Log in Page Discussion History Go to the site toolbox

501 Readings Week 7

From BluWiki

Niemi, R., and Weisberg, H. (2001). “Why is Voter Turnout Low (and Why is it Declining?)” in Niemi, Richard and H Weisberg, eds. (2001). Controversies in Voting Behavior. Washington, D.C: CQ Press.

Niemi and Weisburg (N&W) start by outlining the American electoral puzzle: increasing levels of education, civil rights, women increasingly entering the workplace, and the relaxation of voter registration laws all should have stimulated voter turnout, yet, turnout has steadily decreased since its highpoint in 1960 (22). In examining the extant literature on the topic, they break the work down into three categories of explanatory variables: 1) Individual factors; 2) Societal factors; and 3) institutional factors.

1) Individual factors: Looking at the framework developed by Downs (1957) (amplified by Tullock (1967) and Riker and Ordeshook (1968)), N&W observe that with voting there is a cost (C), voting happens when one feels a candidate will produce a benefit (B) (rational voters only voting when the benefits exceed the costs), times the probability that voting will affect the outcome (P), and any psychological gratification or sense of duty (D) where: PB + D > C. While citizen do not explicitly undertake the calculus, the implication is that, “… low participation and turnout are due to either high costs or low benefits, and correspondingly that declining participation and turnout are due either to increasing costs or decreasing expected benefits,” (23).

In contrast to the rational calculus, however, correlates like education and legal restrictions have moved in the “wrong” direction. Researchers have responded by looking at individual factors that can explain this decline (Starting with Teixeira, 1987). A second approach has been to evaluate the connection between education and voting, and whether increases in higher education lead to higher turnout (Nie et al., 1996). Finding that years of education and the educational environment cancel each other out, questions then surfaced around the application of this finding. A second take on this question argues that turnout is driven by political interest (which comes from factors other than education alone), and greater interest and turnout do not necessarily come from increases in education levels (See Verba, Schlozman, and Brady, 1995 [arguably the most comprehensive study-see N&W, 27]; and Brady, Verba, and Schlozman, 1995). This suggests that the impact of education has been overestimated and the impact of interest underestimated (Brady et al., 1995). Others see a steady turnout and argue that it is caused by problems measuring voter eligibility (McDonald and Popkin, 1999). Still other studies employ selection bias in highlighting how specific factors (like race, gender and marital status) affect voting.

2) Societal factors: Seeing a decline in voting, these works attempt to evaluate “societal” changes that may have caused them. Observing that declines have a strong generational component, for example, these works find it problematic that the individual approach fails to identify what has actually happened (regardless of being able to show which groups occasionally participate and the motivations for withdrawal). For example, N&W observe of Miller and Shanks (1996), a vehement argument that any observed declines are generational (the New Deal cohort never wavered from largely going to the polls), and changes the puzzle. Similarly, Putnam (1995 and 2000) discusses extensively the importance of the component of generational change. As Putnam sees it, turnout is a symptom of broader change on the American political and civic landscape (2000), evident in declines in many forms of political activities.

Neither, however, can say exactly what causes the generational change. This has led scholars like Brehm and Rhan (1997) to return to the individual level in examining concepts of social capital and connectedness. Other works observe that this decline is common in contemporary democracies, and more may be at work than just changes in U.S. society (see Dalton, 1996; Wattenburg, 1997; and Lijphart, 1997). Others still argue that interest has not flagged, but that venues have seen changes. Seen most clearly in Inglehart (1990, 1997), there are parallel shifts between voter decline and increases in “unconventional” or “protest” politics, and that noninstitutional “elite-challenging” form of participation have increased (N&W, 30). Inglehart links these changes to strong generational changes in values, and finds increasing “Post-materialism.” Also included in this literature are works surrounding the effects of negative advertising, the results of which are mixed (N&W, 30).

3) Institutional factors: Turning to institutional factors, this approach takes a historical or comparative view. Acknowledging widespread agreement on low turnout as explained by the legal and institutional environment, at least in part, when compared to other countries. Powell (1986) finds a 14% difference between the US and 20 other democracies in the lack of competitiveness of U.S. elections (10%), and the weaknesses of linkages between parties and groups of citizens (3+%). Franklin argues that the electoral systems proportionality also contributes to differences (Ch5. In N&W), and employs a number of legal factors like compulsory voting, postal voting, and Sunday voting. In this literature, N&W observe that the most cited factor is a legal consideration- voter registration, which is supported by both empirical analysis and the historical record (for more See N&W 31-33).

Putnam, R. (1995). “Tuning in, Tuning out: The Strange Disappearance of Social Capital in America.” in Niemi, Richard and H Weisberg, eds. (2001). Controversies in Voting Behavior. Washington, D.C: CQ Press.

Putnam, in this chapter, takes up declining social capital and its causes. Looking at social life (networks, norms, and trust), Putnam is interested in social capital that serves civic ends. He sees that social trust and civic engagement as strongly correlated and, while causally difficult to sort the direction of the flow, move together. With multiple lines of evidence, Putnam argues that the stock of social capital in America has been shrinking for more than 25 years (40). Putnam, believing that the capital has in fact diminished and that this matters, looks to the causes of this decline in attempting to answer: What can we do about it?

After looking to a number of possible answers, Putnam considers 9 potential influences of the formulation of this capital: 1) Education; 2) Pressures of time and money; 3) Mobility and Suburbanization; 4) The changing role of Women (Happy Women’s Her-story month!); 5) Marriage and Family; 6) the rise of the welfare state; 7) Race and the Civil Rights Revolution; 8) Generational Effects, and 9) the long civic generation (for more see N&W, 43-60). Putnam then reformulates the puzzle and argues that the likely candidate for the anti-civic “X-ray” is television (61).

Rosenstone, S., and Hansen, J. (1993). “Solving the Puzzle of Participation in Electoral Politics.” in Niemi, Richard and H Weisberg, eds. (2001). Controversies in Voting Behavior. Washington, D.C: CQ Press.

Rosenstone and Hansen (R&H) examine the participation puzzle through a lens comparing the solutions, both personal and political. In outlining their book, they look at the empirical evidence for declining participation, some of the proposed solutions, and seek to assess the magnitude of the effects of resources, interests, and strategic mobilization on participation. In the end, they find that, “In every case the changing pattern of mobilization by parties, campaigns, and social movements accounts for at least half of the decline in elector participation since the 1960’s. Explanations that have focused on individual citizens – their demographic characteristics and political beliefs – have missed at least half the story,” (70). The changing patterns seen in the political realm, they argue, give voters fewer: opportunities to take part in elections, chances to share the burden of political involvement, and less opportunity to see the gains of the reward and satisfaction of political activity. They argue that it is not the citizens failing the system, but the system failing them (79)

Putnam, Robert. 1995. “Bowling Alone: America’s Declining Social Capital.” Journal of Democracy. Vol. 6(1): 65-78.

Why should we care about social capital? Norms and networks of civic engagement impact quality of public life and institutions (66). Social capital is important to the resolution of collective action problems, through the development of networks of negotiation and trust, collaboration is promoted (67).

What’s happened to social capital in the U.S.? Despite rising levels of education, American engagement in politics and government has declined and continued to fall over the last 30 years (68). The past three decades have witnessed declining church attendance, union membership, PTA involvement, and membership in and volunteering for civic and fraternal organizations (68 – 69). Familial bonds also represent important forms of social capital. Such things as rising divorce rates and mobility have eroded these bonds, which also impacts American social capital. Amount of Socializing with neighbors has also decreased, but some evidence that socializing with non-neighbor friends has increased. Trust of others has also decreased. This is important because social trust and other types of social capital are correlated: “Members of associations are much more likely than nonmembers to participate in politics, to spend time with neighbors, to express social trust, and so on” (73). At all educational levels, and across many types of memberships, engagement has declined and American social capital has been reduced (72 - 73).

Are these declines countered by increased membership in other new organizations, like the Sierra Club? While these new organizations have political importance, membership in them is of different type than membership in the earlier civic organizations like unions, the PTA, etc. These new organizations do not create social networks, people are tied to the organization but shared membership does not create bonds between members (70 - 71). Additionally, non-profits are not equivalent to secondary associations like the Lions club. Growth of the non-profit sector should not be conflated with growth in new civic organizations (71). What about support groups? May represent an important type of social capital, but they may not create social bridging. These groups seem to not provide community engagement, but the opportunity “for individuals to focus on themselves in the presence of others” (72).

Possible explanations for the decline of social capital: The movement of women into the labor force may have reduced time and energy available for civic engagement. But, the evidence for this is slim (74). Studies have suggested that Americans are more mobile than ever before and this mobility hampers the development of social capital. But, evidence has shown that residential stability has actually risen, not decreased (74 – 75). Other economic and demographic changes have been implicated. Technological changes may be to blame. Our leisure time has been increasingly privatized, especially by the television. This needs further study (75).

Areas for future research: dimensions of social capital, what types of networks and organizations are most effective in fostering the development of social capital? What other sorts of social trends will impact the development or erosion of social capital in America? How will Economic or technological developments impact this? A cost-benefit type analysis of effects of community / civic engagement that also incorporates insights offered by collective action studies. Also need to examine the impacts of public policy on the development of social capital (76).

Aldrich, John H. 1993. “Rational Choice and Turnout.” American Journal of Political Science. Vol. 37(1): 246 – 278.

What: Using rational choice theory to explain voter turn out.

The basic model of turnout: The “fundamental equation of political behavior”: preferences determine behavior (247). Individuals decide a course of action by considering possible outcomes, ranking those outcomes by preference, and considering the probability of a given outcome occurring (247 – 248). The “instrumental” or “investment” theory of turnout: actions are instruments for achieving outcomes. Costs are associated both with voting and with abstention, the “C” term (248). Then, when costs of voting increase, turnout should decline, which is what happens (250). In situations of indifference, even with constant costs associated with voting, those costs will serve as larger barriers to turn out. Put simply, when costs outweigh benefits, turn out will be low. In the reverse, turnout should be higher (250). Voters must also consider the likely actions of others (249).

The Calculus of Voting Model: Anthony Downs adds a preference to the decision calculus. This “D” term is used to account for a voter’s preference for “seeing democracy continue”. Riker and Ordeshook expand on this, terming it “citizen duty” by which the vote is associated with other democratic and political values including an intrinsic value of voting itself. These positive values interact with costs of voting: when values exceed costs, abstention is less preferred (251). Problem with this: if the calculus hinges on the “D” term, this model may have added nothing to the explanation of turnout or to the explanation of how preferences determine action (258). Voting is also part of a collective action (252). Because of this, individuals will also consider the likely behavior of others (252). This “P” term also involves a subjective calculus of the probability of casting a deciding vote (252). Problem with this: no reason for asserting that your vote will make a difference. This may again leave us with a noninstrumental account of turnout (258). Tests of the model find that C, D, and B terms are “strong predictors of turnout”. While tests of the P term have found mixed results (252).

The Minimax Regret Model: Ferejohn and Fiorina argue that people probably find it very difficult to form probability assessments as captured by the “P” term and argue against its inclusion in the model. “Minimax regret” captures the individual’s assessment of likely regret following an action (253). People will choose the action that minimizes the likelihood of various types of regret (256). They argue that this allows a better prediction of turnout than calculus models do (256). Problems with this: the decision calculus involving regret also incorporates a probability estimate (which is paradoxical because “minimax regret” was developed to overcome the probability calculus). Again, the likelihood of an election being decided by one vote is low and the minimax regret model seems to leave this problem untouched. Some voters may vote for a second-choice rather than waste their vote on a sure loser, but this model suggests that voters will always vote for their first-choice (259). “The appearance of wasted voting, therefore, undermines the plausibility of the minimax regret formulation…” (261).

Game-Theoretic Accounts: Based on the notion that assessment of the likely actions of others is done in a strategic fashion. Put simply, these models incorporate the individual’s consideration of the decisionmaking calculus of other individuals. These models involve the same problems as those found in the calculus models because probabilities may be difficult to judge by a given individual (257).

The Rationality of Turnout: Turnout is not a good example of a collective action problem because for most people most of the time, voting is a “low-cost, low-benefit action”. Costs are decreased when multiple contests can be voted on at a single time, and easing registration laws and processes have decreased costs (261). Cost differences between voting and abstaining are not as high as the other models suppose. Information costs are also decreased through “incidental” information, especially in high profile contests, more incidental information exposure is likely to lead to increased turnout. Which we’ve seen, those with more political information are more likely to vote (262). Information shortcuts decrease decisionmaking costs. Benefits are also low; many people don’t perceive much difference between candidates (263). This means that small changes in costs or benefits will alter the turnout decision calculus for most citizens (261). Politicians know this, and this explains the empirical findings (264). First, this helps us understand why so many variables are related to turnout. This may also make it difficult to completely determine who votes and why they vote. Because it is not worth much effort, voters will make errors in their choices (264). Helps us also understand measurement error. All of this points to a “major general conclusion [that] low-cost, low-benefit actions are consistent with the empirical findings” (265). Understanding politicians as strategic actors helps explain increased turnout in close elections, and behavior consistent with the wasted vote hypothesis (266). Because politicians are concerned about close elections, it is to those that resources will be allocated in efforts to stimulate turnout. These resources may result in decreased costs for voters, changing the decision calculus and stimulating turnout regardless of the voters’ perceptions of the closeness of the race (267). Models of rational choice have been developed using this and empirical evidence demonstrates strategic allocation of resources to stimulate turnout (268). Also find that parties, candidates and interest groups make public arguments to inform voters that a vote for a third party candidate is a wasted vote. In essence, strategic politicians inform citizens on how to behave as strategic voters (270).

How can rational choice explain declining turnout? Rational choice accounts tend to be election specific. Nothing in specific elections explains declining turnout and looking at all of them suggests an increasing or erratic pattern in turnout (271). We can explain this by reexamining the “D” term, which can also be expressed as the value associated with voting as a demonstration of party support and affiliation. Doing this, declining party attachment may lead to decreased turnout (272). Declines in turnout might also be explained by perceptions of declining efficacy, enhanced by long-term experience with divided government (273). Further conclusions: “D” term is not “politically inert” and strategic politicians may be able to manipulate the values associated with this term. An expanded “D” term incorporates both long-term and election specific values and allows a new model consistent with explanations for declining turnout (273).

Major lesson: prior work employing and criticizing these models has used an overly narrow interpretation of the theoretical models (274). This narrow interpretation has led to a (false) belief that rational choice fails to explain turnout (276).

Jackman, Robert W. 1993. “Response to Aldrich’s ‘Rational Choice and Turnout’: Rationality and Political Participation.” American Journal of Political Science. Vol. 37(1): 279 – 290.

Aldrich’s arguments and conclusions (that the decision to vote is one with low-costs and low-benefits and so turnout is not a good example of the problem of collective action and that strategic politicians can influence the decision calculus) returns rational choice to Downs’ analysis. Specifically, it provides a reminder of the context of decisionmaking including uncertainty and strategic political considerations and actions (279).

Understanding rationality: need to maintain a distinction between substantive rationality and procedural view of rationality (280). Procedural rationality allows the incorporation of uncertainty and incomplete information – not distinct from the notion of bounded rationality (281, 282). Procedural rationality does not explicitly refer to goals but can incorporate a variety of types of goals (281). Simply, critics of the use of rationality use a narrow interpretation of substantive rationality and misunderstand the importance of procedural rationality in explanations of decisions (283).

Two arguments: First, alternative frameworks are inferior (280). Alternatives to rational-choice frameworks are in actuality, compatible with rational-choice and casting these as such offers an improvement (283). Converse’s model of political learning, recast into a rational-choice framework allows explanation through learning, socialization, strategic politicians, retrospective evaluations, and other political factors and so allows a richer understanding of the “empirical linkage between the age of democratic institutions and partisan stability” (283 – 284). Michels and Edelman both examine the movement of parties towards the center. Both argue that ambiguity is used in order to attract a wider base of support (284 – 285). Understanding this and the impacts on voters through procedural rationality and rational choice offers a richer understanding and allows us to account for a range of other political phenomena (285). Second, the rational choice perspective is optimal for analysis of turnout and of political participation more broadly conceived (280). How can rational choice allow us to understand participation beyond turnout? Activities occurring within institutions are relatively easy to explain through rational-choice perspectives. Critics have argued that less conventional forms of participation (rebellion) are not well accounted for by rational choice because such forms of participation occur under conditions of uncertain and fluid rules (286). But, rational choice accounts can provide understanding of these as long as we do not employ narrow understandings of goals and rationality (287). And, Aldrich’s account fits nicely because studies have found the importance of leadership in these less conventional forms of participation (288).

Brady, Henry E., Sidney Verba, and Kay Lehman Schlozman. 1995. Beyond SES: A Resource Model of Political Participation. American Political Science Review 89(2): 271-294.

Stemming from a query of why people don’t participate in politics the authors speculate that the answers would be “because they can’t, because they don’t want to, or because nobody asked” (271):

- From the first (“can’t”) the authors identify resources (namely time, money, and civic skills) as important factors.

- From the second (“don’t want to”) the authors identify engagement as important (“lack of interest in politics, minimal concern with public issues, a sense that activity makes no difference, and no consciousness of membership in a group with shared political interests” (271)).

- The third (“nobody asked”) concerns “isolation from recruitment networks” (271).

While all three are potentially important the authors focus most on resources in this article.

The authors seek to move beyond arguments about political participation based upon socioeconomic status (SES) components such as education, income and occupation. Through their study the authors intend to explore how resources are related to SES and also to political activity.

Methodologically the authors believe that resources may offer a more secure foundation upon which to base their claims and conclusions. Resources are easier to measure that psychological engagement. Resource models of political participation are linked to two other strands of academic work: stratification theories from sociology and individual choice perspectives from economics.

The Citizen Participation Study (272)

- Survey data used “from a large-scale, two-stage survey of the voluntary activity of the American public” (272). 1st stage: random sample telephone survey of 15,000 people (July-Dec. 1986). 2nd stage: spring 1990, in-person interviews of 2,517 chosen from the original survey population with activists and African-Americans and Latinos preferences disproportionately in the selection.

- Voluntary activity in the study was identified in politics, but also churches and organizations (272). Political participation construed “broadly” – factors considered: voting, electoral activity, contacting public officials, attending protests, formal or informal involvement in local issues.

Defining and Measuring Political Resources (273)

- Time and Money: “Time in the service of political action” was measured by the authors through many aspects including (but not limited to) working in a campaign, letter-writing, attending meetings. Money in donations. While time is more limited and more equally distributed (same number of hours in a day for everyone), money is less equitable.

- Civic Skills: Measured several ways: (1) education – on the basis that education and organizational skills are acquired in school – including aspects such as length/level of education, participation in school government, language at home, and vocabulary; (2) measuring activities from work and voluntary organization that may contribute to the development of civic skills.

The Distribution of Political Resources

- The authors claim that the “presence or absence of resources contributes substantially to individual differences in participation” (274). Resources are not equally distributed, with some socioeconomic groups “better endowed,” however the stratification of resources varies.

- Money and Time: While income is easily measured, free time is not. However, it is generally assumed that those with more money can use their money to gain themselves more free time and therefore those with more money also would be expected to have more time. This does not play out in the authors’ study: “free time and SES are unrelated” (274). Education and income are strongly related, but “there is no such consistent pattern of stratification when it comes to time” (274 – strongly recommended to view Figure 1 on 275). Free time affected by “’life circumstances.’”

- Civic Skills: “Civic skills are, in general, more likely to be possessed by the socioeconomically advantaged” (275). Higher levels of education correlate with speaking English at home, involvement in school gov’t, and better vocabulary. Acquisition of civic skills from work, church and organizations varied in stratification, but generally those with low educations levels had little opportunity to develop civic skills, while those with high levels of education had many opportunities (275 – see also figure 3 on 276). The three institutions varied in how opportunities for civic skill development was achieved. Such opportunity requires “involvement in the institution and a setting that provides the chance to practice some skills” (275). The workplace offers the most opportunity, compared to church and organization.

- The Resource Model: (Figure 3 on 277) “Involvement in institutions – first in school and later on the job, and in church – provides opportunities to acquire the resources relevant to political activity.” (276) Civic skills are central to the model.

Developing Civic Skills: A Learning Model

- Three obstacles to asserting that civic skills foster political participation: (1) Measurement Problem – having the indicators measured in education doesn’t necessarily mean capacity to participate politically, nor do measures of opportunities to develop skills in organizations (skill-acts) necessarily measure civic skills themselves – (2) “correlation is not enough” (277) – spurious correlation may be occurring between civic skills and skill acts – (3) “locus of development problem” (277) – skill acts may develop civic skills, however possessing civic skills may lead to undertaking skill acts.

- Concerning measurement, the authors identify that their measures have face validity. Also modeling may help to resolve questions abot measurement as well as the other two obstacles (Figure 3).

Political Participation and Resources

- Estimating the Model: Using ordinary least squares (OLS) to “regress political acts on free time, family income, skill-acts, language abilities, and formal educational experiences” (279). To address problems of endogeneity the use of a two-stage least squares (2SLS) may be necessary.

Political Resources and Specific Political Acts

- The authors disaggregate political acts to determine whether different acts may be linked to different resources. Three kinds of political acts: voting, investment of money and investment of time.

- Voting: Authors believe a disproportionate amount of attention to education may have been paid in examining proclivity to turn out and vote. While education is important, political interest is very important to voting.

- Money: “The major determinant of giving money is having money” (283). Years of education is also important, but not the other resources of free time or civic skills or to much degree political interest.

- Time: Two-stage process appears to take place. Political interest and resources impact “decision to participate, but constraints on free time control the amount of time-based political activity once this decision is made” (284). The Impact of Resources on Political Activity: The impacts of Church Involvement, Income and Free time are briefly explored to offer some insight into the relation of factors (285).


- “Motivations such as interest are not enough to explain political participation.”

- “The resources of time, money and skills are powerful predictors of political participation in America.”

- Resource model moves beyond SES models in that: (1) by looking at a more general level and specifying resources that can be derived from socioeconomic position the resource model establish mechanisms that can link SES to participation and (2) by focusing on resources not influenced by SES understanding disparity along other divisions may be possible.

- Brief summary of findings: “political interst is especially important for turnout; civic skill, for acts requiring an investment of time; and money, for acts requiring an investment of money.” (285)

Kahn, Kim Fridkin, and Patrick J. Kenny. 1999. Do Negative Campaigns Mobilize or Suppress Turnout? Clarifying the Relationship between Negatively and Participation. American Political Science Review 93(4): 877-889.

Hahn and Kenney explore the effect of negative campaigns upon voters and voter turnout for elections. While studies have been done that indicate negative campaigning turns off voters and negatively effects voter turnout, other studies have indicated that negative campaigning enhances voter turnout. Seeking to better understand both negative campaigning and voter turnout the authors’ study explores campaign tone and voter turnout – while providing a better conceptualization of campaign tone and addressing criticisms of previous research.

The Tone of Campaign Messages

- The authors begin with “the simple premise that negative information is helpful and motivates participation as long as it addresses relevant topics and is presented in an appropriate manner” (878). This premise is supported by research that indicates:

o Because most everyday information is positive, negative information “is more unique, more salient, and more memorable” (878) and could be seen as more exciting or memorable.

o Negative information in campaigns may point out risks, that may engage the risk-adverse.

- Negative messages are therefore expected to “generate more interest and involvement in campaigns” (878) and will instigate higher turnout on the day of elections.

- On the contrary, negative messages that concern “questionable topics” and/or are excessively biased in their presentation alienate voters and decrease turnout. This concerns negative messages that:

o Contain campaign topics that are irrelevant or inappropriate

o Criticize opponents “in an accusatory and ad hominem manner” (878) Design

- The authors identify aspects of their study that “allow us to examine the relationship between the tone of the campaign information and voter turnout more precisely than past studies”: (878)

o Used survey responses from the 1990 NES Senate Election Study, rather than aggregate data that may have differing levels of advertizing tone, turnout. Such factors are held constant in this study.

o Focusing on Senate campaigns allows the authors to use a larger number of cases than presidential elections would allow and to better control for variation among the elections themselves because Senate elections vary less in money spent, quality of candidates, content and tone of messages, and media reporting.

o Measurement of tone of information from both candidates campaign’s and press coverage. Many studies only focused on info from campaigns, not info from media – which studies have shown is a major source of voter information about campaigns.

o By focusing on 1990, a non-presidential campaign year, the authors sought to “avoid contamination by the presidential campaign” (879).

The Tone of Candidate Messages

- The authors measured political tone of television advertisement through content analysis of commercials.

- T.V. commercials were selected because of their centrality in Senate campaigns (largest campaign expenditure), widespread use, and greater effect on changing voter opinion.

- Tone was estimated for each commercial on a scale of 0 (no negative message) to 2 (major emphasis on negativity), and validity of that measurement was checked through correlation with indicator from content analysis of CQ.

The Tone of News Coverage

- Newspapers were chosen to represent media coverage due to studies that demonstrate they devote resources and space to campaigns than television media and that voters rely more on newspapers for local and state elections.

- Content analysis of the press coverage of states with a Senate election in 1990 was covered, using the largest circulating newspaper from each state.

Identifying Mudslinging

- In the analysis of media sources because of the subjective nature of “determining what ‘goes too far’” (881), those managing campaigns were asked to identify mudslinging races and excessive negativity in the media.

- In measuring the relationship between negative info and turnout, the authors controlled for 5 forces that have been indicated as explanatory in influencing the likelihood of voter turnout: (1) closeness of the election; (2) characteristics of concurrent campaigns; (3) attitudes about the candidates and the Senate campaign; (4) demographic characteristics; and, (5) psychological involvement in politics (882, detailed discussion of these factors pp 882-883).


- “People distinguish between legitimate and tempered criticisms, on the one hand, and acrimonious and unjust criticisms, on the other hand” (884):

o Substantive and reasoned criticism is useful to voters, and increases voter turnout

o Mudslinging alienates voters

- Traditional correlates of turnout – education, age, interest in campaigns, and strength of partisan attachment – strongly shaped the likelihood of voting.

- Attitudes about the candidates themselves influenced voter turnout

- Contact with the candidates is important, “significantly elevates turnout” (884).

- “competitiveness does not directly influence voter turnout” (884), in close races candidates will “step up campaign activities” which can increase familiarity and contact with candidates.

The Differential Effect of Negativity on Turnout: The Importance of Voter Characteristics

- Campaign environment does not affect everyone equally:

o Core/habitual voters are immune

o “Campaigns that provide serious, critical, and thought-provoking debate in an interesting and germane manner may engage and activate uninterested and distracted citizens” (885).

o Independents and political novices are more likely to be influence by the tone of a campaign

o Political profile important


- “Negative information does not have a uniform effect” (887)

- “Critical campaign messages are especially useful when they appear relevant and are delivered in an appropriate manner” (887)

- “Responses to the negativity of campaigns depend on political predispositions” (887).

Saunders, Kyle L. and Alan I. Abramowitz. 2004. Ideological Realignment and Active Partisans in the American Electorate. American Politics Research 32(3): 285-309.

Saunders and Abramowitz investigate the ideological characteristics of active partisans in comparison to less active and nonpartisan voters. In an investigation of the political beliefs of active partisans, Saunders and Abramowitz explore four hypotheses about ideology among partisans:

- (1) That “active partisans, especially active Republicans, perceived larger differences between the policy positions of the parties during the Reagan era and especially during the Clinton-Gingrich era than previously;”

- (2) that “ideology played a larger role in motivating active Republicans during the Reagan era and especially during the Clinton-Gingrich era than previously”;

- (3) that “ideology played a larger role in motivating active Republicans than active Democrats during the Reagan era and especially during the Clinton-Gingrich era than previously”; and,

- (4) that, “during the Clinton-Gingrich era, active Republicans were farther to the right of the electorate than active Democrats were to the left of the electorate on most policy issues” (289-290).

Data and Measures

- The authors used data from the NES surveys from 1972-2000, and a cross-section from 1998.

- Measurement of active partisanship was done through NES questions on election-related activities and the standard Party ID Scale

- Policy questions were used to create a liberal-conservative index.

- The existing Liberal-conservative scale from the NES was used to measure ideological orientation – those who did no self-ID were given a middle score.


- In exploring whether “ideological views of active partisans differed systematically from those of inactive partisans and the overall electorate” (291) and the above four hypotheses the authors began with an examination of correlations between activism and policy extremism from 1994-2000 and find “strong support” for the hypothesis about the role of ideology motivating activism.

- Because correlation does not mean causation the authors then undertake a Poisson regression with the dependent variable being an activism/participation index. Ideology proved stronger predictor of activism among Republican identifiers than Democratic.

- In exploring that both Democrats and Republicans hold views on policy issues more extreme than the average citizen, with Republicans having greater separation than Democrats, the authors first examined the issue of Clinton’s impeachment (which strongly supports their hypothesis) and then seven other policy issues (generally consistent with their hypothesis).

- The influence of ideology on activism increasing during the Reagan and Clinton-Gingrich eras, Poisson regressions of “participation among Democratic and Republican identifiers” during four periods – Nixon-Ford, Carter, Reagan-Bush, and Clinton (301) – was tested along with other variables. While the influence of ideology on activism was greatest for Democrats during the Nixon-Ford era, they were highest in Clinton years.

Discussion and Conclusion

- “The evidence presented in this article has demonstrated that ideological convictions can be an important motivation for participation in electoral activities that go beyond the act of voting.” (305)

- “The influence of ideology as a motivation for activism varies over time and between parties depending on the salience of the ideological cues provided by a party’s candidates and office-holders.” (305)

- Reagan’s election and emergence of Newt Gingrich “contributed to an increased emphasis on conservative ideological appeals by Republican candidates and officeholders” (305) and ideology became a “more important motivation for activism among Republicans.”

Burns, N., Schlozman, K., and Verba, S (1997). “The Public Consequences of Private Inequality: Family Life and Citizen Participation.” The American Political Science Review, Vol. 91, No. 2 (Jun., 1997), pp. 373-389

Burns et al., after developing the largest civic participation study to that point (see, N&W 27), look at the consequences of private inequality for the public. More specifically, they employ 380 telephone surveys of married couples to examine the implications of domestic inequalities for political activity. More specifically, this piece is an attempt to examine one common theme in contemporary feminist philosophy: women cannot be equal in the polity until they are equal in the home. After outlining the literature on the topic and acknowledging the weakness in their specificity, they also test resource constraints, and family social structure. After citing studies illustrating inequality and gender roles in politics, Burns et al. look at multiple lines of inequality and attempts to understand the link between the family and politics in assessing the marital effects of domestic inequality on participation (for more on the variables employed and their analysis see 376-383).

They conclude that their data support the belief that the public involvement of family members is affected by the private life of the family. They find that both married partners are affected by domestic arrangements. A husbands control over economics, decision-making, and social status is more relevant for men that women. The implication is that this increases the aggregate husbandly participation. In other words, these factors do not depress wives political activity, but “boost” the activity of their husbands (384).

Site Toolbox:

Personal tools
GNU Free Documentation License 1.2
This page was last modified on 8 March 2010, at 21:43.
Disclaimers - About BluWiki