Charismatic Communication Skill, Media Legitimacy, and Electoral

Charismatic Communication Skill, Media Legitimacy, and Electoral Success Tamir Sheafer The Hebrew University, Jerusalem

ABSTRACT. There is growing interest in how much a political candidate’s media and political skills, sometimes referred to as ‘‘charisma,’’ are critical to their electoral success and political survival. According to one hypothesis, the effects of media skills might be so strong that actors who do not possess media skills quickly vanish from the political arena. However, while many believe that media skills are an important aspect of being elected, few have actually tried to measure their effects. Part of the reluctance, no doubt, stems from methodological difficulties in measuring them. In this study, I look more directly at the media and political skills that are likely to play an important role in electoral competition and suggest a new methodology for measuring such skills. Findings are presented that show that media and political skills, as well as media coverage, can have an important impact on long-term electoral success of Israeli members of the Knesset.

Tamir Sheafer is Assistant Professor, Department of Communication and Department of Political Science, The Hebrew University, Jerusalem, Israel. The author expresses his thanks to Gadi Wolfsfeld and Shaul Tzionit for their contribution and helpful comments on earlier versions of this article. Address correspondence to: Tamir Sheafer, The Hebrew University, Mount Scopus, 91905 Jerusalem, Israel (E-mail: [email protected]). Journal of Political Marketing, Vol. 7(1) 2008 Available online at http://jpolm.haworthpress.com # 2008 by The Haworth Press. All rights reserved. doi: 10.1080/15377850802063983

1

2

JOURNAL OF POLITICAL MARKETING

KEYWORDS. Candidate quality, charisma, media effects, media skills, political campaigns, political communication

The main goal of this study is to analyze and measure the direct influences of media and political skills of political actors, and of their media coverage, on their electoral success. The influence of the first of them in particular is attracting increasing interest due to the popular, professional, and academic understanding that media skills are extremely important for political survival (Altheide and Snow 1979; Iyengar 2001; Meyrowitz 1985). For example, it is widely assumed that one of the major variables that explain the electoral success of leaders such as Reagan, Clinton, Berlusconi, and Blair is their media skills and performance. However, in spite of the recognition of the electoral importance of media skill, few have actually tried to measure it. Part of the reluctance to measure this variable no doubt stems from the difficulties in measuring it. As for now, the field of political communication lacks theoretical conceptualizations and empirical examinations of the role of media skill of political actors in the electoral process. During the last two decades some students of communication have criticized the lack of research that focus on individual political actors, and emphasized the need to advance a different approach, a source-centered, or political-actor-centered approach (Blumler 1993; Gans 1979; Ericson, Baranek, and Chan 1989; Schlesinger 1993). This is what I attempt to develope and test here. Some political science studies did analyze the influence of actual political qualities and skills of candidates on their electoral success. For example, Mondak (1995) conducted direct measures of candidates’ quality, and found that ‘‘the electoral system does work to screen incumbents on the basis of quality’’ (p. 1056). But few of these studies include media-related or even political behavioral qualities. The study presented here is among the first attempts made to empirically analyze the influence of actual media and political skills on the electoral success of political actors. In order to conduct such analyses there is a need to take a more actor-centered approach to the issue. Such Charismatic communication skill approach is conceptualized and tested here. It focuses on certain personal resources of political actors that are expected to be influential in their competition for media coverage. Due to the growing overlap between the political and the political communication arenas (Bennett

Tamir Sheafer

3

2003; Blumler and Kavanagh 1999; Ericson, Baranek, and Chan 1989; Mazzoleni and Schulz, 1999), it is hypothesized that these ‘‘media’’ resources will also be of crucial importance in the electoral contest.

THE CONTEST OVER MEDIA LEGITIMACY IN THE POLITICAL COMMUNICATION ARENA The Arena’s ‘‘Rules of the Game’’ The political communication arena is characterized by a competition between political issues, events, messages, and actors over dominating the arena as much as possible (Blumler and Gurevitch 1986; Gans 1979; Schlesinger 1993). This competition is guided by the arena’s ‘‘rules of the game.’’ The news, according to a popular definition, is what is important and interesting (Cook 1989, 1996; Gans 1979; Gitlin 1980). What is important and interesting is strongly influenced by the arena’s ‘‘rules of the game,’’ which are made by two categories of values. The first category is politicalcultural values, which is an ideological system that comprises the symbolic center of the nation—the consensus—and influences judgments of which events, issues, and actors are important and which are not. Media attention is therefore an inverse function of the distance from this center. For example, the strongest political actors usually receive the highest media attention (Bennett 2003; Cook 1989, 1996; Ericson et al. 1989; Gans 1979; Graber 1993; Hilgartner and Bosk 1988; Shoemaker and Reese 1991; Wolfsfeld 1997). The second category is the media’s professional values and needs, which originate in the needs of media organizations and the values of their employees. Generally, they underline the importance of adapting to media routines, deadlines, and expectations, as well as the taste for drama, conflict, novelty, colorful events, fiction-like story line, strategy, and personalization (Altheide and Snow 1979; Cook 1996; Ericson et al. 1989; Gans 1979; Gitlin 1980; Graber 1993; Mazzoleni and Schulz 1999; Shoemaker and Reese 1991; Wolfsfeld 1997).

Political Resources, Charismatic Resources, and Legitimacy It is claimed here that from the actors’ point of view, the best way to understand their contest within the political communication arena is to focus on their goal, which is to achieve media legitimacy—that is,

4


legitimacy in the political communication arena. Generally, legitimacy can be defined as public acceptance of domination (Mueller 1973; Yatziv 1998). Since no political actor can achieve complete control over the political communication arena, dominating the arena amounts to systematically achieving better media attention–more access to or coverage by the media, as well as better control over media valence. But why are such systematic advantages provided to certain political actors and not others? This question focuses on which individuals are most likely to succeed in the ongoing competition over publicity. Among the most important factors that may help in this competition–the factors that affect media legitimacy–are political standing, charismatic communication skills, and parliamentary activity. Political standing is important because, as mentioned above, the strongest political actors usually receive the highest media attention. That is, political standing is important because the political and cultural values in the ‘‘rules of the game’’ underline the centrality of the national order. Parliamentary activity of political actors should be important, because we expect the media to monitor the work of our elected officials. And charismatic communication skills are important because they represent the actor’s ability to adapt to the media’s professional values and needs—the second category in the ‘‘rules of the game.’’ Clearly, the construct of charismatic communication skill is much more difficult to specify and measure than the other variables. It will therefore be necessary to devote more space in order to explain the development of this variable. Going back to Max Weber, the validity of one of the major claims for legitimacy is based on charismatic grounds (Eisenstadt 1968; Matheson 1987; Muller 1973; Shils 1965; Weber 1947=1986). Weber defines charisma as a relationship between a leader and her=his followers. Later charisma scholars broadened his definition by asking what there was in the political actors themselves, their skills, behavior, performance, etc., which causes the charismatic relationship (Eisenstadt 1968; Gardner and Avolio 1998S; House and Howell 1992; House, Spangler, and Woycke 1991; Shamir, House, and Arthur 1993; Shils 1965). For example, ‘‘behavioral charisma’’ is ‘‘based on the actual or presumed behavior of the leader’’ (House et al. 1991, p. 366). These certain behaviors, skills, and qualities will be called here the ‘‘charismatic communication skill’’ categories. Based on Shils (1965),

Tamir Sheafer

5

it is argued here that the actor who seemed endowed with charisma is in fact demonstrating a higher level of playing by the arena’s ‘‘rules of the game.’’ It is essential to note here that ‘‘charismatic communication skill’’ is not synonymous with ‘‘charisma’’ since the latter includes also the leader-followers relationship, while the first focuses only on the skills or behavior of the leader.

Charismatic Communication Skill and Media Legitimacy Charismatic communication skill is defined as the actor’s demonstrated skills, behaviors, performance, and talent that are pertinent to the main arena in which (s)he operates. Since each arena raises somewhat different requirements (Hilgartner and Bosk, 1988), different ‘‘rules of the game,’’ the composition of the charismatic communication skill categories changes between arenas (for similar arguments regarding charisma see Bass 1988; Eisenstadt 1968; Steyrer 1998). Therefore, in the political communication arena we are looking for the communication and political skills and behaviors that are likely to bring about the attribution of charisma. These are also the skills and behaviors that are likely to make the political actor more newsworthy, to increase and maintain her or his media legitimacy. Although the concept ‘‘charismatic communication and political skills and behaviors’’ might be more appropriate, we would use here ‘‘charismatic communication skill’’ for simplicity. In developing a coherent list of skills, it is necessary to combine the fields of research that focus on the actors and the field of research that explains the arena in which they function. The requirements of the political communication arena are represented by the media’s professional values and needs, as appear in many political communication studies. Analysis of actors’ performance can be found in leadership and charisma studies that focused on behaviors that are likely to bring about attribution of charisma and in political psychology studies that analyze the political performance of political actors. Leadership and charisma research. Some leadership and charisma scholars agree with the claim that specific actor behavior and skills produce the attribution of charisma (Bass 1988; Bryman 1993; House et al. 1991; Shamir et al. 1993). They mention the following behaviors and skills as affecting the attribution of charisma: rhetorical and speech ability (Bryman 1993; Shils 1965); nonverbal and symbolic communication (House and Howell 1992; Shamir et al. 1993); creativity

6


(House and Howell 1992; Shils 1965); energy (House and Howell 1992); and intelligence and cognitive abilities (Bass, 1988). Political psychology research. Renshon (1995; 1996) categorizes political styles and skills of political leaders by analyzing visible political behavior, which is a function of deep character structure (see also Barber 1992; Greenstein 1992; Simonton 1988). He identifies three categories of skills: First, ambition and determination—these refer to the ‘‘capacities to commit oneself to one’s purpose, to endure frustration, and to maintain determination’’ (Renshon 1996, p. 196). This is necessary in the political communication arena, since ‘‘the news is weighted toward sources which are eager to provide information’’ (Gans 1979, p. 117). Second, interpersonal skills—these refer to the actor’s ability to cooperate with others, to form alliances and to persuade. In the political communication arena it is essential to form personal and professional relationship with journalists, as part of the competition over access to the media and media framing. Cooperation with other politicians might be also important. For example, Sellers (2000, p. 24) found that ‘‘If a large number of legislators join in promoting the same message to the national media in Washington, that message is more likely to receive national news coverage.’’ Third, cognitive-creative skills—these help in resolving problems, analyzing different situations the actor is faced with, and developing creative initiatives. They enable the actor to develop a high oratorical ability, or, in Greenstein’s (1995, p. 140) words, ‘‘verbal intelligence.’’ In the political communication arena, these skills nourish the ability to comprehend and develop some of the most important skills required by the ‘‘rules of the game’’ of the arena, and especially by the media’s values and needs. Political communication research. The media’s values and needs determine to a large extent which skills will be important in the arena. Among the relevant skills mentioned in political communication research are the ability to adapt to the routines and formats of news organizations (Ericson et al. 1989; Gans 1979; Shoemaker and Reese, 1991), deep understanding of the ‘‘good story’’ structure (Altheide and Snow 1979; Cook 1996), demonstrated rhetorical abilities (Cook 1989; Gans 1979; Mazzoleni and Shulz 1999) that fit the sound-bites requirements, a flair for dramatic performance (Ericson et al. 1989; Gitlin 1980; Meyrowitz 1985), know-how in creating and promoting media events (Bennett 2003; Gans 1979),

Tamir Sheafer

7

and the ability to develop working relationship with journalists (Bennett 2003; Gans 1979). When developing the charismatic communication skill categories I tried to represent the actor’s skills and behaviors and the specific requirements of the arena and to create categories suitable for empirical analysis. For example, one may find representations of Renshon’s (1996) three categories of skills in the five charismatic skill categories below: ambition and determination (in categories 1 & 2), interpersonal skills (in categories 4 & 5), and cognitive-creative skills (in categories 1, 2 & 3). These are presented here in visible behavioral categories that are relevant to the specific requirements of the political communication arena. Since the emphasis is on skills and behaviors, other possible influential variables, such as physical appearance, were not considered. It should be noted that the five categories do not cover all activities, characteristics, and skills that influence the attribution of charisma. But they do seem to be among the most important ones: (1) political initiative and creativity; (2) communication initiative and creativity; (3) rhetorical and dramatic abilities; (4) cooperation with politicians; and (5) cooperation with journalists. Note that categories 1 & 4 are not direct communication skills. But these are skills required from political actors in the political communication arena in order to make them newsworthy. An analysis of media legitimacy of Members of the Israeli Knesset (MKs), which is presented elsewhere (Sheafer 2001), reveals a strong and statistically significant influence of the charismatic communication skill on the media legitimacy of these MKs. Media legitimacy is defined as achieving higher media prominence (the amount and salience of media coverage) and media valence (the quality of media coverage). The charismatic communication skill is the most important variable that affects media legitimacy, and its effect is much stronger than that of political standing and of the level of parliamentary activity (which does not affect media legitimacy at all).

MEDIA AND POLITICAL SKILL, MEDIA LEGITIMACY, AND THE CONTEST OVER ELECTORAL SUCCESS IN THE POLITICAL ARENA From the perspective of the present study, it is possible to ask two central questions regarding the influence of the political

8


communication arena on the electoral success of political actors. First, what is the influence of actors’ charismatic communication skills on their electoral success? Second, what is the effect of actors’ media legitimacy (media prominence and media valence) on their electoral success? Note that charismatic communication skills and media legitimacy are two separate variables. Although charismatic communication skill strongly influences media legitimacy (Sheafer 2001), media legitimacy is also influenced by political standing. While media legitimacy is a measure of media coverage, charismatic communication skills is a behavioral variable that affects the amount and valence of media coverage, alongside other variables. Electoral success, the dependent variable in this study, is the success of political actors in the parties’ primaries.

The Influence of Charismatic Communication Skills on Electoral Success What kind of quality and performance is required from political actors pursuing electoral success? Many political science and communication scholars believe that the performances required are closely related to political substance and that voters support candidates who perform their political duties well (Jacobson 1992; Mazzoleni and Schulz 1999; Wattenberg 1995; Zaller 2001). Quite similarly, one actorcentered study found that high-quality incumbents face greater electoral success compared with low-quality incumbents (Mondak 1995). Quality is defined in that study in terms of various political and personal traits, such as being effective, hard-working, and well informed. No media-related activities or skills are analyzed in that study. An alternative approach concentrates on the media skills of the actors. Altheide and Snow (1979) claim that ‘‘the criteria used for assessing a candidate’s competence depends on his TV performance’’ (p. 109). Nimmo and Savage (1976) argue that candidates’ image is influenced by evaluations of their communication style (use of media, qualities as a speaker, dramatic persona), in addition to their political role. Just, et al. (1996), who analyzed voter decision-making during the 1992 US presidential campaign, found that voters mentioned most the candidates’ performances as campaigners. In the age of modern media campaigns (Swanson and Mancini, 1996), it can be hypothesized that evaluating a candidate’s performance as a campaigner is actually evaluating her=his charismatic communication skill.

Tamir Sheafer

9

Consequently, it is possible to hypothesize that, H1: There is a positive influence of the charismatic communication skill of the political actor on her=his electoral success.

The Effect of Media Legitimacy on Electoral Success In the age of candidate-centered politics the ‘‘electorate . . . has grown accustomed to looking directly at the candidates through the media’’ (Wattenberg 1995, p. 21). Scholars recognize that the image of the candidates has an impact greater than previously assumed (Boiney and Paletz 1991; Nimmo and Savage 1976; Wattenberg 1995)—as does affection toward them (Just et al. 2001; Shamir and Arian, 2002)— on the vote. And it is generally believed that media coverage affects these variables quite strongly (Hellweg 1995; Just et al. 2001; Nimmo and Savage, 1976). Why media legitimacy has a greater effect in the primaries: Despite a scarcity of empirical and quantitative evidence, many scholars argue that the influence of media coverage is much stronger in the primary campaigns compared with the general elections. There are two main reasons for that. First, party identification, a central variable in the general elections, is not relevant in primaries. Second, there is low level of information about each of the large number of candidates in the primaries, which make any piece of information about them valuable (Bartels 1988; Gelman and King 1993; Hazan 1997). Information is so important because voters are unlikely to support candidates they do not know enough about (Bartels 1988; Cook 1989; Just et al. 1996). Consequently, one can hypothesize that media legitimacy would positively influence electoral success in a primary election. Specifically, H2: There is a positive influence of the media legitimacy of the political actor on her=his electoral success.

THE SETTING: PARTY ELECTIONS IN ISRAEL It is possible to distinguish between two kinds of party elections in the 1999 Israeli election. Labor conducted a closed direct primary, in

10


which all dues-paying party members were allowed to vote (Hazan 1997). Five other parties—Likud, Meretz, Israel Be’aliya, Hadash, and Mafdal—conducted indirect elections, in which only a relatively small number of members (up to a few thousand) who belong to the party’s central committee were allowed to vote. The exact voting method in each of these five parties was different, but generally each voter was asked to select a few candidates (up to 14 in the Likud). The Labor Party’s primary was divided into two levels: national and local. A candidate could run either on the national list or in a district. On the national list, each party member was asked to vote for 9 to 11 candidates and on the districts’ lists for one candidate. The top of the list, up to number 16, was reserved for candidates from the national list.

METHODOLOGY The research group for analyzing the charismatic and political resources and media legitimacy included 91 members of the 14th Israeli Knesset (1996–1999).1 The research group for analyzing the electoral success included 38 members of the original group, who belonged to parties that conducted party primaries or elections.2 A content analysis was conducted of the media coverage of these MKs in the three leading Israeli daily newspapers (Yediot Aharonot, Ma’ariv, & Ha’aretz) and channel 1 the public television station. These are the media of choice for the vast majority of the Israeli public, and hence the decision to analyze them. Sixty days were chosen randomly from the research period, and all newspaper articles and TV news items that appeared during those days and covered MKs, even in a minor way, were analyzed. Overall, 1,809 items were analyzed, but since the unit of analysis was the media coverage of a single MK in a single article, there was a total of 4,355 coding items (usually a few MKs are covered in a single news item).

Dependent Variable Electoral success: Electoral success is defined as the MKs’ relative success in the party elections. For each party, the MK who reached the first place in the primary received the maximum points, which is equal to the party’s number of seats in the Knesset at that time.

Tamir Sheafer

11

For example, Labor had 34 seats, and MK Ben-Ami, who reached the first place, received the score of 34. Then, this number was divided by the party’s number of seats in the Knesset. This allowed for equating the relative success of MKs from large and small parties.

Control Variables Political standing. Political standing may positively influence electoral success, as is the case with the decisive influence of incumbency, which is discussed in numerous studies. All MKs in this study were incumbents, yet political standing differences may account for some of their electoral success. Therefore, it was decided to hold this variable constant. The actor’s political standing represents her=his hierarchic position in the Knesset and in her=his party. The coding of political standing was as follows: high ¼ chairmanship position, party and section leaders; medium ¼ none of the above but more than one term in the Knesset; low ¼ none of the above. Parliamentary activity. Parliamentary activity is introduced in this study as an alternative explanation to charismatic communication skill. It represents quality and electoral success based on substantive activity, rather than on media skill. Three parliamentary activity indicators were analyzed, based on the official statistics provided by the Knesset: (1) proposing a law; (2) initiating a law; (3) passing a law. Each activity was given a different weight, based on the amount of time and energy needed to implement it. For example, it takes more time and energy to mobilize the MKs to pass a law than to propose a law, and therefore, passing a law was given a bigger weight than was given to initiating one. This scale is based on a specific measure developed within the Knesset, designed to evaluate the workload of different MKs. Number of staff members. The number and quality of the MKs’ staff might be a threat to the validity of the concept of charismatic communication skill. One may argue that some of the categories of charismatic communication skill, such as cooperation with journalists, might reflect the skills of the MK’s staff as much as the personal skills of the MK. Consequently, one might surmise that MKs with more savvy staff members would score higher on the charismatic communication skill categories than other MKs. This threat, however, is minimized here. First, during the 14th Knesset MKs received budget sufficient to employ only a single parliamentary assistant.

12


This was usually a student with no prior public relations experience. Therefore, the quality of staff was not a major variable in the Israeli case. Indeed, this is probably a more important variable in other parliaments where members employ more professional staff. Second, some MKs indeed used other resources to employ an additional staff member, or even a few. Most of these were students or other nonprofessionals as well. To control for the number of staff members this variable was added as a control variable to the regression analyses. Third, as can be seen below, the experts that rated the MKs on the charismatic communication skill categories were asked to rate the skills of the MKs, not of their staff. Since most experts knew personally both the MKs and their staff, it was possible for them to make that separation. Success in the previous elections. Success in the previous elections (the 1996 party elections) may positively influence electoral success in the 1999 elections. To control for this possibility, the relative success in the 1996 elections was added to the model. The computation is similar to that conducted for the dependent variable.

Independent Variables Media legitimacy: The media legitimacy of the political actors is represented by the systematic amount, salience, and quality (positive versus negative) of media coverage that they were provided. Media legitimacy is composed of two indexes: media prominence, which represents the actors’ extent of media coverage; and media valence, which represents the actors’ ability to promote media framing that presents them as legitimate actors. Media prominence is composed of a number of variables, among them the number of articles, the section and page in which they were placed, their size, and the prominence of the actor within each item. In the media valence index, the item is marked as positive, as presenting the actor as legitimate, if it presents the MK mainly as (1) a leader or representative of some position (but not when this leadership is due to her=his official role), (2) an expert on the specific issue discussed (but not due to her=his official role), (3) a contributor to society due to her=his innovative, professional and responsible acts, (4) having positive characteristics. The item is marked as negative if the opposite is true, and it is marked as neutral if positive and negative remarks are mixed, or they do not exist. A coding system was devised to measure these various variables

Tamir Sheafer

13

that form media prominence and media valence. The researcher and three more trained coders conducted the content analysis. The intercoder reliability (using Scott’s pi), tested in a session joined by all four coders for 96 coding items, was .90 in the media prominence index, and .76 in the media valence index. The final media legitimacy variable was created by multiplying media prominence and media valence.3 It is important to note that the period of media coverage analyzed here was both relatively long (one year) and, most importantly, ended before the primary campaigns started. As such, it is different from other studies that concentrate on media coverage of candidates only during the campaign. This might be important, as some scholars argue, because it is plausible that many voters make up their minds before the campaign even begins, especially when the candidates, such as the incumbent MKs, are relatively known (Bartels 1993; Finkel 1993; Iyengar and Simon 2000; Lang and Lang in Klapper 1960). In Israeli party primaries this is even more important, because the party members who vote in these elections know most of the incumbent MKs before the campaign begins. Charismatic communication skill. Assessing charismatic communication skills was understandably complicated. The analyses relied on information from experts who knew the actors personally, have often seen them ‘‘in action,’’ and could therefore provide first hand knowledge about their skills in this area. Thirty-nine experts were used for the analysis of the MKs’ charismatic communication skill categories. These experts–MKs, parliamentary and political reporters, and other advisers and professionals–knew the MKs from close proximity, in an unmediated way. In a face-to-face session, each of them rated the MKs on the five skill categories (political initiative and creativity; communication initiative and creativity; rhetorical and dramatic abilities; cooperation with politicians; and cooperation with journalists) and on an additional category defined only as ‘‘charisma.’’ This category was added to examine the extent to which the skill categories represented the perceived charisma of the actor. A unique methodology was developed to enable collecting such a high number of responses from each expert. This methodology and the analysis of the relevant data are presented in the appendix. The analysis shows that the experts rated each MK as a whole. Consequently, I created a single variable named ‘‘charismatic communication skill’’ for each MK.

14


RESULTS The Influence of Charismatic Communication Skill on Electoral Success A hierarchical regression analysis4 was conducted to examine the first hypothesis, which states that there is a positive influence of the charismatic communication skill of the political actor on her=his electoral success. In the first step the four control variables (political standing, parliamentary activity, number of staff and success in the previous party elections) were introduced, while charismatic communication skill, the independent variable, was introduced in the second step. As can be seen in Table 1, this model provides a strong prediction for electoral success, accounting for almost two thirds of the variance in the electoral success of Mks (R2 Adj. ¼ .61). Among the control variables, only success in the previous party elections has a significant and positive contribution to the model (b ¼ .41; p .05). The prediction power of the model with these four variables alone (first block) is weak (R2 Adj. ¼ .15). The association with, and the influence of, parliamentary activity on electoral success is negative, although not significant (b ¼ .03; p > .1). Apparently, being active in the Knesset does not contribute to success in the TABLE 1. Hierarchical Multiple Regression Analysis of Electoral Success as a Function of the Control Variables (first block) and Charismatic Communication Skill (second block)

Political standing Parliamentary activity Number of staff Success in the previous election Charismatic communication skill R2 Change F Change for R2 Change

First Block b (SE)

Second Block b (SE)

.04 (.07) .03 (.04) .08 (.07) .41 (.16)

.04 (.05) .01 (.03) .009 (.05) .19 (.12) .21 (.03) .42 40.62

.24

N ¼ 38. p .05; p .001, two tailed. The control variables entered in the first block and charismatic communication skill in the second block (using SPSS). Global statistics for the whole model: R2 Adj. for the first block ¼ .15 and for the second block ¼ .61; F for the first block ¼ 2.65 and for the second block ¼ 12.79; df ¼ 37.

Tamir Sheafer

15

primaries. The finding that the contribution of political standing is not significant is surprising at first, because the influence of this variable, which usually represents incumbency, is expected to be very strong and positive. However, incumbency was held constant in this study, because all MKs in the research group were incumbent. When introducing charismatic communication skill in the second step, the prediction power of the model increases dramatically (R2 change ¼ .42). The influence of charismatic communication skill on electoral success is proven to be positive and significant (b ¼ .21; p .001), as well as the increment in the prediction power of the model (F change for R2 change ¼ 40.62; p .001). The first hypothesis is therefore strongly supported. Since the levels of MKs’ charismatic communication skill were measured eight months before the primary elections, this single variable demonstrates a very strong forecasting ability.

The Influence of Media Legitimacy on Electoral Success According to the second hypothesis, there is a positive influence of media legitimacy of political actors on their electoral success. This hypothesis is also strongly supported. As can be seen in Table 2, a regression analysis of electoral success as a function of media legitimacy and success in the previous party elections, the only control variable that has a significant influence on electoral success (see Table 1), reveals three major findings: First, media legitimacy predicts one half of the variance in MKs’ electoral success (R2 Adj. ¼ .50). Second, the influence of media legitimacy is positive and significant (b ¼ .44; p .001). Third, the influence of success in the previous party elections only approaches significance at the p .05 level (b ¼ .24; p ¼ .06). TABLE 2. Multiple Regression Analysis of Electoral Success as a Function of Media Legitimacy and Success in the Previous Elections

Media Legitimacy Success in the Previous Elections Number of Cases: 38 R2 Adj.: .50

b (SE)

Statistical Significance

.44 (.09) .24 (.13)

p < .000 p ¼ .061

16


The Combined Influence of the Political Communication Arena on Electoral Success It is interesting to test the influence of the two independent variables originated in the political and political communication arenas (charismatic communication skill and media legitimacy) on electoral success in a single model. However, because there is a high multicollinearity between the independent variables (as noted above, charismatic communication skill strongly influences media legitimacy, see Sheafer 2001), both were combined to a single variable. This is one solution to a multicollinearity problem suggested by Lewis-Beck (1980, p. 61). This combined variable, named political communication legitimacy, was created by multiplying both variables.

FIGURE 1. Electrical Success as a Function of Political Communication Legitimacy

Tamir Sheafer

17

As can be seen in Figure 1, this variable also strongly influences the dependent variable (R2 ¼ .58).

DISCUSSION This is an exploratory study. The goal of the charismatic communication skill approach developed here is to fill a theoretical and methodological lacuna in political communication by analyzing the competition in the political and political communication arenas from the point of view of the political actors. This research theoretically conceptualizes the charismatic communication skill and empirically analyzes the influences of this variable on electoral politics, relying on specially developed methodology. The present research is part of a larger research project. Its first part discusses the role of charismatic communication skill in the political communication arena (Sheafer 2001; see also Sheafer and Wolfsfeld 2004). Its second part, presented here, analyzes more broadly the influence of the political communication arena, and especially charismatic communication skill, on the political arena and the electoral process. The implications for the democratic process are therefore more profound here. Charismatic communication skill is defined as the actor’s demonstrated skills and talents pertaining to the requirements of the political communication arena. Other actor-centered political science studies have analyzed the influence of the quality of actors on their electoral success, but they did not include media-related variables in their analysis, and hardly any other political skill. The methodological development presented in this study helps to demonstrate that charismatic communication skill is observable and measurable. The findings presented show that charismatic communication skill has a strong impact on long-term electoral success. Media legitimacy—the prominence and valence of media coverage—is analyzed here as an independent variable influencing electoral success. When analyzing the direct influence of the political communication arena on electoral politics, most political communication studies use media coverage as an explanatory variable. Yet, there is a main difference between this study and previous ones. It analyzes actual media coverage of political actors through content analysis of the media. Many other voter-centered studies rely instead on

18


surveys, in which respondents report about their media coverage to campaign communication (Iyengar 2001). Scholars argue that authority must be performed in the political arena. But what kind of performance is required? According to many scholars, these are substantive political performances that can be summarized as achieving ‘‘peace, prosperity and moderation’’ (Mazzoleni and Schulz 1999; Zaller 2001). When discussing MKs, the equivalent of this performance of political substance might be their parliamentary activity. But the findings presented here do not support the ‘‘politics of substance’’ hypothesis. MKs’ parliamentary activity has no positive effect on their electoral success. On the contrary, the findings show a negative (although not significant) correlation between the level of parliamentary activity and electoral success. This finding is disturbing, because it shows that doing the parliamentary work one was elected to do might lower the chances of being re-elected. One may argue that the reason for this finding is simply because the media ignore the parliamentary activities of MKs, and therefore the public cannot make electoral decisions based on this information. But the fact is that the voting public in most parties is a small number of activists who have quite good information about the parliamentary activities of their party representatives. Yet, they chose to ignore this information when making their electoral decisions. Although the voting public was not studied here, these findings suggest that voters rely heavily on the candidates’ charismatic communication skill to make political decisions. Since charismatic communication skill is tailored specifically to the needs of the political communication arena and the media organizations, it might suggest, as Cook (1996) has claimed, that the values of media organizations have strongly influenced the public’s political values. The bottom line of these claims is that political or electoral suitability is determined not only by political standards or political substance, but also by communication or media standards, as is represented by the importance of media legitimacy, charismatic communication skill, or their combination.

NOTES 1. This research group included all members of Knesset who were not government ministers, opposition leader, ex-Prime Minister, or Speaker of the House, and had a tenure of at least 6 months at the beginning of the research period.

Tamir Sheafer

19

2. Among the 38 MKs, 16 are from the Labor, 8 from the Likud, 7 from Meretz, 3 from Hadash, 3 from Mafdal, and 2 from Israel Be’aliya. 3. This variable went through a linear logarithmic transformation (Log Base 10). 4. In hierarchical regression analysis the independent variables are introduced in steps, or blocks (in SPSS), according to the theory. The focus is whether the increment of variance accounted for the second block is significant as is the total equation.

REFERENCES Aiken, L. S. and S. G. West. (1991). Multiple regression: Testing and interpreting interactions. Newbury Park: Sage. Altheide, D. L. and R. P. Snow. (1979). Media logic. Beverly Hills: Sage. Bar, A. (1994). Behirot Makdimot (Primaries) Veshitot Aherot [Primary elections and other election methods]. Jerusalem: The Israel Democracy Institute. Barber, J. D. (1992). The presidential character: Predicting performance in the White House, 4th Edition. Englewood Cliffs, NJ: Prentice-Hall, Inc. Bartels, L. M. (1988). Presidential primaries and the dynamics of public choice. Princeton: Princeton University Press. Bartels, L. M. (1993). Message received: The political impact of media coverage. American Political Science Review, 87(2), 267–285. Bass, B. M. (1988). Evolving perspectives on charismatic leadership. In J. A. Conger, and R. N. Kanungo. (Eds.), Charismatic leadership: The elusive factor in organizational effectiveness. San Francisco: Jossey-Bass Publishers. Bennett, W. L. (2003). News: The politics of illusion, 5th edition. New York: Longman. Blumler, J. G. and D. Kavanagh. (1999). The third age of political communication: Influences and features. Political Communication, 16(3), 209–230. Boiney, J. and D. L. Paletz. (1991). In search of the model model: Political science versus political advertising perspectives on voter decision making. In F. Bioccan (Ed.), Television and political advertising (Vol. 1). Hillsdale, NJ: Lawrence Erlbaum Associates, Publishers. Bryman, A. (1993). Charisma and leadership in organizations. London: Sage Publications. Cook, T. E. (1989). Making laws and making news: Media strategies in the U.S. House of Representatives. Washington, DC: The Brookings Institution. Cook, T. E. (1996). Afterword: Political values and production values. Political Communication, 13, 469–481. Eisenstadt, S. N. (1968). Introduction. In M. Weber (Ed.), On charisma and institution building. Chicago: The University of Chicago Press. Ericson, R. V., P. M. Baranek, and J. B. L. Chan. (1989). Negotiating control: A study of news sources. Toronto: University of Toronto Press. Finkel, S. E. (1993). Reexamining the ‘minimal effects’ model in recent presidential campaigns. The Journal of Politics, 55(1), 1–21. Gans, H. (1979). Deciding what’s news. New York: Vintage Books. Gardner, W. L. and B. J. Avolio. (1998). The charismatic relationship: A dramaturgical perspective. Academy of Management Review, 23(1), 32–58.

20


Gelman, A. and G. King. (1993). Why are American presidential election campaign polls so variable when voters are so predictable? British Journal of Political Science, 23, 409–51. Gitlin, T. (1980). The whole world is watching. Berkeley: University of California Press. Graber, D. A. (1993). Mass media and American politics, 4th edition. Washington, DC: CQ Press. Greenstein, F. I. (1992). Can personality and politics be studied systematically? Political Psychology, 13(1), 105–128. Greenstein, F. I. (1995). Political style and political leadership: The case of Bill Clinton. In S. A. Renshon (Ed.), The Clinton presidency: Campaigning, governing and psychology of leadership. Boulder: Westview Press. Hazan, R. Y. (1997). The 1996 intra-party elections in Israel: Adopting party primaries. Electoral Studies, 16(1), 95–103. Hilgartner, S. and C. L. Bosk. (1988). The rise and fall of social problems: A public arenas model. American Journal of Sociology, 94, 1988. House, R. J. and J. M. Howell. (1992). Personality and charismatic leadership. Leadership Quarterly, 3(2), 81–108. House, R. J., W. D. Spangler, and J. Woycke. (1991). Personality and charisma in the U.S. presidency: A psychological theory of leadership effectiveness. Administrative Science Quarterly, 36, 364–396. Iyengar, S. (2004). Engineering consent: The renaissance of mass communications research in politics. In J. T. Jost, M. R. Banaji, and D. Prentice (Eds.), Perspectives in social psychology: THe yin and yang of scientific progress (pp. 247–258). Washington, DC: American Psychological Association. Iyengar, S. and A. F. Simon. (2000). New perspectives and evidence on political communication and campaign effects. Annual Review of Psychology, 51, 149–169. Jacobson, G. C. (1992). The politics of congressional elections (3rd edition). New York: HarperCollins. Just, M. R., A. Crigler, D. Alger, K. Montague, D. West, and T. Cook. (1996). Crosstalk: Citizens, candidates, and the media in a presidential campaign. Chicago: The University of Chicago Press. Klapper, J. T. (1960). The effects of mass communication. New York: Free Press. Lewis-Beck, M. S. (1980). Applied regression: An introduction. Newbury Park: Sage Publications. Lombard, M., J. Snyder-Duch, and C. C. Bracken. (2002). Content analysis in mass communication: Assessment and reporting of intercoder reliability. Human Communication Research, 28(4), 587–604. Matheson, C. (1987). Weber and the classification of form of legitimacy. The British Journal of Sociology, 38(2), 199–215. Mazzoleni, G. and W. Schulz. (1999). ‘Mediatization’ of politics: A challenge for democracy? Political Communication, 16(3), 247–262. Meyrowitz, J. (1985). No sense of place: The impact of electronic media on social behavior. New York: Oxford University Press. Miller, W. E. and J. M. Shanks. (1996). The new American voter. Cambridge, MA: Harvard University Press.

Tamir Sheafer

21

Mondak, J. J. (1995). Competence, Integrity and the electoral success of congressional incumbents. The Journal of Politics, 57(4), 1043–1069. Mueller, C. (1973). The politics of communication. New York: Oxford University Press. Nimmo, D. D. and R. L. Savage. (1976). Candidates and their images: Concepts, methods, and findings. Pacific Palisades, CA: Goodyear. Renshon, S. A. (Ed.) (1995). The Clinton presidency: Campaigning, governing and psychology of leadership. Boulder: Westview Press. Renshon, S. A. (1996). The psychological assessment of presidential candidates. New York: New York University Press. Sellers, P. J. (2000). Manipulating the message in the U.S. Congress. Press=Politics, 5(1), 22–31. Shamir, B., R. J. House, and M. B. Arthur. (1993). The motivational effects of charismatic leadership: A self-concept based theory. Organization Science, 4(4), 577–594. Shamir, M. and A. Arian. (2002). Introduction. In A. Arian and M. Shamir (Eds.): The elections in Israel—2001 [Hebrew] (pp. 7–28). The Israel Democracy Institute. Sheafer, T. (2001). Charismatic Skill and Media Legitimacy: An Actor-Centered Approach to Understanding the Political Communication Competition. Communication Research, 28(6), 711–736. Sheafer, T. and G. Wolfsfeld. (2004). Production Assets, News Opportunities, and Publicity for Legislators: A Study of Israeli Knesset Members. Legislative Studies Quarterly, 29(4), 611–630. Shils, E. (1965). Charisma, order, status. American Sociological Review, 30, 199–213. Shoemaker, P. J. and S. D. Reese. (1991). Mediating the message: Theories of influences on mass media content. White Plains, NY: Longman Publishing Group. Simonton, D. K. (1988). Presidential style: Personality, biography, and performance. Journal of Personality and Social Psychology, 55(6), 928–936. Swanson, D. L. and P. Mancini. (Eds.) (1996). Politics, media and modern democracy. Westport, CT: Praeger. Wattenberg, M. P. (1995). The rise of candidate-centered politics. Cambridge: Harvard University Press. Weber, M. (1947=1986). Types of authority. In B. Kellerman (Ed.), Political leadership: A source book. University of Pittsburgh Press. Wolfsfeld, G. (1997). Media and political conflict: News from the Middle East. Cambridge: Cambridge University Press. Yatziv, G. (1998). Lexicon hevrati [Social lexicon]. Tel Aviv: Administration Library. Young, D. (1999). Model building in regression. New York: Addison-Wesley Longman. Zaller, J. (2001). Monica Lewinsky and the mainsprings of American politics. In W. L. Bennett and R. M. Entman (Eds.), Mediated politics: Communication in the future of democracy. New York: Cambridge University Press.

22


APPENDIX. MEASURING CHARISMATIC COMMUNICATION SKILL Collecting data from the 39 experts about the 91 MKs (the size of the larger research group) required a high number of responses from each expert (91 MKs X 6 categories of charismatic communication skill ¼ 546). To complete this task a unique methodology was developed. Six color-coded packages of cards represented the six categories, a different color for each category. Each package had cards with the names and pictures of all MKs. The researcher defined a category, handed a card-package to the expert, and asked her=him to rate each MK in that category by putting each card in the appropriate cell in a box, marked either high, medium, or low. This process was repeated for each category. The sorting was done only at the end of the session. The score of each MK in each category was calculated as a mean of the scores given to her=him by all 39 experts. This was done only after conducting two procedures. First, I discovered a high degree of agreement between the 39 experts, with the correlations calculated ranging from .79 to .89. I examined only a sample of the total correlations (39 judges X38 X6 categories). We first calculated the percent of agreement between the judges. Then we corrected it by the ordinal correlation coeffient between the disagreements among the same judges. That is, we gave ‘‘credit’’ for judges whose decisions are ‘‘close.’’ This is exactly what is suggested by Lombard et al. (2002 p. 590-91; see also note 3 in their article). Second, I treated raters as if they were scales to see if the raters are reliable as a group on any one item or on the aggregate scale for charismatic communication skill. This procedure resulted in a very high inter-raters reliability for all the six categories and for the aggregate scale (Cronbach coefficient alpha < .95 for all categories). The meaning of this finding is that all experts analyzed the skills of the MKs in a similar manner, disregarding the expert’s occupation, political beliefs, gender or age. Finally, the charismatic communication skill score of each MK was calculated as the mean of the MK’s scores in all six categories. This was done only after I found that the Pearson correlations between all the skill categories are extremely high, with r equals no less than .869. This shows that facing so many MKs, and having to

Tamir Sheafer

23

rate their skills in a short time (as mentioned above, each session lasted about 30 minutes), the experts saw each one of them as a whole. What they saw in their mind while comparing between the MKs was a single unified category that stands for the relevant skills, behaviors and charisma of each MK in the political communication arena. Indeed, I think that a similar thinking process happens to reporters, who have to select in a very short time only few political actors from the many who try to get media coverage. There are, however, two alternative explanations for the scores the experts gave the MKs. Both alternative explanations can be the outcome of biases arising from the apparent speed at which the judgments were made. The first is that the experts rated the overall political strength of the candidate, not his or her charismatic communication skill. This alternative explanation can be statistically tested by analyzing the correlations between the charismatic communication skills and the MK’s political position (representing his or her political strength) and political longevity. I conducted such correlations analyses and found them to be only moderate (correlation coefficients ranged from .38 to .49). Therefore, this alternative explanation can be rejected. Second, when asked about each charismatic skill category, each expert might actually have thought about the MK’s media exposure. This is a common threat to the validity of a research measuring skills or performances. One version of this is the common-method response bias, which ‘‘may inflate or suppress true relationships among variables hypothesized to be causally related. The common method may be a single instrument or questionnaire or document used to derive independent and dependent variables . . .’’ (House et al. 1991, pp. 371–2). To eliminate this threat, I used different methods to measure media legitimacy (content analysis) and charismatic communication skill (questionnaire). Another version is the implicit leadership theory, according to which ‘‘observers of leaders . . . have a theory about leader behavior. . . . If . . . they have some idea about a leader’s performance or effectiveness, they use their implicit leadership theory to infer the behavioral attributes that ‘logically’ led to the perceived performance’’ (House et al. 1991, p. 373). Therefore, if the experts knew how much media exposure each MK got, they might have rated the MK’s skills according to her=his level of media exposure. This threat was minimized here. According to Bryman

24


(1993, p. 11), ‘‘When questionnaire items relate to leaders’ specific behavior patterns, the distorting effects of implicit leadership theories are less pronounced.’’ The charismatic communication skill categories are exactly such specific behavior patterns. Furthermore, to accept this alternative explanation one has to accept the unreasonable assumption that all experts ignored their personal ongoing contacts with the MKs, and preferred to depend on what they saw appearing in the news. It makes little sense to assume, for example, that a reporter who in spite of receiving no cooperation at all from a specific MK will rate her=him high in the category of cooperation with journalists, because that MK received a lot of media exposure. It is also illogical to suggest that an assistant to an MK will ignore the MK’s ability to cooperate with other MKs and instead rate her=his media exposure.