• By 2030, we should be living healthier, more prosperous lives on a greener, cleaner planet helped by robust policies to combat climate change and use precious resources, like water, wisely.

    That is, if world leaders live up to the promise of ambitious development goals to end poverty and hunger, and promote a more sustainable way of life, over the next 15 years.

    But for governments to achieve the Sustainable Development Goals (SDGs), due to be adopted at a U.N. summit this month, there must be a push for better data to allow progress to be tracked and leaders to be held to account, experts say.

    Too often data that is produced is inaccurate, driven by donor priorities, released after a long time lag or omits groups on society's margins. As a result, both rich and poor countries know a lot less about their citizens than they think.

    "There is no data on the value of data," said Elizabeth Stuart, research fellow at Britain's Overseas Development Institute (ODI) thinktank, which analysed data flaws in a report earlier this year.

    "It blows my mind that we do all this policymaking and planning based on guesstimates and extrapolations and interpolations behind the guise of empiricism," Stuart told the Thomson Reuters Foundation.

    Take poverty. The most up-to-date World Bank estimate of the number of people living on less than $1.25 a day is just over 1 billion.

    But that figure is four years old, and could be a quarter less than the real number, because as many as 350 million people worldwide are not counted in household surveys, the ODI says.

    Similarly, some 133,000 women in sub-Saharan Africa may have died from childbirth-related causes in 2013 - or maybe twice as many, according to the ODI.

    WRONG DATA, WRONG POLICY

    Experts agree that even in this patchy state, data has improved since the last U.N. development goals were launched in 2000. But the task ahead - finding data to track 17 goals and 169 targets compared with eight goals and 18 targets previously - is daunting.

    Data gaps are not found only in poorer countries struggling with problems like power cuts, insecurity or a shortage of statisticians to count communities that are nomadic or cut off by mountains, rivers and bad roads.

    A project to road test the SDGs and assess the data available in key areas, such as education, poverty and environmental sustainability, showed Canada lacked data on its Aboriginal people - 4.3 percent of the population.

    Ethnic minorities, women, the elderly and the disabled are least covered by data and risk being left behind unless countries get better at counting them, experts say.

    "The big question for high-income countries is going to be around the 'leave no one behind' agenda, and that becomes a political decision of who gets measured and why they get measured," said Shannon Kindornay, project leader for the Post-2015 Data Test.

    The "what" that gets measured also needs careful attention.

    In Nigeria, net school enrolment rates were used as an indicator of progress on universal primary education.

    Once cash payments were introduced to encourage parents to send their children to school, enrolment rates skyrocketed. But when the authorities later measured literacy rates, they were found to be falling.

    "Parents enrolled their children to collect their checks and other benefits and subsequently withdrew them from school to help at home and on their farms," Nigeria's statistician general, Yemi Kale, told a briefing at London's Chatham House in July.

    By focusing on enrolment rates, the quality of education or whether children were completing their schooling were ignored. "This is an example of wrong data, wrong diagnosis, wrong policy," Kale said.

    "Many African countries often complain that international commitments require us to track set indicators that are more relevant to advanced countries than our own economies," he said.

    A DATA REVOLUTION

    The solution lies in investing more in national statistics offices, better use of already available data, improved data accuracy and the use of alternative data sources, experts say.

    "Botswana's civil registration statistics are over two decades out of date. At the same time the country is being photographed by satellite every 24 hours," said David McNair, who leads advocacy and policy work on transparency at anti-poverty campaign group ONE.

    "These photographs could be used to understand where people are living in poverty, how crops are growing, where governments or donors need to make early interventions to avoid hunger and malnutrition," he told the Thomson Reuters Foundation.

    In Namibia, researchers have used satellite and cell phone data to fight malaria, according to the Data Impacts project.

    By consulting satellite images of conditions in which mosquitoes thrive, like rainfall and the density of vegetation, and cellphone records to track population movements, the authorities were able to distribute bed nets in hotspots.

    In Uganda, another research project mined Facebook data to understand attitudes about contraception methods and family planning among young people.

    "The data revolution is really popular," said Amanda Glassman, director of global health policy at the Center for Global Development.

    "Everyone can agree there's a problem and we should do something about it, but then the solutions tend to focus on these technological quick fixes or on filling a data gap."

    The focus should be on getting basic indicators like gross domestic product or vaccination coverage, right, she said.

    The data revolution may already have begun but to be truly effective, citizen-generated and other types of data should complement traditional methods of data gathering such as household surveys and censuses, experts say.

    Others caution that more data will not guarantee better policymaking and accountability.

    "You can't assume that just because you have the data, governments are going to take notice of it and start changing their policies accordingly," the ODI's Stuart said.

    Tanzania, for example, has been a member of the Open Government Partnership, a global initiative to promote transparency, since 2011. But it passed a law this year which only allows data endorsed by the government's Bureau of Statistics to be published.

    "Paradoxically, some of the governments that have the best policy around open data are also ... clamping down on freedoms," Stuart said.