The open data initiative has gained traction and visibility in recent years, particularly with various governments releasing a wide range of public data. Andrew Stott, the former Director for Transparency and Digital Environment for the UK Government, is an expert in this area who has been deeply engaged with the British government’s transformation towards data-driven governance. We interviewed Andrew when he came to speak at the Talk Big Data seminar on ‘Big Data, Big Brother’ at the University of Essex on 17 November 2016.
Could you tell us about the work you have done and your work at data.gov.uk advocating for open data?
I have had a multi-faceted career in government. My recent work has been about driving the open data revolution, fuelling research, improved services in government, and public participation in improving public services at local level.
To illustrate the impact open data can have on the delivery of services at the local level – two examples might be helpful. First, the publication of data on hospital infection rates in the UK drove down infection rates by 84%, without allocating more budget for this in the health service. It pushed hospital managers to ensure staff took basic hand washing hygiene seriously, and to spend money that was already allocated for deep cleans on actually doing the deep cleans. Second, in Uganda giving mothers in rural villages data on public health and data on the resources that the government was allocating for rural health led to a drop in infant mortality by a third, compared to control villages where this was not done. Mothers became more assertive about ensuring their children receive the inoculations they were entitled to, and about expecting healthcare workers to turn up when they were supposed to.
Much of my work in the British government involved service transformation, transformation of government, data-driven governance, the use of data for operational decisions, and the use of administrative data within government. My final job in the government was to lead the work on open data, much of which has surfaced through data.gov.uk. From my experience, the UK has been an international leader in the area of open data. Since my retirement from public service in 2010, I have continued to provide advice to the Cabinet Office, the Foreign Office, the Department of Transport, as well as other governments bilaterally and through international organisations. This has been focused on open data for transparency and accountability, and open data for business innovation and economic growth.
Open data does not include personal data. Even if personal data is made public, the use of such data is still governed by data protection principles. For example, even if the names of company directors are public information, there is no consent for their information to be used for junk mail. For information about people convicted in court, such data may be relevant only for a restricted amount of time. Sometimes it is possible to anonymise personal data so that it can be freely used; but that needs care: one of the things we did at the start of the data.gov.uk project was to commission academic work with some ‘white hat’ hackers to deepen our understanding about anonymization. We looked into the measures that needed to be taken to satisfy beyond reasonable doubt that the anonymization could not be reversed – not just within the data itself but also by using other available datasets.
What is your sense of public response to the open data initiative?
For a whole host of open data initiatives, there has generally been a positive response.
Take for example the publication of educational performance figures – knowing how your child’s school is performing is important to parents and it gets them more involved in education. Some teachers and educationalists did object to the publication of educational performance figures because they perceived it as telling parents that their children are going to a poorly-performing school. But what is the justification of withholding that knowledge? The evidence has shown that publishing this information drove performance up. Interestingly, the data shows that some of the best school leaders are going to work at schools where they can make a difference. The publication of data also encourages improvement in the accuracy and relevance of the data: since the initial publication of exam results more factors have been explored to study how children perform and how much they advance in a particular school – social class, income, good parenting and so on. That data did not exist until the initial data got published.
So unless you get the data out there, there is not much incentive to improve. Having the data out there, with more people using it, empowers the people within the organisation who want better data to make the case for funding efforts to improve the upstream sources.
What is the relationship between open data and change?
An open government is not just about the government making information available; it is also about the government listening to and engaging with civil society. I have seen a variety in this regard – from governments that publish a lot of information but do not listen, to some other governments that publish information and actively stop civil society organising around that information.
There are relatively few “armchair auditors”, since most people do not like to spend their evenings ploughing through large government datasets on expenditure – but it only takes a few to highlight questionable payments and so make officials look more closely at the justification for everything they spend.
What is interesting here, is how community groups have started using data to agitate for change. They do not look at data in the abstract and think about a cause; they have a cause and want to make an improvement. Having data allows them to lobby more effectively. For example, tools developed in a project about visualising transport data, a project I am involved in for the European Commission, have been used by community groups in Birmingham to visualise road accidents involving children in relation to the location of schools. This has helped to identify unsafe junctions that need to be improved, and safe walking routes.
In the UK the central government has a challenging task to ensure minimum standards and local responsiveness across huge national public service delivery networks. Hence, there has been a self-interest for central government to adopt an open data policy as it creates greater transparency in the performance of public services. This enables local communities to engage and press for local improvements where they are due, which improves performance and encourages greater accountability. For example, health service performance data for individual hospitals was made public on nhs.uk, which prompted patients to question the disparity in performance between hospitals. Similarly, crime data published on police.uk has led to more public engagement in setting local policing priorities. Drawing on another example in Uganda, at one time only 24% of the budget allocated to the schools actually reached the schools. The Government published the budget it was allocating for individual schools, and parents and head teachers could question where the rest of the money had gone if it had not reached the schools. Over five years, the portion of the money actually reaching the local school rose to 80%.
What role does transparency play in the successes that open data brings about?
We are facing a time where there is an erosion of trust in governments by citizens. Therefore, governments need to be more open about what they are doing, how they are performing, and engage citizens more. That is the way – the only way – that they can maintain public trust. Engaging local communities is also the way that services can be developed to meet local needs.
Key insights from the interview
Our interview with Andrew is evidence that information is powerful. An informed citizenry is an empowered citizenry, and informational transparency from the government is a critical driver of public trust. The many examples Andrew shared illustrated that open data can reveal the potential causes of problems, trigger action, and generate positive change.
Disclaimer: The views expressed herein are the author(s) alone.