you're reading...
Data Revolution, Opinion

A new household survey to catalyse the data revolution

Written by Sabina Alkire, Director, Oxford Poverty and Human Development Initiative (OPHI), University of Oxford.

This is the eight post in our blog series on ‘What kind of ‘data revolution’ do we need for post-2015?

Data drive policy. Yet data on poverty and deprivations are weak, limiting our poverty eradication efforts. While there have been huge leaps in the last thirty years, survey data remain severely limited in terms of both frequency and coverage. Meanwhile, data on other phenomena has exploded.

While stock market data are made public every day and labour force surveys are updated every three months, in India – the country with among the highest rates and number of malnourished people – malnutrition data have not been updated since 2006. When we are flooded with data in many domains, it is a travesty that we lack up-to-date information on – and across – key dimensions of poverty like health, wealth, and education. This data is needed to design high impact policies and celebrate policy success.

Indicators on the Millennium Development Goals (MDGs) came from censuses, administrative data, and household surveys; in future this will be supplemented by data from mobile phones and satellites. Yet across all countries – from developing to industrialised – national and international household surveys are a bedrock source of quality data, and remain key to ‘leaving no one behind’.

Data on poverty – which we understand to be multidimensional – should have three characteristics post-2015: be timely; cover multiple dimensions; and show that no one is left behind. A catalytic tool will be a post-2015 survey which generates comparable core indicators, and can reflect national priorities.


In terms of frequency, poverty data lag behind most other economic information, as survey data are published only every 3-10 years. For instance, data on malnutrition, housing, assets, schooling, etc. are updated approximately every five years in international surveys.[1] And some data are not infrequent, but rather missing altogether: in 2011, of the 19 countries of East Asia and the Pacific, data on malnutrition was available for only one of them (ESCAP).

Poverty reduction programmes could be far more cost-effective if they were informed by regular, timely data. It may even cost less to collect this data more regularly than to ‘update’ MDG estimations without new data (effectively making sophisticated office-based guesses as to the situation in each country). And it would have the advantage of accuracy. Infrequent data make it impossible to answer questions like the impact of financial crises or environmental disasters on the poor. Such questions need to be answered swiftly – not several years after the event.

Multiple dimensions

Poverty is more than a lack of income. A million voices have already said so in the World We Want report. If our post-2015 headline indicator of poverty is $1.25/day, and our data only cover expenditure, we will overlook things that matter. And even high quality multi-topic surveys rarely cover violence.

Our poverty data must illuminate peoples’ multidimensional experiences of poverty, including the many disadvantages that batter their lives simultaneously. It matters for high-impact policy. The first key message in The MDGs at Mid-point – a 50-country study on accelerating progress that the UNDP released in 2010 – was that successful countries had addressed different deprivations together.

This means surveys need to capture multiple dimensions of poverty, such as lack of services relating to health, education, and living standards, environmental degradation, assets or consumption, gender discrimination, quality of work, and violence. Gendered data should become the norm. And some questions need to improve, such as years of schooling, which does not proxy educational achievements. Current poverty surveys often omit important dimensions of poverty such as violence, empowerment or informal work. In future surveys, indicators must be well-designed and relevant, and capture data on the ‘missing dimensions’ of poverty that poor people say are important.

Leave no one behind

Data from a survey covering key domains of poverty can be integrated to construct a multidimensional poverty index which starts with each person and the direct deprivations he or she experiences at the same time. That is why over 25 countries and institutions in the global Multidimensional Poverty Peer Network are calling for a headline Multidimensional Poverty Index – or MPI2015+ – as a high-resolution tool to track progress towards the new development goals adopted post-2015.

An MPI2015+ can show how poverty varies between nations but also between groups according to their region, gender, ethnicity and other characteristics such as age, occupational status, disability, and so on. The data revolution must facilitate this sort of disaggregation, to communicate transparently who is most vulnerable and how their poverty is composed. Many actors can then work to ensure that no one is left behind, whilst also breaking apart the current silos of poverty reduction interventions.

A short, powerful survey with national and international indicators

To ensure both universality and national specificity, we need a brief, multi-topic survey that includes indicators on the key goals identified by the post-2015 development discussions (including economic, social and environmental aspects of poverty), and allows space for nationally chosen questions. The survey could be conducted using different institutional arrangements to match different contexts, with different statistical aspirations, capacities, and ownership profiles.

This new survey instrument must be short, powerful and selective – so it can be conducted every 2-3 years.  Its core internationally comparable questionnaire should take no more than 45-60 minutes to complete. The sample surveyed should be representative of the key regions or social groups, and should provide gendered data. A country might decide to append a set of questions (30-45 minutes) that reflect national priorities and the cultural, climactic, and institutional context, as well as participatory inputs on poverty priorities and characteristics.

Such a core questionnaire should not cover all post-2015 goals. Some indicators may require specialised surveys; some may not require updating this frequently; some may be sourced from community, administrative or census data; and some complex indicators may take too long to collect. Focus is essential to any revolution. As the Italian proverb puts it, ‘often she who does too much, does too little.’

Yet this survey could yield poverty data that provide profound insights into the profile of disadvantages poor people experience and the impact of poverty reduction programmes, bolstering the design, targeting and monitoring of future policy interventions. It is not the only tool required for a data revolution, but without such a tool, it is hard to see where the data revolution will begin.

[1] For example the splendid Demographic and Health Survey have been updated every 5.88 years across all countries that have ever updated them (across a total of 155 ‘gaps’ between DHS surveys). Dropping all incidents where 10 or more years have passed between DHS surveys, that average falls only to 5.31 years. (analysing www.measuredhs.com accessed 15 November 2013).



2 thoughts on “A new household survey to catalyse the data revolution

  1. misplaced concreteness? mass poverty ensues from a specific economic system, and therefore its solution lies in political action. what more do we need to know?

    Posted by A. Ercelan | December 19, 2013, 1:39 pm


  1. Pingback: Launch of a new blog series: What kind of data revolution do we need for post-2015? | Post2015.org - what comes after the MDGs? - November 21, 2013


January 1st, 2016

Site facilitated by ODI

This site is facilitated by the Overseas Development Institute.
%d bloggers like this: