Google supercomputers tackle giant drug-interaction data crunch Twenty tips to help you interpret scientific claims How oral sex can cause throat cancer by transmitting HPV

Science has become pervasive in public policy, and all of us who are active in this arena are aware of the intense scrutiny that scientific evidence rightly receives. Yet much of the data that shape and underpin crucial areas of public policy, such as improving health and reducing poverty, are substandard.

A report published this week makes a key recommendation to address this gap in data quality: the establishment of a new international agency, Worldstat. Worldstat would carry out quality control on global statistics. It would assess and improve data-collection practices and monitor for the misuse of statistics. Its role is crucial: unless all countries gather and publish reliable and comparable data on topics such as disease, income and employment, then international comparisons of economic growth, health, life expectancies and so forth cannot be relied on. Nor can such data form a firm basis for action by governments or international agencies.

There are currently serious weaknesses in data quality.

This proposal is part of a wider set of recommendations that have emerged from a year-long process by the Oxford Martin Commission for Future Generations, in which I have participated as a member. It is chaired by Pascal Lamy, former director-general of the World Trade Organization. The commission's 19 members, who hail from many nations and have diverse political and professional backgrounds, collectively have broad experience and expertise. But what brought them all to the table was a shared concern that a prosperous, equitable and sustainable global future is in jeopardy because modern politics and businesses have become too preoccupied with short-term pressures at the expense of long-term needs.

The result of our work is the publication of 'Now for the Long Term', a report that proposes a set of principles aimed at overcoming deep political and cultural barriers that obstruct a longer-term vision (see www.oxfordmartin.ox.ac.uk/commission). It provides practical recommendations for action to deliver progress on climate change, reduce economic inequality, improve corporate practices and address the chronic burden of disease.

Data vary so much across the world that in many fields it is almost impossible to generate reliable comparisons. Often, the information is simply not there. The paucity of economic data on key indices — such as average income in poor countries, particularly in Africa — makes it difficult to assess the true level of inequality and its drivers. Even in the United Kingdom there is concern about proposals to scrap the decadal census in its conventional form; whatever changes are made after the current consultation period must not compromise the collection of key data or the ability to monitor long-term trends.

Unreliable or missing data block good governance. One cannot properly assess the effectiveness of governments and international organizations unless there are reliable performance indicators. Such indicators help to reduce corruption and waste. There are currently serious weaknesses in data quality in areas as diverse as health spending, mortality rates, gender representation and biodiversity; assessments of success and failure in public policy are often based on distorted or subjective perceptions. As a result, indicators should come with a 'health warning' to emphasize their limitations.

Worldstat would not be a substitute for existing institutions such as the United Nations Statistical Commission or the UN Statistics Division (both within the UN Economic and Social Council). Instead, it would complement existing work by focusing on implementing agreed standards and improving the capabilities for archiving and interpreting data, particularly in the developing world. As a separate entity with a budget and resources on a scale comparable with Eurostat, Worldstat could also fast-track international efforts to adopt appropriate and robust indicators for sustainable development and direct attention to capacity building on this front. This week's report highlights, in particular, the need to devise a 'long-term impact index' that could comprehensively measure a country's long-term progress on a much broader range of indicators than the standard measure, gross domestic product.

The report also includes recommendations for corporate reform, such as a voluntary taxation and regulatory exchange, to address tax abuse and avoidance and to harmonize company taxation arrangements. It also offers specific proposals for dealing with youth unemployment and poverty by removing price-distorting subsidies and investing in social protection measures, such as conditional cash-transfer programmes.

The commission's work allowed us to identify the need for Worldstat, and the next step is to work with existing organizations to decide how the agency could be funded and organized, and to which bodies it would be accountable. Yet we are mindful that it is all too easy to propose new international agencies, and that few twentieth-century agencies have closed down, even though some are now anachronistic. To counter such proliferation, we suggest the introduction of 'sunset clauses' that require regular reviews of accomplishments to ensure that publicly funded international institutions are fit for twenty-first-century purposes.

Science and engineering can enhance lives in the developing world and safeguard the welfare of future generations. But there is a dismaying gap between what should be done and what actually happens. 'Now for the Long Term' presents a practical agenda that is designed to help overcome the gap between knowledge and action.

Credit: Oxford Martin School