In Finding Pathways: Mixed-Method Research for Studying Causal Mechanisms, Weller and Barnes seek to explain “how the small-N component of multi-method research can meaningfully contribute and add value to the study of causal mechanisms” (quote from blurb). The book contains nine chapters, including the introduction and concluding chapter. The second chapter lays a foundation by discussing the basics and rationale of multimethod research (MMR). Chapters 3 and 4 provide a very thorough introduction of the case selection procedure that is proposed in the book and a comparison of various case selection approaches. Chapter 5 continues with MMR on non-linear relationships, followed by chapter 6 on using matching techniques for comparative case selection. Chapter 7 turns the tables by closing with a discussion of case selection that elaborates on the benefits of running a regression analysis after the case study. Chapter 8 summarizes the debate and takes a look forward to future MMR studies.
Finding Pathways deals with a salient and widely discussed topic in MMR. It is accessibly written and comes with all replication data online (Stata do-files). This is a great benefit for users who seek guidance on how to implement their own MMR. Substantively, the book raises several interesting points that should guide the future debate about case selection and process tracing in MMR. It extends the perspective to the standard OLS model in the literature and considers non-linear relationships and matching techniques (see also Nielsen’s recently published article). Conceptually, it is also welcomed that they systemize potential settings of mechanism that one might discern via process tracing which goes beyond the usual consideration of a single causal chain.
These positive points of Finding Pathways are somewhat offset by several problems as to the book’s content and structure. First, there are many redundancies pertaining to the presentation of Weller and Barnes’ preferred case selection strategy. It is described in detail in chapter 2 and then again several times in the following chapters combined with the presentation of specific case selection strategies. The repetitious discussion of the general guidelines has little added value and creates the impression that the key points could have been delivered to the reader in a more succinct way.
Second, Finding Pathways does not establish a satisfying link between the conceptual apparatus of the quantitative mediation literature and the qualitative process tracing literature. In particular, chapters 8 and 9 indicate that Weller and Barnes’ own background is quantitative. This is not a problem per se and I believe there is potential for linking research on mediation analysis with qualitative process tracing. However, the link is not straightforward because many issues would have to be sorted out first. Probably, the most important point concerns the definition of a mechanism. In chapter 8, the discussion turns to the “average marginal effect of a mechanism”, i.e., an idea that would make many qualitative researchers openly weep because mechanism and effects are understood to be fundamentally different concepts.
Third, the book offers many interesting arguments, but it should have linked to the existing literature more fully and developed some thoughts further. With respect to the multi-method literature, it is striking that Lieberman’s groundbreaking discussion of nested analysis is not cited once in the book. Weller and Barnes prefer the idea of pathway cases as introduced by Seawright and Gerring and slightly modify it in one respect. The pathway case does not figure in nested analysis, highlighting that it would have been necessary to explain in detail why the pathway case is superior to the strategy laid out in nested analysis that is, much as the pathway case, in my understanding a type of what Weller and Barnes call “residual-based case selection”.
Another example of insufficient anchorage in the existing literature can be found in the chapter on matching. The idea of using matching techniques is very useful, but in the way it is introduced in Finding Pathways, it is only a formalization of traditional case comparisons discussed and used for decades in the qualitative domain. Seminal contributions in this domain are completely ignored in this chapter. Matching-based case selection therefore is not a genuine innovation, but one step further in a decades-old development.
Despite these issues, Finding Pathways is a valuable addition to the multimethod literature, offering many valuable insights; everyone who is implementing and working on MMR should make use of this book.