Background: Community Based Rehabilitation (CBR) is the strategy promoted by the World Health Organization (WHO) and other United Nations (UN) agencies as an effective way to improve the lives and wellbeing of people with disabilities in underserved regions. During the last decade CBR has undergone major reconceptualization, and is now a multi-sectorial approach, as reflected in the new CBR guidelines. Evaluation of Community-based Rehabilitation (CBR) is considered important for developing good practice. However, evaluations remain scarce and as a consequence very little is known about how CBR benefits persons with disabilities and their families. Consensus is lacking about appropriate evaluation methods in CBR. Leading international frameworks such as the United Nations Convention on the Rights of Persons with Disabilities (CRPD) and the WHO CBR Guidelines have participation as one of the core principles in their human rights based approach to disability, including participation in programme evaluation. The WHO CBR Guidelines strongly recommend the application of participatory evaluation (PE) approaches in CBR. However, while there are many models of PE in mainstream development, it is unclear which may be appropriate for use in CBR. The aim of this research is to identify, field test, adapt and assess an existing model of Participatory Evaluation (PE) in a real world environment. This thesis, rather than researching the impact of CBR on people with disabilities, focuses on the evaluation process itself and the variables that affect changes in stakeholders thinking and behavior as a result of engagement in the evaluation. Methodology There were two research components: 1. Selection of PE model to be adapted to CBR: Three steps were taken to provide background for an expert group to select one model for field-testing: An online survey of current evaluation capacities and practices within CBR programmes internationally; A systematic review of PE models used in international development; A Delphi study with CBR experts to derive criteria for good PE models for CBR. The expert group used the research findings and selected Outcome Mapping (OM) as PE model to be implemented and field-tested in a CBR programme in Jamaica. 2. Field testing of the PE model in a Jamaican CBR programme: This research component consisted of three main elements: The implementation and adaptation of PE (OM) in a Jamaican CBR Programme; Interviews and focus groups collecting narratives about the evaluation process from stakeholders were undertaken to explore the usability of the adapted PE model in this programme. Changes in “process use”, i.e. how the stakeholders in the evaluation learned from and acted upon their involvement in the PE processes, were explored; The participatory development of a framework that participants felt could guide PE in CBR, one that can be locally adapted to different situations. Conclusion: The evaluation participants felt there were significant limitations of the OM approach in their setting and therefore proposed a substantially modified model. They favored a more fluid PE framework, which was flexible, adaptive and iterative, rather than a rigid approach, and one that focused on creating a safe space for sharing, learning and taking action. The thesis concludes with a call for more critical and bottom-up approaches of evaluation that move away from control-oriented approaches towards a more experimental and adaptive problem and process-orientated mindset of evaluative thinking.