First and ten leadership: A historiometric investigation of the CIP leadership model

Criticisms of the dominant leadership perspectives in the literature are increasing and, as such, a growing number of scholars are calling for more complex and conceptually sound theories of leadership. With a multi-faceted perspective on effective leadership, detailed conceptual underpinnings, and increasing body of empirical support, the Charismatic, Ideological, and Pragmatic (CIP) model of leadership has the potential to address many of these concerns and substantively contribute to our understanding of effective leadership. Despite such advantages , however, wide scale proliferation of the model remains elusive. As such, this effort provides the first comprehensive review of the CIP model to analyze its potential to expand our understanding of leadership in science and practice. In doing so this review frames and organizes the existing CIP literature, identifies key strengths of the model, addresses key limitations of the model, and outlines future research opportunities that would benefit from adopting a CIP perspective.

Theories of outstanding, historically notable leadership have traditionally emphasized charisma. Recent research, however, suggests that charisma may represent only one pathway to outstanding leadership. Outstanding leadership may also emerge from ideological and pragmatic leadership. This article examines the conditions influencing the emergence and performance of charismatic, ideological, and pragmatic leaders. In particular, different conditions operating at the environmental, organizational, group, and individual levels influence the emergence and performance of each of these three types of leaders. Implications for understanding the origins and impact of charismatic, ideological, and pragmatic leaders are discussed as well.

The first publication to appear of an Archaic settlement in Aetolia with stratified pottery and tools for textile production. (The second half of the publication)

The Leadership Quarterly 22 (2011) 70–91 Contents lists available at ScienceDirect The Leadership Quarterly j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / l e a q u a First and ten leadership: A historiometric investigation of the CIP leadership model Samuel T. Hunter a,⁎, Liliya Cushenbery a, Christian Thoroughgood a, Johanna E. Johnson a, Gina Scott Ligon b a b Pennsylvania State University, USA Villanova University, USA a r t i c l e i n f o Available online 3 February 2011 a b s t r a c t Emerging from the early work of Weber (1924), the charismatic, ideological, and pragmatic (CIP) model of leadership (Mumford, 2006) has enjoyed a recent surge in research attention. According to the model, the three leader types differ in a number of fundamental ways — differences largely tied to how the leaders provide sensemaking to followers. Although these differences are central to the model, these components have yet to be examined directly. As such, the aim of this study was to explicitly test the core tenants of the CIP model. Using a historiometric sample of college and NFL football coaches we found general support for specific predictions made by the model and in the aggregate, the model as a whole. We also examined the unique patterns and features that were used to distinguish among the leader types providing useful insight into how leaders may be categorized. Implications and future directions are discussed. © 2010 Elsevier Inc. All rights reserved. “On the morning of Super Bowl XXXVI, Belichick had no remarkable speeches for the team. He gave the Patriots the major points of the plan once again, guessing it was the twentieth time they had heard these instructions from him” (Holley, 2004, p. 66).– Bill Belichick preparing his team for the Super Bowl — pragmatic leader “I believe God has prepared the leaders of our team for this time. Over the last four years, starting in 2003, we have had the most wins in football, yet each season has ended in disappointment. Until this one” (Whitaker, 2007, p. 291).– Tony Dungee preparing his team for the Super Bowl — ideological leader “We've got only one more half to play in the Coliseum for a long time. Remember it's about us; it's not about them. Show them how we do it and give everything you got — so we can win forever” (Malcomsen, 2008).– Pete Carroll preparing his team for the second half — charismatic leader Recently, the leadership literature has witnessed a reemergence of Weber's (1924) original conceptualization of leadership styles — a reemergence that has expanded our thinking about leaders from largely a charismatic or transformational perspective to include those leaders that are more ideological and pragmatic in nature. In fact, the development of the charismatic, ideological and pragmatic (CIP) model of leadership (Mumford, 2006; Strange & Mumford, 2002; Mumford & Van Doorn, 2001) has resulted in several new studies examining multiple differences among the three leader types; differences that include creative ⁎ Corresponding author. Industrial and Organizational Psychology, Penn State University, 112 Moore Building, State College, PA 16802, USA. Tel.: +1 814 865 0107 (ofﬁce). E-mail address: samhunter@psu.edu (S.T. Hunter). 1048-9843/$ – see front matter © 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.leaqua.2010.12.008 S.T. Hunter et al. / The Leadership Quarterly 22 (2011) 70–91 71 performance, communication exchanges, political behaviors, and Machiavellianism (Bedell, Hunter, Angie & Vert, 2006), among others (cf. Mumford, 2006). Despite a growing body of literature on the model, however, there has yet to be a direct test of the model's core tenants in any of these prior research endeavors. Moreover, previous investigations of the model have been confounded by leadership domain, making it difﬁcult to assess if differences among key outcomes are due to common domain differences (i.e., similar leader types emerging from similar domains) or the leaders themselves. Thus, the aim of this study is to directly examine these tenants and consider the validity of the model using a sample of leaders emerging from a common domain — in this case, using a biographical sample of National Football League (NFL) and college football coaches. We begin with a historical review of the CIP model's emergence. 1. History of the CIP model Max Weber is well known for his impact on modern day perspectives of management, shaping many of our current views on bureaucracy and organizational functioning (Adair-Toteff, 2005; Miller, 1963; Weiss, 1983). One of his more notable contributions included propositions about differing styles or approaches to leadership which he referred to as the three types of management authority (Weber, 1924). The ﬁrst was termed “traditional” and included leaders who emphasized a strong focus on the past. Leaders employing this type of authority were concerned with stability, core values, and customs; they placed a great emphasis on family and tradition. The second type of authority was termed “rational” or legal. Weber suggested that these leaders emphasized logic and led via rational appeals to followers. The rational style, moreover, involved impartiality as leaders often dealt with the pragmatic concerns and problems associated with bureaucracy. Third and ﬁnally, Weber introduced the “charisma” label into the management ethos. He suggested that the charismatic leader was unique from both traditional and rational leaders in that they emphasized change and a break from bureaucracy. Weber was also aware of the transient nature of charisma as well as the unique characteristics of the followers who were most impacted by charismatic leaders. Despite the current emphasis on charismatic leadership in the literature, some have contended that Weber had little interest in charisma and instead saw the largest value in understanding the traditional and rational leadership styles (e.g., Hennis, 1996). Although this point is debatable, the end result is not: charismatic leadership and its visionary counterpart, transformational leadership, have emerged as the dominant ways of conceptualizing how we think of leaders (Hunt, 1999; Lowe & Gardner, 2000). 1.1. Charismatic leadership As alluded to, Weber's work gave way to many of the seminal pieces of leadership literature emerging in the twentieth century. For example, Burns' (1978) introduction of transformational leadership has inﬂuenced a number of scholars who produced several variations of, and expansions on, his core concepts. Some of the most notable include the full range model by Bass (1985) and colleagues (e.g., Avolio, 1999; Bass & Avolio, 1994). Around the similar time period and also largely inﬂuenced by Weber, House (1977) introduced his version of charismatic leadership which has led to several variations including those by Conger and Kanungo (1988; 1998) as well as expanded versions such as that by Shamir, House, and Arthur (1993) and Howell and Shamir (2005). Despite notable differences, these conceptualizations have more recently been viewed under a broader rubric of visionary or charismatic leadership (e.g., Hunt, 1999; Shamir et al., 1993). 1.2. Pragmatic leadership A collection of researchers have returned to Weber's work and suggested that in addition to the charismatic style there are at least two other styles of leading: pragmatic and ideological. In their review of Benjamin Franklin's leadership, for example, Mumford and Van Doorn (2001) observed that Franklin emphasized pragmatic problem solving over vision-based leadership. Similarly, in Collins's (2001) review of 11 outstanding businesses, he found that the organizational leaders were nearly the opposite of the charismatic style and employed behaviors much more akin to the rational (i.e., pragmatic) and traditional leaders described by Weber. Others have recently suggested that an overemphasis on charismatic leadership may be dangerous and, instead, we must turn our attention to leaders who emphasize intelligence and rational decision-making over vision formation (Khurana, 2002). In Hunt and Ropo's (1995) case study of General Motor's CEO, for example, the authors found that more pragmatic behaviors such as structuring and framing problems in a broader system were critical components of outstanding leadership. Finally, in their study of 40 fortune 500 ﬁrms, Pasternack and O'Toole (2002) reported “… our research ﬁnds that the CEOs whose companies are best weathering the recent economic downturn are practicing old-fashioned, pragmatic management … (p. 3).” 1.3. Ideological leadership Building off the “traditional” leadership concept put forth by Weber (1924), Strange and Mumford (2002) quantitatively analyzed the biographies of 60 historical leaders. The researchers observed that ideological leaders engaged followers in very different ways than charismatic leaders, emphasizing strong values and ideological beliefs over inspiration and intellectual stimulation. Despite these differences in style, ideological leaders were not inferior to their charismatic counterparts — they simply differed in their approach to sensemaking. Along similar lines, in a recent study of Chinese CEO's, Tsui, Zhang, Wang, Xin, and Wu (2006) also observed differences in leadership types suggesting that some CEO's emphasized traditions while the others emphasized a more charismatic or 72 S.T. Hunter et al. / The Leadership Quarterly 22 (2011) 70–91 transformational approach. Tsui et al. (2006) noted that “… non-charismatic leaders are associated with strong cultures because of their attention to internal integration issues (institutionalize the values and systematize the processes)” (p. 131). As a whole, these studies highlight the importance of expanding our perspectives on leadership to include those that may not ﬁt cleanly into visionbased molds — a point echoed by a number of inﬂuential scholars (e.g., Hunt, 1999; Yukl, 1999; Yukl, 2006). 2. Charismatic, ideological and pragmatic leadership: deﬁning the CIP model As a result of observations such as those noted above, Mumford (2006) and colleagues (e.g., Bedell-Avers, Hunter, & Mumford, 2009) returned to, and expanded on, Weber's (1924) original classiﬁcation by conducting a series of studies extensively examining this tri-style leadership perspective comprised of charismatic, ideological and pragmatic leaders or more simply termed the CIP model of leadership. We will now explore this model, its core tenets, the results of the studies examining it, and ﬁnally, the critical questions that still remain. 2.1. Sensemaking Although differences among the three leadership styles may be witnessed in multiple contexts, Mumford and colleagues (Strange & Mumford, 2002; Mumford, 2006) suggested that their unique forms of inﬂuence are best observed during times of crisis. That is, the three leader types are theorized to fundamentally differ in how they provide sensemaking to followers — differences most readily observed when subordinates are in the greatest need of comfort, guidance, and direction (Beyer, 1999; Hunt, Boal, & Dodge, 1999; Pillai & Meindl, 1998). Accordingly, leaders are thought to help subordinates make sense of their environments by providing information about the direction they will be taking, and why. This description of future events is known as a prescriptive mental model, and it is this description that helps subordinates make sense of an ambiguous, stressful situation (Habermas & Bluck, 2000). Before they can offer this prescriptive framework, however, leaders must understand the situation as it currently is, a framework known as a descriptive mental model (Johnson-Laird, 1983; Sein & Bostrom, 1989). Descriptive mental models, or ways of seeing the world, are largely derived from past experiences. Illustrating this concept, Ligon, Hunter, and Mumford (2008) found that due largely to differences in early developmental experiences, the three leader types form very different perspectives of the world. As a result they offer varying prescriptive mental models to subordinates during the sensemaking process. Thus, the CIP model emerges from a collection of leaders that take very different approaches to sensemaking. Charismatic leaders, for example, are theorized to develop future-oriented visions that are often centered on a hope for a better future. Classic examples of charismatic leaders include John F. Kennedy, Fiorello La Guardia, and Franklin D. Roosevelt. Ideological leaders, on the other hand, are focused on the past and correcting previous mistakes. Examples of ideological leaders are Michael Collins, Jane Addams, and Lech Walesa. Finally, pragmatic leaders are more problem-solving and present focused; concerned with handling the task at hand in any way necessary. Examples of pragmatic leaders include Alfred Dupont, Thomas Watson, and Sam Walton. Even with this small set of leader exemplars it becomes evident how different – yet highly successful – the three leader types can be. Mumford and colleagues (Mumford, 2006; Bedell-Avers, Hunter & Mumford, 2008) summarized the variations in prescriptive mental models, suggesting that there are at least nine unique features that distinguish the three leader types — features that ultimately deﬁne the CIP framework: (a) time frame, (b) type of experience used, (c) nature of outcomes sought, (d) type of outcomes sought, (e) focus in model construction, (f) locus of causation, (g) controllability of causation, (h) targets of inﬂuence, and (i) crisis conditions associated with emergence and optimal performance. In addition, Mumford and colleagues have suggested that the use and type of emotionality in prescriptive mental models are also key differences among leader types (Mumford, 2006; Mumford et al., 2008). These core features and their associations to each leader type are summarized in Table 1. 3. Research on the CIP model 3.1. Early studies To date, there has been a reasonably strong collection of studies investigating the CIP model — a number approaching 15 empirical efforts. We should note that in this review of previous research we will focus only on those studies that have Table 1 Hypothesized differences for charismatic, ideological and pragmatic leaders. Charismatic Ideological Pragmatic Time frame orientation Type of experienced used Nature of outcomes sought Number of outcomes sought Focus in model construction Locus of causation Controllability of causation Targets of inﬂuence Crisis conditions Use of emotions Future a Past a Present a Positive a Negative a Both a Positive b Transcendent Malleable a Multiple b Few b Variable a External Internal External People a Situations Interactive a High a Low Selective b Masses a Base Cadre Elites b Ordered Chaotic Localized Positive a Negative b Rational Note. Table represents predictions taken from Mumford and colleagues (Bedell-Avers, Hunter and Mumford, 2008; Mumford, 2006; Mumford & Strange, 2002). a Fully supported by the study. b Partially supported by the study. S.T. Hunter et al. / The Leadership Quarterly 22 (2011) 70–91 73 expressly examined the CIP model and not on those studies that are congruent with the model (e.g., Collins, 2001; Pasternack & O'Toole, 2002; Pasternack, Williams, & Anderson, 2001; Tsui et al., 2006; Weber, 1924). As mentioned earlier, one of the earliest studies examining pragmatic leadership was by Mumford and Van Doorn (2001) who examined 10 qualitative case examples of Benjamin Franklin's leadership approach. These cases included the initiation of the volunteer ﬁre department, introduction of paper currency, and development of the University of Pennsylvania. What is witnessed by considering these cases is the notable lack of charismatic or transformational behaviors exhibited by Franklin. Rather, each was initiated via a rational approach to problem-solving and an emphasis on convincing key players or “elites” Franklin saw as necessary to bring about change. Complimenting these results and expanding the CIP model was a study by Strange and Mumford (2002) who investigated differences between charismatic and ideological leaders using the historiometric approach (Simonton, 1990). Speciﬁcally, Strange and Mumford content coded the biographical information of 60 historical leaders, examining selections from both rise to power and in-power time periods. Their results indicated that the two leader types exhibited markedly different behaviors, particularly with regard to those behaviors associated with values, such as value autonomy and value commitment. 3.2. Compilation of studies The compelling results of these two articles and afﬁliated book chapters (e.g., Mumford & Strange, 2002) prompted the comprehensive examination of the CIP model in relation to a host of key organizational and societal outcomes. The methodology employed was historiometric, where 120 historical leader biographies were coded by multiple independent judges who examined a host of key behaviors and outcomes. The collection of studies was published in book format, and contained roughly six unique examinations of the leadership model (Mumford, 2006). For example, Mumford, Strange, Scott, Dailey, and Blair (2006) examined LMX relationship differences among the three leader types, observing that ideological leaders, in particular, depended on closefollower relationships to achieve their goals. Other topics included the investigation of the varying political tactics taken by charismatic, ideological, and pragmatic leaders. In their examination, Mumford, Licuanan, Marcy, Dailey, and Blair (2006) found that pragmatic leaders emphasized rational persuasion whereas charismatic and ideological leaders emphasized tactics that led to conditions of conﬂict and tension — conditions that more ideally suited their leadership styles. Although the book contains several other studies examining key leadership topics including problem-solving, communication strategies, and developmental inﬂuences, the above should sufﬁce to make our basic point: the CIP model was examined quite extensively in this collection. It must be noted that one particularly important observation emerging from the book was, in many ways, a key “non-ﬁnding.” Namely, the authors did not observe general performance differences among the three leader types. This point emerged as a key theme of the book: There are multiple pathways (i.e., styles or ways) to becoming an outstanding leader and we must expand our thinking to include those individuals that may not ﬁt current stereotypical molds — stereotypes often tied to vision-based forms of leadership. 3.3. Experimental studies Following the book and related articles, researchers also examined the CIP model in an experimental context. Bedell-Avers et al. (2008), for example, used a sample of undergraduates to examine differences in problem solving approaches among the three leader types. Participants were initially categorized as being ideological, pragmatic, charismatic, or mixed, and were presented with ambiguous problems from two different domains and within two different contexts. Participants were then asked to provide solutions to these problems which were then assessed for their quality and originality by an independent set of raters. Results indicated that, similar to the book, there were not overall performance differences among the three leadership styles. Rather, performance differences were observed only when the speciﬁc problem types were considered. Pragmatic leaders, for example, were generally more consistent and tended to be fairly adaptable problem solvers across study conditions. Ideological leaders, on the other hand, tended to succeed when they were clearly designated as a leader. Finally, charismatic leaders did well under problem-solving conditions that allowed for more ﬂexibility in their approach. Along similar lines, Hunter, Bedell-Avers, & Mumford (2009) examined performance differences among 240 undergraduate students who engaged in a computerized leadership simulation. The use of the simulation allowed researchers to manipulate the complexity of the task as well as how the problem was framed (e.g., as an ideological problem, pragmatic, problem or charismaticoriented problem). Results indicated that pragmatic leaders were generally consistent in their performance — performing well across all conditions in the study. Ideological leaders, on the other hand, had difﬁculty in conditions that conﬂicted with their beliefs and values, yet succeeded in fairly complex situations where their beliefs helped provide guidance amidst the chaos. Finally, charismatic leaders performed well in several conditions, but had difﬁculty in high-complexity contexts where it proved difﬁcult to solve problems using a largely future-oriented perspective. 3.4. Recent efforts The results of these early studies as well as various book chapters have given way to several recent studies exploring the CIP model. For example, Bedell et al. (2006) examined differences in Machiavellianism among the three leader types using historiometric analysis. The researchers found that pragmatic leaders were the most Machiavellian — likely due to their emphasis on ﬂexibility when attempting to solve problems (Mumford, 2006; Mumford & Van Doorn, 2001). Falling lower on the 74 S.T. Hunter et al. / The Leadership Quarterly 22 (2011) 70–91 Machiavellianism spectrum were the charismatic leaders and ﬁnally ideological leaders who were observed to be the least Machiavellian. The results were consistent with predictions given an ideological leader's uncompromising commitment to their beliefs and values. Another recent study conducted by Bedell-Avers et al. (2009) examined interactions among the three leader types. The researchers asked coders to systematically analyze multiple leader biographies of three leaders during the civil rights movement: W.E.B. Dubois (ideological), Frederick Douglas (charismatic) and Booker T. Washington (pragmatic). The results indicated that pragmatic leaders were the most ﬂexible, getting along with either leader type when it served their problem solving needs. Ideological and charismatic leaders, on the other hand, often conﬂicted quite ﬁercely with other leader types. Finally, Mumford, Antes, Caughron, & Friedrich (2008) recently provided a multilevel framework for understanding the individual, group, organization, and environmental conditions which facilitate the emergence and performance of the three leadership styles, providing some insight into when and where the leader types are most likely to come into power and succeed once positions of inﬂuence are attained. 3.5. Study impetus Upon consideration of the general methodologies employed in the above studies, a few key observations emerge. First, given the sheer volume of research, there is at least a preliminary indication that it is worthwhile to (re)consider Weber's views of management authority. At the same time, however, we must also be both careful and critical in our assessment of the CIP model as much of this research is relatively recent and as such, the model is still in its early stages of development; it is likely to need further reﬁnement and adjustment. Second, studies have relied upon a priori taxonomies of leadership styles (e.g., charismatic versus pragmatic), where leaders where initially categorized (or selected) by a set of judges as being charismatic, ideological and pragmatic, yet this coding process has not been examined in any great detail. As these categorizations were made early in the CIP model development, it is unclear if this coding process made use of any or all of the ten identiﬁed mental model differences (see Table 1) to classify these leaders. Moreover, in previous studies, once a leader was categorized a separate set of coders typically assessed various behavioral differences and outcomes among the three leader types (e.g., creative problem solving, political behavior). What has not been examined, however, is a direct comparison of the theorized mental model differences proposed in Table 1. Thus, the primary aim of this study was to examine if there are, as predicted, theorized mental model differences among the three leader types. A full list of directional hypotheses – derived from the CIP model – may be seen in Table 1. The secondary aim of the study was to aid in the revision and extension of the model by determining which components of the leaders' prescriptive mental models are used more or less heavily to differentiate leaders as being charismatic, ideological, or pragmatic. 4. The CIP model and domain Careful consideration of the previous studies examining the CIP model also reveals a trend that is both valuable and in some ways, problematic. Speciﬁcally, historical leaders were chosen from a wide variety of domains including political, business, military, and social justice. The use of domain breadth in sample selection is beneﬁcial in that it allows for some degree of external validity — that is, the results can be reasonably generalized to multiple domains due to the broad sampling of leaders. What is problematic in many of these studies, however, is that the three leader types tended to emerge from similar domains. For example, charismatic leaders often emerged from political arenas, pragmatic leaders from military or business, and ideological leaders were often associated with social justice endeavors. Thus, although there was a reasonable overlap in the sample (i.e., some military leaders were charismatic, some politicians were ideological), it is still unclear exactly how much variability in the results may be attributed to either domain or leadership style, speciﬁcally. This problem was addressed on a small scale by Bedell-Avers et al. (2009) in her examination of three leaders interacting in a common domain — the civil rights movement. To provide more data around these issues, our study examines multiple leaders within a common leadership domain: football. By examining leaders in the same arena, we can determine with increased accuracy the extent to which behavior may be attributed to the leader versus the domain from which they emerged. Before turning to the methodology and results of the study, however, it is useful to ﬁrst consider the use of sports teams in the study of leadership. The use of sports teams in research has been fairly common and appears to be reasonably accepted as a source for exploring related organizational and managerial phenomena (Avery, Tonidandel, Grifﬁth & Quinones, 2003). Giambatista (2004), for example, examined leader life cycles in NBA teams. Similarly, Dirks (2000) explored the relationship between leadership and trust using a sample of men's college basketball teams. Using NHL teams, Day, Sin and Chen (2004) examined role changes in leadership among team captains and how these changes impacted performance over time. Other studies have examined leadership and related phenomena in baseball teams (Hawkins, & Tolzin, 2002), NBA teams (Avery et al., 2003), softball teams (Shields, Gardner, Bredemeier, & Bostro, 1997), and mixed sports teams (Charbonneau, Barling, & Kelloway, 2001). More central to the present effort, Garland and Barry (1990) examined Chellandurai and Carron's (1978) multidimensional theory of leadership by measuring coaches' personality and leader behaviors in relation to football player performance. Results indicated that coaches had a signiﬁcant impact on the rated performance of the player over the course of the season. As a whole, the above studies highlight an important point — leadership is readily observable in sports teams and, it would seem, the results have applicability to other applied forms of management. S.T. Hunter et al. / The Leadership Quarterly 22 (2011) 70–91 75 5. Method 5.1. Sample The historiometric sample used in the study was comprised of championship winning NFL and college head football coaches. Although the subordinates may differ somewhat from college to professional ranks, the general leadership tasks associated with coaching at the two levels are roughly equivalent — albeit with some notable differences (e.g., slight rule changes, increases in pay). This relative comparability is evidenced by several coaches who have made the transition from college to professional and vice versa (professional to college). Moreover, most professional football players (i.e., subordinates) begin in the college ranks, further providing illustration of continuity between the two levels. Finally, the use of both college and NFL coaches allows us to increase our sample size and, therefore, our capacity to test the study hypotheses. Along these lines, it should be noted that this variable was coded and examined as a potential moderating variable in our analyses. The use of championship caliber coaches ensures that we have a sample of proven “outstanding” leaders within a common domain thereby providing some additional control for the study. In an attempt to also control for time period, we began our search for a sample of coaches starting with the year 1967 — the ﬁrst year of the NFL super bowl (championship game). Although college teams played before this time, based on suggestions by Simonton (1990) we felt it important to control for time and only chose college coaches that won (or tied) national championships from 1967 on. This relatively recent sample of leaders may also provide increased applicability to current and future leaders. It should also be noted that the designation of a clear national champion in college football is a fairly recent phenomenon. Prior to 1992 there was no systematic attempt to allow the best teams in the country to play one another allowing for the emergence of a clear national champion. As a result, in some years there were two national champion head coaches. In these instances, a search was made for the multiple coaches who would reasonably be viewed as outstanding leaders in these shared championship years. This initial search resulted in 66 coaches identiﬁed for possible inclusion in the study. Once our list was comprised, an attempt was made to obtain historical biographies of the outstanding coaches. In some instances, multiple biographies were available and were retained. In other instances, no biography was available and this coach was dropped from the sample. Once the biographies were obtained, they were examined for two major criteria. First, they had to contain a chapter, or preferably multiple chapters, where the leader's approach to interacting, motivating, and coaching his subordinates was detailed. If a book did not contain such information, it was dropped from the study. Second, the chapter had to be of reasonably high quality and provide detailed descriptions of how the leader behaved. Not surprisingly, we found that many coaches' books contained a high amount of superﬂuous information (i.e., pictures, anecdotes, etc.) and in those instances the books were also dropped from the study. The end result was 103 biographies examining 54 coaches (presented in Appendix A). Of these, 29 were college coaches and 25 were NFL coaches. This ﬁnal sample is presented in Table 2. Once the biographies were selected, two independent researchers selected book chapters that best represented a coach's leadership approach. More speciﬁcally, chapters were selected if they 1) contained descriptions of how the coach interacted with players and 2) provided a description of a coach's philosophy toward leading and/or coaching and 3) could be read and understood by raters with little or no football background. There was 80% agreement on chapter selection. In the cases of disagreement, the two researches conferred until a consensus was reached. On average, selected material was 24 pages long and contained 10,351 words. 5.2. Coding procedures With chapters selected, work began on coding the biographical material. Prior to beginning ﬁnal ratings four independent judges, all graduate students in Industrial and Organizational Psychology, participated in 20 h of rater training. Raters were chosen, in part, based on their lack of familiarity with the football domain and, more speciﬁcally, lack of familiarity with the coaches in the study. This limited some of the bias associated with making ratings about known leaders. Upon satisfactory completion of training (ICC above .80 on sample rating materials), raters assessed the biographies for the mental model differences proposed by the CIP model. For example, to assess differences in the “time-frame” mental model component, coders were asked to provide ratings on three questions: 1) to what extent does this leader focus on the future, 2) to what extent does this leader focus on the present and 3) to what extent does the leader focused on the past? See Appendix B for all rating scale items used in the study. By posing these questions as continuous scales we were able to analyze the data using more traditional statistical techniques and avoid the problems often associated with ipsative scales (Chan, 2003). Similar questions were asked for each of the 10 proposed mental model differences. Using ICCs, the average interrater reliability was .78 across all rating material. With respect to the use of multiple biographies, results indicated very consistent results across biographical information (average ICC = .94). Thus, in the instances of multiple biographies for coaches, information was aggregated resulting in a single set of scores for each coach. In addition to the initial four raters, three supplemental raters were given the chapters and asked to simply categorize the leader as being charismatic, ideological, or pragmatic. These raters were also graduate students in Industrial and Organizational Psychology with notable interest and emerging expertise in the study of leadership — although this was their ﬁrst direct exposure to the CIP model. Following protocol set forth by Mumford and colleagues, leaders were also categorized as being personalized or socialized (House & Howell, 1992) — a variable later used as a covariate to control for differences in orientation. These raters were not given material on the theoretical framework of the CIP model and were provided the list of 120 leaders and their respective categorizations used in the Mumford (2006) book. Decisions about ﬁnal categorization were made by using the selection made by 76 S.T. Hunter et al. / The Leadership Quarterly 22 (2011) 70–91 Table 2 Classiﬁcation results. Coach Leader type Coaching level Bill McCartney Bill Walsh Bob Stoops Bobby Bowden Brian Billick Bud Wilkinson Chuck Noll Dan Devine Danny Ford Dennis Erickson Dick Vermeil Don Shula Jim Tressel Joe Paterno John Robinson Lavell Edwards Lloyd Carr Lou Holtz Mack Brown Mike Shanahan Nick Saban Pete Carroll Tom Osborne Vince Dooley Duffy Daugherty George Seifert Hank Stram Joe Gibbs John Madden John McKay Johnny Majors Mike Ditka Mike Holmgren Paul “Bear” Bryant Phillip Fulmer Steve Spurrier Tom Landry Vince Lombardi Ara Parseghian Barry Switzer Bill Belichick Bill Cowher Bill Parcells Bob Devaney Darrell Royal Don James Gene Stallings Gene Stallings Jimmy Johnson Jon Gruden Tom Flores Urban Meyer Weeb Ewbank Woody Hayes Charismatic Charismatic Charismatic Charismatic Charismatic Charismatic Charismatic Charismatic Charismatic Charismatic Charismatic Charismatic Charismatic Charismatic Charismatic Charismatic Charismatic Charismatic Charismatic Charismatic Charismatic Charismatic Charismatic Charismatic Ideological Ideological Ideological Ideological Ideological Ideological Ideological Ideological Ideological Ideological Ideological Ideological Ideological Ideological Pragmatic Pragmatic Pragmatic Pragmatic Pragmatic Pragmatic Pragmatic Pragmatic Pragmatic Pragmatic Pragmatic Pragmatic Pragmatic Pragmatic Pragmatic Pragmatic College NFL College College NFL NFL NFL College College College NFL NFL College College College College College College College NFL College College College College College NFL NFL NFL NFL College College NFL NFL College College College NFL NFL College NFL NFL NFL NFL College College College College NFL NFL NFL NFL College NFL College at least 2 of the 3 judges. There was complete agreement for categorization for approximately 65% of the coaches. There were no instances of complete disagreement, where each judge selected a differing categorization (i.e., judge 1 chose charismatic, judge 2 chose pragmatic and judge 3 chose ideological). Coaches and their respective categorizations are presented in Table 2. When considering the use of historiometric analysis, a few points should be born in mind. First, Hunter, Bedell-Avers and Mumford (2007) have recently argued that to move beyond our current understanding of leadership, we must be creative in our research approaches and turn to alternative methods of assessment and investigation. The historiometric approach stands as one of those unique and emerging methods that allows for the examination of a leader sample not attainable using more traditional approaches such as surveys or even direct interviews. This method, however, also brings with it a share of challenges including author bias and variability in material selected for coding. Fortunately, several of these biases can be controlled for (Simonton, 1990; Mumford, 2006) allowing for more substantive conclusions to be drawn. Thus, the second group of coders also rated eight control variables, including: assessments about the bias of the biographer, nature of the book (autobiographical or biographical), S.T. Hunter et al. / The Leadership Quarterly 22 (2011) 70–91 77 percentage of quoted material, overall quality of the material presented, author's view of the leader, total number of pages, and age of coach at the time of the book. The average ICC across covariates was .91. 5.3. Analyses Data were analyzed using two primary procedures. The ﬁrst was a MANCOVA, used to determine if there were general mean differences in expressions of the leaders' mental models across the three leader types. Follow-up univariate analyses were also conducted to examine the speciﬁc differences in the dependent variables. Moreover, to more accurately assess cell mean differences, 95% conﬁdence intervals were also formed — a task necessary as the application of covariates precludes the use of multiple comparison procedures (Toothaker, 1993). The second primary analysis performed was a discriminant function analysis which was used to determine which components of the leader's expressed mental models most strongly predicted discrimination among the three leader types. 6. Results Descriptive results including means, standard deviations and correlations are presented in Table 3. The results of the MANCOVA are presented in Table 4. With regard to control variables, all covariates were entered simultaneously and assessed for their applicability to the analyses. To maximize degrees of freedom (Tabachnick & Fidell, 2001), only those covariates signiﬁcant at p b .20 were retained in ﬁnal multivariate and univariate analyses, resulting in three retained control variables: leader orientation (personalized versus socialized), degree of author bias, and amount of quoted material in the selection. As may be seen in the table, there was an overall signiﬁcant main effect for leader type F(58, 34). It should also be noted that the interaction term between level of coaching was not signiﬁcant, although it did approach traditional levels of signiﬁcance (F(58, 34) = 4.428, p = .09). Closer inspection of the univariate results, moreover, revealed that across 29 dependent variables only 1 interaction reached signiﬁcance. As such, when taken in conjunction with results from previous studies it appears that the CIP model is relatively robust within the football domain and that the results are applicable across both professional and college samples. 6.1. Leader type Univariate results for leader type are presented in Table 5, along with associated means for the three leader types. One interesting set of results not appearing in the descriptive statistics is the frequency of leader types occurring within college versus professional arenas. In the college domain, the number of leader types was as follows: charismatic = 16, ideological = 6, and pragmatic = 7. Within the professional ranks, the breakdown of leader types was: charismatic = 8, ideological = 8, and pragmatic = 9. Although strong conclusions cannot be drawn from these frequency values (χ2 = 2.92, p = .23), the prevalence of charismatic leaders in the college ranks appears noteworthy. Given that college coaches are essentially leading non-professional subordinates, these results are consistent with others (e.g., Howell & Shamir, 2005) who have suggested that charismatic leaders are best suited to contexts where subordinates are open to guidance and direction via a charismatic leader's vision. 6.1.1. Time frame orientation As may be seen when examining univariate results for leader type in Table 5, results are consistent with predicted theoretical frameworks. Speciﬁcally, charismatic leaders were more future oriented, pragmatic leaders were more present focused and ideological leaders focused on past events — results well illustrated in Fig. 1. Given these results, there appeared to be strong support for this component of the theory. 6.1.2. Type of experience used With regard to the type of experience used, charismatic leaders focused on more positive experiences than ideological leaders — yet did not differ substantially from pragmatic leaders who also appeared to draw on positive experiences when they deemed it necessary. As predicted (see Table 1), ideological leaders used negative experiences signiﬁcantly more than charismatic and pragmatic leaders (Table 5). Finally, pragmatic leaders changed the type of experiences used more often than charismatic leaders, yet did not differ from ideological leaders in this way. These results highlight general support for the theory regarding this component of the CIP model. 6.1.3. Nature of outcomes sought The CIP model predicts that charismatic leaders will seek positive outcomes, ideological leaders will seek transcendent outcomes and pragmatic leaders will be malleable or ﬂexible in the nature of the outcomes sought. Overall, there was general support for this aspect of the model. Charismatic leaders produced the highest means on with respect to seeking positive outcomes and ideological leaders sought transcendent outcomes to a greater degree than pragmatic leaders but not charismatic leaders. Finally, pragmatic leaders produced higher means with respect to being malleable in the nature of the outcomes sought — producing higher means than both ideological and charismatic leaders. 78 Table 3 Means, standard deviations and intercorrelations among study variables. General category Time frame orientation Type of experience used Nature of outcomes sought Number of outcomes sought Focus on model construction Locus of causation Controllability of causation Targets of inﬂuence Crisis conditions Use of emotions Covariates Variable Mean SD 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Future Present Past Positive Negative Both Positive Transcendent Malleable Multiple Few Variable External Internal People Situations Interactive High Low Selective Masses Base Cadre Elites Ordered Chaotic Localized Positive Negative Rationale Autobiography Quoted material Quality of material Years as a coach Author's view of leader Number of words a Age of coach Amount of bias Personalized/socialized 3.49 3.53 2.57 3.73 2.71 3.49 4.66 3.91 2.28 3.61 2.42 2.09 3.01 4.13 3.98 2.66 3.45 4.45 1.37 1.81 3.97 3.22 2.61 4.26 1.86 3.21 3.66 2.53 3.78 .63 2.06 3.71 3.25 4.21 103.51 56.15 3.19 .26 .80 .60 .78 .81 .76 .67 .35 .69 .64 .89 .82 .66 .57 .43 .56 .46 .60 .53 .35 .48 .65 .66 .56 .47 .52 .54 .74 .83 .59 .55 .43 .51 .84 .54 3.47 7.98 .72 .46 1.00 − .08 − .31 .53 − .17 − .18 .46 .24 − .17 .45 − .30 − .19 .00 .20 .21 − .13 − .20 .31 − .27 − .39 .31 − .09 .01 − .01 − .09 .23 .27 − .14 .03 .09 − .02 .06 − .03 .03 .03 − .01 .13 − .22 1.00 − .22 .30 − .07 .09 − .04 − .11 .03 − .02 .11 .19 − .02 .21 − .03 − .14 .01 − .13 − .19 − .01 .04 − .20 − .06 .14 .01 .06 − .10 − .18 .11 − .24 .02 .09 − .03 .22 .16 − .04 .29 .21 1.00 − .36 .40 .08 − .23 .13 .35 − .11 .12 .16 .36 − .37 − .26 .32 .23 − .25 .18 .32 − .14 .01 − .01 − .16 − .46 − .16 .16 .43 − .11 .35 .02 − .15 .11 − .11 − .13 .08 − .24 .08 1.00 − .36 .40 .08 − .23 .13 .35 − .11 .12 .16 .36 − .38 − .27 .33 .23 − .25 .18 .32 − .14 .00 − .01 − .16 .15 − .46 − .16 .43 − .11 .35 .02 − .15 .11 − .11 − .13 .08 − .24 1.00 − .42 − .18 .50 .24 − .13 .50 − .33 − .19 − .09 .26 .29 − .12 − .18 .11 − .26 − .27 .52 − .39 − .13 .10 − .07 .04 .43 − .37 .02 .07 − .03 − .11 .05 .09 .00 .02 .22 1.00 .43 − .45 − .19 .27 − .35 .51 .17 .36 − .16 − .21 .04 − .02 − .03 .17 .28 − .33 .25 .16 − .34 .42 − .14 − .41 .70 − .01 .13 − .04 .19 − .06 − .24 .10 .03 − .13 1.00 − .16 − .19 .13 − .34 .24 .19 .16 − .23 − .07 − .11 .15 .07 − .08 .09 − .15 .16 .36 .01 .00 − .23 − .28 .34 .06 .10 − .05 .22 − .16 − .13 .11 − .09 − .08 1.00 .44 − .13 .37 − .48 − .09 − .18 .31 .29 .05 .09 .13 − .10 − .36 .43 − .18 − .06 .24 − .28 .10 .31 − .41 .11 .11 − .05 .01 .02 .08 − .14 .08 .13 1.0 − .04 .58 − .53 − .22 .10 .08 .10 .16 .11 .19 − .03 .04 .37 − .26 − .17 .02 .03 − .13 .13 − .10 − .27 .31 − .10 − .22 .10 .19 .04 − .01 .15 1.00 − .03 .30 .41 .57 − .23 − .38 .38 .49 − .44 .49 .38 − .12 .14 .21 − .43 .14 − .50 − .30 .28 .32 .14 − .04 − .15 − .03 − .18 − .13 − .03 − .23 1.00 − .73 − .22 .07 .17 .13 .05 .08 .20 − .16 − .12 .41 − .40 − .08 .20 − .18 − .04 .38 − .25 − .15 .15 − .15 − .16 .24 .23 .00 .15 .16 1.00 .23 .16 − .22 − .22 − .02 − .07 − .30 .23 .32 − .50 .26 − .11 − .41 .29 − .06 − .42 .39 .16 − .06 − .01 .03 − .22 − .45 .11 − .24 − .32 1.00 .28 − .21 − .28 .12 .45 − .15 .23 .42 − .18 .13 .19 − .12 − .09 − .33 − .09 .05 .36 − .02 .02 .03 − .15 − .18 .00 − .06 − .12 1.00 − .36 − .46 .48 .56 − .40 .42 .49 − .03 .07 .22 − .25 .13 − .53 .05 .21 .11 .15 − .05 − .09 − .03 − .25 .05 − .02 − .06 1.00 .37 − .41 − .32 .45 − .13 − .26 .30 − .18 − .16 .27 .04 .39 .20 − .32 − .09 − .26 .24 .38 .30 .34 .12 .40 .50 1.00 − .28 − .36 .49 − .40 − .37 .25 − .17 .02 .27 − .20 .32 .25 − .28 − .16 .04 − .03 .09 .04 .10 − .07 .14 .14 Note. Correlations signiﬁcant at p ≤ .05 if above the absolute value of .27. a Number of words presented in 1000 U. S.T. Hunter et al. / The Leadership Quarterly 22 (2011) 70–91 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 Table 3 Means, standard deviations and intercorrelations among study variables. 18 19 20 21 22 23 24 25 26 27 28 29 30 31 33 34 35 36 37 38 1.0 .56 − .52 .41 .33 − .15 .13 .02 − .39 .16 − .58 .10 .03 .16 .21 .15 − .31 − .18 − .16 − .03 − .27 − .30 1.00 − .44 .39 .35 .02 .10 .17 − .19 − .04 − .56 − .04 − .02 .14 .11 − .09 − .19 − .04 − .12 − .06 − .09 − .15 1.00 − .45 − .42 .19 − .03 .11 .48 − .27 .37 .06 .02 − .16 − .04 .07 .12 .08 .10 .15 .08 .20 1.00 .47 − .05 .06 − .06 − .34 .22 − .27 − .01 .01 .27 .04 − .02 − .02 .03 − .08 − .07 .02 − .15 1.00 − .16 − .10 − .14 − .32 .21 − .44 − .08 .23 .14 .12 − .01 − .05 − .03 − .08 .13 − .15 − .09 1.00 − .34 − .04 .08 .07 .00 .39 − .21 − .18 − .03 .09 .00 .14 .26 − .17 .16 .27 1.00 .66 − .23 .18 − .03 − .19 .21 − .08 − .13 .12 .05 − .10 − .15 .06 − .05 − .19 1.00 .01 − .02 − .12 .05 .11 .03 − .08 .03 .14 .03 .02 .00 .13 − .04 1.00 − .74 .39 .13 − .23 − .06 − .14 − .17 .20 .40 .18 .25 .36 .33 1.00 − .20 − .06 .22 − .17 .03 .40 .04 − .26 .02 − .19 − .13 − .03 1.00 − .02 − .09 − .11 − .22 − .07 .39 .09 .21 .16 .13 .25 1.00 − .67 − .05 .10 .13 .06 .01 .08 − .03 .06 .15 1.00 − .12 .09 − .04 .04 .00 − .25 .05 − .05 − .26 1.00 .09 − .13 .01 − .08 − .04 .05 − .12 − .08 1.00 − .07 − .29 .00 − .14 .27 − .24 − .24 1.00 .21 − .24 .07 − .09 − .10 .01 1.00 .14 .32 .09 .44 .42 1.00 .20 .31 .77 .21 1.00 − .01 .27 .73 1.00 − .12 − .02 1.00 .46 S.T. Hunter et al. / The Leadership Quarterly 22 (2011) 70–91 17 79 80 S.T. Hunter et al. / The Leadership Quarterly 22 (2011) 70–91 Table 4 Multivariate analysis of covariance for proposed mental model differences. Covariates Orientation Author bias Quoted material Main effects Level (college or pro) Leader type Interaction Level ⁎ leader type F df p η2p 3.02 1.54 1.57 29, 17 29, 17 29, 17 .01 .17 .17 .83 .73 .73 7.12 4.85 29, 17 58, 34 .00 .00 .92 .89 1.42 58, 36 .09 .71 Note. F = F-value using Pillai's trace; df = degrees of freedom; p = p-value; η2p = partial eta squared effect size. 6.1.4. Number of outcomes sought As predicted by the theoretical framework, we found that charismatic leaders sought more positive outcomes than ideological leaders. In contrast to theorized trends, however, charismatic leaders did not signiﬁcantly differ from pragmatic leaders in this way, although means were in the predicted directions (Table 5). With regard to transcendent outcomes sought, means were again in predicted directions, but ideological leaders did not signiﬁcantly differ from either pragmatic or charismatic leaders. Finally, consistent with the proposed theoretical framework, pragmatic leaders were more malleable in their agendas for goal attainment, differing signiﬁcantly from both ideological and charismatic leaders. Thus, as a whole, there is moderate support for the theory regarding this component of the model. 6.1.5. Focus on model construction According to the CIP model, charismatic leaders and pragmatic leaders are predicted to focus on external factors when constructing their prescriptive mental models in contrast to ideological leaders, who are theorized to be more internally focused. Results of the univariate analyses, however, failed to support this trend and in fact revealed a nearly opposite effect (Table 4). More speciﬁcally, with respect to an external focus, pragmatic leaders produced the highest means followed by ideological leaders, and ﬁnally results indicated that charismatic leaders produced signiﬁcantly lower means than both of the other leader types. As may be surmised given the nature of the items, a similar pattern – albeit in the opposite direction – was observed for internal focus. Notably, with internal focus, ideological leaders produced the lowest means when theory predicted they would produce the highest. Thus, these results do not support the pattern predicted by the CIP model. 6.1.6. Locus of causation As predicted, univariate results revealed that charismatic leaders viewed people as the core of outcome causality to a greater degree than ideological and pragmatic leaders. With regard to viewing the situation as the primary cause of outcomes, ideological leaders produced signiﬁcantly higher means than charismatic leaders but not pragmatic leaders (Table 5). Finally, pragmatic leaders largely viewed causality as the interaction between people and the situation, producing signiﬁcantly higher means than both of the other leader types for this dependent variable. Overall, then, there appears to be general support for the locus of causation component of the CIP theoretical framework. 6.1.7. Controllability of causation The CIP model predicts that charismatic leaders will view the situation as highly controllable — that it can be changed, altered, or adjusted. The results were consistent with this prediction; charismatic leaders producing means higher than both ideological and pragmatic leaders. The model also predicts that ideological leaders will view the situation outside of their control, attributing the context to external drivers such as fate or destiny. The results supported this prediction relative to charismatic leaders but not pragmatic leaders who actually produced means greater than ideological leaders. Finally, pragmatic leaders were predicted to be selective in how they viewed the situation — at times seeing it as controllable, at others seeing it as out of their hands. Means trend in predicted direction, with the strongest support being for pragmatic leaders compared to charismatic leaders. Overall, there is some support for this aspect of the CIP model. 6.1.8. Targets of inﬂuence According to the CIP model, charismatics are theorized to appeal to the masses, while ideological leaders are theorized to appeal to a core base of close individuals. In contrast, pragmatic leaders are theorized to appeal to elites, or those in a position to have substantive inﬂuence in a broader social context. Within the football domain, elites were deﬁned as the star players on the team, regardless of rank or seniority. Generally speaking, the results of this model component support these propositions — charismatic leaders produced means highest on appeals to the masses (Table 4). Similarly, ideological leaders appealed to their core followers to a greater degree than both pragmatic and charismatic leaders. With respect to the football domain directly, core followers were players on the team for a long period of time and were strongly enculturated into the coach's philosophy. Finally, 81 S.T. Hunter et al. / The Leadership Quarterly 22 (2011) 70–91 Table 5 Univariate results for leader type. Dependent variables F Time frame orientation Future 6.62 Present 6.12 Past 9.96 Type of experience used Positive 16.33 Negative 6.25 Both 3.60 Nature of outcomes sought Positive 3.50 Transcendent 1.09 Malleable 6.35 Number of outcomes sought Multiple 13.69 Few 3.37 Variable 4.75 Focus on model construction External 5.64 Internal 6.51 Locus of causation People 4.54 Situations 2.09 Interactive 10.19 Controllability of causation High 9.22 Low 4.06 Selective 8.19 Targets of inﬂuence Masses 3.22 Base Cadre .624 Elites 3.07 Crisis conditions Ordered 1.36 Chaotic .186 Localized 3.96 Use of emotions Positive 7.17 Negative 6.75 Rationale 1.84 df p η2p Charismatic mean (SE) Ideological mean (SE) Pragmatic mean (SE) 2, 45 2, 45 2, 45 .01 .01 .01 .22 .21 .31 3.96 (.16) 3.46 (.18) 2.20 (.15) 3.08 (.21) 3.20 (.14) 3.29 (.14) 3.16 (.18) 3.84 (.13) 2.42 (.17) 2, 45 2, 45 2, 45 .01 .04 .04 .42 .21 .14 4.14 (.13) 2.43 (.14) 3.27 (.15) 2.95 (.16) 3.21 (.17) 3.53 (.19) 3.81 (.14) 2.59 (.15) 3.84 (.17) 2, 45 2, 45 2, 45 .04 .34 .01 .14 .05 .22 4.79 (.07) 4.00 (.14) 1.93 (.14) 4.47 (.09) 3.90 (.18) 2.40 (.17) 4.67 (.08) 3.70 (.16) 2.70 (.15) 2, 45 2, 45 2, 45 .01 .04 .01 .38 .13 .17 4.13 (.13) 2.10 (.16) 1.84 (.14) 3.11 (.16) 2.70 (.20) 2.09 (.18) 3.28 (.14) 2.64 (.18) 2.51 (.16) 2, 45 2, 45 .01 .01 .20 .22 2.71 (.12) 4.29 (.07) 3.10 (.15) 3.86 (.09) 3.33 (.14) 4.05 (.08) 2, 45 2, 45 2, 45 .02 .14 .01 .17 .09 .31 4.26 (.12) 2.49 (.10) 3.09 (.12) 3.86 (.09) 2.80 (.13) 3.46 (.15) 4.04 (.08) 2.75 (.11) 3.93 (.14) 2, 45 2, 45 2, 45 .01 .05 .01 .29 .15 .26 4.78 (.10) 1.20 (.08) 1.52 (.10) 4.30 (.13) 1.43 (.10) 2.03 (.12) 4.01 (.12) 1.54 (.09) 2.07 (.11) 2, 45 2, 45 2, 45 .05 .54 .06 .13 .03 .12 4.13 (.12) 3.16 (.12) 2.62 (.12) 3.60 (.15) 3.25 (.15) 2.39 (.15) 3.93 (.14) 3.37 (.13) 2.88 (.13) 2, 45 2, 45 2, 45 .27 .83 .03 .06 .01 .15 4.41 (.10) 1.81 (.11) 3.49 (.12) 4.15 (.13) 1.92 (.14) 2.99 (.15) 4.21 (.11) 1.85 (.12) 3.06 (.13) 2, 45 2, 45 2, 45 .01 .01 .17 .24 .23 .08 4.09 (.15) 2.20 (.17) 3.69 (.14) 3.13 (.19) 3.19 (.21) 3.71 (.17) 3.57 (.17) 3.37 (.19) 4.05 (.15) Note. F = F-value; df = degrees of freedom; p = p-value; η2p = partial eta squared effect size. Covariates used in analyses: leader orientation, degree of author bias and use of quoted material. mean trends suggest that pragmatic leaders attempted to appeal to the elites to a greater degree than both charismatic and ideological leaders — although this difference was only signiﬁcant when comparing means to charismatic leaders. On the whole, there was relatively strong support for this component of the model. Fig. 1. Mean results for time frame orientation. 82 S.T. Hunter et al. / The Leadership Quarterly 22 (2011) 70–91 6.1.9. Crisis conditions With respect to crisis conditions, the CIP model predicts that charismatic leaders emerge and succeed in more ordered conditions, ideological leaders in more chaotic conditions, and pragmatic leaders in more localized conditions. Results, however, do not support this trend. In fact, for the one dependent variable that was signiﬁcant, localized conditions, means are in directions opposite from what is predicted by the model. Thus, there was no support for this component of the model — although as will be noted in the limitation section, the crises faced by coaches may represent a unique form of crisis relative to those faced by other high-level leaders. 6.1.10. Use of emotions Although not explicitly described as a core prescriptive mental model component, the use of emotions when providing sensemaking is discussed as a critical element for both charismatic and ideological leaders. In contrast, we predicted that pragmatic leaders would use rational appeals rather than emotional appeals to followers. Moreover, given the emphasis on presenting an idealized future, for charismatics we predicted a greater use of positive emotions and given the emphasis on righting the wrongs of the past, we predicted a greater use of negative emotions for ideological leaders. Univariate results for all three dependent variables were generally as predicted, providing support for predictions regarding the use of emotions (Table 5). 6.1.11. Summary Overall, there was fairly strong support for the components of the CIP framework. More speciﬁcally, of the 10 components analyzed, roughly 8 were supported — albeit to varying degrees. When examining the observed effect sizes, these differences were not trivial. These mean differences, however, tell us only part of the story. It is useful, then, to examine which components are most heavily used when discriminating among leader types. We turn now to the results of the discriminant function analysis. 6.2. Discriminant function analyses The results of the discriminant function analyses revealed a number of interesting trends. First, it should be noted that both discriminant functions were signiﬁcant: F1 (r = .93, χ2 = 137.67, p ≤ .001) and F2 (r = .91, χ2 = 63.89, p ≤ .001). Given the signiﬁcance of both functions, it seems relevant to consider each of them to gain a better understanding of what aspects of the leaders' mental models were most heavily used by coders when categorizing the three leader types. Before examining function loadings, however, it is critical to understand what each function represents — a task requiring careful consideration of the group mean centroids. These means, along with other requisite function statistics are presented in Table 6. As may be seen in the table, the function means indicate that the ﬁrst function best represents the distinction of ideological leaders (M = − 4.13) relative to both pragmatic (M = 1.51) and charismatic leaders (M = 1.40) as a whole, who scored very similarly on this function. Again, these means help inform interpretation on the factor loadings, where high scores on this function are indicative of charismatic and pragmatic leaders, while low (negative scores) are representative of ideological leaders. Not surprisingly, the second function essentially represents the distinction between charismatic (M = − 2.01) and pragmatic leaders (M = 2.94) with ideological leaders scoring near zero on the function (M = .07). Again, these means help us understand that high scores are indicative of pragmatic leaders and lower scores (negative loadings) are more indicative of charismatic leaders. Inspection of the factor loadings, presented in Table 7, also highlights a number of interesting trends. When examining ideological leaders, for example, it is evident that what separates them from their charismatic and pragmatic counterparts is a strong emphasis on the past (F1 = − .27), a focus on a few key outcomes (F1 = −.14), and a reasonably high amount of negative emotionality (F1 = − .24). When considering what separates charismatic leaders from ideological and pragmatic leaders, on the other hand, is seems that a strong emphasis on positive experiences (F1 = .36, F2 = −.15), seeking multiple outcomes (F1 = .18, F2 = − .30), offering their vision to a wide array of individuals (F1 = .20, F2 = − .17), and using a strong amount of positive emotions (F1 = .19, F2 = − .16) best categorizes this leader type. Finally, pragmatic leaders appeared fairly malleable, using positive (F1 = .36), negative (F1 = .23, F2 = .12) and combinations (F2 = .21) of experiences when necessary. In addition, pragmatic leaders appear to vary the number of outcomes sought (F2 = .21), and viewing the locus of causality as interactive (F2 = .23). Finally, pragmatic leaders also placed a reasonable amount of focus on the present — something that separated them Table 6 Discriminant function results. Leader type Charismatic Ideological Pragmatic Canonical correlation Wilks lambda Chi-square p-value Function 1 Group centroids Function 2 Group centroids 1.40 − 4.13 1.51 − 2.01 .07 2.94 Function 1 Function 2 .93 .02 137.68 .00 .91 .18 63.89 .00 83 S.T. Hunter et al. / The Leadership Quarterly 22 (2011) 70–91 Table 7 Discriminant function analyses structure matrix. Dependent variable Time frame orientation Future Present Past Type of experience used Positive Negative Both Nature of outcomes sought Positive Transcendent Malleable Number of outcomes sought Multiple Few Variable Focus on model construction External Internal Locus of causation People Situations Interactive Controllability of causation High Low Selective Targets of inﬂuence Masses Base Cadre Elites Crisis conditions Ordered Chaotic Localized Use of emotions Positive Negative Rationale Function 1 Loadings Char and Prag versus Ideo Function 2 Loadings Char versus Prag − .15 .16 − .27 − .25 .15 .03 .36 .23 − .02 − .18 .12 .21 .18 .02 − .03 − .15 − .18 .21 .18 − .14 .00 − .30 .23 .25 − .03 .22 .19 − .16 .12 − .07 .01 − .18 .07 .23 .09 − .05 − .10 − .28 .15 .20 .20 .06 − .08 − .17 .17 .16 .09 − .04 .09 − .10 .06 − .11 .19 − .24 .04 − .16 .07 .11 from both ideological and charismatic leaders (F1 = .16, F2 = .15). On the whole, these results are consistent with predictions, but are also unique in that they reveal which variables were used more heavily than others in the categorization process used by coders. Going further, there is at least some indication that these factor loadings may be reﬂective of which aspects of the prescriptive mental models may be the most salient or distinctive components of the differing types of leaders. 7. Discussion Before turning to the broader implications of the study, it is important to bear in mind a few limitations. The ﬁrst is the somewhat small sample size of football coaches used in the study, which may be viewed as limiting the capacity to observe predicted effects afﬁliated with the model. Although we concede that the sample size is not overwhelming, we do feel justiﬁed with the sample size given that the hypothesized differences examined in this study were the fundamental tenants of the CIP model. As such, these differences were predicted to produce large effect sizes which would require a smaller sample size to detect (Cohen, 1992). Power analyses, furthermore, indicated that the sample size was adequate for examining effects of this predicted size. It should also be noted that although the ﬁnal sample was comprised of 54 coaches, over 100 leader biographies were rated in the coding process. Finally, because results were generally consistent with the predicted model, there is substantive evidence that the sample size was large enough to examine the proposed effects. Second, although the use of leaders from a single domain was an intended aspect of the study, caution is warranted when generalizing the results to domains outside the football arena. In particular, the crises and associated emotional contexts faced by leaders may be somewhat unique with respect to other forms of outstanding leadership; every game may feel like a crisis and emotions typically run very high in sports contexts — particularly the very physical and frequently violent sport of football. When taken in conjunction with previous studies examining the CIP model, however, we do feel that the results of the study have reasonable applicability to other forms of leadership. Moreover, given the prevalence of leadership books and materials afﬁliated with sports 84 S.T. Hunter et al. / The Leadership Quarterly 22 (2011) 70–91 coaches, it seems reasonable to suggest that there may be some leadership lessons to be learned from the sports arena — even if in limited fashion (e.g., Avery et al., 2003; Day et al., 2004; Lansing, 1989). Third, rated material assessing the mental model differences among leader types was tested against categorizations made by a second set of judges and it is important to note that this second set of judges should not be taken as the “correct” categorization. Indeed, it is likely that miscategorizations were made by these coders when they were asked to assess a leader as being charismatic, ideological or pragmatic. However, given the predicted directionality of the results, miscategorizations are likely to have lessened or negated the observed results (i.e., type II errors). As such, the ﬁndings of the study are likely to be more conservative, rather than liberal, estimates. Moreover, because the observed patterns were generally consistent with the CIP model, there is at least partial evidence of rater capacity to make valid judgments about leader types. Fourth, along similar lines, leader types were assessed by the second set of coders in a categorical manner. That is, judges were instructed to code a leader as being ideological, pragmatic, or charismatic rather than rating charisma, ideology, or pragmatism on a continuum. Moreover, there is some evidence that mixed model types may exist (e.g., Bedell-Avers et al., 2009; Hunter et al., 2009). Once again, this miscategorization (e.g., coding a leader as pragmatic when they are mixed) is likely to have introduced greater error into analyses, thereby resulting in a reduction of effects (type II errors) rather than more liberal estimates (type I errors). Because results are consistent with predictions, it seems that most type II errors were avoided. The point remains, however, that some leader types may be mixed and this area stands as an important avenue for future research. Fifth and ﬁnally, the use of biographical information is not without its share of limitations as a research method. The information available to raters is, in many ways, bound by what was chosen by the authors of the books. Moreover, within the leadership arena there is oftentimes a substantial emphasis on heroic leadership — viewing the leader as a driving force behind team success (Meindl, Ehrlich, & Dukerich, 1985). Although we concede these limitations, we do want to note that several author biases were controlled for and that by choosing only championship coaches we have limited many of the confounds associated with using a sample of coaches in general (with various records). Moreover, the use of historiometric analysis allowed us to examine a sample of coaches literally unattainable via other more traditional methods. The fact remains, however, that all methods have their limitations, and the historiometric approach is not without its share of them. 7.1. Conclusions Despite these limitations, we feel that the manuscript makes a number of unique and important contributions to our understanding of leadership. First, this study represents a direct attempt to examine the core tenants of the CIP model — a task yet to be undertaken in the nearly 15 previous studies examining the model. The results of this effort provide general support for the tenants of the model as a whole. More speciﬁcally, of the 10 components of leaders' prescriptive mental models, 8 were found to be signiﬁcant and in the predicted directions. Although predicted, these ﬁndings are hardly trivial and lend substantive credence to an emerging leadership model that signiﬁcantly expands on previous perspectives of viewing leadership. Moreover, these results were obtained in a sample of leaders who have proven that they are capable of outstanding achievement. It is even more noteworthy that the sample was drawn from a common domain and because of this we can more conﬁdently say that it appears leaders can be successful in a variety of ways, employing a number of different styles. In short, these results highlight the emerging conclusion that we can no longer ignore the fact that not all outstanding leaders employ a charismatic leadership style (Yukl, 1999). Our views of leadership must be expanded to include other forms of leadership beyond the more traditional vision-based models. This is not to say that charismatic leadership is ineffective — quite the opposite is true (e.g., Lowe, Kroeck, & Sivasubramaniam, 1996). Rather, our results provide evidence supporting the core theoretical framework of an expanded leadership model, one that includes multiple approaches to outstanding leadership. Second, despite the generally supportive results, not all predictions were observed. In particular, results failed to support the “focus in model construction” and “crisis conditions” components of the model. It must be conceded that the nature of the study precluded rich and detailed investigation of these aspects of the leaders' mental models. Moreover, the crises faced by football coaches may be unique and as such, results may not be applicable to other forms of outstanding leadership. Taken together, this pattern of non-ﬁndings may best be attributed to the study method, rather than incorrect speciﬁcation of theory. The lack of predicted results, however, provides some indication as to areas of future research that might explore speciﬁc boundary conditions or the inclusion of key contextual moderators. Third, given the strong results observed in the use of emotions component of the model, it seems warranted to suggest that this element be included in future discussions of the model. More speciﬁcally, this component of the CIP model has only been discussed peripherally to date. Due to the fairly strong results observed in this study, however, it would seem that the use of emotions when expressing and framing a prescriptive mental model is an important discriminating factor across leader types — if perhaps only in speciﬁc domains (e.g., those that allow for frequently displays of emotion). These results seem to suggest that future revisions to the model include explicit reference to the varied uses of emotional appeals by the three leader styles (George, 2000; McCollKennedy & Anderson, 2002). Fourth and ﬁnally, the results of the discriminant function analyses revealed an interesting trend regarding how the three leader types differ from one another. That is, although the results of the MANOVA and follow-up ANOVAs demonstrated that the three leader types differ on most of the hypothesized mental model components, when examining what was used to discriminate each leader type from the other, a unique pattern emerged for each leader type. Speciﬁcally, charismatic leaders may be summarized as being very positive, appealing to the masses and offering hope for a brighter future. Ideological S.T. Hunter et al. / The Leadership Quarterly 22 (2011) 70–91 85 leaders, on the other hand, are more succinctly described as being focused on a few key outcomes, placing a greater emphasis on the past and drawing on some degree of negativity when making appeals to followers. Finally, pragmatic leaders are aptly summarized as being very malleable; changing and altering their behavior when necessary — a summary consistent with the ﬁndings of Bedell and colleagues' (2006) examination of Machiavellianism, where pragmatic leaders were willing to do whatever it took to accomplish a given objective. The point here is that although the leader types differed on the mental model components predicted by the model, the pattern that deﬁned a leader as pragmatic, charismatic or ideological was unique for each leader type. 7.2. Implications The results of this study have a number of important theoretical implications. The ﬁrst is that we clearly need to expand our view on what it takes to be an outstanding leader. More directly, although it is abundantly clear that vision-based models are essential components of outstanding leadership, they are not the only means of achieving signiﬁcant outcomes. In our sample of winning coaches, for example, we saw a number of leaders who do not lead in a charismatic fashion yet were able to achieve the ultimate outcome in their respective domains. Once again, this is not to discount the impact or importance of vision-based models — rather, we must realize from a theoretical perspective that there is much to be learned from other forms of leadership. With respect to the CIP model directly, our results suggest there is value in taking a sensemaking perspective in leadership theory. More speciﬁcally, by validating the core framework of the CIP model, there is increased evidence that a sensemaking perspective is a useful means of understanding differences among leader types (Gioia & Thomas, 1996; Keller, 2003). This suggests that future leadership frameworks may beneﬁt by considering how leaders form their mental models as well as how they are expressed to followers. Natural extensions of the sensemaking perspective, moreover, include gaining an understanding of the respective mental models of followers and how they resonate, or fail to resonate, with a given leader (Howell & Shamir, 2005). The ﬁnal theoretical implication is associated with the increased frequency of charismatic leaders in the college ranks relative to the professional ranks. Although still speculative given the size of the sample (n = 29 college versus n = 25 of NFL), these results are consistent with the propositions put forth by Howell and Shamir (2005) who suggested that subordinates with low selfconcept clarity would develop stronger relationships with charismatic leaders relative to those subordinates who have a high self concept clarity and identify more strongly with the group as a whole. Moreover, in their multilevel discussion on leader emergence, Mumford et al. (2008) suggested that charismatic leaders would emerge in populations that are more vulnerable in contrast to pragmatic leaders who would emerge in more autonomous populations. Given the contrast of subordinates in the samples – student-athletes (college) versus professional (NFL) – these trends are in the proposed directions and would support the propositions put forth by Howell and Shamir (2005) as well as Mumford et al. (2008). Thus, it seems that more work is warranted in examining the conditions under which the leader types might emerge and be most successful. In addition to the theoretical implications, the results of this study also speak to a number of practical implications as well. The most critical is related to leadership training and development (Day, 2000). Stated simply, emerging research on the CIP model is forcing us to consider the possibility that leadership development programs solely emphasizing vision-based models may not be applicable to all individuals (Pasternack & O'Toole, 2002; Khurana, 2002). In fact, by attempting to develop all individuals into a charismatic or transformational mold, we may be opposing their descriptive mental models — mental models that are crystallized and were likely shaped during early formative stages in life (Ligon et al., 2008). Along similar lines, leadership development programs aimed at more pragmatic or even ideological training may not be suitable for individuals with a charismatic perspective on leading. Although challenging to practitioners and leader development experts, we must explore the possibility that tailored training programs may be required to maximize the potential for leaders falling into differing styles of leadership (Espejo, Day, & Scott, 2005). Another practical implication of the study is derived from considering the results of the discriminant function analysis. Speciﬁcally, the results provide some indication about possible scoring weights with respect to the future development of measures used to distinguish among leader types. Although previous measures have been used with reasonable success (BedellAvers et al., 2008; Hunter et al., 2009) there is room for reﬁnement and improvement. In particular, these measures take upwards of 30 min to administer, require high-levels of cognitive engagement and are fairly taxing on participants overall. More directly, however, given the results of the discriminant function analyses in this study, these measures may beneﬁt from increased sophistication with respect to scoring weights. Thus, future scale development should consider improved scoring methods derived from the results of this study and others to come. 7.3. Future research Turning now to future research opportunities, this study lends credence to the notion that future research on the CIP model, and other models like it (e.g., Collins, 2001; Tsui et al., 2006; Pasternack & O'Toole, 2002), is necessary. That is, by validating the core components of the model, we contribute to the growing evidence supporting the notion that expanded leadership perspectives are critical areas of research. Generally speaking, then, future research should emphasize broader frameworks of leadership as they clearly merit increased exploration (Yukl, 1999). With regard to the CIP model speciﬁcally, it would appear critical to examine the malleability of leader behaviors within and across the styles — a point noted by Weber (1924) in his original classiﬁcation. It is clear, for example, that charismatic leaders can 86 S.T. Hunter et al. / The Leadership Quarterly 22 (2011) 70–91 act in pragmatic ways when solving problems. Similarly, ideological leaders are not always negative and may offer perspectives of hope and optimism by means of a strong connection to beliefs and values. However, it is evident that there are stable differences across the leader types — differences that may be used to reliably categorize the leaders. Thus, future research must explore which behaviors are universal to all forms of leadership and which are more centrally indicative of a given type. The core sensemaking components explored in this study are certainly steps in this direction, but future research is needed to examine the malleability and consistency of other behaviors and cognitions. Along similar lines, future research should examine the mixed-type leader who may share aspects of multiple styles of leadership. It is unclear at this point if such a leader would be superior given their ﬂexibility or if they would lack the strengths associated with having a clear leadership style. There is some evidence suggesting that mixed-type leaders are effective in a variety of conditions. For example, Hunter et al. (2009) found that mixed-type leaders performed well across a variety of conditions; a ﬁnding consistent with research by Bedell-Avers et al. (2009) that also included a mixed-type category and found that they were effective problem solvers. Mumford (2006) also discusses leaders such as Malcolm X who was both ideological and charismatic — yet is historically viewed as a highly effective, albeit controversial, leader (Clarke, 1990). Despite this anecdotal evidence, however, we know very little about this leader type and as such, more work is needed in this area. Shifting focus onto the subordinate, future research would be well served to examine follower characteristics and their impact on leader performance. Within the sports domain, for example, it is apparent that teams with outstanding subordinate talent can be successful with only mediocre leadership (Lazenby, 2002). Conversely, there is some anecdotal indication that great coaches (i.e., outstanding leaders) have little success without requisite subordinate talent. Thus, understanding the role of subordinate KSAs in leadership and team success appears to be a worthwhile endeavor within the sports domain speciﬁcally — with potential research and practical implications beyond this arena. Finally, it is essential that future research examine the congruence in mental models between leader types and followers. Although this has been examined in some ways by Mumford, Strange, et al. (2006) in their investigation of LMX exchanges — more work needs to be done if we continue to place a focus on sensemaking in our leadership models (Baker, 2007). As suggested by Howell and Shamir (2005), aspects of the subordinate such as self-concept clarity and relationship orientation dictate the nature of the ultimate relationship with the leader. Along these lines, by focusing on mental model congruence (and divergence), we may gain a better understanding of the inﬂuence mechanisms at work both with regard to leader–follower inﬂuence as well as inﬂuence from follower to leader (Mohammed & Dumville, 2001). In sum, the results of the study underscore the importance of a growing trend in leadership and management research — expanding our view beyond the dominant vision-based perspectives to include other forms of leadership (Hunter et al., 2007). Again, this is not to the exclusion of charismatic or transformational leadership, but rather our results support a call for the extension of leadership perspectives, more broadly. By ﬁnding general support for expanded perspectives such as those in the CIP model, we hope that others are motivated to consider expanded leadership models and our study is an impetus for growth in this direction. Acknowledgements We would like to thank Brandon Vessey, Pamela Batson, Natalia Ayub, and Zachary Slaybaugh for their efforts on the project. We would also like to thank Dr. Michael Mumford along with the three anonymous reviewers for their suggestions and help in improving the manuscript. Appendix A Allen, G. (with Weiskopf, D.) (1975). Handbook in of winning football. Boston: Allyn and Bacon. Baldoni, J. (2005). How great leaders get great results. New York: McGraw-Hill. Banks, J. (1973). The Darrell Royal story. Austin TX: Shoal Creek Publishers, INC. Bilinski, B. (1989). Champions: Lou Holtz's Fighting Irish. South Bend, IN: Diamond Communications, Inc. Billick, B. & Peterson, J. A. (2001). Competitive leadership: Twelve principles for success. Chicago, IL: Triumph Books. Blount, R. Jr. (1974). About three bricks shy of a load: A highly irregular lowdown on the year the Pittsburgh Steelers were super but missed the bowl. Boston: Brown and Company. Bouchette, E. (1999). Ready or not: Steelers sputtering as the opener dawns. In D. Eyring (Ed.), Cowher power (pp. 70–72). Chicago: Triumph Books. Bouchette, E. (2005). Facing the music: No longer a rookie, Roethlisberger has discovered the price of fame and high expectations. In D. Eyring (Ed.), Cowher power (pp. 118–122). Chicago: Triumph Books. Bowden, B. (with Bowden, C.) (2001). The Bowden Way. Marietta, GA: Longstreet Press, Inc. Brown, M. & Little, B. (2006). One Heartbeat II: The road to the national championship. Albany, TX: Bright Sky Press. Brown, M. & Little, B. (2001). One Heartbeat: A philosophy of teamwork, life, and leadership. Albany, TX: Bright Sky Press. Browning, A. (2001). I remember Paul “Bear” Bryant: Personal memories of college football's most legendary coach as told by the people who knew him best. Nashville, TN: Cumberland House. Bryant, P. W., & Underwood, J. (1974). Bear: The hard life and good times of Alabama's Coach Bryant. Little, Brown, and Company: Boston. Bynum, M. (1993). Pop Warner: Football's greatest teacher. USA: Gridiron Football Properties Corporation. S.T. Hunter et al. / The Leadership Quarterly 22 (2011) 70–91 87 Cavanaugh, J. (1930). The autobiography of Knute K. Rockne. Indianapolis, IN: The Bobbs-Merrill Company. Chastain, B. (2002). The Steve Spurrier story: From Heisman to head bull coach. Lanham, MD: Taylor Trade Publishing. Collier, G. (2006). Mission accomplished: Cowher ﬁnally gets the Steelers that “one for the thumb”. In D. Eyring (Ed.), Cowher power (pp. 132–136). Chicago: Triumph Books. Daugherty, D. (with Diles, D.) (1974). Duffy: An autobiography. Garden City, NY: Doubleday & Company. Denlinger, K. (1994). For the glory: College football dreams and realities inside Paterno's program. New York: St. Martin's Press. Devaney, B. (with Babcock, M., Bryant, D., Parker, V., & York, R.) (1981). Devaney. Lincoln, NE: Bob Devaney. Devine, D. (with Steele, M.R.). Simply Divine: Memoirs of a Hall of Fame Coach. Champaign, IL: Sports Publishing Inc. Devine, D., Onofrio, A. Missouri Power Football: Revised Edition. Columbia, MO: Lucas Brothers Publishers. Dickey, G. (1989). San Francisco 49ers: The super years. San Francisco: Chronicle Books. Ditka, M. (2001). Ditka on his bear days. In R. Wolfe (Ed.), Da coach (pp. 205–216). Chicago: Triumph Books. Ditka, M. (with Pierson, D.) (1986). Ditka: An autobiography. Chicago: Bonus Books. Dooley, V. (with Giles, B.). (2005). Vince Dooley's Tales from the 1980 Georgia Bulldogs. Champaign, IL: Sports Publishing LLC. Dooley, V.J. (with Barnhart, T.). (2005). Dooley: My 40 Years at Georgia. Chicago: Triumph Books. Dunnavant, K. (1996). Coach: The life of Paul “Bear” Bryant. New York: Simon & Schuster Inc. Edwards, L. (as told by Nelson, L.). (1980). LaVell Edwards: Building a Winning Football Tradition at Brigham Young University. Provo, UT: Council Press. Edwards, L. (with Benson, L.). (1995). LaVell: Airing it Out. Salt Lake City, UT: Shadow Mountain. Ellis, S., & Vilona, B. (2006). Pure Gold: Bobby Bowden. Champaign, IL: Sports Publishing L.L.C. Ewbank, W. (as told to Roiter, N.) (1972). Goal to go: The greatest football games I have coached. New York: Hawthorn Books. Facer, D. (2005). BCS breakthrough: Utah's historic 2004 season. United States: Kci Sports Publishing. Flores, T. (with Cooney, F.) (1992). Fire in the iceman: Autobiography of Tom Flores. Chicago: Bonus Books. Flores, T. (with Fulks, M.) (2003). Tales from the Oakland Raiders. Champaign, IL: Sports Publishing L.L.C. Fox, L. (1969). Broadway Joe and his super Jets. New York: Coward-McCann. Freeman, D. H., & Aron, J. (2001). I remember Tom Landry. Champaign, IL: Sports Publishing L.L.C. Fulmer, P. (with Hagwood, J., & Ward, J.). (1999). A Perfect Season. Nashville, TN: Rutledge Hill Press. Fulmer, P. & Gerald, S. (1997). Legacy of Winning: “It Doesn't All Happen On Game Day.” Knoxville, TN: Pressmark International. Gibbs, J. (with Abraham, K.) (2002). Racing to win. Sisters, Oregon: Multnomah Publishers. Gibbs, J. (with Jenkins, J.) (1991). Joe Gibbs: Fourth and one. Nashville, TN: Thomas Nelson. Golenbock, P. (2005). Landry's boys: An oral history of a team and an era. Chicago: Triumph Books. Greenberg, S. & Ratermann, D. (1997 and 2004). I Remember Woody: Recollections of the Man They Called Coach Hayes. Chicago: Triumph Books. Gruden, J. (with Carucci, V.) (2003). Do you love football?!: Winning with heart, passion, and not much sleep. New York: HarperCollins. Havel, C. (1996). Holmgren: The early years. In F. J. Fitzgerald (Ed.), Heir to the legacy: The memorable story of Mike Holmgren's Green Bay Packers (pp. 8–15) Louisville, KY: AdCraft. Hession, J. (1993). Forty Niners: Collector's Edition. U.S.: Foghorn Press. Holley, M. (2004). Patriot reign: Bill Belichick, the coaches, and the players who built a champion. New York: Harper Collins. Holtz, L. (with Heisler, J.). (1989). The ﬁghting spirit: A championship season at Notre Dame. New York: Random House. Holtz. L. (1998). Winning every day: The game plan for success. New York: Harper-Collins. Hornung, P. (1991). Woody Hayes: A reﬂection. Champaign, IL: Sagamore Publishing Inc. Hubbard, S. (1997). Shark among dolphins: Inside Jimmy Johnson's transformation of the Miami Dolphins. New York: Ballantine Books. Isaacson, K. (with Kessenich, T.) (1996). Return to glory: The inside story of the Green Bay Packers return to prominence. Iola, WI: Krause Publications. James, D. (as told to Virgil Parker). (1992). James. Lincoln, NB: Self-published. Johson, J. (as told to Hinton, E.) (1993). Turning the thing around: Pulling America's team of the dumps-and myself out of the doghouse. New York: Hyperion. Krikorian, D. (2003). Pete Carroll: Carroll on top of game. Long Beach Press-Telegram. 120–121. Limprecht, H.J., Denney, J., Silber, H.S. (1972). Bob Devaney: Portrait of a Winner. Omaha, NE: Kratville Publications. Lombardo, J. (2005). A ﬁre to win. New York: Thomas Dunne Books. Madden, J. (with Anderson, D.) (1984). Hey, wait a minute, I wrote a book! New York: Willard Books. Madden, J. (with Anderson, D.) (1996). All Madden: Hey, I'm talking pro football! New York: HarperCollins. Majors, J. (with Byrd, B.). (1986). You can go home again. Nashville, TN: Rutledge Hill Press. Martin, B. (2006). The boys from old Florida: Inside Gator nation. Champaign, IL: Sports Publishing, LLC. McCartney, B. (with Diles, D.). (1990). From ashes to glory. Nashville, TN: Thomas Nelson Publishers. McGuff, J. (1970). Winning it all: The Chiefs of the AFL. Garden City, NY: Doubleday & Company, Inc. McKay, J. (with Perry, J.). (1974). McKay: A coach's story. New York: Atheneum. 88 S.T. Hunter et al. / The Leadership Quarterly 22 (2011) 70–91 McMahon, J. (2001). The monster of the midway. In R. Wolfe (Ed.), Da coach (pp. 93–106). Chicago: Triumph Books. O'brien, M. (1987). Vince: A personal biography of Vince Lombardi. New York: William Morrow and Company. O'Brien, M. (1998). No ordinary Joe: The biography of Joe Paterno. Nashville, TN: Rutledge Hill Press. Osborne, T. (1996). On solid ground. Lincoln, NE: Nebraska Book Publishing. Osborne, T. (1999). Faith in the game: Lessons on football, work, and life. New York: Broadway Books. Pagna, T. (with Best, B.). (1976). Notre Dame's era of Ara. Huntsville, AL: The Strode Publishers, Inc. Parcellis, B. (with Lupica, M.) (1987). Parcells: Autobiography of the biggest Giant of them all. Chicago: Bonus Books. Parcells, B. (with Coplon, J.) (1995). Finding a way to win: The principles of leadership, teamwork, and motivation. New York: Doubleday. Parseghian, A., & Pagna, T. (1971). Parseghian and Notre Dame football. Garden City, NY: Doubleday and Company, Inc. Paterno, J. (with Asbell, B.). (1989). Paterno: By the book. New York: Random House. Quakenbush, R., & Bynum, M. (1988). Knute Rockne: His life and legend. USA: October Football Corp. Rappoport, K. (1974). The Trojans: A story of southern California football. Huntsville, AL: The Strode Publishers. Robinson, J. (1996). Coach to coach: Business lessons from the locker room. San Diego, CA: Pfeiffer & Co. Robinson, J., & Jares, J. (1981). Quest. Royal, D. (with Wheat, J.) (2005). Coach Royal: Conversations with a Texas football legend. Austin, TX: University of Texas Press. Saban, N. (2004). Nick Saban's tiger triumph. Chicago, IL: Triumph Books. Saban, N. (with Curtis, B.) (2005). How good do you want to be? New York: Ballantine Books. Sahadi, L. (1979). Steelers! Team of the decade. New York: Times Books. Seaman, R.S. (1981). The Legend of Bill and Vince Dooley: Bowl Breaking Brothers. Huntsville, AL: The Strode Publishers. Shanahan, M. (with Schefter, A.) (1999). Think like a champion. New York: Harper Business. Shropshire, M. (2004). When the Tuna went down to Texas: How Bill Parcells led the Cowboys back to the promised land. New York: Harper Collins Publishers. Shula D., & Blanchard, K. (1995). Everyone's a coach: You can inspire anyone to be a winner. New York: Harper Business. Shula, D. (with Sahadi, L.) (1973). The Winning Edge. New York: E. P. Dutton & Co., Inc. Silverstein, T. (1996). Always a winner. In F. J. Fitzgerald (Ed.), Heir to the legacy: The memorable story of Mike Holmgren's Green Bay Packers (pp. 35–42) Louisville, KY: AdCraft. Sittler, D. (2001). Heir to the Sooner legacy: The championship story of Oklahoma coach Bob Stoops. USA: Epic Sports. Snapp, S. & Tressel, J. (2006). 12-0: An insider's account of Ohio State's 2006 championship season. Chicago, IL: Triumph Books. Spurrier, S. (with Carlson, N.) (1992). Gators: The inside story of Florida's ﬁrst SEC title. Orlando, FL: Tribune Publishing. Stambroski, J. (1998). J.J. straight talking: Jimmy Johnson's insights, outbursts, kudos, and comebacks. Hollywood, FL: Lifetime Books. Stram, H. (with Sahadi, L.) (1986). They're playing my game. New York: William Morrow and Company, Inc. Switzer, B. (with Shrake, B) (1991). Bootlegger's boy. New York: William Morrow and Company, INC. The Staff of the Mobile Register (1993). Back to glory: The amazing story of Alabama's 1992 championship season. Kansas City, KS: Andrews and McMeel. Towle, M. (2002). I remember: Bud Wilkinson. Nashville, TN: Cumberland House. Triumph Books (2000). Gotta go to work: The Rams unforgettable championship season. Chicago: Triumph Books. Tysiac, K. (2006). Tales from Clemson's 1981 Championship Season. Champaign, IL: Sports Publishing LLC. University of Michigan (1998). 1997 National Champions. Charlotte, NC: UMI Publications, Inc. Vare, R. (1974). Buckeye: A Study of Coach Woody Hayes and the Ohio State Football Machine. New York: Harper & Row, Publishers, Inc. Walsh, B. (with Billick, B. & Peterson, J.) (1998). Finding the winning edge. Champaign, IL: Sports Publishing Inc. Walsh, B. (with Dickey, G.) (1990). Building a Champion. New York: St. Martin's Press. Wells, R. W. (1971). Vince Lombardi: His life and times. Madison, Wisconsin: Prairie Oak Press. Wexell, J. (2004). Tales from behind the steel curtain. Champaign, IL: Sports Publishing L.L.C. Wharton, D. & Klein, G. (1961). Conquest: Pete Carroll and the Trojans' climb to the top of the college football mountain. Chicago: Triumph Books. Wolfe, R. (2000). Jon Gruden: All it takes is all ya got take this job and love it. United States: Lone Wolfe Press. Appendix B For all ratings use the following scale: 5 = Nearly always 4 = The majority of time 3 = Some of the time 2 = Rarely 1 = Never S.T. Hunter et al. / The Leadership Quarterly 22 (2011) 70–91 Time frame To what extent does the leader: Use a future-oriented time-frame? Use a present-oriented time-frame? Use a past-oriented time-frame? Type of experience used To what extent does the leader: Discuss positive prior experiences when interacting with (e.g., motivating) subordinates? Discuss negative prior experiences when interacting with (e.g., motivating) subordinates? Blend both positive and negative prior experiences when interacting with (e.g., motivating) subordinates? Nature of outcomes sought To what extent does the leader: Seek positive outcomes? Seek transcendent (e.g., going beyond normal bounds or expectations) outcomes? Change or alter his discussion of outcomes sought, depending on the situation? Number of outcomes sought To what extent does the leader: Discuss seeking multiple outcomes? Discuss seeking only a few outcomes? Change or alter the number of outcomes sought depending on the situation? Focus in model construction To what extent does the leader: Discuss external inﬂuences/factors when motivating/coaching subordinates? Discuss internal inﬂuences/factors when motivating/coaching subordinates? Locus of causation To what extent does the leader: View people as the cause of outcomes? View situations as the cause of outcomes? View the interaction of people and situations as the cause of outcomes? Controllability of causation To what extent does the leader: Discuss how he and his subordinates have a high degree of control over their destiny? Discuss how he and his subordinates have a low degree of control over their destiny? Change his discussion of controllability depending on the situation? Targets of inﬂuence To what extent does the leader: Spend his time trying to convince/motivate the entire team and coaching staff? Spend his time trying to convince/motivate only the key members of the team? Spend his time trying to convince/motivate the star players on the team? 89 90 S.T. Hunter et al. / The Leadership Quarterly 22 (2011) 70–91 Crisis conditions To what extent does the leader: View the situation as ordered and highly structured? View the situation as chaotic? View the situation as existing only in the localized area (i.e., does not discuss external factors or inﬂuences)? Use of emotional appeals To what extent does the leader: Use emotional appeals to motivate and engage followers? Use rational persuasion to motivate and engage followers? References Adair-Toteff, C. (2005). Max Weber's charisma. Journal of Classical Sociology, 5, 189−204. Avery, D. R., Tonidandel, S., Grifﬁth, K. H., & Quinones, M. A. (2003). The impact of multiple measures of leader experience on leader effectiveness. Journal of Business Research, 56, 673−679. Avolio, B. J. (1999). Full leadership development: Building the vital forces in organizations. Thousand Oaks, CA: Sage. Baker, S. D. (2007). Followership: The theoretical foundation of a contemporary construct. Journal of Leadership & Organizational Studies, 14, 50−60. Bass, B. M., & Avolio, B. J. (1994). Improving organizational effectiveness through transformational leadership (2nd ed.). Mahwah, NJ: Erlbaum. Bedell, K. E., Hunter, S. T., Angie, A. D., & Vert, A. (2006). A historiometric examination of Machiavellianism and a new taxonomy of leadership. Journal of Leadership and Organizational Studies, 12, 50−72. Bedell-Avers, K. E., Hunter, S. T., & Mumford, M. D. (2008). Conditions of problem-solving and the performance of charismatic, ideological, and pragmatic leaders: A comparative experimental study. Leadership Quarterly, 19, 89−106. Bedell-Avers, K. E., Hunter, S. T., & Mumford, M. D. (2009). Charismatic, ideological, and pragmatic leaders: How do they interact? The Leadership Quarterly, 20, 299−315. Beyer, J. M. (1999). Taming and promoting charisma to change organizations. Leadership Quarterly, 10, 307−330. Burns, J. M. (1978). Leadership. New York: Harper & Row. Chan, W. (2003). Analyzing ipsative data in psychological research. Behaviormetrika, 30, 99−121. Charbonneau, D., Barling, J., & Kelloway, E. K. (2001). Transformational leadership and sports performance: The mediating role of intrinsic motivation. Journal of Applied Social Psychology, 31, 1521−1534. Chellandurai, P., & Carron, A. V. (1978). Leadership. Canadian Association for Health, Physical Education and Recreation Sociology of Sport Monograph Series A. Calgary, AB: University of Calgary. Clarke, J. H. (1990). Malcolm X: The man and his times. Trenton, NJ: Africa World Press. Cohen, J. (1992). A power primer. Psychological Bulletin, 112, 155−159. Collins, J. (2001). Good to great: Why some companies make the leap and others don't. New York, NY: Harper Business. Conger, J. A., & Kanungo, R. S. (1988). Toward a behavioral theory of charismatic leadership on organizational settings. Academy of Management Review, 12, 637−647. Conger, J. A., & Kanungo, R. S. (1998). Charismatic leadership in organizations. Thousand Oaks, CA: Sage. Day, D. V. (2000). Leadership development: A review in context. Leadership Quarterly, 11, 581−613. Day, D. V., Sin, H., & Chen, T. T. (2004). Assessing the burdens of leadership: Effects of formal leadership roles on individual performance. Personnel Psychology, 57, 573−605. Dirks, K. T. (2000). Trust in leadership and team performance: Evidence from NCAA basketball. . Espejo, J., Day, E. A., & Scott, G. (2005). Performance evaluations, need for cognition, and the acquisition of complex skill: An attribute-treatment interaction. Personality and Individual Differences, 38, 1867−1877. Garland, D. J., & Barry, J. R. (1990). Personality and leader behaviors in collegiate football: A multidimensional approach to performance. Journal of Research in Personality, 24, 355−370. George, J. M. (2000). Emotions and leadership: The role of emotional intelligence. Human Relations, 53, 1027−1055. Giambatista, R. C. (2004). Jumping through hoops: A longitudinal study of leader life cycles in the NBA. The Leadership Quarterly, 15, 607−624. Gioia, D. A., & Thomas, J. B. (1996). Institutional identity, image, and issue interpretation: Sensemaking during strategic change in academia. Administrative Science Quarterly, 41(3), 370−403. Habermas, T., & Bluck, S. (2000). Getting a life: The emergence of the life story in adolescence. Psychological Bulletin, 126, 748−769. Hawkins, K., & Tolzin, A. (2002). Examining the team/leader interface: Baseball teams as exemplars of postmodern organizations. Group & Organization Management, 27, 97−112. Hennis, W. (1996). Max Webers Wissenschaft vom Menschen: Neue Studien zur Biographie des Werks. Mohr, Siebeck: Tübingen. 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: Southern Illinois University Press. House, R. J., & Howell, J. M. (1992). Personality and charismatic leadership. Leadership Quarterly, 3, 81−108. Howell, J. S., & Shamir, B. (2005). The role of followers in the charismatic leadership process: Relationships and their consequences. Academy of Management Review, 30, 96−112. Hunt, J. G. (1999). Transformation/charismatic leadership's transformation of the ﬁeld: An historical essay. Leadership Quarterly, 10, 129−144. Hunt, J. G., & Ropo, A. (1995). Multi-level leadership: Grounded theory and mainstream theory applied to the case of General Motors. Leadership Quarterly, 6, 379−412. 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. Leadership Quarterly, 10, 423−448. Hunter, S. T., Bedell-Avers, K. E., & Mumford, M. D. (2007). The typical leadership study: Assumptions, implications, and potential remedies. The Leadership Quarterly, 18, 435−446. Hunter, S. T., Bedell-Avers, K. E., & Mumford, M. D. (2009). Impact of situational framing and complexity on charismatic, ideological and pragmatic leaders: Investigation using a computer simulation. The Leadership Quarterly, 20, 383−404. Johnson-Laird, P. (1983). Mental models. Cambridge, MA: Harvard University Press. Keller, T. (2003). Parental images as a guide to leadership sensemaking: An attachment perspective on implicit leadership theories. Leadership Quarterly, 14(2), 141−160. S.T. Hunter et al. / The Leadership Quarterly 22 (2011) 70–91 91 Khurana, R. (2002). The curse of the superstar CEO. Harvard Business Review, 80, 60−66. Lansing, R. K. (1989). Football coach's lessons can lead your team to victory. Management Review, 78, 14. Lazenby, R. (2002). Blood on the horns: The long strange ride of Michael Jordan's Chicago Bulls. Lenexa, KS: Addax Publishing. Ligon, G. S., Hunter, S. T., & Mumford, M. D. (2008). Development of outstanding leadership: A life narrative approach. Leadership Quarterly, 19, 312−334. Lowe, K. B., & Gardner, W. L. (2000). Ten years of The Leadership Quarterly: Contributions and challenges for the future. Leadership Quarterly, 11, 459−514. Lowe, K. B., Kroeck, K. G., & Sivasubramaniam, N. (1996). Effectiveness correlates of transformation and transactional leadership: A meta-analytic review of the MLQ literature. The Leadership Quarterly, 7, 385−425. McColl-Kennedy, J. R., & Anderson, R. D. (2002). Impact of leadership style and emotions on subordinate performance. The Leadership Quarterly, 13, 545−559. Malcomsen, B. (2008). Pete Carroll news — Victory against ASU. http://www.petecaroll.com. Retrieved Dec 10, 2008, from. Meindl, J. R., Ehrlich, S. B., & Dukerich, J. M. (1985). The romance of leadership. Administrative Science Quarterly, 30, 78−102. Miller, S. M. (1963). Max Weber. New York, NY: Crowell Co.. Mohammed, S., & Dumville, B. C. (2001). Team mental models in a team knowledge framework: Expanding theory and measurement across disciplinary boundaries. Journal of Organizational Behavior, 22, 89. Mumford, M. D. (2006). Pathways to outstanding leadership: A comparative analysis of charismatic, ideological, and pragmatic leaders. Mahwah, NJ: Lawrence Erlbaum Associates Publishers. Mumford, M. D., Antes, A. A., Caughron, J. J., & Friedrich, T. L. (2008). Charismatic, ideological and pragmatic leadership: Multi-level inﬂuences on emergence and performance. The Leadership Quarterly, 19, 144−160. Mumford, M. D., Licuanan, B., Marcy, R. T., Dailey, L., & Blair, C. (2006). Political tactics. In M. D. Mumford (Ed.), Pathways to outstanding leadership: A comparative analysis of charismatic, ideological, and pragmatic leadership (pp. 138−166). Mahwah, NJ: Erlbaum Press. Mumford, M. D., & Strange, J. M. (2002). Vision and mental models: The case of charismatic and ideological leadership. In B. J. Avolio & F.J. Yammarino (Eds.), Charismatic and transformational leadership: The road ahead (pp. 109−142). Oxford, England: Elsevier. Mumford, M. D., Strange, J., Scott, G., Dailey, L., & Blair, C. (2006). Leader–follower interactions. In M. D. Mumford (Ed.), Pathways to outstanding leadership: A comparative analysis of charismatic, ideological, and pragmatic leadership (pp. 138−166). Mahwah, NJ: Erlbaum Press. Mumford, M. D., & Van Doorn, J. R. (2001). The leadership of pragmatism: Reconsidering Franklin in the age of charisma. The Leadership Quarterly, 12, 279−309. Pasternack, B. A., & O'Toole, J. (2002). Yellow-light leadership: How the world's best companies manage uncertainty. Strategy and Business, 27, 1−10. Pasternack, B. A., Williams, T. D., & Anderson, P. F. (2001). Beyond the cult of the CEO: Building institutional leadership. Strategy and Business, 22, 1−12. Pillai, R., & Meindl, J. R. (1998). Context and charisma: A “meso” level examination of the relationship of organic structure, collectivism, and crisis to charismatic leadership. Journal of Management, 24, 643−671. Sein, M. K., & Bostrom, R. P. (1989). Individual differences and conceptual models in training novice users. Human–computer interaction, 4, 197−229. Shamir, B., House, R. J., & Arthur, M. B. (1993). The motivational effects of charismatic leadership: A self-concept based theory. Organizational Science, 4, 577−594. Shields, D. L. L., Gardner, D. E., Bredemeier, B. J. L., & Bostro, A. (1997). The relationship between leadership behaviors and group cohesion in team sports. The Journal of Psychology, 131, 196−210. Simonton, D. K. (1990). Psychology, science, and history: An introduction to historiometry. New Haven, CT: Yale University Press. Strange, J. M., & Mumford, M. D. (2002). The origins of vision: Charismatic versus ideological leadership. The Leadership Quarterly, 13, 343−377. Tabachnick, B. G., & Fidell (2001). Using multivariate statistics. Needham Heights, MA: Allyn & Bacon. Toothaker, L. E. (1993). Multiple comparison procedures. Newbery Park, CA: Sage. Tsui, A. S., Zhang, Z., Wang, H., Xin, K. R., & Wu, J. B. (2006). Unpacking the relationship between CEO leadership behavior and organizational culture. Leadership Quarterly, 17, 113−137. Weber, M. (1924). The theory of social and economic organizations. New York: Free Press. Weiss, R. M. (1983). Weber on bureaucracy: Management consultant or political theorist? Academy of Management Review, 8, 244−262. Yukl, G. (1999). An evaluation of conceptual weaknesses in transformational and charismatic leadership theories. Leadership Quarterly, 10, 285−305. Yukl, G. (2006). Leadership in organizations VI. Upper Saddle River, New Jersey: Prentice-Hall.

RELATED PAPERS

RELATED TOPICS

Log In

First and ten leadership: A historiometric investigation of the CIP leadership model

First and ten leadership: A historiometric investigation of the CIP leadership model

Related Papers

RELATED PAPERS

RELATED TOPICS