2017 Evaluation of Aid Analysis

Confronting the Contradiction – an Exploration into the Dual Purpose of Accountability and Learning in Aid Evaluation

Kristina Bjorkdahl, Desmond MchNeil, Hilde Reinersten

Learning is a key purpose of aid evaluation. So why do aid organisations not learn more from their own experiences? More
specifically, why do they not learn more from their own evaluations? Yet learning is but one part of the well-established “dual purpose” of aid evaluation – the other key purpose is accountability.

This study investigate how these two purposes, learning and accountability, often are difficult to reconcile in practice. The report analyses the current and historical practices of aid evaluation in Sweden and Norway. The  empirical material consists of in-depth interviews with senior evaluation managers, a mapping of historical documents and a small sample of evaluation
reports. In addition, we review the existing literature specifically that which addresses the dual purpose of accountability and learning in aid evaluation.

The report was presented during the seminar Two sides of the same coin? Are learning and accountability compatible in aid evaluation?

Main findings (a selection)

  • A sample of evaluation reports shows that while they may clearly contribute to accountability, they to a much lesser extent contribute to learning.
  • The role assigned to and taken by the external consultants directly affects the learning potential. The most fundamental problem with using external consultants is that those who learn the most in the process have no responsibility for applying the lessons.
  • Feeding lessons learned back into the organisation by means of the evaluation reports and related efforts at synthesis and communication remains a considerable challenge for the evaluation staff. Their main experience is that few have the time to read evaluation reports and absorb their content.
  • The choice of evaluation system clearly has implications for how and where evaluation may contribute to either learning, accountability, or both.
  • The end beneficiaries of aid thus hold only a limited role in the donors’ own evaluation systems.
  • Finally, a fundamental problem is that the aid system on all levels displays exaggerated expectations of what aid evaluation may accomplish.

Recommendations

  • We must talk openly about the trade-offs between accountability and learning.
  • We must adjust our expectations to both aid interventions and aid evaluations.

Authors
Hilde Reinertsen, Postdoctoral Fellow, TIK Centre for Technology, Innovation and Culture, University of Oslo
Kristian Bjørkdahl, Senior Researcher, Uni Research Rokkan Centre, Bergen
Desmond McNeill, Research Director, Centre for Development & Environment (SUM), University of Oslo