Introduction: Frontiers of methodological progress in qualitative research.

Blatter, Joachim ; Sager, Fritz

The recent decade has witnessed an unprecedented flurry of methodological reflection on qualitative approaches in Political Science and related disciplines. In the Anglo-Saxon world, much of this methodological reflection has been (framed, at least, as) a reaction to the attempt to establish the epistemology and methodology that underpins quantitative research as the only legitimate one in the Social Sciences (King, Keohane, and Verba 1994). The book Rethinking Social Enquiry: Diverse Tools, Shared Standards, edited by Henry E. Brady and David Collier (2004), contains a broad spectrum of arguments in favor of distinct tools for generating causal inference in small-N studies. At the same time, Alexander George and Andrew Bennett (2005) pointed to the fundamental importance of "causal-process tracing" as a distinct form of inferring causality in case studies. Furthermore, Charles Ragin proposed a configurational/set-theortical alternative to the co-variational/statistical template proposed by King, Keohane and Verba since the end of the 1980s (Ragin 1987), recently summed up in Redesigning Social Inquiry: Fuzzy Sets and beyond (Ragin 2008). As a result, we have now two distinct ways of thinking about causal analysis and the corresponding methodological advice in the Social Sciences. Mahoney and Goertz (2006, forthcoming) argue that these methodological developments reveal that there are two distinct cultures in the Social Sciences - a quantitative and a qualitative culture - with separate approaches to explanation, different conceptions of causation and corresponding different methodologies and research practices, e.g. in respect to case selection and generalization.

Nevertheless, many scholars who perceive their work as "qualitative research" will certainly reject Mahoney and Goertz' implication that an approach to causal analysis that is based on settheory and "configurational thinking" (Ragin 2008: 109-123) represents qualitative research. In the German speaking countries, and probably in continental Europe in general, "qualitative research" is much more associated with hermeneutics and interpretative methods based on a constructivist or conventionalist epistemology. Furthermore, recent text books on qualitative analysis and case study research (Blatter, Janning and Wagemann 2007, Blatter and Haverland forthcoming) emphasize the plurality of epistemological foundations, methodological approaches and specific techniques that are useful and legitimate in the Social Sciences. But emphasizing the plurality of legitimate research approaches is only the first step towards a productive combination of research methods. A further, necessary second step is to strive for an "epistemological middle ground" that rejects any fundamentalist epistemological positions and the corresponding claims of incommensurability (Blatter and Haverland forthcoming).

The articles in this issue of GPS represent the diversity of methodological developments within "qualitative research" and show how it is possible to move towards the "middle ground" between positivist, constructivist and realist epistemologies.

The first two contributions show how far we got in respect to applying the new methodological tools in causal analysis. They address the two major methodological innovations in causal analysis - configurational thinking and causal process tracing - but they do this quite differently. Rihoux, Rezsohazy and Bol (2011) present an extensive review of the existing studies which apply Qualitative Comparative Analysis (QCA) as technique for revealing causal configurations. Julian Junk (2011) presents a paper on the combinations of methods in the form of both triangulation and parallelization.

From a methodological point of view the most interesting (and somehow puzzling) result is the fact that most scholars who apply QCA do this in a way that Rihoux, Rezshohazy and Bol call "inductive" or "soft theoretical." Within their theoretical framework, many scholars actually do not deduce causal configurations from theories but delineate a set of potential causal conditions like a set of "independent" (in both senses: explanatory and autonomous) variables in statistical analysis. Only inductively they end up with revealing causal configurations of individually necessary and jointly sufficient conditions that correspond to a specific result. Quite frankly, we are not sure whether we should interpret the fact that many QCA applications are not really based on configurational thinking as a positive sign which shows how specific methods can escape the confines of epistemological perspectives or as a negative sign for a methodologically unreflective and incoherent state of the art. It is certainly a topic that is worth wile for further methodological discussions.

For QCA the same statement holds as for statistical analysis: Configuration/correlation (alone) is not (proof of) causation! Causal configurations and correlations have to be complemented by theoretically deduced arguments on how and why a causal factor should have a specific effect - or they have to be complemented by causal- process tracing in order to reveal empirically the specific links and connections which lead from cause to effect. Metaphorically speaking: QCA is able to identify one or a few lists of ingredients for producing a specific meal, but it needs theory and/or causal process tracing in order to create a full-fledged recipe! Therefore, it seems no surprise that methodological concept of causal-process tracing has generated the most excitement among those qualitative scholars who strive for explanations. Here, we cannot delve deep into the debate.

Instead, we have included a paper that shows how the debate on causal-process tracing spurs new ways for taking temporality serious in causal analysis and for clarifying the possible ways to "mix" methods. Julian Junk starts his contribution by introducing some distinctions that are necessary for systematizing the confusing debate on "mixed methods": First, the distinction between methods of data generation and methods of data analysis; second, the differentiation between method triangulation and method parallelization. In the main part of his contribution he presents a research project that aims to identify the necessary and sufficient conditions for humanitarian interventions. Briefly, he scrutinizes the theoretical framework of the study, which consists of a detailed chain of logically necessary steps for a humanitarian intervention. In other words, the explanatory framework consists of a specification of causal chains (the temporal succession of preconditions) and causal conjunctions (the temporal co-existence of preconditions). He then shows how different methods are applied for generating the data that is used to pin down the temporal order and development of the proposed preconditions. He concludes by discussing the challenges of both his two types of method triangulation (data generation triangulation and data analysis triangulation) and his three types of method parallelization (multivariate designs, research programs, and causal processes).

The next two contributions illustrate the progress that has been made towards the epistemological middle ground by qualitative researchers who start with ontologies and theories that emphasize the crucial importance of texts and contexts. The first article by Henning Schmidtke and Frank Nullmeier (2011) illustrates that political analysis which is not geared towards explanation plays an important and indispensable role; but is makes also clear that research that tries to reveal "meanings" or "ideational factors" is moving towards the epistemological middle ground by developing methods that straddle the cleavage between quantitative and qualitative research. The second article indicates how qualitative research is able to overcome the cleavage between structuralist and agency-centered research.

Henning Schmidtke and Frank Nullmeier's "political valuation analysis" fills in a void in political analysis that focusses on ideational or communicative factors since valuations - in contrast to arguments, claims, frames or discourses - have not been focused on within the interpretative research agenda. Furthermore, they straddle the qualitative-quantitative dividing line. In line with qualitative reasoning, they thoroughly ground their approach in linguistics, acknowledge the need to reconstruct the manifest content of statements, and reflect on how to translate the statements into the variables of the valuation grammar - and they show how it is possible to end up with quantitative results.

Roland Willner's (2011) contribution points to innovative ways to address the call for "micro-foundations" that has become a boost with the rise of Scientific Realism while at the same time to acknowledge the complexities of the structural contexts in which political decision-making is taking place. Willner presents a heuristic framework for micro-analyses of political organizations focusing on the individual room for maneuvre within organizational structures. He then links his conceptual approach to three different qualitative research designs that he considers fit for micro-political studies, namely qualitative interviewing, ethnographic research, and the documentary method.

With Andreas Balthasar's (2011) paper on policy evaluation we complement this issue on qualitative methods with a kind of research that has become almost fully neglected in Political Science and in the scrutinized methodological debate. Nevertheless, policy evaluation represents certainly one of the major links between the science of politics and the practice of politics. Furthermore, some protagonists of the recent methodological debate have emphasized that their methodological stance is driven by the goal to produce policy-relevant knowledge (Frey and Ledermann 2010; George and Benett 2005: 263-285; Pawson 2006; Sager and Andereggen 2011). Two things are striking when we compare the "critical friend approach" to policy evaluation that Balthasar is proposing with the debate on qualitative ways to draw causal inferences: First, the critical friend approach and causal process tracing emphasize the importance of temporal proximity - the first by emphasizing that evaluations should create feedback during the implementation process in order to allow learning and adaptation, the second by emphasizing the role of temporal proximity as basis for drawing causal inferences. Second, the evaluation designs that Balthasar proposes represent rather classical research designs which try to generate causal conclusions by cross-case comparisons whereas causal-process tracing represents a technique of within-case analysis.

Summing up, we think that this symposium issue of GPS reflects the many directions in which qualitative methods are currently making progress. While the assembled contribution in this issue present exiting highlights of this debate, it is clear that a special issue cannot provide the full picture and that we cannot claim representativity over the field. However, the following articles may contribute to a deeper understanding of the heterogeneity of qualitative approaches and may thus lead to conscience that there is no single qualitative method, but a plethora of approaches that serve both as stand-alone analytical techniques as well as complementary to and in combination with other methods and research approaches.

References

Balthasar, Andreas (2011). Critical Friend Approach: Policy Evaluation Between Methodological Soundness, Practical Relevance, and Transparency of the Evaluation Process, German Policy Studies 7(3):

Blatter, J. and M. Haverland (forthcoming): Designing Case Studies. Explanatory Approaches in Small-N Research. Basingstoke: Palgrave Macmillan.

Blatter, J., F. Janning, and C. Wagemann (2007): Qualitative Politikanalyse. Eine Einfuhrung in Forschungsansatze und Methoden. Wiesbaden: Verlag fur Sozialwissenschaften.

Brady, H. and D. Collier (eds.) (2004): Rethinking Social Inquiry. Diverse Tools, Shared Standards. Lanham: Rowman & Littlefield.

Frey, Kathrin, and Simone Ledermann (2010) (eds.). Evidence-Based Policy. A Concept in Expansion. Symposium Issue of German Policy Studies 6(2).

George, A. and A. Bennett (2005): Case Studies and Theory Development in the Social Sciences. Cambridge: MIT Press.

Junk, Julian (2011). Method Parallelization and Method Triangulation: Method Combinations in the Analysis of Humanitarian Interventions, German Policy Studies 7(3):

King, G., R. O. Keohane, and S. Verba (1994): Designing Social Inquiry. Scientific Inference in Qualitative Research. Princeton: Princeton University Press.

Mahoney, J. and G. Goertz (2006): A Tale of Two Cultures. Contrasting Quantitative and Qualitative Research. Political Analysis 14(3), 227-49.

Mahoney, J. and G. Goertz (forthcoming): A Tale of Two Cultures: Contrasting Qualitative and Quantitative Paradigms. Princeton University Press

Pawson, Ray (2006). Evidence-Based Policy. A Realistic Perspective. London: Sage.

Ragin, C. (1987): The Comparative Method. Moving beyond Qualitative and Quantitative Strategies. Berkeley: University of California Press.

Ragin, C. (2008): Redesigning Social Inquiry. Fuzzy Sets and Beyond. Chicago: University of Chicago Press.

Rihoux, Benoit, Ilona Rezsohazy and Damien Bol (2011). Qualitative Comparative Analysis (QCA) in Public Policy Analysis: an Extensive Review, German Policy Studies 7(3):

Sager, Fritz, and Celine Andereggen (2011). "Dealing with Complex Causality in Realist Synthesis: The Promise of Qualitative Comparative Analysis (QCA) ", American Journal of Evaluation. Early online

Schmidtke, Henning, and Frank Nullmeier (2011). Political valuation analysis and the legitimacy of international organizations. German Policy Studies 7(3):

Willner, Roland (2011) Micro-politics: An Underestimated Field of Qualitative Research in Political Science, German Policy Studies 7(3):

Joachim Blatter University of Lucerne (Switzerland) and Fritz Sager University of Bern (Switzerland)

Blatter, Joachim, Prof. Dr., Email: [email protected] Homepage: http://www.unilu.ch/deu/prof.-dr.-joachim-blatter_254226.html

Sager, Fritz, Prof. Dr., Email: [email protected] Homepage: http://www.kpm.unibe.ch/index.php?bereich=institut&page=mitarbeiter&id=2382&sub=1