These key purposes do not always enjoy equal attention in the evaluation process, particularly as it relates to stakeholder involvement. Both evaluation policies in the South African government’s National Evaluation System (NES) and the authors’ first-hand experiences in conducting government evaluations in South Africasuggest that a greater emphasis has been placed on evaluation processes which support learning for performance improvement, than on traditional accountability purposes, and this has both potential benefits and costs.

With an increasing number of government evaluations being completed by PDG, applying NES policies and guidelines, we have an opportunity to reflect on the implications for accountability of “standardising” and applying utilisation-guided approaches to evaluation as the norm. This paper explores the problem of how progress in the consolidation of evaluation practices, according to NES standards, risks privileging the participation of government officials to the detriment of citizen participation and broader accountability. Looking at the authors’ experience in three evaluations, we conclude that collaborating closely with government in the evaluation of its programmes increases the usefulness of evaluations for government stakeholders, the buy-in to recommendations and the potential for learning in government. Strong collaboration with government, however, can also come at the expense of the opportunities for beneficiaries to use government evaluations to hold government to account.

This paper further argues that deepening the role that beneficiaries and their representatives play in evaluations can allow for greater usefulness of an evaluation for accountability of government to citizens, and still allow for learning and programme improvement.

Written by: Kevin Foster, Cara Hartley and Mike Leslie. 

A copy of Chapter 6 can be downloaded below.