Spatial Causality: A Systematic Review on Spatial Causal Inference

The growing interest in causal inference in recent years has led to new causal inference methodologies and their applications across disciplines and research domains. Yet, studies on spatial causal inference are still rare. Causal inference on spatial processes is faced with additional challenges, such as spatial dependency, spatial heterogeneity, and spatial effects. These challenges can lead to spurious results and subsequently, incorrect interpretations of the outcomes of causal analyses. Recognizing the growing importance of causal inference in the spatial domain, we conduct a systematic literature review on spatial causal inference based on a formal concept mapping. To identify how to assess and control for the adverse effects of spatial influences, we assess publications relevant to spatial causal inference based on criteria relating to application discipline, methods used, and techniques applied for managing issues related to spatial processes. We thus present a snapshot of state of the art in spatial causal inference and identify methodological gaps, weaknesses and challenges of current spatial inference studies, along with opportunities for future research.     

Bridging Positivist and Interpretivist Approaches to Qualitative Methods

The usual juxtaposition of qualitative research against quantitative research makes it easy to miss the fact that qualitative research itself encompasses at least two traditions: positivist and interpretivist. Positivist work seeks to identify qualitative data with propositions that can then be tested or identified in other cases, while interpretive work seeks to combine those data into systems of belief whose manifestations are specific to a case. In this paper, I argue that discovering causal relationships is the province of positivist research, while discovering causal mechanisms is the province of interpretivists. I explain why absolutist claims for one or the other approach are mistaken, and argue that the combination of both makes more sense. Finally, I offer suggestions for combinations of positivist and interpretive work, both at the level of thought experiment and in actual data collection and analysis. Throughout, I draw my examples from recent studies of poverty, a field in which a small but distinguished tradition of qualitative studies of the poor has been joined by a growing body of both positivist and interpretive work.     

Matched wake analysis: Finding causal relationships in spatiotemporal event data

This paper introduces a new method for finding causal relationships in spatiotemporal event data with potential applications in conflict research, criminology, and epidemiology. The method analyzes how different types of interventions affect subsequent levels of reactive events. Sliding spatiotemporal windows and statistical matching are used for robust and clean causal inference. Thereby, two well-described empirical problems in establishing causal relationships in event data analysis are resolved: the modifiable areal unit problem and selection bias. The paper presents the method formally and demonstrates its effectiveness in Monte Carlo simulations and an empirical example by showing how instances of civilian assistance to US forces changed in response to indiscriminate insurgent violence in Iraq.     

One Model Does Not Fit All: The Varied Politics of State Immigrant Policies, 2005–16

This article addresses conceptual and measurement challenges that complicate the study of state immigrant policies. First, given the multiple facets of immigrant‐related policy, policy‐specific effects may be obscured by highly aggregated outcomes variables. Second, variables of interest often capture both time‐varying and time‐invariant effects, potentially producing coefficients that are uninterpretable averages of both processes. This article presents a research design that addresses both of these obstacles and applies it to an original dataset of both integrative and punitive policies adopted over the period 2005–16. The findings suggest that the causal roles of growing immigrant populations, partisanship, and wealth vary across different clusters of immigrant policies and that average, cross‐state effects often differ from within‐state effects. Future research would do well to clearly link theoretical expectations to specific types of policy outcomes and test hypotheses over both integrative and restrictive outcomes.     

Transcending the “Obnoxious Spectator”: a case for processual pluralism in ethnoarchaeology

Several commentators have argued that ethnoarchaeology will only become a productive part of archaeological research once both archaeology and ethnoarchaeology are unified by the axioms of a single theory of behavior. Through an examination of the different roles that ethnoarchaeological research has adopted, I demonstrate that ethnoarchaeology is already theoretically unified by a general concern with analogy. I argue that the problems that many commentators have recognized with ethnoarchaeology’s apparent eclecticism arise from an over-reliance on “core universals” by both processual and postprocessual researchers. Instead of implementing a single unified theory of behavior, I suggest that ethnoarchaeologists should adopt a pluralistic orientation that is sensitive to the contextual applicability of specific causal processes.     

Apparent and actual disease landscapes. Some reflections on the geographical definition of health and disease

Like geographical analysis in general, medical geography is based on the assumption that location is the first clue to causal analysis. Against a background of two empirical studies, one dealing with a group of ALS cases and the other with a group of leukaemia cases, this paper argues that it is insufficient to define disease occurrence in spatial terms, referring to patients' places of residence at the time of onset. Since exposure and effect are related to both time and place, geographical studies of ill health in modern society are complicated by high mobility, long periods of latency and environmental change which might cause a distortion between cause and effect. Therefore a different approach is required.     

Patterns of Vulnerability to AIDS Impacts in Zambian Households

Using household data from Northern Zambia, this article looks at HIV/AIDS impacts on different aspects of people's access to food. The findings draw particular attention to the variances in vulnerability among households burdened by illness and orphans that are headed by men, women and the elderly. It is argued that vulnerability levels to HIV/AIDS impact differ substantially among households and implicitly expose the underlying causal conditions that enable or disable people in their responsiveness. Households affected by HIV/AIDS cannot be treated as a homogeneous group and understanding the differences in vulnerability can play an important policy role in designing targeted support.     

SQUARE PEGS IN ROUND HOLES: PROGRAM EVALUATION AND ORGANIZATIONAL UNCERTAINTY

It is argued that variation in uncertainty about social program goals and causal processes should influence the selection of evaluation designs and methods. Eighteen prominent evaluation reports are examined to contrast predicted evaluation strategies derived from theory about organizational uncertainty with the actual evaluation methods used. It was found that the actual design, measurement, and analysis procedures were frequently different from procedures that were compatible with the kinds of program uncertainty described in the evaluation reports. It is suggested that theory and practice in program evaluation can be improved by systematically incorporating organizational concepts that would contribute to the selection of evaluation methods.     

When Epistemic Communities Fail: Exploring the Mechanism of Policy Influence

Epistemic communities are an established concept in the study of international relations but can also explain policy change at the domestic level. This article asks why some epistemic communities manage to convince decision makers of their preferred policies while others do not. It suggests that the reason lies in the causal mechanism of epistemic communities' influence on decision makers, mediated by decision makers' demand for expert input. Most epistemic communities scholarship focuses on single case studies where the communities' influence on policy was strong and clear, leading to an overestimation of the groups' influence. To help correct this probable bias, this article compares a successful case of epistemic community influence (health technology assessment in Poland) with an unsuccessful example of the same policy (in the Czech Republic). The juxtaposition allows for unpacking of the necessary parts of the causal mechanism (emergence of an epistemic community, its activity, access to decision makers, and successful suasion) and separating them from the crucial scope condition. Decision makers' uncertainty about the policy issue at hand has traditionally been the key scope condition for epistemic communities but the successful Polish case demonstrates that epistemic communities can be influential even in highly certain areas of routine policymaking, leading to a reconceptualization of uncertainty as policymakers' demand for expert input. Demand can originate from various sources, but any change of it affects individual parts of the causal mechanism, leading to a success or failure of epistemic communities' policy enterprise.     

Does Policy Diffusion Need Space? Spatializing the Dynamics of Policy Diffusion

For decades, scholars in multiple disciplines have examined spatial diffusion, or the spatiotemporal properties associated with the diffusion of innovations. These properties include contagious, hierarchical, and relocation diffusion. Each of these refers to a spatial model that epitomizes how innovations spread among geographic locations. Policy diffusion, a separate but homologous research tradition, had its theoretical underpinnings in spatial diffusion. However, contemporary policy diffusion has focused largely on mechanism‐based diffusion. This article demonstrates how exploratory spatial data analysis can be used to uncover spatial policy diffusion properties. In this study, municipal smoking regulation adoptions, religious‐based initiatives, and bag ban and bag fees are examined. This study finds evidence that for each policy more than one property is occurring; therefore, this study proposes that a hybrid model best explains diffusion. This article demonstrates how examining spatial diffusion properties, in addition to diffusion mechanisms, can improve the conceptualization of diffusion theories, enhance mechanism or theory‐based specification of diffusion models, and unravel the specific regional or neighboring causal pathways linking policies between adopting jurisdictions.     

Exploring the links between political structure and metropolitan growth

Considerable scholarship documents a causal link between metropolitan growth and the resultant number and arrangement of local governments in metropolitan areas. Only scant research explores the reverse phenomena, that patterns of local government structure influence metropolitan population growth. Yet ample social science theory and case study evidence suggest support for such a causal link. This study examines empirically the influence of levels and changes in political structure on metropolitan population growth for 129 large US metropolitan areas between 1962 and 1982. The analysis extends previous research in two ways: first, by replacing aggregate measures of political fragmentation with a more sophisticated set of variables that capture different aspects of the multifaceted concept of political structure; and second, by examining the dynamic association between change in political structure and metropolitan growth. The results reveal mixed support for theories linking political structure to metropolitan growth. Of methodological importance is the finding that different dimensions of political structure interact differentially with metropolitan growth, suggesting that traditional aggregate measures of political structure obscure more than they reveal.     

Causal Assessment in Small‐N Policy Studies

The identification of cause‐and‐effect relationships plays an indispensable role in policy research, both for applied problem solving and for building theories of policy processes. Historical process tracing has emerged as a promising method for revealing causal mechanisms at a level of precision unattainable through statistical techniques. Yet historical analyses often produce dauntingly complex causal explanations, with numerous factors emerging as necessary but insufficient causes of an outcome. This article describes an approach that renders complex causal narratives more analytically tractable by establishing measurement criteria for ranking the relative importance of component causes. By focusing on subjectively useful measurement attributes, the approach is well suited to the policy sciences' unique combination of explicitly normative aspirations and a commitment to the systematic assessment of causal claims.     

Geschichte ohne Kausalität. Abgrenzungsstrategien gegen die Wissenschaftssoziologie in zeitgenössischen Ansätzen historischer Epistemologie

History Without Causality. How Contemporary Historical Epistemology Demarcates Itself From the Sociology of Scientific Knowledge. Contemporary proponents of historical epistemology often try to delimit their enterprise by demarcating it from the sociology of scientific knowledge and other sociologically oriented approaches in the history of science. Their criticism is directed against the use of causal explanations which are deemed to invite reductionism and lead to a totalizing perspective on science. In the present article I want to analyse this line of criticism in what I consider are two paradigmatic works of contemporary historical epistemology: Lorraine Daston's und Peter Galison's Objectivity and Hans‐Jörg Rheinberger's Toward a History of Epistemic Things. I first present their arguments against the sociological and causal analysis of scientific knowledge and practice and then try to defend sociological work in the history of science against their charges. I will, however, not do so by defending causal explanations directly. Rather, I will show that the arguments against sociological analysis put forward in contemporary historical epistemology, as well as historical epistemology's own models of historical explanation and narration, bear problematic consequences. I argue that Daston, Galison and Rheinberger fail to create productive resonances between macro‐ and microhistorical perspectives, that they reproduce an internalist picture of scientific knowledge, and finally that Rheinberger's attempt to deconstruct the dichotomy between subject and object leads him to neglect questions about the political dimension of scientific research.     

Michael Oakeshott and Hans-Georg Gadamer on Practices,Social Science,and Modernity

This article compares Michael Oakeshott and Hans-Georg Gadamer, in particular examining the different ways they conceptualise human practices and the relationship between theory and practice. First, I highlight where the two agree. Both are sceptical of causal explanations of human behaviour, and instead advocate understanding human conduct intersubjectively, using Aristotle's concept of ‘practical wisdom’. Second, however, I also highlight important areas of disagreement. Oakeshott maintains that non-philosophical but non-practical theoretical disciplines are possible; by contrast, Gadamer stresses the intrinsically practical nature of all understanding. More practically, they also differ over how useful Aristotelian insights are politically. Gadamer claims (like Aristotle) that we can seek an objective common good; Oakeshott rejects this, due to his commitment to pluralism. Finally, I suggest that these divergences are due to different conceptualisations of ‘experience’ and the Western tradition. First, Oakeshott and Gadamer differ over how authentically we can know human experience, and therefore over how important the right discipline for investigation is. Second, they differ over the degree to which we can interrogate the Western tradition. For Oakeshott, we must accept its pluralist inheritance; for Gadamer a healthy dialogue with it is essential—only thus can we unearth the (Aristotelian) insights vital for us to live well.     

Priming political trust: Evidence from an experiment

Political trust is one of the most researched areas in political science. Yet little is known about what causes political trust to vary. Past research has relied almost solely on survey data, and focused on exploring the correlation between political trust and various micro- and macro-level factors. This research note reports the findings from an experiment designed to examine the causal effect of one of the most commonly cited causes of political trust: political probity. Results show that political trust and trust in specific institutions change when participants read an article about political probity and complete a word-association task. The treatment we used is low cost, straightforward and may be used by other researchers to alter political trust in experiments.     

Max Weber on Causal Analysis,Interpretation, and Comparison

Max Weber's methodological writings offered a model of singular causal analysis that anticipated key elements of contemporary Anglo‐American philosophy of the social and cultural sciences. The model accurately portrayed crucial steps and dimensions of causal reasoning in these disciplines, outlining a dynamic and probabilistic conception of historical processes, counterfactual reasoning, and comparison as a substitute for counterfactual argument. Above all, Weber recognized the interpretation of human actions as a subcategory of causal analysis, in which the agents‘ visions of desired outcomes, together with their beliefs about how to bring them about, cause them to act as they do.     

METAPHYSICS IN HISTORY: NOTES ON THE ORIGINS OF AUTHORITARIANISM AND POPULISM

This article contains an analysis of metaphysics in historical narrative, especially as it pertains to the study of authoritarianism and populism, and a brief exploration of the political implications of metaphysical narratives. The article engages closely with twentieth-century accounts of the origins of authoritarianism and populism and related topics insofar as they are relevant today. Some present-day authors continue to adopt some of the tropes of twentieth-century accounts, though they do so without acknowledging the uncertainties and doubts expressed by twentieth-century historians and social scientists with regard to their own paradigms. The analysis proceeds through an immanent critique, examining the internal contradictions of complex notions. The focus is on teleology and transcendentalism. Teleology occludes short-term causality, contexts, and conjunctures. It entails anachronism, or the retrospective attribution of meaning, and ontological fatalism, which renders historical explanation irrelevant. Eschewing fatalism means allowing for the causal efficiency of intervening conditions, which contradicts the premises of the teleological approach. The reification of stages (or eras) in teleological successions leads to asynchronies, or the coexistence of elements belonging to different totalities. The formulation of origins as predispositions and potentialities entails a transcendental approach. Immanently, there are no potentialities but actual existents immersed in their historical context; these can only be potentialities with respect to a transcendental being or essence. But this approach leads to irresoluble contradictions and an alienated form of history in which human agency and actors themselves are only manifestations of a beyond. The neglect of social antagonisms as immediate causes of authoritarianism entails a specific political position. The postulation of populism as transcendentally equivalent to authoritarianism carries a negative valuation of challenges to liberal democracy. With a less deterministic approach to history, analytical and normative assessments become less predetermined.     

HISTORY'S DEMARCATION PROBLEM

In his 1998 book Heterocosmica: Fiction and Possible Worlds, Lubomír Dole?el put forth a theory of narrative fiction based on the interdisciplinary framework of possible worlds. In Possible Worlds of Fiction and History: The Postmodern Stage, Dole?el takes his earlier theory further and applies it to historiography as well, with the specific aim of showing how the study of history might be defended against the postmodern challenge via the use of possible worlds (PW) semantics. Dole?el's book is essentially an argument against the postmodern views expressed by Roland Barthes and Hayden White, who have claimed that fundamentally, there is no difference between fictional and historical narratives. According to Dole?el, this difference can be saved if the focus of attention is shifted from the textual features of these narratives to the fictional or historical worlds that the narratives project. Dole?el's comparison of fictional and historical worlds to each other is quite illuminating and thorough. However, the question remains whether the application of PW semantics does anything besides offering a detailed analysis of the structure of the different types of narrative worlds. After all, one should not overlook the perhaps more practical way of differentiating between historical and fictional narratives through their institutional status. Furthermore, we argue that by focusing on the properties of the end products, that is, the resulting narratives, Dole?el concedes too much to postmodernists. A stronger way to give postmodernists a taste of their own medicine would be to argue that the rules that historians follow in the process of generating, constructing, and evaluating weighed causal explanations (or historical models of the past) are fundamentally different from whatever rules govern the generation and construction of fiction.     

Population growth among U.S. regions and metropolitan areas: a test for causality

ABSTRACT This paper explores theories of population growth, and implied economic growth, among the major U.S. metro areas and regions. One set of theoretical arguments favor the growth of large cities, while an alternative set of arguments favor the growth of the smaller urban areas. Still another set of arguments combine economic space and urban size in the concept of regional growth centers as the engines of regional growth. Granger causality tests are applied to determine the role of urban size or growth centres as engines of regional growth. The test results indicate no causal relationship exists.