Synthesis Paper

When confidence is detrimental: Influence of overconfidence on leadership effectiveness

Amanda S. Shipman ⁎, Michael D. Mumford University of Oklahoma, OK, USA

a r t i c l e i n f o a b s t r a c t

Available online 8 June 2011 Confident leaders are seen as competent and capable by others. However, excessive amounts of confidence may be detrimental to a leader's performance. The purpose of the current study was to identify indicators of overconfidence and examine the influence that overconfidence has on certain kinds of leader performance. Results indicated two elements of overconfidence: seeing deficiencies and expectations of positive outcomes. Low levels of confidence associated with seeing many deficiencies, is beneficial to performance in leader planning and vision formation. However, high levels of confidence associated with expectations of positive outcomes, are related to effective vision statements. Implications for results are discussed.

© 2011 Elsevier Inc. All rights reserved.

Keywords: Overconfidence Self-confidence Self-efficacy Leader cognition Planning Vision

Self-confidence is traditionally viewed as valuable to leadership performance. Leaders who are confident welcome challenges and set difficult goals (Luthans, Luthans, Hodgetts, & Luthans, 2001). Not only is confidence said to help leaders take the risks needed to persevere in working towards their visions and objectives (Black & Porter, 2000; Northouse, 1997), but a leader's confidence increases the follower's willingness to work toward the leader's objectives (Luthans & Peterson, 2002). The question arises as to whether this leader characteristic is always beneficial. Confidence, when in excess may be the underlying cause for leaders making poor decisions, continuing with failing plans, and ignoring obvious flaws. The repercussions of these actions are more than inconvenient, but quite detrimental as seen throughout history in incidents like the Bay of Pigs debacle and Napoleon's March on Moscow. Given the potentially detrimental consequences of overconfidence in leadership, exploring this topic further is warranted.

Although leadership researchers have not focused specifically on excessive confidence, related topics have been explored in the literature on destructive leadership. Leadership hubris is one such area that has received some attention. Hubris reflects individuals with not only excessive self-confidence, but also puffed up egos and highly positive, unrealistic self-evaluations (see Judge, Piccola, & Kosalka, 2009). Thus, hubris includes overconfidence, but it also extends beyond just overconfidence into constructs like pride and self-worth. Owen and Davidson (2009) described how certain American presidents have demonstrated hubristic behaviors, including Kennedy with the Bay of Pigs fiasco, Nixon with the Watergate scandal, and Bush with the decision to invade Iraq.

Destructive leadership researchers have acknowledged that even leader traits that are considered beneficial can become destructive. Along these lines, theories of charismatic leadership were originally proposed to describe a positive and effective leadership style (e.g., Bass, 1985; Bass & Avolio, 1994; House & Howell, 1992; Yukl, 2002), however, charisma can have a ‘dark side’ where it becomes destructive and ineffective (Conger, 1990; Howell & Avolio, 1992; Padilla, Hogan, & Kaiser, 2007). The destructive charismatic leader may pursue personalized objectives at the expense of others, identify so heavily with his/her vision that gross errors and miscalculations are made, and manipulate others to maintain control (Conger, 1990). Deluga (2001) proposed that destructive charismatics share a similar feature as Machiavellians, which is self-confidence. Machiavellianism is another destructive leadership approach that involves pursuing one's own self-interests using manipulation and deceit (Christie & Geis, 1970).

The Leadership Quarterly 22 (2011) 649–665

⁎ Corresponding author. E-mail address: [email protected] (A.S. Shipman).

1048-9843/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.leaqua.2011.05.006

Contents lists available at ScienceDirect

The Leadership Quarterly

journal homepage: www.elsevier.com/locate/leaqua

Narcissism is another destructive leader characteristic related to confidence. Narcissism is a characterized by inflated views of oneself and excessive self-confidence (Campbell, Goodie, & Foster, 2004; Watson & Biderman, 1994). While there can be negative outcomes associated with hubristic leaders, charismatic leadership, Machiavellianism, and narcissism these characteristics can in fact provide certain benefits to individuals by drawing them to leader roles, encouraging others to look to them for direction and guidance, and giving them perseverance in pursuing risky projects (Bochner, di Salvo, & Jonas, 1975; Brunell et al., 2008; Geis, Krupat, & Berger, 1970; Gervais, Heaton, & Odean, 2007; Glad, 2002; House, 1977; Okanes & Stinson, 1974; Rim, 1966; Trice & Beyer, 1986; Tucker, 1970).

Given that self-confidence is a feature of hubristic leaders (Hayward & Hambrick, 1997), charismatic leadership (Conger & Kanungo, 1987; House, 1977; Shamir, House, & Arthur, 1993), Machiavellianism (Jameson, 1945; Martin & Sims, 1956; Pfiffner, 1951), and narcissism (Campbell et al., 2004; Post, 1993) and knowing that destructive leadership is possible with these types of leaders, it appears that confidence is linked to destructive leadership under some circumstances. The destructiveness or the ‘dark side’ of confidence may be overconfidence. Leaders who are on the ‘dark side’ of confidence may be prone to destructive leadership patterns that impede their leadership effectiveness.

While related destructive leadership characteristics have received some attention by leadership researchers, a specific focus on overconfidence is lacking. In this respect, a clear conceptualization of overconfidence has not been proposed. In order to develop a clear conceptualization, relevant literature on overconfidence must be explored to inform predictions as to what the dimensions of overconfidence are within the context of leadership. The first purpose of this study is to identify these indicators by reviewing literature on confidence, overconfidence, and related traits and behaviors associated with destructive leadership.

The influence that overconfidence has on leadership effectiveness may be detrimental, but this relationship has not been explored in the literature. While leaders who exhibit confidence to others tend to be perceived as more effective than leaders who do not, the question arises as to whether leaders are confident when engaged in intrapersonal, cognitive activities. In order to answer this question, a distinction must be made between perceptions of leadership effectiveness and actual leadership performance (Kaiser, Hogan, & Craig, 2008). Therefore, the current study considers it critical to measure leadership effectiveness without relying on perceptions in an effort to better explain the true relationship between overconfidence and leadership effectiveness. In sum, the second purpose of the current study is to explore what influence overconfidence has on leadership effectiveness in a purely cognitive activity as well as an in an activity involved in interpersonal influence.

1. Leader self-confidence: defining and measurement

While self-confidence in leaders is considered helpful (Bass, 1990; Yukl, 2002), the clarity of the concept is rather vague. Generally speaking, self-confidence involves belief in one's capability to be successful and self-perceptions of competence in knowledge, skills, and abilities. Put more simply, confidence involves a judgment as to whether an individual believes they can do something (Hollenbeck & Hall, 2004). Self-efficacy, on the other hand, is described as an estimate of one's capability to perform successfully in order to obtain one's objectives (Bandura, 1997; Gist & Mitchell, 1992). Clearly self-confidence and self-efficacy are related concepts. The general trend appears that self-confidence and self-efficacy can be considered one and the same (Hannah, Avolio, Luthans, & Harms, 2008; Yukl, 2002), therefore, for the purposes of this study they are considered synonymous.

Researchers have taken different approaches to conceptualizing leader self-confidence or self-efficacy. Popper, Amit, Gal, Mishkal-Sinai, and Lisak (2004) described self-confidence as composed of several personality traits: internal locus of control, low neuroticism, and general self-efficacy. Another trait self-confidence is said to be related to is self-esteem (Blaine & Crocker, 1993), where those with low self-esteem tend to have lower self-confidence. These researchers describe confidence in terms of the related personality traits. While self-confidence may be correlated with certain personality traits, this relationship may be relatively weak since leader efficacy tends to be specific to the leadership domain (Chemers, Watson, & May, 2000). More recently, Hannah et al. (2008) took a different approach and described the areas that a leader may be confident: thought, action, self- motivation, and means. Palis and Green (2002) took a behavioral approach and proposed a three-part taxonomy of behaviors associated with a confident leader: direction-setting, gaining commitment, and overcoming obstacles. Thus, these behaviors are indicators of confidence in a leader.

Although there are variations for understanding self-confidence, the measurement is often relatively consistent. Researchers rely heavily on self-report measures (e.g., Paulson & Betz, 2004) where leaders indicate level of certainty in their ability to accomplish a given task or act as a leader in general. In Chemers et al. (2000) study leaders rated how confident they were in different leadership skills such as communication, initiative, and influence. Additionally, leaders rated how confident they were in their general leadership ability (e.g., “I am confident of my ability to influence a group I lead”). While self-reports are useful in some settings, a different approach might be more appropriate to assess a leader's overconfidence. An actual measure of overconfidence based on indicators within the leader's activities may limit the social desirability influences associated with self-reports.

2. Conceptualizing Leader Overconfidence

To develop a definition of overconfidence, recall the simple description of confidence suggested by Hollenbeck and Hall (2004), that is, confidence is a judgment of whether one is capable to do something. Those who are confident judge themselves to be capable whereas those who are not confident have low estimates of their capabilities. The question arises as to how this definition of confidence translates to overconfidence. Given that confidence involves a judgment it is possible to be more or less accurate in this judgment. Overconfidence comes from inaccuracies in one's confidence judgment. Thus, overconfidence can be defined as

650 A.S. Shipman, M.D. Mumford / The Leadership Quarterly 22 (2011) 649–665

overestimation in judging one's ability to do something. This definition is consistent with how researchers in judgment and decision making have conceptualized overconfidence. Overconfidence here is typically assessed by asking participants to estimate how well they will perform a given task in the laboratory and then comparing this estimate to actual performance. If confidence scores are greater than performance scores the participant is described as overconfident (see Fischer, 1982). Thus, predicting one will perform better than one actually does is considered overconfidence.

Bearing in mind the proposed definition of overconfidence, the relationship between overconfidence and other leadership concepts becomes clearer. First, narcissists have inflated perceptions of themselves compared to more objective measures in different domains such as task performance, physical attractiveness, or personality traits (Robins & Paulhus, 2001). In this regard, narcissists tend to overestimate themselves (Campbell et al., 2004) or are overconfident. A second related area of leadership research is self-other agreement. This line of research generally argues that accurate self-evaluations are helpful to leaders, while overestimation of oneself is not. Leaders who appraise themselves closely to how subordinates and supervisors appraise the leader tend to have higher performance evaluations than those who overestimate themselves (Atwater & Yammarino, 1992; Yammarino & Atwater, 1997). Even though these studies do not conceptualize overestimation as overconfidence, given the proposed definition of overconfidence, overestimation can be seen as a form of overconfidence in terms of oneself, similar to narcissism.

To further elaborate on what overconfidence entails and subsequently measure this construct, it is necessary to consider what an overconfident leader might do or consider more specifically how a leader might be overconfident. Yukl (2002) described generally what leadership overconfidence entails: denial of potential flaws and overly optimistic beliefs. These two elements, while somewhat general, present a useful preliminary conceptualization of overconfidence.

First, overconfidence might entail denial of potential flaws. Dorner and Schaub (1994) describe how individuals who are seeking to maintain feelings of competence may make the mistake of avoiding potential errors or problems. Along similar lines, researchers in cognitive psychology suggest that overconfidence occurs when people choose to ignore disconfirming evidence in favor of focusing on confirming evidence (e.g., Klayman & Ha, 1989; Koriat, Lichtenstein, & Fischhoff, 1980; Wason, 1968). Ignoring potential problems or negative outcomes might be manifested in several different ways. More specifically, being overconfident in terms of denying potential flaws may make a leader predict few or no problems associated with their plans or the problems the leader predicts may be inconsequential or not given appropriate weight in subsequent decision making. Additionally, leaders may fail to consider how their followers may react poorly to proposed ideas. These types of specific behaviors may all fit within the broader leadership overconfidence dimension of failing to see deficiencies.

Behaviors consistent with this overconfidence dimension have been attributed to destructive leaders. These types of behavior are characteristic of hubristic leaders, who are said to become unwilling or unable to see changes in the environment (Kroll, Toombs, & Wright, 2000). Narcissists may ignore or fail to consider problems that are evident in reality but do not coincide with their idealized views of themselves or their ideas (Kohut, 1966). Similarly, destructive charismatics attempt to cover up mistakes or failures and also seek to limit contradictory or disconfirming evidence (Yukl, 1999). These points led to the first hypothesis.

Hypothesis 1. Leaders who are overconfident will fail to see deficiencies and problems.

The second type of overconfidence suggested by Yukl (2002) is overly optimistic beliefs. Researchers have found that a typical confidence error is overestimation of the probability of obtaining desirable outcomes (Bazerman & Neale, 1983; Bazerman & Sondak, 1988; Neale & Bazerman, 1985; Thompson & Hastie, 1990). The distinction between confident leaders and overconfident leaders in terms of this dimension is that overconfident leaders have an excessive amount of positive expectations. These expectations could include predictions of unrealistic positive reactions from others, overly optimistic beliefs about the implementation of one's ideas, or excessive pride with one's work.

Overconfidence of this type can be seen in hubristic leaders, where expectations of positive outcomes shade into arrogance. In this respect, hubristic executives are more likely to take on overpriced mergers because of their unrealistic positive expectations (Kroll et al., 2000). Destructive charismatic leaders are susceptible to having excessive positive expectations. These leaders tend to exaggerate their positive achievements when pursuing their own self-interests (Yukl, 1999). Even though Luthans et al. (2001) argued that optimism is beneficial to leaders, these authors still make the caveat that excessive optimism could be detrimental to leader performance. Being certain and positive about one's decisions and judgments even when they are poor quality is a characteristic seen by narcissistic leaders (Hogan, Raskin, & Fazzini, 1990). Taking these points together the second hypothesis is proposed.

Hypothesis 2. Leaders who are overconfident will express expectations of positive outcomes.

3. Leader performance and overconfidence

The way in which leadership performance is operationalized varies widely. Thus, it is useful to clarify what aspect of performance will be the focus for a given study. The domains of performance of interest in the present study are planning and vision formation. Planning represents a cognitive activity critical to leadership effectiveness and vision formation represents a cognitive activity that subsequently is used to influence others. Thus, a vision is used in interpersonal domains of a leadership influence.

Before delving more deeply into these leader activities and exploring how they will likely be affected by overconfidence, it is useful to discuss how researchers have described the influence of confidence on leader performance. Typically leadership studies of confidence explore differences between high and low confidence, concluding that high confidence is better than low confidence for leader performance. For example, self-confidence is related to ratings of leader potential (Chan & Drasgow, 2001) and ratings of

651A.S. Shipman, M.D. Mumford / The Leadership Quarterly 22 (2011) 649–665

performance as assessed by peers and supervisors (Luthans & Peterson, 2002; Robertson & Sadri, 1993). The benefit of leader self- confidence to performance should not be discounted. However, these studies tend to look at leader performance interpersonally and as viewed by others, such as perceptions of leader potential (Chan & Drasgow, 2001), follower engagement (Luthans & Peterson, 2002), and peer ratings of leader effectiveness (Chemers et al., 2000). In one study by Hendricks and Payne (2007) leader confidence was related to peer's perception of leader effectiveness but not the objective performance on a leader task. It is likely that confidence helps in interpersonal domains to increase followers’ certainty in the leader's abilities.

However, the leader's role is not exclusive to interpersonal domains. Leadership scholars recognize that being a leader requires complex cognitive and social problem-solving skills (Yukl, 2002). In fact, the cognitive activities leaders engage in are a critical aspect of their performance (Mumford, Friedrich, Caughron, & Byrne, 2007). In this respect, many models of leadership include cognitive activities (e.g., Conway, 2000; Dennison, Hooijberg, & Quinn, 1995; Hooijberg, Hunt, & Dodge, 1997; Hunt & Phillips, 1991; Mumford et al., 2007; Mumford & Connelly, 1991; Quinn, 1988).

To determine how overconfidence might influence the leader's cognitive activities, it is useful to consider leader cognition in more detail. Before leaders can exercise influence attempts they engage in sensemaking, where they attempt to understand a crisis or turbulent situation by applying some mental model that helps in determining how to solve the key problems and obtain desired goals (Drazin, Glynn, & Kazanjian, 1999; Holyoak & Thagard, 1997; Largan-Fox & Code, 2000). In working through these mental operations leaders develop problem solutions and develop plans. Planning represents mentally simulating future actions to be taken (Noice, 1991; Xiao, Milgram, & Doyle, 1997) and is critical to leader performance (Marta, Leritz, & Mumford, 2005; Miller & Cardinal, 1994). The influence that overconfidence has on developing plans has not been studied in the literature. Xiao et al. (1997) found that leaders who became overly optimistic when forecasting, or predicting future outcomes of ideas, failed to develop backup plans or alternative solution pathways. Mumford et al. (2007) suggested that this optimism can lead to errors in leader problem-solving. Specific problems with overoptimism might include creating unrealistic expectations, failing to take necessary steps to reach goals, and deceiving oneself about the amount of progress made toward accomplishing goals (Luthans et al., 2001). Taking these points together led to the third hypothesis.

Hypothesis 3. Leaders who indicate overconfidence associated with expectations of positive outcomes will develop less effective plans.

Another error in cognitive processing involves failure to consider information that discredits or disconfirms ideas (Dorner & Schaub, 1994). This evaluation of ideas is considered critical to creative problem-solving (Halpern, 2003; Wakefield, 2003). By scrutinizing and finding flaws in one's ideas individuals can discard poor ideas (Simonton, 1999) and improve workable ideas (Baer, 2003; Gorman & Plucker, 2003). Thus, ignoring deficiencies in one's ideas is likely to result in errors and subsequently poor planning performance. This idea led to the fourth hypothesis.

Hypothesis 4. Leaders who indicate overconfidence associated with failure to see deficiencies will develop less effective plans. Another element of leader performance is developing vision statements. A vision represents a leader's idealized image of the

future that can be used to influence others (Conger, 1999). An effective vision will be appealing to followers emotionally (Bono & Ilies, 2006). More specifically, effective visions appeal to followers on an emotional level in order to induce motivation to commit to the leader's objectives. In this respect, leaders use visions to exercise influence attempts on followers. By articulating an effective vision, leaders can elicit not only follower motivation (Sosik, Kahai, & Avolio, 1999), but also follower satisfaction (Dumdum, Lowe, & Avolio, 2002). Thus, effective visions are motivational and elicit positive responses from followers. These positive reactions could be gained by making visions memorable or making follower's feel included in the vision. Given that visions are used to influence others, this element of leader performance is more interpersonally based and social in nature unlike the cognitive activity of planning. For this reason, the nature of the relationship between confidence and visions might operate somewhat differently.

To maintain a future-oriented outlook needed to develop a vision and convince followers that actions will lead to positive outcomes, Popper and Mayseless (2006) argued that a leader needs optimism. In this respect, optimism has been demonstrated to be a predictor of a leader's ability to form a vision that inspires emotionally driven action (George, 2000; Wunderley, Reddy, & Dember, 1998). Using emotional contagion, leaders can increase emotional arousal in their followers which increases their confidence in the leader (Hannah et al., 2008). In this respect, emotional contagion occurs as leaders can transmit positive emotions to followers (Bono & Ilies, 2006; Naidoo & Lord, 2008). Norman, Luthans, and Luthans (2005) proposed that through emotional contagion a leader's hope can be transmitted to inspire hope in the followers which encourage them to persevere despite adversity or risk. Taking these points together the positive element of overconfidence seems likely to benefit the emotional appealing aspect of the effective vision and the following hypothesis is offered.

Hypothesis 5. Leaders who indicate overconfidence associated with expectations of positive outcomes will develop more emotionally appealing vision statements.

On the other hand, visions not only need to invoke an emotional response, but also need to induce commitment to the leader's cause. In order to commit to the vision, people must believe that the vision is attainable (Hollenbeck & Klein, 1987; van Knippenberg, 2000). In this regard, effective visions should describe appropriate response to the situation at hand (Hunt, Boal, & Dodge, 1999). Effective visions will describe the leader's vision in a feasible way, such that followers believe it is attainable. Additionally, the vision ought to make followers believe the leader's ideas are useful and high quality. In creating a vision with this

652 A.S. Shipman, M.D. Mumford / The Leadership Quarterly 22 (2011) 649–665

practical element, a leader who sees problems or deficiencies might have more realistic expectations to communicate to followers. Carnevale and Pruitt (1992) suggested that maintaining a realistic and modest perspective makes a leader more likely to consider all the parties involved and develop creative solutions to problems. Maintaining a realistic perspective is not easy for destructive leaders and this hurts their performance. Along these lines, Conger (1990) suggested that the destructive leader may fail to develop effective visions. More specifically, the destructive charismatic leader develops poor visions when they have the inability to see potential problems. These points led to the last hypothesis.

Hypothesis 6. Leaders who indicate overconfidence associated with failure to see deficiencies will develop less practical vision statements.

4. Method

4.1. Sample

Participants in this study included 159 undergraduate students (71 males, 88 females) with an average age of 18.7 years. Students participated to earn credit in their introductory psychology course. To sign up to participate in the study, students accessed a website containing brief summaries of available studies. Similar to other subject pools of this nature, students had scores one quarter standard deviation above national average for freshmen at four year institutions.

4.2. Procedure

Participants took part in the study in university classrooms. Participants completed a paper-and-pencil in-basket task using the leadership scenario developed by Strange & Mumford (2005). See Fig. 1 for the full scenario. The scenario asks participants to assume the role of principal for an experimental high school with the objective of improving standardized tests scores. This scenario provides a novel, ill-defined problem in which there could be multiple solution pathways, conditions that provide participants with the opportunity to engage in creative problem-solving (Mumford, Whetzel, & Reiter-Palmon, 1997). Solving these types of problems is critical to leadership performance (Mumford, Zaccaro, Harding, Jacobs, & Fleishman, 2000).

Written instructions informed participants they would be working with an external consulting company that would help them develop their ideas for the school. The consulting company provided a way to instruct the participants as to what they needed to do as leader in the scenario. All of the instructions and exercises participants completed were under the guise of the consulting

You have been appointed as the Principal of the state’s experimental school in Tulsa, Oklahoma called “Oklahoma Excel.” The school is part of a national study to increase achievement in schools in the United States. Funding for the OE School will be allotted in accordance with a federal grant distributed by the National Education Agency to each State Department of Education. Each state is awarded funding for one experimental school, and Tulsa’s OE School is Oklahoma’s representation in the national study. The goal of each experimental state school is to develop and implement a new type of educational program that increases students’ academic performance. At the end of the 2007-2008 school year, OE will be evaluated in reference to the students of the other states’ experimental schools as well as in relation to the students of traditional Oklahoma public schools.

This evaluation of the students in the experimental schools will be based on improvement of the students in the schools. Each student will take a pre-test over material selected by the National Education Agency at the beginning of the school year. This will assess the increases in academic performance for each experimental school. These tests will be administered in all of the experimental schools, and the improvement scores will be compared across students of all the states. The material on the test will be benchmarked by the National Standards of Education General Guidelines (for example, all students should read grade level).

The evaluation of OE students compared to other students in Oklahoma will be based on scores of the Oklahoma Standardized Test. All students in Oklahoma are required to take this test, and the material covered on it is general. It assesses writing skills, reading comprehension, mathematic skills, and analytical skills. There are also subtests on sciences, social studies, geography and a foreign language component that assesses fluency. This test is essentially how OE students are compared to students in traditional schools in Oklahoma.

After these comparisons to other states’ experimental schools and other Oklahoma traditional public schools, the National Education Agency will rank the most successful states in terms of experimental school accomplishment. The states with the most successful experimental schools will receive additional federal funding for the next school year in order to spread the new curriculum around the state for comprehensive state scholastic improvement.

Therefore, the Oklahoma State Board of Education is hopeful for dramatic improvements of students in your OE School. You are feeling additional motivation for success because, last year, students in Oklahoma Public Schools ranked 47th nationally in academic performance on Standardized Tests. The state is also currently ranked 49th in funding for education. With these poor rankings in mind, you are determined to give students at OE the best chance at success. Doing so would lead to exciting effects in Oklahoma and increase our state’s national standing.

OE will be a secondary school with students of grades 9-12. You have a projected enrollment of 400 students from varied ethnic backgrounds (73% Caucasian, 13% Native American, 10% African American, 3% Hispanic, and 1% Other). Also, a principle concern of yours will be to make sure that your teaching method helps members of special populations, including gifted students and academically disabled students. Since the State Department of Education is so interested in OE, it is willing to provide maximum support. This includes providing enough teachers for a 20:1 ratio of students to instructors. Also, they are willing to pay the teachers above average salaries. Because of these optimal teaching conditions, you will be able to recruit high-caliber instructors who are motivated to make your school a success.

Your first task as Principal of OE is to generate a plan on how to “achieve academic excellence.” This plan should include a number of elements including teaching strategies, process improvement ideas, special activities or programs, etc. One of your priorities is to develop some teaching method ideas for the school. To date, you have material that a team of teachers has gathered for you. This consists of educational models that have been successful in the past. You should review this material, and from that, develop a new plan for you school’s method of teaching. Your job is to design a school instructional program that will encourage maximum proficiency of general education requirements from students at OE.

Fig. 1. Task description.

653A.S. Shipman, M.D. Mumford / The Leadership Quarterly 22 (2011) 649–665

company helping the participant assume the leadership role and was solely through a packet of instructions and exercises (i.e., participants did not interact with an actual person). After reading information about the problem scenario, participants received instructions from the consulting company to generate ideas and were given one page to do so. Next, participants revised their ideas using three prompter questions again through instructions from the consulting company, one page given to each question. Fig. 2 provides the prompter questions as seen by participants. Responses to these open-ended questions served as the basis for assessing overconfidence. After revising their ideas, participants generated a 1–2 page final plan and then lastly a 1–2 page vision for the school. The vision was operationalized as a speech to be given to parents, faculty, and students. This operationalization for visions has been shown appropriate and effective in other leadership studies (Ellis, 2001; Shipman et al., 2010; Strange & Mumford, 2005). This task took participants 1.5 h to complete.

4.3. Control measures

During the second half of the study participants completed a battery of covariate measures relevant to confidence and leader cognitive problem-solving including intelligence, personality, narcissism, self-esteem, and previous leadership positions. Lastly participants completed a measure of task engagement. Intelligence tends to be related to cognitive problem-solving, where more intelligent individuals tend to perform problems-solving activities more effectively (Ackerman & Humphreys, 1990). To measure intelligence, the Wonderlic Personnel Test (Wonderlic, 1992) was administered. This 50-item test asks analogical reasoning questions and has a split-half reliability above .70. Additional evidence bearing on the validity of this measure is provided by Bell, Lassiter, Leverett, and Matthews (2002), Dodrill and Warner (1988), McKelvie (1989) and Wonderlic (1992).

The Big 5 personality traits (e.g., openness, conscientiousness, extraversion, agreeableness, and neuroticism) have been shown related to leader self-efficacy or confidence (Hendricks & Payne, 2007). In fact, some researchers describe emotional stability as an underlying component of confidence (Judge, Erez, Bono, & Thoresen, 2003; Judge, Locke, & Durham, 1997; Popper et al., 2004). In this respect, individuals with low levels of emotional stability are likely to have decreased functioning when faced with stressful and uncertain situations (Sarason, 1975). Extraversion, agreeableness, and openness have been found to be positively related to leader efficacy (Hendricks & Payne, 2007). To measure the Big 5 personality traits, the Goldberg Adjective Checklist was used (Goldberg, 1992). This measure presents 100 adjectives that participants rate on a 9-point scale as to how accurately each adjective describes them. The internal consistency for each subscale is around .80. Further evidence bearing on the validity of this measure has been collected (see Becker, Billings, Eveleth, & Gilbert, 1997; Conway & Peneno, 1999; Goldberg, 1992; Reysen, 2005; Saucier, 2002).

Client OES Principal Report Idea Revision: Part III

# 3 Now that you have really fleshed out your ideas, what do you think you will really need to remember when you try to implement or execute your ideas? In other words, what things do you need to consider or be aware of when actions are taken to carry out your plan?

Client OES Principal Report Idea Revision: Part II

#2 Consider the outcomes you predicted on the last page. Will your ideas need to be adjusted or changed because of these predicted outcomes? Do you feel certain your ideas will work? Why or why not? You may want to consider if anything will stand in your way, how people will react, and how long it is going to take to get the school going.

Client OES Principal Report Idea Revision: Part I

#1 What do you expect would be the outcomes of your ideas? Predict what you think would realistically occur if you ideas were implemented. Consider the effects of your plan and important outcomes. Include all details of what you think will really happen once your ideas for OES will be made possible.

Fig. 2. Idea revision questions to assess overconfidence.

654 A.S. Shipman, M.D. Mumford / The Leadership Quarterly 22 (2011) 649–665

As has been discussed previously, narcissistic individuals tend to have inflated levels of confidence (e.g., Campbell et al., 2004). To measure narcissism the Narcissistic Personality Inventory (NPI) was used (Raskin & Hall, 1979). The NPI includes pairs of statements that participants must choose between where each pair offers a narcissistic response and a neutral response (e.g., Superiority is something that you acquire with experience/Superiority is something you are born with; I prefer to blend in with the crowd/I like to be the center of attention). The internal consistencies of the underlying scales are around .8 and .7. Recent research has provided validation evidence for this measure (Corry, Merritt, Mrug, & Pamp, 2008).

Self-esteem is the last personality variable used as a covariate in this study. Self-esteem is argued to be related to confidence, such that individuals with more self-esteem tend to be more confident (Popper et al., 2004). To measure self-esteem, the Rosenberg Self-Esteem scale was administered to participants (Rosenberg, 1965). This 10-item measure has an internal consistency of .77. It asks participants to rate on a scale of 1–5 how much they agree that statements describe themselves (e.g., I am able to do things as well as most other people; I feel I have a number of good qualities). See Byrne (1996) and Gray-Little, Williams, and Hancock (1997) for additional validation evidence for this measure.

Participants completed a demographics sheet in which they indicated the number of previous leadership positions they had held. Those who have had prior leadership experience are not only more likely to emerge in leadership positions (Atwater, Dionne, Avolio, Camobreco, & Lau, 1999), but are also more likely to develop leader efficacy (Popper & Mayseless, 2006). A final control measure was engagement in the experimental task. This measure included four questions to be completed on a 5-point Likert scale indicating how much participants engaged in the task (e.g., Did you find the task interesting? When completing the task did you want to generate a good solution?). Participants engaged in the task are likely to perform at higher levels than those who are not engaged.

4.4. Overconfidence, plans, and visions

Thirteen variables were generated to indicate overconfidence based on the hypothesized elements: expectations of positive outcomes and seeing deficiencies. See Table 1 for a full list of overconfidence indicators and operational definitions. These variables included: 1) amount of obstacles discussed, 2) understanding of obstacles, 3) amount of weaknesses discussed, 4) understanding of weaknesses, 5) amount of adjustments discussed, 6) appropriateness of adjustments, 7) amount of positive reactions discussed, 8) legitimacy of positive reactions, 9) amount of negative reactions discussed, 10) legitimacy of negative reactions, 11) confidence, 12) excitement, and 13) optimism. Effectiveness of plans was assessed using two variables: quality and originality. These variables are useful to assess performance in creative problem-solving, which is called for with complex, ill-defined problems (Besemer &

Table 1 Overconfidence variables.

Variable Definition

Amount of obstacles The number of obstacles predicted that could stand in the way of the participant's ideas. Obstacles are things that are outside of the participant's control.

Understands obstacles The extent to which the participant understands the obstacles. Explains logically why obstacles are or are not likely to occur. Demonstrates that obstacles have been carefully considered and whether they could impact ideas.

Amount of weaknesses The number of weaknesses described associated with the participant's ideas. Weaknesses are flaws, drawback, or errors in the ideas (as opposed to external things that could stand in their way).

Understands weaknesses The extent to which the participant understands the weaknesses. Explains logically why weaknesses are or are not going to be a problem. Demonstrates that weaknesses have been carefully considered and whether they could impact ideas.

Amount of adjustments Suggests adjustments or considers potential changes that could be made to ideas. Discusses contingencies (e.g., if this happens then I could do this).

Appropriate adjustments The extent to which adjustments or lack of needed adjustments are justified. Provides well-thought out reasons and logical explanations for needed adjustments or why adjustments should not be needed.

Amount of positive reactions The amount of positive reactions predicted to occur in response to ideas. Legitimate positive reactions The extent to which positive reactions are justified. Positive reactions may not be

expected and this should be justified and appropriate as well. Amount of negative reactions The amount of negative reactions predicted to occur in response to ideas. Legitimate negative reactions The extent to which negative reactions are acknowledged and carefully considered.

Logically justifies why a negative reaction may be likely to occur or why a negative reaction is unlikely. Places some importance on considering negative reactions and how they might impact proposed ideas.

Confidence The degree of confidence or certainty expressed. Excitement The degree of excitement and enthusiasm expressed by the participant about

their ideas. Optimism The degree to which the participant feels optimistic that success is in the future. May

predict success toward themselves, their ideas, or the school.

655A.S. Shipman, M.D. Mumford / The Leadership Quarterly 22 (2011) 649–665

O'Quin, 1999; Ford & Gioia, 2000; Mumford & Gustafson, 1988; Ward, Smith, & Finke, 1999), as is the case in the current study. Lastly, seven variables were used to assess the effectiveness of the vision: 1) inclusiveness, 2) people's reaction, 3) motivational impact, 4) feasibility, 5) utility, 6) emotional appeal, and 7) memorable (Bono & Ilies, 2006; Hunt et al., 1999; Naidoo & Lord, 2008). All of these variables were coded on a 5-point Likert scale where a score of one represented poor quality and five represented excellent quality.

To assess overconfidence, plans, and visions, four psychology graduate students received a 20-hour rater training program. This program began with a familiarization of the rating process. Next, raters learned about the variables of interest through operational definitions and benchmark examples of high, medium, and low scores for each variable. See Figs. 3–5 for example definitions and benchmarks for overconfidence, plans, and visions. After familiarization with the variables of interests the raters practiced making ratings independently. After this practice rating, raters discussed their ratings and were given feedback as to what appropriate ratings would be.

To determine interrater agreement an rwg was calculated for each variable (James, Demaree, & Wolf, 1984, 1993). For overconfidence the interrater agreements were: amount of obstacles (rwg=.76), understands obstacles (rwg=.77), amount of weaknesses (rwg=.80), understands weaknesses (rwg=.65), amount of adjustments (rwg=.74), appropriate adjustments (rwg=.65), amount of positive reactions (rwg=.71), legitimate positive reactions (rwg=.77), amount of negative reactions (rwg=.82), legitimate negative reactions (rwg=.73), confidence (rwg=.82), excitement (rwg=.83), and optimism (rwg=.84). Plan rwg’s were for quality (rwg =.80) and for originality (rwg=.77). Lastly, the interrater agreement for visions were inclusiveness (rwg=.81), people's reaction (rwg=.84), motivational impact (rwg=.82), feasibility (rwg=.78), utility (rwg=.75), emotional appeal (rwg=.78) and memorable (rwg=.82).

4.5. Analyses

A series of analyses were conducted to test the hypotheses. First, correlations were found between study variables. Second, the scores on the overconfidence variables were factor analyzed. By examining correlations among study variables and factor analyzing the overconfidence variables the first two hypotheses about the nature of overconfidence could be examined. To test the influence of overconfidence on performance, a separate multivariate analysis of covariance was used for planning performance and vision performance. In order to treat overconfidence as the fixed factor in the multivariate analyses, scores on each overconfidence factor were categorized based on the distribution of scores into one of three categories: less than negative one standard deviation, between negative one and positive one standard deviation, and greater than positive one standard deviation. For expectations of positive outcomes, high scores indicated overconfidence and for seeing deficiencies low scores indicated overconfidence. The first multivariate analysis of covariance examined the influence of the overconfidence factors on plan, where

Understanding Obstacles Legitimate Positive Reactions - The extent to which the participant understands the obstacles. Explains logically

why obstacles are or are not likely to occur. Demonstrates that obstacles have been carefully considered and whether they could impact ideas.

- Does the participant describe obstacles? If not, do they logically justify why the ideas won’t have obstacles? Does the participant explain the obstacles?

- Does the participant demonstrate that these obstacles have been carefully thought out and considered? Or are the obstacles discounted?

- Are the obstacles logical and reasonable?

- Does the participant discuss how actions may need to be modified depending on the outcomes of other actions?

- The extent to which positive reactions are justified. Positive reactions may not be expected and this should be justified and appropriate as well.

- Does the participant describe expected positive reactions to their ideas? Are these positive reactions justified and well-thought out? Do they make sense?

- Does the participant describe why positive reactions may not be likely? - Are these explanations realistic and legitimate?

1 – No understanding. Does not demonstrate understanding of obstacles. Discounts obstacles as unimportant without considering them at all. Does not anticipate any obstacles or anticipates unimportant obstacles. Offers illogical or poorly thought-out explanation for why there are no obstacles.

1 – Not at all. Positive reactions are not justified at all. Positive reactions are very unlikely and unrealistic.

2 – Very low understanding. 2

3 – Somewhat understanding. Describes obstacles somewhat and obstacles are somewhat logical. Offers somewhat logical explanation for why there are no obstacles.

3 – Somewhat. Describes briefly why positive reactions are expected. Expected positive reactions are somewhat realistic and logical.

4 – Understands obstacles. 4

5 – High understanding. Describes obstacles that are logical and make sense. Elaborates on obstacles demonstrating that these obstacles are well-thought out and have been carefully considered. Offers highly logical explanation for why there are no obstacles.

5 – Very much. Logically justifies why positive reactions are expected. Or logically justifies why positive reactions should not be expected.

Fig. 3. Example overconfidence benchmark rating scales.

656 A.S. Shipman, M.D. Mumford / The Leadership Quarterly 22 (2011) 649–665

Quality Originality - The overall quality of the plan

- Completeness – were the critical issues understood? Was the most relevant issue at hand addressed?

- Coherence – was the response coherent? Was it well thought out and logical?

- Usefulness – is the response actually feasible and appropriate for addressing the problem?

- The extent to which the plan is original and creative

- Unexpected – was the problem approached in a novel, imaginative, unpredictable, or innovative manner?

- Elaborative/descriptive – was a rich answer provided—one that helps the reader to visualize the solution for addressing the problem?

1 – Poor quality. The plan is haphazard and fragmented and does not address any of the key issues; it does not provide key information in a logical manner.

1 – Poor originality. The plan is very predictable and is given in basic terms with no elaboration. The plan only uses bare ideas and is commonplace and ordinary.

2 – Poor to average quality. A few key issues may be addressed; however, a clear plan is still not presented; key parts of the plan are unclear.

2 - Poor to average originality. The plan presents ideas in a slightly unique manner. The plan mostly provides common ideas that do not reflect much elaboration or description.

3 – Average quality. The plan is presented in a logical form; a number of key issues may still be missing or vague, but overall the plan addresses some of the major issues of the problem and is presented clearly and coherently.

3 – Average originality. The plan contains something that makes it different from the typical plan. The approach is original and contains some descriptive information. Description and elaboration are present but not entirely complete.

4 – Average to excellent quality. Many of the key is sues are addressed in the plan and plan is feasible; however, some information may seem unimportant to the plan or is not completely thought out.

4 - Average to excellent originality. The plan contains something that makes it different from the typical solution. The approach is original and contains some descriptive information. Description and elaboration are present but not entirely complete.

5 – Excellent quality. The plan is presented so that is exceptionally coherent and clear and addresses the key issues in a manner that is feasible.

5 – Excellent originality. The plan is exceptionally unique. The participant includes characteristics or details that make the plan unique to him/her. The plan clearly reflects an unexpected understanding approach to the problem and goes beyond the norm and presents new ideas that are highly descriptive.

Fig. 4. Example plan benchmark rating scales.

Motivation Impact Utility - The speech would likely motivate and inspire the audience - The extent to which the speech outlines a plan that sounds useful and

like it would actually work. If the speech has good utility it would likely result in the audience’s involvement in making the ideas in the speech a reality.

1 – Very low motivational impact. People would not be motivated or inspired by this speech.

1 – Poor speech utility. The speech would not cause change, and would not be successful. No focus on encouraging followers to participate in implementation of the plan. Students attending this school would not find social or academic success

2 – Low to moderate 2 - Poor to average

3 – Moderate motivational impact. People would be somewhat motivated or excited by this speech.

3 – Average speech utility. The speech would probably encourage followers to minimally engage in implementing the plan. Students attending this school would find average academic and social success in this school

4 – Moderate to high 4 - Average to excellent

5 – Very high motivational impact. People would be very motivated or inspired by this speech.

5 – Excellent speech utility. The speech would cause a great deal of change, and be very successful. The encouragement would most likely result in active followers. Students attending this school would find high levels of academic and social success

Fig. 5. Example vision benchmark rating scales.

657A.S. Shipman, M.D. Mumford / The Leadership Quarterly 22 (2011) 649–665

overconfidence scores were treated as fixed variables and plan scores were treated as dependent variables. The second multivariate analysis of covariance used vision factor scores as the dependent variables and the overconfidence factors scores again as the fixed variables.

5. Results

5.1. Overconfidence

Hypotheses 1 and 2 predicted dimensions of overconfidence as failure to see deficiencies and expectations of positive outcomes. To address these hypotheses several analyses were conducted. First was factor analysis and next was correlations between study variables. The exploratory factor analysis with direct oblimin rotation on the overconfidence factors yielded three factors with eigenvalues greater than one. The scree plot revealed that a two-factor solution was more appropriate. Running a two- factor confirmatory factor analysis revealed two distinct factors with clean loadings. These two factors were labeled seeing deficiencies, accounting for 39.62% of variance and expectations of positive outcomes, accounting for 23.56% of variance. The variables loading on the seeing deficiencies factor included amount of obstacles (r=.74), understands obstacles (r=.84), amount of weaknesses (r=.81), understands weaknesses (r=.81), amount of adjustments (r=.66), appropriate adjustments (r=.64), amount of negative reactions (r=.69), and legitimate negative reactions (r=.71). The second factor, labeled expectations of positive outcomes had the following variables loadings: amount of positive reactions (r=.79), legitimate positive reactions (r=.71), confidence (r=.65), excitement (r=.72) and optimism (r=.78). This factor solution provided some support for Hypotheses 1 and 2 describing the structure of overconfidence.

To further explore Hypotheses 1 and 2, correlations between the overconfidence factors and relevant constructs were examined. Correlations between study variables can be seen in Table 2. The Narcissism Personality Inventory (NPI) was scored for two previously identified factors: leadership/authority and exhibitionism/entitlement (Corry et al., 2008; Kubarych, Deary, & Austin, 2004). Leadership/authority items generally reflect belief that oneself is an effective leader and has influence over others. Exhibitionism/entitlement scores generally indicate a desire for attention, belief about deserved respect, and belief in one's attractiveness. Although there has been some debate as the factor structure of the NPI, recently researchers have demonstrated better fitting models using the 2-factor solution. An exploratory factor analysis of the current dataset demonstrated further evidence for a 2-factor solution being more appropriate than a 4-factor solution. Both of the leadership/authority and exhibitionism/entitlement were negatively correlated with the seeing deficiencies overconfidence factor. The leadership/ authority factor was approaching significance (r=−.15, p=.06), while the exhibitionism/entitlement factor was significantly related (r=−.19, p=.02). Thus, seeing few deficiencies (e.g., overconfidence) was related to higher narcissism scores. The NPI scores were not significantly correlated with expectations of positive outcomes, the second overconfidence factor.

The observed relationship between the seeing deficiencies overconfidence factor and the elements of narcissism, although marginal in one case, provides some evidence that this factor is an element of overconfidence, providing some additional support for Hypothesis 1. However, the lack of relationship between the expecting positive outcomes factor and narcissism does not provide evidence supporting this factor as an element of overconfidence and thus, does not support Hypothesis 2. However, narcissism and overconfidence are not considered synonymous or the only relevant construct to overconfidence.

Correlations of other theoretically relevant constructs with the overconfidence factors were also examined. Neuroticism was negatively correlated with both seeing deficiencies (r=−.19, p=.01) and expectations of positive outcomes (r=−.16, p=.05). Thus, those who were more neurotic were more overconfident in terms of expectations of positive outcomes, but were also more likely to see deficiencies or were less confident. The relationship between seeing deficiencies and neuroticism was unexpected. Given that previous studies have found neuroticism to be negatively related to confidence, we expected it to also be negatively related to overconfidence. With seeing deficiencies, those who were overconfident (i.e., saw few to no deficiencies) tended to be less neurotic than those who were not confident (i.e., saw lots of deficiencies). However, this overconfidence factor likely involves some element of being risk-aversive or cautious and neuroticism tends to negatively correlate with risk-taking behavior (e.g., Lauriola & Levin, 2001).

As seen in Table 2, there were no other significant correlations between the overconfidence scores and personality variables. Additionally, intelligence, self-esteem, number of past leadership positions, and task engagement were not significantly correlated with either overconfidence factor.

5.2. Effects of overconfidence on plan

Hypotheses 3 and 4 predict a negative influence of seeing deficiencies and expectations of positive outcomes respectively on planning performance. To analyze the effects of overconfidence on planning performance a multivariate analysis of covariance was performed. As described above, overconfidence was treated as a fixed factor by assigning low, medium, or high confidence based on the distribution of scores for each factor. Thus, scores greater than one standard deviation for expectations of positive outcomes were considered overconfident and for seeing deficiencies scores lower than negative one standard deviation were considered overconfident. The results for the MANCOVA are reported in Table 3. All control measures described earlier were included as covariates, but only significant measures were retained. As may be seen, intelligence was the only significant covariate retained related to effectiveness of plans (F(2, 147)=8.31, pb .01, η2=.10). The MANCOVA for overconfidence on plan performance was significant for the seeing deficiencies overconfidence factor (F(2, 148)=11.24, pb .01, η2=.13 ) using Roy's largest root. The

658 A.S. Shipman, M.D. Mumford / The Leadership Quarterly 22 (2011) 649–665

Table 2 Means, standard deviations, and intercorrelations for study variables.

Mean SD 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

1 Intelligence 3.28 .66 1.00 2 Narcissism: arrogance/leadership 3.34 .63 .11 1.00 3 Narcissism: exhibition/entitlement 22.31 5.00 .08 .26 1.00 4 Openness to experience 70.23 28.21 .16 .23 .14 1.00 5 Conscientiousness 26.33 18.46 .00 .21 .04 .28 1.00 6 Agreeableness 132.14 19.93 −.15 .12 −.01 .37 .41 1.00 7 Extraversion 124.82 16.96 .15 .44 .33 .30 .14 .27 1.00 8 Neuroticism 142.23 16.91 −.05 .06 .04 .03 .11 .26 .18 1.00 9 Self-esteem 115.27 20.21 −.05 .28 .21 .12 .28 .37 .28 .31 1.00 10 Leadership experience 5.03 3.89 −.05 .27 .04 .13 .09 .15 .14 .00 .17 1.00 11 Task engagement 13.55 2.51 −.04 .11 −.06 .13 .16 .17 .03 −.01 .10 .05 1.00 12 Overcon: seeing deficiencies 96.41 17.64 −.03 −.15 −.19 −.02 .03 .02 −.14 −.19 −.10 −.02 −.05 1.00 13 Overcon: expecting pos. outcomes 40.31 6.39 −.04 .00 .11 .05 −.05 .07 .10 −.16 .05 .09 .09 −.06 1.00 14 Plan quality 2.00 .56 .20 −.05 −.13 −.05 .05 .10 −.05 −.20 −.04 .06 .08 .30 .13 1.00 15 Plan originality 2.00 .56 .25 −.09 −.15 −.06 −.10 −.07 −.11 −.11 −.17 −.05 .01 .33 .13 .58 1.00 16 Speech appeal .00 .98 .07 .06 −.15 .09 .09 .15 .05 −.07 .07 .16 .11 .19 .15 .36 .28 1.00 17 Speech practicality .00 .97 −.03 .05 −.05 .00 .00 .13 −.04 −.20 .16 .10 .09 .33 .17 .39 .26 .58 1.00

Note. N=159. All correlations with an absolute value greater than .15 significant at p≤ .05, correlations with an absolute value greater than .13 are approaching significance at p≤ .10.

6 5 9

A .S.

Sh ip m an

, M .D . M u m ford

/ Th

e Lead

ersh ip

Q u arterly

2 2 (2 0 1 1 ) 6 4 9 – 6 6 5

univariate analyses revealed that the multivariate significance was due to both quality (F(2, 148)=9.62, pb .01) and originality (F(2, 148)=9.95, pb .01). Cell means had the same pattern for both quality and originality respectively, where those with low confidence generated the best plans (M=3.80, SE=.17, M=3.87, SE=.17) and the worst plans came from those with excessive confidence (M=3.00, SE=.15, M=2.90, SE=.14). These findings provide support for Hypothesis 3.

This effect should be interpreted in light of a significant interaction between both overconfidence factors (F(4, 148)=2.86, p=.03, η2=.07). Here the univariate analysis revealed a significant effect on quality only (F(4, 148)=2.85, p=.03). Cell means indicated that plans highest in quality were generated by those who had both low confidence in terms of seeing deficiencies and low confidence in terms of expecting positive outcomes (M=4.17, SE=.27). Thus, lower levels of confidence help plan performance. This finding is indirectly expected based on Hypotheses 3 and 4, where these hypotheses predicted overconfidence to be related to poor performance, but not specifically that low confidence results in the most effective plans.

However, the lowest performance in plan quality was not from those overconfident on both factors, as expected based on Hypotheses 3 and 4, but rather from those who had moderate confidence in terms of seeing deficiencies and low confidence in terms of expecting positive outcomes (M=2.82, SE=.15). Thus, those who saw some deficiencies in their problem-solving and expected few positive outcomes had the lowest quality plans. These individuals expressed low positivity and saw some deficiencies with their ideas, perhaps these expectations made these individuals less interested in developing a high quality plan detailing these ideas. These results provide partial support for Hypotheses 3 and 4 in that the best plan performance came from those who were not overconfident on both factors, but the worst performance did not come from those who were overconfident on both factors.

5.3. Effects of overconfidence on vision

The remaining Hypotheses 5 and 6 made predictions about the influence of overconfidence on vision performance. In order to simplify the analyses and subsequent interpretation for the influence of overconfidence on vision, the seven vision variables were factor analyzed. An exploratory factor analysis using direct oblimin rotation on the vision variables supported a two-factor solution. Although the scree plot for the two-factor solution was somewhat unclear, the two-factor solution evidenced two theoretically distinct factors with clean loadings. The two factors were labeled inspirational appeal, accounting for 67.77% of variance and practicality, accounting for 15.37% of variance. The first factor, inspirational appeal included inclusiveness (r=.59), people's reaction (r=.71), motivational impact (r=.93), emotional appeal (r=1.02), and memorable (r=.84). The second factor, labeled practicality included feasibility (r=−.81) and utility (r=−.98).

Hypothesis 5 predicted a positive relationship between the expectations of positive outcomes overconfidence factor and vision effectiveness, while Hypothesis 6 predicted a negative relationship between seeing deficiencies overconfidence factor and vision

Table 4 Multivariate analysis of covariance examining effects of overconfidence factors on vision.

F df p η2

Covariates Self-esteem 6.67 2, 146 .002 .08 Neuroticism 2.70 2, 146 .071 .04

Main effects Seeing deficiencies 8.35 2, 147 .000 .10 Expecting positive outcomes 3.61 2, 147 .029 .05

Interactions Seeing deficiencies by expecting positive outcomes 2.23 4, 147 .0680 .06

Note: F=F Ratio; df=Degrees of Freedom; p=Significance Level using Roy's Largest Root; η2=Effect Size (eta squared).

Table 3 Multivariate analysis of covariance examining effects of overconfidence factors on plan.

F df p η2

Covariates Intelligence 8.31 2, 147 .000 .10

Main effects Seeing deficiencies 11.24 2, 148 .000 .13 Expecting positive outcomes .76 2, 148 .471 .01

Interactions Seeing deficiencies by expecting positive outcomes 2.86 4, 148 .0250 .07

Note: F=F Ratio; df=Degrees of Freedom; p=Significance Level using Roy's Largest Root; η2=Effect Size (eta squared).

660 A.S. Shipman, M.D. Mumford / The Leadership Quarterly 22 (2011) 649–665

effectiveness. The influence of overconfidence on vision performance was assessed using another multivariate analysis of covariance. The results of this analysis can be seen in Table 4. Again, all control measures were included as covariates, but only significant measures were retained. Here the significant covariate measure was self-esteem (F(2, 146)=6.67, pb .01, η2=.08) and an approaching significant measure of neuroticism (F(2, 146)=2.70, p=.07, η2=.04). Those with higher self-esteem tended to have better speeches while those with higher levels of neuroticism had worse speeches.

A significant effect was seen for overconfidence on vision performance for seeing deficiencies (F (2, 147)=8.35, pb .01, η2=.10) using Roy's Largest Root. The univariate analyses revealed this significance for inspirational appeal (F(2, 147)=5.63, pb .01) and practicality (F(2, 147)=7.81, pb .01). The pattern was consistent in the cell means for both practicality and inspirational appeal where those who were overconfident in terms of seeing deficiencies generated the worst visions (M=−.76, SE=.24, M=−.41, SE=.24) compared to those who had low and moderate confidence (M=.36, SE=.19, M=.30, SE=.10). Thus, those who were overconfident and saw few deficiencies had the worst vision statements. These results supported Hypothesis 6.

The multivariate analysis also indicated a significant effect for expecting positive outcomes (F(2, 147)=3.61, p=.03, η2=.05). The univariate revealed this was true for inspirational appeal only (F(2, 147)=3.54, p=.03). The cell means indicated that the most appealing vision statements were generated by those who were overconfident in terms of their expectations of positive outcomes (M=.58, SE=.28) while less appealing visions were developed by those with moderate to low levels of expectations about positive outcomes (M=−.19, SE=.19). In other words, those who were overconfident and expected a great deal of positive outcomes generated the most inspirationally appealing vision statements. These findings provide support for Hypothesis 5.

6. Discussion

Before turning to the broader implications of the current study a few limitations should be noted. First, an undergraduate sample served as the study participants. While this sample has been effective at assuming leadership roles and in particular this leadership scenario (e.g., Strange & Mumford, 2005), the extent to which this sample is generalizable may be questioned. Although significant effects were found using the multivariate analysis of covariance, the effect sizes were relatively small. This is a limitation of the current research and future research should make efforts to use an alternate improved study design and/or measures to ascertain whether the effect sizes are accurately reflecting the population or actually larger.

This study did not address situational variables, such as followers or environmental conditions that may influence overconfidence. Follower confidence is argued to have a critical influence on leader confidence over time (Wood, 2007). On the other hand, situational variables may be less critical to studies of overconfidence if this leader characteristic is more trait-like and stable. In Hannah et al.'s (2008) recent review of leadership self-efficacy they describe this leader characteristic as being somewhat stable (i.e., more stable than moods) but less rigid than personality (Parker, 1998). It could be that leaders with varying levels of relatively stable confidence are influenced by different situations. Thus, those who have a tendency to be confident may become overconfident in certain situations while those who have a tendency to be less confident may become more confident in other situations. This relationship would be a valuable avenue for future research.

In addition to the situation, the nature of the task itself may influence the leader's confidence. The confidence level may be so dependent or localized to the task itself that the influence of personality is limited. Along these lines, Chemers et al. (2000) suggested that leader confidence may have weak relationships with personality traits since it is specific to the leader domain. If this is the case, the limited number of significant correlations between the overconfidence factors and theoretically relevant personality variables can be explained. These questions should be explored in future research.

Although the nature of overconfidence should be further explored, this study offers a preliminary contribution to that effort. In this regard, a conceptual framework for understanding and measuring overconfidence was proposed. Measuring overconfidence based off of observable indicators is likely more appropriate and accurate than using a self-report or peer assessments method. Within this framework two elements of overconfidence were examined: seeing deficiencies and expecting positive outcomes. These elements were proposed based on the relevant literature (e.g., Yukl, 2002) and showed some structural validity based on the factor analysis results. Additionally, these elements of overconfidence, particularly the seeing deficiencies element indicated at least some construct validity as seen in relationships with narcissism, which is theoretically related to overconfidence (Felfe & Schyns, 2006; Raskin, Novacek, & Hogan, 1991). The expectations of positive outcomes overconfidence factor was only found negatively related to neuroticism, which is in line with Popper et al.'s (2004) conceptualization of confidence.

Additional validation evidence is needed, particularly in regards to the expectations of positive outcomes element. One implication flowing from this study is that overconfidence is a multidimensional construct with the dimensions operating somewhat independently of each other since these two demonstrated different relationships with the performance criteria. This point is important to bear in mind as it suggests that a leader can be overconfident in one respect and not confident in the other.

Confidence has been shown to positively influence leader performance interpersonally (e.g. Chemers et al., 2000; Luthans & Peterson, 2002; Robertson & Sadri, 1993). However, this leader characteristic may not always enhance leader performance in non- interpersonal domains. A strength of the current study is that leader performance was directly assessed, rather than assessing perceptions of leader performance. The influence that overconfidence has on leader performance in cognitive activities, such as planning was examined. The results here indicate that overconfidence is largely detrimental to performance, and in fact, low levels of confidence, associated with seeing a great deal of deficiencies is critical to developing effective plans. This relationship can most likely be explained by a lack of confidence inducing more intensive processing which is beneficial to planning (Mumford & Van Doorn, 2001). Given the findings of the current study it would not be appropriate to conclude that low levels of confidence results

661A.S. Shipman, M.D. Mumford / The Leadership Quarterly 22 (2011) 649–665

in the best leader performance. The logical argument could be proposed that seeing a great deal of deficiencies in one's ideas helps the leader adjust and refine ideas accordingly. By working through the potential problems with ideas the leader may actually become quite confident afterwards.

When a leader has moderate confidence in terms of seeing deficiencies paired with low confidence in terms of expectations of positive outcomes plan performance suffers. The reason for this decrement to performance is uncertain, however, it could be because a moderate amount of critical analysis paired with a lack of enthusiasm or optimism reflects leaders detached and minimally interested in their idea development. These leaders may see enough problems and obstacles with their ideas that they feel disinterested in putting forth the effort to further develop the ideas since they do not also expect positive outcomes. However, the question why this particular combination is detrimental to plan performance is not answered by the current study.

On the other hand, overconfidence associated with expectations of positive outcomes is neither detrimental nor helpful to plan performance. This lack of relationship is meaningful in and of itself as some researchers suggest that being confident in terms of optimism motivates a leader which enhances their performance (e.g., Luthans et al., 2001; Peterson, 2000). This does not appear to be the case with cognitive planning performance. A leader does not need the motivation associated with optimism in order to develop a highly effective plan. Being an optimistic leader does not guarantee better performance than the less positive counterpart. What does differentiate leader performance in planning is a focus on the negative (i.e., deficiencies).

Being extremely confident in terms of expecting positive outcomes does enhance performance in another area of leadership performance. That is, in articulating an inspirational vision statement. In other words, excessive optimism helps a leader develop a vision that is appealing and motivating to followers. It is likely that a leader has to believe their ideas are ultimately going to be effective and result in positive outcomes in order to deliver a vision that inspires, motivates, and generates confidence in followers. Along these lines, Wunderley et al. (1998) found a positive correlation between leader's optimism and inspiring a shared vision. This relationship may best be explained through emotional contagion, whereby, the optimism the leader has becomes contagious to the followers (Bono & Ilies, 2006; Naidoo & Lord, 2008). However, excessive optimism must be paired with a critical evaluation of ideas. The current study found that, like plans, visions are hurt by overconfidence in terms of failing to see deficiencies.

6.1. Practical implications

The results of the current study have certain practical implications for real world leaders. Certain individuals may be more likely to emerge as leaders given their confidence. However, these leaders must be encouraged to keep their confidence at reasonably low levels when engaged in critical cognitive activities, such as planning. Specifically, leaders should not let confidence blind themselves to obstacles, problems or errors. One solution suggested to remedy hubristic leaders is to encourage introspection and limit beliefs about infallibility (Kroll et al., 2000). Thus, organizations may put systems in place, such that leaders are encouraged or required to critically self-evaluate and revise their work based off of potential problems. Leaders that take these steps may augment their confidence in their work, increasing their expectations of positive outcomes. This optimism is useful to the leader when they begin their influence attempts on others, where they can use their optimism to inspire and appeal emotionally to their followers.

Given that leaders will need to critically evaluate their own work, yet also influence others to support their ideas presents a potentially challenging aspect of being a leader. It may be the case that leaders must engage in a certain amount of acting in order to fully convince others to pursue the leader's objectives. On the other hand, leaders may develop authentic positive expectations as a result of their previous critical analysis, giving them the positivity needed to inspire others to believe the leader's ideas. The antecedents preceding overconfidence and the relationship between the types of overconfidence are not explored in the current study, but these questions may be addressed in future studies of overconfidence.

In conclusion, leader overconfidence is a valuable avenue for leadership research. Self-confident individuals may excel interpersonally and give those around them the impression of competence. In this respect, self-confidence has predominantly been considered a positive leader characteristic and beneficial to leader performance. This study takes a different perspective and considers the ‘dark side’ to leader confidence and demonstrates that global statements about the benefit of self-confidence are not appropriate. In fact, leaders who have low levels of confidence associated with seeing many deficiencies in ideas develop better plans and vision statements. Thus, a leader may portray confidence to their followers but in order to generate effective problem solutions and vision statements, the leader must become highly critical and evaluative. On the other hand, overconfidence associated with expectations of positive outcomes may be helpful to leaders when they must develop an inspiring vision and influence their followers. While the elements of overconfidence and the relationship between overconfidence and leader performance requires more research, this study offers some preliminary evidence in this regard.

References

Ackerman, P. L., & Humphreys, L. G. (1990). Individual differences theory in industrial and organizational psychology. In M. D. Dunnette & L. M. Hough (Eds.), Handbook of industrial and organizational psychology, Vol. 1. (pp. 223–282). Palo Alto, CA: Consulting Psychological Press.

Atwater, L. E., Dionne, S. D., Avolio, B., Camobreco, J. F., & Lau, A. W. (1999). A longitudinal study of the leadership development process: Individual differences predicting leader effectiveness. Human Relations, 52, 1543–1562.

Atwater, L. E., & Yammarino, F. J. (1992). Does self-other agreement on leadership perceptions moderate the validity of leadership and performance predictions? Personnel Psychology, 45, 141–164.

Baer, J. (2003). Evaluative thinking, creativity, and task specificity: Separating wheat from chaff is not the same as finding needle in haystacks. In M. A. Runco (Ed.), Critical creative processes (pp. 129–152). Cresskill, NJ: Hampton. Citation.

Bandura, A. (1997). Self-efficacy: The exercise of control. New York: Freeman.

662 A.S. Shipman, M.D. Mumford / The Leadership Quarterly 22 (2011) 649–665

Bass, B. M. (1985). Leadership and performance beyond expectations. New York: Free Press. Bass, B. M. (1990). Bass and Stogdill's handbook of leadership: Theory, research, and managerial applications. New York: Free Press. Bass, B. M., & Avolio, B. J. (1994). Improving organizational effectiveness through transformational leadership. Thousand Oaks, CA: Sage. Bazerman, M. H., & Neale, M. A. (1983). Heuristics in negotiations: Limitations to dispute resolution potential. In M. H. Bazerman & R. Lewicki (Eds.), Negotiating in

organizations (pp. 51–67). Beverly Hills, CA: Sage. Bazerman, M. H., & Sondak, H. (1988). Judgmental limitations in diplomatic negotiations. Negotiation Journal, 4, 303–317. Becker, T. E., Billings, R. S., Eveleth, D. M., & Gilbert, N. W. (1997). Validity of scores on three attachment style scales: Exploratory and confirmatory evidence.

Educational and Psychological Measurement, 57, 477–493. Bell, N. J., Lassiter, K. S., Leverett, K. S., & Matthews, T. D. (2002). Validity of the Wonderlic personnel test as a measure of fluid or crystallized intelligence:

Implications for career assessment. North American Journal of Psychology, 4, 113–120. Besemer, S. P., & O'Quin, K. (1999). Confirming the three-factor creative product analysis matrix model in an American sample. Creativity Research Journal, 12,

287–296. Black, J. S., & Porter, L. W. (2000). Management: Meeting new challenges. Upper Saddle River, NJ: Prentice Hall. Blaine, B., & Crocker, J. (1993). Self-esteem and self-serving biases in reactions to positive and negative events: An integrative view. In R. F. Baumeister (Ed.), Self-

esteem: The puzzle of low self-regard (pp. 55–85). New York: Plenum. Bochner, A. P., di Salvo, V., & Jonas, T. (1975). A computer-assisted analysis of small group process: An investigation of two Machiavellian groups. Small Group

Behavior, 6, 187–203. Bono, J. E., & Ilies, R. (2006). Charisma, positive emotions and mood contagion. The Leadership Quarterly, 17, 317–334. Brunell, A. B., Gentry, W. A., Campbell, W. K., Hoffman, B. J., Kunhert, K. W., & DeMarree, K. G. (2008). Leader emergence: The case of the narcissistic leader.

Personality and Social Psychology Bulletin, 34, 1663–1676. Byrne, B. M. (1996). Measuring self-concept across the life span. Hyattsville, MD: American Psychological Association. Campbell, W. K., Goodie, A. S., & Foster, J. D. (2004). Narcissism, confidence, and risk attitude. Journal of Behavioral Decision Making, 17, 297–311. Carnevale, P. J., & Pruitt, D. G. (1992). Negotiation and mediation. Annual Review of Psychology, 43, 531–582. Chan, K., & Drasgow, F. (2001). Toward a theory of individual differences and leadership: Understanding the motivation to lead. The Journal of Applied Psychology,

86, 481–498. Chemers, M. M., Watson, C. B., & May, S. T. (2000). Dispositional affect and leadership effectiveness: A comparison of self-esteem, optimism, and efficacy.

Personality and Social Psychology Bulletin, 26, 267–277. Christie, R., & Geis, F. (1970). Studies in Machiavellianism. New York: Academic Press. Conger, J. A. (1990). The dark side of leadership. Organizational Dynamics, 19, 44–55. Conger, J. A. (1999). Charismatic and transformational leadership in organizations: An insider's perspective on these developing streams of research. The

Leadership Quarterly, 10, 145–179. Conger, J. A., & Kanungo, R. N. (1987). Toward a behavioral theory of charismatic leadership. Academy of Management Review, 12, 637–647. Conway, J. M. (2000). Managerial performance development constructs and personality correlates. Human Performance, 13, 23–46. Conway, J. M., & Peneno, G. M. (1999). Comparing structured interview question types: Construct validity and applicant reactions. Journal of Business and

Psychology, 13, 485–506. Corry, N., Merritt, R. D., Mrug, S., & Pamp, B. (2008). The factor structure of the narcissistic personality inventory. Journal of Personality Assessment, 90, 564–573. Deluga, R. J. (2001). American presidential Machiavellianism: Implications for charismatic leadership and rated performance. The Leadership Quarterly, 12,

339–363. Dennison, D. R., Hooijberg, R., & Quinn, R. E. (1995). Paradox and performance: Toward a theory of behavioral complexity in managerial leadership. Organizational

Science, 6, 524–540. Dodrill, C. B., & Warner, M. H. (1988). Further studies of the Wonderlic Personnel Test as a brief measure of intelligence. Journal of Consulting and Clinical Psychology,

56, 145–147. Dorner, D., & Schaub, H. (1994). Errors in planning and decision making and the nature of human information processing. Applied Psychology: An International

Review, 43, 433–453. Drazin, R., Glynn, M. A., & Kazanjian, R. K. (1999). Multilievel theorizing about creativity in organizations: A sensemaking perspective. Academy of Management

Review, 24, 286–307. Dumdum, U. R., Lowe, K. B., & Avolio, B. J. (2002). A meta-analysis of transformational and transactional leadership correlates of effectiveness and satisfaction: An

update and extension. In B. J. Avolio & F. J. Yammarino (Eds.), Transformational and charismatic leadership: The road ahead, Vol. 2. (pp. 35–66). Oxford, UK: Elsevier Science.

Ellis, J. J. (2001). Founding brothers: The revolutionary generation. New York: Alfred Knopf. Felfe, J., & Schyns, B. (2006). Personality and the perception of transformational leadership: The impact of extraversion, neuroticism, personal need for structure,

and occupational self-efficacy. Journal of Applied Social Psychology, 36, 708–739. Fischer, G. W. (1982). Scoring-rule feedback and the overconfidence syndrome in subjective probability forecasting. Organizational Behavior and Human

Performance, 29, 352–369. Ford, C. M., & Gioia, D. A. (2000). Factors influencing creativity in the domain of managerial decision-making. Journal of Management, 26, 705–732. Geis, F. L., Krupat, E., & Berger, D. (1970). Taking over in group discussion. In R. Christie & F. L. Geis (Eds.), Studies in Machiavellianism (pp. 392). New York:

Academic Press. George, J. M. (2000). Emotion and leadership: The role of emotional intelligence. Human Relations, 53, 1027–1055. Gervais, S., Heaton, J. B., & Odean, T. (2007). Overconfidence, investment policy, and manager welfare. Working paper: Duke University. Gist, M. E., & Mitchell, T. R. (1992). Self-efficacy: A theoretical analysis of its determinants and malleability. Academy of Management Review, 17, 183–211. Glad, B. (2002). Why tyrants go too far: Malignant narcissism and absolute power. International Society of Political Psychology, 23, 1–37. Goldberg, L. R. (1992). The development of markers for the Big-Five factor structure. Psychological Assessment, 4, 26–42. Gorman, M. E., & Plucker, J. A. (2003). Teaching invention as critical creative processes: A course on techno scientific creativity. In M. A. Runco (Ed.), Critical creative

processes (pp. 275–302). Cresskill, NJ: Hampton. Gray-Little, B., Williams, V. S. L., & Hancock, T. D. (1997). An item response theory analysis of the Rosenberg Self-Esteem Scale. Personality and Social Psychology

Bulletin, 23, 443–451. Halpern, D. F. (2003). Thinking critically about creative thinking. In M. A. Runco (Ed.), Critical creative processes (pp. 189–208). Cresskill, NJ: Hampton. Hannah, S. T., Avolio, B. J., Luthans, F., & Harms, P. D. (2008). Leadership efficacy: Review and future directions. The Leadership Quarterly, 19, 669–692. Hayward, M. L. A., & Hambrick, D. C. (1997). Explaining the premiums paid for large acquisitions: Evidence of CEO hubris. Administrative Science Quarterly, 42,

103–127. Hendricks, J., & Payne, S. (2007). Beyond the big five: Leader goal orientation as a predictor of leadership effectiveness. Human Performance, 20, 317–343. Hogan, R., Raskin, R., & Fazzini, D. (1990). The dark side of charisma. In K. E. Clark & M. B. Clark (Eds.), Measures of leadership (pp. 343–354). West Orange, NJ:

Leadership Library of America. Hollenbeck, G. P., & Hall, D. T. (2004). Self-confidence and leader performance. Organizational Dynamics, 33, 254–269. Hollenbeck, J. R., & Klein, H. J. (1987). Goal commitment and the goal-setting process: Problems, prospects, and proposals for future research. The Journal of Applied

Psychology, 72, 212–220. Holyoak, K. J., & Thagard, P. (1997). The analogical mind. The American Psychologist, 52, 35–44. Hooijberg, R., Hunt, J. G., & Dodge, G. E. (1997). Leadership complexity and development of the leaderplex model. Journal of Management, 23, 375–408. House, R. J. (1977). A 1976 theory of charismatic leadership. In J. G. Hunt & L. L. Larson (Eds.), Leadership: The cutting edge (pp. 189–207). Carbondale, IL: Southern

Illinois University Press.

663A.S. Shipman, M.D. Mumford / The Leadership Quarterly 22 (2011) 649–665

House, R. J., & Howell, J. M. (1992). Personality and charismatic leadership. The Leadership Quarterly, 3, 81–108. Howell, R. J., & Avolio, B. J. (1992). The ethics of charismatic leadership: Submission or liberation? Academy of Management Executive, 6, 43–54. Hunt, J. G., Boal, K. B., & Dodge, G. E. (1999). The effects of visionary and crisis-responsive charisma on followers: An experimental examination of two kinds of

charismatic leadership. The Leadership Quarterly, 10, 423–448. Hunt, J. G., & Phillips, R. L. (1991). Leadership in battle and garrison: A framework for understanding the differences and preparing for both. In R. Gal & D.

Mangeldorrf (Eds.), Handbook of military psychology (pp. 411–429). Chichester, UK: Wiley. James, L. R., Demaree, R. G., & Wolf, G. (1984). Estimating within-group interrater reliability with and without response bias. The Journal of Applied Psychology, 69,

85–98. James, L. R., Demaree, R. G., & Wolf, G. (1993). rwg: An assessment of within-group interrater agreement. The Journal of Applied Psychology, 78, 306–309. Jameson, S. H. (1945). Principles of social interaction. American Sociological Review, 10, 6–12. Judge, T. A., Erez, A., Bono, J. E., & Thoresen, C. J. (2003). The core self-evaluations scale: Development of a measure. Personnel Psychology, 56, 303–331. Judge, T. A., Locke, E. A., & Durham, C. C. (1997). The dispositional causes of job satisfaction: A core evaluations approach. Research in Organizational Behavior, 19,

151–188. Judge, T. A., Piccola, R. F., & Kosalka, T. (2009). The bright and dark sides of leader traits: A review and theoretical extension of the leader trait paradigm. The

Leadership Quarterly, 20, 855–875. Kaiser, R. B., Hogan, R., & Craig, S. B. (2008). Leadership and the fate of organizations. The American Psychologist, 63, 96–110. Klayman, J., & Ha, Y. (1989). Hypothesis testing in rule discovery: Strategy, structure, and content. Journal of Experimental Psychology. Learning, Memory, and

Cognition, 15, 596–604. Kohut, H. (1966). Forms and transformations of narcissism. Journal of the American Psychoanalytic Association, 14, 243–272. Koriat, A., Lichtenstein, S., & Fischhoff, B. (1980). Reasons for confidence. Journal of Experimental Psychology: Human Learning and Memory, 6, 107–118. Kroll, M. J., Toombs, L. A., & Wright, P. (2000). Napoleon's tragic march home from Moscow: Lessons in hubris. Academy of Management Executive, 14, 117–128. Kubarych, T. S., Deary, I. J., & Austin, E. J. (2004). The Narcissistic Personality Inventory: Factor structure in a non-clinical sample. Personality and Individual

Differences, 36, 857–872. Largan-Fox, J., & Code, S. (2000). Team mental models: Techniques, methods, and analytical approaches. Human Factors, 42, 242–271. Lauriola, M., & Levin, I. P. (2001). Personality traits and risky decision-making in a controlled experimental task: An exploratory study. Personality and Individual

Differences, 31, 215–226. Luthans, F., Luthans, K. W., Hodgetts, R. M., & Luthans, B. C. (2001). Positive approach to leadership (PAL): Implications for today's organizations. The Journal of

Leadership Studies, 8, 3–20. Luthans, F., & Peterson, S. J. (2002). Employee engagement and manager self-efficacy: Implications for managerial effectiveness and development. The Journal of

Management Development, 21, 376–387. Marta, S., Leritz, L. E., & Mumford, M. D. (2005). Leadership skills and group performance: Situational demands, behavioral requirements, and planning. The

Leadership Quarterly, 16, 97–120. Martin, N. H., & Sims, J. H. (1956). Thinking ahead: Power tactics. Harvard Business Review, 34, 25–36. McKelvie, S. J. (1989). The Wonderlic Personnel Test: Reliability and validity in an academic setting. Psychological Reports, 65, 161–162. Miller, C. C., & Cardinal, L. B. (1994). Strategic planning and firm performance: A synthesis of more than two decades of research. Academy of Management Journal,

37, 1649–1665. Mumford, M. D., & Connelly, M. S. (1991). Leaders as creators: Leader performance and problem-solving in ill-defined domains. The Leadership Quarterly, 2,

289–315. Mumford, M. D., Friedrich, T. L., Caughron, J. J., & Byrne, C. E. (2007). Leader cognition in real-world settings: How do leaders think about crises? The Leadership

Quarterly, 18, 515–543. Mumford, M. D., & Gustafson, S. B. (1988). Creativity syndrome: Integration, application, and innovation. Psychological Bulletin, 103, 27–43. Mumford, M. D., & Van Doorn, J. R. (2001). Performance in planning: Reconsidering Franklin in the age of charisma. The Leadership Quarterly, 12, 451–495. Mumford, M. D., Whetzel, D. L., & Reiter-Palmon, R. (1997). Thinking creatively at work: Organizational influences on creative problem-solving. Journal of Creative

Behavior, 31, 7–17. Mumford, M. D., Zaccaro, S. J., Harding, F. D., Jacobs, T. O., & Fleishman, E. A. (2000). Leadership skills for a changing world: Solving complex social problems. The

Leadership Quarterly, 11, 11–36. Naidoo, L. J., & Lord, R. G. (2008). Speech imagery and perceptions of charisma: The mediating role of positive affect. The Leadership Quarterly, 19, 283–296. Neale, M. A., & Bazerman, M. H. (1985). The effects of framing and negotiator overconfidence on bargaining behaviors and outcomes. Academy of Management

Journal, 28, 34–49. Noice, H. (1991). The role of explanations and plan recognition in the learning of theatrical scripts. Cognitive Science, 15, 425–460. Norman, S., Luthans, B., & Luthans, K. (2005). The proposed contagion effect of hopeful leaders on the resiliency of employees and organizations. Journal of

Leadership and Organizational Studies, 12, 55–64. Northouse, P. G. (1997). Leadership theory and practice. Thousand Oaks, CA: Sage Publications. Okanes, M. M., & Stinson, J. E. (1974). Machiavellianism and emergent leadership in a management simulation. Psychological Reports, 35, 255–259. Owen, D., & Davidson, J. (2009). Hubris syndrome: An acquired personality disorder? A study of US Presidents and UK Prime Ministers over the last 100 years.

Brain, 132, 1396–1406. Padilla, A., Hogan, R., & Kaiser, R. B. (2007). The toxic triangle: Destructive leaders, susceptible followers, and conducive environments. The Leadership Quarterly, 18,

176–194. Palis, L. L., & Green, S. G. (2002). Leadership self-efficacy and managers’ motivation for leading change. Journal of Organizational Behavior, 23, 215–235. Parker, S. K. (1998). Enhancing role-breadth self-efficacy: The roles of job enrichment and other organizational interventions. The Journal of Applied Psychology, 83,

835–852. Paulson, A. M., & Betz, N. E. (2004). Basic confidence predictors of career decision-making self-efficacy. Career Development Quarterly, 52, 354–362. Peterson, C. (2000). The future of optimism. The American Psychologist, 55, 44–55. Pfiffner, J. M. (1951). The supervision of personnel: Human relations in the management of men. Englewood Cliffs, NJ: Prentice-Hall. Popper, M., Amit, K., Gal, R., Mishkal-Sinai, M., & Lisak, A. (2004). The capacity to lead: Major psychological differences between leaders and nonleaders. Military

Psychology, 16, 1531–7876. Popper, M., & Mayseless, O. (2006). The building blocks of leader development: A psychological conceptual framework. Leadership and Organization Development

Journal, 28, 664–684. Post, J. (1993). Current concepts of the narcissistic personality: Implications for political psychology. Political Psychology, 14, 99–121. Quinn, R. E. (1988). Beyond rational management: mastering the paradoxes and competing demands of high performance. San Francisco: Jossey-Bass. Raskin, R. N., & Hall, C. S. (1979). A narcissistic personality inventory. Psychological Reports, 45, 590. Raskin, R., Novacek, J., & Hogan, R. (1991). Narcissistic self-esteem management. Journal of Personality and Social Psychology, 60, 911–918. Reysen, S. (2005). Construction of a new scale: The Reysen likability scale. Social Behavior and Personality, 33, 201–208. Rim, Y. (1966). Machiavellianism and decisions involving risk. The British Journal of Social and Clinical Psychology, 5, 35–50. Robertson, I. T., & Sadri, G. (1993). Managerial self-efficacy and managerial performance. British Journal of Management, 4, 37–45. Robins, R. W., & Paulhus, D. L. (2001). The character of self-enhancers: Implications for organizations. In B. W. Roberts & R. Hogan (Eds.), Personality psychology in

the workplace (pp. 193–219). Washington, DC: American Psychological Association. Rosenberg, M. (1965). Measurement of self-esteem. In M. Rosenberg (Ed.), Society and the adolescent self image (pp. 297–307). New York: Princeton University

Press. Sarason, I. G. (1975). Anxiety and self preoccupation. In I. G. Sarason & C. D. Spielberger (Eds.), Stress and anxiety, Vol. 2. (pp. 27–44). Washington DC: Hemisphere.

664 A.S. Shipman, M.D. Mumford / The Leadership Quarterly 22 (2011) 649–665

Saucier, G. (2002). Orthogonal markers for orthogonal factors: The case of the Big Five. Journal of Research in Personality, 36, 1–31. Shamir, B., House, R. J., & Arthur, M. B. (1993). The motivational effects of charismatic leadership: A self-concept based theory. Organization Science, 4, 1–17. Shipman, A., Byrne, C. L., & Mumford, M. D. (2010). Leader vision formation and forecasting: The effects of forecasting extent, resources, and timeframe. The

Leadership Quarterly, 21, 439–456. Simonton, D. K. (1999). Significant samples: The psychological study of eminent individuals. Psychological Methods, 4, 425–451. Sosik, J. M., Kahai, S. S., & Avolio, B. J. (1999). Leadership style, anonymity, and creativity in group decision support systems. Journal of Creative Behavior, 33,

227–257. Strange, J. M., & Mumford, M. D. (2005). The origins of vision: Effects of reflection, models, and analysis. The Leadership Quarterly, 16, 121–148. Thompson, L., & Hastie, R. (1990). Judgment tasks and biases in negotiation. In B. H. Sheppard, M. H. Bazerman, & R. J. Lewicki (Eds.), Research on negotiation in

organizations, Vol. 2. (pp. 31–54). Greenwich, CT: JAI Press. Trice, H. M., & Beyer, J. M. (1986). Charisma and its routinization in two social movement organizations. In L. L. Cummings & B. M. Staw (Eds.), Research in

organizational behavior (Vol 8) (pp. 113–164). Greenwich, CT: JAI Press. Tucker, R. C. (1970). The theory of charismatic leadership. In D. A. Rustow (Ed.), Philosophers and kings: Studies in leadership (pp. 69–94). New York: Braziller. van Knippenberg, D. (2000). Work motivation and performance: A social identity perspective. Applied Psychology: An International Review, 49, 357–371. Wakefield, J. F. (2003). The development of creative thinking and critical reflection. In M. A. Runco (Ed.), Critical creative processes (pp. 253–274). Cresskill, NJ:

Hampton. Ward, T. B., Smith, S. M., & Finke, R. A. (1999). Creative cognition. In R. J. Sternberg (Ed.), Handbook of creativity (pp. 189–212). New York: Cambridge University

Press. Wason, P. C. (1968). Reasoning about a rule. The Quarterly Journal of Experimental Psychology, 20, 273–281. Watson, P. J., & Biderman, M. D. (1994). Narcissistic Traits Scale: Validity evidence and sex differences in narcissism. Personality and Individual Differences, 16,

501–504. Wonderlic, E. (1992). Wonderlic Personnel Test. IL: Libertyville. Wood, D. (2007). Using PRISM to compare the explanatory value of general and role-contextualized trait ratings. Journal of Personality, 75, 1103–1126. Wunderley, L. J., Reddy, W. B., & Dember, W. N. (1998). Optimism and pessimism in business leaders. Journal of Applied Social Psychology, 28, 751–760. Xiao, Y., Milgram, P., & Doyle, D. J. (1997). Capturing and modeling planning expertise in anesthesiology: Results of a field study. In C. E. Zsambok & G. Klein (Eds.),

Naturalistic decision making (pp. 197–205). Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. Yammarino, F. Y., & Atwater, L. E. (1997). Do managers see themselves as others see them? Organizational Dynamics, 25, 35–44. Yukl, G. (1999). An evaluation of conceptual weaknesses in transformational and charismatic leadership theories. The Leadership Quarterly, 10, 285–305. Yukl, G. (2002). Leadership in organizations (5th Ed.). Upper Saddle River, NJ: Prentice Hall.

665A.S. Shipman, M.D. Mumford / The Leadership Quarterly 22 (2011) 649–665