Explaining Usage Patterns in Open Government Data: The Case of Data.Gov.UK
Jonathan Bright
University of Oxford - Oxford Internet Institute
Helen Zerlina Margetts
Oxford Internet Institute, University of Oxford
Ning Wang
University of Oxford - Oxford Internet Institute
Scott A. Hale
Oxford Internet Institute, University of Oxford
June 3, 2015
Abstract:
Open data is one of the most significant current trends in public administration, with over 100 current open government projects around the world investing considerable time and effort in opening up public sector information for free re-use. The movement, whilst attracting widespread support, has also proved highly controversial, with critics highlighting both the high financial burden it appears to place on government departments and its uncertain political consequences; with some going so far to claim that the rhetoric of government transparency and open innovation has been co-opted into the service of neoliberal deregulation.
Despite the significance of the movement and the importance of the critiques, "little systematic OD research has been performed to date" (Peled 2013, 187), with both supportive and critical authors largely relying on anecdotes and one-off case studies. The major reason for this lies in the openness of open data itself: with no requirement to create an account before accessing data, there are no records of who is using it or for what purpose. Hence it is hard to systematically answer many of the major questions being raised about open data.
The aim of this article is to take a step towards remedying this deficit. Based on a unique dataset created through data downloaded and scraped from the UK’s open data portal website, we develop and test analytical models which seek to explain the amount of times individual datasets have been downloaded. We explore factors relating to both the economic sustainability of open data (asking whether extra departmental effort in curating datasets results in extra download levels) and its political implications (asking whether datasets relating to financial transparency or datasets with profit making potential are downloaded more). Our results challenge some of the key critiques currently live in the open data debate, and also suggest a way forward for the area in terms of financial sustainability and its contribution to democratic accountability.
Despite the significance of the movement and the importance of the critiques, "little systematic OD research has been performed to date" (Peled 2013, 187), with both supportive and critical authors largely relying on anecdotes and one-off case studies. The major reason for this lies in the openness of open data itself: with no requirement to create an account before accessing data, there are no records of who is using it or for what purpose. Hence it is hard to systematically answer many of the major questions being raised about open data.
The aim of this article is to take a step towards remedying this deficit. Based on a unique dataset created through data downloaded and scraped from the UK’s open data portal website, we develop and test analytical models which seek to explain the amount of times individual datasets have been downloaded. We explore factors relating to both the economic sustainability of open data (asking whether extra departmental effort in curating datasets results in extra download levels) and its political implications (asking whether datasets relating to financial transparency or datasets with profit making potential are downloaded more). Our results challenge some of the key critiques currently live in the open data debate, and also suggest a way forward for the area in terms of financial sustainability and its contribution to democratic accountability.
