Negativity in Day-To-Day Politics. Examining Politicians Who Play the Attack Game. Pauline Ketelaars. University of Antwerp

Transcription

1 Negativity in Day-To-Day Politics Examining Politicians Who Play the Attack Game Pauline Ketelaars University of Antwerp Media, Movements & Politics ( Paper prepared for the ECPR Joint Sessions in Nothingham, April 2017 First draft: Please do not cite. During the past two decades there has been a spectacular increase of research on negativity in the political realm. This literature has shown that politicians often stress the weaknesses and faults of their competitors instead of emphasizing their own strengths and achievements. The large majority of studies on attack politics focus on elections, studying how politicians go negative during electoral races and examining possible effects of negative campaigning on voter turnout and political attitudes (for overviews see Lau and Rovner 2009; Lau, Sigelman, and Rovner 2007). But negativity in politics is not limited to elections. Politicians do not save up their attacks for this three month period every four or five years. Generating negative attention for political opponents is, as a Belgian politician pointed out during an interview: part of the political game. Moreover, it can be argued that negativity in day-to-day politics is gaining ground. Scholars have suggested that political parties increasingly may go negative in response to a changing political environment in which voter loyalties decline and protest parties gain ever more support (Mair, Müller, and Plasser 2004). Additionally, through the process of mediatization (Strömbäck 2008) politicians are nowadays involved in a permanent campaign, constantly seeking media attention 1

2 and continually defending their seat against future challengers. Nevertheless, studies that investigate politicians use of negativity outside electoral races are extremely rare. Broadly speaking, scholars of attack politics have tried to answer three questions: which politicians or parties attack their opponents, under what circumstances do they go negative, and what are the effects of attack politics? (Lau and Rovner 2009). This study concentrates on the first question and tries to explain why some politicians are more likely to go negative than others. The aim of the study is twofold. First, it tests whether the factors that scholars have found to lead to negativity during election campaigns can be extrapolated to politics in general. Second, it tries to contribute to a more general understanding of the determinants of negativity by exploring differences between individual politicians. Up to now, research in European democracies has foremost studied negativity at the party-level (e.g. vanheerde-hudson 2011; Elmelund-Præstekær 2008; Schweitzer 2010; Walter 2014). These studies have found that particularly opposition parties (Hansen and Pedersen 2008), extreme parties (Elmelund- Præstekær 2010), and parties with low coalition potential are likely to go negative (Walter, van der Brug, and van Praag 2014; Walter and van der Brug 2013). While there are contributions that use individuals instead of parties as units of analysis, these works almost exclusively focus on differences between males and females (Herrnson and Lucas 2006; Walter 2013; Carlson 2001). As a consequence, the extent to which individual politicians differ in negativity, and how we can explain intra-party variation, remains largely unclear (see Dolezal, Ennser-Jedenastik, and Müller 2015 for a notable exception). Although US studies generally examine individual candidates, they typically look at a candidate s incumbency status and his or her ranking in the polls (Lau and Pomper 2004; Kahn and Kenney 2004); elements that foremost dig into the circumstances when politicians go negative instead of clarifying exactly who is prone to attack political rivals. The data for this study consist of surveys with 230 Belgian politicians cabinet members and members of parliament (MPs) administered in Contrary to the bulk of negativity research, this 2

3 study puts the decisions that political actors have to make in their day-to-day work center stage. Negativity scholars generally examine attack politics through means of content analyses of, for instance, media coverage, political ads, or press releases (for exceptions, see Herrnson and Lucas 2006; Theilmann and Wilhite 1998). The cognitive considerations that politicians make, which inevitably precede the recorded public action, are usually not captured. Therefore, this study directly asks politicians to what extent they are inclined to go negative. The results show that, in their day-to-day work, MPs of opposition parties are more negative than representatives of parties in government. I also find that male politicians are more focused on exhibiting the weaknesses of their rivals than females. Contrary to previous studies, there are no signs that politicians of extreme parties differ from center parties. Furthermore, politicians who specialize in one or two policy areas tend to be significantly less negative than politicians who prefer to speak and act upon a wide range of issues. Additionally, the more a politician is focused on realizing policy goals, the less he or she is prone to harm political rivals. As such, the variation in negativity between politicians is mostly explained by the roles they take up in parliament. Defining Negativity And Why Politicians Go Negative In the literature on negative campaigning we can distinguish between scholars who use evaluative definitions of negativity and those who use directional definitions (Walter and Vliegenthart 2010). Evaluative definitions refer to negative politics as illegitimate and unfair practices, labelling politicians critiques negative when they can be considered deceptive, untruthful or irrelevant. Negativity, then, is related to manipulation, lying and dirty politics (Davis and Ferrantino 1996; Jamieson, Waldman, and Sherr 2000). These definitions, however, are susceptible to measurement bias. After all, when is an attack unfair? And when is criticism dishonest or irrelevant? Directional definitions, on the other hand, define negativity as all forms of criticism or attack on opponents (see e.g. Elmelund-Præstekær 2010; Geer 2006; Buell and Sigelman 2008), without differentiating between appropriate critiques and mudslinging (Kahn 3

4 and Kenney 1999). Geer (2006, 23), for example, defines negativity as any criticism leveled by one candidate against another during a campaign. The advantage of directional definitions is that they are clear and neutral: they allow to measure negativity without making normative judgements. Hence, in line with the majority of scholars and attuned to the current study I define negativity as the extent to which politicians intend to demonstrate the weaknesses of political opponents, regardless of whether generating attention for these weaknesses can be considered fair or not. There are various reasons why we can expect politicians to use and look for negative information about their political rivals. Like regular citizens, politicians are likely to be biased towards negative messages (Lewicka, Czapinski, and Peeters 1992; Robinson-Riegler and Winton 1996). Psychologists have provided ample evidence that people emotionally, cognitively and behaviorally respond stronger and faster to negative information than to positive information (Baumeister et al. 2001; Rozin and Royzman 2001; Ito et al. 1998). Research in political communication as well, has confirmed a negativity bias. Negative news stories are more arousing and attention grabbing than positive news stories (Soroka and McAdams 2015). Politicians tendency to focus on the weaknesses of their opponents, hence, is not necessarily a product of strategic considerations alone. They are humans, after all, and subject to the same biases as their electorate. Nevertheless, politicians also have tactical reasons to attack their rivals. In order to get people s support, politicians want to maximize their appeal. On the one hand, they can do this by self-praise, sending positive messages about themselves, their actions and their party i.e. positive campaigning (Gregory 2015). On the other hand, they can try to reduce the attractiveness of their competitors (Harrington and Hess 1996). Politicians might fear that the flaws of their opponents will go unnoticed by the electorate if they do not disseminate negative information they have about them. Although research has not produced consistent evidence that negative campaigning works (Lau, Sigelman, and Rovner 2007) and though it might actually have a boomerang effect (Byrne and Hart 2009) campaign 4

5 managers indeed believe that negativity can help their candidates to gain support (Perloff and Kinsey 1992). Furthermore, by going negative representatives can increase their chance to generate media attention (Haynes and Rhine 1998). Media attention is essential for a politician s electoral success. Voters are unlikely to vote for candidates they know little about and most people obtain their knowledge about politicians via the news (Arnold 2004). As journalists generally provide more extensive coverage of negative messages than of positive messages (Hansen and Pedersen 2008), politicians are encouraged to attack their challengers, hoping that the attack will generate some coverage and that the public will remember their name. Explaining Variation in Negativity While politicians have both natural and strategic incentives to go negative, not every politician or political party is equally prone to attack political rivals. Scholars on negative campaigning have found that negativity is, in part, determined by the electoral context. Politicians are for instance more likely to go negative at the end of a campaign, when they are attacked by opponents, when there are many candidates, and when it is a close race (Lau and Pomper 2004; Hale, Fox, and Farmer 1996; Peterson and Djupe 2005; Buell and Sigelman 2008). However, studying negative campaigning in the Netherlands, Germany and Britain, Walter, van der Brug and Praag (2014) find that compared to party characteristics the electoral context is not so important to explain why politicians attack each other. This study therefore concentrates on features of parties and politicians, and it tests whether the variables that have been found to affect negativity during elections also matter outside electoral races. One of the most consistent findings in US research on election campaigns is that challenger candidates are more likely to attack their opponents than incumbents (Kahn and Kenney 1999; Hale, Fox, and Farmer 1996; Druckman, Kifer, and Parkin 2010). Similarly, European scholars have found by and large that opposition parties more often go negative than government parties (Elmelund-Præstekær 2008, 5

6 2010; Hansen and Pedersen 2008; Walter and van der Brug 2013). We can expect this difference to exist in day-to-day politics as well. Even outside the electoral race, opposition parties will try to make clear why they should be ones in office and why a change of course is needed. By pointing to the flaws of the politicians and parties in government they can show that they deserve the electorate s support rather than the ruling parties (Lau and Rovner 2009). Furthermore, government parties are less likely to go negative because they are the ones with policy-making power. They are therefore more inclined to promote themselves and to look for good news in order to show that their approach is working. Rather than gathering negative information, they want to claim credit for things that are going well (Thesen 2013). Opposition parties, on the contrary, explicitly have an attacking position in the public debate. They are expected to control the government and to hold it responsible for negative developments and social problems. We can expect that opposition parties emphasize negative information and search for ammunition to attack the government simply because it is their job to do so (Elmelund-Præstekær 2010). Additionally, because government actors are the ones in power, they are more newsworthy for journalists than politicians from the opposition. Compared to opposition MPs, government MPs have less incentives to attract journalists attention with negative stories about political rivals. Finally, we can expect government parties to be less negative because they will try to avoid to reveal negative aspects of the other parties in government. While opposition parties can essentially attack any other political party, government parties will mostly try to refrain from attacking each other. In sum, the first hypothesis of this study is: H1: Politicians from opposition parties are more focused on demonstrating the weaknesses of political opponents than politicians from government parties. Another consistent finding in the US literature is that, during electoral races, Republican candidates tend to be more negative than Democrats (Lau and Pomper 2001; Hale, Fox, and Farmer 1996; Theilmann and 6

7 Wilhite 1998; Ansolabehere and Iyengar 1997). This result, however, is difficult to be generalized to political systems with multiple parties, such as the Belgian system. Translating these American findings for a study of four election campaigns in Denmark, Elmelund-Præstekær (2010, 142) suggests that the intensity of a party s ideology can be expected to affect negativity: the more ideologically extreme a party is, the more it disagrees with other parties on political issues. Extreme parties can be expected to be more negative than mainstream parties because they dispute more policies of more parties. Elmelund- Præstekær found that extreme leftwing and extreme rightwing parties produced more negative statements in TV presentation broadcasts. However, only the extreme left parties were more likely to communicate negative statements during debates, and there was no difference with non-extreme parties regarding advertisements and manifestos. Walter and van der Brug (2013) also theorize that extremer parties are more likely to attack their opponents. They argue, however, that the differences between parties are a consequence of the variation in coalition potential. They hold that the further a party is positioned from the median party, the less likely it is to be part of the government coalition, and the more likely it is to go negative during the electoral race. If a party is not expected to be part of the government, it has less to lose by attacking others in the campaign. During everyday politics, however, it is clear who is part of the coalition and who is not. In the context of the current study a party s coalition potential cannot be a motivation to refrain from negativity. Yet, following Elmelund-Præstekær (2010), I expect extreme parties to be more attentive to the weaknesses of political opponents because their ideologies differ the most from other parties and because they are most likely to disagree with others on various issues. H2: Politicians from extreme parties are more focused on demonstrating the weaknesses of political opponents than politicians from center parties. 7

8 Intra-Party Differences When trying to explain variation in negativity, very few scholars address differences at the individual level. European scholars have almost exclusively used parties as units of analysis (e.g. Elmelund-Præstekær 2008, 2010; Walter and van der Brug 2013; Walter, van der Brug, and van Praag 2014; Hansen and Pedersen 2008; Schweitzer 2010; vanheerde-hudson 2011). Although scholars in the US usually examine individual candidates, they measure determinants that are foremost connected to the electoral race, such as the timing, how many candidates are in the race, who is behind in the polls, or a candidate s funding (e.g. Skaperdas and Grofman 1995; Lau and Pomper 2001, 2004; Peterson and Djupe 2005). These variables mainly provide insight into when politicians attack each other, rather than which politicians play the attack game. There is one important exception to scholars tendency to focus on party-level characteristics and contextual variables: various studies have addressed the difference in negativity between male and female candidates. According to Grossmann (2012) gender is one of the most heavily studied determinants of negativity. This literature, however, has produced very inconsistent results. Some studies find no significant effect of gender (Proctor, Schenck-Hamlin, and Haase 1994; Lau and Pomper 2001, 2004; Walter 2013). 1 Some prove that males are more negative (Bystrom 2006; Carlson 2001; Trent and Sabourin 1993), and that the difference with women even increases when the electoral race gets more competitive (Kahn and Kenney 2000). But other studies conclude that females exhibit a greater proportion of negativism than men (Druckman, Kifer, and Parkin 2010; Evans and Clark 2016; Kahn 1993). For various reasons we might expect males to be more focused on the weaknesses of their opponents than females. Firstly, in general men are more dominant and aggressive than women, and so male politicians might instinctively be more inclined to attack other politicians. Surveying various candidates in the US, Herrnson and Lucas (2006) confirm that female politicians are less likely to approve negative campaign strategies than their male counterparts. Yet, they note that in some cases, for instance 8

9 when the opponent faces a documented case of sexual harassment, women are more likely to attack. Secondly, female candidates might not necessarily be less aggressive than males, they may also be more hesitant to attack others because of gender stereotypes held by the public (Lawless 2004). Voters view women as more kind, honest, caring and sympathetic (Fridkin and Kenney 2009). This can sometimes be an advantage for female politicians, but they also have to be careful not to violate sex-role expectations. Krupnikov and Bauer (2014) show that women are indeed disproportionately punished by voters for negative ads compared to men, but only when two conditions are met: when the woman is perceived as the instigator of the negativity and when she is from a different party than the voter. On the other hand, one could argue that female politicians have incentives to follow a more negative strategy. Women in politics might want to show the public that they are tough enough to be in politics (Lau and Pomper 2001). Furthermore, female politicians are traditionally in a disadvantaged position compared to men: they are the natural underdog. Like other disadvantaged politicians, such as seat challengers and candidates losing in the polls, women might therefore be more inclined to focus on other candidates weaknesses (Druckman, Kifer, and Parkin 2010). Evens and Clark (2016) for instance show that female politicians send more attack-style tweets than males. Similarly, there is evidence that men less frequently use negative advertisements, especially when they race against a female politician, because they fear to be perceived as beating up on a woman (Kahn 1993, 491). What, then, can we expect from male and female politicians in their day-to-day work? Given the lack of the electoral competition, women in everyday politics have less incentives to be the attacking underdog or to show that they are tough enough for politics. When the election campaign is over once female candidates have won their seat these motivations largely disappear, and there is no reason why they should be more negative than male politicians. However, I do expect male politicians to be more negative. The stereotypical softer roles for women and the more aggressive roles for men also apply outside the election campaign. 9

10 H3: Male politicians are more focused on demonstrating the weaknesses of political opponents than female politicians. As stated above, apart from studies that go into differences between male and female politicians, research into features of politicians that can explain variation within parties is extremely rare. One of the few exceptions are Harrington and Hess (1996), who theorize that candidates with more attractive personal attributes communicate less negative and more positive campaign messages. Testing this theory in an experiment with political consultants, Theilmann and Wilhite (1998), however, do not find evidence for Harrington and Hess theory. A more recent exception is the work of Dolezal, Ennser-Jedenastik and Müller (2015). They point out that, although negative campaigns can have backlash effects for representatives, political parties do not want to refrain from campaigning negatively because they want to make voters aware of the faults of other parties. These authors approach negative campaigning as a collective action problem: it is important for parties to do but it is potentially damaging for individual politicians. Their analyses show that parties try to minimize the costs of negative campaigning through a division of labor in which party leaders and public office holders, such as cabinet members, leave the dirty work for floor leaders and general secretaries. While this study makes an important contribution, it is limited to the examination of six high offices: heads of government, speakers of parliament, cabinet members, party leaders, party floor leaders, and party secretaries. These functions only cover four percent of the individuals that are included in the study s dataset. The rest of the representatives were put into a single category of other politicians, and were not further discussed in the article. The current study takes the next step and considers differences between all representatives, accounting for each elected politician individually in the analyses. Furthermore, while Dolezal et al. (2015) only account for politicians positional roles institutionally driven formal functions I will especially consider their preference roles (Searing 1994). Each politician 10

11 takes on only one positional role, like minister or member of parliament. The variation between individual politicians based on these fixed roles is therefore limited, and in this classification regular MPs all fall under the same category. Preference roles, on the other hand, are informal roles MPs can choose to take on in addition to their positional role, and are much less constrained by institutions and therefore may be shaped by the motivations and preferences of the MPs themselves (Vonno 2012, 121). These roles have proven to be important to understand the behavior and activities of parliamentarians, and scholars have produced various typologies of legislative roles (for an overview see Andeweg 2014). Dolezal, Ennser-Jedenastik and Müller (2015) theorize that the differences they find between public office holders and party office holders are a consequence of the different incentive structures that come with these positions. They argue that public office holders have less incentives to pursue attack politics. The job of prime ministers, for instance, is to keep the government running and to resolve conflict: they should rather be motivated to avoid negativity. Furthermore, they discuss the extent to which politicians are available for party activity and claim that politicians with more time flexibility and accessibility, such as floor leaders and party secretaries, are better able to contribute to the public political debate and to generate negative attention for political rivals. In the following I apply these ideas to representatives in general, concentrating on two preference roles that I expect to be related to the attack behavior of politicians. First, I consider the extent to which politicians specialize in certain policy areas (Searing 1994; Navarro 2012). While some politicians dedicate themselves to one or two issues, others prefer to speak and act upon a wide range of issues from different policy areas. These preferences affect a politician s incentives and availability to attack opponents. Searing describes specialists as parliamentarians who take a distinctly serious and systematic approach to influence. They aim to influence policy carefully, quietly, persistently, and effectively (1987, 439). Specialists are pragmatic, want to influence the decision-making process and do not mind getting into technical details (Navarro 2012). They differ 11

12 considerably from politicians who are on the other side of the spectrum, what we might call generalists. These politicians would describe themselves a kind of freewheeling, freelancing trouble shooter and they prefer to be in the chamber and around the lobby and corridors where they can criticize ministers and launch ideas (Searing 1987, 434/441). Second, I account for the extent to which representatives focus on achieving policy goals. Politicians who are mainly concerned with making policy, also called policy advocates, are generally in a cross party mode, in which ministers and MPs combine to interact on the basis of cross-party interests (Andeweg 1997, 115). The loyalty of policy advocates lies less with the parliament or with their party and its program: it lies with certain interests regardless of who is defending them. They therefore have less incentives to be a party warrior (Sevenans, Walgrave, and Vos 2015) and can be expected to be less focused on demonstrating the weaknesses of opponents. Furthermore, policy advocates are mainly busy with governing-related tasks such as developing bills and amendments, often technical and slow processes, which makes them less time flexible and available for party work. In sum, both specialists and policy advocates are expected to be less partisan in their goals and in their behavior. I expect them to be less available and less interested in the genuine party work than other representatives, and therefore to be less focused on criticizing politicians from other parties. They are not so much interested in the ideological struggle between parties, but more in the policy areas they have specialized in or the policy goals they want to achieve. The final hypotheses of this study are: H4: The more a politician is specialized in a few policy areas, the less he or she is focused on demonstrating the weaknesses of political opponents. H5: The more a politician wants to achieve policy goals, the less he or she is focused on demonstrating the weaknesses of political opponents. 12

13 Data and Methods The data consist of surveys with Belgian cabinet members and members of parliament (MPs). The survey, administered on ipads, is part of a series of face-to-face interviews with politicians conducted between March and July In total, 272 out of the population of 413 politicians participated, leading to a response rate of 65.8%, which is very decent for elite research. 230 politicians from 11 political parties answered all questions that are used in this study. The dependent variable negativity is measured via the following question: Parties have different goals. Within a faction, a division of labor may occur, whereby certain members of the faction focus on one goal, whereas other members focus on other goals. Can you indicate the extent to which you, compared to your colleague faction members, focus on the following: Demonstrate the weaknesses of other parties. The variable is measured via a slider with a scale from 0 (I focus on this goal very little) to 100 (I focus on this goal very much). There are five independent variables. First, I code whether the politician is part of a government party (0) or an opposition party (1). Second, I categorize parties regarding their ideological extremity, going from 0 to 3. Third, I distinguish between female (0) and male (1) politicians. Fourth, to measure the degree to which politicians specialize in certain policy areas, the following question was asked: Some politicians specialize in one or two policy areas, while others prefer to speak and act upon a wide range of issues from different policy areas. Where would you place yourself on the following scale? The variable was measured via a slider with a scale going from 0 to 10. The variable was recoded so that 0 measures a large number of policy issues and 10 measures a small number of policy issues. Finally, politicians level of policy advocacy was operationalized via the following question: Parties have different goals. Within a faction, a division of labor may occur, whereby certain members of the faction focus on one goal, whereas other members focus on other goals. Can you indicate the extent to which you, compared to your 13

14 colleague faction members, focus on the following: Realize policy goals. The variable is again measured via a slider with a scale from 0 (I focus on this goal very little) to 100 (I focus on this goal very much). Additionally, the models include four control variables. To account for personal characteristics that might affect politicians inclination to attack opponents, I control for the extent to which politicians see themselves as a critical/quarrelsome person on the one hand, and a sympathetic/warm person on the other. The question wording was: The following questions are designed to assess certain personality traits and have been used extensively in a wide variety of international studies, including a number with MPs and CEOs. Although some of the traits may seem contradictory, we would like you to rate the extent to which a pair of traits applies to you, even if one characteristic applies more strongly than the other. I see myself as: Critical, quarrelsome ; Sympathetic, warm (1: strongly disagree, 2: disagree, 3: somewhat disagree, 4: neither agree nor disagree, 5: somewhat agree, 6: agree, 7: strongly agree). 2 Finally, I control for age and for whether the politician is a cabinet member or a MP. The descriptives of all variables are shown in Table 1. Table 1. Variable Descriptives (N=230) Mean Std. Dev. Min. Max. Negativity Opposition Extremity Male Specialist Policy Advocate Critical/quarrelsome Sympathetic/warm Age MP Results Belgian politicians generally say that, in comparison to their faction colleagues, they do not so much focus on demonstrating the weaknesses of other parties. On average they give a score of 38 points on the 14

15 Frequency negativity scale going from 0 to 100. It is difficult to draw conclusions from this figure alone, since politicians were explicitly asked to compare themselves with other members of the faction the variable measures a politician s relative negativity within a party and because I am not able to compare this result with results from other studies or countries. Nevertheless, the fact that the politicians in the sample on average give a relatively low score makes sense. First, Belgium is a typical European multiparty system. In multiparty systems, attacking opponents is more tricky than in two-party systems like the United States, where the large majority of negativity studies have been done. While the potential costs of criticizing other politicians in multiparty systems are entirely at the expense of the attacker, the benefits are shared with other parties. As a consequence, two-party systems seems to produce more negativity (Elmelund- Præstekær 2010). Furthermore, the questions were asked in a non-electoral context, and we can expect politicians to attack their competitors less outside the context of elections. As shown in Figure 1, however, the extent to which politicians are inclined to roll out the flaws of political opponents varies greatly. The surveyed representatives are dispersed across the scale. Table 2 shows regression models that try to explain this variation. Because politicians are nested in political parties, multilevel models with random intercepts are used to account for the nested structure of the data Negativity: Demonstrate weaknesses of other parties compared to faction colleagues Figure 1. Frequency distribution of the dependent variable (Negativity) Note: The responses on the scale were categorized for this figure. 15

16 The first three hypotheses of this study aimed to test whether factors that have shown to affect negativity during election campaigns also correlate with negativity during day-to-day politics. H1 expected politicians from opposition parties to be more prone to attack political rivals than representatives from government parties. The results in Model 3 of Table 2 confirm this hypothesis. Although the question to measure the dependent variable asked representatives to compare themselves to colleagues within their faction, we still find that MPs from opposition parties on average are significantly more likely to go negative than politicians from the ruling parties (B=6.119; p=.034). Marginal effects based on Model 3, keeping other variables at their means, show that politicians from government parties overall score 36 on negativity while politicians from opposition parties score 42 (results not in table). Second, I expected that politicians from ideologically extreme parties are more disposed to focus on demonstrating weaknesses of others than politicians from center parties. I do not find evidence for this hypothesis in Table 2. Extreme parties do not differ from moderately left and right parties, nor from center parties. Analyses with each category of extremity separately as a dummy variable in the regressions give the same results. Hypothesis 2, therefore, has to be rejected. Thirdly, I expected male politicians to be more negative than female politicians in H3. This hypothesis can be confirmed based on the regression results. Male politicians are more focused on demonstrating the weaknesses of other parties than female politicians (B=6.925; p=.014). Marginal means indicate that females on average score 34 on negativity, while males generally have a score of 41. Hence, two of the three determinants of negative campaigning that are tested in this study also explain negativity outside the electoral race. Furthermore, I formulated two hypotheses that did not receive scholarly attention yet. I expected that representatives who specialize in one or two policy areas are less disposed to confront political rivals with their incompetence and mistakes. The regression results show that specialists are indeed less negative than MPs who focus on a wide range of policy issues (B=-1.397; p=.010). Marginal effects show that politicians who are very much specialized (a score of 10 on the specialist scale) on average score 31 16

17 on Negativity while very generalist politicians averagely put the slider on 45. This confirms the fourth hypothesis. Finally, we can see in Table 2 that policy advocates are less keen on generating negative attention for opponents than colleagues who are not so much aimed at realizing policy goals (B=-.243; p=.001). Politicians who are very much committed to achieve policy goals score 30 on Negativity while politicians who are not at all interested in accomplishing policy aims have an average Negativity score of 54. Hypothesis 5 can be confirmed as well. Combining these results and using marginal means, we learn that on a scale from 0-100, male politicians from opposition parties who focuses on a wide range of issues and who are not policy advocates, on average score 68 on Negativity. Female politicians from governing parties, specialized in one issue and strongly focused on realizing policy goals on average score Table 2. Multilevel linear regressions with random intercepts and with Negativity as dependent variable MODEL 1 MODEL 2 MODEL 3 Party-level Opposition Party (3.062)*** 8.382(3.130)** (2.894)* Extreme Party (1.984) 2.083(2.114) (1.872) Politician-level Male 6.753(2.874)* (2.809)* Specialist (.561)** (.542)* Policy Advocate -.234(.077)** (.073)** Controls Critical/quarrelsome (.924)*** Sympathetic/warm (1.435) Age (.145) MP (7.402) Constant (2.832)*** (7.120)*** ( ) Prob > chi² Wald chi² (df) (2) 39.05(5) (9) Variation party-level (e. m.).000 (5.250) (5.250).000 (5.250) Variation individual-level (e.m.) (22.314) (22.314) (22.314) Notes: N politicians = 237; N parties = 11; e.m. = empty model between brackets; significant results are in boldface Most of the control variables do not significantly affect the dependent variable: members of parliament are not more negative than cabinet members and negativity has nothing to do with age or with being a sympathetic or warm person. But there is one exception. The more politicians describe 17

18 themselves as critical and quarrelsome, the more they are prone to attack their opponents (B=3.976; p=.000). Accordingly, politicians inclination to attack political rivals is not only a consequence of strategic considerations or of a certain division of labor within parties, but it is also a result of personal characteristics that make them more disposed to focus on flaws of opponents. Looking at Figure 2, which displays predicted probabilities, we can see that this is an important predictor compared to other variables tested in this study. The figure also shows that the individual-level variables policy advocate, specialist, and critical/quarrelsome are the three most important factors that account for the extent to which politicians aim to demonstrate the weaknesses of other political parties Figure 2. Predicted probabilities for negativity per variable Notes: Calculated based on the regression in Table 2, Model 3; all other variables are held at mean; 95 percent confidence intervals are shown; black line indicates predicted probabilities for the reference group. Discussion and Conclusion [To be written]. 18

19 Endnotes 1. Walter s (2013) research suggests that female party leaders are more likely to go negative. However, the difference between male and female leaders disappears in her study when Margaret Thatcher is left out of the analysis. 2. Politicians were asked to rate applicability for ten traits, in the following order: Extravert, enthusiastic; Critical, quarrelsome; Dependable, self-disciplined; Anxious, easily upset; Open to new experiences, complex; Reserved, quiet; Sympathetic, warm; Disorganized, careless; Calm, emotionally stable; Conventional, uncreative. References Andeweg, Rudy Role Specialisation or Role Switching? Dutch Mps between Electorate and Executive. The Journal of Legislative Studies 3 (1): Roles in Legislatures. In The Oxford Handbook of Legislative Studies, edited by Shane Martin, Thomas Saalfeld, and Kaare Strøm, Oxford University Press. Ansolabehere, Stephen, and Shanto Iyengar Going Negative. New York: Free Press. Arnold, R. Douglas Congress, the Press, and Political Accountability. Princeton University Press. Baumeister, Roy F., Ellen Bratslavsky, Catrin Finkenauer, and Kathleen D. Vohs Bad Is Stronger than Good. Review of General Psychology 5 (4): doi: / Buell, Emmett H., and Lee Sigelman Attack Politics: Negativity in Presidential Campaigns Since First Edition edition. Lawrence, Kan: University Press Of Kansas. Byrne, Sahara, and Philip Solomon Hart The Boomerang Effect A Synthesis of Findings and a Preliminary Theoretical Framework. Annals of the International Communication Association 33 (1): doi: /

20 Bystrom, Dianne Advertising, Web Sites and Media Coverage: Gender and Communication along the Campaign Trail. In Gender and Elections, Shaping the Future of American Politics, Cambridge: Cambridge University Press. Carlson, Tom Gender and Political Advertising across Cultures: A Comparison of Male and Female Political Advertising in Finland and the US. European Journal of Communication 16 (2): doi: / Davis, Michael L., and Michael Ferrantino Towards a Positive Theory of Political Rhetoric: Why Do Politicians Lie? Public Choice 88 (1/2): Dolezal, Martin, Laurenz Ennser-Jedenastik, and Wolfgang C Müller Who Will Attack the Competitors? How Political Parties Resolve Strategic and Collective Action Dilemmas in Negative Campaigning. Party Politics, November, doi: / Druckman, James N., Martin J. Kifer, and Michael Parkin Timeless Strategy Meets New Medium: Going Negative on Congressional Campaign Web Sites, Political Communication 27 (1): doi: / Elmelund-Præstekær, Christian Negative Campaigning in a Multiparty System. Representation 44 (1): Beyond American Negativity: Toward a General Understanding of the Determinants of Negative Campaigning. European Political Science Review 2 (1): doi: /s Evans, Heather K., and Jennifer Hayes Clark You Tweet Like a Girl! : How Female Candidates Campaign on Twitter. American Politics Research 44 (2): doi: / x Fridkin, Kim Leslie, and Patrick J. Kenney The Role of Gender Stereotypes in U.S. Senate Campaigns. Politics & Gender 5 (3): doi: /s x

21 Geer, John G In Defense of Negativity: Attack Ads in Presidential Campaigns. University of Chicago Press. Gregory, Peter A Comparing the Effectiveness of Positive and Negative Political Campaigns. Inquiries Journal 7 (11). Grossmann, Matt Who (or What) Makes Campaigns Negative? The American Review of Politics 33 (Spring): Hale, Jon F., Jeffrey C. Fox, and Rick Farmer Negative Advertisements in U.S. Senate Campaigns: The Influence of Campaign Context. Social Science Quarterly 77 (2): Hansen, Kasper M., and Rasmus Tue Pedersen Negative Campaigning in a Multiparty System. Scandinavian Political Studies 31 (4): doi: /j x. Harrington, Jr., Joseph E., and Gregory D. Hess A Spatial Theory of Positive and Negative Campaigning. Games and Economic Behavior 17 (2): doi: /game Haynes, Audrey A., and Staci L. Rhine Attack Politics in Presidential Nomination Campaigns: An Examination of the Frequency and Determinants of Intermediated Negative Messages Against Opponents. Political Research Quarterly 51 (3): doi: / Herrnson, Paul S., and Jennifer C. Lucas The Fairer Sex?: Gender and Negative Campaigning in U.S. Elections. American Politics Research 34 (1): doi: / x Hudson, Jennifer vanheerde The Americanization of British Party Advertising? Negativity in Party Election Broadcasts, British Politics 6 (1): doi: /bp Ito, T. A., J. T. Larsen, N. K. Smith, and J. T. Cacioppo Negative Information Weighs More Heavily on the Brain: The Negativity Bias in Evaluative Categorizations. Journal of Personality and Social Psychology 75 (4):

22 Jamieson, Kathleen H., Paul Waldman, and Susan Sherr Eliminate the Negative? Categories of Analysis for Political Advertisements. In Crowded Airwaves: Campaign Advertising in Elections, edited by James A. Thurber, Candice J. Nelson, and David A. Dulio, Brookings Institution Press. Kahn, Kim Fridkin Gender Differences in Campaign Messages: The Political Advertisements of Men and Women Candidates for U.S. Senate. Political Research Quarterly 46: Kahn, Kim Fridkin, and Patrick J. Kenney Do Negative Campaigns Mobilize or Suppress Turnout? Clarifying the Relationship between Negativity and Participation. The American Political Science Review 93 (4): doi: / How Negative Campaigning Enhances Knowledge of Senate Elections. In Crowded Airwaves, edited by James A. Thurber, Nelson, Candice J., and David A. Dulio, Washington, D.C.: Brookings Inst When Do Candidates Go Negative? In No Holds Barred: Negativity in U.S. Senate Campaigns, edited by Kim Fridkin Kahn and Patrick J. Kenney, Pearson/Prentice Hall. Krupnikov, Yanna, and Nichole M. Bauer The Relationship Between Campaign Negativity, Gender and Campaign Context. Political Behavior 36 (1): doi: /s Lau, Richard R., and Gerald M. Pomper Negative Campaigning by US Senate Candidates. Party Politics 7 (1): doi: / Negative Campaigning: An Analysis of U.S. Senate Elections. Rowman & Littlefield. Lau, Richard R., and Ivy Brown Rovner Negative Campaigning. Annual Review of Political Science 12 (1): doi: /annurev.polisci Lau, Richard R., Lee Sigelman, and Ivy Brown Rovner The Effects of Negative Political Campaigns: A Meta-Analytic Reassessment. Journal of Politics 69 (4): doi: /j x. 22

23 Lawless, Jennifer L Women, War, and Winning Elections: Gender Stereotyping in the Post- September 11th Era. Political Research Quarterly 57 (3): doi: / Lewicka, Maria, Janusz Czapinski, and Guido Peeters Positive-Negative Asymmetry or When the Heart Needs a Reason. European Journal of Social Psychology 22 (5): doi: /ejsp Mair, Peter, Wolfgang C. Müller, and Fritz Plasser Political Parties and Electoral Change: Party Responses to Electoral Markets. SAGE. Navarro, J The Cognitive Rationality of Role Choices: Evidence from the European Parliament. In Parliamentary Roles in Modern Legislatures, edited by M Blomgren and O Rozenberg, London: Routledge. Perloff, Richard M., and Dennis Kinsey Political Advertising as Seen by Consultants and Journalists. Journal of Advertising Research 32 (3): Peterson, David A. M., and Paul A. Djupe When Primary Campaigns Go Negative: The Determinants of Campaign Negativity. Political Research Quarterly 58 (1): doi: / Proctor, David E., William J. Schenck-Hamlin, and Karen A. Haase Exploring the Role of Gender in the Development on Negative Political Advertisements. Women & Politics 14 (2): doi: /j014v14n02_01. Robinson-Riegler, Gregory L., and Ward M. Winton The Role of Conscious Recollection in Recognition of Affective Material: Evidence for Positive-Negative Asymmetry. The Journal of General Psychology 123 (2): doi: / Rozin, Paul, and Edward B. Royzman Negativity Bias, Negativity Dominance, and Contagion. Personality and Social Psychology Review 5 (4): doi: /s pspr0504_2. 23

24 Schweitzer, Eva Johanna Global Patterns of Virtual Mudslinging? The Use of Attacks on German Party Websites in State, National and European Parliamentary Elections. German Politics 19 (2): doi: / Searing, Donald New Roles for Postwar British Politics: Ideologues, Generalists, Specialists, and the Progress of Professionalization in Parliament. Comparative Politics 19 (4): doi: / Westminster s World: Understanding Political Roles. Harvard University Press. Sevenans, Julie, Stefaan Walgrave, and Debby Vos Political Elites Media Responsiveness and Their Individual Political Goals: A Study of National Politicians in Belgium. Research & Politics 2 (3): 1 7. Skaperdas, Stergios, and Bernard Grofman Modeling Negative Campaigning. The American Political Science Review 89 (1): doi: / Soroka, Stuart, and Stephen McAdams News, Politics, and Negativity. Political Communication 32 (1): doi: / Strömbäck, Jesper Four Phases of Mediatization: An Analysis of the Mediatization of Politics. The International Journal of Press/Politics 13 (3): Theilmann, John, and Allen Wilhite Campaign Tactics and the Decision to Attack. The Journal of Politics 60 (4): doi: / Thesen, Gunnar When Good News Is Scarce and Bad News Is Good: Government Responsibilities and Opposition Possibilities in Political Agenda-Setting. European Journal of Political Research 52 (3): doi: /j x. Trent, Judith S., and Teresa Sabourin Sex Still Counts: Women s Use of Televised Advertising during the Decade of the 80 s. Journal of Applied Communication Research 21 (1): doi: /

25 Vonno, Cynthia M. C. van Role-Switching in the Dutch Parliament: Reinvigorating Role Theory? The Journal of Legislative Studies 18 (2): doi: / Walter, Annemarie S Women on the Battleground: Does Gender Condition the Use of Negative Campaigning? Journal of Elections, Public Opinion and Parties 23 (2): doi: / Negative Campaigning in Western Europe: Similar or Different? Political Studies 62 (April): doi: / Walter, Annemarie S., and Wouter van der Brug When the Gloves Come off: Inter-Party Variation in Negative Campaigning in Dutch Elections, Acta Politica 48 (4): doi: /ap Walter, Annemarie S., Wouter van der Brug, and Philip van Praag When the Stakes Are High: Party Competition and Negative Campaigning. Comparative Political Studies 47 (4): doi: / Walter, Annemarie S., and Rens Vliegenthart Negative Campaigning across Different Communication Channels: Different Ball Games? The International Journal of Press/Politics 15 (4): doi: /