The Challenges of #Data4Impact

We often assume that more Open Data is better – but is it?

Without a doubt, the production and use of data has added value for development practitioners, governments, and citizens. Accessible and transparent information allows non-traditional stakeholders – such as academics, journalists, and citizens – to conduct research and contribute to domestic and international dialogues. Traditional stakeholders also benefit; transparency means heightened visibility and recognition for progress, standardization allows for easy sharing of a re-useable product, and the analysis of large datasets is increasingly tied to more informed decision making.

However, bottlenecks continue to hinder the production and use of Open Data:

Standardization – While organizations and governments are increasingly adopting data reporting standards, (such as the IATI Standard for development assistance information), creating “clean” datasets can be a time-consuming process.

Development Gateway is contributing to ongoing efforts in making IATI data downloads and uploads more accurate and user-friendly. However, these efforts are a two-way street – further adoption of the IATI standard by governments and organizations will make sharing even easier, as more stakeholders will be reporting in a similar fashion.

Timeliness – Especially when used for decision-making, datasets need to be updated consistently and frequently, requiring resources and man-hours.

Context – Datasets should not be interpreted in a vacuum; location and context are needed to ensure accurate and efficient policies.

Why is context so important? Take this research on a transit study performed for Côte d’Ivoire. Researchers for the study used cell phone traffic from a leading mobile network provider in the country to analyze commuting patterns and chokepoints in the nation’s capital. The winning research paper used this data to propose a model for public transportation expansion. But what these datasets failed to show (and what was therefore unaccounted for in the proposal) was the widespread civil unrest occurring throughout the data collection phase. Without incorporating the on-the-ground reality, the public transport expansion model did not holistically reflect the needs of the citizens of Abidjan.

Applicability – Finally, perhaps the most complex bottleneck to tackle is the production of datasets that are valuable for, and used by, practitioners, governments, and citizens; how can we produce the “right” kind of data, and how can we encourage uptake?

Development Gateway recently conducted an in-depth analysis of government data use and uptake in Nepal. While the full report is forthcoming, a major takeaway from our work was the gap between the reporting of data, and the demand and use of data. Officials often perceived data reports as a process necessary because of protocol, but not necessarily as a useful tool to inform decision-making and resource allocation. Further efforts will thus to happen to make data “relevant,” and as a vital part of policymaking.

How then can we resolve these data bottlenecks?

1. Continue to iterate on and improve open data standards and technologies.
2. Encourage direct and indirect stakeholders to contribute to the data dialogue and demand accurate, relevant data, and more rigorous data usage.
3. Groundtruth quantitative data – citizen voice is key for providing context to datasets and fostering informed decision-making. More Open Data is better than no open data – when it’s accessible, accurate, and actionable.

Image source: www.epicgraphic.com