This is the response to the National Data Strategy and its consultation document based on presentation and discussions at the December meeting. You will need to access the consultation document to fully understand the structure of this response and of course you need to read the National Data Strategy itself. Apologies for the font differences.
- To what extent do you agree with the following statement: Taken as a whole, the missions and pillars of the National Data Strategy focus on the right priorities.
West Midlands Open Data Forum (WMODF) welcomes the publication of the National Data Strategy and is delighted to respond to the consultation. We trust our views will constructively improve the strategy and will be reflected in the implementation plan.
WMODF promotes the release, re-use and integration of open data to benefit communities, businesses and public services in the West Midlands area. We have a broad and active membership covering public, private and third sectors and have regular meetings (currently via Zoom).
To inform our response to the consultation we held an open online event for members and guests on 18th November 2020 informed by two external speakers, Rachel Coldicutt from Careful Industries and Dr Milly Zimeta from the Open Data Institute. We decided to respond to specific areas of the strategy that impacted directly on local/regional issues and the interests of our members.
We feel that the strategy is an important step forward but lacks a vision and only offers a framework for the critical changes needed to take full economic and social advantage of a data-driven UK. Over the ten years of our forum’s existence we are concerned that we have seen much discussion about the need for a national framework and common standards and infrastructure but patchy progress at a local level. We believe that a national strategy can only be delivered through local plans, coordination and commitment.
Various comments (from Chat and anonymised):
The national data strategy is ambitious in addressing a huge issue within the public sector where data isn’t being shared – but the presentation raises a bigger even more important point about the data we do not already know and are NOT currently being fed into decision-making – and that bigger point isn’t addressed in the strategy.
VCS, community groups etc. have extensive data on their provision that does not influence public realm programmes
We are in the middle of data-driven economy. How can we ensure shared data to be used responsibly by both public and private sectors? The lack trust and transparency of the current ecosystem must be addressed, otherwise it will be harder to encourage the sharing IMO.
My experience is that understanding of data protection is lacking and that is hampering data sharing. People who have data are so scared of sharing data incorrectly that they don’t share any at all unless instructed by a central body to do so.
the strategy addresses operational bottlenecks I’m seeing within the way I do my work – and the presumption in favour of sharing addresses that. However, it certainly doesn’t address bigger and more important question raised in this discussion.
We see culture and leadership as a major gap in the strategy, only briefly mentioned…this is critical
in a framework strategy there should be some kind of vision of what the ultimate goal is. I guess they mean the opportunities: though I’m not sure all of these are right and am surprised there’s not more specific consultation on these. I’m sure in earlier consultations ‘Sustainability’ was up there – though I notice they’ve gone for ‘growth’
Re: Q 8 What could central government do beyond existing schemes to tackle the particular barriers that small and medium-sized enterprises (SMEs) face in using data effectively? Why just SMEs and not the not-for-profit sector?
I think a major problem for employers (public sector/not-for-profit is that they don’t know what’s possible, what they want to do as an organisation, and therefore who/what skills they need to employ. A data scientist might not be the first hire. A colleague of mine noted the strategy seems to also focus on building data science skills over broader data literacy. It’s like telling everyone they need to know calculus rather than basic numeracy
I’ve found that if you say Data Analysis to most managers in councils they think it means taking some numbers and making a graph so senior managers will understand it.
Agree with X (above). The lack of understanding over data analysis and insight can mean a lack of skills development in the right area, particularly data literacy. They think if you know a piece of software, or can make a chart, that’s enough.
Some smaller, community focused or managed, groups do not access funds to enhance their data / digital provision. therefore they continue to make do
There is a growing recognition in CS that data scientists training should also include responsible innovation and ethics, whose importance cannot be underestimated
We have had such a long history of producing efficient data scientists or ML modellers, but neglecting the social responsibility training in their curriculum. We’ve learnt hard lessons from this
They’re going to hire for Excel skills, not SQL, R, Python &c or knowledge of how to approach the data best.
Qu. 1: Missions and pillars. The pillars summarise the key issues effectively. We feel they should include the existing declared commitment to open access as the default position and a recognition that attitudes to data usage need to change as well as skills development.
Mission 1: Unlocking the value should recognise that there is data held at local level which may not be known but has an important role in community delivery such as the provision of lunch clubs contributing to adult care. New areas are arising such as the collection and analysis of changed day care provision for adults with learning difficulties during the COVID crisis which Midlands Mencap used to provide greater contact and identify improvements to future provision.
Mission 2: the data regime needs to recognise the unfortunately increasing distrust of authoritative data (fake news), the concern that data can often be partial and inaccurate particularly for those who aren’t able to engage and the specific issues for minority communities.
Mission 3: A transformation of usage needs to build on clear and required standards and an acceptance of openness of data as the starting point. We are acutely aware that open data is not a priority for many of the managers with whom we have contact – in fact it is seen as a threat.
Mission 4 Yes
Mission 5 – We have a lot to learn from our partners across the world
2. We are interested in examples of how data was or should have been used to deliver public benefits during the coronavirus (COVID-19) crisis, beyond its use directly in health and social care. Please give any examples that you can, including what, if anything, central government could do to build or develop them further.
Please note that we are only looking for examples outside health and social care data. Health and social care data will be covered in the upcoming Data Strategy for Health and Social Care.
In monitoring local authority care prevention grant provision (maximisation of income element), one local authority in the West Midlands only wants a report from a group of local voluntary organisations on the maximisation of income for individuals through benefits which has been raised and is attributable to their grant funding, which is what is requested – this cumulative figure would be about £2m in the past 7 months; if we include work undertaken through other grants by the same organisations in multiple-funded organisations, data in which the council don’t seem interested, the actual figure is nearer £20m – this represents data on the impact of Covid in this area and the effects on a number of people who have lost work or income as they were on zero hour contracts or self-employed in service/hospitality sector when the Covid crisis started, and the effects on their families
Several colleagues made these observations about skills gaps”I think a major problem for employers (public sector/not-for-profit is that they don’t know what’s possible, what they want to do as an organisation, and therefore who/what skills they need to employ. A data scientist might not be the first hire…A colleague of mine noted the strategy seems to also focus on building data science skills over broader data literacy. It’s like telling everyone they need to know calculus rather than basic numeracy” and “I’ve found that if you say Data Analysis to most managers in councils they think it means taking some numbers and making a graph so senior managers will understand it” and “Agree with X. The lack of understanding over data analysis and insight can mean a lack of skills development in the right area, particularly data literacy. They think if you know a piece of software, or can make a chart, that’s enough” and “Some smaller, community focused or managed, groups do not access funds to enhance their data / digital provision. therefore they continue to make do” and “There is a growing recognition in CS that data scientists training should also include responsible innovation and ethics, whose importance cannot be underestimated” and “We have had such a long history of producing efficient data scientists or ML modellers, but neglecting the social responsibility training in their curriculum. We’ve learnt hard lessons from this” and “They’re going to hire for Excel skills, not SQL, R, Python &c or knowledge of how to approach the data best”
3. If applicable, please provide any comments about the potential impact of the proposals outlined in this consultation on individuals with a protected characteristic under the Equality Act 2010.
4. We welcome any comments about the potential impact of the proposals outlined in this consultation on the UK across all areas, and any steps the government should take to ensure that they take account of regional inequalities and support the whole of the UK.
We agree that there are many currently hidden economic and social benefits in correlating data across (and within) organisations and sectors.
Our concern is that local opportunities and issues are rarely recognised in national analyses. Birmingham has a leading and growing community of digital businesses and expertise which needs incentives to turn their resources to this area.
As our ODI Leeds colleagues put it “if the UK’s data policy is debated and set in London then the requirements and concerns of businesses in and close to London will be weighted more heavily in that policy. Businesses with their senior team elsewhere will be disadvantaged”
There can be unexpected benefits from linking datasets
From May 20th WMODF meeting: Tom Forth gave an introduction to the Index of Deprivation based on the entry in Wikipedia which he had recently updated. There are issues with UK data such as different measures by each home nation which prevents comparisons. Also because by England’s standards nearly all of Birmingham is deprived, mapping data like Covid-19 deaths onto local deprivation was unhelpful. There was a need to recalibrate.
ONS reports used age standardised mortality rates and set least deprived areas to zero on their charts. Tom had worked on different analyses and his code has been uploaded to github. ONS analyses show increased deprivation implies higher mortality but his recalibrations for Birmingham show no pattern i.e. the virus affects everyone.
• In France there is much more data at Departement level
• There is no definitive list of care homes to assist in locating mortality hotspots
• Covid testing is skewed towards health and care staff, many of whom live in the more deprived areas.
5. Which sectors have the most to gain from better data availability? Please choose all relevant options, below.
This list is based on the Standardised Industry Classification (SIC) framework for categorising work of organisations.
- Charity or Non-Profit
- Public Sector/Central or Local Government, including Defence
- Information and Communication
- Professional, Scientific and Technical Activities
- Arts, Entertainment and Recreation
- Human Health and Social Work Activities
- Wholesale and Retail Trade
- Other: Social enterprises
We have emphasised the areas in which we have the most expertise
6. What role do you think central government should have in enabling better availability of data across the wider economy?
We need clear standards and allocated responsibilities to avoid wasted effort.
An example would be local authority data structures in the West Midlands. Effective implementation however must be at a local and or regional level. In the next report, responsibility needs to be determined, local commitment required and resources identified, particularly for the public and third sectors. Different incentives are needed for the private sector to recognise the wider benefits of sharing data.
The view of our Forum, based on many years of monitoring activity in local as well as in our region is that there has been little regard to the efficiency and application benefits of sharing data ad infrastructure. Clearly this is likely to continue unless there a strong external push at a national or regional level to gain the benefits of cooperative infrastructure and data sharing.
6a. How should this role vary across sectors and applications?
We want common data structures across all sectors but recognise that different information will be valuable to different sectors e.g. data about the individual is esp valuable to health and other Public Sector whereas data about land has great importance for local govt but also private sector e.g. construction. Therefore the enabling role of central govt should be sensitive to the priorities of the different sectors while maintaining a firm steer on basic standards
The basic data is immensely flawed because people protect it and don’t expose it
LAs do not encourage using us as the people who can comment on their data
Exceptions are used to defeat the broad rule
Concern, largely by public sector managers, about security stifles data sharing
Data foundations: The true value of data can only be fully realised when it is fit for purpose, recorded in standardised formats on modern, future-proof systems and held in a condition that means it is findable, accessible, interoperable and reusable. By improving the quality of the data, we can use it more effectively, and drive better insights and outcomes from its use.
7. To what extent do you agree with the following statement: The government has a role in supporting data foundations in the wider economy.
Data Orchard: Here’s a summary of a local data sharing/covid recovery response from our website. Note the dashboard already includes some regional data and could be fairly easily replicable to other areas:
“As Herefordshire recovers from the impact of the lockdown and manages the impact of Covid-19 on the community and economy, public services, business and communities need good information on what is happening. The Covid recovery dashboard pulls together a range of data to help people understand what is happening in Herefordshire right now. It is kindly supported by Herefordshire Community Foundation where we’re working in partnership on data sharing in relation to economic and social recovery at a local level.”
We agree with ODI Leeds that “Government organisations should have websites from the day they are created, with regular updates on their work, the data that they are using to take and justify decisions, and links to other organisations that they work with”
8. What could central government do beyond existing schemes to tackle the particular barriers that small and medium sized enterprises (SMEs) face in using data effectively?
We ask why include just SMEs and not the whole not-for-profit sector? Open and share all the data unless there is a good reason not to share specific pieces of data, and make that open to scrutiny
The Smart Data Review in 2019 consulted on ways to make evolving schemes more coordinated across banking, finance, telecoms and energy. The focus of Smart Data is citizens asking their providers to share information about them with third parties.
9. Beyond existing Smart Data plans, what, if any, further work do you think should be done to ensure that consumers’ data is put to work for them?
10. How can the UK’s data protection framework remain fit for purpose in an increasingly digital and data driven age?
In section 7.1.2 of the framework National Data Strategy we lay out the functions of the Centre for Data Ethics and Innovation (CDEI), set up in 2018 to advise the Government on the use of data-driven technologies and Artificial Intelligence (AI).
11. To what extent do you agree with the following statement: the functions for the Centre for Data Ethics and Innovation (CDEI) should be Artificial Intelligence (AI) monitoring, partnership working and piloting and testing potential interventions in the tech landscape?
We agree with ODI Leeds: The CDEI seems to have largely ignored the opinions it received in response to its founding consultation. Except for one meeting in Edinburgh it has never held a meeting outside of London. It seems to have only just created a website, which remains hard to find and does not project any independence from the UK government. There is little knowledge of its existence. Even to those of us who are interested in it, it remains unclear why it was set up and what it does.
This and the other institutions mentioned in the strategy i.e. Alan Turing Institute, the National Innovation Centre for Data, the ODI, the Data Skills Taskforce, the AI Council, the UK Cyber Security Council and others should be measured on how they lead by example in terms of data publishing, sharing and use of data. The government should regularly check that their location represents, and can work with, a data industry that is widely distributed across the UK.
11a. How would a change to statutory status support the CDEI to deliver its remit?
12. We have identified five broad areas of work as part of our mission for enabling better use of data across government:
1. Quality, availability and access
2. Standards and assurance
3. Capability, leadership and culture
4. Accountability and productivity
5. Ethics and public trust
We want to hear your views on any actions you think will have the biggest impact for transforming government’s use of data.
the five broad areas of work are unfortunately all of key importance and success can only follow the coordinated addressing of them all
We agree with ODI Leeds: A Tableau [or other tool] dashboard with a spreadsheet download button would probably have been cheaper, better and would have been available much earlier. It would have also served as an example to local government and other institutions that cannot afford to develop custom dashboards for their needs, but could learn from central government using widely available tools.
13. The Data Standards Authority is working with a range of public sector and external organisations to create a pipeline of data standards and standard practices that should be adopted.
We welcome your views on standards that should be prioritised, building on the standards which have already been recommended.
We are aware of work underway to develop data standards and data practices. Data that Government requires as returns from public sector organisations is well defined and standardised in the main but this data sometimes is not made available for wider use and standards may still need coordination across different government departments.
The infrastructure on which data relies is the virtualised or physical data infrastructure, systems and services that store, process and transfer data. This includes data centres (that provide the physical space to store data), peering and transit infrastructure (that enable the exchange of data), and cloud computing that provides virtualised computing resources (for example servers, software, databases, data analytics) that are accessed remotely.
14. What responsibilities and requirements should be placed on virtual or physical data infrastructure service providers to provide data security, continuity and resilience of service supply?
14a. How do clients assess the robustness of security protocols when choosing data infrastructure services? How do they ensure that providers are keeping up with those protocols during their contract?
15. Demand for external data storage and processing services is growing. In order to maintain high standards of security and resilience for the infrastructure on which data use relies, what should be the respective roles of government and data service providers, their supply chain and their clients?
16. What are the most important risk factors in managing the security and resilience of the infrastructure on which data use relies?
For example, the physical security of sites, the geographic location where data is stored, the diversity and actors in the market and supply chains, or other factors.
- To what extent do you agree with the following statement: The government should play a greater role in ensuring that data use does not negatively contribute to carbon usage?
- Strongly disagree
- Somewhat disagree
- Neither agree nor disagree
- Somewhat agree
- Strongly agree
Please explain your answer here. If applicable, please indicate how the government can effectively ensure that data does not negatively contribute to carbon usage.
If govt doesn’t do this, who will?
18. How can the UK improve on current international transfer mechanisms, while ensuring that the personal data of UK citizens is appropriately safeguarded?
We will seek EU ‘data adequacy’ to maintain the free flow of personal data from the EEA and we will pursue UK ‘data adequacy’ with global partners to promote the free flow of data to and from the UK and ensure it will be properly protected.
19. What are your views on future UK data adequacy arrangements (e.g. which countries are priorities) and how can the UK work with stakeholders to ensure the best possible outcome for the UK?
We agree with ODI Leeds view “We are open to this arrangement if those promoting “UK data adequacy” can clearly give examples where this would help UK businesses more than the cost of dealing with double regulations. Until then, the UK should aim for EU data adequacy above all, even if our efforts are not initially recognised. Unilaterally emulating the EU’s regulations on data would be a cheap and easy option that would reduce the need for extra bureaucracy within the UK”