Knowledge for Development

Challenges for Impact Evaluation of Agricultural Projects

Author: E. Sadoulet, A. de Janvry

Date: 11/01/2012

Introduction:

The difficulties of impact assessment are considered. It is concluded that methodological pluralism is needed, including use of Randomized Control Trials as one option in a portfolio of methods and not always as the superior approach. Each case is different. Hence, each case needs to be carefully diagnosed (through case studies, descriptive statistics) to understand in particular who has adopted and why non-adoption occurs.


 

E. Sadoulet and A. de Janvry. FERNI Policy Brief 29. Fondation pour les études et recherches sur le développement international. 08/2011.

http://www.ferdi.fr/uploads/sfCmsContent/html/111/B29.pdf

Impact analysis is all in design and robustness checks. Not everything can be measured for impact. It is consequently better not to do it when it cannot be done well than to do it wrong. Impact analysis is an art, not a mechanical application of methods. Hence, high quality impact evaluation can be very useful to help define priorities for agricultural investment, identify its effectiveness, and assist in improving implementation, if rigorously done when it can be done.


Download PDF (315.08 kB)