n Journal of Public Administration - Assessing the impact of complex social interventions

Special issue 2
  • ISSN : 0036-0767



Many social interventions do not lend themselves easily to impact evaluation. A successful impact assessment of a relatively simple linear cause-effect relationship is at best not a straightforward matter. Most policy interventions are, however, typically complex organisational and institutional programmes that have ambitious multi-sectoral transformation agendas. Such complex interventions make life even more difficult for the evaluator who wishes to conduct an impact assessment. Unlike the everyday examples of instantaneous linear impact, the impact of social programmes is often only evident after some time has lapsed. It is usually the accumulated result of various effects or outcomes that together produce the benefits to the target group. Moreover, it is made up of very different kinds of mutually reinforcing effects and often at different levels. Subsequently, they normally comprise of various potentially overlapping and cumulative non-linear causal relationships.

Complex policy interventions are needed to assess the impact of these highly sophisticated and specialised evaluation designs and methodologies. This contribution compares and assesses three such evaluation designs, namely experimental and quasi-experimental designs, evaluative case studies and outcome monitoring and assessment.
It is argued that the evaluative case study design emphasises contextual explanations that are essentially narrative in nature. Moreover, it is embedded in a distinct theoretical framework and is potentially a much more appropriate and fruitful approach to assessing the impact of complex social interventions than the other designs that were assessed.

Loading full text...

Full text loading...


Article metrics loading...


This is a required field
Please enter a valid email address
Approval was a Success
Invalid data
An Error Occurred
Approval was partially successful, following selected items could not be processed due to error