Individual Voting Behavior Research Articles

Nomination Choices: Caucus and Primary Outcomes, 1976-88

Primaries and caucuses now determine the Democratic and Republican parties' presidential nominations. While several political scientists have investigated the voting behavior of individual primary voters (Abramson et al. 1992; Bartels 1988; Geer 1989; Keeter and Zukin 1983; Marshall 1984; Norrander 1986, 1992; Wattier 1983a, 1983b; Williams et al. 1976), aggregate results of primaries and caucuses have scarcely been studied. Primary and caucus outcomes foster publicity, contributions, and, perhaps, voter support for some candidates in what is commonly called momentum, while forcing other candidates from the field. Aggregate vote totals allocate convention delegates for each state, and delegate totals ultimately decide the nominations. While individual decisions make up aggregate outcomes, we are left to make inferential leaps from studies of individual voters when what we really need to understand is the collective outcome. In this vein, presidential primary researchers need to follow the lead of congressional scholars who find it necessary to understand both the aggregate dynamics of and the individual voting behavior that underlie the loss of seats by the presidential party in offyear congressional elections. Flaws in three previously published aggregate analyses of presidential primary and caucus returns hamper us in our ultimate goal of making generalizations about the outcomes of these vital contests. Two of the works ignore caucuses (Bartels 1988; Grush 1980); two analyze only candidates from one party in one election year (Grush 1980; Parent, Jillson, and Weber 1987); and one ignores the dynamic elements of nomination contests (Parent, Jillson, and Weber 1987).

Electoral Dynamics in South Korea Since 1981

The analysis of election results from South Korea presented in this paper is one part of a cross-national study of electoral dynamics in several countries, including the Nordic countries, Britain, Italy, the United States, and Japan. The object of the study is primarily to assess the general applicability of the logit methodology for ecological analysis of election results, formulated by Thomsen (1987). In a "most different systems research design," as recommended by Przeworski and Teune (1970), South Korea was chosen as a supposed extreme case marked by volatile electoral behavior during the critical transformation from authoritarian regime to emergent democracy (Billet, 1990; Han, 1989). It is, of course, possible to find more extreme cases in third world countries, but one advantage with the case of Korea is that direct election fraud in counting the votes has been a minor problem in recent elections (C. Park, 1988a, p. 64). The general approach in the analysis is to identify possible causes of the electoral dynamics at four different levels: the national, the regional, the local district, and the individual levels. While actual election results are well suited for the study of electoral dynamics on the first three levels, the estimation of individual voting behavior by the method of ecological inference is more controversial. However, in a country like South Korea, where survey results on individual voter mobility is either not available or of questionable validity, we believe that ecological estimates of voter mobility can provide important clues concerning the possible motives of the individual voters. 1

A Federalist Explanation of Municipal Elections

This paper suggests that individual voting behavior in municipal elections is most closely associated with voter concern with municipal economic development and basic city services. Redistributive issues and race are, as such, irrelevant in local elections. Candidates for local office should, therefore, avoid such issues and associate themselves with economic growth potential and better provision of services. To test this fiscal federal assertion, a panel survey of registered voters in Houston, Texas, was conducted during the city’s 1985 mayoral election. The findings support the assertion that municipal electoral politics are limited by the very policies with which municipal governments are charged.

Alienation, indifference and the choice of ideological position

In this paper we describe a simple model of individual voting behavior and present its implications for the candidate positioning problem under both vote and plurality maximization. Under our assumptions, some voters at the extremes of the ideological spectrum typically will not vote because they are alienated by the equilibrium location of candidates. There will also be some voters in the middle of the ideological spectrum who will not vote because they are indifferent between the equilibrium locations of the candidates. Both the abstention from alienation and from indifference arise explicitly from utility maximization. Once we allow for alienation and indifference, the two alternative candidate objective functions (vote maximization and plurality maximization) yield different outcomes. In particular, we show that under vote maximization the Median Voter (or Minimum Differentiation) outcome will not arise. On the other hand, under plurality maximization, the Median Voter outcome may or may not hold, depending on the distribution of voter preferences.

Drawing aggregate inferences from individual level data: the utility of the notion of a normal vote

Abstract. Instead of using an all‐encompassing model of individual voting behaviour for drawing aggregate inferences from individual level data, this article proposes an alternative approach by employing a normal vote‐model. Three levels of abstraction are distinguished in the notion of a normal vote: the normal vote‐concept, the normal vote‐approach and the normal vote‐model. A discussion of the most detailed empirical elaboration of the notion of the normal vote, that by Converse, is given and then an inventory of criticisms is made. Although valid in their own right, these criticisms fall short of damaging the very concept of a normal vote. Finally, it is argued that if one intends to transfer the concept to an electoral context other than the American, normal vote‐approaches (and consequently normal vote‐models) may differ across countries and over time.

Party strategies and the use of televised campaign debates

Abstract. This paper examines the evolution of debates as a significant element of the strategy of election campaigns in a number of countries. Analyzing survey data from two Canadian cases (the elections of 1979 and 1984), it is found that the effects of debates on individual voting behaviour are primarily reinforcing, but that debates have nevertheless grown considerably in importance in strategic decision making by parties and candidates.

Electoral Consequences of "Winning" Televised Campaign Debates

This study explores the long-neglected impact of a candidate being perceived as the “winner” of political debates on individual voting behaviour in West Germany. While there has been much research on the effects of debates on perceptions of candidate policy stands and personality atttributes, there has been little work that attempts to guage how much “winning” the debate helps a candidate at the polls. I investigate this process with straightforward probit models of the vote choice, including party identification and candiate evaluation as well as whether the individual judged a particular candidate as the “winner”, or “fared best” in the debate. These models are run for the 1972, 1976, 1980, and 1983 elections in West Germany. The results show that there is a consistent impact of “winning” the debate on individual voting choices in West Germany; controlling for party and other factors, there is a significantly higher probability of voting for the party of the politician the respondent feels won the contest. Debates, then, do ultimately make a substabtial electoral difference.

Comparing the Policy Positions of Supreme Court Justices from Different Periods

T HHE small size of the Supreme Court creates a number of difficulties for analysis of decision making on the Court. One of these difficulties concerns explanation of differences among justices in their voting behavior. The membership of 435 in the House of Representatives allows sophisticated analysis of characteristics associated with individual voting behavior. In contrast, the Supreme Court's membership of 9 largely precludes even the most elementary analysis of such characteristics. Largely for this reason, some students of the Court have undertaken analyses that go beyond the nine justices who sat at a single time to include all the justices who served during a period of several decades. C. Neal Tate (1981) and S. Sidney Ulmer (1986), for instance, have taken this approach in examining relationships between personal background characteristics and voting behavior (see also Handberg 1986). The benefit of this approach in increasing the number of justices to analyze is counterbalanced by a problem of comparability. The most systematic basis for comparisons among justices, and the one that scholars typically use, is their voting support for particular values or types of claims made by litigants. However, since justices who sat during different parts of a period faced different sets of cases, such measures of their voting behavior are not fully comparable. Differences among justices in support for certain claims or values might result from differences in the case stimuli to which they responded rather than from true differences in their policy positions. If one justice supported challenges to government economic regulation more often than another, that difference might stem at least in part from the greater difficulty of supporting such challenges in the cases that the second justice faced.

Groups and political behavior: Legitimation, deprivation, and competing values

This essay asks what might be most usefully studied in future analyses of the impact of groups on the voting behavior of individuals. A dimension of group analysis so far neglected concerns some value orientations that affect the relevance of group consciousness to voting or other political behavior. In particular, the legitimacy of group political action and of group politics more generally is of central importance. Connected to such evaluations is the character of group comparisons, such as in group-focused relative power deprivation. Also of relevance are competing political value constellations, especially individualism and majoritarianism. By the use of a variety of National Election Studies and Wisconsin survey data, these themes are explored empirically to show the extent of a group focus that goes beyond the usual measures of group consciousness. Relative approval of group-based pluralism is also shown to affect the patterns of relationships of major predictors of turnout. And finally, a LISREL analysis is presented that shows that a group focus is as important as a party focus in affecting the level of political alienation pertinent to voter participation.

Great Debates: The Televised Leadership Debates of 1979

AbstractThis article examines the structure and setting of Encounter '79, the televised debates between party leaders which took place a week prior to the 1979 federal election, and considers the possible impact of this event on individual voting behaviour and on the outcome of that election. Using data from the 1979 National Election Study to measure exposure to the debates and perceptions of their content, it is argued that much of the potential effect is limited by public exposure to other political information through television and by the tendencies of such events to reinforce existing attitudes. The direct effects, if any, are more likely to be on participation than on voting decisions.

Identification, Agreement and Government Performance: The Relative Impact on Voting*

Debating explanations of electoral behavior, American scholars have focused on three main theories: The identification model relying on underlying loyalty of voters towards specific parties; the political agreement or proximity model assuming a rational calculation of parties' ideological positions or stand on salient political issues as the yardstick for choice of party; and the investment model relying on voters' ability to calculate which government alternative will bring most utility for the individual voter. Examining these theories with the use of Norwegian data from the elections of 1965, 1969 and 1977, we find that the identification model is far the most powerful in predicting individual voting behavior. This model also has an edge in explaining support for the individual parties and the total distribution of voters. However, at the major postwar government election in 1965, the investment model certainly is of importance, and at the election in 1977 the significance of the proximity model has increased.

Partisanship, Voting Behavior, and Election Outcomes in Canada

Perhaps the most salient feature of Canadian federal elections in the recent past has been the similarity of their outcomes. In the period since the end of World War II, the Liberal party has dominated the electoral arena, winning seven of ten general elections. This party has formed the government of Canada continuously since 1963, its share of the popular vote varying on average by only 4.8 percent over the last five elections. At present, however, the basis of that Liberal dominance is imperfectly understood since relatively few studies have been made of the psychological basis of electoral choice in Canada or of the relationships between individual voting behavior and election outcomes.' To enhance understanding of these phenomena, this study will examine the nature of partisanship in the Canadian electorate and the interplay between voters' partisan attachments and the stimuli provided by short-term forces generated during particular election campaigns. The pattern of electoral outcomes in Canada, together with the strength and salience of certain social cleavages (e.g., region, religion, ethnicity) might lead one to assume that voters' partisan predispositions are highly stable-reflecting, and in turn reinforcing, the enduring political significance of these cleavages. Yet, research in Canada and other political settings suggests that a note of caution may be in order, as interpretations of voting behavior in Western democracies have undergone substantial revision since the advent of the earliest American studies. The American Voter (1960), and some of its predecessors,2

Dynamic models of the voter's decision calculus: Incorporating retrospective considerations into rational-choice models of individual voting behavior

Since Kramer's article (1971), a growing body of literature indicates that U.S. national elections can be viewed as referenda on the performance of incumbent administrations. Retrospective considerations, however, have not been explicitly incorporated into a spatial model of party competition.

On the Decline of Competition in Congressional Elections

Several authors have observed a decline in the number of competitive congressional districts during the past two decades. Various explanations have been proposed for this change. Among these are theories attributing major causal significance to changing methods of drawing district boundaries, and increasing control of campaign resources by incumbents. These theories are examined critically and arguments are advanced for their rejection. The principal cause of the decline of competition for congressional seats appears to rest on a change in individual voting behavior.

On the Decline of Competition in Congressional Elections

Several authors have observed a decline in the number of competitive congressional districts during the past two decades. Various explanations have been proposed for this change. Among these are theories attributing major causal significance to changing methods of drawing district boundaries, and increasing control of campaign resources by incumbents. These theories are examined critically and arguments are advanced for their rejection. The principal cause of the decline of competition for congressional seats appears to rest on a change in individual voting behavior.

Predicting the outcome of the senate trial of Richard M. Nixon

A prediction model for individual voting behavior in the U.S. Senate is developed and tested. The predictors are empirically derived indices which are combined using a unit weighting scheme. This model is then tested on recent roll call votes and a prediction is made about the results of the impeachment trial of Richard M. Nixon.

An Aggregate Data Analysis of Turnout and Party Voting in Local Elections

Much has been written about individual voting behaviour in Britain and the United States but relatively little work has been done with aggregate data in the study of total electoral situations. A partial correlation and multiple regression analysis of local elections in a large English city shows that class, housing tenure, age, and coloured immigrants are closely related to voting patterns of total electorates, but geographical mobility and density of hoeing occupation are not. The political variables of marginality, size of majority, and number of candidates are not noticeably related to either turnout or party voting, although the presence of a Liberal candidate does increase turnout in some circumstances. Housing tenure has a particularly strong and complex relationship with voting patterns which suggests that the concept of housing class may be useful in the analysis of voting as well as more general social patterns. There is no persistent pattern of support for the Liberals who gain or lose votes according to the popularity of the two main parties.

Electoral Choice in the American States: Incumbency Effects, Partisan Forces, and Divergent Partisan Majorities

It is a relatively common occurrence in American politics for state electorates to divide their partisan majorities between different parties, depending on the office contest. Observations concerning these divergent aggregate patterns are usually accompanied by speculation that the theoretical propositions on individual voting behavior, developed and tested in the context of presidential voting, hold less relevance for voting in statewide contests. Evidence presented in this paper does not bear out that view of state elections. The candidate incumbency context of state elections is introduced as an aid in predicting the partisan direction of split-ticket voting at the state level. Setting respondents in various conflict situations with respect to (1) basic party loyalties, (2) net assessments of presidential candidates, and (3) incumbent partisanship yields reasonably accurate specification of split-ticket voting patterns in gubernatorial, senatorial, and presidential contests; it also suggests at least one source of disparate partisan majorities among state electorates.

Senatorial decision making: II. Prediction

A prediction model for individual voting behavior in the United States Senate is developed. It's predictors are based upon the results of a three-mode factor analysis of 419 roll calls from 1964 through 1969. The model is the multiple logistic and it is tested and validated on 18 roll calls from 1970. The parameter estimation and the validation is conducted through the use of Tukey's Jackknife. Generally, the fit is quite good, despite the fact that the test roll calls were chosen for their difficulty. An especially careful analysis is done on the nomination of G. Harrold Carswell as Associate Justice of the U.S. Supreme Court in which the votes of 94 percent of the voting Senators were correctly predicted.

Ecological Inference and Electoral Analysis

Ecological Inference and Electoral Analysis This paper examines the utility of a relatively new statistical method for studying the relationships between individuals' sociological characteristics and their voting behavior during periods when only aggregate data are available. Almost all of the propositions relating sociological factors and voting behavior are based on the post-I940 period when sample surveys began providing reliable individual-level data. Yet three decades constitute an extremely weak foundation for generalizations and, in order to have as large a data base as possible underlying our generalizations, we must extend our studies of individual voting behavior back into time. Moreover, it is clear that certain large-scale transformations in sociological groups' voting preferences take several decades to work themselves out. If we are to gain an adequate understanding of the dynamics of voting behavior over long stretches of time, we must be able to estimate the individual-level relationships from aggregate data. As Robinson has shown, we cannot directly infer individual relationships from aggregate ones. The fact that a certain set of geographical areas with a certain pattern of demographic and social characteristics behaves in a certain way cannot be construed as meaning that individuals in those areas possessing most or all of the various characteristics all behave in the same manner. Even though, for example, districts with a high proportion of black residents tend to return large Democratic majorities, there is still no airtight logic, no matter how great the suspicion, for inferring that blacks tend to vote Democratic. Fortunately, devices for circumventing the ecological fallacy have been developed.2 The one we will examine here is Goodman's technique, which uses ecological regression coefficients to estimate individual voting proportions.3 To understand how Goodman's technique