or more passport benefit or free school meals?

Mr Marshall:

No; a school’s performance is measured at a certain level, and the children from that school come from various output areas. Effectively, we map educational disadvantage using educational outcome rather than family income. However, I appreciate that there is correlation between income and educational outcome.

Mr Weir:

When measuring the results, do you use geographical statistics on where the children come from to map the situation precisely?

Dr Marshall:

That is exactly how we do it. A school will have a performance level, and that is apportioned across its catchment area. At secondary school level, we use the pupil’s outcome; for example, we take account of the type of house that a pupil with GCSEs lives in. The measure is as accurate as we can make it.

One item that has been raised during the consultation is using English as an additional language as a measure of educational disadvantage. Our view is that that is not a direct measure of educational disadvantage. Rather, we measure outcomes, such as examination performance at Key Stage 2, absenteeism, special needs, and so on, as the primary measures. We try to make it domain specific.

Mr McLaughlin:

Your outline of the restructuring and the comparator with the previous system is very helpful. The restructuring is positive, and that type of methodology is the way forward. However, I am still intrigued as to why you do not wish to use the receipt of free school meals as an indicator of deprivation. There is a wider and spatial significance in the cohort of young people who receive free school meals, because it affects all sorts of issues.

You mentioned using English as an additional language, which throws up another question. Having English is part of the essential skills component of the various jobseeker programmes. I believe that having information on the receipt of free school meals is relevant and cross-cutting, and would ensure that you are aware of people’s situation at the earliest possible stage.

Dr Marshall:

The Department of Education uses information on free school meals, so Departments can act outside of —

Mr McLaughlin:

Are they members of your steering group?

Dr Marshall:

Yes.

Mr McLaughlin:

Am I right to assume that they have signed off on this approach?

Dr Marshall:

Yes.

Ms Purvis:

Are you saying that, for primary schools, you do not look at the individual outcomes for the pupils but map the areas around the school and the school’s performance?

Dr Marshall:

Effectively, we know what wards, output areas, or super output areas that the children in a class come from. We cannot link children with results at primary school level yet; however, we can do so for children at secondary school level. Therefore, at primary school level we apportion the performance.

Ms Purvis:

That could be skewed. Why, for example, do you not use outcome at Key Stage 1?

Dr Marshall:

We use Key Stage 2 outcomes, but, as yet, we do not use Key Stage 1 outcomes. I can check with my colleagues in education, but I think that there are issues around the detail of recording.

We have heard it said before that perhaps we should be trying to measure educational attainment not just at the ages of 10 and 15 but as early as possible. That is an issue, and, if we could gather that information, it would be a positive inclusion.

Mr McQuillan:

In 2011, when the RPA kicks in, surely the figures will be skewed again. However, there is going to be a census in 2011, and that should bring the figures back into line.

If you do not get the tax credit information, which is very important to the exercise, is it worthwhile doing it at all? After all, the report will not be published until mid-2010, and the new census will be taken six months after that. Therefore, the information in your report will be out of date almost before it is published.

Dr Caven:

The new census period will have begun; however, it takes time, particularly at that level of disaggregation, to get the census results out; and all the information must be collated. The census may be taken in 2011, but it could be 2014 before the exercise is carried out again. A census can be produced very quickly, depending on the amount of money that is thrown at it.

Mr McQuillan:

I understand, I was not thinking about it in that way.

The Chairperson:

Thank you very much for coming, and for your patience.