The end of the week sees the meeting to agree the Sustainable Development Goals. Don'™t expect too many surprises; the document to be signed off has already been published. There are 17 proposed goals, with 169 targets. The good news is the impressive scope; the bad news is the impressive scope. The proposed goals cover new and important areas, and adopt new approaches, but the sheer number risks people picking and choosing between myriad targets. Pinning accountability for delivery, including through the development of national plans, is going to be the challenge.

So watch out for press coverage, and an overheated blogosphere over the next few days.  Over the next few months, and indeed years, many discussions will flesh out what the SDGs will really mean in practice. With that implementation question in mind, a couple of reflections:

First, the context is changing. Consider a couple of points;

• Since 2001 the number of low income countries has halved, with so-called '˜graduation'™ to middle income country status. With that rise in we
•  Overseas aid is falling as a share of external finance to poorer countries. Overseas Development Assistance (ODA) is worth about a third the value of remittances. Of course ODA remains important for many Ministries in poor countries, but while the MDGs aimed to align donor efforts, SDG success relies on national ownership
•  school enrolment at primary level in 1999 was 84% and by 2015 it had reached 93% . However, one in six children in low and middle income countries were not expected to reach the last primary grade. The policy problem is not simply one of access to interventions, it is of the effectiveness of systems.

Both the policies that the SDGs promote, and the accountability framework implied, need to reflect these changing realities. this will mean effective national ownership, moving beyond 'access' and towards more effective policies for children.

Second, one of the real advances over the Millennium Development Goals has been the recognition of the need for greater data disaggregation. Average progress often masks inequities. Proposed data disaggregation takes a number of forms -“ there has been a great recognition in the SDG debates about gender, understanding differences between women and men/ girls and boys. Some of that emphasis was there in the MDGs too, but it'™s been given greater importance in the SDGs. There is a recognition that age matters and poverty measures are intended to be broken down by age and gender;

Goal 1, target 1 .2 ; œ'By 2030, reduce at least by half the proportion of men, women and children of all ages living in poverty in all its dimensions according to national definitions.'

So it is not only a positive advance that we should begin to know more about child poverty from these SDG related measures, but that it will become more obvious that children are over represented among the poorest people.

Third, while there have been human development gains in the past decades, big problems remain. More young children are surviving to 5 years, with child mortality falling from 90 to 43 deaths per 1,000 live births between 1990 and 2015, but that global rate is still more than 10 times higher than the equivalent rate in the UK of 4 per 1,000 live births. And while building new services creates opportunities it hasn'™t always ensured quality, nor done so with equity. There are big challenges to grapple with to convert the policy efforts towards increased enrolment, and more basic infrastructure towards better child wellbeing. Key messages from our longitudinal research on that front: invest early, and sustain interventions, address the multidimensional disadvantage which otherwise undermines children'™s development. One key innovation in recent years has extended access to social protection systems, which help households manage risk in many countries. Ensuring good coverage for households with children is foundational to supporting the efforts of other policy interventions.

Fourth, should be the role of science in supporting better policy delivery for children? Research not only helps understand the world, but it helps policy makers formulate better responses. Data collection is expensive, but hardly as expensive as ineffective schools or other social policies.

So how can research help? The potential of evidence based policy has been recognised in the SDGs, with a call for a ˜data revolution™ to support better policy by more and better data to support policy makers and to encourage accountability. But data alone won'™t necessarily empower citizens or decision makers unless capacity exists within ministries and elsewhere to use that data effectively: evidence based policy requires more than evidence alone.

So, what would the SDGs have to do to create progress for poor children? From the experience of  Young Lives, I think the following will be critical;  

• Create a climate for greater attention to be given to children, with a particular emphasis on the earliest phase of life when more happens in children'™s lives than is really de
• Encourage systems building of interventions for children. Since children'™s lives are multidimensional, so too ought to be the systems which support them. Early childhood development and good social protection services which support wider he

Research can help inform options, but policy choices are political, not just technical. This brings me back to square one;  for the SDGs to advance progress for children, accountability will be the key.