بحران جاذبه ارزش ها و رفتار رای دادن در انتخابات ریاست جمهوری سال 2004
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|3349||2009||17 صفحه PDF||سفارش دهید||محاسبه نشده|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : The Leadership Quarterly, Volume 20, Issue 2, April 2009, Pages 70–86
This study extends Pillai and Williams [1998, Pillai, R., Williams, E.A., Lowe, K.B., & Jung, D.I. (2003). Personality, transformational leadership, trust, and the 2000 U.S. presidential vote. The Leadership Quarterly, 14, 161–192] and examines leadership in the context of the 2004 U.S. presidential election. Data were collected at two time periods from respondents in three locations across two major regions of the U.S. Our results indicate that respondents' perception of crisis was related to charismatic leadership in the negative direction for the incumbent George W. Bush and in the positive direction for the challenger John Kerry. For Bush and Kerry the relationship between crisis and voting behavior was mediated by charismatic leadership. For Bush, decisiveness was related to charismatic leadership, which in turn predicted voting behavior. For Kerry, decisiveness and charismatic leadership predicted voting behavior. Implications of the findings for leadership research, in particular with respect to an incumbent and the challenger to an incumbent leader, are discussed.
نتیجه گیری انگلیسی
Overall, our findings suggest that situational and contextual elements play an important role in follower perceptions of charismatic leadership and effectiveness. This is one of the unique contributions of this research in the context of a presidential election. The findings show that both leadership ratings and party identification are related to voting preference and choice, and that, in combination, these variables can predict the vote for a particular candidate. This result replicates the findings of the Pillai & Williams (1998) study of the 1996 election and the Pillai et al. (2003) study of the 2000 election, reemphasizing the importance of leadership evaluations and party identification on voting behavior. The results also show that voters may be partisan when it comes to evaluating their leaders: Democrats saw John Kerry as significantly more charismatic than Republicans and the same held true for the Republicans with respect to George W. Bush. These perceptions may be deeply entrenched as one opinion poll conducted in June 2005, nearly 7 months after this very tight 2004 election, found that virtually all of Bush and Kerry voters (94%) still agreed with their vote with only 4% of Bush voters and 3% of Kerry voters indicating they regretted their vote. This investigation goes beyond the earlier studies and examines the role of crisis, value congruence, and decisiveness in the voting decision. The results of our direct effects tests show that respondents' perceptions of crisis are related to attributed charisma and to voting behavior. Both of these relationships were, as hypothesized, negative for Bush and positive for Kerry. The results of our tests of the more sophisticated mediation model was also supported with the relationship between crisis perceptions and voting behavior mediated by attributed charisma for both candidates. These findings have significant implications for those seeking to understand how voters make decisions as well as those seeking to influence those voting decisions. Assumptions such as crisis favors the Republican (conservative) party or that crisis favors the challenger are too simplistic. A more complex consideration of the interaction between the leaders positioning in the context (challenger versus incumbent), the subjective appraisal of the situation (crisis perception), and perception of leader characteristics (attributed charisma) is warranted. Our findings of direct (in different directions) and mediating effects for both candidates suggest that crisis does create differential challenges and opportunities for incumbents and challengers. However, the ability to ameliorate or capitalize on circumstances is impacted by the extent to which leadership perceptions are simultaneously developed or managed. This finding would appear to offer rich opportunities for social information processing where we might seek to deconstruct the cognitive processes that integrate individual perceptions of crisis with the process of attributing charisma to leaders. Building on our model and the results of our first set of results, our second set of tests investigated the direct effects of attributed charisma and value congruence on voting behavior as well as the potential for attributed charisma to mediate the value congruence to voting behavior relationship. Our results show that both value congruence and attributed charisma have direct effects on voting behavior but that attributed charisma does not mediate the value congruence to voting behavior relationship. Thus, a simpler model than that for crisis perceptions is indicated here. Again, our results have significant implications for those seeking to understand how voters make decisions as well as those seeking to influence those voting decisions. Our results suggest that voters consider the leader attributes, as measured by attributed charisma somewhat independently of their assessment of congruence between their values and those of the leader. Since much of the leadership literature suggests that the sine qua non of leadership is aligning their message with the aspirations of followers, future research might explore how or why voters separate their assessment of leader charisma from their evaluation of the leader's values. It may well be that followers focus on the ability to deliver summary outcomes rather than the underlying processes when evaluating leadership abilities (e.g., can he get us out of the war even if my reasons for wanting this are different than the leader's?). From an applied perspective the results suggest that those wishing to influence voting behavior can simultaneously focus on communicating value congruence and on boosting charismatic impressions of leaders without an overriding concern for the potential costs of mixed messages. Our third and final set of tests incorporated decisiveness to investigate the direct effects of decisiveness and attributed charisma on voting behavior as well as the potential for attributed charisma to mediate the decisiveness to voting behavior relationship. Our results show that both decisiveness and attributed charisma have direct effects on voting behavior. However, our results for tests of mediation were mixed with attributed charisma a mediator for the incumbent Bush but not for the challenger Kerry. Thus, the interpretation and implications of this result are less straightforward than our other tests of hypotheses. One speculative explanation for this finding is that over an extended time period, follower observations of a leader's decisiveness are a primary input to attributions of leader charisma. Thus, for the incumbent, decisiveness and attributed charisma share considerable variance in follower evaluations. However, for challengers where follower observations have been fewer, of a shorter duration, and not “battle tested” decisiveness and charisma may still be relatively separate cognitive categorizations. 7.1. Theoretical and practical implications For incumbent leaders, crisis events will likely lead to decreased perceptions of charisma when leaders are perceived to be unsuccessful in their policies. This is probably due to the fact that the voters often blame the incumbents for the ongoing crisis. We recognize that this is speculative and that we did not test for this effect in the current study. The steady decline in President Bush's leadership ratings prior to and following the mid-term elections in 2006 is probably a reflection of blame attribution by voters for the country's problems. These findings are consistent with prior research which found a negative correlation between perceptions of crisis and perceptions of charisma of the existing unit leaders (Pillai & Meindl, 1998) and also a study of the California governor recall election of 2003 (Bligh et al., 2004). In the latter study they found that perceptions of crisis decreased charismatic attributions of the incumbent, Gray Davis, who was probably being blamed for the crisis by some of the voters. Our findings seem to indicate that the relationship between crisis and charisma and the resulting effects on outcomes may operate very differently depending on other characteristics of the leader and the situation, as Shamir & Howell (1999) have previously proposed. An understanding of incumbency effects perhaps helps to explain how the process works. In Anderson & Glomm's (1992) discussion of the first mover advantage, a candidate who appears to avoid taking strong positions on key issues, or based on the current study, one who may be seen as indecisive, may be seen as the second mover and thus have a disadvantage. Since incumbents might have the higher probability of being elected than challengers, the perception of being slow to act or being seen as less decisive than the incumbent may allow the incumbent benefit from a charisma advantage. This may be reflected in political analysts' reports that voters believed they knew exactly what President Bush stood for but they never clearly understood how Kerry would solve the crises posed by the escalating violence in Iraq, the war on terrorism, and a weak national economy. The Bush administration was the first one since Herbert Hoover to preside over a simultaneous decline in payroll jobs and the stock market. Obviously, Kerry was not an incumbent leader who was being blamed for the perceived crisis; rather, some of his supporters saw him as an alternative to the status quo. However, in a study conducted before the 2004 election, Landau et al. (2004) found that mortality salience and reminders of 9/11 increased support for Bush and the desire to vote for him in the forthcoming election but caused less favorable attitudes toward and reduced the inclination to vote for Kerry. This reflects terror management theory which posits that in times of crisis, people experience heightened fears of death, which causes them to turn to charismatic leaders to feel protected (Landau et al., 2004). In the end, given the crisis context of the situation, voters believing they knew where Bush stood, may have preferred “the devil you know to the devil you don't.” Identification with the leader and alignment between the values of the leader and followers is one of the important aspects of charismatic leadership. The interest that presidential leadership generates might be paralleled to that of leaders of prominent multi-nationals. In many cases there is similar media attention given to high profile CEOs and Presidential candidates, suggesting that the generalizability of our findings may extend to high profile leadership in organizational contexts. As Chen et al. (2007) argue, charismatic leadership constructions emerge and are sustained when the leader is perceived as looking out for the interests of the follower. Considerable media attention was given to the importance of values in the 2004 presidential election, and values emerged as a top concern of voters. Indeed James McGregor Burns is cited in Riggio (2004) as suggesting that the greater than expected mobilization of right-wing and evangelical Christians was accomplished by a focus on moral values rather than on substantive issues such as education and the economy. Of those who said that “moral values” was their top concern, 78% voted for the president (Fineman, 2004). Our study shows that in general, leadership evaluations and value congruence influenced reported voting behavior. The relationship between value congruence and the emergence of charismatic leadership is consistent with past research (Jung & Avolio, 2000). Both Democrats and Republicans who identified with their leader's values and saw him as charismatic were influenced to vote for him. It would be interesting to see how this influences the 2008 presidential vote. The present research added to findings from the Pillai et al. (2003) study which found support for predictions that empathy, achievement orientation, and proactivity would impact charisma and transformational leadership by showing that leader decisiveness played an important role in influencing ratings of charismatic leadership for both candidates. The Republicans perceived Bush as decisive and the Democrats perceived Kerry as decisive and this influenced their leadership evaluations of the candidates and allowed them to emerge as leaders among members of their party. It is possible that followers share perceptions of leadership characteristics and qualities that are then associated with leadership emergence. Through these series of studies, we are beginning to understand which characteristics are associated with presidential leadership. The role of decisiveness and charisma therefore appears to be critical to understanding the role of leadership perceptions in the voting decision. Further analysis revealed that decisiveness mediated the effects of crisis on reported voting behavior for Kerry. His ratings on decisiveness were statistically significantly lower than the ratings for Bush (3.07 vs. 3.22, see Table 1). It therefore appears that perceptions of Kerry as less decisive than Bush were more important to the voting decision than the belief that a crisis situation existed. The ratings of Kerry and Bush on attributed charisma were not statistically significantly different from each other. The mediating effects of attributed charisma on the relationship between crisis and voting for the incumbent and challenger suggest that charisma was more important to the voting decision than the belief that a crisis situation existed. Thus, even though the effects of crisis were negative for Bush's leadership ratings and reported voting for Bush, and positive for Kerry, at the end of the day it appears that Bush's charisma and perceived decisiveness were the determining factors in his success in the election. In order to understand leader emergence and the advantage that incumbency effects provide, Cohen, Solomon, Maxfield, Pyszczynski, & Greenberg (2004) conducted an experiment to examine the effects of a subtle reminder of death on voting intentions for the 2004 U.S. presidential election. This research on mortality salience shows that the preference for charismatic leadership from political leaders increases when terror management needs are activated by the reminder of mortality (Cohen et al., 2004). When subjects were presented with a mortality salience induction condition and exposed to candidate statements with task-oriented, relationship-oriented, and charismatic messages, they were more likely to express a preference and vote for the political leader with a charismatic message (Cohen et al., 2004). Thus, it appears that Bush was favored over Kerry following a reminder of death, suggesting that President Bush's re-election may have been facilitated by nonconscious concerns about mortality in the aftermath of September 11, 2001. 7.2. Limitations and strengths of the study 7.2.1. Limitations The current research sampled a limited set of voters in a large “Blue” (Democratic Party) state (California with 55 electoral votes), a large “Red” (Republican Party) state (Florida with 27 electoral votes) and a medium sized Red state (North Carolina with 15 electoral votes). Respondents were limited to a small section of the population since all were students in business schools and the average age of our sample was 25.5 years which is younger than the average age of the U.S. population. Another limitation of this study is that our measures of decisiveness, value congruence, and leadership were obtained at one point in time. We therefore could not establish causality in the modeling of the relationships. The nature of our study necessitated asking respondents to report on decisiveness, crisis, and leadership variables observed. While research by Crampton & Wagner (1994) and Spector (1994) suggests that the bias caused by self reports might be overstated it is possible that common source or method variance affected our results. We recognize that there might be an important gap between how people view these presidential candidates as merely “public figures” and how people rate them as supervisory leaders. Thus, the approach we took in having respondents describe each candidate as if they were the direct report or follower limits the conclusions that we can make and limits the generalizability of our findings. Research that has concluded that the much of the variance in the results across studies relating traits to leadership was found to be due to methodological artifacts caution that their results generalize to leadership perceptions only (Judge, Colbert & Ilies, 2004). We also echo this caution that our results may generalize to leadership perceptions only. Similar to the approach employed by House et al. (1991), we employed a number of techniques in order to reduce biased responses due to the single-source approach employed and the approach that asked respondents to rate the candidate as if they were a direct follower. While we attempted to capture respondents' evaluation of the decisiveness of the candidates we cannot be certain that our approach to have them think of themselves as a close follower was effective in having them be able to effectively ascertain the extent to which this was a trait of the candidate. Recent research on presidents highlights the fact that it is common to have biographers rate presidential personality but there are often only a few biographers available to provide such ratings. Other approaches include having individuals read biographies and then provide ratings (Rubenzer & Faschingbauer, 2004). Our approach was to have respondents rate how they perceived the candidates based on exposure to information in the media which is similar to the above approach but less reliable. Thus, our conclusions are based on respondent perceptions rather than a more objective approach. Our measure of decisiveness also mixed observation type questions with those that captured the nature of the candidates in order to have respondents think broadly about how decisions were made. Thus, our measure of decisiveness might have more accurately captured decisiveness of actions than been reflective of decisiveness as a trait. The variables considered in the study were limited in scope and excluded potentially important factors that might affect the reported results. While we measured voter perceptions of crisis, attributed charisma, and leader decisiveness we did not measure voter knowledge of or involvement with political issues. It may well be that voters who intensely scrutinize various media to evaluate candidate characteristics and positions on political issues would report a different pattern of relationships than more casual observers of the political landscape. Such a finding would be consistent with the performance appraisal literature where subject matter experts report a different set of relationships between behavior and performance ratings than do novice observers. Future research might seek to control for intensity of voter involvement to determine how political awareness impacts ratings of crisis, decisiveness, and attributed charisma. We also would suggest that intensity of involvement might influence follower liking of the leader (c.f. Brown and Keeping, 2005) which may impact perceptions of value congruence and to some extent attributed charisma. Future research might also examine the traits of the respondents which might influence the way that the candidates were perceived. 7.2.2. Strengths In alignment with one report by the U.S. census bureau (U.S. Census Bureau, 2000) that by November 2000 about 40% of the voting age population would be ages 25 to 44, election reports by the major media outlets indicated that this age group represented a large segment of the electorate. Exit polling for the 2004 U.S. presidential election indicated that approximately 46% of the individuals reporting that they voted in the 2004 presidential election were between the ages of 18 and 44 (approximately 95% of our respondents fell into this age range). Thus, the age of our sample is representative of a substantial portion of the U.S. population while the education level is likely above the national average. Although, future research needs to sample voters from across the country representing all elements of the voting population, the fact that the basic findings regarding party identification, leadership and voting behavior have been remarkably stable across three presidential elections provides some degree of confidence that future research will find similar results if the study is expanded to include a more diverse demographic and geographic sample. Further, all the analyses include registered voters and those who voted in the election, a departure from most previous studies that have focused largely on voting preferences and not on actual voting behavior measured after the election. We included a SDRS scale with five items to allow us to examine the extent to which social desirability in responding was present. A small percentage of the respondents (less than 18%) responded in an extreme manner. The correlations of each of the items with our main study variables were not significant, indicating that reports are not inflated by biased responses. We also employed a number of corrections to partial out the effect of bias on our results, including separating scale items and partial correlation adjustments (Podsakoff et al., 2003). Harman's one factor test indicated that no single general factor was underlying our data (Podsakoff et al., 2003). Polls have been shown to be more accurate close to an election (Crespi, 1988). Our preliminary results indicated that the correlations between intent to vote and reported vote were .74 and .77 for Bush and Kerry respectively. Thus, it appears that the two week post election window in this study allowed the respondents to accurately recall their decisions and also captured that portion of the sample population that changed their voting inclinations as the polling date neared. Studies that examined charisma and the presidency with respect to the post 9/11 era indicated that the rhetoric of President Bush had become stronger and his charisma had increased in the eyes of the U.S. public, with increasingly favorable opinions regarding the president's leadership (Bligh et al., 2004). It therefore suggests that voters may have been clear in their choice for president far in advance of the 2004 election. 7.3. Implications for future research Future research might examine the extent to which leadership ratings of incumbent presidents rise and fall in conjunction with the popularity ratings and the extent to which this affects the way that the electorate votes. For instance, shortly after the Gulf War, President George H. Bush's job approval ratings were at an all time high. However, by the 1992 elections, his leadership ratings had declined significantly due to the attribution that he failed to stimulate a sluggish economy. Shortly after combat operations were deemed over in Iraq, positive public opinion of George W. Bush's presidency approached unprecedented levels and he received accolades in the press and from the public for being a leader with a clear vision, the necessary decisiveness to wage a tough war on terrorism, and empathy for the negative consequences for some of his decisions. This was not the case a few months before the 2000 election when his legitimacy and fitness for the position were being questioned. This was also true in the period immediately following the very closely contested 2000 election which culminated in election recount fiascos. A question that was raised by political pundits during the 2004 election was whether election history (with George H. Bush) would repeat itself with the current president (George W. Bush) in the face of daily attacks in Iraq that resulted in the deaths of coalition forces and Iraqis on a regular basis and also the uncertainty of the economy. President Bush has repeatedly expressed the desire to be perceived as a highly transformational leader who has strong values, a vision for the country and is willing to take bold decisive action. Clearly Bush appeared decisive in comparison to his opponent (Riggio, 2004). It may well be that a platform of perceived decisiveness and transparent values, rather than a position on any single political issue, is what propelled Bush to victory in the final analysis, despite the misgivings of almost half the electorate. Bush was also helped by the lack of a strong challenge from the Democratic candidate to change the status quo. However, history may judge his decisiveness to be symptomatic of a stubbornness and unwillingness to change in the face of evidence that his strategies were not particularly successful. This may explain the steady decline of his approval ratings and the defeat of his party in the 2006 mid-term elections. What was once perceived as strength in the wake of the September 11 attacks is now being perceived as stubbornness and fanatical adherence to a failed policy. Simonton (1993) noted that since few voters know the candidates personally they may rely on inferences in judgments about the traits held. Most voters make inferences about personality, policy, and leadership based on their perceptions and limited familiarity with specific issues although this may be changing in the age of “Google” and “YouTube” where every controversial comment made by a candidate is reported and analyzed. While, there may be potential problems inherent in studying leadership at a distance, studies have been conducted that support this methodology (Simonton, 1993). Voters are routinely called upon to elect governmental representatives such as the presidents without having had direct contact and this is the case for the majority of the voting population. Between the extended exposure and detailed analysis provided by 24-hour news networks and the salience of the critical issues being decided in the 2004 U.S. presidential election, it is likely that voters were well informed in their voting decision. It is possible that with the extensive access to information via the media and the internet that characterize modern elections that voters are better able than ever before to assess their candidates as if they were in direct communication. It may also be argued that voters felt far more comfortable with the candidate whom they felt they knew (President Bush) and trusted him to lead the nation through difficult times rather than the challenger (John Kerry) who had yet to be tested in a national crisis. The large voter turnout reported in the news suggests that the issues in the 2004 election drove voting behavior over and above party affiliation. It also appears that voters closely identified with the candidates. Values appeared to play a major role in the voting decision in 2004. We asked respondents to indicate on the post-election surveys the top five issues that drove their decision to elect one candidate over the other. Approximately one-third identified the values of the candidate or “integrity” as a key factor. Approximately 12% identified the war in Iraq as an important factor and 10% identified “terrorism” as an important factor. Previous research on leadership supports arguments that personal characteristics of followers are antecedents of leadership evaluations (Bass, 1998 and Dvir and Shamir, 2003). In future research we hope to include personality characteristics from the Big Five in our model to help explain how respondent characteristics influence leadership attributions. This paper is more narrowly focused on contextual factors such as crisis and behavioral style as a reflection of decisiveness but we hope to expand the model in future research on presidential candidates and CEOs. Independent ratings from various sources might be also useful in future research to evaluate leadership attributes. Future research will need to build on the moral values and contextual elements and include more personality characteristics to fully understand the phenomenon of electing the U.S. President. It is possible that certain personality characteristics that are considered significant at a particular time for presidential leadership emergence may not be as significant at another time. The 2008 presidential election which had no incumbent candidates in the running for the first time in several decades provides a rich opportunity for examining the factors that facilitate the emergence of leadership at the national level. Overall, the series of studies of leadership and voting behavior carried out during the last three elections have provided important additive and incremental knowledge that provide a better understanding of what influences the American voter as they elect arguably the most powerful leader in the world every four years. Currently, the authors are analyzing data collected during the 2008 elections.