This chapter extends the analysis from Chapter 7 to multi-case settings and demonstrate how we can use the approach to undertake mixed-method analysis. We show how, when analyzing multiple cases, we can update our theory from the evidence and then use our updated theory to draw both population- and case-level inferences. While single-case process tracing is entirely theory-informed, mixed-data inference is thus also “data”-informed. We show how the approach can integrate information across any arbitrary mix of data structures, such as “thin” data on causes and outcomes in many cases and “thicker” process evidence on a subset of those cases.

We connect the literature on causal models to qualitative inference strategies used in process tracing. The chapter outlines a procedure for drawing case-level causal inferences from a causal model and within-case evidence. We also show how a key result from the causal-models literature provides a condition for when the observation of a node in a causal model (a “clue”) may be (or certainly will not be) informative, and we extract a set of implications for process-tracing methods.

We apply the causal-model-based approach to process tracing to two major substantive issues in comparative politics: the relationship between inequality and democratization and the relationship between institutions and growth. Drawing on case-level data, we use qualitative restrictions on causal types together with flat priors to draw inferences about a range of causal queries. The applications illustrate the different types of learning that can be gleaned from information on moderators and mediators, as well as the scope for learning from historical data when researchers have informative beliefs about confounding processes.

This chapter shows how we can integrate inferences across models. We provide four examples of situations in which, by combining models, researchers can learn more than they could from any single model. Examples include situations in which researchers seek to integrate inferences from experimental and observational data, learn across settings, or integrate inferences from multiple studies.

This chapter defines causal questions in the language of causal models. We describe major families of causal queries, including case-level queries about causes of effects, case-level attribution, population-level effects, and effects along causal pathways. We show how these can all be defined as queries about the values of nodes in a causal model.

This chapter integrates the notion of “theory” into the causal-model framework that we use in the book. We describe an approach in which theoretical claims are thought of as model justifications within a hierarchy of causal models. The approach has implications for the consistency of inferences across models and for assessing when and how theory is useful for strengthening causal claims.

This chapter argues for the utility of causal models as a framework for choosing research strategies and drawing causal inferences. It provides a roadmap for the rest of the book. The chapter highlights the approach’s payoffs for qualitative analysis, for combining intensive and extensive empirical strategies, and for making research design choices.

This chapter is the first of three in which we investigate how causal models can inform research design. In this chapter, we draw out the implications of the causal-model approach for clue-selection strategies, for figuring out which pieces of evidence are likely to be most informative about a question of interest. We demonstrate procedures for assessing which clues minimize expected posterior variance and how to construct an optimal decision tree for determining a dynamic clue-gathering strategy.

The chapter outlines social media and qualitative research. It describes social media for data collection and different qualitative research approaches to data collection. The chapter describes social media as a phenomenon for research and outlines different levels of social media utilization: individual, work-practice and supra-organizational levels. Vignettes for the different levels are provided and the need for qualitative research concluded.

In collaboration with the HR team of a large IT service provider, this chapter relates to a study of fifty individuals who have been identified as high performers by their employer and the search for indicators and patterns of sustainable high performance.

The research design consisted of initial interviews at a virtual day, attendance of 2.5-day off-site coaching workshops and up to 60-minute follow-up interviews. During the workshop days, 24-hour heart rate variability (HRV) measurements were collected – a well-established biomarker of well-being, strain and recovery. As HRV data are difficult to analyze without contextual information, interviews, focus-group sessions, participatory observation and debriefing interviews were combined in order to contextualize the quantitative measurements and involve the participants in the interpretation and sense-making of the findings.

The methodological goal of this chapter is to demonstrate how orchestrating, improvising and performing a mixed-method study has been essential to validate, augment and complement quantitative data. The study results depend on the ability of the researchers to skilfully and empathetically engage with the interviewees and to engage them as participants in the interpretation of their data and thus as co-producers of meaning.

Recognizing the pervasive influence of modern digital technologies, this chapter argues for the supremacy of strategy work in terms of giving shape and effect to the associated agenda for strategic, organizational and technological change. The chapter focuses on the theory and practice of action research as a Mode 2 approach to knowledge production as managers co-inquire into the practice of strategizing. The discussion speaks directly to the practice of action research in government organizations, of enhancing strategy work and its related outcomes, and the broader outcomes of co-inquiry. The chapter affirms the central role of action research in knowledge production and emphasizes how the practice of action research is itself being transformed by enabling digital technologies during the current COVID-19 pandemic. The contention throughout is that good practice informs research and good research informs practice.