A Seemingly Unrelated Regression Model for Analyzing Multiparty Elections

This paper develops an estimator for models of election returnsin multiparty elections. It shares the same functional formasthe Katz–King estimator but is computationally simpler,can be used with any number of parties, and is based on moreconventional distributional assumptions. Small sample propertiesof the estimator are derived, which makes it particularly usefulin many of the applications where there are a relatively smallnumber of voting districts. The distributional assumptions arecontained in two elements. The first treats the observed votesas the outcomes resulting from sampling the voters in each district.The second stochastic element arises from the usual treatmentof the stochastic term in a regression model, namely, the inabilityof the included variables and the linear form to match the underlyingprocess perfectly. The model is then used to analyze the 1993Polish parliamentary elections. The results from this analysisare used to develop Monte Carlo experiments comparing severaldifferent yet feasible estimators. The conclusion is that anumber of accessible estimators, including the standard seeminglyunrelated regression model and the Beck–Katz model withpanel-corrected standard errors, are all good choices.     

Selection Bias and Continuous-Time Duration Models: Consequences and a Proposed Solution

This article analyzes the consequences of nonrandom sample selection for continuous-time duration analyses and develops a new estimator to correct for it when necessary. We conduct a series of Monte Carlo analyses that estimate common duration models as well as our proposed duration model with selection. These simulations show that ignoring sample selection issues can lead to biased parameter estimates, including the appearance of (nonexistent) duration dependence. In addition, our proposed estimator is found to be superior in root mean-square error terms when nontrivial amounts of selection are present. Finally, we provide an empirical application of our method by studying whether self-selectivity is a problem for studies of leaders' survival during and following militarized conflicts.     

An Alternative Approach to Estimating Who is Most Likely to Respond to Changes in Registration Laws

Scholars often seek to understand which individuals are most responsive to the change in some treatment. Such work inevitably faces issues of identification. When the dependent variable is binary, the assumption that the largest effect occurs where p = 0.5 is also encountered. I apply Manski’s [(1995). Identification problems in the social sciences. Cambridge: Harvard University Press] non-parametric Bounds approach, which relaxes the functional form and distributional assumptions found in traditional models, in an attempt to resolve the long standing debate on which types of individuals are most affected by changes in registration laws. Under the standard assumption that treats the selection of registration laws as exogenous, the results revise the current understanding. By exploring the power of various behavioral assumptions, new insights into the study of policy changes emerge, calling into question some of the assumptions that are standard in the literature.

Using Auxiliary Data to Estimate Selection Bias Models, with an Application to Interest Group Use of the Direct Initiative Process

Recent work in survey research has made progress in estimatingmodels involving selection bias in a particularly difficultcircumstance—all nonrespondents are unit nonresponders,meaning that no data are available for them. These models arereasonably successful in circumstances where the dependent variableof interest is continuous, but they are less practical empiricallywhen it is latent and only discrete outcomes or choices areobserved. I develop a method in this article to estimate thesemodels that is much more practical in terms of estimation. Themodel uses a small amount of auxiliary information to estimatethe selection equation parameters, which are then held fixedwhile estimating the equation of interest parameters in a maximum-likelihoodsetting. After presenting Monte Carlo analyses to support themodel, I apply the technique to a substantive problem: Whichinterest groups are likely to to be involved in support of potentialinitiatives to achieve their policy goals?     

Political interest and the decision to vote: a self-selection problem

ABSTRACT

Political interest is frequently considered as one of the most important driving factors of voter turnout. This article uses data from the German Socio-Economic Panel to investigate the causal relationship between political interest and the decision to vote. It introduces a recursive simultaneous equation model which is estimated via bivariate probit regression and thus avoids major drawbacks of the method of instrumental variables. The results suggest that the effect of political interest on the decision to vote is not causal, but rather results from self-selection.     

The Epistemological Foundations of Public Choice Theory

The new political economy - public choice - has contributed considerably to a long-awaited inter-marriage between political science and economics. The economic approach to the interpretation of the public sector - politics as well as administration - has resulted in a number of interestingly relevant models. At the same time modelling political behaviour on the basis of the economic man perspective gives rise to puzzling difficulties creating much controversy. The article attempts to establish in principle the ethically neutral and scientifically objective nature of the public choice approach. It elaborates on the crucial concept of self-interest in the public choice behaviour assumptions in order to show that the criticism from the public interest adherents may be countered.     

Random Coefficient Models for Time-Series-Cross-Section Data: Monte Carlo Experiments

Jonathan N. Katz Division of the Humanities and Social Sciences, California Institute of Technology, Pasadena, CA 91125 e-mail: jkatz{at}caltech.edu e-mail: nathaniel.beck{at}nyu.edu (corresponding author) This article considers random coefficient models (RCMs) fortime-series–cross-section data. These models allow forunit to unit variation in the model parameters. The heart ofthe article compares the finite sample properties of the fullypooled estimator, the unit by unit (unpooled) estimator, andthe (maximum likelihood) RCM estimator. The maximum likelihoodestimator RCM performs well, even where the data were generatedso that the RCM would be problematic. In an appendix, we showthat the most common feasible generalized least squares estimatorof the RCM models is always inferior to the maximum likelihoodestimator, and in smaller samples dramatically so. Authors' note: We gratefully acknowledge the financial supportof the National Science Foundation. Katz also acknowledges thesupport of the Center for Advanced Study in the Behavioral Sciences.We are thankful to Jake Bowers, Rob Franzese, Andy Gelman, SandyGordon, Bill Greene, and Luke Keele for comments; to Larry Bartelsfor always reminding us that our judgment may outperform thedata; as well as to the anonymous reviewers of Political Analysis.     

Building policy scales from manifesto data: A referential content validity approach

Researchers using scales based on MRG/CMP/MARPOR's manifesto dataset face a bewildering array of different scales. The validation of these scales has tended to focus on external, convergent validity. The actual content of these scales has received less attention and the choice of the manifesto components which make up these scales has often been conducted by either opaque or questionable methods. This article develops a critique of existing methods of component selection and proposes a new method of component selection based on the covariance of components with ‘naïve’ provisional scales, which are refined in an iterative process. It uses this method to construct a set of comparable one (general left–right) and two (economic and social) dimensional scales – filling a gap in the existing body of scales that will allow researchers to compare dimensionality across models without inadvertently comparing different assumptions that underlie the construction of the scales.     

Gendering women and politics in contemporary China

This article suggests moving beyond "state-society" models of Chinese politics in order to more effectively integrate "gender" as a category of analysis. The article examines a number of empirical examples of women’s activism in Hong Kong prior to the 1997 handover to China as a way of gaining insights into which variables might more effectively nuance "state-society" models so as to highlight the gendered aspects of women and politics. In the Hong Kong case, such variables include ethnicity, class, race, nationalism, and feminism. The article finds 1)that a number of these variables also apply equally well to research on women and politics in Hong Kong and to gender studies focusing on the contemporary People’s Republic of China; and 2)that disaggregating concepts like "state" and “society” using variables like gender, race, and class provides a more complex understanding of the process of politics in Chinese societies. She received her Ph.D. degree in Political Science from the University of Wisconsin-Madison. She has taught in both Hong Kong and the United States. Her work on the women’ movement in pre-1997 Hong Kong will soon be published in an edited volume from University of British Columbia Press.     

Instrumental Variables Estimation in Political Science: A Readers’ Guide

The use of instrumental variables regression in political science has evolved from an obscure technique to a staple of the political science tool kit. Yet the surge of interest in the instrumental variables method has led to implementation of uneven quality. After providing a brief overview of the method and the assumptions on which it rests, we chart the ways in which these assumptions are invoked in practice in political science. We review more than 100 articles published in the American Journal of Political Science, the American Political Science Review, and World Politics over a 24‐year span. We discuss in detail two noteworthy applications of instrumental variables regression, calling attention to the statistical assumptions that each invokes. The concluding section proposes reporting standards and provides a checklist for readers to consider as they evaluate applications of this method.     

International Organization as a Seal of Approval: European Union Accession and Investor Risk

Much of the literature on international institutions argues that membership regularizes expectations about members' future behavior. Using the accession of the postcommunist countries as a test case, this article argues that the EU can send strong signals to financial markets about the trajectory of a particular country. Examining spreads on sovereign debt from 1990 to 2006, this article shows that closing negotiation chapters on domestic economic policy—in other words, receiving a seal of approval from Brussels that previously existing policy reform is acceptable to the wider EU—substantially decreases perceptions of default risk in those countries. That decrease operates independently from policy reform that the country has taken and is also distinct from selection processes (modeled here with new variables, including UNESCO World Heritage sites and domestic movie production, that proxy for cultural factors). Thus, this particular international organization has played an important role in coordinating market sentiment on members, conferring confidence that policy reform alone could not accomplish.     

OLS is AOK for ACE: A Regression-Based Approach to Synthesizing Political Science and Behavioral Genetics Models

There is a growing interest in empirically exploring the biological underpinnings of political attitudes and behavior. Heritability studies are a primary vehicle for conducting such investigations and data sets rich in political phenotypes are becoming broadly accessible. A bottleneck exists, however, in exploiting these opportunities because they involve a statistical re-tooling for political scientists and require a conceptual shift that has substantial implications for the field’s traditional theoretical models. Methodologically, most twin studies rely on structural equation models unfamiliar to political scientists. We show this methodological bottleneck is easily navigable; it is the lesser discussed shift in theoretical assumptions poses the larger problem to integrating biological elements into the study of political attitudes and behavior. To address these issues we provide a detailed introduction to a regression-based method for analyzing genetic influence on political attitudes and behaviors that will be methodologically intuitive to political scientists with even minimum quantitative training. In doing so, we provide a platform for bridging important conceptual divides between political science and behavioral genetics.     

Means of support,fertility, and social reproduction among peasants: three approaches

This article examines interpretations regarding the relationship between means of support and fertility, taking into account views on social reproduction among peasants, with emphasis on the way they are expressed in terms of indicators. The author begins with a brief summary of theories put forth by Malthus, and then examines Neo-Malthusian models which use as an independent variable, the amount of land and as a dependent variable, the number of children. He notes the existence of a positive ratio between these variables, and then demonstrates that in Chayanov's model, the number of children is used as an independent variable, while the idea of a positive ratio with the amount of land is still maintained. He compares these assumptions with those of Palerm, who suggests the existence of an inverse ratio between the amount of land available and fertility; this leads to policy implications contrary to those arising from neo-Malthusian models. The paper ends with a discussion of assumptions about the rationality, motivations, and guiding values the different approaches attribute to economic and demographic behavior among peasants.     

Congressional Intelligence Oversight: The Electoral Disconnection

This article seeks to reconcile congressional oversight models in theory with oversight realities in intelligence. For nearly three decades, political scientists have argued that Congress controls the bureaucracy – and in surprisingly efficient ways. Yet the history of intelligence oversight suggests the opposite. We take a fresh look at the logic and empirics of police patrol and fire alarm models and find that neither explains intelligence oversight well. Both rely on assumptions, such as the presence of strong and plentiful interest groups, which characterize domestic policy but not US intelligence policy. Our data – comparing committee hearing activities, legislative productivity, and interest groups across different policy domains between 1985 and 2005 – reveal that oversight varies dramatically by policy issue, and that intelligence almost always ranks at the bottom. Ironically, the same electoral incentives that generate robust oversight in some policy areas turn out to be far weaker in intelligence.     

Managerial Quality,Administrative Performance and Trust in Governance: Can We Point to Causality?

The relationship between managerial quality, administrative performance and citizens' trust in government and in public administration systems is a field of study that so far has not received adequate scholarly attention. This article explores some interrelationships between these variables and empirically tests between causality, if it exists, between performance and trust. Applying a technique of structural equation modelling (SEM) with LISREL 8.3 the study examined a sample of 345 Israeli citizens and compared three alternative models. The second model that showed a quality → performance → trust relationship fitted the data best. However, the third model also had some advantages worthy of elaboration. Thus, we concluded that administrative performance may be treated as a precondition to trust in governance rather than trust serving as the precondition to performance. The article ends with further discussion of the findings and their meaning in light of the democratic, bureaucratic and new public management theory.     

Political Choice, Public Policy, and Distributional Outcomes

I address the functioning of the U.S. governing system by analyzing distributional outcomes from 1947 to 2000. The key question is whether public policy influences distributional outcomes. The macropolitics model and power resource theory suggest that left policies should equalize the distribution of income. I utilize single equation error correction models to assess the impact of policy on income inequality through two mechanisms—market conditioning and redistribution. Since nearly every government action influences markets in some way, I examine policy in the aggregate rather than focusing only on policies explicitly designed to redistribute income. The analysis indicates that policy influences inequality through both mechanisms, with left policy producing more equality. The results are consistent with power resource theory and strongly support the macropolitics model. Furthermore, I find that market conditioning is as important as, and works in tandem with, explicit redistribution.     

Bias in Conditional and Unconditional Fixed Effects Logit Estimation

Fixed-effects logit models can be useful in panel data analysis,when N units have been observed for T time periods. There aretwo main estimators for such models: unconditional maximum likelihoodand conditional maximum likelihood. Judged on asymptotic properties,the conditional estimator is superior. However, the unconditionalestimator holds several practical advantages, and thereforeI sought to determine whether its use could be justified onthe basis of finite-sample properties. In a series of MonteCarlo experiments for T < 20, I found a negligible amountof bias in both estimators when T 16, suggesting that a researchercan safely use either estimator under such conditions. WhenT < 16, the conditional estimator continued to have a verysmall amount of bias, but the unconditional estimator developedmore bias as T decreased.     

Patterns of democracy reconsidered: The ambiguous relationship between corporatism and consensualism

Arend Lijphart's Patterns of Democracy, similar to most of his work, elicited fierce scientific debate. This article replicates some of the analyses proposed in its second edition (published in 2012) in the light of the critiques received by the first edition (published in 2009). It primarily examines the relationship between institutional setup and interest group representation, disentangling the effect of consensualism from that of corporatism on issues such as macroeconomic performance and governance capabilities. The article further deepens our understanding of the complex causal mechanisms connecting these variables, proposing a more sophisticated empirical investigation that emphasises selection effects and conjunctural causation.     

Adjustment of an expologistic function to the evolution of Mexico's total population, 1930-1945

The author makes the case that none of the mathematical models currently in use for estimating population dynamics can be suitably applied to Mexico's natural growth "due to the fact that the assumptions underlying [their] mathematical representations do not correspond to the observed dynamics of the components of natural growth. The purpose of this article is to adjust a mathematical function to the evolution of Mexico's total population between 1930 and 1985, adequately reproducing the evolution of natality and mortality observed in that period." (SUMMARY IN ENG)     

Black-box models of candidate evaluation

All contemporary models of candidate evaluation are memory-based models in that they treat the direction and strength of evaluation as a function of the mix of positively and negatively valued (valenced) information retrieved from memory. Yet, oddly enough, despite the assumption that memory mediates judgment, none of the major models looks at the processes involved in what information voters recall and how that evidence was integrated into a summary evaluation. In this sense then, political science models of vote choice are black-box models: They are silent about how voters actually go about interpreting information and integrating the evidence into a summary evaluation of the candidates. In this article we critique the major political science models, call attention to the implicit assumptions they make about what evidence is assumed to be in memory, and conclude with an argument for introducing process into our explanations of vote choice.