CONTEST FRAMING AND ITS EFFECTS ON VOTER (DE)MOBILISATION
NEWS EXPOSURE AND ITS IMPACT ON VOTING TURNOUT IN THE 2008 AUSTRIAN
ELECTIONS IRIS HÖLLER
GÜNTHER LENGAUER
Abstract
This article investigates the impact of news exposure on voting turnout in the 2008 Austrian elections by specifically focusing on horse race, conflict and drama levels to capture the nature and effects of contest framing in the campaign coverage. This study rests on the analytical linkage of extensive content analyses of newspaper and TV news coverage and a representative post election survey comprising the Austrian electorate. This investigation first contrasts the magnitude of contest framing in tabloid and quality news and then applies logistic regression analyses, outlining its (de)mobilisation effects on voters to answer the guiding questions: To what extent is the election campaign portrayed as a contest and how does this affect the (de)mobilisation of the electorate? Thereby, we contrast the effects of sheer news exposure with the impact of exposure levels regarding contest framing by the media to learn what is more effective. The findings firstly show that tabloid news is more contest-oriented in their reporting than quality news. Secondly, dissonant to our expectations, we find that whereas general news exposure holds no mobilising power regarding the Austrian electorate, horse race framing by the media even shows a reversed mobilisation
effect by turning voters off.
Günther Lengauer is Principal Investigator of the Austrian National Election Study (AUTNES) and Senior Post-Doc Researcher at the Department of Political Science at the University of Innsbruck; e-mail: guenther.lengauer@uibk.ac.at.
Iris Höller is Pre-Doc Researcher at the Department of Political Science at the University of Innsbruck; e-mail: iris.hoeller@uibk.ac.at.
rN ■
ro r^.
si ü
o
rN i—i o rN
o~>
£
£ 3
a
0
£ *
1 *
(R 0 C
! m
Introduction
Over the last decades voting behaviour has become increasingly volatile (Dalton and Wattenberg 2000; Norris, LeDuc and Niemi 2010). Against this background, it is widely assumed that rather instant campaign and media effects gain in importance as determinants of electoral behaviour (e.g., Ridout 2004). Simultaneously, mass media have impressively emerged and established themselves as primary sources of political and electoral information (Norris 2000; Plasser and Plasser 2002). In line with these driving forces, the focus on the potential impact of news coverage on voters has moved to the center of political and communication research.
Voting turnout "is mainly about how elections appear to people" (Franklin 2004, 6). Thereby, the public appearance of contemporary elections is mainly coined by depictions conveyed and transformed by the media, above all mass media. Consequently, media use, its frequency and the associated portrayals of electoral campaigns may critically contribute to political perception and behaviour of the electorate. Thereby, earlier research has taken two cardinal perspectives: The first points to media's contribution to political cynicism, alienation, apathy and demobilisation of the electorate (e.g., Cappella and Jamieson 1997; Patterson 2002; Delli Carpini 2004), and the second camp antithetically suggests that media exposure positively contributes to democratic engagement and political activism (e.g., Norris 2000; 2006; Adriaansen, Van Praag and De Vreese 2010). Thereby, a vast body of existing evidence is exclusively based on general media exposure, without considering news content (e.g., Norris 2006). Referring to the complexity of the interplay between communications and citizens' involvement in political and civic life, Delli Carpini, however, notes that "the impact of the media is tied in part to the tone and content of the information provided" (2004, 398). Consequently, to properly examine convincing explanations of news effects on political attitudes or behaviour, it is necessary to additionally measure actual media content parameters and link them to the intensity the voters are exposed to this content (e.g., De Vreese y and Semetko 2004; Slater 2004; Elenbaas and De Vreese 2008).
To conceptualise hypotheses predicting news effects in election campaigns, it seems particularly fruitful to investigate media framing, in particular the impact of generic media frames (De Vreese 2005a). Generic frames such as "conflict" or "horse race" have not only been shown to be relevant characteristics of contemporary media portrayals of politics, but also may therefore impinge on the perception of election campaigns. Most recent framing research shows that not only issue-specific framing might have an impact on voting behaviour, but also generic framing (e.g., De Vreese 2005b; Schuck, Vliegenthart and De Vreese 2011). ^	In the context of generic media framing, we can state that little scientific atten-
tion has yet been devoted to contest framing in explaining the electorate's turnout to vote. In our conceptualisation the magnitude of contest framing is determined by the levels of dramatisation, confrontation and horse race in electoral reporting. This study particularly examines the relationship between exposure levels to contest framing in newspapers and on TV news and voting turnout in the 2008
G
C >
Austrian Parliamentary Elections. In particular, we draw on an extensive content analysis of the newspaper and TV news coverage of the 2008 Austrian elections ^ and on a representative post election survey among Austrian voters. Thereby, we
contrast the magnitude of contest framing in tabloid and quality news and then apply logistic regression analyses, outlining its (de)mobilisation effects on voters to LO answer the guiding questions: To what extent is the election campaign portrayed as ^ a contest and how does this affect the (de)mobilisation of the electorate? Thereby, we contrast the effects of sheer news exposure with the effects of concrete contest framing exposure levels to learn what is more effective and what appears as a more reliable measure of news effects, general media exposure or exposure to specific media content.
The Predictors of Voting Turnout
When investigating media effects on turnout, first a fundamental set of robust and reliable predictors of voting turnout beyond media-related factors that put them in a larger explanatory context needs to be identified and extracted from earlier research. The general question to be examined is why some individuals vote and others do not. As a crucial reason explaining non-voting, Blais (2007, 631) stated plainly "because it does not matter." The relevance of going to the polls is not exclusively driven by media-related influences. But in conjunction with individual political predispositions and sociodemographic characteristics, we suppose media exposure and media content to have a significant impact on the perception of whether the election matters to the people and whether the electorate believes that there is something at stake, as voters are predominantly informed by the mass media about politics and elections. This is the point of departure for our analysis.
Previous research has revealed a number of individual- and system-level factors involved in affecting voter turnout (Wattenberg 2002; Franklin 2004; Seeber and Steinbrecher 2011). Sociodemographic characteristics and individual predispositions are predominantly relevant for this investigation that implements media exposure variables to explain turnout. For example, Valentino, Beckmann and Buhr (2001) have shown in an experimental design that the strength of media exposure effects might be affected by levels of sophistication or political involvement. Regarding political attitudes, Brady, Verba and Schlozman stated that "what matters most for going to the polls are not the resources at voters' disposal but, rather, their civic orientations" (1995, 283). Earlier research has shown that amongst the most reliable predictors of voting turnout are sociodemographics of age, education, income, or gender, and civic orientations such as political interest or party identification (Franklin 2004; Norris 2004; Seeber and Steinbrecher 2011). To establish a comprehensive research design, we extend the list of well established predictors of voting turnout regarding sociodemographics and civic orientations with individual media exposure variables that are the center stage in our explanatory models of voting turnout in the 2008 Austrian elections to filter out media-related effects on turnout.
Media Framing and Its Effects
The effects of media content, particularly media framing, are increasingly at the center of interest in political and communication science. Framing research basically differentiates between issue-specific and generic frames. The latter are at the focus of interest here, as they "transcend thematic limitations and can be identified in relation to different topics, some even over time and in different cultural
contexts" (De Vreese 2005a, 54). That implies that generic frames are particularly applicable for investigating election campaigns in their entirety and are not limited to specific debates or actors.
Earlier research on generic frame-related effects on voter (de)mobilisation has predominantly focused on strategy and conflict framing. Thereby, mostly political cynicism was applied as the central dependent variable of interest. The vast majority of studies on strategy framing comes to the conclusion that high levels of strategic framing correlate with high degrees of political cynicism (Cappella and Jamieson 1997; Elenbaas and De Vreese 2008). This corrosive relationship between media coverage and political orientations has to some extent been qualified by studies showing that framing effects might be moderated by political predispositions or sociodemographic characteristics of the recipients (Valentino, Beckmann and Buhr 2001; De Vreese 2005b). For the 2000 Danish referendum campaign on the introduction of the Euro De Vreese and Semetko (2004) showed by combining a two-wave panel study and a content analysis of national news that turnout was unaffected by the level of strategic news. Recently, Schuck, Vliegenthart and De Vreese (2011) found that exposure to conflict framing had a positive effect on the intention to vote in the 2009 European Parliamentary Elections. In contrast, they also verified that horse race framing, operationalised as references to parties' standing in the polls, had no statistically significant effect on the voting intention. These conflicting conclusions based on inconclusive empirical evidence might be partly due to methodological and operational inconsistencies in previous research. In total, empirical evidence on effects of generic news framing on voter mobilisation remains fragmented and rather inconclusive.
To refine the investigation of generic news framing effects, our analysis focuses on a set of generic news frames that are to model the contest aspects of campaigns. By referring to Blais, who stated that "turnout is higher when the election is perceived to be important and close" (2007, 633), we assume that turnout is related to whether the people perceive that there is something at stake and that their voting y decision is particularly relevant, as the election is portrayed as being contested and thereby their vote may make a difference.
Contest Framing in the News
Earlier definitions of contest framing by the media mostly referred to single, unidimensional indicators. For example, Cottle and Rai (2006, 172) restricted their "contest frame" to conflictual news stories that are framed in terms of binary opposition. And Hanggli and Kriesi (2010, 144) perceived and operationalised "contest frames" as reports that "focus on the actors involved or on the contest as such" and defined contest as absence of substance (issue-related discussions). This approach is vastly equivalent to the horse race dimension of news reporting. To transfer and integrate these isolated elements and rather narrow definitions of contest framing, in our definition we expand horse race and conflict by the dimension of dramatisation to achieve a more comprehensive and triangular framework of contest framing in political news. Consequently, we differentiate between levels of confrontation ("contest of ideas"), dramatisation ("contest of emotions"), and the level of horse race ("contest of odds") to capture (a) the salience of contest framing in electoral news coverage and (b) its effect on voting turnout.
c >
II
Figure 1: Triangular Framework of Contest Framing in Election News
/ Dramatisation	
l "Contest of Emotions"	
// \ .223** .247*	
/ Horse Race	Confrontation \
1 "Contest of Odds" I .096** j	"Contest of Ideas"/
Note: ** Spearman's rho correlation coefficients are significant at the 0.01 level.
The dimensions of confrontation, dramatisation, and horse race are not necessarily mutually exclusive, rather they variably emerge simultaneously in electoral news reports and thus may strengthen or weaken the contest nature interactively (indicated by areas of overlap in Figure 1). However, they are still distinct, as they address and add different aspects of contest in the generic framing of electoral news. For example, depictions of conflict or horse race are not necessarily also framed as dramatic and vice versa. Although there exists significant statistical overlap between the selected indicators of contest framing (indicated by the given correlation coefficients in Figure 1), their salience in news reporting may vary significantly. Consequently, we perceive contest framing as a mix of variable levels of dramatisation, confrontation and horse race that simultaneously coin news coverage and constitute different levels of contest framing.
The introduced indicators of contest framing are defined as bipolar continuums comprising also their conceptual antitheses. The level of confrontation ranges from conflict to consent, the level of dramatisation ranges from dramatised/arousing/ emotional to sober/unemotional reporting, and the level of horse race comprises the spectrum from sheer horse race to substantial policy discussions.
As our analysis comprises tabloid and quality news, we initially compile and contrast contest framing levels along these two types of media outlets to outline an empirical baseline regarding the salience of contest framing in the electoral coverage. By doing so, we refer to tabloid news as the tabloid press and commercial TV news, following Dahlgren who coined the term "tele-tabloids" (1996, 60) for private TV news. Equivalently, as quality news, we define the coverage of quality papers and public service TV news. First, we expect tabloid news to be more permeated by contest framing, as horse race, drama and conflict are prominent and frequently cited characteristics of market-driven journalism representing the tabloidisation of politics (McManus 1994; Esser 1999; Sparks and Tulloch 2000). The phenomenon of tabloidisation is primarily linked to the logic of tabloid and commercial news. A Swedish study conducted by Stromback (2008) showed that tabloid news (commercial television news and tabloid papers) tended to frame politics more as a game and less as issue-centred debates than quality papers and public service
TV news. Additionally, dramatised scandal framing was much more prominent in tabloid than in quality news. Strömbäck and Van Aelst (2009) reported for their comparative investigation of election coverage in Sweden and Belgium that the gaming and horse race aspect were significantly more common in tabloids and on commercial TV news than in quality newspapers and on public service TV. From these preliminary results, we expect tabloid news to be more contest-centred than quality news in their electoral coverage (Hypothesis 1).
Political News Exposure. Turning to our effect testing models, we start with the implementation of the individual, however general, exposure to political news. Overall, there exists a rather established stimulating relation between public affairs exposure by the media and political participation. For example, Norris (2000) reported a positive nexus between TV and newspaper usage in European Elections and voting turnout. De Vreese and Boomgaarden (2006, 331) showed a positive effect of news exposure on the turnout intention regarding an EU referendum on enlargement in the Netherlands and Denmark and stated that "this suggests that the relationships between news watching and knowledge and participation are rather more positive than negative. Accordingly, it is less consequential whether people watch the news on a public or a commercial station, but rather whether people do watch the news at all or turn to entertainment programming." Based on this knowledge, we initially postulate the following hypothesis regarding the effects of sheer news exposure: The higher the exposure to political news, the higher the likelihood to turn out to vote (Hypothesis 2). Turning to effects of exposure to concrete media content, we now focus on our three earlier introduced dimensions of contest framing.
Confrontation - "The Contest of Ideas." The dimension of confrontation displays the level of contest of ideas by referring to conflict or consensus in the depiction of politics in election campaigns. De Vreese (2006) gives insights that news foci on disagreement, conflict and diverging opinions and positions may hold mobilising y power. Controversy and conflict framing heat up the contest and may boast the notion that something is at stake, as conflictual, contesting positions emerge. A story is considered confrontational when controversies or conflicts are explicitly stressed and these references are more salient than references to consensus and cooperation. Conflict-centred reporting may relate to the depiction of dissenting or clashing sides, disputes, controversy, disagreement, discordance or confrontation. In contrast, the consensus dimension comprises accordance, consonances, conformities, dispute settlements, agreement, willingness to cooperate or compromise, approval or reconciliation. ^	Previous empirical research on news framing has impressively demonstrated
that conflict is a dominant and vital media frame when displaying politics (Neu-man, Just and Crigler 1992; Strömbäck and Dimitrova 2006; Canel, Holtz-Bacha and Mancini 2007). Conflict is further inherent to politics and embodied in political reasoning and decision making (Lupia, McCobbins and Popkin 2000). Schuck, Vliegenthart and De Vreese (2011) have shown for the 2009 European Parliamentary
0
c >
Elections in 21 member states that conflict framing held a mobilising effect on turnout. Consequently, we assume that conflict, as a contest of ideas that is inherent to p^ democratic decision making may have positive effects on the mobilisation of the
electorate. The more voters are exposed to conflict framing in the news, the more likely they perceive the campaign as contested and relevant and consequently the C^ more likely they turn out to vote (Hypothesis 3).	^
Dramatisation - "The Contest of Emotions." Paletz and Entman (1981, 17) concluded that "drama is a defining characteristic of news. An event is particularly newsworthy if it has some elements of a dramatic narrative." Confirmingly, Bennett (2009, 40) depicts dramatisation as one of four major "information biases that matter" in contemporary political journalism. He states that "news dramas emphasize crisis over continuity (...). News dramas downplay complex policy information" (2009, 41). In this sense, dramatisation can be interpreted as a means of displaying a "contest of emotions." We take the term "immediate emotion" (Bennett 2009, 42) as the central characteristic of our applied definition of dramatisation. Consequently, dramatisation refers to the nature of emotionalisation and arousal within election campaigns that primarily highlights that there is something at stake by triggering emotions, anger, excitement, accentuating dramatic consequences, polarisation, focusing on appealing, agitating or escalating and arousing depictions. On the other end of the continuum, non-drama is characterised by neutral, sober, cool and distant, unemotional, de-escalating or not agitating depictions of politics. Drama may signify that there is something at stake as well as the closeness of the race. Derived from that we postulate that the more voters are exposed to dramatised news on politics, the more likely they perceive the campaign as contested and relevant and consequently the more likely they turn out to vote (Hypothesis 4).
Horse Race - "The Contest of Odds." One of the most prominent indicators of the contemporary media logic that can be extracted from previous research is the so-called horse race frame (Stromback and Dimitrova 2006; Schuck, Vliegenthart and De Vreese 2011). This notion shares a great deal of common ground with the sometimes even changeable applied concepts of game framing (Patterson 1993; Esser and Hemmer 2008) or strategy framing (Cappella and Jamieson 1997; Valentino, Beckmann and Buhr 2001). The area of overlap among these concepts, on which we focus, describes a portrayal of politics in a depoliticised way, lacking policy relevance and substance. Thereby, politics is portrayed as a competitive game or horse race by mostly applying sports metaphors of (predicted) winners and losers concerning the protagonists' odds and projections of the outcome. We perceive the contest of odds as an integral dimension of contest framing. This kind of media framing with above average audience appeal (Iyengar, Norpoth and Hahn 2004) has usually been linked to cynical and disaffected attitudes towards politics and election campaigns (Cappella and Jamieson 1997; Valentino, Beckmann and Buhr 2001). Schuck, Vliegenthart and De Vreese (2011), however, reported that references to opinion polls predicting the outcome had no significant effect on turnout in the 2009 European Parliamentary Elections. Inconclusive empirical evidence might be due to inconsistencies in operational definitions of earlier research. We interpret horse race as a framing device that primarily highlights the contest character of election campaigns. As such, we hypothesise that horse race framing may function as a mobilising factor which does not only activate the contest notion but also may encourage voters to go to the polls by suggesting that every single vote may make a difference. Consequently, we postulate that the more voters are exposed to horse
race framing, the more likely they perceive the campaign as contested and relevant and consequently the more likely they turn out to vote (Hypothesis 5).
Study Design and Method
In order to establish a direct link between political news content and individual exposure to this information, we utilised a post election survey that was conducted as face-to-face interviews (CAPI) under the auspices of the Austrian National Election Study (AUTNES) and comprised 1,165 eligible Austrian voters. A well documented problem of turnout questions in election surveys is over-reporting and turnout bias (e.g., Duff et al. 2007). The results of our study are not seriously biased by over-reporting, as the surveyed turnout lies by 85.4 percent, which is equivalent to a rather moderate level of over-reporting of 4.5 percentage points.
The applied media content analysis comprised four major Austrian daily newspapers with the highest readership in the tabloid (Kronen Zeitung, Österreich) and the quality press (Der Standard, Die Presse) segment. Additionally, it also covers the evening TV newscasts of the public service (ORF Zeit im Bild) and private sector (ATV Aktuell) with the highest national viewership (Plasser and Lengauer 2010). The analysis included the total coverage on Austrian domestic and foreign politics. Thereby, the selection criterion was exclusively topic-driven and no sectional restrictions were applied. This content analysis focused on the final six weeks of the 2008 Austrian election campaign (TV news: Sunday, August 17 to Saturday, September 27, 2008; Newspapers: Monday, August 18 to Sunday, September 28, 2008). Election Day was Sunday, September 28, 2008. In total, 4,712 news items have been identified and coded. A total of 2,281 (48.4 percent) of all news items referred to the tabloid news segment (Kronen Zeitung - 1,174 items; Österreich - 979 items; ATV Aktuell - 128 items). Another 2,431 items (51.6 percent) have been published by national quality news (Der Standard - 1,063 items; Die Presse - 1,101 items; ORF Zeit im Bild - 267 items). In a series of pre-tests, intercoder realiability and validity of the data were tested. The validity test showed an average researcher-coder y concordance of 0.82 for the variables utilised in this analysis. The average Holsti measure for intercoder reliability of the applied framing variables ranged from 0.71 (dramatisation), 0.76 (conflict) to 0.78 (horse race) and leveled off at 0.75 on average. The media content analysis was conducted by the Media Analysis Team of the Austrian National Election Study (AUTNES).
Measures
The applied bipolar coding strategy (conflict vs. consensus; horse race vs. policy; drama vs. non-drama) not only allowed to question the orthodoxy of most unipolar approaches (e.g., measuring levels of conflict only), but also enabled to expand the focus to its antithesis. Levels of confrontation, horse race, and dramatisation were coded on a tripartite Likert-scale likewise ranging from -1 (predominantly consensus-centred; policy-centred; unemotional/sober), 0 (ambivalent or not applicable) to +1 (predominantly conflict-centred; horse race-centred; dramatised). These absolute measures of the three contest framing indicators were the point of departure for
...........................................................
establishing a measure that weights actual media content with individual exposure to this information. Thereby, the individual exposure to political news regarding the qq six analysed media outlets was compiled for each respondent and transferred into
a score ranging from 0 (never) to 1 (on a daily basis). This resulted in an additive index that represents the individual news exposure score (INES). For our regression models including contest framing scores, we computed additive frame exposure indexes by weighing the outlet-bound degree of confrontation, dramatisation, and horse race (shown in Table 1) with the usage of these media outlets. This procedure provided a single and individual measure for the actual exposure to confrontation (individual confrontation exposure score - ICES), horse race (individual horse race exposure score - IHES), and dramatisation (individual dramatisation exposure score - IDES). Applying such an integrative measure controls for the individual usage of tabloid or quality news and its associated, diverging levels of contest framing in the coverage on the micro-level (the media outlet level).
Hypothesis Testing Logic
Starting from there, we computed binary logistic regression models with turnout (yes/no) as the dependent variable. Our modelling follows a step by step procedure. Our basic model (1) explains voting turnout by including sociodemographic characteristics and individual political predispositions only. Sociodemographics such as age, gender or education have long been known to affect turnout (Lazars-feld, Berelson and Gaudet 1944; Wolfinger and Rosenstone 1980; Rosenstone and Hansen 1993; Franklin 2004; Blais 2007). Based on these insights we include age (measured in years), gender (male, female) and education (dichotomised in at least general qualification for university entrance and lower levels of education) in our basic explanatory model.
Primarily following the socio-psychological approach, we also identified central, individual civic orientations that may crucially affect the propensity to vote (Campbell et al. 1960; Aldrich and Simon 1986). Individual predispositions of voters relate to the psychological engagement and suggest that it does matter whether people care about politics. It is postulated that the higher the affirmative ties to the political system and politics, the higher the likelihood to go to the polls. Additionally inspired by the rational choice theorem (Downs 1957), we also assume that people who think that their vote does make a difference are more likely to go to the polls. This notion is part of the concept of political efficacy, which strongly correlates with political participation (Almond and Verba 1963; Shaffer 1981; Powell 1986; Rosenstone and Hansen 1993). Besides perceptions of political efficacy, trust in politics and political institutions appear to be another vital dimension of affirmative civic orientations (Shaffer 1981). For example, Gronlund and Setala (2007) analysed European Social Survey data in 22 countries and found that trust in national parliaments had a positive impact on turnout. Correspondingly, Cox (2003) found that voting turnout in the 1999 European Parliament election was strongly and positively correlated with trust in political institutions. Van der Eijk and others have repeatedly argued that the meaning and importance of party identification measures in a European context is doubtful (Van der Eijk and Niemoller 1983; Van der Eijk and Franklin 1996). To capture levels of political involvement, we therefore draw on general interest in politics. Interest in politics has evolved as a consistent determinant of voter turnout (e.g., Verba, Schlozman, and Brady 1995; Caballero 2005). "The more interested one is, the more likely one is to vote" (Blais 2007, 631). In this context, earlier research also repeatedly pointed to the fact that political interest is not only
a relevant and direct indicator of voting turnout, but additionally may also be seen as a key motivational factor regarding news consumption in the first place (Delli Carpini 2004; Strömbäck and Shehata 2010; Boulianne 2011). From a longitudinal perspective it has also been shown that political interest has even become a more powerful determinant of news consumption in high-choice media environments over time (Strömbäck, Djerf-Pierre, and Shehata 2012). Although earlier studies have mostly emphasised that the relationship between political interest and news media usage is reciprocal, they have mainly concluded that the impact of political interest on news exposure is stronger than vice versa (Strömbäck and Shehata 2010; Boulianne 2011). Based on this evidence it might be expected that individual degrees of political interest may also control and moderate the effectiveness of news exposure as well as contest framing effects on voting turnout. To account for this factor, we refine our testing models by additionally controlling for potential interaction effects between political interest and our measures of political news and contest framing exposure. To complete the list of potential predictors of voting turnout, we additionally install government approval (specific mode) and satisfaction with democracy (general mode) as proxies reflecting the satisfaction with the performance of the political system and as such as a measure of affirmative civic orientations. Regarding the United States, research has shown that voter participation between 1960 and 1997 was affected by both public approval and disapproval rates toward the incumbent president (Cebula 2005). Regarding satisfaction with democracy, earlier research repeatedly reported a positive nexus between satisfied voters and turnout (Grönlund and Setälä 2007; Schuck, Vliegenthart and De Vreese 2011; Seeber and Steinbrecher 2011).
All non-metric variables were converted into dichotomous dummy variables (1/0). Due to survey data limitations we had to rely on a single-item question regarding political efficacy, which asked whether people think who they vote for does or does not make a difference (external efficacy). Government approval rates were dichotomised into a two-dimensional measure (with the reference group 1 - approval; and 0 - disapproval). For measuring satisfaction with democracy we dichotomised the applied four-item scale. For measuring interest in politics, the respondents were divided in a group that is rather highly interested in politics and one with minor interest in politics (1/0). The measure of trust in politics consists of a four-item index containing reported levels of trust toward the national parliament, the government, political parties and politicians.
In our regression models 2 and 3 we additionally incorporate and contrast the effects of media-related indicators. We investigate to what extent general exposure to political news (INES) and specific contest framing exposure levels (ICES, IDES, ^ and IHES) lead to an increase of the explanatory power of our regression models regarding turnout by additionally controlling for interaction effects between political interest and news exposure as well as contest framing exposure. The core independent variable in model 2 is exposure to political news. Respondents indicated for each news outlet comprised in our media content analysis how frequently they use any of them to gather political information (daily, several times a week, rarely, never). We recoded these answers in a normalised index ranging from 0 (never) to 1 (daily). We computed a simple additive news exposure score by adding up these normalised indices for each news outlet (INES). Then, the qq outlet-specific contest framing scores were weighted by the individual exposure
0 £
C >
scores which resulted in a single measure for confrontation, dramatisation, and horse race exposure, dependent on which news outlets were used how frequently fO (ICES, IDES, and IHES).	00
Findings
We start with a look at the pooled generic framing structures of the leading Austrian news outlets in the 2008 election campaign. As Table 1 illustrates, conflict (indicated by a positive confrontation score=.284) is the most salient indicator of contest framing in the 2008 Austrian electoral coverage, followed by levels of horse race (score=-.030) and drama (score=-.275) which are both negative in total. The positive mean scores representing the levels of confrontation in the media coverage indicate that conflict framing is clearly more salient than its antithesis of consensus framing. Secondly, the horse race aspect of politics is not prevalent, but still prominently displayed in the Austrian electoral coverage. Almost half of all reports focus on horse race instead of policy debates (score=-.030). Thirdly, dramatisation appears to be the least salient contest indicator. The dramatisation score of -.275 signifies that the majority of the news items published in the final six weeks of the electoral race is not prevalently marked by arousing, emotional depictions of politics.
Table 1: Contest Framing Scores in the Austrian Tabloid and Quality News
Contest Framing Scores (-1 to +1)		Confrontation	Dramatisation	Horse Race
News Coverage (total)		.284	-.275	-.030
Tabloid News	Kronen Zeitung (Paper)	.318	.069	.004
	Österreich (Paper)	.289	-.156	.229
	ATV Aktuell (TV News)	.227	.102	.313
	Tabloid News (total)	.300	-.037	.118
Quality News	Der Standard (Paper)	.274	-.449	-.198
	Die Presse (Paper)	.278	-.463	-.157
	ORF Zeit im Bild (TV News)	.214	-.835	-.109
	Quality News (total)	.269	-.498	-.170
Mann-Whitney U Test				
Tabloid vs. Quality News		-2.023	-16.712	-10.364
(z value, significance)		p=.043	p=.000	p=.000
Note: As not all indicators were normally distributed, we applied Mann-Whitney U statistics for testing the significance of differences in the distributions.
The degree of contest framing in Austrian news varies considerably when comparing tabloid and quality news outlets, especially regarding levels of dramatisation and horse race. First, however, we draw our attention to confrontation framing of political news during the 2008 Austrian election campaign. Both, tabloid and quality news are characterised by a clear dominance of conflict over consensus framing (score=.269/.300). However, in total tabloid news still focuses more on confrontation than quality news on a moderately significant level (Mann-Whitney U Test: z=-2.023; p=.043).
Regarding the level of dramatisation, our analysis shows political reporting to be predominantly sober and unemotional in tone (tabloid news score=-.037; quality news score=-.498). Nonetheless, dramatisation is significantly more salient in tabloid news than in quality news (z=-16.712; p=.000). Looking at the level of horse race reporting, we can also state that tabloids news concentrates significantly more on winning and losing aspects of politics than quality news (z=-10.364; p=.000). Moreover, all tabloid news outlets even puts more emphasis on horse race than on policy debates (mean horse race score=.118), whereas all quality news outlets prevalently promote substantive, policy-relevant coverage instead of showing winning and losing or race schemas (mean=-.170).
We can conclude that whereas conflict is a dominant generic framing feature of all news formats, dramatisation and horse race appear to be a means of reporting predominantly applied by tabloid news. Validated for all three dimensions, we can sum up that the level of contest framing in quality news is significantly lower than in tabloid news. Thus, hypothesis 1 is strongly supported by our empirical data.
These empirically outlined, overall magnitudes and measures of contest framing in the Austrian news coverage are the basis for now investigating their impact on individual voting turnout. Consequently, we combine these contest framing scores in the news with our survey measures of individual political news exposure in order to appropriately assess the impact of exposure to contest framing on voting turnout. For doing so, our contest framing scores for all investigated media outlets are weighted by the individual exposure to these media outlets. Table 2 shows binary regression models, explaining voting turnout and thereby considering socio-demographics and civic orientations as controlling variables, and news exposure and contest framing exposure as our central variables of interest. We applied a comparative procedure to elucidate the explanatory power added by media exposure-related predictors in our turnout models.
Model 1 is our point of departure for explaining voting turnout. It includes sociodemographic characteristics and basic civic orientations reflecting levels of y political involvement and affirmation. It explains about 22 percent of the variance in reported turnout of Austrian voters (see Nagelkerke R2). Thereby, interest in politics, political efficacy, trust in politics, and satisfaction with democracy appear as significant and robust predictors of voting turnout in the 2008 Austrian elections. The more Austrian voters are interested in politics, the more they think their vote does make a difference. The more they trust political institutions, and the more they are satisfied with how democracy works, the more likely they cast their votes. In contrast, the tested sociodemographics (age, gender and education) are non-factors in explaining voting turnout in contemporary Austria.
To estimate the additional effect of media-exposure related variables, we now proceed to model 2, which additionally regards general news exposure levels (individual news exposure scores - INES). It shows that the additional consideration of general news exposure does not add significant explanatory power to our turnout regression model. Differing levels of individual news exposure do not affect the likelihood to vote. Consequently, hypothesis 2 is not supported by our findings. In
c >
the Austrian case, higher levels of news exposure are not associated with higher levels of voting turnout. We computed a model additionally testing the interaction qq between political interest and news exposure. As this procedure did not result in
Table 2: Regression Models Explaining Effects of Contest Framing Exposure on
Voting Turnout	LO
00
Dependent Variable Voting Turnout: yes (1), no (0)		Model 1	Model 2	Model 3
		Beta Exp(B) (SE)	Beta Exp(B) (SE)	Beta Exp(B) (SE)
Constant		-.212 .809 (.380)	-.508 .602 (.417)	-.348 .706 (.425)
Socio Demographics	Age	.008 1.008 (.006)	.008 1.008 (.006)	.004 1.004 (.007)
	Gender	-.017 .983 (.214)	-.014 .986 (.215)	.014 1.014 (.217)
	Education	-.177 .838 (.255)	-.200 .819 (.256)	-.319 .727 (.261)
Civic Orientations	Political Efficacy	1.031*** 2.804 (.213)	1.048*** 2.853 (.214)	1.029*** 2.798 (.216)
	Government Approval	.196 1.216 (.262)	.156 1.169 (.263)	.131 1.140 (.265)
	Satisfaction with Democracy	.583** 1.792 (.222)	.575** 1.777 (.223)	.587** 1.799 (.226)
	Interest in Politics	1.342*** 3.828 (.288)	1.276*** 3.582 (.291)	1.214*** 3.366 (.292)
	Trust in Politics	.907*** 2.477 (.229)	.886*** 2.424 (.230)	.875*** 2.399 (.232)
News Exposure	News Exposure (INES)		.230 1.259 (.135)	
Contest Framing Exposure	Confrontation Exposure (ICES)			1.041 2.832 (.726)
	Dramatisation Exposure (IDES)			-.013 .987 (.480)
	Horse Race Exposure (IHES)			-2.624* .073 (1.163)
Nagelkerke R2/Incremental R2 (%)		.225/-	.230/0.5	.242/1.7*
Log Restricted-Likelihood		623.847	620.918	613.827
Number of valid cases		1,041	1,041	1,041
Note: ***p<0.001; **p<0.01, *p<0.05 level. The variance inflation factor (VIF) does not indicate a multicollinearity problem in any regression model (the VIF of all independent variables included is < 2.07).
significant additional effects, we refrained from reporting this extended model. We also tested the effectiveness of general news exposure on turnout for heavy and light news users (split electorate by the median of the individual news exposure score - 1.33) as well as for heavy tabloid news users and others and found no variance (not shown in tables). These results qualify some earlier findings on the mobilising effect of news exposure and indicate, at least for the Austrian context, that sheer exposure to political news, even when controlling for heavy and light (tabloid or quality) news usage, might be a too general and cursory factor to be accurate. This is implicitly substantiated by De Vreese and Boomgarden (2006) who found that media exposure is effective, when the outlets carry a clearly one-sided information flow. And Newton (1999) noted that "it seems to be the content of the media, rather than the form which is important" (p. 577).
Thus, to refine our analysis, we expand our search for news effects to a more sophisticated and specific level - actual content characteristics of political news representing the contest aspects of the campaign. Consequently, we now call our attention to the effects of contest framing exposure. To avoid multicollinearity problems, we decided to not integrate general news exposure and contest framing exposure variables at once in our explanatory model. Instead, we contrast them in a comparative procedure to bring fact to face the strength of their effects. In model 3, we add our frame-based news exposure measures regarding individual confrontation exposure scores (ICES), individual dramatisation exposure scores (IDES), and individual horse race exposure scores (IHES) as explanatory variables to our basic model. It shows that horse race framing is the only contest framing indicator that constitutes a significant predictor of voting turnout, whereas confrontation and dramatisation framing are not effective. Therefore, hypotheses 3 and 4, expecting high exposure levels to confrontation and dramatisation to mobilise voters, are not supported. Moreover, also against our expectations, exposure to horse race framing actually lowers the chance to go to the polls. Hypothesis 5 is not supported either as horse race framing is rather turning Austrian voters off. The more voters y are exposed to the contest in the form of a horse race, the more they are inclined to turn their back on going to the polls.
To control for potential interaction effects between individual levels of political interest and contest framing effects initialised by the news coverage, we augmented model 3 by the product variables of political interest and ICES, IDES and IHES. As we found no significant interactions, we can state that political interest does not bias the nexus between all applied variables measuring contest framing exposure and voting turnout. Consequently, we abstained from reporting the extended interaction effect model in detail.
To summarise, we can conclude that contest framing by Austrian media holds no mobilising power at all. Our results are in contrast to some of the earlier evidence on voter mobilisation in the context of European parliamentary elections (Norris
/^l^l^l• Krnurl/ x/liAfrûnfnorf anrl Fiû \/mûOû /111 1 \ anrl r<Arml">ArcifA triât mntûcf fram_
c >
2000; Schuck, Vliegenthart and De Vreese 2011) and corroborate that contest framing does not hold a universal mobilising effect. It might be rather context-sensitive and even result in a reversed effect, at least as far horse race framing is concerned.
Depicting the campaign predominantly as a horse race moderately contributes to the demobilisation of Austrian voters. General exposure to political news remains qq effectless and so do levels of confrontation and dramatisation. Our findings support
the notion that general news exposure might be a too cursory and fuzzy measure to provide accurate estimates of news exposure effects. Measures of exposure of r^ specific news content yield more realistic and more reliable representations of what the people are exposed to in what intensity, whereas general media usage and exposure might be blurred by avoidance of political news in general or by an insufficient juxtaposition of tabloid or quality news users. Most people do not exclusive use tabloid or quality news, at least in the investigated Austrian case.
Conclusion and Discussion
The here presented study comparatively investigated news effects on the (de)mobilisation of the electorate in the 2008 Austrian Parliamentary Elections on two different levels: Firstly, the general exposure to political news in newspapers and on television news; secondly, the exposure to specific content characteristics that reflect campaigns as a "contest." We hypothesised that generic framing depicting election campaigns predominantly as a contest would mobilise voters by connoting that something is at stake than rather turning them off. Our study does not confirm a positive nexus between contest framing exposure and voting turnout in all applied indicators (confrontation, dramatisation, horse race) in the context of the 2008 Austrian Parliamentary Elections. Instead, we report a reversed mobilisation effect of horse race framing exposure. Supportingly, Schuck, Vliegenthart and De Vreese (2011) suggest that horse race coverage may not offer a substantive basis to actually engage voters (see also Valentino, Beckman and Buhr 2001). This is partly also in line with a recent study showing that substantive news, as the antithesis of horse race news, had a positive effect on civic orientations by lowering the levels of political cynicism among young citizens in the 2006 Dutch election campaign (Adriaansen et al. 2010).
Nonetheless, our findings qualify earlier research on horse race, strategy and conflict framing effects on political engagement to some extent and corroborate that such effects are not universal, but rather context-sensitive. Additionally, in-consinstencies of the existing empirical evidence might be also due to different operationalisations of horse race in this study and strategy framing elsewhere. Moreover, different methods of data collection were applied, including experiments (Cappella and Jamieson 1997; Valentino, Beckmann and Buhr 2001) or combining survey and content analysis instruments (Schuck, Vliegenthart and De Vreese 2011). Furthermore, earlier European framing effect studies mostly concentrated on European parliamentary elections instead of national elections. Last but not least, our integrative approach combining actual exposure levels with the framing salience in the news outlets on the micro level may also account for more nuanced insights on news framing effects regarding voter mobilisation.
Our findings, based on explanatory models placing media exposure effects in a larger context and controlling for socio-demographics and civic orientations appear to be rather robust. In additional tests, we found that effects and non-effects of news and news framing exposure are alike for party identifiers versus non-identifiers or heavy (tabloid) news users versus others.
This investigation carries relevant implications for political communication. Our findings point to the fact that media contest framing may in fact alienate voters and erodes electoral participation instead of holding a mobilising effect. The
journalistic and also political attempts to foster attention and to activate voters and the audience likewise by framing the campaign as a heated, dramatised and conflictual contest fails to have the desired impact and even end up with a reversed effect. Consequently, contest framing in political communication may work for the media to attract audience(s), but it does not work for democracy and the electoral mobilisation in the Austrian context.
Our study enriches empirical evidence on news effects that focus on concrete media content characteristics by showing that conflict is not a universal mobilising factor. Under conditions of European Parliamentary Elections it obviously appears as a mobilising factor, but not so much in the Austrian National Parliamentary Election context. This points to the need of more investigations on the level of national elections to broaden the empirical foundation in European political communication contexts. Our approach also enhances research in this area from a methodological point of view. It offers an integrative and rather realistic measure of content-related exposure effects as media-outlet specific levels of generic framing are weighted by the actual and individual exposure to these frames.
Our study is, however, characterised by some limitations. It has a static focus and does not allow depicting changes of the propensity to vote over time. Further research on contest framing should apply dynamic panel designs to enhance the focus on changes in the course of election campaigns.
Acknowledgement
This research was carried out under the auspices of the Austrian National Election Study (AUTNES), a National Research Network (NFN) sponsored by the Austrian Research Fund (FWF) (Project: Lengauer - Media Coverage and Effects, S10904-G11).
References:
Adriaansen, Maud, Philip van Praag and Claes H. De Vreese. 2010. Substance Matters: How News Content can Reduce Political Cynicism. International Journal of Public Opinion Research 22, 4, 433-457.
Aldrich, John H. and Dennis M. Simon. 1986. Turnout in American National Elections. In S. Long (ed.),
Research in Micropolitics (volume 1), 271-301. Greenwich, Conneticut: JAI Press. Almond, Gabriel and Sidney Verba. 1963. The Civic Culture. Political Attitudes and Democracy in Five
Nations. Princeton: Princeton University Press. Bennett, Lance W. 2009. News. The Politics of Illusion. New York: Pearson Longman. Blais, André. 2007. Turnout in Elections. In R. J. Dalton and H.-D. Klingemann (eds.), The Oxford
Handbook of Political Behavior, 621-635. Oxford: Oxford University Press. Boulianne, Shelley. 2011. Stimulating or Reinforcing Political Interest: Using Panel Data to Examine Reciprocal Effects Between News Media and Political Interest. Political Communication 28, 2, 147-162.
Brady, Henry, Sidney Verba and Kay Schlozmann. 1995. Beyond SES: A Resource Model of Political Participation. American Political Science Review 89, 2, 271-294.
u
(6
Caballero, Claudio. 2005. Nichtwahl. In J. W. Falter and H. Schoen (eds.), Handbuch Wahlforschung, 329-365. Wiesbaden: VS Verlag für Sozialwissenschaften.
C
^	Campbell, Angus, Philip E. Converse, Warren E. Miller and Donald E. Stokes. 1960. The American Voter.
New York: John Wiley and Sons. Canel, Maria Jose, Christina Holtz-Bacha and Paolo Mancini. 2007. Conflict as a Frame in Television Coverage of Politics: A Comparative Study in Italy, Spain, and Germany. Paper presented at the
qq	Annual meeting of the International Communication Association, San Francisco, May 2007.
Cappella, Joseph and Kathleen H. Jamieson. 1997. Spiral of Cynicism. The Press and the Public Good.
Oxford: Oxford University Press.	^
Cebula, Richard J. 2005. Strong Presidential Approval or Disapproval Influencing the Expected	00
Benefits of Voting and the Voter Participation Rate. Atlantic Economic Journal 33, 2, 159-167.
Cottle, Simon and Mugdha Rai. 2006. Between Display and Deliberation: Analyzing TV News as Communicative Architecture. Media Culture and Society 28, 2, 163-189.
Cox, Michaelene. 2003. When Trust Matters: Explaining Differences in Voter Turnout. JCMS: Journal of Common Market Studies 41, 4, 757-770.
Dahlgren, Peter. 1996. Television and the Public Sphere. Citizenship, Democracy and the Media. Thousand Oaks: Sage.
Dalton, Russell J. and Martin P. Wattenberg. 2000. Parties Without Partisans: Political Change in Advanced Industrial Democracies. Oxford: Oxford University Press.
De Vreese, Claes H. 2005a. News Framing: Theory and Typology. Information Design Journal + Document Design 13, 1, 51-62.
De Vreese, Claes H. 2005 b. The Spiral of Cynicism Reconsidered. European Journal of Communication 20, 3, 283-301.
De Vreese, Claes H. 2006. Political Parties in Dire Straits? Consequences of National Referendums for Political Parties. Party Politics 12, 5, 581-598.
De Vreese, Claes H. and Hajo G. Boomgaarden. 2006. News, Political Knowledge and Participation: The Differential Effects of News Media Exposure on Political Knowledge and Participation. Acta Politica 41, 4, 317-341.
De Vreese, Claes H. and Holli A. Semetko. 2004. News Matters: Influences on the Vote in the Danish 2000 Euro Referendum Campaign. European Journal of Political Research 43, 699-722.
Delli Carpini, Michael X. 2004. Mediating Democratic Engagement: The Impact of Communications on Citizens' Involvement in Political and Civic Life. In L. L. Kaid (ed.), Handbook of Political Communication Research, 395-434. Mahwah: Lawrence Erlbaum.
Downs, Anthony. 1957. An Economic Theory of Democracy. New York: Harper and Brothers.
Duff, Brian, Michale J. Hanmer, Won-Ho Park and Ismail K. White. 2007. Good Excuses: Understanding who Votes With an Improved Turnout Question. Public Opinion Quarterly 71, 1, 67-90.
Elenbaas, Matthijs and Claes H. De Vreese. 2008. The Effects of Strategic News on Political Cynicism and Vote Choice Among Young Voters. Journal of Communication 58, 3, 550-567.
Esser, Frank. 1999. Tabloidization of News: A Comparative Analysis of Anglo-American and German Press Journalism. European Journal of Communication 14, 3, 291-324.
Esser, Frank and Katharina Hemmer. 2008. Characteristics and Dynamics of Election News Coverage in Germany. In J. Strômbâck and L. L. Kaid (eds.), The Handbook of Election Coverage around the World, 289-307. Mahwah, NJ: Lawrence Erlbaum.
Franklin, Mark N. 2004. Voter Turnout and the Dynamics of Electoral Competition in Established Democracies Since 1945. Cambridge: Cambridge University Press.
Grônlund, Kimmo and Maija Setâlâ. 2007. Political Trust, Satisfaction and Voter Turnout. Comparative European Politics 5, 4, 400-422.
Hânggli, Regula and Hanspeter Kriesi. 2010. Political Framing Strategies and Their Impact on Media Framing in a Swiss Direct-Democratic Campaign. Political Communication 27, 2, 141-157.
Iyengar, Shanto, Helmut Norpoth and Kyu S. Hahn. 2004. Consumer Demand for Election News: The Horserace Sells. The Journal of Politics 66, 1, 157-175.
Lazarsfeld, Paul F., Bernard R. Berelson and Hazel Gaudet. 1944. The People's Choice. New York: Columbia University Press.
Lupia, Arthur, Mathew D. McCubbins and Samuel L. Popkin. 2000. Elements of Reason: Cognition, Choice, and the Bounds of Rationality. Cambridge: Cambridge University Press.
McManus, John H. 1994. Market-Driven Journalism: Let the Citizen Beware?Thousand Oaks: SAGE Publications.
Neuman, Russel W., Marion R. Just and Ann N. Crigler. 1992. Common Knowledge. News and the Construction of Political Meaning. Chicago: University of Chicago Press.
Newton, Kenneth. 1999. Mass Media Effects: Mobilization or Media Malaise? British Journal of Political Science 29, 3, 577-599.
Norris, Pippa. 2000. A Virtuous Circle? Political Communications in PostindustrialSocieties. New York: Cambridge University Press.
Norris, Pippa. 2004. Electoral Engineering: Voting Rules and Political Behavior. New York: Cambridge University Press.
Norris, Pippa. 2006. Did the Media Matter? Agenda-Setting, Persuasion and Mobilization Effects in the British General Election Campaign. British Politics 1, 2, 195-221.
Norris, Pippa, Lawrence LeDuc and Richard G. Niemi. 2010. Introduction: Building and Sustaining Democracy. In L. LeDuc, R. G. Niemi and P. Norris (eds.), Comparing Democracies 3, 1-21. Thousand Oaks: Sage.
Paletz, David and Robert Entman. 1981. Media Power Politics. New York: Free Press.
Patterson, Thomas E. 1993. Out of Order. New York: Vintage.
Patterson, Thomas E. 2002. The Vanishing Voter: Public Involvement in an Age of Uncertainty. New York: Alfred A. Knopf.
Plasser, Fritz and Günther Lengauer. 2010. Die österreichische Medienarena: Besonderheiten des politischen Kommunikationssystems. In F. Plasser (ed.), Politik in der Medienarena, 19-52. Vienna: Facultas.
Plasser, Fritz and Gunda Plasser. 2002. Global Political Campaigning: A Worldwide Analysis of Campaign Professionals and Their Practices. Santa Barbara: Praeger Publishers.
Powell, Bingham G. 1986. American Voter Turnout in Comparative Perspective. The American Political Science Review 80, 1, 17-43.
Ridout, Travis. 2004. How Campaigns Matter: The Relative Impacts of Television Advertising,
Candidate Appearances and News Coverage in the 2004 Presidential Nomination Race. Paper presented at the annual meeting of the Midwest Political Science Association, Seattle, USA, April 2004.
Rosenstone, Steven J. and John M. Hansen. 1993. Mobilization, Participation and Democracy in America. New York: Macmillan.
Schuck, Andreas, Rens Vliegenthart and Claes De Vreese. 2011. Who's Afraid of Conflict? How Conflict Framing in Campaign News Coverage Mobilized Voters in the 2009 European Parliamentary Elections. Paper presented at the annual meeting of the International Communication Association, Boston, USA, May 2011.
Seeber, Gilg and Markus Steinbrecher. 2011. Why Europeans (Don't) Turnout to Vote. In M. Maier, J. Strömbäck and L.L. Kaid (eds.), Political Communication in European Parliamentary Elections, 213232. Farnham: Ashgate.
Shaffer, Stephen D. 1981. A Multivariate Explanation of Decreasing Turnout in Presidential Elections, 1960-1976. American Journal of Political Science 25, 1, 68-95. £	Slater, Michael D. 2004. Operationalizing and Analyzing Exposure: The Foundation of Media Effects
Research. Journalism & Mass Communication Quarterly 81, 1, 168-183.
Sparks, Colin and John Tulloch. 2000. Tabloid Tales: Global Debates Over Media Standards. New York: Rowman & Littlefield Publishers. Q	Strömbäck, Jesper. 2008. Swedish Election News Coverage. In J. Strömbäck and L. L. Kaid (eds.), The
Handbook of Election News Coverage Around the World, 160-174. New York: Routledge.
Strömbäck, Jesper and Daniela V. Dimitrova. 2006. Political and Media Systems Matter. A
Comparison of Election News Coverage in Sweden and the United States. The International Journal of Press/Politics 11, 4, 131-147.
Strömbäck, Jesper, Monika Djerf-Pierre and Adam Shehata. 2012. The Dynamics of Political Interest
Oand Media Consumption: A Longitudinal Perspective. International Journal of Public Opinion Research. First published online: June 19, 2012.
Strömbäck, Jesper and Adam Shehata. 2010. Media Malaise or a Virtuous Circle? Exploring the ^	Causal Relationship Between News Media Exposure, Political News Attention and Political
Interest. European Journal of Political Research 49, 5, 575-597. Strömbäck, Jesper and Peter Van Aelst. 2009. Exploring Some Antecedents of the Media's Framing of Election News: A Comparison of Swedish and Belgian Election News. The International Journal of Press/Politics 15, 1, 41-59.
II
O
cr>
Valentino, Nicholas A., Matthew N. Beckmann and Thomas A. Buhr. 2001. A Spiral of Cynicism for
Some: The Contingent Effects of Campaign News Frames on Participation and Confidence in	-,-h
Government. Political Communication 18, 4, 347-367.	^
Van der Eijk, Cees and Mark M. Franklin. 1996. Choosing Europe?: The European Electorate and National Politics in the Face of Union. Ann Arbor: The University of Michigan Press.
Van der Eijk, Cees and Broer Niemöller. 1983. Electoral Change in the Netherlands: Empirical Results and Methods of Measurement. Amsterdam: CT Press.
Verba, Sidney, Kay L. Schlozman and Henry Brady. 1995. Voice and Equality: Civic Voluntarism in American Politics. Cambridge, Mass.: Harvard University Press.
Wattenberg, Martin P. 2002. Where Have all the Voters Gone? Cambridge, Mass.: Harvard University Press.
Wolfinger, Raymond and Steven Rosenstone. 1980. Who Votes? New Haven: Yale University Press.