Working Papers
1997
78 records found
Recovering a Basic Space From a Set of Issue Scales
Poole, Keith T.
Submitted: 1997-01-29
Keywords: predictive dimensions, basic space, issue scales, Eckart-Young Decomposition, alternating least squares
Abstract: (click to show/hide) This paper develops a procedure for estimating the basic dimensions underlying a set of issue or attribute scales. A simple Hinich- Ordeshook spatial theory of voting is used to model Converse's fundamental insight that individuals' positions on issues are bundled together, and the knowledge of one or two issue positions makes the remaining positions very predictable. The model assumes that individuals' positions on a set of issue or attribute dimensions are determined by the individuals' positions on a small number of underlying evaluative or basic dimensions. The procedure developed in this paper for estimating these basic dimensions is, in effect, a method of performing singular value decomposition of a matrix with missing elements. Monte Carlo testing shows that the procedure reliably reproduces the missing elements. Because of this reliability, the estimation procedure can be used to produce Eckart-Young lower rank approximations. A number of applications to political data are shown and discussed.
Poole, Keith T.
Submitted: 1997-01-29
Keywords: predictive dimensions, basic space, issue scales, Eckart-Young Decomposition, alternating least squares
Abstract: (click to show/hide) This paper develops a procedure for estimating the basic dimensions underlying a set of issue or attribute scales. A simple Hinich- Ordeshook spatial theory of voting is used to model Converse's fundamental insight that individuals' positions on issues are bundled together, and the knowledge of one or two issue positions makes the remaining positions very predictable. The model assumes that individuals' positions on a set of issue or attribute dimensions are determined by the individuals' positions on a small number of underlying evaluative or basic dimensions. The procedure developed in this paper for estimating these basic dimensions is, in effect, a method of performing singular value decomposition of a matrix with missing elements. Monte Carlo testing shows that the procedure reliably reproduces the missing elements. Because of this reliability, the estimation procedure can be used to produce Eckart-Young lower rank approximations. A number of applications to political data are shown and discussed.
Generalized Substantively Reweighted Least Squares Regression
Gill, Jeff
Submitted: 1997-01-29
Keywords: Linear Models, Robust Procedures, Data Analysis, Outlier Identification
Abstract: (click to show/hide) Linear modeling often employs robust and resistant techniques to compensate for undesirable properties in the data. Conversely, Substantive Weighted Least Squares differs from these techniques since it seeks to analyze what makes the outliers distinguishable in their use of resources. SWLS does not see outliers as becoming potentially unbounded or even that they are necessarily undesirable elements of the data. SWLS runs consecutive weighted OLS models downweighting each case whose jacknifed residual is less than a specific threshold. Final iteration significant variables are identified as those which have a greater effect on higher performing cases and therefore provide prescriptive recommendations. GSRLS generalizes the SWLS technique by using transformations relating the jackknifed residuals to a common tabular distribution. This allows alpha-level positive outlier identification. Here, GSRLS is first placed in a theoretical context and further explored through monte-carlo simulation. In general, GSRLS can be seen as a data-analytic tool that exploits certain characteristics of the linear model to find variable influence on successful cases.
Gill, Jeff
Submitted: 1997-01-29
Keywords: Linear Models, Robust Procedures, Data Analysis, Outlier Identification
Abstract: (click to show/hide) Linear modeling often employs robust and resistant techniques to compensate for undesirable properties in the data. Conversely, Substantive Weighted Least Squares differs from these techniques since it seeks to analyze what makes the outliers distinguishable in their use of resources. SWLS does not see outliers as becoming potentially unbounded or even that they are necessarily undesirable elements of the data. SWLS runs consecutive weighted OLS models downweighting each case whose jacknifed residual is less than a specific threshold. Final iteration significant variables are identified as those which have a greater effect on higher performing cases and therefore provide prescriptive recommendations. GSRLS generalizes the SWLS technique by using transformations relating the jackknifed residuals to a common tabular distribution. This allows alpha-level positive outlier identification. Here, GSRLS is first placed in a theoretical context and further explored through monte-carlo simulation. In general, GSRLS can be seen as a data-analytic tool that exploits certain characteristics of the linear model to find variable influence on successful cases.
Getting the Mean Right is a Good Thing: Generalized Additive Models
Beck, Nathaniel, Jackman, Simon
Submitted: 1997-01-30
Keywords: non-parametric regression, scatterplot smoothing, local fitting, splines, non-linearity, Perot, incumbency, cabinet duration, democratic peace
Abstract: (click to show/hide) This is a substantial revision of the paper submitted as beck96. A shorter version of this paper is under consideration at a political science journal of note. Theory: Social scientists almost always use statistical models positing the dependent variable as a linear function of X, despite suspicions that the social and political world is not so parsimonious. Generalized additive models (GAMs) permit each independent variable to be modelled non-parametrically while requiring that the independent variables combine additively, striking a sensible balance between the flexibility of non-parametric techniques and the ease of interpretation and familiarity of linear regression. GAMs thus offer social scientists a practical methodology for improving on the extant practice of ``linearity by default''. Method: We present the statistical concepts and tools underlying GAMs (e.g., scatterplot smoothing, non-parametrics more generally, and accompanying graphical methods), and summarize issues pertaining to estimation, inference, and the statistical properties of GAMs. Monte Carlo experiments assess the validity of tests of linearity accompanying GAMs. Re-analysis of published work in American politics, comparative politics, and international relations demonstrates the usefulness of GAMs in social science settings. Results: Our re-analyses of published work show that GAMs can extract substantive mileage beyond that yielded by linear regression, offering novel insights, particularly in terms of modelling interactions. The Monte Carlo experiments show there is little danger of GAMs spuriously finding non-linear structures. All data analysis, Monte Carlo experiments, and statistical graphs were generated using S-PLUS, Version 3.3. The routines and data are available at ftp://weber.uscd.edu/pub/nbeck/gam.
Beck, Nathaniel, Jackman, Simon
Submitted: 1997-01-30
Keywords: non-parametric regression, scatterplot smoothing, local fitting, splines, non-linearity, Perot, incumbency, cabinet duration, democratic peace
Abstract: (click to show/hide) This is a substantial revision of the paper submitted as beck96. A shorter version of this paper is under consideration at a political science journal of note. Theory: Social scientists almost always use statistical models positing the dependent variable as a linear function of X, despite suspicions that the social and political world is not so parsimonious. Generalized additive models (GAMs) permit each independent variable to be modelled non-parametrically while requiring that the independent variables combine additively, striking a sensible balance between the flexibility of non-parametric techniques and the ease of interpretation and familiarity of linear regression. GAMs thus offer social scientists a practical methodology for improving on the extant practice of ``linearity by default''. Method: We present the statistical concepts and tools underlying GAMs (e.g., scatterplot smoothing, non-parametrics more generally, and accompanying graphical methods), and summarize issues pertaining to estimation, inference, and the statistical properties of GAMs. Monte Carlo experiments assess the validity of tests of linearity accompanying GAMs. Re-analysis of published work in American politics, comparative politics, and international relations demonstrates the usefulness of GAMs in social science settings. Results: Our re-analyses of published work show that GAMs can extract substantive mileage beyond that yielded by linear regression, offering novel insights, particularly in terms of modelling interactions. The Monte Carlo experiments show there is little danger of GAMs spuriously finding non-linear structures. All data analysis, Monte Carlo experiments, and statistical graphs were generated using S-PLUS, Version 3.3. The routines and data are available at ftp://weber.uscd.edu/pub/nbeck/gam.
Estimating the Probability of Events That have Never Occurred: When Does Your Vote Matter?
Gelman, Andrew, King, Gary, Boscardin, John
Submitted: 1997-02-14
Keywords: conditional probability, decision analysis, elections, electoral campaigning, forecasting, political science, presidential elections, rare events, rational choice, subjective probability, voting power
Abstract: (click to show/hide) Researchers sometimes argue that statisticians have little to contribute when few realizations of the process being estimated are observed. We show that this argument is incorrect even in the extreme situation of estimating the probabilities of events so rare that they have never occurred. We show how statistical forecasting models allow us to use empirical data to improve inferences about the probabilities of these events. Our application is estimating the probability that your vote will be decisive in a U.S.\ presidential election, a problem that has been studied by researchers in political science for more than two decades. The exact value of this probability is of only minor interest, but the number has important implications for understanding the optimal allocation of campaign resources, whether states and voter groups receive their fair share of attention from prospective presidents, and how formal ``rational choice'' models of voter behavior might be able to explain why people vote at all. We show how the probability of a decisive vote can be estimated empirically from state-level forecasts of the presidential election and illustrate with the example of 1992. Based on generalizations of standard political science forecasting models, we estimate the (prospective) probability of a single vote being decisive as about 1 in 10 million for close national elections such as 1992, varying by about a factor of 10 among states. Our results support the argument that subjective probabilities of many types are best obtained via empirically-based statistical prediction models rather than solely mathematical reasoning. We discuss the implications of our findings for the types of decision analyses that are used in public choice studies.
Gelman, Andrew, King, Gary, Boscardin, John
Submitted: 1997-02-14
Keywords: conditional probability, decision analysis, elections, electoral campaigning, forecasting, political science, presidential elections, rare events, rational choice, subjective probability, voting power
Abstract: (click to show/hide) Researchers sometimes argue that statisticians have little to contribute when few realizations of the process being estimated are observed. We show that this argument is incorrect even in the extreme situation of estimating the probabilities of events so rare that they have never occurred. We show how statistical forecasting models allow us to use empirical data to improve inferences about the probabilities of these events. Our application is estimating the probability that your vote will be decisive in a U.S.\ presidential election, a problem that has been studied by researchers in political science for more than two decades. The exact value of this probability is of only minor interest, but the number has important implications for understanding the optimal allocation of campaign resources, whether states and voter groups receive their fair share of attention from prospective presidents, and how formal ``rational choice'' models of voter behavior might be able to explain why people vote at all. We show how the probability of a decisive vote can be estimated empirically from state-level forecasts of the presidential election and illustrate with the example of 1992. Based on generalizations of standard political science forecasting models, we estimate the (prospective) probability of a single vote being decisive as about 1 in 10 million for close national elections such as 1992, varying by about a factor of 10 among states. Our results support the argument that subjective probabilities of many types are best obtained via empirically-based statistical prediction models rather than solely mathematical reasoning. We discuss the implications of our findings for the types of decision analyses that are used in public choice studies.
Appendix to 'A Computable Equilibrium Model for the Study of Political Economy'
Freeman, John R., Houser, Daniel
Submitted: 1997-03-14
Keywords: political economy, computable political economic equilibrium, simulation
Abstract: (click to show/hide) This is an appendix to 'Freeman, John R. and Daniel Houser: A Computable Equilibrium Model for the Study of Political Economy.' [cf. note 9]
Freeman, John R., Houser, Daniel
Submitted: 1997-03-14
Keywords: political economy, computable political economic equilibrium, simulation
Abstract: (click to show/hide) This is an appendix to 'Freeman, John R. and Daniel Houser: A Computable Equilibrium Model for the Study of Political Economy.' [cf. note 9]
Antitrust and Markets
Hayes, Jeffery W.
Submitted: 1997-03-18
Keywords: antitrust, principal-agent theory, financial markets
Abstract: (click to show/hide) Antitrust Division of the Department of Justice in order to test a theory of differential information asymmetry which posits that different political actors are capable of controlling different kinds of bureaucratic performance. Information asymmetry between the bureaucracy and political principals is an essential but infrequently tested foundation of principal--agent theory. I address the problem by generating a novel outcome measure of performance based upon how financial markets react to antitrust agency actions. By comparing measures of internal agency processes to an outcome-based market measure, we can shed light on the problem of information asymmetry when we realize that controlling processes requires less information resources than shaping outcomes. It is thus possible to make inferences not only about who controls the bureaucracy but also about what control means to the President and Congress. My preliminary findings suggest that a critical perspective on the meaning of bureaucratic performance illuminates interesting aspects of regulatory change. Examining the new market measure, I demonstrate that the entrenchment of Chicago school antitrust ideas resulted in a massivedecline in enforcement against major firms in the U.S. economy. I then argue that this decline in government intervention in the economy had stabilized by the late 1970s, well before Ronald Reagan's so-called antitrust revolution. Finally, I show that this second claim is not inconsistent with the notion of political control of the bureaucracy. By modeling a mid-range theory of differential information asymmetry, I find both a long-run equilibrium relationship between presidential ideology and agency outcomes as well as a consistent effect of Congressional ideology on agency processes. I conclude with a discussion of the implications of these findings for assessments of antitrust policy and bureaucracy research more generally.
Hayes, Jeffery W.
Submitted: 1997-03-18
Keywords: antitrust, principal-agent theory, financial markets
Abstract: (click to show/hide) Antitrust Division of the Department of Justice in order to test a theory of differential information asymmetry which posits that different political actors are capable of controlling different kinds of bureaucratic performance. Information asymmetry between the bureaucracy and political principals is an essential but infrequently tested foundation of principal--agent theory. I address the problem by generating a novel outcome measure of performance based upon how financial markets react to antitrust agency actions. By comparing measures of internal agency processes to an outcome-based market measure, we can shed light on the problem of information asymmetry when we realize that controlling processes requires less information resources than shaping outcomes. It is thus possible to make inferences not only about who controls the bureaucracy but also about what control means to the President and Congress. My preliminary findings suggest that a critical perspective on the meaning of bureaucratic performance illuminates interesting aspects of regulatory change. Examining the new market measure, I demonstrate that the entrenchment of Chicago school antitrust ideas resulted in a massivedecline in enforcement against major firms in the U.S. economy. I then argue that this decline in government intervention in the economy had stabilized by the late 1970s, well before Ronald Reagan's so-called antitrust revolution. Finally, I show that this second claim is not inconsistent with the notion of political control of the bureaucracy. By modeling a mid-range theory of differential information asymmetry, I find both a long-run equilibrium relationship between presidential ideology and agency outcomes as well as a consistent effect of Congressional ideology on agency processes. I conclude with a discussion of the implications of these findings for assessments of antitrust policy and bureaucracy research more generally.
Changing Minds? Not in Congress!
Poole, Keith T.
Submitted: 1997-03-19
Keywords: rank orderings, ideological consistency
Abstract: (click to show/hide) This paper shows a variety of evidence that members of Congress are ideologically consistent. The bulk of the evidence is from an analysis of rank-orderings of the members of the Post-World War II Congresses (80-104). A simple method of estimating rank orders that maximizes the correct classification of the observed roll call voting choices is shown in an appendix. A joint rank-ordering of all members of the House, Senate, and the 9 Presidents from Eisenhower to Clinton, accounts for 86.1 percent of the 6.9 million choices during the period. The Spearman correlation between the relative rank orderings within the two Chambers of the 123 members who served in both the House and Senate during the period was .92. Based upon the roll call voting record, once elected to Congress, members adopt a consistent ideological position and maintain it over time. There may be changing minds, but they are not in Congress.
Poole, Keith T.
Submitted: 1997-03-19
Keywords: rank orderings, ideological consistency
Abstract: (click to show/hide) This paper shows a variety of evidence that members of Congress are ideologically consistent. The bulk of the evidence is from an analysis of rank-orderings of the members of the Post-World War II Congresses (80-104). A simple method of estimating rank orders that maximizes the correct classification of the observed roll call voting choices is shown in an appendix. A joint rank-ordering of all members of the House, Senate, and the 9 Presidents from Eisenhower to Clinton, accounts for 86.1 percent of the 6.9 million choices during the period. The Spearman correlation between the relative rank orderings within the two Chambers of the 123 members who served in both the House and Senate during the period was .92. Based upon the roll call voting record, once elected to Congress, members adopt a consistent ideological position and maintain it over time. There may be changing minds, but they are not in Congress.
A Simulated Maximum Likelihood application to the 1988 Democratic Primary
Lawrence, Eric D.
Submitted: 1997-03-28
Keywords: simulated maximum likelihood, multinomial probit, vote choice models
Abstract: (click to show/hide) The multinomial probit model has some appealing advantages over models that do not allow for correlated errors, such as multinomial logit and conditional logit. With a few exceptions, however, multinomial probit models have not been estimated for vote choice models because of the computational costs inherent in evaluating high dimensional integrals. This paper explains one recently developed approach, simulated maximum likelihood combined with the GHK simulator, that makes it feasible to accurately estimate multinomial probit models. The method is demonstrated on a model of the 1988 Democratic Super Tuesday primary.
Lawrence, Eric D.
Submitted: 1997-03-28
Keywords: simulated maximum likelihood, multinomial probit, vote choice models
Abstract: (click to show/hide) The multinomial probit model has some appealing advantages over models that do not allow for correlated errors, such as multinomial logit and conditional logit. With a few exceptions, however, multinomial probit models have not been estimated for vote choice models because of the computational costs inherent in evaluating high dimensional integrals. This paper explains one recently developed approach, simulated maximum likelihood combined with the GHK simulator, that makes it feasible to accurately estimate multinomial probit models. The method is demonstrated on a model of the 1988 Democratic Super Tuesday primary.
The Changing Economic Preferences of the American Public: 1976-1991
Sekhon, Jasjeet
Submitted: 1997-03-28
Keywords: Macroeconomic Preferences, Monetary Policy, Hermite Polynomials, Splines, Bootstrap
Abstract: (click to show/hide) I show that the public indeed does have coherent preferences over macroeconomic tradeoffs, and these preferences have changed in ways consistent with not only economic theory but also with the changes which occurred in the American political system during the 1980s. In particular, most people learned something new about the state of the world in the late 1970s, and began to reject classical Keynesian explanations about economic reality. Individuals were becoming more sympathetic to the economic platform of the Republican party---i.e., they began to favour price stability. Moreover, the results support the notion that poor Americans do not hold government policy responsible for their personal economic plight (Hochschild 1981, Lane 1962).
Sekhon, Jasjeet
Submitted: 1997-03-28
Keywords: Macroeconomic Preferences, Monetary Policy, Hermite Polynomials, Splines, Bootstrap
Abstract: (click to show/hide) I show that the public indeed does have coherent preferences over macroeconomic tradeoffs, and these preferences have changed in ways consistent with not only economic theory but also with the changes which occurred in the American political system during the 1980s. In particular, most people learned something new about the state of the world in the late 1970s, and began to reject classical Keynesian explanations about economic reality. Individuals were becoming more sympathetic to the economic platform of the Republican party---i.e., they began to favour price stability. Moreover, the results support the notion that poor Americans do not hold government policy responsible for their personal economic plight (Hochschild 1981, Lane 1962).
Partisan and Ideological Trends: Causality and Sophistication in the Electorate
Box-Steffensmeier, Janet M., De Boef, Suzanna
Submitted: 1997-04-01
Keywords: political sophistication, macropartisanship, macroideology, ARFIMA models, Granger causality
Abstract: (click to show/hide) Studying the liberal-conservative dimensions of political competition deserves high-priority to gain insight into political change, as emphasized by Eisinga, Franses, and Ooms (1997, 3) and Inglehart and Klingemann (1976, 272). Yet the relationship between ideological and partisan movements in the American electorate has largely gone uninvestigated (but see Box-Steffensmeier, Knight, and Sigelman 1996). We investigate the relationship between trends in macropartisanship and macroideology for more and less politically sophisticated adults. We argue that only a portion of the electorate is involved with and attentive to the political environment, able to organize political debate in terms of liberal and conservative referents, and in turn, can link their ideological and partisan identifications. Using CBS and New York Times survey data on partisanship and ideology we find a causal relationship between ideology and partisanship only for the more politically sophisticated respondents. Mutual causality at both short and long lags characterizes the relationship between partisan and ideological change for adults with education beyond high school. In addition to the increased level of political sophistication that characterizes those for whom the series are linked, these respondents are more likely, by wide margins, to have claimed to have voted than less sophisticated respondents. Thus, any linkage has political implications. The incentives for politicians to link popular ideological sentiment with partisanship are strong. The people who put them in office (or kick them out) are the same folks who connect ideology and partisanship and who pay attention to politics.
Box-Steffensmeier, Janet M., De Boef, Suzanna
Submitted: 1997-04-01
Keywords: political sophistication, macropartisanship, macroideology, ARFIMA models, Granger causality
Abstract: (click to show/hide) Studying the liberal-conservative dimensions of political competition deserves high-priority to gain insight into political change, as emphasized by Eisinga, Franses, and Ooms (1997, 3) and Inglehart and Klingemann (1976, 272). Yet the relationship between ideological and partisan movements in the American electorate has largely gone uninvestigated (but see Box-Steffensmeier, Knight, and Sigelman 1996). We investigate the relationship between trends in macropartisanship and macroideology for more and less politically sophisticated adults. We argue that only a portion of the electorate is involved with and attentive to the political environment, able to organize political debate in terms of liberal and conservative referents, and in turn, can link their ideological and partisan identifications. Using CBS and New York Times survey data on partisanship and ideology we find a causal relationship between ideology and partisanship only for the more politically sophisticated respondents. Mutual causality at both short and long lags characterizes the relationship between partisan and ideological change for adults with education beyond high school. In addition to the increased level of political sophistication that characterizes those for whom the series are linked, these respondents are more likely, by wide margins, to have claimed to have voted than less sophisticated respondents. Thus, any linkage has political implications. The incentives for politicians to link popular ideological sentiment with partisanship are strong. The people who put them in office (or kick them out) are the same folks who connect ideology and partisanship and who pay attention to politics.
Breaking Up Isn't So Hard to Do: Ecological Inference and Split-Ticket Voting in the 1988 Presidential Election
Burden, Barry C., Kimball, David
Submitted: 1997-04-01
Keywords: ecological inference, split-ticket voting
Abstract: (click to show/hide) This method uses Gary King's (1997) solution to the ecological inference problem to examine split-ticket voting patterns in the 1988 elections. Earlier studies of split-ticket voting used either aggregate data, which suffer from the ecological fallacy, or survey data, which suffer from misreporting and small, unrepresentative sample sizes within states and districts. This paper produces accurate estimates of the proportions of voters splitting their ballots in each state and district for the first time. With these results we test several competing theories of split-ticket voting and divided government. We find, contrary to Fiorina's (1996) balancing argument, that voters are not intentionally splitting their tickets to produce moderate policies. In most cases split outcomes are the result of lopsided congressional campaigns that feature well-funded, high quality candidates versus unknown competitors.
Burden, Barry C., Kimball, David
Submitted: 1997-04-01
Keywords: ecological inference, split-ticket voting
Abstract: (click to show/hide) This method uses Gary King's (1997) solution to the ecological inference problem to examine split-ticket voting patterns in the 1988 elections. Earlier studies of split-ticket voting used either aggregate data, which suffer from the ecological fallacy, or survey data, which suffer from misreporting and small, unrepresentative sample sizes within states and districts. This paper produces accurate estimates of the proportions of voters splitting their ballots in each state and district for the first time. With these results we test several competing theories of split-ticket voting and divided government. We find, contrary to Fiorina's (1996) balancing argument, that voters are not intentionally splitting their tickets to produce moderate policies. In most cases split outcomes are the result of lopsided congressional campaigns that feature well-funded, high quality candidates versus unknown competitors.
Tau-b or Not Tau-b: Measuring Alliance Portfolio Similarity
Signorino, Curtis S., Ritter, Jeffery M.
Submitted: 1997-04-02
Keywords: alliances, tau-b, similarity, utility, risk, ordinal
Abstract: (click to show/hide) The pattern of alliance commitments among states is commonly assumed to reflect the extent to which states have common or conflicting security interests. For the past twenty years, Kendall's tau-b has been used to measure the similarity between two nations' ``portfolios'' of alliance commitments. Widely employed indicators of systemic polarity, state utility, and state risk propensity all rely upon tau-b. We demonstrate that tau-b is inappropriate for measuring the similarity of states' alliance commitments. We develop an alternative measure of alliance portfolio similiarity, S, which avoids many of the problems associated with tau-b, and we use data on alliances among European states to compare the effects of S versus tau-b in measures of utility and risk propensity. Finally, we identify several problems with inferring state interest from alliance commitments and we provide a method to overcome those problems using S in combination with data on alliances, trade, UN votes, diplomatic missions, and other types of state interaction.
Signorino, Curtis S., Ritter, Jeffery M.
Submitted: 1997-04-02
Keywords: alliances, tau-b, similarity, utility, risk, ordinal
Abstract: (click to show/hide) The pattern of alliance commitments among states is commonly assumed to reflect the extent to which states have common or conflicting security interests. For the past twenty years, Kendall's tau-b has been used to measure the similarity between two nations' ``portfolios'' of alliance commitments. Widely employed indicators of systemic polarity, state utility, and state risk propensity all rely upon tau-b. We demonstrate that tau-b is inappropriate for measuring the similarity of states' alliance commitments. We develop an alternative measure of alliance portfolio similiarity, S, which avoids many of the problems associated with tau-b, and we use data on alliances among European states to compare the effects of S versus tau-b in measures of utility and risk propensity. Finally, we identify several problems with inferring state interest from alliance commitments and we provide a method to overcome those problems using S in combination with data on alliances, trade, UN votes, diplomatic missions, and other types of state interaction.
A New Approach for Modeling Strategic Voting in Multiparty Systems
Alvarez, R. Michael, Nagler, Jonathan
Submitted: 1997-04-04
Keywords: multinomial probit, strategic voting
Abstract: (click to show/hide) Whether voters vote strategically, using their vote to best further their interests, or vote sincerely, voting for their first choice among the alternatives, is a question of longstanding interest. We offer two innovations in searching for the answer to this question. First, we begin with a more consistent model of sincere voting in multiparty democratic systems than has been presented in the literature to date. Second, we incorporate new operationalizations of the objective potential for strategic behavior than have been used in the past. We offer a test of strategic voting in the 1987 British General Election based on the varience in strategic setting across constituencies in Britain. Prepared for presentation at the Annual Meeting of the Midwest Political Science Association. This is one of many papers by the authors; the ordering of names reflects alphabetic convention. We thank Jonathan Katz and Guy Whitten for supplying helpful data for this project. We also thank Gary Cox and Jonathan Katz for discussions of this subject. Last, we thank Shaun Bowler for his work with us on a related project.
Alvarez, R. Michael, Nagler, Jonathan
Submitted: 1997-04-04
Keywords: multinomial probit, strategic voting
Abstract: (click to show/hide) Whether voters vote strategically, using their vote to best further their interests, or vote sincerely, voting for their first choice among the alternatives, is a question of longstanding interest. We offer two innovations in searching for the answer to this question. First, we begin with a more consistent model of sincere voting in multiparty democratic systems than has been presented in the literature to date. Second, we incorporate new operationalizations of the objective potential for strategic behavior than have been used in the past. We offer a test of strategic voting in the 1987 British General Election based on the varience in strategic setting across constituencies in Britain. Prepared for presentation at the Annual Meeting of the Midwest Political Science Association. This is one of many papers by the authors; the ordering of names reflects alphabetic convention. We thank Jonathan Katz and Guy Whitten for supplying helpful data for this project. We also thank Gary Cox and Jonathan Katz for discussions of this subject. Last, we thank Shaun Bowler for his work with us on a related project.
Markov Chain Models for Rolling Cross-section Data: How Campaign Events and Political Awareness Affect Vote Intentions and Partisanship in the United States and Canada
Mebane, Walter R., Wand, Jonathan
Submitted: 1997-04-07
Keywords: Markov chains, rolling cross-section data, macro data, measurement error, categorical data, ordinal data, panel data, survey data, party identification, American politics, Canadian politics, estimation
Abstract: (click to show/hide) We use a new approach we have developed for estimating discrete, finite-state Markov chain models from ``macro'' data to analyze the dynamics of individual choice probabilities in two collections of rolling cross-sectional survey data that were designed to support investigations of what happens to voters' information and preferences during campaigns. Using data from the 1984 American National Election Studies Continuous Monitoring Study, we show that not only did individual party identification vary substantially during the year, but the dynamics of party identification changed significantly in response to the conclusion of the Democratic party's nomination contest. Party identification appears to have measurement error only when the model misspecifies the dynamics. There are rapid oscillations among some categories of partisanship that may reflect individual stances regarding not only competition between the parties but also competition among party factions. Using data from the 1993 Canadian Election Study, we show that the critical events that shaped voting intentions in the election varied tremendously depending on an individual's level of political awareness, and that the effects of awareness varied across regions of the country.
Mebane, Walter R., Wand, Jonathan
Submitted: 1997-04-07
Keywords: Markov chains, rolling cross-section data, macro data, measurement error, categorical data, ordinal data, panel data, survey data, party identification, American politics, Canadian politics, estimation
Abstract: (click to show/hide) We use a new approach we have developed for estimating discrete, finite-state Markov chain models from ``macro'' data to analyze the dynamics of individual choice probabilities in two collections of rolling cross-sectional survey data that were designed to support investigations of what happens to voters' information and preferences during campaigns. Using data from the 1984 American National Election Studies Continuous Monitoring Study, we show that not only did individual party identification vary substantially during the year, but the dynamics of party identification changed significantly in response to the conclusion of the Democratic party's nomination contest. Party identification appears to have measurement error only when the model misspecifies the dynamics. There are rapid oscillations among some categories of partisanship that may reflect individual stances regarding not only competition between the parties but also competition among party factions. Using data from the 1993 Canadian Election Study, we show that the critical events that shaped voting intentions in the election varied tremendously depending on an individual's level of political awareness, and that the effects of awareness varied across regions of the country.
A Statistical Model for Multiparty Electoral Data
Katz, Jonathan, King, Gary
Submitted: 1997-04-08
Keywords: multiparty elections, compositional data, multivariate-t
Abstract: (click to show/hide) This paper proposes an internally consistent and comprehensive statistical model for analyzing multiparty, district-level aggregate election data. This model can be used to explain or predict how the geographic distribution of electoral results depends upon economic conditions, neighborhood ethnic compositions, campaign spending, and other features of the election campaign or characteristics of the aggregate areas. We also provide several new graphical representations for help in data exploration, model evaluation, and substantive interpretation. The model is more general, but we apply it resolve a controversy over the size of and trend in the electoral advantage of incumbency in Great Britain.
Katz, Jonathan, King, Gary
Submitted: 1997-04-08
Keywords: multiparty elections, compositional data, multivariate-t
Abstract: (click to show/hide) This paper proposes an internally consistent and comprehensive statistical model for analyzing multiparty, district-level aggregate election data. This model can be used to explain or predict how the geographic distribution of electoral results depends upon economic conditions, neighborhood ethnic compositions, campaign spending, and other features of the election campaign or characteristics of the aggregate areas. We also provide several new graphical representations for help in data exploration, model evaluation, and substantive interpretation. The model is more general, but we apply it resolve a controversy over the size of and trend in the electoral advantage of incumbency in Great Britain.
Economic Voting: Enlightened Self-Interest and Economic Reference Groups
Nagler, Jonathan, Willette, Jennifer R., Jackman, Simon
Submitted: 1997-04-09
Keywords: elections, economy, presidential elections, economic voting
Abstract: (click to show/hide) One of the more robust findings over the last 50 years in research on elections has been the importance of macroeconomic conditions on voting in U.S. presidential elections. An important finding in that research was made by Steven Weatherford in a 1978 article demonstrating that working class voters are more sensitive to economic conditions than are middle class voters in their vote choice. Weatherford's result was based on the 1956 through 1960 elections. We extend Weatherford's analysis for the 1956 thru 1992 elections. We are unable to produce evidence that poor voters are consistently more sensitive to the economy than are middle class and rich voters in their electoral behavior. We also offer a new theory of economic voting. We propose that voters vote based on the economic performance of their economic reference group - rather than on their own personal finances or on the state of the national economy. We offer a very preliminary and very crude initial test of this theory using NES data for 1956 to 1992.
Nagler, Jonathan, Willette, Jennifer R., Jackman, Simon
Submitted: 1997-04-09
Keywords: elections, economy, presidential elections, economic voting
Abstract: (click to show/hide) One of the more robust findings over the last 50 years in research on elections has been the importance of macroeconomic conditions on voting in U.S. presidential elections. An important finding in that research was made by Steven Weatherford in a 1978 article demonstrating that working class voters are more sensitive to economic conditions than are middle class voters in their vote choice. Weatherford's result was based on the 1956 through 1960 elections. We extend Weatherford's analysis for the 1956 thru 1992 elections. We are unable to produce evidence that poor voters are consistently more sensitive to the economy than are middle class and rich voters in their electoral behavior. We also offer a new theory of economic voting. We propose that voters vote based on the economic performance of their economic reference group - rather than on their own personal finances or on the state of the national economy. We offer a very preliminary and very crude initial test of this theory using NES data for 1956 to 1992.
Voter Turnout and the Life Cycle: A Latent Growth Curve Analysis
Plutzer, Eric
Submitted: 1997-04-09
Keywords: turnout, life-cycle, random effects model, HLM, latent growth models
Abstract: (click to show/hide) The distinctive relationship between age and voter turnout has intrigued students of electoral behavior since at least the early 1960s. Nevertheless, political scientists actually know little about how individuals acquire the habit of voting during young adulthood. Moreover, previous speculations and explanations are all questionable because they are based on data and models that are inappropriate for what is essentially a developmental process. Problems include confounding age with generational effects, assumptions of reversibility of gains in participation from key life events, and a failure account for the fact that an individual's probability of turnout at any particular age is a function of two distinct latent variables: their turnout rate in the very first elections, and their subsequent rate of increase. Theory construction is muddled because these two variables are negatively correlated and have different predictors. This study uses longitudinal data covering young voters over their first four presidential elections and uses latent growth curve models -- a special case of multi-level or Hierarchical Linear Models which are finding wide applicability in the social sciences. Given appropriate data, this approach permits statistical models that better correspond to life-cycle hypotheses. The findings clarify the role of parental influence, marriage and parenthood, while raising questions about the costs of mobility.
Plutzer, Eric
Submitted: 1997-04-09
Keywords: turnout, life-cycle, random effects model, HLM, latent growth models
Abstract: (click to show/hide) The distinctive relationship between age and voter turnout has intrigued students of electoral behavior since at least the early 1960s. Nevertheless, political scientists actually know little about how individuals acquire the habit of voting during young adulthood. Moreover, previous speculations and explanations are all questionable because they are based on data and models that are inappropriate for what is essentially a developmental process. Problems include confounding age with generational effects, assumptions of reversibility of gains in participation from key life events, and a failure account for the fact that an individual's probability of turnout at any particular age is a function of two distinct latent variables: their turnout rate in the very first elections, and their subsequent rate of increase. Theory construction is muddled because these two variables are negatively correlated and have different predictors. This study uses longitudinal data covering young voters over their first four presidential elections and uses latent growth curve models -- a special case of multi-level or Hierarchical Linear Models which are finding wide applicability in the social sciences. Given appropriate data, this approach permits statistical models that better correspond to life-cycle hypotheses. The findings clarify the role of parental influence, marriage and parenthood, while raising questions about the costs of mobility.
Who Needs Ecological Regression? Measuring the Constitutionality of Majority-Minority Districts
Epstein, David, O'Halloran, Sharyn
Submitted: 1997-04-09
Keywords: voting rights, districting, ecological regression
Abstract: (click to show/hide) According to the Supreme Courts interpretation of the 1965 Voting Rights Act, minority voters must have an equal opportunity to elect the representative of their choice. Yet the key term candidate of choice has never been fully defined, the elections analyzed in voting rights cases are usually for different offices than the ones being challenged and, worst of all, current methods for determining the point of equal opportunity rely heavily on ecological regression. In fact, the ecological fallacy is especially pernicious in voting rights cases, for it can be triggered by exactly the phenomena that the VRA sought to promote. This paper offers an alternative measure of equal opportunity that circumvents ecological regression in favor of probit or logit analyses of electoral results. It also provides a way to compare the likely substantive policy impacts of competing districting schemes. These techniques are then applied to the analysis of elections to the South Carolina State Senate.
Epstein, David, O'Halloran, Sharyn
Submitted: 1997-04-09
Keywords: voting rights, districting, ecological regression
Abstract: (click to show/hide) According to the Supreme Courts interpretation of the 1965 Voting Rights Act, minority voters must have an equal opportunity to elect the representative of their choice. Yet the key term candidate of choice has never been fully defined, the elections analyzed in voting rights cases are usually for different offices than the ones being challenged and, worst of all, current methods for determining the point of equal opportunity rely heavily on ecological regression. In fact, the ecological fallacy is especially pernicious in voting rights cases, for it can be triggered by exactly the phenomena that the VRA sought to promote. This paper offers an alternative measure of equal opportunity that circumvents ecological regression in favor of probit or logit analyses of electoral results. It also provides a way to compare the likely substantive policy impacts of competing districting schemes. These techniques are then applied to the analysis of elections to the South Carolina State Senate.
Heterogeneity and Bias in Models of Vote Choice
Berinsky, Adam
Submitted: 1997-04-21
Keywords: voting models, selection bias, heteroskedasticity, missing data
Abstract: (click to show/hide) Voters in the United States do not behave in a homogenous manner. Voting models typically account for such heterogeneity by seeking to decompose the process of vote choice into a number of distinct components. By examining voting choice data in this way, researchers are able to ascertain reasonable estimates of the average effect of various socio-economic and political variables on the candidate selection process. Models of this sort, while plausible, may not properly reflect the true heterogeneity of the American voter. At their core, simple models assume that voters use a common and uniform decision rule when deciding between candidates. But it is possible, if not likely, that different groups and classes of citizens use differently tructured processes to determine their choice of candidates. Searchers have attempted to account for this heterogeneity in a variety of ways. Rivers(1988) and Jackson (1992), for example, have accounted for differences in the voting behavior of individuals by allowing the mean effect of theoretically important variables to vary across individuals. While these approaches are extremely promising, in this paper I will take a different approach and examine three more subtle forms of heterogeneity in the vote choice process: (1) heterogeneity induced by non-random selection from the full population of citizens into the vote choice model sample; (2) heterogeneity due to the interaction of selection bias and non-constant variance; and (3) heterogeneity in the patterns of missing data across groups of the respondents. While much of the discussion in the paper is focused on the first two forms of heterogeneity, it is the third form of heterogeneity - one not typically addressed in the political science literature - that is the most important determinant of the degree of bias in vote choice models. Thus, heterogeneity within the sample of respondents affects the vote choice model estimates, just not in the way I originally envisioned. It is not just heterogeneity in the variance term, or in the selection into the vote choice process that poses a threat to accurate estimates of the power of the predictors in our vote choice models. Rather, it is the failure to preserve or account for the heterogeneity of the paths by which people answer survey questions that is the real bogeyman of vote choice models.
Berinsky, Adam
Submitted: 1997-04-21
Keywords: voting models, selection bias, heteroskedasticity, missing data
Abstract: (click to show/hide) Voters in the United States do not behave in a homogenous manner. Voting models typically account for such heterogeneity by seeking to decompose the process of vote choice into a number of distinct components. By examining voting choice data in this way, researchers are able to ascertain reasonable estimates of the average effect of various socio-economic and political variables on the candidate selection process. Models of this sort, while plausible, may not properly reflect the true heterogeneity of the American voter. At their core, simple models assume that voters use a common and uniform decision rule when deciding between candidates. But it is possible, if not likely, that different groups and classes of citizens use differently tructured processes to determine their choice of candidates. Searchers have attempted to account for this heterogeneity in a variety of ways. Rivers(1988) and Jackson (1992), for example, have accounted for differences in the voting behavior of individuals by allowing the mean effect of theoretically important variables to vary across individuals. While these approaches are extremely promising, in this paper I will take a different approach and examine three more subtle forms of heterogeneity in the vote choice process: (1) heterogeneity induced by non-random selection from the full population of citizens into the vote choice model sample; (2) heterogeneity due to the interaction of selection bias and non-constant variance; and (3) heterogeneity in the patterns of missing data across groups of the respondents. While much of the discussion in the paper is focused on the first two forms of heterogeneity, it is the third form of heterogeneity - one not typically addressed in the political science literature - that is the most important determinant of the degree of bias in vote choice models. Thus, heterogeneity within the sample of respondents affects the vote choice model estimates, just not in the way I originally envisioned. It is not just heterogeneity in the variance term, or in the selection into the vote choice process that poses a threat to accurate estimates of the power of the predictors in our vote choice models. Rather, it is the failure to preserve or account for the heterogeneity of the paths by which people answer survey questions that is the real bogeyman of vote choice models.
Heterogeneity and Individual Party Identification
Box-Steffensmeier, Janet M., Smith, Renee M.
Submitted: 1997-05-01
Keywords: heterogeneity, party identification, macropartisanship, panel data, Wiley-Wiley, Monte Carlo, beta-logistic, Markov
Abstract: (click to show/hide) Box-Steffensmeier and Smith (1996) suggest that heterogeneity in individual-level party identification accounts for aggregate dynamics in macropartisanship. Wiley-Wiley estimates of partisan persistence suggesting a very high degree of individual-level partisan persistence have been made under the assumption of no heterogeneity. Stratifying panel data by subgroups based on information, interest, and age, shows some heterogeneity in persistence even when the Wiley-Wiley estimator is used. Analytical and Monte Carlo results show, however, that the Wiley-Wiley estimator is biased upward when heterogeneity is present. Given these problems, we estimate a beta-logistic model of heterogeneity and persistence in individual-level party identification and show (a) heterogeneity in the probabilities of persistent response does exist and (b) a portion of that heterogeneity is systematically explained by interest in political campaigns in the 1990-91-92 ANES panel three-wave panel. Our estimates indicate Markov models assuming true state dependence may not be needed. Further, we find that our estimate of one of the parameters of the beta distribution is consistent with the estimate of that parameter that would be derived from our previous aggregate-level analysis.
Box-Steffensmeier, Janet M., Smith, Renee M.
Submitted: 1997-05-01
Keywords: heterogeneity, party identification, macropartisanship, panel data, Wiley-Wiley, Monte Carlo, beta-logistic, Markov
Abstract: (click to show/hide) Box-Steffensmeier and Smith (1996) suggest that heterogeneity in individual-level party identification accounts for aggregate dynamics in macropartisanship. Wiley-Wiley estimates of partisan persistence suggesting a very high degree of individual-level partisan persistence have been made under the assumption of no heterogeneity. Stratifying panel data by subgroups based on information, interest, and age, shows some heterogeneity in persistence even when the Wiley-Wiley estimator is used. Analytical and Monte Carlo results show, however, that the Wiley-Wiley estimator is biased upward when heterogeneity is present. Given these problems, we estimate a beta-logistic model of heterogeneity and persistence in individual-level party identification and show (a) heterogeneity in the probabilities of persistent response does exist and (b) a portion of that heterogeneity is systematically explained by interest in political campaigns in the 1990-91-92 ANES panel three-wave panel. Our estimates indicate Markov models assuming true state dependence may not be needed. Further, we find that our estimate of one of the parameters of the beta distribution is consistent with the estimate of that parameter that would be derived from our previous aggregate-level analysis.
Structural Shifts And Deterministic Regime Switching in Aggregate Data Analysis
Tam, Wendy
Submitted: 1997-06-04
Keywords: ecological inference
Abstract: (click to show/hide) It is common for the only available data to be aggregated at a level above the microeconomic unit in question. Analyzing aggregate data with ecological regression implicitly assumes constancy of parameters across aggregate units. This assumption is rarely tenable in aggregate data analysis since the aggregation process often generates new macro-level observations where the parameters vary across the aggregate units. Standard ecological regression estimates are not useful when they are employed on data with changing structural parameters. A switching regressions context is proposed where the state-defining variable is deterministic and measures homogeneity between the macro-level units.
Tam, Wendy
Submitted: 1997-06-04
Keywords: ecological inference
Abstract: (click to show/hide) It is common for the only available data to be aggregated at a level above the microeconomic unit in question. Analyzing aggregate data with ecological regression implicitly assumes constancy of parameters across aggregate units. This assumption is rarely tenable in aggregate data analysis since the aggregation process often generates new macro-level observations where the parameters vary across the aggregate units. Standard ecological regression estimates are not useful when they are employed on data with changing structural parameters. A switching regressions context is proposed where the state-defining variable is deterministic and measures homogeneity between the macro-level units.
Non-Parametric Analysis of Binary Choice Data
Poole, Keith T.
Submitted: 1997-06-16
Keywords: discrete choice analysis, non-parametric unfolding
Abstract: (click to show/hide) This paper shows a general non-parametric technique for maximizing the correct classification of binary choice or two-category data. Two general classes of data are analyzed. The first consists of binary choice matrices such as congressional roll calls or preferential rank ordering of stimuli gathered from individuals. For this class of data a general non-parametric unfolding procedure is developed. To unfold binary choice data two subproblems must be solved. First, given a set of chooser or legislator points a cutting plane through the space for the binary choice must be found such that it divides the legislators into two sets that reproduce the actual choices as closely as possible. Second, given a set of cutting planes for the binary choices a point for each chooser or legislator must be found that reproduces the actual choices as closely as possible. Solutions for these two problems are shown in this paper. The second class of data analyzed consists of a two-category dependent variable and a set of independent variables. This class of data is a subset of the binary choice unfolding problem. The cutting plane procedure can be used to estimate a cutting plane through the space of the independent variables that maximizes the number of correct classifications. The normal vector to this cutting plane closely corresponds to the beta vector from a standard probit, logit, or linear probability analysis.
Poole, Keith T.
Submitted: 1997-06-16
Keywords: discrete choice analysis, non-parametric unfolding
Abstract: (click to show/hide) This paper shows a general non-parametric technique for maximizing the correct classification of binary choice or two-category data. Two general classes of data are analyzed. The first consists of binary choice matrices such as congressional roll calls or preferential rank ordering of stimuli gathered from individuals. For this class of data a general non-parametric unfolding procedure is developed. To unfold binary choice data two subproblems must be solved. First, given a set of chooser or legislator points a cutting plane through the space for the binary choice must be found such that it divides the legislators into two sets that reproduce the actual choices as closely as possible. Second, given a set of cutting planes for the binary choices a point for each chooser or legislator must be found that reproduces the actual choices as closely as possible. Solutions for these two problems are shown in this paper. The second class of data analyzed consists of a two-category dependent variable and a set of independent variables. This class of data is a subset of the binary choice unfolding problem. The cutting plane procedure can be used to estimate a cutting plane through the space of the independent variables that maximizes the number of correct classifications. The normal vector to this cutting plane closely corresponds to the beta vector from a standard probit, logit, or linear probability analysis.
Why Study Only Presidential Campaigns? Statewide Races as a Laboratory for Campaign Analysis
Alvarez, R. Michael
Submitted: 1997-06-19
Keywords: presidential campaigns, statewide campaigns, campaign dynamics, voter learning, study design, case studies
Abstract: (click to show/hide) Political campaigns play a central role in democratic politics since they are an important source of contact between citizens and voters. But the literature has been quite pessimistic about whether political campaigns can influence the preferences and behavior of voters. In this paper I argue that one of the primary reasons for this pessimism stems from the consistent and lasting focus on presidential campaigns. While presidential campaigns are an important aspect of the American political process, they make poor laboratories for the study of campaigns. Instead I argue that statewide political campaigns provide a much better laboratory for the study of campaigns. The paper presents a series of empirical analyses of statewide campaigns and concludes with a discussion of different designs for the study of statewide campaigns.
Alvarez, R. Michael
Submitted: 1997-06-19
Keywords: presidential campaigns, statewide campaigns, campaign dynamics, voter learning, study design, case studies
Abstract: (click to show/hide) Political campaigns play a central role in democratic politics since they are an important source of contact between citizens and voters. But the literature has been quite pessimistic about whether political campaigns can influence the preferences and behavior of voters. In this paper I argue that one of the primary reasons for this pessimism stems from the consistent and lasting focus on presidential campaigns. While presidential campaigns are an important aspect of the American political process, they make poor laboratories for the study of campaigns. Instead I argue that statewide political campaigns provide a much better laboratory for the study of campaigns. The paper presents a series of empirical analyses of statewide campaigns and concludes with a discussion of different designs for the study of statewide campaigns.
Estimating the Same Quantities from Different Levels of Data: Time Dependence and Aggregation Bias in Event Process Models
Alt, James E., King, Gary, Signorino, Curtis S.
Submitted: 1997-06-28
Keywords: event process, exponential, gamma, time dependence, aggregation
Abstract: (click to show/hide) 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 models and theoretical processes that are not constrained by the particular forms of data that happen to be available.
Alt, James E., King, Gary, Signorino, Curtis S.
Submitted: 1997-06-28
Keywords: event process, exponential, gamma, time dependence, aggregation
Abstract: (click to show/hide) 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 models and theoretical processes that are not constrained by the particular forms of data that happen to be available.
Pattern Recognition of International Crises using Hidden Markov Models
Schrodt, Philip A.
Submitted: 1997-06-30
Keywords: hidden Markov models, event data, early warning, international crisis, sequence analysis, Middle East, WEIS, BCOW
Abstract: (click to show/hide) Event data are one of the most widely used indicators in quantitative international relations research. To date, most of the models using event data have constructed numerical indicators based on the characteristics of the events measured in isolation and then aggregated. An alternative approach is to use quantitative pattern recognition techniques to compare an existing sequence of behaviors to a set of similar historical cases. This has much in common with human reasoning by historical analogy while providing the advantages of systematic and replicable analysis possible using machine-coded event data and statistical models. This chapter uses "hidden Markov models" Ñ- a recently developed sequence- comparison technique widely used in computational speech recognition Ñ- to measure similarities among international crises. The models are first estimated using the Behavioral Correlates of War data set of historical crises, then applied to an event data set covering political behavior in the contemporary Middle East for the period April 1979 through February 1997. A split-sample test of the hidden Markov models perfectly differentiates crises involving war from those not involving war in the cases used to estimate the models. The models also provide a high level of discrimination in a set of test cases not used in the estimated, and most of the erroneously-classified cases have plausible distinguishing features. The difference between the war and nonwar models also correlates significantly with a scaled measure of conflict in the contemporary Middle East. This suggests that hidden Markov models could be used to develop conflict measures based on event similarities to historical conflicts rather than on aggregated event scores.
Schrodt, Philip A.
Submitted: 1997-06-30
Keywords: hidden Markov models, event data, early warning, international crisis, sequence analysis, Middle East, WEIS, BCOW
Abstract: (click to show/hide) Event data are one of the most widely used indicators in quantitative international relations research. To date, most of the models using event data have constructed numerical indicators based on the characteristics of the events measured in isolation and then aggregated. An alternative approach is to use quantitative pattern recognition techniques to compare an existing sequence of behaviors to a set of similar historical cases. This has much in common with human reasoning by historical analogy while providing the advantages of systematic and replicable analysis possible using machine-coded event data and statistical models. This chapter uses "hidden Markov models" Ñ- a recently developed sequence- comparison technique widely used in computational speech recognition Ñ- to measure similarities among international crises. The models are first estimated using the Behavioral Correlates of War data set of historical crises, then applied to an event data set covering political behavior in the contemporary Middle East for the period April 1979 through February 1997. A split-sample test of the hidden Markov models perfectly differentiates crises involving war from those not involving war in the cases used to estimate the models. The models also provide a high level of discrimination in a set of test cases not used in the estimated, and most of the erroneously-classified cases have plausible distinguishing features. The difference between the war and nonwar models also correlates significantly with a scaled measure of conflict in the contemporary Middle East. This suggests that hidden Markov models could be used to develop conflict measures based on event similarities to historical conflicts rather than on aggregated event scores.
Forecasting Time Series
Hinich, Melvin J.
Submitted: 1997-07-08
Keywords: forecast, autoregressive, vector AR, state space, linear
Abstract: (click to show/hide) The limits of forecasting a linear times series system are discussed. A stable autoregressive linear system can only be accurately predicted for a few steps ahead of the last observation. If the time series is a deterministic trend plus random fluctuations then the trend can be predicted as long as it is stable.
Hinich, Melvin J.
Submitted: 1997-07-08
Keywords: forecast, autoregressive, vector AR, state space, linear
Abstract: (click to show/hide) The limits of forecasting a linear times series system are discussed. A stable autoregressive linear system can only be accurately predicted for a few steps ahead of the last observation. If the time series is a deterministic trend plus random fluctuations then the trend can be predicted as long as it is stable.
A Theory of Nonseparable Preferences in Survey Responses
Lacy, Dean
Submitted: 1997-07-11
Keywords: nonseparable-preferences, framing, experiments, question-order-effects
Abstract: (click to show/hide) This paper presents a model of individual-level responses to issue questions in public opinion surveys when respondents have nonseparable preferences. The model implies two results: responses will change depending on the order of questions and vary over time. Each of these conclusions is consistent with empirical findings that are often cited to support the argument that people are irrational or lack fixed and well-formed preferences. Results from an experiment reveal that question-order effects occur on issues for which people have nonseparable preferences, and order effects do not occur on issues for which most people have separable preferences.
Lacy, Dean
Submitted: 1997-07-11
Keywords: nonseparable-preferences, framing, experiments, question-order-effects
Abstract: (click to show/hide) This paper presents a model of individual-level responses to issue questions in public opinion surveys when respondents have nonseparable preferences. The model implies two results: responses will change depending on the order of questions and vary over time. Each of these conclusions is consistent with empirical findings that are often cited to support the argument that people are irrational or lack fixed and well-formed preferences. Results from an experiment reveal that question-order effects occur on issues for which people have nonseparable preferences, and order effects do not occur on issues for which most people have separable preferences.
A Statistical Model for Multiparty Electoral Data
Katz, Jonathan, King, Gary
Submitted: 1997-07-16
Keywords: multiparty elections, compositional data, multivariate-t
Abstract: (click to show/hide) We propose an internally consistent and comprehensive statistical model for analyzing multiparty, district-level aggregate election data. This model can be used to explain or predict how the geographic distribution of electoral results depends upon economic conditions, neighborhood ethnic compositions, campaign spending, and other features of the election campaign or characteristics of the aggregate areas. We also provide several new graphical representations for help in data exploration, model evaluation, and substantive interpretation. Although the model applies more generally, we use it to resolve an important controversy over the size of and trend in the electoral advantage of incumbency in Great Britain. Contrary to previous analyses, which are all based on measures now known to be biased, we demonstrate that the incumbency advantage is about 1% for the major parties and 4% for the Liberal party and its successors. Also contrary to previous research, we show that these effects have not grown in recent years. Finally, we are able to estimate from which party each party's incumbency advantage is predominantly drawn.
Katz, Jonathan, King, Gary
Submitted: 1997-07-16
Keywords: multiparty elections, compositional data, multivariate-t
Abstract: (click to show/hide) We propose an internally consistent and comprehensive statistical model for analyzing multiparty, district-level aggregate election data. This model can be used to explain or predict how the geographic distribution of electoral results depends upon economic conditions, neighborhood ethnic compositions, campaign spending, and other features of the election campaign or characteristics of the aggregate areas. We also provide several new graphical representations for help in data exploration, model evaluation, and substantive interpretation. Although the model applies more generally, we use it to resolve an important controversy over the size of and trend in the electoral advantage of incumbency in Great Britain. Contrary to previous analyses, which are all based on measures now known to be biased, we demonstrate that the incumbency advantage is about 1% for the major parties and 4% for the Liberal party and its successors. Also contrary to previous research, we show that these effects have not grown in recent years. Finally, we are able to estimate from which party each party's incumbency advantage is predominantly drawn.
Recent Developments in Econometric Modelling: A Personal Viewpoint
Maddala, G.S.
Submitted: 1997-07-17
Keywords: dynamic panel data models, dynamic models with limited dependent variables, unit roots, cointegration, VAR's, Bayesian
Abstract: (click to show/hide) The quotation above (more than three thousand years ago) essentially summarizes my perception of what is going on in econometrics. Dynamic economic modelling is a comprehensive term. It covers everything except pure cross-section analysis. Hence, I have to narrow down the scope of my paper. I shall not cover duration models, event studies, count data and Markovian models. The areas covered are: dynamic panel data models, dynamic models with limited dependent variables, unit roots, cointegration, VAR’s and Bayesian approaches to all these problems. These are areas I am most familiar with. Also, the paper is not a survey of recent developments. Rather, it presents what I feel are important issues in these areas. Also, as far as possible, I shall relate the issues with those considered in the work on Political Methodology. I have a rather different attitude towards econometric methods which my own colleagues in the profession may not share. In my opinion, there is too much technique and not enough discussion of why we are doing what we are doing. I am often reminded of the admonition of the queen to Pollonius in Shakespeare’s Hamlet, “More matter, less art.”
Maddala, G.S.
Submitted: 1997-07-17
Keywords: dynamic panel data models, dynamic models with limited dependent variables, unit roots, cointegration, VAR's, Bayesian
Abstract: (click to show/hide) The quotation above (more than three thousand years ago) essentially summarizes my perception of what is going on in econometrics. Dynamic economic modelling is a comprehensive term. It covers everything except pure cross-section analysis. Hence, I have to narrow down the scope of my paper. I shall not cover duration models, event studies, count data and Markovian models. The areas covered are: dynamic panel data models, dynamic models with limited dependent variables, unit roots, cointegration, VAR’s and Bayesian approaches to all these problems. These are areas I am most familiar with. Also, the paper is not a survey of recent developments. Rather, it presents what I feel are important issues in these areas. Also, as far as possible, I shall relate the issues with those considered in the work on Political Methodology. I have a rather different attitude towards econometric methods which my own colleagues in the profession may not share. In my opinion, there is too much technique and not enough discussion of why we are doing what we are doing. I am often reminded of the admonition of the queen to Pollonius in Shakespeare’s Hamlet, “More matter, less art.”
Modeling Multilevel Data Structures
Jones, Bradford S., Steenbergen, Marco R.
Submitted: 1997-07-17
Keywords: multilevel models, random coefficients, contextual analysis, comparative
Abstract: (click to show/hide) Although integrating multiple levels of data into an analysis can often yield better inferences about the phenomenon under study, traditional methodologies used to combine multiple levels of data are problematic. In this paper, we discuss several methodologies under the rubric of multilevel analysis. Multilevel methods, we argue, provide researchers, particularly researchers using comparative data, substantial leverage in overcoming the typical problems associated with either ignoring multiple levels of data, or problems associated with combining lower-level and higher-level data (including overcoming implicit assumptions of fixed and constant effects). The paper discusses several variants of the multilevel model and provides an application of individual-level support for European integration using comparative political data from Western Europe.
Jones, Bradford S., Steenbergen, Marco R.
Submitted: 1997-07-17
Keywords: multilevel models, random coefficients, contextual analysis, comparative
Abstract: (click to show/hide) Although integrating multiple levels of data into an analysis can often yield better inferences about the phenomenon under study, traditional methodologies used to combine multiple levels of data are problematic. In this paper, we discuss several methodologies under the rubric of multilevel analysis. Multilevel methods, we argue, provide researchers, particularly researchers using comparative data, substantial leverage in overcoming the typical problems associated with either ignoring multiple levels of data, or problems associated with combining lower-level and higher-level data (including overcoming implicit assumptions of fixed and constant effects). The paper discusses several variants of the multilevel model and provides an application of individual-level support for European integration using comparative political data from Western Europe.
The Analysis of Binary Time-Series--Cross-Section Data and/or The Democratic Peace
Beck, Nathaniel, Katz, Jonathan
Submitted: 1997-07-18
Keywords: binary dependent variable, time-series--cross-section data, serially correlated errors, event history analysis
Abstract: (click to show/hide) The analysis of binary time-series--cross-section (BTSCS) data almost invariably ignores temporal dependence. Using Monte Carlo we show that ordinary probit standard errors underestimate variability in the presence of serially correlated errors. This underestimate, while severe, is smaller than in a corresponding OLS analysis. The simulations show that the standard errors can be partially corrected using Huber's method. We then discuss a variety of other methods for allowing temporal dependency in BTSCS estimation. The simulations show that using a lagged dependent variable will not be the panacea that it is for continuous data. We briefly examine the ``general estimating equation approach.'' We then note the equivalence of BTSCS and event history data, and thus show that common event history techniques which allow for ``duration dependence'' can be used for temporally dependent BTSCS data. The methods are used to re-analyze Oneal and Russett's study of the role of democracy and trade in facilitating peace. After correcting for temporal dependence we still find support for the democratic peace hypothesis but no longer find support for the liberal trade hypothesis.
Beck, Nathaniel, Katz, Jonathan
Submitted: 1997-07-18
Keywords: binary dependent variable, time-series--cross-section data, serially correlated errors, event history analysis
Abstract: (click to show/hide) The analysis of binary time-series--cross-section (BTSCS) data almost invariably ignores temporal dependence. Using Monte Carlo we show that ordinary probit standard errors underestimate variability in the presence of serially correlated errors. This underestimate, while severe, is smaller than in a corresponding OLS analysis. The simulations show that the standard errors can be partially corrected using Huber's method. We then discuss a variety of other methods for allowing temporal dependency in BTSCS estimation. The simulations show that using a lagged dependent variable will not be the panacea that it is for continuous data. We briefly examine the ``general estimating equation approach.'' We then note the equivalence of BTSCS and event history data, and thus show that common event history techniques which allow for ``duration dependence'' can be used for temporally dependent BTSCS data. The methods are used to re-analyze Oneal and Russett's study of the role of democracy and trade in facilitating peace. After correcting for temporal dependence we still find support for the democratic peace hypothesis but no longer find support for the liberal trade hypothesis.
Too many Variables? A Comment on Bartels' ModelAveraging Proposal
Erikson, Robert S., Wright, Gerald C., McIver, John P.
Submitted: 1997-07-18
Keywords: Bayes Factor, Bayesian Information Criterion, Bayesian statistics, model averaging, model specification, specification uncertainty, Bartels
Abstract: (click to show/hide) Abstract: Bartels (1997) popularizes the procedure of model- averaging (Raftery, 1995, 1997), making some important innovations of his own along the way. He offers his methodology as a technology for exposing excessive specification searches in other peoples' research. As a demonstration project, Bartels applied his version of model- averaging to a portion of our work on state policy and purports to detect evidence of considerable model uncertainty. . In response, we argue that Bartels' extensions of model averaging methodology are ill-advised, and show that our challenged findings hold up under the scrutiny of the original Raftery-type model averaging.
Erikson, Robert S., Wright, Gerald C., McIver, John P.
Submitted: 1997-07-18
Keywords: Bayes Factor, Bayesian Information Criterion, Bayesian statistics, model averaging, model specification, specification uncertainty, Bartels
Abstract: (click to show/hide) Abstract: Bartels (1997) popularizes the procedure of model- averaging (Raftery, 1995, 1997), making some important innovations of his own along the way. He offers his methodology as a technology for exposing excessive specification searches in other peoples' research. As a demonstration project, Bartels applied his version of model- averaging to a portion of our work on state policy and purports to detect evidence of considerable model uncertainty. . In response, we argue that Bartels' extensions of model averaging methodology are ill-advised, and show that our challenged findings hold up under the scrutiny of the original Raftery-type model averaging.
Coordinating Voting in American Presidential and House Elections
Mebane, Walter R.
Submitted: 1997-07-21
Keywords: coordinating voting, moderating voting, probabilistic voting, spatial voting, retrospective voting, presidential elections, congressional elections, split-ticket voting, pivotal voter theorem, beta distribution, multinomial logit, maximum likelihood
Abstract: (click to show/hide) I describe and estimate a probabilistic voting model designed to test whether individuals' votes for President and for the House of Representatives are coordinated with respect to two cutpoints on a single spatial dimension, in the way that Alesina and Rosenthal's pivotal voter theorem suggests they should be. In my model the cutpoints are random variables about which each individual has a subjective probability distribution. Each person's probabilistic coordinating voting behavior occurs relative to the cutpoints' expected values under the distribution. The model implements the idea the pattern of coordination depends on an individual's evaluation of the economy. The economic bias in the coordinating pattern implies that voters punish a Democratic President for success in improving the economy. The economically successful Democratic President can avoid losses only if the voters who rate the economy as having improved also believe that the policy position of the Democratic party has shifted to the right.
Mebane, Walter R.
Submitted: 1997-07-21
Keywords: coordinating voting, moderating voting, probabilistic voting, spatial voting, retrospective voting, presidential elections, congressional elections, split-ticket voting, pivotal voter theorem, beta distribution, multinomial logit, maximum likelihood
Abstract: (click to show/hide) I describe and estimate a probabilistic voting model designed to test whether individuals' votes for President and for the House of Representatives are coordinated with respect to two cutpoints on a single spatial dimension, in the way that Alesina and Rosenthal's pivotal voter theorem suggests they should be. In my model the cutpoints are random variables about which each individual has a subjective probability distribution. Each person's probabilistic coordinating voting behavior occurs relative to the cutpoints' expected values under the distribution. The model implements the idea the pattern of coordination depends on an individual's evaluation of the economy. The economic bias in the coordinating pattern implies that voters punish a Democratic President for success in improving the economy. The economically successful Democratic President can avoid losses only if the voters who rate the economy as having improved also believe that the policy position of the Democratic party has shifted to the right.
Legislative Entrepreneurship and Campaign Finance
Wawro, Gregory
Submitted: 1997-07-21
Keywords: campaign finance, fixed effects, panel data, selection bias
Abstract: (click to show/hide) Drawing on models of service--induced and investor PAC campaign contributions, I analyze the role that legislative entrepreneurship plays in PACs' contribution decisions. I explore the possibility that PACs use campaign contributions to invest in members of Congress with the expectation that members will reciprocate by engaging in entrepreneurial behavior to the benefit of PACs. To determine whether a relationship exists between legislative entrepreneurship and PAC contributions I compute measures of entrepreneurial behavior for individual members of the U.S. House using detailed data on bill sponsorship and congressional hearings from the 97th through the 101st Congress. In order to cleanly estimate the effects of legislative entrepreneurship, we need to account for unobservable member--specific factors that enter into the PAC contribution calculus. To account for such factors I employ panel data methods which require very few assumptions about the data and provide a way to test whether the manipulations of the data that are required for a panel analysis introduce bias.
Wawro, Gregory
Submitted: 1997-07-21
Keywords: campaign finance, fixed effects, panel data, selection bias
Abstract: (click to show/hide) Drawing on models of service--induced and investor PAC campaign contributions, I analyze the role that legislative entrepreneurship plays in PACs' contribution decisions. I explore the possibility that PACs use campaign contributions to invest in members of Congress with the expectation that members will reciprocate by engaging in entrepreneurial behavior to the benefit of PACs. To determine whether a relationship exists between legislative entrepreneurship and PAC contributions I compute measures of entrepreneurial behavior for individual members of the U.S. House using detailed data on bill sponsorship and congressional hearings from the 97th through the 101st Congress. In order to cleanly estimate the effects of legislative entrepreneurship, we need to account for unobservable member--specific factors that enter into the PAC contribution calculus. To account for such factors I employ panel data methods which require very few assumptions about the data and provide a way to test whether the manipulations of the data that are required for a panel analysis introduce bias.
Congressional Campaign Contributions, District Service and Electoral Outcomes in the United States: Statistical Tests of a Formal Game Model with Nonlinear Dynamics
Mebane, Walter R.
Submitted: 1997-07-22
Keywords: congressional elections, campaign contributions, campaign finance, district service, intergovernmental transfers, formal model, game theory, Cournot-Nash equilibrium, Nash equilibrium, differential equations, dynamical system, nonlinear dynamics, Hopf bifurcation, normal form, Whitney embedding theorem, divergence theorem, Liouville's theorem, multivariate normal distribution, maximum likelihood, Wald test, stability, asymptotic stability
Abstract: (click to show/hide) Using a two-stage game model of congressional campaigns, the second stage being a system of ordinary differential equations, I argue that candidates, political parties and financial contributors interact strategically in American congressional elections in a way that is inherently nonlinear. Congressional races in which the incumbent faces a challenge are generated by dynamical systems that have Hopf bifurcations: a small change in the challenger's quality or in the type of district service can change a stable incumbent advantage into an oscillating race in which the incumbent's chances are uncertain. The normal form equations for such a system inspire a statistical model that can recover qualitative features of the dynamics from cross-sectional data. I estimate and test the model using data from the 1984 and 1986 election periods for political action committee campaign contributions, intergovernmental transfers and general election vote shares.
Mebane, Walter R.
Submitted: 1997-07-22
Keywords: congressional elections, campaign contributions, campaign finance, district service, intergovernmental transfers, formal model, game theory, Cournot-Nash equilibrium, Nash equilibrium, differential equations, dynamical system, nonlinear dynamics, Hopf bifurcation, normal form, Whitney embedding theorem, divergence theorem, Liouville's theorem, multivariate normal distribution, maximum likelihood, Wald test, stability, asymptotic stability
Abstract: (click to show/hide) Using a two-stage game model of congressional campaigns, the second stage being a system of ordinary differential equations, I argue that candidates, political parties and financial contributors interact strategically in American congressional elections in a way that is inherently nonlinear. Congressional races in which the incumbent faces a challenge are generated by dynamical systems that have Hopf bifurcations: a small change in the challenger's quality or in the type of district service can change a stable incumbent advantage into an oscillating race in which the incumbent's chances are uncertain. The normal form equations for such a system inspire a statistical model that can recover qualitative features of the dynamics from cross-sectional data. I estimate and test the model using data from the 1984 and 1986 election periods for political action committee campaign contributions, intergovernmental transfers and general election vote shares.
Testing Theories Involving Strategic Choice: The Example of Crisis Escalation
Smith, Alastair
Submitted: 1997-07-23
Keywords: Strategic choice, Bayesian model testing, Markov chain Monte Carlo simulation, multi-variate probit, crisis escalation, war
Abstract: (click to show/hide) If we believe that politics involves a significant amount of strategic interaction then classical statistical tests, such as Ordinary Least Squares, Probit or Logit, cannot give us the right answers. This is true for two reasons: The dependent variables under observation are interdependent-- that is the essence of game theoretic logic-- and the data is censored -- that is an inherent feature of off the path expectations that leads to selection effects. I explore the consequences of strategic decision making on empirical estimation in the context of international crisis escalation. I show how and why classical estimation techniques fail in strategic settings. I develop a simple strategic model of decision making during crises. I ask what this explanation implies about the distribution of the dependent variable: the level of violence used by each nation. Counterfactuals play a key role in this theoretical explanation. Yet, conventional econometric techniques take no account of unrealized opportunities. For example, suppose a weak nation (B) is threatened by a powerful neighbor (A). If we believe that power strongly influences the use of force then the weak nation realizes that the aggressor's threats are probably credible. Not wishing to fight a more powerful opponent, nation B is likely to acquiesce to the aggressor's demands. Empirically, we observe A threaten B. The actual level of violence that A uses is low. However, the theoretical model suggests that B acquiesced precisely because A would use force. Although the theoretical model assumes a strong relationship between strength and the use of force, traditional techniques find a much weaker relationship. Our ability to observe whether nation A is actually prepared to use force is censored when nation B acquiesces. I develop a Strategically Censored Discrete Choice (SCDC) model which accounts for the interdependent and censored nature of strategic decision making. I use this model to test existing theories of dispute escalation. Specifically, I analyze Bueno de Mesquita and Lalman's (1992) dyadically coded version of the Militarized Interstate Dispute data (Gochman and Moaz 1984). I estimate this model using a Bayesian Markov chain Monte Carlo simulation method. Using Bayesian model testing, I compare the explanatory power of a variety of theories. I conclude that strategic choice explanations of crisis escalation far out-perform non-strategic ones.
Smith, Alastair
Submitted: 1997-07-23
Keywords: Strategic choice, Bayesian model testing, Markov chain Monte Carlo simulation, multi-variate probit, crisis escalation, war
Abstract: (click to show/hide) If we believe that politics involves a significant amount of strategic interaction then classical statistical tests, such as Ordinary Least Squares, Probit or Logit, cannot give us the right answers. This is true for two reasons: The dependent variables under observation are interdependent-- that is the essence of game theoretic logic-- and the data is censored -- that is an inherent feature of off the path expectations that leads to selection effects. I explore the consequences of strategic decision making on empirical estimation in the context of international crisis escalation. I show how and why classical estimation techniques fail in strategic settings. I develop a simple strategic model of decision making during crises. I ask what this explanation implies about the distribution of the dependent variable: the level of violence used by each nation. Counterfactuals play a key role in this theoretical explanation. Yet, conventional econometric techniques take no account of unrealized opportunities. For example, suppose a weak nation (B) is threatened by a powerful neighbor (A). If we believe that power strongly influences the use of force then the weak nation realizes that the aggressor's threats are probably credible. Not wishing to fight a more powerful opponent, nation B is likely to acquiesce to the aggressor's demands. Empirically, we observe A threaten B. The actual level of violence that A uses is low. However, the theoretical model suggests that B acquiesced precisely because A would use force. Although the theoretical model assumes a strong relationship between strength and the use of force, traditional techniques find a much weaker relationship. Our ability to observe whether nation A is actually prepared to use force is censored when nation B acquiesces. I develop a Strategically Censored Discrete Choice (SCDC) model which accounts for the interdependent and censored nature of strategic decision making. I use this model to test existing theories of dispute escalation. Specifically, I analyze Bueno de Mesquita and Lalman's (1992) dyadically coded version of the Militarized Interstate Dispute data (Gochman and Moaz 1984). I estimate this model using a Bayesian Markov chain Monte Carlo simulation method. Using Bayesian model testing, I compare the explanatory power of a variety of theories. I conclude that strategic choice explanations of crisis escalation far out-perform non-strategic ones.
Estimation and Strategic Interaction in Discrete Choice Models of International Conflict
Signorino, Curtis S.
Submitted: 1997-07-23
Keywords: discrete choice, strategic, QRE, logit, international relations
Abstract: (click to show/hide) Typical applications of logit and probit to theories of international conflict do not capture the structure of the strategic interdependence implied by those theories. In this paper I demonstrate how to use a game-theoretic solution concept, the quantal response equilibrium (QRE), to derive strategic discrete choice models of international conflict, where the structure of the strategic interaction is incorporated directly in the statistical model. I demonstrate this for a crisis interaction model and use monte carlo analysis to show that logit provides estimates with incorrect substantive interpretations and fitted values that are often far from the true values. Finally, I reanalyze a well-known game-theoretic model of war, Bueno de Mesquita and Lalman's (1992) international interaction game, using this method. My results indicate that their model does not explain international interaction as well as they claim.
Signorino, Curtis S.
Submitted: 1997-07-23
Keywords: discrete choice, strategic, QRE, logit, international relations
Abstract: (click to show/hide) Typical applications of logit and probit to theories of international conflict do not capture the structure of the strategic interdependence implied by those theories. In this paper I demonstrate how to use a game-theoretic solution concept, the quantal response equilibrium (QRE), to derive strategic discrete choice models of international conflict, where the structure of the strategic interaction is incorporated directly in the statistical model. I demonstrate this for a crisis interaction model and use monte carlo analysis to show that logit provides estimates with incorrect substantive interpretations and fitted values that are often far from the true values. Finally, I reanalyze a well-known game-theoretic model of war, Bueno de Mesquita and Lalman's (1992) international interaction game, using this method. My results indicate that their model does not explain international interaction as well as they claim.
Congress and the Environment: A Longitudinal Analysis
Shipan, Charles R., Lowry, William R.
Submitted: 1997-08-01
Keywords: Congress, interest group ratings, environment
Abstract: (click to show/hide) ach year since 1970 the League of Conservation Voters has identified a series of key votes on public health, energy, wilderness, and other environmental issues and used these votes to calculate "ratings" for members of Congress. While these ratings are useful for comparing the level of environmental support by members within a given year and a given chamber, they are less useful for making cross-chamber or longitudinal comparisons, as they are based on different votes and thus computed on a different scale in each year and in each chamber. In this paper we make use of an innovative new methodological approach developed by Groseclose, Levitt, and Snyder (1997) to adjust these scores in a way that allows for interchamber and intertemporal comparisons. We then use these adjusted scores to examine patterns in congressional voting on environmental issues. In addition to presenting these trends, we propose and evaluate some explanations for the most striking trend: the finding that the parties have diverged on environmental issues over the past twenty-five years.
Shipan, Charles R., Lowry, William R.
Submitted: 1997-08-01
Keywords: Congress, interest group ratings, environment
Abstract: (click to show/hide) ach year since 1970 the League of Conservation Voters has identified a series of key votes on public health, energy, wilderness, and other environmental issues and used these votes to calculate "ratings" for members of Congress. While these ratings are useful for comparing the level of environmental support by members within a given year and a given chamber, they are less useful for making cross-chamber or longitudinal comparisons, as they are based on different votes and thus computed on a different scale in each year and in each chamber. In this paper we make use of an innovative new methodological approach developed by Groseclose, Levitt, and Snyder (1997) to adjust these scores in a way that allows for interchamber and intertemporal comparisons. We then use these adjusted scores to examine patterns in congressional voting on environmental issues. In addition to presenting these trends, we propose and evaluate some explanations for the most striking trend: the finding that the parties have diverged on environmental issues over the past twenty-five years.
Indecision Theory: An Informational Model of Roll-Off
Katz, Jonathan, Ghirardato, Paolo
Submitted: 1997-08-05
Keywords: voting, formal theory, decision theory, uncertainty, abstention
Abstract: (click to show/hide) We address the so-called "roll-off" phenomenon: Selective abstention in multiple elections. We present a discuss a novel model of decision making by voters that explains this as a result of differences in quality and quantity of information that the voters have about each election. In doing so we use a spatial model that differs from the Euclidean one, and is more naturally applied to modeling differences in information.
Katz, Jonathan, Ghirardato, Paolo
Submitted: 1997-08-05
Keywords: voting, formal theory, decision theory, uncertainty, abstention
Abstract: (click to show/hide) We address the so-called "roll-off" phenomenon: Selective abstention in multiple elections. We present a discuss a novel model of decision making by voters that explains this as a result of differences in quality and quantity of information that the voters have about each election. In doing so we use a spatial model that differs from the Euclidean one, and is more naturally applied to modeling differences in information.
The Political Entropy of Vote Choice: An Empirical Test of Uncertainty Reduction
Gill, Jeff
Submitted: 1997-08-05
Keywords: Entropy, Voting Under Uncertainty, Proximity Spatial Voting Model, Heteroscedastic Probit
Abstract: (click to show/hide) Recent literature in voting theory has developed the idea that individual voting preferences are probabilistic rather than strictly deterministic. This work builds upon spatial voting models (Enelow and Hinich 1981, Ferejohn and Fiorina 1974, Davis, DeGroot and Hinich 1972, Farquharson 1969) by introducing probabilistic uncertainty into the calculus of voting decision on an individual level. Some suggest that the voting decision can be modeled with traditional probabilistic tools of uncertainty (Coughlin 1990, Coughlin and Nitzen 1981). Entropy is a measure of uncertainty that originated in statistical thermodynamics. Essentially, entropy indicates the amount of uncertainty in probability distributions (Soofi 1992), or it can be thought of as signifying a lack of human knowledge about some random event (Denbigh and Denbigh, 1985). Entropy in statistics developed with Kolmogorov (1959), Kinchin (1957), and Shannon (1948), but has rarely been applied to social science problems. Exceptions include Darcy and Aigner's (1980) use of entropy to analyze categorical survey responses in political science, and economic applications by Theil (1967) and Theil and Fiebig (1984). I examine voters' uncertainty as they assess candidates, and measure policy positions. I then test whether or not these voters minimize the cost of voting (specifically the cost of information) by determining a maximum entropy selection. Except for the inclusion of entropy terms, this approach is similar to others in the recent literature. In this paper I develop a measure to aggregate evaluation of issue uncertainty and corresponding vote choice where the uncertainty parameterization is derived from an entropy calculation on a set of salient election issues. The primary advantage of this approach is that it requires very few assumptions about the nature of the data. Using 1994 American National Election Study survey data from the Center for Political Studies, I test the hypothesis that the ``Contract with America'' reduced voter uncertainty about the issue positions of Republican House candidates. The entropic model suggests that voters used the written and explicit Republican agenda as a means of reducing issue uncertainty without substantially increasing time spent evaluating candidate positions.
Gill, Jeff
Submitted: 1997-08-05
Keywords: Entropy, Voting Under Uncertainty, Proximity Spatial Voting Model, Heteroscedastic Probit
Abstract: (click to show/hide) Recent literature in voting theory has developed the idea that individual voting preferences are probabilistic rather than strictly deterministic. This work builds upon spatial voting models (Enelow and Hinich 1981, Ferejohn and Fiorina 1974, Davis, DeGroot and Hinich 1972, Farquharson 1969) by introducing probabilistic uncertainty into the calculus of voting decision on an individual level. Some suggest that the voting decision can be modeled with traditional probabilistic tools of uncertainty (Coughlin 1990, Coughlin and Nitzen 1981). Entropy is a measure of uncertainty that originated in statistical thermodynamics. Essentially, entropy indicates the amount of uncertainty in probability distributions (Soofi 1992), or it can be thought of as signifying a lack of human knowledge about some random event (Denbigh and Denbigh, 1985). Entropy in statistics developed with Kolmogorov (1959), Kinchin (1957), and Shannon (1948), but has rarely been applied to social science problems. Exceptions include Darcy and Aigner's (1980) use of entropy to analyze categorical survey responses in political science, and economic applications by Theil (1967) and Theil and Fiebig (1984). I examine voters' uncertainty as they assess candidates, and measure policy positions. I then test whether or not these voters minimize the cost of voting (specifically the cost of information) by determining a maximum entropy selection. Except for the inclusion of entropy terms, this approach is similar to others in the recent literature. In this paper I develop a measure to aggregate evaluation of issue uncertainty and corresponding vote choice where the uncertainty parameterization is derived from an entropy calculation on a set of salient election issues. The primary advantage of this approach is that it requires very few assumptions about the nature of the data. Using 1994 American National Election Study survey data from the Center for Political Studies, I test the hypothesis that the ``Contract with America'' reduced voter uncertainty about the issue positions of Republican House candidates. The entropic model suggests that voters used the written and explicit Republican agenda as a means of reducing issue uncertainty without substantially increasing time spent evaluating candidate positions.
A Process Control Model of Legislative Productivity: Testing the Effects of Congressional Reform
Gill, Jeff, Thurber, James A.
Submitted: 1997-08-08
Keywords: Queueing Theory, Software Simulation, Congressional Reform, Productivity Equilibrium Model, Committee Efficiency
Abstract: (click to show/hide) We examine the effects of congressional reform on legislative productivity using a completely new methodology in political science based on queueing theory and industrial simulation and control software. The foundation of our analysis is the development of a process control model of legislative development. The model establishes a status quo productivity equilibrium based on empirical data from the first 100 days of the $103^{rd}$ House of Representatives, then stresses the system using the mandated productivity of the first 100 days of the $104^{th}$ House of Representatives. We compare the agenda based distribution of bill assignments from the ``Contract with America'' with a uniform assignment and find that requiring a stable legislative system to greatly increase productivity has substantial effects on members' allocation of time. In particular, members are likely to reduce time considering legislation and increasingly rely upon partisan cues for vote decisions. The methodology is sufficiently general that it can be applied to almost any legislative setting. Our application focuses on the feedback response from an electoral shift, but the methodology can address any productivity question. Since all legislative bodies have defined processes by which initiatives flow, the modeling and simulating of these processes can illuminate efficiencies and inefficiencies. Queueing theory addresses the prevalent and generalizable scenario in which demand for legislative outcomes exceeds the short-term capacity of a legislative system.
Gill, Jeff, Thurber, James A.
Submitted: 1997-08-08
Keywords: Queueing Theory, Software Simulation, Congressional Reform, Productivity Equilibrium Model, Committee Efficiency
Abstract: (click to show/hide) We examine the effects of congressional reform on legislative productivity using a completely new methodology in political science based on queueing theory and industrial simulation and control software. The foundation of our analysis is the development of a process control model of legislative development. The model establishes a status quo productivity equilibrium based on empirical data from the first 100 days of the $103^{rd}$ House of Representatives, then stresses the system using the mandated productivity of the first 100 days of the $104^{th}$ House of Representatives. We compare the agenda based distribution of bill assignments from the ``Contract with America'' with a uniform assignment and find that requiring a stable legislative system to greatly increase productivity has substantial effects on members' allocation of time. In particular, members are likely to reduce time considering legislation and increasingly rely upon partisan cues for vote decisions. The methodology is sufficiently general that it can be applied to almost any legislative setting. Our application focuses on the feedback response from an electoral shift, but the methodology can address any productivity question. Since all legislative bodies have defined processes by which initiatives flow, the modeling and simulating of these processes can illuminate efficiencies and inefficiencies. Queueing theory addresses the prevalent and generalizable scenario in which demand for legislative outcomes exceeds the short-term capacity of a legislative system.
Heterogeneity, Salience, and Voter Decision Rules for Candidate Preference
Glasgow, Garrett
Submitted: 1997-08-10
Keywords: voter behavior, decision rules, rank ordered logit, salience, issue voting
Abstract: (click to show/hide) Voters in American Presidential elections display a wide variety of decision rules when choosing a candidate. One form of this heterogeneity is differential weighting of issues used to make a vote choice. The structure of this heterogeneity and differential salience of issues has important implications for the American political process. Determining the nature of these heterogeneous preferences is vital to understanding electoral politics in the United States. An empirical technique for modeling and exploring heterogeneity is developed and applied to the 1980 NES Panel Study. I show that heterogeneity in voter decision rules is widespread, and that while many voters rely on non-issue considerations when determining candidate preference, issue voting does play a role in the decision rules of many voters.
Glasgow, Garrett
Submitted: 1997-08-10
Keywords: voter behavior, decision rules, rank ordered logit, salience, issue voting
Abstract: (click to show/hide) Voters in American Presidential elections display a wide variety of decision rules when choosing a candidate. One form of this heterogeneity is differential weighting of issues used to make a vote choice. The structure of this heterogeneity and differential salience of issues has important implications for the American political process. Determining the nature of these heterogeneous preferences is vital to understanding electoral politics in the United States. An empirical technique for modeling and exploring heterogeneity is developed and applied to the 1980 NES Panel Study. I show that heterogeneity in voter decision rules is widespread, and that while many voters rely on non-issue considerations when determining candidate preference, issue voting does play a role in the decision rules of many voters.
Minority Representation in Multi-member Districts
Gerber, Elisabeth R., Morton, Becky, Rietz, Thomas
Submitted: 1997-08-13
Keywords: cumulative voting, multi-member districts, minority representation, laboratory elections
Abstract: (click to show/hide) We present a theoretical and experimental examination of cumulative voting versus straight (non-cumulative) voting in multi-member district elections. Cumulative voting has been proposed as a method for increasing minority representation. Given the recent court rulings against racial gerrymandering to achieve minority representation in single-member districts, the effect of multi- member district elections on minority representation is an important issue. We present a model of voting in double-member district elections with two majority candidates and one minority candidate and consider the voting equilibria under the two voting systems. In straight voting, we find that while an equilibrium always exists where the two majority candidates are expected to win the two seats, equilibria also exist where minority candidates may be elected. In cumulative voting, we find that equilibrium minority candidate wins are also possible but are less likely when minority voters prefer one majority candidate over another. We then describe experimental evidence on voting behavior and outcomes in straight and cumulative voting elections. We find that minority candidates win significantly more seats in cumulative than in straight voting elections, as predicted, but win fewer elections when minority voters prefer one majority candidate over another.
Gerber, Elisabeth R., Morton, Becky, Rietz, Thomas
Submitted: 1997-08-13
Keywords: cumulative voting, multi-member districts, minority representation, laboratory elections
Abstract: (click to show/hide) We present a theoretical and experimental examination of cumulative voting versus straight (non-cumulative) voting in multi-member district elections. Cumulative voting has been proposed as a method for increasing minority representation. Given the recent court rulings against racial gerrymandering to achieve minority representation in single-member districts, the effect of multi- member district elections on minority representation is an important issue. We present a model of voting in double-member district elections with two majority candidates and one minority candidate and consider the voting equilibria under the two voting systems. In straight voting, we find that while an equilibrium always exists where the two majority candidates are expected to win the two seats, equilibria also exist where minority candidates may be elected. In cumulative voting, we find that equilibrium minority candidate wins are also possible but are less likely when minority voters prefer one majority candidate over another. We then describe experimental evidence on voting behavior and outcomes in straight and cumulative voting elections. We find that minority candidates win significantly more seats in cumulative than in straight voting elections, as predicted, but win fewer elections when minority voters prefer one majority candidate over another.
The Spatial Theory of Voting and the Presidential Election of 1824
Jenkins, Jeffery A., Sala, Brian R.
Submitted: 1997-08-15
Keywords: spatial voting theory, ideological voting, presidential selection, Nominate scores
Abstract: (click to show/hide) One recent analysis claims that in at least five p residential contests since the end of World War II a relatively minor vote shift in a small number of states would have produced Electoral College deadlock, leading to a House election for president (Longley and Peirce 1996). A presidential contest in the House would raise fundamental questions from agency theory - do members "shirk" the collective preferences of their constituent-principals on highly salient votes and, if so, what explains the choices they do make? Can vote choices be rationalized in a theory of ideological voting, or are legislators highly susceptible to interest-group pressures and enticements? We apply a spatial-theoretic model of voting to the House balloting for president in 1825 in order to test competing hypotheses about how MCs would likely vote in a presidential ballot. We find that a sincere voting model based on ideal points for MCs and candidates derived from Nominate scores closely matches the choices made by MCs in 1825.
Jenkins, Jeffery A., Sala, Brian R.
Submitted: 1997-08-15
Keywords: spatial voting theory, ideological voting, presidential selection, Nominate scores
Abstract: (click to show/hide) One recent analysis claims that in at least five p residential contests since the end of World War II a relatively minor vote shift in a small number of states would have produced Electoral College deadlock, leading to a House election for president (Longley and Peirce 1996). A presidential contest in the House would raise fundamental questions from agency theory - do members "shirk" the collective preferences of their constituent-principals on highly salient votes and, if so, what explains the choices they do make? Can vote choices be rationalized in a theory of ideological voting, or are legislators highly susceptible to interest-group pressures and enticements? We apply a spatial-theoretic model of voting to the House balloting for president in 1825 in order to test competing hypotheses about how MCs would likely vote in a presidential ballot. We find that a sincere voting model based on ideal points for MCs and candidates derived from Nominate scores closely matches the choices made by MCs in 1825.
Conflict, Information, and Lobbying Coalitions
Esterling, Kevin M.
Submitted: 1997-08-18
Keywords: Policy Alliances, Organizational Deliberation, Nested Logit
Abstract: (click to show/hide) This paper explains lobbying organizations' choice to join alliances on policy matters with respect to 1) the degree of the organization's access to external information sources, and 2) the amount of internal organizational conflict and deliberation. An informational view of lobbying suggests that the more informed an organizational actor is, the more likely it will gain access to governmental decision makers; and greater access to the government will decrease the utility of joining a cooperative lobbying effort. In addition, internal conflict in the definition of a policy position will limit an organization's ability to take any position on a policy issue, while successful internal deliberation will augment a lobbying organization's ability to find cooperation partners. Outcome and explanatory data are taken from an existing dataset housed at ICPSR. Nested logit maximum likelihood estimates for the trichotomous-choice cooperation model are presented and interpreted. Support is lent to both the internal conflict and the informational theories of cooperation in policy lobbying. In particular, the model results suggest that organizations predisposed to internal conflict find both non-policy lobbying and cooperative lobbying appealing, suggesting that these organizations only sometimes successfully deliberate over policy. And consistent with the information view of lobbying, greater access to information sharply decreases the utility of lobbying cooperatively.
Esterling, Kevin M.
Submitted: 1997-08-18
Keywords: Policy Alliances, Organizational Deliberation, Nested Logit
Abstract: (click to show/hide) This paper explains lobbying organizations' choice to join alliances on policy matters with respect to 1) the degree of the organization's access to external information sources, and 2) the amount of internal organizational conflict and deliberation. An informational view of lobbying suggests that the more informed an organizational actor is, the more likely it will gain access to governmental decision makers; and greater access to the government will decrease the utility of joining a cooperative lobbying effort. In addition, internal conflict in the definition of a policy position will limit an organization's ability to take any position on a policy issue, while successful internal deliberation will augment a lobbying organization's ability to find cooperation partners. Outcome and explanatory data are taken from an existing dataset housed at ICPSR. Nested logit maximum likelihood estimates for the trichotomous-choice cooperation model are presented and interpreted. Support is lent to both the internal conflict and the informational theories of cooperation in policy lobbying. In particular, the model results suggest that organizations predisposed to internal conflict find both non-policy lobbying and cooperative lobbying appealing, suggesting that these organizations only sometimes successfully deliberate over policy. And consistent with the information view of lobbying, greater access to information sharply decreases the utility of lobbying cooperatively.
The Structure of Signaling: A Combinatorial Optimization Model with Network-Dependent Estimation
Esterling, Kevin M., Lazer, David, Carpenter, Daniel
Submitted: 1997-08-18
Keywords: lobbying models, combinatorial optimization, count models, network-dependent estimation, structural autocorrelation
Abstract: (click to show/hide) This paper examines the relationship between lobbyists' contact-making behavior and their long-term access to the government. Specifically: 1) Do lobbyists establish social contacts in an individually-rational manner to best receive information from each other? And, 2) does the resulting network position condition their access to the government? We begin by wedding rational choice models to network analysis with a formal model of lobbyists' choice of contacts in a network, adopting the classic combinatorial optimization approach of Boorman (1975). The model predicts that when the demand for political information is low, a cocktail equilibrium prevails: lobbyists will invest their time in gaining "weak tie" acquaintances rather than in gaining "strong tie" trusted partners. When the demand for information in a policy domain is high, then both cocktail equilibria and "chum" equilibria (all strong-tie networks) prevail. We then turn to an empirical analysis of lobbyist contact-making and access, using the data of Laumann and Knoke in The Organizational State. We analyze the communication structure of the policy domains in health policy, using count data models that are adjusted for "structural autocorrelation" by the networks we study. The results support the cocktail equilibrium hypothesis, and offer a result that portends rich questions for future research: Washington lobbyists appear to overinvest in strong ties, in general reducing their credibility with the government in the long-term, as well as reducing the informational efficiency of the overall communication network.
Esterling, Kevin M., Lazer, David, Carpenter, Daniel
Submitted: 1997-08-18
Keywords: lobbying models, combinatorial optimization, count models, network-dependent estimation, structural autocorrelation
Abstract: (click to show/hide) This paper examines the relationship between lobbyists' contact-making behavior and their long-term access to the government. Specifically: 1) Do lobbyists establish social contacts in an individually-rational manner to best receive information from each other? And, 2) does the resulting network position condition their access to the government? We begin by wedding rational choice models to network analysis with a formal model of lobbyists' choice of contacts in a network, adopting the classic combinatorial optimization approach of Boorman (1975). The model predicts that when the demand for political information is low, a cocktail equilibrium prevails: lobbyists will invest their time in gaining "weak tie" acquaintances rather than in gaining "strong tie" trusted partners. When the demand for information in a policy domain is high, then both cocktail equilibria and "chum" equilibria (all strong-tie networks) prevail. We then turn to an empirical analysis of lobbyist contact-making and access, using the data of Laumann and Knoke in The Organizational State. We analyze the communication structure of the policy domains in health policy, using count data models that are adjusted for "structural autocorrelation" by the networks we study. The results support the cocktail equilibrium hypothesis, and offer a result that portends rich questions for future research: Washington lobbyists appear to overinvest in strong ties, in general reducing their credibility with the government in the long-term, as well as reducing the informational efficiency of the overall communication network.
The "Miracle" Revisited: An Examination of The Micro-Foundations of Aggregate Public Opinion
Berinsky, Adam
Submitted: 1997-08-18
Keywords: public opinion, heteroskedastic probit, ordered probit, selection bias, item non-response
Abstract: (click to show/hide) One of the best-known findings in the public opinion literature is that individual responses to survey questions, by and large, both exhibit little constraint and are highly unstable over time. One response to this bleak finding has been to search for coherence and stability at the aggregate level. Scholars who adopt this approach -- most notably Page and Shapiro (1992) -- argue that though most individuals are poorly informed about politics and may have unstable attitudes, the "miracle TRUNCATED.
Berinsky, Adam
Submitted: 1997-08-18
Keywords: public opinion, heteroskedastic probit, ordered probit, selection bias, item non-response
Abstract: (click to show/hide) One of the best-known findings in the public opinion literature is that individual responses to survey questions, by and large, both exhibit little constraint and are highly unstable over time. One response to this bleak finding has been to search for coherence and stability at the aggregate level. Scholars who adopt this approach -- most notably Page and Shapiro (1992) -- argue that though most individuals are poorly informed about politics and may have unstable attitudes, the "miracle TRUNCATED.
Macroideology, Macropartisanship, and the Gender Gap
Box-Steffensmeier, Janet M., De Boef, Suzanna, Lin, Tse-Min
Submitted: 1997-08-19
Keywords: Gender gap, macropartisanship, macroideology, fractional integration, Granger causality
Abstract: (click to show/hide) Ever since the late 1970s, women have been significantly more likely to identify with the Democratic Party than man. Explanations of the phenomenon usually link the rise of the gender gap to the political swing to conservatism and its attacks on the welfare state. We propose to study this linkage by way of examining time series data aggregated from CBS News/New York Times National Surveys. Specifically, we propose to test two alternative perspectives that predict such a linkage: that women are more compassionate than men and that women are more welfare-dependent than men. On the basis of these perspectives, we develop hypotheses concerning the causality between the gender gap in macropartisanship and conservative macroideology. We apply Granger causality tests after fitting the appropriate AutoRegressive Fractionally Integrated Moving Average (ARFIMA) models to the relevant time series. Our findings confirm that the aggregate partisan gender gap since the late 1970s has been a direct response to the dynamics of conservative macroideology. While feminine compassion may be a universal factor, the evidence for its influence behind the gender gap is most prevalent among partisans who are less welfare-dependent. Among welfare-dependent partisans, our evidence suggests that self- interested behavior is more apparent.
Box-Steffensmeier, Janet M., De Boef, Suzanna, Lin, Tse-Min
Submitted: 1997-08-19
Keywords: Gender gap, macropartisanship, macroideology, fractional integration, Granger causality
Abstract: (click to show/hide) Ever since the late 1970s, women have been significantly more likely to identify with the Democratic Party than man. Explanations of the phenomenon usually link the rise of the gender gap to the political swing to conservatism and its attacks on the welfare state. We propose to study this linkage by way of examining time series data aggregated from CBS News/New York Times National Surveys. Specifically, we propose to test two alternative perspectives that predict such a linkage: that women are more compassionate than men and that women are more welfare-dependent than men. On the basis of these perspectives, we develop hypotheses concerning the causality between the gender gap in macropartisanship and conservative macroideology. We apply Granger causality tests after fitting the appropriate AutoRegressive Fractionally Integrated Moving Average (ARFIMA) models to the relevant time series. Our findings confirm that the aggregate partisan gender gap since the late 1970s has been a direct response to the dynamics of conservative macroideology. While feminine compassion may be a universal factor, the evidence for its influence behind the gender gap is most prevalent among partisans who are less welfare-dependent. Among welfare-dependent partisans, our evidence suggests that self- interested behavior is more apparent.
Concordance, Projection, and Citizen Perceptions of Roll Call Voting
Wilson, J. Matthew, Gronke, Paul
Submitted: 1997-08-21
Keywords: roll call votes, congress, representation
Abstract: (click to show/hide) Democratic government is permised on some level of representation of constituent opionion by elected legislators. Our work examines this relationship from the perspective of the individual constituent. We assess the degree and causes of constituent knowledge about particular roll call votes. In the current work, we select a high profile domestic policy issue -- the 1994 Crime Bill -- as our focus. We show how that citizen misperception of their representative's position on this piece of legislation was rampant, with false positives (erroneously attributing support) far outweighing false negatives (erroneously attributing opposition). These descriptive results alone stand in constrast to previous findings regarding the 1991 Gulf War vote and the 1992 vote to confirm Clarence Thomas to the Supreme Court (Alvarez and Gronke, 1996a, 1996b). Next, in a series of causal models, we show how concordance and projection dominate citizen attributions for this particular bill. These processes may work to the benefit of incumbents, since people who are unable to recall a particular vote are most likely to project their own attitudes onto the representative.
Wilson, J. Matthew, Gronke, Paul
Submitted: 1997-08-21
Keywords: roll call votes, congress, representation
Abstract: (click to show/hide) Democratic government is permised on some level of representation of constituent opionion by elected legislators. Our work examines this relationship from the perspective of the individual constituent. We assess the degree and causes of constituent knowledge about particular roll call votes. In the current work, we select a high profile domestic policy issue -- the 1994 Crime Bill -- as our focus. We show how that citizen misperception of their representative's position on this piece of legislation was rampant, with false positives (erroneously attributing support) far outweighing false negatives (erroneously attributing opposition). These descriptive results alone stand in constrast to previous findings regarding the 1991 Gulf War vote and the 1992 vote to confirm Clarence Thomas to the Supreme Court (Alvarez and Gronke, 1996a, 1996b). Next, in a series of causal models, we show how concordance and projection dominate citizen attributions for this particular bill. These processes may work to the benefit of incumbents, since people who are unable to recall a particular vote are most likely to project their own attitudes onto the representative.
Economics, Entitlements and Social Issues: Voter Choice in the 1996 Presidential Election
Alvarez, R. Michael, Nagler, Jonathan
Submitted: 1997-08-21
Keywords: elections, issues, ideology, economic voting, economy, multinomial probit
Abstract: (click to show/hide) In this paper we examine three sets of explanations for the outcome of the 1996 presidential election campaign. First, we look at the effects of voter perceptions of the national economy on voter support for Clinton. Second we look at the effects of candidate and voter positions on a number of issues and on ideology. Last, we seek to understand whether other issues --- social issues such as abortion as well as issues revolving around entitlements and taxation --- played significant roles in this election. Thus this work extends the work of Alvarez and Nagler (1995), and enriches it with analysis of a more comprehensive set of issues considered. In the end, we are able to pull together each of these different sets of explanations into a consistent analysis of the 1996 presidential election which shows why Clinton won this race, but which also helps us understand why it was that both Dole and Perot fell so far from electoral victory.
Alvarez, R. Michael, Nagler, Jonathan
Submitted: 1997-08-21
Keywords: elections, issues, ideology, economic voting, economy, multinomial probit
Abstract: (click to show/hide) In this paper we examine three sets of explanations for the outcome of the 1996 presidential election campaign. First, we look at the effects of voter perceptions of the national economy on voter support for Clinton. Second we look at the effects of candidate and voter positions on a number of issues and on ideology. Last, we seek to understand whether other issues --- social issues such as abortion as well as issues revolving around entitlements and taxation --- played significant roles in this election. Thus this work extends the work of Alvarez and Nagler (1995), and enriches it with analysis of a more comprehensive set of issues considered. In the end, we are able to pull together each of these different sets of explanations into a consistent analysis of the 1996 presidential election which shows why Clinton won this race, but which also helps us understand why it was that both Dole and Perot fell so far from electoral victory.
The Consequences of Majority-Minority Districts for Representation: Evidence of Partisan Mobilization, Countermobilization and Demobilization
Brandt, Patrick T., Bailey, Michael
Submitted: 1997-08-21
Keywords: Multinomial probit, panel data methods, simulated maximum likelihood, probability simulation, redistricting
Abstract: (click to show/hide) Few analyses of the effects of race-based congressional redistricting have used survey data to analyze the implications of redistricting. This type of micro-level data can add significant intuition to aggregate data analysis. This paper looks at whether voters respond to redistricting by mobilizing, demobilizing, or countermobilizing using panel data from the 1990-1992 National Election Study. A 2-period vote choice model is estimated using a multiperiod multinomial probit model, and controlling for the effects of redistricting. Results show that the presence of black Democratic candidates in majority-minority districts after redistricting reduces turnout by white voters for the Democratic candidates.
Brandt, Patrick T., Bailey, Michael
Submitted: 1997-08-21
Keywords: Multinomial probit, panel data methods, simulated maximum likelihood, probability simulation, redistricting
Abstract: (click to show/hide) Few analyses of the effects of race-based congressional redistricting have used survey data to analyze the implications of redistricting. This type of micro-level data can add significant intuition to aggregate data analysis. This paper looks at whether voters respond to redistricting by mobilizing, demobilizing, or countermobilizing using panel data from the 1990-1992 National Election Study. A 2-period vote choice model is estimated using a multiperiod multinomial probit model, and controlling for the effects of redistricting. Results show that the presence of black Democratic candidates in majority-minority districts after redistricting reduces turnout by white voters for the Democratic candidates.
Linking Representation and House Member Behavior to Constituents' Voting Behavior
Box-Steffensmeier, Janet M., Kimball, David, Tate, Katherine
Submitted: 1997-08-21
Keywords: representation, NES, legislative activity, Congress, public approval
Abstract: (click to show/hide) How one is represented should shape one's opinion about the incumbent's performance in office, one's opinion about Congress as an institution, and ultimately, one's underlying attitude toward democratic practices. In this paper, we develop and test a model in which voters evaluate members of Congress on the basis of their performance in office in addition to other factors. Adding contextual data to the 1994 National Election Study, we analyzed the effects of legislative activity, descriptive characteristics, legislative positions, and constituency variables on constituency recognition of the incumbent and vote choice. We found that the activities of legislators, such as speech making and bill sponsorship, committees membership, and campaign spending, have a significant impact on the attitudes of constituents toward their representatives. We also found party affiliation of House members as well as their gender and race to be important cues that influence their constituents' evaluations. Our research identifies the type of legislative activities that benefit House members the most politically. It also establishes that while voters continue to rely on informational short cuts, such as party membership, gender, and information gathered through campaigns, they also appear responsive to the representative's legislative behavior. Overall, active legislators have more knowledgeable, satisfied, and loyal constituents.
Box-Steffensmeier, Janet M., Kimball, David, Tate, Katherine
Submitted: 1997-08-21
Keywords: representation, NES, legislative activity, Congress, public approval
Abstract: (click to show/hide) How one is represented should shape one's opinion about the incumbent's performance in office, one's opinion about Congress as an institution, and ultimately, one's underlying attitude toward democratic practices. In this paper, we develop and test a model in which voters evaluate members of Congress on the basis of their performance in office in addition to other factors. Adding contextual data to the 1994 National Election Study, we analyzed the effects of legislative activity, descriptive characteristics, legislative positions, and constituency variables on constituency recognition of the incumbent and vote choice. We found that the activities of legislators, such as speech making and bill sponsorship, committees membership, and campaign spending, have a significant impact on the attitudes of constituents toward their representatives. We also found party affiliation of House members as well as their gender and race to be important cues that influence their constituents' evaluations. Our research identifies the type of legislative activities that benefit House members the most politically. It also establishes that while voters continue to rely on informational short cuts, such as party membership, gender, and information gathered through campaigns, they also appear responsive to the representative's legislative behavior. Overall, active legislators have more knowledgeable, satisfied, and loyal constituents.
Economic Conditions and Presidential Elections
Nagler, Jonathan, Willette, Jennifer R.
Submitted: 1997-08-21
Keywords: Elections, economic voting, replication
Abstract: (click to show/hide) One of the more robust findings over the last 50 years in research on elections has been the importance of macroeconomic conditions on voting in U.S. presidential elections. An important contribution to that literature was made by Steven Weatherford in a 1978 article demonstrating that working class voters are more sensitive to economic conditions than are middle class voters in their vote choice. Weatherford's result was based on the 1956 through 1960 elections. We replicate Weatherford's result for 1960, and show that the substantive finding is extremely sensitive to the definition of class. When using occupation groups as the measure of class, we are able to essentially replicate Weatherford's result. However, using income as the measure of class we do not find any evidence to support the same finding for 1960. We then extend the analysis to cover the period 1956 thru 1996 using both an income-based measure of class and an occupation-based measure of class. We show that there does not appear to be a clear pattern distinguishing levels of economic voting between working-class and middle-class voters; though using the occupation-based measure working class voters appear more sensitive to the economy in recent elections. Finally, we offer a new theory of economic voting. We propose that voters vote based on the economic performance of their economic reference group - rather than on their own personal finances or on the state of the national economy.
Nagler, Jonathan, Willette, Jennifer R.
Submitted: 1997-08-21
Keywords: Elections, economic voting, replication
Abstract: (click to show/hide) One of the more robust findings over the last 50 years in research on elections has been the importance of macroeconomic conditions on voting in U.S. presidential elections. An important contribution to that literature was made by Steven Weatherford in a 1978 article demonstrating that working class voters are more sensitive to economic conditions than are middle class voters in their vote choice. Weatherford's result was based on the 1956 through 1960 elections. We replicate Weatherford's result for 1960, and show that the substantive finding is extremely sensitive to the definition of class. When using occupation groups as the measure of class, we are able to essentially replicate Weatherford's result. However, using income as the measure of class we do not find any evidence to support the same finding for 1960. We then extend the analysis to cover the period 1956 thru 1996 using both an income-based measure of class and an occupation-based measure of class. We show that there does not appear to be a clear pattern distinguishing levels of economic voting between working-class and middle-class voters; though using the occupation-based measure working class voters appear more sensitive to the economy in recent elections. Finally, we offer a new theory of economic voting. We propose that voters vote based on the economic performance of their economic reference group - rather than on their own personal finances or on the state of the national economy.
Beyond Ordinary Logit: Taking Time Seriously in Binary Time-Series--Cross-Section Models
Beck, Nathaniel, Katz, Jonathan, Tucker, Richard
Submitted: 1997-08-22
Keywords: binary time-series--cross-section data, logit/probit, temporal dependence, grouped duration models, complementary log-log, cubic spline, economic interdependence, democratic peace, war
Abstract: (click to show/hide) Researchers typically analyze time-series--cross-section data with a binary dependent variable (BTSCS) using ordinary logit or probit. However, BTSCS observations are likely to violate the independence assumption of the ordinary logit or probit statistical model. It is well known that if the observations are temporally related that the results of an ordinary logit or probit analysis may be misleading. In this paper, we provide a simple diagnostic for temporal dependence and a simple remedy. Our remedy is based on the idea that BTSCS data is identical to grouped duration data. This remedy does not require the BTSCS analyst to acquire any further methodological skills and it can be easily implemented in any standard statistical software package. While our approach is suitable for any type of BTSCS data, we provide examples and applications from the field of International Relations, where BTSCS data is frequently used. We use our methodology to re-assess Oneal and Russett's (1997) findings regarding the relationship between economic interdependence, democracy, and peace. Our analyses show that 1) their finding that economic interdependence is associated with peace is an artifact of their failure to account for temporal dependence and 2) their finding that democracy inhibits conflict is upheld even taking duration dependence into account.
Beck, Nathaniel, Katz, Jonathan, Tucker, Richard
Submitted: 1997-08-22
Keywords: binary time-series--cross-section data, logit/probit, temporal dependence, grouped duration models, complementary log-log, cubic spline, economic interdependence, democratic peace, war
Abstract: (click to show/hide) Researchers typically analyze time-series--cross-section data with a binary dependent variable (BTSCS) using ordinary logit or probit. However, BTSCS observations are likely to violate the independence assumption of the ordinary logit or probit statistical model. It is well known that if the observations are temporally related that the results of an ordinary logit or probit analysis may be misleading. In this paper, we provide a simple diagnostic for temporal dependence and a simple remedy. Our remedy is based on the idea that BTSCS data is identical to grouped duration data. This remedy does not require the BTSCS analyst to acquire any further methodological skills and it can be easily implemented in any standard statistical software package. While our approach is suitable for any type of BTSCS data, we provide examples and applications from the field of International Relations, where BTSCS data is frequently used. We use our methodology to re-assess Oneal and Russett's (1997) findings regarding the relationship between economic interdependence, democracy, and peace. Our analyses show that 1) their finding that economic interdependence is associated with peace is an artifact of their failure to account for temporal dependence and 2) their finding that democracy inhibits conflict is upheld even taking duration dependence into account.
Measuring Party Cohesion on Roll Call Votes with an Application to the Labor Committee of the Chilean Senate
Londregan, John B.
Submitted: 1997-08-22
Keywords: Maximum Likelihood, Roll Call Voting, Chile
Abstract: (click to show/hide) I introduce measures of two forms of party cohesion, affinity, in which members of the same party share a similar ideological outlook, and would vote alike in any event, and discipline, in which legislators of the same party compromise their basic ideological positions on party votes. These measures are based on maximum likelihood estimates of a spatial model of voting. Applied to the Labor Committee of the Chilean Senate the analysis identifies substantial affinity among elected Senators from the ruing Concertacion coalition, while the Institutional Senators exhibit marked differences in their ideological affinities. Neither of the discipline measures exceeds the threshold of tatistical significance.
Londregan, John B.
Submitted: 1997-08-22
Keywords: Maximum Likelihood, Roll Call Voting, Chile
Abstract: (click to show/hide) I introduce measures of two forms of party cohesion, affinity, in which members of the same party share a similar ideological outlook, and would vote alike in any event, and discipline, in which legislators of the same party compromise their basic ideological positions on party votes. These measures are based on maximum likelihood estimates of a spatial model of voting. Applied to the Labor Committee of the Chilean Senate the analysis identifies substantial affinity among elected Senators from the ruing Concertacion coalition, while the Institutional Senators exhibit marked differences in their ideological affinities. Neither of the discipline measures exceeds the threshold of tatistical significance.
Partisan Strength and Uncertain Presidential Evaluations
Gronke, Paul
Submitted: 1997-08-22
Keywords: presidential approval, uncertainty, partisanship
Abstract: (click to show/hide) American presidents, as do all democratic political leaders, rely on popular support in order to promote their political agenda, gain legislative victories, and succeed at the ballot box. Presidential approval, however, displays more than just a mean value, it also has a variance. And even a well regarded political leader would prefer to avoid widely variant support. At the individual level, variance is analogous to the level of uncertainty that an individual has about presidential performance This paper demonstrates the central role that partisan attachments play in fostering clarity in presidential approval. In general, respondents with stronger partisan attachments, combined with issue positions favorable to the president, are far more certain in their approval response. Fascinating variations in the role that party played during the Reagan years, compared to Carter, Bush, and Clinton, suggest a complex interaction between partisan ties, presidential performance, and the particular occupant of the oval office. The paper draws on data from the National Election Studies, 1980--1994. Ordinary least squares regression models are estimated, and clear evidence of heteroskedasticity is shown. A more general model that includes both a model for the mean and for the variance is presented and estimated using the same set of data. The main hypotheses regarding partisan strength and response uncertainty are confirmed.
Gronke, Paul
Submitted: 1997-08-22
Keywords: presidential approval, uncertainty, partisanship
Abstract: (click to show/hide) American presidents, as do all democratic political leaders, rely on popular support in order to promote their political agenda, gain legislative victories, and succeed at the ballot box. Presidential approval, however, displays more than just a mean value, it also has a variance. And even a well regarded political leader would prefer to avoid widely variant support. At the individual level, variance is analogous to the level of uncertainty that an individual has about presidential performance This paper demonstrates the central role that partisan attachments play in fostering clarity in presidential approval. In general, respondents with stronger partisan attachments, combined with issue positions favorable to the president, are far more certain in their approval response. Fascinating variations in the role that party played during the Reagan years, compared to Carter, Bush, and Clinton, suggest a complex interaction between partisan ties, presidential performance, and the particular occupant of the oval office. The paper draws on data from the National Election Studies, 1980--1994. Ordinary least squares regression models are estimated, and clear evidence of heteroskedasticity is shown. A more general model that includes both a model for the mean and for the variance is presented and estimated using the same set of data. The main hypotheses regarding partisan strength and response uncertainty are confirmed.
Heterogeneity and Disperson in the Beta-Binomial Model
Palmquist, Bradley
Submitted: 1997-08-23
Keywords: beta-binomial, overdispersion, underdispersion, count models, abortion
Abstract: (click to show/hide) Count variables built up from sums of independent and identically distributed (IID) binary random variables can be easily modeled by the binomial distribution. But what happens to sums of binary random variables if they are not IDD? King (1989) and others have presented the Beta Binomial and Extended Beta Binomial distributions as a way of handling the overdispersion that results from heterogeneity. This model seems to work well for some examples such as the distribution of state by state totals like the number of school districts banning a book in a given year. Heterogeneity of book banning rates across states would produce overdispersion (greater variability than expected from a binomial model). But another obvious example, the heterogeneity among Senators in vote counts aggregated by roll call, cannot be directly modeled by the Beta Binomial models in the same way. In the toxicology literature, from which political scientists have borrowed the Beta Binomial models, the heterogeneity observed is across units (``litters"), not within. Under the most reasonable assumptions, heterogeneity among Senators in response probabilities either produces pure Binomial variation in vote counts or contributes to underdispersion from roll call to roll call. These results are shown analytically and by simulation. Then a preliminary analysis of data of this type -- repeated votes on abortion in the Senate from 1974 to 1994 -- is presented.
Palmquist, Bradley
Submitted: 1997-08-23
Keywords: beta-binomial, overdispersion, underdispersion, count models, abortion
Abstract: (click to show/hide) Count variables built up from sums of independent and identically distributed (IID) binary random variables can be easily modeled by the binomial distribution. But what happens to sums of binary random variables if they are not IDD? King (1989) and others have presented the Beta Binomial and Extended Beta Binomial distributions as a way of handling the overdispersion that results from heterogeneity. This model seems to work well for some examples such as the distribution of state by state totals like the number of school districts banning a book in a given year. Heterogeneity of book banning rates across states would produce overdispersion (greater variability than expected from a binomial model). But another obvious example, the heterogeneity among Senators in vote counts aggregated by roll call, cannot be directly modeled by the Beta Binomial models in the same way. In the toxicology literature, from which political scientists have borrowed the Beta Binomial models, the heterogeneity observed is across units (``litters"), not within. Under the most reasonable assumptions, heterogeneity among Senators in response probabilities either produces pure Binomial variation in vote counts or contributes to underdispersion from roll call to roll call. These results are shown analytically and by simulation. Then a preliminary analysis of data of this type -- repeated votes on abortion in the Senate from 1974 to 1994 -- is presented.
Early Warning of Conflict in Southern Lebanon using Hidden Markov Models
Schrodt, Philip A.
Submitted: 1997-08-24
Keywords: hidden Markov models, event data, early warning, international crisis, sequence analysis, Middle East, WEIS, BCOW
Abstract: (click to show/hide) This paper extends earlier work on the application of hidden Markov models (HMMs) to the problem of forecasting international conflict. HMMs are a sequence comparison method widely used in computerized speech recognition as a computationally efficient method of generalizing a set of sequences observed in a noisy environment. The technique is easily be adapted to work with sequences of international event data. The paper provides a theoretical "micro-foundation" for the use of sequence comparison in conflict early- warning based on coadaptation of organizational standard operating procedures. The left-right (LR) HMM used in speech recognition is first extended to a left-right-left (LRL) model that allows a crisis to escalate and de-escalate. This model is tested for its ability to correctly discriminate between BCOW crisis that involve and do not involve war. The LRL model provides slightly more accurate classification than the LR model. The interpretation of the hidden states in the LRL models, however, is more ambiguous than in the LR model. The HMM is then applied to the problem of forecasting the outbreak of armed violence between Israel and Arab forces in south Lebanon during the period 1979 to 1997 (excluding 1982-1985). An HMM is estimated using six cases of "tit-for-tat" escalation, then fitted to the entire time period. The model identifies about half of the TFT conflictsÑincluding all of the training casesÑthat occur in the full sequence, with only one false positive. This result suggests that HMMs could be used in an event-based monitoring system. However, the fit of the model is very sensitive to the number of days in a sequence when no events occurred, and consequently the fit measure is ineffective as an early warning indicator. Nonetheless, in a subset of models, the maximum likelihood estimate of the sequence of hidden Markov states provides a robust early warning indicator with a three to six-month lead. These models are valid in a split-sample test, and the patterns of cross-correlation of the individual states of the model are consistent with the theoretical expectations. While this approach clearly needs further validation, it appears promising. The paper concludes with observations on the extent to which the HMM approach can be generalized to other categories of conflict, some suggestions on how the method of estimation can be improved, and the implications that sequence-based forecasting techniques have for theories of the causes of conflict.
Schrodt, Philip A.
Submitted: 1997-08-24
Keywords: hidden Markov models, event data, early warning, international crisis, sequence analysis, Middle East, WEIS, BCOW
Abstract: (click to show/hide) This paper extends earlier work on the application of hidden Markov models (HMMs) to the problem of forecasting international conflict. HMMs are a sequence comparison method widely used in computerized speech recognition as a computationally efficient method of generalizing a set of sequences observed in a noisy environment. The technique is easily be adapted to work with sequences of international event data. The paper provides a theoretical "micro-foundation" for the use of sequence comparison in conflict early- warning based on coadaptation of organizational standard operating procedures. The left-right (LR) HMM used in speech recognition is first extended to a left-right-left (LRL) model that allows a crisis to escalate and de-escalate. This model is tested for its ability to correctly discriminate between BCOW crisis that involve and do not involve war. The LRL model provides slightly more accurate classification than the LR model. The interpretation of the hidden states in the LRL models, however, is more ambiguous than in the LR model. The HMM is then applied to the problem of forecasting the outbreak of armed violence between Israel and Arab forces in south Lebanon during the period 1979 to 1997 (excluding 1982-1985). An HMM is estimated using six cases of "tit-for-tat" escalation, then fitted to the entire time period. The model identifies about half of the TFT conflictsÑincluding all of the training casesÑthat occur in the full sequence, with only one false positive. This result suggests that HMMs could be used in an event-based monitoring system. However, the fit of the model is very sensitive to the number of days in a sequence when no events occurred, and consequently the fit measure is ineffective as an early warning indicator. Nonetheless, in a subset of models, the maximum likelihood estimate of the sequence of hidden Markov states provides a robust early warning indicator with a three to six-month lead. These models are valid in a split-sample test, and the patterns of cross-correlation of the individual states of the model are consistent with the theoretical expectations. While this approach clearly needs further validation, it appears promising. The paper concludes with observations on the extent to which the HMM approach can be generalized to other categories of conflict, some suggestions on how the method of estimation can be improved, and the implications that sequence-based forecasting techniques have for theories of the causes of conflict.
Evaluating Measures of Ideology
Bishin, Benjamin G., King, Gary, Zeng, Langche
Submitted: 1997-08-24
Keywords: FILTER, ADA, NOMINATE, ideology
Abstract: (click to show/hide) A vigorous debate has arisen over the metric used to measure ideology (Jackson and Kingdon 1992, Poole and Rosenthal 1985, Snyder 1991, Krehbiel 1993). Ideology is difficult to measure because legislator's statements may be politically motivated and insincere. This paper evaluates the accuracy of NOMINATE and ADA scores by comparing them to an independent measure, based on background characteristics, developed herein. By Forecasting the Ideology of Legislators Through Elite Response (FILTER), this measure avoids the problems inherent in use of the roll call vote metric. In addition, the FILTER methodology is generalizable to studies of other deliberative bodies. The results show that FILTER scores are highly correlated with NOMINATE and ADA scores.
Bishin, Benjamin G., King, Gary, Zeng, Langche
Submitted: 1997-08-24
Keywords: FILTER, ADA, NOMINATE, ideology
Abstract: (click to show/hide) A vigorous debate has arisen over the metric used to measure ideology (Jackson and Kingdon 1992, Poole and Rosenthal 1985, Snyder 1991, Krehbiel 1993). Ideology is difficult to measure because legislator's statements may be politically motivated and insincere. This paper evaluates the accuracy of NOMINATE and ADA scores by comparing them to an independent measure, based on background characteristics, developed herein. By Forecasting the Ideology of Legislators Through Elite Response (FILTER), this measure avoids the problems inherent in use of the roll call vote metric. In addition, the FILTER methodology is generalizable to studies of other deliberative bodies. The results show that FILTER scores are highly correlated with NOMINATE and ADA scores.
Non-Compulsory Voting in Australia?: what surveys can (and can't) tell us
Jackman, Simon
Submitted: 1997-08-25
Keywords: turnout, Australian politics, compulsory voting, political participation, counter-factuals, surveys, non-response, measurement error, social-desirability heuristic, question-order effects, simulation, parametric bootstrap
Abstract: (click to show/hide) Compulsory voting has come under close scrutiny in recent Australian political debate, and influential voices within the (conservative) Coalition government have called for its repeal. Conventional wisdom holds that a repeal of compulsory voting would result in a sizeable electoral boost for the Coalition; the proportion of Coalition voters who would not vote is thought to be smaller than the corresponding proportion of Labor voters. But estimates of Coalition gains under a return to voluntary turnout are quite rough-and-ready, relying on methods hampered by critical shortcomings. In this paper I focus on assessing the counter-factual of non-compulsory turnout via surveys: while turnout is compulsory in Australia, responding to surveys isn't, and the problems raised by high rates of non-response are especially pernicious in attempting to assess the counter-factual of voluntary turnout. Among survey respondents, social-desirability and question-order effects also encourage over-reports of the likelihood of voluntarily turning out. Taking non-response and measurement error into consideration, I conclude that survey-based estimates (a) significantly \emph{under-estimate} the extent to which turnout would \emph{decline} under a voluntary turnout regime; but (b) \emph{over-estimate} the extent to which a fall in turnout would work to the advantage of the Coalition parties. Nonetheless, the larger of the Coalition parties --- the Liberal Party --- unambiguously increases its vote share under a wide range of assumptions about who does and doesn't voluntarily turnout.
Jackman, Simon
Submitted: 1997-08-25
Keywords: turnout, Australian politics, compulsory voting, political participation, counter-factuals, surveys, non-response, measurement error, social-desirability heuristic, question-order effects, simulation, parametric bootstrap
Abstract: (click to show/hide) Compulsory voting has come under close scrutiny in recent Australian political debate, and influential voices within the (conservative) Coalition government have called for its repeal. Conventional wisdom holds that a repeal of compulsory voting would result in a sizeable electoral boost for the Coalition; the proportion of Coalition voters who would not vote is thought to be smaller than the corresponding proportion of Labor voters. But estimates of Coalition gains under a return to voluntary turnout are quite rough-and-ready, relying on methods hampered by critical shortcomings. In this paper I focus on assessing the counter-factual of non-compulsory turnout via surveys: while turnout is compulsory in Australia, responding to surveys isn't, and the problems raised by high rates of non-response are especially pernicious in attempting to assess the counter-factual of voluntary turnout. Among survey respondents, social-desirability and question-order effects also encourage over-reports of the likelihood of voluntarily turning out. Taking non-response and measurement error into consideration, I conclude that survey-based estimates (a) significantly \emph{under-estimate} the extent to which turnout would \emph{decline} under a voluntary turnout regime; but (b) \emph{over-estimate} the extent to which a fall in turnout would work to the advantage of the Coalition parties. Nonetheless, the larger of the Coalition parties --- the Liberal Party --- unambiguously increases its vote share under a wide range of assumptions about who does and doesn't voluntarily turnout.
Pauline, the Mainstream, and Political Elites: the place of race in Australian political ideology
Jackman, Simon
Submitted: 1997-08-25
Keywords: public opinion, political ideology, political elites, race, immigration, Australian politics, factor analysis, ideological locations, density estimation, plotting highest density regions
Abstract: (click to show/hide) An often heard claim in the current ``race debate'' is that Australia's major political parties are out of touch with ``mainstream'' Australia on issues related to race. Parallel surveys of the electorate and candidates in the 1996 Federal election allow this claim to be tested, with items tapping general ideological dispositions, but including questions about Aboriginal Australians, immigration, and links with Asia. I make three critical findings: \begin{itemize} \item the electorate holds quite conservative opinions on these issues relative to the candidates, and is quite distant from ALP candidates in particular; \item attitudes on racial issues are a powerful component of the electorate's otherwise relatively loosely organized political ideology, so much so that any categorisation of Australian political ideology ignoring race must be considered incomplete; \item racial attitudes cut across other components of the electorate's ideology, placing all the parties under internal ideological strains, but the ALP appears particularly vulnerable on this score. \end{itemize} While the data show the Coalition to be the net beneficiary of the ideological tensions posed by race, the formation of Pauline Hanson's One Nation party has exposed the Coalition's vulnerability to race as a cross-cutting political issue. Racial issues thus have many characteristics of a realigning dimension in Australian politics.
Jackman, Simon
Submitted: 1997-08-25
Keywords: public opinion, political ideology, political elites, race, immigration, Australian politics, factor analysis, ideological locations, density estimation, plotting highest density regions
Abstract: (click to show/hide) An often heard claim in the current ``race debate'' is that Australia's major political parties are out of touch with ``mainstream'' Australia on issues related to race. Parallel surveys of the electorate and candidates in the 1996 Federal election allow this claim to be tested, with items tapping general ideological dispositions, but including questions about Aboriginal Australians, immigration, and links with Asia. I make three critical findings: \begin{itemize} \item the electorate holds quite conservative opinions on these issues relative to the candidates, and is quite distant from ALP candidates in particular; \item attitudes on racial issues are a powerful component of the electorate's otherwise relatively loosely organized political ideology, so much so that any categorisation of Australian political ideology ignoring race must be considered incomplete; \item racial attitudes cut across other components of the electorate's ideology, placing all the parties under internal ideological strains, but the ALP appears particularly vulnerable on this score. \end{itemize} While the data show the Coalition to be the net beneficiary of the ideological tensions posed by race, the formation of Pauline Hanson's One Nation party has exposed the Coalition's vulnerability to race as a cross-cutting political issue. Racial issues thus have many characteristics of a realigning dimension in Australian politics.
Overtime Inference from Cross-Sectional Surveys
Schuessler, Alexander, Penubarti, Mohan
Submitted: 1997-08-26
Keywords: opinion polls, ecological inference, EI
Abstract: (click to show/hide) To derive panel inferences at the micro level from cross-sectional surveys invites an ecological inference problem. Drawing on King's EI method we derive an application that allows us to estimate overtime change from cross-sectional opinion surveys. We validate our application in panel data where magnitudes of micro-level change are known. We subsequently apply the method to independent surveys on presidential approval. In doing so, we detect micro-level volatility that has been unavailable to previous researchers, leading them to derive imprecise conclusions.
Schuessler, Alexander, Penubarti, Mohan
Submitted: 1997-08-26
Keywords: opinion polls, ecological inference, EI
Abstract: (click to show/hide) To derive panel inferences at the micro level from cross-sectional surveys invites an ecological inference problem. Drawing on King's EI method we derive an application that allows us to estimate overtime change from cross-sectional opinion surveys. We validate our application in panel data where magnitudes of micro-level change are known. We subsequently apply the method to independent surveys on presidential approval. In doing so, we detect micro-level volatility that has been unavailable to previous researchers, leading them to derive imprecise conclusions.
The Coalition-oriented Evolution of Vote Intentions across Regions and Levels of Political Awareness during the 1993 Canadian Election Campaign: Quotidian Markov Chain Models using Rolling Cross-section Data
Wand, Jonathan, Mebane, Walter R.
Submitted: 1997-08-28
Keywords: Markov chains, rolling cross-section data, macro data, categorical data, survey data, Canadian politics, strategic voting, coalitions, estimation
Abstract: (click to show/hide) We use survey data collected in Ontario and Quebec during the 1993 Canadian federal election to assess the extent to which voters were sensitive to the distribution of positions in special institutions that would possibly be created to handle negotiations between Quebec and the rest of Canada following a referendum on Quebec sovereignty expected after the election. We draw on a theory of coalition-oriented voting developed by Austin-Smith and Banks (1988) to argue that voters' anticipations regarding those institutions contributed to the catastrophic losses suffered by the Progressive Conservative party. We use a method we have developed for estimating discrete, finite-state Markov chain models from ``macro'' data to analyze the dynamics of individual choice probabilities in daily rolling cross-sectional survey data from 1993 Canadian Election Study. We allow each transition matrix to be updated as a function of daily vote support for either the Bloc or Reform to test for reactive coalition-oriented voting. We find significant reactive voting among Quebecois non-sovereigntists. The timing of these reactions depended on the individual's level of political awareness. In contrast, we find no evidence of reactive voting among either Quebecois sovereigntists or Ontario voters.
Wand, Jonathan, Mebane, Walter R.
Submitted: 1997-08-28
Keywords: Markov chains, rolling cross-section data, macro data, categorical data, survey data, Canadian politics, strategic voting, coalitions, estimation
Abstract: (click to show/hide) We use survey data collected in Ontario and Quebec during the 1993 Canadian federal election to assess the extent to which voters were sensitive to the distribution of positions in special institutions that would possibly be created to handle negotiations between Quebec and the rest of Canada following a referendum on Quebec sovereignty expected after the election. We draw on a theory of coalition-oriented voting developed by Austin-Smith and Banks (1988) to argue that voters' anticipations regarding those institutions contributed to the catastrophic losses suffered by the Progressive Conservative party. We use a method we have developed for estimating discrete, finite-state Markov chain models from ``macro'' data to analyze the dynamics of individual choice probabilities in daily rolling cross-sectional survey data from 1993 Canadian Election Study. We allow each transition matrix to be updated as a function of daily vote support for either the Bloc or Reform to test for reactive coalition-oriented voting. We find significant reactive voting among Quebecois non-sovereigntists. The timing of these reactions depended on the individual's level of political awareness. In contrast, we find no evidence of reactive voting among either Quebecois sovereigntists or Ontario voters.
Monetary Policy and Wage/Price Bargaining: Macro-Institutional Interactions in the Traded, Public, and Sheltered Sectors
Franzese, Robert
Submitted: 1997-09-05
Keywords: monetary policy, central bank independence, corporatism, sectoral structure
Abstract: (click to show/hide) Prepared for Hall, Peter and David Soskice, eds., _Varieties of Capitalism: The Chalenges Facing Contemporary Political Economies_ (forthcoming, title preliminary). This chapter considers the politico-economic management of unemployment and inflation in developed capitalist democracies, focusing on the institutional and structural features of labor and goods markets and the credibility and conservatism of the monetary-policy authority. It reviews the central-bank-independence (CBI) and coordinated-wage-/price-bargaining (CWB) literatures and then offers a synthesis and extension which emphasizes that the degrees of CBI and CWB interact, with each other and with the sectoral structure of the economy, to structure the incentives facing the politico-economic actors involved in monetary policy and wage/price bargaining. The empirical records of 21 OECD countries in the post-Bretton Woods era are then used to evaluate the emergent hypotheses. The conclusion addresses two questions of pressing intellectual and practical concern: the likely impact of and independent European central bank and the roots of the "collapse" of CWB.
Franzese, Robert
Submitted: 1997-09-05
Keywords: monetary policy, central bank independence, corporatism, sectoral structure
Abstract: (click to show/hide) Prepared for Hall, Peter and David Soskice, eds., _Varieties of Capitalism: The Chalenges Facing Contemporary Political Economies_ (forthcoming, title preliminary). This chapter considers the politico-economic management of unemployment and inflation in developed capitalist democracies, focusing on the institutional and structural features of labor and goods markets and the credibility and conservatism of the monetary-policy authority. It reviews the central-bank-independence (CBI) and coordinated-wage-/price-bargaining (CWB) literatures and then offers a synthesis and extension which emphasizes that the degrees of CBI and CWB interact, with each other and with the sectoral structure of the economy, to structure the incentives facing the politico-economic actors involved in monetary policy and wage/price bargaining. The empirical records of 21 OECD countries in the post-Bretton Woods era are then used to evaluate the emergent hypotheses. The conclusion addresses two questions of pressing intellectual and practical concern: the likely impact of and independent European central bank and the roots of the "collapse" of CWB.
Precise, Second-Order Correct Estimates of the Natural Rate of Unemployment via Bootstrap Calibration
Sekhon, Jasjeet
Submitted: 1997-09-11
Keywords: Natural Rate of Unemployment, NAIRU, Unemployment, Bootstrap, Calibration, Confidence Interval, Likelihood, Marginal Likelihood, Conditional Likelihood, Profile Likelihood
Abstract: (click to show/hide) The natural rate of unemployment, which is usually considered to be the nonaccelerating inflation rate of unemployment (NAIRU), is an important and often used economic variable (Gordon 1997). In this paper I present second-order correct $O_{p}(n^{-2})$ confidence regions for estimates of the NAIRU, obtained via bootstrap calibration. These confidence regions are three times smaller than those provided in recent econometric work (Staiger, Stock and Watson 1997a, 1997b). The confidence regions are sufficiently precise to support use of the NAIRU for a variety of analytical and policy purposes, including monetary policy, as determined by a criterion suggested by Krueger (1997).
Sekhon, Jasjeet
Submitted: 1997-09-11
Keywords: Natural Rate of Unemployment, NAIRU, Unemployment, Bootstrap, Calibration, Confidence Interval, Likelihood, Marginal Likelihood, Conditional Likelihood, Profile Likelihood
Abstract: (click to show/hide) The natural rate of unemployment, which is usually considered to be the nonaccelerating inflation rate of unemployment (NAIRU), is an important and often used economic variable (Gordon 1997). In this paper I present second-order correct $O_{p}(n^{-2})$ confidence regions for estimates of the NAIRU, obtained via bootstrap calibration. These confidence regions are three times smaller than those provided in recent econometric work (Staiger, Stock and Watson 1997a, 1997b). The confidence regions are sufficiently precise to support use of the NAIRU for a variety of analytical and policy purposes, including monetary policy, as determined by a criterion suggested by Krueger (1997).
Generally Speaking... The Temporal Dimensions of Party Identification
Timpone, Richard J., Neely, Francis K.
Submitted: 1997-09-17
Keywords: party identification, measurement, structural equation models
Abstract: (click to show/hide) The measurement and stability of party identification has been an area of important debate for decades. We hypothesize that a systematic problem may exist in the traditional 7-point scale that has not yet been fully examined. We posit that the follow- up question asking independents the direction of their leaning is asking respondents a fundamentally different question than the other questions of general allegiance and strength. In short, the direction of leaning may be providing more direct evaluation of short-term factors such as candidate evaluation and vote choice. If this is the case, the direction of causality for these individuals may be mis-specified in some traditional studies of voting behavior. We examine this question by using the NES Gulf War Panel Study to investigate the systematic determinants of change in individual party identification through structural equation modeling. While short-term forces do not appear to significantly influence the strength or direction of partisanship for individuals identifying with the parties, they are significantly related to direction of leaning for partisans. This differential endogeneity between short-term forces and party identification has important implications for the measurement and use of the concept of partisanship. It also helps illuminate earlier work into 'paradoxical' relationships such as the intransitivity of candidate support and comparisons of different measures.
Timpone, Richard J., Neely, Francis K.
Submitted: 1997-09-17
Keywords: party identification, measurement, structural equation models
Abstract: (click to show/hide) The measurement and stability of party identification has been an area of important debate for decades. We hypothesize that a systematic problem may exist in the traditional 7-point scale that has not yet been fully examined. We posit that the follow- up question asking independents the direction of their leaning is asking respondents a fundamentally different question than the other questions of general allegiance and strength. In short, the direction of leaning may be providing more direct evaluation of short-term factors such as candidate evaluation and vote choice. If this is the case, the direction of causality for these individuals may be mis-specified in some traditional studies of voting behavior. We examine this question by using the NES Gulf War Panel Study to investigate the systematic determinants of change in individual party identification through structural equation modeling. While short-term forces do not appear to significantly influence the strength or direction of partisanship for individuals identifying with the parties, they are significantly related to direction of leaning for partisans. This differential endogeneity between short-term forces and party identification has important implications for the measurement and use of the concept of partisanship. It also helps illuminate earlier work into 'paradoxical' relationships such as the intransitivity of candidate support and comparisons of different measures.
The Economic Sophistication of Public Opinion in the United States
Sekhon, Jasjeet
Submitted: 1997-09-18
Keywords: Public Opinion, Economic Sophistication, Survey of Consumer Attitudes and Behavior (SCAB), Natural Rate of Unemployment, NAIRU, Unemployment, Bootstrap, Bootstrap Confidence Region
Abstract: (click to show/hide) I show that the public does indeed have coherent and sophisticated reactions to macroeconomic variables. These reactions are consistent with economic theory. Individuals form evaluations and expectations in a way which is sensitive to the complex trade-off between unemployment and inflation as determined by the nonaccelerating inflation rate of unemployment (NAIRU). The primary dataset used in this analysis has 69,680 observations and is compiled by merging 113 individual level ``Surveys of Consumer Attitudes and Behavior'' from 1976:01 to 1991:12. The data analysis makes extensive use of bootstrap methods to create confidence regions and to conduct hypothesis tests.
Sekhon, Jasjeet
Submitted: 1997-09-18
Keywords: Public Opinion, Economic Sophistication, Survey of Consumer Attitudes and Behavior (SCAB), Natural Rate of Unemployment, NAIRU, Unemployment, Bootstrap, Bootstrap Confidence Region
Abstract: (click to show/hide) I show that the public does indeed have coherent and sophisticated reactions to macroeconomic variables. These reactions are consistent with economic theory. Individuals form evaluations and expectations in a way which is sensitive to the complex trade-off between unemployment and inflation as determined by the nonaccelerating inflation rate of unemployment (NAIRU). The primary dataset used in this analysis has 69,680 observations and is compiled by merging 113 individual level ``Surveys of Consumer Attitudes and Behavior'' from 1976:01 to 1991:12. The data analysis makes extensive use of bootstrap methods to create confidence regions and to conduct hypothesis tests.
The Impact of Political Campaigns on the Effects of Political Sophistication
Fournier, Patrick P.
Submitted: 1997-09-18
Keywords: campaign, sophistication, information, heterogeneity, individual deviation, aggregate deviation, Canadian politics
Abstract: (click to show/hide) [none provided]
Fournier, Patrick P.
Submitted: 1997-09-18
Keywords: campaign, sophistication, information, heterogeneity, individual deviation, aggregate deviation, Canadian politics
Abstract: (click to show/hide) [none provided]
The Resurgence of Nativism in California? The Case of Proposition 187 and Illegal Immigration
Alvarez, R. Michael, Butterfield, Tara L.
Submitted: 1997-09-25
Keywords: two-stage probit, discrete choice, binary probit, propositions and initiatives, economic voting, illegal immigration, immigration reform, California politics
Abstract: (click to show/hide) We argue that support among California voters for Proposition 187 in 1994 was an example of cyclical nativism. This nativism was provoked primarily by California's economic downturn during the early 1990s. We develop four specific hypotheses to explain how poor economic conditions in California and the consequent nativistic sentiments would result in support for Proposition 187: 1) voters who believe that California's economic condition is poor will be more likely to support Proposition 187; 2) voters who perceive themselves as being economically threatened by illegal immigrants will be more likely to support Proposition 187; 3) voters with lower levels of education are more economically vulnerable and will be more likely to support Proposition 187; 4) voters in Southern California feel more directly affected by illegal immigration and will be more likely to support Proposition 187. To test these hypotheses, we analyze voter exit poll data from the 1994 California election. We utilize a two-stage probit model to allow for the endogeneity which results from the politicization of illegal immigration during this election. We find support for our hypotheses in the data. These findings cause us to conclude that nativism, fueled by economic conditions, was a salient factor leading many Californians to support Proposition 187.
Alvarez, R. Michael, Butterfield, Tara L.
Submitted: 1997-09-25
Keywords: two-stage probit, discrete choice, binary probit, propositions and initiatives, economic voting, illegal immigration, immigration reform, California politics
Abstract: (click to show/hide) We argue that support among California voters for Proposition 187 in 1994 was an example of cyclical nativism. This nativism was provoked primarily by California's economic downturn during the early 1990s. We develop four specific hypotheses to explain how poor economic conditions in California and the consequent nativistic sentiments would result in support for Proposition 187: 1) voters who believe that California's economic condition is poor will be more likely to support Proposition 187; 2) voters who perceive themselves as being economically threatened by illegal immigrants will be more likely to support Proposition 187; 3) voters with lower levels of education are more economically vulnerable and will be more likely to support Proposition 187; 4) voters in Southern California feel more directly affected by illegal immigration and will be more likely to support Proposition 187. To test these hypotheses, we analyze voter exit poll data from the 1994 California election. We utilize a two-stage probit model to allow for the endogeneity which results from the politicization of illegal immigration during this election. We find support for our hypotheses in the data. These findings cause us to conclude that nativism, fueled by economic conditions, was a salient factor leading many Californians to support Proposition 187.
Primary Election Systems and Policy Divergence
Morton, Becky, Gerber, Elisabeth R., Fowler, James
Submitted: 1997-10-03
Keywords: election laws, candidate nomination procedures, policy divergence, United States Congress
Abstract: (click to show/hide) We examine how differences in the institutions that regulate candidate nomination procedures - specifically direct primary election laws -- affect elite control over candidate nominations and ultimately affect candidate policy divergence. We hypothesize that in more closed primary systems, control over candidate nominations by ideological extremists will translate into a higher likelihood that extreme candidates win in the general election. We hypothesize that in more open systems, participation by a wider spectrum of the electorate means that candidates must appeal to more moderate voters, leading to the election of more moderate candidates. Using pooled cross-section time-series regression analysis, we find that US Representatives from states with closed primaries take policy positions that are furthest from their districts' estimated median voter's ideal positions. Representatives from states with semi-closed primaries are the most moderate. We conclude that the opportunities for strategic behavior created by electoral institutions have important consequences for electoral outcomes.
Morton, Becky, Gerber, Elisabeth R., Fowler, James
Submitted: 1997-10-03
Keywords: election laws, candidate nomination procedures, policy divergence, United States Congress
Abstract: (click to show/hide) We examine how differences in the institutions that regulate candidate nomination procedures - specifically direct primary election laws -- affect elite control over candidate nominations and ultimately affect candidate policy divergence. We hypothesize that in more closed primary systems, control over candidate nominations by ideological extremists will translate into a higher likelihood that extreme candidates win in the general election. We hypothesize that in more open systems, participation by a wider spectrum of the electorate means that candidates must appeal to more moderate voters, leading to the election of more moderate candidates. Using pooled cross-section time-series regression analysis, we find that US Representatives from states with closed primaries take policy positions that are furthest from their districts' estimated median voter's ideal positions. Representatives from states with semi-closed primaries are the most moderate. We conclude that the opportunities for strategic behavior created by electoral institutions have important consequences for electoral outcomes.
Estimating the Probability of Events That have Never Occurred: When Does Your Vote Matter?
Gelman, Andrew, King, Gary, Boscardin, John
Submitted: 1997-10-27
Keywords: conditional probability, decision analysis, elections, electoral campaigning, forecasting, political science, presidential elections, rare events, rational choice, subjective probability, voting power
Abstract: (click to show/hide) Researchers sometimes argue that statisticians have little to contribute when few realizations of the process being estimated are observed. We show that this argument is incorrect even in the extreme situation of estimating the probabilities of events so rare that they have never occurred. We show how statistical forecasting models allow us to use empirical data to improve inferences about the probabilities of these events. Our application is estimating the probability that your vote will be decisive in a U.S. presidential election, a problem that has been studied by political scientists for more than two decades. The exact value of this probability is of only minor interest, but the number has important implications for understanding the optimal allocation of campaign resources, whether states and voter groups receive their fair share of attention from prospective presidents, and how formal ``rational choice'' models of voter behavior might be able to explain why people vote at all. We show how the probability of a decisive vote can be estimated empirically from state-level forecasts of the presidential election and illustrate with the example of 1992. Based on generalizations of standard political science forecasting models, we estimate the (prospective) probability of a single vote being decisive as about 1 in 10 million for close national elections such as 1992, varying by about a factor of 10 among states. Our results support the argument that subjective probabilities of many types are best obtained via empirically-based statistical prediction models rather than solely mathematical reasoning. We discuss the implications of our findings for the types of decision analyses that are used in public choice studies.
Gelman, Andrew, King, Gary, Boscardin, John
Submitted: 1997-10-27
Keywords: conditional probability, decision analysis, elections, electoral campaigning, forecasting, political science, presidential elections, rare events, rational choice, subjective probability, voting power
Abstract: (click to show/hide) Researchers sometimes argue that statisticians have little to contribute when few realizations of the process being estimated are observed. We show that this argument is incorrect even in the extreme situation of estimating the probabilities of events so rare that they have never occurred. We show how statistical forecasting models allow us to use empirical data to improve inferences about the probabilities of these events. Our application is estimating the probability that your vote will be decisive in a U.S. presidential election, a problem that has been studied by political scientists for more than two decades. The exact value of this probability is of only minor interest, but the number has important implications for understanding the optimal allocation of campaign resources, whether states and voter groups receive their fair share of attention from prospective presidents, and how formal ``rational choice'' models of voter behavior might be able to explain why people vote at all. We show how the probability of a decisive vote can be estimated empirically from state-level forecasts of the presidential election and illustrate with the example of 1992. Based on generalizations of standard political science forecasting models, we estimate the (prospective) probability of a single vote being decisive as about 1 in 10 million for close national elections such as 1992, varying by about a factor of 10 among states. Our results support the argument that subjective probabilities of many types are best obtained via empirically-based statistical prediction models rather than solely mathematical reasoning. We discuss the implications of our findings for the types of decision analyses that are used in public choice studies.
Not Asked and Not Answered: Multiple Imputation for Multiple Surveys
Gelman, Andrew, King, Gary, Liu, Chuanhai
Submitted: 1997-10-27
Keywords: Bayesian inference, cluster sampling, diagnostics, hierarchical models, ignorable nonresponse, missing data, political science, sample surveys, stratified sampling, multiple imputation
Abstract: (click to show/hide) We present a method of analyzing a series of independent cross-sectional surveys in which some questions are not answered in some surveys and some respondents do not answer some of the questions posed. The method is also applicable to a single survey in which different questions are asked, or different sampling methods used, in different strata or clusters. Our method involves multiply-imputing the missing items and questions by adding to existing methods of imputation designed for single surveys a hierarchical regression model that allows covariates at the individual and survey levels. Information from survey weights is exploited by including in the analysis the variables on which the weights were based, and then reweighting individual responses (observed and imputed) to estimate population quantities. We also develop diagnostics for checking the fit of the imputation model based on comparing imputed to non-imputed data. We illustrate with the example that motivated this project --- a study of pre-election public opinion polls, in which not all the questions of interest are asked in all the surveys, so that it is infeasible to impute each survey separately.
Gelman, Andrew, King, Gary, Liu, Chuanhai
Submitted: 1997-10-27
Keywords: Bayesian inference, cluster sampling, diagnostics, hierarchical models, ignorable nonresponse, missing data, political science, sample surveys, stratified sampling, multiple imputation
Abstract: (click to show/hide) We present a method of analyzing a series of independent cross-sectional surveys in which some questions are not answered in some surveys and some respondents do not answer some of the questions posed. The method is also applicable to a single survey in which different questions are asked, or different sampling methods used, in different strata or clusters. Our method involves multiply-imputing the missing items and questions by adding to existing methods of imputation designed for single surveys a hierarchical regression model that allows covariates at the individual and survey levels. Information from survey weights is exploited by including in the analysis the variables on which the weights were based, and then reweighting individual responses (observed and imputed) to estimate population quantities. We also develop diagnostics for checking the fit of the imputation model based on comparing imputed to non-imputed data. We illustrate with the example that motivated this project --- a study of pre-election public opinion polls, in which not all the questions of interest are asked in all the surveys, so that it is infeasible to impute each survey separately.
Do Voters Learn from Presidential Election Campaigns?
Alvarez, R. Michael, Glasgow, Garrett
Submitted: 1997-10-27
Keywords: random effects panel models, content analysis, presidential election campaigns, voter decisionmaking, voter learning
Abstract: (click to show/hide) We present a model of voter campaign learning which is based on Bayesian learning models. This model assumes voters are imperfectly informed and that they incorporate new information into their existing perceptions about candidate issue positions in a systematic manner. Additional information made available to voters about candidate issue positions during the course of a political campaign will lead voters to have more precise perceptions of the issue positions of the candidates involved. We use panel survey data from the 1976 and 1980 presidental elections, combined with content analyses of the media during these same elections. Our primary analysis is conducted using random effects panel models. We find that during each of these campaigns many voters became better informed about the positions of candidates on many issues and that these changes in voter information are directly related to the information flow during each presidential campaign.
Alvarez, R. Michael, Glasgow, Garrett
Submitted: 1997-10-27
Keywords: random effects panel models, content analysis, presidential election campaigns, voter decisionmaking, voter learning
Abstract: (click to show/hide) We present a model of voter campaign learning which is based on Bayesian learning models. This model assumes voters are imperfectly informed and that they incorporate new information into their existing perceptions about candidate issue positions in a systematic manner. Additional information made available to voters about candidate issue positions during the course of a political campaign will lead voters to have more precise perceptions of the issue positions of the candidates involved. We use panel survey data from the 1976 and 1980 presidental elections, combined with content analyses of the media during these same elections. Our primary analysis is conducted using random effects panel models. We find that during each of these campaigns many voters became better informed about the positions of candidates on many issues and that these changes in voter information are directly related to the information flow during each presidential campaign.
The Reciprocal Relationship Between State Defense Interest and Committee Representation in Congress
Carsey, Thomas, Rundquist, Barry
Submitted: 1997-11-04
Keywords: Distributive Politics, LISREL, Pooled Time Series
Abstract: (click to show/hide) Does prior representation of a state on a Congressional defense committee lead to higher levels of per capita defense contract, or do higher levels of prior per capita contract awards to a state increase its probability of being represented on a defense committee? To solve this puzzle, we estimate a cross-lagged three-equation model on data from all 50 states from 1963 to 1989 using maximum likelihood within LISREL. We find a substantial reciprocal but non-confounding relationship between representation and the allocation of benefits for the House, but not for the Senate. Thus, for the House, this more appropriate model of distributive politics in Congress supports both the committee-induced benefits hypothesis and the recruitment hypothesis. Further, the paper elaborates on how this reciprocal relationship plays out over time.
Carsey, Thomas, Rundquist, Barry
Submitted: 1997-11-04
Keywords: Distributive Politics, LISREL, Pooled Time Series
Abstract: (click to show/hide) Does prior representation of a state on a Congressional defense committee lead to higher levels of per capita defense contract, or do higher levels of prior per capita contract awards to a state increase its probability of being represented on a defense committee? To solve this puzzle, we estimate a cross-lagged three-equation model on data from all 50 states from 1963 to 1989 using maximum likelihood within LISREL. We find a substantial reciprocal but non-confounding relationship between representation and the allocation of benefits for the House, but not for the Senate. Thus, for the House, this more appropriate model of distributive politics in Congress supports both the committee-induced benefits hypothesis and the recruitment hypothesis. Further, the paper elaborates on how this reciprocal relationship plays out over time.
A Test for Conformity in Voting Behavior
Coleman, Stephen
Submitted: 1997-11-10
Keywords: voting, turnout, political parties, conformity, entropy, mathematical model, United States, Germany, Japan, Russia
Abstract: (click to show/hide) This working paper shows how to test for the effect of social conformity on voting behavior, including its effect on voter turnout, the distribution of votes among political parties, and the relationship between turnout and party voting. A mathematical model is constructed that leads to specific predictions that are tested on national elections in several countries. Examples are shown graphically from presidential elections in the United States and parliamentary elections in Japan, Germany, Austria, and Russia. The predictions and results contrast with predictions of rational voting theory.
Coleman, Stephen
Submitted: 1997-11-10
Keywords: voting, turnout, political parties, conformity, entropy, mathematical model, United States, Germany, Japan, Russia
Abstract: (click to show/hide) This working paper shows how to test for the effect of social conformity on voting behavior, including its effect on voter turnout, the distribution of votes among political parties, and the relationship between turnout and party voting. A mathematical model is constructed that leads to specific predictions that are tested on national elections in several countries. Examples are shown graphically from presidential elections in the United States and parliamentary elections in Japan, Germany, Austria, and Russia. The predictions and results contrast with predictions of rational voting theory.
The Diffusion of Democracy, 1946-1994
O'Loughlin, John, Ward, Michael D., Lofdahl, Corey L., Cohen, Jordin S., Brown, David S., Reilly, David, Gleditsch, Kristian S., Shin, Michael E.
Submitted: 1997-11-12
Keywords: Spatial diffusion, , exploratory spatial data analysis, spatial statistics, regional effects, democracy, measures of democracy, space-time autocorrelation
Abstract: (click to show/hide) Research to date on democratization neglects the interconnections between temporal and spatial components that influence this process. This article presents research that reveals the relationship between the temporal and spatial aspects of democratic diffusion in the world-system since 1946. We provide strong and consistent evidence of temporal cascading of democratic and autocratic trends as well as strong spatial association (or autocorrelation) of authority structures. The analysis uses an exploratory data approach in a longitudinal framework to understand global and regional trends in democratization. Our work also reveals discrete changes in regimes that run counter to the dominant aggregate trends of democratic waves or sequences. We demonstrate how the ebb and flow of democracy varies between the world's regions. We conclude that further modeling of the process of regime change from autocracy to democracy as well as reversals should start from a "domain-specific" position that disaggregates the globe into its regional mosaics.
O'Loughlin, John, Ward, Michael D., Lofdahl, Corey L., Cohen, Jordin S., Brown, David S., Reilly, David, Gleditsch, Kristian S., Shin, Michael E.
Submitted: 1997-11-12
Keywords: Spatial diffusion, , exploratory spatial data analysis, spatial statistics, regional effects, democracy, measures of democracy, space-time autocorrelation
Abstract: (click to show/hide) Research to date on democratization neglects the interconnections between temporal and spatial components that influence this process. This article presents research that reveals the relationship between the temporal and spatial aspects of democratic diffusion in the world-system since 1946. We provide strong and consistent evidence of temporal cascading of democratic and autocratic trends as well as strong spatial association (or autocorrelation) of authority structures. The analysis uses an exploratory data approach in a longitudinal framework to understand global and regional trends in democratization. Our work also reveals discrete changes in regimes that run counter to the dominant aggregate trends of democratic waves or sequences. We demonstrate how the ebb and flow of democracy varies between the world's regions. We conclude that further modeling of the process of regime change from autocracy to democracy as well as reversals should start from a "domain-specific" position that disaggregates the globe into its regional mosaics.
Iff the Assumption Fits...: A Comment on the King Ecological Inference Solution
Tam, Wendy
Submitted: 1997-12-08
Keywords: ecological inference
Abstract: (click to show/hide) 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.
Tam, Wendy
Submitted: 1997-12-08
Keywords: ecological inference
Abstract: (click to show/hide) 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.
Conditional Party Government And Member Turnout On Senate Recorded Votes, 1873-1935
Sala, Brian R., Forgette, Richard
Submitted: 1997-12-29
Keywords: rational voter model, time series, roll-call voting
Abstract: (click to show/hide) According to the conditional party government thesis, party members bond or precommit themselves to supporting "party" positions under certain circumstances. A test of this thesis asks whether party members are more likely to participate in a roll call vote when the question has been identified by party leaders as important to the party. We show that party leadership signals systematically affected member turnout levels in the U.S. Senate during 1873-1935. On average, two-party turnout on party-salient votes rose by more than five members during 1873-1923 and more than three members during 1923-35 relative to "non-salient" votes. These results also provide evidence of cohesive partisan behavior in the Senate well before the parties began the regular practice of designating floor leaders and whips.
Sala, Brian R., Forgette, Richard
Submitted: 1997-12-29
Keywords: rational voter model, time series, roll-call voting
Abstract: (click to show/hide) According to the conditional party government thesis, party members bond or precommit themselves to supporting "party" positions under certain circumstances. A test of this thesis asks whether party members are more likely to participate in a roll call vote when the question has been identified by party leaders as important to the party. We show that party leadership signals systematically affected member turnout levels in the U.S. Senate during 1873-1935. On average, two-party turnout on party-salient votes rose by more than five members during 1873-1923 and more than three members during 1923-35 relative to "non-salient" votes. These results also provide evidence of cohesive partisan behavior in the Senate well before the parties began the regular practice of designating floor leaders and whips.
