Master’s Thesis
The Effect of Politicians’ Emotional Displays On Candidate Evaluations
Liselotte J. Pelle (10704868) Graduate School of Communication
June 25, 2015
Author note
This research is part of my Master’s Thesis at the Graduate School of Communication, where I follow the Master’s programme Communication Science. Dr. Bert N. Bakker is supervisor of this Master’s Thesis.
Abstract
In this study, I have examined the influence of a politician’s emotional display of enthusiasm, anger, and sadness on voters’ evaluation of the politician’s likeability, trustworthiness, and authenticity. Using a between-subjects experimental design, each participant watched a pre-existing video of a US politician, who displayed no emotion (control condition), enthusiasm, anger, or sadness. Subsequently, participants reported their own emotions and evaluations of the politician in the video. I find that participants who watched an enthusiastic politician, experienced enthusiasm themselves, and
evaluated the politician as more likeable and trustworthy than participants in the control condition. An angry politician evoked anger and was evaluated as less likeable and less trustworthy by participants compared to the control condition. Unfortunately, I did not find any significant results for the influence of sadness on candidate evaluations. Likewise, I did not find any significant effects for the evaluation of the politician’s authenticity. This study indicates that discrete emotions have separate effects on distinct candidate evaluations. Consequently, my research adds to the expanding literature on the role of emotions in politics and provides implications for the literature on emotional contagion and affect-as-information theory.
Acknowledgments
I want to give special thanks to the Methods Desk Communication Science for their help with the explanation of the mediation analyses and the subsequent interpretation. In addition, I want to express my gratitude to my supervisor, Dr. Bert N. Bakker, who tirelessly read the draft versions of my Master’s thesis and provided critical and constructive advice throughout the process. Finally, I want to show my appreciation to my friend, Sophie van den Noort, who offered invaluable suggestions and advice to improve the final product.
The Effect of Politicians’ Emotional Displays On Candidate Evaluations On July 17th 2014, flight MH17 crashed near the border of Ukraine and Russia, killing 298 people of which 193 were Dutch. During the United Nations Security Council meeting that followed this event, the Dutch Foreign Affairs Minister at the time, Frans Timmermans, gave an emotional speech. In his speech he put the collective feeling of loss that was felt in the Netherlands into words, while also expressing his anger about the injustice of innocent people going on a holiday being shot out of the sky. The speech was broadcasted on national television, shared on different social network sites, and has been viewed almost 1.2 million times on YouTube (The Daily Conversation, 2014). After Timmermans gave this speech, the Labor Party (Timmermans’ party) gained three seats according to the opinion polls at that time and left many with a positive attitude towards the current government (“Kiezer terug”, 2014). Minister Timmermans was even elected ‘Politician of the Year 2014’ by the parliamentary press and by the public (Van der Laan, 2014; Isitman, 2014). Given this case, the question arises whether politicians’ emotional displays evoke emotions in the people they represent and how these emotions affect the evaluations of politicians. This research will examine the effects of the display of discrete emotions on these political candidate evaluations. To this end, I will look further into the theoretical framework proposed by Glaser and Salovey (1998, p. 156), which led them to identify a need for more research to identify “psychological mechanisms by which a target’s displays of emotion influence judgments of that target”.
Previous research has found that both positively and negatively valenced
emotions have an influence on candidate evaluations (Abelson, Kinder, Peters, & Fiske, 1982; McHugo, Lanzetta, Sullivan, Masters, & Englis, 1985; Sullivan & Masters, 1988;
Lewis, 2000; Schaubroeck & Shao, 2012). Sullivan and Masters (1988) found that positive emotions lead to positive attitudes toward a politician. However, in the broader range of emotions, these are discrete and can have different effects on candidate
evaluations (Lerner & Keltner, 2000; Madera & Smith, 2009). Looking at sadness and anger, which are both negatively valenced emotions, Madera and Smith (2009) found that politicians who displayed sadness were evaluated more favorably than politicians who showed anger. This suggests that discrete emotions have different effects on candidate evaluations. Of these candidate evaluations, the most studied type is politicians’ likeability, while other studies have shown that traits such as trustworthiness and authenticity are also very important to have as a (political) leader (Kinder, Peters, Abelson, & Fiske, 1980; Parry-Giles, 2001; Louden & McCauliff, 2004). However, previous studies have not examined the influence of discrete emotions on the evaluations of these traits. The current research will build on the existing literature and add to it by showing politicians’ displays of discrete emotions to evoke discrete emotions from voters. Subsequently, this study will look at the influence of these emotions on
evaluations of likeability, trustworthiness, and authenticity. Taking this into account, the following research question was formulated:
RQ: How do distinct emotional displays by a politician influence voters’ evaluations of this politician?
Emotional Appeals By Politicians and Emotional Voters
When politicians express emotions, research has found that people observing these emotional displays experience similar emotions (McHugo et al., 1985; Sullivan &
Masters, 1988; Lewis, 2000; Schaubroeck & Shao, 2012). This transferal of emotions can be explained by the theory of emotional contagion, which indicates that people respond automatically, unintentionally and without their knowing to facial expressions of others by imitating this expression (Hatfield, Cacioppo, & Rapson, 1994). Through this process, people who respond to a happy facial expression will experience a happy feeling
themselves and people who respond to an angry facial expression will experience anger themselves (Sullivan & Masters, 1988; Ryan, 2012). Additionally, voters who empathize with a politician actually make an effort to internalize and experience the emotion that he or she expresses (Davis, 1996). McHugo et al. (1985) found that participants experienced similar emotions as the emotion expressed by a politician in a video clip. In this study, McHugo and his colleagues measured emotions through the observed contraction of the facial muscles of their respondents, their psychophysiological responses and their self-reported emotions. Likewise, Sullivan and Masters (1988) found that voters, who
described the emotional display of a political leader as happy, reported to be happier after being exposed to this display.
However, emotions do not only differ in their valence. If this were the case, then all negatively valenced emotions would have the same effect on outcomes. Madera and Smith (2009) found that politicians who display sadness are evaluated more positively than politicians displaying anger, which shows that the display of discrete emotions leads to different candidate evaluations (Lerner & Keltner, 2000). Next to the fact that research has found that (discrete) emotions have distinct effects, experiencing discrete emotions has different causes. Therefore, emotions are often examined in research as discrete (Lench, Flores, & Bench, 2011). Discrete emotions depend largely on the individual
experience and interpretation of situations, after which people react emotionally (Aarøe, 2011). Cognitive appraisal theory has provided a distinction between discrete emotions. According to this theory, anger is evoked when a threat is attributable to a certain source and when a person has control over the situation, whereas anxiety is evoked when a person is uncertain about the situation and does not have control (Lerner & Keltner, 2000; Tiedens & Linton, 2001). Sadness is evoked when a person experiences irreversible loss (Lazarus, 1991; Lazarus, 2006). Next to these two discrete negative emotions, anger and sadness, the current research will also look at a discrete positive emotion, enthusiasm. Enthusiasm is evoked when a person “is making reasonable progress toward the attainment of a goal” (Lazarus, 2006, p.16). This study expects that through emotional contagion, the display of discrete emotions will lead to the experience of discrete emotions.
Emotional Voters and Candidate Evaluations
In addition to the expectation that discrete emotions are evoked after watching the emotional display of a politician, the current research will examine the influence of these emotions on candidate evaluations. During election campaigns, much information is shared but most people forget this information pretty quickly. Their positive or negative affective response to this information, however, is stored in their so-called ‘on-line tally’ until they have to make a decision or give an opinion (e.g. Lodge, Steenbergen, & Brau, 1995; Kim & Garett, 2012). When people are later asked to make a judgment about a certain topic or person, they use their past feelings stored in the ‘on-line tally’ to make a decision. Cassino and Lodge (2007) found that people used their previously saved
affective response towards the politician as information to determine whether to evaluate him or her more positively or negatively. Similar results were found by Coleman and Wu (2010), who argue that when politicians display emotions, people use a similar affective manner to process these emotions as when they process their own emotions. These emotions are subsequently used to make political judgments.
The affect-as-information theory indicates that emotion is the most important information people use to make evaluations (Schwarz, 1990). The theory points out that emotion will be used when the evoked feelings are deemed relevant for the judgment (Schwarz, 1990). Emotions influence candidate evaluations (Abelson et al., 1982; Capelos, 2010; Cassino & Lodge, 2007; Ottati, Steenbergen, & Riggle, 1992; Wu & Coleman; 2014). For instance, Sullivan and Masters (1988) found that political leaders who evoked positive emotions in voters through their emotional display, also brought about more favorable attitudes towards them.
The previously discussed studies have focused primarily on the likeability of politicians. Next to being likeable, voters indicate that political candidates should be trustworthy (Kinder et al., 1980) and authentic (Parry-Giles, 2001; Louden & McCauliff, 2004). Trustworthiness is a characteristic of someone “whom people perceive as honest and sincere” (Priester & Petty, 2003, p. 408). Authenticity is defined as a characteristic of someone who behaves “in accord with their true self” (Kernis, 2003, p. 14). To perceive someone as authentic does therefore not only mean that someone displays an emotion, but that this emotion is in line with the self as the politician has presented him or herself to the public (Salmela, 2005). In this study, I will move beyond examining likeability of
politicians and also study the effect of discrete emotions on politicians’ trustworthiness and authenticity. In the next paragraph, I will present my testable hypotheses.
Emotional Politicians, Emotional Voters, and Candidate Evaluations
Building upon emotional contagion and affect-as-information theory, I theorize in the remainder of this paragraph how the displays of distinct emotions by a politician evoke distinct emotions among voters and accordingly influence their evaluations of the politician’s likeability, trustworthiness, and authenticity. To do this, I will use the model illustrated in Figure 1.
Figure 1. Theoretical Model Thesis
Enthusiasm is a mobilizing emotion that induces a feeling of intense enjoyment or approval (Huddy, Feldman, & Cassese, 2007). Feeling enthusiasm towards a political candidate leads to a positive evaluation of this politician (Marcus, Neuman, & MacKuen, 2000). Sullivan and Masters (1988) found that politicians who display positive emotions (such as enthusiasm and happiness) are evaluated more positively (Newcombe &
Ashkanasy, 2002). People who feel enthusiastic are more trusting (Dunn & Schweitzer, Emotional reaction voter Candidate evaluations Emotional display of politician Emotional contagion Affect-as-information
2005) and politicians who are positive towards their opponents are perceived as more trustworthy (Combs & Keller, 2010). Therefore, I expect that politicians who display enthusiasm will be evaluated more positively in terms of trustworthiness and likeability. Seeing as enthusiasm is an accepted emotion to display for a political leader, I expect political leaders who display enthusiasm to be evaluated more positively in terms of authenticity.
H1: Politicians who display enthusiasm will invoke enthusiasm among voters (a) and will be evaluated as more likeable (b), more trustworthy (c), and more authentic (d).
Similar to enthusiasm, anger is a mobilizing emotion (Valentino, Brader, Groenendyk, Gregorowicz, & Hutchings, 2011). However, anger induces a feeling of annoyance or displeasure. Displaying anger has been found to result in negative
evaluations of likeability (Schaubroeck & Shao, 2012). Moreover, angry people are less trusting and perceive others as less trustworthy (Dunn & Schweitzer, 2005). Lastly, it is accepted for politicians to show anger, as it is part of the set of behaviors of an authentic leader (Yagil & Medler-Liraz, 2014; Schaubroeck & Shao, 2012). Therefore, I expect that when people experience anger after seeing an angry politician they will evaluate this politician as less likeable, less trustworthy, but more authentic.
H2: Politicians who display anger will invoke anger among voters (a) and will be evaluated as less likeable (b), less trustworthy (c), but more authentic (d).
Sadness is an emotion that induces a feeling of sorrow and unhappiness. However, expressing sadness indicates sympathy, warmth, affiliation and concern (Madera & Smith, 2009). Madera and Smith (2009) found that politicians who displayed sadness were evaluated more favorably in terms of likeability than politicians who displayed anger. Looking at trustworthiness, research shows that sad people are more trusting (Dunn & Schweitzer, 2005). Glaser and Salovey (1998) suggest that a political leader who displays sadness “is more likely to be seen as caring and compassionate, but perhaps also weak” in terms of leadership (p. 168). Then again, when this sadness is displayed during a time of crisis and indicates empathy towards others, the expression of this emotion is congruent with the role of a leader, which in turn could lead to a positive evaluation of authenticity (Madera & Smith, 2009). Yagil and Medler-Liraz (2014, p. 66) found that leaders who deviate from rules of emotional expression “become more
accessible”, which could also lead to a positive evaluation of authenticity. Therefore, I expect politicians who display sadness to be evaluated as more likeable, more
trustworthy, and more authentic.
H3: Politicians who display sadness will invoke sadness among voters (a) and will be evaluated as more likeable (b), more trustworthy (c), and more authentic (d).
Method Design
This study used a between-subjects experimental design. The between-subjects factor was the type of emotional display by the politician, which was divided in four
categories: enthusiasm, anger, sadness, and neutral (control condition). Each participant was randomly assigned to one condition and viewed one YouTube video only.
Participants
The 116 participants that took part in my experiment were all recruited to
participate through social media, e-mail and personal conversation between May 5th and May 12th, 2015. Out of 223 participants, 119 completed the survey. This high dropout rate (53.4%) was caused by a malfunction in displaying the videos on a tablet or mobile phone, which caused participants to quit after having filled out part of the survey. Of the 119 participants that completed the survey, three participants did not agree with the informed-consent terms and did not finish the survey. Of the resulting 116 participants, 37 were males and 79 were females, with an age range between 18 and 66 (M = 34.02, SD = 14.52). The majority of the participants were educated at an undergraduate level (bachelor and hoger beroepsonderwijs). On a 0 – 10 scale, participants’ ideology was more liberal than conservative on average (M = 4.81, SD = 2.21). The participants were randomly assigned to one of the four different conditions (neutral n = 27, enthusiasm n = 25, anger n = 31, sadness n = 33).
Materials
Display of emotion by politician. This study used pre-existing videos of a US politician (Newt Gingrich) selected from YouTube to evoke the distinct emotions displayed by the politician. To ensure high external validity, real television material was selected. The choice to present my Dutch participants with an American politician was
instigated by Sullivan and Masters (1988), who showed that previous attitudes towards the politician influenced the evaluation of that politician. To prevent this from happening, the pre-test also examined previous knowledge of this politician and found that indeed, the participants did not know Newt Gingrich.
The video clips were selected based on the politician’s distinct emotional displays of enthusiasm, anger, and sadness (Ekman & Oster, 1979; Masters, Sullivan, Lanzetta, McHugo, & Englis, 1986; see Table 1 in appendix). Enthusiasm is displayed by smiling (raising corners of the mouth) and showing teeth (Ekman, Friesen, & Ancoli, 1980; Dimberg & Thunberg, 1998) and raising of the eyebrows (Du, Tau, & Martinez, 2014). These facial features could be accompanied by large hand gestures. Anger is displayed through frowning (lowering eyebrows) and lowering of the mouth corners (Ekman et al., 1980; Masters et al., 1986; Dimberg & Thunberg, 1998; Du et al., 2014). In addition, balling of the fist or pointing at someone could signal anger. Sadness is displayed through protruding lips and lowering of the eyebrows (Du et al., 2014), in addition to watery eyes. Moreover, hands could be used to wipe away tears or to cover the face. Since research shows that people react more emotionally without the interference of cognitive
information in the voice of a politician, the sound of the video clips was muted (McHugo et al., 1985).
The average length of the video clips was 36 seconds. Each YouTube video portrayed the same male politician, displaying four different emotional facial expressions (neutral, enthusiasm, anger, and sadness). For the control condition, the video depicted Newt Gingrich not displaying any facial expressions and maintained a non-emotional manner. The enthusiasm condition depicted the politician laughing, using large gestures,
nodding, and smiling. The anger condition depicted the politician looking stern, and using his hands to show his frustration. For sadness, the video portrayed the politician tearing up, and wiping a tear away (see Table 3 in appendix for links to the stimulus material).
The stimulus material was pre-tested using a sample of undergraduate students (N = 5). In this pre-test, I tested whether the videos depicted the intended emotions. The undergraduate students were asked to rate the intensity of the emotions displayed by the politician on a 7-point Likert scale, from not at all (1) to very much (7). The findings showed that the participants indicated the politicians displaying the intended emotions in the YouTube videos. In the enthusiasm condition, participants reported that the politician expressed more enthusiasm than the other conditions. Similarly, more anger was reported in the anger condition than in the other conditions and more sadness was reported in the sadness condition than in the other conditions (see Table 2 in appendix). Furthermore, the pre-test indicated that people did not find Newt Gingrich particularly attractive, or knew him, so they did not have any preconceived notions about him. Seeing as this research wanted to solely examine the influence of discrete emotional display and exclude any prior attitudes or knowledge about the politician, this outcome indicated we could use the stimulus material.
Dependent measures.
Experienced emotions after seeing an emotional display. Voters’ self-reported emotion was measured using 15 items. Participants were asked to rate how strongly they felt each emotion at that moment on a seven-point Likert scale ranging from not at all (1)
to very strongly (7) (see McHugo et al., 1985). Out of these 15 items, three dimensions were computed. The experienced enthusiasm was made up of the emotions happiness, enthusiasm, pleasure, pride, and surprise (α = .90, M = 2.30, SD = .05). The experienced anger was made up of the emotions anger, disgust, frustration, and irritation (α = .88, M = 2.84, SD = .20). The experienced sadness was made up of the emotions sadness, shame, guilt, and shyness (α = .81, M = 2.18, SD = .02).
To ensure that the three experienced emotions were independent in each condition, I looked at the correlations. Within the control condition, enthusiasm and sadness correlated (r = .46, n = 27, p = .016) and anger and sadness correlated (r = .40, n = 27, p = .040), which are both moderate associations. In the control condition,
enthusiasm and anger did not correlate (p = .624). In the enthusiasm condition, enthusiasm and anger correlated (r = -.45, n = 25, p = .024), and anger and sadness correlated (r = .73, n = 25, p = .000), which are moderate and strong associations
respectively. Enthusiasm and sadness did not correlate (r = -.11, n = 25, p = .595). In the anger condition, enthusiasm and anger correlated (r = .48, n = 31, p = .006), enthusiasm and sadness correlated (r = .41, n = 31, p = .024), and anger and sadness correlated (r = .45, n = 31, p = .012), which are all moderate associations. In the sadness condition, enthusiasm and anger (p = .256), and enthusiasm and sadness did not correlate (p = .279). Anger and sadness, however, did correlate (r = .60, n = 33, p = .000), which is a strong association. These findings suggest that instead of being unemotional as a baseline, people experience multiple discrete emotions at the same time. These are then influenced by the display of a politician. Looking at the enthusiasm condition, for example, people who experienced enthusiasm, experienced less anger, and people who experienced anger,
also experienced sadness. However, the current research is interested in comparing these emotions to the control condition, and whether these emotions differed significantly from the baseline.
Voters’ assessment of politician’s likeability. The likeability of the politician was measured using the ‘feeling thermometer’, which asked participants to indicate on a 0-100 scale how much they liked the politician (Sullivan & Masters, 1988).
Voters’ assessment of politician’s trustworthiness. The perceived
trustworthiness of the politician was measured by using the developed scale of Wheeless and Grotz (1977, see Table 4 in appendix). Participants were asked to rate their
agreement with the statements on a 7-point Likert scale, ranging from strongly disagree (1) to strongly agree (7). One of the 14 items read as follows: “On a scale from 1 to 7, how do you evaluate this politician? Trustworthy – untrustworthy”. The mean of the scale was 3.97 (SD = .99). Internal consistency reliability indicates a Cronbach’s alpha of 0.91, which is very high.
Voters’ assessment of politician’s authenticity. To measure the politician’s perceived authenticity, a selection was made from the questions proposed by Louden & McCauliff (2004, see Table 5 in appendix). Participants were asked to answer these questions on a 7-point Likert scale based on their impression of the politician, ranging from No, not at all (1) to Yes, absolutely (7). An example of 1 of the 10 items read as follows: “Does the politician show his true self?”. The mean of the scale was 3.96 (SD = .83). Internal consistency reliability indicates a Cronbach’s alpha of 0.74, which is quite high.
Manipulation check. To ensure that the emotional displays of the politician were correctly observed by the participants, the participants were asked to rate on a 7-point Likert scale, ranging from not at all (1) to very much (7), how much the political leader displayed each of the 15 distinct emotions provided.
Procedure
After reading the introductory text, following the ethical guidelines, they were asked to give their permission to participate in the experiment. Participants were instructed to focus on the politician’s face and asked to be in a peaceful surrounding to ensure that they would not be disturbed while watching the video. In addition,
participants were notified that no sound was included in the video. They were then randomly assigned to one of the four conditions. After watching each separate video clip, participants were asked to report how strongly they felt each distinct emotion.
Subsequently, they were asked to evaluate the politician’s likeability, trustworthiness, and authenticity. A manipulation check was included to ensure that the participants observed the intended emotion displayed by the politician. Lastly, participants filled out their demographic information, such as gender, age, education level, ideology, and political sophistication. Following ethical procedures of the Amsterdam School of Communication Research, all participants were shown the same video about a young polar bear. This video was selected, because it is completely unrelated to politics and showed an endearing animal.
Results Randomization
To ensure that the conditions were randomized, a Chi-Square test was conducted with the variables condition and age. The results showed that there was no difference between the condition groups in terms of age, X2 (111) = 117.39, p = .321. Similarly, no differences were found for gender (Fisher-exact p = .057), education (X2 (15) = 18.71, p = .227), and ideology (X2 (30) = 19.82, p = .921) between the conditions. Randomization has been successful, since all of the tests were not significant.
Subsequently, the manipulation check was examined. Manipulation was
successful, seeing as participants in the enthusiasm condition perceived the politician to be more enthusiastic than participants in the other conditions, participants in the anger condition perceived the politician to be more angry than participants in the other
conditions, and participants in the sadness condition perceived the politician to be more sad than participants in the other conditions (see Table 6 in appendix for results).
Main Analyses
To examine whether the condition had an effect on the participant’s perception of the politician as likeable, a two-way analysis of variance (ANOVA) was conducted with condition as independent variable and likeability as dependent variable. A main effect was found, F(3, 112) = 4.85, p = .003, η2 = .12. The effect size is weak, but does suggest that there is an effect of the display of the politician’s emotion on how much participants like the politician. Contrary to the expectation of hypothesis 1(b), the post-hoc test indicated that participants in the enthusiasm condition did not evaluate the politician
more positively compared to participants in the control condition (Mdifference = 4.25, p = .438). Secondly, participants in the anger condition liked the politician less than
participants in the control condition (Mdifference = -14.60, p = .006). Accordingly, I confirm hypothesis 2(b), which stated that seeing an angry politician leads to a decrease in the likeability of this politician (see Table 7 in appendix). Lastly, I did not find evidence for hypothesis 3(b) that participants in the sadness condition liked the politician significantly more than participants in the control condition (Mdifference = -4.97, p = .332).
To examine whether the display of a politician’s emotion had an effect on the participant’s perception of the politician as trustworthy, a two-way analysis of variance was conducted with condition as independent variable and trustworthiness as dependent variable. A main effect was found, F(3, 112) = 5.50, p = .001, η2 = .13. Similar to
likeability, the effect size is weak but the effect is apparent. In contrast to the expectation of hypothesis 1(c), the subsequent post-hoc tests indicated that participants in the
enthusiasm condition did not think that the politician was more trustworthy than
participants in the control condition (Mdifference = -.30, p = .257). Participants in the anger condition trusted the politician less than participants in the control condition (Mdifference = -.95, p = .000). Accordingly, this confirms hypothesis 2(c), which stated that seeing an angry politician leads to a decrease in the trustworthiness of this politician (see Table 8 in appendix). Finally, participants in the sadness condition did not evaluate the politician as more trustworthy than participants in the control condition (Mdifference = -.25, p = .306). These findings disconfirm hypothesis 3(c), because participants in the sadness condition did not think the politician was more trustworthy.
Finally, to examine whether the condition had an effect on the participant’s perception of the politician as authentic, an analysis of variance was conducted with condition as independent variable and authenticity as dependent variable. No main effect was found, F(3, 112) = .75, p = .524, η2 = .02. The lack of a main effect disconfirms hypotheses 1(d), 2(d), and 3(d), since the participants in the enthusiasm condition, anger condition, and sadness condition did not think the politician was more authentic than participants in the control condition.
Experienced emotions after seeing an emotional display. First, it was
examined whether the display of emotion by the politician significantly evoked emotions in the participants. Multiple ANOVAs were conducted with condition as independent variable and the three dimensions of emotions (enthusiasm, anger, and sadness) as dependent variables. A main effect was found of the condition on the evoking of
enthusiasm, F(3, 112) = 14.44, p = .000, η2 = .28, which is a moderate effect size. In line with hypothesis 1(a), participants in the enthusiasm condition experienced more
enthusiasm than participants in the control condition (Mdifference = 1.97, p = .000), the anger condition (Mdifference = 1.72, p =.000), and the sadness condition (Mdifference = 1.49, p = .000; see Table 9 in appendix). Turning to anger, a main effect with a medium effect size, of condition on the evoked anger was found, F(3, 112) = 18.67, p = .000, η2 = .33. In accordance with hypothesis 2(a), participants in the anger condition experienced more anger than participants in the control condition (Mdifference = 2.08, SD = 1.39), in the sadness condition (Mdifference = 2.09, p = .000), and in the enthusiasm condition (Mdifference = 2.26, p = .000; see Table 10 in appendix). However, no main effect of condition on the
evoking of sadness was found, F(3, 112) = 2.69, p = .050, η2 = .07. Disconfirming hypothesis 3(a), participants in the sadness condition did not experience more sadness than participants in the control condition (Mdifference = .60, p = .065), participants in the anger condition (Mdifference = -.06, p = .845), or participants in the enthusiasm condition (Mdifference = .66, p = .050; see Table 11 in appendix).
Seeing as participants in the sadness treatment did not experience more sadness than the other conditions, the following mediation analyses will solely be conducted for the enthusiasm and anger treatments.
Mediation analyses. In order to test for mediation, and calculate confidence intervals for the indirect effects, I used the SPSS macro PROCESS provided by Preacher and Hayes (2004). When this confidence interval does not include zero, it can be stated that mediation is present. To be able to examine each treatment separately as an
independent variable (the display of anger, sadness or enthusiasm), the variable condition had to be dummy coded. This resulted in three treatment conditions: the display of anger (compared to the control condition), the display of sadness (compared to the control condition), and the display of enthusiasm (compared to the control condition) by the politician. Voters’ assessment of the politician’s likeability was the dependent variable, and each self-reported emotion (the dimension of anger, sadness, or anger) was separately used in the analysis as a mediator, while controlling for the other two experimental
conditions.
To study this mediation model, I will first report the direct effect of the
path c). Second, I will report the coefficient that indicates how much the politician’s emotional display predicts the participants’ experienced emotion (also called path a). Subsequently, I will report the coefficient that shows how much the participants’ experienced emotion predicts the participants’ assessment of the politician (also called path b). Finally, I will report the indirect effect of the politician’s emotional display on the participants’ assessment of the politician, when it is mediated by one of the
experienced emotions (also called path c’).
I started with the indirect effect of politician’s display of enthusiasm on the participants’ assessment of the politician’s likeability, mediated by experienced emotion. The direct effect of display of enthusiasm on the politician’s likeability was not
significant (c = 4.25, p = .438). The display of the politician’s enthusiasm did significantly predict participants’ experienced enthusiasm (a = 1.97, p = .000). In
addition, the participants’ experienced enthusiasm significantly predicted the assessment of the politician as likeable (b = 7.43, p = .000). When experienced enthusiasm was added to the analysis as a mediator, the effect of the enthusiasm treatment on the
participants’ assessment of the politician as likeable was still not significant (c’ = -10.36, p = .069). The indirect effect of the enthusiasm treatment on the participants’ assessment of the politician as likeable when mediated by experienced enthusiasm was significant, 14.61 (95% CI [8.43, 24.22]). Since 0 is not part of this confidence interval, this indicates a significant mediation effect (p < .05). These findings confirm hypothesis 1, because experienced enthusiasm significantly influences the participants’ assessment of the politician as more likeable. Participants who saw an enthusiastic politician and
experienced enthusiasm rated him 14.61 points higher, on average, on the 100-point scale than participants in the control condition.
Subsequently, I tested the anger treatment as the independent variable, voters’ assessment of politician’s likeability as dependent variable, and self-reported anger as mediator. Participants in the anger condition evaluated the politician as less likeable compared to the control condition (c = -14.60, p = .006). Participants in the anger
condition experienced more anger compared to the control condition (a = 2.08, p = .000). In addition, the participants’ experienced anger predicted the assessment of the politician as less likeable compared to the control condition (b = -6.65, p = .000). When
experienced anger was added to the analysis as a mediator, the effect of the anger treatment on the participants’ assessment of the politician as likeable was not significant (c’ = -.77, p = .883). The indirect effect of the anger treatment on the participants’ assessment of the politician as likeable when mediated by experienced anger was significant, -13.83 (95% CI [-22.43, -7.47]). Since 0 is not part of this confidence interval, this indicates a significant mediation effect (p < .05). These findings confirm hypothesis 2, because experienced anger significantly influences the participants’
assessment of the politician as less likeable. Participants who saw an angry politician and experienced anger rated him 13.83 points less, on average, on the 100-point scale than participants in the control condition.
To measure the indirect effect of politician’s display of emotion on the participants’ assessment of the politician’s trustworthiness, mediated by experienced emotion, I started with the display of enthusiasm. The direct effect of display of
addition, the participants’ experienced enthusiasm significantly predicted the assessment of the politician as trustworthy (b = .25, p = .001). When experienced enthusiasm was added to the analysis as a mediator, the effect of the enthusiasm treatment on the participants’ assessment of the politician as trustworthy was significant (c’ = -.78, p = .008). The indirect effect of the enthusiasm treatment on the participants’ assessment of the politician as trustworthy when mediated by experienced enthusiasm was significant, .48 (95% CI [.24, .83]). Since 0 is not part of this confidence interval, this indicates a significant mediation effect (p < .05). These findings confirm hypothesis 1, because experienced enthusiasm significantly influences the participants’ assessment of the politician as more trustworthy. On average, participants who saw an enthusiastic politician and experienced enthusiasm rated the politician .48 points higher on a 1-7 Likert-scale than participants in the control condition.
Subsequently, I tested the anger treatment as the independent variable, voters’ assessment of politician’s trustworthiness as dependent variable, and self-reported anger as mediator. Participants in the anger condition evaluated the politician as less
trustworthy compared to the control condition (c = -.95, p = .000). In addition, the participants’ experienced anger significantly predicted the assessment of the politician as less trustworthy compared to the control condition (b = -.20, p = .002). When
experienced anger was added to the analysis as a mediator, the effect of the anger treatment on the participants’ assessment of the politician as trustworthy was not significant (c’ = -.53, p = .053). The indirect effect of the anger treatment on the
anger was significant, -.42 (95% CI [-.79, -.15]). Since 0 is not part of this confidence interval, this indicates a significant mediation effect (p < .05).
These findings confirm hypothesis 2, because experienced anger significantly influences the participants’ assessment of the politician as less trustworthy. On average, participants who saw an angry politician and experienced anger rated the politician .42 points less on a 1-7 Likert-scale than participants in the control condition.
Turning to the evaluation of the politician’s authenticity, no significant findings were found for display of enthusiasm or anger.
Discussion
Summarizing the major findings of this study, I have found, first of all, that the display of the distinct emotions enthusiasm and anger by a politician evokes similar distinct self-reported emotions in the voters. Unfortunately, this was not the case for the distinct emotion sadness, which can be explained by two reasons. One reason could be that the operationalization of the sadness condition prevented the successful transmission of sadness to the voters. The other reason could be that sadness is not a ‘contagious’ emotion. From the findings in this research, I have to conclude that some discrete
emotions (enthusiasm and anger) are contagious and can be picked up by voters, whereas sadness is not. These findings suggest that emotions should indeed be studied as discrete and can be elicited by the facial expressions of politicians. Examining the correlations between the experienced emotions, the associations suggest that people are not
unemotional, but experience multiple discrete emotions at the same time. The display of emotion by a politician not only evokes the displayed discrete emotion, but can also
influence the other experienced emotions. More research is necessary to examine the process through which this occurs.
Secondly, these evoked distinct emotions significantly influence the subsequent evaluations of the politician. In line with the literature, participants who observed an enthusiastic politician experienced enthusiasm themselves and subsequently evaluated the politicians as more likeable and trustworthy than participants in the control condition (Sullivan & Masters, 1988). In contrast, participants who observed an angry politician experienced anger and this influenced their evaluations of the political candidate as less likeable and less trustworthy than participants in the control condition rated him
(Schaubroeck & Shao, 2012). This suggests that people indeed use their emotions as information to make (political) judgments.
Unfortunately, the display of sadness did not lead to any significant effects on self-reported sadness by the participants or any significant effects on the subsequent candidate evaluations. Perhaps this was caused due to the fact that the sadness
manipulation was not strong enough. Or it is possible that emotions need to be felt very strongly to be able to influence subsequent evaluations of the candidate. Enthusiasm and anger are mobilizing emotions (Huddy et al, 2007; Valentino et al., 2011), which could possibly evoke stronger distinct emotions than sadness can. An additional issue might have been that people need a relationship with, or at least know, a person before they can feel empathy for them or feel sadness (Davis, 1996). Therefore, a suggestion for future research is to study a politician who is well known and compare him or her with an unknown politician.
The evaluation of authenticity did not result in any significant findings either. Similar to the evoking of sadness, perhaps participants need a relationship or at least a little history with a politician to be able to judge whether he or she is authentic or not. Especially questions such as ‘Is public and private character authentic?’ might need a little more context than solely the emotional display of a politician. Even though an emotion can also be evaluated as ‘authentic’, even there a “wider context of the person’s identity” is needed (Salmela, 2005, p. 217). The difference between the evaluations of likeability and trustworthiness, and the evaluation of authenticity, is that liking and trusting a person can be an emotional judgment, whereas authenticity is more of a cognitive judgment. This suggests that some evaluations are influenced by experienced emotions, whereas other candidate evaluations are not.
What is interesting to think about is whether the emotional response to a politician displaying his emotion will lastingly change the evaluation of the politician. As
previously discussed, information that is shared during election campaigns, for example, is quickly forgotten. However, the way it made people feel is not and stored in people’s on-line tally (Lodge et al., 1995; Kim & Garret, 2012). As Schwarz (1990) explains with his affect-as-information theory, when these feelings are relevant for a question about their evaluation of the politician, people will use these emotions to make a judgment. In that case, the question is whether the emotions that are used to make a judgment are the emotions people felt most recently or with the most intensity. Future research could examine this by conducting a longitudinal study that investigates candidate evaluations over time when participants are exposed to different emotional displays.
Future research should ask a politician (or multiple politicians) who is better known to participate in their experiment to measure authenticity. In this way, it can be investigated whether previous attitudes toward a politician influence the experienced emotion and subsequent evaluations. Other discrete (negative) emotions should be
examined as well, seeing as sadness did not influence candidate evaluations. It is possible that only certain discrete emotions can evoke similar feelings in the public and play a role in candidate evaluations. For example, anxiety indicates that a person is uncertain about a situation and lacks control (Lerner & Keltner, 2000; Tiedens & Linton, 2001). When a politician displays anxiety, this could lead to negative evaluations of the politician’s likeability, trustworthiness, and authenticity, because a politician is supposed to have control. Similarly, hope includes a “level of uncertainty about the candidate and how he or she will perform in the future” (Just, Crigler, & Belt, 2007, p. 236). For hope to blossom, however, a longer relationship with or a certain to a politician is necessary. Therefore, I expect the display of hope to not have any effects on candidate evaluations unless previous attitudes exist.
Furthermore, gender could be an interesting factor to include. Perhaps emotional female politicians do not evoke the same responses as male politicians, because there are different gender stereotypes that are also at work in politics (see Matsumoto, Hee Yo, & Fontaine, 2008 for more about display rules). It is more accepted for women to display sadness, for example, whereas it is more accepted for men to show anger (Schaubroeck & Shao, 2012). Therefore, I would expect women who display sadness to be evaluated more likeable, more trustworthy, and more authentic than men displaying sadness. In contrast, I
would expect men who display anger to be evaluated as more likeable, more trustworthy, and more authentic than women displaying anger.
All in all, this study adds important conclusions to the expanding literature on the role of emotion in politics. Politicians who display distinct emotions evoke similar distinct emotions. These distinct emotions have specific effects on the traits (likeability, trustworthiness, and authenticity) studied in the current research. Considering anger for example, this emotion could be a sign of strong leadership, but this study shows it results in negative likeability and trustworthiness evaluations. It appears that Frans Timmermans has been fortunate in his display of sadness and anger, and leaves a need for more
research on this subject. Considering the findings of this study, it is clear that a politician should think twice before expressing his or her emotion.
References
Aarøe, L. (2011). Investigating Frame Strength: The Case of Episodic and Thematic Frames. Political Communication, 28, 207–226.
Abelson, R. P., Kinder, D. R., Peters, M. D., & Fiske, S. T. (1982). Affective and semantic components in political person perception. Journal of Personality and Social Psychology, 42(4), 619–630. doi:10.1037/0022-3514.42.4.619
Capelos, T. (2010). Feeling the Issue: How Citizens' Affective Reactions and Leadership Perceptions Shape Policy Evaluations. Journal of Political Marketing, 9, p. 9–33. Cassino, D., & Lodge, M. (2007). The Primacy of Affect in Political Evaluations. In
Neuman, W. R., Marcus, G. E., Crigler, A. N., & MacKuen, M. (Eds.), The Affect Effect: Dynamics of Emotion in Political Thinking and Behavior. Chicago:
University of Chicago Press.
Coleman, R., & Wu, H. D. (2010). Proposing Emotion as a Dimension of Affective Agenda Setting: Separating Affect into Two Components and Comparing Their Second-Level Effects. Journalism & Mass Communication Quarterly, 87, 315–327. doi:10.1177/107769901008700206
Combs, D. J. Y., & Keller, P. S. (2010). Politicians and Trustworthiness: Acting Contrary to Self-Interest Enhances Trustworthiness. Basic and Applied Social Psychology, 32(October 2014), 328–339. doi:10.1080/01973533.2010.519246
Davis, M. H. (1996). Empathy: A Social Psychological Approach. Westview Press: Boulder, CO.
Dimberg, U., & Thunberg, M. (1998). Rapid facial reactions to emotional facial expressions. Scandinavian Journal of Psychology, 39, 39–45.
Du, S., Tao, Y., & Martinez, A. M. (2014). Compound facial expressions of emotion. Proceedings of the National Academy of Sciences of the United States of America, 111(15), E1454–E1462.
Dunn, J. R., & Schweitzer, M. E. (2005). Feeling and believing: the influence of emotion on trust. Journal of Personality and Social Psychology, 88(5), 736–748.
doi:10.1037/0022-3514.88.5.736
Ekman, P., & Oster, H. (1979). Facial Expressions of Emotion. Annual Review of Psychology, 30, 527–554.
Ekman, P., Friesen, W. V., & Ancoli, S. (1980). Facial Signs of Emotional Experience. Journal of Personality and Social Psychology, 39(6), 1125–1134.
Glaser, J., & Salovey, P. (1998). Affect in electoral politics. Personality and Social Psychology Review : An Official Journal of the Society for Personality and Social Psychology, Inc. doi:10.1207/s15327957pspr0203_1
Hatfield, E., Cacioppo, J., & Rapson, R. (1994). Emotional Contagion. Cambridge University Press: New York.
Huddy, L., Feldman, S., & Cassese, E. (2007). On the Distinct Political Effects of
Anxiety and Anger. In Neuman, W.R., Marcus, G.E., Crigler, A.N., & MacKuen, M. (Eds.), The Affect Effect: Dynamics of Emotion in Political Thinking and Behavior. Chicago: University of Chicago Press.
Isitman, E. (2014, December 15). En weer is de titel 'beste politicus van het jaar' voor Timmermans. Elsevier. Retrieved from:
http://www.elsevier.nl/Politiek/achtergrond/2014/12/En-weer-is-de-titel-beste-politicus-van-het-jaar-voor-Timmermans-1666516W/
Just, M. R., Crigler, A. N., & Belt, T. L. (2007). Don't Give Up Hope: Emotions,
Candidate Appraisals, and Votes. In Neuman, W.R., Marcus, G.E., Crigler, A.N., & MacKuen, M. (Eds.), The Affect Effect: Dynamics of Emotion in Political Thinking and Behavior. Chicago: University of Chicago Press.
Kernis, M. H. (2003). Toward a Conceptualization of Optimal Self-Esteem. Psychological Inquiry, 14(1), 1–26. doi:10.1207/S15327965PLI1401_01 Kiezer terug bij VVD en PvdA na vliegramp MH17 (2014, August 3). Het Parool.
Retrieved from:
http://www.parool.nl/parool/nl/3587/POLITIEK- BINNENLAND/article/detail/3710166/2014/08/03/Kiezer-terug-bij-VVD-en-PvdA-na-vliegramp-MH17.dhtml
Kim, Y. M., & Garrett, K. (2012). On-line and Memory-based: Revisiting the
Relationship Between Candidate Evaluation Processing Models. Political Behavior, 34, 345–368. doi:10.1007/s11109-011-9158-9
Kinder, D. R., Peters, M. D., Abelson, R. P., & Fiske, S. T. (1980). Presidential prototypes. Political Behavior, 2(4), 315–337.
Lazarus, R. S. (1991). Emotion and Adaptation. New York, NY: Oxford University Press.
Lazarus, R. S. (2006). Emotions and Interpersonal Relationships: Toward a Person-Centered Conceptualization of Emotions and Coping. Journal of Personality, 74(1), 9–46.
Lench, H. C., Flores, S. A., & Bench, S. W. (2011). Discrete Emotions Predict Changes in Cognition, Judgment, Experience, Behavior, and Physiology: A Meta-Analysis of Experimental Emotion Elicitations. Psychological Bulletin, 137(5), 834–855. Lerner, J. S., & Keltner, D. (2000). Beyond valence: Toward a model of emotion-specific
influences on judgement and choice. Cognition & Emotion, 14, 473–493.
Lewis, K. M. (2000). When leaders display emotion: How followers respond to negative emotional expression of male and female leaders. Journal of Organizational
Behavior, 21, 221–234. doi:10.1002/(SICI)1099-1379(200003)21:2<221::AID-JOB36>3.3.CO;2-S
Lodge, M., Steenbergen, M. R., & Brau, S. (1995). The Responsive Voter: Campaign Information and the Dynamics of Candidate Evaluation. American Political Science Review, 89(2), 309–326. doi:10.2307/2082427
Louden, A., & McCauliff, K. (2004). “The ‘authentic candidate’: Extending candidate image assessment,” In Hacker, K. (Ed.), Presidential candidate images. Oxford: Rowman & Littlefield, pp. 85–103.
Madera, J. M., & Smith, D. B. (2009). The effects of leader negative emotions on evaluations of leadership in a crisis situation: The role of anger and sadness. The Leadership Quarterly, 20(2), p. 103–114.
Marcus, G. E., Neuman, W. R., & MacKuen, M. (2000). Affective Intelligence and Political Judgment. Chicago: University of Chicago Press.
Masters, R. D., Sullivan, D. G., Lanzetta, J. T., McHugo, G. J., & Englis, B. G. (1986). The facial displays of leaders: Toward an ethology of human politics. Journal of Social and Biological Systems, 9(4), 319–343.
Matsumoto, D., Hee Yo, S., & Fontaine, J. (2008). Mapping Expressive Differences Around the World: The Relationship Between Emotional Display Rules and Individualism Versus Collectivism. Journal of Cross-Cultural Psychology, 39(1), 55–74.
McHugo, G. J., Lanzetta, J. T., Sullivan, D. G., Masters, R. D., & Englis, B. G. (1985). Emotional reactions to a political leader’s expressive displays. Journal of
Personality and Social Psychology, 49(6), 1513–1529. doi:10.1037/0022-3514.49.6.1513
Newcombe, M. J., & Ashkanasy, N. M. (2002). The role of affect and affective
congruence in perceptions of leaders: An experimental study. Leadership Quarterly, 13, 601–614. doi:10.1016/S1048-9843(02)00146-7
Ottati, V. C., Steenbergen, M. R., & Riggle, E. (1992). The cognitive and affective components of political attitudes: Measuring the determinants of candidate evaluations. Political Behavior, 14(4), 423–442. doi:10.1007/BF00992043
Parry-Giles, S. J. (2001). Political Authenticity, Television News, and Hillary Rodham Clinton. In Hart, R. P., & Sparrow, B. H. (Eds.) Politics, Discourse, and American Society: New Agendas, 12: 211–244. Oxford: Rowman & Littlefield Publishers. Preacher, K. J., & Hayes, A. F. (2004). SPSS and SAS procedures for estimating indirect
effects in simple mediation models. Behavior Research Methods, Instruments, & Computers, 36(4), 717–731.
Priester, J. R., & Petty, R. E. (2003). The Influence of Spokesperson Trustworthiness on Message Elaboration, Attitude Strength, and Advertising Effectiveness. Journal of Consumer Psychology, 13(4), 408–421. doi:10.1207/S15327663JCP1304_08 Ryan, T. (2012). What Makes Us Click? Demonstrating Incentives for Angry Discourse
with Digital-Field Experiments. The Journal of Politics, 74(4), 1138–1152. Salmela, M. (2005). What is emotional authenticity? Journal for the Theory of Social
Behaviour, 35(3), 209–230.
Schaubroeck, J. M., & Shao, P. (2012). The role of attribution in how followers respond to the emotional expression of male and female leaders. The Leadership Quarterly, 23, 27–42.
Schwarz, N. (1990). Feelings as information: informational and motivational functions of affective states. In Sorrentino, R., & Higgins, E.T. (Eds.) Handbook of
Motivationand Cognition: Foundations of Social Behavior, 2: 527–61. New York: Guilford.
Sullivan, D. G., & Masters, R. D. (1988). “Happy Warriors”: Leaders’ Facial Displays , Viewers' Emotions , and Political Support. American Journal of Political Science, 32(2), 345–368. doi:10.2307/2111127
The Daily Conversation (2014, July 14). Emotional MH17 Crash Speech By Netherlands Delegate to UN [Video file]. Retrieved from:
https://www.youtube.com/watch?v=QcGCBHNcKyI on June 11th, 2015. Tiedens, L. Z., & Linton, S. (2001). Judgment Under Emotional Certainty and
Uncertainty: The Effects of Specific Emotions on Information Processing. The Journal of Personality and Social Psychology, 81(6), 973–988.
Valentino, N. A., Brader, T., Groenendyk, E. W., Gregorowicz, K., & Hutchings, V. L. (2011). Election Night’s Alright for Fighting: The Role of Emotions in Political Participation. The Journal of Politics, 73(1), 156–170.
doi:10.1017/S0022381610000939
Van der Laan, S. (2014, December 11). Parlementaire pers: Timmermans politicus van het jaar. Elsevier. Retrieved from:
http://www.elsevier.nl/Politiek/achtergrond/2014/12/Parlementaire-pers-kiest-Timmermans-tot-politicus-van-het-jaar-1664556W/
Wheeless, L. R., & Grotz, J. (1977). The Measurement of Trust and Its Relationship to Self-Disclosure. Human Communication Research, 3(3), 250–257.
Wu, H. D., & Coleman, R. (2014). The Affective Effect on Political Judgment: Comparing the Influences of Candidate Attributes and Issue Congruence. Journalism & Mass Communication Quarterly, 1077699014538825–. doi:10.1177/1077699014538825
Yagil, D., & Medler-Liraz, H. (2014). Feel Free, Be Yourself: Authentic Leadership, Emotional Expression, and Employee Authenticity. Journal of Leadership & Organizational Studies, 21(1), 59–70.
Appendix Table 1. The criteria for selecting the YouTube videos
Criteria for Classifying Facial Displays
Facial Features Enthusiasm/happiness Anger Sadness
Eyelids Wide, normal or
slightly closed
Opened wide Relaxed upper eyelid
Eyebrows Raised Lowered Inner corner raised
Eye orientation Focused then cut off Staring Down
Mouth corners Retracted and/or raised
Forward or lowered
Lowering of the lower lip
Teeth showing Upper or both Lower None
Head motion
Lateral Side-to-side None None
Vertical Up-down None Pushing up of the
chin Head orientation
To body Normal to trunk Forward
from trunk
Down
Angle to vertical Up Down Down
Source: Masters et al., 1986
Table 2. Results of pre-test YouTube videos
Perceived emotion Condition M SD
Anger Control 1.60 0.89 Anger 5.60 0.55 Sadness 1.20 0.45 Enthusiasm 1.00 0.00 Happiness Control 2.00 1.73 Anger 1.00 0.00 Sadness 1.60 0.89 Enthusiasm 4.60 1.95 Disgust Control 1.80 0.84 Anger 3.20 1.79 Sadness 2.20 1.64 Enthusiasm 1.20 0.45 Sadness Control 1.20 0.45 Anger 1.60 0.89 Sadness 4.40 1.67 Enthusiasm 1.00 0.00 Fear Control 1.40 0.55 Anger 1.40 0.89 Sadness 2.20 1.79 Enthusiasm 1.00 0.00
Surprise Control 1.80 0.84 Anger 1.00 0.00 Sadness 1.00 0.00 Enthusiasm 2.60 1.52 Enthusiasm Control 2.40 1.14 Anger 1.20 0.45 Sadness 1.40 0.55 Enthusiasm 6.20 0.84 Pleasure Control 2.00 1.41 Anger 1.00 0.00 Sadness 1.20 0.45 Enthusiasm 5.00 1.58 Shame Control 1.20 0.45 Anger 1.40 0.89 Sadness 1.40 0.89 Enthusiasm 1.20 0.45 Guilt Control 1.40 0.55 Anger 1.40 0.89 Sadness 2.80 1.79 Enthusiasm 1.00 0.00 Shyness Control 1.40 0.89 Anger 1.00 0.00 Sadness 1.40 0.55 Enthusiasm 1.40 0.89 Frustration Control 1.40 0.55 Anger 4.40 1.52 Sadness 2.20 1.64 Enthusiasm 1.00 0.00 Hope Control 2.20 1.30 Anger 1.00 0.00 Sadness 1.40 0.55 Enthusiasm 2.40 1.67
Table 3. Links to the YouTube videos
Condition Link
Neutral https://www.youtube.com/watch?v=aEwYE4C1VCw
Anger https://www.youtube.com/watch?v=NNEonBFtz_I
Sadness https://www.youtube.com/watch?v=ipmrZs_WO7w
Table 4. Individualized Trust Scales (ITS)
On a scale from 1 to 7, how do you evaluate the politician? Trustworthy – untrustworthy
Trustful of this person – distrustful of this person Confidential – divulging
Benevolent – exploitive Safe – dangerous Candid – deceptive Not deceitful – deceitful Straightforward- tricky Respectful – disrespectful Considerate – inconsiderate Honest – dishonest Reliable – unreliable Sincere – insincere Careful – careless
Source: Wheeles and Grotz (1977)
Table 5. Measuring authenticity
Authenticity judgments
Personal Qualities and Traits Are they who they say they are?
Do the candidates know who they are? Do the candidates reveal their true selves? Is the candidate consistent with his or her core self?
Is the candidate comfortable in his or her own skin?
Are candidates an enactment of themselves? Competence
Can they be counted on?
Do they believe in what they say they believe? Is the candidate consistent with his or her central beliefs?
Do the candidates do what they say? Communicative behavior
Is their interaction authentic?
Is public and private character authentic? Is the candidate too political?
Table 6. Manipulation check – Which emotion does the politician display?
Emotion displayed
by politician Fisher-exact p Condition Not shown Shown
Anger 62.38 .000 Control 26 1 Anger 7 24 Sadness 32 1 Enthusiasm 24 1 Happiness 40.53 .000 Control 27 0 Anger 31 0 Sadness 23 10 Enthusiasm 10 15 Disgust 33.28 .000 Control 22 5 Anger 10 21 Sadness 28 5 Enthusiasm 24 1 Sadness 24.54 .000 Control 25 2 Anger 30 1 Sadness 19 14 Enthusiasm 25 0 Fear 9.16 .020 Control 27 0 Anger 23 8 Sadness 30 3 Enthusiasm 22 3 Surprise 3.63 .303 Control 26 1 Anger 29 2 Sadness 32 1 Enthusiasm 21 4 Enthusiasm 43.95 .000 Control 21 6 Anger 28 3 Sadness 26 7 Enthusiasm 3 22 Pleasure 42.41 .000 Control 26 1 Anger 31 0 Sadness 31 2 Enthusiasm 9 16 Shame 6.00 .107 Control 25 2 Anger 24 7 Sadness 24 9 Enthusiasm 23 2 Guilt 11.42 .008 Control 24 3 Anger 24 7 Sadness 23 10 Enthusiasm 25 0
Shyness 6.84 .073 Control 20 7 Anger 28 3 Sadness 22 11 Enthusiasm 22 3 Frustration 30.97 .000 Control 19 8 Anger 7 24 Sadness 20 13 Enthusiasm 23 2 Hope 19.60 .000 Control 13 14 Anger 30 1 Sadness 26 7 Enthusiasm 17 8 Pride 22.22 .000 Control 17 10 Anger 24 7 Sadness 28 5 Enthusiasm 7 18 Irritation 29.60 .000 Control 19 8 Anger 8 23 Sadness 24 9 Enthusiasm 23 2
Table 7. Assessment of politician’s likeability
Condition M SD
Control 45.19 18.78
Anger 30.58 18.06
Sadness 40.21 21.02
Enthusiasm 49.44 20.71
Table 8. Assessment of politician’s trustworthiness
Condition M SD
Control 4.35 1.13
Anger 3.41 .72
Sadness 4.10 1.04
Enthusiasm 4.06 .82
Table 9. Experienced enthusiasm
Condition M SD
Control 1.67 .86
Anger 1.92 .88
Sadness 2.15 1.08
Table 10. Experienced anger Condition M SD Control 2.32 1.39 Anger 4.40 1.45 Sadness 2.32 1.24 Enthusiasm 2.14 1.38
Table 11. Experienced sadness
Condition M SD
Control 1.84 1.04
Anger 2.51 1.28
Sadness 2.45 1.41
