Presidential Approval: the case of George W. Bush

To read the full-text of this research, you can request a copy directly from the authors.


Abstract We use a Bayesian dynamic linear model to track approval for George W. Bush over time. Our analysis deals with several issues that have been usually addressed separately in the extant literature. First, our analysis uses polling data collected at a higher frequency than is typical, using over 1,100 published national polls, and data on macro-economic conditions collected at the weekly level. By combining this much poll information, we are much better poised to examine the public’s reactions to events over shorter time scales than can the typical analysis of approval that utilizes monthly or quarterly approval. Second, our statistical modeling explicitly deals with the sampling error of these polls, as well as the possibility of bias in the polls due to house eects.,Indeed, quite aside from the question of “what drives approval?”, there is considerable interest in the extent to which polling organizations systematically diverge from one another in assessing approval for the president. These bias parameters are not only necessary parts of any realistic model of approval that utilizes data from multiple polling organizations, but easily estimated via the Bayesian dynamic linear model. The determinants of presidential approval is one of the chestnut problems in the study of American politics. For political methodologists it is also one of the chestnut quantitative issues, and clearly the leading example for our study of time series methods, dating back at least to Hibbs (1973). While the earliest studies were for the US, the industry has branched out to most democ- racies where a polling industry provides repeated measures of approval. If we take the subfield as starting with Mueller’s (1973) work, the original question related to the impact do exoge- nous events (war for Mueller) on presidential approval. While the impact of events has been of some continued interest (particularly in the work of MacKuen 1983), most work has been on the more general determinants of approval, and particularly the economic determinants of approval. Methodologically, most analyses have used Gallup polls to measure approval, with almost all analysts treating the more or less monthly polls as monthly data. With very few exceptions analysts have ignored measurement error in this monthly data. Also, with few exceptions, analysts have not combined information from dierent,polling houses, nor have they treated the unit of analysis as anything other than the month. This makes sense in the context of

No full-text available

Request Full-text Paper PDF

To read the full-text of this research,
you can request a copy directly from the authors.

... Variational approximations are broadly applicable and useful for any model where standard sampling based approaches to posterior approximation are infeasible or severely limiting. This includes many statistical models for political texts (Quinn et al., 2010; Proksch and Slapin, 2009), the measurement of preferences in both political institutions and the public (Clinton, Jackman and Rivers., 2004; Levendusky, Pope and Jackman, 2008), and the estimation of complex models that vary over time and space (Hoff and Ward, 2004; Beck, Jackman and Rosenthal, 2006; Shor et al., 2007). ...
Markov chain Monte Carlo (MCMC) methods have facilitated an explosion of interest in Bayesian methods. MCMC is an incredibly useful and important tool but can face difficulties when used to estimate complex posteriors or models applied to large data sets. In this paper, we show how a recently developed tool in computer science for fitting Bayesian models, variational approximations, can be used to facilitate the application of Bayesian models to political science data. Variational approximations are often much faster than MCMC for fully Bayesian inference and in some instances facilitate the estimation of models that would be otherwise impossible to estimate. As a deterministic posterior approximation method, variational approximations are guaranteed to converge and convergence is easily assessed. But variational approximations do have some limitations, which we detail below. Therefore, variational approximations are best suited to problems when fully Bayesian inference would otherwise be impossible. Through a series of examples, we demonstrate how variational approximations are useful for a variety of political science research. This includes models to describe legislative voting blocs and statistical models for political texts. The code that implements the models in this paper is available in the supplementary material.
... The polls were conducted by more than 140 different polling or news organizations, so it is impossible to estimate whether a particular polling firm is more or less accurate for black and female candidates. For more on estimating house effects, see Beck, Jackman, and Rosenthal (2006). 7 Among others, this criterion removes Barack Obama's 2004 Illinois Senate race against Alan Keyes because both are black and Susan Collins' 2002 Senate race against Chellie Pingree because both are women. ...
The 2008 election renewed interest in the Wilder or Bradley effect, the gap between the share of survey respondents expressing support for a candidate and the candidate’s vote share. Using new data from 180 gubernatorial and Senate elections from 1989 to 2006, this paper presents the first large-sample test of the Wilder effect. It demonstrates a significant Wilder effect only through the early 1990s, when Wilder himself was Governor of Virginia. Evidence from the 2008 presidential election reinforces this claim. Although the same mechanisms could affect female candidates, this paper finds no such effect at any point in time. It also shows how polls’ overestimation of front-runners’ support can exaggerate estimates of the Wilder effect. Together, these results accord with theories emphasizing how short-term changes in the political context influence the role of race in statewide elections. The Wilder effect is the product of racial attitudes in specific political contexts, not a more general response to underrepresented groups.
... The " memory " in the process immediately post-9/11 was rather short, whereas during more mundane times, the memory may be longer. Nathaniel Beck and colleagues, using a Kalman filter, insert a jump at 9/11, which does leads to a far better fit (Beck, Jackman, and Rosenthal 2006). Inserting the jump at 9/11, though, assumes the answer to the question we are asking. ...
Full-text available
Many social processes are stable and smooth in general, with discrete jumps. We develop a sequential segmentation spline method that can identify both the location and the number of discontinuities in a series of observations with a time component, while fitting a smooth spline between jumps, using a modified Bayesian Information Criterion statistic as a stopping rule. We explore the method in a large-n, unbalanced panel setting with George W. Bush's approval data, a small-n time series with median DW-NOMINATE scores for each Congress over time, and a series of simulations. We compare the method to several extant smoothers, and the method performs favorably in terms of visual inspection, residual properties, and event detection. Finally, we discuss extensions of the method. © The Author 2009. Published by Oxford University Press on behalf of the Society for Political Methodology. All rights reserved. For Permissions, please email: [email protected] /* */
This study investigates the transition of postwar Japanese public opinion on constitutional revision by comprehensively analyzing the results of opinion polls on the constitution conducted by seven major polling institutes. A “pooling the polls” method, derived from a dynamic linear model, allows us to estimate the latent approval and disapproval rates for constitutional revision in consideration of variations in question items and wording, house and mode effects, and sampling errors. The results imply that, while voters perceived “full revision” of the constitution as a political issue in the 1950s, they came to consider Article 9 as the most important problem by far after the 1960s. We also find that citizens became clearly aware of issues other than Article 9 in the 1990s and 2000s, and that they focused on the problem of Article 9 again, following the Koizumi administration. Moreover, the effects of question items and wording suggest that voters regard the second (but not the first) paragraph of Article 9 as a political issue, and that priming respondents with questions containing the word “war” increases the disapproval rate for the revisions in Article 9.
Scholars have recently argued that prime ministers reshuffle their cabinets strategically. Although some scholars assume that cabinet reshuffles help prime ministers increase their government's popularity, this assumption has not been tested formally because of the endogeneity problem. In Japan, polling firms sometimes provide respondents with cues about a reshuffle when asking about cabinet approval following reshuffles, while others do not. I utilized this convention in the Japanese media to test the assumption that reshuffles increase cabinet approval ratings. Applying a dynamic linear model to pooled poll data from 2001 to 2015, I achieved high internal, external, and ecological validity. The analyses show that cues about reshuffles increase cabinet approval ratings by 2.4 percentage points on average, and the credible interval of the effect does not include zero. This result reinforces the findings of previous research on the theory of cabinet management. © The Author(s) 2018. Published by Oxford University Press on behalf of the American Association for Public Opinion Research.
The Kalman filter is useful to estimate dynamic models via maximum likelihood. To do this the model must be set up in state space form. This article shows how various models of interest can be set up in that form. Models considered are Auto Regressive-Moving Average (ARMA) models with measurement error and dynamic factor models. The filter is used to estimate models of presidential approval. A test of rational expectations in approval shows the hypothesis not to hold. The filter is also used to deal with missing approval data and to study whether interpolation of missing data is an adequate technique. Finally, a dynamic factor analysis of government entrepreneurial activity is performed. Appendices go through the mathematical details of the filter and show how to implement it in the computer language GAUSS.
This article reports on an examination of the movement of presidential popularity for the years 1963-1980. In particular, it compares public responsiveness to changing economic conditions, on the one hand, with responsiveness to dramatic events of a high political nature, on the other. The evidence indicates that each of these categories of influence plays an equal part. In addition, the analysis measures the temporal dimension of the various substantive influences on presidential popularity. The data suggest that the identifiable influences have distinctive dynamics, a fact leading to speculation that those influences may be of different characters.
Poll results vary over the course of a campaign election and across polling organisations, making it difficult to track genuine changes in voter support. I present a statistical model that tracks changes in voter support over time by pooling the polls, and corrects for variation across polling organisations due to biases known as ‘house effects’. The result is a less biased and more precise estimate of vote intentions than is possible from any one poll alone. I use five series of polls fielded over the 2004 Australian federal election campaign (ACNielsen, the ANU/ninemsn online poll, Galaxy, Newspoll, and Roy Morgan) to generate daily estimates of the Coalition's share of two-party preferred (2PP) and first preference vote intentions. Over the course of the campaign there is about a 4 percentage point swing to the Coalition in first preference vote share (and a smaller swing in 2PP terms), that begins prior to the formal announcement of the election, but is complete shortly after the leader debates. The ANU/ninemsn online poll and Morgan are found to have large and statistically significant biases, while, generally, the three phone polls have small and/or statistically insignificant biases, with ACNielsen and (in particular) Galaxy performing quite well in 2004.* An earlier version of this paper was prepared for the annual meeting of the Australasian Political Studies Association, University of Adelaide, 29 September–1 October 2004.
Bayesian simulation is increasingly exploited in the social sciences for estimation and inference of model parameters. But an especially useful (if often overlooked) feature of Bayesian simulation is that it can be used to estimate any function of model parameters, including “auxiliary” quantities such as goodness-of-fit statistics, predicted values, and residuals. Bayesian simulation treats these quantities as if they were missing data, sampling from their implied posterior densities. Exploiting this principle also lets researchers estimate models via Bayesian simulation where maximum-likelihood estimation would be intractable. Bayesian simulation thus provides a unified solution for quantitative social science. I elaborate these ideas in a variety of contexts: these include generalized linear models for binary responses using data on bill cosponsorship recently reanalyzed in Political Analysis, item—response models for the measurement of respondent's levels of political information in public opinion surveys, the estimation and analysis of legislators' ideal points from roll-call data, and outlier-resistant regression estimates of incumbency advantage in U.S. Congressional elections
Bayesian statistics have made great strides in recent years, developing a class of methods for estimation and inference via stochastic simulation known as Markov Chain Monte Carlo (MCMC) methods. MCMC constitutes a revolution in statistical practice with effects beginning to be felt in the social sciences: models long consigned to the "too hard" basket are now within reach of quantitative researchers. I review the statistical pedigree of MCMC and the underlying statistical concepts. I demonstrate some of the strengths and weaknesses of MCMC and offer practical suggestions for using MCMC in social-science settings. Simple, illustrative examples include a probit model of voter turnout and a linear regression for time-series data with autoregressive disturbances. I conclude with a more challenging application, a multinomial probit model, to showcase the power of MCMC methods.
We show how to use the Gibbs sampler to carry out Bayesian inference on a linear state space model with errors that are a mixture of normals and coefficients that can switch over time. Our approach simultaneously generates the whole of the state vector given the mixture and coefficient indicator variables and simultaneously generates all the indicator variables conditional on the state vectors. The states are generated efficiently using the Kalman filter. We illustrate our approach by several examples and empirically compare its performance to another Gibbs sampler where the states are generated one at a time. The empirical results suggest that our approach is both practical to implement and dominates the Gibbs sampler that generates the states one at a time.
Horserace Polling and Survey Method Effects: an analysis of the 2000 campaign
  • Monika Mcdermott
  • Kathleen A Frankovic
McDermott, Monika and Kathleen A. Frankovic. 2003. "Horserace Polling and Survey Method Effects: an analysis of the 2000 campaign." Public Opinion Quarterly 67:244-264.