1.3 Research Methodology
Qualitative Case Study Method
In order to investigate and empirically analyze these theoretical considerations, I undertake qualitative research along the lines of a “structured, focused comparison” (George/Bennett 2005; Gerring 2007) of a controllable number of case studies. This methodology was chosen, because, first, the study focuses on identifying processes and patterns, making such an approach appropriate (Ragin 1987; Cramer 2006:136). Second, many of the potentially decisive mechanisms entailed in state-making and state-breaking are hard to grasp without detailed investigation, which rules out statistical methods in favor of a small-N comparative analysis (Acemoglu et al. 2003:84). Third, the decision to investigate the role of war for state trajectories entails a small-N study, as it has convincingly been argued that “[m]ost theories of war are best tested by case-study methods” (van Evera 1997:30). Last, but not least, the decision to apply this approach was also based on the fact that while the rare process of ‘autonomous recovery’ as allegedly observed in Somaliland is not fit for statistical testing, it holds great potential for qualitative analysis and theory development.
A central advantage of the case study approach is that it does not run the risk of ‘conceptual stretching’ (Sartori 1970, 1984), a problem frequently encountered by statistical studies that subject quite dissimilar cases to ‘one-size-fits-all’ analytical frameworks. Small-N studies allow for high levels of internal (or conceptual) validity in that they provide the researcher with the opportunity to identify and assess those indicators that best capture the underlying theoretical concepts. They also invite conceptual refinement largely characterized by a higher level of validity – however, over a more limited number of observations as compared to large-N studies. Moreover, in comparison to statistical studies, case studies dispose of the advantage that they are superior in accommodating multifaceted causal relations and equifinality (von Bertalanffy 1968). Yet, in order to leverage these advantages, a number of issues need to be taken into consideration when choosing cases.
According to orthodox procedure, the “fundamental problem of inference” (King et al. 1994:91) should be counteracted by ensuring variation in the independent variable (ibid.:140f.). Such a ‘controlled comparison’, however, is largely incompatible with social science research, for which case-selection on the dependent variable is a legitimate alternative as long as sufficient variation in the values of the dependent variable is ensured (Ragin 2004:7).66 Given the vastly diverging state trajectories the Somali territories have experienced, this condition is met. Apart from sufficient variation on the dependent variable, cases have to fulfill the demands of a ‘most similar research design’ (Przeworski/Teune 1970:32f.) in order to not only facilitate the identification of the particular processes and patterns, but also to allow for their identification in the first place. The Somali cases clearly fulfill this precondition. Taking these aspects into consideration valuable insights can be gained from research that is rooted in “more pared down but detailed sets of comparisons and contrasts” (Cramer 2006:136).
The research presented in this thesis is comparative in time, comparing past processes of state-making in Somalia with the more recent one experienced by Somaliland. Such longitudinal studies ensure a limited variation of background conditions, as many variables can be held constant, thereby facilitating the identification of processes and patterns responsible for diverging state-making outcomes. Moreover, the project is also comparative on an intra-case basis, as it compares processes of state-making and state-breaking amongst different geographical polities within the territorial confines of the former Republic of Somalia (a national comparative study). The division of the case into ‘sub-cases’ or ‘subunits’ along chronological and geographical lines, allows an increase in the number of observations, by “making many observations from few” (King et al. 1994:217ff.; Lijphart 1971), thereby confronting the problem of under determinacy of research design.
[su_button url=”https://saxafimedia.com/state-making-in-somalia-and-somaliland/10/” style=”soft” size=”12″ wide=”yes” center=”yes” text_shadow=”0px 0px 0px #FFFFFF” rel=”lightbox”]CONTINUE READING ON THE NEXT PAGE >[/su_button]
