Dear Lady Armstrong

‘Levelling up’ and public services inquiry

Thank you for giving the Office for Statistics Regulation (OSR) the opportunity to comment on the following questions:

  • Are there gaps in regional/local data sets (particularly for health, education, justice and wellbeing outcomes) which may impact upon the Government’s ability to target investment, and for Government success to be measured?
  • How can we ensure transparency and access to the data sets that the Government is using, to ensure Parliament can scrutinise decisions as to where money is prioritised, and the effectiveness of such investments?

OSR’s Strategic Business Plan sets out our vision and priorities for 2020-2025 and how we will contribute to fostering the UK Statistics Authority’s ambitions for the statistics system, as set out in the Authority Strategy. One of OSR’s ambitions is that, by 2025, the statistical system will provide much a richer picture of the UK’s changing economy and society. Statistics should not simply focus on the average, but instead provide disaggregated and granular insight into how different communities, places and people are doing. As regulator of government statistics, we will assess whether statistics provide regional, local and disaggregated pictures of society. We will challenge producers to produce more granular statistics, about ethnicity for example, and to respond to changing aspects of social identity and economic activity. Our systemic reviews will uncover areas where the needs of a wide range of users are not being met, and challenge producers to address these needs and gaps. Where producers do not address these issues, we will continue to highlight our concerns and criticism publicly.

There are a number of gaps in data – both in terms of personal characteristics and low levels of geography – which our previous work has highlighted; we also consider a lack of timeliness or comparability across the UK to be characteristic of data gaps. I wrote a blog on data gaps to initiate some work OSR is doing on the process of demystifying data gaps, building on our growing understanding of how different statistical producers have addressed them. We are still building up our case examples of what works to address gaps, but one striking early conclusion is the importance of collaboration between different producers. And, over recent years, we are pleased that departments are innovating in data collection or processing methods to fill gaps.

This letter addresses your questions about health, education, justice and wellbeing data in more detail, but gaps in data are a widespread problem. During the past year, for example, we have: reviewed statistics and found data gaps in areas such as the Office for National Statistics’ (ONS) internet access and use statistics; seen that the ONS has the potential to do more analysis on alternative methods with the microdata provided by the Valuation Office Agency for the Consumer Price Inflation, including Owner-Occupied Housing Costs (CPIH); and despite positive changes, noted that data gaps remain in Housing and Planning statistics.

Data gaps in public health, health care and social care

As our report on Adult Social Care statistics noted, “While there is rightly a focus on delivery, a scarcity of funding has led to under investment in data and analysis, making it harder for individuals and organisations to make informed decisions. This needs to be addressed” across both social care and health care to ensure the system is sustainable for the management of future public health crises. During 2020, we have called for better statistics to understand and take action to manage the COVID-19 pandemic. The immense effort required to re-work existing, or develop new, data collections has shown the administrative data systems used to collect information about public health are not suitable for the timely production of official statistics. The development of the UK COVID-19 data dashboard is a welcome innovation, but it would have been good to see data of sufficient granularity to meet the needs of the public sooner.

Our report about social care statistics in England highlighted a form of levelling up – that is, the work required to bring social care statistics to the levels of granularity and comprehensiveness of hospital care statistics. Additionally, the lack of joined-up data between health and social care has been a major data gap with serious consequences for many people during the pandemic. As well as improvements to social care, the opportunity afforded by the NHS White Paper in England could also enable the levelling up, in this sense, of primary health care data and mental health care data, as our report on adult mental health statistics in England outlined. Investment in IT infrastructure and staff skills will be needed to be able to accurately capture improvements in care outcomes that are provided by services such as these which take place out of hospital. We will continue to push for better social care data, working with technology leaders, such as NHSX.

Data gaps in education

Our report on the public value of post-16 education and skills in England noted “that better information about applicants to university would help shed light on social mobility. We also found that there are information gaps surrounding the further education workforce and workforce skills, which make planning to meet future demand difficult.” Across the UK, we found that “Significant gaps exist in statistics and data on individual student circumstances, in particular, about whether students are care leavers or have care experience. This information is self-reported and the data quality can be poor.” We continue to work with statistics producers to improve the information available.

Data gaps in poverty

Poverty remains a significant issue for the UK and has the potential to be of greater importance as we adjust to life following COVID-19, which is why we launched a systemic review on the coherence of poverty statistics in Autumn 2020. The volume of official data is difficult to navigate and does not reflect the changing nature of poverty or the unavoidable costs faced by low-income households, to the extent that stakeholders have developed new metrics outside the world of official statistics, including the Joseph Rowntree Foundation’s destitution study and the Social Metrics Commission’s framework for measuring poverty. Our report on poverty statistics will be published shortly, which we can share with the Committee.

Data gaps in rough sleeping

We worked with statistics producers to encourage the development of new and improved data and statistics on rough sleeping, especially during the pandemic. We reported this in a blog, which noted that “robust statistical evidence needed to answer key questions about the experiences of UK rough sleepers since the start of the pandemic is still lacking. New management information on the numbers of rough sleepers, and those at risk of rough sleeping, who have been provided with emergency accommodation since the start of lockdown is now being collected by UK councils. However, this management information is not always recorded consistently, and in many cases remains unpublished.”

Closing the data gaps

There are some examples of good work being undertaken to close data gaps, helping to provide a high-quality evidence base to inform levelling up.

The ONS publishes information about societal and personal well-being in the UK looking beyond what we produce, to areas such as health, relationships, education and skills, what we do, where we live, our finances and the environment. The data are available at local authority level and address the widespread concern that standard measures of the economy do not reflect the underlying welfare and well-being of the population.

The ONS also publishes data for the environmental accounts, including access to green space in Great Britain in 2020, at a local geographic level and includes Natural England survey data on garden access in England, broken down by personal characteristics such as age and ethnicity. Again, this work addresses a concern that focusing on traditional measures of the economy does not capture the environmental consequences of economic activity.

There are endeavours to address data gaps in a range of other policy areas:

  • Health: We were pleased to see that the ONS is developing a Health Index for England, which should allow for benchmarking the progress of local authorities. The Health Index is an Experimental Statistic to measure a broad definition of health, in a way that can be tracked over time and compared between different areas. The domains include healthy people, healthy places and healthy lives.
  • Deprivation: The Ministry of Housing, Communities and Local Government (MHCLG) publishes English Indices of Multiple Deprivation. OSR reviewed the outputs and noted that MHCLG has worked with the University of Sheffield and MySociety to develop a new mapping tool which allows users to visualise the statistics at different geographical scales including new geographies that the statistics have not been presented by before: Westminster Parliamentary Constituencies and Travel to Work areas.
  • Justice: In collaboration with Administrative Data Research UK, the Ministry of Justice is undertaking a data linkage project called Data First. OSR’s review, The Public Value of Justice Statistics, highlighted the need for statistics that move from counting people as they interact with specific parts of the justice system to telling stories about the journeys people take. Data First will anonymously link data from across the family, civil and criminal courts in England and Wales, enabling research on how the justice system is used and enhancing the evidence base to understand what works to help tackle social and justice policy issues.


OSR’s work is for statistics for the public good and one of our key tenets is to insist on transparency of data. Sometimes, particularly during the pandemic, the use of data has not consistently been supported by transparent information being provided in a timely manner. As a result, there is potential to confuse the public and undermine confidence in the statistics. It is important that data are shared in a way that promotes transparency and clarity. It should be published in a clear and accessible form with appropriate explanations of context and sources. It should be made available to all at the time the information is referenced publicly.

I hope this is useful to the Committee, and please do let me know if there is anything further I can do to assist with this inquiry or others.


Yours sincerely

Ed Humpherson

Director General for Regulation