“Geeks Meet Government”
Tag-line of Rewired State hack-days
It is hard to understand how OGD is being used without having an understanding of who uses OGD and what factors motivate their engagement with open data, and government data in particular. Whilst making no claims to be a representative sample, this study’s survey data suggests most OGD users are male (1-to-6; a similar ratio was observed at OGD events), better educated than the population average, and working either in private sector SME firms (39%), or in the public sector (34%) with only a few (3%) from the voluntary sector. Asked about labels they would ascribe to themselves (from a pre-specified list), survey respondents in general associated strongly with the labels ‘Citizen’, ‘Open Data Advocate’, ‘IT Specialist’, ‘Data Specialist’ and ‘Web Developer’, but very weakly with labels including ‘Policy Maker’, ‘Politician’, and surprisingly, given the proportion of the sample government employers, ‘Civil Servant’.
O’Reilly blogger Nat Torkington (2010) has suggested there are five types of people with an interest in OGD: “…  low-polling governments who want to see a PR win from opening their data,  transparency advocates who want a more efficient and honest government,  citizen advocates who want services and information to make their lives better,  open advocates who believe that governments act for the people therefore government data should be available for free to the people, and  wonks who are hoping that releasing datasets …will deliver…economic benefits to the country”. However, whilst statements that would capture each of these motivations were included in the survey, rather than finding neat divisions between different motivations for working with OGD, this study finds a range of overlapping motivations, many tied together by an ideological focus on digitized and efficient government. Nat’s list notably omits policy, social and business entrepreneurs interested in exploring how OGD may be exploited. These groups featured strongly in the survey sample and in participant-observer research environments.
Exploratory analysis of respondents’ self-assessed motivations for engaging with OGD led to the identification of six overlapping motivational clusters: government focused; technology innovation focused; reward focused; digitizing government focused; problem solving; and social or public sector entrepreneurialism. Using multidimensional scaling we can plot the relationship between these motivational clusters in two-dimensional space as in Fig. 4 below.
The small cluster of OGD users who can be described as ‘Problem solvers (5)’, engaged with OGD because it was the best tool to help them meet a pre-defined goal (their own, or given by a client/colleague). They may have learnt new skills to work with the data. This group’s interest in government data and the technologies for using it is based on its functional value for specific tasks. A separate cluster (6), are also interested in OGD for its functional value, but have more direct interest in government data, often as small private-sector businesses or social enterprises providing services to government, or with core business based around PSI.
The remaining motivational clusters are more closely interrelated: few individuals can be isolated who are driven by just one set of these motivations. For example, engagement with OGD may take place both interests in technological innovation and political views regarding the current inefficiency and unaccountability of the state. The arrows in Fig. 4 draw on both correlation between clusters and qualitative data to indicate common relationships between these clusters. Some OGD users had prior interests in the accountability and efficiency of government, but for many OGD users a focus on government data emerges from the availability of that data. The fact that many of the technologically skilled OGD users who have developed a focus on government data work within small entrepreneurial private sector businesses may explain the focus on government efficiency, and on reforms designed to reduce the size of the state and to prioritize private sector innovation in many of the statements made in survey responses and interviews. The relationship between a ‘Reward’ motivation (3) and ‘Technology innovation’ motivation (2) may be partially understood through the lens of open-source and online hacker culture (Castells 2001, p.47), where “prestige, reputation and social esteem” are gained by contributions to the community. In the case of OGD, where ‘the community’ increasingly includes governmental decision makers and budget holders, there are also potential power and patronage factors involved.
The most significant motivational cluster (4) has been labeled ‘Digitising government’ to capture a set of beliefs concerning use of technology to reform government. Respondents in this motivational cluster are likely to believe that “Releasing OGD is a matter of principle” and “Government currently makes poor use of data”. However, it may also be possible to more richly conceptualize this category by drawing on Kling and Lacono’s (2003; 1988b) notion of ‘Computerization Movements’ as way of describing a “movement whose advocates focus on computer-based systems as instruments to bring about a new social order” (1995, p.122). They note “few CM activists, including those who publish their arguments, assert their key ideological themes directly” (1988a, p.235). This supports an identification of this ‘Digitising government’ cluster as motivated by a belief in the capacity of OGD to bring governmental and political reform, whilst the potentially divisive details of such reform are left unelaborated. The overlap between an OGD CM and those motivated by an interest in linked-data and semantic-web technologies (motivational cluster 2) is notable.
In summary, OGD users span a wide range of contexts, although currently the voluntary-sector is under-represented. Multiple motivations drive engagement with OGD, and Fig. 4 presents these on axes relating to their government or technology focus. Further insights into the nature of OGD users will be gained as this study turns to explore OGD use in practice.
 Space precludes the presentation of the full exploratory analysis here. Full details are presented at http://www.practicalparticipation.co.uk/odi/survey where the underlying data is also available. In summary: survey respondents rated their motivations on a scale from ‘Not-at-all-important’ to ‘Very-Important’. The correlation matrix of response was visualized and candidate motivation clusters identified before detailed checks on cluster coherence were made (including checking against factor and bootstrapped cluster analysis, and checking interpretability against qualitative data). A composite scale for each cluster was created summing responses for each item in that cluster. The inverse of the correlation matrix of composite clusters provides a distance matrix for multidimensional-scaling.