Skip to main content
Normal View

Thursday, 11 Apr 2019

Vote 4 - Central Statistics Office

Mr. Pádraig Dalton (Director General, Central Statistics Office) called and examined.

This morning we will have two sessions. The first is the 2017 Appropriation Accounts, Vote 4 - Central Statistics Office. The second is Vote 8 - Comptroller and Auditor General. We want to complete both sessions before the voting block at approximately 1 p.m. Members and witnesses will be pleased to hear that we are allocating one hour and 15 minutes to this session. The second session will cover the same period. Immediately after lunch, we will have a final session with Professor Chris Fitzpatrick on the National Paediatric Hospital Development Board. That will be after the voting session.

The first session today is to discuss the Vote of the Central Statistics Office. I welcome from the CSO Mr. Pádraig Dalton, director general, Ms Maria Hurley, Ms Jennifer Banim and Mr. Richard McMahon. This is Mr. Dalton's first appearance before the committee.

I advise the witnesses that by virtue of section 17(2)(l) of the Defamation Act 2009, they are protected by absolute privilege in respect of their evidence to the committee. However, if they are directed by the committee - that means me, as Chairman, acting on behalf of the committee - to cease giving evidence on a particular matter and they continue to so do, they are entitled thereafter only to a qualified privilege in respect of their evidence. They are directed that only evidence connected with the subject matter of these proceedings is to be given and asked to respect the parliamentary practice to the effect that, where possible, they should not criticise or make charges against any person, persons or entity by name or in such a way as to make him, her or it identifiable.

Members are reminded of the provisions of Standing Order 186 to the effect that the committee should refrain from inquiring into the merits of a policy or policies of the Government, a Minister of the Government or the merits of the objectives of such policies. While we expect witnesses to answer questions asked by the committee clearly and with candour, they can and should expect to be treated fairly and with respect and consideration at all times in accordance with the witness protocol.

I ask Mr. McCarthy, Comptroller and Auditor General, to make his opening statement.

Mr. Seamus McCarthy

The 2017 Appropriation Accounts, Vote 4 - Central Statistics Office, records gross expenditure of a little over €46 million. Appropriations-in-aid were €1.5 million, resulting in net voted expenditure of €44.7 million. The surplus for surrender at year-end was €3.9 million.

Members will note that the gross expenditure in 2017 was some €30 million less than in 2016. The difference is mainly due to the additional costs incurred by the CSO in carrying out the five-yearly census of population in 2016.

The appropriation account is presented under a single programme entitled Delivery of Annual Statistical Programme. An analysis of the expenditure is given in note 3 under standard administrative headings used in appropriation accounts. This also includes explanations for significant variances in spending on individual subheads relative to the amount provided. I issued a clear audit opinion on the appropriation account and there were no matters that I considered required reporting to Dáil Éireann.

Thank you, Mr. McCarthy. I ask Mr. Dalton to make his opening statement. I note the CSO has submitted an appendix, which we will note and publish. It is not necessary to read the detail of that document into the public record as it is an appendix.

Mr. Pádraig Dalton

I thank the committee for the invitation to discuss the 2017 appropriation accounts for the Central Statistics Office. I am joined by several colleagues. Maria Hurley is assistant director general with responsibility for the corporate affairs directorate, which includes the finance, governance and human resources functions. Jennifer Banim is assistant director general of the economic statistics directorate. Richard McMahon is assistant director general of the social and demography directorate. I propose to briefly outline the role of the Central Statistics Office as well as our strategy and achievements during 2017. I will also give an overview of the CSO 2017 appropriation account and outline CSO expenditure in 2017.

The CSO is Ireland's national statistical institute and is responsible for the production and oversight of the production of all official statistics for Ireland. The CSO is an independent office of the Civil Service under the aegis of the Taoiseach. The role of the director general of the CSO is prescribed in the Statistics Act 1993, which provides that the officeholder acts independently and exercises sole responsibility in professional statistical matters, including statistical methodology, professional standards, timing and content of statistical releases and methods of dissemination.

The National Statistics Board, with the agreement of the Taoiseach, has the general function of guiding the overall strategic direction of the CSO. This independent position reflects international best practice for the organisation of official statistics. The functions of the CSO, as set out in the Statistics Act, include "the collection, compilation, extraction and dissemination for statistical purposes of information relating to economic, social and general activities and conditions in the State." This mandate was broadened under EU Regulation No. 223/2009, which makes the CSO director general responsible for the co-ordination and oversight of the quality of all European official statistics compiled by public authorities in Ireland. Confidence in the quality and independence of official statistics is crucial as this information serves as an objective input to policy development, oversight and governance at national and international levels.

Under EU regulation, each member state was invited to commit to improve or maintain the conditions for the implementation of the European statistics code of practice via a commitment on confidence. In May 2017, the Government adopted a commitment on confidence in statistics. In signing this declaration, the Government reaffirmed the CSO's independence, impartiality and commitment to quality.

From a governance and planning perspective, good policies start with good data. Good data are the lifeblood of decision-making and the basis for accountability. The role of the CSO is to make sense of data by providing high quality statistics, independent insight and information for effective debate and decision-making across government, business and society. The work we do creates good quality statistics that help people to understand the changes taking place in Ireland's economy and society.

The key deliverables and objectives for the CSO are informed by national and European policy. The CSO delivers a demanding and growing statistical work programme each year. In 2017, a total of 345 electronic releases and publications were published, including 13 outputs on census 2016. We are currently preparing for the next census in 2021. In addition to electronic releases and publications, the CSO transmits data to EUROSTAT, the Central Bank, the European Central Bank, the Organisation for Economic Co-operation and Development and the World Health Organization, among others.

The European statistical system has a significant impact on the work of the CSO. Approximately 73% of the statistics we produced in 2017 were required under EU statistical legislation. The CSO is an active member of the European statistical system and is centrally involved in key committees and executive bodies across a range of domains.

We delivered across all our strategic aims in 2017. Alongside our 2017 statistical work programme, we completed the census 2016 project during 2017, producing some new style results publications and more than 30 individual infographics to support the communication of key releases. We published the first estimate of GNI*. This is a modified indicator of the overall size or level of the economy. We also published a modified total domestic demand indicator that focuses on activity within the Irish economy. This is designed to exclude significant globalisation effects that disproportionately affect the Irish economic results.

We established the Irish statistical system co-ordination directorate, which is leading on key strategic issues such as the development of the national data infrastructure and the co-ordination of professional statistical services to the broader Irish public system. We continued to explore the use of secondary and blended data sources to compile new analysis, progressing the evolution of official statistics through the use of emerging data resources while focusing on burden reduction for respondents.

We introduced a more cost-effective data collection methodology in our household survey domain, namely, computer assisted telephone interviewing, which has afforded respondents a greater degree of flexibility in the manner in which they engage with the office and support our field force in concentrating its efforts on respondent recruitment in an environment of persistently falling response rates.

The CSO ran the Civil Service employee engagement survey in 2017, which operated across the entire Civil Service under action 25 of the Civil Service renewal plan and provided important insights into the experiences and opinions of those working in the Civil Service. The results were compiled by the office before being supplied to the Department of Public Expenditure and Reform for analysis and publication and the resulting outputs were used to inform further development towards the reform of the Civil Service. We capitalised on new approaches to communication and user engagement by developing a new style of infographic and developing user applications on Brexit, house prices and the consumer price index among others.

The CSO also improved access to data for researchers by launching a remote access to microdata solution for researchers in 2017, significantly improving data security. From an internal statistical quality perspective, during 2017 the CSO continued its investment in the new quality and methodology support function, headed by a senior statistician and tasked with underpinning the CSO's efforts to consistently refine its approach to quality management in official statistics and its commitment to continuous methodological improvement. Overall, 2017 was a year of transition for the CSO, maintaining the historic focus on the systematic production of key economic and social indicators, while striving to broaden the focus and reach of official statistics to a more mainstream audience in response to the public demand for access to impartial and high-quality data to inform evaluation and decision making.

The net outturn for the CSO in 2017 was €44.727 million against an estimate provision of €48.584 million. This lower than anticipated expenditure resulted in a surrender of €3.856 million back to the Exchequer at the end of December 2017. The bulk of the saving related to expenditure on salaries, wages and allowances that are payable to permanent staff assigned to the office. The 2017 appropriation account was audited by the Office of the Comptroller and Auditor General.

I will briefly outline the office’s expenditure in 2017. First, with respect to salaries, the bulk of the office's expenditure is expended on salaries, wages and allowances that are paid to permanent staff assigned to the CSO. In 2017, this expenditure amounted to €36.080 million against an estimate of €38.983 million. Savings of €2.903 million were realised. The estimate allocation was based on staff numbers of 755 but the actual number serving throughout the year was lower at 736. This was mainly due to difficulties and delays in the filling of vacancies at statistician level and in IT technical grades during the year.

With respect to other non-pay, in line with other Votes, the CSO has several non-pay administrative subheads: A(ii) travel and subsistence; A(iii) training and development and incidental expenses; A(iv) postal and telecommunications services; A(v) office equipment and external IT services; A(vi) office premises expenses; A(vii) consultancy services and value for money and policy reviews; and A(viii) collection of statistics. Non-pay administrative expenditure of the office in 2017 amounted to €10.189 million against an estimate of €10.989 million. Savings of €800,000 were returned to the Exchequer. The detail of this expenditure is set out in appendix A, which has been circulated to the committee.

Our strategic goals were captured in our statement of strategy 2016 to 2019, which was published in late 2016. In this, we said that we aim to turn data and statistics into information and knowledge for all; to increase the use of secondary data sources; to continue to build the capacity of our people; to modernise our statistical processes and systems, and to co-ordinate, oversee and assure the quality of all official statistics produced in Ireland. The CSO is operating in an increasingly challenging environment. We are experiencing growing user demand, both nationally and from Europe, where EUROSTAT and the European Central Bank are the two main sources. The CSO is increasingly being asked to take on new survey work or to incorporate existing studies into its programme of official statistics, as can be seen by the recent announcement to embed the Growing Up in Ireland survey from 2023 and the decision to task the CSO with conducting a new sexual violence survey, a modern successor to the 2002 SAVI report.

The complexity of the world we are trying to measure is also increasing with emerging issues such as globalisation and digitalisation, which result in complex economic transactions necessitating sophisticated analysis to decipher. We are experiencing transformational change in the range and complexity of the data sources available to describe our economy and our society. We are witnessing the use of statistics and data by a growing variety of stakeholders, sometimes with agenda-driven emphasis. We are experiencing growing levels of scrutiny of data to ascertain its reliability, accuracy and veracity. In this environment, the CSO is responding by maintaining a focus on its important heritage of impartiality and objectivity in the formulation of transparently produced, accurate and reliable official statistics, while exploring the boundaries of technology, data sources and statistical methods and techniques to persistently enhance the quality and insight afforded by our outputs. To do this, we are committed to building an innovative, agile and high-performing organisation to fully exploit the potential of the data and technological revolutions. Our aim is to deliver a broader range of high-quality information on societal and economic change for Ireland, which is demonstrably independent and without bias, and which can withstand scrutiny as regards data sources, statistical methodology and the impartiality of its compilers. In so doing, we can support the process of guiding Ireland’s future development for the benefit of all our citizens. That concludes my statement. I thank the members for their attention. My colleagues and I are happy to take any questions they have to offer.

I thank Mr. Dalton for that. The first speaker is Deputy Aylward, who has a maximum of 20 minutes. Each other speaker will have ten minutes. We will conclude this session in an hour's time because there is another session to be dealt with before lunch. I ask members to be as tight as possible in their contributions.

I welcome Mr. Dalton and his officials. I acknowledge the important role the CSO plays, as stated by Mr. Dalton, with respect to statistics, information and data. I note at national level CSO official statistics inform decision making across a range of areas, including construction, health, welfare, environment and the economy and at European level they provide an accurate picture of Ireland's economic and social performance and enable comparison between Ireland and other countries. The CSO is playing a very important role in that respect. I want to ask Mr. Dalton how we are doing economically in the context of Europe, how will we fare in the next few years and in the context of Brexit, the elephant in the room and what is on everyone's lips, what effect does he envisage it will have on our economy and on our future dealings with Europe?

Mr. Pádraig Dalton

The focus here is on the economic statistics. As the Deputy said, the most recent set of figures show that our economy on a GDP basis is growing at 6.7% but as we know the GDP figures have an impact. Globalisation impacts those figures. We have now produced a broader range of stations and GNI* is one of those where we net out the impact of globalisation and modify domestic demand, personal consumption and expenditure. When we consider those indicators it is clear that the domestic economy is growing in the order of about 4%, which is consistent with what we are seeing in the labour market.

In terms of the future, as the Deputy will be aware, the CSO does not forecast. We are a little bit boring. We advise what has happened and what the current situation is. Forecasts and predictions of the future rest with the research community and economists so that is not really our bailiwick.

The Deputy mentioned Brexit. Obviously, it is relevant to us, particularly around the collection of statistics and especially trade data. As I presume the Deputy will have heard from previous witnesses, if and when the UK leave the European Union the collection of trade data will move from what is called Intrastat to Extrastat but we are prepared for that. It will mean a change in the data source but we will still have data to compile the trade data. As for the impact of Brexit, I do not know of anybody who is able to accurately predict what the fallout will be.

People are saying it could result in a 3% to 4% reduction in our GDP and the loss of 70,000 to 80,000 jobs. If there is a hard Border and we fall over the cliff, so to speak, does Mr. Dalton believe that is a possibility? I am not asking him to predict but does he believe it is a possible it could have that serious effect on our economy?

Mr. Pádraig Dalton

The only thing I can say is that the CSO would not have a view on that because that is not our role. I am sure there are others who can give the Deputy opinions on that. It is not a role for the CSO to predict the future.

It is all guess work at the moment.

Mr. Pádraig Dalton

It is all guess work.

I note the CSO had a surplus of just under €3.94 million and a reduction in staff of 62. That is quite a number of personnel. Given the CSO's job of gathering statistics and material, 62 people is quite a number to short of, having regard to its surplus of just under €4 million.

Mr. Pádraig Dalton

As always, the devil is in the detail. The year 2016 was an atypical one because it was a census year. As the Deputy will be aware, for the census of population, we recruit a significant number of staff for a short period of time.

Throughout 2017, staff were redeployed out of the CSO. This is the normal cyclical change in the number of staff in the CSO. In the period up to a census year, we build up and in the year after the census, we begin to lose staff. If I talk about our core numbers, excluding census, the underlying direction of travel for staffing numbers in the CSO is in an upward direction. Part of the problem we have had, which I alluded to in my report, is that the type of skill sets for which we are looking in particular involve data analysts and those on the IT or technology front. As members will be aware, the labour market for data analysts is very competitive so it is difficult to get as many data analysts in-----

Is Mr. Dalton saying the graduates the CSO needs are not coming through?

Mr. Pádraig Dalton

The graduates are there but, unfortunately, we are competing with the private sector. There is a significant number of organisations in Ireland that are recruiting graduates on the data analytics front. Statisticians are essentially data analysts. There are also growing demands for data analysts across the public service so it is a far more crowded labour market than was the case previously. We are doing our best. A competition is under way with over 200 applicants. We will be running interviews in the next period of time. We will put a panel in place and try to bring in as many of those statisticians and data analysts as we can.

Does this shortage of staff have implications for CSO in terms of carrying out its work?

Mr. Pádraig Dalton

If it continued, it could be a problem but right now, it is not a problem. We are well resourced. The Deputy can see that we have the money. The problem is more about getting the staff. We have changed our approach to the recruitment of statisticians. We are running those competitions ourselves now, which has proved far more successful over the past couple of iterations. The technology side will also be difficult. The technology is almost more difficult than the data analytical side. We are experiencing more difficulties in getting the technological staff we need. As the Deputy is aware, the nature of the skill sets in technology changes very quickly as technology develops so that might be the area where we have some issues. I am quite confident on the data analytical side that we will get the staff.

I must move on. I have a quick question about the contingent liabilities. I know Mr. Dalton cannot speak about individual cases. How many cases relate to the contingent liabilities? What is the general nature of the pending legal proceedings?

Mr. Pádraig Dalton

There were five cases in 2017 and six in 2018.

Can Mr. Dalton explain what this litigation involved? I am not asking about individual cases.

Mr. Pádraig Dalton

Unfortunately, we are not in a position to disclose what those liabilities are.

I thought that might be the case.

The committee has heard a great deal about the PULSE system, including its capabilities and lack of capabilities. I note the CSO suspended publishing statistics from PULSE in 2014 and resumed publishing statistics from it in 2017. On its website, the CSO states that:

The CSO recognises that the deferral of these important statistics results in an information gap and is a source of frustration to users. The CSO has taken the decision to resume publication of Recorded Crime statistics under a new category "Under Reservation".

I want to ask Mr. Dalton about this category. The CSO suspended publication of PULSE statistics because of question marks about the system and then started publishing them again in 2017 under a category entitled "Under Reservation". Will Mr. Dalton explain what this category means?

Mr. Pádraig Dalton

Quite simply, under reservation means that we have concerns about the data source. The quarterly data we publish relates to recorded crime and the only source of recorded crime statistics is PULSE. The Deputy will be aware of the issues that have arisen with the underlying quality of some of the data on the PULSE system. Most recently, there were questions around homicide statistics but there have other areas as well. We suspended publication because we felt we could not stand over the statistics back in 2014. We suspended a second time in 2016. Even when we were publishing, we were adding heavy caveats so if one looked at the release even back at that time-----

Will Mr. Dalton identify the reason the CSO suspended publication of the statistics? What was the problem with PULSE?

Mr. Pádraig Dalton

If we go back to the inspectorate report in 2014, there was a broad range of issues. There were issues with classification of certain crimes. Some crimes were not classified in the right category. There were some cases where crimes were not captured on PULSE at all. There were some duplication issues. There was no clarity on the application of the crime recording rules. The crime recording rules effectively constitute a manual that provides absolute clarity. They are supposed to provide clarity on how any particular crime is supposed to be accounted for in a recorded crime system like PULSE. Obviously, they have these crime recording rules in other jurisdictions. We are working continuously with An Garda Síochána to try to improve the quality of its data for its purposes first of all because it needs good quality data from PULSE for its own reasons. Then we would become downstream beneficiaries. One of the things we are working on in particular involves the crime recording rules. We are trying to encourage them and we will work with An Garda Síochána to develop this crime recording rules document.

There are two other things we want An Garda Síochána to do. We want it to introduce a formal quality management framework. The quality division in the CSO has met An Garda Síochána to explain what the quality management framework would be like and to outline how it might go about improving the quality of the data on PULSE over time. Finally, we really feel it is important that An Garda Síochána appoints a single person at a very senior level to take point on data-related issues. At the minute, An Garda Síochána has an interim chief data officer. I think the intention is make that a permanent post. This is incredibly important for us but it is important that this role be at a very senior level in An Garda Síochána so that the person can influence the development and the move towards improving crime data.

Under reservation is a term used by EUROSTAT for government finance statistics. If EUROSTAT has a concern about a country's government finance statistics, it can put this reservation on the data. What it effectively means is that there are concerns about the data and EUROSTAT cannot fully stand over the data. We were left in this situation where there was a vacuum. Our concern is that vacuums normally get filled with anecdotal information or poor quality or subjective data. The longer a vacuum is left, the greater the risk that this occurs so we took the decision to resume publication but we needed to be clear with users that there was an issue with the data. That is why we put in the classification "Under Reservation".

What we have also done to help users is produce three quality reports on the data coming from PULSE. We try to give a clear signal to users that the figures on a certain classification coming from PULSE may well be an understatement, for example, burglaries. We give a clear statement in the quality report as to our assessment of the accuracy of the particular statistic on whichever part or type of crime it is. We could have left a vacuum and said we were not going to publish but we felt transparency was important. It is the best data that can be published right now.

What is the percentage of error? It looks like there are deficiencies in the entire system and that PULSE is not up to scratch. We have been talking about this since 2014. Should a new system be put in place? An Garda Síochána and PULSE are the most important elements for law in this country. Is PULSE outdated? There are deficiencies if people are looking at data coming from that system that are not up to scratch. We have already seen how major mistakes were made with breathalyser tests. Have all these been rectified?

Mr. Pádraig Dalton

There is a broad range of issues. There is no silver bullet. There is no one solution. Sometimes we jump to the conclusion that the problem is the technology and the system. I do not necessarily think that the problem lies in the system. It is about how the system is used and how the data are being captured within the system so that is very often-----

Is Mr. Dalton talking about the personnel who are feeding the system?

Mr. Pádraig Dalton

Obviously, data must be inputted into PULSE and that is centralised in Castlebar.

Mr. Richard McMahon

It is in the Garda information and security centre in Castlebar. I will elaborate on some of the problems with the PULSE system. There are probably four main types of problems. The first relates to the difference between the number of crimes that are reported and, subsequently, the percentage of those crimes that are actually recorded on PULSE.

Another problem was with the completeness of the recording of the details of crime, for example, the relationship between the victim and the offender, which is important information in terms of crimes like domestic abuse, and the motive of the crime, which is important information in terms of hate and discrimination crimes. The timeliness of the recording of crimes was also a problem.

Our quality reviews have undertaken in-depth studies of these quality characteristics. It is important to note that we have seen some improvement between the first and third quality reports and that the PULSE system is being adapted by the Garda to address some of the issues raised in the quality reports. However, significant gaps still exist. Until those gaps and problems are-----

Is the CSO identifying the gaps and telling the Garda where it has found deficiencies and problems? Are those problems then being rectified?

Mr. Richard McMahon

We are identifying them. We publish a detailed quality report where we indicate the data sources that we have examined and the issues that have been identified and make recommendations to address the problems. Those reports are published. We have close co-operation with the Garda and run through the reports with it to outline the difficulties and our proposals on improving the quality of the statistics.

How does the PULSE system compare with the system in London, Germany or the US? It is a broad question, but we are a small country and should be able to have a system in place that is accurate and fool proof, so to speak.

Mr. Richard McMahon

It is difficult to compare different police systems internationally. It is fair to say that similar problems exist in other jurisdictions.

Mr. Pádraig Dalton

It is not necessarily about the system. Rather, the issue is what is going into the system. This relates to one of our key recommendations, namely, the importance of creating the criminal recording rules document. If there was clarity in document form about how different crimes should be recorded and that document was made available to everyone who had to input the data, one would introduce a systematic way of consistently capturing types of information. The problem is that there is not really clarity at the moment and no document is in use. This is why we have prioritised this document as one of the three actions that we want to push with the Garda. We have been clear. We are working with the Garda and are being open and transparent with it about what we view as the quality issues. We have also been open and clear with it about what we see as the solutions. Of course, we are not criminologists. Our focus is on trying to help the Garda improve its data. The key point is that it needs to improve the quality of its PULSE data for its own purposes first. If it does so, then that will have a benefit for the Garda and the recording of crime statistics.

Am I hearing this right? Is Mr. Dalton saying that the system is only as good as what is being put into it? If so, is the problem with the personnel who are feeding in the information, in that they are not up to scratch professionally? I do not mean to cast aspersions, but is the computer system not working because information is not being presented and fed into it properly?

Mr. Pádraig Dalton

I would not criticise any individual.

I did not mean any person.

Mr. Pádraig Dalton

I understand. It has to do with clarity around the rules. For example, when the CSO compiles statistics, we must use manuals and clear guidelines, which come from EUROSTAT. Every member state must compile statistics in the same way. My point in this instance is that there is not enough clarity around the crime recording rules. Therefore, it becomes difficult for members of the Garda who are trying to put data into the PULSE system to do so in a consistent way across the regions.

I do not understand. I am from Kilkenny. Waterford is next door. Is data fed into the central system in Dublin from every station in the country? Is that how it works? If someone is caught speeding tonight and is breathalysed, does the individual garda put that information into the system or is it done through a centre in, for example, Kilkenny?

Mr. Pádraig Dalton

No. I am not an expert in this but, as I understand it, the data are put into the PULSE system centrally from Castlebar. If a garda has a recorded crime that he or she wants to put into the system, my understanding is that the garda provides that information to his or her-----

Must the individual garda on duty do it?

Mr. Pádraig Dalton

The garda would fill out the relevant documentation, which I presume would go through the appropriate line management or supervision in the local station or region. It would then go to the Garda Information Services Centre, GISC, in Castlebar.

Mr. Richard McMahon

The general procedure is for the garda to phone in the details to the GISC in Castlebar.

Mr. Richard McMahon

Those data are then inputted into the PULSE system.

Are some areas better at feeding information in than others? Do we have that statistic? Are there problems in some areas and none in others? I am just trying to see how the system works.

Mr. Richard McMahon

As far as I am aware, there is no specific regional variation.

Mr. Pádraig Dalton

It would be helped by clarity on the crime recording rules. That is the key focus. If we can achieve clarity in the crime recording rules among all gardaí, we will see an improvement in the consistency of the recording of data within the PULSE system.

In speaking generally with gardaí, I have been told that, if someone is caught for speeding tonight, that information will be in the system within one or two hours. I believed the garda fed it in by computer rather than phone, but maybe I was wrong.

Mr. Seamus McCarthy

From other work that we have done, there is some automatic uploading at local level.

I was thinking that. Rather than phoning-----

Mr. Seamus McCarthy

Hand-held instruments are used for certain functions. For more serious crimes, a different range of information may need capturing. The process might be relatively straightforward with an incidence of speeding. For example, one would use the car registration and so on to capture information.

Next are Deputies Connolly and Murphy. These are ten-minute slots.

I welcome the witnesses. It is good to see gender representation. I quoted the CSO's opening statement yesterday: "good policies start with good data" and "Good data is the lifeblood of decision making and the basis for accountability." The CSO is the most important organisation or body to appear before us. In a time of post-truths and fake news, we must be able to rely on statistical information. An intelligent Government should form policy using statistical information. I will base my questions on that idea, so I will ask the witnesses to bear with me. The witnesses also stated that they aimed to turn statistics into information and knowledge for all. That is very good.

The CSO is before us with its accounts and no issues have been raised in that regard, so I will zone in on the risk register. If we cannot believe statistics, we are in serious trouble. I will revert to the Garda issue in a minute. Regarding the CSO's own risk register, what risks has it identified to assure the public that the quality of its work is further above par than any other organisation?

Mr. Pádraig Dalton

The vast majority of the risks in the CSO are statistical in nature. Of our budget, 78% is attributable to wages and salaries. We do not have a programme budget, only an administrative one. We put the risks across three fronts - financial, statistical and technological. The types of risk we have identified relate to ensuring that we have a budget that supports our corporate objective of providing data for decision makers and for accountability. We always concentrate on ensuring that we have the appropriate funding in place.

Mr. Dalton mentioned three areas - financial, statistical and technological - but what particular risks has the CSO identified?

Mr. Pádraig Dalton

Cybersecurity is certainly one risk about which we are concerned. Like any organisation, we are a very-----

What other risks? I am being conscious to focus my questions.

Mr. Pádraig Dalton

We want to ensure compliance with the quality management framework that we have put in place.

What does that mean?

Mr. Pádraig Dalton

The quality management framework is a formal quality process with which all statistical areas in the CSO must comply.

Who does the formal quality aspect?

Mr. Pádraig Dalton

Every statistical area has to have a suite of formal checks and balances in place to ensure data quality.

Mr. Pádraig Dalton

The first thing, on the economics statistics side, is that EUROSTAT and the European Central Bank, ECB, visit the CSO. We are audited on that side. Every statistic we compile and send to EUROSTAT has to be verified by it independently.

What happens on the financial side?

Mr. Pádraig Dalton

On the financial side, we have our own finance section and that is independently reviewed by the Comptroller and Auditor General.

There is a management board over the CSO. Is that correct? I saw that mentioned in some of the briefing materials.

Mr. Pádraig Dalton

Yes, that is correct.

What is the management board and who comprises it?

Mr. Pádraig Dalton

The management board is made up of the director general-----

That is Mr. Dalton.

Mr. Pádraig Dalton

That is correct. I and my colleagues here are all members of the management board. There are three other members. We have the chief information officer, the director of business statistics and the assistant director general with responsibility for statistical system co-ordination.

Is there any independent or peer review, or whatever the appropriate term is, of what the CSO is producing?

Mr. Pádraig Dalton


What is that procedure?

Mr. Pádraig Dalton

There are a number of them. We are peer reviewed by EUROSTAT and we have what are called Excessive Deficit Procedure, EDP, visits and gross national income, GNI, audits. They are independently carried out. We also have a review against the European statistics code of practice peer review. That is normally undertaken by a three person team. The last time it was comprised the former heads of statistics from Estonia, Spain and Latvia. All three were independent and all of those reports are published. We also have-----

Are the minutes of that management board published?

Mr. Pádraig Dalton

They are all published internally.

That is good. I just wanted to clarify that.

Mr. Pádraig Dalton

Yes, they are all published internally.

In his opening statement, Mr. Dalton spoke about staff and recruitment. It is an important issue. What is the total number of staff in the CSO?

Mr. Pádraig Dalton

We had about 736 people on a full-time equivalent basis in 2017. We have many more staff now, about 860 and climbing.

I am sorry, what did Mr. Dalton say?

Mr. Pádraig Dalton

We have about 860 staff and climbing.

Does the CSO have a full complement of staff to carry out its work?

Mr. Pádraig Dalton

We do at present, although we are carrying vacancies within the organisation.

How many vacancies are there?

Mr. Pádraig Dalton

We have vacancies for about 30 staff. Our vacancies go beyond the CSO, of course. We are now providing services to broader systems by seconding professional statisticians to Government Departments and we have a number of vacancies-----

Just give me a minute because I lose my train of thought. There are 860 staff in the CSO and 30 vacancies.

Mr. Pádraig Dalton

Yes, approximately 30.

Mr. Dalton stated there is some complexity with that total. Regarding the vacancies, how long have those posts been vacant?

Mr. Pádraig Dalton

The vacancies vary in duration. None of them is much more than 12 months. Some of them are, however, now at that point and many of those are in the technology side where we are having difficulties recruiting.

I understand. The CSO is the most important body to come before us. We have had body after body before us giving us reassurances. I am only trying to establish the facts.

Mr. Pádraig Dalton

Of course.

There are 30 vacancies and some of those have remained open for longer than a year.

Ms Maria Hurley

As a rapidly evolving organisation, we are continually reviewing our staffing profile-----

I understand that. How long have the 30 vacancies been unfilled? What is the longest that any of one those vacancies has remained unfilled?

Ms Maria Hurley

May I distinguish between-----

Ms Hurley certainly can.

Ms Maria Hurley

The management board might decide it wishes to fill a newly created post. In that context, we have one vacancy where, in principle, a decision has been taken by the management board to fill the post but that post had not previously been filled. It is a post we wish to fill but it is not a currently vacant post. Regarding posts actually vacant, where there was an existing person in situ, that is constantly evolving. We would not have any posts in that category for more than 12 or 15 months.

Now, 12 and 15 months is a long time. How many positions have been vacant between 12 and 15 months?

Ms Maria Hurley

It is a very small number. I do have the exact number figures but it is in single digits.

Perhaps we could get a note on that later through the Chair.

That is fine.

How are those vacancies affecting the ability of the CSO to deliver its service? What are the risks in having those vacancies? What measures are being taken to fill those vacancies and what procedures are being used?

Mr. Pádraig Dalton

We have a growing demand for data and those vacancies will limit our ability in that area. It is not limiting our ability too much at the moment because many of these posts are roles designed to help us meet new demands such as the Growing Up in Ireland survey and the sexual violence survey. We are also trying to continually exploit administrative data, in other words data held by Government bodies, and we are also seeking to exploit big data. These vacancies are slowing down the speed at which we can progress some of these corporate issues. We have a broad range of recruitment competitions under way in the CSO across all grades. We have been filling posts right since the end of last year. We have a figure in mind and we are trying to fill 75 posts between now and the end of this year in a broad range of grades and skill sets.

I ask Mr. Dalton to hold on for a second. The CSO wants to fill 75 posts.

Mr. Pádraig Dalton

We want to fill an additional 75 posts.

Are those 75 posts additional to the 30 posts that are vacant?

Mr. Pádraig Dalton

No, we are including those 30 posts.

The CSO, therefore, wants to fill 75 posts, including the 30 that are vacant, by the end of the year.

Mr. Pádraig Dalton

Yes, by the end of the year. That is the intention and it is a big ask.

That is fine. Let me go back to what was stated about the CSO saving money on consultants. I am always delighted when we save money on consultants, bearing in mind that we have the consultants for the national children's hospital beside us here. The CSO saved money and then it went back and carried out an employee engagement survey in 2017. I like to see that. I often think it is possible to find much more money if people on the ground are listened to. I state that given my 17 years at local level. We could have avoided many problems if that had been done. What feedback has come from staff within the CSO and been listened to and heard to improve the organisation's service? What help is given to staff to upskill given the CSO has vacancies? What has been learned in general from that engagement with employees? Please do not tell me what has been done with all departments. I am just interested in the witnesses' department.

Mr. Pádraig Dalton

That is fine. We have had strong figures on engagement. It was more than 70%, which is well above the Civil Service average.

Is that in the witnesses' own department?

Mr. Pádraig Dalton

Yes, I am not talking about anybody else.

There was a 70% engagement rate among the CSO's own staff.

Mr. Pádraig Dalton

Yes, that is correct. The vast majority of the CSO's figures are above the Civil Service average. Those that are below, and I think that is what Deputy Connolly is interested in, include one on involvement climate, the second is-----

What was the first one?

Mr. Pádraig Dalton

Involvement climate. The second aspect is innovation.

I am sorry, I heard "climate change". What does Mr. Dalton mean by "involvement climate"?

Mr. Pádraig Dalton

It refers to the extent to which staff feel they have a say in developments locally.

What percentage of staff feel involved?

Mr. Pádraig Dalton

I do not have the figure off the top of my head, but I think it was roughly 51%. A score of 50 means that the same number of staff stated they were happy with the level of involvement as stated they were not happy. It is an index. The figure was about 52%

I am being a bit unfair because we are under pressure, but I really would like information on the last time there was engagement. Staff matters have nothing to do with me, clearly, but what has to do with us is accountability, governance and learning all of the time. In that context, when was the last staff engagement, what issues emerged and what was the involvement of staff rate in that? Before the Chair stops me, I also want to ask about the Sexual Abuse and Violence in Ireland, SAVI, survey on sexual violence.

Mr. Pádraig Dalton

The first thing is we have an internal-----

If Mr. Dalton will give me just one second. I am very familiar with this. I know the CSO has got it, so I just want an outline of what it is now doing. I ask because I saw a time span for the completion of this research that concerned me. It is way in the future. When is it starting, when is it going to be completed and does the CSO have enough staff for SAVI?

Mr. Pádraig Dalton

We have started the work on the SAVI survey. We expect to be in the field next year with a pilot. International best practice is to conduct a pilot, so our intention is to be in the field with a pilot in the summer months of 2020, ideally. Our hope then is to be in field with the live survey in the summer of 2021. We do not want to undertake a survey of this nature in the dark winter months. That is also the guidance internationally. We are now engaging with a broad range of users. Our staff were in the UK last week meeting expects including Professor Liz Kelly and Professor Sylvia Walton. We are trying to set up a meeting with Professor Hannah McGee to learn from the experience of the first SAVI report back in 2002.

We also had engagement with the broader range of stakeholders to outline what we are planning to do, and we will be setting up a stakeholder group to start that engagement process towards the design of the questionnaire, the design of the samples and so on. The international research work is under way, and we will moving throughout this year into the other phases of the project.

I will come back to it. My time has run out.

I picked out exactly the same sentence about planning and good governance and the point that policies start with good data. We have never had a greater body of data available to us and the ability to manipulate it because, obviously, we have IT systems that allow us to do that. At the same, we have bigger traffic jams, a shortage of school places, schools in the wrong places at the wrong time and a housing mismatch. If it does anything, this highlights that we are not using some of material we should be using in an effective way to deal with these issues. However, the information is there. I cannot resist saying that because it drives me nuts.

In regard to the PULSE system, the old adage "rubbish in, rubbish out" comes to mind, in that, if the material is not inputted and catalogued correctly, it will not give accurate results. At the moment, how confident is the CSO in the Garda statistics? Are there flaws in the PULSE system in addition to the personnel input?

Mr. Pádraig Dalton

Yes, there are still issues with the statistics we are compiling from the PULSE system, which means there are problems with the data that reside within the PULSE system. The answer is "Yes", which is why statistics are published under reservation.

The witnesses told us a senior appointment would be needed to get to grips with that. Is there a timeline that the Garda and the CSO are working to so those statistics can be relied on fully?

Mr. Pádraig Dalton

The issue of the CSO lifting the reservation will be down to the speed at which the Garda can address the issue within the PULSE system. The CSO cannot predict how long it will take to lift the reservation. Equally, the chief data officer of the CSO would have no role as that is an internal matter for the Garda.

It is a very serious matter when the CSO cannot stand over the crime statistics. Mr. Dalton told us it came to light by virtue of some of the things that ended up in the public domain. Would Mr. Dalton have noticed there was a mismatch in the statistics in any case?

Mr. Pádraig Dalton

It really came from the inspectorate report back in 2014. To be honest, it would have been impossible for the CSO to identify the issues the inspectorate report came up with because we would not have had access to the data required to get into that level of detail, and we would never have been in a position to get the access to the information that was required to unearth those issues. We do now, however, because, subsequently, following on from all of the engagement with the Garda Síochána, gardaí now make all information available to us to assess the quality of Garda data accurately.

We are signed up to EU Regulation No. 223/2009, and the statistics are Europe-wide statistics. How many other similar bodies to the CSO would have a problem with something as significant as their crime statistics? Are we outliers?

Mr. Pádraig Dalton

No, there are issues with recorded crime statistics in a number of countries throughout Europe. It is a recognised issue. It can be quite difficult to get the necessary information on the recorded crime statistics from what is a system that is designed for administrative purposes rather than for statistical purposes. We are not necessarily an outlier. It is definitely an issue in the UK as well, as far as I am aware.

Mr. Richard McMahon

The crime statistics in the UK were downgraded from official statistics to unofficial statistics, basically, arising from the quality concerns around the data there.

I want to address two other issues. One is the consumer price index, for which specific categories of things are counted. Is the cost of accommodation counted in the consumer price index?

Ms Jennifer Banim

Rent is included.

What other range is there in that basket?

Mr. Pádraig Dalton

The basket of goods for the consumer price index is determined by the household budget survey. I can give this information to the Deputy as the basket of goods that is measured in the consumer price index is publicly available. As it is quite a long list, I can send on to the Deputy a link to the basket of goods so she can see what is in it.

The census of population is produced every five years, although there have been couple of small variations in that. One of those variations was 100 years ago because we had a civil war that did not allow the census to be taken that particular year. Therefore, there will be a gap before the next release of 100-year data, which is to be in 2026. What kind of lead-in will the CSO need? Does the CSO require additional personnel to have the census in a format that can be handed to a body like the National Archives so it can put that online or publicly release it?

Mr. Pádraig Dalton

The National Archives already has the forms from 1926. We are currently working with the National Archives on a project to protect all the census forms, not just those from 1926 onwards. There is a project starting that will look at the whole issue of digitising the census forms from 1926 to make sure they are available once the 100 years has elapsed in 2026. I am not familiar with the costs because it will not be a cost incurred by the CSO and it will be an issue for the National Archives to digitise and publish those data.

Does Mr. Dalton know if that work has started?

Mr. Pádraig Dalton

I think it is under way. We have work under way with the National Archives on the 1951 to 1986 censuses, so there is a collaboration under way.

Mr. Richard McMahon

On the 1926 project, work has got under way in regard to assessing the volume of work that will be involved. There is a pilot project which is looking at a number of batches to see what will be involved in retrieving forms and putting them into a format where they can be digitised.

It will not be released prior to 2026.

Mr. Pádraig Dalton

The legislation states it is a 100-year rule.

That is very hard.

The CSO's annual report from 2017 contains a very big appendix with all the timelines of the different reports. One part refers to "System Health Accounts". If the witnesses do not have the answer now, they can send it on to the committee. There is a timescale of a year and a half, or 78 weeks. What are those statistics and why does it take a year and a half to get them? Surely the CSO should be more worried about that.

Mr. Pádraig Dalton

I will let my colleague answer that. Primarily, they are exclusively administrative data sources, so these would be data that are made available from the administrative system. Very often, when compiling data from administrative data sources, we cannot publish those data until they have actually been finalised. Ms Banim has responsibility for the system health accounts and she can give more detail.

Ms Jennifer Banim

It will not take long. They are very important statistics and it is relatively new for the CSO to be publishing them in the format they are currently going out in, with additional detail. It is about what is spent on health, who is spending it and where it is being spent. They are a very valuable set of tables for health policymakers and decision-makers in that space. As Mr. Dalton said, they are produced from a blend of data, primarily data from the HSE.

We also do some collection from private health providers and we use other administrative data for estimating out-of-pocket expenditure on health. We are trying to really show the broad expenditure on health. Timeliness is an issue. We have worked on it and we have tried to improve it. We have brought it back and it is a focus for improving that timeliness even more so.

In the CSO report published last May, the figures were for 2014. We are a long way back on-----

Ms Jennifer Banim

We have more recent data since then.

I ask for an up to date note on when the 2015, 2016, 2017 and 2018 figures-----

Ms Jennifer Banim

I think they are out. I think maybe we published a back series. We got back-data because it was a new product from CSO. We got back-data from the HSE and we were able to do retrospection, but we do have more recent data than that.

I have a non-financial question on county income and GDP. I regularly see on a regional basis and sometimes on a county basis, average income per household. It outlines how well off a county is and how far below Dublin we all are in annual income. They always seem to be quite a few years old when we get them. Can important statistics like that not be more up to- date? I think they relate to annual household income. I do not see it exactly here; it may be the one that was referred to.

Ms Jennifer Banim

I know the issue the Chairman is referring to. It is county incomes and value added across the counties. The figures are benchmarked against our regular annual production cycle. In June 2019 we will be producing definitive figures for 2018 and then we work on from that on things like the county incomes and the value added. What we have done in the last iterations of those county incomes, in acknowledgement of the fact that they are important to people to have up-to-date data, is we have put in provision for very preliminary estimates for the latest year, if I can put it that way. So when we are going out we put in the latest year as well in there. Just as the Chairman said, people need more timely data on it. For the annual data it can be difficult to get the sources, but we can always do something on partial sources and we do that where we can.

I ask Ms Banim to send us a note on the last set of-----

Ms Jennifer Banim

We will send you the latest.

----- county and regional income figures.

Ms Jennifer Banim

I think the publication is planned around now. We can send the committee a link to that.

I ask for it to show where we are in the more recent estimates even if they are not finalised. Many counties would like to know where they stand on that figure.

In his opening statement Mr. Dalton mentioned telephone interviews. Are they calls to landlines or mobile phones? If they are calls to mobile phones, the CSO could be talking to somebody on holidays in the Canaries. How does it regionalise that? What quality control does it have?

Mr. Pádraig Dalton

That is in relation to the labour force survey. If I was included in the labour force survey, I am included for five consecutive quarters. In the first quarter an actual CSO employee, one of our field force who do a fantastic job for us, calls to my house. They do the first interview, what we call a wave 1 interview. That wave 1 interview is done face-to-face.

Therefore it starts with a physical meeting.

Mr. Pádraig Dalton

The subsequent interviews are done through computer-assisted telephone interviewing.

Therefore the CSO knows whom it is interviewing and where they are located.

Mr. Pádraig Dalton


That is my question.

I ask for a detailed note in respect of the State balance sheet, which is often referred to here. I want to know where the universities and institutes of technologies are on the State balance sheet. Time is tight now. We would prefer a proper note on that.

The issue of the housing agencies came up here recently. We got some reports, some of the big ones, around the State balance sheet. I ask for a detailed note on that.

The CSO was involved in the application to have Irish Water put on the State balance sheet, which failed abysmally at EUROSTAT level. How did all the brains in the CSO get that so obviously wrong? EUROSTAT rejected it; how did the CSO get it so wrong? It should not have. We have had this before. The CSO should know the rules of what is on and off the balance sheet.

Mr. Pádraig Dalton

I will let Ms Banim get into-----

If we are out of time, I ask for the CSO to send us a note.

Mr. Pádraig Dalton

Of course.

It did not help in how the CSO was seen. I have noted that the CSO frequently uses the word "independent". It looked like a pawn of the Government in that. It was making a case that did not make sense to people because most of Irish Water's income was coming from public sources. The bid did not succeed. I know the Department of Finance was probably very supportive of it. Before this committee its officials would have said the CSO had great experts who knew all about EUROSTAT as part of putting the proposal together.

Mr. Pádraig Dalton

It was the first major classification that was done under the new ESA 2010 rules. It is the role of the national statistical institute to do the preliminary classification. As always EUROSTAT has the final say. Our assessment of it was based on the business case that was there because all that was available at the time was the business case. Every argument we made we were able to link back to the ESA 2010 regulation and also the Manual on Government Debt and Deficit, MGDD, statistics. Every part of the assessment, every criterion, we were able to point to the reason we assigned certain things to being on or off balance sheet.

EUROSTAT in its response, perhaps, was not able to articulate its rationale in the same way in that it was not able to link it exactly to the various elements of ESA 2010. In its response, it said that it has to take an holistic approach without defining what that holistic approach was. There were issues of interpretation around ESA 2010 because it was the first time it was being implemented and we took the view-----

Is Mr. Dalton saying EUROSTAT got it wrong?

Mr. Pádraig Dalton

I will let Ms Banim answer.

Ms Jennifer Banim

As Mr. Dalton has already said and as EUROSTAT acknowledged, it was an extremely difficult classification decision. We had a process for it and we went through it. We can send it on because all the documentation and all our thinking are publicly available on the Internet. Some of the issues that were borderline or on-balance issues on Irish Water are still being thrashed out at the different technical groups in EUROSTAT. There are a number of states that agree with our interpretation; EUROSTAT has its interpretation and other member states have different interpretations again. What I would say is that some of the fundamentals that classification decision raised still rumble on. I would still say that on the process, on our interpretation of the process, what we did we did absolutely by the book and by the rules. That is the way we have to do it.

Ms Banim is essentially saying the CSO based its element of the process on the business case that came from the line Department. It was making a business case that it wanted it to succeed. It looks like the CSO took what it was saying as-----

Ms Jennifer Banim

We would have to take the information that was provided from it.

It is always back to that.

Ms Jennifer Banim

We questioned that and we put tolerances around it. If one is constructing a business case that one is going to be building out and if it is going to be very close to the margins, if I can call it, around the rules around these decisions, the likelihood is that a prudent decision would mean that the classification would be on balance sheet. It is very difficult in these, what are called, ex ante classifications.

The CSO will send us a note on it.

Ms Jennifer Banim

We have a lot of information on it.

Given that domestic water charges are now gone, I do not understand why somebody is trying to make a case at this stage.

Ms Jennifer Banim

It is not so much the case on that particular classification; it is around making sure that the guidelines and the standards are crystal clear so that we will never be in that situation again and no other member state will be in that situation again.

Does Ms Banim mean that if domestic water charges come back, the CSO will get it right the next time?

Ms Jennifer Banim

It is for all classifications.

Mr. Pádraig Dalton

Ms Banim is saying that whole process raised issues around the interpretation of ESA 2010. Those interpretations do not just apply for Irish Water; they apply for all Government classification decisions. Therefore there are technical issues outstanding that still require clarification.

The CSO will send us a note.

We have six minutes left in this session and I call Deputy Connolly.

Other people have asked about the criteria for the Garda. It is under reservation and it is a very serious matter. We have heard the Policing Authority report today. In that context, when will the reservation be lifted? What are the criteria for lifting the reservation?

I presume the CSO did not seek to do the SAVI report. Following pressure the Government took a policy decision to do it. What expertise does the CSO have in that regard? I am concerned about the date. I realise it has to be done right. I read it at the time.

A feature that jumped out besides the prevalence of the violence was the therapeutic nature of the way the research was carried out. Follow-up research was done and, because of the sensitive way it was carried out, it was found to be extremely therapeutic. Men and women reported stuff they had never reported before on a telephone. Could Mr. Dalton refer to that type of qualitative research and expertise?

I have a practical question on the islands. I was reading something for tomorrow on the Aran Islands. Many of the CSO's figures were very helpful but I understand it does not give a statistical breakdown island by island. Is that the case?

My last question is on staff, the importance of the CSO, and risks. Having sat here for two and a half years, I believe we learn most from staff if we listen to them and if what is required is in place. On the last occasion Mr. Dalton spoke to the staff - I refer to the questionnaire - what issues arose and how is he dealing with them? How is he dealing with recruitment issues and in-service building up of skills when there are difficulties outside? Those are questions but I would like the witnesses to start with the easy ones, on the criteria and the islands.

Mr. Pádraig Dalton

On crime, I cannot give an answer as to when the reservation will be lifted. It depends on how quickly the Garda-----

What are the criteria?

Mr. Pádraig Dalton

The criteria have been set out. If the Deputy looks at the quality report, she will see the criteria we use. It identifies the issues we are looking at from a quality perspective.

So they are set out in the report.

Mr. Pádraig Dalton

They are set out.

If the Garda does not meet the standard, then this reservation-----

Mr. Pádraig Dalton


So, theoretically, it could be there for a long time.

Mr. Pádraig Dalton

Theoretically, it could be. I want to be clear with the Deputy that it will not be lifted next year or, in all likelihood, the year after. A number of years' work is required to address the issue.

Has the pathway been set out clearly?

Mr. Pádraig Dalton

There is a pathway. We have an action plan in place with the Garda.

What about the islands?

Mr. Pádraig Dalton

On the SVS, we have a lot of experience with household surveys, of course, but this is a very different survey, as the Deputy has rightly pointed out. We are going to engage with international experts and all the relevant stakeholders, including the NGOs, because we understand the really important role they have to play in helping this survey to be successful.

At this point, I just want to know whether the CSO will have the right staff. I want a period of time.

Mr. Pádraig Dalton

Yes. We will have the right staff. We are going to be in the field on the pilot next year and we expect to be live in the field around summer 2021.

When will the report be completed?

Mr. Pádraig Dalton

The report will be completed within 12 months of that but hopefully much faster.

What about the last question, on the islands?

Mr. Pádraig Dalton

On the islands, let me refer to the ability to give a breakdown. If one is producing data from a sample survey, one cannot drill down. As the Deputy knows from the labour force survey, we cannot produce statistics below regional level because, as one drills down through a sample, the coefficient of variation or sampling error becomes a factor. In other words, the quality deteriorates. One cannot always produce data for small areas. What we are doing about this is increasingly using administrative data to allow us to produce more refined statistical outputs.

Secondary sources.

Mr. Pádraig Dalton

Yes, secondary sources. Secondary sources are basically data held by Departments, Civil Service bodies, local authorities and public service bodies. We have a legal right of access to those data. We can integrate them and provide new statistical analysis. One sees a lot of that. Almost 55% of all CSO outputs now incorporate what we call blended administrative data. It involves administrative data linked with other administrative data, or administrative data linked with statistical data. We are continually moving in that direction.

Does Ms Hurley want to say something about staffing and risk?

Ms Maria Hurley

On the first topic, to offer the Deputy some reassurance regarding vacancies we will provide her with a detailed breakdown. In broad terms, it is important to understand that more than half of our medium-term vacancies relate to specialist technology posts. One of the greatest difficulties we have had in this regard concerns the lack of availability of panels of staff ready to be appointed and the necessity, in some cases, to run tailored competitions.

On the statistical side, we run statistical competitions every year. It happened last year that our full panel of statisticians was exhausted by the autumn. Therefore, it is only vacancies that have crystallised on the statistical side since the autumn that continue into to this year. Added to that, it is important to note that, because we supply statisticians for secondment to the wider statistical system, we have a constant catalogue of requests coming in for statisticians to send on secondment. That has also created backlogs.

Turning to the other dimension, concerning risk management and the extent to which we engage with staff, whether it be through the Civil Service employee engagement survey, CSEES, or other channels, we have already published our action plan further to the CSEES survey. The action plan was developed in consultation with colleagues from across the organisation. There are now active working groups considering what actions are required to give effect to the objectives set out in the action plan. A key channel in the delivery of those actions is our engagement and innovation board, which is a long-standing structure in the CSO. It is a cross-staffing grade structure specifically targeted at affording a discussion forum for matters affecting staff in all areas of the business, both statistical and administrative.

The engagement and innovation board is just one of the forums we use to engage with staff. We have grade-based forums, a senior management group forum, a general management form, and grade-based forms for all the administrative grades. At present, we are in the process of completing a significant round of consultations in the preparation of the CSO people strategy to take us forward for the next several years in respect of the supports we are putting in place to develop our staff in terms of their technical skills and aptitude to deliver the supports required for the business and also to enable them to fulfil their full potential. I will follow up in writing with further detail, as the Deputy requested. I hope, however, my answer gives some illustration of the extensive variety of consultations and forums that are available to enable the flow of information in both directions across the entire organisation.

I thank all the witnesses. They may send the information directly to the committee. We appreciate that.

This was a short session, but on a clear, concise area. I thank Mr. Dalton and his three colleagues for the information provided at today's meeting and the information we will receive in writing from them in due course.

The witnesses withdrew.
Sitting suspended at 11.45 a.m. and resumed at 11.50 p.m.