Using community-generated data to deliver and track the Sustainable Development Goals at the local level


National progress can mask a disparate picture within a country, but if communities can take ownership of identifying and tracking local progress toward the Sustainable Development Goals we are in a powerful position to ensure no one is left behind.

The project, led by the Open Institute, with Development Initiatives working as the partner responsible for data quality and analysis, has begun with a pilot study into community-generated data in Lanet Umoja, Kenya. The study’s aims were simple: explore whether communities can assess and identify their own local development priorities using the Sustainable Development Goals (SDGs), and help them to generate timely and accurate data that feeds into local and national development intervention plans. This would help ensure the SDGs are met for their community, and that the data is available and accessible in an open format through Lanet Open County portal.

The study is due for completion in October 2016 but already it makes a strong case for the benefits of community-generated data. Two rounds of data collection covering 9,136 households have shown that this local approach is more cost-effective in collecting census data. Furthermore, the method used in this exercise demonstrates the value of collecting SDG data based on local interpretations of the Global Goals, making them more relevant to people’s aspirations for social progress. Just ten months in and already the project has led to improved buy-in to the Global Goals – a better understanding of what they are and how they can be useful as a planning and participatory tool. The project has already led to tangible changes in people’s lives – with Lanet Umoja residents drinking and cooking with clean water for the first time.

  • The community generated data on number of households and household sizes is real data, not based on a sample survey of households in the wider population.
  • Household budget data is recent and collected by knowledgeable residents, who can better cross-check the responses to expenditure module of the report. This serves to reduce reporting errors on key aspects of the survey. Household expenditure structure is not extrapolated from a large sample: rather, it is built bottom-up from the collected data as reported by each individual household. By cross-referencing with a real dependency ratio calculated from this data, consumption per capita makes more sense than when calculated across a large population using inferences based on average household sizes.
  • Overall, the database from this exercise can be built overtime to capture real social progress – in access to basic social services, for instance – and a measure of improvements to living standards in the long run.

Read our case study in full