Contextualizing leaders' interpretations of proactive followership

Research context

We investigated leaders' interpretations of followership by drawing from the experiences of highly competitive sport team coaches. The rationale for this choice is that leader–follower interactions are particularly salient when people are engaged in problems that require social coordination, as their interactions with one another can have immediate ramifications for collective outcomes (Van Vugt et al., 2008). Moreover, deliberate and purposeful interactions frequently occur between leaders and followers when group members are bound together by a high degree of task and outcome interdependence. In addition, team members' roles differ according to specialized functions and hierarchical positions, offering additional contextual information that may be relevant to understanding how follower behaviors are perceived. Although several scholars have suggested that sport offers a fertile context for studying a range of issues relevant to organizations (Day, Gordon, & Fink, 2012; Wolfe et al., 2005), the group properties of sport teams are particularly similar to those found in localized work teams (Kozlowski & Bell, 2003). Head coaches of highly competitive sport teams share many of the same duties and responsibilities as leaders who manage localized work groups, including designing and monitoring training activities, team member recruitment, delegating roles, and hiring staff.

Participants

After obtaining approval from an institutional ethics board, full-time professional head coaches of Canadian Interuniversity Sport programs were purposefully recruited for their insights related to followership in a team setting. Fourteen coaches of male (n = 8) and female (n = 6) sport teams from seven Canadian universities agreed to participate. Half of the participants were the same gender as the team they coached, but there was an uneven distribution of male (n = 12) and female (n = 2) coaches. The number of female participants (14.5 percent) is, unfortunately, fairly representative of the current ratio of male/female head coaches in the Canadian Interuniversity Sport (19 percent, Donnelly, Kidd, & Norman, 2011). The sports coached by the participants included basketball (n = 5), football (n = 3), hockey (n = 3), and volleyball (n = 3). On average, participants had 15 years (SD = 8.97) of professional coaching experience at the interuniversity level. We used publicly available information to direct our recruiting efforts to individuals whose experiences reflected coaching excellence (e.g., team championship(s), coach of the year award(s), and experience coaching for the Canadian national team). Within our sample, 11 coaches had been officially recognized as coaches of the year in their conference (e.g., Ontario University Association and Canada West), 10 had won provincial or national championships, and five had coached for their respective national team. We selectively invited coaches on the basis of these accolades to capture the perspectives of individuals who were considered to be experts in their ability to manage group–member interactions in a high-performance team setting and would have sufficient personal experiences to draw from when recounting their interpretations of followership in different contexts. We sought a relatively heterogeneous sample in terms of sport type and gender of the teams coached to enhance transferability of our findings (Patton, 2002).

Development of the interview guide

Semi-structured interviews were conducted to capture contextually situated descriptions of followership. To enhance the depth and richness of participants' responses, each interview began with rapport building questions to put participants at ease and promote a conversational flow (e.g., “Can you tell me about your previous experiences as a coach at this level?”). Next, questions were posed to elicit participants' general thoughts on the topic of followership (e.g., “Within your team, who do you consider to occupy a followership role?”). Our key questions explored leaders' interpretations of followership behavior (please refer to Appendix A for these key questions). These questions were designed to tap into participants' episodic memories to gather contextually rich descriptions related to acts of followership. That is, participants were encouraged to discuss their actual experiences (e.g., “Can you tell me about an example of an individual who was a particularly effective follower?”), which were followed up by elaborative probes (“Can you tell me about a specific example of this individual's behavior?”) and contrast probes (“Can you tell me about an example of an individual who was a particularly ineffective follower?”). Based on these initial responses, we then explored whether leaders' expectations of followership behavior varied across contexts (“Can you share any experiences where followers engaged in that behavior in a different context?”) and then followed up with questions to gather their reactions to those behaviors (How did you react in that situation?). From a social constructivist perspective, semi-structured interviews offer the flexibility to explore the range and variation in which acts of followership emerge. For example, many initial descriptions of followership behavior were restricted to athletes in their first year with the team (i.e., rookies). These responses were then followed up by questions asking about other instances where follower–leader dynamics emerged. Although each interview covered the same key areas of interest, the order of questions varied by what was deemed to be the most appropriate given the nature of participants' responses. The researcher responsible for conducting the interviews was experienced in semi-structured interviewing.

Collection and analysis of data

The conduct and corresponding analysis of interviews followed a reflexive iterative framework (Srivastava & Hopwood, 2009). Based on this framework, the interviewer kept a journal denoting relevant observations, which was updated following each interview to gauge how incoming data aligned with our interest in articulating a contextualized representation of followership. This provided a way to explore emergent areas of interest while informing our decision as to when there were enough data to discern core consistencies in leaders' interpretations of followership. At this point, when subsequent interviews were no longer providing novel information related to the research questions that we set out to explore, data collection ceased.

The first six interviews were conducted in person, with the next eight interviews conducted via telephone to expand our geographical reach to target coaches who were considered to be at the upper echelon of their respective league. The duration of interviews did not substantively differ as a function of whether they were conducted in person (M = 46 minutes) or via telephone (M = 42 minutes). Although some critics argue that in-person interviews are superior to telephone interviews in terms of the depth and quality of information obtained, a review comparing these modes of interviewing (i.e., phone versus face to face) found no empirical support to claims of superiority (Novick, 2008). The interviews were all digitally audio-recorded and transcribed verbatim, which resulted in a total of 232 pages of transcribed data (excluding initial questions aimed at rapport development).

Interviews were analyzed in a manner whereby emerging insights were compared with existing theory (Srivastava & Hopwood, 2009). Despite the relatively fluid nature of this process, our thematic analysis proceeded in several stages. The lead researcher—who collected and transcribed all of the data—began examining transcripts in their entirety to have a sense of the dominant viewpoints communicated. The understanding gleaned from these initial readings facilitated an open-coding process, which required the researcher to identify regions of text where there was a noticeable shift in meaning, that is, when the content contained within a passage of text no longer conveyed information about the same idea or topic. These segments of text were coded based on their content to form the initial categories of data. Despite the benefits of fragmenting transcripts into more manageable units of data, a valid concern is losing contextually relevant details. To address this concern, the researcher then revisited each transcript to ensure no contextually relevant details were lost while parsing out segments of text. These coded passages were then compared with one another to search for common (and distinctive) properties to organize into coherent themes. To understand the essential properties within each theme, pertinent scholarly works were consulted to facilitate a conceptual understanding of these data-driven categories (Srivastava & Hopwood, 2009).

At this point, a second member of the research team independently reviewed 40 percent (112/282) of the coded quotes allocated to themes to ensure that the raw data reflected the theme in which they were housed. Over the course of several meetings, the researchers met to discuss the nature of each theme to resolve any potential discrepancies. No systematic issues in the coding of categories were identified, and the research team came to a consensus on how to slightly revise the description of several themes to better exemplify the range of content captured by participants' descriptions (i.e., ensuring each theme's heading accurately depicted its substantive content). The final stage of analysis was an interactive process where all members of the research group deliberated on how the themes could be best positioned together to articulate leaders' interpretations of followership, which continued until a consensus among the research team was achieved.

Personal qualities associated with followership

Beyond this core quality, coaches detailed characteristics that differentiated between what they perceived as effective and ineffective followership. In response to queries about team members who they considered to display exemplary followership, coaches viewed ideal followership as exhibiting (i) a collective orientation, (ii) active independent thought in the context of team values, (iii) relational transparency, and (iv) an ability to process self-related information accurately.

Thematic summary

Leaders' interpretations of what constitutes appropriate followership covered an array of characteristics. These qualities depict a follower who is proactive in his or her responsibilities, accountable, willing to challenge leader ideas and offer alternative insights, and display an overt commitment to supporting team efforts. However, portraying ideal followership as a set of qualities rests on the assumption that these qualities will translate into context-appropriate behaviors—yet not all situations are conducive to proactive follower input (Chan, 2006; Whiting et al., 2012). The following section illustrates how proximal contextual factors that are salient during leader–follower interactions constrained when proactive follower behaviors were viewed favorably by coaches.

Proximal contextual factors influencing perceptions of followership

Coaches' perceptions of appropriate followership depended upon a number of proximal contextual factors, and these factors were interrelated in a number of ways. Endorsements of followership behaviors were context specific; coaches coveted individuals who were able to understand the changing role dynamics in the group and act accordingly. Leaders' construal of follower behaviors depended upon interplay among the following contextual features: (i) presence of third-party observers, (ii) performance versus learning contexts, (iii) stage in the decision-making process, (iv) suitability of the targeted issue, and (v) relational dynamics.

Thematic summary

Leaders coveted followers who had an acute awareness of which contexts permitted more proactive role engagement, versus contexts that required committed deference to leader decisions. In this sense, they desired individuals who could flexibly adapt to the situation at hand. For example, although leaders endorsed active and independent thought as a quality of followership, this was based on the supposition that followers would not constructively resist them during inopportune moments (e.g., performance-critical moments and after consensus on a decision had been reached). Likewise, relational transparency was a coveted followership trait, but there was a time and a place for initiating interpersonal dialogue. As Hannah, Sumanth, Lester, and Cavarretta (2014) suggested, failing to distinguish between the antecedent characteristics of good leadership and how these traits manifest into acts of leadership masks the complexities and context-embedded nature of a leader's actions. Although the point made by Hannah et al. was directed toward the study of leadership, it is equally applicable to our discussion of followership. In sum, contextualizing acts of followership offers insight into when leaders are likely to endorse proactive followership behaviors in a team setting.

Theoretical implications

The context-specificity of followership tended to revolve around proactive follower behaviors, which entailed unsolicited advice or input from followers. Consistent with these findings, previous work has shown that both political skill (Sun & van Emmerik, 2015) and an ability to gauge the current situation (Chan, 2006) moderate the relationship between proactivity and supervisor performance evaluations. Although proactive behavior is a way for followers to influence the leadership process (Carsten et al., 2010; Oc et al., 2014), the potential rewards of proactive followership are not without risk (Grant & Ashford, 2008). One way to understand why these behaviors are a double-edged sword in the context of leader–follower dynamics is by revisiting a defining feature of proactivity, which is that these behaviors are self-initiated. This means that the motives underlying proactive followership are largely unknown to observers, leaving the intentions behind such behavior open to interpretation. Considering that people attend to salient contextual cues in the environment when making inferences about others' motives (Grant et al., 2009), attempts to influence a leader's decisions in public may not be viewed as a thoughtful enterprise to benefit achievement of collective goals (i.e., prosocial motive), but perhaps as an attempt to subvert his or her leadership in front of others (i.e., egoistic motive). Similar reasoning is applicable to why leaders were adamant that followers required awareness and respect toward the topics that were non-negotiable and at which stage in the decision-making process they simply needed to accede to the final decision. In other words, despite the best of intentions, overzealous follower behaviors (e.g., offering an alternative strategy and continued insistence) may be construed as acts of insubordination when enacted at inopportune moments.

Moreover, leaders' negative views toward proactive followership in situations where performance outcomes were of immediate concern are supported by temporal models of teamwork. Marks et al. (2001) suggested that the pursuit of common team goals represents a cyclical process of moving back and forth between action and transition phases. Action phases entail situations where performance outcomes are emphasized, whereas transition phases are characterized by a greater concern toward continued improvement as team members reflect and plan for future action phases. Within this model, formulating strategies and goals through discussions among team members is crucial during transition phases but can be detrimental during action phases because of the high degree of coordination required among individuals. As such, the demands placed on leaders during action phases requires commitment-oriented acts of followership, as additional debate and discussion from followers is likely to hinder, rather than enhance, team functioning. A caveat to this finding, however, is that leaders primarily referenced athletes' followership behaviors when discussing this contextual boundary, which has potentially weaker parallels to a supervisor–employee relationship than the coach–assistant coach relationship. In addition, organizational teams may not regularly encounter situations where there is a clear distinction between emphasizing immediate performance outcomes versus strategic development. Although this contextual boundary may be less applicable to the leader–follower dynamics found in many organizational settings, it applies to certain specialized groups (e.g., a military unit and a surgical team in a hospital). For example, in medical teams, performance outcomes are heavily emphasized during a crisis event, but learning is emphasized during debriefing periods (Hunt, Shilkofski, Stavroudis, & Nelson, 2007).

In addition to the different ways leaders construed acts of followership (i.e., in a positive or negative manner) across contexts, leaders described instances where pre-existing interpersonal dynamics influenced the degree to which acts of followership influenced their decisions. That is, leaders were more receptive to being influenced and challenged by followers who were closer to them in group status (i.e., prestige associated with formal rank). This aligns with Oc and Bashshur's (2013) model of social influence, which posits that followers who occupy higher ranks of status in a group may have a greater capacity to influence their leaders because of the strength and heightened psychological immediacy such positions afford. This has interesting implications for how followers may formulate their attempts to influence others in multi-tier hierarchical groups, as a low status follower may find better success by attempting to influence their most proximal leaders who can then relay such messages up the hierarchy. An interesting twist, however, is that although leaders favored proactive followership behaviors enacted by high-status group members, leaders indicated that they reacted more harshly to contextual violations enacted by higher status group members.

Limitations and considerations for future work

In recognizing that interpretive qualitative research seeks to generate contextually rich knowledge based on participants' recall of events, these findings must be considered within several boundaries. Although we deemed our sample size appropriate by the consistency and range in which contextual factors were identified—which were additionally supported by existing theoretical accounts of why interpretations of behaviors are sensitive to contextual factors—we studied a select number of leaders' experiences that were limited to groups characterized by high degrees of task and outcome interdependence. However, we echo Williams (2000) who argued that although interpretive modes of research emphasize the subjectivity of participants' experiences, if researchers carefully consider the structural properties shared between group contexts when extrapolating insights generated in the context of highly competitive sport teams to another domain (e.g., military and businesses), these insights can usefully inform both theoretical and practical issues that may remain overlooked in less descriptive modes of inquiry. When generalizing these findings to organizational teams, it is important to consider that in sport teams, leader–follower interactions occur frequently, primarily in person, team members are under constant scrutiny for their actions (i.e., roles are contingent upon frequent performance evaluations), and roles are guided by a formalized status hierarchy. In addition, metrics for performance in sport are readily available, but outcomes produced by individual efforts are difficult to isolate from collective efforts. It is also important to consider the nature of the leader–follower relationships described in the current study. Although the use of student participants is common in (experimental) studies of proactive behaviors (e.g., Maynes & Podsakoff, 2014, Studies 1 and 2; Oc et al., 2014, Study 2; Whiting et al., 2012), leaders' insights pertaining to followership were partly based on coach–athlete relationships, which may differ from traditional supervisor–subordinate working relationships. Nevertheless, our key personality characteristics associated with effective followership aligns well with the qualities identified in studies of the workplace (Carsten et al., 2010; Dulebohn et al., 2012; Luthans & Avolio, 2003). In addition, a supplementary analysis of the themes related to the contextual boundaries of proactive followership revealed that coaching staff- and athlete-specific references were generally equivalent in terms of frequency, with the exception of the performance versus learning contexts and stage in the decision-making process themes. Taken together, the findings of the current study are most applicable to understanding the leader–follower dynamics that occur within localized work groups/teams that share these aforementioned properties.

A related consideration is that relying on leaders' recall and interpretation of past events puts an inherent constraint on the claims that can be made from these data. We only illuminated one side of the Uhl-Bien et al. (2014) leadership process approach by capturing leaders' perspectives of followership. A reasonable extension is to explore whether other contextual demands are brought into focus when explored through the lens of a follower's perspective. In addition, there may be additional and meaningful contextual factors that are shielded by cognitive blind spots, which are not accessible through semi-structured interviews (Schacter, 1999). From this standpoint, capturing in situ observations of critical incidents related to acts of followership may help bridge the gap between relying on how leaders and followers recall previous events.

Another boundary pertains to the exploratory nature of the present findings, as we did not test the proposed relationships between contextual factors and leaders' interpretation of followership. Manipulating these aforementioned contextual variables via experimental designs could help to discern whether, in fact, certain proximal contextual factors cause leaders to interpret follower motives in a different light (e.g., communal and self-serving) and whether this influences how leaders respond to acts of followership. Longitudinal designs, however, offer a more ecologically valid assessment to track how leaders respond to contextualized acts of followership over a period of time, albeit at the expense of experimental controls.

An attractive aspect of applying a more contextualized lens to the study of followership is that it opens several possibilities for expanding upon previous findings. As one example, Tepper et al. (2006) found that followers were rated positively when they used negotiation tactics in the context of a higher quality leader–follower relationship, but such negotiation tactics were futile in the context of a lower quality leader–follower dynamics, even though the frequency with which followers negotiated were similar. A reasonable question to pose is why are negotiation tactics ineffectual for followers in low LMX? One explanation is that perhaps followers who negotiate in context-appropriate situations foster a higher quality LMX, whereas followers who negotiate at inopportune moments undermine the quality of LMX.

Practical implications

Leaders viewed followership as a critical aspect of the leadership process, as they relied on followership to offer alternative insights or solutions to problems while also expecting them to collectively support the directives established by the leadership. Contextualizing acts of followership revealed how proactive attempts to influence leaders at the wrong time and place can render a well-intentioned behavior as disruptive in the eyes of a leader. An implication of this perspective is that followership requires an acute awareness and ability to calibrate one's actions in accord with the circumstances, as individuals need to understand when, and under which circumstances, certain patterns of followership are expected (e.g., overt commitment/compliance and proactive feedback). From a group dynamics perspective, it would be advantageous to clarify the contextual boundaries pertaining to when certain acts of followership are permitted, desired, or admonished. Proactively addressing such expectations might be especially effective in groups that operate with a multi-tier leadership structure, as inappropriate followership behaviors from people with intermediary formal authority—in our study, this referred to team captains and assistant coaches—may be the most disruptive to leadership efforts. A practical concern, however, is that the pejorative nature of being labeled a “follower” has tempered attempts to popularize the importance of followership (Hoption, Christie, & Barling, 2012). Given leaders' specific expectations for how they wanted followers to act in certain situations, it would be prudent for leaders to open a dialogue pertaining to which patterns of followership they view as being supportive—rather than undermining—their leadership efforts. In addition, teams may benefit from reframing the terminology surrounding the role of followers. In accord with the leadership process approach (Uhl-Bien et al., 2014), one solution is to convey that people engage in acts of followership, just as others engage in acts of leadership, and to explain that these roles are mutually supportive of one another as people shift between acts of leading and following. Focusing on establishing and clarifying expectations pertaining to effective acts of followership, rather than what makes a good follower, may encourage group members to view followership not as a static role of subordination but as an essential aspect of cooperative group efforts that is entrenched within the leadership process.