The consequences of continuous technological improvements, increasing accountability, and changes in working life require new responses from higher education. New modes in the delivery of education, pedagogy, and teaching are often called for (e.g., Barber et al. 2013; Fullan and Scott 2009; Kezar 2011). However, acknowledging the importance of change does not automatically result in successful innovations. University teacher’s involvement appears to be a key factor for sustainable educational change (e.g., Emo 2015; Van Driel et al. 1997), and connecting these professionals has been identified as one of the basic requirements for complex change processes to achieve innovation (e.g., Kotter 2012).
In this regard, it has been suggested that encouraging university teachers to work on an innovative task together does play a crucial role in achieving educational change. Innovative tasks are defined here as highly novel, complex, and low-structured (De Dreu and Weingart 2003; Devine 2002; Hoegl et al. 2003). Fullan (2010) suggests that working together on an innovative task potentially offers university teachers the opportunity of combining multiple inputs in identifying the need for innovation, developing ownership, and designing, implementing, and evaluating solutions. Roxå and Mårtensson (2015) argue that collaboration between university teachers takes place in various forms and that collaborative forms of innovation are mainly characterized by a shared responsibility for educational development. However, Vangrieken et al. (2015) question the extent to which collaborating university teachers in fact share and follow up on this responsibility. Cox (2004) observed that university teacher’s work tradition is largely solitary, with high levels of individual autonomy. This work tradition also exists in secondary education, with Brouwer et al. (2012) finding that when teachers collaborate, their interdependency is seldom task-related, leading to a low shared outcome responsibility and accountability. Yet, research outside the educational domain also provides ample evidence that teams of professionals that are interdependent and share responsibility can be very successful in tackling innovative tasks (e.g., Lee et al. 2010; Zaccaro et al. 2008).
Such teams are not groups of individual professionals who are loosely coupled for reasons such as simply working in the same department or sharing an interest (Katzenbach and Smith 1993). In contrast, a team is “a collection of individuals who are interdependent in their tasks, who share responsibility for outcomes, who see themselves and who are seen by others as an intact social entity embedded in one or more larger social systems” (Cohen & Bailey, 1997, p. 241). This task interdependency and shared responsibility distinguishes teams from other forms of collaboration (Katzenbach and Smith 1993). This study uses a team perspective to investigate teacher collaboration for innovation in higher education, first, because teams appear to be very effective for achieving innovation (Zaccaro et al. 2008). Second, Roxå and Mårtensson (2015) reasoned that collaborative forms for innovation in higher education are mainly characterized by sharing a responsibility for educational development. A team approach is in line with that reasoning, because in such an approach, teachers formally share responsibility and task interdependency exists. Third, studies on teacher collaboration are hardly ever conducted in higher education (Kezar 2014; Vangrieken et al. 2015), while overall, there are few studies on effective conditions and outcomes of teacher collaboration for educational innovation (Little 2006; Vescio et al. 2008). Therefore, this study builds on studies performed in organizational and educational contexts (e.g., Crow and Pounder 2000). When describing the context, higher education literature takes precedence, although studies conducted in secondary education were also used where the former was lacking, since the setting has similar features (e.g., Kezar 2014), such as the tradition of professionals working solitarily and autonomously (Brouwer et al. 2012; Cox 2004; Van Waes et al. 2015) and the tendency of teachers to avoid change (Furco and Moely 2012; Van Eekelen et al. 2006).
Research shows that effective teams with an innovative task can adapt to new situations and develop new knowledge together through engaging in team learning behaviors, which in turn explains their success (Lee et al. 2010; Srivastava et al. 2006). Decuyper et al. (2010) define such behaviors in terms of three essential learning behaviors: sharing, constructive conflict, and co-construction. Team members who demonstrate all three learning behaviors can build new knowledge, solve complex problems, and develop innovative solutions collectively (Paavola et al. 2004; Van den Bossche et al. 2006). Team effectiveness depends on these team learning behaviors and also encompasses interpersonal factors (Edmondson 1999). In this regard, team psychological safety and team efficacy appear to be conditional for team learning behaviors (Edmondson 1999; Gully et al. 2002).
Despite evidence that a team approach can be successful for innovation in different contexts, simply bringing together university teachers in teams with an innovative task may not be enough (e.g., Fullan and Scott 2009). University teachers’ team learning behavior needs to be encouraged for a number of reasons. Firstly, Cox (2004) showed that university teachers operate in a long tradition of solitariness and are not used to sharing and discussing their practices together. It appears that they need to feel safe in a social sense before they engage in such collaborative learning behaviors (Roxå and Mårtensson 2009; Van Waes et al., 2015). Secondly, establishing teams for innovative purposes does not automatically mean that the team members will acknowledge the innovative features of their team task. This is reflected in the review of Timperley et al. (2007), which showed that collaborating secondary school teachers tend to seek support for the status quo and marginalized or even ignored new ideas. This adherence to routine might be also present in higher educational contexts, which could hinder university teacher’s need to engage in team learning behaviors. Furco and Moely (2012) showed that university teachers need support from their faculty in taking the perceived risk of sharing practices and co-constructing new methods. Therefore, several authors claim that the readiness of university teacher teams to become engaged in collaborative learning depends heavily on how this is encouraged and facilitated by team leadership (Furco and Moely 2012; Kezar 2005; Roxå and Mårtensson 2009).
Many studies across a wide variety of settings have shown that team leadership behavior plays an important role in fueling team learning behavior (Bucic et al. 2010; Burke et al. 2006; Harris 2011). For example, Lee et al. (2010) found that leaders who inspired and encouraged team members in developing new ideas and trying different approaches supported team learning behaviors in innovative IT teams. Additionally, Bryman (2007) and Van Ameijde et al. (2009) showed that sharing such leadership behaviors can motivate university teachers facing innovative tasks collaboratively, because it provides a sense of team ownership (e.g., Carson et al. 2007; Pearce et al. 2008). However, most studies do not integrate multiple types of leadership behaviors or are mainly focused on the leadership behaviors of a single vertical leader (Avolio et al. 2009). Such research also predominately focuses on team performance as an outcome and not on team learning (Burke et al. 2006; Nicolaides et al. 2014). Moreover, research on the influence of team leadership on teacher team learning is lacking, because empirical studies on stimulating collaborative teacher learning are limited (Little 2006; Vescio et al. 2008) and mostly do not concern teams but collaboration forms with lower levels of task interdependence and shared responsibility (Brouwer et al. 2012; Vangrieken et al. 2015). In addition, these studies have mainly been conducted in primary and secondary education, and rarely in higher education (Kezar 2011; Vangrieken et al. 2015). For these reasons, this study aims to understand how university teacher teams established by the organization learn to deal with their task together. This will be done by exploring the role of team members’ perceptions of learning, their task, and the leadership behaviors in their team.
Learning in teacher teams
Decuyper et al. (2010) identify three essential team learning behaviors: sharing, constructive conflict, and co-construction. Sharing is defined as “the process of communicating knowledge, competencies, opinions, or creative thoughts of one member to other team members, who were not previously aware that these were present in the team” (Decuyper et al. 2010, p. 116). Constructive conflict is viewed as “a conflict or an elaborated discussion that stems from diversity and open communication, and leads to further communication and some kind of temporary agreement” (Decuyper et al. 2010, p. 117). In a constructive conflict, differences are “negotiated by arguments and clarifications” (Van den Bossche 2006, p. 91). Finally, co-construction refers to the process of developing shared knowledge and building shared meaning “by refining, building on, or modifying an original offer in some way” (Baker, 1994, in Van den Bossche et al. 2006, p. 495). According to Van den Bossche (2006), “the outcome of this process is that ‘new’ meanings emerge in the collaborative work that were not previously available to the team” (p. 91). All three team learning behaviors are deemed relevant for developing innovative solutions collectively: sharing the available cognitive resources and unique expertise, integrating different viewpoints, and collaboratively building new knowledge (Lee et al. 2010; Paavola et al. 2004).
It is crucial not only to consider the cognitive aspect of team learning but also to include the social process as well (Van den Bossche et al. 2006). Roxå and Mårtensson (2009) showed that university teachers needed to feel safe and to experience mutual trust to engage in collaborative learning behavior, because performing such behaviors is risky and causes uncertainty. This phenomenon has been studied extensively in primary and secondary education, where teachers’ traditional work climate is characterized by ignoring differences and pursuing support and consensus, rather than questioning and seeking professional disagreements (Hargreaves 2001). Kwakman (2003) showed that secondary school teachers preferred sharing views only and that they perceived disagreements as threatening instead of viewing them as opportunities to examine opposite views. Van Eekelen et al. (2006) also found that secondary school teachers preferred a predictable, routine approach to work, and therefore tend to avoid risks. According to them, this anxiety caused by change was due to low self-efficacy (Bandura 1997). Therefore, in this paper, we argue that university teacher team learning should be studied as both a cognitive and a social process (e.g., Roxå and Mårtensson 2009; Van den Bossche et al. 2006).
Across domains, psychological safety and team efficacy have been consistently identified as important interpersonal factors for team learning (Decuyper et al. 2010; Gully et al. 2002). Edmondson (1999) referred to team psychological safety as a “sense of confidence that the team will not embarrass, reject or punish someone for speaking up” (p. 354) and added that team members do not feel rejected when putting themselves at risk, for example, by seeking feedback, admitting errors or asking for help. Team efficacy is defined as the collective perceived ability to work together to achieve goals (Bandura 1997). Collins and Parker (2010) showed that a strong belief in a team’s abilities leads to more ambitious goals, to the development of strategies to achieve those goals, and to increased persistence in the face of setbacks. Psychological safety and team efficacy might also lead to social support for university teacher team members in taking risks and overcoming problems.
The role of task perception on team learning behavior
We argue that a team’s task is not an objective fact: it depends on its members’ perceptions of the task, given their attitudes and work experience. Imants et al. (2013) showed that secondary school teachers’ perceptions of their collaborative task regarding educational change strongly influenced their attitudes towards engaging in collaborative learning behaviors. Similarly, research in other domains shows that recognizing task features, such as interdependency and innovativeness, can be expected to support team learning behaviors (Hoegl et al. 2003; Van Eekelen et al. 2006). Task interdependency means that “one perceives that one is linked with others in a way so that one cannot succeed unless they do (and vice versa) and/or that one must coordinate one’s efforts with the efforts of others to complete a task” (Johnson and Johnson 2003, p. 173). When members see that their effort is needed, they increase their contribution, which subsequently benefits team learning (Johnson and Johnson 2003). Task innovativeness contains three elements: novelty, structure, and complexity. Task novelty is the perceived amount of new or unknown task elements (Edmondson et al. 2007; Hoegl et al. 2003). Task structure refers to the extent to which the task, methods, and outcomes are observed as prescribed/given or open/unpredictable (Ellström 2001). Task complexity involves recognizing the task’s difficulty and any absence of standard solutions (Cooke et al. 2001; De Dreu and Weingart 2003).
In contrast to perceiving a task as repetitive or routine, innovative tasks are perceived as highly novel, low-structured, and highly complex (De Dreu and Weingart 2003; Devine 2002; Hoegl et al. 2003), which is argued to trigger team learning behaviors (Edmondson et al. 2003). Perceiving task novelty as high could stimulate team members to collectively develop new solutions instead of adhering to the status quo (Hoegl et al. 2003). Observing low-structured tasks implies that team members may feel the need to clarify their task, develop new methods, and deal with ambiguous outcomes together (Devine 2002). If team members perceive task complexity as high, this suggests they are more likely to collaborate in order to solve their difficult and unstandardized problem (Cooke et al. 2001; De Dreu and Weingart 2003; Van der Haar et al. 2013).
Ellström (2001) explains the influence of task perception on team learning using two learning levels: adaptive and developmental learning. If team members perceive their task as routine, they are more likely to engage in “adaptive learning.” In that case, sharing is considered to be enough for success (Paavola et al. 2004). In contrast, perceiving the task as innovative means it is likely that “developmental learning” will occur (Ellström 2001) for which constructive conflicts and co-constructions are necessary, in addition to sharing (Paavola et al. 2004). Team members’ task perception is thus argued to play a role in performing team learning behaviors (Hoegl et al. 2003; Imants et al. 2013). London (2014) also argues that future research should focus on the role of team leadership behavior in influencing team members’ task perception and, subsequently, their engagement in either adaptive or developmental learning.
The role of team leadership behavior in supporting university teacher team learning
Team leadership is repeatedly identified as a critical factor in supporting team learning behaviors (Burke et al. 2006; Harris 2011). Team leadership behaviors refer to the processes of influencing and facilitating, that is, “influencing others to understand and agree about what needs to be done and how it can be done effectively; (…) facilitating individual and collective efforts to accomplish a shared objective” (Yukl, 2002, in Ensley et al. 2006, p. 220). This definition includes two team leadership perspectives: style (i.e., transformational and transactional) and source (i.e., vertical and shared). Bass and Avolio (1994) operationalized the transformational leadership style as leaders who motivate members via behaviors such as articulating a vision, setting high expectations, questioning the status quo, and supporting the individual needs. The transactional leadership style consists of behaviors that establish agreements on the tasks, the necessary facilities, and the rewards for achieving them (Bass and Avolio 1994). Transactional leaders also actively monitor team members’ performance and take action when mistakes are made. When dealing with an innovative task, it is unlikely that a single, vertical leader will have all the answers and will be able to perform both leadership styles adequately (Day et al. 2004; Timperley et al. 2007). Therefore, a shared team leadership approach is suggested, which Carson et al. (2007) described as team leadership behaviors that stem from multiple sources: the team members themselves.
Thus far, there is no consensus on which style and source of leadership is most effective in supporting team learning (Burke et al. 2006; Nicolaides et al. 2014; Zaccaro et al. 2008). On the one hand, vertical transformational leadership behaviors appear to stimulate team learning on innovative tasks. Moolenaar et al. (2010) found that such behaviors supported secondary school teachers’ recognition of innovative task features and their sense of urgency and willingness to collectively develop new knowledge. Furthermore, vertical transformational leadership behaviors appear to support team learning via promoting team psychological safety and team efficacy. Lee et al. (2010), for instance, showed that team leaders who advise and provide new information build interpersonal safety and trust, which explained 69% of the team knowledge sharing variance (e.g., Edmondson 1999; Srivastava et al. 2006). On the other hand, Timperley et al. (2007) stressed that vertical transactional leadership behaviors are also promising for structuring collaborative teacher learning in (secondary) education. Bucic et al. (2010) and Mebane and Galassi (2003) demonstrated that both vertical transformational and transactional leadership styles encouraged university teacher team learning by challenging teachers to share and by structuring the task. The reviews of London (2014) and Nicolaides et al. (2014) suggest that the most effective leadership style depends on the team’s situation, such as the task features or perceived task features, but supporting evidence is limited.
To date, studies that include both leadership styles are scarce and mainly focus on a single, vertical leader (Avolio et al. 2009). However, Van Ameijde et al. (2009) found that teams in higher education benefit from shared leadership behaviors; specifically, this is the case if such behaviors involve collectively building ownership and trust, decision-making, and monitoring performance. Sharing these leadership behaviors and giving university teachers the space to perform them provides them with a degree of autonomy and influence, which can motivate them to solve complex problems collectively (Bryman 2007). Despite the promising influence of shared team leadership, more evidence on how shared leadership behaviors fuel teacher team learning is necessary (Nicolaides et al. 2014). Therefore, this study draws upon both vertical and shared leadership, by considering that transformational and transactional leadership behaviors can both stem from a vertical leader and be shared by the team members (e.g., Bryman 2007; Ensley et al. 2006).