Data for sustainable development
On 25 September 2015, the United Nations General Assembly adopted a fifteen year program for development. Building on the experience of the Millennium Development Goals, the new Sustainable Development Goals will involve hundreds of statistical indicators, many of which have never been systematically collected in so many countries with such widely varying capacities. In step with the General Assembly, a consortium of multilateral and bilateral agencies, civil society organizations, and private companies announced a Global Partnership for Sustainable Development Data to mobilise support for better statistics.
Two years earlier a high level panel commissioned by the UN Secretary General anticipated the need for more and better statistics to monitor the post-2015 development agenda. In their report, A New Global Partnership, they called for '…a data revolution for sustainable development, with a new international initiative to improve the quality of statistics and information available to people and governments.' They noted, 'a true data revolution would draw on existing and new sources of data to fully integrate statistics into decision making, promote open access to and use of data, and ensure increased support for statistical systems.'
Embracing the data revolution
The data revolution is underway. Big data, crowd sourced data and remote sensing data are already part of public discourse. But how will they change people’s lives, especially in the least developed countries? And will the change be for the better? A report by the Secretary General’s independent expert advisory group, A World That Counts, sees opportunities in new technologies, new methods and new partnerships between data producers and data users – public and private – to foment a data revolution for sustainable development to the benefit of all. But they warn '…the data revolution comes with a range of new risks.'
Among these are the risk that new actors with access to vast quantities of data will undermine traditional modes of governance and standards for statistics. Another risk is that traditional producers of official statistics will fail to innovate and seize the opportunities of the data revolution. The new demand for statistics of all types increases the incentives to produce more and faster, but it also increases the risk of untested short cuts, sloppy methods, private data hoarding, misrepresentation of facts and a lack of accountability.
Trust but verify
WorldStat – a working name – is a response both to the opportunities and the risks inherent in a data revolution. Now for the Long Term defines its purpose as '…to undertake quality control of global statistics, assess practices of both official and non-official producers of influential data, regulate misuse, and improve data collection.' This idea was subsequently fleshed out by a working group. They describe the proposed agency as an 'independent watchdog' working in the public interest to examine statistical weaknesses, publicise political interference and expose misuse of statistics.
But more than just barking, WorldStat should encourage greater efficiency, innovation and collaboration about statistical producers, and encourage the development of statistics to address emerging issues. Deviating from the Commission’s original formulation, the working group suggests that WorldStat would not have a proscriptive or regulatory function. Its authority would be based on its independence and adherence to scientific standards and ethical principles. Nor would it try to replace the role of the United Nations Statistical Commission or any existing agencies in promulgating international statistical standards. However, It would look closely at how standards are implemented, drawing attention to data quality issues, good and bad practices, and gaps in data and standards.
The working group has circulated a paper describing the new agency and is soliciting expressions of interest from potential hosts. It suggests a start-up team of a manager, three to four full time technical staff, one or two IT staff plus administrative support and three to four consultants. Governance would be provided through a board comprising of recognised international figures and donors and an advisory board of senior statisticians. Securing financial support for the estimated budget of $2.7 to $4.0 million in the first three years is, of course, essential. In October the working group will review the progress made on all fronts and adjust the plan accordingly. The Oxford Martin School will continue to lead this work and coordinate the activities of the working group going forward.
The views expressed in the Opinion section of StatsLife are solely those of the original authors and other contributors. These views and opinions do not necessarily represent those of The Royal Statistical Society.