Impact assessment, or impact evaluation, is research that tries to establish which changes in the lives of clients or communities were caused by a programme, such as a microfinance or savings scheme, rather than by other factors. The central problem is attribution: incomes may rise for many reasons, so the study must estimate what would have happened without the programme.
The most rigorous method is the randomised controlled trial, which compares a randomly chosen treatment group with a control group. Quasi-experimental methods, such as matching or difference-in-differences, are used when randomisation is not possible. A set of randomised studies of microcredit published together in 2015 found modest average effects on business activity and little evidence of large gains in household income, which reshaped debate about the sector. Impact assessment differs from impact monitoring, which tracks outcome indicators routinely without attempting to prove cause. Rigorous studies are expensive and slow, so most institutions rely on monitoring and use full assessments selectively.
Ref 1120000182