2015. Issue-Specific Opinion Change: The Supreme Court and Health Care Reform. Forthcoming at Public Opinion Quarterly (with David M. Glick). Appendix | Bibtex | AbstractDid the Supreme Court decision in the Affordable Care Act case change public opinion about health care reform? We utilize a multi-wave panel design with observations collected just hours before and after the Court's decision to address macro-level questions concerning the Court's effect on opinion about health care reform generally and the individual mandate specifically. We show that support for health care reform remained constant despite significant positive movement on the mandate. We further exploit features of the panel to analyze this micro-level change and test hypotheses related to cognitive models, individual attributes, and assessments of the Court's legitimacy. Despite some evidence of micro-level variation, our findings ultimately point to a decision that induced a general, persistent, and relatively unconditional uptick in support for the provision the Court deemed constitutional.

Keywords: Supreme Court, Public opinion, Panel data, Health care, Mandate, Mixed-effects model

2015. Comparing Membership Interest Group Networks Across Space and Time, Size, Issue and Industry. Forthcoming at Network Sciece (with Janet M. Box-Steffensmeier). Appendix | Bibtex | AbstractWe compare and contrast the network formation of interest groups across industry and issue area. We focus on membership interest groups, which by virtue of representing the interests of voluntary members, face particular organizational and maintenance constraints. To reveal their cooperative behavior we build a network data set based on cosigner status to United States Supreme Court amicus curiae briefs and analyze it with exponential random graph models (ERGMs) and multidimensional scaling. We find that while many of the same factors shape membership networks, religious, labor and political organizations do not share the same struc- ture as each other or as the business, civic and professional groups. Our methodological approach culminates in a clear and compact spatial representation of network similarities and differences.

Keywords: Interest groups, membership organizations, coalition strategies, amicus curiae briefs, exponential random graph models, multidimensional scaling

2015. Chief Justice Roberts's Health Care Decision Disrobed: The Microfoundations of the Court's Legitimacy. Forthcoming at American Journal of Political Science (with David M. Glick). Appendix | Data | Bibtex | AbstractThe 2012 challenge to the Affordable Care Act was an unusual opportunity for people to form or reassess opinions about the Supreme Court. We utilize panel data coupled with as-if random assignment to reports that Chief Justice Roberts' decision was politically motivated to investigate the microfoundations of the Court's legitimacy. Specifically, we test the effects of changes in individuals' ideological congruence with the Court and exposure to the non-legalistic account of the decision. We find that both affect perceptions of the Court's legitimacy. Moreover, we show that these mechanisms interact in important ways and that prior beliefs that the Court is a legalistic institution magnify the effect of updating ones ideological proximity to the Court. While we demonstrate that individuals can and did update their views for multiple reasons, we also highlight constraints which allow for aggregate stability in spite of individual level change.

Keywords: Supreme Court, Legitimacy, Diffuse support, Public opinion, Ideology, Panel data, Health care

*Winner of the 2015 Best Article Award by the Law and Courts Section of the American Political Science Association.

2015. Mass Preferences on Shared Representation and the Composition of Legislative Districts. American Politics Research, X(X): 1-28 (with Todd Makse). Appendix | Bibtex | AbstractScholars of redistricting often discuss communities of interest as a guideline for drawing districts, but scholarship offers little guidance on how citizens construe communities and interests in the context of representation. In this paper, we seek to better understand how citizens' perceptions of people and places affect preferences regarding representation. Using an original survey conducted in fifteen Massachusetts communities, we explore whether citizens have meaningful preferences about the communities with whom they share the same representative. To the extent they do, we test whether these preferences are driven by considerations of geography or other factors such as partisanship, race, and socioeconomic status. Our findings not only offer the opportunity to refine the concept of communities of interest to account for voter preferences, but more broadly speak to the literature on the increasingly political nature of residential preferences and their impact on political attitudes, participation, and voting behavior.

Keywords: Communities of interest, Redistricting, Representation, Residential preference, Massachusetts

2014. Following the Money: Super PACs and the 2012 Presidential Nomination. Presidential Studies Quarterly, 44(3): 410-430 (with Corwin D. Smidt). Bibtex | AbstractThe entrance of Super Political Action Committees (Super PACs), outside groups with no caps on fundraising or independent expenditures, prompts a reexamination of the role of money in campaigns and elections. We investigate the influence of Super PAC expenditures in the 2012 Republican nomination contest. A compressed calendar makes nomination campaigns expensive and money crucial, especially for lesser-known candidates, such that outside expenditures likely made a difference. Indeed, we find Super PACs helped to extend Santorum’s long-shot candidacy but also helped Romney by weakening momentum from Gingrich and Santorum wins. Using panel data of candidate dynamics, we also find that candidate and Super PAC expenditures within various key primary states reactively complement each other. However, we do not find dispositive evidence that Super PACs coordinate with campaigns, thereby acting, at least in this context, within the bounds of their legally mandated independence.

Keywords: Super PACs, Seemingly unrelated regression, Campaign fundraising, Campaign expenditures, 2012 nomination contest, Coordination

2014. Deus ex Machina: Candidate Web Presence and the Presidential Nomination Campaign. Political Research Quarterly, 67(1): 108-122 (with Corwin D. Smidt and Costas Panagopoulos). Appendix | Bibtex | AbstractUsing data from the 2008 presidential nomination contest, we offer systematic tests of the relationships between traditional campaign factors, the Internet and campaign performance. We find that claims of the Internet’s relevance to modern campaigns are warranted, as it is a unique facet of campaigns and significantly improves candidates’ financial and electoral support. The Internet is especially helpful to candidates in generating small-donor contributions and in maximizing contributions after early primary victories. Overall, these findings suggest that the Internet offers a viable mechanism for long-shot candidates to overcome the resource demands of the current presidential nomination system.

Keywords: Internet, Web presence, Campaign contributions, Polls, Presidential primary, Long-shots, Frontrunners, Campaign dynamics

*Winner of the 2014 Best PRQ Article Award.

2014. The Evolution and Formation of Amicus Curiae Networks. Social Networks, 36: 82-96 (with Janet M. Box-Steffensmeier). Appendix | Bibtex | AbstractThis paper sheds light on two age-old questions of interest group behavior: how have interest group coalition strategies changed over time and which factors determine whether interest groups work together? Through the creation of a new network measure of interest group coalitions based on cosigner status to United States Supreme Court amicus curiae briefs, we illuminate the central players and overall characteristics of this dynamic network from 1930 to 2009. We present evidence of an increasingly transitive network resembling a host of tightly grouped factions and leadership hub organizations employing mixed coalition strategies. We also model the attribute homophily and structure of the present-day network. We find assortative mixing of interest groups based on industry area, budget, sales and membership.

Keywords: Interest groups, Coalition strategies, Amicus curiae briefs, Social networks, Exponential random graph model

2013. Quality Over Quantity: Amici Influence and Judicial Decision Making. American Political Science Review, 107(3): 1-15 (with Janet M. Box-Steffensmeier and Matthew P. Hitt). Data | Bibtex | AbstractInterest groups often make their preferences known on cases before the U.S. Supreme Court via amicus curiae briefs. In evaluating the case and related arguments, we posit that judges take into account more than just the number of supporters for the liberal and conservative positions. Specifically, judges’ decisions may also reflect the relative power of the groups. We use network position to measure interest group power in U.S. Supreme Court cases from 1946 to 2001. We find that the effect of interest group power is minimal in times of heavily advantaged cases. However, when the two sides of a case are approximately equal in the number of briefs, such power is a valuable signal to judges. We also show that justice ideology moderates the effect of liberal interest group power. The results corroborate previous findings on the influence of amicus curiae briefs and add a nuanced understanding of the conditions under which the quality and reputation of interest groups matter, not just the quantity.

Keywords: Interest groups, Supreme Court decisions, Amicus curiae briefs, Social networks, Eigenvector centrality, Ideology

2013. Crowdsourcing Panel Studies and Real-Time Experiments in MTurk. The Political Methodologist, 20(2): 27-33 (with David M. Glick). Bibtex | AbstractWhile researchers conducting quick experiments and pilot studies currently appear to make the most use of Amazon’s Mechanical Turk (MTurk) as a subject recruitment tool, it is capable of supporting more ambitious research designs, many of which would be otherwise infeasible for most researchers. Specifically, researchers with a modest budget can use MTurk to recruit participants for customized and flexible panel studies. Because it is a cloud based environment, it is easy to recontact respondents, which helps limit panel attrition. Moreover, when used in tandem with online survey software, panel waves can be quickly fielded around imminent and recent political events, rather than at constant intervals or other times determined well in advance. Thus, MTurk’s attributes allow researchers to affordably collect individual level data for pre-post comparisons that can be combined with real-time experimental treatments. In this piece we briefly discuss our own experience conducting panel studies in MTurk and provide some basic instructions for researchers looking to do the same. We utilize the design and data from one of our own recent studies to discuss how we took advantage of MTurk and suggest some avenues for future research.

Keywords: MTurk, Cloud computing, Panel data, Real-time experiments, Foreseeable events

2012. More Bang for the Buck: Campaign Spending and Fundraising Success. American Politics Research, 40(6): 949-975 (with Corwin D. Smidt). Appendix | Bibtex | AbstractCan candidates spend their way into financial success? We propose that the 2007 presidential money primary offers unprecedented leverage to evaluate spending’s influence since it allows for sharper controls of confounding factors. Our results demonstrate that greater candidate spending on fundraising- related efforts is associated with significant future financial benefits. We estimate that, prior to the primaries, increases in spending have an equal or larger payoff than increases in a candidate’s viability and find different types of spending are beneficial for frontrunner and long-shot candidates. The results consistently indicate greater early spending works to advantage candidates, suggesting a lack of initial resources is a significant obstacle for candidates who seek to financially benefit from their campaign’s performance.

Keywords: Money primary, Campaign expenditures, Viability, Frontrunners, Long-shots, Campaign fundraising, Campaign dynamics

2012. Still Part of the Conversation: Iowa and New Hampshire’s Say within the Invisible Primary. Presidential Studies Quarterly, 42(3): 597-621 (with Corwin D. Smidt). Bibtex | AbstractWe propose that the extant literature has underestimated the central roles of Iowa and New Hampshire within the invisible primary and, thus, party nominations. Since candidates and the news media focus disproportionately on these states early in the nomination season, impressions of candidate performance within these states have a disproportionate influence on the invisible primary long before their actual outcomes are observed. Using a Bayesian vector autoregression we find that polls within Iowa and New Hampshire have a more consistent influence on candidates’ levels of national news media coverage and national polling than vice versa. We also find that campaign contributions are as responsive to early state polls as they are to national forces or campaign activities. Although these findings do not dispute that candidates need a broad basis of national support to win a party’s nomination, they explain why candidates continue to campaign early and intensely in these first-in-the-nation contests.

Keywords: Invisible primary, Bayesian vector autoregression, Campaign contributions, Polling, Early states, New Hampshire, Iowa, News media

2011. Riding the Waves of Money: Contribution Dynamics in the 2008 Presidential Nomination Campaign. Journal of Political Marketing, 10: 1-23 (with Corwin D. Smidt). Bibtex | AbstractThe 2008 primary was the most nuanced and expensive nomination contest in history. We investigate how this massive battle for contributions played out over 2007 and the first half of 2008 by analyzing the daily dynamics of candidate contributions using the Federal Elections Commission’s collection of individual contributions. Not surprisingly, Giuliani and Clinton were the leaders in contribution momentum during the latter parts of the so-called money primary. This pattern abruptly changed in 2008 as both parties experienced a structural change in contribution flows. While Iowa and New Hampshire placements helped their causes, the South Carolina primary was by far the most rewarding early contest for Obama and McCain. Furthermore, primary victories do not benefit all candidates equally, as Clinton and Huckabee gained far less than their counterparts in response to their early victories.

Keywords: Presidential nomination contests, Momentum, Campaign contributions, Early states, Campaign dynamics, Bayesian state space model, Kalman filter

2009. Speeding Up R for Windows. The Political Methodologist, 17(1): 4-11 (with Joshua A. Morris). Bibtex | AbstractTo what extent do different Windows PC characteristics increase the modeling efficiency of R? Do some programs or versions of R run better on different PCs? And for which kinds of models do enhanced PCs and clusters diminish processing time? This research note seeks to provide novice to intermediate level R users with a framework for understand- ing the benefits of explicit parallel processing and upgrades in PC hardware for large datasets and computationally burdensome models. We compare the relative benefits of each optimization with simple efficiency tests. In addition, we provide basic R code to make the transition to parallel processing easier for novice users without networked labs or cluster access.

Keywords: Computing, Computational clusters, R, CRAN, Parallel processing, Memory

2007. Changing Horses in Wartime? The 2004 Presidential Election. Political Behavior, 29: 279-304 (with Herbert F. Weisberg). Bibtex | AbstractThe literature makes clear that foreign policy affects voting, but it does not lead to clear expectations as to how a war will affect voting. Will views about the advisability of the war predominate? Or will the indirect effect through the incumbent’s image be more important? Will a war crowd out other potential issues, particularly domestic ones? This paper addresses these questions through a series of focused analyses of NES survey data. We find that an increase in strong Republican partisans clinched the election for President Bush. The Iraq War was not a direct vote gainer for the President, but the larger War on Terrorism burnished his image as a leader, at least long enough to win the election. Likewise, the cultural war allowed President Bush to retain some of the votes that he might otherwise have lost due to the unpopularity of the Iraq War.

Keywords: Presidential election, Foreign policy, Wartime elections, NES, Iraq War, War on Terrorism, Culture war, Voting behavior


2014. Applied Social Science Methodology. Under Contract at Cambridge University Press (with John Gerring).


Data collection:
Interest Group Networks
The Visible Primary
Public Opinion around Foreseeable Events
The Ohio Political Survey (TOPS)

PolMeth XXV: Learning from the Campaign Environment
PolMeth XXVI: Issues, Media and Momentum in the 2008 Primary

Resources & Tutorials

Importing and Exporting Data
Guide on Fixed Width Data Importing (pdf)
Addendum on Importing Data (pdf)

An Introduction to Using R (pdf) | Accompanying R Script (txt) | Accompanying Data (zip)
OLS Data Analysis in R (pdf) | Accompanying R Script (txt) | Accompanying Data (zip)
An Introduction to Complex Surveys in R (pdf) | Accompanying Data & R Script File (txt)

An Introduction to Stata (pdf) | Accompanying Data & .Do File (zip)