Working Papers
1998
63 records found
Information Asymmetries and Simultaneous versus Sequential Voting
Morton, Becky, Williams, Kenneth C.
Submitted: 1998-01-19
Keywords: sequential voting, simultaneous voting, information aggregation, Presidential primaries, uniform election days
Abstract: (click to show/hide) We theoretically and empirically compare sequential with simultaneous voting elections and the impact of the representativeness of early voters in sequential voting on the electoral outcome when voters have asymmetric information about the candidates. We use a simple three-candidate model where one candidate is a Condorcet winner, i.e. would defeat either opponent in a pairwise competition. However, under complete information multiple equilibria exist in which any of the three candidates could win election. Theoretically, in simultaneous voting elections with voters asymmetrically informed about the candidates, the candidate better known is more likely to win, regardless of whether this candidate is the Condorcet winner or not. In sequential voting, early voters should vote "informatively" and multiple equilibria exist. Using laboratory elections, we investigate our theoretical predictions and consider which of the equilibrium outcomes are more likely. Better known candidates are more likely to win in simultaneous voting, regardless of candidate type. Early voters in sequential voting elections vote informatively and, when given detail on voting by early voters, later voters appear to infer information about the candidates from early voting. The Condorcet winner is more likely to win in sequential voting elections than in simultaneous voting elections when that candidate is less well known. If early voters are not representative of the voting population, there is evidence that their most preferred candidate is more likely to win if they are able to identify their first preference. However, non-representativeness of early voters increases the likelihood that the Condorcet winner will win in sequential voting. For information contact: Rebecca-Morton@uiowa.edu
Morton, Becky, Williams, Kenneth C.
Submitted: 1998-01-19
Keywords: sequential voting, simultaneous voting, information aggregation, Presidential primaries, uniform election days
Abstract: (click to show/hide) We theoretically and empirically compare sequential with simultaneous voting elections and the impact of the representativeness of early voters in sequential voting on the electoral outcome when voters have asymmetric information about the candidates. We use a simple three-candidate model where one candidate is a Condorcet winner, i.e. would defeat either opponent in a pairwise competition. However, under complete information multiple equilibria exist in which any of the three candidates could win election. Theoretically, in simultaneous voting elections with voters asymmetrically informed about the candidates, the candidate better known is more likely to win, regardless of whether this candidate is the Condorcet winner or not. In sequential voting, early voters should vote "informatively" and multiple equilibria exist. Using laboratory elections, we investigate our theoretical predictions and consider which of the equilibrium outcomes are more likely. Better known candidates are more likely to win in simultaneous voting, regardless of candidate type. Early voters in sequential voting elections vote informatively and, when given detail on voting by early voters, later voters appear to infer information about the candidates from early voting. The Condorcet winner is more likely to win in sequential voting elections than in simultaneous voting elections when that candidate is less well known. If early voters are not representative of the voting population, there is evidence that their most preferred candidate is more likely to win if they are able to identify their first preference. However, non-representativeness of early voters increases the likelihood that the Condorcet winner will win in sequential voting. For information contact: Rebecca-Morton@uiowa.edu
A Comparison of Methods for the Analysis of Time Allocation: Donut Shops, Speed Traps, and Paperwork
Brehm, John, Gates, Scott, Gomez, Brad
Submitted: 1998-03-05
Keywords: additive logistic, allocation of time, compliance, Dirichlet, Liouville
Abstract: (click to show/hide) Supervisors in public bureaucracies serve a variety of roles, only few of which have been subject to systematic academic scrutiny. In prior work (Brehm and Gates 1997), we dispense with the coercive capacities of supervisors, but this leaves other supervisory functions as potential levers on subordinate compliance. In this paper, we consider the capacity of supervisors as coordinators of subordinate tasks. We first develop propositions from our previously published formal models, and then test these propositions using data from the 1977 Police Services Study (Ostrom, Parks, and Whittaker 1988). We explore the allocation of time by bureaucratic subordinates using a method for compositional data analysis, obtaining maximum likelihood estimates of Dirichlet distributions. We evaluate the propositions using computer simulations of the predicted distributions and conclude that, although supervisors have some capacities to influence subordinate allocation of time across tasks, the major influences stem from inter-subordinate contact and subordinate preferences.
Brehm, John, Gates, Scott, Gomez, Brad
Submitted: 1998-03-05
Keywords: additive logistic, allocation of time, compliance, Dirichlet, Liouville
Abstract: (click to show/hide) Supervisors in public bureaucracies serve a variety of roles, only few of which have been subject to systematic academic scrutiny. In prior work (Brehm and Gates 1997), we dispense with the coercive capacities of supervisors, but this leaves other supervisory functions as potential levers on subordinate compliance. In this paper, we consider the capacity of supervisors as coordinators of subordinate tasks. We first develop propositions from our previously published formal models, and then test these propositions using data from the 1977 Police Services Study (Ostrom, Parks, and Whittaker 1988). We explore the allocation of time by bureaucratic subordinates using a method for compositional data analysis, obtaining maximum likelihood estimates of Dirichlet distributions. We evaluate the propositions using computer simulations of the predicted distributions and conclude that, although supervisors have some capacities to influence subordinate allocation of time across tasks, the major influences stem from inter-subordinate contact and subordinate preferences.
Comparing Measures of Issue Salience in a Spatial Voting Model
Glasgow, Garrett
Submitted: 1998-03-12
Keywords: none submitted
Abstract: (click to show/hide) Few fields of empirical study of voting behavior have generated as many mixed results as the study of issue salience. Voters employing differential weighting of issues when determining candidate preference has long been postulated in theoretical works on the spatial voting model (Davis, Hinich, and Ordeshook 1970, Enelow and Hinich 1984), yet empirical research has failed to consistently uncover salience effects among voters. Much of this inconsistency is due to the use of different measures of issue salience and different models to test for salience effects in each study. What accounts for these different findings? There appear to be two different factors that affect each of these studies to different degrees. The first is the limitations of respondents in reporting their cognitive processes. Psychological research demonstrates that individuals are often unaware or unable to accurately report their decision making rules, a limitation that has severe implications for any attempt to measure the weight individuals assign to particuluar issues when determining candidate preference. The second is the cost of gathering information about candidates that can be employed in the choice rule for candidate preference. An individual who places a great deal of weight on a particular issue, but has no information on candidate positions on that issue, will be forced to rely on less salient issues to determine candidate preference. The cognitive and informational limitations of respondents and the implications that they have for measurement of issue salience are examined and accounted for in a spatial voting model of candidate preference. The tendency for individuals for focus on one or a few issues when determining candidate preference is confirmed, but the hypothesised informational effects fail to emerge.
Glasgow, Garrett
Submitted: 1998-03-12
Keywords: none submitted
Abstract: (click to show/hide) Few fields of empirical study of voting behavior have generated as many mixed results as the study of issue salience. Voters employing differential weighting of issues when determining candidate preference has long been postulated in theoretical works on the spatial voting model (Davis, Hinich, and Ordeshook 1970, Enelow and Hinich 1984), yet empirical research has failed to consistently uncover salience effects among voters. Much of this inconsistency is due to the use of different measures of issue salience and different models to test for salience effects in each study. What accounts for these different findings? There appear to be two different factors that affect each of these studies to different degrees. The first is the limitations of respondents in reporting their cognitive processes. Psychological research demonstrates that individuals are often unaware or unable to accurately report their decision making rules, a limitation that has severe implications for any attempt to measure the weight individuals assign to particuluar issues when determining candidate preference. The second is the cost of gathering information about candidates that can be employed in the choice rule for candidate preference. An individual who places a great deal of weight on a particular issue, but has no information on candidate positions on that issue, will be forced to rely on less salient issues to determine candidate preference. The cognitive and informational limitations of respondents and the implications that they have for measurement of issue salience are examined and accounted for in a spatial voting model of candidate preference. The tendency for individuals for focus on one or a few issues when determining candidate preference is confirmed, but the hypothesised informational effects fail to emerge.
Strategic Allocation of Party Money
Glasgow, Garrett
Submitted: 1998-03-17
Keywords: none submitted
Abstract: (click to show/hide) This paper models and empirically examines the strategic behavior of American political parties in House elections. In each election cycle, political parties must decide which candidates competing for seats will receive aid from party coffers. Parties pursue two goals when distributing funds; they wish to maximize their representation in the House by winning seats, and they wish to fill those seats with candidates who are likely to vote in a way that furthers the party's goals. I assume that in each district both the Republicans and the Democrats pursue these goals by sending a signal about the strength of their support for their respective candidates. This signal takes the form of monetary contributions to a candidate's campaign -- the greater the contribution, the stronger the signal of support for the candidate. These signals attract other contributions to the candidate receiving party support, as other investors in candidates regard the party's signal as credible evidence that the candidate is of high quality. These added contributions increase the likelihood of the candidate winning the seat. Private investors wish to contribute to candidates who already have a good chance of winning the election, as contributions to winning candidates give them access to a legislator, while contributions to a losing candidate are lost. Parties know that they can improve the chances of victory for a candidate by offering aid and thus sending a signal of support, but incentives to win a maximal number of seats and reward party loyalists are balanced with the desire to maintain a credible signal. As party resources are not infinite, each party would like to distribute its limited resources in such a way as to maximize the chances of achieving its goals. A game-theoretic model that captures the decision problem faced by each party is developed and empirically tested. I conclude that parties do not always distribute funds in a rational manner (in particular, too many resources are allocated to incumbents in secure districts), but as electoral outcomes in a particular district become uncertain, parties begin to adhere to the predictions of the game-theoretic model much more closely.
Glasgow, Garrett
Submitted: 1998-03-17
Keywords: none submitted
Abstract: (click to show/hide) This paper models and empirically examines the strategic behavior of American political parties in House elections. In each election cycle, political parties must decide which candidates competing for seats will receive aid from party coffers. Parties pursue two goals when distributing funds; they wish to maximize their representation in the House by winning seats, and they wish to fill those seats with candidates who are likely to vote in a way that furthers the party's goals. I assume that in each district both the Republicans and the Democrats pursue these goals by sending a signal about the strength of their support for their respective candidates. This signal takes the form of monetary contributions to a candidate's campaign -- the greater the contribution, the stronger the signal of support for the candidate. These signals attract other contributions to the candidate receiving party support, as other investors in candidates regard the party's signal as credible evidence that the candidate is of high quality. These added contributions increase the likelihood of the candidate winning the seat. Private investors wish to contribute to candidates who already have a good chance of winning the election, as contributions to winning candidates give them access to a legislator, while contributions to a losing candidate are lost. Parties know that they can improve the chances of victory for a candidate by offering aid and thus sending a signal of support, but incentives to win a maximal number of seats and reward party loyalists are balanced with the desire to maintain a credible signal. As party resources are not infinite, each party would like to distribute its limited resources in such a way as to maximize the chances of achieving its goals. A game-theoretic model that captures the decision problem faced by each party is developed and empirically tested. I conclude that parties do not always distribute funds in a rational manner (in particular, too many resources are allocated to incumbents in secure districts), but as electoral outcomes in a particular district become uncertain, parties begin to adhere to the predictions of the game-theoretic model much more closely.
Models of Monetary Policy Decision-Making: Arthur Burns and the Federal Open Market Committee
Chappell, Jr., Henry W., McGregor, Rob Roy, Vermilyea, Todd
Submitted: 1998-04-01
Keywords: monetary policy, Federal Reserve, median voter
Abstract: (click to show/hide) This paper investigates decision-making within the Federal Open Market Committee of the Federal Reserve, focusing on the competing pressures of majority rule, consensus-building, and the power of the Chairman. To undertake this analysis, we have constructed a data set recording desired Federal funds rates for each member of the Committee over the 1970-1978 period. We empirically link individuals' policy preferences to adopted policies using generalized versions of the median voter model and alternative specifications. Our results confirm a persistent attraction of the median voter's ideal point; they also confirm a disproportionate influence of the Chairman in the policy process. The voting weight of the Chairman is estimated to be between 0.38 and 0.58 in preferred specifications. Results also suggest that district Federal Reserve Bank presidents have somewhat greater influence over adopted policies than Governors.
Chappell, Jr., Henry W., McGregor, Rob Roy, Vermilyea, Todd
Submitted: 1998-04-01
Keywords: monetary policy, Federal Reserve, median voter
Abstract: (click to show/hide) This paper investigates decision-making within the Federal Open Market Committee of the Federal Reserve, focusing on the competing pressures of majority rule, consensus-building, and the power of the Chairman. To undertake this analysis, we have constructed a data set recording desired Federal funds rates for each member of the Committee over the 1970-1978 period. We empirically link individuals' policy preferences to adopted policies using generalized versions of the median voter model and alternative specifications. Our results confirm a persistent attraction of the median voter's ideal point; they also confirm a disproportionate influence of the Chairman in the policy process. The voting weight of the Chairman is estimated to be between 0.38 and 0.58 in preferred specifications. Results also suggest that district Federal Reserve Bank presidents have somewhat greater influence over adopted policies than Governors.
GEE Models of Judicial Behavior
Zorn, Christopher
Submitted: 1998-04-02
Keywords: generalized estimating equations, time-series cross-sectional data, temporal dependence, heterogeneity, judicial decision making
Abstract: (click to show/hide) The assumption of independent observations in judicial decision making flies in the face of our theoretical understanding of the topic. In particular, two characteristics of judicial decision making on collegial courts introduce heterogeneity into successive decisions: individual variation in the extent to which different jurists maintain consistency in their voting behavior over time, and the ability of one judge or justice to influence another in their decisions. This paper addresses these issues by framing judicial behavior in a time-series cross-section context and using the recently developed technique of generalized estimating equations (GEE) to estimate models of that behavior. Because the GEE approach allows for flexible estimation of the conditional correlation matrix within cross-sectional observations, it permits the researcher to explicitly model interjustice influence or over-time dependence in judicial decisions. I utilize this approach to examine two issues in judicial decision making: latent interjustice influence in civil rights and liberties cases during the Burger Court, and temporal consistency in Supreme Court voting in habeas corpus decisions in the postwar era. GEE estimators are shown to be comparable to more conventional pooled and TSCS techniques in estimating variable effects, but have the additional benefit of providing empirical estimates of time- and panel- based heterogeneity in judicial behavior.
Zorn, Christopher
Submitted: 1998-04-02
Keywords: generalized estimating equations, time-series cross-sectional data, temporal dependence, heterogeneity, judicial decision making
Abstract: (click to show/hide) The assumption of independent observations in judicial decision making flies in the face of our theoretical understanding of the topic. In particular, two characteristics of judicial decision making on collegial courts introduce heterogeneity into successive decisions: individual variation in the extent to which different jurists maintain consistency in their voting behavior over time, and the ability of one judge or justice to influence another in their decisions. This paper addresses these issues by framing judicial behavior in a time-series cross-section context and using the recently developed technique of generalized estimating equations (GEE) to estimate models of that behavior. Because the GEE approach allows for flexible estimation of the conditional correlation matrix within cross-sectional observations, it permits the researcher to explicitly model interjustice influence or over-time dependence in judicial decisions. I utilize this approach to examine two issues in judicial decision making: latent interjustice influence in civil rights and liberties cases during the Burger Court, and temporal consistency in Supreme Court voting in habeas corpus decisions in the postwar era. GEE estimators are shown to be comparable to more conventional pooled and TSCS techniques in estimating variable effects, but have the additional benefit of providing empirical estimates of time- and panel- based heterogeneity in judicial behavior.
Voting, Abstention, and Individual Expectations in the 1992 Presidential Election
Herron, Michael C.
Submitted: 1998-04-07
Keywords: voting, abstention, selection bias, 1992 election
Abstract: (click to show/hide) This paper develops and applies to the 1992 presidential election a statistical model of voting and abstention in three--candidate elections. The model allows us to estimate key preference--related covariates in 1992, the extent to which abstention rates were correlated with political preferences, and the impact on abstention rates of expectations regarding the election winner. Throughout this paper, we contrast our results with those in Alvarez and Nagler (1995), a study of the 1992 election that does not incorporate abstention, and in so doing we illustrate the selection bias risked by presidential election voting research that ignores abstention. Our results highlight the importance of retrospective voting in 1992, and we identify numerous policy issues, for example, the death penalty, environmental spending, and social security, that individuals used to distinguish the three candidates in the 1992 election. Abortion, we find, played only a minor role in candidate choice. We find support for the angry voting hypothesis, namely, that angry individuals often supported the independent candidate, Ross Perot. Concerning abstention, we find that supporters of the Democratic challenger Bill Clinton abstained at higher rates than supporters of Perot and the incumbent president George Bush. And, we find that expectations concerning the likelihood that Clinton was going to be victorious in 1992 influenced abstention rates. Namely, Clinton supporters who believed that Clinton was likely to win voted at higher rates than individuals who believed otherwise. The opposite relation holds for Bush supporters: such individuals, when they predicted a Clinton victory, frequently abstained from voting. The results in this paper suggests that empirical voting studies should explicitly model the impact of expectations on voting and abstention and, more generally, should model abstention as a viable, individual--level
Herron, Michael C.
Submitted: 1998-04-07
Keywords: voting, abstention, selection bias, 1992 election
Abstract: (click to show/hide) This paper develops and applies to the 1992 presidential election a statistical model of voting and abstention in three--candidate elections. The model allows us to estimate key preference--related covariates in 1992, the extent to which abstention rates were correlated with political preferences, and the impact on abstention rates of expectations regarding the election winner. Throughout this paper, we contrast our results with those in Alvarez and Nagler (1995), a study of the 1992 election that does not incorporate abstention, and in so doing we illustrate the selection bias risked by presidential election voting research that ignores abstention. Our results highlight the importance of retrospective voting in 1992, and we identify numerous policy issues, for example, the death penalty, environmental spending, and social security, that individuals used to distinguish the three candidates in the 1992 election. Abortion, we find, played only a minor role in candidate choice. We find support for the angry voting hypothesis, namely, that angry individuals often supported the independent candidate, Ross Perot. Concerning abstention, we find that supporters of the Democratic challenger Bill Clinton abstained at higher rates than supporters of Perot and the incumbent president George Bush. And, we find that expectations concerning the likelihood that Clinton was going to be victorious in 1992 influenced abstention rates. Namely, Clinton supporters who believed that Clinton was likely to win voted at higher rates than individuals who believed otherwise. The opposite relation holds for Bush supporters: such individuals, when they predicted a Clinton victory, frequently abstained from voting. The results in this paper suggests that empirical voting studies should explicitly model the impact of expectations on voting and abstention and, more generally, should model abstention as a viable, individual--level
The Two Faces of Public Opinion
Berinsky, Adam
Submitted: 1998-04-13
Keywords: public opinion, selection bias, item non-response, social desirability, bivariate probit
Abstract: (click to show/hide) In this paper I trace out the aggregate effects of the social forces in the survey interview that might influence the opinions which individuals express. First, I advance the "Mediated Communication" theory of the survey response, which builds on existing models of public opinion in the political science literature by accounting for effects related to the social context of the survey setting. I then discuss how the aggregation process could compound these individual-level effects to create an opinion signal which is a poor representation of the collective public's policy preferences. As an illustration of these effects and the resulting difficulties involved in measuring aggregate opinion on socially sensitive issues, I use National Elections Study (NES) data from 1990-1994 to show that public opinion polls overstate support for school integration. Specifically, individuals who harbor anti-integration sentiments are likely to hide their socially unacceptable opinions behind a mask of indifference. Finally, in order to confirm the validity of these findings, I show that the same methods which predict that opinion polls understate true opposition to government involvement in school integration also predict the results of the 1989 New York City mayoral election -- an election where the charged racial atmosphere made accurate polling difficult, if not impossible -- more accurately than the marginals of the pre-election polls taken in the weeks leading to the election. All told, these results suggests that survey questions on school integration -- and more generally questions on racial attitudes -- may provide an inaccurate picture of true public sentiment on such sensitive issues.
Berinsky, Adam
Submitted: 1998-04-13
Keywords: public opinion, selection bias, item non-response, social desirability, bivariate probit
Abstract: (click to show/hide) In this paper I trace out the aggregate effects of the social forces in the survey interview that might influence the opinions which individuals express. First, I advance the "Mediated Communication" theory of the survey response, which builds on existing models of public opinion in the political science literature by accounting for effects related to the social context of the survey setting. I then discuss how the aggregation process could compound these individual-level effects to create an opinion signal which is a poor representation of the collective public's policy preferences. As an illustration of these effects and the resulting difficulties involved in measuring aggregate opinion on socially sensitive issues, I use National Elections Study (NES) data from 1990-1994 to show that public opinion polls overstate support for school integration. Specifically, individuals who harbor anti-integration sentiments are likely to hide their socially unacceptable opinions behind a mask of indifference. Finally, in order to confirm the validity of these findings, I show that the same methods which predict that opinion polls understate true opposition to government involvement in school integration also predict the results of the 1989 New York City mayoral election -- an election where the charged racial atmosphere made accurate polling difficult, if not impossible -- more accurately than the marginals of the pre-election polls taken in the weeks leading to the election. All told, these results suggests that survey questions on school integration -- and more generally questions on racial attitudes -- may provide an inaccurate picture of true public sentiment on such sensitive issues.
The Revolution Against Affirmative Action in California: Politics, Economics, and Proposition 209
Alvarez, R. Michael, Butterfield, Tara L.
Submitted: 1998-04-14
Keywords: discrete choice, endogeneity, generalized extreme value, affirmative action, race and politics, California politics
Abstract: (click to show/hide) We consider two possible explanations --- economic anxiety and racial division --- for the appeal of Proposition 209 to California voters during the 1996 election. To test these hypotheses, we analyze voter exit poll data from teh 1996 California election. We utiliize a two--stage logit model to allow for the endogeneity of candidate endorsements. We find support for the second of our two hypotheses, which leads us to conclude that racial division fueled by a fear of arbitrary exclusion prompted voter support for Proposition 209.
Alvarez, R. Michael, Butterfield, Tara L.
Submitted: 1998-04-14
Keywords: discrete choice, endogeneity, generalized extreme value, affirmative action, race and politics, California politics
Abstract: (click to show/hide) We consider two possible explanations --- economic anxiety and racial division --- for the appeal of Proposition 209 to California voters during the 1996 election. To test these hypotheses, we analyze voter exit poll data from teh 1996 California election. We utiliize a two--stage logit model to allow for the endogeneity of candidate endorsements. We find support for the second of our two hypotheses, which leads us to conclude that racial division fueled by a fear of arbitrary exclusion prompted voter support for Proposition 209.
Operationalizing and Testing Spatial Theories of Voting
Quinn, Kevin M., Martin, Andrew D.
Submitted: 1998-04-15
Keywords: spatial voting, factor analysis, multinomial probit, multinomial logit, Bayesian inference, model comparison, Bayes factors, MCMC, Dutch politics, Danish politics
Abstract: (click to show/hide) Spatial models of voting behavior provide the foundation for a substantial number of theoretical results. Nonetheless, empirical work involving the spatial model faces a number of potential difficulties. First, measures of the latent voter and candidate issue positions must be obtained. Second, evaluating the fit of competing statistical models of voter choice is often more complicated than previously realized. In this paper, we discuss precisely these issues. We argue that confirmatory factor analysis applied to mass-level issue preference questions is an attractive means of measuring voter ideal points. We also show how party issue positions can be recovered using a variation of this strategy. We go on to discuss the problems of assessing the fit of competing statistical models (multinomial logit vs. multinomial probit) and competing explanations (those based on spatial theory vs. those derived from other theories of voting such as sociological theories). We demonstrate how the Bayesian perspective not only provides computational advantages in the case of fitting the multinomial probit model, but also how it facilitates both types of comparison mentioned above. Results from the Netherlands and Denmark suggest that even when the computational cost of multinomial probit is disregarded, the decision whether to use multinomial probit (MNP) or multinomial logit (MNL) is not clear-cut.
Quinn, Kevin M., Martin, Andrew D.
Submitted: 1998-04-15
Keywords: spatial voting, factor analysis, multinomial probit, multinomial logit, Bayesian inference, model comparison, Bayes factors, MCMC, Dutch politics, Danish politics
Abstract: (click to show/hide) Spatial models of voting behavior provide the foundation for a substantial number of theoretical results. Nonetheless, empirical work involving the spatial model faces a number of potential difficulties. First, measures of the latent voter and candidate issue positions must be obtained. Second, evaluating the fit of competing statistical models of voter choice is often more complicated than previously realized. In this paper, we discuss precisely these issues. We argue that confirmatory factor analysis applied to mass-level issue preference questions is an attractive means of measuring voter ideal points. We also show how party issue positions can be recovered using a variation of this strategy. We go on to discuss the problems of assessing the fit of competing statistical models (multinomial logit vs. multinomial probit) and competing explanations (those based on spatial theory vs. those derived from other theories of voting such as sociological theories). We demonstrate how the Bayesian perspective not only provides computational advantages in the case of fitting the multinomial probit model, but also how it facilitates both types of comparison mentioned above. Results from the Netherlands and Denmark suggest that even when the computational cost of multinomial probit is disregarded, the decision whether to use multinomial probit (MNP) or multinomial logit (MNL) is not clear-cut.
Uncertainty and Candidate Personality Traits
Alvarez, R. Michael, Glasgow, Garrett
Submitted: 1998-04-16
Keywords: uncertainty, direct measures of uncertainty, survey response, ordered probit, candidate evaluation, candidate traits, presidential elections
Abstract: (click to show/hide) Recently, some scholars have focused attention on the role of uncertainty in elections (Alvarez 1997, Bartels 1986, Franklin 1991). They reveal that there is a great deal of uncertainty about the issue positions of candidates, and thus the costs of issue voting are burdensome for the average citizen. Further, this uncertainty affects how voters evaluate candidates in two ways. First, voters are less likely to evaluate a candidate in terms of an issue when they are uncertain about the candidate's position on that issue. Second, uncertainty about candidate issue positions has a negative impact on voter evaluations of a candidate. However, it is important to realize that for most individuals, information about the personality traits of candidates comes from the same sources as information about the issue positions of the candidates, generally media outlets. This means that information about the the personalities of candidates is passed through the same noisy channels as information about their issue positions, and is thus subject to the same types of distortions and biases that contribute to the cost of issue information. Although it is likely easier to interpret than issue information, trait information is still subject to uncertainty. In this paper we introduce direct survey measures of candidate personality trait uncertainty. Using survey data drawn from the 1995 and 1996 National Election Studies, we first establish that the direct measure of uncertainty used in this paper is a valid measure. We then examine the effect of trait opinions on candidate evaluations and test the effects that uncertainty about those opinions has on the use of traits in candidate evaluation.
Alvarez, R. Michael, Glasgow, Garrett
Submitted: 1998-04-16
Keywords: uncertainty, direct measures of uncertainty, survey response, ordered probit, candidate evaluation, candidate traits, presidential elections
Abstract: (click to show/hide) Recently, some scholars have focused attention on the role of uncertainty in elections (Alvarez 1997, Bartels 1986, Franklin 1991). They reveal that there is a great deal of uncertainty about the issue positions of candidates, and thus the costs of issue voting are burdensome for the average citizen. Further, this uncertainty affects how voters evaluate candidates in two ways. First, voters are less likely to evaluate a candidate in terms of an issue when they are uncertain about the candidate's position on that issue. Second, uncertainty about candidate issue positions has a negative impact on voter evaluations of a candidate. However, it is important to realize that for most individuals, information about the personality traits of candidates comes from the same sources as information about the issue positions of the candidates, generally media outlets. This means that information about the the personalities of candidates is passed through the same noisy channels as information about their issue positions, and is thus subject to the same types of distortions and biases that contribute to the cost of issue information. Although it is likely easier to interpret than issue information, trait information is still subject to uncertainty. In this paper we introduce direct survey measures of candidate personality trait uncertainty. Using survey data drawn from the 1995 and 1996 National Election Studies, we first establish that the direct measure of uncertainty used in this paper is a valid measure. We then examine the effect of trait opinions on candidate evaluations and test the effects that uncertainty about those opinions has on the use of traits in candidate evaluation.
Who Votes By Mail? A Dynamic Model of the Individual-Level Consequences of Vote-by-Mail Systems
Berinsky, Adam, Burns, Nancy, Traugott, Michael
Submitted: 1998-04-17
Keywords: turnout, vote-by-mail, duration analysis, continuous-time multistate duration model
Abstract: (click to show/hide) Throughout the years, a number of changes have been proposed to electoral laws with the aim of increasing voter turnout and altering the composition of the electorate to make it more reflective of the voting age population. The most recent of these innovations is voting-by-mail (VBM). While the use of VBM has spread through the United States, little empirical evaluation of the impact of VBM has been undertaken to date. The analysis presented here fills this gap in our knowledge by assessing the impact of VBM on the Oregon electorate through a multistate duration analysis (Heckman and Singer, 1984; Heckman and Walker, 1986, 1991) that takes into account other factors associated with election administration and characteristics of individual voters. This methodology has the added advantage of providing a reasonable basis for extrapolation of these effects to other jurisdictions. The results of our research suggest that VBM does increase voter turnout in the aggregate, although its effects are not uniform across all groups in the electorate. More importantly, it does not seem to exert any influence on the partisan composition of the electorate. From a methodological perspective, the use of a multistate duration analysis provides a promising approach to extrapolating the impact of a policy change from one jurisdiction to another when appropriate data are available in each.
Berinsky, Adam, Burns, Nancy, Traugott, Michael
Submitted: 1998-04-17
Keywords: turnout, vote-by-mail, duration analysis, continuous-time multistate duration model
Abstract: (click to show/hide) Throughout the years, a number of changes have been proposed to electoral laws with the aim of increasing voter turnout and altering the composition of the electorate to make it more reflective of the voting age population. The most recent of these innovations is voting-by-mail (VBM). While the use of VBM has spread through the United States, little empirical evaluation of the impact of VBM has been undertaken to date. The analysis presented here fills this gap in our knowledge by assessing the impact of VBM on the Oregon electorate through a multistate duration analysis (Heckman and Singer, 1984; Heckman and Walker, 1986, 1991) that takes into account other factors associated with election administration and characteristics of individual voters. This methodology has the added advantage of providing a reasonable basis for extrapolation of these effects to other jurisdictions. The results of our research suggest that VBM does increase voter turnout in the aggregate, although its effects are not uniform across all groups in the electorate. More importantly, it does not seem to exert any influence on the partisan composition of the electorate. From a methodological perspective, the use of a multistate duration analysis provides a promising approach to extrapolating the impact of a policy change from one jurisdiction to another when appropriate data are available in each.
Follow the Leader? Presidential Approval, Perceived Presidential Support, and Representatives'Electoral Fortunes'
Gronke, Paul, Koch, Jeffery, Wilson, J. Matthew
Submitted: 1998-04-17
Keywords: congress, 1994, presidential approval, projection, negative voting
Abstract: (click to show/hide) The relationship between presidential approval and congressional incumbent electoral success has been long established. Surprisingly, the individual level dynamics of this process have been largely unexamined. Drawing on a new set of questions included in the 1993, 1994, and 1996 National Election Studies, we explore the degree to which citizen perceptions of member support for Clinton's legislative program mediate the impact of presidential approval on evaluations and choice. We find that the degree to which individuals thought their members supported the President's legislative program functions just as we hypothesize, enhancing or ameliorating the impact of presidential approval on affective attachments to the member, evaluation of the incumbent's job performance, and congressional vote choice.
Gronke, Paul, Koch, Jeffery, Wilson, J. Matthew
Submitted: 1998-04-17
Keywords: congress, 1994, presidential approval, projection, negative voting
Abstract: (click to show/hide) The relationship between presidential approval and congressional incumbent electoral success has been long established. Surprisingly, the individual level dynamics of this process have been largely unexamined. Drawing on a new set of questions included in the 1993, 1994, and 1996 National Election Studies, we explore the degree to which citizen perceptions of member support for Clinton's legislative program mediate the impact of presidential approval on evaluations and choice. We find that the degree to which individuals thought their members supported the President's legislative program functions just as we hypothesize, enhancing or ameliorating the impact of presidential approval on affective attachments to the member, evaluation of the incumbent's job performance, and congressional vote choice.
Parties, Issue Spaces, and Voting: A Comparative Perspective
Alvarez, R. Michael, Nagler, Jonathan, Willette, Jennifer R.
Submitted: 1998-04-20
Keywords: elections, parties, issues, comparative
Abstract: (click to show/hide) An important property of any party system is the set of choices it presents to the electorate. In this paper we analyze the distribution of the parties in the multidimensional issue space, and introduce the notion of compactness of the party system. We show how compactness can be measured using standard survey items found on national election surveys. By measuring the spacing of the parties relative to the distribution of the voters, we are able to compute a metric-free measure of compactness of the party system. Comparing the compactness of party systems across countries allows us to determine the relative amount of issue choice afforded voters in different polities. We test the impact compactness of the party space has on voter choice in four countries: the United States, the Netherlands, Canada, and Great Britain. We demonstrate that the more compact the issue space on any issue, the less voters weight that issue in making their vote decision. Thus we provide evidence for theories of issue voting.
Alvarez, R. Michael, Nagler, Jonathan, Willette, Jennifer R.
Submitted: 1998-04-20
Keywords: elections, parties, issues, comparative
Abstract: (click to show/hide) An important property of any party system is the set of choices it presents to the electorate. In this paper we analyze the distribution of the parties in the multidimensional issue space, and introduce the notion of compactness of the party system. We show how compactness can be measured using standard survey items found on national election surveys. By measuring the spacing of the parties relative to the distribution of the voters, we are able to compute a metric-free measure of compactness of the party system. Comparing the compactness of party systems across countries allows us to determine the relative amount of issue choice afforded voters in different polities. We test the impact compactness of the party space has on voter choice in four countries: the United States, the Netherlands, Canada, and Great Britain. We demonstrate that the more compact the issue space on any issue, the less voters weight that issue in making their vote decision. Thus we provide evidence for theories of issue voting.
Mandate Elections and Policy Change in Congress
Peterson, David A.M., Stimson, James, Gangl, Amy E., Grossback, Lawrence J.
Submitted: 1998-04-20
Keywords: duration, discrete time, fractional polynomial, mandate
Abstract: (click to show/hide) We postulate that members of Congress react to occasional elections which are unusual in outcome, unexpected, and carry a clear message about the direction of voter preferences. Our cases are 1964/65, 1980/81, and 1994/95. That reaction is movement in the direction of the perceived mandate, where both aggregate outcomes and individual behavior deviate from long-term norms under the temporary influence of perception of dramatic movements in voter preferences. We undertake analyses that show aggregate shifts in Congressional outcomes and individual departures from personal equilibria. We conclude with an event history analysis designed to capture the duration of this temporary phenomenon.
Peterson, David A.M., Stimson, James, Gangl, Amy E., Grossback, Lawrence J.
Submitted: 1998-04-20
Keywords: duration, discrete time, fractional polynomial, mandate
Abstract: (click to show/hide) We postulate that members of Congress react to occasional elections which are unusual in outcome, unexpected, and carry a clear message about the direction of voter preferences. Our cases are 1964/65, 1980/81, and 1994/95. That reaction is movement in the direction of the perceived mandate, where both aggregate outcomes and individual behavior deviate from long-term norms under the temporary influence of perception of dramatic movements in voter preferences. We undertake analyses that show aggregate shifts in Congressional outcomes and individual departures from personal equilibria. We conclude with an event history analysis designed to capture the duration of this temporary phenomenon.
Duration Models and Proportional Hazards in Political Science
Box-Steffensmeier, Janet M.
Submitted: 1998-04-20
Keywords: (none submitted)
Abstract: (click to show/hide) In recent years political scientists have increasingly adopted a wide range of techniques for modeling duration data. A key assumption of all these approaches is that the hazard ratios (i.e., the conditional relative risks across substrata) are proportional to one another, and that this proportionality is maintained over time. Estimation of proportional hazards (PH) models when in fact hazards are non-proportional results in coefficient biases and decreased power of significance tests. In particular, misspecified PH models will overestimate the impact of variables whose associated hazards are increasing, while coefficient estimates for covariates in which the hazards are converging will be biased towards zero. We investigate the proportionality assumption of two widely used duration models, the Weibull parametric model and Cox's (1972) semiparametric approach, in the context of a duration modelof Supreme Court retirements. We address the potential problems with incorrectly assuming proportionality, illustrate a range of techniques for testing the proportionality assumption, and conclude with a number of means for accurately and efficiently estimating these models in the presence of non-proportional hazards.
Box-Steffensmeier, Janet M.
Submitted: 1998-04-20
Keywords: (none submitted)
Abstract: (click to show/hide) In recent years political scientists have increasingly adopted a wide range of techniques for modeling duration data. A key assumption of all these approaches is that the hazard ratios (i.e., the conditional relative risks across substrata) are proportional to one another, and that this proportionality is maintained over time. Estimation of proportional hazards (PH) models when in fact hazards are non-proportional results in coefficient biases and decreased power of significance tests. In particular, misspecified PH models will overestimate the impact of variables whose associated hazards are increasing, while coefficient estimates for covariates in which the hazards are converging will be biased towards zero. We investigate the proportionality assumption of two widely used duration models, the Weibull parametric model and Cox's (1972) semiparametric approach, in the context of a duration modelof Supreme Court retirements. We address the potential problems with incorrectly assuming proportionality, illustrate a range of techniques for testing the proportionality assumption, and conclude with a number of means for accurately and efficiently estimating these models in the presence of non-proportional hazards.
Duration Models and Proportional Hazards in Political Science
Zorn, Christopher, Box-Steffensmeier, Janet M.
Submitted: 1998-04-20
Keywords: (none submitted)
Abstract: (click to show/hide) In recent years political scientists have increasingly adopted a wide range of techniques for modeling duration data. A key assumption of all these approaches is that the hazard ratios (i.e., the conditional relative risks across substrata) are proportional to one another, and that this proportionality is maintained over time. Estimation of proportional hazards (PH) models when in fact hazards are non-proportional results in coefficient biases and decreased power of significance tests. In particular, misspecified PH models will overestimate the impact of variables whose associated hazards are increasing, while coefficient estimates for covariates in which the hazards are converging will be biased towards zero. We investigate the proportionality assumption of two widely used duration models, the Weibull parametric model and Cox's (1972) semiparametric approach, in the context of a duration modelof Supreme Court retirements. We address the potential problems with incorrectly assuming proportionality, illustrate a range of techniques for testing the proportionality assumption, and conclude with a number of means for accurately and efficiently estimating these models in the presence of non-proportional hazards.
Zorn, Christopher, Box-Steffensmeier, Janet M.
Submitted: 1998-04-20
Keywords: (none submitted)
Abstract: (click to show/hide) In recent years political scientists have increasingly adopted a wide range of techniques for modeling duration data. A key assumption of all these approaches is that the hazard ratios (i.e., the conditional relative risks across substrata) are proportional to one another, and that this proportionality is maintained over time. Estimation of proportional hazards (PH) models when in fact hazards are non-proportional results in coefficient biases and decreased power of significance tests. In particular, misspecified PH models will overestimate the impact of variables whose associated hazards are increasing, while coefficient estimates for covariates in which the hazards are converging will be biased towards zero. We investigate the proportionality assumption of two widely used duration models, the Weibull parametric model and Cox's (1972) semiparametric approach, in the context of a duration modelof Supreme Court retirements. We address the potential problems with incorrectly assuming proportionality, illustrate a range of techniques for testing the proportionality assumption, and conclude with a number of means for accurately and efficiently estimating these models in the presence of non-proportional hazards.
A Theory of Nonseparable Preferences in Survey Responses (Revised, with New Evidence)
Lacy, Dean
Submitted: 1998-04-20
Keywords: surveys, nonseparable preferences, question-order effects, temporal instability
Abstract: (click to show/hide) This paper presents two models of individual-level responses to issue questions in public opinion surveys when respondents have nonseparable preferences. Both models imply that even when survey respondents have fixed preferences, their responses will change depending on the order of questions, and responses may vary over time. Results from two survey experiments reveal that question-order effects occur on issues for which people have nonseparable preferences, and order effects do not occur on issues for which most people have separable preferences.
Lacy, Dean
Submitted: 1998-04-20
Keywords: surveys, nonseparable preferences, question-order effects, temporal instability
Abstract: (click to show/hide) This paper presents two models of individual-level responses to issue questions in public opinion surveys when respondents have nonseparable preferences. Both models imply that even when survey respondents have fixed preferences, their responses will change depending on the order of questions, and responses may vary over time. Results from two survey experiments reveal that question-order effects occur on issues for which people have nonseparable preferences, and order effects do not occur on issues for which most people have separable preferences.
Persuasion Through the Purse: How Political Contributions Crowd Out Information
Bennedsen, Morton, Feldmann, Sven E.
Submitted: 1998-04-21
Keywords: lobbying, campaign contributions, information
Abstract: (click to show/hide) Interest groups can influence political decisions in two distinct ways: by offering contributions to political actors and by providing them with relevant information that is advantageous for the group. We analyze the conditions under which interest groups are more inclined to use one or the other channel of in uence. First, we identify an indirect cost of searching for information in the form of an information externality that increases the cost of offering contributions. We then show that an extreme interest group might find it beneficial to abandon information search altogether and instead seek in uence solely via contributions. Thus, our analysis lends support to a rather cynical view of lobbying wherein groups provide little or no useful information.
Bennedsen, Morton, Feldmann, Sven E.
Submitted: 1998-04-21
Keywords: lobbying, campaign contributions, information
Abstract: (click to show/hide) Interest groups can influence political decisions in two distinct ways: by offering contributions to political actors and by providing them with relevant information that is advantageous for the group. We analyze the conditions under which interest groups are more inclined to use one or the other channel of in uence. First, we identify an indirect cost of searching for information in the form of an information externality that increases the cost of offering contributions. We then show that an extreme interest group might find it beneficial to abandon information search altogether and instead seek in uence solely via contributions. Thus, our analysis lends support to a rather cynical view of lobbying wherein groups provide little or no useful information.
A Random Effects Approach to Legislative Ideal Point Estimation
Bailey, Michael
Submitted: 1998-04-21
Keywords: ideal points, random effects models, Bayesian estimation, em algorithm
Abstract: (click to show/hide) Conventionally, scholars use either standard probit/logit techniques or fixed-effect methods to estimate legislative ideal points. However, these methods are unsatisfactory when a limited number of votes are available: standard probit/logit methods are poorly equipped to handle multiple votes and fixed-effect models disregard serious ``incidental parameter'' problems. In this paper I present an alternative approach that moves beyond single-vote probit/logit analysis without requiring the large number of votes needed for fixed-effects models. The method is based on a random effects, panel logit framework that models ideal points as stochastic functions of legislator characteristics. Monte Carlo results and an application to trade politics demonstrate the practical usefulness of the method.
Bailey, Michael
Submitted: 1998-04-21
Keywords: ideal points, random effects models, Bayesian estimation, em algorithm
Abstract: (click to show/hide) Conventionally, scholars use either standard probit/logit techniques or fixed-effect methods to estimate legislative ideal points. However, these methods are unsatisfactory when a limited number of votes are available: standard probit/logit methods are poorly equipped to handle multiple votes and fixed-effect models disregard serious ``incidental parameter'' problems. In this paper I present an alternative approach that moves beyond single-vote probit/logit analysis without requiring the large number of votes needed for fixed-effects models. The method is based on a random effects, panel logit framework that models ideal points as stochastic functions of legislator characteristics. Monte Carlo results and an application to trade politics demonstrate the practical usefulness of the method.
Is Instrumental Rationality a Universal Phenomenon?
Bennett, D. Scott, Stamm, Allan C.
Submitted: 1998-04-22
Keywords: rational, expected utility, preferences, game theory
Abstract: (click to show/hide) This paper examines whether the expected utility theory of war explains international conflict equally well across all regions and time-periods as a way of examining whether instrumental rationality is a universal phenomenon. In the rational choice literature, scholars typically assume that decision-makers are purposive egoistic decision-makers with common preferences across various outcomes. However, critics of the assumption have suggested that preferences and decision structures vary as a function of polity type, culture and learning among state leaders. There have been few attempts to directly examine this assumption and evaluate whether it seems empirically justified. In this paper we attempt to test the assumption of common instrumental rationality, examining several competing hypotheses about the nature of decision making in international relations and expectations about where and when instrumental rationality should be most readily observable. In particular, we want to explore the effects of regional learning to discover if there is a difference by region and over time in the outbreak of war and the predictions of the expected utility model. We find important differences both over regions and over time in how the predictions of expected utility theory fit actual conflict occurrence.
Bennett, D. Scott, Stamm, Allan C.
Submitted: 1998-04-22
Keywords: rational, expected utility, preferences, game theory
Abstract: (click to show/hide) This paper examines whether the expected utility theory of war explains international conflict equally well across all regions and time-periods as a way of examining whether instrumental rationality is a universal phenomenon. In the rational choice literature, scholars typically assume that decision-makers are purposive egoistic decision-makers with common preferences across various outcomes. However, critics of the assumption have suggested that preferences and decision structures vary as a function of polity type, culture and learning among state leaders. There have been few attempts to directly examine this assumption and evaluate whether it seems empirically justified. In this paper we attempt to test the assumption of common instrumental rationality, examining several competing hypotheses about the nature of decision making in international relations and expectations about where and when instrumental rationality should be most readily observable. In particular, we want to explore the effects of regional learning to discover if there is a difference by region and over time in the outbreak of war and the predictions of the expected utility model. We find important differences both over regions and over time in how the predictions of expected utility theory fit actual conflict occurrence.
Is Instrumental Rationality a Universal Phenomenon?
Bennett, D. Scott, Stam, III, Allan C.
Submitted: 1998-04-22
Keywords: rational, expected utility, preferences, game theory
Abstract: (click to show/hide) This paper examines whether the expected utility theory of war explains international conflict equally well across all regions and time-periods as a way of examining whether instrumental rationality is a universal phenomenon. In the rational choice literature, scholars typically assume that decision-makers are purposive egoistic decision-makers with common preferences across various outcomes. However, critics of the assumption have suggested that preferences and decision structures vary as a function of polity type, culture and learning among state leaders. There have been few attempts to directly examine this assumption and evaluate whether it seems empirically justified. In this paper we attempt to test the assumption of common instrumental rationality, examining several competing hypotheses about the nature of decision making in international relations and expectations about where and when instrumental rationality should be most readily observable. In particular, we want to explore the effects of regional learning to discover if there is a difference by region and over time in the outbreak of war and the predictions of the expected utility model. We find important differences both over regions and over time in how the predictions of expected utility theory fit actual conflict occurrence.
Bennett, D. Scott, Stam, III, Allan C.
Submitted: 1998-04-22
Keywords: rational, expected utility, preferences, game theory
Abstract: (click to show/hide) This paper examines whether the expected utility theory of war explains international conflict equally well across all regions and time-periods as a way of examining whether instrumental rationality is a universal phenomenon. In the rational choice literature, scholars typically assume that decision-makers are purposive egoistic decision-makers with common preferences across various outcomes. However, critics of the assumption have suggested that preferences and decision structures vary as a function of polity type, culture and learning among state leaders. There have been few attempts to directly examine this assumption and evaluate whether it seems empirically justified. In this paper we attempt to test the assumption of common instrumental rationality, examining several competing hypotheses about the nature of decision making in international relations and expectations about where and when instrumental rationality should be most readily observable. In particular, we want to explore the effects of regional learning to discover if there is a difference by region and over time in the outbreak of war and the predictions of the expected utility model. We find important differences both over regions and over time in how the predictions of expected utility theory fit actual conflict occurrence.
Cosponsorship Coalitions in the U.S. House of Representatives
Grant, J. Tobin, Pellegrini, Pasquale (Pat) A.
Submitted: 1998-04-22
Keywords: clustering, coalitions, cosponsorship, duration models, hazard models, heterogeneity, spatial models
Abstract: (click to show/hide) urrent theories and methods for studying of cosponsorship assume that the decision to cosponsor is identical to decision to vote. In this paper we develop a new theory of cosponsorship that identifies where along the ideological spectrum cosponsors of a bill are more likely to be. Moreover, we predict that members with organizational ties to the sponsor are more likely to cosponsor than other members. To test this theory, we employ a spatial duration model. This method has recently been used by geographers to estimate areas that are more likely to experience an "event." Using this technique permits a statistical test that supports our substantive hypotheses that cosponsorship coalitions are shaped by the characteristics of the location of the bill, the shared ties to the sponsor, and the policy area. In addition, more active sponsors are associated with wider and less clustered coalitions. These findings demonstrate that theories of the voting decision are not applicable to cosponsorship.
Grant, J. Tobin, Pellegrini, Pasquale (Pat) A.
Submitted: 1998-04-22
Keywords: clustering, coalitions, cosponsorship, duration models, hazard models, heterogeneity, spatial models
Abstract: (click to show/hide) urrent theories and methods for studying of cosponsorship assume that the decision to cosponsor is identical to decision to vote. In this paper we develop a new theory of cosponsorship that identifies where along the ideological spectrum cosponsors of a bill are more likely to be. Moreover, we predict that members with organizational ties to the sponsor are more likely to cosponsor than other members. To test this theory, we employ a spatial duration model. This method has recently been used by geographers to estimate areas that are more likely to experience an "event." Using this technique permits a statistical test that supports our substantive hypotheses that cosponsorship coalitions are shaped by the characteristics of the location of the bill, the shared ties to the sponsor, and the policy area. In addition, more active sponsors are associated with wider and less clustered coalitions. These findings demonstrate that theories of the voting decision are not applicable to cosponsorship.
Cointegration Tests when Data are Near-Integrated
De Boef, Suzanna, Granato, Jim
Submitted: 1998-04-22
Keywords: time series, near-integration, ECMs, DickeyFuller tests, Monte Carlo
Abstract: (click to show/hide) Testing theories about political change requires analysts to make assumptions about the nature of the memory of their time series. Applied analyses are often based on inferences that the time series of interest are integrated and cointegrated. Typically these analyses rest on Dickey-Fuller pretests for unit roots and tests for cointegration based on the residuals from a cointegrating regression in the context of the Engle-Granger two-step methodology. We argue that this approach is not a good one and use Monte Carlo results to show that these tests can lead analysts to falsely conclude that the data are cointegrated (or nearly- cointegrated) when the data is near-integrated and not cointegrating. Further, analysts are likely to falsely conclude the relationship is not cointegrating when it is. We show how inferences are highly sensitive to sample size and the signal to noise ratio in the data. We suggest that analysts use the single equation error correction test for cointegrating relationships, and that caution be used in all cases where near-integration is a reasonable alternative to unit roots. Finally, we suggest that in many cases analysts can drop the language of cointegration and adopt single equation error correction models when the theory of error correction is relevant.
De Boef, Suzanna, Granato, Jim
Submitted: 1998-04-22
Keywords: time series, near-integration, ECMs, DickeyFuller tests, Monte Carlo
Abstract: (click to show/hide) Testing theories about political change requires analysts to make assumptions about the nature of the memory of their time series. Applied analyses are often based on inferences that the time series of interest are integrated and cointegrated. Typically these analyses rest on Dickey-Fuller pretests for unit roots and tests for cointegration based on the residuals from a cointegrating regression in the context of the Engle-Granger two-step methodology. We argue that this approach is not a good one and use Monte Carlo results to show that these tests can lead analysts to falsely conclude that the data are cointegrated (or nearly- cointegrated) when the data is near-integrated and not cointegrating. Further, analysts are likely to falsely conclude the relationship is not cointegrating when it is. We show how inferences are highly sensitive to sample size and the signal to noise ratio in the data. We suggest that analysts use the single equation error correction test for cointegrating relationships, and that caution be used in all cases where near-integration is a reasonable alternative to unit roots. Finally, we suggest that in many cases analysts can drop the language of cointegration and adopt single equation error correction models when the theory of error correction is relevant.
The Presidential Election of 1988: Low Voter Turnout and the Defeat of Michael Dukakis
Herron, Michael C.
Submitted: 1998-05-18
Keywords: turnout, abstention, 1988 election, Heckman model
Abstract: (click to show/hide) This paper studies voting and abstention in the 1988 presidential election and in particular focuses on one of the election's more noteworthy features: voter turnout in 1988 was 50.1%, low even by American presidential election standards. In light of such a low turnout rate, we ask the following two questions: did low voter turnout in 1988 harm the Democratic candidate Michael Dukakis or his Republican counterpart George Bush? And, if the full franchise had voted in 1988, who would have won the election? According to our estimates, individuals who abstained from voting in 1988 were primarily supporters of Dukakis, and we find that Dukakis would almost certainly have won the 1988 election had the full franchise voted. The bottom line is that Bush greatly benefited from low voter turnout, and we conclude that, on account of the dismal turnout rate, the 1988 election outcome was not representative of average citizen preferences. Our findings, based on data from the 1988 American National Election Study, are consonant with other turnout--related research, in particular Radcliff's aggregate study of presidential elections from 1928 to 1980. In contrast with Radcliff's analysis, though, this paper's use of individual--level as opposed to aggregate data allows us to describe the key determinants of voting and abstention in 1988. Our results highlight the deleterious consequences of low voter turnout and imply that efforts to stimulate voting may improve the likelihood that presidential election outcomes are representative of citizen, as opposed to solely voter, preferences.
Herron, Michael C.
Submitted: 1998-05-18
Keywords: turnout, abstention, 1988 election, Heckman model
Abstract: (click to show/hide) This paper studies voting and abstention in the 1988 presidential election and in particular focuses on one of the election's more noteworthy features: voter turnout in 1988 was 50.1%, low even by American presidential election standards. In light of such a low turnout rate, we ask the following two questions: did low voter turnout in 1988 harm the Democratic candidate Michael Dukakis or his Republican counterpart George Bush? And, if the full franchise had voted in 1988, who would have won the election? According to our estimates, individuals who abstained from voting in 1988 were primarily supporters of Dukakis, and we find that Dukakis would almost certainly have won the 1988 election had the full franchise voted. The bottom line is that Bush greatly benefited from low voter turnout, and we conclude that, on account of the dismal turnout rate, the 1988 election outcome was not representative of average citizen preferences. Our findings, based on data from the 1988 American National Election Study, are consonant with other turnout--related research, in particular Radcliff's aggregate study of presidential elections from 1928 to 1980. In contrast with Radcliff's analysis, though, this paper's use of individual--level as opposed to aggregate data allows us to describe the key determinants of voting and abstention in 1988. Our results highlight the deleterious consequences of low voter turnout and imply that efforts to stimulate voting may improve the likelihood that presidential election outcomes are representative of citizen, as opposed to solely voter, preferences.
Indifference, Voting, and Abstention in the 1976 Presidential Election
Herron, Michael C.
Submitted: 1998-05-19
Keywords: 1976 election, indifference, abstention, turnout, voting
Abstract: (click to show/hide) This paper develops a statistical model of voting and abstention and applies it to the presidential election of 1976, a contest between incumbent president Gerald Ford and Democratic challenger Jimmy Carter. Our model is grounded in random utility theory, and, unlike many extant models of voting and abstention, its treatment of turnout focuses on the distinction between political extremists and individuals who were close to indifferent between Ford and Carter. We expect that individuals close to indifferent abstained at lower rates than political extremists. And, in light of research which highlights relatively high abstention rates among Democratic supporters, our model allows politically--left extremists to abstain at different rates than politically--right extremists. We uncover some evidence that indifference between Ford and Carter exerted a downward influence on voting propensity in 1976. However, there is much stronger evidence that individuals who were politically--left extremists abstained at higher rates than all others. We also find that individuals who anticipated a close election in 1976 voted at higher rates than those who expected a lopsided victory. The value of the paper's model is its focus on the relation between abstention and strength of preference. Generalizations and applications of the model to additional presidential elections should foster a determination of whether indifference is as important to abstention as is politically--left extremism.
Herron, Michael C.
Submitted: 1998-05-19
Keywords: 1976 election, indifference, abstention, turnout, voting
Abstract: (click to show/hide) This paper develops a statistical model of voting and abstention and applies it to the presidential election of 1976, a contest between incumbent president Gerald Ford and Democratic challenger Jimmy Carter. Our model is grounded in random utility theory, and, unlike many extant models of voting and abstention, its treatment of turnout focuses on the distinction between political extremists and individuals who were close to indifferent between Ford and Carter. We expect that individuals close to indifferent abstained at lower rates than political extremists. And, in light of research which highlights relatively high abstention rates among Democratic supporters, our model allows politically--left extremists to abstain at different rates than politically--right extremists. We uncover some evidence that indifference between Ford and Carter exerted a downward influence on voting propensity in 1976. However, there is much stronger evidence that individuals who were politically--left extremists abstained at higher rates than all others. We also find that individuals who anticipated a close election in 1976 voted at higher rates than those who expected a lopsided victory. The value of the paper's model is its focus on the relation between abstention and strength of preference. Generalizations and applications of the model to additional presidential elections should foster a determination of whether indifference is as important to abstention as is politically--left extremism.
Sensitivity of GARCH Estimates: Effects of Model Specification on Estimates of Macropartisan Volatility
Maestas, Cherie, Gleditsch, Kristian S.
Submitted: 1998-05-24
Keywords: volatility of aggregate partisanship, ARCH/GARCH
Abstract: (click to show/hide) This paper explores the volatility of aggregate partisanship using a generalized autoregressive conditional heteroskedasticity (GARCH) model of the variance. We are particularly interested in how different specifications of the mean model affect the variance estimates. Modeling the variance of macropartisanship is theoretically interesting because such a model can capture periods of greater and lesser volatility in aggregate party identification. However, given the widespread debate over the dynamic properties of the aggregate partisanship time series, a range of plausible specifications for the mean model should be considered before drawing conclusions about variance estimates. We find similar estimates of the variance effects using ARMA-GARCH, ARFIMA-GARCH, ARIMA-GARCH and ECM-GARCH models. Weak ties to party consistently predict greater volatility in all four models, while presidential election quarters are associated with greater volatility in three of the four models. Counter to our expectations, the candidate centered era of the last few decades is associated with lower average variance. Finally, all four models indicate that volatility tends to persist beyond the duration of the shock that sparks it.
Maestas, Cherie, Gleditsch, Kristian S.
Submitted: 1998-05-24
Keywords: volatility of aggregate partisanship, ARCH/GARCH
Abstract: (click to show/hide) This paper explores the volatility of aggregate partisanship using a generalized autoregressive conditional heteroskedasticity (GARCH) model of the variance. We are particularly interested in how different specifications of the mean model affect the variance estimates. Modeling the variance of macropartisanship is theoretically interesting because such a model can capture periods of greater and lesser volatility in aggregate party identification. However, given the widespread debate over the dynamic properties of the aggregate partisanship time series, a range of plausible specifications for the mean model should be considered before drawing conclusions about variance estimates. We find similar estimates of the variance effects using ARMA-GARCH, ARFIMA-GARCH, ARIMA-GARCH and ECM-GARCH models. Weak ties to party consistently predict greater volatility in all four models, while presidential election quarters are associated with greater volatility in three of the four models. Counter to our expectations, the candidate centered era of the last few decades is associated with lower average variance. Finally, all four models indicate that volatility tends to persist beyond the duration of the shock that sparks it.
Binomial-Beta Hierarchical Models for Ecological Inference
King, Gary, Rosen, Ori, Tanner, Martin A.
Submitted: 1998-05-28
Keywords: ecological inference, aggregation, MCMC, hierarchical models, iterative simulation
Abstract: (click to show/hide) We develop a binomial-beta hierarchical model for ecological inference, using insights from King's (1997) ecological inference model and from the literature on hierarchical models based on Markov chain Monte Carlo algorithms (Tanner, 1996). Models in the framework we provide appear to scale up well, to have few numerical difficulties, and to recognize and avoid automatically problems with multiple modes and some other statistical issues.
King, Gary, Rosen, Ori, Tanner, Martin A.
Submitted: 1998-05-28
Keywords: ecological inference, aggregation, MCMC, hierarchical models, iterative simulation
Abstract: (click to show/hide) We develop a binomial-beta hierarchical model for ecological inference, using insights from King's (1997) ecological inference model and from the literature on hierarchical models based on Markov chain Monte Carlo algorithms (Tanner, 1996). Models in the framework we provide appear to scale up well, to have few numerical difficulties, and to recognize and avoid automatically problems with multiple modes and some other statistical issues.
No Evidence on Proximity vs. Directional Voting
Lewis, Jeffrey B., King, Gary
Submitted: 1998-06-05
Keywords: spatial models, voting, elections, decision models
Abstract: (click to show/hide) The directional and proximity models offer dramatically different theories for how voters make decisions. We demonstrate here that the empirical tests in the large and growing literature on this subject amount to theoretical debates about which statistical assumption is right. The key statistical assumptions in this literature have not been empirically tested, and indeed turn out to be effectively untestable with existing methods and data. Unfortunately, these assumptions are also crucial since changing them leads to different conclusions about voter decision processes.
Lewis, Jeffrey B., King, Gary
Submitted: 1998-06-05
Keywords: spatial models, voting, elections, decision models
Abstract: (click to show/hide) The directional and proximity models offer dramatically different theories for how voters make decisions. We demonstrate here that the empirical tests in the large and growing literature on this subject amount to theoretical debates about which statistical assumption is right. The key statistical assumptions in this literature have not been empirically tested, and indeed turn out to be effectively untestable with existing methods and data. Unfortunately, these assumptions are also crucial since changing them leads to different conclusions about voter decision processes.
Estimating Time-Varying Parameters with Flexible Least Squares
Wood, B. Dan
Submitted: 1998-07-02
Keywords: time series, time-varying parameters, stochastic parameters, flexible least squares
Abstract: (click to show/hide) A common assumption among time series analysts is that estimated coefficients remain constant through time. Yet this strong assumption often has little grounds in substantive theory or empirical tests. If coefficients vary through time in an infinite time sequence, but are estimated with constant coefficient methods in a finite time sequence, then this can lead to significant information loss, as well as to errors of inference. This paper demonstrates a method for exploring the relative stability of time series coefficients, Flexible Least Squares (FLS). In particular, FLS is superior to other such methods, in that it enables the analyst to diagnose the magnitude of coefficient variation, as well as detect which particular coefficients are changing. FLS also provides an estimated vector of time-varying coefficients that can be used for exploratory or descriptive purposes. FLS properties are demonstrated through simulation analysis and an evaluation of the time-varying equilibrium between federal revenues and expenditures from 1904-1996.
Wood, B. Dan
Submitted: 1998-07-02
Keywords: time series, time-varying parameters, stochastic parameters, flexible least squares
Abstract: (click to show/hide) A common assumption among time series analysts is that estimated coefficients remain constant through time. Yet this strong assumption often has little grounds in substantive theory or empirical tests. If coefficients vary through time in an infinite time sequence, but are estimated with constant coefficient methods in a finite time sequence, then this can lead to significant information loss, as well as to errors of inference. This paper demonstrates a method for exploring the relative stability of time series coefficients, Flexible Least Squares (FLS). In particular, FLS is superior to other such methods, in that it enables the analyst to diagnose the magnitude of coefficient variation, as well as detect which particular coefficients are changing. FLS also provides an estimated vector of time-varying coefficients that can be used for exploratory or descriptive purposes. FLS properties are demonstrated through simulation analysis and an evaluation of the time-varying equilibrium between federal revenues and expenditures from 1904-1996.
A Monte Carlo Comparison of Methods for Compositional Data Analysis
Brehm, John, Gates, Scott, Gomez, Brad
Submitted: 1998-07-08
Keywords: Compositional data, Dirichlet, Additive Logistic, Monte Carlo, Police Behavior
Abstract: (click to show/hide) This paper offers an explication of two techniques for compositional data analysis, which involve non-negative data belonging to mutually exclusive and exhaustive categories. The Dirichlet distribution is a multivariate generalization of the beta distribution that offers considerable flexibility, and ease of use, but requires a strong form of an ``independence of irrelevant alternatives'' (IIA) assumption. The second application, proposed by Aitchison (1986) and applied to political data by Katz and King (1997), is the additive logistic method. This approach addresses the strong IIA assumption, but cannot handle strong forms of independence (Rayens and Srinivasen 1994). Monte Carlo simulations are employed on compositional data to explore the limits of applications of the two methods. Data on police officers' allocation of time across a variety of tasks (Ostrom et al. 1988) is used in this analysis. Comparing both common covariates and unique covariates. When the composites are influenced by common covariates, there appears to be no advantage in the use of additive logistic methods over the Dirichlet. Similarly, the additive logistic and Dirichlet methods appear to be equally successful at estimating the effects of the unique covariates on composites. From these simulation results we conclude that the additive logistic method offers little advantage over the Dirichlet, and suffers from several disadvantages.
Brehm, John, Gates, Scott, Gomez, Brad
Submitted: 1998-07-08
Keywords: Compositional data, Dirichlet, Additive Logistic, Monte Carlo, Police Behavior
Abstract: (click to show/hide) This paper offers an explication of two techniques for compositional data analysis, which involve non-negative data belonging to mutually exclusive and exhaustive categories. The Dirichlet distribution is a multivariate generalization of the beta distribution that offers considerable flexibility, and ease of use, but requires a strong form of an ``independence of irrelevant alternatives'' (IIA) assumption. The second application, proposed by Aitchison (1986) and applied to political data by Katz and King (1997), is the additive logistic method. This approach addresses the strong IIA assumption, but cannot handle strong forms of independence (Rayens and Srinivasen 1994). Monte Carlo simulations are employed on compositional data to explore the limits of applications of the two methods. Data on police officers' allocation of time across a variety of tasks (Ostrom et al. 1988) is used in this analysis. Comparing both common covariates and unique covariates. When the composites are influenced by common covariates, there appears to be no advantage in the use of additive logistic methods over the Dirichlet. Similarly, the additive logistic and Dirichlet methods appear to be equally successful at estimating the effects of the unique covariates on composites. From these simulation results we conclude that the additive logistic method offers little advantage over the Dirichlet, and suffers from several disadvantages.
Modeling Time Series Count Data: A State-Space Approach to Event Counts
Brandt, Patrick T., Williams, John T., Fordham, Benjamin
Submitted: 1998-07-08
Keywords: Poisson models, event counts, state-space models, Kalman filter, non-normal time series
Abstract: (click to show/hide) This is a revised version, dated July 16, 1998. Time series count data is prevalent in political science. We argue that political scientists should employ time series methods to analyze time series count data. A simple state-space model is presented that extends the Kalman filter to count data. The properties of this model are outlined and further evaluated by a Monte Carlo study. We then show how time series of counts present special problems by turning to two replications: the number of hospital deaths that are the subject of a recent criminal court case, and Pollins (1996) MIDs data from international relations.
Brandt, Patrick T., Williams, John T., Fordham, Benjamin
Submitted: 1998-07-08
Keywords: Poisson models, event counts, state-space models, Kalman filter, non-normal time series
Abstract: (click to show/hide) This is a revised version, dated July 16, 1998. Time series count data is prevalent in political science. We argue that political scientists should employ time series methods to analyze time series count data. A simple state-space model is presented that extends the Kalman filter to count data. The properties of this model are outlined and further evaluated by a Monte Carlo study. We then show how time series of counts present special problems by turning to two replications: the number of hospital deaths that are the subject of a recent criminal court case, and Pollins (1996) MIDs data from international relations.
Rational Expectations Coordinating Voting in American Presidential and House Elections
Mebane, Walter R.
Submitted: 1998-07-08
Keywords: coordinating voting, probabilistic voting, spatial voting, retrospective voting, policy moderation, presidential elections, congressional elections, ticket splitting, rational expectations, voter equilibrium, Bayesian-Nash equilibrium, generalized extreme value model, nonparametric, Monte Carlo integration, maximum likelihood
Abstract: (click to show/hide) I define a probabilistic model of individuals' presidential-year vote choices for President and for the House of Representatives in which there is a coordinating (Bayesian Nash) equilibrium among voters based on rational expectations each voter has about the election outcomes. I estimate the model using data from the six American National Election Study Pre-/Post-Election Surveys of years 1976--1996. The coordinating model passes a variety of tests, including a test against a majoritarian model in which there is rational ticket splitting but no coordination. The results give strong individual-level support to Alesina and Rosenthal's theory that voters balance institutions in order to moderate policy. The estimates describe vote choices that strongly emphasize the presidential candidates. I also find that a voter who says economic conditions have improved puts more weight on a discrepancy between the voter's ideal point and government policy with a Democratic President than on a discrepancy of the same size with a Republican President.
Mebane, Walter R.
Submitted: 1998-07-08
Keywords: coordinating voting, probabilistic voting, spatial voting, retrospective voting, policy moderation, presidential elections, congressional elections, ticket splitting, rational expectations, voter equilibrium, Bayesian-Nash equilibrium, generalized extreme value model, nonparametric, Monte Carlo integration, maximum likelihood
Abstract: (click to show/hide) I define a probabilistic model of individuals' presidential-year vote choices for President and for the House of Representatives in which there is a coordinating (Bayesian Nash) equilibrium among voters based on rational expectations each voter has about the election outcomes. I estimate the model using data from the six American National Election Study Pre-/Post-Election Surveys of years 1976--1996. The coordinating model passes a variety of tests, including a test against a majoritarian model in which there is rational ticket splitting but no coordination. The results give strong individual-level support to Alesina and Rosenthal's theory that voters balance institutions in order to moderate policy. The estimates describe vote choices that strongly emphasize the presidential candidates. I also find that a voter who says economic conditions have improved puts more weight on a discrepancy between the voter's ideal point and government policy with a Democratic President than on a discrepancy of the same size with a Republican President.
Time to Give: PAC Motivations and Electoral Timing
McCarty, Nolan, Rothenberg, Lawrence S.
Submitted: 1998-07-09
Keywords: Interest Groups, Campaign Finance, Tobit, GHK Simulation
Abstract: (click to show/hide) There has been much discussion about how members of Congress desire money early in the campaign season. However, to date, models of how contributions are allocated during the electoral cycle have been lacking. Our analysis attempts to remedy this gap by providing and testing a model which specifies how the process by which bargaining between members of Congress and organized interests produces the pattern of donations observed over the course of the electoral cycle. The results suggest that strategic incumbents can receive money early in the campaign if they desire but that they are generally unwilling to pay the price of lower aggregate fundraising and greater provision of access. These findings, in turn, buttress earlier empirical findings that question the instrumental value of early money; in particular, they imply that challengers have reasonably rational and informed expectations about how much money members of Congress are capable of raising over the electoral cycle and that the value of stockpiling money early is not sufficiently high to induce reelection-seeking incumbents to lower their access price significan
McCarty, Nolan, Rothenberg, Lawrence S.
Submitted: 1998-07-09
Keywords: Interest Groups, Campaign Finance, Tobit, GHK Simulation
Abstract: (click to show/hide) There has been much discussion about how members of Congress desire money early in the campaign season. However, to date, models of how contributions are allocated during the electoral cycle have been lacking. Our analysis attempts to remedy this gap by providing and testing a model which specifies how the process by which bargaining between members of Congress and organized interests produces the pattern of donations observed over the course of the electoral cycle. The results suggest that strategic incumbents can receive money early in the campaign if they desire but that they are generally unwilling to pay the price of lower aggregate fundraising and greater provision of access. These findings, in turn, buttress earlier empirical findings that question the instrumental value of early money; in particular, they imply that challengers have reasonably rational and informed expectations about how much money members of Congress are capable of raising over the electoral cycle and that the value of stockpiling money early is not sufficiently high to induce reelection-seeking incumbents to lower their access price significan
Panel Effects in the American National Election Studies
Bartels, Larry M.
Submitted: 1998-07-11
Keywords: panel attrition, panel conditioning, fractional pooling, two-stage auxiliary instrumental variables, American National Election Studies
Abstract: (click to show/hide) Parallel panel and fresh cross-section samples in recent NES surveys provide valuable leverage for assessing the magnitude of biases in statistical analyses of survey data due to panel attrition and panel conditioning. My analysis employing a variety of typical regression models suggests that, on average, panel biases reduce the inferential value of panel data by about ten percent. Biases in individual coefficients are rarely statistically "significant," even when panel and cross-section respondents have markedly different characteristics. Thus, while I propose adjustments for panel effects in both cross-sectional and dynamic analyses, such adjustments are unlikely to be necessary in typical applications using NES or similar data
Bartels, Larry M.
Submitted: 1998-07-11
Keywords: panel attrition, panel conditioning, fractional pooling, two-stage auxiliary instrumental variables, American National Election Studies
Abstract: (click to show/hide) Parallel panel and fresh cross-section samples in recent NES surveys provide valuable leverage for assessing the magnitude of biases in statistical analyses of survey data due to panel attrition and panel conditioning. My analysis employing a variety of typical regression models suggests that, on average, panel biases reduce the inferential value of panel data by about ten percent. Biases in individual coefficients are rarely statistically "significant," even when panel and cross-section respondents have markedly different characteristics. Thus, while I propose adjustments for panel effects in both cross-sectional and dynamic analyses, such adjustments are unlikely to be necessary in typical applications using NES or similar data
A Method-Matching Approach to Maximum Likelihood Estimation of the Beta Distribution
Paolino, Philip
Submitted: 1998-07-11
Keywords: maximum-likelihood, beta distribution, Monte Carlo
Abstract: (click to show/hide) The beta distribution is a flexible distribution that can produce a uniform, unimodal, or bimodal distribution of points that can be either symmetric or skewed, but because the two shape parameters in a standard beta distribution do not correspond to the mean and the variance of the distribution, it it not obvious how one tests for the statistical significance of independent variables upon the mean or variance. In this paper, I will first discuss a "standard" approach to this problem as well as develop a "moment-matching" approach. Second, I will use Monte Carlo simulations to examine how well these approaches reproduce the true values of the function given different sample sizes and conditions. Third, I will present some empirical results using the moment-matching approach and compare these results with those obtained from the "standard" approach. From this work, I conclude that while the "moment-matching" approach produces reasonable estimates under the most common situations, the "standard" approach, using a Wald test to evaluate statistical significance, generally outperforms the "moment-matching" approach. As such, while the "moment-matching" approach has the attractive feature of allowing the researcher to estimate a variable's effect upon the mean or variance directly, its use is probably limited to instances where the researcher has a very good reason for wanting to constrain certain parameters to having zero effect upon either the mean or varianc
Paolino, Philip
Submitted: 1998-07-11
Keywords: maximum-likelihood, beta distribution, Monte Carlo
Abstract: (click to show/hide) The beta distribution is a flexible distribution that can produce a uniform, unimodal, or bimodal distribution of points that can be either symmetric or skewed, but because the two shape parameters in a standard beta distribution do not correspond to the mean and the variance of the distribution, it it not obvious how one tests for the statistical significance of independent variables upon the mean or variance. In this paper, I will first discuss a "standard" approach to this problem as well as develop a "moment-matching" approach. Second, I will use Monte Carlo simulations to examine how well these approaches reproduce the true values of the function given different sample sizes and conditions. Third, I will present some empirical results using the moment-matching approach and compare these results with those obtained from the "standard" approach. From this work, I conclude that while the "moment-matching" approach produces reasonable estimates under the most common situations, the "standard" approach, using a Wald test to evaluate statistical significance, generally outperforms the "moment-matching" approach. As such, while the "moment-matching" approach has the attractive feature of allowing the researcher to estimate a variable's effect upon the mean or variance directly, its use is probably limited to instances where the researcher has a very good reason for wanting to constrain certain parameters to having zero effect upon either the mean or varianc
Listwise Deletion is Evil: What to Do About Missing Data in Political Science
King, Gary, Honaker, James, Joseph, Anne, Scheve, Kenneth
Submitted: 1998-07-13
Keywords: missing data, imputation, IP, EM, EMs, EMis, data augmentation, MCMC, importance sampling, item nonresponse
Abstract: (click to show/hide) We address a substantial discrepancy between the way political scientists analyze data with missing values and the recommendations of the statistics community. With a few notable exceptions, statisticians and methodologists have agreed on a widely applicable approach to many missing data problems based on the concept of ``multiple imputation,'' but most researchers in our field and other social sciences still use far inferior methods. Indeed, we demonstrate that the threats to validity from current missing data practices rival the biases from the much better known omitted variable problem. This discrepancy is not entirely our fault, as the computational algorithms used to apply the best multiple imputation models have been slow, difficult to implement, impossible to run with existing commercial statistical packages, and demanding of considerable expertise on the part of the user (indeed, even experts disagree on how to use them). In this paper, we adapt an existing algorithm, and use it to implement a general-purpose, multiple imputation model for missing data. This algorithm is between 20 and 100 times faster than the leading method recommended in the statistics literature and is very easy to use. We also quantify the considerable risks of current political science missing data practices, illustrate how to use the new procedure, and demonstrate the advantages of our approach to multiple imputation through simulated data as well as via replications of existing research.
King, Gary, Honaker, James, Joseph, Anne, Scheve, Kenneth
Submitted: 1998-07-13
Keywords: missing data, imputation, IP, EM, EMs, EMis, data augmentation, MCMC, importance sampling, item nonresponse
Abstract: (click to show/hide) We address a substantial discrepancy between the way political scientists analyze data with missing values and the recommendations of the statistics community. With a few notable exceptions, statisticians and methodologists have agreed on a widely applicable approach to many missing data problems based on the concept of ``multiple imputation,'' but most researchers in our field and other social sciences still use far inferior methods. Indeed, we demonstrate that the threats to validity from current missing data practices rival the biases from the much better known omitted variable problem. This discrepancy is not entirely our fault, as the computational algorithms used to apply the best multiple imputation models have been slow, difficult to implement, impossible to run with existing commercial statistical packages, and demanding of considerable expertise on the part of the user (indeed, even experts disagree on how to use them). In this paper, we adapt an existing algorithm, and use it to implement a general-purpose, multiple imputation model for missing data. This algorithm is between 20 and 100 times faster than the leading method recommended in the statistics literature and is very easy to use. We also quantify the considerable risks of current political science missing data practices, illustrate how to use the new procedure, and demonstrate the advantages of our approach to multiple imputation through simulated data as well as via replications of existing research.
Aggregate Voting Data and Implied Spatial Voting
Herron, Michael C.
Submitted: 1998-07-15
Keywords: spatial voting, aggregate data, ecological inference, micro-foundations
Abstract: (click to show/hide) The paper draws attention to the micro--foundations of aggregate voting data by introducing the concept of an implied spatial voting model. The adjective ``implied'' refers to the fact that this paper's spatial theory primitives, which describe how individual--level preferences are distributed across and within voting districts, are implied by or derived from aggregate voting data. The key idea proposed here is that, given an observed distribution of aggregate voting data, it is possible to derive features of an individual--level, spatial voting model capable of generating the observed data. Thus, an implied spatial voting model is an inverse image of an observed, aggregate vote share distribution. We provide numerical examples of how spatial voting models can be implied by aggregate voting data and we then analyze aggregate data and National Election Study survey data from the 1980, 1984, and 1988 presidential elections. And, to demonstrate that implied spatial voting models can be calculated from aggregate data alone, we consider presidential elections 1928--1960 and the Chicago mayoral elections of 1983 and 1987. This paper's focus on the micro--foundations of aggregate data highlights the limitations inherent in aggregate data analyses. In particular, the paper discusses identification problems, in part a consequence of the lack of scale and location invariance in preference orderings and in part a consequence of the lack of individual--level information in aggregate data, that affect movement between individual--level theories like spatial voting theory and aggregate voting data.
Herron, Michael C.
Submitted: 1998-07-15
Keywords: spatial voting, aggregate data, ecological inference, micro-foundations
Abstract: (click to show/hide) The paper draws attention to the micro--foundations of aggregate voting data by introducing the concept of an implied spatial voting model. The adjective ``implied'' refers to the fact that this paper's spatial theory primitives, which describe how individual--level preferences are distributed across and within voting districts, are implied by or derived from aggregate voting data. The key idea proposed here is that, given an observed distribution of aggregate voting data, it is possible to derive features of an individual--level, spatial voting model capable of generating the observed data. Thus, an implied spatial voting model is an inverse image of an observed, aggregate vote share distribution. We provide numerical examples of how spatial voting models can be implied by aggregate voting data and we then analyze aggregate data and National Election Study survey data from the 1980, 1984, and 1988 presidential elections. And, to demonstrate that implied spatial voting models can be calculated from aggregate data alone, we consider presidential elections 1928--1960 and the Chicago mayoral elections of 1983 and 1987. This paper's focus on the micro--foundations of aggregate data highlights the limitations inherent in aggregate data analyses. In particular, the paper discusses identification problems, in part a consequence of the lack of scale and location invariance in preference orderings and in part a consequence of the lack of individual--level information in aggregate data, that affect movement between individual--level theories like spatial voting theory and aggregate voting data.
The Problem with Quantitative Studies of International Conflict
Beck, Nathaniel, King, Gary, Zeng, Langche
Submitted: 1998-07-15
Keywords: Conflict, logit, neural networks, forecasting, Bayesian analysis
Abstract: (click to show/hide) Despite immense data collections, prestigious journals, and sophisticated analyses, empirical findings in the literature on international conflict are frequently unsatisfying. Statistical results appear to change from article to article and specification to specification. Very few relationships hold up to replication with even minor respecification. Accurate forecasts are nonexistent. We provide a simple conjecture about what accounts for this problem, and offer a statistical framework that better matches the substantive issues and types of data in this field. Our model, a version of a ``neural network'' model, forecasts substantially better than any previous effort, and appears to uncover some structural features of international conflict.
Beck, Nathaniel, King, Gary, Zeng, Langche
Submitted: 1998-07-15
Keywords: Conflict, logit, neural networks, forecasting, Bayesian analysis
Abstract: (click to show/hide) Despite immense data collections, prestigious journals, and sophisticated analyses, empirical findings in the literature on international conflict are frequently unsatisfying. Statistical results appear to change from article to article and specification to specification. Very few relationships hold up to replication with even minor respecification. Accurate forecasts are nonexistent. We provide a simple conjecture about what accounts for this problem, and offer a statistical framework that better matches the substantive issues and types of data in this field. Our model, a version of a ``neural network'' model, forecasts substantially better than any previous effort, and appears to uncover some structural features of international conflict.
Democracy and Exchange Rates: An Experimental Study
Freeman, John R., Hays, Jude, Stix, Helmut
Submitted: 1998-07-17
Keywords: Markov switching model, exchange rates, comparative democracy, political economy
Abstract: (click to show/hide) The world's financial markets are becoming increasingly liberalized and interconnected. There is much debate about whether this development is socially desirable. Of increasing interest in this debate are the implications of the globalization of finance for democracy. The relationship between the workings of currency markets and democratic institutions is studied. The economic literature on exchange rate determination is briefly reviewed. The Markov switching model is considered as one of the most useful with which to analyze the politics of exchange rate determination. Next, the political science literature is discussed, including the research on electoral systems and comparative democracy. Out of this discussion emerge several competing propositions about how political (re)equilibration affects currency markets, more specifically, what the Markov switching framework should show about the impact of electoral outcomes and political polls on compound returns (the log difference of the exchange rate) in some or all democracies. A design for testing these propositions then is laid out and implemented. The results support the view that democratic politics affects currency markets. In particular, opinion polls about chief executive and government performance have a direct effect on the probabilities of switches between currency regimes. This suggests that these polls cause currency traders to revise their expectations about the stability of governments and (or) the content of public policy. In addition, the results refute claims that pluralist and majoritarian forms of democracy are more likely to be a source of trader uncertainty and hence regime shifts than corporatist and consensual forms of democracy. There is some evidence that (democratic) institutional "incoherency" (Garrett, 1998) is a source of market uncertainty and therefore that the effects of opinion polls and other political variables on the probabilities of regime shifts are greater in the respective countries.
Freeman, John R., Hays, Jude, Stix, Helmut
Submitted: 1998-07-17
Keywords: Markov switching model, exchange rates, comparative democracy, political economy
Abstract: (click to show/hide) The world's financial markets are becoming increasingly liberalized and interconnected. There is much debate about whether this development is socially desirable. Of increasing interest in this debate are the implications of the globalization of finance for democracy. The relationship between the workings of currency markets and democratic institutions is studied. The economic literature on exchange rate determination is briefly reviewed. The Markov switching model is considered as one of the most useful with which to analyze the politics of exchange rate determination. Next, the political science literature is discussed, including the research on electoral systems and comparative democracy. Out of this discussion emerge several competing propositions about how political (re)equilibration affects currency markets, more specifically, what the Markov switching framework should show about the impact of electoral outcomes and political polls on compound returns (the log difference of the exchange rate) in some or all democracies. A design for testing these propositions then is laid out and implemented. The results support the view that democratic politics affects currency markets. In particular, opinion polls about chief executive and government performance have a direct effect on the probabilities of switches between currency regimes. This suggests that these polls cause currency traders to revise their expectations about the stability of governments and (or) the content of public policy. In addition, the results refute claims that pluralist and majoritarian forms of democracy are more likely to be a source of trader uncertainty and hence regime shifts than corporatist and consensual forms of democracy. There is some evidence that (democratic) institutional "incoherency" (Garrett, 1998) is a source of market uncertainty and therefore that the effects of opinion polls and other political variables on the probabilities of regime shifts are greater in the respective countries.
A Dynamic Panel Analysis of Campaign Contributions in Elections for the U.S. House of Representatives
Himmelberg, Charles P., Wawro, Gregory
Submitted: 1998-07-17
Keywords: campaign finance, panel data methods, dynamics, GMM estimators, censoring
Abstract: (click to show/hide) Political scientists have recognized the importance of dynamics in understanding the role of campaign finance in congressional elections. Yet for the most part, researchers have not exploited available data to its fullest or used appropriate methods to answer questions of interest. Though the Federal Election Commission's reporting and disclosure requirements enable us to use panel data models, researchers have ignored these powerful tools. One of the main advantages of panel data methods is that they enable us to account for unobserved individual and temporal effects that, if not accounted for, might lead us to incorrect inferences. In this paper we describe the problems with estimating dynamic panel models and discuss techniques that correct for these problems. We apply recently developed panel data methods to estimate a dynamic model of campaign finance and assess the usefulness of these methods by examining the robustness of results obtained with more traditional methods. We examine the relationship between past and current campaign contributions to incumbents and challengers during the 1986 through 1992 election cycles. Dynamic panel estimators give results that differ in substantively interesting ways from those given by standard estimators. In particular, the estimates obtained from dynamic panel methods suggest that challengers who are successful fundraisers can cut into the fundraising efforts of incumbents.
Himmelberg, Charles P., Wawro, Gregory
Submitted: 1998-07-17
Keywords: campaign finance, panel data methods, dynamics, GMM estimators, censoring
Abstract: (click to show/hide) Political scientists have recognized the importance of dynamics in understanding the role of campaign finance in congressional elections. Yet for the most part, researchers have not exploited available data to its fullest or used appropriate methods to answer questions of interest. Though the Federal Election Commission's reporting and disclosure requirements enable us to use panel data models, researchers have ignored these powerful tools. One of the main advantages of panel data methods is that they enable us to account for unobserved individual and temporal effects that, if not accounted for, might lead us to incorrect inferences. In this paper we describe the problems with estimating dynamic panel models and discuss techniques that correct for these problems. We apply recently developed panel data methods to estimate a dynamic model of campaign finance and assess the usefulness of these methods by examining the robustness of results obtained with more traditional methods. We examine the relationship between past and current campaign contributions to incumbents and challengers during the 1986 through 1992 election cycles. Dynamic panel estimators give results that differ in substantively interesting ways from those given by standard estimators. In particular, the estimates obtained from dynamic panel methods suggest that challengers who are successful fundraisers can cut into the fundraising efforts of incumbents.
A Statistical Assessment of The Spatial Model of Ideology
Ghobarah, Hazen
Submitted: 1998-07-20
Keywords: spatial theory, macroµ-stability, ideology, maximum likelihood, multi-dimensional scaling
Abstract: (click to show/hide) The spatial model of ideology (Hinich and Munger, 1994) specifies a formal framework for linking positions of the electorate, the parties, and the candidates on a plethora of issues to positions on a few ideological dimensions- perhaps just one or two dimensions. While extant tests of this model have relied on cross-sectional survey data, this study utilizes a panel. The panel format allows a direct examination of the stability, and indeed the reality, of the parameters and the cognitive processes that are posited by the formal model. Given the available variables in the panel, I operationalize one model for party competition and another for presidential candidates. The results of both are supportive of the linkage model. The statistical methodology used in this study is no more complex than the model requires; it includes maximum likelihood factor analysis and a customized multi-dimensional scaling procedure.
Ghobarah, Hazen
Submitted: 1998-07-20
Keywords: spatial theory, macroµ-stability, ideology, maximum likelihood, multi-dimensional scaling
Abstract: (click to show/hide) The spatial model of ideology (Hinich and Munger, 1994) specifies a formal framework for linking positions of the electorate, the parties, and the candidates on a plethora of issues to positions on a few ideological dimensions- perhaps just one or two dimensions. While extant tests of this model have relied on cross-sectional survey data, this study utilizes a panel. The panel format allows a direct examination of the stability, and indeed the reality, of the parameters and the cognitive processes that are posited by the formal model. Given the available variables in the panel, I operationalize one model for party competition and another for presidential candidates. The results of both are supportive of the linkage model. The statistical methodology used in this study is no more complex than the model requires; it includes maximum likelihood factor analysis and a customized multi-dimensional scaling procedure.
Inferring Micro- from Macrolevel Change: Ecological Panel Inference in Surveys
Penubarti, Mohan, Schuessler, Alexander
Submitted: 1998-07-20
Keywords: Ecological panel inference (EPI), surveys, public opinion
Abstract: (click to show/hide) To draw panel inferences at the microlevel from cross-sectional surveys invites an ecological inference problem. In this paper we derive from King's ecological inference solution a method of ecological panel inference (EPI) which allows researchers to estimate microlevel change from macrolevel measures of change. We verify our approach in panel data where magnitudes of microlevel change are known, and we subsequently apply and illustrate our method using public opinion data on presidential approval. EPI should be of interest to researchers seeking to explain microlevel change in the absence of microlevel data. It should equally be of interest to researchers seeking to explain macrolevel change as it makes visible to them the microlevel components that drive such aggregate-level change.
Penubarti, Mohan, Schuessler, Alexander
Submitted: 1998-07-20
Keywords: Ecological panel inference (EPI), surveys, public opinion
Abstract: (click to show/hide) To draw panel inferences at the microlevel from cross-sectional surveys invites an ecological inference problem. In this paper we derive from King's ecological inference solution a method of ecological panel inference (EPI) which allows researchers to estimate microlevel change from macrolevel measures of change. We verify our approach in panel data where magnitudes of microlevel change are known, and we subsequently apply and illustrate our method using public opinion data on presidential approval. EPI should be of interest to researchers seeking to explain microlevel change in the absence of microlevel data. It should equally be of interest to researchers seeking to explain macrolevel change as it makes visible to them the microlevel components that drive such aggregate-level change.
Time Series Models for Discrete Data: solutions to a problem with quantitative studies of international conflict
Jackman, Simon
Submitted: 1998-07-21
Keywords: categorical time series, dependent binary data, Markov regression models, latent autoregressive process, Markov Chain Monte Carlo, international conflict, democratic peace
Abstract: (click to show/hide) Discrete dependent variables with a time series structure occupy something of a statistical limbo for even well-trained political scientists, prompting awkward methodological compromises and dubious substantive conclusions. An important example is the use of binary response models in the analysis of longitudinal data on international conflict: researchers understand that the data are not independent, but lack any way to model serial dependence in the data. Here I survey methods for modeling categorical data with a serial structure. I consider a number of simple models that enjoy frequent use outside of political science (originating in biostatistics), as well as a logit model with an autoregressive error structure (the latter model is fit via Bayesian simulation using Markov chain Monte Carlo methods). I illustrate these models in the context of international conflict data. Like other re-analyses of these data addressing the issue of serial dependence, \citeaffixed{beck:btscs}{e.g.,}, I find economic interdependence does not lessen the chances of international conflict. Other findings include a number of interesting asymmetries in the effects of covariates on transitions from peace to war (and vice versa). Any reasonable model of international conflict should take into account the high levels of persistence in the data; the models I present here suggest a number of methods for doing so.
Jackman, Simon
Submitted: 1998-07-21
Keywords: categorical time series, dependent binary data, Markov regression models, latent autoregressive process, Markov Chain Monte Carlo, international conflict, democratic peace
Abstract: (click to show/hide) Discrete dependent variables with a time series structure occupy something of a statistical limbo for even well-trained political scientists, prompting awkward methodological compromises and dubious substantive conclusions. An important example is the use of binary response models in the analysis of longitudinal data on international conflict: researchers understand that the data are not independent, but lack any way to model serial dependence in the data. Here I survey methods for modeling categorical data with a serial structure. I consider a number of simple models that enjoy frequent use outside of political science (originating in biostatistics), as well as a logit model with an autoregressive error structure (the latter model is fit via Bayesian simulation using Markov chain Monte Carlo methods). I illustrate these models in the context of international conflict data. Like other re-analyses of these data addressing the issue of serial dependence, \citeaffixed{beck:btscs}{e.g.,}, I find economic interdependence does not lessen the chances of international conflict. Other findings include a number of interesting asymmetries in the effects of covariates on transitions from peace to war (and vice versa). Any reasonable model of international conflict should take into account the high levels of persistence in the data; the models I present here suggest a number of methods for doing so.
Bias and Responsiveness in Multiparty and Multigroup Representation
Monroe, Burt L.
Submitted: 1998-07-21
Keywords: partisan bias, responsiveness, seats and votes, electoral systems, compositional data, JudgeIt
Abstract: (click to show/hide) There is an extensive and expanding literature that examines methods for estimating the responsiveness and partisan bias of two-party electoral systems. Attempts to extend these methods into the multiparty domain appropriate for the vast majority of electoral systems, or to the analysis of the representation of other types of groups (e.g., regions, ethnic groups), have been limited. I describe index, multiyear, uniform swing, and variable swing methods -- along with novel graphical displays -- for analyzing seats-votes curves, bias, and responsiveness in multiparty systems. The variable swing method is a multiparty generalization of Gelman and King's "JudgeIt" model. Examples discussed include elections in the UK, Mauritius, and Costa Rica, and geographic representation worldwide. In comparing the various methods it is argued that variable swing is ideal for most applications, that uniform swing and index methods provide useful answers to a limited set of questions despite faulty assumptions, and that multiyear methods are generally not useful.
Monroe, Burt L.
Submitted: 1998-07-21
Keywords: partisan bias, responsiveness, seats and votes, electoral systems, compositional data, JudgeIt
Abstract: (click to show/hide) There is an extensive and expanding literature that examines methods for estimating the responsiveness and partisan bias of two-party electoral systems. Attempts to extend these methods into the multiparty domain appropriate for the vast majority of electoral systems, or to the analysis of the representation of other types of groups (e.g., regions, ethnic groups), have been limited. I describe index, multiyear, uniform swing, and variable swing methods -- along with novel graphical displays -- for analyzing seats-votes curves, bias, and responsiveness in multiparty systems. The variable swing method is a multiparty generalization of Gelman and King's "JudgeIt" model. Examples discussed include elections in the UK, Mauritius, and Costa Rica, and geographic representation worldwide. In comparing the various methods it is argued that variable swing is ideal for most applications, that uniform swing and index methods provide useful answers to a limited set of questions despite faulty assumptions, and that multiyear methods are generally not useful.
Modeling Direction and Intensity in Ordinal Scales with Midpoints
Jones, Bradford S., Sobel, Michael E.
Submitted: 1998-07-21
Keywords: adjacent category logit, log-linear models, public opinion, Congress
Abstract: (click to show/hide) Political opinion analysts are frequently work with semantically balanced ordinal scales. Such survey items are frequently used to measure candidate evaluations, public spending preferences, positions on social issues, and candidate and party placement. Because of the special nature of these survey items (semantically balanced about a midpoint), researchers may be interested in understanding how both the response direction and response intensity varies over time and/or across covariate classes. That is, trends may be found in the tendency for respondents to choose categories above vs. below the midpoint (the response direction) and trends may be found in the tendency for respondents to choose between or among category labels above or below the midpoint. And while political analysts are commonly interested in response intensity and direction, traditional methods used to model distributions on semantically balanced ordinal scales are problematic. In this paper, we discuss a class of models originally developed by Sobel (1995, 1997, 1998) that allows researchers to simultaneously model direction and intensity in ordinal scales with midpoints. Specifically, we parameterize the model as an adjacent category logit model. Numerous parsimonious models may be arrived at that describe trends in the response direction and response intensity. Because the adjacent category logit model is linear in the logits, we estimate the model using log-linear models. We present an application of the models to data on approval ratings of House incumbents. We find that the trends in response directions (the tendency for respondents to evaluate the incumbent favorably or not favorably) increase through the 1980s, peaking in the late Eighties, and are now declining over the 1990s. With regard to response intensity, (that is, the tendency to respond in the extreme categories vs. the moderate categories), we find that intensity increases during most presidential election cycles and vanishes during midterm election years. We argue this finding is related to the different levels of political information citizens are exposed to in presidential vs. midterm election cycles.
Jones, Bradford S., Sobel, Michael E.
Submitted: 1998-07-21
Keywords: adjacent category logit, log-linear models, public opinion, Congress
Abstract: (click to show/hide) Political opinion analysts are frequently work with semantically balanced ordinal scales. Such survey items are frequently used to measure candidate evaluations, public spending preferences, positions on social issues, and candidate and party placement. Because of the special nature of these survey items (semantically balanced about a midpoint), researchers may be interested in understanding how both the response direction and response intensity varies over time and/or across covariate classes. That is, trends may be found in the tendency for respondents to choose categories above vs. below the midpoint (the response direction) and trends may be found in the tendency for respondents to choose between or among category labels above or below the midpoint. And while political analysts are commonly interested in response intensity and direction, traditional methods used to model distributions on semantically balanced ordinal scales are problematic. In this paper, we discuss a class of models originally developed by Sobel (1995, 1997, 1998) that allows researchers to simultaneously model direction and intensity in ordinal scales with midpoints. Specifically, we parameterize the model as an adjacent category logit model. Numerous parsimonious models may be arrived at that describe trends in the response direction and response intensity. Because the adjacent category logit model is linear in the logits, we estimate the model using log-linear models. We present an application of the models to data on approval ratings of House incumbents. We find that the trends in response directions (the tendency for respondents to evaluate the incumbent favorably or not favorably) increase through the 1980s, peaking in the late Eighties, and are now declining over the 1990s. With regard to response intensity, (that is, the tendency to respond in the extreme categories vs. the moderate categories), we find that intensity increases during most presidential election cycles and vanishes during midterm election years. We argue this finding is related to the different levels of political information citizens are exposed to in presidential vs. midterm election cycles.
Representative Bureaucracy and Distributional Equity: Addressing the Hard Question
Wrinkle, Robert D., Meier, Kenneth J., Polinard, J.L.
Submitted: 1998-07-21
Keywords: representative bureaucracy, organizational outputs, education policy, minorities
Abstract: (click to show/hide) Research on representative bureaucracy has failed to deal with whether or not representative bureaucracies produce minority gains at the expense of nonminorities. Using a pooled time series analysis of 350 school districts over six years, this study examines the relationship between representative bureaucracy and organizational outputs for minorities and nonminorities. Far from finding that representative bureaucracy produces minority gains at the expense of nonminorities, this study finds both minority and nonminority students perform better in the presence of a representative bureaucracy. This finding suggests an alternative hypothesis to guide research, that representative bureaucracies are more effective than their nonrepresentative counterparts.
Wrinkle, Robert D., Meier, Kenneth J., Polinard, J.L.
Submitted: 1998-07-21
Keywords: representative bureaucracy, organizational outputs, education policy, minorities
Abstract: (click to show/hide) Research on representative bureaucracy has failed to deal with whether or not representative bureaucracies produce minority gains at the expense of nonminorities. Using a pooled time series analysis of 350 school districts over six years, this study examines the relationship between representative bureaucracy and organizational outputs for minorities and nonminorities. Far from finding that representative bureaucracy produces minority gains at the expense of nonminorities, this study finds both minority and nonminority students perform better in the presence of a representative bureaucracy. This finding suggests an alternative hypothesis to guide research, that representative bureaucracies are more effective than their nonrepresentative counterparts.
Making the Most of Statistical Analyses: Improving Interpretation and Presentation
King, Gary, Tomz, Michael, Wittenberg, Jason
Submitted: 1998-08-10
Keywords: presentation, interpretation, simulation
Abstract: (click to show/hide) e demonstrate that social scientists rarely take full advantage of the information available in their statistical results. As a consequence, they miss opportunities to present quantities that are of greatest substantive interest for their research, and to express their degree of certainty about these quantities. In this paper, we offer an approach, built on the technique of statistical simulation, to extract the currently overlooked information from any statistical method, no matter how complicated, and to interpret and present it in a reader-friendly manner. Using this technique requires some sophistication, which we try to provide herein, but its application should make the results of quantitative articles more informative and transparent to all. To illustrate our recommendations, we replicate the results of several published works, showing in each case how the authors' own conclusions can be expressed more sharply and informatively, and how our approach reveals important new information about the research questions at hand. We also offer very easy-to-use software that implements our suggestions.
King, Gary, Tomz, Michael, Wittenberg, Jason
Submitted: 1998-08-10
Keywords: presentation, interpretation, simulation
Abstract: (click to show/hide) e demonstrate that social scientists rarely take full advantage of the information available in their statistical results. As a consequence, they miss opportunities to present quantities that are of greatest substantive interest for their research, and to express their degree of certainty about these quantities. In this paper, we offer an approach, built on the technique of statistical simulation, to extract the currently overlooked information from any statistical method, no matter how complicated, and to interpret and present it in a reader-friendly manner. Using this technique requires some sophistication, which we try to provide herein, but its application should make the results of quantitative articles more informative and transparent to all. To illustrate our recommendations, we replicate the results of several published works, showing in each case how the authors' own conclusions can be expressed more sharply and informatively, and how our approach reveals important new information about the research questions at hand. We also offer very easy-to-use software that implements our suggestions.
The Number of Parties: New Evidence from Local Elections
Benoit, Kenneth
Submitted: 1998-08-11
Keywords: electoral systems, regression analysis, Duverger, political parties, Hungary
Abstract: (click to show/hide) Theory: Duverger's ``Law'' concerning the structural and psychological consequences of electoral rules has been much studied in both single cases and in multinational samples, but these suffer from several common theoretical and empirical shortcomings that make their estimates suspect. Besides resort to experimental data, another solution is to select a carefully controlled election dataset where the precise nature of the processes generating the data is understood. Local elections provide a means to control social cleavages as well as to provide a potentially large number of observations. Hypotheses: The size of electoral districts, as well as the type of electoral formula, will influence the number of parties that compete, the concentration of support for these parties, and the number of parties that win seats, even when the elections are confined to one country at the subnational level. In addition, the greater number of observations should provide very precise estimates of these effects. Methods: Regression analysis of district magnitude with an interactive term characterizing rules as proportional or plurality. The data come from 8,377 Hungarian local elected bodies consisting of municipal councils, county councils, town councils, and mayors. Results: The results extend previous research on Duverger's effects, providing more precise estimates that may be compared directly to previous results. In addition, the analysis of rare multi-member plurality elections reveals a counter-intuitive result about candidate and party entry in response to these rules, suggesting several directions for future investigation of MMP rules.
Benoit, Kenneth
Submitted: 1998-08-11
Keywords: electoral systems, regression analysis, Duverger, political parties, Hungary
Abstract: (click to show/hide) Theory: Duverger's ``Law'' concerning the structural and psychological consequences of electoral rules has been much studied in both single cases and in multinational samples, but these suffer from several common theoretical and empirical shortcomings that make their estimates suspect. Besides resort to experimental data, another solution is to select a carefully controlled election dataset where the precise nature of the processes generating the data is understood. Local elections provide a means to control social cleavages as well as to provide a potentially large number of observations. Hypotheses: The size of electoral districts, as well as the type of electoral formula, will influence the number of parties that compete, the concentration of support for these parties, and the number of parties that win seats, even when the elections are confined to one country at the subnational level. In addition, the greater number of observations should provide very precise estimates of these effects. Methods: Regression analysis of district magnitude with an interactive term characterizing rules as proportional or plurality. The data come from 8,377 Hungarian local elected bodies consisting of municipal councils, county councils, town councils, and mayors. Results: The results extend previous research on Duverger's effects, providing more precise estimates that may be compared directly to previous results. In addition, the analysis of rare multi-member plurality elections reveals a counter-intuitive result about candidate and party entry in response to these rules, suggesting several directions for future investigation of MMP rules.
Forecasting Parliamentary Outcomes in Multiparty Elections: Hungary 1998
Benoit, Kenneth
Submitted: 1998-08-16
Keywords: computer simulation, resampling, election forecasting, electoral systems, Hungary
Abstract: (click to show/hide) Forecasting seat outcomes in legislative elections in countries with stable, two-party systems is sufficiently challenging as to have proven elusive for much of democratic experience. Forecasting an election in a relatively new democracy with a fluid multi-party system, therefore, would seem on its face to be a hopeless objective. In this paper I attempt to demonstrate that election forecasting in such an environment is in fact quite feasible, using data from previous elections, opinion poll research, and computer simulation models to predict the outcome of the Hungarian parliamentary elections which took place in May 1998. First, I discuss the general problems with election forecasting, and then outline a strategy for dealing with each. I outline a forecasting method in detail, which I apply to Hungary's case to generate a prediction published in December 1997. The remainder of the paper compares the actual results of the election to the author's forecasts published before the election, identifying areas for improvement in the basic forecasting model but also proving that accurate forecasting of final outcomes in multiparty elections is possible in practice.
Benoit, Kenneth
Submitted: 1998-08-16
Keywords: computer simulation, resampling, election forecasting, electoral systems, Hungary
Abstract: (click to show/hide) Forecasting seat outcomes in legislative elections in countries with stable, two-party systems is sufficiently challenging as to have proven elusive for much of democratic experience. Forecasting an election in a relatively new democracy with a fluid multi-party system, therefore, would seem on its face to be a hopeless objective. In this paper I attempt to demonstrate that election forecasting in such an environment is in fact quite feasible, using data from previous elections, opinion poll research, and computer simulation models to predict the outcome of the Hungarian parliamentary elections which took place in May 1998. First, I discuss the general problems with election forecasting, and then outline a strategy for dealing with each. I outline a forecasting method in detail, which I apply to Hungary's case to generate a prediction published in December 1997. The remainder of the paper compares the actual results of the election to the author's forecasts published before the election, identifying areas for improvement in the basic forecasting model but also proving that accurate forecasting of final outcomes in multiparty elections is possible in practice.
Airline Deregulation: A Financial Markets Perspective on Who Mattered When
Hayes, Jeffery W.
Submitted: 1998-08-17
Keywords: deregulation
Abstract: (click to show/hide) Deregulation has been one of the most important developments in the American administrative state during the post-war era. According to the existing literature, a host of factors played a role in the airline reform process specifically. Almost four decades after the deregulation movement began, the task is to re-examine the important developments and parse the more robust explanations. To what extent did different political, bureaucratic, judicial, and industry actors contribute to airline deregulation during the 1970s? A thorough analysis of this phenomenon entails tests of numerous related theories, including the influence of intellectual ideas, the contagion effects of other deregulation initiatives, and the intriguing but untested ``deregulatory snowball'' theory which posits a specific structural form for the reform process over time. Empirical obstacles, however, have prevented the resolution of many of the provocative research issues surrounding deregulation. Financial market data are an ideal resource to fill this lacuna since reform had significant and negative economic ramifications for the regulated air carriers. A financial markets perspective on regulatory reform yields illuminating results. To begin with, a broad range of institutions contributed to the likelihood of airline deregulation. Congressional influence was especially important. On the other hand, the impact of the CAB was generally more limited than many analysts have believed. Interestingly, the influence of Congress and the CAB were inversely related over time, suggesting the agency's limited role solely as an initial policy entrepreneur while Congress' importance grew as statutory deregulation became the strategic choice of reformers. Additionally, we find evidence that both the politics of ideas and legislative contagion were relevant factors in the reform debate. On the other hand, the deregulatory snowball hypothesis is clearly rejected for the case of the airlines. These findings add a new layer of understanding to the literature on political control of regulatory policy.
Hayes, Jeffery W.
Submitted: 1998-08-17
Keywords: deregulation
Abstract: (click to show/hide) Deregulation has been one of the most important developments in the American administrative state during the post-war era. According to the existing literature, a host of factors played a role in the airline reform process specifically. Almost four decades after the deregulation movement began, the task is to re-examine the important developments and parse the more robust explanations. To what extent did different political, bureaucratic, judicial, and industry actors contribute to airline deregulation during the 1970s? A thorough analysis of this phenomenon entails tests of numerous related theories, including the influence of intellectual ideas, the contagion effects of other deregulation initiatives, and the intriguing but untested ``deregulatory snowball'' theory which posits a specific structural form for the reform process over time. Empirical obstacles, however, have prevented the resolution of many of the provocative research issues surrounding deregulation. Financial market data are an ideal resource to fill this lacuna since reform had significant and negative economic ramifications for the regulated air carriers. A financial markets perspective on regulatory reform yields illuminating results. To begin with, a broad range of institutions contributed to the likelihood of airline deregulation. Congressional influence was especially important. On the other hand, the impact of the CAB was generally more limited than many analysts have believed. Interestingly, the influence of Congress and the CAB were inversely related over time, suggesting the agency's limited role solely as an initial policy entrepreneur while Congress' importance grew as statutory deregulation became the strategic choice of reformers. Additionally, we find evidence that both the politics of ideas and legislative contagion were relevant factors in the reform debate. On the other hand, the deregulatory snowball hypothesis is clearly rejected for the case of the airlines. These findings add a new layer of understanding to the literature on political control of regulatory policy.
Candidate Positioning in U.S. House Elections
Ansolabehere, Stephen D., Snyder, Jr., James M., Steward, III, Charles
Submitted: 1998-08-17
Keywords: spatial, candidate-competition, congressionalelections, representation
Abstract: (click to show/hide) Relying on two new, unique data sets on the policy stances of candidates for the U.S. Congress, we analyze the ideological positioning of House candidates running for office from 1874 to 1996. We argue that throughout this period congressional candidates have primarily espoused the ideology associated with the national party, moderating very little to accommodate local ideological conditions. District-by-district competition exerts some pressure on candidates to fit with their constituents, and there have been times in American history when this pressure has been more acute than others. From the 1940s to 1970s, candidates became much more responsive to district interests, but that degree of responsiveness wanted in the 1980s and 1990s.
Ansolabehere, Stephen D., Snyder, Jr., James M., Steward, III, Charles
Submitted: 1998-08-17
Keywords: spatial, candidate-competition, congressionalelections, representation
Abstract: (click to show/hide) Relying on two new, unique data sets on the policy stances of candidates for the U.S. Congress, we analyze the ideological positioning of House candidates running for office from 1874 to 1996. We argue that throughout this period congressional candidates have primarily espoused the ideology associated with the national party, moderating very little to accommodate local ideological conditions. District-by-district competition exerts some pressure on candidates to fit with their constituents, and there have been times in American history when this pressure has been more acute than others. From the 1940s to 1970s, candidates became much more responsive to district interests, but that degree of responsiveness wanted in the 1980s and 1990s.
Nonnested Model Testing for World Politics: Assessing Binary Choice Models
Clarke, Kevin A.
Submitted: 1998-08-17
Keywords: nonnested hypothesis testing, Cox test, model selection
Abstract: (click to show/hide) he major goal of this project is to introduce and develop a methodology of nonnested hypothesis testing that researchers in world politics will find useful. I make use of both the Cox test for nonnested hypotheses and the Vuong test for nonnested model selection. I argue for a sequential approach where the Vuong test will be used depending upon the outcome of the Cox test. In keeping with the goal of making this methodology useful for world politics research, I discuss both tests in the context of binary choice models, specifically probits. I apply the methodology developed to the problem of testing alternative models of the escalation of great power militarized disputes.
Clarke, Kevin A.
Submitted: 1998-08-17
Keywords: nonnested hypothesis testing, Cox test, model selection
Abstract: (click to show/hide) he major goal of this project is to introduce and develop a methodology of nonnested hypothesis testing that researchers in world politics will find useful. I make use of both the Cox test for nonnested hypotheses and the Vuong test for nonnested model selection. I argue for a sequential approach where the Vuong test will be used depending upon the outcome of the Cox test. In keeping with the goal of making this methodology useful for world politics research, I discuss both tests in the context of binary choice models, specifically probits. I apply the methodology developed to the problem of testing alternative models of the escalation of great power militarized disputes.
Listwise Deletion is Evil: What to Do About Missing Data in Political Science (revised)
King, Gary, Honaker, James, Joseph, Anne, Scheve, Kenneth
Submitted: 1998-08-19
Keywords: missing data, imputation, IP, EM, EMs, EMis, data augmentation, MCMC, importance sampling, item nonresponse
Abstract: (click to show/hide) We propose a remedy to the substantial discrepancy between the way political scientists analyze data with missing values and the recommendations of the statistics community. With a few notable exceptions, statisticians and methodologists have agreed on a widely applicable approach to many missing data problems based on the concept of ``multiple imputation,'' but most researchers in our field and other social sciences still use far inferior methods. Indeed, we demonstrate that the threats to validity from current missing data practices rival the biases from the much better known omitted variable problem. As it turns out, this discrepancy is not entirely our fault, as the computational algorithms used to apply the best multiple imputation models have been slow, difficult to implement, impossible to run with existing commercial statistical packages, and demanding of considerable expertise on the part of the user (even experts disagree on how to use them). In this paper, we adapt an existing algorithm, and use it to implement a general-purpose, multiple imputation model for missing data. This algorithm is between 65 and 726 times faster than the leading method recommended in the statistics literature and is very easy to use. We also quantify the considerable risks of current political science missing data practices, illustrate how to use the new procedure, and demonstrate the advantages of our approach to multiple imputation through simulated data as well as via replications of existing research. We also offer easy-to-use public domain software that implements our approach.
King, Gary, Honaker, James, Joseph, Anne, Scheve, Kenneth
Submitted: 1998-08-19
Keywords: missing data, imputation, IP, EM, EMs, EMis, data augmentation, MCMC, importance sampling, item nonresponse
Abstract: (click to show/hide) We propose a remedy to the substantial discrepancy between the way political scientists analyze data with missing values and the recommendations of the statistics community. With a few notable exceptions, statisticians and methodologists have agreed on a widely applicable approach to many missing data problems based on the concept of ``multiple imputation,'' but most researchers in our field and other social sciences still use far inferior methods. Indeed, we demonstrate that the threats to validity from current missing data practices rival the biases from the much better known omitted variable problem. As it turns out, this discrepancy is not entirely our fault, as the computational algorithms used to apply the best multiple imputation models have been slow, difficult to implement, impossible to run with existing commercial statistical packages, and demanding of considerable expertise on the part of the user (even experts disagree on how to use them). In this paper, we adapt an existing algorithm, and use it to implement a general-purpose, multiple imputation model for missing data. This algorithm is between 65 and 726 times faster than the leading method recommended in the statistics literature and is very easy to use. We also quantify the considerable risks of current political science missing data practices, illustrate how to use the new procedure, and demonstrate the advantages of our approach to multiple imputation through simulated data as well as via replications of existing research. We also offer easy-to-use public domain software that implements our approach.
Rivalry, Reciprocity, and the Dynamics of Presidential-Congressional Institution Building
Krause, George
Submitted: 1998-08-20
Keywords: Institution Building, Presidential-Congressional Relations, Prisoner's Dilemma, Nonmyopic Equilibria, Theory of Moves, Johansen Cointegration Procedure, FIML, Vector Error Correction Mechanisms (VECMs), Innovation Accounting
Abstract: (click to show/hide) A central feature of the development of the presidential and congressional branches has been the process of institution building. This phenomenon represents the size and scope of a branch's formal institutional apparatus that is reflected by the resources it utilizes for operational and functional purposes. In this study, a simple dynamic Prisoner's Dilemma game-theoretic model, based on the Theory of Moves (Brams 1994), is set forth to explain this process. This theoretical model produces two Nonmyopic Equilibria (NMEs): (1) a Contractionary Equilibrium where both the president and Congress expend fewer resources; and (2) an Expansionary Equilibrium where each institution expends greater resources. The theoretical predictions derived from this positive model suggest that variations in the institutional expenditures by each branch will exhibit a stable long-run equilibrium relationship that is consistent with these NME's. Using constant-dollar annual data on Executive Office of the President and Legislative Branch expenditures for the 1939-1997 period, a Vector Error Correction Mechanism (VECM) model, that is derived from the game-theoretic model noted above, is employed to empirically account for both short-run and long-run movements as well as long-run equilibrium relations. The statistical evidence supports the predictions of the theoretical model. Specifically, the historical evolution of presidential and congressional institution building represents a conflict situation where neither institution has a permanent advantage over the other due to its equal power and farsightedly rational behavior. Contrary to existing research on this topic, the empirical findings reveal that both presidential and congressional efforts at institution building do not just emanate from within each respective branch, but instead are very responsive to one another with respect to these activities. This, in turn, suggests that causal (temporal) sequence of institution building is in stark contrast from the conventional wisdom of an "opportunistic" presidency that exploits Congress.
Krause, George
Submitted: 1998-08-20
Keywords: Institution Building, Presidential-Congressional Relations, Prisoner's Dilemma, Nonmyopic Equilibria, Theory of Moves, Johansen Cointegration Procedure, FIML, Vector Error Correction Mechanisms (VECMs), Innovation Accounting
Abstract: (click to show/hide) A central feature of the development of the presidential and congressional branches has been the process of institution building. This phenomenon represents the size and scope of a branch's formal institutional apparatus that is reflected by the resources it utilizes for operational and functional purposes. In this study, a simple dynamic Prisoner's Dilemma game-theoretic model, based on the Theory of Moves (Brams 1994), is set forth to explain this process. This theoretical model produces two Nonmyopic Equilibria (NMEs): (1) a Contractionary Equilibrium where both the president and Congress expend fewer resources; and (2) an Expansionary Equilibrium where each institution expends greater resources. The theoretical predictions derived from this positive model suggest that variations in the institutional expenditures by each branch will exhibit a stable long-run equilibrium relationship that is consistent with these NME's. Using constant-dollar annual data on Executive Office of the President and Legislative Branch expenditures for the 1939-1997 period, a Vector Error Correction Mechanism (VECM) model, that is derived from the game-theoretic model noted above, is employed to empirically account for both short-run and long-run movements as well as long-run equilibrium relations. The statistical evidence supports the predictions of the theoretical model. Specifically, the historical evolution of presidential and congressional institution building represents a conflict situation where neither institution has a permanent advantage over the other due to its equal power and farsightedly rational behavior. Contrary to existing research on this topic, the empirical findings reveal that both presidential and congressional efforts at institution building do not just emanate from within each respective branch, but instead are very responsive to one another with respect to these activities. This, in turn, suggests that causal (temporal) sequence of institution building is in stark contrast from the conventional wisdom of an "opportunistic" presidency that exploits Congress.
If the Assumption Fits...: A Comment on the King Ecological Inference Solution
Cho, Wendy K. T.
Submitted: 1998-08-20
Keywords: ecological inference
Abstract: (click to show/hide) I examine a recently proposed solution to the ecological inference problem (King 1997). It is asserted that the proposed model is able to reconstruct individual-level behavior from aggregate data. I discuss in detail both the benefits and limitations of this model. The assumptions of the basic model are often inappropriate for instances of aggregate data. The extended version of the model is able to correct for some of these limitations. However, it is difficult in most cases to apply the extended model properly.
Cho, Wendy K. T.
Submitted: 1998-08-20
Keywords: ecological inference
Abstract: (click to show/hide) I examine a recently proposed solution to the ecological inference problem (King 1997). It is asserted that the proposed model is able to reconstruct individual-level behavior from aggregate data. I discuss in detail both the benefits and limitations of this model. The assumptions of the basic model are often inappropriate for instances of aggregate data. The extended version of the model is able to correct for some of these limitations. However, it is difficult in most cases to apply the extended model properly.
Intrainstitutional Mobility in the Postreform House of Representatives
Wawro, Gregory
Submitted: 1998-08-26
Keywords: legislative entrepreneurship, legislative organization, maximum likelihood methods, mobility analysis, career concerns, vacancy competition
Abstract: (click to show/hide) Theory: When deciding whom to promote to prestigious positions within the House, members will favor those individuals who are the most likely to use the resources associated with prestigious positions to produce legislation when there is substantial demand for it. Members will select those individuals who have demonstrated a propensity for engaging in legislative entrepreneurship because they are the most qualified in this regard. Hypothesis: "The job ladders hypothesis": Members who engage in legislative entrepreneurship are more likely to move up the job ladder to prestigious positions within the committee and party hierarchies in the House. Method: I develop measures of legislative entrepreneurship using data on the characteristics of bills sponsored by members and members' testimony before committees. I develop a statistical model that addresses the problems of analyzing intrainstitutional mobility and the problems with assessing entrepreneurial ability. With this model I perform a multivariate analysis to assess the effects of legislative entrepreneurship while accounting for other variables that previous studies have found to affect intrainstitutional mobility. Results: Engaging in legislative entrepreneurship increases the probability that members of the majority party will advance to full committee, subcommittee, and party leadership positions.
Wawro, Gregory
Submitted: 1998-08-26
Keywords: legislative entrepreneurship, legislative organization, maximum likelihood methods, mobility analysis, career concerns, vacancy competition
Abstract: (click to show/hide) Theory: When deciding whom to promote to prestigious positions within the House, members will favor those individuals who are the most likely to use the resources associated with prestigious positions to produce legislation when there is substantial demand for it. Members will select those individuals who have demonstrated a propensity for engaging in legislative entrepreneurship because they are the most qualified in this regard. Hypothesis: "The job ladders hypothesis": Members who engage in legislative entrepreneurship are more likely to move up the job ladder to prestigious positions within the committee and party hierarchies in the House. Method: I develop measures of legislative entrepreneurship using data on the characteristics of bills sponsored by members and members' testimony before committees. I develop a statistical model that addresses the problems of analyzing intrainstitutional mobility and the problems with assessing entrepreneurial ability. With this model I perform a multivariate analysis to assess the effects of legislative entrepreneurship while accounting for other variables that previous studies have found to affect intrainstitutional mobility. Results: Engaging in legislative entrepreneurship increases the probability that members of the majority party will advance to full committee, subcommittee, and party leadership positions.
Direction and Intensity of Russian Macroeconomic Evaluations
Jones, Bradford S., Willerton, John P., Sobel, Michael E.
Submitted: 1998-08-30
Keywords: Russia, public opinion, log linear models
Abstract: (click to show/hide) The Russian macroeconomy has exhibited volatility since the transformation from the Soviet Union to the Russian Federation. Much is known about the Russian public opinion climate during the end of the Soviet era and the beginning of the Russian Federation era; however, less well understood is the nature of Russians' macroeconomic evaluations during this on-going transformation. In this paper, we analyze Russians' assessments of the macroeconomy using Russian public opinion data asking respondents to assess the Russian national economy. We establish four testable hypotheses. First, we hypothesize that the direction of Russian opinion will be asymmetrically more negative than positive across all periods in the study. Second, we hypothesize that economic assessments will vary by residential region. Specifically, we contend the response distribution for respondents from Moscow and St. Petersburg (MSP) will differ from respondents from other residential regions. Third (and related to the second), we hypothesize that the response distributions for MSP respondents will be temporally heterogenous while the response distribution for respondents outside MSP will be temporally homogenous. Fourth, we hypothesize that despite the poor performance of the economy during the Russian Federation transition, Russian public opinion will not exhibit extreme negativity in macroeconomic evaluations. Using published survey data collected from the bi -monthly\textsl{Russian Public Opinion Monitor} conducted by the Russian Center for Public Opinion Research (VCIOM), for the period January 1994 to July 1996, we examine both the direction and intensity of Russian opinion toward the state of the national economy by estimating the distribution on the response variable using an adjacent category logit model (Jones and Sobel 1998, Sobel 1995, 1997, 1998). From our analysis, we find first that the direction of Russians' evaluation of the macroeconomy is consistently negative rather than positive---a finding that corroborates extant research; however, the directional nature of economic assessments displays significant residential variation between MSP and the rest of the country. Second, we find significant residential variation in economic assessments. Specifically, the response distribution for MSP respondents can be distinguished from the response distribution from respondents in other residential regions, and also, the response distribution for MSP respondents displays considerable temporal heterogeneity. We argue this variability tends to follow changes in the macroeconomic and political environments. Third, we do not find support for the hypothesis of temporal homogeneity in the response distribution for respondents outside of MSP. Nevertheless, residents in other cities and in rural regions seem not to be as responsive to macroeconomic changes over the period, thus eliciting milder temporal variability than MSP respondents. Fourth, we find that in terms of the response distribution, the intensity of Russian pessimism (or optimism) is \textsl{not} extreme.
Jones, Bradford S., Willerton, John P., Sobel, Michael E.
Submitted: 1998-08-30
Keywords: Russia, public opinion, log linear models
Abstract: (click to show/hide) The Russian macroeconomy has exhibited volatility since the transformation from the Soviet Union to the Russian Federation. Much is known about the Russian public opinion climate during the end of the Soviet era and the beginning of the Russian Federation era; however, less well understood is the nature of Russians' macroeconomic evaluations during this on-going transformation. In this paper, we analyze Russians' assessments of the macroeconomy using Russian public opinion data asking respondents to assess the Russian national economy. We establish four testable hypotheses. First, we hypothesize that the direction of Russian opinion will be asymmetrically more negative than positive across all periods in the study. Second, we hypothesize that economic assessments will vary by residential region. Specifically, we contend the response distribution for respondents from Moscow and St. Petersburg (MSP) will differ from respondents from other residential regions. Third (and related to the second), we hypothesize that the response distributions for MSP respondents will be temporally heterogenous while the response distribution for respondents outside MSP will be temporally homogenous. Fourth, we hypothesize that despite the poor performance of the economy during the Russian Federation transition, Russian public opinion will not exhibit extreme negativity in macroeconomic evaluations. Using published survey data collected from the bi -monthly\textsl{Russian Public Opinion Monitor} conducted by the Russian Center for Public Opinion Research (VCIOM), for the period January 1994 to July 1996, we examine both the direction and intensity of Russian opinion toward the state of the national economy by estimating the distribution on the response variable using an adjacent category logit model (Jones and Sobel 1998, Sobel 1995, 1997, 1998). From our analysis, we find first that the direction of Russians' evaluation of the macroeconomy is consistently negative rather than positive---a finding that corroborates extant research; however, the directional nature of economic assessments displays significant residential variation between MSP and the rest of the country. Second, we find significant residential variation in economic assessments. Specifically, the response distribution for MSP respondents can be distinguished from the response distribution from respondents in other residential regions, and also, the response distribution for MSP respondents displays considerable temporal heterogeneity. We argue this variability tends to follow changes in the macroeconomic and political environments. Third, we do not find support for the hypothesis of temporal homogeneity in the response distribution for respondents outside of MSP. Nevertheless, residents in other cities and in rural regions seem not to be as responsive to macroeconomic changes over the period, thus eliciting milder temporal variability than MSP respondents. Fourth, we find that in terms of the response distribution, the intensity of Russian pessimism (or optimism) is \textsl{not} extreme.
Direction and Intensity of Russian Macroeconomic Evaluations
Jones, Bradford S., Willerton, John P., Sobel, Michael E.
Submitted: 1998-08-30
Keywords: Russia, public opinion, log linear models
Abstract: (click to show/hide) The Russian macroeconomy has exhibited volatility since the transformation from the Soviet Union to the Russian Federation. Much is known about the Russian public opinion climate during the end of the Soviet era and the beginning of the Russian Federation era; however, less well understood is the nature of Russians' macroeconomic evaluations during this on-going transformation. In this paper, we analyze Russians' assessments of the macroeconomy using Russian public opinion data asking respondents to assess the Russian national economy. We establish four testable hypotheses. First, we hypothesize that the direction of Russian opinion will be asymmetrically more negative than positive across all periods in the study. Second, we hypothesize that economic assessments will vary by residential region. Specifically, we contend the response distribution for respondents from Moscow and St. Petersburg (MSP) will differ from respondents from other residential regions. Third (and related to the second), we hypothesize that the response distributions for MSP respondents will be temporally heterogenous while the response distribution for respondents outside MSP will be temporally homogenous. Fourth, we hypothesize that despite the poor performance of the economy during the Russian Federation transition, Russian public opinion will not exhibit extreme negativity in macroeconomic evaluations. Using published survey data collected from the bi -monthly\textsl{Russian Public Opinion Monitor} conducted by the Russian Center for Public Opinion Research (VCIOM), for the period January 1994 to July 1996, we examine both the direction and intensity of Russian opinion toward the state of the national economy by estimating the distribution on the response variable using an adjacent category logit model (Jones and Sobel 1998, Sobel 1995, 1997, 1998). From our analysis, we find first that the direction of Russians' evaluation of the macroeconomy is consistently negative rather than positive---a finding that corroborates extant research; however, the directional nature of economic assessments displays significant residential variation between MSP and the rest of the country. Second, we find significant residential variation in economic assessments. Specifically, the response distribution for MSP respondents can be distinguished from the response distribution from respondents in other residential regions, and also, the response distribution for MSP respondents displays considerable temporal heterogeneity. We argue this variability tends to follow changes in the macroeconomic and political environments. Third, we do not find support for the hypothesis of temporal homogeneity in the response distribution for respondents outside of MSP. Nevertheless, residents in other cities and in rural regions seem not to be as responsive to macroeconomic changes over the period, thus eliciting milder temporal variability than MSP respondents. Fourth, we find that in terms of the response distribution, the intensity of Russian pessimism (or optimism) is \textsl{not} extreme.
Jones, Bradford S., Willerton, John P., Sobel, Michael E.
Submitted: 1998-08-30
Keywords: Russia, public opinion, log linear models
Abstract: (click to show/hide) The Russian macroeconomy has exhibited volatility since the transformation from the Soviet Union to the Russian Federation. Much is known about the Russian public opinion climate during the end of the Soviet era and the beginning of the Russian Federation era; however, less well understood is the nature of Russians' macroeconomic evaluations during this on-going transformation. In this paper, we analyze Russians' assessments of the macroeconomy using Russian public opinion data asking respondents to assess the Russian national economy. We establish four testable hypotheses. First, we hypothesize that the direction of Russian opinion will be asymmetrically more negative than positive across all periods in the study. Second, we hypothesize that economic assessments will vary by residential region. Specifically, we contend the response distribution for respondents from Moscow and St. Petersburg (MSP) will differ from respondents from other residential regions. Third (and related to the second), we hypothesize that the response distributions for MSP respondents will be temporally heterogenous while the response distribution for respondents outside MSP will be temporally homogenous. Fourth, we hypothesize that despite the poor performance of the economy during the Russian Federation transition, Russian public opinion will not exhibit extreme negativity in macroeconomic evaluations. Using published survey data collected from the bi -monthly\textsl{Russian Public Opinion Monitor} conducted by the Russian Center for Public Opinion Research (VCIOM), for the period January 1994 to July 1996, we examine both the direction and intensity of Russian opinion toward the state of the national economy by estimating the distribution on the response variable using an adjacent category logit model (Jones and Sobel 1998, Sobel 1995, 1997, 1998). From our analysis, we find first that the direction of Russians' evaluation of the macroeconomy is consistently negative rather than positive---a finding that corroborates extant research; however, the directional nature of economic assessments displays significant residential variation between MSP and the rest of the country. Second, we find significant residential variation in economic assessments. Specifically, the response distribution for MSP respondents can be distinguished from the response distribution from respondents in other residential regions, and also, the response distribution for MSP respondents displays considerable temporal heterogeneity. We argue this variability tends to follow changes in the macroeconomic and political environments. Third, we do not find support for the hypothesis of temporal homogeneity in the response distribution for respondents outside of MSP. Nevertheless, residents in other cities and in rural regions seem not to be as responsive to macroeconomic changes over the period, thus eliciting milder temporal variability than MSP respondents. Fourth, we find that in terms of the response distribution, the intensity of Russian pessimism (or optimism) is \textsl{not} extreme.
Estimating voter preference distributions from individual-level voting data (with application to split-ticket voting
Lewis, Jeffrey B.
Submitted: 1998-09-15
Keywords: split ticket voting, ideal point estimation, spatial voting models, EM algorithm
Abstract: (click to show/hide) In the last decade a great deal of progress has been made in estimating spatial models of legislative roll-call voting. There are now several well-known and effective methods of estimating the ideal points of legislators from their roll-call votes. Similar progress has not been made in the empirical modeling of the distribution of preferences in the electorate. Progress has been slower, not because the question is less important, but because of limitations of data and a lack of tractable methods. In this paper, I present a method for inferring the distribution of voter ideal points on a single dimension from individual-level voting returns on ballot propositions. The statistical model and estimation technique draw heavily on the psychometric literature on test taking and, in particular, on the work of Bock and Aitkin (1981}. The method yields semi-parametric estimates of the distribution of voters along an unobserved spatial dimension. The model is applied to data from the 1992 general election in Los Angeles County. I present the distribution of voter ideal points of each of 17 Congressional districts. Finally, I consider the issue of split-ticket voting estimating for two Congressional districts the distribution of voters that split their tickets and of those that did not.
Lewis, Jeffrey B.
Submitted: 1998-09-15
Keywords: split ticket voting, ideal point estimation, spatial voting models, EM algorithm
Abstract: (click to show/hide) In the last decade a great deal of progress has been made in estimating spatial models of legislative roll-call voting. There are now several well-known and effective methods of estimating the ideal points of legislators from their roll-call votes. Similar progress has not been made in the empirical modeling of the distribution of preferences in the electorate. Progress has been slower, not because the question is less important, but because of limitations of data and a lack of tractable methods. In this paper, I present a method for inferring the distribution of voter ideal points on a single dimension from individual-level voting returns on ballot propositions. The statistical model and estimation technique draw heavily on the psychometric literature on test taking and, in particular, on the work of Bock and Aitkin (1981}. The method yields semi-parametric estimates of the distribution of voters along an unobserved spatial dimension. The model is applied to data from the 1992 general election in Los Angeles County. I present the distribution of voter ideal points of each of 17 Congressional districts. Finally, I consider the issue of split-ticket voting estimating for two Congressional districts the distribution of voters that split their tickets and of those that did not.
Measuring the Electoral and Policy Impact of Majority-Minority Voting Districts: Candidates of Choice, Equal Opportunity, and Representation
Epstein, David, O'Halloran, Sharyn
Submitted: 1998-09-15
Keywords: voting rights act, ecological regression
Abstract: (click to show/hide) The Voting Rights Act guarantees minority voters an "equal opportunity to elect the candidate of their choice." Yet the implementation of this requirement is beset with technical difficulties: first, current case law provides no clear definition as to who qualifies as a candidate of choice of the minority community; second, traditional techniques for estimating equal opportunity rely heavily on ecological regression, which is prone to statistical bias; and third, no attempt is made to systematically evaluate the impact of alternative districting strategies on the substantive representation of minority interests, rather than just descriptive representation. This paper offers an alternative approach to majority-minority districting that 1) explicitly defines the term "candidate of choice;" 2) determines the point of equal opportunity without relying on ecological regression; and 3) estimates the expected impact of competing districting schemes on substantive representation. It then applies this technique to a set of alternative districting plans for the South Carolina State Senate.
Epstein, David, O'Halloran, Sharyn
Submitted: 1998-09-15
Keywords: voting rights act, ecological regression
Abstract: (click to show/hide) The Voting Rights Act guarantees minority voters an "equal opportunity to elect the candidate of their choice." Yet the implementation of this requirement is beset with technical difficulties: first, current case law provides no clear definition as to who qualifies as a candidate of choice of the minority community; second, traditional techniques for estimating equal opportunity rely heavily on ecological regression, which is prone to statistical bias; and third, no attempt is made to systematically evaluate the impact of alternative districting strategies on the substantive representation of minority interests, rather than just descriptive representation. This paper offers an alternative approach to majority-minority districting that 1) explicitly defines the term "candidate of choice;" 2) determines the point of equal opportunity without relying on ecological regression; and 3) estimates the expected impact of competing districting schemes on substantive representation. It then applies this technique to a set of alternative districting plans for the South Carolina State Senate.
Negotiating Coalitions
Bottom, William P., Miller, Gary J., Holloway, James, McClurg, Scott D.
Submitted: 1998-09-15
Keywords: Game theory, Experimental Design, Coalition Formation, Negotiation, Risk
Abstract: (click to show/hide) Game theory's best efforts have done little but verify the undecidability of coalitional problems. The typical solution concept specifies the hypothesized distribution for each of several viable coalition structures--but cannot choose among the coalition structures. For example, the bargaining set presumes that bargaining proceeds by objection and counter-objection until potential coalition members are indifferent between the coalitions that they pivot between. Thus, the bargaining set makes a clear distributional hypothesis, but thereby gives up any leverage on which coalition will occur. In this paper, we explore how risk preferences and the nature of coalitional goods influence the coalition-building process. We test a variety of potential explanations with data collected in an experimental setting. Foremost among our conclusions is that the coalitions which form among inexperienced subjects are affected by their risk preferences. We further find that this effect disappears among experienced subjects. We conclude the paper by discussing some of the explanations for and questions stemming from our results.
Bottom, William P., Miller, Gary J., Holloway, James, McClurg, Scott D.
Submitted: 1998-09-15
Keywords: Game theory, Experimental Design, Coalition Formation, Negotiation, Risk
Abstract: (click to show/hide) Game theory's best efforts have done little but verify the undecidability of coalitional problems. The typical solution concept specifies the hypothesized distribution for each of several viable coalition structures--but cannot choose among the coalition structures. For example, the bargaining set presumes that bargaining proceeds by objection and counter-objection until potential coalition members are indifferent between the coalitions that they pivot between. Thus, the bargaining set makes a clear distributional hypothesis, but thereby gives up any leverage on which coalition will occur. In this paper, we explore how risk preferences and the nature of coalitional goods influence the coalition-building process. We test a variety of potential explanations with data collected in an experimental setting. Foremost among our conclusions is that the coalitions which form among inexperienced subjects are affected by their risk preferences. We further find that this effect disappears among experienced subjects. We conclude the paper by discussing some of the explanations for and questions stemming from our results.
Economic Performance, Job Insecurity, and Electoral Choice
Lacy, Dean, Mughan, Anthony
Submitted: 1998-09-17
Keywords: economic voting, economic insecurity, Perot, turnout, multinomial probit, 1996 election
Abstract: (click to show/hide) The mass political economy literature concentrates on egocentric and sociotropic evaluations of short-term economic performance. Scant attention is paid to other economic concerns people may have. In a neo-liberal economic climate characterized by a downsized labor market and the retrenchment of government welfare entitlements, one such widely-publicized concern is job insecurity. We show that job insecurity is a novel form of discontent that is independent of the retrospective evaluations of short-term performance that are the stuff of the mainstream mass political economy literature. At the same time, the political effects of job insecurity are distinctive. In a multinomial probit model of electoral choice in the 1996 U.S. presidential election, job insecurity is associated with support for the third-party candidate, Ross Perot, but, contrary to conventional wisdom, has no implications for turnout. Traditional retrospective evaluations of economic performance explain the major-party vote and abstention.
Lacy, Dean, Mughan, Anthony
Submitted: 1998-09-17
Keywords: economic voting, economic insecurity, Perot, turnout, multinomial probit, 1996 election
Abstract: (click to show/hide) The mass political economy literature concentrates on egocentric and sociotropic evaluations of short-term economic performance. Scant attention is paid to other economic concerns people may have. In a neo-liberal economic climate characterized by a downsized labor market and the retrenchment of government welfare entitlements, one such widely-publicized concern is job insecurity. We show that job insecurity is a novel form of discontent that is independent of the retrospective evaluations of short-term performance that are the stuff of the mainstream mass political economy literature. At the same time, the political effects of job insecurity are distinctive. In a multinomial probit model of electoral choice in the 1996 U.S. presidential election, job insecurity is associated with support for the third-party candidate, Ross Perot, but, contrary to conventional wisdom, has no implications for turnout. Traditional retrospective evaluations of economic performance explain the major-party vote and abstention.
