Political Analysis

Published by Oxford University Press (OUP)
Online ISSN: 1476-4989
Print ISSN: 1047-1987
Combined RNC/DNC credentialled blog citation network aggregated over all time points, edges weighted by time-frequency. Purple node is credentialled by both the RNC and DNC. 
Methods for analysis of network dynamics have seen great progress in the past decade. This article shows how Dynamic Network Logistic Regression techniques (a special case of the Temporal Exponential Random Graph Models) can be used to implement decision theoretic models for network dynamics in a panel data context. We also provide practical heuristics for model building and assessment. We illustrate the power of these techniques by applying them to a dynamic blog network sampled during the 2004 US presidential election cycle. This is a particularly interesting case because it marks the debut of Internet-based media such as blogs and social networking web sites as institutionally recognized features of the American political landscape. Using a longitudinal sample of all Democratic National Convention/Republican National Convention–designated blog citation networks, we are able to test the influence of various strategic, institutional, and balance-theoretic mechanisms as well as exogenous factors such as seasonality and political events on the propensity of blogs to cite one another over time. Using a combination of deviance-based model selection criteria and simulation-based model adequacy tests, we identify the combination of processes that best characterizes the choice behavior of the contending blogs.
The minimization approach to balancing a sequential experiment with discrete covariates
Consistent marginal and joint distributions of two binary covariates
Improvements in balance and precision in MVN-correlated data (r ¼ 0:6) when an extreme outlier arrives early (top row) or late (bottom row) in experiment. See Section 4.3.  
Sequential blocking outperforms complete randomization in the presence of extreme outliers
In typical political experiments, researchers randomize a set of households, precincts, or individuals to treatments all at once, and characteristics of all units are known at the time of randomization. However, in many other experiments, subjects "trickle in" to be randomized to treatment conditions, usually via complete randomization. To take advantage of the rich background data that researchers often have (but underutilize) in these experiments, we develop methods that use continuous covariates to assign treatments sequentially. We build on biased coin and minimization procedures for discrete covariates and demonstrate that our methods outperform complete randomization, producing better covariate balance in simulated data. We then describe how we selected and deployed a sequential blocking method in a clinical trial and demonstrate the advantages of our having done so. Further, we show how that method would have performed in two larger sequential political trials. Finally, we compare causal effect estimates from differences in means, augmented inverse propensity weighted estimators, and randomization test inversion.
A hypothetical distribution of feeling thermometer data across two states with two districts each 
Summaries of hypothetical thermometer data in Table 1 using conventional approaches 
A hypothetical Democratic primary as contests between issues 
Features of various scoring models 
This article presents a new model for scoring alternatives from “contest” outcomes. The model is a generalization of the method of paired comparison to accommodate comparisons between arbitrarily sized sets of alternatives in which outcomes are any division of a fixed prize. Our approach is also applicable to contests between varying quantities of alternatives. We prove that under a reasonable condition on the comparability of alternatives, there exists a unique collection of scores that produces accurate estimates of the overall performance of each alternative and satisfies a well-known axiom regarding choice probabilities. We apply the method to several problems in which varying choice sets and continuous outcomes may create problems for standard scoring methods. These problems include measuring centrality in network data and the scoring of political candidates via a “feeling thermometer.” In the latter case, we also use the method to uncover and solve a potential difficulty with common methods of rescaling thermometer data to account for issues of interpersonal comparability.
Many social processes are stable and smooth in general, with discrete jumps. We develop a sequential segmentation spline method that can identify both the location and the number of discontinuities in a series of observations with a time component, while fitting a smooth spline between jumps, using a modified Bayesian Information Criterion statistic as a stopping rule. We explore the method in a large-n, unbalanced panel setting with George W. Bush's approval data, a small-n time series with median DW-NOMINATE scores for each Congress over time, and a series of simulations. We compare the method to several extant smoothers, and the method performs favorably in terms of visual inspection, residual properties, and event detection. Finally, we discuss extensions of the method. © The Author 2009. Published by Oxford University Press on behalf of the Society for Political Methodology. All rights reserved. For Permissions, please email: [email protected] /* */
In this article, we respond to Shultziner's critique that argues that identical twins are more alike not because of genetic similarity, but because they select into more similar environments and respond to stimuli in comparable ways, and that these effects bias twin model estimates to such an extent that they are invalid. The essay further argues that the theory and methods that undergird twin models, as well as the empirical studies which rely upon them, are unaware of these potential biases. We correct this and other misunderstandings in the essay and find that gene-environment (GE) interplay is a well-articulated concept in behavior genetics and political science, operationalized as gene-environment correlation and gene-environment interaction. Both are incorporated into interpretations of the classical twin design (CTD) and estimated in numerous empirical studies through extensions of the CTD. We then conduct simulations to quantify the influence of GE interplay on estimates from the CTD. Due to the criticism's mischaracterization of the CTD and GE interplay, combined with the absence of any empirical evidence to counter what is presented in the extant literature and this article, we conclude that the critique does not enhance our understanding of the processes that drive political traits, genetic or otherwise.
Ordinal variables—categorical variables with a defined order to the categories, but without equal spacing between them—are frequently used in social science applications. Although a good deal of research exists on the proper modeling of ordinal response variables, there is not a clear directive as to how to model ordinal treatment variables. The usual approaches found in the literature for using ordinal treatment variables are either to use fully unconstrained, though additive, ordinal group indicators or to use a numeric predictor constrained to be continuous. Generalized additive models are a useful exception to these assumptions. In contrast to the generalized additive modeling approach, we propose the use of a Bayesian shrinkage estimator to model ordinal treatment variables. The estimator we discuss in this paper allows the model to contain both individual group—level indicators and a continuous predictor. In contrast to traditionally used shrinkage models that pull the data toward a common mean, we use a linear model as the basis. Thus, each individual effect can be arbitrary, but the model "shrinks" the estimates toward a linear ordinal framework according to the data. We demonstrate the estimator on two political science examples: the impact of voter identification requirements on turnout and the impact of the frequency of religious service attendance on the liberality of abortion attitudes.
Many claims about political behavior are based on implicit assumptions about how people think. One such assumption, that political actors use identical conjectures when assessing others’ strategies, is nested within applications of widely used game-theoretic equilibrium concepts. When empirical findings call this assumption into question, the self-confirming equilibrium (SCE) concept provides an alternate criterion for theoretical claims. We examine applications of SCE to political science. Our main example focuses on the claim of Feddersen and Pesendorfer that unanimity rule can lead juries to convict innocent defendants (1998. Convicting the innocent: The inferiority of unanimous jury verdicts under strategic voting. American Political Science Review 92:23–35). We show that the claim depends on the assumption that jurors have identical beliefs about one another's types and identical conjectures about one another's strategies. When jurors’ beliefs and conjectures vary in ways documented by empirical jury research, fewer false convictions can occur in equilibrium. The SCE concept can confer inferential advantages when actors have different beliefs and conjectures about one another.
In a standard spatial model of Parliamentary roll call voting with quadratic utility, the ligislator votes for the policy outcome corresponding to Yea if her utility for Yea is greater than her utility for Nay. The voting decision of the legislator is modeled as a function of the difference between these two utilities. This utility difference has a simple geometric interpretation that can be exploited to estimate legislator ideal points and roll call parameters in a standard framework where the stochastic portion of the utility function is normally distributed. The geometry is almost identical to that used in Poole (2000) to develop a non-parametric unfolding of binary choice data and the algorithms developed by Poole (2000) can be easily modified to implement the standard maximum likelihood model.
Testing and estimating formal models of political behavior has not advanced as far as theoretical applications. One of the major literatures in formal theory is the spatial model of electoral competition which has its origins in the work of Black (1948) and Downs (1957). These models are used to make predictions about the policy positions candidates take in order to win elections. A lack of data on these candidate positions, especially challengers who never serve in Congress, has made direct testing of these models on congressional elections difficult. Recently, researchers have begun to incorporate campaign finance into the standard Downsian model. These models of position-induced contributions examine the tradeoff that candidates make between choosing positions favorable to interest group contributors and positions favorable to voters. A major premise of these models is that interest group contributions are based on the policy positions of candidates. This has been borne out empirically in the case of incumbents, but not challengers. To test key hypotheses of these models, we develop a simple spatial model of position-induced campaign contributions where the PAC's decision to contribute or abstain from each race is a function of the policy distance between the PAC and the candidates. We use data from political action committee contributions in order to estimate the locations of incumbents, challengers and PACs. Our model reliably estimates the spatial positions as well as correctly predicts nearly 74 percent of the contribution and abstention decisions of the PACs. Conditional upon making a contribution, we correctly predict the contribution in 94 percent of the cases. These results are strong evidence for position-induced campaign contributions. Furthermore, our estimates of candidate positions allow us to address issues of platform convergence between candidates.
Indicators of Social Capital: Roper Particpation Trends
DBB Needham Indicators of Social Capital: Civic Participation
Civic Engagement, 1972:3-2000:2
Direction of Granger Causality Between Civic Engagement and Trust
Interpersonal Trust 1972:2-2000:4
Interest in social capital has grown as it has become apparent that it is an important predictor of collective well-being. Recently, however, attention has shifted to how levels of social capital have changed over time. But focusing on how a society moves from one level of social capital to another over time requires better macro level measures.Better measures are required to test even basic hypotheses such as the establishing the direction of causality between the two components of social capital. In the following analysis, I develop macro measures of social capital through the development of longitudinal measures of civic engagement and interpersonal trust. I,then, use these measures to test a basic assumption about social capital. I, first, perform a direction of causality test to substantiate the causal direction between the two components of social capital. Second, I model civic engagement as a function of the time and monetary-related resources required for civic participation and interpersonal trust as a function of long term trends in civic engagement and a set of controls for collective experiences. The result is more than just the ¯rst over time measures of social capital, but also an increase in our understanding of social capital as a macro process with complex causes and effects.
Questions of causation are important issues in empirical research on political behavior. Most of the discussion of the econometric problems associated with multi-equation models with reciprocal causation has focused on models with continuous dependent variables (e.g. Markus and Converse 1979; Page and Jones 1979). Since many models of political behavior involve discrete or dichotomous dependent variables, this paper turns to two techniques which can be employed to estimate reciprocal relationships between dichotomous and continuous dependent variables. One technique which I call two-stage probit least squares (2SPLS) is very similar to familiar two-stage instrumental variable techniques. The second technique, called two-stage conditional maximum likelihood (2SCML), may overcome problems associated with 2SPLS, but has not been used in the political science literature. First I show the properties of both techniques using Monte Carlo simulations. Then, I apply these techniques to an empirical example which focuses on the relationship between voter preferences in a presidential election and the voter's uncertainty about the policy positions taken by the candidates. This example demonstrates the importance of these techniques for political science research.
Dyadic data are common in the social sciences, although inference for such settings involves accounting for a complex clustering structure. Many analyses in the social sciences fail to account for the fact that multiple dyads share a single member, and that errors are thus likely correlated across these dyads. To address this clustering, we propose a nonparametric sandwich-type robust variance estimator for linear regression with dyadic data. We enumerate conditions for estimator consistency and extend our results to repeated observations, including directed dyads and longitudinal data. We examine empirical performance with simulations and an application in the study of international relations.
Examples of the method.  
Synder and Groseclose (2000) develop and apply an innovative method for detecting and estimating the frequency and magnitude of party influence in congressional roll call voting. This paper presents a framework for assessing to coefficient that the authors interpret as "party influence." The analysis reveals that, and shows why, the coefficient manifests two troublesome characteristics. The coefficient cannot discriminate between disparate types of party influence because the mapping between a types of partisan influence and signs of the coefficient is not one-to-one. Similarity, the coefficient has a responsiveness problem in that a marginal increase in one party's influence can cause the estimate of the coefficient to increase, decrease, or remain constant. Because the literature on parties in Congress emphasizes majority-constant. Because the literature on parties in Congress emphasizes majority-party strength, the inability of the coefficient to isolate party-specific effects is a serious drawback in the ongoing hunt for genuine party discipline.
Unemployment insurance policies are multidimensional objects, with variable waiting periods, eligibility duration, benefit levels, and asset tests, making intertemporal or international comparisons very difficult. Furthermore, labor market conditions, such as the likelihood and duration of unemployment, matter when assessing the generosity of different policies. In this article, we develop a new methodology to measure the generosity of unemployment insurance programs with a single metric. We build a first model with all characteristics of the complex unemployment insurance policy. Our model features heterogeneous agents that are liquidity constrained but can self-insure. We then build a second model, similar in all aspects but one: the unemployment insurance policy is one-dimensional (no waiting periods, eligibility limits, or asset tests, but constant benefits). We then determine which level of benefits in this second model makes society indifferent between both policies. We apply this measurement strategy to the unemployment insurance program of the United Kingdom.
Mean squared errors
This paper outlines a nonstationary, heterogeneous Markov model designed to estimate entry and exit transition probabilities at the micro-level from a time series of independent cross-sectional samples with a binary outcome variable. The model has its origins in the work of Moffitt (1993) and shares features with standard statistical methods for ecological inference. We show how ML estimates of the parameters can be obtained by the method-of- scoring, how to estimate time-varying covariate effects, and how to include non-backcastable variables in the model. The latter extension of the basic model is an important one as it strongly increases its potential application in a wide array of research contexts. The example illustration uses survey data on American presidential vote intentions from a five-wave panel study conducted by Patterson (1980) in 1976. We treat the panel data as independent cross sections and compare the estimates of the Markov model with the observations in the panel. Directions for future work are discussed.
The efficiency gains of the dynamic model over a model in which the ideal points are assumed iid. PSD denotes the posterior standard deviation.
Estimated posterior distribution of the location of the median justice for the dynamic ideal point model.
At the heart of attitudinal and strategic explanations of judicial behavior is the assumption that justices have policy preferences. In this paper we employ Markov chain Monte Carlo (MCMC) methods to fit a Bayesian measurement model of ideal points for all justices serving on the U.S. Supreme Court from 1953 to 1999. We are particularly interested in determining to what extent ideal points of justices change throughout their tenure on the Court. This is important because judicial politics scholars oftentimes invoke preference measures that are time invariant. To investigate preference change, we posit a dynamic item response model that allows ideal points to change systematically over time. Additionally, we introduce Bayesian methods for fitting multivariate dynamic linear models (DLMs) to political scientists. Our results suggest that many justices do not have temporally constant ideal points. Moreover, our ideal point estimates outperform existing measures and explain judicial behavior quite well across civil rights, civil liberties, economics, and federalism cases.
Binary, count, and duration data all code for discrete events occurring at points in time. Although a single data generation process can produce any of these three data types, the statistical literature is not very helpful in providing methods to estimate parameters of the same process from each. In fact, only a single theoretical process exists for which known statistical methods can estimate the same parameters --- and it is generally limited to count and duration data. The result is that seemingly trivial decisions about which level of data to use can have important consequences for substantive interpretations. We describe the theoretical event process for which results exist, based on time-independence. We also derive a new set of models for a time-dependent process and compare their predictions to those of a commonly used model. Any hope of avoiding the more serious problems of aggregation bias in events data is contingent on first deriving a much wider arsenal of statistical mode...
The dominant methodology for short-term forecasting of electoral outcomes uses trial-heat polls, where respondents report their current electoral preferences (not their election-day predictions). Election markets, where self-selected participants trade shares of candidates at prices predictive of election-day results, provide an alternative method that often produces more accurate forecasts. Consequently, increasing attention is being paid to this methodology. However, it is poorly understood and lacks theoretical justification. Surprisingly, the rationale for forecasting using trial-heat polls has not been completely developed either. We develop the justification for using both election markets and public opinion polls to forecast electoral outcomes, giving conditions under which each method performs ideally. For the ideal case, we prove (under the reasonable assumption that market participants are aware of the poll results) that the mean square prediction error for the market forecast is smaller than that of any forecast based on one or more polls. The case in which the assumptions supporting each method fail is also considered. It is often reasonable to expect that the best case results hold approximately, in which case the market forecast should also beat any poll-based forecast. We also compare the bias and variance of market and poll-based forecasts; our results suggest the utility of using the series of market prices to study the course of campaigns.
I examine a recently proposed solution to the ecological inference problem (King 1997). It is asserted that the proposed model is able to reconstruct individual-level behavior from aggregate data. I discuss in detail both the benefits and limitations of this model. The assumptions of the basic model are often inappropriate for instances of aggregate data. The extended version of the model is able to correct for some of these limitations. However, it is difficult in most cases to apply the extended model properly.
Measures of change in pairs of attitudinal variables can provide important insights into the structure of the political belief systems of mass publics. Panel data reveal evidence of the greater centrality of some idea elements rather than others in the context of short-term dynamic constraint. Specification of the theoretically relevant voter attributes makes it possible to test for expected structural differences connecting policy related predispositions and policy preferences; specification also makes it possible to test propositions involving the reciprocal effects of attitudes and emerging vote preferences. Some of the more helpful specifications disclose the extent to which population heterogeneity produces a blurred image of relationships when analysis is based on the total electorate rather than limited to voters or subsets of voters specified by theoretical criteria.
Normal Mixture: The Claw Density
Normal Mixture: Asymmetric Double Claw
Normal Mixture: Discrete Comb Density
Normal Mixture: Discrete Comb Density, Magnified Peaks
GENOUD for 1986 Labor PACs and Highways Transfers Data
We describe a new computer program that combines evolutionary algorithm methods with a derivative-based, quasi-Newton method to solve difficult unconstrained optimization problems. The program, called GENOUD (GENetic Optimization Using Derivatives), effectively solves problems that are nonlinear or perhaps even discontinuous in the parameters of the function to be optimized. When a statistical model's estimating function (for example, a log-likelihood) is nonlinear in the model's parameters, the function to be optimized will usually not be globally concave and may contain irregularities such as saddlepoints or discontinuous jumps. Optimization methods that rely on derivatives of the objective function may be unable to find any optimum at all. Or multiple local optima may exist, so that there is no guarantee that a derivative-based method will converge to the global optimum. We discuss the theoretical basis for expecting GENOUD to have a high probability of finding global optima. We conduct Monte Carlo experiments using scalar Normal mixture densities to illustrate this capability. We also use a system of four simultaneous nonlinear equations that has many parameters and multiple local optima to compare the performance of GENOUD to that of the Gauss-Newton algorithm in SAS's PROC MODEL.
There has been much discussion about how members of Congress desire money early in the campaign season. However, to date, models of how contributions are allocated during the electoral cycle have been lacking. Our analysis attempts to remedy this gap by providing and testing a model which specifies how the process by which bargaining between members of Congress and organized interests produces the pattern of donations observed over the course of the electoral cycle. The results suggest that strategic incumbents can receive money early in the campaign if they desire but that they are generally unwilling to pay the price of lower aggregate fundraising and greater provision of access. These findings, in turn, butress earlier empirical findings that question the instrumental value of early money; in particular, they imply that challengers have reasonably rational and informed expectations about how much money members of Congress are capable of raising over the electoral cycle and that the ...
We study rare events data, binary dependent variables with dozens to thousands of times fewer ones (events, such as wars, vetoes, cases of political activism, or epidemiological infections) than zeros (“nonevents”). In many literatures, these variables have proven difficult to explain and predict, a problem that seems to have at least two sources. First, popular statistical procedures, such as logistic regression, can sharply underestimate the probability of rare events. We recommend corrections that outperform existing methods and change the estimates of absolute and relative risks by as much as some estimated effects reported in the literature. Second, commonly used data collection strategies are grossly inefficient for rare events data. The fear of collecting data with too few events has led to data collections with huge numbers of observations but relatively few, and poorly measured, explanatory variables, such as in international conflict data with more than a quarter-million dyads, only a few of which are at war. As it turns out, more efficient sampling designs exist for making valid inferences, such as sampling all available events (e.g., wars) and a tiny fraction of nonevents (peace). This enables scholars to save as much as 99% of their (nonfixed) data collection costs or to collect much more meaningful explanatory variables. We provide methods that link these two results, enabling both types of corrections to work simultaneously, and software that implements the methods developed.
Constituent Indicators and the Legislative Productivity Index (Dark Line is the LPI)  
presents further tests of divergent validity. First, we report the correlation of
Models of Productivity (1st -108th Congresses) 
Models of Productivity (81st -108th Congresses) 
Models of Productivity (45th -103rd Congresses) 
We measure legislative productivity for the entire history of the U.S. Congress. Current measures of legislative productivity are problematic because they measure productivity for a limited number of decades and because they are based on different aspects of productivity. We provide a methodology for measuring (1) a Legislative Productivity Index (LPI) and (2) a Major Legislation Index (MLI). We use the W-CALC algorithm of Stimson (1999, Public opinion in America: Moods, cycles, and swings. 2nd ed. Boulder, CO: Westview Press) to combine information from previously used indicators of productivity into measures of the LPI and the MLI. We provide examinations of content, convergent, and construct validity. The construct validity model includes potential determinants of legislative productivity. We conclude that the LPI and the MLI are superior measures of productivity than other measures used in the literature.
A time series (t = 921) of weekly survey data on vote intentions in the Netherlands for the period 1978–1995 shows that the percentage of undecided voters follows a cyclical pattern over the election calendar. The otherwise substantial percentage of undecided voters decreases sharply in weeks leading up to an election and gradually increases afterwards. This article models the dynamics of this asymmetric electoral cycle using artificial neural networks, with the purpose of estimating when the undecided voters start making up their minds. We find that they begin to decide which party to vote for nine weeks before a first order national parliamentary election and one to four weeks before a second order election, depending on the type of election (European Parliament, Provincial States, City-councils). The effect of political campaigns and the implications for political analysis are discussed.
In monetary policy, decision makers seek to influence the expectations of agents in ways that can avoid making abrupt, dramatic, and unexpected decisions. Yet in October 1979, Chairman Paul Volcker led the Federal Reserve's Federal Open Market Committee (FOMC) unanimously to shift its course in managing U.S. monetary policy, which in turn eventually brought the era of high inflation to an end. Although some analysts argue that “the presence and influence of one individual” - namely, Volcker - is sufficient to explain the policy shift, this overlooks an important feature of monetary policymaking. FOMC chairmen-however, omnipotent they may appear-do not act alone. They require the agreement of other committee members, and in the 1979 revolution, the decision was unanimous. How, then, did Chairman Volcker manage to bring a previously divided committee to a consensus in October 1979, and moreover, how did he retain the support of the committee throughout the following year in the face of mounting political and economic pressure against the Fed? We use automated content analysis to examine the discourse of the FOMC (with this discourse recorded in the verbatim transcripts of meetings). In applying this methodology, we assess the force of the arguments used by Chairman Volcker and find that deliberation in the FOMC did indeed “matter” both in 1979 and 1980. Specifically, Volcker led his colleagues in coming to understand and apply the idea of credible commitment in U.S. monetary policymaking.
In an earlier report, two of us (Bowers and Ensley, 2003, National Election Studies Technical Report, www.umich.edu/∼nes ) provided a general framework for understanding the particular strategy outlined by Fogarty et al. (in this issue). Fogarty et al.'s strategy is to make the face-to-face variables more like the random digit dial (RDD) telephone variables by trimming the ends in order to reduce the variance of the face-to-face (FTF) variables. Perhaps some scholars will want the FTF variables to look like the RDD variables, but that would be a fix for a specific research question. Given the significant differences in the representativeness of the samples, the processes of survey nonresponse, and the quality and character of the responses between data taken from a National Area Probability sample in person and data taken from an RDD telephone sample, research questions involving comparisons with other years in the 50-year time series will require different remedies.
Since the inception of the American National Election Study (ANES) in the 1940s, data have been collected via face-to-face interviewing in the homes of members of area probability samples of American adults, the same gold-standard approach used by the U.S. Census Bureau, other federal agencies, and some nongovernment researchers for many of the most high-profile surveys conducted today. This paper explores whether comparable findings about voters and elections would be obtained by a different, considerably less expensive method: Internet data collection from nonprobability samples of volunteer respondents. Comparisons of the 2000 and 2004 ANES data (collected via face-to-face interviewing with national probability samples) with simultaneous Internet surveys of volunteer samples yielded many differences in the distributions of variables and in the associations between variables (even controlling for differences between the samples in reported interest in politics). Accuracy was higher for the face-to-face/probability sample data than for the Internet/volunteer sample data in 88% of the possible comparisons. This suggests that researchers interested in assuring the accuracy of their findings in describing populations should rely on face-to-face surveys of probability samples rather than Internet samples of volunteer respondents.
Example of Spatial Analysis - Iowa 18 
Influences on Change in Incumbent Victory Margins, 2000-2002
In this article, I use geographic information systems to develop a continuous measure of district continuity and change following the 2000-02 congressional redistricting cycle. The new measure provides details of where the new population in a district came from and how the old population was distributed within new districts. This measure is then used to demonstrate the independent and interactive influence of district change on competition for congressional elections. © The Author 2005. Published by Oxford University Press on behalf of the Society for Political Methodology. All rights reserved.
National Election Studies (NES) Show Card for In-person Respondents
Comparability of Mean Self-Placements on Issues from 1988-2000
Can researchers draw consistent inferences about the U.S. public's issue attitudes when studying survey results from both the in-person and telephone interview modes of the 2000 National Election Studies (NES) survey? We address this question through an analysis contrasting the distribution of issue attitudes across modes in the dual sample design of the 2000 NES. We find clear differences across mode even when applying a method devised by the NES to improve comparability by recoding issue attitude scales from the in-person mode. We present an alternative method of recoding these scales, which substantially improves comparability between modes. Through an analysis of the covariance structure of the issues and simple models of vote choice, we discuss the implications of the results for the study of issue attitudes in the 2000 NES.
Although political scientists have begun to investigate the properties of Internet surveys, much remains to be learned about the utility of the Internet mode for conducting major survey research projects such as national election studies. This paper addresses this topic by presenting the results of an extensive survey comparison experiment conducted as part of the 2005 British Election Study. Analyses show statistically significant, but generally small, differences in distributions of key explanatory variables in models of turnout and party choice. Estimating model parameters reveals that there are few statistically significant differences between coefficients generated using the in-person and Internet data, and the relative explanatory power of rival models is virtually identical for the two types of data. In general, the in-person and Internet data tell very similar stories about what matters for turnout and party preference in Britain. Determining if similar findings obtain in other countries should have high priority on the research agenda for national election studies.
In this paper, we present data from a three-mode survey comparison study carried out in 2010. National surveys were fielded at the same time over the Internet (using an opt-in Internet panel), by telephone with live interviews (using a national RDD sample of landlines and cell phones), and by mail (using a national sample of residential addresses). Each survey utilized a nearly identical questionnaire soliciting information across a range of political and social indicators, many of which can be validated with government data. Comparing the findings from the modes to each other and the validated benchmarks, we demonstrate that a carefully executed opt-in Internet panel produces estimates that are as accurate as a telephone survey and that the two modes differ little in their estimates of other political indicators and their correlates.
Studies on national identity differentiate between nationalistic attitudes and constructive patriotism (CP) as two more specific expressions of national identity and as theoretically two distinct concepts. After a brief discussion of the theoretical literature, the following questions are examined: (1) Can nationalism and CP be empirically identified as two distinct concepts?; (2) Is their meaning fully or partially invariant across countries?; and (3) Is it possible to compare their means across countries? Data from the International Social Survey Program (ISSP) 2003 National Identity Module are utilized to answer these questions in a sample of 34 countries. Items to measure nationalism and CP are chosen based on the literature, and a series of confirmatory factor analyses to test for configural, measurement (metric), and scalar invariance are performed. Full or partial metric invariance is a necessary condition for equivalence of meaning across cultures and for a meaningful comparison of associations with other theoretical constructs. Scalar invariance is a necessary condition for comparison of means across countries. Findings reveal that nationalism and CP emerge as two distinct constructs. However, in some countries, some items that were intended to measure one construct also measure the other construct. Furthermore, configural and metric invariance are found across the full set of 34 countries. Consequently, researchers may now use the ISSP data to study relationships among nationalism, CP, and other theoretical constructs across these nations. However, the analysis did not support scalar invariance, making it problematic for comparing the means of nationalism and CP across countries.
In this study, we propose a model of individual voter behavior that can be applied to aggregate data at the district (or precinct) levels while accounting for differences in political preferences across districts and across voters within each district. Our model produces a mapping of the competing candidates and electoral districts on a latent "issues" space that describes how political preferences in each district deviate from the average voter and how each candidate caters to average voter preferences within each district. We formulate our model as a random-coefficients nested logit model in which the voter first evaluates the candidates to decide whether or not to cast his or her vote, and then chooses the candidate who provides him or her with the highest value. Because we allow the random coefficient to vary not only across districts but also across unobservable voters within each district, the model avoids the Independence of Irrelevant Alternatives Assumption both across districts and within each district, thereby accounting for the cannibalization of votes among similar candidates within and across voting districts. We illustrate our proposed model by calibrating it to the actual voting data from the first stage of a two-stage state governor election in the Brazilian state of Santa Catarina, and then using the estimates to predict the final outcome of the second stage.
Illustration of RD using age as a forcing variable. The red circles depict average voting rates among observed voters, grouped by year of age, which has been rescaled so that zero (age 55) is the point of discontinuity. The blue circles depict average voting rates among counterfactual voters. The red circles to the left of the age cutoff (where age equals 0) represent the treatment group, which received the experimental mailings. The red circles to the right of the cutoff represent the control group, which received no experimental mailings. The size of the circles is proportional to the number of observations in each age group.
Comparison between RD estimates and experimental benchmarks, full sample
Local linear regression estimates and confidence intervals, by bandwidth. The graph shows local linear regression estimates of the treatment effect, using various age bandwidths between 0.5 and 20 years. Dotted lines denote 95% CIs. The horizontal dashed line denotes the experimental benchmark estimate from Table 1, column d ( ˆ b 1 5 10.0). The vertical line denotes the ImbensKalyanaraman estimate of optimal bandwidth.
Comparison between local linear regression estimates and experimental benchmarks, with different bandwidth selection algorithms
Comparison between experimental benchmarks and local regression estimates using optimal bandwidth, for various simulated age thresholds
Regression discontinuity (RD) designs enable researchers to estimate causal effects using observational data. These causal effects are identified at the point of discontinuity that distinguishes those observations that do or do not receive the treatment. One challenge in applying RD in practice is that data may be sparse in the immediate vicinity of the discontinuity. Expanding the analysis to observations outside this immediate vicinity may improve the statistical precision with which treatment effects are estimated, but including more distant observations also increases the risk of bias. Model specification is another source of uncertainty; as the bandwidth around the cutoff point expands, linear approximations may break down, requiring more flexible functional forms. Using data from a large randomized experiment conducted by Gerber, Green, and Larimer (2008), this study attempts to recover an experimental benchmark using RD and assesses the uncertainty introduced by various aspects of model and bandwidth selection. More generally, we demonstrate how experimental benchmarks can be used to gauge and improve the reliability of RD analyses.
Katz and King have previously proposed a statistical model for multiparty election data. They argue that ordinary least‐squares (OLS) regression is inappropriate when the dependent variable measures the share of the vote going to each party, and they recommend a superior technique. Regrettably, the Katz–King model requires a high level of statistical expertise and is computationally demanding for more than three political parties. We offer a sophisticated yet convenient alternative that involves seemingly unrelated regression (SUR). SUR is nearly as easy to use as OLS yet performs as well as the Katz–King model in predicting the distribution of votes and the composition of parliament. Moreover, it scales easily to an arbitrarily large number of parties. The model has been incorporated into Clarify , a statistical suite that is available free on the Internet.
We use an analogy with the normal distribution and linear regression to demonstrate the need for the Generalized Event Count (GEC) model. We then show how the GEC provides a unified framework within which to understand a diversity of distributions used to model event counts, and how to express the model in one simple equation. Finally, we address the points made by Christopher Achen, Timothy Amato, and John Londregan. Amato's and Londregan's arguments are consistent with ours and provide additional interesting information and explanations. Unfortunately, the foundation on which Achen built his paper turns out to be incorrect, rendering all his novel claims about the GEC false (or in some cases irrelevant).
2001 CSLP Telephone Survey Response Information
¾ -Comparison Across Recruitment Methods -
This article provides a basic report about subject recruitment processes for Web-based surveys. Using data from our ongoing Internet Survey of American Opinion project, two different recruitment techniques (banner advertisement and subscription campaigns) are compared. This comparison, together with a typology of Web-based surveys, provides insight into the validity and generalizability of Internet survey data. The results from this analysis show that, although Internet survey respondents differ demographically from the American population, the relationships among variables are similar across recruitment methods and match those implied by substantive theory. Thus, our research documents the basic methodology of subject acquisition for Web-based surveys, which, as we argue in our conclusion, may soon become the survey interview mode of choice for social scientists.
Embedding experiments within surveys has reinvigorated survey research. Several survey experiments are generally embedded within a survey, and analysts treat each of these experiments as self-contained. We investigate whether experiments are self-contained or if earlier treatments affect later experiments, which we call “experimental spillover.” We consider two types of bias that might be introduced by spillover: mean and inference biases. Using a simple procedure, we test for experimental spillover in two data sets: the 1991 Race and Politics Survey and a survey containing several experiments pertaining to foreign policy attitudes. We find some evidence of spillover and suggest solutions to avoid bias.
This paper develops and tests arguments about how national-level social and institutional factors shape the propensity of individuals to form attachments to political parties. Our tests employ a two-step estimation procedure that has attractive properties when there is a binary dependent variable in the first stage and when the number of second-level units is relatively small. We find that voters are most likely to form party attachments when group identities are salient and complimentary. We also find that institutions that assist voters in retrospectively evaluating parties—specifically, strong party discipline and few parties in government—increase partisanship. These institutions matter most for those individuals with the fewest cognitive resources, measured here by education.
Actors, Incentives, Capabilities, and Consequences in the Data Supply Chain
Statistical Capabilities in the Government of India
This paper examines the construction and use of data sets in political science. We focus on three interrelated questions: How might we assess data quality? What factors shape data quality? and How can these factors be addressed to improve data quality? We first outline some problems with existing data set quality, including issues of validity, coverage, and accuracy, and we discuss some ways of identifying problems as well as some consequences of data quality problems. The core of the paper addresses the second question by analyzing the incentives and capabilities facing four key actors in a data supply chain: respondents, data collection agencies (including state bureaucracies and private organizations), international organizations, and finally, academic scholars. We conclude by making some suggestions for improving the use and construction of data sets.
What do we really know about applicants to graduate school? How much information is in an applicant's file? What do we learn by having graduate admissions committees read and score applicant files? In this article, I develop a statistical model for measuring applicant quality, combining the information in the committee members' ordinal ratings with the information in applicants' GRE scores. The model produces estimates of applicant quality purged of the influence of committee members' preferences over ostensibly extraneous applicant characteristics, such as gender and intended field of study. An explicitly Bayesian approach is adopted for estimation and inference, making it straightforward to obtain confidence intervals not only on latent applicant quality but over rank orderings of applicants and the probability of belonging in a set of likely admittees. Using data from applications to a highly ranked political science graduate program, I show that there is considerable uncertainty in estimates of applicant quality, making it impossible to make authoritative distinctions as to quality among large portions of the applicant pool. The multiple rater model I develop here is extremely flexible and has applications in fields as diverse as judicial politics, legislative politics, international relations, and public opinion.
While Herron (2004, Political Analysis 12:182–190) is correct that sensitivity to changes in underlying scale and how they affect estimates and inferences is generally important, our assumption in Rothenberg and Sanders (2000, American Journal of Political Science 44:310–319) that W-NOMINATE scales can be directly compared from one Congress to another to study legislative shirking is quite defensible because scale variability is not a substantial problem. Not only are the assumptions in our original analysis regarding variability very reasonable, because any variability is quite small, but effects on consistency are marginal and, to the degree that they are relevant, indicate that our test of the shirking hypothesis is conservative. Furthermore, even generous estimates of variability in W-NOMINATE between one immediate Congress and another have little impact on results. In addition, Herron's analysis includes an unaddressed censoring problem that again, while unlikely to have much substantive relevance, indicates that Rothenberg and Sanders have worked against themselves in trying to find shirking. In conclusion, the issues that Herron highlights are of marginal consequence for the original analysis and, to the extent they matter, only buttress the findings generated and the inferences drawn.
Lee (2008) and others have estimated the incumbency advantage in US elections using a regression discontinuity design (RDD). The idea is that very close election outcomes are the equivalent of a randomized coin flip. Caughey and Sekhon (2011) present evidence that this benign interpretation does not hold for US House elections. First, they show evidence of a distributional imbalance in incumbent vote share; there are more barely-winning candidates of the incumbent party than of the non-incumbent party, rather than the similar frequencies that would result from an unbiased coin flip. Potentially of greater threat, they show a considerable imbalance in some key pre-election covariates (e.g., candidate spending) when barely winning Democrats are compared to barely losing Democrats.. The impression is that the incumbent party can exploit some sort of advantage when the election is perceived to be close, and that this distorts estimates of the incumbency advantage using RDD. We challenge Caughey and Sekhon's claim. Barely winning Democrats are more likely to be incumbents than are barely losing Democrats. This is the source of the covariate imbalance between close Democratic wins and close Democratic losses. When the comparison is between barely winning incumbents and barely losing incumbents, , the covariance imbalance vanishes. The distributional imbalance (more incumbent party close winners than close losers) is in part a natural result of the fact that (as Snyder et al 2011 explain) for declining values of the incumbent party vote, cases are increasingly rare. We show that initial findings of Lee hold up nicely in terms of incumbent party wins and losses. There is no special reason to dismiss RDD as a tool for estimating the incumbency advantage in the US.
This article demonstrates how the selection of cases for study on the basis of outcomes on the dependent variable biases conclusions. It first lays out the logic of explanation and shows how it is violated when only cases that have achieved the outcome of interest are studied. It then examines three well-known and highly regarded studies in the field of comparative politics, comparing the conclusions reached in the original work with a test of the arguments on cases selected without regard for their position on the dependent variable. In each instance, conclusions based on the uncorrelated sample differ from the original conclusions.
I analyze how the diffusion of power in parliaments affects voter choice. Using a two-step research design, I first estimate an individual-level model of voter choice in 14 parliamentary democracies, allowing voters to hold preferences both for the party most similar to them ideologically and for the party that pulls policy in their direction. While in systems in which power is concentrated the two motivations converge, in consensual systems they diverge: since votes will likely be watered down by bargaining in the parliament, outcome-oriented choice in consensual systems often leads voters to endorse parties whose positions differ from their own views. In the second step, I utilize institutional measures of power diffusion in the parliament to account for the degree to which voters in different polities pursue one motivation versus the other. I demonstrate that the more power diffusion and compromise built into the political system via institutional mechanisms, the more voters compensate for the watering down of their vote by endorsing parties whose positions differ from their own views.
Descriptive statistics of agency characteristics
Agency characteristics and expert opinions
The study of bureaucracies and their relationship to political actors is central to understanding the policy process in the United States. Studying this aspect of American politics is difficult because theories of agency behavior, effectiveness, and control often require measures of administrative agencies' policy preferences, and appropriate measures are hard to find for a broad spectrum of agencies. We propose a method for measuring agency preferences based upon an expert survey of agency preferences for 82 executive agencies in existence between 1988 and 2005. We use a multirater item response model to provide a principled structure for combining subjective ratings based on scholarly and journalistic expertise with objective data on agency characteristics. We compare the resulting agency preference estimates and standard errors to existing alternative measures, discussing both the advantages and limitations of the method.
This article addresses the lack of cohesion in econometric model building. This incoherence contributes to model building based on statistical criteria—correcting residuals—and not theoretical criteria. The models we build, therefore, are not valid replications of theory. To deal with this problem, an agenda for model building is outlined and discussed. Drawing on the methodological approaches of Hendry, Qin, and Favero (1989), Hendry and Richard (1982, 1983), Sargan (1964), and Spanos (1986), this agenda incorporates a “general to simple” modeling philosophy, a battery of diagnostic tests, reduction theory, and the development of models that include short-term and long-term parameters. A comparison is made between a model based on this agenda and a model based on corrected residuals. The findings show that the agenda-based model outperforms the residual correction model.
Political scientists lack methods to efficiently measure the priorities political actors emphasize in statements. To address this limitation, I introduce a statistical model that attends to the structure of political rhetoric when measuring expressed priorities: statements are naturally organized by author. The expressed agenda model exploits this structure to simultaneously estimate the topics in the texts, as well as the attention political actors allocate to the estimated topics. I apply the method to a collection of over 24,000 press releases from senators from 2007, which I demonstrate is an ideal medium to measure how senators explain their work in Washington to constituents. A set of examples validates the estimated priorities and demonstrates their usefulness for testing theories of how members of Congress communicate with constituents. The statistical model and its extensions will be made available in a forthcoming free software package for the R computing language.
Statistical models are often extended to explore the aggregate impact of policy reforms. After discussing these techniques and the incorporation of prediction uncertainty, this article examines the effects of registration reform in an analytic framework that explicitly considers the two stages that defined electoral participation throughout the 20th century in the United States—registration and then voting. Using selection bias techniques, the effects of counterfactual registration reform conditions are explored on the aggregate level of participation and the nature of representation in the electoral process. These offer a richer baseline of the impact of policy changes than previous work by directly exploring the expected level of dropoff in going to the polls by new registrants. Results indicate that the dropoff between registration and voting would be expected to increase as more individuals become registered. In addition, while turnout due to reforms among projected “new registrants” shows potentially larger biases than those among existing registrants, because of the different bases of registration the changes would still lead to a modest reduction in the disparity between actual group sizes and their role in elections.
Top-cited authors
Thomas Brambor
  • Lund University
Stefano M. Iacus
  • Harvard University
Jeroen K. Vermunt
  • Tilburg University
Kevin M. Quinn
  • University of California, Berkeley
Michael J. Gilligan
  • New York University