Data and Development

This blog was originally published by Development Studies Association in their May 2018 bulletin.

The media has been full of concern about Facebook, and issues around privacy, personal data. How an online identity can be used, not only to sell us products we didn’t know we needed – but potentially to sway elections. The idea that if you don’t pay for a product, you are the product, casts a shadow over the role of data in everyone’s lives. But a different and more positive side of the data revolution is underway. And the twin issues of who owns data and what it can be used for – and the fundamental importance of identity – are absolutely central and deserve more academic attention. The idea of a data revolution that could be harnessed to deliver on the SDGs and ensure no one is left behind is a core part of Agenda 2030. At the first UN World Data Forum, held in Cape Town in January 2017, it was clear that the statistical community was moving fast to embrace new opportunities: joining up data from different sources; making new partnerships, being open to combining informal data with official statistics; using innovative methods of data collection.

Governments in the shape of National Statistical Offices and line ministries plus civil society organisations (some with their own datasets and others looking for data) were also busy exploring how the data revolution could help them deliver better on their poverty-related objectives. Academics concerned with data science, mapping technology and innovative uses of data were much in demand. But otherwise, the academic community who could potentially use the outputs of the data revolution – who could provide a stronger evidence base that would help to deliver on SDG 1 and SDG 10 – as well as potentially demonstrating how aid can work – seemed rather thin on the ground. It will be interesting to see whether the academic community is more engaged at the Second World Data Forum in Dubai during October 2018. Or at the fifth Knowledge, Policy and Statistics meeting in Incheon a month later.

The data revolution has sparked welcome attention to the need for disaggregated data – especially to ensure that no one is left behind – a key component of the agenda 2030 commitments agreed by leaders from every country at the UN in September 2015. Nearly 30 years on from Naila Kabeer’s influential paper ‘Monitoring Poverty as if Gender Mattered’, gender identity has moved on from being a matter of checklist to a matter of reflex within the development community. Recently the issues of disability (thanks to CSOs such as ADD and Sightsavers and the efforts of the Washington Group) and age (thanks to HelpAge and the new Titchfield Group) are fast becoming identities that are increasingly taken into account in programme design. LGBT issues are also being prioritised by the more progressive donors.

This represents real progress in terms of how the data revolution is helping to deliver on efforts to address multidimensional poverty and exclusion. But there is an underlying principle that – for practical and academic reasons – really needs more attention. To look at the practicalities first: we all, as individuals, have multiple identities. These may be related to ethnicity, nationality, class or caste, religious persuasion, gender, sexual orientation, age, disability status. Or any of the many identities (some transient) which help to distinguish or define us. These can appear trivial: supporting a particular sports team or type of music. But at particular points in people’s lives, these identities and identifications can be enormously important to the individual – as well as shaping how the individual is perceived by the outside world. From a statistician’s point of view, to count all of these identities is fiendishly difficult and often expensive. From the perspective of organisations or governments trying to address the multidimensional nature of poverty and disadvantage, all of these identities can be potentially crucial influences on how social and economic interventions may work. And there is a real problem of people with limited resources and limited attention, unwittingly setting one identity against another – because they can only afford to gather data on a limited number of disaggregations. CSOs passionately arguing for the interests of their constituency, or bureaucrats having to prioritise can so easily act in a way that seems to elevate, for example age over LGBT status or disability over ethnicity. So the point of principle that needs to be explored and understood is fairly simple – and we should have learned this as we have gradually got to grips with the issue of gender.

The fundamental principle is that a person’s identity should never be the cause of their immiseration or exclusion. Can the academic community through its analysis and its teaching help to build on our shared progress on gender, to sensitise people to this broader principle? If we can do this, if we can help people broaden their understanding that a single identity – be it gender or disability or age – should not result in people being disadvantaged, to a broader awareness that identity-based vulnerability or exclusion must always be looked for and guarded against, then we will help to ensure that the data revolution uses identity in a way that can help to deliver an end to poverty and greater equity in every country by 2030.