Rapid assessment and response: sound methodology for producing timely responses

Abstract

McKeganey concludes his Commentary on Rapid Assessment and Response (RAR) with what I believe is a valid call for rigorous and independent evaluations of the results of rapid assessment (McKeganey, 2000). Evaluation, however, is also needed for other types of action research. He joins the ranks of a small group of critical observers of rapid assessment who note the enthusiasm of its proponents, their general lack of perspective, and their unrealistic expectations as to what it can achieve. McKeganey’s commentary on RAR, however, is not consistent with RAR as discussed by Stimson and colleagues (Rhodes et al., 1999; Stimson et al., 1999) and in the RAR Guides produced by the World Health Organization (WHO, 1998; WHO/UNAIDS, 1998; WHO/UNICEF, 1998). His commentary on rapid assessment is also not consistent with my experience with rapid assessment or rapid appraisal during the last two decades (Beebe, 1995, 2000). Therefore, I do not believe many of his specific issues with RAR are appropriate. McKeganey appears to have concluded that there is no significant difference in the methods employed by RAR and traditional social research. According to him the only real difference is RAR’s reliance on less information collected more quickly. When he refers to a singular ‘‘assessor’’ implementing RAR he misses what may be the most significant difference — teams and not individuals implement RARs and these teams should include ‘‘insiders’’. The WHO Guides (1998) consistently refer to RAR teams and include explicit reference to building the RAR team. Rapid assessment substitutes intensive team interaction during the collection and analysis of data for the prolonged fieldwork normally associated with social research. Intensive teamwork is especially critical for data collection based on triangulation and during the iterative process of data analysis and additional data collection. Team interaction among team members, including ‘‘insiders’’ from the group under investigation, can increase the power of triangulation exponentially. The process benefits from the different