Examining Effective Technology Project Leadership Traits And Behaviors

ARTICLE IN PRESS

Computers in Human Behavior Computers in Human Behavior xxx (2004) xxx–xxx www.elsevier.com/locate/comphumbeh

Examining effective technology project leadership traits and behaviors Kenneth David Strang 44 Sunset Drive, Fredericton, NB, Canada, E3A 9H7

Abstract Effective project team leadership theory is explored from the perspective of leader traits, skills, roles, and behaviors. Existing leadership traits and behaviors are examined from organization science empirical studies, they are differentiated from management functions, and the gap between extant theory and the project management domain is highlighted through this research. Leadership principles are examined from actual organizational work settings, specifically dynamic projects, with a view to discovering what actually happens as compared with espoused theory. The underlying research question is whether these theories hold up and can be effectively integrated to the project management domain. The significance of leadership behavior differences at the project and organizational levels is grounded on empirical evidence. The reflection of integrated organizational science and project leadership theory using qualitative and quantitative research methods contributes to the body of knowledge by sharing and learning in professional and academic communities of practice. Design/methodology/approach. Contemporary and alternative trait-based leadership theories are discussed, such as task/relationship orientations, emotionality maturity, Leadership Virtual Reality, Level 5 Leader Trait Hierarchy, charisma, emotional maturity, and personality traits are discussed. Multiple theories and typologies such as Managerial Roles, Leadership Roles Model, and Competing Values Framework, as well as the Meta-Category Leadership Taxonomy are explained using models, diagrams, and matrices. Theoretically sampled case studies are analyzed using several typologies to explore leadership behavior and traits across leaders in different projects and organizations. Participant assessments and 360° peer reviews are triangulated with qualitative and quantitative statistical measures to test hypotheses and analyze correlation between leader skills, traits, and behaviors across their projects. E-mail address: [email protected] (K.D. Strang). 0747-5632/$ - see front matter Ó 2004 Published by Elsevier Ltd. doi:10.1016/j.chb.2004.10.041

ARTICLE IN PRESS 2

K.D. Strang / Computers in Human Behavior xxx (2004) xxx–xxx

Findings and research value. This research identified a gap between extant organizational leadership theory and applied leadership behavior in effective projects. The preliminary results are surprising in a few cases but they generally support the proposition that effective leadership behaviors in any context are partly explained by leader traits, skills, and personality. The most unusual finding was these projects actually show that leadership principles from management science are not universally applicable (observable) in effective and efficient projects – instead the project leaders exhibited self-management theory and applied specific leader behaviors according to the situation. These findings, albeit on a small statistical scale of global significance, are a catalyst for continued and broader reflective leadership research using case study methods and hypotheses replication with these constructs. Ó 2004 Published by Elsevier Ltd. Keywords: Technology project leadership; Leader traits; Technology manager roles

1. Introduction The paper begins with a brief theoretical review of relevant contemporary leadership traits, skills, and behaviors principles, including interesting emergent views, and ending with relevant typologies for assessing leader behavior, traits, and skills in order to test the hypotheses. It is posited the gap between extant theory and leadership in technology projects lacks practitioner reflection-in-action, in that existing literature does not prove what effective global project managers actually do within the dynamic situational context of multiple competing demands. To examine this, technology case studies are analyzed quantitatively (using ANOVA-2), then qualitatively (using various leadership roles model typologies), to test the hypotheses. The paper concludes with a synopsis of the research and results of the hypotheses testing, followed by cited references. 1.1. Discovering why leadership is valuable at all organizational levels Managing and leading people and projects to the satisfaction of the sponsor(s), the team, and the stakeholders, generally requires interpersonal ability, technical competencies, and cognitive aptitude, along with the capability to understand the situation and people then dynamically integrate appropriate leadership behaviors (Strang, 2003a). Leading and managing functions are complimentary yet interdependent, and both are needed in the same position (Kotter, 1990), at all levels in the organization. Variations of leadership trait theory emphasizes the social capability of leadership effectiveness as part of a managers role-set in that ‘‘networking, the development and maintenance of cooperative organizational relationships, has been linked to managerial effectiveness’’ (Yukl, 1994 in: Schneider, 2002, p. 213); furthermore, organizational leadership ‘‘has been viewed as an interpersonal managerial role having to do with motivating subordinates’’ (Mintzberg, 1973 in: Schneider, 2002, p. 213).

ARTICLE IN PRESS K.D. Strang / Computers in Human Behavior xxx (2004) xxx–xxx

3

Cultivating relationships across the organization, particularly vertical and horizontal, multilevel and multi-domain, impact leadership, and can result in a reputation for effectiveness by meeting the varying expectations of supervisors, subordinates, and peers (Pfeffer & Salanick, 1975; Schermerhorn, Hunt, & Osborn, 2003; Tsui, 1984; Graen & Uhl-Bien, 1991 in: Schneider, 2002, p. 213). Thus there is a multifaceted dependency between leadership traits, role-sets, relationship building skills, and stakeholder expectation management, with the key point being that motivational leadership behavior across broad stakeholders increases leader effectiveness (see Exhibit 1). Leadership is typically classified as a characteristic or trait (ability to lead), an object/role set (behavior category required to get work done), adjective (type of managerial or executive quality), the application of power, or as a process (motivating, influencing, inspiring, coaching, mentoring, facilitating, etc.). Leadership is also classified by the factors that impact it, which from a meta-level standpoint are: the organizational context, including the larger macro-environment, the business situation (task environment), and the internal organizational and/or project characteristics. The diagram illustrates an abstraction of the leadership epistemologies, shown within the high-level environmental force categories that impact them. The focus of this paper is exploring effective leadership skills, traits, and behaviors, in terms of theory and practice, using reflective case studies. Literature reveals the importance of a combination of skills and traits for effectively managing and leading people in all types of organizations, all sizes, all forms, ranging from public to private to non-profit sectors, including business units, alliances, divisions, departments, projects, and so on. Effectiveness in this sense generally means improving productivity, increasing satisfaction, for both individuals and groups, as well as being promoted to higher levels or growing a small business into a larger one. On the other hand, scenarios involving self-managed teams, communities of practice, leadership substitutes (cohesive/mature team, task autonomy, bureau-

Power: •Expert / Referent •Formal / Legitimate •Relations / Networks •Reward / Coercive •Influence / Charisma

Behaviors: •Role Sets •Functions •Boundaries •Patterns

Traits: •Cognitive / IQ •Emotive / EQ •Knowledge / Competency •Attributes •Maturity

Processes: •Practices •Methodologies •Contingency Models •Situational Techniques

This Paper

Exhibit 1. Leadership epistemological contexts and disciplines.

ARTICLE IN PRESS 4

K.D. Strang / Computers in Human Behavior xxx (2004) xxx–xxx

cratic environment, etc.), or just plain good luck, can reduce or eliminate the need for significant management or leadership (Strang, 2004a). Nevertheless, based upon research and experience, situations where management and leadership are not needed are rare, yet even when they exist, alternative forms of self-management and dynamic informal leadership processes (i.e., influence and persuasion) can improve performance and satisfaction (Strang, 2004a). Repeated behaviors integrated together can improve leader effectiveness, such as possessing a high degree of self-confidence and stress tolerance were found to increase the capacity of leaders to perform cognitive functions, especially in crises events (Mumford & Connelly, 1991). To a finite extreme, it is proposed that optimal leadership behavior means balancing behaviors and functions in complex combinations customized for the situation, in an optimal manner to address the needs of the stakeholders, as well as to compensate for weaknesses in the leader/manager, as well as leverage or adapt environmental neutralizers/constraints. This may result in a leader purposefully developing/increasing certain traits to balance off others, as part of the perceived customized situational behavior, or in an effort to cognitively overcome weaknesses, or leverage/innovate strengths in the leader (self), in stakeholders (followers), or in the context (environment). Some research has investigated this and found leaders in a team (as well as teams of leaders such as senior management) compensate for, develop and enhance memberÕs behavioral strengths and weaknesses, in a holistic manner, to increase overall effectiveness and performance (Bradford & Cohen, 1984; Schermerhorn et al., 2003). Exhibit 2 depicts a systemic model showing a process view of a leader balancing skills, traits, and behavior interactions in a project situation (as a point in time from an organizational level of analysis) while interacting with various contextual variables and competing demands, to achieve desirable outcomes, by motivating and influencing followers and stakeholders. The focus of this research is how leadership Delay Primary/direct impact Secondary/indirect impact Feedback/learning la De

Context variables: Substitutes Neutralizers Turbulence/Stability

y

Leader Traits: Cognitive, Emotive, Knowledge

Intervening variables: Follower Commitment Task Structure Relationships

Project Leader Behavior

Delay

Delay

Delay

Leader Power

This Paper

y la De

Outcomes: Production Success Follower Motivation Stakeholder Satisfaction

Success Criteria

Exhibit 2. Leadership processes systemic model.

ARTICLE IN PRESS K.D. Strang / Computers in Human Behavior xxx (2004) xxx–xxx

5

behavior, traits, and skills interact in different combinations by leaders across different projects. Additionally, as the model implies, leaders are driven to balance behavior and functions to deal with competing values from the organization (time, events, culture, and macro-environment), from the team, and from other stakeholders, such as: focusing on task structure versus interpersonal relationships; democracy versus bureaucracy; controlling versus empowering; directing versus influencing; contingency planning versus risk taking; stability versus change orientation; just to name a few leader behavioral continua (Lombardo & McCall, 1978; McCall, Lombardo, & Morrison, 1988; Quinn, 1988).

2. Methods Leadership can be studied from the input, output, and cause/effect perspectives whereby approaches differ depending on the level of analysis, such as intra-individual (psychology/self), dyadic (leader/follower), leader-team (social-psychology/group interactions), or organizational versus project domains (and others). Most leadership theory, whether studied at the organizational or project team level, is approached from a leader-to-follower performance causality, which is the approach taken in this paper, but recent studies have also examined leadership from other interesting angles such as from a follow-to-follower and follower-to-leader impact. Exhibit 3 illustrates the concept that leader disciplines are analyzed from mainly four theoretical levels: 1. community, organizational, societal – organizational behavior, transformational change processes, systemic macro and task views of leadership; 2. team dynamics/group processes – social-psychology, group influences/interactions, group decision-making;

Organization Business Unit, Firm, Community, Society

This paper Group Department, Team, Community of Practice Levels of Analysis

Dyadic Leader-Follower

Intraindividual Self, Psychoanalytic

Exhibit 3. Leadership levels of analysis.

ARTICLE IN PRESS 6

K.D. Strang / Computers in Human Behavior xxx (2004) xxx–xxx

3. dyadic (two entities view) – leader–follower and more recently follower-to-leader, processes/relationships; 4. intrinsic view/self-leadership (intra-individual) – psychological, motivation, time management, self-management, individual decision making. Contemporary leadership research has tended to use multiple levels of analysis (individual, dyad, and group) simultaneously to assess hierarchical boundary domains of one or more leader theories, and these studies found leader behaviors differed between higher-level (dyad group) and cross-boundary-level (leaders in boundary spanning positions) within the same study and as compared to assertions previously documented in the literature (Dansereau, Alutto, & Yammarino, 1984); thus multiple levels of leadership analysis can reveal unexpected and enriched correlation (Dubin, 1976). This paper uses mainly the group and intra-individual trait/personality levels to analyze leader behavior and skills. Furthermore, as the diagram suggests, the various levels analyzed can and usually do interact with one another, thus complicating the study of leadership at any point or situation, and across time periods. This research is primarily focused on effective leadership behavior, traits, skills, and their interaction with one another as units of analysis. The leaders selected were considered effective. Since their project deliverables satisfied all sponsors and most stakeholders. 2.1. Research questions It is propositioned here that effective project leaders know their own behavior, trait, skill, strengths, and weaknesses, they improve weak areas and leverage the strong ones, they balance skills and behaviors according to the situation, taking into account themselves, the environmental context (organization, etc.), the group/team (followers), the sponsors, the stakeholders, and of course the priority of each in this circumstance. Management science advocates that effective leaders are not necessarily born with all the necessary skills and propensities to lead but they will leverage their innate capability, learn appropriate skills, and act out appropriate behaviors, and this is applicable to both organizational and project leaders. To explore this, a theoretical review of leadership trait, skill, behavior principles and taxonomies is documented. These are then applied to theoretically sampled project case studies for analysis. It is propositioned here that effective leaders will differ from ineffective leaders in their natural leadership propensity, especially in project management contexts which can be very demanding due to time, cost, schedule, and budget constraints, as compared with business-as-usual, which requires active (not passive), relationship and change-oriented (not just task focused) behaviors, possibly all in the same project and situation. Therefore, effective project leaders will display more relationship and change-oriented behaviors and in turn they will likely have more extrovert personalities. The research hypotheses examined are:

ARTICLE IN PRESS K.D. Strang / Computers in Human Behavior xxx (2004) xxx–xxx

7

(1) Organizational leadership principles such as contingency, situational, and trait theories will apply and be observable in the project domain, for project leaders; therefore, effective project managers typically do not practice alternative theoretical leadership trait and skill principles – such as personality-driven charisma and emotionality maturity, since it is propositioned project managers utilize core competencies linked to management science theory not emergent leadership epistemology. (2) Since project leaders are essentially managers, they will adopt more managerial roles (directing, monitoring, etc.) than leadership roles (facilitating, mentoring, etc.), and both dimensions in combinations, but they will not apply multiple managerial and leadership role sets concurrently. (3) Effective project leaders have little time for training, and by sheer industry demand, the natural selection process favors those whom have a personal competency and personality combination that is conducive and complimentary to the dynamic project management role; therefore, effective project managers will display more relations and change-oriented traits, supported by extrovert and affective-dominate personality profiles.

2.2. Research methodology The cases contain six project leaders managing six different projects within four different organizations in roughly the same multi-year time range. Six leaders are used since it is recommended that between four and 10 cases be used in this type of research (Eisenhardt, 1989; Scho¨n, 1983; Yin, 1994). Although that guideline was recommended for theory-building exercises, by including six leaders, this will increase internal validity and reliability by reducing perception biases, broadening the standpoint, and triangulating the data sources. Construct validity is achieved by applying the management science principles and leadership models which have been used and well-publicized at the individual level of analysis (Yin, 1994), which also allows for referential integrity comparisons of these basic leadership theoretical principles between general management and the project domain. Flowcharts, process models, categorized/ordered matrices and other graphics are used to enrich theory discourse (Miles & Huberman, 1984). The direct and direct-participant observation techniques were used, supplemented by project documents and direct verification of some facts (by interviewing the leader for some variables), along with confirmation of certain facts using 360° interviews (talking with team members, peers), for data and method triangulation. The personality profile tests were conducted by human resource personnel and/or external service providers and confirmed by the subject along with at least other peer and/ or co-worker (to ensure the ratings were representative). Leader behavior observations were triangulated from the different sources to establish a consensus. Observation recordings were made using positive integer scales (Likert fashion), and the later decision was to include only the purposeful, obvious, ‘‘high-probability’’ observations in this analysis. This means that only ‘‘high’’ observations of particular leader

ARTICLE IN PRESS 8

K.D. Strang / Computers in Human Behavior xxx (2004) xxx–xxx

behaviors were included in these taxonomies, such as a 5–7 on a Likert scale of 1–7 (where 7 is always and/or high) for a particular variable. This paper is part of a series of academic research studies involving 21 international project cases across multiple industries. The project cases were organized into a research database for this research, thus a system key index is listed which can collate cases across the research papers, and all personal/company identifying information is removed to protect confidentiality and competitive advantages. Although the direct and direct-participant observation techniques were used, the author was not a subject for the unit of analysis in these particular cases. The case data is a point-intime representation, observed on or about the duration of the project (and in some cases including data gathering/observations conducted shortly afterwards, but no more than three months beyond project completion). Senior students and professionals assisted with the research on either a volunteer or paid basis. Ethical human resource selection procedures were applied, involving criteria such as high marks in social science research methods courses, as well as grade confirmation and reference checking.

3. Effective leadership as traits, skills or behaviors? Many writers have argued for and against whether leadership is an inborn trait or that it can be developed. In general, the consensus is that leadership can be developed and is a combination of qualities, knowledge, and processes (Adair, 1997; Collins, 2001; Yukl, 1998). Leaders need more than pure ability, skill, and innate attributes – as pointed out in early management science research based upon 124 trait studies, ‘‘a person does not become a leader by virtue of the possession of some combination of traits. . . the pattern of personal characteristics [and behavior] of the leader must bear some relevant relationship to the characteristics, activities, and goals of the followers’’ (Stogdill, 1948, p. 64). Later research reviews of 163 trait studies covering 1949–1970 by the same author confirmed traits and skills were related (increases the likelihood that a leader would be successful) to leader effectiveness (Stogdill, 1974). Traits are considered to be patterns of individual attributes, such as skills, values, needs, and behaviors, which are relatively stable in the sense that they tend to repeat over time. Skills are abilities to do things, satisfy needs, and make decisions, in an effective manner. The Exhibit 4 portrays a model of the leadership skill, trait, and behavior interaction process as a systemic model. The consensus from literature is that there are patterns of specific leadership skills, activities, and behaviors which are related to leader effectiveness, namely: acquiring status through demonstrating group facilitation and performance, intelligence, task understanding, initiative, persistence, self-confidence, sensitive to others needs, willingness to accept responsibility, ability to exert dominance and control as necessary (Stogdill, 1948). The four basic traits which seem to be shared by most successful leaders are (Hellriegel, Slocum, & Woodman, 1992): intelligence, maturity/broad

ARTICLE IN PRESS K.D. Strang / Computers in Human Behavior xxx (2004) xxx–xxx

Situational demands

Technical skills

Task activities

Social skills

Relations activities

Cognitive skills

Change activities

9

Leader behavior

Traits:stable, repeating behavior patterns

Exhibit 4. Leadership skill–trait–behavior process model.

interests, strong inner motivation/drives (self-confidence and self-efficacy), and consideration of others needs/values (emotive sensitivity). Most writers typically discuss several broad meta-categories of effective leader traits, such as: cognitive ability, intelligence, emotive, affective ability, and technical knowledge, and functional capability. It is proposed here from research and experience that knowledge and capacity to act are in the cognitive category, but they could just as easily be treated as a separate group as they are often approached in knowledge management (Strang, 2003b). In the organizational and project domains, a leadership skill is considered an ability to translate knowledge into action that results in a desired performance (Schermerhorn et al., 2003). Competency is typically viewed as a cumulative term in leadership theory to encompass knowledge, skills, attitudes, and behaviors, while an applied project management action science definition transforms this to three dimensions of project management knowledge, performance, and personal competence (Boyatzis, 1982; Parry, 1993; Spencer & Spencer, 1993). Personal competency is most relevant to this research, which generally includes the following factored elements: achievement and action, helping others and human service, impact and influence, managerial, cognitive, and personal effectiveness (PMI, 2002, 2003). The table in Exhibit 5 summarizes the common traditional and emerging leader traits and skills (contents adapted from: Katz & Kahn, 1978; Stogdill, 1974; Yukl, 1971, 1998, pp. 243–258). The project management personal competency factors tend to be already captured in the more refined elements defined from management science leadership theory as described in the table, and are thus not further segregated – defining more detailed leadership elements (root definitions) will improve the quantitative and qualitative behavioral analysis applied in this research. What is interesting about effective leadership traits from emergent research is the lack of a specific skill namely intelligence, which according to the literature means a leader does not have to be highly intelligent to be effective (and in fact studies show that high intelligence could be a weakness if applied incorrectly at higher management levels), but instead a leader must be able to function at a moderately high

ARTICLE IN PRESS 10

K.D. Strang / Computers in Human Behavior xxx (2004) xxx–xxx

Exhibit 5 Synopsis of effective leader traits and skills from management science studies Traditional effective leadership traits

Traditional effective leadership skills

Emergent effective leadership traits

Organized (also an administrative ability)

Ambitious, achievement oriented

Moderately high achievement orientation

Clever (intelligent) Conceptually skilled Knowledgeable about the work

Assertive Decisive Dominant (power motivation) Energetic (high activity levels) Alert to social environment Cooperative

Socially skilled Fluent in speaking (could also be a technical skill) Diplomatic Tactful Creative Persuasive

Dependable Self-confident Adaptable to situations Tolerant of stress Persistent Willing to assume responsibility

Internal locus of control High energy level Emotional maturity Socialized power motivation Personal integrity Self-confidence Stress tolerance Low need for affiliation

cognitive level to balance skills, contexts, and make decisions (McCall et al., 1988; Yukl, 1998). Other research and experience find that part of a leaderÕs task is to manage complexity and this individual trait is therefore related to intelligence. In addition, attaining a leadership position and being effective may be a reflection of an individualÕs needs or motivations. The search for associations between personal attributes and effectiveness of leaders has been a major objective of research, but this has led to inconsistencies in relation to individual differences as an explanation for leadership effectiveness. Social skills, such as the ability to perceive stakeholder and follower needs and responsiveness, and leader flexibility in response, has been linked to effective leader interpersonal skills. It has been speculated that socially developed leaders have more experiences, cause–effect process knowledge, perception skills, and understanding of organizational context (Zaccaro, Gilbert, Thor, & Mumford, 1991). Cognitive/conceptual ability has been described as a social complexity ability to think in a multidimensional, abstract manner, while synthesizing information at various levels of abstraction (Jacobs & Jaques, 1987). This is certainly characteristic of many managerial and project manager roles. Contemporary leadership studies segment leader skills into the following metacategories (Katz & Kahn, 1978; Yukl, 1971):  social/interpersonal/emotive – understanding and ability to modify human behavior, use communications processes, social and cultural understanding, influence skills, diplomacy, cooperation, disposition, empathy;  cognitive/conceptual – analytic ability, logical thinking, abstract concept understanding and articulation, creativity, problem solving, reasoning, capacity to act;

ARTICLE IN PRESS K.D. Strang / Computers in Human Behavior xxx (2004) xxx–xxx

11

 technical/knowledge/administrative/process – method knowledge, ability to use tools, physical capability, linguistic ability, experience, knowledge to apply.

3.1. Alternative theoretical leadership trait and skill typologies Some research exploring leader personality traits correlate the following with leadership effectiveness: conscientiousness, extroversion, agreeableness, open to experience and low on neuroticism (a narcissistic tendency). Other traits linked to leader effectiveness are humility and humor (mentioned later in the Level 5 Hierarchy) and the trait that has attracted the most attention in recent years is charisma – regarded by many writers as the essence of transformational and organizational change leadership. 3.1.1. Leadership virtual reality Several interesting taxonomies have been emerged from the literature for describing and studying leadership traits. One interesting perception is the five virtual reality environments which describes an evolutionary leadership maturity paradigm shift from the industrial age of management science through to a high-quality, high-trust, and high-morality based (perhaps altruistic) developmental ambition (Fairholm, 1998). The five virtual reality environments are considered streams of ideas, possibly viewable as a continuum from managerial control to spiritual holism, with some historical chronology (but they are not sequential stages), as summarized below: 1. leadership as management – scientific management principles emphasizing the leader functions and task orientation, such as from writers like Taylor, Fayol, Mintzberg; 2. leadership as excellent management – transactional approach, total quality management (Juran), emphasizing integrity, stakeholder expectations management; 3. values leadership – relational view, incorporating group dynamics and cultural aspects, visioning, from writers such as Hart, Fairholm, and (Walker, Hampson, & Peters, 2000); 4. trust leadership – transformational approach, applying motivation, behavioral psychology, group culture, possibly some of the charismatic tendencies; 5. spiritual leadership – the triple bottom line concept (Walker, 2000) combined with ethics, incorporating morality principles from philosophy and theology, advocating that ‘‘leadership comes out of the leaderÕs true self, his/her inner spiritual core’’ (Fairholm, 1998).

3.1.2. Level 5 hierarchy It was mentioned that early management science theory posited that effective leadership was not based solely on traits. The level 5 hierarchy model is an interesting description of a leadership taxonomy which somewhat parallels the leader trait/ maturity concepts evident in transformational, charismatic, virtual reality leadership

ARTICLE IN PRESS 12

K.D. Strang / Computers in Human Behavior xxx (2004) xxx–xxx

theories. In this theory, leadership maturity evolves from level 1, highly capable individual with talent, skills, knowledge, to level 2, contributing team member with group cooperation skills, to level 3 as a competent manager with people organization skills, to level 4 as an effective leader having transactional and motivational qualities, finally to level 5 leadership, balancing ‘‘personal humility plus professional will’’ (Collins, 2001, pp. 69–70). The theoretical model is shown in Exhibit 6. What is significant about the level 5 hierarchy is it applies to managers, nonmanagers, informal or formal leaders (even non-leaders and community champions), and it combines many of the proven principles of other organizational behavioral theories, such as visioning, positive attitude, work ethic, motivation, cultural appreciation, and so on. It is not a sequential scale (as the diagram indicates, leaders can jump levels), and it is empirically based, not theoretical, meaning that the theory summarizes study findings as a type of leadership typology, but it does not describe applied theory. Its credibility should be noted since it used a 30-year longitudinal and objective study of 1435 Fortune 500 companies, to determine what senior managers (CEOÕs etc., such as Iaacocco at Chrysler, Smith at Kimberly-Clark) did (or did not do) to transform companies. Two key propositions the authors construct from their research were (Collins, 2001): 1. leadership is not necessarily genetic, as the model suggests, it can be developed if one has intrinsic attitude (possibly form positive life experiences and quality parental upbringing), albeit there were only 11 level 5 leaders found; 2. ‘‘abilities’’ in the level 5 hierarchy are not sufficient by themselves to transform a company into a success – other macro and task environmental factors need to be supportive or conducive to allowing this change.

Le ad er sc an

sk ip

sta ge s

LEVEL 5LEADER Enduring personal humility, professional will LEVEL 4 EFFECTIVE LEADER Catalyzes commitment, compelling vision, stimulates group to high performance standards

LEVEL3 COMPETENT MANAGER Organizes people, resources toward effective, efficient pursuit of predetermined objectives LEVEL2 CONTRIBUTING TEAM MEMBER Contributes to achievement of group objectives, works effectively with others in group setting LEVEL1 HIGHLY CAPABLE INDIVIDUAL Makes productive contribution through talent, knowledge, skills and good work habits

Exhibit 6. Level 5 leader trait hierarchy.

ARTICLE IN PRESS K.D. Strang / Computers in Human Behavior xxx (2004) xxx–xxx

13

3.1.3. Emotional maturity and personality An interesting measurement of leadership traits and skills is the theory of emotional maturity (initially developed by Peter Salovey and John Mayer then extended by Daniel Goleman). Emotional maturity has also been termed as emotional quotient, emotional intelligence, and affective capability/behavior. There are four stages of emotional maturity: identifying, using, understanding, and managing emotions, in terms of reactions to external realities and in helping others manage their relationships. Research findings indicate leaders with high emotional intelligence have potential to be more effective leaders (Goleman, 1998, 2000). Emotional maturity theory overlaps with the alternative trait views such as the higher stages in the level 5 hierarchy as well as the values/trust elements of virtual reality leadership theory. Emotional maturity also parallels and integrates several detailed behaviors (such as concern for team members) in the relationship meta-categories, and it can be viewed as a social skill. Emerging research has found empirical support and construct validity for the Big Five model of personality, which consists of five personality traits associated with effective leadership: emotional stability, extraversion, openness to experience, agreeableness, and conscientiousness (Tett, Jackson, & Rothstein, 1991; Viswesvaran & Deniz, 2000). This construct also overlaps with other trait theories, and although both interesting and relevant to studying leader trait behaviors, it is less used than the MBTI (discussed below); furthermore, it has limitations such as a small number of categories, quantitative measurement of its qualitative psychometric properties (the same problem as with most other taxonomies), and the lack of field experience knowledge applying it as compared with other the leader personality constructs. The Myers–Briggs Type Indicator (MTBI) is a similar measure of emotional maturity which originated from the psychology social science field, and it is grounded upon principles such as: behaviors are not random but have patterns, they reflect preferences, they are fairly orderly and consistent over time – as developed by Jung (1971), Katherine Briggs (1926), Myers and McCaulley (1989), and later Consulting Psychologists Press (1975). The MBTI offers a logical model and assessment construct of human behavior, it identifies and measures the strength of preferences that comprise various personality types, but it does not assess intelligence, emotional stability, capacity to learn, maturity, or skills (Myers & McCaulley, 1989). The outcome variable of an MBTI test is a personality type, which represents a specific alignment/combination of preferences compromising a basic orientation toward the outer or inner world: Extraversion (E) vs Introversion (I); the kind of perception used: Sensing (S) vs Intuition (N); the type of judgment used: Thinking (T) vs Feeling (F); and the basic lifestyle adopted for dealing with the environment: Judgment (J) vs Perception (P). There are dominant, auxiliary, tertiary, and inferior functions, and they indicate a set of preferences and use, which is the application of MBTI here, being that the dominant function is the one that is most conscious, well-developed, and most often used, by the leaders. The analysis for this paper applies MBTI as substitute measure of emotional maturity since that construct was documented and available as a common baseline for all leaders in the case studies, although its limitations are recognized. The MTBI

ARTICLE IN PRESS 14

K.D. Strang / Computers in Human Behavior xxx (2004) xxx–xxx

is proposed as an excellent complimentary dimension which has been successfully applied in other leadership studies to correlate detailed applied leader skills and behaviors (Charon, 2003). However, caution in applying MBTI has been argued due to its conceptual and psychometric weaknesses, because the raw scores can be slight for any personality dimension (Johnson, Mauzey, Johnson, Murphy, & Zimmerman, 2001; Michael, 2003), which means that summary indicators can stereotype and bias assessment, and most importantly, this tool does not analyze how the organizational context impacts the behavior and interaction of its constituent elements. 3.1.4. Charisma Charismatic leadership has been described as an effective leader trait and behavior, while the term itself is based on Greek etymological origins of its meaning as a special gift (trait) that select individuals possess which gives them the capacity to do extraordinary things. According to research, charismatic leaders possess specific personality traits (or their combination), and display unique behavioral patterns; they become strong role models for the beliefs and values they declare; they appear competent; they focus on ideological and moral goals; they communicate high expectations for followers; they exhibit confidence in follower ability to meet expectations; and they have profound and unusual effects on followers (House, 1977). Contemporary theory writers state that charisma is the result of follower perceptions and attributions influenced by actual leader traits and behavior, the situational context, as well as individual and collective follower needs (Yukl, 1998, pp. 297–298). Charisma could be associated with another form of leadership epistemology: transformational contingency processes. Charisma could be considered part of the change-oriented behaviors meta-category, as well it could be considered related to the traits of a level 5 hierarchy leader. Charisma as a trait or process (depending on the research context and perspective) is not specifically studied further in this paper. 3.2. Leadership or management behaviors? Management science, the field of study where traditional leadership theories have matured in their articulation (mostly during the 20th century: Yukl, 1998), describes leading as a function within management, with management itself defined as being variations of planning, organizing, leading, controlling, and staffing (Koontz, OÕDonnell, & Weihrich, 1980). Management is generally accepted to refer to a broad supervisory role, usually accountable for human assets, capital budgeting, work activities, and/or other resources, involving the application of some or all of the functions named above (Mintzberg, 1975; Zalesnik, 1977), while leadership is in comparison more specifically focused on subordinate/follower motivation. Some research delineates managers and leaders as being at opposite poles, at extreme ends of a spectrum, and provides a dozen examples starting with ‘‘the manager administers; the leader innovates’’ and ending with ‘‘the manager does things right; the leader does the right thing’’ (Bennis, 1989, p. 45), and in later studies researchers confirm that managers solve routine problems, but leaders identify problems to be solved (Bennis & Nanus, 1997, p. 39).

ARTICLE IN PRESS K.D. Strang / Computers in Human Behavior xxx (2004) xxx–xxx

15

An alternative perspective of leading and managing propositions it is necessary to achieve a balance between management and leadership and as such they are divided into two separate processes and competencies respectively, which could be applied (balanced) by a single individual (Kotter, 1998). While it is acknowledged not everyone can lead and manage effectively, Kotter and other writers argue that companies should recruit and develop executives with combined leader-manager competencies (Duffy, 2001; Kotter, 1998). Although some managers are leaders, and some leaders are managers, according to accepted theory, leading and managing are not identical (Goleman, 1998; Lombardo & McCall, 1978; Verma & Wideman, 1994; Yukl, 1998; Zalesnik, 1977). While managing includes leading it also requires significant cognitive and general-purpose skills necessary for planning, budgeting, controlling, organizing, and so on. Leading on the other hand involves a high degree of social-psychology aptitude to support relationship building, communications and change management using influence, inspirational and motivational techniques. Exhibit 7 suggests a balanced integration of key management functions with leadership traits and skills. Leaders tend to focus on effectiveness, do the right things, formulate strategies, and establish goals, while managers strive towards achieving efficiency, do things right, organize, plan, and so on (Bennis, 1989). A consequential definition of leadership that differentiates follower behavior from management outcomes, describes it as an ‘‘influential increment over and above mechanical compliance with routine direction of the organization’’ (Katz & Kahn, 1966 p. 302 in: Schneider, 2002, p. 210) – this supports the process view of leadership theory from management science (Yukl,

Organizing Resolving

Staffing Recruiting

Planning Learn

Visioning

Communicate Innovate Inspire

Monitoring

Leading

Controlling

Motivating Management & Leadership Balance

Exhibit 7. Integrated leadership skills and management functions.

ARTICLE IN PRESS 16

K.D. Strang / Computers in Human Behavior xxx (2004) xxx–xxx

1998). KotterÕs definition of managing frames it as Ôcoping with complexityÕ, which is in keeping with his assertion that a key driver in the business environment is the emergence of large, complex enterprises, and thus the emphasis for strong management (Kotter, 1990). Leadership, he explains, is about ‘‘coping with change’’, driven by competition, technological change, and new workforce demographics. Despite their uniqueness in terms of process, there seems to be common trait, skill and behavior elements connecting leading and managing, namely: communicating, relationship building, people motivating, work promoting, and change making. This convergence of managing and leading has been depicted as four primary overlapping leaders skill processes: decision-making, information exchanging, influencing, and building relationships (Yukl, 1998, pp. 34–35). The difference then between leading and managing arises through the method and context through in which they are accomplished; this is the contingency, trait, skill, behavioral, and integrated perspectives of organizational leadership theory. 3.3. Analytical typologies of leader traits, skills, and behaviors This section will review several appropriate and credible management science methodologies and constructs for analyzing leadership trait, skills, and behavior. Most have been used for manager-level analysis in utility and service sectors, but these taxonomies are very generic and therefore should be suitable for multiple industries as well as at the project level. So rather than reinvent the wheel, and also to provide for opportunity to replicate earlier research (and allow for increased external validity of this paperÕs findings), existing frameworks will be used for this paper. Many of these are based on job functions and role categories. Observation of behaviors, such as job function studies, have proven most useful to reveal the true nature of manager and leader tasks used in actual work practice. Several well-known studies were completed using the survey/factor analysis method (Hemphill, 1959; Mahoney, Jerdee, & Carroll, 1965; Page & Tornow, 1987; Tornow & Pinto, 1976). These studies evolved into two dimensional taxonomies based upon task-oriented behavior (deliverable focus, utilizing resources efficiently, maintaining stable and reliable operations, and so on), and also on relations-orientated behavior (improving teamwork, helping and motivating people, etc.). Subsequent research using various methods advanced theory into the benchmark Managerial Position Description Questionnaire (MPDQ) which defined a management role as: supervising, planning/organizing, decision making/problem solving, monitoring indicators, controlling, representing, coordinating, consulting, administering and this was again revised to add: clarifying roles/objectives, informing, delegating, supporting, motivating/inspiring, conflict resolution/team building, developing/mentoring, praising/recognizing, rewarding (Bass & Stogdill, 1990). These leader behavior studies append substantial and relevant detail to the broader Exhibit7: Integrated Leadership Skills and Management Functions theoretical model presented earlier. Several credible measurement frameworks for studying leadership behavior are already well-documented in the literature (see: Yukl, Gordon, & Taber, 2002).

ARTICLE IN PRESS K.D. Strang / Computers in Human Behavior xxx (2004) xxx–xxx

17

3.3.1. Managerial roles Other researchers, such as Mintzberg, used direct observation rather than questionnaires to produce taxonomies of manager/leader roles. One good role observation study documents a useful and widely applied construct consisting of three manager function categories with ten roles (explained below from: Mintzberg, 1975). Subsequent researchers have built on this framework. (1) information processing (a) disseminator – essentially communications; (b) monitor – meetings, environmental scanning; (c) spokesperson – pubic relations, shareholder/owner relations; (2) decision-making (a) (b) (c) (d)

entrepreneur – innovations, continuous improvement and learning; disturbance handler – problem analysis and resolution; resource allocator – planning, staffing; negotiator – contracts, conflict resolution;

(3) interpersonal (a) liaison – relationships, networks, alliances; (b) figurehead – formal authority, role model; (c) leader – in MintzbergÕs view to integrate subunits and hire, train, direct, praise, promote, fire. He noted effective leaders and managers are those whom can be introspective about their jobs, can find systematic ways to share privileged information, possess a Ôbig pictureÕ mentality, and make room in (gain control of) their busy schedules for their own time to think, support subordinates/peers, and organize structure/tasks (Mintzberg, 1998). 3.3.2. Leadership roles model QuinnÕs 8 Leadership Roles Model of managerial behavior is also very appropriate to assess leadership behavior. This model is proposed to be credible since it was developed from earlier theoretical models (Quinn, Faerman, Thompson, & McGrath, 1996; Quinn & Robrbaugh, 1983), namely the 10 Managerial Roles (Mintzberg, 1975) discussed above. The leader roles construct is designed around four leader quadrants: open systems (illustrating entrepreneurship, innovativeness and adaptive leadership), human relations (exhibiting people/relational leadership), internal process (emphasizing stability leadership and possibly basic managerial functionality), and rational-goal philosophy (suggesting project and task orientation). These quadrants are differentiated by two dimensions: a polarization of internal versus external focus (horizontal axis) and flexibility versus control orientation

ARTICLE IN PRESS 18

K.D. Strang / Computers in Human Behavior xxx (2004) xxx–xxx

Monitor

In te St ab rna lP ilit y Le roc es ad s er sh ip

R Ta atio sk na lG Le oa ad l er sh ip

Internal Focus

Ad Op ap en Sy tiv e st e Le ad ms er sh ip

Coordinator

Stability

Facilitator

Innovator

Hu Pe m a n op Re le Le sou ad r er ces sh ip

Mentor

Flexibility

(vertical axis). This is abstracted in the next exhibit, which combines MintzbergÕs and QuinnÕs ideas. As the diagram (Exhibit 8) illustrates, each of the four quadrants of the Leadership Roles Model contains two leadership roles (which closely parallel MintzbergÕs managerial roles discussed earlier), each one respectively closer to a particular dimension of internal/external focus and flexibility/stability continua. It is suggested that the way to interpret the Leadership Roles Model is to view each of these roles as being an approximate continua on the internal/external focus and flexibility/stability dimensions of leader behavior, for either a brief time, or as a habitual practice. The subject does not necessarily require a title according to the Leadership Roles Model – the captions simply refer to anyone practicing the typical leadership functions as described. A leader or manager will usually exhibit several of these role behaviors in their job but perhaps one or a few of them would be dominant at a single point of time or during a period of a career (or even during a phase of a project). The purpose of the Leadership Roles Model is to define a set of overlapping roles that emphasize the need for a leader to have both an internal and external unit/organizational focus a well as accommodate stability along with the need for adaptation and growth (Denison, Hooijberg, & Quinn, 1995, p. 528). What is also important to keep in mind is that models such as these typify job roles in an organization, with employment performance responsibilities, thus they are not representative of individual behaviors in a family, social or volunteer setting.

Broker

Director

Exhibit 8. Technology project leadership roles.

External Focus

Producer

ARTICLE IN PRESS K.D. Strang / Computers in Human Behavior xxx (2004) xxx–xxx

19

The adaptive leadership quadrant contains the innovator and broker roles that are linked to open systems theory (cognition focus, decision making, problem solving, conflict resolution, evolution, knowledge creation/management). An innovator is creative, inspires and encourages others, while facilitating organizational change. A broker maintains external relationships/networks, is politically astute, negotiates to acquire resources, and continuously scans the macro-environment for partnership opportunities. The task leadership quadrant comprises the producer and director roles that are tied to rational goal theory (contingency approaches, mostly focused on performance and effectiveness). Both roles in this quadrant strive to achieve goals external to the group. The producer is task-oriented, deliverable-focused, and thus motivates team members to complete the work as required. A director is more focused on governance and process in the sense that in this role objectives are communicated, rules/policies are defined, applied, and expectations are clarified. The stability leadership quadrant includes the coordinator and monitor roles that are associated with internal process theory (originating from classical management science typifying functional and bureaucracy characteristics, along with rules oriented behavior and macro-environmental stability focus). As such, both roles have a control and internal focus, and as evidenced from the literature, they are obviously typical of pure managerial behaviors. Coordinators maintain structure, schedule, coordinate functions, manage crises, as well as apply rules and standards. The monitor collects and distributes information, measures performance and efficiency, while providing a sense of continuity and stability. The people leadership quadrant decomposes into the facilitator and mentor roles that are tied to human relations theory (grounded on organizational behavior principles which address social behavior and emotional aspects, cultural and group norms, informal communications, motivation, and employee satisfaction). Both of these roles advocate internal relationship and team building. The facilitator is group focused, encourages expression of opinion, seeks consensus, negotiates compromise, fosters collective effort, and builds a cohesive team. A mentor is individually focused, used active listening skills, develops people through affective and empathetic behaviors, while being considerate, sensitive, caring, and fair. 3.3.3. Integrating leader roles, competing values, and behaviors Various leadership competing values and behavior models have been successfully used in the study of ethics, project management, manager/leader effectiveness, organizational design, and in fact it is purposeful to assess leadership using alternative methodologies such as observation and/or interviews, as well as applying multiple methods, but with consistent factor terminology across studies (Avolio, Bass, & Jung, 1999; Yukl et al., 2002). As implied earlier, the Leader Roles Model is credible construct and it maps into the generally-accepted Mintzberg managerial roles model (Mintzberg, 1975), as well as integrates with the transactional/transformational leadership continua (Bass, 1985). Exhibit 9 illustrates an integrative perception of this. For example, inspirational motivation and intellectual stimulation of transformational leadership can

ARTICLE IN PRESS 20

K.D. Strang / Computers in Human Behavior xxx (2004) xxx–xxx

map into the open systems quadrant as the role of innovator. In fact, it has been argued that the type of leadership is contextually sensitive (depends on the situation), is related to the task or behavior to be performed, and is also subject to characteristics and priorities of the leader and follower/group (Schein, 1985). The key principle gained from QuinnÕs Leadership Roles Model and Competing Values Framework is congruent with KotterÕs (and others) thinking that leaders and ‘‘master managers’’ must holistically balance between managerial philosophies, roles, traits, and behaviors without too much emphasis on any one model or aspect of theory (Quinn, 1988; Quinn et al., 1996). 3.3.4. Leadership behaviors taxonomy An emergent construct for analyzing leader skills, traits, and behaviors is the Leadership Behaviors Taxonomy (published 1981 and updated in: Yukl, 1998). This taxonomy was originally built from earlier leadership studies and models, based on a combination of leadership analysis approaches including factor analysis, observation/judgmental classification of leader actions, as well as theoretical deduction (Yukl, 1987). This model originally contained 19 leader behavior variables and was further expanded to 30, which supports more detailed analysis since there are more elements to classify observations. Earlier and alternative research taxonomies typically contained two meta-categories but this particular one was refined to incorporate additional epistemology, namely recognizing a third Ôchange-oriented behaviorÕ dimension along with task and relations (Yukl, 1998, pp. 59–62, 495–497), therefore it is that which makes this framework rational for use in descriptive project

Exhibit 9 Technology manager roles, values, and leadership behavior alignment Technology manager and leader roles

Competing values

Transformational behaviors

1. Innovator (creative problem solving, change, adaptation) 2. Broker (power, influence, resource acquisition) 3. Facilitator (conflict management, participative decision making) 4. Mentor (human resources development)

Open systems, adaptive leadership Open systems, adaptive leadership Human relations, people leadership

Inspirational motivation, Intellectual stimulation Inspirational motivation, Intellectual stimulation Individualized consideration, Support of others

Human relations, people leadership

Individualized consideration, Support of others

5. Coordinator (task analysis, coordination, financial control) 6. Monitor (information management, critical thinking) 7. Producer (productivity, efficiency) 8. Director (planning, goal setting)

Internal process, stability leadership Internal process, stability leadership Rational goal, task leadership Rational goal, task leadership

Transactional behaviors Focus on task performance to achieve organizational goals Focus on task performance to achieve organizational goals Focus on task performance to achieve organizational goals Focus on task performance to achieve organizational goals

ARTICLE IN PRESS K.D. Strang / Computers in Human Behavior xxx (2004) xxx–xxx

21

leadership research. The taxonomy shown in Exhibit 10 segments leader behaviors into the three functional categories, namely: task orientation, relationship focus, and change/adaptation management, which relate back to the ‘‘activities’’ shown in Exhibit 4: Leadership Skill–Trait–Behavior Process Model. The grouping of leader behaviors into general categories is done to facilitate descriptive factor analysis; even within the categories each behavior is itself generalized, so these taxonomies are guidelines only, as well, individual behaviors overlap and could be associated with multiple categories. The behaviors in the Ôtask-orientedÕ category have been associated with transactional leadership, while those in ÔchangeorientedÕ group are aligned with transformational leadership (Yukl, 1998, pp 60–62). A further generalization of this might be that the ÔrelationsÕ category includes team and participative leadership activities. However, Yukl considers participative leadership distinct from task and relationship-oriented behaviors (Yukl, 1998, p. 123) yet the earlier researchers recorded them into the task category. Experience suggests participative leadership is often change oriented.

Exhibit 10 YuklÕs leadership behaviors taxonomy Task focus

Relations emphasis

Change orientation

Organizes work activities to improve effectiveness. Develops milestones and action plans for a project. Determines what resources are needed to do a project. Clarifies role expectations for project members. Clarifies quality standards for task performance. Facilitates collection and dissemination of information. Actively monitors operations and performance. Resolves immediate questions or problems from team.

Actively listens attentively to a personÕs concerns. Actively provides support and encouragement. Socializes with team beyond work to build relationships. Publicly recognizes contributions and accomplishments. Provides individual role and/or behavior coaching. Consults with members on decisions affecting them. Helps team members (as a group) resolve conflicts. Keeps team (group) informed about upcoming actions.

Leads by example and models exemplary behavior. Encourages viewing things from multiple perspectives. Expresses confidence team can attain objectives. Creates sense of urgency, promotes change. Studies other projects to get ideas for improvements. Envisions exciting new possibilities for the organization. Develops strategies linked to organizationÕs vision. Builds coalition of stakeholders to get change approved. Creates task force to guide implementation of change. Suggests symbolic changes that affect the work. Empowers members to implement new strategies. Announces, celebrates progress supporting changes. Encourages/facilitates learning by team members. Experiments with new approaches.

22

Exhibit 11 Theory-sampled project case study summaries by context variable

Leader #1

Organizational sector

American public mutual

Org Stage in Life Cycle Structural orientation Organization topology Governance Line of business affecting project Strategic driver

Critical environmental constraint Project sponsor title Project leader title Project leader affective profile (with style & perceived Myers–Briggs rating) Project charter

Case B #E8504

Case C #C9803

Case D #C0109

Leader #2

Leader #3

Leader #4

Leader #5

American public mutual

Canadian public mutual

Canadian privately held

Early/New BU Bureaucracy Multidivisional Democratic Individual auto insurance

Mature BU Bureaucracy Multidivisional Democratic

Market diversification to new state Legal: BlileyÕ & SarbOx Acts, NAIC/NCSL

Core competency maintenance Legal (same as left) & competition CIO Project Manager Democratic, committed, intense, male, 30Õs, ISTJ, sensor Technology (software) conversion to improve rules handling & setup, policy admin.

Mature Mechanistic Functional Democratic Group health insurance Product development

Growth Adhocracy Product Adhocratic Group life insurance Core competency maintenance Competition (time to market)

International International (European, (European, North & South North & South American) American) Consortium Consortium Development (some partners) Adhocracy Adhocracy Matrix Matrix Democratic Democratic Infrastructure construction

VP BU Manager Participative, extrovert, female, 40Õs, ESFP, feeler Process and automation improvements to inbound/ outbound call center for expansion.

Lack of skilled IT resources at local prices Senior Manager Manager Laissiez-faire, religious, reserved, male, 50s, INTJ, intuitor. Technology (software) development: online quote & claims system, policy binder, MIB reports

Manager Project Director Autocratic, methodical, female, 30Õs, ISTJ, thinker Technology (software) development: report writing for actuarial data mining, policy binder

Leader #6

Project cost, time, quality.

Improved technology

Quality/scope: performance penalties Senior Manager Project Manager Autocratic, creative, male, 30s, INTP, thinker. Process improvements to maintain schedule and increase quality (avoid penalty)

Distributed field staff (lack of systems) Manager Project Director Autocratic, assertive, female, 30Õs, ISFJ, sensor Database analysis to capture, analyze, find nonconformances.

ARTICLE IN PRESS

Case A #C0001

K.D. Strang / Computers in Human Behavior xxx (2004) xxx–xxx

Context variables by project, leader

ARTICLE IN PRESS K.D. Strang / Computers in Human Behavior xxx (2004) xxx–xxx

23

Exhibit 12 Sample of observed leadership behaviors across project cases Count of observed high leadership behaviors by category across projects (30 work-day period, concurrent for leaders in same case) Task (1) Organizes work activities to improve effectiveness. (2) Develops milestones and action plans for a project. (3) Determines what resources are needed to do a project. (4) Clarifies role expectations for project members. (5) Clarifies quality standards for task performance. (6) Facilitates collection and dissemination of information. (7) Actively monitors operations and performance. (8) Resolves immediate questions or problems from team. (9) Actively listens attentively to a personÕs concerns. (10) Actively provides support and encouragement. (11) Socializes with team beyond work to build relationships. (12) Publicly recognizes contributions and accomplishments. Relations focus (13) Provides individual role and/or behavior coaching. (14) Consults with members on decisions affecting them. (15) Helps team members (as a group) resolve conflicts. (16) Keeps team (group) informed about upcoming actions. (17) Leads by example and models exemplary behavior. (18) Encourages viewing things from multiple perspectives. (19) Expresses confidence team can attain objectives.

Case A leader #1

Case B leader #2

Case C Leader #3

Case D leader #4

Case D leader #5

Case D leader #6

4

19

3

10

5

9

2

3

2

7

2

4

5

1

5

6

3

4

2

1

0

6

1

2

1

2

0

5

2

1

3

1

1

6

0

0

2

6

4

7

0

4

4

1

0

4

1

1

29

7

1

2

0

0

2

4

0

0

0

0

4

6

0

1

1

0

2

1

0

0

0

0

2

3

0

10

0

0

6

2

0

1

1

0

2

2

0

2

1

0

4

4

3

2

2

1

3

3

1

1

3

2

2

1

2

0

0

0

3

1

0

1

1

0

(continued on next page)

ARTICLE IN PRESS 24

K.D. Strang / Computers in Human Behavior xxx (2004) xxx–xxx

Exhibit 12 (continued) Count of observed high leadership behaviors by category across projects (30 work-day period, concurrent for leaders in same case)

Case A leader #1

Case B leader #2

Case C Leader #3

Case D leader #4

Case D leader #5

Case D leader #6

(20) Creates sense of urgency, promotes change. (21) Studies other projects to get ideas for improvements. (22) Envisions exciting new possibilities for the organization. (23) Develops strategies linked to organizationÕs vision. (24) Builds coalition of stakeholders to get change approved.

0

1

0

3

1

3

2

1

2

2

1

0

3

0

1

0

1

0

2

0

1

0

1

0

3

1

1

1

1

0

2

1

0

1

0

0

1

0

1

0

1

0

2

0

1

0

1

0

4

2

0

0

1

0

0

2

0

1

0

0

2

1

2

0

2

0

Change emphasis (25) Creates task force to guide implementation of change. (26) Suggests symbolic changes that affect the work. (27) Empowers members to implement new strategies. (28) Announces and celebrates progress supporting changes. (29) Encourages/facilitates learning by team members. (30) Experiments with new approaches.

Exhibit 13. Observed meta-category project leader practices.

ARTICLE IN PRESS K.D. Strang / Computers in Human Behavior xxx (2004) xxx–xxx

25

Exhibit 14 Observed leader behavior descriptive statistics Behavior

Leaders

Sum

Average

Variance

Quantitative Analysis

Task

6

162

27

191.6

Relations

6

99

16.5

386.3

Change

6

84

14

Most frequent behavior observed for all leaders, moderate variance Significant relations behavior variance between leaders but low mean Low ÔchangeÕ behavior overall, not much variance between leaders

65.2

3.3.5. Summary of leadership trait and behavior typologies These leadership trait and behavior taxonomies, like others in the literature, are not necessarily exhaustive of all possible leader behaviors, since the studies were focused on managerial level individuals, and thus they will not typically include selfmotivation, team dynamics, organizational context, or other macro-level or project domain perspectives. Also these leader behavior classification schemes are descriptive in nature, not predictive or prescriptive of what a leader should necessarily do in a situation. Nevertheless, taxonomies are useful for case study documentation, as well as comparative and exploratory purposes. YuklÕs 30 Leadership Behaviors Taxonomy is an excellent typology but QuinnÕs Leadership Roles Model may possibly be more mature and tested to the extent that it has been successfully used to explore leadership and organizational culture (Buenger, Daft, Conlon, & Austin, 1996; Denison et al., 1995; Egri & Herman, 2000; Quinn, 1988; Quinn et al., 1996). Both Yukl and Quinn emphasize that these models are convenient frameworks for understanding leadership behavior tendencies with the general idea that effective leaders need to have skills in all the roles, and as mentioned earlier, perform several of these behaviors concurrently, not just one at a time. In fact, the practical principle suggested by the Leadership Roles Model research is that a leader needs to perform all of these roles simultaneously (to various degrees), which also corroborates earlier literature (Denison et al., 1995, p. 528).

4. Results and discussions Previous sections described several alternative methodologies and taxonomies for analyzing the proposition that effective project leaders cognitively balance and integrate traits, skills and behaviors according to the situation, taking into account themselves, the environmental context, the group/team, the sponsors, the stakeholders, and of course the priority of each in this circumstance. A great deal of research was completed but only relevant analysis is included in this paper. All of these taxonomies discussed herein were used to examine the same leadership behaviors, traits, and skills in the case studies for each leader, and they furnished interesting results; however, due to space constraints only the Leadership Behaviors Taxonomy (which provided the most revealing and comprehensive synthesis in this scenario) will be documented and discussed here.

26

Sum of squares

Behavior categories

571

Leader traits

Degrees of freedom

Mean squares

Critical F value (a 0.01)

P value

Computed F statistic

Quantitative analysis

2

285.5

7.559493

0.231271

1.701092

1537.167

5

307.4333

5.636366

0.19421

1.831778

Accept H0, means same, leader traits not explained by leadership behavior. Accept H0, means same, leadership behavior not explained by leader traits.

Unexplained correlation

1678.333

10

167.8333

Note. The above is not the hypothesized result of this research (H0 is merely the statistical formula constraint), since the belief is that behavior is attributed to leader traits. Additionally, a large portion (1678 which is > 50%) is unexplained variance.

Total

3786.5

17

ARTICLE IN PRESS

Factor

K.D. Strang / Computers in Human Behavior xxx (2004) xxx–xxx

Exhibit 15 Observed leader behavior analysis of variance by meta-categories

Factor

Sum of squares

Degrees of freedom

Mean squares

Critical F value (a 0.01)

P value

548.5833

29

18.91667

1.844398

2.67E

Leader traits

153.7167

5

30.74333

3.145971

0.00067

Unexplained correlation

975.45

145

1677.75

179

Total

6.727241

Note. This finding supports the proposition that leadership behavior is attributed to leader personality traits, and vice versa, leader traits are explained through behavior. A much larger portion of correlation is now explained by using detailed observations.

05

Quantitative analysis

2.81195

Reject H0, means different, leader traits variation is explained by behavior Reject H0, means different, behavior variation is attributed to leader traits

4.569976

ARTICLE IN PRESS

Behaviors (detailed)

Computed F statistic

K.D. Strang / Computers in Human Behavior xxx (2004) xxx–xxx

Exhibit 16 Observed leader behavior analysis of variance by detailed factors

27

ARTICLE IN PRESS 28

K.D. Strang / Computers in Human Behavior xxx (2004) xxx–xxx

Exhibit 17 Descriptive statistical summary of leader traits correlated to behaviors Measure Mean Standard error Standard deviation Sample variance Kurtosis Skewness Confidence Level (0.95)

Leader 1

Leader 2

Leader 3

Leader 4

Leader 5

Leader 6

3.433333333 0.914359688

2.566666667 0.656853057

1.033333333 0.242037963

2.333333333 0.499041226

1.1 0.205387215

1.033333333 0.366666667

5.00815427

3.597732364

1.32569652

2.733361368

1.124952106

2.008316044

25.0816092

12.94367816

1.757471264

7.471264368

1.265517241

4.033333333

25.50361217 4.878060033 1.870076559

15.36046504 3.568140106 1.343416076

1.804675462 1.457333302 0.495023486

0.687469146 1.252332371 1.020654466

3.899142375 1.659860465 0.420064249

7.875343652 2.605763314 0.749917945

As stated earlier, the case studies were chosen based on theory sampling methodology and are shown in Exhibit 11, with key points summarized (rather than taking space here to describe them in detail narrative form). The construct for the organizational leadership variables used in the table (leftmost column), is similar in concept to that used by other management science researchers, and in this case the taxonomy framework was derived from other research (Strang, 2004b). The table in Exhibit 12 shows a sample of ‘‘high’’ leader behavior observations. The chart in Exhibit 13 displays a graphical view of the project leader behavior frequency observation counts grouped by the three meta-leadership categories (plus the grand total of all three). This emphasizes the high degree of task and relations behaviors by all leaders, the relatively low use of change oriented practices, as well as the individual differences between these cases. The exception to this was leader # 1 who displayed a much more balanced number of task, relations, and change-oriented behavior, and in particular, emphasized her team relations skills. In fact leader # 1 was the only individual of these research subjects noted for utilizing more relations behavior (in total) as compared with task and change management. Relations and change behaviors were infrequent for leaders # 3, # 4, # 5, and # 6. 4.1. Analytical methods results The two-way upper-tailed analysis of variance (ANOVA-2) test, without replication, is included in this paper as a statistical test to compare the relationship of leader traits and leadership behaviors. In this situation the ANOVA-2 statistical test is analogous to comparing means across samples of populations, which is similar in methodology but exactly opposite in hypothesis testing, to the v2 upper-tailed contingency distribution to evaluate two qualitative variables for independence. In a two-way upper-tailed ANOVA-2 test, if the computed value is greater than the F statistic (at the chosen significance level that in this case was alpha 0.01), it indicates there is a dependency between the treatment, blocking factor, response variables, in essence the variation is explained by both the treatment and blocking factor variables.

Exhibit 18 Descriptive statistics of leadership behaviors correlated to leader traits sorted by Kurtosis Standard Errora

Standard deviationa

Sample variancea

(26) Suggests symbolic changes that affect the work. (18) Encourages viewing things from multiple perspectives. (17) Leads by example and models exemplary behavior. (15) Helps team members (as a group) resolve conflicts. (30) Experiments with new approaches. (20) Creates sense of urgency, promotes change. (8) Resolves immediate questions or problems from team. (16) Keeps team (group) informed about upcoming actions. (13) Provides individual role and/or behavior coaching. (7) Actively monitors operations and performance. (21) Studies other projects to get ideas for improvements. (25) Creates task force to guide implementation of change. (23) Develops strategies linked to organizationÕs vision. (27) Empowers members to implement new strategies. (11) Socializes with team beyond work to build relationships. (3) Determines what resources are needed to do a project. (28) Announces and celebrates progress supporting changes. (12) Publicly recognizes contributions and accomplishments. (29) Encourages/facilitates learning by team members. (10) Actively provides support and encouragement. (6) Facilitates collection and dissemination of information. (1) Organizes work activities to improve effectiveness. (19) Expresses confidence team can attain objectives. (22) Envisions exciting new possibilities for the organization. (2) Develops milestones and action plans for a project. (5) Clarifies quality standards for task performance. (24) Builds coalition of stakeholders to get change approved. (4) Clarifies role expectations for project members. (14) Consults with members on decisions affecting them. (9) Actively listens attentively to a personÕs concerns.

0.500 0.833 2.167 1.167 1.167 1.333 1.833 2.667 1.000 3.833 1.333 0.667 0.667 0.667 2.000 4.000 1.167 0.500 0.500 1.000 1.833 8.333 1.000 0.833 3.333 1.833 1.167 2.000 1.667 6.500

0.224 0.401 0.401 0.401 0.401 0.558 0.703 0.494 0.516 1.046 0.333 0.333 0.333 0.333 1.000 0.730 0.654 0.342 0.342 0.683 0.946 2.418 0.447 0.477 0.803 0.703 0.401 0.856 0.919 4.624

0.548 0.983 0.983 0.983 0.983 1.366 1.722 1.211 1.265 2.563 0.816 0.816 0.816 0.816 2.449 1.789 1.602 0.837 0.837 1.673 2.317 5.922 1.095 1.169 1.966 1.722 0.983 2.098 2.251 11.327

0.300 0.967 0.967 0.967 0.967 1.867 2.967 1.467 1.600 6.567 0.667 0.667 0.667 0.667 6.000 3.200 2.567 0.700 0.700 2.800 5.367 35.067 1.200 1.367 3.867 2.967 0.967 4.400 5.067 128.300

Statistics calculated to E

3.333 2.390 2.390 2.390 2.390 1.875 1.731 1.550 0.781 0.571 0.300 0.300 0.300 0.300 0.300 0.586 1.240 1.429 1.429 1.429 1.693 1.855 2.500 2.552 2.666 2.723 3.603 3.657 3.797 4.857

09 level of accuracy and then rounded to three decimal points for display purposes.

Skewnessa 0.000 0.456 0.456 0.456 0.456 0.523 0.731 0.075 0.889 0.366 0.857 0.857 0.857 0.857 1.102 0.943 1.354 1.537 1.537 1.537 1.459 1.370 1.369 1.586 1.675 1.435 1.438 1.755 1.882 2.185

Confidence level (0.95)a 0.575 1.032 1.032 1.032 1.032 1.434 1.808 1.271 1.327 2.689 0.857 0.857 0.857 0.857 2.571 1.877 1.681 0.878 0.878 1.756 2.431 6.214 1.150 1.227 2.064 1.808 1.032 2.201 2.362 11.887

29

a

Kurtosisa (sort column)

ARTICLE IN PRESS

Meana

K.D. Strang / Computers in Human Behavior xxx (2004) xxx–xxx

Behavior

ARTICLE IN PRESS 30

K.D. Strang / Computers in Human Behavior xxx (2004) xxx–xxx

This type of analysis is related to the concept underlying the Within-And-BetweenAnalysis (Dansereau et al., 1984; Yammarino & Bass, 1990) being used in contemporary leader behavior studies, and it is suitable for the statistical analytic comparison desired in this research. As mentioned earlier, the case study analysis is supplemented with quantitative statistical analysis for triangulation of data and method, to strengthen internal validity and reliability. The goal is to determine if these qualitative observations show how effective leadership behavior varies across leader personalities. The purpose is to determine if these qualitative observations show leadership behavior is mostly explained/impacted by the leader personalities. An additional objective of the methodology is to supplement the qualitative case analysis with quantitative analysis at summary and detailed levels, by challenging the ANOVA-2 null hypothesis to show that there is variation in leadership behaviors explained by different leader dominant personality traits. 4.2. Quantitative case study analysis The following three exhibits show summary level descriptive statistics and ANOVA-2 results. The first table Exhibit 7 reveals that for all six leaders, there were more task-focused behaviors observed (162) at a high level, and the average was almost double (27), and with moderate variance (meaning most leaders performed similar behavior amounts in this category), as compared with the other two meta-categories. As explained earlier, after all ratings of leader behavior were recorded, it was decided not to include observations less than a definite ‘‘high level’’ in this analysis, to increase precision, enhance statistical credibility, and to reduce author/observer perceptional bias errors (see Exhibit 14). An interesting and unexpected observation was the significant relations behavior variance between the leaders (386.3), and combined with the moderate average (16.5) this might be interpreted to signify that several leaders performed low amounts of consideration for team members, while others performed very high amount, which does concur with the actual self-reflected experience. The low ÔchangeÕ behavior overall, and little variance between leaders, indicates as expected, that these behaviors were not commonly observed, either during the recording period, nor for the duration of these projects (again, self-reflected). The reason for low change behaviors is self-reflected to be resulting from a stable organizational environment and the nature of the projects did not involve significant process or role changes. This should not be interpreted to indicate this is a benchmark of leadership or projects in general, it is simply the case here. The table in Exhibit 15 illustrates summary behavior meta-category findings, which emphasizes the importance of decomposing factor analysis variables to their lowest possible and practical level. The reason for the statistical lack of dependency correlation among leader traits and behavior categories is that summary level metadata (low number of observation demarcations/classes) will not be sufficient for statistical significance. Therefore, these results were considered misleading because the nature of v2 (not shown here) and ANOVA-2 are numerically and probability driven

ARTICLE IN PRESS K.D. Strang / Computers in Human Behavior xxx (2004) xxx–xxx

31

therefore larger amounts of observation and sample strata provide more accurate correlation analysis. The next table (Exhibit 16) supports the research proposition by illustrating the ANOVA-2 results using detailed task, relations, and change behaviors (30 classes) correlated with leader personality traits (skills, abilities, motives), and vice versa. The statistical findings are credible by standard measures of a low ‘‘P’’ value (lower than 0.001). Essentially this could be interpreted as the traits of the six leaders have a variation which is explained by their detailed behaviors. Also, since the statistical means are different (as measured by the formulas), the detailed behavior variations are attributed to differences in leader traits. This corroborates the assertion above, in particular, that different leaders acted with significantly different amounts of relations behavior (high variance between the leaders). On a related note, the ANOVA-2 of detailed behaviors shows that a much larger portion of correlation is now explained by using detailed observations as compared with the three meta-categories. Also what is not observable in the statistical tabulations and analysis is the direct-participant-observed fact that certain leaders combined these behaviors more smoothly and somewhat more effectively than others, particularly task with change, and also relations with change, thus making it difficult in some cases to differentiate between the category applied, even at the detail level. For example, Leader 1 spent a lot of time communicating with people on the team both in the work context and at other times. She actively listened attentively, spoke of visions, and encouraged members to try new things, etc., so it was difficult to determine if this was a high observation of relations or change oriented activity. Other leaders seemed to do many of the behaviors superficially in all categories, but it was difficult to consider these behaviors moderate or high so they were not recorded in this analysis. The next table (Exhibit 17) shows several descriptive statistical measures of the leader traits correlated to the detailed behaviors. Of particular interest here at the descriptive level are the Kurtosis and standard deviation. The Kurtosis (K) factor was one of a number of statistical distribution measures taken on the comparison of leader traits with behaviors. A positive high K factor indicates a relatively peaked as compared to normal distribution, while a negative K suggests a relatively flat distribution. Leader 1 and Leader 2 show a high distribution peak when their traits are tested against the behaviors, with similar higher values for their standard deviations, statistical confidence levels, and skewness. The other leaders all exhibited positive K values but they were typically an order of magnitude lower. This is shown in the next exhibit with items of significant interest highlighted. The next exhibit (Exhibit 18) lists similar descriptive statistical measures for the leadership behaviors across the leaders (at the 0.95 confidence level), but this time it is ordered by K factor, then by the standard deviation. The ‘‘Behavior’’ column correlates back to the earlier Exhibit 12: Sample of Observed Leadership Behaviors Across Project Cases, which describes the specific behavior observed, and their metacategories. This analysis illustrates certain leadership behaviors that were fairly low

ARTICLE IN PRESS 32

K.D. Strang / Computers in Human Behavior xxx (2004) xxx–xxx

in distribution peaks (top of table), and those that were highly peaked (bottom portion of the exhibit). Relations behaviors (# 9–16) from Exhibit 18 (except # 15 ‘‘Helps team members. . .’’) tended to have rather moderate K factor observation distributions, while several of the change orientations were negative, specifically # 26 ‘‘Suggests symbolic changes that affect the work’’ ( 3.33), # 18 ‘‘Encourages viewing things from multiple perspectives’’ ( 2.39), # 17 ‘‘Leads by example and models exemplary behavior’’ ( 2.39), # 15 ‘‘Helps team members (as a group) resolve conflicts’’ ( 2.39), and # 30 ‘‘Experiments with new approaches’’ ( 2.39). At the other end of the contingency spectrum, certain leader behaviors produced high K factors, specifically, # 19 ‘‘Expresses confidence team can attain objectives’’ (2.5), # 22 ‘‘Envisions exciting new possibilities for the organization’’ Ô(2.55), # 2 ‘‘Develops milestones and action plans for a project’’, (2.67), # 15 ‘‘Clarifies quality standards for task performance (2.72), # 24 ‘‘Builds coalition of stakeholders to get change approved’’, (3.6), # 4 ‘‘Clarifies role expectations for project members’’ (3.66), # 14 ‘‘Consults with members on decisions affecting them’’ (3.8), and # 19 ‘‘Actively listens attentively to a personÕs concerns’’ (4.86). It was interesting to observe that the last item, #9, an important team relations leadership behavior, had both a high K value, and a high standard deviation (11.33) correlated to leader traits (across leaders). Overall, this contingency distribution analysis could be interpreted to mean that project leaders achieved outcomes using different combinations of specific leader behaviors, with each leader having different dominant personality traits, yet all accomplished relative successful deliverable production. Some of the correlation was statistically explained, but there was a noted difference in team satisfaction, and motivation, which was not captured in the observed behaviors or project outcomes, whereby the higher satisfied team members were those under leaders 1 and 2, yet some of the leaders did not exhibit trait personalities leaning toward supporting relationship oriented behaviors. For example, leader # 2 was less extroverted and less feeling, than leader # 1, yet both exhibited positive behaviors (but not necessarily correlated/explained by the observed traits). This suggests that an effective project leader can assess the circumstances, consciously select a behavior or behaviors to apply in the situation (individual or group), and perform positive behaviors, without having a natural personality trait pattern/propensity to behave in this way. Therefore, effective project leadership skills and behaviors could be learned and applied as a cognitive mental model. Other studies in the literature concluded that variance in leadership behavior was due to perceptual interpretations at the individual level (which may not reflect actual or intended behavior), and leader behavior differs for different subordinates (Avolio, Yammarino, & Bass, 1991; Yammarino & Bass, 1990), which implies that the leader manages their traits, skills, and personality, to generate appropriate actions according to the situation, subordinate(s) and the context. Multicultural and multi-country leadership studies confirm that project leadership success (from the stakeholder perspective) is evaluated and rated by various factors at the individual stakeholder mental models or normative group shared normative vision (Diallo & Thuillier, 2004).

ARTICLE IN PRESS K.D. Strang / Computers in Human Behavior xxx (2004) xxx–xxx

33

4.3. Qualitative case study analysis The exhibit shows the qualitative assessment of leadership behavior by categorizing each leader into a dominant role set based on the Leadership Roles Model discussed earlier. When this quantitative data above is considered with the qualitative case reflections, it is obvious that the behaviors and traits of the first two leaders had something to do with the positive effects on the project outcomes and team satisfaction. These two projects were judged to be more successful in terms of team satisfaction, cohesive performance than the other leader projects. However, all leaders achieved effective on time production of deliverables, but, for the other leaders, with lower relations behaviors, their success in task outcomes was at the cost of less satisfying relations, a bit more overtime, and less enthusiasm towards their next project. Change was not a significant factor in these projects. Exhibit 19 shows the qualitative results of the leader role set analysis (details not included in this paper), thus showing how quantitative and qualitative measures corroborate leader traits and behaviors. The most difficult leader to categorize was leader # 3, a creative intuitive type, but after considerable self-reflection, and peer opinion, it became obvious that this leader chose to play a monitor behavior role despite the fact that he was an innovator in other project contexts. Transformational (change oriented) research suggests that leader and follower motive patterns affect relationships between leadership and

Leader#1

Monitor

In te St ab rna lP ilit y Le roc es ad s er sh ip

R Ta atio sk na lG Le oa ad l er sh ip

Internal Focus

Coordinator

Leader#3

Innovator Ad Op ap en Sy tiv e st e Le ad ms er sh ip

Stability

Facilitator

Leader#5

Hu Pe m a n op Re le Le sou ad r er ces sh ip

Mentor

Flexibility

Leader#2

Broker

External Focus

Producer

Director

Exhibit 19. Leadership roles model analysis.

Leaders #4 + #6

ARTICLE IN PRESS 34

K.D. Strang / Computers in Human Behavior xxx (2004) xxx–xxx

outcomes, meaning that traits and behaviors are explained by more than individual perceptual phenomenon (Wofford, Whittington, & Goodwin, 2001); therefore, motivation may be the source of some variation between the leaders and meta-categories, which is ratified by the author as a participant self-reflection. What is also interesting about this analysis is the leader trait/personality does not necessarily predict the role model that will be used by a leader in a particular project context. A good case in point of this is that leader # 5 tended to have personality traits favoring a director or coordinator role, and this was mentioned by other team members. For example, leader # 5 behaved as an innovator in this project by inspiring and encouraging others toward required quality process changes, yet in other projects he was more focused on governance and process (director role). Much the same could be said about leader # 3, who played a monitor role in the observed context, while in other situations he commonly played an innovator role. Other leaders were also observed to play different roles beyond the project context. Thus the inductive verdict is that leaders play multiple roles in different contexts, which can also differ (slightly or significantly) depending on the time, the situation, and the context itself. These qualitative assessments partly defend similar concepts from other quantitative research that individual, dyadic, and team-based leadership assessments explain variability in the responses as differences in rater perceptions, and that leadership behavior is very much individually (trait) driven and not explained at the dyad or group level of analysis (Avolio & Yammarino, 1990; Avolio et al., 1991; Bass & Avolio, 1994; Yammarino & Bass, 1990)s. However other contextual elements will most certainly make it difficult for any specific causality variable to be isolated and explained.

5. Conclusions This research has shown quantitative and qualitative support for the proposition that effective project leaders know their own behavior, traits, and skills, they cognitively balance these traits and behaviors according to the situation, taking into account themselves, the organizational context, the group/team, the stakeholders, and of course the priority of each of these elements in the circumstance. The leaders in the case studies demonstrated traits, skills, roles, and behaviors affirming empirical management science studies, and their conduct differed from typical organizational level and project management functions. The leadership behaviors usually conformed to typologies such as the Managerial Roles, the Leadership Roles Model, the Competing Values Framework, as well as the Meta-Category Leadership Taxonomy. This concurs in part with the hypothesis that project leaders adopt more managerial roles (directing, monitoring, etc) than leadership roles (facilitating, mentoring, etc.), in that these leaders practiced different roles in several role sets, but the interesting finding was that several leaders exhibited multiple role sets concurrently.

ARTICLE IN PRESS K.D. Strang / Computers in Human Behavior xxx (2004) xxx–xxx

35

Furthermore, the leaders exhibited some traits found in alternative theories research such as Leadership Virtual Reality, Level 5 Leader Trait Hierarchy, and charisma leadership. Thus, unfortunately the first hypothesis is only partially proven here, being that organizational leadership principles are applied and observable in the project domain, but a few project managers did exhibit emergent psychoanalytic leadership trait and skill principles, primarily emotionality maturity and self-monitoring techniques. The theoretically sampled case project leaders achieved outcomes with high variances of leader behaviors within the three meta categories of task orientation, relations focus, and change emphasis. The interesting finding was all leaders exhibited different behaviors in different combinations with their dominant personality traits (some of the correlation was explained as noted earlier), yet all accomplished deliverable production. Direct observation of all cases and leaders by the author definitely concurs with this finding that leader traits and their behaviors applied to project situations is a conscious, cognitively complex, and contingency procedure, which is different for each situation faced, and not necessarily a pattern directly associated with their personality traits (although some leaders tended to be more predictable than others in terms of their behaviors for many types of project situations). The main outcome variable for leaders having lower relations behaviors, which statistically showed some dependency/attribution to personality traits, was less team satisfaction, and lower future motivation to continue to perform, whereby the higher satisfied team members were those with leaders exhibiting higher relations behaviors, but not necessarily having extroverted, sensing, or feeling personality traits. The research conclusion is that an effective project leader can assess the circumstances, consciously select skill(s) to apply in the situation, and act out beneficial project leadership behaviors, without necessarily having a contextually-advantageous personality propensity to behave in a supportive way, which repudiates the hypothesis that effective project leaders need a personal competency and personality combination that is conducive and complimentary to the dynamic project management role. This research disagrees with some aspects of contemporary management science theory in that project managers do not require an emphasis on relations behaviors and change oriented traits, supported by extrovert and affective-dominate personality profiles. This assertion is supported by psychology theory that argues certain personality traits and skills associated with leadership style are intrinsic to each person; therefore, the other capabilities must be developed (Jung, 1971; Collins, 2001). Therefore, inductively, it is propositioned that effective project leadership behaviors can be learned and applied. In the same manner, leadership skills, behaviors, and processes could be proximally learned and improved at any level in the organization or community. Both of these propositions are posited contingent on the leader having at least a moderate level of leadership knowledge, cognitive functioning, emotional stability, and contextual skills. These findings, albeit on a small statistical scale of significance, are a catalyst for continued and broader global (cross-industry, inter-cultural) research, from multiple stakeholder perspectiveÕs of leadership theories using case analysis methods and hypotheses replication with these constructs and alternative taxonomies. As well,

ARTICLE IN PRESS 36

K.D. Strang / Computers in Human Behavior xxx (2004) xxx–xxx

it is recommend to supplement this research with empirical studies at the dyadic and organizational levels of analysis (in addition to individual and group levels used here).

Acknowledgments Many thanks to Dr. Jerry Hunt (Director, Institute for Leadership Research, Management Ph.D. Coordinator at Rawls College of Business in Lubbock TX) and Dr. Dinese Rousseau (Carnegie Mellon University at Pittsburg, PA) for reviewing several of my papers and providing constructive feedback. I also thank Dr. Alison M. Konrad (Richard Ivey School of Business at London, Ontario) for reviewing several research papers in this series.

References Adair, J. (1997, March). Developing leadership skills. In Paper presented at the Human Resource Development Week. Avolio, B. J., Bass, B. M., & Jung, D. I. (1999). Re-examining the components of transformational and transactional leadership using the multifactor leadership questionnaire. Journal of Occupational and Organizational Psychology, 72(4), 441–462. Avolio, B. J., & Yammarino, F. J. (1990). Operationalizing charismatic leadership using a level of analysis framework. Leadership Quarterly, 1, 193–208. Avolio, B. J., Yammarino, F. J., & Bass, B. M. (1991). Identifying common methods variance with data collected from a single source: An unresolved sticky issue. Journal of Management, 17, 571–587. Bass, B. M. (1985). Leadership and performance beyond expectations. New York, USA: Free Press. Bass, B. M., & Avolio, B. J. (1994). Improving organisational effectiveness through transformational leadership. London: Sage. Bass, B. M., & Stogdill, R. M. (1990). Bass & StogdillÕs handbook of leadership: Theory, research, and managerial applications (3rd ed.). New York: Free Press. Bennis, W. (1989). On becomming a leader. San Francisco: Peresus Publishing. Bennis, W., & Nanus, B. (1997). Leaders strategies for taking charge (3rd ed.). New York: Harper Business. Boyatzis, R. E. (1982). The competent manager: A model for effective performance. New York, NY: Wiley. Bradford, D. L., & Cohen, A. R. (1984). Managing for excellence: The guide to developing high performance organizations. New York, NY: Wiley. Buenger, V., Daft, R. L., Conlon, E. J., & Austin, J. (1996). Competing values in organizations: contextual influences and structural consequences. Organization Science, 7(5), 557–576. Charon, L. d. (2003). A transformational leadership development program: Jungian psychological types in dynamic flux. Organization Development Journal, 21(3), 9–21. Collins, J. (2001). Level 5 leadership: The triumph of humility and fierce resolve. Harvard Business Review, 79(1), 66–76. Dansereau, F., Alutto, J. A., & Yammarino, F. J. (1984). Theory testing in organizational behavior: The variant approach. Englewood Cliffs, NJ: Prentice-Hall. Denison, D. R., Hooijberg, R., & Quinn, R. E. (1995). Paradox and performance: toward a theory of behavioral complexity in managerial leadership. Organizational Science, 6(5), 524–540. Diallo, A., & Thuillier, D. (2004). The success dimensions of international development projects: The perceptions of african project coordinators. International Journals of Project Management, 22(1), 19–31.

ARTICLE IN PRESS K.D. Strang / Computers in Human Behavior xxx (2004) xxx–xxx

37

Dubin, R. (1976). Theory building in applied areas. In M. D. Dunnette (Ed.), Handbook of industrial and organizational psychology. Chicago, IL: Rand McNally. Duffy, J. (2001). Maturity models: blueprints for evolution. Strategy & Leadership, 29(6), 19. Egri, C. P., & Herman, S. (2000). Leadership in the north american environmental sector: Values, leadership styles, and contexts of environmental leaders and their organizations. Academy of Management Journal, 43(4), 571–605. Eisenhardt, K. M. (1989). Building theory from case study research. Acade my of Management Review, 14, 488–511. Fairholm, G. W. (1998). Leadership as an exercise in virtual reality. Learning & Organizational Development Journal, 19(4), 187–193. Goleman, D. (1998). What makes a leader? Harvard Business Review, 76(6), 92–102. Goleman, D. (2000). Leadership that gets results. Harvard Business Review, 78(2), 78–90. Graen, G. B., & Uhl-Bien, M. (1991). Relationship-based approach to leadership: development of leadermember exchange theory of leadership over 25 years: applying a multi-level multi-domain approach. Leadership Quarterly, 6, 219–247. Hellriegel, D., Slocum, J. W., & Woodman, R. W. (1992). Organizational behavior (6th ed.). St. Paul, MN: West Publishing Company. Hemphill, J. K. (1959). Job descriptions for executives. Harvard Business Review, 37, 55–67. House, R. J. (1977). In J. G. a. L. Hunt (Ed.), Leadership: The cutting edge. Carbondale, IL: Southern Illinois University Press. Jacobs, T. O., & Jaques, E. (1987). Leadership in complex systems. In J. Zeidner (Ed.), Human productivity enhancement: Organizations, personnel, and decision making. New York, USA: Praeger. Johnson, W. L., Mauzey, E., Johnson, A. M., Murphy, S. D., & Zimmerman, K. J. (2001). A higher order analysis of the factor structure of the Myers–Briggs type indicator. Measurement and Evaluation in Counseling and Development, 34, 96–116. Jung, C. G. (1971). Psychological types. In R. Hull (Ed.), The basic writings of Jung. Princeton, NJ: Princeton University Press. Katz, D., & Kahn, R. L. (1978). The social psychology of organizations. New York: Wiley. Koontz, H., OÕDonnell, C., & Weihrich, H. (1980). Management (7th ed.). New York: McGraw-Hill. Kotter, J. P. (1990). What leaders really do. Harvard Business Review, 68(3). Kotter, J. P. (1998). What leaders really do. In H. B. Review (Ed.), Harvard business review on leadership (pp. 37–60). Boston, MA: Harvard Business School Publishing. Lombardo, M. M., & McCall, M. W. J. (1978). Leadership. In M. W. J. McCall & M. M. Lombardo (Eds.), Leadership: Where else can we go?. Durham, NC: Duke University Press. Mahoney, T. A., Jerdee, T. H., & Carroll, S. J. (1965). The jobs of management. Industrial Relations Journal, 4, 97–110. McCall, M. W. J., Lombardo, M. M., & Morrison, A. (1988). Lessons of experience. Lexington, MA: Lexington Books. Michael, J. (2003). Using the Myers–Briggs type indicator as a tool for leadership development. Apply with caution. Journal of Leadership & Organizational Studies, 10(1), 68. Miles, M. B., & Huberman, A. M. (1984). Analyzing qualitative data: A source book for new methods. Beverly Hills, CA: Sage. Mintzberg, H. (1975). The managerÕs job – folklore and fact. Harvard Business Review, 53(4). Mintzberg, H. (1998). The managerÕs job – folklore and fact. In H. B. Review (Ed.), Harvard business review on leadership (pp. 1–36). Boston, MA: Harvard Business School Publishing. Mumford, M. D., & Connelly, M. S. (1991). Leaders as creators: leader performance and problem solving in ill-defined domain. Leadership Quarterly, 2(4), 289–315. Myers, I. B., & McCaulley, M. H. (1989). A guide to the development and use of the Myers–Briggs type indicator. Palo Alto, CA: Consulting Psychologists Press. Page, R. & Tornow, W. W. (1987). Managerial job analysis: Are we farther along? In Paper presented at the society for industrial and organizational psychology, Atlanta, GA, USA. Parry, S. B. (1993). Just what is a competency and why should you care?. Training(June), 58–64.

ARTICLE IN PRESS 38

K.D. Strang / Computers in Human Behavior xxx (2004) xxx–xxx

PMI (2002). Project manager competency development framework. Newton Square, PA: Project Management Institute. PMI (2003). Organizational project management maturity model. Newton Square, PA: Project Management Institute. Quinn, R. E. (1988). Beyond rational management: Mastering the paradoxes and competing demands of high performance. San Francisco, CA: Jossey-Bass. Quinn, R. E., Faerman, S. R., Thompson, M. P., & McGrath, M. R. (1996). Becoming a master manager: A competency framework. New York, USA: Wiley. Quinn, R. E., & Robrbaugh, J. (1983). A spatial model of effectiveness criteria towards a competing values approach to organizational analysis. Management Science, 29(3). Schein, E. H. (1985). Organisational culture and leadership. San Francisco: Jossey-Bass. Schermerhorn Hunt & Osborn (2003). Organizational behavior (8th ed.). New York, NY: Wiley. Schneider, M. (2002). A stakeholder model of organizational leadership. Organizational Science, 13(2), 209–220. Scho¨n, D. A. (1983). The reflective practitioner – How professionals think in action. Aldershot, UK: BasiAshgate ARENA. Spencer, L. M., & Spencer, S. M. (1993). Competence at work: Models for superior performance. New York, NY: Wiley. Stogdill, R. M. (1948). Personal factors associated with leadership. Journal of Psychology, 25, 35–71. Stogdill, R. M. (1974). Handbook of leadership: A survey of the literature. New York, USA: Free Press. Strang, K. D. (2003a, September 23). Achieving organizational learning across projects. In Paper presented at the PMI North America global congress 2003, Baltimore, MD, USA. Strang, K. D. (2003b). Practical explorations in knowledge management maturity (Essay). Melbourne, Australia: RMIT. Strang, K. D. (2004a). Exploring dynamic project leadership processes: A social science perspective. Newton Square, PA: In-press. Strang, K. D. (2004b). Examining effective and ineffective transformational project leadership. Team Performance Management; Emerald Publishing Group; in press. Tett, R. P., Jackson, D. N., & Rothstein, M. (1991). Personality measures as predictors of job performance: a meta-analytic review. Personnel Psychology, 44, 703–742. Tornow, W. W., & Pinto, P. R. (1976). Development of a managerial taxonomy: A system for describing, classifying, and evaluating executive positions. Journal of Applied Social Psychology, 61, 410–418. Verma, V. K., & Wideman, R. M. (1994). Project manager to project leader? And the rocky road between. In Paper presented at the proceedings of the 25th annual PMI seminars/symposium on project management, Upper Darby, PA, USA. Viswesvaran, C., & Deniz, A. O. (2000). Measurement error in big five factors personality assessment: Reliability generalization across studies and measures. Educational and Psychological Measurement, 60, 224–236. Walker, D. H. T. (2000). Client/customer or stakeholder focus. Iso14000 ems as a construction industry case study. TQM, 12(1), 18–25. Walker, D. H. T., Hampson, K. D., & Peters, R. J. (2000). Relationship-based procurement strategies for the 21st century. Canberra, Australia: AusPrint. Wofford, J. C., Whittington, J. L., & Goodwin, V. L. (2001). Follower motive patterns as situational moderators for transformational leadership effectiveness. Journal of Managerial Issues, 13(2), 196–212. Yammarino, F., & Bass, B. M. (1990). Transformational leadership and multiple levels of analysis. Human Relations, 43, 975–995. Yin R. K. (1994). Case study research: Design and methods (2nd ed., Vol. 5). Thousand Oaks, CA: Sage Publications Inc.. Yukl, G. (1971). Toward a behavioral theory of leadership. Yukl, G. (1987). A new taxonomy for integrating diverse perspectives on managerial behavior. In Paper presented at the Leadership Developments, New York, NY, USA. Yukl, G. (1998). Leadership in organizations (4th ed.). Eaglewood Cliffs, NJ: Prentice-Hall.

ARTICLE IN PRESS K.D. Strang / Computers in Human Behavior xxx (2004) xxx–xxx

39

Yukl, G., Gordon, A., & Taber, T. (2002). A hierarchical taxonomy of leadership behavior: integrating a half-century of behavior research. Journal of Leadership & Organizational Studies, 9(1), 15–33. Zaccaro, S. J., Gilbert, J. A., Thor, K. K., & Mumford, M. D. (1991). Leadership and social intelligence: linking social perceptiveness and behavioral flexibility to leader effectiveness. Leadership Quarterly, 2(4), 317–342. Zalesnik, A. (1977). Managers and leaders are they different. Harvard Business Review, 55(3). Dr. Kenneth David Strang is a professional international project manager with 21 years experience in various countries, sectors, and industries, especially insurance, financial, academic, construction, technology, and government. He is also an academic researcher, teaches graduate level at several universities, and designs courses (face-to-face plus internet-based). His academic research interests are: leadership, project management, organizational learning, knowledge management, ethics, organizational behavior, strategy, systemic models, and web-based adult educational psychology. Kenneth completed his Doctoral in Organizational Project Management (honors) at RMIT University in Australia. He has a Master of Business Administration (honors), he holds a Fellow Life Management Institute designation (with distinction), specializing in multi-line Life Insurance Actuary and Pension Planning Systems, he has a Bachelor of Science (honors), and a Business Technology Diploma (honors). Kenneth is certified by PMI as a Project Management Professional. He is also an active board director in several Canadian and international associations, and he often volunteers for non-profit corporations such as United Way, Insync, Learning Network, School District Education Councils, Project Management Institutes, IEEE, Unicef, and others.