Working Papers
2001
38 records found
The Foundations of Latino Voter Partisanship
Alvarez, R. Michael, Bedolla, Lisa Garcia
Submitted: 2001-03-08
Keywords: Latino, Partisanship, 2000 presidential election
Abstract: (click to show/hide) Traditionally, the Latino electorate has been considered to be Democratic in partisan affiliation. However, during the 2000 presidential election there were many efforts made by the Republican party to court Latino voters, suggesting that perhaps Latino voters may becoming more Republican in orientation. Using a telephone survey of Latino likely voters conducted in the 2000 election, we examine three different sets of correlates of Latino voter partisanship: social and demographic, issue and ideological, and economic. We find that in Latino voter partisanship is strongly structured by social and demographic, as well as issue and ideological, factors. We also find that while it is unlikely that changes in economic factors or abortion attitudes will significantly change which parties the different Latino nation-origin groups identify with, it is possible that changes in ideological positoins regarding the role of government in providing social services could result in significant changes in Latino party identification.
Alvarez, R. Michael, Bedolla, Lisa Garcia
Submitted: 2001-03-08
Keywords: Latino, Partisanship, 2000 presidential election
Abstract: (click to show/hide) Traditionally, the Latino electorate has been considered to be Democratic in partisan affiliation. However, during the 2000 presidential election there were many efforts made by the Republican party to court Latino voters, suggesting that perhaps Latino voters may becoming more Republican in orientation. Using a telephone survey of Latino likely voters conducted in the 2000 election, we examine three different sets of correlates of Latino voter partisanship: social and demographic, issue and ideological, and economic. We find that in Latino voter partisanship is strongly structured by social and demographic, as well as issue and ideological, factors. We also find that while it is unlikely that changes in economic factors or abortion attitudes will significantly change which parties the different Latino nation-origin groups identify with, it is possible that changes in ideological positoins regarding the role of government in providing social services could result in significant changes in Latino party identification.
Zen and the Art of Policy Analysis: A Response to Nielsen and Wolf
Meier, Kenneth J., Eller, Warren, Wrinkle, Robert D., Polinard, J. L.
Submitted: 2001-03-12
Keywords: education, pooled time series
Abstract: (click to show/hide) Neilsen and Wolf (N.d.) lodge several criticism of Meier, Wrinkle and Polinard (1999). Although most of the criticisms deal with tangential issues rather than our core argument, their criticisms are flawed by misguided estimation strategies, erroneous results, and an inattention to existing theory and scholarship. Our re-analysis of their work demonstrates these problems and presents even stronger evidence for our initial conclusion–both minority and Anglo students perform better in schools with more minority teachers.
Meier, Kenneth J., Eller, Warren, Wrinkle, Robert D., Polinard, J. L.
Submitted: 2001-03-12
Keywords: education, pooled time series
Abstract: (click to show/hide) Neilsen and Wolf (N.d.) lodge several criticism of Meier, Wrinkle and Polinard (1999). Although most of the criticisms deal with tangential issues rather than our core argument, their criticisms are flawed by misguided estimation strategies, erroneous results, and an inattention to existing theory and scholarship. Our re-analysis of their work demonstrates these problems and presents even stronger evidence for our initial conclusion–both minority and Anglo students perform better in schools with more minority teachers.
Public Opinion During The Vietnam War: A Revised Measure Of The Public Will
Berinsky, Adam
Submitted: 2001-04-02
Keywords: non-response, public opinion
Abstract: (click to show/hide) The conception of opinion polls as "broadly representative" of public sentiment has long pervaded academic and popular discussions of polls. In 1939, polling pioneer George Gallup advanced the virtues of surveys as a means for political elites to assess the collective "mandate of the people." If properly designed and conducted, Gallup argued, polls would act as a "sampling referendum" and provide a more accurate measure of popular opinion than more traditional methods, such as reading mail from constituents and attending to newspapers (see also Gallup and Rae 1940; for a contrary view, see Ginsberg 1986, Herbst 1993). More recently, Verba has argued, "sample surveys provide the closest approximation to an unbiased representation of the public because participation in a survey requires no resources and because surveys eliminate the bias inherent in the fact that participants in politics are self-selected -- surveys produce just what democracy is supposed to produce -- equal representation of all citizens"(1996, 3). Thus, while surveys may be limited in several respects they appear to provide a requisite egalitarian compliment to traditional forms of political participation. Through opinion polls, the voice of "the people," writ broadly, may be heard. Or maybe not. In this paper, I reconsider this conventional wisdom. Specifically, I demonstrate that the imbalance in political rhetoric surrounding the Vietnam War disadvantaged those groups who were the natural opponents of the War. I investigate the effect of accounting for "don't know" responses on the shape of public opinion on the Vietnam issue using a number of datasets from the 1960s and early 1970s and find that analyses that use very different data sources converge to the same conclusion. The process of collecting opinion on Vietnam excluded a dovish segment of the population from the collective opinion signal in the early part of the war. However, this bias shrank over time as anti-war messages became more common in the public sphere. To use the language of Verba, Schlozman, and Brady, the "voice" of those who abstained from the Vietnam questions was different from those who responded to such items. So while there may indeed have been a "silent majority" -- as President Nixon maintained during the early years of his presidency -- it was a majority that opposed, rather than supported, the war.
Berinsky, Adam
Submitted: 2001-04-02
Keywords: non-response, public opinion
Abstract: (click to show/hide) The conception of opinion polls as "broadly representative" of public sentiment has long pervaded academic and popular discussions of polls. In 1939, polling pioneer George Gallup advanced the virtues of surveys as a means for political elites to assess the collective "mandate of the people." If properly designed and conducted, Gallup argued, polls would act as a "sampling referendum" and provide a more accurate measure of popular opinion than more traditional methods, such as reading mail from constituents and attending to newspapers (see also Gallup and Rae 1940; for a contrary view, see Ginsberg 1986, Herbst 1993). More recently, Verba has argued, "sample surveys provide the closest approximation to an unbiased representation of the public because participation in a survey requires no resources and because surveys eliminate the bias inherent in the fact that participants in politics are self-selected -- surveys produce just what democracy is supposed to produce -- equal representation of all citizens"(1996, 3). Thus, while surveys may be limited in several respects they appear to provide a requisite egalitarian compliment to traditional forms of political participation. Through opinion polls, the voice of "the people," writ broadly, may be heard. Or maybe not. In this paper, I reconsider this conventional wisdom. Specifically, I demonstrate that the imbalance in political rhetoric surrounding the Vietnam War disadvantaged those groups who were the natural opponents of the War. I investigate the effect of accounting for "don't know" responses on the shape of public opinion on the Vietnam issue using a number of datasets from the 1960s and early 1970s and find that analyses that use very different data sources converge to the same conclusion. The process of collecting opinion on Vietnam excluded a dovish segment of the population from the collective opinion signal in the early part of the war. However, this bias shrank over time as anti-war messages became more common in the public sphere. To use the language of Verba, Schlozman, and Brady, the "voice" of those who abstained from the Vietnam questions was different from those who responded to such items. So while there may indeed have been a "silent majority" -- as President Nixon maintained during the early years of his presidency -- it was a majority that opposed, rather than supported, the war.
Selection Bias in Studies of Sanctions Efficacy
Nooruddin, Irfan
Submitted: 2001-04-05
Keywords: sanctions, strategic censoring, censored probit
Abstract: (click to show/hide) Sanctions rarely work but they continue to be used frequently by policymakers. Previous research on the determinants of sanctions identifies various factors that are thought to contribute to sanctions success but do not give us an answer to the original puzzle of why this ineffective policy is so commonly used. I argue that this is because studies of sanctions have ignored the problem of strategic censoring by focusing only on cases of observed sanctions. In this paper, I develop a unified model of sanction imposition and success and test it using a simultaneous equation censored probit model. The selection- corrected sanction model finds that the process by which sanctions are imposed is linked to the process by which some succeed while others fail, and that the unmeasured factors that lead to sanction imposition are negatively related to their success.
Nooruddin, Irfan
Submitted: 2001-04-05
Keywords: sanctions, strategic censoring, censored probit
Abstract: (click to show/hide) Sanctions rarely work but they continue to be used frequently by policymakers. Previous research on the determinants of sanctions identifies various factors that are thought to contribute to sanctions success but do not give us an answer to the original puzzle of why this ineffective policy is so commonly used. I argue that this is because studies of sanctions have ignored the problem of strategic censoring by focusing only on cases of observed sanctions. In this paper, I develop a unified model of sanction imposition and success and test it using a simultaneous equation censored probit model. The selection- corrected sanction model finds that the process by which sanctions are imposed is linked to the process by which some succeed while others fail, and that the unmeasured factors that lead to sanction imposition are negatively related to their success.
Racial and Ethnic Heterogeneity and Competition in House
Branton, Regina P., Jones, Bradford S.
Submitted: 2001-04-13
Keywords: House elections, race and politics
Abstract: (click to show/hide) The principal focus of this paper is to examine how a U.S. House district's level of racial and ethnic heterogeneity is related to various indicators of electoral competition. Prior research examining the relationship between race and electoral competition has tended to focus on how a district's African American population is related to electoral outcomes. As much of this literature has focused on the important issue of racial redistricting, the primary interest in the distribution of the African American population has been reasonable (and appropriate given the research questions asked). The focus here is not directly on matters pertaining to redistricting and as such, we argue that the exclusive focus on black-white competition belies the fact that the United States is a considerably diverse country, in terms of the distribution of racial and ethnic minority groups. To understand how racial and ethnic heterogeneity impacts electoral competition in the House, we collected data on the distribution of whites, African Americans, Latinos, Asians, and Native Americans residing in U.S. House districts (using U.S. Census Bureau data). We then use an indicator measuring the degree of racial and ethnic fractionalization in district. This measure is used as a covariate in various models of incumbent electoral success. The data we use are longitudinal data on incumbent electoral success during the period 1972 to 1998. We estimate several models of electoral competition and find that racial and ethnic heterogeneity is associated with greater electoral volatility in primary elections and less volatility in general elections. Because minority group preferences tend toward the Democratic party, these relationships are more pronounced for Democrats than for Republicans. The implications of the differences in the kinds of districts Democrats represent compared to the kinds of districts Republicans represent are then discussed.
Branton, Regina P., Jones, Bradford S.
Submitted: 2001-04-13
Keywords: House elections, race and politics
Abstract: (click to show/hide) The principal focus of this paper is to examine how a U.S. House district's level of racial and ethnic heterogeneity is related to various indicators of electoral competition. Prior research examining the relationship between race and electoral competition has tended to focus on how a district's African American population is related to electoral outcomes. As much of this literature has focused on the important issue of racial redistricting, the primary interest in the distribution of the African American population has been reasonable (and appropriate given the research questions asked). The focus here is not directly on matters pertaining to redistricting and as such, we argue that the exclusive focus on black-white competition belies the fact that the United States is a considerably diverse country, in terms of the distribution of racial and ethnic minority groups. To understand how racial and ethnic heterogeneity impacts electoral competition in the House, we collected data on the distribution of whites, African Americans, Latinos, Asians, and Native Americans residing in U.S. House districts (using U.S. Census Bureau data). We then use an indicator measuring the degree of racial and ethnic fractionalization in district. This measure is used as a covariate in various models of incumbent electoral success. The data we use are longitudinal data on incumbent electoral success during the period 1972 to 1998. We estimate several models of electoral competition and find that racial and ethnic heterogeneity is associated with greater electoral volatility in primary elections and less volatility in general elections. Because minority group preferences tend toward the Democratic party, these relationships are more pronounced for Democrats than for Republicans. The implications of the differences in the kinds of districts Democrats represent compared to the kinds of districts Republicans represent are then discussed.
Uncertainty and American Public Opinion
Alvarez, R. Michael, Brehm, John, Wison, Catherine
Submitted: 2001-04-14
Keywords: uncertainty, opinion
Abstract: (click to show/hide) In this paper we describe three distinct causes (uncertainty, ambivalence and equivocation) of individual level variation in responses to survey questions soliciting opinion on public policy issues. We evaluate the predictive power of these causes using the heteroscedastic probit model, an inferential statistical technique that is well-suited to the study of variability in public opinion. We estimate this model using survey data on four public policy issues: abortion, affirmative action, school prayer and english-only mandates. Our results indicate that for these policy issues individual uncertainty is the primary cause of survey response variability.
Alvarez, R. Michael, Brehm, John, Wison, Catherine
Submitted: 2001-04-14
Keywords: uncertainty, opinion
Abstract: (click to show/hide) In this paper we describe three distinct causes (uncertainty, ambivalence and equivocation) of individual level variation in responses to survey questions soliciting opinion on public policy issues. We evaluate the predictive power of these causes using the heteroscedastic probit model, an inferential statistical technique that is well-suited to the study of variability in public opinion. We estimate this model using survey data on four public policy issues: abortion, affirmative action, school prayer and english-only mandates. Our results indicate that for these policy issues individual uncertainty is the primary cause of survey response variability.
Estimating Risk and Rate Levels, Ratios, and Differences in Case-Control Studies
King, Gary, Zeng, Langche
Submitted: 2001-05-06
Keywords: Logistic Models, Case-Control Studies, Relative Risk, Odds Ratio, Risk Ratio, Risk Difference, Hazard Rate, Rate Ratio, Rate Difference
Abstract: (click to show/hide) Classic (or ``cumulative'') case-control sampling designs do not admit inferences about quantities of interest other than risk ratios, and then only by making the rare events assumption. Probabilities, risk differences, and other quantities cannot be computed without knowledge of the population incidence fraction. Similarly, density (or ``risk set'') case-control sampling designs do not allow inferences about quantities other than the rate ratio. Rates, rate differences, cumulative rates, risks, and other quantities cannot be estimated unless auxiliary information about the underlying cohort such as the number of controls in each full risk set is available. Most scholars who have considered the issue recommend reporting more than just risk and rate ratios, but auxiliary population information needed to do this is not usually available. We address this problem by developing methods that allow valid inferences about all relevant quantities of interest from either type of case-control study when completely ignorant of or only partially knowledgeable about relevant auxiliary population information.
King, Gary, Zeng, Langche
Submitted: 2001-05-06
Keywords: Logistic Models, Case-Control Studies, Relative Risk, Odds Ratio, Risk Ratio, Risk Difference, Hazard Rate, Rate Ratio, Rate Difference
Abstract: (click to show/hide) Classic (or ``cumulative'') case-control sampling designs do not admit inferences about quantities of interest other than risk ratios, and then only by making the rare events assumption. Probabilities, risk differences, and other quantities cannot be computed without knowledge of the population incidence fraction. Similarly, density (or ``risk set'') case-control sampling designs do not allow inferences about quantities other than the rate ratio. Rates, rate differences, cumulative rates, risks, and other quantities cannot be estimated unless auxiliary information about the underlying cohort such as the number of controls in each full risk set is available. Most scholars who have considered the issue recommend reporting more than just risk and rate ratios, but auxiliary population information needed to do this is not usually available. We address this problem by developing methods that allow valid inferences about all relevant quantities of interest from either type of case-control study when completely ignorant of or only partially knowledgeable about relevant auxiliary population information.
How much does a vote count? Voting power, coalitions, and the Electoral College
Gelman, Andrew, Katz, Jonathan
Submitted: 2001-05-08
Keywords: coalition, decisive vote, electoral college, popular vote, voting power
Abstract: (click to show/hide) In an election the probability that a single voter is decisive is affected by the electoral system -- that is, the rule for aggregating votes into a single outcome. Under the assumption that all votes are equally likely (i.e., random voting), we prove that the average probability of a vote being decisive is maximized under a popular-vote (or simple majority) rule and is lower under any coalition system, such as the U.S. Electoral College system, no matter how complicated. Forming a coalition increases the decisive vote probability for the voters within a coalition, but the aggregate effect of coalitions is to decrease the average decisiveness of the population of voters. We then review results on voting power in an electoral college system. Under the random voting assumption, it is well known that the voters with the highest probability of decisiveness are those in large states. However, we show using empirical estimates of the closeness of historical U.S. Presidential elections that voters in small states have been advantaged because the random voting model overestimates the frequencies of close elections in the larger states. Finally, we estimate the average probability of decisiveness for all U.S. Presidential elections from 1960 to 2000 under three possible electoral systems: popular vote, electoral vote, and winner-take-all within Congressional districts. We find that the average probability of decisiveness is about the same under all three systems.
Gelman, Andrew, Katz, Jonathan
Submitted: 2001-05-08
Keywords: coalition, decisive vote, electoral college, popular vote, voting power
Abstract: (click to show/hide) In an election the probability that a single voter is decisive is affected by the electoral system -- that is, the rule for aggregating votes into a single outcome. Under the assumption that all votes are equally likely (i.e., random voting), we prove that the average probability of a vote being decisive is maximized under a popular-vote (or simple majority) rule and is lower under any coalition system, such as the U.S. Electoral College system, no matter how complicated. Forming a coalition increases the decisive vote probability for the voters within a coalition, but the aggregate effect of coalitions is to decrease the average decisiveness of the population of voters. We then review results on voting power in an electoral college system. Under the random voting assumption, it is well known that the voters with the highest probability of decisiveness are those in large states. However, we show using empirical estimates of the closeness of historical U.S. Presidential elections that voters in small states have been advantaged because the random voting model overestimates the frequencies of close elections in the larger states. Finally, we estimate the average probability of decisiveness for all U.S. Presidential elections from 1960 to 2000 under three possible electoral systems: popular vote, electoral vote, and winner-take-all within Congressional districts. We find that the average probability of decisiveness is about the same under all three systems.
The Binomial-Beta Hierarchical Model for Ecological Inference Revisited and Implemented via the ECM Algorithm
Mattos, Rogerio, Veiga, Alvaro
Submitted: 2001-05-21
Keywords: ecological inference, hierarchical models, binomial-beta distribution, ECM Algorithm
Abstract: (click to show/hide) The binomia-beta hierarchical model is a recent contribution to ecological inference. Developed for the 2x2 tables case and under a bayesian perspective, the model is based on compounding the binomial and the beta distributions into a hierarchical structure to describe the behavior of aggregate variables. From a sample of aggregate observations, inference with this model can be made with regard to the values of the unobservable disaggregate variables. The paper discusses some issues regarding the construction of this EI model: First, previous uses of compounded binomial and beta distributions in the EI literature are reviewed; second, a faster approach to use the model in practice, based on posterior maximization implemented via the ECM algorithm, is proposed and illustrated with an application to a real dataset; finally, limitations regarding the use of marginal posteriors for binomial probabilities as elements of inference (basically, the failure to respect the accounting identity) instead of the predictive densities for the binomial proportions are pointed, together with suggestions of principles for EI model building in general.
Mattos, Rogerio, Veiga, Alvaro
Submitted: 2001-05-21
Keywords: ecological inference, hierarchical models, binomial-beta distribution, ECM Algorithm
Abstract: (click to show/hide) The binomia-beta hierarchical model is a recent contribution to ecological inference. Developed for the 2x2 tables case and under a bayesian perspective, the model is based on compounding the binomial and the beta distributions into a hierarchical structure to describe the behavior of aggregate variables. From a sample of aggregate observations, inference with this model can be made with regard to the values of the unobservable disaggregate variables. The paper discusses some issues regarding the construction of this EI model: First, previous uses of compounded binomial and beta distributions in the EI literature are reviewed; second, a faster approach to use the model in practice, based on posterior maximization implemented via the ECM algorithm, is proposed and illustrated with an application to a real dataset; finally, limitations regarding the use of marginal posteriors for binomial probabilities as elements of inference (basically, the failure to respect the accounting identity) instead of the predictive densities for the binomial proportions are pointed, together with suggestions of principles for EI model building in general.
The Rules of Inference
King, Gary, Epstein, Lee
Submitted: 2001-06-25
Keywords: inference, empirical research, legal research
Abstract: (click to show/hide) Although the term "empirical research" has become commonplace in legal scholarship over the past two decades, law professors have, in fact, been conducting research that is empirical-i.e., learning about the world using quantitative data or qualitative information-for almost as long as they have been conducting research. For just as long, however, they have been proceeding with little awareness of, much less compliance with, the rules of inference, and without paying heed to the key lessons of the revolution in empirical analysis that has been taking place over the last century in other disciplines. The sustained, self-conscious attention to the methodology of empirical analysis so present in the journals in traditional academic fields is virtually nonexistent in the nation's law reviews. As a result, readers learn considerably less accurate information about the empirical world than the studies' stridently stated, but overly confident, conclusions suggest. To remedy this situation-both for the producers and consumers of empirical work-we adapt the rules of inference used in the natural and social sciences to the special needs, theories, and data in legal scholarship, and explicate them with extensive illustrations from existing research. We also offer suggestions on how to reorganize the infrastructure of teaching and research at law schools so that it can better support the creation of first-rate empirical research without compromising other important objectives.
King, Gary, Epstein, Lee
Submitted: 2001-06-25
Keywords: inference, empirical research, legal research
Abstract: (click to show/hide) Although the term "empirical research" has become commonplace in legal scholarship over the past two decades, law professors have, in fact, been conducting research that is empirical-i.e., learning about the world using quantitative data or qualitative information-for almost as long as they have been conducting research. For just as long, however, they have been proceeding with little awareness of, much less compliance with, the rules of inference, and without paying heed to the key lessons of the revolution in empirical analysis that has been taking place over the last century in other disciplines. The sustained, self-conscious attention to the methodology of empirical analysis so present in the journals in traditional academic fields is virtually nonexistent in the nation's law reviews. As a result, readers learn considerably less accurate information about the empirical world than the studies' stridently stated, but overly confident, conclusions suggest. To remedy this situation-both for the producers and consumers of empirical work-we adapt the rules of inference used in the natural and social sciences to the special needs, theories, and data in legal scholarship, and explicate them with extensive illustrations from existing research. We also offer suggestions on how to reorganize the infrastructure of teaching and research at law schools so that it can better support the creation of first-rate empirical research without compromising other important objectives.
Automated Coding of International Event Data Using Sparse Parsing Techniques
Schrodt, Philip A.
Submitted: 2001-06-28
Keywords: event data, natural language processing, conflict, content analysis, open source
Abstract: (click to show/hide) "Event data" record the interactions of political actors reported in sources such as newspapers and news services; this type of data is widely used in research in international relations. Over the past ten years, there has been a shift from coding event data by humans -- typically university students -- to using computerized coding. The automated methods are dramatically faster, enabling data sets to be coded in real time, and provide far greater transparency and consistency than human coding. This paper reviews the experience of the Kansas Event Data System (KEDS) project in developing automated coding using "sparse parsing" machine coding methods, discusses a number of design decisions that were made in creating the program, and assesses features that would improve the effectiveness of these programs.
Schrodt, Philip A.
Submitted: 2001-06-28
Keywords: event data, natural language processing, conflict, content analysis, open source
Abstract: (click to show/hide) "Event data" record the interactions of political actors reported in sources such as newspapers and news services; this type of data is widely used in research in international relations. Over the past ten years, there has been a shift from coding event data by humans -- typically university students -- to using computerized coding. The automated methods are dramatically faster, enabling data sets to be coded in real time, and provide far greater transparency and consistency than human coding. This paper reviews the experience of the Kansas Event Data System (KEDS) project in developing automated coding using "sparse parsing" machine coding methods, discusses a number of design decisions that were made in creating the program, and assesses features that would improve the effectiveness of these programs.
Monitoring conflict using automated coding of newswire reports
Schrodt, Philip A., Gerner, Deborah J., Simpson Gerner, Erin M.
Submitted: 2001-06-28
Keywords: event data, natural language processing, conflict, content
Abstract: (click to show/hide) his paper discusses the experience of the Kansas Event Data System (KEDS) project in developing event data sets for monitoring conflict levels in five geographical areas: the Levant (Arab-Israeli conflict), Persian Gulf, former Yugoslavia, Central Asia (Afghanistan, Armenia-Azerbijan, former Soviet republics), and West Africa (Liberia, Sierra Leone). These data sets were coded from commercial news sources using the KEDS and TABARI automated coding systems. The paper discusses our experience in developing the dictionaries required for this coding, the problems with the number of reported events in the various areas, and provides examples of the statistical summaries that can be produced from event data. We also compare the coverage of the Reuters and Agence France Presse news services for selected years in the Levant and former Yugoslavia. We conclude with suggestions for four topics where additional efforts that could be usefully undertaken by multiple research projects.
Schrodt, Philip A., Gerner, Deborah J., Simpson Gerner, Erin M.
Submitted: 2001-06-28
Keywords: event data, natural language processing, conflict, content
Abstract: (click to show/hide) his paper discusses the experience of the Kansas Event Data System (KEDS) project in developing event data sets for monitoring conflict levels in five geographical areas: the Levant (Arab-Israeli conflict), Persian Gulf, former Yugoslavia, Central Asia (Afghanistan, Armenia-Azerbijan, former Soviet republics), and West Africa (Liberia, Sierra Leone). These data sets were coded from commercial news sources using the KEDS and TABARI automated coding systems. The paper discusses our experience in developing the dictionaries required for this coding, the problems with the number of reported events in the various areas, and provides examples of the statistical summaries that can be produced from event data. We also compare the coverage of the Reuters and Agence France Presse news services for selected years in the Levant and former Yugoslavia. We conclude with suggestions for four topics where additional efforts that could be usefully undertaken by multiple research projects.
Analyzing the Dynamics of International Mediation Processess in the Middle East and the former Yugoslavia
Gerner, Deborah J., Schrodt, Philip A.
Submitted: 2001-06-28
Keywords: mediation, event data, cross-correlation, conflict, Middle East
Abstract: (click to show/hide) This paper discusses a new National Science Foundation-funded project that will examine the dynamics of third-party international mediation using statistical time-series analyses of political event data. Third-party mediation was attempted in over half of the conflicts in the post-WWII period and it is likely that the use of mediation has increased following the end of the Cold War. Surprisingly, there have been few systematic studies on mediation. Those that do exist have generally focused on relatively static contextual factors such as the the conflict's attributes and the prior relationship between the mediator and protagonists rather than on dynamic factors' both contextual and process that may contribute to the success or failure of mediation activities. In contrast, the extensive qualitative literature provides numerous hypotheses about dynamic aspects of mediation. This, however, primarily consists of case studies, often by mediation practitioners, that exhibit little cumulation and, when taken as a whole, are rife with contradictory assertions. The project will formally test a number of the hypotheses embedded in the theoretical and qualitative literatures on mediation, using automated coding of event data from news-wire sources and employing time-series and event- history methods. A system of specialized event codes that a sensitive to mediation activities will be developed, then events will be coded from news reports using the TABARI machine coding program. The research will look at the factors that influence (1) whether mediation is accepted by the parties in a conflict, (2) whether formal agreements are reached, and (3) whether the agreements actually reduce the level of conflict. The project will initially focus on conflicts in the Middle East, a region where the principal investigators have substantial field experience. After refining the statistical tests on the Middle East case, the analysis will be extended to event data on conflicts in the former Yugoslavia and West Africa. The paper presents the results of an empirical "plausibility probe" based on existing WEIS-coded event data for the Levant and the former Yugoslavia. It employs a simple measure of third-party mediation efforts as the independent variables and Goldstein-scaled cooperation as the dependent variable. In the Levant, we find a weak but consistent pattern of mediation correlating with past conflictual activity, and resulting in later increases in cooperation. In the former Yugoslavia, the analysis shows strikingly different results for the mediation efforts the UN, European states, and the US. All three respond to increased conflict, but the UN efforts correlate with greater conflict, the US efforts with greater cooperation, and the European efforts have no effect. These results are consistent with many of the qualitative assessments of these efforts, and suggest that the event data approach will produce credible results
Gerner, Deborah J., Schrodt, Philip A.
Submitted: 2001-06-28
Keywords: mediation, event data, cross-correlation, conflict, Middle East
Abstract: (click to show/hide) This paper discusses a new National Science Foundation-funded project that will examine the dynamics of third-party international mediation using statistical time-series analyses of political event data. Third-party mediation was attempted in over half of the conflicts in the post-WWII period and it is likely that the use of mediation has increased following the end of the Cold War. Surprisingly, there have been few systematic studies on mediation. Those that do exist have generally focused on relatively static contextual factors such as the the conflict's attributes and the prior relationship between the mediator and protagonists rather than on dynamic factors' both contextual and process that may contribute to the success or failure of mediation activities. In contrast, the extensive qualitative literature provides numerous hypotheses about dynamic aspects of mediation. This, however, primarily consists of case studies, often by mediation practitioners, that exhibit little cumulation and, when taken as a whole, are rife with contradictory assertions. The project will formally test a number of the hypotheses embedded in the theoretical and qualitative literatures on mediation, using automated coding of event data from news-wire sources and employing time-series and event- history methods. A system of specialized event codes that a sensitive to mediation activities will be developed, then events will be coded from news reports using the TABARI machine coding program. The research will look at the factors that influence (1) whether mediation is accepted by the parties in a conflict, (2) whether formal agreements are reached, and (3) whether the agreements actually reduce the level of conflict. The project will initially focus on conflicts in the Middle East, a region where the principal investigators have substantial field experience. After refining the statistical tests on the Middle East case, the analysis will be extended to event data on conflicts in the former Yugoslavia and West Africa. The paper presents the results of an empirical "plausibility probe" based on existing WEIS-coded event data for the Levant and the former Yugoslavia. It employs a simple measure of third-party mediation efforts as the independent variables and Goldstein-scaled cooperation as the dependent variable. In the Levant, we find a weak but consistent pattern of mediation correlating with past conflictual activity, and resulting in later increases in cooperation. In the former Yugoslavia, the analysis shows strikingly different results for the mediation efforts the UN, European states, and the US. All three respond to increased conflict, but the UN efforts correlate with greater conflict, the US efforts with greater cooperation, and the European efforts have no effect. These results are consistent with many of the qualitative assessments of these efforts, and suggest that the event data approach will produce credible results
Inferring Transition Probabilities from Repeated Cross Sections: A Cross-level Inference Approach to US Presidential Voting
Pelzer, Ben, Eisniga, Rob, Franses, Philip Hans
Submitted: 2001-07-04
Keywords: Markov model, repeated cross sections, cross-level inference, ecological inference
Abstract: (click to show/hide) This paper outlines a nonstationary, heterogeneous Markov model designed to estimate entry and exit transition probabilities at the micro-level from a time series of independent cross-sectional samples with a binary outcome variable. The model has its origins in the work of Moffitt (1993) and shares features with standard statistical methods for ecological inference. We show how ML estimates of the parameters can be obtained by the method-of- scoring, how to estimate time-varying covariate effects, and how to include non-backcastable variables in the model. The latter extension of the basic model is an important one as it strongly increases its potential application in a wide array of research contexts. The example illustration uses survey data on American presidential vote intentions from a five-wave panel study conducted by Patterson (1980) in 1976. We treat the panel data as independent cross sections and compare the estimates of the Markov model with the observations in the panel. Directions for future work are discussed.
Pelzer, Ben, Eisniga, Rob, Franses, Philip Hans
Submitted: 2001-07-04
Keywords: Markov model, repeated cross sections, cross-level inference, ecological inference
Abstract: (click to show/hide) This paper outlines a nonstationary, heterogeneous Markov model designed to estimate entry and exit transition probabilities at the micro-level from a time series of independent cross-sectional samples with a binary outcome variable. The model has its origins in the work of Moffitt (1993) and shares features with standard statistical methods for ecological inference. We show how ML estimates of the parameters can be obtained by the method-of- scoring, how to estimate time-varying covariate effects, and how to include non-backcastable variables in the model. The latter extension of the basic model is an important one as it strongly increases its potential application in a wide array of research contexts. The example illustration uses survey data on American presidential vote intentions from a five-wave panel study conducted by Patterson (1980) in 1976. We treat the panel data as independent cross sections and compare the estimates of the Markov model with the observations in the panel. Directions for future work are discussed.
Using Auxiliary Data to Estimate Selection Bias Models
Boehmke, Frederick
Submitted: 2001-07-06
Keywords: selection bias, two-stage estimation, survey design, initiative, interest groups
Abstract: (click to show/hide) Recent work has made progress in estimating models involving selection bias of a particularly strong nature: all nonrespondents are unit nonresponders, meaning that no data is available for them. These models are reasonable successful in circumstances where the dependent variable of interest is continuous, but they are less practical empirically when it is latent and only discrete outcomes or choices are observed. I develop a method in this paper to estimate these models that is much more practical in terms of estimation. The model uses a small amount of auxiliary information to estimate the selection equation parameters which are then held fixed to estimate the equation of interest parameters in a maximum likelihood setting. After presenting monte carlo analysis to support the model, I apply the technique to a substantive problem: which interest groups are likely to to be involved in support of potential initiatives to achieve their policy goals.
Boehmke, Frederick
Submitted: 2001-07-06
Keywords: selection bias, two-stage estimation, survey design, initiative, interest groups
Abstract: (click to show/hide) Recent work has made progress in estimating models involving selection bias of a particularly strong nature: all nonrespondents are unit nonresponders, meaning that no data is available for them. These models are reasonable successful in circumstances where the dependent variable of interest is continuous, but they are less practical empirically when it is latent and only discrete outcomes or choices are observed. I develop a method in this paper to estimate these models that is much more practical in terms of estimation. The model uses a small amount of auxiliary information to estimate the selection equation parameters which are then held fixed to estimate the equation of interest parameters in a maximum likelihood setting. After presenting monte carlo analysis to support the model, I apply the technique to a substantive problem: which interest groups are likely to to be involved in support of potential initiatives to achieve their policy goals.
Mixed Logit Models in Political Science
Glasgow, Garrett
Submitted: 2001-07-08
Keywords: mixed logit, discrete choice, heterogeneity
Abstract: (click to show/hide) Mixed logit (MXL) is a general discrete choice model that is applicable to a wide range of political science problems. Mixed logit assumes the unobserved portions of utility are a mixture of an IID extreme value term and another multivariate distribution selected by the researcher. This general specification allows MXL to avoid imposing the independence of irrelevant alternatives (IIA) property on the choice probabilities. Further, and more importantly, MXL is a flexible tool for examining heterogeneity in individual behavior through random-coefficients specifications. Three empirical examples are presented. Two are drawn from studies of voting behavior. The first uses data from the 1987 British general election and examines heterogeneity in the impact of social class on voting, and the second uses data from the 1992 U.S. presidential election and examines heterogeneity in the impact of union membership on voting. A third example examines heterogeneity in the factors that lead to various Congressional career decisions. These empirical examples demonstrate the utility of mixed logit in political science research. This paper has both a methodological and substantive contribution for political science. Methodologically, it expands the tool set available to researchers for studying various phenomena in political science. More importantly, this paper contributes substantively by allowing for more realistic models of individual behavior. Most models currently used in political science assume the independent variables have a homogeneous effect on the dependent variable. This assumption is usually made to keep models tractable, even though few believe it is an accurate description of behavior. MXL is a tractable way to relax this assumption and study heterogeneity in a variety of settings.
Glasgow, Garrett
Submitted: 2001-07-08
Keywords: mixed logit, discrete choice, heterogeneity
Abstract: (click to show/hide) Mixed logit (MXL) is a general discrete choice model that is applicable to a wide range of political science problems. Mixed logit assumes the unobserved portions of utility are a mixture of an IID extreme value term and another multivariate distribution selected by the researcher. This general specification allows MXL to avoid imposing the independence of irrelevant alternatives (IIA) property on the choice probabilities. Further, and more importantly, MXL is a flexible tool for examining heterogeneity in individual behavior through random-coefficients specifications. Three empirical examples are presented. Two are drawn from studies of voting behavior. The first uses data from the 1987 British general election and examines heterogeneity in the impact of social class on voting, and the second uses data from the 1992 U.S. presidential election and examines heterogeneity in the impact of union membership on voting. A third example examines heterogeneity in the factors that lead to various Congressional career decisions. These empirical examples demonstrate the utility of mixed logit in political science research. This paper has both a methodological and substantive contribution for political science. Methodologically, it expands the tool set available to researchers for studying various phenomena in political science. More importantly, this paper contributes substantively by allowing for more realistic models of individual behavior. Most models currently used in political science assume the independent variables have a homogeneous effect on the dependent variable. This assumption is usually made to keep models tractable, even though few believe it is an accurate description of behavior. MXL is a tractable way to relax this assumption and study heterogeneity in a variety of settings.
Bayesian Learning about Ideal Points of U.S. Supreme Court Justices, 1953-1999
Martin, Andrew D., Quinn, Kevin M.
Submitted: 2001-07-09
Keywords: item response models, dynamic linear models, Markov chain Monte Carlo
Abstract: (click to show/hide) At the heart of attitudinal and strategic explanations of judicial behavior is the assumption that justices have policy preferences. These preferences have been measured in a handful of ways, including using factor analysis and multidimensional scaling techniques (Schubert, 1965, 1974), looking at past votes in a single policy area (Epstein et al., 1989), content-analyzing newspaper editorials at the time of appointment to the Court (Segal and Cover, 1989), and recording the background characteristics of the justices (Tate and Handberg, 1991). In this manuscript we employ Markov chain Monte Carlo (MCMC) methods to Þt Bayesian measurement models of judicial preferences for all justices serving on the U.S. Supreme Court from 1953 to 1999. We are particularly interested in considering to what extent ideal points of justices change throughout their tenure on the Court, and how the proposals over which they are voting also change across time. To do so, we Þt four longitudinal item response models that include dynamic speciÞcations for the ideal points and the case-speciÞc parameters. Our results suggest that justices do not have constant ideal points, even after controlling for the types of cases that come before the Court.
Martin, Andrew D., Quinn, Kevin M.
Submitted: 2001-07-09
Keywords: item response models, dynamic linear models, Markov chain Monte Carlo
Abstract: (click to show/hide) At the heart of attitudinal and strategic explanations of judicial behavior is the assumption that justices have policy preferences. These preferences have been measured in a handful of ways, including using factor analysis and multidimensional scaling techniques (Schubert, 1965, 1974), looking at past votes in a single policy area (Epstein et al., 1989), content-analyzing newspaper editorials at the time of appointment to the Court (Segal and Cover, 1989), and recording the background characteristics of the justices (Tate and Handberg, 1991). In this manuscript we employ Markov chain Monte Carlo (MCMC) methods to Þt Bayesian measurement models of judicial preferences for all justices serving on the U.S. Supreme Court from 1953 to 1999. We are particularly interested in considering to what extent ideal points of justices change throughout their tenure on the Court, and how the proposals over which they are voting also change across time. To do so, we Þt four longitudinal item response models that include dynamic speciÞcations for the ideal points and the case-speciÞc parameters. Our results suggest that justices do not have constant ideal points, even after controlling for the types of cases that come before the Court.
An Estimator for Some Binary-Outcome Selection Models without Exclusion Restrictions
Sartori, Anne E.
Submitted: 2001-07-09
Keywords: selection bias, discrete choice, small-sample properties
Abstract: (click to show/hide) This paper provides a new estimator for selection models with dichotomous dependent variables when identical factors affect the selection equation and the equation of interest. Such situations arise naturally in game-theoretic models where selection is typically nonrandom and identical explanatory variables influence all decisions under investigation. When its own identifying assumption is reasonable, the estimator allows the researcher to avoid the painful choice among identifying from functional form alone (using a Heckman-type estimator), adding a theoretically unjustified variable to the selection equation in a mistaken attempt to "boost" identification, or giving upon estimation entirely. The paper compares the small-sample properties of the estimator with those of the Heckman- type estimator and ordinary probit using Monte Carlo methods. A brief analysis of the causes of enduring rivalries and war, following Lemke and Reed (2001),
Sartori, Anne E.
Submitted: 2001-07-09
Keywords: selection bias, discrete choice, small-sample properties
Abstract: (click to show/hide) This paper provides a new estimator for selection models with dichotomous dependent variables when identical factors affect the selection equation and the equation of interest. Such situations arise naturally in game-theoretic models where selection is typically nonrandom and identical explanatory variables influence all decisions under investigation. When its own identifying assumption is reasonable, the estimator allows the researcher to avoid the painful choice among identifying from functional form alone (using a Heckman-type estimator), adding a theoretically unjustified variable to the selection equation in a mistaken attempt to "boost" identification, or giving upon estimation entirely. The paper compares the small-sample properties of the estimator with those of the Heckman- type estimator and ordinary probit using Monte Carlo methods. A brief analysis of the causes of enduring rivalries and war, following Lemke and Reed (2001),
Time Series Cross-Sectional Analyses with Different Explanatory Variables in Each Cross-Section
Girosi, Federico, King, Gary
Submitted: 2001-07-11
Keywords: Bayesian hierarchical model, time series, cross-section
Abstract: (click to show/hide) The current animosity between quantitative cross-national comparativists and area studies scholars originated in the expanding geographic scope of data collection in the 1960s. As quantitative scholars sought to include more countries in their regressions, the measures they were able to find for all observations became less comparable, and those which were available (or appropriate) for fewer than the full set were excluded. Area studies scholars appropriately complain about the violence these procedures do to the political reality they find from their in depth analyses of individual countries, but as quantitative comparativists continue to seek systematic comparisons, the conflict continues. We attempt to eliminate a small piece of the basis of this conflict by developing models that enable comparativists to include different explanatory variables, or the same variables with different meanings, in the time-series regression in each country. This should permit more powerful statistical analyses and encourage more context-sensitive data collection strategies. We demonstrate the advantages of this approach in practice by showing how out-of-sample forecasts of mortality rates in 25 countries, 17 age groups, and 17 causes of death in males and 20 in females from this model out-perform a standard regression approach.
Girosi, Federico, King, Gary
Submitted: 2001-07-11
Keywords: Bayesian hierarchical model, time series, cross-section
Abstract: (click to show/hide) The current animosity between quantitative cross-national comparativists and area studies scholars originated in the expanding geographic scope of data collection in the 1960s. As quantitative scholars sought to include more countries in their regressions, the measures they were able to find for all observations became less comparable, and those which were available (or appropriate) for fewer than the full set were excluded. Area studies scholars appropriately complain about the violence these procedures do to the political reality they find from their in depth analyses of individual countries, but as quantitative comparativists continue to seek systematic comparisons, the conflict continues. We attempt to eliminate a small piece of the basis of this conflict by developing models that enable comparativists to include different explanatory variables, or the same variables with different meanings, in the time-series regression in each country. This should permit more powerful statistical analyses and encourage more context-sensitive data collection strategies. We demonstrate the advantages of this approach in practice by showing how out-of-sample forecasts of mortality rates in 25 countries, 17 age groups, and 17 causes of death in males and 20 in females from this model out-perform a standard regression approach.
A Seemingly Unrelated Regression Model for Analyzing Multiparty Elections
Jackson, John
Submitted: 2001-07-11
Keywords: elections, multiparty, methods
Abstract: (click to show/hide) This paper presents a model for analyzing the returns in multiparty elections. The dependent variable is the log of the ratio of each party's vote to the vote share of a base party, as in the King-Katz model. The estimator is a version of the seemingly unrelated regression model, thereby taking advantage of the properties and computational convenience of the linear model. The error structure is composed of two elements. The first is the conventional SUR type errors that are homoscedastic across voting districts. The second is a district specific error structure that is derived by treating the observed votes as a sample from a multivariate normal distribution of true party support. The paper derives the small sample properties of the estimator, which are important in many applications where there are not a large number of districts. The model is applied to the 1993 Polish parliamentary elections. The results from this analysis form the basis for monte carlo simulations comparing several different estimators.
Jackson, John
Submitted: 2001-07-11
Keywords: elections, multiparty, methods
Abstract: (click to show/hide) This paper presents a model for analyzing the returns in multiparty elections. The dependent variable is the log of the ratio of each party's vote to the vote share of a base party, as in the King-Katz model. The estimator is a version of the seemingly unrelated regression model, thereby taking advantage of the properties and computational convenience of the linear model. The error structure is composed of two elements. The first is the conventional SUR type errors that are homoscedastic across voting districts. The second is a district specific error structure that is derived by treating the observed votes as a sample from a multivariate normal distribution of true party support. The paper derives the small sample properties of the estimator, which are important in many applications where there are not a large number of districts. The model is applied to the 1993 Polish parliamentary elections. The results from this analysis form the basis for monte carlo simulations comparing several different estimators.
How Factual is your Counterfactual?
King, Gary, Zeng, Langche
Submitted: 2001-07-12
Keywords: counterfactual, causality, forecasting, democracy
Abstract: (click to show/hide) Inferences about counterfactuals are essential for prediction, answering ``what if'' questions, and estimating causal effects. However, when the counterfactuals posed are too far from the data at hand, conclusions drawn from well-specified statistical analyses become based on speculation and convenient but indefensible model assumptions rather than empirical evidence. Yet, standard model outputs do not reveal the degree of model-dependence, and so this problem can be hard to detect, regardless of its severity. We develop easy-to-apply methods to evaluate counterfactuals that do not require sensitivity testing over specified classes of models. One analysis with these methods applies to the class of all models, for any smooth conditional expectation function, and to the set of all possible dependent variables, given only the choice of a set of explanatory variables. We illustrate by studying the scholarly literatures that try to assess the effects of changes in the degree of democracy in a country (on any dependent variable); we find widespread evidence that scholars are inadvertently drawing conclusions based more on their hypotheses than on their empirical evidence.
King, Gary, Zeng, Langche
Submitted: 2001-07-12
Keywords: counterfactual, causality, forecasting, democracy
Abstract: (click to show/hide) Inferences about counterfactuals are essential for prediction, answering ``what if'' questions, and estimating causal effects. However, when the counterfactuals posed are too far from the data at hand, conclusions drawn from well-specified statistical analyses become based on speculation and convenient but indefensible model assumptions rather than empirical evidence. Yet, standard model outputs do not reveal the degree of model-dependence, and so this problem can be hard to detect, regardless of its severity. We develop easy-to-apply methods to evaluate counterfactuals that do not require sensitivity testing over specified classes of models. One analysis with these methods applies to the class of all models, for any smooth conditional expectation function, and to the set of all possible dependent variables, given only the choice of a set of explanatory variables. We illustrate by studying the scholarly literatures that try to assess the effects of changes in the degree of democracy in a country (on any dependent variable); we find widespread evidence that scholars are inadvertently drawing conclusions based more on their hypotheses than on their empirical evidence.
Fission and Fusion in a Party System
Benoit, Kenneth, Laver, Michael
Submitted: 2001-07-12
Keywords: party systems, dynamic models, power indexes, legislatures
Abstract: (click to show/hide) Existing work on party systems typically involves essentially static models and pays little attention to the dynamics of party splits and fusions. Our approach explores these dynamics by setting out a simple model of legislative behavior in a parliament responsible for making and breaking governments. This model abandons the unitary actor assumption about political parties models individual legislators as utility-maximizing agents tempted to defect to other parties if this would increase their expected payoffs. We first set out a dynamic model of party fission and fusion couched in these terms and discuss this analytically. We then explore unanswered questions computationally by generating a novel type of "metadata" set, comprising the entire universe of possible legislative party systems in parliaments with up to 10 parties, generating a total of 6,292,018 theoretically possible non-equivalent legislatures. Using this metadata set and building on analytic results, we set out to characterize what makes certain parties "attractive" to legislators from other parties in a dynamic system. The results reveal an inherent instability in party systems and identify legislative configurations more prone to fission and fusion. They also strikingly highlight the role of the largest party, regardless of it size, as being attractive to potential defectors from other parties. Finally, they highlight the relatively weak position of the second-largest party. This provides an intriguing new interpretation of the potential for intense competition between the largest two parties for the role of the largest party, in a generalization to multiparty systems of the "all or nothing" competition endemic in two-party systems.
Benoit, Kenneth, Laver, Michael
Submitted: 2001-07-12
Keywords: party systems, dynamic models, power indexes, legislatures
Abstract: (click to show/hide) Existing work on party systems typically involves essentially static models and pays little attention to the dynamics of party splits and fusions. Our approach explores these dynamics by setting out a simple model of legislative behavior in a parliament responsible for making and breaking governments. This model abandons the unitary actor assumption about political parties models individual legislators as utility-maximizing agents tempted to defect to other parties if this would increase their expected payoffs. We first set out a dynamic model of party fission and fusion couched in these terms and discuss this analytically. We then explore unanswered questions computationally by generating a novel type of "metadata" set, comprising the entire universe of possible legislative party systems in parliaments with up to 10 parties, generating a total of 6,292,018 theoretically possible non-equivalent legislatures. Using this metadata set and building on analytic results, we set out to characterize what makes certain parties "attractive" to legislators from other parties in a dynamic system. The results reveal an inherent instability in party systems and identify legislative configurations more prone to fission and fusion. They also strikingly highlight the role of the largest party, regardless of it size, as being attractive to potential defectors from other parties. Finally, they highlight the relatively weak position of the second-largest party. This provides an intriguing new interpretation of the potential for intense competition between the largest two parties for the role of the largest party, in a generalization to multiparty systems of the "all or nothing" competition endemic in two-party systems.
Alternative Models of Dynamics in Binary Time-Series--Cross-Section Models: The Example of State Failure
Beck, Nathaniel, Jackman, Simon, Epstein, David, O'Halloran, Sharyn
Submitted: 2001-07-14
Keywords: dynamic probit, btscs, state failure, Gibbs sampling, MCMC, transitional models, discrete data, ROC, correlated binary data, generalized residuals
Abstract: (click to show/hide) This paper investigates a variety of dynamic probit models for time-series--cross-section data in the context of explaining state failure. It shows that ordinary probit, which ignores dynamics, is misleading. Alternatives that seem to produce sensible results are the transition model and a model which includes a lagged \emph{latent} dependent variable. It is argued that the use of a lagged latent variable is often superior to the use of a lagged realized dependent variable. It is also shown that the latter is a special case of the transition model. The relationship between the transition model and event history methods is also considered: the transition model estimates an event history model for both values of the dependent variable, yielding estimates that are identical to those produced by the two event history models. Furthermore, one can incorporate the insights gleaned from the event history models into the transition analysis, so that researchers do not have to assume duration independence. The conclusion notes that investigations of the various models have been limited to data sets which contain long sequences of zeros; models may perform differently in data sets with shorter bursts of zeros and ones.
Beck, Nathaniel, Jackman, Simon, Epstein, David, O'Halloran, Sharyn
Submitted: 2001-07-14
Keywords: dynamic probit, btscs, state failure, Gibbs sampling, MCMC, transitional models, discrete data, ROC, correlated binary data, generalized residuals
Abstract: (click to show/hide) This paper investigates a variety of dynamic probit models for time-series--cross-section data in the context of explaining state failure. It shows that ordinary probit, which ignores dynamics, is misleading. Alternatives that seem to produce sensible results are the transition model and a model which includes a lagged \emph{latent} dependent variable. It is argued that the use of a lagged latent variable is often superior to the use of a lagged realized dependent variable. It is also shown that the latter is a special case of the transition model. The relationship between the transition model and event history methods is also considered: the transition model estimates an event history model for both values of the dependent variable, yielding estimates that are identical to those produced by the two event history models. Furthermore, one can incorporate the insights gleaned from the event history models into the transition analysis, so that researchers do not have to assume duration independence. The conclusion notes that investigations of the various models have been limited to data sets which contain long sequences of zeros; models may perform differently in data sets with shorter bursts of zeros and ones.
Is All Politics and Economics Local? National Elections and Local
Wawro, Gregory, Himmelberg, Charles P.
Submitted: 2001-07-14
Keywords: elections, economic conditions, voting behavior, aggregation
Abstract: (click to show/hide) Scholars have long sought to understand the causal relationships between economics and political participation. Of particular concern has been how economic experiences have affected individuals' decisions to participate in elections and cast votes for candidates of different political parties. Practically all of the studies on elections in the United States have focused on national aggregate economic conditions and national aggregate political outcomes, while only a handful of studies have focused on whether state and local economic conditions affect federal elections. The conclusion one would reach from these studies is that the adage ``all politics is local'' does not apply to economics and elections. In fact, despite the findings of some early studies (e.g. Tufte 1975), recent research would lead us to conclude that economic conditions have no direct effects on congressional elections (Erikson 1990; Alesina and Rosenthal 1995). According to these recent studies, the economy is related to congressional elections only indirectly through its effects on presidential elections. And even in presidential elections, a key economic indicator---unemployment---appears to have little to no effect on presidential elections. In this paper, we question the conclusions of previous studies by considering how the failure to correctly model vote shares at the local level could produce misleading results on the effects for economic conditions on elections in local analysis. We develop a model for local vote shares by adapting a model derived in the empirical literature on demand for differentiated products. Our model explicitly accounts for nonlinearity and aggregation in vote share functions and so avoids some of the problems of standard linear specifications of vote shares that are common in the literature. We estimate our model using data at the local level to assess the impact of economic conditions on presidential vote shares and turnout in the 1992 election. We find that local unemployment does affect presidential votes and these effects vary by demographic groups in interesting ways.
Wawro, Gregory, Himmelberg, Charles P.
Submitted: 2001-07-14
Keywords: elections, economic conditions, voting behavior, aggregation
Abstract: (click to show/hide) Scholars have long sought to understand the causal relationships between economics and political participation. Of particular concern has been how economic experiences have affected individuals' decisions to participate in elections and cast votes for candidates of different political parties. Practically all of the studies on elections in the United States have focused on national aggregate economic conditions and national aggregate political outcomes, while only a handful of studies have focused on whether state and local economic conditions affect federal elections. The conclusion one would reach from these studies is that the adage ``all politics is local'' does not apply to economics and elections. In fact, despite the findings of some early studies (e.g. Tufte 1975), recent research would lead us to conclude that economic conditions have no direct effects on congressional elections (Erikson 1990; Alesina and Rosenthal 1995). According to these recent studies, the economy is related to congressional elections only indirectly through its effects on presidential elections. And even in presidential elections, a key economic indicator---unemployment---appears to have little to no effect on presidential elections. In this paper, we question the conclusions of previous studies by considering how the failure to correctly model vote shares at the local level could produce misleading results on the effects for economic conditions on elections in local analysis. We develop a model for local vote shares by adapting a model derived in the empirical literature on demand for differentiated products. Our model explicitly accounts for nonlinearity and aggregation in vote share functions and so avoids some of the problems of standard linear specifications of vote shares that are common in the literature. We estimate our model using data at the local level to assess the impact of economic conditions on presidential vote shares and turnout in the 1992 election. We find that local unemployment does affect presidential votes and these effects vary by demographic groups in interesting ways.
Delaying Justice(s): A Duration Analysis of Supreme Court Nominations
Shipan, Charles R., Shannon, Megan L.
Submitted: 2001-07-16
Keywords: Hazard Model, Spatial Model, Supreme Court Nominations
Abstract: (click to show/hide) Presidents have great success when nominating justices to the Supreme Court, with confirmation being the norm and rejection being the rare exception. In this paper we show that while the end result of the confirmation process is that the nominee taking a seat on the Court, there is a great deal of variance in the amount of time it takes the Senate to approve the nominee. To derive a theoretical explanation of this underlying dynamic in the confirmation process, we draw on a spatial model of presidential nominations to the Court. We then use a hazard rate model to test this explanation, using data on all Supreme Court nominations and confirmations since the end of the Civil War. The hazard model is superior to alternative models such as probit, where information on right-censored nominations in our data would be lost. More specifically, the Cox proportional hazards model is a better fit for our data as compared to the Weibull, exponential, and log-logistic hazard models. Our paper thus makes two key contributions. First, it identifies the political factors that influence Supreme Court confirmations and the duration of the confirmation process. Second, it demonstrates the ways in which the nomination process affects the confirmation process.
Shipan, Charles R., Shannon, Megan L.
Submitted: 2001-07-16
Keywords: Hazard Model, Spatial Model, Supreme Court Nominations
Abstract: (click to show/hide) Presidents have great success when nominating justices to the Supreme Court, with confirmation being the norm and rejection being the rare exception. In this paper we show that while the end result of the confirmation process is that the nominee taking a seat on the Court, there is a great deal of variance in the amount of time it takes the Senate to approve the nominee. To derive a theoretical explanation of this underlying dynamic in the confirmation process, we draw on a spatial model of presidential nominations to the Court. We then use a hazard rate model to test this explanation, using data on all Supreme Court nominations and confirmations since the end of the Civil War. The hazard model is superior to alternative models such as probit, where information on right-censored nominations in our data would be lost. More specifically, the Cox proportional hazards model is a better fit for our data as compared to the Weibull, exponential, and log-logistic hazard models. Our paper thus makes two key contributions. First, it identifies the political factors that influence Supreme Court confirmations and the duration of the confirmation process. Second, it demonstrates the ways in which the nomination process affects the confirmation process.
Analyzing the dynamics of international mediation processes
Schrodt, Philip A., Gerner, Deborah J.
Submitted: 2001-07-16
Keywords: event data, cross-correlation, mediation, Cox proportional hazard, pattern recognition
Abstract: (click to show/hide) This paper presents initial results from a project that will formally test a number of the hypotheses embedded in the theoretical and qualitative literatures on mediation, using automated coding of event data from news-wire sources. In contrast to most of the existing quantitative literature, which emphasizes the structural aspects of mediation, we will focus on the dynamics. The initial part of the paper focuses on two issues of design. First, we discuss the advantages of generating data using fully automated methods, which increases the transparency and replicability of the research. This transparency is extended to the development of more complex variables that cannot be captured as single events: these are defined as pattern of the underlying event data. We also suggest that these can be usefully studied using conventional inferential statistics rather than computational pattern recognition. Second, we justify the "statistical case study" approach which focuses on a small number of cases that are limited in geographical and temporal scope. While the risk of this approach is that one will find patterns of behavior that apply only in those circumstances, we point out that the more conventional large-N time-series cross-sectional studies also carry inferential risks. The statistical tests reported in this paper look at three different issues using data on the Israel-Lebanon and Israel-Palestinian conflicts in the Levant (1979-1999), and the Serbia-Croatia and Serbia-Bosnia conflicts in the Balkans (1991-1999). First, cross- correlation is used to look at the effects of mediation on the level of violence over time. Second, we test the "sticks-or-carrots" hypothesis on whether mediation is more effective in reducing violence if accompanied by cooperative or conflictual behavior by the mediator. Finally, we estimate Cox proportional hazard models to assess the factors that influence (1) whether mediation is accepted by the parties in a conflict, (2) whether formal agreements are reached, and (3) whether the agreements reduce the level of conflict. Future work in the project involves development of a new event coding scheme specifically designed for the study of mediation, and expansion of the list of cases to include other mediated conflicts in the Middle East and West Africa.
Schrodt, Philip A., Gerner, Deborah J.
Submitted: 2001-07-16
Keywords: event data, cross-correlation, mediation, Cox proportional hazard, pattern recognition
Abstract: (click to show/hide) This paper presents initial results from a project that will formally test a number of the hypotheses embedded in the theoretical and qualitative literatures on mediation, using automated coding of event data from news-wire sources. In contrast to most of the existing quantitative literature, which emphasizes the structural aspects of mediation, we will focus on the dynamics. The initial part of the paper focuses on two issues of design. First, we discuss the advantages of generating data using fully automated methods, which increases the transparency and replicability of the research. This transparency is extended to the development of more complex variables that cannot be captured as single events: these are defined as pattern of the underlying event data. We also suggest that these can be usefully studied using conventional inferential statistics rather than computational pattern recognition. Second, we justify the "statistical case study" approach which focuses on a small number of cases that are limited in geographical and temporal scope. While the risk of this approach is that one will find patterns of behavior that apply only in those circumstances, we point out that the more conventional large-N time-series cross-sectional studies also carry inferential risks. The statistical tests reported in this paper look at three different issues using data on the Israel-Lebanon and Israel-Palestinian conflicts in the Levant (1979-1999), and the Serbia-Croatia and Serbia-Bosnia conflicts in the Balkans (1991-1999). First, cross- correlation is used to look at the effects of mediation on the level of violence over time. Second, we test the "sticks-or-carrots" hypothesis on whether mediation is more effective in reducing violence if accompanied by cooperative or conflictual behavior by the mediator. Finally, we estimate Cox proportional hazard models to assess the factors that influence (1) whether mediation is accepted by the parties in a conflict, (2) whether formal agreements are reached, and (3) whether the agreements reduce the level of conflict. Future work in the project involves development of a new event coding scheme specifically designed for the study of mediation, and expansion of the list of cases to include other mediated conflicts in the Middle East and West Africa.
Is the Road to Bad Inference Paved with Good Intentions? An Audit of Vote Intention Survey Items
Arceneaux, Kevin
Submitted: 2001-07-17
Keywords: vote intention, survey response, statistical inference
Abstract: (click to show/hide) It is not always possible to conduct surveys near an election. This is especially true in Europe where the timing of elections varies widely from country to country. Not only do elections rarely occur at the same time across countries, election dates are not fixed within most European countries, making it nearly impossible to plan post-election surveys in Europe. Even in the United States, where elections are fixed events, off-election surveys are often used. Because there is no ongoing campaign, these surveys rely on vote intention questions as a measure of vote choice. The typical wording of a vote intention question is, “If there were a general election tomorrow, which party would you support?” Many researchers, especially users of Eurobarometer data, rely on these questions as a measure of vote choice. However, it is unclear if these questions are in fact valid indicators of voting behavior. Drawing on information processing theories, this paper develops a theoretical framework to identify the conditions under which vote intention measures accurately vote choice. Hypotheses generated by this framework are tested across multiple countries and years using Eurobarometer and American National Election Studies survey data.
Arceneaux, Kevin
Submitted: 2001-07-17
Keywords: vote intention, survey response, statistical inference
Abstract: (click to show/hide) It is not always possible to conduct surveys near an election. This is especially true in Europe where the timing of elections varies widely from country to country. Not only do elections rarely occur at the same time across countries, election dates are not fixed within most European countries, making it nearly impossible to plan post-election surveys in Europe. Even in the United States, where elections are fixed events, off-election surveys are often used. Because there is no ongoing campaign, these surveys rely on vote intention questions as a measure of vote choice. The typical wording of a vote intention question is, “If there were a general election tomorrow, which party would you support?” Many researchers, especially users of Eurobarometer data, rely on these questions as a measure of vote choice. However, it is unclear if these questions are in fact valid indicators of voting behavior. Drawing on information processing theories, this paper develops a theoretical framework to identify the conditions under which vote intention measures accurately vote choice. Hypotheses generated by this framework are tested across multiple countries and years using Eurobarometer and American National Election Studies survey data.
Random Coefficient Models for Time-Series--Cross-Section Data: The 2001 Version
Beck, Nathaniel, Katz, Jonathan
Submitted: 2001-07-17
Keywords: random coefficients, generalized least squares, empirical Bayesian, Stein-rule, TCSC
Abstract: (click to show/hide) This paper considers random coefficient models (RCMs) for time-series--cross-section data. These models allow for unit to unit variation in the model parameters. After laying out the various models, we assess several issues in specifying RCMs. We then consider the finite sample properties of some standard RCM estimators, and show that the most common one, associated with Hsaio, has very poor properties. These analyses also show that a somewhat awkward combination of estimators based on Swamy's work performs reasonably well; this awkward estimator and a Bayes estimator with an uninformative prior (due to Smith) seem to perform best. But we also see that estimators which assume full pooling perform well unless there is a large degree of unit to unit parameter heterogeneity. We also argue that the various data driven methods (whether classical or empirical Bayes or Bayes with gentle priors) tends to lead to much more heterogeneity than most political scientists would like. We speculate that fully Bayesian models, with a variety of informative priors, may be the best way to approach RCMs.
Beck, Nathaniel, Katz, Jonathan
Submitted: 2001-07-17
Keywords: random coefficients, generalized least squares, empirical Bayesian, Stein-rule, TCSC
Abstract: (click to show/hide) This paper considers random coefficient models (RCMs) for time-series--cross-section data. These models allow for unit to unit variation in the model parameters. After laying out the various models, we assess several issues in specifying RCMs. We then consider the finite sample properties of some standard RCM estimators, and show that the most common one, associated with Hsaio, has very poor properties. These analyses also show that a somewhat awkward combination of estimators based on Swamy's work performs reasonably well; this awkward estimator and a Bayes estimator with an uninformative prior (due to Smith) seem to perform best. But we also see that estimators which assume full pooling perform well unless there is a large degree of unit to unit parameter heterogeneity. We also argue that the various data driven methods (whether classical or empirical Bayes or Bayes with gentle priors) tends to lead to much more heterogeneity than most political scientists would like. We speculate that fully Bayesian models, with a variety of informative priors, may be the best way to approach RCMs.
Detection of Multinomial Voting Irregularities
Mebane, Walter R., Sekhon, Jasjeet, Wand, Jonathan
Submitted: 2001-07-17
Keywords: outlier detection, robust estimation, overdispersed multinomial, generalized linear model, 2000 presidential election, voting irregularities
Abstract: (click to show/hide) We develop a robust estimator for an overdispersed multinomial regression model that we use to detect vote count outliers in the 2000 presidential election. The count vector we model contains vote totals for five candidate categories: Buchanan, Bush, Gore, Nader and ``other.'' We estimate the multinomial model using county-level data from Florida. In Florida, the model produces results for Buchanan that are essentially the same as in a binomial model: Palm Beach County has the largest positive residual for Buchanan. The multinomial model shows additional large discrepancies that almost always hurt Gore or Nader and help Bush or Buchanan.
Mebane, Walter R., Sekhon, Jasjeet, Wand, Jonathan
Submitted: 2001-07-17
Keywords: outlier detection, robust estimation, overdispersed multinomial, generalized linear model, 2000 presidential election, voting irregularities
Abstract: (click to show/hide) We develop a robust estimator for an overdispersed multinomial regression model that we use to detect vote count outliers in the 2000 presidential election. The count vector we model contains vote totals for five candidate categories: Buchanan, Bush, Gore, Nader and ``other.'' We estimate the multinomial model using county-level data from Florida. In Florida, the model produces results for Buchanan that are essentially the same as in a binomial model: Palm Beach County has the largest positive residual for Buchanan. The multinomial model shows additional large discrepancies that almost always hurt Gore or Nader and help Bush or Buchanan.
Pre-Election Polls in Nation and State: A Dynamic Bayesian Hierarchical Model
Franklin, Charles
Submitted: 2001-07-17
Keywords: campaigns, polling, aggregation, Bayesian, hierarchical models
Abstract: (click to show/hide) A vast number of national trial heat polls are conducted in the months preceding a presidential election. But as was dramatically demonstrated in 2000, candidates must win states to win the presidency, not just win popular votes. The density of state level polling is much less than that for the nation as a whole. This makes efforts to track candidate support at the state level, and to estimate campaign effects in the states, very difficult. This paper develops a Bayesian hierarchical model of trial heat polls which uses state and national polling data, plus measures of campaign effort in each state, to estimate candidate support between observed state polls. At a technical level, the Bayesian approach provides not only estimates of support but also easily understood estimates of the uncertainty of those estimates. At an applied level, this method can allow campaigns to target polling in states that are most likely to be changing while being alerted to potential shifts in states that are not as frequently polled.
Franklin, Charles
Submitted: 2001-07-17
Keywords: campaigns, polling, aggregation, Bayesian, hierarchical models
Abstract: (click to show/hide) A vast number of national trial heat polls are conducted in the months preceding a presidential election. But as was dramatically demonstrated in 2000, candidates must win states to win the presidency, not just win popular votes. The density of state level polling is much less than that for the nation as a whole. This makes efforts to track candidate support at the state level, and to estimate campaign effects in the states, very difficult. This paper develops a Bayesian hierarchical model of trial heat polls which uses state and national polling data, plus measures of campaign effort in each state, to estimate candidate support between observed state polls. At a technical level, the Bayesian approach provides not only estimates of support but also easily understood estimates of the uncertainty of those estimates. At an applied level, this method can allow campaigns to target polling in states that are most likely to be changing while being alerted to potential shifts in states that are not as frequently polled.
Practical Maximum Likelihood
Altman, Micah, McDonald, Michael P.
Submitted: 2001-07-22
Keywords: maximum likelihood, optimization, statistical computation, numerical stability, accuracy
Abstract: (click to show/hide) Maximum likelihood estimation is now widely used in political science, providing a general statistical framework in which we build and test increasingly complex models of politics. The modern development of maximum likelihood is attributable to Fisher, and the approach dominated mathematical statistics during the twentieth century. \ More attention has been paid to the development of complex statistical models than to the necessary details of their estimation. In this article we discuss some of the art and practice of MLE: -Estimation: We discuss how to choose algorithms for MLE estimations, methods for setting algorithm parameters appropriately, and how to formulate likelihood functions for efficient and accurate estimation. -Tests of Estimation: Methods of statistical inference assume that a global maximum of the likelihood function has been found. There are however, few general guarantees that likelihood functions are single-peaked. Furthermore, no MLE software currently in use by political scientists verifies that global maximum of the likelihood function has been reached. We provide tests of global optimality, drawing from current research in statistics, econometrics, and computer science. -MLE Based Inference: Standard errors produced by MLE's can be misleading, and lead to unreliable inferences, when the likelihood function is not well behaved around its maximum. We illustrate the consequences of unreliable methods, and discuss more robust methods of calculating
Altman, Micah, McDonald, Michael P.
Submitted: 2001-07-22
Keywords: maximum likelihood, optimization, statistical computation, numerical stability, accuracy
Abstract: (click to show/hide) Maximum likelihood estimation is now widely used in political science, providing a general statistical framework in which we build and test increasingly complex models of politics. The modern development of maximum likelihood is attributable to Fisher, and the approach dominated mathematical statistics during the twentieth century. \ More attention has been paid to the development of complex statistical models than to the necessary details of their estimation. In this article we discuss some of the art and practice of MLE: -Estimation: We discuss how to choose algorithms for MLE estimations, methods for setting algorithm parameters appropriately, and how to formulate likelihood functions for efficient and accurate estimation. -Tests of Estimation: Methods of statistical inference assume that a global maximum of the likelihood function has been found. There are however, few general guarantees that likelihood functions are single-peaked. Furthermore, no MLE software currently in use by political scientists verifies that global maximum of the likelihood function has been reached. We provide tests of global optimality, drawing from current research in statistics, econometrics, and computer science. -MLE Based Inference: Standard errors produced by MLE's can be misleading, and lead to unreliable inferences, when the likelihood function is not well behaved around its maximum. We illustrate the consequences of unreliable methods, and discuss more robust methods of calculating
Distortion magnified: New Labour and the British
Johnston, Ron, Rossiter, David, Charles, Pattie, Dorling, Danny
Submitted: 2001-07-26
Keywords: electoral bias, Britain, gerrymander
Abstract: (click to show/hide) UK election results show not only the characteristic disproportionality associated with plurality systems but also considerable bias in the allocation of seats relative to votes for the two main political parties (Conservative and Labour). Over the period 1950-1997 (the 1950 election was the first using constituencies defined by independent Boundary Commissions) this bias both increased and shifted from favouring the Conservatives to favouring Labour. By 1997, Labour would have won 82 more seats than Conservative with equal vote shares - the largest bias recorded for the period: and then in 2001 the pro-Labour bias increase to 141. This paper explores the reasons for this shift, using a procedure developed by Brookes for measuring and decomposing bias. Labour benefited because of the geography of iots successful campaigns in 1997 and 2001.
Johnston, Ron, Rossiter, David, Charles, Pattie, Dorling, Danny
Submitted: 2001-07-26
Keywords: electoral bias, Britain, gerrymander
Abstract: (click to show/hide) UK election results show not only the characteristic disproportionality associated with plurality systems but also considerable bias in the allocation of seats relative to votes for the two main political parties (Conservative and Labour). Over the period 1950-1997 (the 1950 election was the first using constituencies defined by independent Boundary Commissions) this bias both increased and shifted from favouring the Conservatives to favouring Labour. By 1997, Labour would have won 82 more seats than Conservative with equal vote shares - the largest bias recorded for the period: and then in 2001 the pro-Labour bias increase to 141. This paper explores the reasons for this shift, using a procedure developed by Brookes for measuring and decomposing bias. Labour benefited because of the geography of iots successful campaigns in 1997 and 2001.
Ticket-Splitting and Strategic Voting in Mixed Electoral Systems
Gschwend, Thomas
Submitted: 2001-08-22
Keywords: Ticket Splitting, Strategic Voting, Mixed Electoral Systems, MNL, Multiple Imputation
Abstract: (click to show/hide) This work attempts to refocus the discussion about strategic voting from its narrow focus on single-member district systems. It provides several contribution to the literature on strategic voting, ticket-splitting and on electoral systems. My first contribution is to allow the electoral institutions to vary, thereby opening up the possibility to provide different incentives to operate at the same time for the same voter. I offer a theory that particular institutions not only determine the \emph{degree} of strategic voting, but also the \emph{kind} of strategies voters employ. In mixed electoral systems strategic voting has two facets. Strategic voters employ either a \emph{wasted-vote strategy} or a \emph{coalition insurance strategy}. My second contribution is to provide evidence that people vary in their \emph{proclivity} to vote strategically, as determined by various motivational factors as well as their capability to comprehend the strategic implications that are offered by particular electoral rules. Evidence supporting these contributions is stemming from an appropriate choice-model using individual-level data from the 1998 German National Election Study
Gschwend, Thomas
Submitted: 2001-08-22
Keywords: Ticket Splitting, Strategic Voting, Mixed Electoral Systems, MNL, Multiple Imputation
Abstract: (click to show/hide) This work attempts to refocus the discussion about strategic voting from its narrow focus on single-member district systems. It provides several contribution to the literature on strategic voting, ticket-splitting and on electoral systems. My first contribution is to allow the electoral institutions to vary, thereby opening up the possibility to provide different incentives to operate at the same time for the same voter. I offer a theory that particular institutions not only determine the \emph{degree} of strategic voting, but also the \emph{kind} of strategies voters employ. In mixed electoral systems strategic voting has two facets. Strategic voters employ either a \emph{wasted-vote strategy} or a \emph{coalition insurance strategy}. My second contribution is to provide evidence that people vary in their \emph{proclivity} to vote strategically, as determined by various motivational factors as well as their capability to comprehend the strategic implications that are offered by particular electoral rules. Evidence supporting these contributions is stemming from an appropriate choice-model using individual-level data from the 1998 German National Election Study
An Agenda for New Political Methodology: Microfoundations and ART
Achen, Christopher H.
Submitted: 2001-08-29
Keywords: logit, probit, scobit, microfoundations
Abstract: (click to show/hide) The last two decades have brought revolutionary change to the field of political methodology. Steady gains in theoretical sophistication have combined with explosive increases in computing power to produce a profusion of new estimators for applied political researchers. Attendance at the annual Summer Meeting of the Methodology Section has multiplied many times, and section membership is among the largest in APSA. All these are signs of success. Yet there are warning signs, too. This paper, written to appear in the {\em Annual Review of Political Science}, attempts to critically summarize current developments in the young field of political methodology. It focuses on recent generalizations of dichotomous dependent variable estimators like logit and probit, arguing that even our best new work stands in need of firmer connection to credible models of human behavior and more sophisticated work habits for discovering reliable empirical generalizations.
Achen, Christopher H.
Submitted: 2001-08-29
Keywords: logit, probit, scobit, microfoundations
Abstract: (click to show/hide) The last two decades have brought revolutionary change to the field of political methodology. Steady gains in theoretical sophistication have combined with explosive increases in computing power to produce a profusion of new estimators for applied political researchers. Attendance at the annual Summer Meeting of the Methodology Section has multiplied many times, and section membership is among the largest in APSA. All these are signs of success. Yet there are warning signs, too. This paper, written to appear in the {\em Annual Review of Political Science}, attempts to critically summarize current developments in the young field of political methodology. It focuses on recent generalizations of dichotomous dependent variable estimators like logit and probit, arguing that even our best new work stands in need of firmer connection to credible models of human behavior and more sophisticated work habits for discovering reliable empirical generalizations.
Stochastic Dependence in Competing Risks
Gordon, Sanford C.
Submitted: 2001-09-05
Keywords: Competing risks, duration models, survival models, event history, random effects, frailty models, unobserved heterogeneity, Monte Carlo simulation, Congress, legislative position-taking, cabinet survival, numeric integration, Markov Chain Monte Carlo
Abstract: (click to show/hide) The term "Competing Risks" describes duration models in which spells may terminate via multiple outcomes: The term of a cabinet, for example, may end with or without an election; wars persist until the loss or victory of the aggressor. Analysts typically assume stochastic independence among risks, the duration modeling equivalent of independence of irrelevant alternatives. However, many political examples violate this assumption. I review competing risks as a latent variables approach. After discussing methods for modeling dependence that place restrictions on the nature of association, I introduce a parametric generalized dependent risks model in which inter-risk correlation may be estimated and its significance tested. The method employs risk-specific random effects drawn from a multivariate normal distribution. Estimation is conducted using numerical methods and/or Bayesian simulation. Monte Carlo simulation reveals desirable large sample properties of the estimator. Finally, I examine two applications using data on cabinet survival and legislative position taking.
Gordon, Sanford C.
Submitted: 2001-09-05
Keywords: Competing risks, duration models, survival models, event history, random effects, frailty models, unobserved heterogeneity, Monte Carlo simulation, Congress, legislative position-taking, cabinet survival, numeric integration, Markov Chain Monte Carlo
Abstract: (click to show/hide) The term "Competing Risks" describes duration models in which spells may terminate via multiple outcomes: The term of a cabinet, for example, may end with or without an election; wars persist until the loss or victory of the aggressor. Analysts typically assume stochastic independence among risks, the duration modeling equivalent of independence of irrelevant alternatives. However, many political examples violate this assumption. I review competing risks as a latent variables approach. After discussing methods for modeling dependence that place restrictions on the nature of association, I introduce a parametric generalized dependent risks model in which inter-risk correlation may be estimated and its significance tested. The method employs risk-specific random effects drawn from a multivariate normal distribution. Estimation is conducted using numerical methods and/or Bayesian simulation. Monte Carlo simulation reveals desirable large sample properties of the estimator. Finally, I examine two applications using data on cabinet survival and legislative position taking.
Aggregation and Dynamics of Survey Responses: The Case of Presidential Approval
Alvarez, R. Michael, Katz, Jonathan
Submitted: 2001-10-01
Keywords: presidential approval, integration, time-series, fractional integration, surveys
Abstract: (click to show/hide) In this paper we critique much of the empirical literature on the important political science concept of presidential approval. We first argue that dynamics attributed to the aggregate presidential approval series are often logically inconsistent and always substantively implausible. In particular, we show that is no way for a bounded series, such as the approval series, to be integrated. However, even in non-integrated models often lead to implausible substantive findings due to aggregation both across Presidential administrations and from models of individual level behavior to aggregate survey marginals. We argue that using individual-level survey responses is superior for methodological and theoretical reasons, and we provide an example of such an analysis using Gallup Organization survey data.
Alvarez, R. Michael, Katz, Jonathan
Submitted: 2001-10-01
Keywords: presidential approval, integration, time-series, fractional integration, surveys
Abstract: (click to show/hide) In this paper we critique much of the empirical literature on the important political science concept of presidential approval. We first argue that dynamics attributed to the aggregate presidential approval series are often logically inconsistent and always substantively implausible. In particular, we show that is no way for a bounded series, such as the approval series, to be integrated. However, even in non-integrated models often lead to implausible substantive findings due to aggregation both across Presidential administrations and from models of individual level behavior to aggregate survey marginals. We argue that using individual-level survey responses is superior for methodological and theoretical reasons, and we provide an example of such an analysis using Gallup Organization survey data.
Individual Choice and Ecological Analysis
McCue, Kenneth F.
Submitted: 2001-12-02
Keywords: ecological regression, voter transitions, multivariate multinomial, split-ticket voting, aggregation bias, liner probability model
Abstract: (click to show/hide) The use of the linear probability model in aggregate voting analysis has now received widespread attention in political science. This article shows that when the linear probability model is assumed to be consistent for the choice of the individual, it is actually a member of a general class of models for estimating individual responses from aggregate data. This class has the useful property that it defines the aggregate analysis problem as a function of the individual choice decisions, and allows the placement of most aggregate voting models into a common probabilistic framework. This framework allows the solution of such problems as inference of individual responses from aggregate data, estimation of the transition model, and the joint estimation and inference from individual and aggregate data. Examples with actual data are provided for these techniques with excellent results.
McCue, Kenneth F.
Submitted: 2001-12-02
Keywords: ecological regression, voter transitions, multivariate multinomial, split-ticket voting, aggregation bias, liner probability model
Abstract: (click to show/hide) The use of the linear probability model in aggregate voting analysis has now received widespread attention in political science. This article shows that when the linear probability model is assumed to be consistent for the choice of the individual, it is actually a member of a general class of models for estimating individual responses from aggregate data. This class has the useful property that it defines the aggregate analysis problem as a function of the individual choice decisions, and allows the placement of most aggregate voting models into a common probabilistic framework. This framework allows the solution of such problems as inference of individual responses from aggregate data, estimation of the transition model, and the joint estimation and inference from individual and aggregate data. Examples with actual data are provided for these techniques with excellent results.
Heterogeneity in Discrete Choice Models
Glasgow, Garrett
Submitted: 2001-12-12
Keywords: heterogeneity, discrete choice, logit, probit, ambivalence
Abstract: (click to show/hide) Nearly all empirical studies of individual behavior in political science have sought to estimate the mean relationship between some variables of interest. While such studies are vital for determining aggregate relationships between variables of interest, they are an incomplete picture of individual behavior. In particular, we generally do not pay attention to the possibility of heterogeneity, or individual-level variation in the relationships we estimate. Ignoring heterogeneity in our models means we are ignoring valuable information about individual behavior. This paper demonstrates that examining heterogeneity in discrete choice models is both important substantively and feasible methodologically. Possible sources of heterogeneity are discussed, and it is shown that these sources of heterogeneity are observationally equivalent in most cases, meaning it is generally not possible to determine the source of heterogeneity in our empirical models. Several empirical models for examining heterogeneity are described. An empirical example studying heterogeneity in union voting in the 1992 US presidential election demonstrates the
Glasgow, Garrett
Submitted: 2001-12-12
Keywords: heterogeneity, discrete choice, logit, probit, ambivalence
Abstract: (click to show/hide) Nearly all empirical studies of individual behavior in political science have sought to estimate the mean relationship between some variables of interest. While such studies are vital for determining aggregate relationships between variables of interest, they are an incomplete picture of individual behavior. In particular, we generally do not pay attention to the possibility of heterogeneity, or individual-level variation in the relationships we estimate. Ignoring heterogeneity in our models means we are ignoring valuable information about individual behavior. This paper demonstrates that examining heterogeneity in discrete choice models is both important substantively and feasible methodologically. Possible sources of heterogeneity are discussed, and it is shown that these sources of heterogeneity are observationally equivalent in most cases, meaning it is generally not possible to determine the source of heterogeneity in our empirical models. Several empirical models for examining heterogeneity are described. An empirical example studying heterogeneity in union voting in the 1992 US presidential election demonstrates the
