Research on Open Data and Transparency

Leave a comment

Some collected new (ish) work on FOI and Open Data


Some collected new (ish) work on FOI and Open Data. All are free to access on the links…

Worthy, Ben and John, Peter and Vannoni, Matia, Transparency at the Parish Pump: A Field Experiment to Measure the Effectiveness of Freedom of Information Requests (December 4, 2015). Available at SSRN: (see also FOI man’s excellent summary here and this longer assessment 201607-parishcouncils)

Worthy, Ben, Freedom of Information and the Media (September 12, 2016). Available at SSRN:

Worthy, Ben and Hazell, Robert, Disruptive, Dynamic and Democratic? Ten Years of Freedom of Information in the UK (December 28, 2015). Parliamentary Affairs, Forthcoming. Available at SSRN:

Worthy, Ben, The Impact of Open Data in the UK: Complex, Unpredictable and Political (March 5, 2015). Public Administration 93 (3): 788-805, 2015. Available at SSRN:

Leave a comment

New Paper: The Impact of Open Data in the UK: Complex, Unpredictable and Political

revenue expenditure01Image


This article examines the democratic impact of the UK Coalition Government’s Transparency Agenda, focusing on the publication of local government spending data as the first steps in an evolving ‘ecology’ of Open Data transparency. It looks at whether the Open Data has driven accountability, participation and information transmission. Rather than forging new ‘performance regimes’ or bringing mass use and involvement, the publication of spending data adds a further element of political ‘turbulence’ that can ‘punctuate’ the ‘equilibrium’ of local politics (Hale et al 2013). The evidence finds that the spending data, so far, has driven some accountability but less participation or information transmission. Taken together, assessment of the three objectives reveals that the use and impact Open Data is far more complex, more unpredictable and more political than the rhetoric around Open Data indicates. The danger is that the gap between aims and impact invites disappointment from supporters.

Worthy, Ben, The Impact of Open Data in the UK: Complex, Unpredictable and Political (March 5, 2015). Public Administration 93 (3): 788-805, 2015.

Paper Available here :

Leave a comment

Who Is Using Open Data? Some More Thoughts



By their very nature users of open data are hard to track down. Looking at the most viewed and downloaded datasets for raises more interesting questions that it answers. There are a few interesting signs, such as the type of people and organizations requesting datasets from

My own research on local government publication of all spending over £500 tells us a little about who is using it. The findings are based on a mixture of FOI requests, surveys and interviews (you can see more stats here).

First, interest in the local spending data on local authority websites seems low. Based on Freedom of Information (FOI) requests to local authorities, the average number of page views was 200, pretty low compared to visits to other sites (note visits and page views are different):

Site Visits per month 540,609 161,101
TheyWorkForYou 200-300,000
WhatDoTheyKnow 100-200,000

(Sources: Escher 2011, 2011, 2013)

Second, those who are using it seem to be a rather diverse bunch. Here’s a comparison of users of FOI at local government level (from this report) and the spending data.

Estimated breakdown of requester groups: FOI and spending data (%) null

The biggest user group appears to be business. The data may be very useful for tenders, contacts or competitors. As Jo Bates points out here, this raises some interesting questions as to who ultimately benefits from open data. Businesses are also big users of FOI at local level, from small ‘one person’ companies to multi-nationals.

Another significant user group is media. Authorities reported a ‘flurry’ of interest from journalists but one that often died down. A search of regional newspapers found 148 articles specifically using the term ‘spending over £500’ between May 2010 and August 2012. The spending data was used to question authorities, focusing on odd spending at a low level, such as this.

NGOs made up just 5% of all users of the spend data. High profile campaigns in the London borough of Barnet sat alongside smaller more focused uses by community groups over particular salaries or contracts. There’s also been some interesting innovations with, for example, Shelter using open data to build a searchable portal on homelessness.

The public were supposed to be the beneficiaries of the new data. Exactly who the ‘public’ users are remains a bit of a mystery. Like FOI, they may be a hard core group of activists with a wider, looser group made up of the curious or those seeking very specific information. There are a few, but not many, Armchair Auditors, who have not yet moved from settee to spending data.

One problem is that all these discrete categories are all a little too neat. Research by Tim Davies on found a mixed group of activists and ‘techy’ people, where the campaigner and person were one and the same person.

Another problem is that the research only tells us about those directly using the sites. From what we know from communication theory, opinion formers pass on information to others. Who, if anyone, is this data trickling down to? Not all users are equal – a journalist can have thousands reading a story. More importantly, who is using the new apps that and sites such as Openly Local or, and the Local Government Association’s LG Inform tool?

So different groups are using it different ways. There is no ‘average’ or ‘person on the street’ data users. This makes the future of open data very unpredictable. Just as it should be.

Leave a comment

Who Is Using Local Spending Data

Here’s some of the results of my survey of English Local Authorities on Open Data (£500 spending). You can also see it as a pdf Survey of English Local Authorities on Open Data

Conducted: April-October 2013 via online survey (BOS)

Population: 102 (2 x excluded)

1.     What type of transparency information is of most interest to users?

user interest

2. Which group or groups are the primary users of the data?


primary user pie chart



3. How would you characterise the level of use of the spending data?

q3 use level


4. What type of transparency will this data create?

q4 type of rans

 5. What type of transparency will this data create?


 q5 barriers


Pageviews of Spending Data on Selected Local Authority Websites January-December 2012[1]

Council Type Page Views per month (average)
District 35
District 87
District 99
District 341
Borough 355
District 437
City 450
District 528
District 610
District 636
City 2021


Comparative Visits for January 2012 on government and non-government Open Data sites

Site Visits per month
Local Council Average 200 540, 609 161,101
TheyWorkForYou 200-300,000
WhatDoTheyKnow 100-200,000

( Esche 2011)

For more details

Worthy, Ben, ‘David Cameron’s Transparency Revolution? The Impact of Open Data in the UK’ (November 29, 2013). Available at SSRN:

Worthy, Ben ‘Evidence to PASC on the Aims of Open Data’

Worthy, Ben ‘Evidence to PASC on the Impact of Open Data’




[1] Based on FOI requests to English local authorities in 2013 asking for hits/views on spending data pages of their website.

Leave a comment

Who’s Doing What? PEW on American attitudes to Government Open Data

Some interesting research from PEW on American attitudes to government Open Data-see the full report here

Here’s some patterns on usage


Few Americans think governments are very effective in sharing data they collect with the public:

Somewhat larger numbers could think of examples in which their local government either did or did not do a good job providing information to the public:

  • 19% of all Americans could think of an example where the local government did a good job providing information to the public about data it collects.
  • 19% could think of an example where local government did not provide enough useful information about data and information to the public.

Relatively few Americans reported using government data sources for monitoring what is going on:

  • 20% have used government sources to find information about student or teacher performance.
  • 17% have used government sources to look for information on the performance of hospitals or health care providers.
  • 7% have used government sources to find out about contracts between government agencies and outside firms.

Here’s some thing else interesting-what does the American public think the benefits will be?


Leave a comment

What can Open Data do for Parliament?

Leave a comment

New Research on Open Data: Making Transparency Stick and International Comparisons


Some autumn reading. Here’s a conference paper examining what could make Open Data succeed or fail over the next few years.Ben Worthy Making Transparency Stick-The Complex Dynamics of Open Data.

Here’s the abstract

This paper examines the complex dynamics of Open Data reform in the UK, assessing the chances of the policy ‘sticking’ or failing over time. Using the ideas of Patashnik and Zelizer (2013) on what makes policies succeed or fail post-enactment, it begins by looking at the unique features of Open Data. The broad but vague vision of the reform, its symbolism and ‘voteless’ status and the multi-instrument, multi-actor approach all make Open Data exceptional. The paper then examines how these play into the three factors that make a policy ‘stick’ or fail over time: the resources re-allocated by the policy, interpretation of its success by different actors and the institutional support it receives. It concludes by arguing that Open Data is likely to benefit from leadership and the ongoing innovation but may be threatened by resistance, manipulation of the aims and the underlying assumptions, which invite disappointment.’

You can also download it here Worthy, Ben, Making Transparency Stick: The Complex Dynamics of Open Data (September 17, 2014). Available at SSRN:

Paper two is this great piece by Tim Davies on international comparisons you can read here

New national Open Government Data (OGD) initiatives are being launched on an almost monthly basis. From Brazil to Burkina Faso, governments are establishing open data portals and committing to make machine-readable datasets available for re-use. Similar patterns are being replicated at the local level, with municipalities and sub-national states also establishing their own open data projects. At first glance, many of these local and national initiatives appear almost identikit copies of each other: using the same data portal software, and selecting similar datasets for their initial launch. Increasingly, the preparation and launch of open data initiatives follows a orthodox approach involving hackathons, training events and outreach activities, designed to build interest in, and demand for, newly available open data. Yet, the countries launching these OGD initiatives are vastly different: in their levels of development, their political structures, and their public policy priorities. This raises important questions about the nature of open data policy, and policy transfer.’

Davies, Timothy Glyn, Open Data Policies and Practice: An International Comparison (September 5, 2014). Available at SSRN:

Leave a comment

Revolution or No Change? The PASC report on Open Data


The Public Administration Select Committee has examined in detail the progress of the Government’s Open Data agenda (you can see the evidence here). Its final report has made some important points about how the new agenda is developing:

First, the report argues that it’s hard to tell what impact the new Open Data reforms are having: ‘It is very difficult to assess the performance of Government in enhancing accountability through opening up its data […] measuring progress on this important agenda is difficult if not impossible’. Consequently ‘this allows supporters of open data to claim the revolution is well under way and the sceptics to say nothing has changed’.

Second, the government itself is still unclear on what it all means: ‘The concept of open data is poorly defined’. Something is happening here but no one, so to speak, knows what it is. The government seems to be making several assumptions about what will happen. It expects economic growth without recognising that innovation comes from unexpected areas. It wishes to increase accountability but, as the report argues ‘simply putting data “out there” is not enough to keep Government accountable’. It is trying to encourage more public participation but ‘Government…make too much use of jargon’ which can ‘alienate and confuse people who do not have expert knowledge of the technical terms’.

Third, as a result of the above, implementation and impact is uneven across government. Some departments and bodies may be fully behind it but others may be simply republishing old data (as happened in the US with Without clarity or standard measures, it’s very hard to know.

So what should be done? The report offers a number of suggestions (see also the report’s conclusions)-

  • There should be a central champion for Open Data in the Cabinet Office, helped by a streamlined panel of experts to replace the advisory board. At present there are simply too many competing groups and bodies.
  •  Government should ‘clarify’ what it means by Open Data and explain the benefits to a sceptical (and sometimes concerned) public.
  •  To ensure consistency the Government should adopt a ‘star-rating system for engagement…for measuring, and reporting to Parliament on, Departments’ progress on increasing accountability through open data. The Government should expect Departments to set out plans to move towards Five Star Engagement for all their data releases’. This can be used to create a definitive list of what data departments should publish.
  • There needs to be awareness of the pitfalls and dangers around privacy but ‘the recent controversy over demonstrates the danger that concerns about privacy will unduly undermine the case for open data’.

Hidden away inside are further interesting recommendations:

  •  Open Companies? In terms of who should be open the report says ‘Open data principles should be applied not only to government departments but also to the private companies with which they make contracts’. The Committee goes on to recommend ‘that companies that…provide contracted or outsourced goods and services should be required to make all data open on the same terms as the sponsoring department.’ This is interesting in the light of discussion around extending FOI.
  •  What should be opened up? The report argues that ‘core data needs to be released fast and, above all, free’ including Postcode Address File (PAF) that has been the subject of intense debate for a number of years. This core data provides the building blocks for innovation.

Overall, the report offers a kind of ‘optimistic concern’, speaking of the huge opportunity Open Data presents but also its concern that it is not being fully exploited:

‘Today there are unparalleled opportunities to harvest unused knowledge that otherwise goes to waste, which can be used to empower citizens, to improve public services and to benefit the economy and society as a whole’

We will await the government’s response…

You can see my written evidence here, oral evidence from Helen Margetts, Tom Steinberg, Rufus Pollock and me here and Nigel Shadbolt here.

Leave a comment

Who is using all this Open Data?


As politicians promise more and more openness, there is one big gap in our understanding: Who is using all this information?

Work on Freedom of Information showed that nailing down users is tricky. Our best guess is FOI users are a diverse mix of public, businesses, NGOs and (a few) journalists (see this attempt at a typology of FOI users written with Greg Michener). But because we don’t know, all sorts of myths and claims fill this gap in our knowledge. Tony Blair believes FOI is ‘mostly’ used by journalists and David Cameron probably thinks similarly, given that he believes those using FOI are ‘furring up the arteries of government’-see this table from our evidence  to the Justice Select Committee for our best guess.

Estimated Requester Groups to Local and Central UK government compared with EU Access to Documents legislation (%)

Requester group Local Government Central Government EU Access to documents Legislation
Public 37 39 32
Journalist 33 8 3
Business 22 8 -8
Academic 1-2 13 23

So what about Open Data? My own research and other work in to the impact of publishing all spending over £500 points to a few interesting things. First, local use seems low. Most authorities appear to registering a few hundred views a month of their online spending information, though a few are far higher (see also this research from the Finnish Institute and this transparency survey from the LGA).

Estimated Breakdown of Requester Groups-FOI and Spending Data (%)

Requester group Local Government FOI Local Government Spending Data
Public 37 23
Journalist 33 31
Business 22 39
Academic 1-2

Second, local users are diverse. We seem to be seeing a mixture of businesses, NGOs, members of the public and a sprinkling of journalists. Who the ‘public’ are remains a bit of a mystery. Like FOI, they may be a hard core group of activists with a wider, looser group made up of the curious or those seeking very specific information. Research into the government’s data portal by Tim Davies found a similar mixed group of activists and ‘techy’ people.

So what does this mean? It means that any impact on local government is likely to be uneven and unpredictable. It could create odd or perverse incentives. A journalist highlighting spend data on hotel bills will have a very different reaction from an authority than a business looking for spending on stationary or a member of the public seeking allotment costs.

It also quite fragile. Like participation in politics generally, all this Open Data it depends on a few people getting active and involved. Whether such uneven and fragile work can create an army of auditors or bring big changes to how local authorities work is another matter.

The low use also tells us that this spending information isn’t precisely what is wanted. As the Local Government Association has pointed out, and ten years of FOI tell us, people want information of value to them-‘real time’ bin collection activity and costs not spreadsheets.  But that’s not to say they cannot be merged, as I argue in my research.

Finally, if you are an Open Data user can you help with my survey?


Where are the Armchair Auditors?

armchair by Didouner One of the most eye-catching ideas from the government’s Transparency Agenda is that of the ‘Armchair Auditor’. Armchair Audit is a form of ‘crowd sourced’ accountability where citizens use newly ‘opened up’ information on spending, contracts and crime to hold public bodies to account. But has it happened?

David Cameron hoped a critical mass of citizen watchdogs would become a new force for accountability.As he put it a ‘whole army of effective armchair auditors looking over the books’ would act as a check on ‘waste’. Citizens being ‘able to see how your taxes are being spent’, ‘judge standards in your local schools and hospitals’ and ‘find out just how effective the police are at fighting crime’ would drive better performance and efficiency.

Much of the emphasis has been financial and local due to the publication of local authority spending over £500 (see this LGA survey for background). Eric Pickles put it bluntly-the data would ‘unleash an army of armchair auditors and quite rightly make those charged with doling out the pennies stop and think twice about whether they are getting value for money’.

Yet the army has not appeared, as the BBC pointed out. This is not to say there are no auditors. Adrian Short has created an open access software taken up by Auditors in the‘North’ (Hull and Lincolnshire) and on the Isle of Wight. In London one particularly high profile case led to Eric Pickles publicly praising Armchair Auditors in their fight against Barnet’s ‘easycouncil’-see this account here. Journalists offered tips and ran storiesand the Taxpayers Alliance promoted it as a grassroots weapon.

But this vanguard shows little sign of becoming a citizen audit army. There are a number of reasons for this, some obvious, some less so.

The first and most obvious barrier is the information. It is not yet consistent so questioning and understanding it is not easy or possible. The Reluctant Armchair Auditor and the Public Accounts Committee have expressed concern at the quality and comparability of information. This may improve in the future, as this FOI response from DCLG suggests, with new applications making it easier to compare.

The second obvious point is that to be an Armchair Auditor you need to be a particular type of person. Politically, you need to be engaged and interested in local government, understand how local government works and have a driving reason to dedicate yourself to it. To have all these traits in combination is rare-the public are, by most measures less engaged, though there may be glimmers of hope for local government. Add to this the time and the need to access technological know-how and this narrows down the potential pool of Armchair Auditors to a select few. Even then there is no guarantee as this would- be-but-frustrated Auditor explains.

The third barrier is what you do with the information. Like FOI, it is hoped that information equals accountability. But there is a missing link here-what do you do with the information once you have it? Do you pass it to the authority itself? A member of the opposition? The press? Or do you self-publish and hope it is picked up? As this discussion on the E-democracy blog shows, it is not clear exactly where this new information fits in. Again like FOI, the data is likely to work alongside other accountability tools-the activists in Barnet have used a combination of FOI requests, the local media and judicial review.

A final point is that Armchair Auditors, like all accountability mechanisms, work best in certain situations. A high profile controversy or scandal, as in Barnet, appears to be one way of kick starting them. A lack of activity from the official opposition or local press may drive groups to do it themselves- so auditors can fill accountability vacuums. The paradox is that for Auditors to then be effective they often need to use these other channels that failed.

David Cameron pointed to the MPs’ expenses scandal as being an example of what could happen when information was released. Yet this scandal actually illustrates how complex accountability can be. It was driven not by citizens but by determined journalists pushing a case through appeal and the courts. The key moment came via a good old fashioned (paid for) leak. It took four years, persistence and a whole range of accountability tools working in a very particular context.

As this paper argues, crowd sourcing exercises are often fragile and subject to bias, relying on a ‘tiny subset’. Auditors, like many other such initiatives, are better seen as a compliment to existing systems than a replacement. Auditors will rise up in certain situations and the new information will form part of the armoury of activists, journalists and NGOs. However, the citizen army is likely to remain in the barracks.

This post was originally published on the ODI blog