Understanding Interaction Models: Improving Empirical Analyses

2006 ◽  
Vol 14 (1) ◽  
pp. 63-82 ◽  
Author(s):  
Thomas Brambor ◽  
William Roberts Clark ◽  
Matt Golder

Multiplicative interaction models are common in the quantitative political science literature. This is so for good reason. Institutional arguments frequently imply that the relationship between political inputs and outcomes varies depending on the institutional context. Models of strategic interaction typically produce conditional hypotheses as well. Although conditional hypotheses are ubiquitous in political science and multiplicative interaction models have been found to capture their intuition quite well, a survey of the top three political science journals from 1998 to 2002 suggests that the execution of these models is often flawed and inferential errors are common. We believe that considerable progress in our understanding of the political world can occur if scholars follow the simple checklist of dos and don'ts for using multiplicative interaction models presented in this article. Only 10% of the articles in our survey followed the checklist.

2018 ◽  
Vol 27 (2) ◽  
pp. 163-192 ◽  
Author(s):  
Jens Hainmueller ◽  
Jonathan Mummolo ◽  
Yiqing Xu

Multiplicative interaction models are widely used in social science to examine whether the relationship between an outcome and an independent variable changes with a moderating variable. Current empirical practice tends to overlook two important problems. First, these models assume a linear interaction effect that changes at a constant rate with the moderator. Second, estimates of the conditional effects of the independent variable can be misleading if there is a lack of common support of the moderator. Replicating 46 interaction effects from 22 recent publications in five top political science journals, we find that these core assumptions often fail in practice, suggesting that a large portion of findings across all political science subfields based on interaction models are fragile and model dependent. We propose a checklist of simple diagnostics to assess the validity of these assumptions and offer flexible estimation strategies that allow for nonlinear interaction effects and safeguard against excessive extrapolation. These statistical routines are available in both R and STATA.


1982 ◽  
Vol 15 (03) ◽  
pp. 439-448 ◽  
Author(s):  
Kathleen A. Frankovic

The relationship between sex and political behavior has been discussed only in passing in the political science literature, if it is discussed at all. There has been little evidence from the 1940s to the 1970s that gender plays a role in determining issue positions, candidate evaluations, or candidate preference, as a quick perusal of some well-read political science works would confirm.Berelson, Lazarsfeld and McPhee inVoting, the seminal work on opinion formation in a presidential campaign, discovered no relationship between vote preference and sex. The researchers did discover a difference in reported interest in the 1948 campaign between men and women early in the campaign, but even that difference faded as the election drew near.Campbell, Converse, Miller and Stokes inThe American Votermake no mention of either the existence or non-existence of sex difference in policy or candidate preference. They focus instead on differences in turnout, involvement and efficacy. Although they suggest that at one time opinions about Prohibition may have separated men from women, the authors conclude, “In the current era, there is no reason to believe that womenas womenare differentially attracted to one of the political parties.”Pomper inVoters' Choicedoes find a relationship between sex and one issue dimension–the dimension of war and peace. But Pomper, Campbell and Berelson agree that any sex differences cannot, by definition, be long-lasting. Berelson cites the lack of differentiation in the way policies affect the sexes, the lack of differences in intergenerational transmission, and the lack of segregation between the sexes.


2017 ◽  
Author(s):  
Luke Keele ◽  
Randolph T. Stevenson

Social scientists use the concept of interactions to study effect dependency. Such analyses can be conducted using standard regression models. However, an interaction analysis may represent either a causal interaction or effect modification. Under causal interaction, the analyst is interested in whether two treatments have differing effects when both are administered. Under effect modification, the analysts investigates whether the effect of a single treatment varies across levels of a baseline covariate. Importantly, the identification assumptions for these two types of analysis are very different. In this paper, we clarify the difference between these two types of interaction analysis. We demonstrate that this distinction is mostly ignored in the political science literature. We conclude with a review of several applications.


2020 ◽  
pp. 135406612092260
Author(s):  
Stephen Aris

IR has long been concerned about its claim on disciplinary status. This includes concerns about its differentiation from Political Science and a divide between scholars who advocate a narrow disciplinary approach and others who conceive of IR as a pluri-disciplinary concept. Although these dilemmas revolve around its position vis-à-vis other disciplines, the vast majority of the recent disciplinary-sociology debates have focused on the extent of IR scholarship’s intradisciplinary fragmentation, along epistemological, topical, national, status and other lines. However, the sociology of science literature stresses that disciplines are the product of not only internal practice but also their knowledge relations to and differentiation from other disciplines. In short, intradisciplinary fragmentation cannot be considered as detached from a discipline’s relations to other disciplines – and, by extension, the differentiated knowledge relationships held by distinct intradisciplinary fragments to other disciplines. Taking this into account, this article uses bibliometric analysis of journals as a proxy for analysing the relationship between IR’s intradisciplinary make-up and its interdisciplinary relations to eight cognate disciplines between 2013 and 2017. Three distinct modes of bibliometric analysis are operationalised to map three different aspects of interdisciplinary knowledge practice: (inter)disciplinary debates (direct citation), multidisciplinary knowledge bases (bibliographic coupling) and interdisciplinary knowledge production (co-citation). On this basis, the article asks, one, whether and how differences in the interdisciplinary knowledge relations practised by IR scholarship correlate with intra-IR lines of fragmentation. And two, what are the implications for how IR’s socio-intellectual composition is understood and its disciplinary status evaluated?


2019 ◽  
Vol 17 (4) ◽  
pp. 436-446 ◽  
Author(s):  
James Dennison

In this article, I offer a review of the uses and findings regarding public issue salience in the political science literature, with a focus on electoral behaviour. I argue that in spite of the increased use of issue salience in recent years, with impressive explanatory results, the concept of issue salience remains underspecified and, at times, contradictory and that its antecedents remain relatively unknown. This is likely to have led to serious shortcomings when attempting to explain recent changes to party systems and electoral results in advanced democracies.


2012 ◽  
Vol 106 (3) ◽  
pp. 607-621 ◽  
Author(s):  
NADIA URBINATI

Freedom as non-domination has acquired a leading status in political science. As a consequence of its success, neo-roman republicanism also has achieved great prominence as the political tradition that delivered it. Yet despite the fact that liberty in the Roman mode was forged not only in direct confrontation with monarchy but against democracy as well, the relationship of republicanism to democracy is the great absentee in the contemporary debate on non-domination. This article brings that relationship back into view in both historical and conceptual terms. It illustrates the misrepresentations of democracy in the Roman tradition and shows how these undergirded the theory of liberty as non-domination as a counter to political equality as a claim to taking part inimperium. In so doing it brings to the fore the “liberty side” of democratic citizenship as the equal rights of all citizens to exercise their political rights, in direct or indirect form.


2016 ◽  
Vol 14 (1) ◽  
pp. 121-131 ◽  
Author(s):  
Marc Lynch

A decade ago, very few political scientists had either the opportunity or the incentive to engage with the political public in a direct, unmediated way. Today, there is a dense and eclectic ecosystem of political science and international relations-focused blogs and online publications, where good work can easily find an audience through social media. There are multiple initiatives dedicated to supporting academic interventions in the public sphere, and virtually every political or cultural magazine of note now offers a robust online section featuring commentary and analysis in which political scientists are well represented. This has transformed publication for a broader public from something exotic to something utterly routine. I discuss how these changes have affected individual scholars, the field of political science, and the political world with which we are engaged.


Author(s):  
David Francis Taylor

This book explores how the works of William Shakespeare, John Milton, Jonathan Swift, and others were taken up by caricaturists as a means of helping the eighteenth-century British public make sense of political issues, outrages, and personalities. The first in-depth exploration of the relationship between literature and visual satire in this period, the book explores how great texts, seen through the lens of visual parody, shape how we understand the political world. It offers a fascinating, novel approach to literary history.


2009 ◽  
Vol 11 (4) ◽  
pp. 1-31 ◽  
Author(s):  
Oded Shenkar ◽  
Ilgaz Arikan

This paper broadens the scope and depth of business alliance research by way of interdisciplinary enrichment. The paper draws on the political science literature on nation-state alliances to generate insights into the establishment, operations and performance of inter-firm alliances. Shared theory bases of game theory and transaction cost economics, as well as theories, variables and research findings indigenous to political science are posited as a platform from which propositions regarding inter-firm alliances are derived.


2016 ◽  
Vol 1 (1) ◽  
pp. 15-26
Author(s):  
Nikos Kaplantzis

As a response to calls for political research to do more than refer to visuals and for visual research to focus on the political, this paper discusses a Ricoeurian narrative-communicative action approach to the construction of political space applied to images, even though, until today, very little attention has been given to Ricoeur’s conception of the relationship between hermeneutics and visual theory. An updated reworking of Paul Ricoeur's critical hermeneutics offers a better basis for reconstructing visual (political) studies by sharpening the focus on the ideas of embodied imaginary and iconic augmentation. Ricoeur offers an explicit connection to visual political studies in the direction of pointing out the ways in which images, scenes, and narratives attempt to convey ideology, balancing a hermeneutics of suspicion with a hermeneutics of faith, illustrating the aporias, the opening and closing of possibilities from iconic image to ideograph and identity.


Sign in / Sign up

Export Citation Format

Share Document