Different leadership theories conceptualize the phenomenon through different levels of analysis (Yulk, 2010). As illustrated in Figure 3, there are four hierarchically order levels that can be used to describe the leadership process: individual, dyadic, group and organization. These four levels can be organized in relation to the type of relationship it produces (individual, interpersonal and relational), which can be used to map these hierarchical levels into the classification utilized in Table 2.
The individual level considers intra-individual processes, such as traits, skills, motivation, and values to understand how the individual leader influences the leadership process (Yulk, 2010). The focus here is usually on the behavior, style or the role of the leader. Examples of leadership theories that utilize this level to explain leadership include very old approaches, such as great man theories and trait theories, as well as very recent developments in the field, such as ethical leadership and the skill-based model of leadership (Bass, 1990; Mumford et al, 2000; Northhouse, 2007).
The dyadic level explains leadership through the relationship between the leader and his or her follower (Yulk, 2010). Theories that conceptualize leadership as a dyadic process consider issues concerned with how the leader influences, interacts, motivates, inspires and develops the follower. The focus of these theories might be in the behavior of the leader as the source of motivation, inspiration, influence; or the focus might be in the changes that occur in the follower due to the interaction with the leader. A large proportion of leadership theories analyze leadership through this lens, including authentic leadership, full-range leadership (transformational and transactional leadership), leadership member-exchange theory and servant leadership (Yulk, 2010; Northhouse, 2007).
The group level focuses on the collective processes that are involved in leadership (Yulk, 2010). The theories that study leadership through this lens explore team performance through processes that involve all the subordinates of the leader. Thus, it considers not only the behaviors that the leader utilized to influence a single individual, as theories in the dyadic level, but also behaviors that promote the interaction, commitment and confidence among all members of the team. Theories that conceptualize leadership at this level of analysis include team leadership, social exchange theory and cross-cultural leadership, shared leadership and distributed leadership.
The organizational level of analysis considers the influence that leaders might have throughout the whole organization (Yukl, 2010). Theories that conceptualize through this lens take into account the survival, adaptation, culture and strategy of the organization as pertaining to the influence and direction set up by the leader (or leaders, in the case of the top team) of the organization. Examples of leadership theory that investigate leadership at this level include strategic leadership and flexible leadership theory.
LEVELS OF ANALYSIS USED BY LEADERSHIP THEORIES ONTOLOGY AND EPISTEMOLOGY
Ontology refers to our assumptions and beliefs about the nature of reality, while epistemology relates to how we gain, understand and communicate our knowledge about the ontological assumptions we have (Hatch & Cunliffe, 2006; Hunt, 1991). The ontological and epistemological stances that researchers, consciously or unconsciously, employ in their research influence the type of questions and hypothesis they will formulate about a phenomenon and the selection of methodology they will utilize to establish their answers. Hence, the perspective that different scholars have made about the ontology and epistemology of leadership might be an important factor to understand the assortment of leadership theories.
Hunt (1991) has organized the knowledge of leadership into six ontological/epistemological stances, in a continuum that range from more objective approaches to more subjectivist approaches (see Table 3 below). This continuum moves from a very static view of leadership, in which the relationship between the leader and follower is predetermined by particular patterns and principles that can be generalized and calculated, to more fluid understandings of the role of the leader, the follower and the environment in constructing and enacting this relationship.
Interestingly, there is not a fair distribution of leadership theories within this continuum. The majority of leadership theories could be classified as pertaining to the machine and/or organism categories. When Hunt (1991) published his book, all of the major leadership theories he analyzed fitted in these groups, and he stated that “the subjectivist leadership perspective largely have been neglected” (p. 51); this reality has not been drastically changed, although more subjective theoretical and empirical inquiries are becoming less uncommon (e.g. Barge, Fairhurst, 2008; Shamir, Dayan-Horesh & Adler, 2005).
More importantly, even Hunt’s (1991) ontological and epistemological portrayal of leadership knowledge contains an implicit ontology that he might not have realized at the time. Drath and colleagues (2008) have argued recently that there is an underlying ontological commitment beneath the majority of leadership theories and models that in fact brings together the seemingly vast and fragmented leadership literature. This essential assumption indicates that leadership is composed by a tripod between three entities: leader (or leaders), followers, and a common objective shared by them. This ontology, which was named the tripod ontology, constitutes the basis of the majority of leadership theories (see Figure 4A). Different theories and definitions, which seem to focus on distinctive aspects of leadership, in fact, “arrange and rearrange the entities of the tripod in various ways” (Drath et al, 2008: 637).
This is not, however, the only way in which leadership can be conceptualized. Drath and colleagues (2008) suggest a new ontological assumption that they believe can assimilate recent theories of leadership (such as shared leadership, relational leadership and complexity leadership) that do not utilize the tripod ontology and can also promote novel insights about the phenomenon. This new ontology, which was named DAC ontology, is composed of three elements: direction, alignment and commitment (see Figure 4B). According to these authors, the DAC ontology has the capability of promoting more integration not only among leadership theories but also between theory and practice due to its focus on the long term outcomes produced by the collective attainment of direction, alignment and commitment, as outcomes are not restricted to a level of analysis and are more connected to the language and interests of practitioners.