Joseph A. Maxwell

CAUSAL EXPLANATION, QUALITATIVE RESEARCH, AND SCIENTIFIC INQUIRY IN EDUCATION

A National Research Council report, Scientific Research in Education, has elicited considerable criticism from the education research community, but this criticism has not focused on a key assumption of the report—its Humean, regularity conception of causality. It is argued that this conception, which also underlies other arguments for “scientifically-based research,” is narrow and philosophically outdated, and leads to a misrepresentation of the nature and value of qualitative research for causal explanation. An alternative, realist approach to causality is presented that supports the scientific legitimacy of using qualitative research for causal investigation, reframes the arguments for experimental methods in educational research, and can support a more productive collaboration between qualitative and quantitative researchers.

Using Qualitative Methods for Causal Explanation

The view that qualitative research methods can be used to identify causal relationships and develop causal explanations is now accepted by a significant number of both qualitative and quantitative researchers. However, this view is still controversial, and a comprehensive justification for this position has never been presented. This article presents such a justification, addressing both recent philosophical developments that support this position and the actual research strategies that qualitative researchers can use in causal investigations.

Reemergent Scientism, Postmodernism, and Dialogue Across Differences

The reemergence of a narrowly defined “scientifically based research” that marginalizes qualitative approaches represents a major threat to qualitative research. A postmodern perspective not only challenges this essentialist definition of “science” but also critiques tendencies of qualitative researchers to essentialize proponents of “science”. These arguments raise the issue of dialogue across the differences between researchers working in different paradigms. Such dialogue can promote a better understanding of the value of qualitative research; it can also sensitize qualitative researchers to the importance of validity concerns and alternative interpretations in their work. Addressing the latter issue strengthens the argument that at least some qualitative research can be fully “scientific” without giving up the essential characteristics of qualitative inquiry.

The Importance of Qualitative Research for Causal Explanation in Education

The concept of causation has long been controversial in qualitative research, and many qualitative researchers have rejected causal explanation as incompatible with an interpretivist or constructivist approach. This rejection conflates causation with the positivist theory of causation, and ignores an alternative understanding of causation, variously known as a “generative,” “process,” or “realist” approach, which is influential in philosophy and is becoming widespread in the social sciences and history. This alternative approach to causation is compatible with the practice and “theory-in-use” of many qualitative researchers and enables qualitative researchers to credibly make and support causal claims. It also enables them to defend their work against the dismissal of qualitative methods by proponents of “science-based” research, since the latter assumes a positivist understanding of causation. Adequate causal explanations in the social sciences depend on the in-depth understanding of meanings, contexts, and processes that qualitative research can provide.

Expanding the History and Range of Mixed Methods Research

Presentations of the history and range of mixed methods research presented in textbooks, handbooks, and journal articles have typically ignored a great deal of earlier and contemporary research that integrated qualitative and quantitative approaches, but did not explicitly identify itself as “mixed methods.” This article reviews earlier research, in both the natural and social sciences, that clearly integrated qualitative and quantitative approaches and methods, and discusses some contemporary research traditions that use such integration without labeling this “mixed methods.” Important implications of these studies and traditions for the conceptualization and conduct of mixed methods research are discussed.

A Study of Non-Volunteer Faculty in a Problem-based Curriculum.

No abstract available.

Causation in educational research

This book addresses an important and currently contentious issue in educational research, and the author’s goal is to provide guidance for educational researchers in making causal inferences. He draws on an extensive range of work on causality, both philosophical and methodological; challenges the current dominance of quantitative methods for inferring causality; and argues for the value of qualitative methods for this purpose. Before proceeding, I want to acknowledge my own perspective. I have written a fair amount on this topic (Maxwell 2004a, 2004b, 2008; Maxwell and Mittapalli 2010), and have strong opinions on some of the issues that Morrison discusses. My review is obviously influenced by these views; I have tried to indicate where different views exist. The main strengths of this book, in my opinion, are as follows:

Book Review: Bergman, M. M. (Ed.). (2008). Advances in Mixed Method Research. Thousand Oaks, CA: SAGE:

Manfred Max Bergman has edited an important work for mixed method researchers, one that can advance the field in several ways. The book consists of 11 chapters, plus an introduction, ‘‘Whither mixed methods?’’ by Bergman. The chapters are grouped into two sections, one on the theory of mixed method design and the other on applications in mixed method design. However, I see the chapters as falling into two distinct types that do not completely correspond to this division. First, there are descriptions of ‘‘advances’’ in the usual sense—new approaches, developments, or applications in mixed method design or methods (the chapters by Creswell et al., Tashakkori and Teddlie, De Leeuw and Hox, Widmer et al., and Niglas et al.). Second, there are contributions that are mainly critiques of some aspect of the current state of mixed method research (the chapters by Bergman, Hammersley, Fielding, Brannen, Bryman, and Pawson). Because there has been little of this sort of critique in the mixed method literature, I will devote more of my review to the latter chapters. In the first group of chapters, Creswell, Plano Clark, and Garrett extend their previous typology of mixed method designs by further explicating how each of these types should be conducted and how problems in such studies can be addressed. Tashakkori and Teddlie review the literature on inference quality in qualitative and quantitative research and present a synthesis of typologies of quality issues and criteria. Researchers who are drawn to such typologies will find these contributions extremely useful, and even those who are not can gain valuable ideas. De Leuuw and Hox discuss what they call ‘‘mixed mode’’ survey research: the use of different quantitative strategies in a survey. They argue that many of the issues that arise in this approach are similar to those in mixed method research, and understanding how mixed mode researchers deal with these issues can be valuable to mixed method researchers. Widmer, Hirschi, Serduelt, and Voegeli describe APES (the actor-process-event scheme), a strategy for converting qualitative case study data into quantitative network displays. Finally, Niglas, Kaipainen, and Kippar present ESO (Exploratory Soft Ontology), a software tool for multiperspective knowledge construction. Fully understanding the latter two contributions requires more quantitative sophistication than I possess. In the second group of contributions, all the authors see the integration of quantitative and qualitative methods and data as more problematic than is often assumed. Bergman, in ‘‘The Straw Men of the Qualitative-Quantitative Divide and Their Influence on Mixed Method Research,’’ claims that mixed method research, in delineating the relative strengths of qualitative and quantitative methods, is simply perpetuating highly questionable assumptions about the paradigmatic unity and identity of each approach. (Bergman could have noted Hammersley’s earlier work (1992) on this issue.) He argues that we need to get beyond general statements of the differences between the two approaches and look at how the latter are actually combined in specific studies. Although I strongly agree with the latter point, I wish Bergman had addressed the distinction between variance and process approaches to explanation, which I see as a key difference between the two types of research (Maxwell, 2004). Journal of Mixed Methods Research Volume 3 Number 4 October 2009 411-413