Mixed Methods - Epistemological and Methodological Issues 1 |
|
| Chair | Dr Leila Akremi (Technical University Berlin, Germany ) |
| Coordinator 1 | Dr Susanne Vogl (University of Vienna, Austria) |
Social sciences are often considered as a multiparadigmatic research field, wherein at least two paradigms can be identified. The reference here is to the so-called “qualitative” and “quantitative” paradigms. Different meanings and uses of the term paradigm can be found within the social sciences and Morgan [1] identifies four main connotations, but the focus here will be on paradigms as epistemological stances – hence, using Kuhn’s [2] original notion.
Is the field of mixed methods research (MMR) moving toward a so-called “third paradigm” or is it rather going “beyond paradigms”? We intend here to question whether MMR can be recognized as a “third paradigm”, with the implicit acceptance of the “qualitative” and “quantitative” strands as paradigms. The main objective of this contribution is thus to explore the epistemological approaches adopted within the field of MMR in social sciences.
While current debates largely focus on the methodological level, the real struggles happen at the ontological and epistemological ones. What should indeed be asked by social scientists in these regards, before going to the epistemological level, has rather to do with ontology. If it is stated only one ontology for social phenomena, integration of the approaches is feasible and in some cases – depending on the research question – even desirable. From this point of view, it can be affirmed that qualitative or quantitative methods are more likely styles of research, rather than distinct paradigms. Thus, neither qualitative nor quantitative research is superior to the other, while a researcher is only facing a phenomenon and the methodological challenges that are related to the process of knowing and understanding it. To adopt this last position would mean to embrace an aparadigmatic viewpoint.
Moreover, questioning how the MMR community deals with epistemology, different stances and standpoints can be identified. In general, a philosophical approach which is commonly related to MMR is pragmatism, in particular in the version of Dewey’s strumentalism [3]. However, pragmatism may relate to two different logics. It may help the construction of a “third paradigm”, wherein from the epistemological stances coming from a specific philosophical tradition it is possible to deduce also methodological standards and guidelines. The second chance is that of an “everyday pragmatism”, wherein an argument is made for the utility of research means for research ends, moving thus “beyond paradigms”. We ask then within this contribution which are the methodological consequences of this “practical” choose of the most adequate method referring to the contingent situation.
Then, other epistemological positions may be found as well, such as the endorsement of a “qualitative” or a “quantitative” strand or even the lack of declaration of any philosophical stance. Hence, we question also the possible methodological consequences of these last positions.
[1] Morgan D., (2007), Paradigms lost and pragmatism regained: Methodological implications of combining qualitative and quantitative methods, in “Journal of Mixed Methods Research”, 1:1.
[2] Kuhn T., (1970), “The structure of scientific revolutions”, Chicago: University of Chicago Press.
[3] Dewey J., (1938), “Logic: The theory of inquiry", Rinehart&Winston.
This paper addresses key issues in the debate about the epistemological foundations of mixed-methods-research: It is often assumed that research methods have to be or are based on differ-ent epistemological assumptions about the nature of reality; and that different research meth-ods (namely qualitative and quantitative approaches) refer to diverging, perhaps even incom-patible philosophical word views (e.g. “positivism” or “constructivism”). My paper attempts to discuss some crucial weaknesses in these arguments, by showing that
1. the term “paradigm” is (for good reasons) a highly contested concept in the philosophy of science which cannot easily be applied to research methods,
2. epistemological paradigms which are usually referred to in the debate either hardly existed as consistent bodies of epistemological propositions made by a certain school; or often were ephemeral and transient phenomena in the history of philosophical thought; or did not entail or lead to many of the claims which are attributed to them;
3. that (paradigm-bound) epistemological assertions and arguments put forward in the methodological debate frequently lack reference to research practice in terms of used data, methods of analysis or inferences drawn from data and analysis results.
There is a serious danger that the reference to epistemological paradigms is misused for legit-imatory purposes, by allowing researchers to dismiss important research questions, practices and techniques which they feel not comfortable with. In my paper, I will propose an alterna-tive to a “foundationalist approach” re. the use of epistemology in research: instead of a seek-ing a solid epistemological fundament by clinging to a certain philosophical tradition, re-searchers may be better advised to draw on epistemological arguments from different school of thoughts in a flexible way.
Many different quality standards, special analyzing tools and different solutions of methodological problems have been developed in the long tradition of quantitative and qualitative research. Because of these well-reasoned differences in epistemological and methodological aspects, there are many social scientists that are skeptical about integrating both approaches. A short look on the wide landscape of textbooks for methods in social sciences is enough to understand that quantitative approaches often ignore qualitative analysis strategies, while qualitative textbooks offensively defend their approach against quantitative aspects. To speak with Hageman-White one could say, that qualitative Research is facing the dramatic challenge to be “Not-Not-Qualitative”. To mix quantitative and qualitative research questions is e.g. seen as a “deadly sin” (Przyborski/Wohlrab-Sahr 2014). However, textbooks about mixed methods often focus on the aspect of research design, neglecting these fundamental differences in regard to epistemological and methodological concepts (vgl. Bryman 2008). Thus, the integration of both approaches is either reduced to organizational problems or dominated by a perspective in which one approach is reduced as an assistant for the other one. However, it is quite obvious that urgent methodological problems (e.g. how to generalize results, how to sample, how to find causal relations, and so on) and their specific solutions cannot simply be mixed without to destroy the foundation of their scientific justification. In this paper I want to argue that this is neither possible nor necessary. To mix methods is not a methodological, but a theoretical problem. Thus, mixed methods designs should not focus too much on technical aspects, but rather on the theoretical justification of the planned research design. Still many qualitative researchers think that their research is supposed to be free of theoretical assumptions in the beginning. This inductivistic self-missunderstanding restrains the possibility to combine qualitative and quantitative research methods. In the paper I want to show that it is fundamentally important to clarify theoretical assumptions in the very beginning of a mixed methods design to make well justified decisions in which case qualitative and quantitative methods are useful tools for a successful research. The big challenge of mixed methods is not to find new methodological principles, but rather to find a common understanding of theory.
This paper explores the potential of Andrew Abbott’s methodological work as a basis for conceptualizing mixed methods and multimethod research (MMMR). Surprisingly, although Abbott’s ideas concerning method-integration are very similar to influential approaches to MMMR – especially those highlighting a “dialectical” (Greene 2007) or “complementary” (Kelle 2008) relation between different methods – this parallel has hardly received any serious attention in the MMMR literature (a notable exception: Maxwell/Mittapali 2010).
The papers main focus will be on Abbott’s notion of “fractals”. It implies that the distinctions often used to separate methodological or epistemological “paradigms” (e.g. realism vs. constructionism, positivism vs. interpretivism) are actually reciprocal in nature, thus reappearing within the supposedly homogenous and distinctive camps (e.g. realist approaches incorporating aspects of social construction in their arguments).
From this fractal characteristic of “basic debates” in methodology follows a principal complementarity of different “methodological traditions” (e.g. standard causal analysis, ethnography, small-N comparisons). According to Abbott, these methodological traditions can be seen as engaging in cycles of mutual criticism – similar to notions of triangulation in MMMR.
But Abbott does not only provide a methodological account of the interrelation of existing research approaches. He also stresses the possibility of applying methodological distinctions as “fractal heuristics”: Basic methodological debates may be used as conceptual tools to systematically develop innovative research perspectives and designs.
The paper will highlight the following three aspects of Abbott's approach as applied to MMMR:
(1) It provides a methodological framework for method integration which is not centered on the qualitative/quantitative methods distinction but rather places it within a broader methodological context.
(2) The concepts of “fractals” and “cycles of criticism” highlight the non-paradigmatic nature of methodological traditions, thus fostering the convergence or reorganization of established “paradigms”.
(3) The concept of “fractal heuristics” outlines possible methods for creatively developing MMMR designs ‘outside the box’ of qualitative and quantitative research traditions.
The aim of the paper is to reflect on my methodological stance in the context of recent changes in British social sciences, where quantitative teaching and research methods have received more institutional support recently (i.e. via the Q-Step Programme). As an academic applying different methods, I have been constantly expected to position myself as either ‘mixed method’ or ‘quantitative’ researcher. This paper uses an autobiographical inquiry to reflect on my own methodological standing. Such method allows fleshing out interconnections between my methodological identity and wider social structures, including the methodological landscape where I develop as an academic and a researcher.
The landscape shaping my methodological identity has been shifting and I argue that there have been key recent developments. 1) First is the mixed-method movement which emerged in late 1990s, after scientists from the both ‘qualitative’ and ‘quantitative’ sides started engaging in a dialogue on building more applied research together and as well as more research of this type started being commissioned by governmental institutions. 2) At the same time a wider debate about methodological crisis of social sciences in the UK emerged, and the need for developing more comprehensive and creative research, mixing techniques traditionally recognised as ‘quantitative’ or ‘qualitative’. 3) Both these strands criticised new quantitative methods of inquiry and and new sources of transactional and online data (‘Big Data’), recognising the need for a novel epistemological framing and reflexivity in doing quantitative social science.
The paper concludes with critical reflections on the epistemological role of methods in knowledge production and the advantages of methodological hesitation and reflexivity, which can be facilitated by mixing methods.