2014 Conference

Visions in Methodology (VIM) 2014

Faculty of Social Science, McMaster University

Hamilton, ON, CANADA

May 20-22, 2014

Sponsored by

The Visions in Methodology (VIM) workshop brings together junior and senior women faculty working in political methodology.

The two and a half day experience is highlighted by:
IMG_1720cropped

  • Scholarly research presentations
  • Feedback on research from conference participants
  • Sessions on career and gender topics
  • Oral autobiographies by senior scholars

Program Co-Chairs: Michelle Dion (McMaster) and Laura Stephenson (Western Ontario)
Program Committee: Jan Box-Steffensmeier (Ohio State), Sara McLaughlin Mitchell (Iowa), Megan Shannon (Florida State), and Caroline Tolbert (Iowa)

2014 Featured Senior Scholars: Saundra Schneider (Michigan State) and Elisabeth Gidengil (McGill)

Participants will arrive by 5pm, May 20 and depart the morning of May 23.

The workshop at McMaster University is part of a broader effort to support women in the field of political methodology. In addition to providing a forum to share scholarly work, VIM connects women in a field where they are under-represented. VIM began as an implementation of recommendations from a National Academy of Sciences report, the APSA Workshop on the Advancement of Women in Academic Political Science, and the 2006 Political Methodology Long Range Strategic Planning Committee Report. VIM provides opportunities for networking, mentoring, and scholarly progress for women in the political methodology community.

Call for Participation [now closed]
We welcome paper presentation proposals addressing survey design and effects, applying advanced statistical methods to substantive issues, and evaluating or developing new experimental or quantitative methods.

Application Information: Women faculty at all ranks are encouraged to apply, though preference will be given to those at earlier stages of their careers or those for whom this would who be their first or second VIM meeting. To apply, send a curriculum vitae and a title and abstract (100-150 words) for the research presentation to dionm@mcmaster.ca. Priority consideration will be given to those applications received by November 15, 2013, though the final closing date will be December 15, 2013. [now closed]

Draft program (2014/05/15)

Recommended readings for professional development discussions

Networking and collaboration

Strategies for engaging in meaningful service

Leaning in? Leaning out?

Strategies for minimizing implicit bias

Presentations accepted for VIM 2014

How partisans stereotype female candidates: Untangling the relationship between partisanship, gender stereotypes, and support for female candidates
Nichole Bauer (PhD candidate, Political Science, Indiana University)
Existing research offers conflicting conclusions on whether feminine stereotypes help, hurt, or have no effect on how voters perceive female candidates (Dolan 2014; Hayes 2011; Brooks 2011). Feminine stereotypes characterize women as nurturing and sensitive – qualities generally not valued in political leadership (Huddy and Terkildsen 1993b). Complicating the relationship between female candidates and feminine stereotypes is that they overlap with partisan stereotypes (Winter 2010). This study uses an original survey experiment to clarify the relationship between partisanship and gender stereotypes. The results uncover several findings. First, partisanship and gender can combine to affect how feminine stereotypes become activated during a campaign. Second, when feminine stereotypes are activated voters use them to evaluate Democrat female candidates more positively than Republican female candidates on caring and trustworthiness. Third, activating masculine stereotypes leads voters to make similar inferences about the leadership abilities of both Democrat and Republican female candidates.
Discussant: Elizabeth Goodyear-Grant, Queen’s University

Variations in the Separation of Powers and Budget Allocations
Constanza Figueroa-Schibber (PhD candidate, Political Science, Washington University in St. Louis)
In separation of powers systems (presidential democracy), we assume that legislative and executive branches have to concur in order for policies to be enacted. When deciding on the nation’s spending priorities, two factors determine the balance struck across budget categories — the ideological divergence between the president and the assembly and the formal and informal powers ascribed to each of them. In many presidential regimes, specially in Latin America, it has been argued that legislators are marginalized in the policy-making process, especially the budget-making process. This raises a fundamental question that has implications for the quality of representation: Do the preferences of democratically elected legislators get reflected in the distribution of public expenditures? In this study I explore how institutional arrangements and inter-branch ideological divergence affect the success of the majority of legislators at determining spending priorities. Looking at 15 Latin American democracies for the past 15 years, I develop new measures to estimate both the level of success of the majority of legislators in the assembly and congruence between the median legislator and the president. Using a Bayesian Multivariate Probit model with random effects, I find that the spending priorities of legislators (as captured with survey data) can be reflected in budget outcomes when the legislative median is distant from the president but only where the institutions that govern the budget-making process minimize the president’s powers or generate incentives for an interbranch compromise.
Discussant: Laurie Beaudonnet, Université de Montréal

“You cheated on me!” Causes and consequences of cheating in online surveys
Carol Galais (Post-Doctoral Fellow, Political Science, Université de Montréal)
Co-authors: Eva Anduiza (Universitat Autònoma de Barcelona)
Online surveys are increasingly popular for their relatively low costs. Previous research has paid some attention to the problems of bias on samples in online surveys. However, less is known about other sources of error that may also be important in this survey methodology. Online surveys are self-administered by respondents willing to receive incentives for completing questionnaires. Thus, they may rush to finish them as soon as possible, not reading the questions carefully and/or clicking random responses. Thus, online surveys may suffer from this type of measurement error derived from cheating (answering without reading). Trap questions have become customary to identify and control for this source of error. In this paper we analyze the causes and consequences of cheating in online surveys using an online panel survey with four waves carried out in Spain between 2011 and 2012. Our data show relatively low levels of cheating behavior, that changes along time. Age, education, time spent online, interest in politics, party closeness, thinking about the survey as a way to express opinions and conscientiousness reduce the likelihood of cheating. The consequences of cheating do not seem dramatic in our case, but vary depending on the type of question and the survey wave. Cheaters may pose a problem when focusing on factual information or building scales.
Discussant: Karen Bird, McMaster University

Predicting unreported conflict fatalities with log-linear multiple recapture models
Anita Gohdes (PhD candidate, Political Science, University of Mannheim)
NOTE: For a copy of this paper, please email the author (anita.gohdes@uni-mannheim.de).
In any conflict, incidences of violence are not reported with equal probability, many acts of violence are purposely hidden from view, or fail to be registered in the chaos of fighting. Yet body counts – or thresholds thereof – form the basis of the majority of studies investigating the onset, diffusion and termination of conflict episodes. Researchers thus face the challenge of working with information that reflects the data generating process of reported violence, not actual conflict patterns. In this paper, I show how multiple recapture methods (MR) make use of incomplete convenience data on violence to model the reporting process, and then predict what violence went unreported. First, I draw biased convenience samples from a simulated population of violent incidences in order to approximate conflict reporting. Based on these samples, I model the reporting process and predict the number of incidences that went ‘unreported’ in the samples. The simulations demonstrate how the method picks up varying levels of underreporting across space and time, and works under conditions of both biased and unbiased reporting. I then apply the method to reported violence from the ongoing conflict in Syria.
Discussant: Jakana Thomas, Michigan State University

A discrete choice model for ordered nests
Laura Grigolon (Assistant Professor, Economics, McMaster)
I introduce a new discrete choice model of demand, the Ordered Nested Generalized Extreme Value (ONGEV) model. This model allows for segmentation in markets with di.erentiated products where consumers are likely to substitute to neighboring seg- ments. The model represents a tractable extension of the nested logit model, in which substitution patterns across all segments, neighboring or not, are instead symmetric. I apply the model to the automobile market where segments are ordered from small-size to luxury. The nested logit model is rejected against the ONGEV model. The implied substitution patterns illustrate the presence of relevant neighboring segment e.ects when consumers substitute outside their segment.
Discussant: Saundra Schneider, Michigan State University

Representing the Underrepresented: Minority Group Representation through Speech in the U.S. House
Nicole Kalaf-Hughes (Assistant Professor, Political Science, Bowling Green State University)
Much of the research on minority representation in the U.S. House has focused on how group preferences are reflected in bill sponsorship, cosponsorship, and recorded votes. However, left unexamined is how legislators can represent group interests through speech on the floor – specifically one-minute speeches. I turn to these speeches and examine the conditions under which members speak on behalf of minority groups. In an analysis of speeches from the 109th Congress, I look specifically at the Latino population, and find the effect of descriptive and substantive representatives to vary across issue area in terms of when a member delivers a speech and the content they choose to deliver. The research in this paper has implications for our understanding of legislative position-taking before a recorded vote, as well as offering a more nuanced discussion of how minorities are best represented in Congress.
Discussant:Eline de Rooij, Simon Fraser University/p>

A Dynamic Hierarchical Bayesian Measurement Model of Political Risk
Jane Lawrence Sumner (PhD candidate, Political Science, Emory University)
Political risk has long been theorized as a key factor that drives the allocation of foreign direct investment (FDI). Frequently defined as the likelihood a host government will engage in actions that threaten an investor’s profitability, political risk is thought to be remedied by political institutions that allow countries to make more credible commitments to honor their pre-investment promises. As a latent variable, scholars have struggled with ways to link institutions to FDI flows in the absence of a measure of political risk. This paper presents a dynamic hierarchical model of political risk that aims to address this problem by producing from observable indicators a single time-series cross-sectional measure of political risk. Additionally, this model produces estimates that allow scholars to measure within-country variation in political risk, a previously unestimated quantity. The model itself is flexible enough to be of use to comparative scholars more generally to estimate latent variables at multiple levels of a hierarchy from observable indicators produced at different levels of that hierarchy.
Discussant: Sandra Marquart-Pyatt, Michigan State University

What words can tell: Effects of emotive and vague words on voters’ interpretation and evaluation of election campaign proposals
Elina Lindgren (PhD candidate, Political Science, University of Gothenburg)
During campaigns, politicians frequently use emotive and vague words such as freedom, change and equality, to attract attention and muster citizen support. We know from numerous studies that creative use of such words can impact voters’ evaluation of political information and candidates. However, we know less about the precise mechanisms that are at play when citizens’ attitudes change. This paper looks further into one such mechanism, namely that emotive and vague words impact the way voters interpret parties’ intentions with campaign proposals. More specifically, the paper addresses the questions of whether emotive and vague words affect voters’ interpretation of intentions with campaign policy proposals, and what impact this have on whether they like the proposed policies or not. The paper make a difference between emotive words that are neutral to ideology, and those that are associated with left or right politics, and compares the effects that these different types of words have on interpretation. Results from experimental analyses indicate that emotive and vague words influence peoples’ interpretation of what proposed policies intends to and what outputs they expect, and that this in turn affect whether they like the proposed policies or not.
Discussant: Tina Fetner, McMaster University

Can Voting Aid Applications Mobilize Citizens? A Field Experiment in a Low Income Neighborhood, in the Context of the 2014 Quebec Election
Valérie-Anne Mahéo (PhD candidate, Political Science, McGill University)
Across Western democracies, Internet and Voting Aid Applications (VAA) are increasingly used to inform and mobilize the public. However, these ‘apps’ and websites mostly preach to the converted. They are used by educated and interested individuals, while less educated and politically marginalized citizens do not know of their existence. Hence, we do not exactly know what is the full political potential of these online information tools. If these voting apps were made available to those most in need of information and mobilization, could they stimulate political attitudes and electoral participation? Using a randomized field experiment, in the context of the provincial electoral campaign in Quebec, in 2014, I evaluate whether the use of a VAA can have meaningful effects on political knowledge, interest, information seeking behavior and electoral behavior. This experiment innovates in the sense that it reaches an under-­‐mobilized and under-­‐researched section of the population: the socio-­‐economically disadvantaged citizens.
Discussant: Heather Ondercin, University of Mississippi

Best of All Plausible Worlds? Checking Robustness of Time-Series Cross-Sectional Models with Fictitious Plausible Alternate Treatments
Evangeline Reynolds (PhD candidate, Political Science, University of Illinois)
This paper describes a new robustness framework for observational panel data analyses, motivated by the potential outcomes framework. Based on knowledge of the data generating process for the key explanatory variable, a set of plausible but fictitious alternate treatments is created. Estimating models with many fictitious data vectors in addition to the true independent variable yields a distribution of test statistics which could have resulted under the plausible alternate treatments. The researcher adjusts her confidence in rejecting the null according to the true treatment test statistic’s position within this distribution. This approach does not afford us with a one-size-fits all routine, therefore I demonstrate the method using three examples from the literature.
Discussant: Tiffany D. Barnes, University of Kentucky

Timing of Congressional Position Taking: An Estimator for a Spatial Duration Model with Competing Risks
Emily Schilling (PhD candidate, Political Science, University of Iowa)
Previous research on the timing of position taking has focused on the factors that lead a particular member of Congress to announce their position on a particular issue. It is obvious from these earlier works that position announcements are strategic in nature but for the most part none of this work has captured the interaction between legislators. Using position on NAFTA, this paper will adapt a log-normal spatial duration model, looking to see how ideologically similar legislators will signal to one another on when they should announce their position. Using the imputation algorithm developed in Boehmke, Hays, and Schilling (w.p.), this paper will be able to account for the competing risks. The imputation algorithm imputes values of the errors through rejection sampling to obtain draws of the dependent variable that are greater than the censoring point. Previous work on competing risks cannot be used in this situation since they do not allow for the spatial dependence that exists between legislators. Legislators at each time period can announce in favor or in opposition of NAFTA. When legislators announce their position, this will increase the likelihood that those legislators that are similar to them will announce their position as well.
Discussant: Regina Branton, University of North Texas

Courts and Conflict: Examining the Causal Mechanisms of Independent Judiciaries and Domestic Conflict
Jacqueline Sievert (PhD candidate, Political Science, SUNY at Buffalo)
Can adopting judicial independence reduce the likelihood of experiencing domestic conflict? This paper develops a game theoretic model that shows that authoritarian leaders can allow judicial independence to learn about the resolve of an aggrieved group within society. This in turn allows the regime to determine what level of concessions will satisfy the group’s demands and thereby prevent mobilization against the state. As the difference between the level of concessions required to satisfy groups of different levels of resolve grows, authoritarian regimes become more likely to allow independent judicial decision making. Additionally, the more important the issue is to the regime, the less likely they are to allow independent decision making. These findings have important implications both for understanding variation in judicial independence and the likelihood of civil unrest and are tested using propensity score matching to estimate the effect of the adoption of independent judiciaries on domestic conflict.
Discussant: Shawna Metzger, National University of Singapore

Support for Public Provision of a Private Good with Top-Up and Opt-Out: A Controlled Laboratory Experiment
Stephanie Thomas (PhD candidate, Economics, McMaster University)
Co-authors: Neil Buckley (York University); Katherine Cuff (McMaster University); Jeremiah Hurley (McMaster); Stuart Mestelman (McMaster University); and David Cameron (McMaster University)
This paper presents the results of a revealed-choice experiment testing the theoretical predictions of political economy models regarding public support for a publicly provided private good financed with proportional income taxes when individuals can purchase the good privately and either continue to consume public provision (‘top-up’) or forego public provision (‘opt-out’), but in each case continue to pay income taxes. Our laboratory results confirm behavior is consistent with the predicted majority-preferred tax rate under mixed financing with top-up, but we identify preferences for significantly higher rates of public provision than predicted under mixed financing with opt-out. Using non-parametric regression analysis, we explore the relationship between individuals’ top-up and opt-out decisions and both their income levels and the implemented tax rates.
Discussant: Mirya Holman, Florida Atlantic University

Cuing the Gap: Gender and Pscyhological Orientations to Politics
Melanee Thomas (Assistant Professor, Political Science, University of Calgary)
Co-authors: Allison Harell (Université du Québec à Montréal) and Tania Gosselin (Université du Québec à Montréal)
Using a unique experimental design, we examine the ways in which gendered messages about women’s political engagement influences their reported levels of interest, knowledge and efficacy. We use two forms of gendered messages: supply-side arguments in which we attribute women’s underrepresentation to their lack of interest and demand-side arguments that attribute women’s underrepresentation to institutional discrimination. These messages are contrasted against a control group that received no message. Our findings show that framing women’s political underrepresentation as a question of demand and/or institutional biases eliminates the gender gap in political knowledge, in part because it may disrupt dominant narratives about women in politics. We also find that emotions may be a moderating variable, though not in the direction the literature suggests. Finally, contrary to survey research, our results suggest that other historically underrepresented political identities – in this case, non-official language speakers and non-citizens – suppress men’s, but not women’s, political knowledge.
Discussant: Joanne Miller, University of Minnesota

2014 Participants

Lee Ann Banaszak, Professor, Political Science, Penn State
Tiffany Barnes, Assistant Professor, Political Science, Kentucky
Nichole Bauer, PhD candidate, Political Science, Indiana
Laurie Beaudonnet, Post-Doctoral Fellow, Political Science, Montreal
Katrine Beauregard, PhD candidate, Political Science, Calgary
Karen Bird, Associate Professor, Political Science, McMaster
Regina Branton , Assistant Professor, Political Science, North Texas
Rebecca Casey, graduate student, Sociology, McMaster
Meryll David-Ismayil, Political Science
Eline de Rooij, Assistant Professor, Political Science, Simon Fraser
Claudia Diaz-Rios, PhD student, Political Science, McMaster
Michelle Dion, Associate Professor, Political Science, McMaster
Claire Evans, Undergraduate, Political Science, Lousiana State
Tina Fetner, Associate Professor, Sociology, McMaster
Constanza Figueroa-Schibber, PhD candidate, Political Science, Washington U. in St. Louis
Carol Galais , Post-Doctoral Fellow, Political Science, Montréal
Elisabeth Gidengil, Professor, Political Science, McGill
Kelly Gleason, PhD candidate, Political Science, Wisconsin-Milwaukee
Anita Gohdes, PhD candidate, Political Science, Mannheim
Nicole Goodman, Assistant Professor, Political Science, McMaster
Elizabeth Goodyear-Grant, Associate Professor, Political Science, Queen’s
Alicia Gopaul, graduate student, Political Science, McMaster
Laura Grigolon, Assistant Professor, Economics, McMaster
Mirya Holman, Assistant Professor, Political Science, Florida Atlantic
Nicole Kalaf-Hughes, Assistant Professor, Political Science, Bowling Green State
Jane Lawrence Sumner, PhD candidate, Political Science, Emory
Alison Leanage, graduate student, Sociology, McMaster
Hyunji Lee, Instructor, Political Science, British Columbia
Elina Lindgren, PhD candidate, Political Science, Gothenburg
Valérie-Anne Mahéo, PhD candidate, Political Science, McGill
Sandra Marquart-Pyatt, Assistant Professor, Sociology, Michigan State
Shawna Metzger, Assistant Professor, Political Science, National U. of Singapore
Joanne Miller, Associate Professor, Political Science, Minnesota
Sara Mitchell, Professor, Political Science, Iowa
Amanda Murdie, Assistant, Political Science, Missouri
Shane Nordyke, Assistant Professor, Political Science, South Dakota
Heather Ondercin, Assistant Professor, Political Science, Mississippi
Ruth Repchuck, PhD student, Sociology, McMaster
Evangeline Reynolds, PhD candidate, Political Science, Illinois
Halina Sapeha, PhD candidate, Political Science, McMaster
Emily Schilling, PhD candidate, Political Science, Iowa
Saundra Schneider, Professor, Political Science, Michigan State
Jessica Shearer, Health Policy, McMaster
Jacqueline Sievert, PhD candidate, Political Science, SUNY at Buffalo
Diana Singh, PhD student, Sociology, McMaster
Laura Stephenson, Associate Professor, Political Science, Western
Jakana Thomas, Assistant Professor, Political Science, Michigan State
Melanee Thomas, Assistant Professor, Political Science, Calgary
Stephanie Thomas, PhD candidate, Economics, McMaster
Rochelle Wijesingha, graduate student, Sociology, McMaster
Marisa Young, Assistant Professor, Sociology, McMaster