Open Geographical Data, Policy Transparency and Accountability

Abstract

Open data is often assumed to automatically and directly improve policy transparency and accountability. Recently, a number of international aid transparency initiatives have emerged that use open geographical data disclosure as a strategy to promote transparency and accountability. The assumed direct causal relationship between open data disclosure and transparency and accountability however, has not yet thoroughly been studied. The objective of this research is to explore the evidence on the extent to which open aid data projects can contribute to improving transparency and accountability, and in the ways system design and the use of geographical information contribute to those objectives.

First, it was theorized that the relationship between open data disclosure and transparency is not direct but rather a cyclical communication process. In order to arrive at transparency, information must be disclosed in an uncolored, unbiased and complete manner. The information must then be interpreted by a user in an effective and meaningful way. This implies that the information must be relevant to the user and that the user has the capabilities and skills to access and process the data. The user’s perception of the usefulness and usability of the information is therefore regarded as a determining factor for the degree to which open data can contribute to transparency. Transparency in turn is a necessary precondition for accountability.

Second, a number of exploratory interviews were held with (potential) stakeholders in the transparency and accountability field. The results show that although participants intended to make use of open aid data for their professional activities, they all experienced major difficulties in accessing the data. The most frequently reported issues are technical difficulties, excessive use of jargon and open data use being too time consuming. Despite those technical usability issues, all participants felt that there must be important information to be found in the dataset, even though they were not able to retrieve it themselves.

The third research phase was a platform assessment, in which respondents were asked to carry out a predefined search scenario on two open aid data platforms and subsequently rate usability and usefulness items. One platform makes use of geographical information technology and the other does not. Although the results in terms of platform appreciation were inconclusive, respondents’ platform preference did suggest that the simulated information need may influence the perceived usability of open aid data platforms.

The two research phases yielded very different results: actual users (interview participants) generally reported having issues with ease of use elements of open aid data which caused them to refrain from using the data entirely. The platform assessment results however, did not indicate any major ease of use issues. Furthermore, the assessment results on the effect of the use of geographical information technology and visualizations remains inconclusive.

This study shows that the relationship between open data disclosure and transparency cannot be assumed to be causal or direct: stakeholders who are willing to make use of the data have postponed or even refrained from using open aid data. This suggests that effective open data disclosure has to be tailored to the information need of the (potential) user. It is recommended that open data architects define and target one or several specific user groups. Their known information needs may help improve platform design and dataset contents to real information needs.