» Home
Method
Projects that are part of Institutions for Fragile States usually advance to completion in three phases.
Knowledge Inventory & Issue Identification
The program uses various methods to assemble information essential to the development of projects and to refinement of research questions. Because of the newness of the subject matter, the information scholars often seek is initially unavailable, locked in the experience of practitioners who have no time to write or reflect. Thus, in many, but not all instances, a question begins its life as the subject of a second-year MPA policy workshop. Expert conferences provide a second avenue for developing our information base and for refining questions is to convene an expert conference. The WWS-Bobst initiative conducted three such workshops in 2006-7, on civil service reform, constitution writing, and systems of representation.
A new program for practitioner-affiliates offers a third way to develop our knowledge base. Each year, two practitioners join us at Princeton for six weeks to build on the materials they have assembled and to reflect upon their experiences.
Oral Histories/Ethnographic Research
In post-conflict settings and fragile states, we often lack the information either to generate sensible propositions about policy consequences or to test these. Part of the problem stems from the absence of detailed descriptions of the policies and institutions put in place. By definition, these interventions are new and the people who know the most about their design are practitioners who have much to do. A first step in trying to determine “what works” is to interview key participants about the choices they made, the obstacles they encountered, and program performance to date. Good analysis rests in part on good ethnographic investigation.
The program’s oral history project supports the development of institutional ethnographies both as a precursor to further analysis and as a source of practically-relevant information and pedagogical case studies.
Natural Experiments
It is often difficult to distill reliable generalizations from ethnographic research, including oral histories, because we can’t control for all the factors that are likely to impinge on outcomes. This problem is particularly acute in post-conflict settings where it if difficult to draw clear inferences about institutional performance because of the volatility of the contexts, the likelihood that conjunctions of circumstances and interactions among conditions matter, and the fact that most institutions and policies have not been in place very long. However, in some instances it may still be possible to carry out analysis of “natural experiments”—especially when an institution is put in place in some communities but not others or when we can assess before/after changes. This kind of research is important because it can tell us more about how much we should depend on the observations offered by the practitioners on the ground.
In those areas in which we can identify “natural experiments,” the program assembles a team of graduate student and faculty researchers to work in tandem with counterparts from host countries to carry out micro-statistical studies, including before-after analyses of changes in institutional design or within-country comparisons of variations in management practices or policies. The program follows the protocols established for equivalent projects by our Oxford partners. Teams report to the larger research group on a semi-annual basis, either at Princeton or at Oxford.
Utumishi House (Public Service Ministry), Dar es Salaam, Tanzania (A. Schalkwyk, 2008)