Background: Implementation of multifaceted interventions typically involves many diverse elements working together in interrelated ways, including intervention components, implementation strategies, and features of local context. Given this real-world complexity, implementation researchers may be interested in a new mathematical, cross-case method called Coincidence Analysis (CNA) that has been designed explicitly to support causal inference, answer research questions about combinations of conditions that are minimally necessary or sufficient for an outcome, and identify the possible presence of multiple causal paths to an outcome. CNA can be applied as a standalone method or in conjunction with other approaches and can reveal new empirical findings related to implementation that might otherwise have gone undetected.Methods: We applied CNA to a publicly available dataset from Sweden with county-level data on human papillomavirus (HPV) vaccination campaigns and vaccination uptake in 2012 and 2014 and then compared CNA results to the published regression findings.Results: The original regression analysis found vaccination uptake was positively associated only with the availability of vaccines in schools. CNA produced different findings and uncovered an additional solution path: high vaccination rates were achieved by either (1) offering the vaccine in all schools or (2) a combination of offering the vaccine in some schools and media coverage.Conclusions: CNA offers a new comparative approach for researchers seeking to understand how implementation conditions work together and link to outcomes.

Coincidence Analysis (CNA) is a configurational comparative method of causal data analysis that is related to Qualitative Comparative Analysis (QCA) but, contrary to the latter, is custom-built for analyzing causal structures with multiple outcomes. So far, however, CNA has only been capable of processing dichotomous variables, which greatly limited its scope of applicability. This paper generalizes CNA for multi-value variables as well as continuous variables whose values are interpreted as membership scores in fuzzy sets. This generalization comes with a major adaptation of CNA’s algorithmic protocol, which, in an extended series of benchmark tests, is shown to give CNA an edge over QCA not only with respect to multi-outcome structures but also with respect to the analysis of non-ideal data stemming from single-outcome structures. The inferential power of multi-value and fuzzy-set CNA is made available to end users in the newest version of the R package cna.

In recent years, proponents of configurational comparative methods (CCMs) have advanced various dimensions of robustness as instrumental to model selection. But these robustness considerations have not led to computable robustness measures, and they have typically been applied to the analysis of real-life data with unknown underlying causal structures, rendering it impossible to determine exactly how they influence the correctness of selected models. This article develops a computable criterion of fit-robustness, which quantifies the degree to which a CCM model agrees with other models inferred from the same data under systematically varied threshold settings of fit parameters. Based on two extended series of inverse search trials on data simulated from known causal structures, the article moreover provides a precise assessment of the degree to which fit-robustness scoring is conducive to finding a correct causal model and how it compares to other approaches of model selection.

Some methodologists take the search target of Qualitative Comparative Analysis (QCA) to be causal INUS-conditions, others contend that QCA should instead be used to search for some form of sufficiency that is more substantive than mere Boolean sufficiency. While the notion of an INUS-condition has a long and uncontroversial definitional history, Adrian Dusa, in a recent paper, is the first to explicitly define a notion of substantive sufficiency, which he labels robust sufficiency. Dusa’s definition, however, is vacuous in real-life research contexts. As an alternative, the first part of this paper non-vacuously defines robust sufficiency and supplies a corresponding notion of minimality, which—I argue—captures Dusa’s conceptual intentions. In the second part, I then report and discuss the results of a series of simulation experiments benchmarking the performance of the different QCA solution types in recovering robust sufficiency and minimality.

Various analytical tools originally developed for theories of mechanistic explanation have recently been imported into the ongoing debate on the hypothesis of (extracerebrally) extended cognition (HEC). One such tool that appears particularly relevant to that debate is Craver’s mutual manipulability account of constitution (MM), most of all because it promises to settle the debate on experimental grounds. This paper investigates whether it is possible to deliver on that promise. We first find that, far from grounding an experimental evaluation of HEC, MM is conceptually incompatible with both internalist and externalist accounts of cognition. Next, we propose a suitable modification of MM, MM*, but it turns out that MM* presupposes rather than produces clarity on the extension of cognition. Moreover, subject to MM* the inference to constitution is radically empirically underdetermined. Finally, we argue that our results can be generalized and conclude that, for principled reasons, it is impossible to experimentally determine whether cognitive processes have extracerebral constituents. Determining the extension of cognition is an inherently pragmatic matter.

Consistency and coverage are two core parameters of model fit used by configurational comparative methods (CCMs) of causal inference. Among causal models that perform equally well in other respects (e.g., robustness or compliance with background theories), those with higher consistency and coverage are typically considered preferable. Finding the optimally obtainable consistency and coverage scores for data $\text{δ}$ , so far, is a matter of repeatedly applying CCMs to $\text{δ}$ while varying threshold settings. This article introduces a procedure called ConCovOpt that calculates, prior to actual CCM analyses, the consistency and coverage scores that can optimally be obtained by models inferred from $\text{δ}$ . Moreover, we show how models reaching optimal scores can be methodically built in case of crisp-set and multi-value data. ConCovOpt is a tool, not for blindly maximizing model fit, but for rendering transparent the space of viable models at optimal fit scores in order to facilitate informed model selection—which, as we demonstrate by various data examples, may have substantive modeling implications.