Ghana, DFID, Impact Evaluation of MVP-SADA Northern Ghana Millennium Village

2011-2016; Ghana; DFID

The MVP approach was piloted in Kenya and Ethiopia and in 2006, reached nearly half a million people across 10 countries in sub-Sahara Africa. In 2011 the UK Department for International Development (DFID) agreed to provide a grant of £11.5 million to implement a new Millennium Village in Northern Ghana. The MVP ran for five years from 2012 to 2016 with interventions targeting a total population of nearly 30,000 people.


The impact evaluation in Northern Ghana is an independent evaluation of the MVP approach. The MVP approach draws on the ‘poverty trap’ theory presented by Professor Jeffrey Sachs and collaborators. The poverty trap theory suggests that, amongst other factors, high transport costs, poor agricultural productivity and the burden of disease locks poor countries into a cycle of poverty. It argues that small investments are unlikely to yield significant benefits in the long term and what is needed is a ‘big push’ in order to lift poor rural communities out of poverty. A set of interventions is recommended to facilitate this ‘big push’ such as stimulating improved agricultural productivity and market development, enabling people in rural areas to save and accumulate wealth, and investing in health and family planning.

Over the years, the MVP approach has been the subject of considerable debate and received much press coverage and online interest. This scrutiny demonstrates the need for a robust, methodologically rigorous, and independent impact evaluation to analyse how the approach delivers against the MDGs. In collaboration with the Institute of Development Studies and PDA Ghana, we assembled an expert team with skills in evaluating impact deploying both quantitative and qualitative research methods. The evaluation uses a quasi-experimental approach, with a Difference-in-Difference design to compare the change in outcomes in MVP with the same outcomes in areas not covered by the project (a comparison group). A number of survey rounds were conducted including household surveys (sampling over 2,000 households), adult (male/female) surveys, facilities surveys, village surveys, blood tests, anthropometric measurements, cognitive tests and time preference/ expectation surveys. Qualitative research methods were also been used to explain the impacts and particularly differential impacts on specific socio-economic groups. The methods included wealth and well-being assessments, focus groups using participatory research, institutional assessments (of community and district government), and immersion studies (the ‘reality checks’ to capture the lived realities of local people). The impact evaluation also conducted a cost-effectiveness analysis.

Read the final report, executive summary, briefing papers and annexes on our resources page.

For more information please contact:

Chris Barnett
T: +44(0)1273 765250