No full-text available
To read the full-text of this research,
you can request a copy directly from the author.
Expressive Bayesian Voters, Their Turnout Decisions, and Double Probit: Empirical Implications of a Theoretical Model
Abstract
Abstract Voting is an expressive act. Since people are not born wanting,to express themselves,politi- cally, the desire to vote must be acquired, either by learning about the candidates, by using party identi…cation as a cognitive shortcut, or by contact from a trusted source. Modeled as Bayesian updating, this simple explanatory framework has dramatic implications for the understanding,of voter turnout.,It mathematically,implies the main,empirical general- izations familiar from the literature, it predicts hitherto unnoticed patterns that appear in turnout data, it provides a better …tting statistical model (“double probit”) for sample surveys of turnout, and it allows researchers to forecast turnout patterns in new elections when,circumstances,change. Thus the case is strengthened,for the Bayesian voter model,as a central organizing principle for public opinion and voting behavior. In political life, noneconomic and/or nonegoistic motives seem to be even more important.,Self–interest cannot explain even the very basic fact that most people choose to vote....In order to do justice to these empirical facts, in my opinion we need to replace the one–motive theory of purely egoistic economic,motivation with less restrictive motivational,assumptions. — Harsanyi (1969, 519) 1,Introduction Voter turnout,in American,elections is quite low by international standards.,A little
... Sobre los factores actitudinales, tenemos expectativas de que jueguen un rol promotor del participación electoral: a medida de que los individuos estén más atentos a la política, sean más cercanos a los partidos, y sean capaces de definirse ideológicamente, esperamos que sus probabilidades de participar en las elecciones aumenten (Palfrey & Poole, 1987;Achen, 2006;Bowler & Donovan, 2012;Falk Pedersen et al, 2014). Asimismo, esperamos que las generaciones más jóvenes sean más sensibles a estos factores, particularmente la atención política, en la medida que el involucramiento con la política a los que estos indicadores se asocian puede actuar cómo un sustituto de la falta de experiencia política de los más jóvenes (Lyons & Alexander, 2000;Sandell, 2008). ...
... Respecto a los indicadores actitudinales y políticos, consideramos un indicador binario de identificación con partidos políticos y una variable que captura a quienes son capaces de ubicarse a sí mismos en el espectro ideológico de izquierda-derecha. Esperamos que estas variables actúen como indicadores de desafección y apatía con el sistema político (Achen, 2006;Falk Pedersen et al, 2014). Además, aprovechando el módulo de evaluación de personajes políticos presente en las encuestas CEP, hemos construido un indicador de atención a la política nacional. ...
... Una excepción esAchen (2006), quien por razones teóricas hipotetiza un efecto lineal de la edad sobre la proba- ...
En esta tesis se explora la participación electoral en Chile en el período post-autoritario, poniendo �énfasis en la composición del electorado y los efectos de la reforma electoral de 2011, que reemplazó el sistema de inscripción voluntaria y votación obligatoria por uno de inscripción automática y votación voluntaria. Un primer capítulo se dedica a describir el contexto histórico y las particularidades propias de la relación de los chilenos con el voto, con especial dedicación a las elecciones municipales. Luego, un segundo capítulo discute las principales teorías que explican la asistencia de los ciudadanos a las urnas, as�� como intenta deducir cómo estas ideas podrían aplicarse para el caso chileno. Un tercer capítulo aborda la importancia de las instituciones electorales en los niveles de participación y la composición de los electorados, centrando la discusión en las diferencias que se dan en regímenes de voto voluntario y compulsivo. De la literatura revisada en el segundo y tercer capítulo se extraen hipótesis empíricas a contrastar en el caso de estudio. Luego, una nueva sección describe la metodología utilizada para analizar el caso de estudio. Se propone una aproximación cuasi-experimental que compare los patrones de participación en la �ultima elección municipal con voto obligatorio (2008) y la primera con voto voluntario (2012) mediante el uso de encuestas post-electorales. En el capítulo siguiente se exponen los resultados de los análisis estadísticos. A partir de estos resultados se ofrecen una serie de conclusiones sustantivas relativas a los niveles de participación y representación electoral, a saber: (1) la importancia de las generaciones políticas, (2) el rol del status socio-económico, y (3) la selectividad actitudinal del electorado. Finalmente, se discuten las implicancias normativas de estos resultados para el sistema democrático y la sociedad chilena.
... The variables are included typically for the sake of a better statistical fit within the sample, but they lack power for policy and intervention analysis where behavioral responses are important modeling considerations. Achen (2006) uses an EITM framework to link an ex ante theoretical prediction, based on Bayesian analysis, with an applied statistical analogue-"double-probit." He finds the fit of his model is superior to a traditional and rival applied statistical specification. ...
... where F (·) is the cdf with meanû n+1 and variance 2 n+1 . For theoretical convenience, Achen (2006) assumeŝ u n+1 is nonnegative: the voter only votes for the party valued higher than another. Following Downs (1957), Achen (2006) suggests the utility of voting in period n + 1 is the difference between the expected benefit of voting, E (u n+1 ), and the cost of voting: ...
... For theoretical convenience, Achen (2006) assumeŝ u n+1 is nonnegative: the voter only votes for the party valued higher than another. Following Downs (1957), Achen (2006) suggests the utility of voting in period n + 1 is the difference between the expected benefit of voting, E (u n+1 ), and the cost of voting: ...
An important disconnect exists between the current use of formal modeling and applied statistical analysis. In general, a lack of linkage between the two can produce statistically significant parameters of ambiguous origin that, in turn, fail to assist in falsifying theories and hypotheses. To address this scientific challenge, a framework for unification is proposed. Methodological unification leverages the mutually reinforcing properties of formal and applied statistical analysis to produce greater transparency in relating theory to test. This framework for methodological unification, or what has been referred to as the empirical implications of theoretical models (EITM), includes (1) connecting behavioral (formal) and applied statistical concepts, (2) developing behavioral (formal) and applied statistical analogues of these concepts, and (3) linking and evaluating the behavioral (formal) and applied statistical analogues. The elements of this EITM framework are illustrated with examples from voting behavior, macroeconomic policy and outcomes, and political turnout.
... The first lies in the gradual clarification of interests and preferences. Younger citizens may not have a clear sense of their own current and future political interests (Achen 2006) -at the age of 18-20 they may be in university, in military service or in an apprentice position and lacking a clear sense of whether they are on track to benefit or not from high tax rates, tough workplace regulations or most government policies. These clarify over time, and seem to do so in ways consistent with Bayesian updating (Achen 2006). ...
... Younger citizens may not have a clear sense of their own current and future political interests (Achen 2006) -at the age of 18-20 they may be in university, in military service or in an apprentice position and lacking a clear sense of whether they are on track to benefit or not from high tax rates, tough workplace regulations or most government policies. These clarify over time, and seem to do so in ways consistent with Bayesian updating (Achen 2006). Likewise, new citizens may not fully appreciate how political parties differ (Plutzer 2002) so, even with clear values, it is difficult to map those values on to particular candidates or parties -diminishing the motivation to vote. ...
... when an individual is exposed to more advertisements, it increases their considerations for the advertised party. The model also has a political information interpretation as in Matsusaka (1995) and Achen (2006). ...
... Likely, those who are more interested in politics and completing surveys are less influenced by advertising because they are more knowledgeable about the candidates (Matsusaka 1995, Achen 2006). ...
This paper studies the channels through which political television advertising influences individuals' voting decisions. Recent studies find that advertising primarily persuades voters to change their choices in candidates but does not mobilize people to vote. In contrast, I find that a quarter of the advertising effect is mobilization, and three-quarters are persuasion. To decompose the advertising effect into its components, this paper develops a discrete choice model that analyzes the decisions of turnout and candidate choice jointly as a Markov chain. Current studies often estimate the persuasion and mobilization effects in two static models. The non-trivial mobilization effect is a result of measuring advertising exposure over four-week windows instead of over several months as in other studies. Further, using a Markov chain to model monthly voting intentions leading up to Election Day allows me to measure the advertising effects based on prior voting intentions. In the context of the 2008 presidential election, the magnitude of the advertising effect is not large enough to overcome the significant deficit for McCain, but it can potentially change the outcomes in close elections such as 2000 and 2016
... While most applications in the field of political behavior have been developed in reference to party identification (see Achen, 1992Achen, , 2002Gerber & Green, 1998;Green et al., 2002;Grynaviski, 2006;Bartels & Achen, 2006), scholars have been increasingly using this device as a theoretical mechanism to explain other types of political behavior. For example, Bartels (1993) employs it for media effects, Alvarez (1998) for campaign learning, and Achen (2006) for voter turnout. In fact, it can be safely argued today that it stands as a general model of opinion formation applicable to any type of attitudes or opinions (Bartels, 2002;Bartels & Achen, 2006;Kim et al., 2010). ...
... Some work of Achen also shares this assumption, particularly his formal model of voter turnout. Some of Achen's work also shares this assumption, particularly his formal model of voter turnout(seeAchen, 2006). Specifically, he assumes that voters' experiences with each incumbent party may vary in time, but they are common for all voters within each time period. ...
This dissertation is composed of three papers that study citizens' opinion formation process considering the contextual conditions surrounding them as a key explanatory source of cognitive variation. Taken together, the papers indicate that political regimes that provide abundant cues and easy-to-process political information favor the organization of political opinions and attitudes. My first paper proposes a conceptual framework that interprets the political environment in terms of its degree of complexity and applies it to the structure of citizens' opinions about political issues. My second paper develops and tests hypotheses about how institutional and party system factors encourage the alignment of citizens' political evaluations in accordance with their partisan dispositions. My third paper is an empirical examination of the contentious claim that citizens update their opinions in unbiased fashion. I test this hypothesis by estimating the influence that party identification has over the rate of change of individuals' political opinions on salient political issues. The first two papers employ a cross-national design which studies individuals' political cognition through the lens of the micro-macro linkage, meaning the connection between system-level properties of a polity and the individuals nested in that polity. The third paper studies individuals' opinions considering the passage of time as a key source of heterogeneity. The three papers identify important, and previously unaccounted, empirical patterns such as the varying role of cognitive resources on the level of association of citizens' opinions, the strong influence that clarity of responsibility exercises over the association between partisanship and political evaluations, and the demonstration that individuals update their opinions with strong partisan biases. All papers rely on multilevel modeling techniques, and engage in innovative uses of this statistical framework. This includes dealing with multivariate dependent variables, capturing latent variables, and accounting for the initial conditions problem of longitudinal data. A practical consequence that follows from this dissertation is that when political institutions are designed, we should consider the type of information that average citizens will be exposed to. This can help engage a portion of the electorate that would otherwise not care about public affairs.
... While most applications in the political behavior field have been developed in reference to party identification (see Achen, 1992Achen, , 2002Gerber & Green, 1998;Green et al., 2002;Grynaviski, 2006;Bartels & Achen, 2006), scholars have been increasingly using this device as a theoretical mechanism able to explain other types of political behavior. For example, Bartels (1993) employs it for media effects, Alvarez (1998) for campaign learning, and Achen (2006) for turnout. In fact, it can be safely argued today that it stands as a general model of opinion formation applicable to any type of attitudes and opinions (Bartels, 2002;Bartels & Achen, 2006;Bullock, 2009). ...
... Some work of Achen also shares this assumption, particularly his formal model of voter turnout(seeAchen, 2006). Specifically, he assumes that voters' experiences, as well as their variance, with each incumbent party might vary in time, but is common for all voters within each time period. ...
The incorporation of Bayesian updating mechanisms into models of opinion formation has become a new and rich field of research. In essence, these models claim that voters update their political opinions or evaluations by averaging their previous opinion (prior) and the information received from the political environment with each element weighted by their degree of precision. While this might not seem particularly controversial, some theoretical developments within this approach have become so. One of these is Gerber & Green's (1998, 1999) argument that voters' evaluations do not only follow Bayes rule, but that they update their evaluations in an unbiased fashion, that is, independently of their own orientations and affiliations. The following paper attempts to empirically evaluate this theoretical prediction considering voters' opinions about a set of highly salient issues measured in the ANES 2008-2009 survey panel. For this I em-ploy an autoregressive random coefficient estimation procedure that can estimate the effect of partisanship over the rates of opinion change while accounting for heterogeneous prior opinion effects. Contrary to the unbiased voter hypothesis, results indicate that for some variables -i.e. evaluations of political candidates-the rate of opinion change is substantially and statistically different across partisan groups. I even find polarization trends for several of the opinion vari-ables considered in the study. These results call into attention the need to complement the Bayesian rational learning model with additional features that can accommodate more easily heterogeneous patterns of opinion updating.
... A standard education question reads: ''What is the highest grade of school or year of college you have completed?'' College students, if they interpret the question literally, should answer 12th grade or one, two, or three years of college; if the usual three-to five-point education scale is then created, freshmen would be classified as ''high school graduates'' and sophomores, juniors, and seniors would be considered as having ''some college.'' 2 Yet many of these students will complete college and even earn advanced degrees (as also noted by Achen, 2006). Is it sensible to classify them by the level they have attained so far? ...
... 23 Given the similarity of registration and turnout among college students, we focus on turnout. This, plus the well-known problems with finding good identification restrictions and the sensitivity of such models to core assumptions, makes a selection model in the vein of Timpone (1998) problematic (Achen, 2008). 24 Using frequency of following the campaign in place of political discussions yields nearly identical results. ...
Objectives. Traditional theories of turnout are of limited applicability to college students: the concepts and measures associated with these theories were not designed with students in mind, and factors not considered by the traditional theories are relevant. We offer a new theoretical perspective for understanding college student turnout and test it using a new data set.
Methods. We conducted a phone survey of college students following the November 2004 U.S. presidential election. We test our theoretical expectations using descriptive statistics and logistic regression models.
Results. We find that the usual demographic factors are of little relevance. Turnout is, however, related to the same motivational factors that stimulate older voters, as well as to mobilization by parties. College-specific factors, such as distance between home and college, are also relevant. Turnout was also higher for students whose hometown or college town was in a battleground state.
Conclusion. The results suggest new approaches to the study of young people and how and when they become part of the voting public.
... My probabilistic decision process is similar in spirit to the sampling of "considerations" in the survey environments of Zaller (1992) and Zaller and Feldman (1992), in which a voter's positive considerations for the advertised party increase with exposure to ads. The model likewise has a political information interpretation, as in Matsusaka (1995) and Achen (2006). Empirically, partisans, older, and more educated individuals vote more often. ...
This paper studies the channels through which political television advertising influences individuals' voting decisions. Scholars are interested to learn whether advertising primarily persuades people to change their choices of candidates or mobilizes people to vote. I find that advertising does both: about 60% to 70% of advertising's effect is persuasion, and 30% to 40% of it is mobilization. Advertising's effects are stronger on those who did not plan to vote for a major-party candidate. To decompose the impact into its components, the present paper estimates a multinomial probit model that permits analysis of decisions of turnout and candidate choice jointly in a Markov chain framework. In contrast to most studies that estimate the effects of aggregated exposure to advertising on voters' choices on Election Day, I study how advertising influences peoples' monthly voting intentions leading up to Election Day. In the context of the 2008 presidential election, the magnitude of the advertising effect is not large enough to overcome John McCain's significant deficit, but it potentially could have changed the outcomes of other close elections such as those in 2000 and 2016.
... Voting is the most common form of political participation and most citizens do not actively participate in politics in ways other than voting. Furthermore, low voter turnout is often considered to be bad for democracy (Achen, 2006). Due to the importance of voting in a democracy, scholars have investigated the impacts of personality traits on voter turnout and provided empirical evidence for the relationship between the Big Five personality traits and voter participation. ...
This study examines whether the Big Five personality traits have different effects on male and female turnout. Previous research has reported an association between personality traits and turnout, but their results have been inconsistent. Nevertheless, there is a solid evidence of gender differences in personality traits and past studies have not taken into consideration the option that personality–turnout relationship might be gender-differentiated. The current study empirically finds that conscientiousness and emotional stability can significantly increase female turnout, but have no effect on male turnout. Furthermore, openness to experience exerts opposite effects on male and female turnout. As openness to experience increases, men become more likely to vote, whereas women become less likely to cast their ballots. However, extraversion and agreeableness are not associated with turnout, regardless of gender. To sum up, this study provides robust evidence that the effects of personality traits on turnout vary by gender and suggests that any future study of the topic must include interaction between gender and personality in order to estimate the effect of personality on turnout in a more accurate manner.
... The EITM linkage is the behavioral concept of learning combined with the empirical concept of discrete choice (see Hetherington (2001), Plutzer (2002), and Gerber, Green, and Shachar (2003)). Other examples include Achen (2006), Dhillon and Peralta (2002), and Shachar and Nalebuff (1999). In these latter papers, the authors develop a Bayesian learning model and link it to a discrete choice empirical model. ...
We provide a framework for the Empirical Implications of Theoretical Models (EITM) initiative. The objective of EITM is to encourage political and social scientists to test empirical models that are directly connected to a formal model. As scholars merge formal and empirical analysis they minimize non-falsifiable research practices and lay the foundation for social scientific cumulation. Our EITM framework involves three steps. The first step is for researchers to unify theoretical mechanisms and applied statistical concepts. Step two is to develop measurable devices ("analogues") for these mechanisms and concepts. The final step is to unify the analogues. The significance of the EITM framework is that it encourages scholars to use a set of plausible facts or axioms and then model them in a rigorous mathematical manner to identify causal relations that explain empirical regularities.
... Despite CYZ's claim that the attack-only model is a new alternative to existing techniques , it is neither new nor simpler than estimating the full structural model. Indeed, it is not even new to Achen (2006), as cited by the authors. Estimating the attack-only model via MLE is exactly what was recommended in Signorino (1999, 285, see equation 20), in Signorino (2003, 323, see equation 3), and in Signorino and Yilmaz (2003, 555). ...
The recent article by Carrubba, Yuen, and Zorn (2007) (CYZ) attempts to relate the strategic random utility models in Signorino
(1999, 2002, 2003) and in Signorino and Yilmaz (2003) to existing game theory practice and to existing statistical techniques.
It contributes to this literature by reminding us that comparative statics analysis can be applied to the equilibria of these
models. There are a number of claims in CYZ, however, that require clarification. In particular, the article's primary claim
is that comparative statics analysis, in combination with one of three proposed statistical estimators, provides a simpler
alternative to methods previously advocated. This claim (or combination of claims) is incorrect. When one examines the procedure
CYZ recommends, it is no simpler for substantive researchers than anything previously recommended. Moreover, none of the proposed
estimators are new: they are exactly the same methods introduced in Signorino (1999, 2003), in Signorino and Yilmaz (2003),
in Signorino, Walker, and Bas (2002), and in Bas, Signorino, and Walker (2007).
Rather than adhering to the cost-benefit calculus often employed in the study of turnout, a recent stream in this literature refers to voting as habit forming. The empirical findings supporting this developmental approach are still scarce, however. Using voting-age discontinuities among almost equally aged individuals, this study enables the identification of the effect of voting in one election on turnout in future elections. Tracing individuals for more than 30 years of their lives, the long-term effect of early voting experiences on people's turnout profiles is also examined. The findings show early voting experiences shape future voting profiles. Moreover, casting a ballot does not boost non-electoral participation.
En este trabajo se explora el efecto de las instituciones electorales en la composición de los electorados, particularmente el tipo de inscripción y el carácter de voluntariedad/obligatoriedad del voto. Explotando la reciente reforma electoral en Chile a través de un diseño cuasi-experimental, se comparan la composición de los electorados antes y después de la reforma, en base a los principales predictores sugeridos por la literatura especializada. Entre los hallazgos centrales se encuentran: un fuerte efecto de la edad en ambos contextos normativos, la importancia de la educación y de atención política tras la reforma, y la heterogeneidad de estos efectos entre las generaciones políticas. Adicionalmente, se cuantifica el efecto de la reforma electoral para distintos grupos, mostrando que ésta no afectó a todos por igual. Finalmente, se concluye ofreciendo una discusión de las implicancias democráticas de la reforma electoral y sus consecuencias en el sistema político.
This paper uses discontinuities imposed by voting-age restrictions to identify the effect of past eligibility on subsequent participation decisions and partisan identification. It compares participation decisions and partisan affiliations of individuals who turned 18 just before past elections with those who turned 18 just after. It presents three main findings. First, past presidential election eligibility increases the probability of subsequent participation. For example, the point estimates indicate that 2000 presidential election eligibility increased participation in the 2004 presidential election by 3.0-4.5 percent, which suggests that voting in the 2000 presidential election increased 2004 participation by about 5 percentage points. Second, past presidential election eligibility affects partisan identification. Third, these effects continue to persist for several election cycles after a voter first becomes eligible.
growing number of Americans rely exclusively on mobile technology as their primary means of communication. Political campaigns and voter mobilization groups must reevaluate how they connect with this segment of voters who aren't reachable through conventional landline and U.S. mail outreach. This study demonstrates that text messaging is a powerful new tool that can be used to reach and harness the voting power of young people. ! Working through independent registration efforts, Working Assets, the Student PIRGs, and Mobile Voter registered nearly 150,000 new voters leading up to the 2006 election. Over 12,000 of these new registrants who provided cell phone numbers were selected for an experiment to test the effectiveness of text messaging for mobilizing voters. ! Half of the participants received a text message reminder to vote on the Monday before Election Day; half did not. Members of the treatment group, who received the reminder, voted at a rate of 56.3%; members of the control group, who did not receive the reminder, voted at a rate of 53.2%—a difference of 3.1 percentage points. Statistically, there is over a 99% chance that the text messages had a net benefit on turnout. ! Because some individuals (about 20%) in the treatment group did not receive the text message (for reasons such as an incorrect phone number), the personal effect of text messaging is greater than the aggregated effect of 3.1 percentage points. On a personal level, a text message recipient was 4.2 percentage points more likely to vote than an individual who did not receive a text message. Messages that were short and to the point were most effective.
Displaying a political yard sign is a conspicuous element of the election experience but an understudied act of political participation. We argue that the study of spatial patterns in the dissemination of yard signs speaks to the debate over “context” as a cause of political participation. We ask whether evidence of such contextual effects exist, and if so, how they operate: through the micro-social environment and interpersonal communication, or the meso-social environment and impersonal information cues? To address these questions, we utilize an original, geo-coded observational dataset of more than 25,000 households in Franklin County, Ohio from the 2008 presidential election cycle. In addition to traditional predictors of participation, we find evidence that spatial dependence exists in the distribution of yard signs at both the property level and at the Census block level, indicating that both contextual mechanisms are simultaneously important in creating the “sign wars” often observed in neighborhoods.
Beginning in 1999, Curtis Signorino challenged the use of traditional logits and probits analysis for testing discrete-choice, strategic models. Signorino argues that the complex parametric relationships generated by even the simplest strategic models can lead to wildly inaccurate inferences if one applies these traditional approaches. In their stead, Signorino proposes generating stochastic formal models, from which one can directly derive a maximum likelihood estimator. We propose a simpler, alternative methodology for theoretically and empirically accounting for strategic behavior. In particular, we propose carefully and correctly deriving one's comparative statics from one's formal model, whether it is stochastic or deterministic does not particularly matter, and using standard logit or probit estimation techniques to test the predictions. We demonstrate that this approach performs almost identically to Signorino's more complex suggestion.
This paper uses the consolidation of polling places in Los Angeles County during the October 2003 gubernatorial recall election to study the costs of voting. The consolidation afforded an opportunity to observe a natural experiment: those whose polling places were changed in the consolidation should be statistically comparable to those whose polling places are not changed. Thus, we may observe both groups' turnout rates and ascribe any differences to the treatment of the change in polling locus. We find evidence that changing polling place locations does decrease turnout overall by a substantial 1.88 percentage points; a drop in polling place turnout of 3.05 percentage points is offset by an increase in absentee voting of 1.19 percentage points.
J/47Te address the effects of negative campaign advertising on turnout. Using a unique jfjJ \ experimental design in which advertising tone is manipulated within the identical X V t audiovisual context, we find that exposure to negative advertisements dropped intentions to vote by 5%. We then replicate this result through an aggregate-level analysis of turnout and campaign tone in the 1992 Senate elections. Finally, we show that the demobilizing effects of negative campaigns are accompanied by a weakened sense of political efficacy. Voters who watch negative advertisements become more cynical about the responsiveness of public officials and the electoral process. It is generally taken for granted that political campaigns boost citizens' involvement-their in- terest in the election, awareness of and informa- tion about current issues, and sense that individual opinions matter. Since Lazarsfeld's pioneering work (Berelson, Lazarsfeld, and McPhee 1954; Lazarsfeld, Berelson, and Gaudet 1948), it has been thought that campaign activity in connection with recurring elec- tions enables parties and candidates to mobilize their likely constituents and "recharge" their partisan sen- timents. Voter turnout is thus considered to increase directly with "the level of political stimulation to which the electorate is subjected" (Campbell et al. 1966, 42; Patterson and Caldeira 1983). The argument that campaigns are inherently "stim- ulating" experiences can be questioned on a variety of grounds. American campaigns have changed dra-
Why is turnout higher in some countries and/or in some elections than in others? Why does it increase or decrease over time? To address these questions, I start with the pioneer studies of Powell and Jackman and then review more recent research. This essay seeks to establish which propositions about the causes of variations in turnout are consistently supported by empirical evidence and which ones remain ambiguous. I point out some enigmas and gaps in the field and suggest directions for future research. Most of the research pertains to established democracies, but analyses of nonestablished democracies are also included here.
The empirical study of political participation is approximately 30 years old. The earliest studies concentrated on explaining why some people chose to vote and others did not. Gradually, inquiries into political participation looked at other behavior such as campaigning, making financial contributions, attending meetings, and so forth. Most of these somewhat broader inquiries into political participation were embedded in studies of voting behavior (Berelson, Lazarsfeld, & McPhee, 1954; Campbell, Gurin, & Miller, 1954; Campbell, Converse, Miller, & Stokes, 1960; Lazarsfeld, Berelson, & Gaudet, 1944). As the findings began to accumulate from a variety of studies done in a variety of countries, it became possible for scholars to prepare summary statements of the findings with respect to political participation (Lane, 1959; Milbrath, 1965).
This paper presents an alternative interpretation of the role played by issues and people's evaluations of the parties' positions on different issues in the electoral process. The statistical model presented and estimated in the paper departs from previous studies by hypothesizing a model in which people's issue positions, their evaluations of the parties' positions, and the strength of their party identifications are simultaneously related to each other and are thus endogenous to the electoral process. Expected votes are then determined by party evaluations. The model is estimated with data from the 1964 Election Study of the Survey Research Center. The most important influences are the effects of people's issue positions on their evaluations of the parties and the influence of these evaluations on party identifications and on voting decisions. The paper concludes by discussing the estimated model's implications for evaluations of the role of the electoral process in the determination of public policy.
Non-voters and voters, when equated as to sex, religion, education, and income, differ significantly. The non-voters took part minimally in organized activities and exposed themselves relatively less to politics in the mass media. Their political information was comparatively meager, and their attitudes uncrystallized. Non-voters indentifies themselves more closely with non-political leaders and were generally more isolated and immobile and most concerned with economic problems.
Survey data are used to test the basic tenets of rational choice theory. We find that (1) voters' preferences are orderly but voters do not always have preferences. When the perceived difference between two candidates is below some threshold, voters are indifferent between them. (2) Most voters choose candidates whose issue positions are congruent with their own. (3) Lack of knowledge about candidates affects choice and utility through two routes—indifference and risk aversion.
Most voters have highly structured utility functions, but the thresholds and risk aversion in these functions place relatively unknown candidates at a disadvantage in presidential primaries and other electoral situations. Moreover, the importance of knowledge in these functions suggests that social choice theorists must take seriously the classic concerns of public opinion research—the operation of the media and the dynamics of information in elections.
I examine the impact of long-term partisan loyalties on perceptions of specific political figures and events. In contrast to the notion of partisanship as a simple “running tally” of political assessments, I show that party identification is a pervasive dynamic force shaping citizens' perceptions of, and reactions to, the political world. My analysis employs panel data to isolate the impact of partisan bias in the context of a Bayesian model of opinion change; I also present more straightforward evidence of contrasts in Democrats' and Republicans' perceptions of “objective” politically relevant events. I conclude that partisan bias in political perceptions plays a crucial role in perpetuating and reinforcing sharp differences in opinion between Democrats and Republicans. This conclusion handsomely validates the emphasis placed by the authors of The American Voter on “the role of enduring partisan commitments in shaping attitudes toward political objects.”
After a virtual neglect for several decades, in the last twenty years renewed interest has been shown in a general theory of social behavior. Most of this theoretical work has depended on two main postulates. One is the functionalist (sometimes called structural-functional) approach to the explanation of social institutions, based on the assumption that the social institutions of a given society can best be understood in terms of their social functions, that is, in terms of the contributions they make to the maintenance of social systems as a whole. For lack of an established technical term, we shall call the other postulate the conformist approach to the explanation of individual behavior, it is based on the assumption that uniformities of individual behavior in a given society can best be understood in terms of certain commonly accepted social values, which most members of the society tend to internalize during their socialization process.
The abstract for this document is available on CSA Illumina.To view the Abstract, click the Abstract button above the document title.
The paradox of not voting is examined in a model where voters have uncertainty about the preferences and costs of other voters. In game-theoretic models of voter participation under complete information, equilibrium outcomes can have substantial turnout even when voting costs are relatively high. In contrast, when uncertainty about preferences and costs is present, only voters with negligible or negative net voting costs participate when the electorate is large.
This article presents a model of individuals' party identification that contrasts with previous models. Past models, with the few recent exceptions noted, assume a hierarchical relationship either from identification to other aspects of political behavior, such as the perception and evaluation of issues and candidates, or from these behaviors to party identifications. The model discussed here places party within a dynamic concept of the electoral process and tests several hypotheses about factors producing changes in identifications. The first factor, consistent with the spatial-type issue voting models, estimates the effects of the relative proximity of each party to the individual's own policy preferences. Second, we examine the effect of the actual voting decision on subsequent identifications, with the expectation that if votes differ from previous identifications, there is a resulting shift in partisanship. Finally, we examine the hypothesis that identifications become less susceptible to change as people age and accumulate political experience. When combined with other research, the results indicate a model of the electoral process in which party identifications are both influenced by circumstances specific to each election and influence other behaviors. This nonrecursive model has a number of implications for the development and evolution of individual and aggregate partisanship. These implications are discussed at the end of the article.
In their book Non-Voting , published in 1924, Charles E. Merriam and Harold F. Gosnell reported that many persons otherwise eligible to vote had been disfranchised by Chicago's registration requirements. Their data showed that “there were three times as many adult citizens who could not vote because they had failed to register as there were registered voters who had failed to vote in the particular election” and that “entirely different reasons [for not voting] were emphasized by those who were not registered than by those who were registered but did not vote …” Their observation can hardly be said to have been influential. Until very recently most students of voting have paid little attention to the temporally prior act of registration.
Failure to do so has had important consequences. It has made it easy to discount unduly the significance of political influences on the size and composition of electorates; easy to argue unrealistically about the value of efforts to increase the turnout of voters; and easy to be puzzled about some aspects of the behavior of voters.
Theory: Characterizing voters as rational actors who update their party affiliations based on a Bayesian assimilation of new information, Achen (1992) showed that a revisionist model of party identification generates, among its empirical implications, stable partisanship among adults. The model further implies that susceptibility to partisan change declines with age. The significance of Achen's model for the study of party identification leads us to examine more closely its underlying assumptions and the empirical ramifications of this and other learning models. Method: This essay develops a more general learning model, based upon the Kalman filter, that encompasses the Achen model as a special case. Results: We show that the Achen assumption of a fixed party benefit level leads to implausible implications about how voters learn from the history of party performance. When party benefit levels are allowed to vary over time, models of voter learning no longer imply that partisan attitudes, even among the older segments of the population, remain stable in the wake of new information about the parties. We conclude by discussing the empirical viability of our revised learning model and its implications for the study of partisan attitudes.
A model of the strategic calculus of voting that incorporates the behavior of groups is presented. If candidates choose different positions, positive turnout is possible in (partial) equilibrium and is generally larger than that without group rationality. Nevertheless, if candidates are probability maximizers and a unique (general) equilibrium exists, the candidates will choose equivalent positions and turnout will be zero even with group rationality. But if candidates care about both policy and winning, they will choose different positions and cause positive turnout. This turnout will be larger than that would occur without group rationality. Group rationality, alone, cannot explain positive turnout in a (general) equilibrium theory of elections; it does imply larger turnout in conjunction with policy-oriented candidates than would be predicted solely by individualistic rational behavior.
Logit and probit, the two most common techniques for estimation of models with a dichotomous dependent variable, impose the assumption that individuals with a probability of .5 of choosing either of two alternatives are most sensitive to changes in independent variables. This assumption is imposed by the estimation technique because both the logistic and normal density functions are symmetric about zero. Rather than let methodology dictate substantive assumptions, I propose an alternative distribution for the disturbances to the normal or logistic distribution. The resulting estimator developed here, scobit (or skewed-logit), is shown to be appropriate where individuals with any initial probability of choosing either of two alternatives are most sensitive to changes in independent variables. I then demonstrate that voters with initial probability of voting of less than .5 are most sensitive to changes in independent variables. And I examine whether individuals with low levels of education or high levels of education are most sensitive to changes in voting laws with respect to their probability of voting.
There is something of a paradox in the findings of public opinion research that those individuals who shift their political
position from one election to another frequently have less information about the political situation and the issues than persons
whose attitudes and positions remain relatively unchanged. This runs contrary to the common assumption that changes in political
attitudes and behavior occur in response to communication of information and arguments about the principal candidates and
the issues on which they are divided. Here is an extended discussion of the question.
Turning out to vote is the most common and important act of political participation in any democracy. Voting is also less well understood and explained empirically than other political acts engaged in regularly by citizens. Turnout, however, presents a special problem for rational choice theories of politics, for it is taken to be the paradigmatic example of the problem of collective action, in which, although all may benefit from voting, it is rarely in the individual's self-interest to vote. This paper begins by examining the problem of explaining turnout. A basic form of rational choice models of turnout is developed--basic in the sense that it is common to all such models. This basic model is shown to be incomplete, and the two most important models, the calculus of voting and the minimax regret model, are illustrated as alternative ways to complete this basic model, along with mention of game-theoretic models. Their strengths and weaknesses are then assessed. The remainder of the paper argues that rational choice accounts of turnout are possible. The first step is to argue that turnout is not an especially problematic version of the collective action problem because it is, for many, a low cost, low (expected) benefit decision. A "strategic politicians" account of turnout and campaigns is examined next. A reinterpretation of the intrinsic benefits of voting is then considered and is used to examine the most important substantive problem in the turnout literature, its decline. These steps, I argue, make theories of ordinary political decisions at once both more political and more integrated into the politics of the larger system.
Rational choice theories have difficulty accounting for individual participation in collective action. Voting is among the acts hardest to explain; consumption benefits have been proposed to account for nonzero turnout. The model presented here establishes a political investment basis for these benefits. I argue that voters often act not in isolation but as part of groups with shared interests. While individuals still vote because of consumption benefits arising from the act of voting, some of these consumption benefits are provided by group leaders out of collective benefits received by the group in return for its votes. The collective benefit comes from a candidate adopting policy positions closer to those of the group. In passing, I show that the widespread availability of public opinion polling increases the candidate's incentive to shift position. Finally, some data from recent off-year elections are shown to be consistent with the model.
The apparent decline in voter participation in national elections since 1972 is an illusion created by using the Bureau of the Census estimate of the voting-age population as the denominator of the turnout rate. We construct a more accurate estimate of those eligible to vote, from 1948–2000, using government statistical series to adjust for ineligible but included groups, such as noncitizens and felons, and eligible but excluded groups, such as overseas citizens. We show that the ineligible population, not the nonvoting, has been increasing since 1972. During the 1960s the turnout rate trended downward both nationally and outside the South. Although the average turnout rates for presidential and congressional elections are lower since 1972 than during 1948–70, the only pattern since 1972 is an increased turnout rate in southern congressional elections. While the voting age was lowered to 18 in 1971, the lower turnout rate of young voters accounts for less than one-fourth of reduced voter participation.
For the most part, scholars who study American political parties in the electorate continue to characterize them as weak and in decline. Parties on the elite level, however, have experienced a resurgence over the last two decades. Such a divergence between elite behavior and mass opinion is curious, given that most models of public opinion place the behavior of elites at their core. In fact, I find that parties in the electorate have experienced a noteworthy resurgence over the last two decades. Greater partisan polarization in Congress has clarified the parties' ideological positions for ordinary Americans, which in turn has increased party importance and salience on the mass level. Although parties in the 1990s are not as central to Americans as they were in the 1950s, they are far more important today than in the 1970s and 1980s. The party decline thesis is in need of revision.
The decision to vote is affected by two elements of the election context. One is the frequency of elections. Presidential and state primaries divert resources away from the general election and reduce turnout among the peripheral electorate who are most dependent on a mobilization effort. Taken together, spring and fall primaries lowered general election by five percentage points nationwide in the 1976, 1980, and 1984 elections. A second element of context is the attractiveness of statewide offices on the presidential year ballot. Gubernatorial races increase the probability of voting by 6% in those states that still elect governors in presidential years. Thus, the postwar shift of gubernatorial races to the congressional election year is one explanation for declining turnout. Senatorial races do not attract additional voters to the November election. These hypotheses are tested on a pooled sample of the 1976, 1980, and 1984 CPS election studies.
In a democracy, voting is the most fundamental act of political
participation and therefore holds a central location in the study of
political behavior. One significant research tradition focuses on the
relationship between registration and turnout and raises important and
related questions. How do registration laws influence turnout levels?
What types of people are most affected by them? What are the partisan
implications of registration requirements? Spanning the late nineteenth
through early twenty-first centuries, this essay places these questions
in the appropriate theoretical context and then answers them. a
This paper reframes our inquiry into voter turnout by making aging the lens through which the traditional resource and cost measures of previous turnout research are viewed, thereby making three related contributions. (1) I offer a developmental theory of turnout. This framework follows from the observation that most citizens are habitual voters or habitual nonvoters (they display inertia). Most young citizens start their political lives as habitual nonvoters but they vary in how long it takes to develop into habitual voters. With this transition at the core of the framework, previous findings concerning costs and resources can easily be integrated into developmental theory. (2) I make a methodological contribution by applying latent growth curve models to panel data. (3) Finally, the empirical analyses provide the developmental theory with strong support and also provide a better understanding of the roles of aging, parenthood, partisanship, and geographic mobility.
Low voter turnout is a serious democratic problem for five reasons: (1) It means unequal turnout that is systematically biased against less well-to-do citizens. (2) Unequal turnout spells unequal political influence. (3) U.S. voter turnout is especially low, but, measured as percent of voting-age population, it is also relatively low in most other countries. (4) Turnout in midterm, regional, local, and supranational elections--less salient but by no means unimportant elections--tends to be especially poor. (5) Turnout appears to be declining everywhere. The problem of inequality can be solved by institutional mechanisms that maximize turnout. One option is the combination of voter-friendly registration rules, proportional representation, infrequent elections, weekend voting, and holding less salient elections concurrently with the most important national elections. The other option, which can maximize turnout by itself, is compulsory voting. Its advantages far outweigh the normative and practical objections to it.
This article develops a model that simultaneously considers individual turnout and vote choice while also accounting for uncertainty about candidates. The theoretical develop-ment of this model implies that the effects of uncertainty on turnout vary with the strength of individual preferences. Application of the model to individual choice in the 1996 American presidential election confirms that decreasing uncertainty about the character traits of the candidates decreases the probability of abstention for individuals with strong preferences but increases the probability of abstention for individuals with weak preferences.
s Abstract The past two decades have brought revolutionary change to the field of political methodology. Steady gains in theoretical sophistication have combined with explosive increases in computing power to produce a profusion of new estimators for applied political researchers. Attendance at the annual Summer Meeting of the Methodology Section has multiplied many times, and section membership is among the largest in APSA. All these are signs of success. Yet there are warning signs, too. This paper attempts to critically summarize current developments in the young field of political methodology. It focuses on recent generalizations of dichotomous-dependent-variable estimators such as logit and probit, arguing that even our best new work needs a firmer connection to credible models of human behavior and deeper foundations in reliable empirical generalizations.
Habit is a frequently mentioned but understudied cause of political action. This article provides the first direct test of the hypothesis that casting a ballot in one election increases one's propensity to go to the polls in the future. A field experiment involving 25,200 registered voters was conducted prior to the November general election of 1998. Subjects were randomly assigned to treatment conditions in which they were urged to vote through direct mail or face-to-face canvassing. Compared to a control group that received no contact, the treatment groups were significantly more likely to vote in 1998. The treatment groups were also significantly more likely to vote in local elections held in November of 1999. After deriving a statistical estimator to isolate the effect of habit, we find that, ceteris paribus, voting in one election substantially increases the likelihood of voting in the future. Indeed, the influence of past voting exceeds the effects of age and education reported in previous studies.
There are several major insights which this game theoretic analysis has produced. First, we have shown that equilibria exist with substantial turnout even when both the majority is much larger than the minority and the costs of voting are exceptionally high. For example, in large electorates using the status quo rule, we show mixed-pure equilibria with turnout roughly equal to twice the size of the minority when the cost is nearly equal to 37% (e
–1 100) of the reward (B). Second, with large electorates the many equilibria appear to reduce to just two types, the type just mentioned and a type with almost no turnout. Third, we have shown that turnout may rise as the costs of voting rise. This results when all members of a team adjust their turnout probabilities so that the probability of being pivotal increases to match the increased cost of voting. We have also shown that turnout is nearly invariant with costs in large electorates where turnout probabilities approach one or zero. Fourth, the actual split of the vote is likely to be a biased measure of the actual distribution of preferences in the electorate. Because majorities have greater incentives to free-ride, they will turn out less heavily than minorities. Elections can be relatively close, even when one alternative is supported by a substantial majority of the electorate. The probability that the majority will win does not seem to be closely related to the size of the electorate or its size relative to the minority. However, turnout is quite strongly correlated with the relative sizes of the minority and the majority.