Printer Friendly

The relevance of emotions in presidential public appeals: anger's conditional effect on perceived risk and support for military interventions.

Research on political discourse and communication has shown that emotioninducing appeals are selectively deployed by political leaders in attempting to boost their approval ratings, garner support for their policy goals, and/or divert the public's attention during times of domestic turmoil (De Castella, McGarty, and Musgrove 2009). Within the U.S. context, presidents often resort to emotive appeals in wielding the bully pulpit with the expectation that it may help rally public support for their political and policy agendas, including the use of military force. (1) For example, in his remarks on March 18, 2011, President Barack Obama sought empathy from Americans as he conveyed key events unfolding in Libya, particularly in describing how civilians were suffering at the hands of Muammar Qadhafi's oppressive, violent regime (Obama 2011):
   Over the last several weeks, the world has watched events unfold in
   Libya with hope and alarm. Last month, protesters took to the
   streets across the country to demand their universal rights and a
   government that is accountable to them and responsive to their
   aspirations. But they were met with an iron fist ... Instead of
   respecting the rights of his own people, Qadhafi chose the path of
   brutal suppression. Innocent civilians were beaten, imprisoned, and
   in some cases killed. Peaceful protests were forcefully put down.
   Hospitals were attacked and patients disappeared. A campaign of
   intimidation and repression began.


President Obama then noted further threats of violence facing the people of Libya and offered a justification for possible U.S. intervention (Obama 2011):
   And just yesterday, speaking of the city of Benghazi, a city of
   roughly 700,000 people, he [Qadhafi] threatened, and I quote: "We
   will have no mercy and no pity"--no mercy on his own citizens ...
   Now, here's why this matters to us. Left unchecked, we have every
   reason to believe that Qadhafi would commit atrocities against his
   people. Many thousands could die. A humanitarian crisis would
   ensue.


To what extent might these types of emotion-inducing presidential appeals influence the public mood and lend credence to a president's policy objectives, both with respect to a particular policy option (here, military intervention in civil conflict) as well as to the president's broader policy agenda (i.e., attempting to take advantage of such emotion-driven public rallies to seek out other foreign and domestic policy objectives)? Presumably, such presidential appeals could lead to emotional arousal and in turn increase public support for specific policies. However, past works are split on whether and how emotive stimuli matter and it is not known if emotional arousal may lead to a spillover effect that influences public support for other policies connected to a president's broader policy agenda. Furthermore, prior studies on emotions have generally focused on military interventions where a direct national security threat exists, failing to consider other instances where there is no direct or immediate threat to the United States (e.g., military action taken for humanitarian purposes). Further still, one must consider the extent to which such potential opinion shifts might actually help facilitate an intervention, particularly with regard to a president's strategic positioning at home and abroad (see Edwards 2009; 2012).

In this study, we employ an experimental design situated in the context of a best test "least likely" case (2) to investigate whether the thematic relevance of emotive stimuli (i.e., when emotional responses are appraised as being policy relevant versus policy irrelevant) embedded in presidential speeches influences people's risk perceptions and support regarding military interventions in civil conflict. For instance, does anger tend to lower people's perceptions of risk and increase support for an aggressive foreign policy option regardless of its source? Or do the thematic underpinnings of anger matter? That is, do the specific contents of a presidential speech that trigger a target emotion--such as an emotion-inducing story embedded in a presidential speech concerning a civil conflict abroad--affect public opinion? We argue that the effectiveness of presidential public appeals in garnering support for a given policy partly depends on the thematic relevance of the emotion-eliciting cues presidents employ.

Thematic Relevance of Emotions

The systematic examination of the link between emotions and politics is a growing area of research. (3) Scholars have examined emotions across a wide range of topics such as political awareness and attentiveness (e.g., Marcus, Neuman, and MacKuen 2000); political participation and mobilization (Valentino et al. 2011); political information processing and decision making (e.g., Redlawsk 2002); risk perceptions (e.g., Lerner et al. 2003); political attitudes (e.g., Brady and Sniderman 1985; Banks and Valentino 2012); reactions to political issues and events (e.g., Conover and Feldman 1986); policy preferences (e.g., Huddy et al. 2005); and conflict behavior, peace negotiations, and conflict resolution (e.g., Halperin et al. 2011; Sabucedo et al. 2011). These studies demonstrate that emotions have a significant influence over cognitive and behavioral processes and outcomes, including memory elicitation, evaluation, political judgment, and political action (see Marcus 2000; Brader 2006).

Initial conceptualizations of emotion portray it as falling along a single bipolar dimension that runs from pleasant to unpleasant (valence) and gains intensity (arousal) as one shifts from the center toward the poles (Brader 2006, 52; see, e.g., Zajonc 1998; Ferguson and Wee 2000). However, some scholars prefer a two-dimensional valence model that conceptualizes emotions as positive and negative (Huddy, Feldman, and Cassese 2007, 204; see, e.g., Tellegan, Watson, and Clark 1999; Watson and Clark 1992). According to the latter model, positive emotions are associated with the approach system motivating one to achieve positive outcomes for pleasure and reward whereas negative emotions are linked to the avoidance system activated to elude negative outcomes in order to protect against pain and harm (e.g., Cacioppo, Gardner, and Berntson 1999; Watson et al. 1999).

In comparison to valence-based approaches, emotion-specific approaches propose that different emotions sharing the same valence (such as anger and fear) may nevertheless have dissimilar (or even opposite) effects on decision-making processes and outcomes (see, e.g., Bodenhausen, Sheppard, and Kramer 1994; DeSteno et al. 2000; Huddy, Feldman, and Cassese 2007; Lerner and Keltner 2000). To elaborate, although fear and anger are both negative emotions, fear is linked to appraisals of uncertainty and lack of control whereas anger is associated with appraisals of certainty about the source of a threat and feelings of personal control over the situation (Lazarus 1991; Huddy et al. 2005; Lerner and Keltner 2000; 2001; Smith and Ellsworth 1985). Such distinct appraisal tendencies associated with different emotions are also connected to variations in information acquisition patterns and cognitive processing, as well as assessments of risk. Specifically, several studies find that anxiety and fear are likely to raise one's level of cognitive effort, vigilance, and perceived risks whereas higher levels of anger tend to trigger more superficial information searches, heuristic-based cognitive processing, and lower risk assessments (Bodenhausen, Sheppard, and Kramer 1994; Lerner and Keltner 2000; 2001; Lerner et al. 2003; Marcus, Neuman, and MacKuen 2000; Mackuen et al. 2010; Valentino et al. 2008). Regarding policy preferences, studies show that anxiety and fear elicit a preference for more precautionary and defensive policy actions while anger increases people's support for retaliatory and aggressive policy responses (Huddy, Feldman, and Cassese 2007; Sadler et al. 2005; Skitka et al. 2006). Anger is also more closely associated with a reluctance to consider alternatives, an unwillingness to engage in diplomacy and negotiation, and the rejection of compromise in dealing with political conflicts (Sabucedo et al. 2011; Mackuen et al. 2010). (4)

If anger is likely to increase support for aggressive policies and decrease risks associated with such policy options, a president who plans to resort to the use of force to deal with a conflict situation may strategically seek to invoke feelings of anger among citizens to garner public approval. Going back to our research question, does it matter whether presidents choose to invoke anger using emotive cues directly related to the targeted policy in their public appeals or can anger evoked even about unrelated policy issues be effective and, if so, what would be the implications for their policy agendas? To answer this question, we consider two alternative theoretical perspectives: (1) the independence of emotions versus (2) the thematic relevance of emotions.

Applying the former perspective, one may argue that the effects of emotions are independent from their sources such that a particular emotional state (such as being angry) will have a uniform effect on one's decision-making irrespective of the thematic content of the emotive trigger as its source (e.g., Gasper and Clore 1998; Goldberg, Lerner, and Tetlock 1999; Schwarz and Clore 1983; 1996). According to this view, appraisal tendencies generated by specific emotions can persist and spill over to influence one's political judgment on a given issue even when the target of judgment is unrelated to the emotion-eliciting stimulus (Gasper and Clore 1998; Goldberg, Lerner, and Tetlock 1999; Johnson and Tversky 1983).

Alternatively, it is plausible to argue that the effects of a certain emotional state on one's decision making is conditional on whether the specific source of that emotional state is thematically related to the policy issue at hand. Schwarz and Clore (1983; 1996) suggest that although people have a tendency to attribute their affective states to the current object of attention (even when the actual source of their feelings is completely unrelated to that object), any misattributions typically disappear when people become aware of the true source of their affective states (see also Keltner, Locke, and Aurain 1993; Schwarz, Servay, and Kumpf 1985; Siemer and Reisenzein 1998). Furthermore, Rosenboim et al. (2012) point out that intense emotions are likely to be particularly resistant to misattribution since their sources tend to be highly salient. Accordingly, the extent to which an emotional state influences subsequent political judgments and assessments of risk may be contingent on whether the individual perceives an external stimulus as connected to and/ or responsible for such emotional state (Hirt et al. 1997; Keltner, Locke, and Aurian 1993; Pham 1998). Within such context, an individual may consider a given emotion to be an irrelevant source of information if the decision-making domain under evaluation is unrelated to the emotional state.

Of these two contending theoretical perspectives, we expect the thematic relevance of emotions (rather than the independence of emotions) is more likely the primary underlying mechanism regarding the effects emotions have on public reactions to presidential messages. Specifically, we posit that because an emotive trigger is embedded in a given cognitive context, the relevance of that context to a policy issue under consideration is likely to accentuate (or diminish) the impact of the induced emotional state on an individual's risk perceptions and policy preferences (see Sirin and Geva 2013). Our approach aligns well with appraisal theories from psychology, which emphasize that a range of cognitive dimensions defines the patterns of appraisal underlying different emotions, thereby differentiating one's emotional experiences and the effects of such emotions (Folkman et al. 1986; Lazarus 1984; 1991; 1994; Lerner and Keltner 2000; 2001; Smith and Ellsworth 1985). Appraisal theories conceptualize that the brain draws on contextual information in determining which emotions to arouse and how to cope with them (Folkman and Lazarus 1988; Smith and Ellsworth 1985). In line with this perspective, we argue that one's coping mechanism for dealing with anger depends on the appraisals attached to it--specifically, whether individuals appraise their feelings of anger as policy relevant or not.

In the context of military interventions, if the source of a certain emotional state is also thematically relevant to the issue (such as getting angry after exposure to an emotion-inducing story embedded in a presidential speech addressing a civil conflict abroad), we believe the relevance of such source is likely to amplify the impact of the elicited emotion on people's risk perceptions and policy preferences. By comparison, if the source of a certain emotional state is not thematically relevant to the issue (such as getting angry after exposure to a presidential speech addressing an incident involving crime in U.S. cities), the disconnect between the source of the emotional state (crime in U.S. cities) and the policy issue to be considered (military intervention in civil conflict) may subdue the impact of the elicited emotion on one's policy preferences.

With all these considerations in mind, we hypothesize as follows:
   Emotion-inducing presidential appeals for taking military action
   that include thematically relevant triggers of anger are likely to
   prompt (H1) lower risk perceptions and (H2) higher policy support
   among individuals (i.e., as compared to presidential appeals that
   include thematically irrelevant triggers of anger or appeals that
   do not include any emotive triggers).


Methodology

To test our hypotheses, we conducted a study that involved a pretest and two experiments with a total of 1,187 undergraduate students. Of these participants, 233 participated in the pretest, 456 took part in Experiment 1 and 498 in Experiment 2. (5) Experiment 2 serves as a robustness check of our initial findings by incorporating some key alterations to the research design as well as a number of additional posttest measures for gathering further evidence and ruling out competing explanations.

Experiment 1

Experiment 1 involves a between-groups factorial design consisting of three conditions: (1) a thematically relevant emotive trigger condition, (2) a thematically irrelevant emotive trigger condition, and (3) an emotion-neutral control condition. We randomly assigned the participants to the experimental conditions.

For the manipulation of experimental factors, we designed hypothetical presidential speeches. (6) Specifically, participants read a short presidential speech addressing two separate issues: (1) crime in U.S. cities and (2) the possibility of a military intervention in a civil conflict abroad. (7) In the thematically relevant emotive trigger condition, an emotion-inducing story was embedded in the speech as part of the civil conflict issue. In contrast, in the thematically irrelevant emotive trigger condition, the same emotion-inducing story was embedded in the speech as part of the issue concerning crime in U.S. cities. No emotion-inducing story was provided in the emotion-neutral control condition.

Below is the text of the emotion-neutral speech (with placement of the emotion-inducing story for the thematically relevant and irrelevant conditions noted in bold):
   In a recent press conference, President Obama addressed two key
   policy issues: Violent crime in U.S. cities is a major issue. We
   must consider the need for increasing law enforcement in our major
   cities like Charlotte given the violent crime that has recently
   arisen. [Emotion-inducing story embedded here in the thematically
   irrelevant condition]. My administration has proposed a number of
   key initiatives to help put an end to such crime. We should
   consider reforming our criminal justice system to restore safety in
   U.S. cities.

   Violence abroad is another major issue. We must consider the
   possibility of a military intervention in the country of Malawi
   given the violent civil conflict that has recently erupted there.
   [Emotion-inducing story embedded here in the thematically-relevant
   condition]. My administration has proposed a number of key
   initiatives to help put an end to such violence. We should consider
   taking military action to restore peace in Malawi.


The embedded emotion-inducing story read as follows:
   Just a couple of days ago, a family visiting a well-known area was
   brutally slaughtered in broad daylight by a gang of criminals.
   According to witnesses, law enforcement officials did not respond
   in time to their calls for help. One of the murdered family members
   was only an 8-month old baby still clutched in her dead mother's
   arms. The murdered father was a young doctor who had recently
   volunteered overseas for the charitable organization, Doctors
   Without Borders, and the family had been celebrating his time off.
   Who will seek justice for them? Unfortunately, we have been
   witnessing a long string of these violent and escalating acts.


To maximize internal validity and control, all of the wording used in the scenarios across the different experimental conditions was kept constant except for the induction and placement of the story. After their exposure to the presidential speech, participants rated on a seven-point scale their risk perceptions and support concerning a possible military intervention in said civil conflict.

As previously mentioned, our experimental design is situated in the context of a best test "least-likely" case to test our theoretical expectations in the most conservative manner. As Levy (2002, 442-43) suggests, "the strongest support for a theory comes when a case is least likely for a particular theory and most likely for the rival theory, and when observations are consistent with the predictions of the theory but not those of its competitor." Research shows that people are relatively uninformed about foreign policy issues and may often feel detached from the foreign policy domain (Holsti 1996; Jentleson 1992; Page and Shapiro 1992). The U.S. public is particularly known to be wary of and at times oblivious to humanitarian interventions in civil conflict where there is no direct threat to national security and/or direct benefits to U.S. interests (Jentleson and Britton 1998; Russett and Nincic 1976). In contrast, because domestic policy issues generally have a more direct effect on people's lives than foreign policy ones (barring a national security threat), people tend to view domestic issues as more salient and familiar (see Villalobos and Sirin 2012). One might thus expect that an emotion-inducing presidential public appeal for taking military action in a civil conflict abroad could be largely ineffective even when it is thematically relevant. By comparison, having high policy salience and representing a direct threat to one's well-being, an emotional arousal regarding the issue of crime in U.S. cities could not only be effective in garnering policy support, but also spill over to subsequent policy judgments about other issues regardless of the thematic relevance. If this is the case, the experimental observations would negate the tenets of the "thematic relevance of emotions" approach. Otherwise, if an emotion-inducing appeal is nevertheless effective in generating support for the thematically relevant (and not so salient) policy proposal of humanitarian action in a civil conflict while emotions triggered by the thematically irrelevant (but highly salient) domestic issue do not spill over to affect said foreign policy judgments, such finding would constitute strong evidence supporting the validity and scope of our theoretical expectations.

To ensure that our experimental design is internally valid, we conducted several manipulation checks. (8) Specifically, we asked the pretest participants how angry they felt about the information they were exposed to. The response options ranged from 1 (not at all angry) to 5 {very angry) on a five-point scale. For our analyses, we subsequently rescaled all of the variables in our models to range from 0 to 1 for ease of comparison and interpretation. The results of our analysis of variance (ANOVA) with a Bonferroni adjustment for post hoc multiple comparisons of group means demonstrate that participants in both the thematically relevant (TR) and thematically irrelevant (TI) emotive trigger conditions expressed significantly higher levels of anger compared to the emotion-neutral control (C) condition ([M.sub.TR] = .67; [M.sub.TI] = .72; [M.sub.C] = .35; p < .001). This indicates that the emotive trigger embedded in the presidential speech was indeed effective in inducing anger. Furthermore, there was no statistically significant difference regarding the level of anger induced between the thematically relevant and irrelevant emotive trigger conditions (p = .61).

In the pretest, we also asked the participants whether they experienced several other emotions--specifically fear, worry, and anxiety--in reaction to the information they read. This is because individuals can potentially experience several emotions at any given moment and one emotion may nullify or mediate the impact of another (Izard et al. 2000). However, we find that the reported levels of fear, worry, and anxiety in the experimental conditions are not significantly different from the emotion-neutral control condition (p > .10). Such results show that anger is the dominant emotion induced by the experimental stimuli, thereby confirming the internal validity of our experiment.

Experiment 1 Results

To test our hypotheses, we conducted ordered logistic regression analyses (given our ordinal-level dependent variables of perceived risk and policy support). We included gender, party identification, and ideological preference as control variables in our models. (9) As sensitivity analyses, we also conducted analysis of covariance (ANCOVA), followed by Bonferroni for multiple group comparisons. The results remain robust to this alternative method of analysis. (10)

For our first hypothesis, our expectation was that presidential emotive appeals with thematically relevant triggers would prompt lower risk perceptions for taking military action in a civil conflict abroad (as compared to presidential appeals that include thematically irrelevant triggers of anger or appeals that do not include any emotive triggers). However, the results (presented in column 1 of Table 1) demonstrate that there is no statistically significant difference between our experimental groups. Therein, estimated marginal means (using ANCOVA with Bonferroni) indicate that risk perceptions are similarly high across all conditions ([M.sub.TR] = .73; [M.sub.TI] = .73; [M.sub.C] = .70). Such results suggest that, particularly in the current post-9/11 context vis-a-vis the costly U.S. military interventions in Iraq and Afghanistan, members of the public are well aware and unlikely to change their perceptions of the risks involved with taking military action.

At first blush, these results may seem contradictory to prior works that found anger tends to lower risk perceptions. For instance, Huddy, Feldman, and Cassese (2007) demonstrate that anger directed at terrorists and Saddam Hussein was strongly associated with lower risk assessments concerning the Iraq War. However, such findings are about military action in the presence of perceived direct threats to national and personal security. In our case, we put the effects of anger to the test in the case of a humanitarian intervention for a civil conflict abroad where there is no direct threat and/or benefit involved. As such, this finding contributes to the literature by indicating that anger's role in diminishing risk perceptions may be limited to foreign policies designed to serve national security interests. Otherwise, risk perceptions are likely to remain high for humanitarian policies despite heightened feelings of anger. But are such high risk perceptions enough to curtail people from supporting military interventions in civil conflict when stimulated by anger?

For our second hypothesis, we proposed that emotive presidential appeals for taking military action that include thematically relevant triggers of anger are likely to prompt higher support among individuals (as compared to presidential appeals that include thematically irrelevant triggers of anger or appeals that do not include any emotive triggers). In line with H2, the results (presented in column 2 of Table 1) show that participants who were exposed to thematically relevant emotive triggers expressed a significantly higher tendency to support military intervention in said civil conflict compared to those in the emotion-neutral control group (p < .05). As for the substantive effects, the probability of moderately to completely supporting military action increases by about 20% upon one's exposure to the thematically relevant emotive trigger. On the other hand, the effect of thematically irrelevant emotive triggers is not statistically different from the control condition (p > .05). Estimated marginal means (using ANCOVA with Bonferroni) further demonstrate that support for military action among participants exposed to the thematically relevant emotive trigger is significantly higher than those in the other two conditions ([M.sub.TR] = .52; [M.sub.TI] = .44; [M.sub.C] = .39).

Experiment 2

In order to check the robustness of our findings for Experiment 1, gather further evidence, and rule out some competing explanations, we conducted a second experiment that incorporates some key alterations to the research design as well as a number of additional posttest measures. A total of 498 respondents took part in Experiment 2. We randomly assigned the participants to four conditions: thematically relevant emotive trigger condition, thematically irrelevant emotive trigger condition, thematically relevant emotion-neutral control condition, and thematically irrelevant emotion-neutral control condition.

In Experiment 1, respondents read both policy passages (i.e., a domestic policy passage about high crime in U.S. cities and a foreign policy passage about a violent conflict abroad) in the presidential speech while one of the passages (in the thematically relevant and irrelevant conditions) included an emotion-inducing story. This design makes the passage with the emotion-inducing story lengthier, which may convey greater importance and increase its policy salience. One may further suggest that this design also makes the emotion-inducing story more vivid, which may lead to some confounding effects. Indeed, scholars suggest that vivid depictions of a scenario (in particular, violent imagery) make that scenario more accessible in a viewer's mind compared to a scenario that does not contain such vivid imagery (see Gross 2008, 173). These issues could cause respondents to change their foreign policy attitudes not because they are angry, but because the salience of the policy issue at hand and the vividness of the passage make it more accessible and persuasive. For Experiment 2, to ensure that the experimental results reflect the effects of thematic relevance and not issue salience or vividness, we not only separated the domestic and foreign policy passages by showing each respondent only one passage with/ without the emotion-inducing story but also revised the presidential speech so that the emotion-neutral control conditions contained exactly the same number of words as the conditions that contained the emotion-inducing story. In redesigning the experimental conditions, we took special care to make them of equal length while keeping the content as identical as possible. (11)

In addition to these alterations to the experimental design, we also expanded the posttest questionnaire and introduced some changes to our analytical models for hypothesis testing. Besides the controls for age, gender, party identification, ideology, and income that we used in the first experiment, Experiment 2 also includes several other key control measures. To begin with, we controlled for how strongly participants identified with those who live in high-crime cities as well as residents of countries experiencing violent conflict. This is because, as Intergroup Emotions Theory suggests, identifying with a group that is threatened likely determines whether a threat provokes anger and, in turn, action. For instance, Van Zomeren, Spears, and Leach (2008) find that a strong group identity increases collective action tendencies via group-based anger. Group identification may thus confound the experimental effects if participants perceived the foreign policy passage as an issue concerning an "other" group while perceiving the domestic policy passage as an issue that directly concerns "us." We further controlled for participants' race/ethnicity (non-white versus white). While it is unlikely that many of the participants would identify with the victims of the violent civil conflict abroad depicted in the presidential speech, residents of high-crime areas in the United States tend to be from historically disadvantaged minority groups (e.g., see Sampson, Raudenbush, and Earls 1997). Minority participants might thus identify more strongly with the victims in the emotion-inducing story and react differently to the experimental treatments. With these possibilities in mind, adding these controls to the analyses allows us to more effectively tease out the experimental effects.

The posttest questionnaire for Experiment 2 also included measures to control for the target of anger. Whether someone is getting angry about the policy issue the president describes or is getting angry at the president himself is an important difference. We partially addressed this in the initial experiment by controlling for party identification and ideology. In Experiment 2, we directly asked the participants, "Setting aside the policy issue you just read about, how angry/anxious/enthusiastic did the President make you feel?" Adding these specific measures to the questionnaire to control for the target of anger proved to be very helpful in checking the robustness of our experimental findings.

In Experiment 2, we expanded our manipulation check of participants' emotional reactions to the experimental conditions by adopting key emotional prompts from Marcus, Neuman, and MacKuen's (2000) study to better capture the multidimensionality of emotions (see also MacKuen et al. 2010; Marcus et al. 2006). Developing the theory of Affective Intelligence, Marcus, Neuman, and MacKuen (2000) suggest that emotional preconscious appraisals systems shape one's consciousness, motivational states, as well as political attitudes and behavior. According to this view, there are three concurrent appraisals: enthusiasm (a reaction to familiar and rewarding contexts), anxiety (a reaction to novel, uncertain, and threatening circumstances), and aversion (a reaction to recurrent punishing or nonrewarding intrusions). Therein, enthusiasm is associated with the pursuit of positive goals and habitual choices whereas anxiety prompts increased attentiveness to information and leads to focused, deliberative engagement with the task at hand. By comparison, aversion (of which anger constitutes an intrinsic part) marks strong normative disapproval and seeks to neutralize disliked elements (see Marcus et al. 2006; MacKuen et al. 2010). In the pretest of Experiment 1, we asked participants whether they experienced anger, fear, worry, and anxiety in reaction to the presidential speech, yet did not include a full scale of possible emotional responses (including other key markers of aversion besides anger and the markers for the latent factor of enthusiasm), which would allow us to generate index measures along the primary dimensions of emotions. In Experiment 2, we included measures for all key markers that the theory of Affective Intelligence recommends.

Our factor analysis confirms the presence of three distinct emotional responses aroused: aversion, anxiety, and enthusiasm. Following Marcus, Neuman, and MacKuen's (2000) methodological approach, our principal components analysis with varimax rotation yielded a three-factor model with eigenvalues of 4.77, 2.04, and 1.29. Therein, the items "angry," "frustrated," disgusted," "bitter," and "resentful" clearly define the aversion dimension. "Anxious," "worried," and "afraid" fit well into the anxiety dimension, while "proud," "hopeful," and "enthusiastic" load well on the enthusiasm dimension. We then used these items to generate three separate scales of emotional response. Cronbach's alpha values are .92 for aversion, .83 for anxiety, and .76 for enthusiasm, indicating that these scales are reliable. (12)

We next conducted ANOVA for pairwise comparisons of the levels of aversion, anxiety, and enthusiasm between experimental conditions with Bonferroni correction to conservatively estimate the differences in means. (13) The results demonstrate that there are no statistically significant differences in either anxiety or enthusiasm across experimental conditions (p > .05). Regarding the level of aversion aroused in reaction to the presidential speech, the thematically relevant and thematically irrelevant emotive trigger conditions significantly differ from both thematically relevant and irrelevant control conditions (p < .05), whereas the control conditions do not differ from one another (p > .05). (14) The manipulations in Experiment 2 thus worked as intended in triggering aversion, thereby confirming the internal validity of the experiment.

Experiment 2 Results

To analyze the treatment effects in Experiment 2, we employ ordered logistic regression. (15) As we predicted, similar to the results of Experiment 1, only the thematically relevant emotive trigger condition in Experiment 2 has a positive and statistically significant effect on support for military intervention in civil conflict (p < .05). Regarding the substantive effects, exposure to the thematically relevant emotive trigger increases the probability of moderately to completely supporting military action by about 27%. By comparison, neither the thematically irrelevant emotive trigger condition nor the thematically relevant emotion-neutral control condition significantly differs from the thematically irrelevant emotion-neutral control condition. Once again, risk perceptions are essentially similar across experimental conditions. (16)

While the results concerning policy support thus far seem quite consistent with our theoretical expectations, we must still demonstrate it is anger itself, and not some other factor such as vividness or issue salience, that explains such differences across experimental conditions. To examine the presumed causal mechanism, we conducted a path analysis by employing generalized structural equation models (Baron and Kenny 1986; Hayes and Preacher 2010). (17) Using ordinary least squares regression, we first estimated the effect of the thematically relevant emotion-inducing presidential speech on our mediator, aversion (path a). We then estimated the effect of aversion on support for military intervention while controlling for thematic relevance (path b) and estimated the direct effect of the thematically relevant emotion-inducing presidential speech on the outcome variable while controlling for aversion (path C), using ordered logistic regression. (18) Upon obtaining path coefficients, we tested the significance of the indirect effects using a bootstrap procedure that yields bias-corrected confidence intervals. The results indicate that aversion indeed has a significant mediating effect regarding the link between the thematically relevant emotion-inducing presidential speech and support for military interventions in civil conflict abroad (p < .05), which further corroborates our hypothesis on policy support. (19)

As our final set of analyses, we also tested the effect of thematically relevant versus irrelevant emotive triggers in the domestic policy domain (i.e., the presidential speech addressing the issue of high crime in U.S. cities). The results demonstrate that only the thematically relevant emotive trigger condition affected people's domestic policy preferences. Specifically, only those who were exposed to the presidential speech on high crime in U.S. cities that contained the emotion-inducing story were more likely to support increasing law enforcement in the country. The results further indicate that participants in the thematically relevant emotive trigger condition perceived a significantly lower risk of increasing law enforcement in U.S. cities than those in the other experimental conditions. (20) Therefore, our theoretical mechanism concerning the thematic relevance of emotive triggers in presidential public appeals also applies to the domestic policy domain, here with regards not only to policy support but also risk perceptions.

These findings suggest that via strategic use of thematically relevant emotive appeals, the president may potentially acquire some flexibility in the domestic arena to act with the public's consent on major crime-related issues, which may include proposing specific policy changes nationwide or taking direct action such as deploying the National Guard to deal with domestic strife. (21) While previous studies point to a relatively higher public deference given to presidents in dealing with foreign policy issues (Peake 2001; Peterson 1994), this additional finding indicates that emotion-inducing, thematically relevant appeals may also help presidents garner public leeway in the domestic arena--both by increasing policy support and decreasing risk perceptions. This represents an important new insight that merits further inquiry for exploring its full potential.

Discussion

Taken together, the results of our experiments provide a number of important insights. First, it appears that members of the public--at least in the current context--are highly sensitive to the risks of military interventions abroad and are unlikely to be swayed by emotion-inducing appeals to set their misgivings of such risks aside. Second, contrary to the notion that emotion-inducing appeals can result in a spillover effect wherein the public lends credence to a president's political agenda across policy domains (i.e., the independence of emotions), we find that people are able to compartmentalize their emotions within the policy sphere that is thematically relevant. This demonstrates a certain level of sophistication among the populace. Nevertheless, despite such nuanced capabilities, members of the public remain quite susceptible to emotion-inducing appeals in terms of allowing their triggered anger to heighten their support for a military intervention even while remaining well aware of the high risks involved. Such findings indicate that emotive stimuli can indeed potentially serve as an influential rhetorical tool when it comes to public appeals for military action. The question that remains, however, is: can increased support actually help facilitate a president's foreign policy agenda?

Such opinion movement likely matters most in cases where presidents stand in a strategically advantageous position to take military action and can thus use the increase in support--if only symbolically--to help justify and facilitate their decisions to do so. As Edwards (2009; 2012) points out, the strategic positioning of the president greatly matters and often determines at the outset whether a president even has the opportunity to affect the output of government, leaving the president's persuasive skills over the public and key political actors as more of a secondary, supplemental--and often quite limited--tool for moving forward the policy agenda. In other words, rather than employing public appeals with the specific intent to create opportunities for their preferred policies, presidents are better off using their appeals strategically--mainly by exploiting existing opportunities--to help facilitate policy changes where possible within the context of the American political system and the specific political environment they find themselves in (Edwards 2003; 2009). It is especially in cases where public opinion already lies in the president's favor and is malleable for generating further support that presidents can take advantage of such circumstances to help justify and facilitate their goals (Edwards 2015; see also Canes-Wrone 2001; 2004). Moreover, when presidents are able to make public appeals amid fewer countervailing agents or opposing messages (e.g., such as during a foreign policy crisis where the narrative provides more of an us versus them rather than left versus right mentality), such conditions may help bolster a president's ability to garner additional public support (Rottinghaus 2009; 2010; see also Zaller 1994). This is particularly so in the foreign policy realm where the degree of public and congressional deference to the president's policy agenda is greater than in the domestic arena (Peake 2001; Peterson 1994; see also Holsti 1996).

Of course, an increase in public support may not reach or surpass a desired threshold (e.g., majority opinion) and even if it does, it may not be enough to help the president take action, especially if the president is relying on a polarized, gridlocked Congress to endorse a policy, much less provide any "potential for conversion" (Edwards 2009, 186). Too often, presidents have fallen to the temptation of "overreach" by overestimating their ability to create, rather than facilitate, opportunities while downplaying the limitations they face in attempting to push policy changes through Congress, particularly under divided government (Edwards 2009; 2012). Such legislative stumbling blocks can sometimes be circumvented, however. Historically, when situated strategically (e.g., with support from key allies abroad and at least the support of their political base at home, among other factors), presidents have from time to time acted unilaterally in overseas interventions with limited interference from the U.S. Congress (Fisher 2000; see also Howell 2003; Mayer 2001; Warber 2006). Nevertheless, presidents risk losing political capital and public confidence if they act unilaterally without broad public support or at the least the perception of effective opinion leadership. Ideally, by achieving an increase in support--perhaps even a moderate, temporary one--presidents can use such shift to their advantage in creating a narrative to justify their actions, whether it be with the consent of the legislature or otherwise unilaterally. In line with our results here, one can argue that anger stimulated through public appeals may serve the president in just such a manner provided the circumstances are ripe for taking military action.

Conclusion

In this study, we explored the conditionality in how presidential public appeals with emotion-inducing cues work. Presidents can stoke anger through their speeches but does the manner in which they do it matter for their policy agendas? We find that the induction of anger via thematically relevant emotive triggers leads to higher levels of support for military interventions in civil conflict even though people's risk perceptions--which we found to be high across all conditions--remain unaffected. As we expected, contrary to the significant effects of the thematically relevant emotive triggers, the effects of anger on policy support observed in the thematically irrelevant emotive trigger conditions do not differ from the emotion-neutral control conditions. Therefore, it is the particular feelings of anger aroused by a trigger specific to the policy at hand that significantly affect people's preferences (here, military intervention in a civil conflict) rather than the general feelings of anger (brought on by nonrelevant emotive triggers) that individuals may have in the context of their personal lives.

The findings of this study bear a number of important policy implications. Civil conflicts constitute the most common and violent forms of large-scale armed conflicts in the world today (Hoeffler 2013). Such conflicts often require immediate humanitarian action by the international community. However, scholars find that when it comes to public support for military interventions, the U.S. public is particularly prudent about interventions in civil wars (Jentleson and Britton 1998; Russett and Nincic 1976). Especially when such interventions are mainly for humanitarian purposes (i.e., in the absence of any direct national interests), the public is sensitive to the costs and risks involved with taking action. Our findings suggest that although people's risk perceptions may remain high even after a presidential appeal for a humanitarian intervention, public support may nevertheless increase and bestow more legitimacy for a president to take swift action.

As mentioned above, however, one should keep in mind that emotion-inducing presidential appeals do not always engender strong enough public support to help facilitate policy action. With respect to intervening in civil conflicts for humanitarian purposes, since the onset of the Syrian uprising that began in March 2011, President Obama attempted several times to convince the U.S. public about the need to take action in Syria given the ongoing violence that has left thousands of civilians suffering at the hands of their own government. Although public support for such action somewhat increased, the majority was still opposed to U.S. military involvement in Syria, even after intelligence reports confirmed the use of chemical weapons on civilians by the Assad regime--a "red line" President Obama had previously laid out (Dugan 2013). It was only after ISIS's presence in Syria became perceived as a more serious, direct threat to U.S. national security that public--and congressional--support for military action rose more notably. Yet even then, the U.S. public remained generally reluctant to support any action beyond conducting air strikes in the region (Jones and Newport 2014).

One future avenue of research may thus involve investigating why the U.S. public is supportive of certain humanitarian interventions while opposed to others. This could involve exploring shifts in perceived risk and support by situating the experimental context in a setting where there is a direct threat to the United States and/or direct benefits to U.S. national interests. More broadly, future studies may delve more deeply into the extent to which the context of today's polarized political environment has limited the president's ability to work with the Congress in resolving foreign policy issues while providing incentives for presidents to increasingly act in a more unilateral manner and what that means with regards to the relevance of public opinion leadership for the presidency moving forward.

Another important question to consider is whether the pattern we find here holds for other emotions beyond anger. In other words, is such sandboxing of emotional influence specific to anger or is it generalizable to other emotions (in particular, anxiety)? Marcus et al. (2005) find that the effect of extrinsic anxiety on political tolerance (i.e., anxiety primed by a story that is unrelated to the immediate tolerance situation) is comparable to intrinsic anxiety. Future replications of our study across different policy contexts and political issues to induce other key target emotions would help advance our understanding of the generalizability and limitations concerning thematic relevance.

Given the prevalent use of emotion-inducing appeals by political leaders, our investigation of how and to what extent the thematic relevance of emotive triggers drives people's risk perceptions and policy preferences addresses an important gap in the literature and constitutes a means for better understanding and predicting public reactions to presidential appeals. Furthermore, by taking into account variations in the thematic relevance of emotional states, we propose a novel supposition about the makeup of emotions, their specificity, and their potential effects on decision making. In the broader context of presidential leadership, we maintain that the potential for emotionally stimulated upticks in public support to help justify and facilitate a military intervention depends highly on the president's strategic positioning at home and abroad. Accordingly, we offer a new contribution to various areas of research, including risk perceptions and risk communication, public support for military interventions, and the public presidency.

DOI: 10.1111/psq.12349

References

Averill, James R. 1983. "Studies on Anger and Aggression: Implications for Theories of Emotion." American Psychologist 38 (11): 1145-60.

Banks, Antoine J., and Nicholas Valentino. 2012. "Emotional Substrates of White Racial Attitudes." American Journal of Political Science 56 (2): 286-97.

Baron, Reuben M., and David A. Kenny. 1986. "The Moderator-Mediator Variable Distinction in Social Psychological Research: Conceptual, Strategic, and Statistical Considerations." Journal of Personality and Social Psychology 51 (6): 1173-82.

Bodenhausen, Galen V., Lori A. Sheppard, and Geoffrey P. Kramer. 1994. "Negative Affect and Social Judgment: The Differential Impact of Anger and Sadness." European Journal of Social Psychology 24 (1): 45-62.

Brader, Ted. 2006. Campaigning for Hearts and Minds: How Emotional Appeals in Political Ads Work. Chicago: University of Chicago Press.

Brader, Ted, and George E. Marcus. 2013. "Emotion and Political Psychology." In Oxford Handbook of Political Psychology, eds. Leonie Huddy, Jack S. Levy, and David O. Sears. New York: Oxford University Press, 165-204.

Brader, Ted, George E. Marcus, and Krystin L. Miller. 2011. "Emotion and Public Opinion." In Oxford Handbook of American Public Opinion and the Media, eds. Robert Y. Shapiro and Lawrence R. Jacobs. Oxford, UK: Oxford University Press, 384-401.

Brady, Henry E., and Paul M. Sniderman. 1985. "Attitude Attribution: A Group Basis for Political Reasoning." American Political Science Review 79 (4): 1061-78.

Cacioppo, John T., Wendi L. Gardner, and Gary G. Berntson. 1999. "The Affect System Has Parallel and Integrative Processing Components: Form Follows Function." Journal of Personality and Social Psychology 76 (5): 839-55.

Canes-Wrone, Brandice. 2001. "The President's Legislative Influence from Public Appeals." American Journal of Political Science 45 (2): 313-29.

--. 2004. "The Public Presidency, Personal Approval Ratings, and Policy Making." Presidential Studies Quarterly 34 (3): 477-92.

--. 2006. Who Leads Whom? Presidents, Policy, and the Public. Chicago: University of Chicago Press.

Conover, Pamela Johnston, and Stanley Feldman. 1986. "Emotional Reactions to the Economy: I'm Mad as Hell and I'm not Going to Take it Anymore." American Journal of Political Science 30 (1): 50-78.

De Castella, Krista, Craig McGarty, and Luke Musgrove. 2009. "Fear Appeals in Political Rhetoric about Terrorism: An Analysis of Speeches by Australian Prime Minister Howard." Political Psychology 30 (1): 1-26.

DeSteno, David, Richard E. Petty, Duane T. Wegener, and Derek D. Rucker. 2000. "Beyond Valence in the Perception of Likelihood: The Role of Emotion Specificity." Journal of Personality and Social Psychology 78 (3): 397-416.

Dugan, Andrew. 2013. "U.S. Support for Action in Syria is Low vs. Past Conflicts." Gallup. http:// www.gallup.com/poll/164282/support-syria-action-lower-past-conflicts.aspx (accessed July 7, 2015).

Edwards, George C., III. 2003. On Deaf Ears: The Limits of the Bully Pulpit. New Haven, CT: Yale University Press.

--. 2009. The Strategic President: Persuasion and Opportunity in Presidential Leadership. Princeton, NJ: Princeton University Press.

--. 2012. Overreach: Leadership in the Obama Presidency. Princeton, NJ: Princeton University Press.

--. 2015. "Predicting the Success of Presidential Leadership." Presented at the Annual Meeting of the Southern Political Science Association, New Orleans, LA, January 16, 2015.

Ferguson, Eva Dreikurs, and Beth Eva Ferguson Wee. 2000. Motivation: A Biosocial and Cognitive Integration of Motivation and Emotion. New York: Oxford University Press.

Fisher, Louis. 2000. Congressional Abdication on War and Spending. College Station: Texas A&M University Press.

Folkman, Susan, and Richard S. Lazarus. 1988. "Coping as a Mediator of Emotion." Journal of Personality and Social Psychology 54 (3): 466-75.

Folkman, Susan, Richard S. Lazarus, Christine Dunkel-Schetter, Anita DeLongis, and Rand J. Gruen. 1986. "Dynamics of a Stressful Encounter: Cognitive Appraisal, Coping, and Encounter Outcomes." Journal of Personality and Social Psychology 50 (5): 992-1003.

Gasper, Karen, and Gerald L. Clore. 1998. "The Persistent Use of Negative Affect by Anxious Individuals to Estimate Risk." Journal of Personality and Social Psychology 74 (5): 1350-63.

Goldberg, Julie H., Jennifer S. Lerner, and Philip E. Tetlock. 1999. "Rage and Reason: The Psychology of the Intuitive Prosecutor." European Journal of Social Psychology 29 (5-6): 781-95.

Gross, Kimberly. 2008. "Framing Persuasive Appeals: Episodic and Thematic Framing, Emotional Response, and Policy Opinion." Political Psychology 29 (2): 169-92.

Halperin, Erin, Alexandra G. Russell, Carol S. Dweck, and James J. Gross. 2011. "Anger, Hatred, and the Quest for Peace: Anger Can be Constructive in the Absence of Hatred." Journal of Conflict Resolution 55 (2): 274-91.

Hayes, Andrew F., and Kristopher J. Preacher. 2010. "Quantifying and Testing Indirect Effects in Simple Mediation Models When the Constituent Paths are Nonlinear." Multivariate Behavioral Research 45 (4): 627-60.

Hirt, Edward R., Gary M. Levine, Hugh E. McDonald, and R. Jeffrey Melton. 1997. "The Role of Mood in Quantitative and Qualitative Aspects of Performance: Single or Multiple Mechanisms?" Journal of Experimental Social Psychology 33 (6): 602-29.

Hoeffler, Anke. 2013. "Can International Interventions Secure the Peace?" Economic Research Southern Africa (ERSA), Working Paper 314.

Holsti, Ole R. 1996. Public Opinion and American Foreign Policy. Ann Arbor: University of Michigan Press.

Howell, William G. 2003. Power without Persuasion: The Politics of Direct Presidential Action. Princeton, NJ: Princeton University Press.

Huddy, Leonie, Stanley Feldman, and Erin Cassese. 2007. "On the Distinct Political Effects of Anxiety and Anger." In The Affect Effect: Dynamics of Emotion in Political Thinking and Behavior, eds. George E. Marcus, W. Russell Neuman, and Michael MacKuen. Chicago: University of Chicago Press, 202-30.

Huddy, Leonie, Stanley Feldman, Charles Taber, and Gallya Lahav. 2005. "Threat, Anxiety, and Support of Antiterrorism Policies." American Journal of Political Science 49 (3): 593-608.

Izard, Carroll E., Brian P. Ackerman, Kristen M. Schoff, and Sarah E. Fine. 2000. "Self-Organization of Discrete Emotions, Emotion Patterns, and Emotion-Cognition Relations." In Emotion, Development, and Self-Organization, eds. Marc D. Lewis and Isabela Granic. New York: Cambridge University Press, 15-36.

Jentleson, Bruce W. 1992. "The Pretty Prudent Public: Post-Vietnam American Opinion on the Use of Military Force." International Studies Quarterly 36 (1): 49-74.

Jentleson, Bruce W., and Rebecca L. Britton. 1998. "Still Pretty Prudent: Post-Cold War American Public Opinion on the Use of Military Force." Journal of Conflict Resolution 42 (4): 395-417.

Johnson, Eric J., and Amos Tversky. 1983. "Affect, Generalization, and the Perception of Risk." Journal of Personality and Social Psychology 45 (1): 20-31.

Jones, Jeffrey M., and Frank Newport. 2014. "Slightly Fewer Back ISIS Military Action vs. Past Actions." Gallup. http://www.gallup.com/poll/177263/slightly-fewer-back-isis-military-actionpast-actions.aspx (accessed July 7, 2015).

Keltner, Dacher, Kenneth D. Locke, and Paul C. Aurain. 1993. "The Influence of Attributions on the Relevance of Negative Feelings to Personal Satisfaction." Personality and Social Psychology Bulletin 19 (1): 21-29.

Lazarus, Richard S. 1984. "On the Primacy of Cognition." American Psychologist 39 (2): 124-29.

--. 1991. "Cognition and Motivation in Emotion." American Psychologist 46 (4): 352-67.

--. 1994. Emotion and Adaptation. New York: Oxford University Press.

Lerner, Jennifer S., Roxana M. Gonzalez, Deborah A. Small, and Baruch Fischhoff. 2003. "Effects of Fear and Anger on Perceived Risks of Terrorism: A National Field Experiment." Psychological Science 14 (2): 144-50.

Lerner, Jennifer S., and Dacher Keltner. 2000. "Beyond Valence: Toward a Model of Emotion-Specific Influences on Judgment and Choice." Cognition & Emotion 14 (4): 473-93.

--. 2001. "Fear, Anger, and Risk." Journal of Personality and Social Psychology 81 (1): 146-59.

Levy, Jack S. 2002. "Qualitative Methods in International Relations." In Millennial Reflections on International Studies, eds. Michael Brecher and Frank P. Harvey. Ann Arbor: University of Michigan Press, 432-54.

MacKuen, Michael B., Jennifer Wolak, Luke Keele, and George E. Marcus. 2010. "Civic Engagements: Resolute Partisanship or Reflective Deliberation." American Journal of Political Science 54 (2): 440-58.

Marcus, George E. 2000. "Emotions in Politics." Annual Review of Political Science 3 (1): 221-50.

Marcus, George E., John L. Sullivan, Elizabeth Theiss-Morse, and Daniel Stevens. 2005. "The Emotional Foundation of Political Cognition: The Impact of Extrinsic Anxiety on the Formation of Political Tolerance Judgments." Political Psychology 26 (6): 949-63.

Marcus, George E., Michael MacKuen, Jennifer Wolak, and Luke Keele. 2006. "The Measure and Mismeasure of Emotion." In Feeling Politics: Emotion in Political Information Processing, ed. David Redlawsk. New York: Palgrave Macmillan, 31-45.

Marcus, George E., W. Russell Neuman, and Michael MacKuen. 2000. Affective Intelligence and Political Judgment. Chicago: University of Chicago Press.

Masters, Roger D., and Denis G. Sullivan. 1989. "Nonverbal Displays and Political Leadership in France and the United States." Political Behavior 11 (2): 123-56.

Mayer, Kenneth R. 2001. With the Stroke of a Pen: Executive Orders and Presidential Power. Princeton, NJ: Princeton University Press.

Nabi, Robin L. 1999. "A Cognitive Functional Model for the Effects of Discrete Negative Emotions on Information Processing, Attitude Change and Recall." Communication Theory 9 (3): 292-320.

Obama, Barack. 2011. Remarks on the Situation in Libya. March 18, 2011. The American Presidency Project, http://www.presidency.ucsb.edu/ws/Ppid590162 (accessed July 7, 2015).

Page, Benjamin I., and Robert Y. Shapiro. 1992. The Rational Public: Fifty Years of Trends in Americans' Policy Preferences. Chicago: University of Chicago Press.

Peake, Jeffrey S. 2001. "Presidential Agenda Setting in Foreign Policy." Political Research Quarterly 54 (1): 69-86.

Peterson, Paul E. 1994. "The President's Dominance in Foreign Policy Making." Political Science Quarterly 109 (2): 215-34.

Pham, Michel T. 1998. "Representativeness, Relevance, and the Use of Feelings in Decision Making." Journal of Consumer Research 25 (2): 144-59.

Redlawsk, David P. 2002. "Hot Cognition or Cool Consideration? Testing the Effects of Motivated Reasoning on Political Decision Making." The Journal of Politics 64 (4): 1021-44.

Rosenboim, Mosi, Uri Benzion, Shosh Shahrabani, and Tal Shavit. 2012. "Emotions, Risk Perceptions, and Precautionary Behavior under the Threat of Terror Attacks: A Field Study among Israeli College Students." Journal of Behavioral Decision Making 25 (3): 248-56.

Rottinghaus, Brandon. 2009. "Strategic Leaders: Determining Successful Presidential Opinion Leadership Tactics through Public Appeals." Political Communication 26 (3): 296-316.

--. 2010. The Provisional Pulpit: Modern Presidential Leadership of Public Opinion. College Station: Texas A&M University Press.

Russett, Bruce, and Miroslav Nincic. 1976. "American Opinion on the Use of Military Force Abroad." Political Science Quarterly 91 (3): 411-31.

Sabucedo, Jose Manuel, Mar Duran, Monica Alzate, and Maria-Soledad Rodriguez. 2011. "Emotional Responses and Attitudes to the Peace Talks with ETA." Revista Latinoamericana de Psicologia 43 (2): 289-96.

Sadler, Melody S., Megan Lineberger, Joshua Correll, and Bernadette Park. 2005. "Emotions, Attributions, and Policy Endorsements in Response to the September 11th Terrorist Attacks." Basic and Applied Social Psychology 27 (3): 249-58.

Sampson, Robert J., Stephen W. Raudenbush, and Felton Earls. 1997. "Neighborhoods and Violent Crime: A Multilevel Study of Collective Efficacy." Science 277 (5328): 918-24.

Schwarz, Norbert, and Gerald L. Clore. 1983. "Mood, Misattribution, and Judgments of WellBeing: Informative and Directive Functions of Affective States." Journal of Personality and Social Psychology 45 (3): 513-23.

--. 1996. "Feelings and Phenomenal Experiences." In Social Psychology: Handbook of Basic Principles, eds. E. Tory Higgins and Arie W. Kruglanski. New York: Guilford, 433-65.

Schwarz, Norbert, Wolfgang Servay, and Martin Kumpf. 1985. "Attribution of Arousal as a Mediator of the Effectiveness of Fear-Arousing Communications." Journal of Applied Social Psychology 15 (2): 178-88.

Siemer, Matthias, and Rainer Reisenzein. 1998. "Effects of Mood on Evaluative Judgments: Influence of Reduced Processing Capacity and Mood Salience." Cognition & Emotion 12 (6): 783-805.

Sirin, Cigdem V., and Nehemia Geva. 2013. "Examining the Distinct Effects of Emotive Triggers on Public Reactions to International Terrorism." Terrorism and Political Violence 25 (5): 709-33.

Skitka, Linda J., Christopher W. Bauman, Nicholas P. Aramovich, and G. Scott Morgan. 2006. "Confrontational and Preventative Policy Responses to Terrorism: Anger Wants a Fight and Fear Wants 'Them' to Go Away." Basic and Applied Social Psychology 28 (4): 375-84.

Smith, Craig A., and Phoebe C. Ellsworth. 1985. "Patterns of Cognitive Appraisal in Emotion." Journal of Personality and Social Psychology 48 (4): 813-38.

Sullivan, Denis, and Roger Masters. 1988. "Happy Warriors: Leaders' Facial Displays, Viewers Emotions, and Political Support." American Journal of Political Science 32 (2): 345-68.

Tavris, Carol. 1982. Anger: The Misunderstood Emotion. New York: Simon & Schuster.

Tellegan, Auke, David Watson, and Lee Ann Clark. 1999. "On the Dimensional and Hierarchical Structure of Affect." Psychological Science 10 (4): 297-303.

Valentino, Nicholas A., Antoine J. Banks, Vincent L. Hutchings, and Anne K. Davis. 2009. "Selective Exposure in the Internet Age: The Interaction between Anxiety and Information Utility." Political Psychology 30 (4): 591-613.

Valentino, Nicholas A., Ted Brader, Eric W. Groenendyk, Krysha Gregorowicz, and Vincent L. Hutchings. 2011. "Election Night's Alright for Fighting: The Role of Emotions in Political Participation." The Journal of Politics 73 (1): 156-70.

Valentino, Nicholas A., Vincent L. Hutchings, Antoine J. Banks, and Anne K. Davis. 2008. "Is a Worried Citizen a Good Citizen? Emotions, Political Information Seeking, and Learning via the Internet." Political Psychology 29 (2): 247-73.

Van Zomeren, Martijn, Russell Spears, and Colin Wayne Leach. 2008. "Exploring Psychological Mechanisms of Collective Action: Does Relevance of Group Identity Influence How People Cope with Collective Disadvantage?" British Journal of Social Psychology 47 (2): 353-72.

Villalobos, Jose D., and Cigdem V. Sirin. 2012. "Agenda Setting from the Oval Office: An Experimental Examination of Presidential Influence over the Public Agenda." International Journal of Public Opinion Research 24 (1): 21-41.

Warber, Adam L. 2006. Executive Orders and the Modern Presidency: Legislating from the Oval Office. Boulder, CO: Lynne Rienner Publishers.

Watson, David, and Lee Anna Clark. 1992. "Affects Separable and Inseparable: On the Hierarchical Arrangement of the Negative Affects." Journal of Personality and Social Psychology 62 (3): 489-505.

Watson, David, David Wiese, Jatin Vaidya, and Auke Tellegen. 1999. "The Two General Activation Systems of Affect: Structural Findings, Evolutionary Considerations, and Psychobiological Evidence." Journal of Personality and Social Psychology 76 (5): 820-38.

Way, Baldwin, and Roger Masters. 1996. "Emotion and Cognition in Political-Information Processing." Journal of Communications 46 (3): 48-65.

Weitzer, Ronald, and Steven A. Tuch. 2006. Race and Policing in America: Conflict and Reform. New York: Cambridge University Press.

Zajonc, Robert B. 1998. "Emotions." In The Handbook of Social Psychology, Vol. 1, eds. Daniel T. Gilbert, Susan T. Fiske, and Gardner Lindzey. New York: McGraw-Hill, 591-632.

Zaller, John. 1994. "Elite Leadership of Mass Opinion: New Evidence from the Gulf War." In Taken by Storm: The Media, Public Opinion, and U.S. Foreign Policy in the Gulf War, eds. W. Lance Bennett and David L. Paletz. Chicago: University of Chicago Press, 186-209.

Supporting Information

Additional supporting information may be found in the online version of this article:

Appendix

(1.) In addition to verbal emotion-inducing appeals, several studies show that a leader's nonverbal display behavior (such as facial displays of happiness/reassurance, anger/threat, or fear/evasion) also evokes differential emotional responses and helps shape trait attributions and political attitudes among the public (e.g., Sullivan and Masters 1988; Masters and Sullivan 1989; Way and Masters 1996). For instance, Sullivan and Masters (1988) suggest that President Reagan was effective in communicating positive emotions and generating favorable attitudes in viewers through nonverbal facial expressive displays during his televised appearances.

(2.) Utilizing a best test "least likely" case provides a means for testing the validity and scope of our theoretical expectations in the most conservative manner (see the "Methodology" section for full details).

(3.) For recent, comprehensive reviews of the literature on the role of emotions in politics, see Brader, Marcus, and Miller (2004) and Brader and Marcus (2013).

(4.) Several studies suggest that cognitive and behavioral consequences of anger also depend on the intensity of anger (Averill 1983; Nabi 1999; Tavris 1982). According to this view, milder forms of anger may actually yield constructive decision-making processes and outcomes such as developing a moderate willingness to carefully consider a situation and/or motivation for problem solving. On a parallel basis, Halperin et al. (2011) argue that in the absence of hatred, anger can indeed be constructive and increase support for compromises in the resolution of intergroup conflict.

(5.) We provide all of the experimental materials in the supplementary appendix (available upon request).

(6.) A natural step in the political emotions literature has been the move toward manipulating emotions in a way that is not only high in internal validity but also externally realistic. For example, the induction task Valentino, Banks, Hutchings, and Davis (2009) employ is a more realistic manipulation of anxiety based on one's real world experiences about politics. With that in mind, we created our experimental scenarios in the form of presidential speeches in close concert to the type of content found in real presidential speeches by using the Miller Center's American President: Presidential Speech Archive (http://millercenter.org/president/ speeches) with a focus on crime in U.S. cities and conflict abroad as issues presidents often address in their public appeals.

(7.) Instead of using a fictional country name for the venue of the civil conflict in the presidential speech, we chose to refer to a real country, Malawi, to help ensure the experimental scenario was realistic. Our objective was for the participants to perceive it as a real country but not be too familiar with it in order to avoid any preexisting attitudes that might confound the effects of our experimental stimuli.

(8.) See Tables A1-A2 in the supplementary appendix for all the statistics regarding these manipulation checks.

(9.) We coded the gender variable as 1 for male and 0 for female. The party identification variable is coded as 1 for Democrat, 2 for Independent, and 3 for Republican. The ideology variable ranges from 1 (very conservative) to 5 (very liberal). As mentioned above, we then rescaled the measures to range from 0 to 1 for ease of comparison and interpretation.

(10.) See Tables A3-A4 in the supplementary appendix.

(11.) See the supplementary appendix for the full wording of each condition.

(12.) These scales and all of the other measures employed in our analyses for Experiment 2 are transformed to range from 0 to 1 for ease of comparison and interpretation.

(13.) These ANOVA results are available in Tables A5 and A6 of the supplementary appendix.

(14.) We found that the level of aversion in Experiment 2 was higher in the thematically irrelevant emotive trigger condition ([M.sub.TI] = .61) than the thematically relevant one ([M.sub.TR] = .54; p < .05). This suggests that we are testing our hypotheses in a conservative manner because the intensity of aversion was actually higher for the thematically irrelevant emotive trigger condition compared to the thematically relevant condition, which may counteract the proposed effects of the thematic relevance of aversion on risk perceptions and policy preferences.

(15.) Replicating the analyses using ANCOVA yields results similar to those using logistic regression (see Tables A7-A12 in the supplementary appendix).

(16.) The results of both Experiments 1 and 2 thus demonstrate that the thematic relevance of anger that presidential speeches trigger increases support for military intervention in civil conflict but does not affect people's risk perceptions. To further explore the process behind policy support, Experiment 2 included two additional posttest measures: perceived costs and benefits associated with intervening militarily in said civil conflict. Analyzing the experimental data using these additional cost and benefit measures reveals that the thematically relevant emotive trigger condition does not change the perceptions of costs associated with military interventions in civil conflict but does significantly increase the perceived benefits of such foreign policy action (see Table A13 in the supplementary appendix). These results may partly explain why support increases despite unchanged perceptions of risk. Further research should investigate such possibility.

(17.) As previously mentioned, the data for emotional response measures were collected from a separate pool of participants in the pretest for Experiment 1. By comparison, participants in Experiment 2 responded to both the emotional response and policy questions following their exposure to the presidential speech, which allowed us to conduct path analysis.

(18.) The path analysis also includes all of the control variables used in our previous analyses.

(19.) We conducted path analysis only for policy support and not for perceived risk given that the experimental results already demonstrated that perceived risk of military intervention remained unchanged across conditions.

(20.) This finding that risk perceptions decreased for domestic crime but not for a military conflict abroad makes sense. Presumably, particularly when one is emotionally moved by anger, an increase in law enforcement in the domestic realm may seem less likely to lead to direct, deadly violence between law enforcement officials and local criminals compared to the risks involved amid the fog of war in a military conflict abroad.

(21.) Aside from the initial public support a president might garner, the president should nevertheless be careful about the manner and extent to which law enforcement increases might be implemented. In particular, the president should consider the possibility of a subsequent backlash (perhaps in the form of lost trust and/or overall support) in cases where actions taken by law enforcement officials may be seen as encroaching on civil rights and liberties in local communities--especially racial/ethnic minority communities historically wary of possible acts of police brutality or other misconduct (e.g., see Weitzer and Tuch 2006).

Jose D. Villalobos is an associate professor of political science at the University of Texas, El Paso. His publications include Czars in the White House: The Rise of Policy Czars as Presidential Management Tools (coauthor) and articles in journals such as The Journal of Politics and Public Administration. Cigdem V. Sirin is an associate professor of political science at the University of Texas, El Paso. Her publications include articles in journals such as The Journal of Politics, Political Psychology, and International Studies Quarterly.

Caption: FIGURE 1. Experiment 2: Path Analysis of Thematic Relevance, Aversion, and Policy Support Note: Path coefficients (with standard errors in parentheses) estimated via generalized structural equation modeling. The significance of indirect effects tested via bootstrapping with bias-corrected confidence intervals. *** p < .001; ** p < .01; * p < .05 (two-tailed).
TABLE 1
Experiment 1: The Effect of Emotive Triggers in a Presidential
Speech on Perceived Risk of and Support for Military Intervention in
Civil Conflict Abroad

                          Perceived Risk    Policy Support

Thematically relevant          -.25             82 ***
emotive trigger                (.21)             (.21)
Thematically irrelevant         .05               .33
emotive trigger                (.21)             (.20)
Gender                         .38*               .13
                               (.18)             (.18)
Party identification            .01              -.43
                               (.23)             (.23)
Ideology                       -.26              -.35
                               (.43)             (.43)
N                               452               452
Log likelihood                -704.79           -837.21
LR [chi square]                7.04              20.28

Note: Ordered logistic regression. Emotion-neutral control condition
is the baseline comparison category.  *** p < .001; ** p < .01; * p
< .05 (two-tailed).

TABLE 2
Experiment 2: Factor Analysis of Emotional Responses to the
Presidential Speech

Emotional Response   Factor 1:   Factor 2:   Factor 3:
                     Aversion     Anxiety    Enthusiasm

Angry                   .87
Frustrated              .81
Disgusted               .86
Bitter                  .82
Resentful               .81
Anxious                             .73
Worried                             .86
Afraid                              .87
Proud                                           .82
Hopeful                                         .84
Enthusiastic                                    .80

Note: Principal components factor analysis, varimax rotation.

TABLE 3
Experiment 2: The Effect of Emotive Triggers in a Presidential
Speech on Perceived Risk of and Support for Military Intervention in
Civil Conflict Abroad (Foreign Policy Domain)

                                   Perceived Risk    Policy Support

Thematically relevant                   -.25             1.09 *
  emotive trigger                       (.24)             (.24)
Thematically irrelevant                  .01               .36
  emotive trigger                       (.25)             (.24)
Thematically relevant                   -.29               .46
  emotion-neutral control               (.24)             (.24)
Identification with residents           .69 *             -.06
  of violent countries                  (.34)             (.34)
Identification with residents           -.19               .18
  of high-crime cities                  (.34)             (.34)
Enthusiasm towards the president        -.20            1.92 ***
                                        (.35)             (.36)
Anger towards the president             -.27              -.50
                                        (.33)             (.33)
Anxiety towards the president            .45               .10
                                        (.34)             (.34)
Age                                     1.44              -.59
                                        (.74)             (.76)
Gender                                  .38 *              .06
                                        (.17)             (.17)
Party identification                    -.06               .08
                                        (.24)             (.24)
Ideology                                 .37              -.17
                                        (.42)             (.42)
Minority race/ethnicity                 -.21               .14
                                        (.30)             (.30)
Income                                  -.17              -.35
                                        (.27)             (.27)
N                                        473               473
Log likelihood                         -794.70           -857.85
LR [chi square]                         19.00             62.35

Note: Ordered logistic regression. Thematically irrelevant
emotion-neutral control is the baseline  comparison category.
*** p < .001; ** p < .01; * p < .05 (two-tailed).

TABLE 4
Experiment 2: The Effect of Emotive Triggers in a Presidential
Speech on Perceived Risk of and Support for Increasing Law
Enforcement in U.S. Cities (Domestic Policy Domain)

                                   Perceived Risk    Policy Support

Thematically relevant                  -.55 *            .71 **
  emotive trigger                       (.24)             (.24)
Thematically irrelevant                 -.15               .33
  emotive trigger                       (.24)             (.24)
Thematically relevant                   .001              -.17
  emotion-neutral control               (.24)             (.24)
Identification with residents            .33              -.01
  of violent countries                  (.34)             (.34)
Identification with residents           -.26               .26
  of high-crime cities                  (.34)             (.34)
Enthusiasm towards the president        -.57            2.10 ***
                                        (.34)             (.35)
Anger towards the president              .57              -.48
                                        (.32)             (.33)
Anxiety towards the president            .35               .24
                                        (.34)             (.35)
Age                                   -2.05 **            1.35
                                        (.77)             (.72)
Gender                                  -.26             .57 ***
                                        (.17)             (.17)
Party identification                    -.33              -.33
                                        (.23)             (.23)
Ideology                               1.10 **          -1.44 ***
                                        (.41)             (.43)
Minority race/ethnicity                  .18              -.18
                                        (.30)             (.30)
Income                                   .11              -.02
                                        (.27)             (.27)
N                                        473               473
Log likelihood                         -837.95           -842.19
LR [chi square]                         38.70             78.07

Note: Ordered logistic regression. Thematically irrelevant
emotion-neutral control is the baseline  comparison category.
*** p < .001; ** p < .01; * p < .05 (two-tailed).
COPYRIGHT 2017 Center for the Study of the Presidency
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2017 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Villalobos, Jose D.; Sirin, Cigdem V.
Publication:Presidential Studies Quarterly
Geographic Code:1USA
Date:Mar 1, 2017
Words:11350
Previous Article:The president, the court, and policy implementation.
Next Article:Dwight Eisenhower and American Foreign Policy During the 1960s: An American Lion in Winter.
Topics:

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters