Bayesian Factor Analysis for Mixed Ordinal and Continuous Responses

Many situations exist in which a latent construct has both ordinaland continuous indicators. This presents a problem for the appliedresearcher because standard measurement models are not designedto accommodate mixed ordinal and continuous data. I addressthis problem by formulating a measurement model that is appropriatefor such mixed multivariate responses. This model unifies standardnormal theory factor analysis and item response theory modelsfor ordinal data. I detail a Markov chain Monte Carlo algorithmfor model fitting. I apply the model to cross-national dataon political-economic risk and find that the model works well.Software for fitting this model is publicly available in theMCMCpack (Martin and Quinn 2004, "MCMCpack 0.4–8") R package.     

Editorial

This issue marks the first anniversary of Parliamentary Affairs'new editorial team. In our first outing we explained how wewanted to develop the journal. Founded in the aftermath of theSecond World War by Stephen King-Hall, an ex-MP and progenitorof the Hansard Society, Parliamentary Affairs was initiallydesigned to overcome widespread ignorance and low-level cynicismabout Parliament by—effectively—celebrating theBritish way of doing politics. Like many of his generation,King-Hall was confident that an educated     

Parties in Elections, Parties in Government, and Partisan Bias

Adam Meirowitz Department of Politics, Princeton University, Princeton, NJ 08544 e-mail: ameirowi{at}princeton.edu Thomas Romer Department of Politics and Woodrow Wilson School of Public and International Affairs, Princeton University, Princeton, NJ 08544 e-mail: romer{at}princeton.edu Political parties are active when citizens choose among candidatesin elections and when winning candidates choose among policyalternatives in government. But the inextricably linked institutions,incentives, and behavior that determine these multistage choicesare substantively complex and analytically unwieldy, particularlyif modeled explicitly and considered in total, from citizenpreferences through government outcomes. To strike a balancebetween complexity and tractability, we modify standard spatialmodels of electoral competition and governmental policy-makingto study how components of partisanship—such as candidateplatform separation in elections, party ID-based voting, nationalpartisan tides, and party-disciplined behavior in the legislature—arerelated to policy outcomes. We define partisan bias as the distancebetween the following two points in a conventional choice space:the ideal point of the median voter in the median legislativedistrict and the policy outcome selected by the elected legislature.The study reveals that none of the party-in-electorate conditionsis capable of producing partisan bias independently. Specifiedcombinations of conditions, however, can significantly increasethe bias and/or the variance of policy outcomes, sometimes insubtle ways.     

Federalism and the Controversy Over the New Media in West Germany

The possibilities of expanding the television network by meansof cable and satellite service have paved the way for the Länderto determine relevant organizational structures. The newly completedState Media Treaty seems to represent a definite breakthroughin regulatory policy. The legal regulation of the new media—cableand satellite television—faces further complications,however, insofar as nine Land media laws impose different legalrequirements on the new broadcasters. Even so, differences amongthe Länder in the area of cable TV are no longer as greatas they were in the past, and a satellite agreement reachedby the Länder provides for more uniform regulation. Also,the Federal Constitutional Court's 1986 decision emphasizingthe importance of a uniform system of broadcasting and requiringdual private and public broadcasting may have settled the fundamentalpolitical dispute among the Länder over the new media.     

Introduction: Gender, Citizenship and Participation: Opportunities and Obstacles in the Twenty-first Century Politics

THIS special half issue of Parliamentary Affairs brings togetherseveral pertinent themes that have been the source of debate—academicand broader—in recent years. We hear much about the ‘crisisof participation’ in which ‘traditional’ formsof political activity attract the participation of ever-decreasingnumbers, although the extent to which this is a ‘natural’development of social change or the result of political bankruptcyremains to be decided. While, for     

Explaining Variance; Or, Stuck in a Moment We Can't Get Out Of

e-mail: bfbraum{at}fas.harvard.edu Political scientists overwhelmingly seek evidence of causationby measuring changes in the mean of the distribution of thedependent variable. This article points out that some causalrelationships produce changes in the variance, not the mean,of that distribution. It makes the case for variance-alteringcausation by demonstrating its empirical relevance to politicalscientists. The article also lays out an array of causal mechanisms—underthe general headings of aggregation, contagion, and constraint—inorder to demonstrate the logical coherence of variance-alteringcausation and the many ways in which it can arise. The discussionhighlights the often-stringent empirical and logical requirementsthat must be met if the researcher hopes to make concrete predictionsabout changes in variance.     

Welfare Reform: Revolution or Retrenchment?

This article introduces the two main themes of this specialissue of Publius: The Journal of Federalism, workfare and federalism.At this early stage in the history of the 1996 Welfare ReformAct, these themes must be stated as questions rather than answers,as hypotheses not conclusions. What does the new welfare policyseek to accomplish? What are the effects—intended andunintended—of its limited devolution of power to the states?     

Introduction

PARLIAMENTARY AFFAIRS was first published by the Hansard Societyin the winter of 1947, one of the coldest in Britain’shistory. At the time representative politics in general—andwhat many still referred to as ‘the mother of Parliaments’in particular—apparently enjoyed considerable popularprestige. It was not just Winston Churchill who, as he pointedout in the journal’s first issue, was conscious of ‘theunique and decisive contribution made by Parliament to the Britishwar effort’. Those busy building Labour’s post-warNew Jerusalem were, moreover, impressed by the extent to whichsuch a long-established political system could be put to     

Perceptions of Federal Aid Impacts on State Agencies: Patterns, Trends, and Variations Across the 20th Century

American federalism is a dynamic process involving the viewsand interactions among state and national officials. Based onmultiple mail surveys of state agency heads across 75 years,this article traces the perspectives of state executives towarda core component of state–national relationships—federalaid. The time frame dates from the 1920s and covers a periodin which federal aid programs to the states grew enormously,as did state administrative establishments. There was a long-termrise in the perceived intrusiveness of federal aid as well asincreased policy distortion effects. Despite substantial fluctuationsin perceived aid impacts, there was a four-decade consistencyin the penetration of federal aid into and across the existing3,000 state agencies. Furthermore, when perceptions of nationalinfluence are combined in an index of perceived national fiscalinfluence, there was a roller coaster effect with an overallsecular "decline" in national influence since 1974.     

Quantitative Leverage Through Qualitative Knowledge: Augmenting the Statistical Analysis of Complex Causes

Social scientific theories frequently posit that multiple causalmechanisms may produce the same outcome. Unfortunately, it isnot always possible to observe which mechanism was responsible.For example, IMF scholars conjecture that nations enter IMFagreements both out of economic need and for discretionary domesticpolitical reasons. Typically, though, all we observe is thefact of agreement, not its cause. Partial observability probitmodels (Poirier 1980, Journal of Econometrics 12:209–217; Braumoeller2003, Political Analysis 11:209–233) provide one methodfor the statistical analysis of such phenomena. Unfortunately,they are often plagued by identification and labeling difficulties.Sometimes, however, qualitative studies of particular casesenlighten us about causes when quantitative studies cannot.We propose exploiting this information to lend additional structureto the partial observability approach. Monte Carlo simulationreveals that by anchoring "discernible" causes for a handfulof cases about which we possess qualitative information, weobtain greater efficiency. More important, our method provesreliable at recovering unbiased parameter estimates when thepartial observability model fails. The paper concludes withan analysis of the determinants of IMF agreements.

A member shall be entitled to purchase the currencies of othermembers from the Fund ...[provided] the member represents thatit has a need to make the purchase because of its balance ofpayments or its reserve position or developments in its reserves. —InternationalMonetary Fund Articles of Agreement [IMF] negotiations sometimesenable government leaders to do what they privately wish todo, but are powerless to do domestically. —Robert Putnam(1988, p. 457)

     

Testing the Predictions of the Multidimensional Spatial Voting Model with Roll Call Data

e-mail: gjeong{at}artsci.wustl.edu This paper develops a procedure for locating proposals and legislatorsin a multidimensional policy space by applying agenda-constrainedideal point estimation. Placing proposals and legislators onthe same scale allows an empirical test of the predictions ofthe spatial voting model. I illustrate this procedure by testingthe predictive power of the uncovered set—a solution conceptof the multidimensional spatial voting model—using rollcall data from the U.S. Senate. Since empirical tests of thepredictive power of the uncovered set have been limited to experimentaldata, this is the first empirical test of the concept's predictivepower using real-world data. Author's note: An earlier version of this paper was presentedat the 2006 Annual Meeting of Political Methodology Society.I am grateful to Andrew Martin, Gary Miller, Dan O'Neill, DavidPark, Robert Walker, and three anonymous reviewers for theirhelpful comments. I am especially indebted to Gary Miller forhis insights and advice. All remaining errors are my own.     

Strategic Interaction and Candidate Competition in U.S. House Elections: Empirical Applications of Probit and Strategic Probit Models

In recent work, Signorino (American Political Science Review93:279–297, 1999; International Interactions 28:93–115,2002) has sought to test statistical models derived from extensive-formgames in the context of international relations research focusingon conflict and interstate bargaining. When two or more actorsinteract with one another under conditions of uncertainty, Signorinodemonstrates that it is necessary to incorporate such strategicinteraction into the underlying model to avoid potential threatsto statistical inference. Outside the realm of internationalrelations research, however, there have been limited applicationsof Signorino's strategic probit model in understanding strategicinteraction. In this article, I present an empirical comparisonof probit and strategic probit models in the context of candidatecompetition in House elections during the 1990s. I show thatincumbent spending deters challenger entry and factors suchas minority party affiliation and redistricting significantlyaffect incumbent career decisions, findings that run counterto those reported in the nonstrategic model. Overall, the resultsillustrate that failing to account for strategic interactioncan lead to biased and inaccurate estimates related to challengerand incumbent entry decisions.     

Transitional justice in ongoing conflicts

I am honoured to inaugurate the first edition of the InternationalJournal of Transitional Justice. As a prosecutor in my native Argentina, I was closely involvedin its process of transitional justice. Argentina's militarydictatorship – an organized bureaucracy which came topower in 1976 with national and     

How Smart Is my Dummy? Time Series Tests for the Influence of Politics

Kevin Grier Department of Economics, 335 Hester Hall, University of Oklahoma, Norman, OK 73019 e-mail: angus{at}ou.edu Of necessity, many tests for political influence on policiesor outcomes involve the use of dummy variables. However, itis often the case that the hypothesis against which the politicaldummies are tested is the null hypothesis that the interceptis otherwise constant throughout the sample. This simple nullcan cause inference problems if there are (nonpolitical) interceptshifts in the data and the political dummies are correlatedwith these unmodeled shifts. Here we present a method for morerigorously testing the significance of political dummy variablesin single equation models estimated with time series data. Ourmethod is based on recent work on detecting multiple regimeshifts by Bai and Perron. The article illustrates the potentialproblem caused by an overly simple null hypothesis, expositsthe Bai and Perron model, gives a proposed methodology for testingthe significance of political dummy variables, and illustratesthe method with two examples.

Before the curse of statisticsfell upon mankind we lived a happy, innocent life —HilaireBelloc, On Statistics

     

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.     

Which Electoral Formula Is the Most Proportional? A New Look with New Evidence

A ranking exists in electoral systems research of differentelectoral formulas—the mathematical functions governingthe conversion of votes into legislative seats—in termsof both proportionality of seats and votes and favorabilityto the largest party. I reexamine this issue with new methodsand new evidence, attempting to cross-validate previous rankingsusing a larger and more controlled data set and more preciseparametric methods than have been applied previously. The resultsby and large confirm previous knowledge but also illuminateseveral important new facets obscured in previous investigations.For example, at common ranges of district magnitude (from 5to 15 seats), it is shown that electoral formula may matterat least as much as district magnitude in shaping proportionality.     

Replication with Attention to Numerical Accuracy

Numerical issues matter in statistical analysis. Small errorsoccur when numbers are translated from paper and pencil intothe binary world of computers. Surprisingly, these errors maybe propagated and magnified through binary calculations, eventuallyproducing statistical estimates far from the truth. In thisreplication and extension article, we look at one method ofverifying the accuracy of statistical estimates by running thesesame data and models on multiple statistical packages. We findthat for two published articles, Nagler (1994, American Journalof Political Science 38:230–255) and Alvarez and Brehm(1995, American Journal of Political Science 39:1055–1089),results are dependent on the statistical package used. In thecourse of our replications, we uncover other pitfalls that mayprevent accurate replication, and make recommendations to ensurethe ability for future researchers to replicate results.     

In Search of the Uncovered Set

e-mail: nmiller{at}umbc.edu This paper pursues a number of theoretical explorations andconjectures pertaining to the uncovered set in spatial votinggames. It was stimulated by the article "The Uncovered Set andthe Limits of Legislative Action" by W. T. Bianco, I. Jeliazkov,and I. Sened (2004, Political Analysis 12:256–78) thatemployed a grid-search computational algorithm for estimatingthe size, shape, and location of the uncovered set, and it hasbeen greatly facilitated by access to the CyberSenate spatialvoting software being developed by Joseph Godfrey. I bring tolight theoretical considerations that account for importantfeatures of the Bianco, Jeliazkov, and Sened results (e.g.,the straight-line boundaries of uncovered sets displayed insome of their figures, the "unexpectedly large" uncovered setsdisplayed in other figures, and the apparent sensitivity ofthe location of uncovered sets to small shifts in the relativesizes of party caucuses) and present theoretical insights ofmore general relevance to spatial voting theory.     

A Robust Transformation Procedure for Interpreting Political Text

Georg Vanberg Department of Political Science, University of North Carolina, Chapel Hill, NC 27599-3265 e-mail: gvanberg{at}unc.edu (corresponding author) In a recent article in the American Political Science Review,Laver, Benoit, and Garry (2003, "Extracting policy positionsfrom political texts using words as data," 97:311–331)propose a new method for conducting content analysis. TheirWordscores approach, by automating text-coding procedures, representsan advance in content analysis that will potentially have alarge long-term impact on research across the discipline. Toallow substantive interpretation, the scores produced by theWordscores procedure require transformation. In this note, weaddress several shortcomings in the transformation procedureintroduced in the original program. We demonstrate that theoriginal transformation distorts the metric on which contentscores are placed—hindering the ability of scholars tomake meaningful comparisons across texts—and that it isvery sensitive to the texts that are scored—opening upthe possibility that researchers may generate, inadvertentlyor not, results that depend on the texts they choose to includein their analyses. We propose a transformation procedure thatsolves these problems. Authors' note: We would like to thank Ken Benoit, Michael Laver,three anonymous referees, and the editor for comments on earlierversions of this article.     

A Comment on Diagnostic Tools for Counterfactual Inference

Alexander Michaelides London School of Economics, Department of Economics, Houghton Street, London WC2A 2AE, UK e-mail: a.michaelides{at}lse.ac.uk We evaluate two diagnostic tools used to determine if counterfactualanalysis requires extrapolation. Counterfactuals based on extrapolationare model dependent and might not support empirically validinferences. The diagnostics help researchers identify thosecounterfactual "what if" questions that are empirically plausible.We show, through simple Monte Carlo experiments, that thesediagnostics will often detect extrapolation, suggesting thatthere is a risk of biased counterfactual inference when thereis no such risk of extrapolation bias in the data. This is becausethe diagnostics are affected by what we call the n/k problem:as the number of data points relative to the number of explanatoryvariables decreases, the diagnostics are more likely to detectthe risk of extrapolation bias even when such risk does notexist. We conclude that the diagnostics provide too severe atest for many data sets used in political science. Author's note: We thank Komei Fukuda, Don Green, Alan Gerber,and Jasjeet Sekhon for their generous help, Mike Kane for assistancewith R programming, and five anonymous referees for constructivecomments.