Abstract
Despite knowing that a shared vision can result in a number of positive outcomes for groups and teams, we have a very limited understanding of what potentially contributes to group and team members sharing the same vision. This study seeks to contribute to our very limited understanding of what potentially contributes to team members sharing their team’s vision by focusing on team members’ expectations of their teammates engaging in discretionary behaviors directed toward the organization, the degree to which team members trust their teammates, and the degree to which team managers provide their teams autonomy. Using 197 team members across 46 teams/team managers and 13 organizations, I find that team members’ expectations of their teammates engaging in citizenship behaviors directed toward the organization is positively associated with team members sharing their team’s vision, but I find no association for team members’ expectations of their teammates engaging in deviant behaviors directed toward the organization and sharing their team’s vision. However, I find that trust weakens and team autonomy strengthens the negative relationship between team members’ expectations of their teammates’ deviant behaviors directed toward the organization and sharing their team’s vision. I find no conditional effects of trust and team autonomy on the relationship between OCBO expectations and shared vision. These results suggest that both properties of individual team members as well as the team may play an important role in sharing a team’s vision.
“In order to serve its purpose, a vision has to be a shared vision.” ∼ Warren G. Bennis
The importance of group sharedness is emphasized in the popular press (e.g., Lencioni, 2002) and in the classroom (e.g., Wheelan, 2013). This is especially true about sharing a vision, or “the extent to which members of an entity agree on a vision for the future” (Berson et al., 2015: 84). Unlike other forms of shared cognition, shared vision is prescriptive rather than descriptive (Mumford & Strange, 2013). At the organizational-level, a shared vision “…embodies the collective goals and aspirations of the members of an organization” (Tsai & Ghoshal, 1998: 467) and is often reflected in organizations’ mission statements, such as Nike’s (2024) mission to “Bring inspiration and innovation to every athlete in the world” and The Walt Disney Company’s (2024) mission “…to entertain, inform and inspire people around the globe through the power of unparalleled storytelling…” Groups and teams within organizations have shared visions as well, described by Pearce and Ensley (2004: 260) as “…a common mental model of the future state of the team or its tasks…“, like team goals (e.g., Hofhuis et al., 2018). Although we know shared vision is associated with positive performance outcomes, such as value co-creation (Chi et al., 2022) and improved service performance (Eldor, 2020), our knowledge about shared vision remains quite limited (Matsuo, 2023), especially about what contributes to sharing a vision (Boyatzis et al., 2015). This is problematic from both a theoretical and practical perspective because researchers need to know how groups and teams can successfully develop a shared vision amongst their members so that researchers can provide evidence-based recommendations to leaders and managers as to how they can successfully develop a shared vision within their groups and teams.
The majority of research on sharing a vision within a team has focused on its outcomes (e.g., Mai, Wu, & Wang, 2022; Su et al., 2022) or its moderating effects (e.g., Resick et al., 2023; Zhang et al., 2012). While theoretical work has identified a number of potential antecedents of shared cognition (e.g., shared vision) within teams, such as individual characteristics, team characteristics, task characteristics, and work structure (Salas et al., 2007), the very limited research that has empirically examined the antecedents of shared vision has adopted a relatively narrow focus. For example, Preston and Karahanna (2009) focused on how a shared vision can be developed between a Chief Information Officer and an organization’s top management team. Thus, the focus was between a specific individual and a specific type of team within an organization. Another example is Zasa, Verganti, and Bellis’ (2022) focus on where innovators and collaborators are positioned within social networks on the formation of a shared vision. Again, the focus is on specific individuals and their relations to others. In this study, I adopt a more generalizable approach to understanding the potential antecedents of sharing a vision by focusing on three characteristics common across most teams, expectations (Salas et al., 2007), trust (Salas et al., 2005), and autonomy (Hollenbeck et al., 2012), each of which are reflective of or influence social activities of team members.
Social exchange is a group of theories that attempt to explain the social activities between two or more entities (Cropanzano et al., 2017), such as between organizations and employees (e.g., Zagenczyk et al., 2013) and between teams and their members (e.g., Seers, 1989). The basic prediction of social exchange is that positive/negative social exchanges will result in positive/negative social outcomes (Cropanzano et al., 2017). Sharing a vision is based on social activity, just like all forms of cognition (Levine & Resnick, 1993). Indeed, Pearce and Ensley (2004) found that vision and teamwork behaviors were reciprocally related. Importantly, social activities need not be limited to the specific interactions that a focal team member has with their teammates. An important component of teamwork is monitoring teammates’ behaviors (Rousseau et al., 2006; Salas et al., 2009). Employees can engage in behaviors directly related to the task or engage in discretionary behaviors (Rotundo & Sackett, 2002), but team members may be particularly attuned to the perceived positive and negative discretionary behaviors of teammates toward the organization since these are the types of behaviors that employees focus on for job performance (Dalal et al., 2009) and the organization has the formal authority to reward and sanction the team because of such behaviors by team members. These discretionary behaviors can serve as social cues for expectations of future team member behaviors (Salancik & Pfeffer, 1978), and based on the basic predictions of social exchange, whether these expectations are for positive or negative discretionary behaviors should influence the degree to which team members share the vision of their team.
However, the social cues provided by team members’ discretionary behaviors do not exist in a vacuum (Salancik & Pfeffer, 1978). The utility of social exchange theories is a function of the degree to which they consider both individual and environmental factors (Cropanzano et al., 2017). Central to all theories of social exchange is the individual factor of trust (Cropanzano et al., 2017), or “the willingness of a party to be vulnerable to the actions of another party based on the expectation that the other will perform a particular action important to the trustor, irrespective of the ability to monitor or control that other party” (Mayer et al., 1995, p. 712). Trust is an important element in more precisely predicting the outcomes of social exchange because it can influence both the activity and desirability of social exchange outcomes (Cropanzano et al., 2017). Thus, social exchange suggests that the degree to which team members trust their team members may play a critical role in more accurately understanding the relationship between discretionary behavioral expectations and sharing a team’s vision.
Likewise, examining the relationship between discretionary behavioral expectations and sharing a team’s vision cannot ignore the environment of the team because it plays an important role in fully understanding teams and their members (Salas et al., 2007). One of the most important environmental factors of teams is how organizations and their managers design teams, as demonstrated by the many studies examining how teams are designed in the workplace (e.g., see Stewart’s, 2006, meta-analysis of team design). Social exchange suggests that particular attention should be paid to team design choices that influence how much team members focus on teammates’ behaviors and, subsequently, their expectations of teammates’ future behaviors. Ryu et al.’s (2022) meta-analysis found that team autonomy, defined as “the amount of collective control and discretion given to a team in the form of setting their own work schedules, determining ways to conduct their tasks, and/or making decisions related to their work” (414), was positively associated with a variety of social activities, such as monitoring the team. Thus, as one of the three dimensions that describe nearly all teams (Hollenbeck et al., 2012), it is important to consider the degree to which a single leader makes decisions (i.e., low team autonomy) or the team as a whole makes decisions (i.e., high team autonomy; Hollenbeck et al., 2012) in order to more accurately understand the relationship between team members’ discretionary behavioral expectations of their teammates and sharing their team’s vision because team autonomy influences the degree to which team members focus on the activities of other team members.
Investigating the relationships between team members’ discretionary behavioral expectations of their teammates and sharing their team’s vision, and how both trust and team autonomy may influence these respective relationships, has the potential to make several significant contributions to theory and practice. First, many models of group development vary in their degree of generalizability and focus (Chang et al., 2006). This study attempts to contribute to the group development literature by taking a more generalizable approach to understanding how factors that are common to most teams and their members impact the degree to which team members share the same vision as their teams. Second, Phillips et al. (2014) argued that there has been much more of a focus on how individuals visually perceive and judge other individuals versus groups. This study adopts a people perception approach (Phillips et al., 2014) by focusing on how team members perceive their teammates’ discretionary behaviors as a basis for discretionary behavioral expectations of their team as a whole and judging whether they want to share the same vision as their team based on those expectations. Third, this study offers a more balanced view of trust by demonstrating its potential “bright” and “dark” sides (Gargiulo & Ertug, 2006; Skinner et al., 2014). Fourth, although Ryu et al.’s (2022) meta-analysis of team autonomy highlights its “bright” side, this study suggests team autonomy might also have a “dark” side.
Theory and Hypotheses
Team Members’ Discretionary Behavioral Expectations and Sharing Their Team’s Vision
Social information processing theory (Salancik & Pfeffer, 1978) posits that individuals pay particular attention to their immediate social environment and that cues from their immediate social environment influence their attitudes. Subsequent research (e.g., Robinson & O'Leary-Kelly, 1998) suggests that the perceived cues from the immediate social environment can separately influence the affective, behavioral, and cognitive components of an attitude (Rosenberg & Hovland, 1960). Within an organizational environment, employees are frequently engaged in social exchanges with their organizations (e.g., Zagenczyk et al., 2021), supervisors (e.g., Aryee et al., 2007), co-workers (e.g., Cruz et al., 2022), and team members (e.g., Seers et al., 1995). Employees also observe the social exchanges of others within the organization (e.g., Mitchell et al., 2015; Tang et al., 2022) and many employees focus on discretionary social exchanges, such as discretionary performance (Dalal et al., 2009; Rotundo & Sackett, 2002). Discretionary performance can consist of behaviors that are positive, which are known as citizenship behaviors (Bateman & Organ, 1983), or behaviors that are negative, which are known as deviant behaviors (Robinson & Bennett, 1995). Irrespective of whether discretionary behaviors are positive or negative, they can be directed toward individuals or organizations, but employees focus more on discretionary behaviors directed toward organizations (Dalal et al., 2009).
Organizational citizenship behaviors directed toward the organization (OCBO; Lee & Allen, 2002) are likely to be salient social cues within a team environment because OCBO can be reflective of positive social exchange relationships team members have with their organization (Kurtessis et al., 2017). In contrast, deviant behaviors directed toward the organization (OD; Robinson & Bennett, 1995) are likely to be salient social cues within a team environment because OD can be reflective of negative social exchange relationships team members have with their organization (Mackey et al., 2021). Although team members may be less likely to see all the negative discretionary behaviors of their teammates toward the organization (Carpenter et al., 2017; Cruz, 2022), team members monitor their teammates’ performance (Rousseau et al., 2006; Salas et al., 2009) and are likely to be especially observant of such behaviors if they do occur since individuals tend to focus on negative events more than positive events (Baumeister et al., 2001). Thus, this study follows others (e.g., Schabram et al., 2018) by focusing on team members’ perceptions of their teammates’ discretionary behaviors toward the organization.
Team members’ perceptions of their teammates’ positive (OCBO) and negative (OD) discretionary behaviors toward the organization are very likely to be compared to social exchange norms within the organization. Norms are informal rules or policies that help guide behaviors and are often formed based on explicit statements about behaviors, past events that have occurred, initial behaviors within groups, and/or behaviors from the past (Feldman, 1984). Assuming the organization is not an organization of corrupt individuals or a corrupt organization (Pinto et al., 2008), social exchange norms that exist within many organizations focus on promoting behaviors that do not harm the organization, but rather, help the organization (Feldman, 1984). The legitimacy of these norms influences individuals to voluntarily follow the norms (Tyler, 2006) and these social exchange norms therefore serve as an important social comparison of team members’ OCBO and OD expectations of their teammates.
Individuals typically want to develop and maintain relationships with social exchange partners who provide value and social exchange partners are avoided who engage in behaviors that threaten value (Kurzban & Leary, 2001). These social exchange partners are not just individuals, but groups or teams as well. For example, teams that engage in deviance toward organizations are associated with less trust in the team by the members of those teams (Schabram et al., 2018). This is not to say that all forms of behavior that deviate from norms are perceived negatively. Deviating from less important norms is tolerated more so than deviating from more important norms (Schein, 1988) and if deviating from norms helps accomplish tasks (Wheelan, 2013). However, individuals expect other individuals to engage in similar behaviors as they would (Ladbury & Hinsz, 2018) and the lower rates of deviant behaviors (Greco et al., 2015) compared to the much higher rates found in many studies of citizenship behaviors (e.g., Cruz & Pinto, 2019) suggest that typical social exchange behaviors are to engage in more citizenship and less deviance. Given that norms impact judgments (Asch, 1956) and influence shared intentions (Bagozzi & Lee, 2002), team members who do and do not engage in similar behaviors can serve as anchoring social cues for expectations of future behaviors and whether team members perceive their teams as valuable social exchange partners (Ballinger & Rockmann, 2010).
These anchoring social cues are important because whether team members share the same cognitions can be based on such things as common experiences, interactions, and comparisons (Tindale et al., 2001). It is therefore not surprising that Pearce and Ensley (2004) found that teamwork behaviors are associated with sharing a vision. However, whether those behaviors are positive or negative and the positive or negative behavioral expectations that result may have a differential impact on sharing a team vision. Team members’ OCBO expectations of their teammates, based on prior social cues of similar behaviors, may lead team members to perceive their team as being a more valuable social exchange partner (Cropanzano et al., 2017) because citizenship behaviors can increase social value (Bolino et al., 2002) and organizations can reward those teams accordingly. These positive aspects of the team as a social exchange partner can lead team members to engage in active and desirable responses toward the team (Cropanzano et al., 2017), like sharing the team’s vision. In contrast, team members who expect their teammates to engage in deviant behaviors toward the organization, based on prior social cues of similar behaviors, may see their team as a less valuable social exchange partner and (in)actively engage in undesirable responses toward the team (Cropanzano et al., 2017), such as less sharing of their team’s vision. Indeed, research suggests that deviant behaviors can hurt social exchange relationships (Welsh et al., 2022) and lower goal commitment (Aubé & Rousseau, 2011; Whiteoak, 2007). I therefore hypothesize:
Team members’ OCBO expectations of their teammates is positively associated with team members sharing their team’s vision.
Team members’ OD expectations of their teammates is negatively associated with team members sharing their team’s vision.
The Moderating Effects of Trust
Trust is central to both social exchange (Colquitt et al., 2011; Cropanzano et al., 2017; Fulmer & Gelfand, 2012) and social value (Coleman, 1988; Putnam, 2000; Tsai & Ghoshal, 1998). Trust is generally argued to represent a focal party’s perception that another party has or will engage in behaviors towards the focal party or others in such a way that the behaviors were or will be beneficial (e.g., Lewicki & Bunker, 1996; Mayer et al., 1995). This perception is critical because it allows the trustor to be vulnerable to the trustee (Mayer et al., 1995). Trust has been further broken down into cognition-based trust and affective-based trust with the former indicative of reliability and dependability and the latter indicative of care and concern (McAllister, 1995). Employees can trust other employees, their teams, and the organization itself (Fulmer & Gelfand, 2012), with trust within teams being especially critical for work relationships (Costa et al., 2018).
Trust in its general form, as well as in its cognitive- and affective-based forms, has been found to be a key variable for strengthening the beneficial aspects and weakening the detrimental aspects of teams (e.g., Brahm & Kunze, 2012; Cheung et al., 2016; Parayitam & Dooley, 2007). Trust should have a similar beneficial influence on the positive relationship between OCBO expectations of teammates and sharing a team’s vision. Predictability is related to trust, but they are not the same because a party can be predictable yet untrustworthy if that party predictably behaves in a way that can be negative for the focal party (Mayer et al., 1995). The positive valence of OCBO expectations (Bateman & Organ, 1983) should lead team members with higher trust to be more cognizant of the perceived predictability of positive social exchange behaviors of their team members toward the organization, and correspondingly, more strongly recognize how those expectations can potentially be beneficial to their team, because individuals are motivated to view relationship partners positively (Simpson, 2007) and higher trust leads individuals to focus more on the positive aspects of social relationships (Rempel et al., 2001). Thus, team members’ perceptions of the social value of the team should be even stronger and team members should therefore share their team’s vision to an even stronger degree. In contrast, team members with lower trust in their teammates may focus less on the predictability of their teammates’ positive social exchanges with the organization and how those positive social exchanges are potentially valuable to them and the team because individuals with less trust focus less on the positive aspects of social relationships (Rempel et al., 2001). I therefore hypothesize:
Trust moderates the positive relationship between team members’ OCBO expectations of their teammates and sharing their team’s vision such that the positive relationship is stronger when trust is higher and weaker when trust is lower. Although individuals tend to focus on negative events more than positive events, the fact that individuals with higher levels of trust view relationships more positively (Simpson, 2007) and focus more on the positive aspects of social relationships (Rempel et al., 2001) suggests that higher levels of trust may “buffer” the negative effect of team members’ OD expectations of their teammates on sharing their team’s vision. Indeed, individuals in close relationships may give each other the benefit of the doubt (Faulkner, 2018) and team members may be more likely to give their teammates the benefit of the doubt because team members are often more attached to their teams than their organizations (Riketta & Van Dick, 2005). This may be especially true if team members’ OD expectations are based more on minor forms of OD (Robinson & Bennett, 1995) or if the behaviors benefit the team and its members (Warren, 2003). However, it could also be because of dynamics within the team. Schabram et al. (2018) found that members of teams engaging in coordinated OD trust their teams just as much as members of honest teams. These streams of research suggest that team members are more forgiving of their teammates if they do expect their teammates to engage in negative social exchanges with their organization, such as OD expectations, and this may be especially true when they are complicit with the social exchange behaviors or believe the behaviors are benefiting the team in some way. Thus, team members will see their teams as more socially valuable social exchange partners and therefore more likely to engage active and desirable responses toward the team (Cropanzano et al., 2017), such as sharing their team’s vision. In contrast, team members who do not trust their teammates as much will not be as forgiving or complicit and the social value of the team will further decrease because of higher expectations of OD. I therefore hypothesize:
Trust moderates the negative relationship between team members’ OD expectations of their teammates and sharing their team’s vision such that the negative relationship is weaker when trust is higher and stronger when trust is lower.
The Moderating Effects of Team Autonomy
Johns (2006; 2018) argued that context plays an important role in understanding organizational behavior. Similarly, Cropanzano et al. (2017) argued that understanding the context is important for more accurately predicting the outcomes of social exchange and Bechky (2003) argued that this is particularly true for sharing, which is an important outcome of social exchange. One of the most important contextual features in the workplace is the degree to which employees are provided autonomy (Hackman & Oldham, 1975, 1976) and autonomy is important for sharing a vision (Salberg et al., 2019). Just as with individuals, one of the most important contextual features of teams is the degree to which teams are provided autonomy (Hollenbeck et al., 2012) because autonomy is important for teams to conduct their work (Wheelan, 2013). The process of deciding how much autonomy should be provided to teams involves making decisions about multiple aspects of team autonomy, such as how much say teams should have in how they complete their work and how much responsibility teams should have for their products or services (Cruz & Pil, 2011). The leaders or managers of teams are often those responsible for making such decisions (Hall, February 12, 2023). It is therefore not surprising that leadership plays an important role in sharing a vision (Berson et al., 2016), especially within teams (Berson et al., 2015). For example, Miller (2014) found in family businesses that strong authority is negatively associated with a shared vision. Thus, the degree of autonomy team leaders or managers provide their teams is likely to influence the relationships between team members’ OCBO and OD expectations of their teammates and the degree to which team members share their team’s vision, respectively.
Prior research has found that autonomy is positively associated with positive social exchanges, such as engaging in various forms of citizenship behaviors (e.g., Park, 2016; Pattnaik & Sahoo, 2021; Yang et al., 2017). Carpenter et al.’s (2014) meta-analysis found that others in the workplace are quite cognizant of a focal employee’s citizenship behaviors and team members engaging in an increased level of OCBO, as a result of the increased team autonomy granted by team managers, gives their fellow team members increased opportunities to observe such behaviors, and subsequently, form stronger expectations of team members’ OCBO. Moreover, Ryu et al. (2022) found in their meta-analysis of team autonomy that team autonomy was positively associated with several task-focused team functioning constructs, including coordination, team monitoring, and norms. This suggests that team members’ OCBO expectations of their teammates are likely to be even more salient in a team environment characterized by higher autonomy because team members are more cognizant of how past behaviors and expectations formed from those behaviors help or hinder the team and its members from accomplishing team tasks. When behaviors and subsequent expectations are positive in nature, like with citizenship behaviors, they can enhance the perceived value of social exchange partners (Bolino et al., 2002). Thus, a higher level of team autonomy provided by team managers can lead team members’ OCBO expectations of their teammates to be even more salient and further increase the perceived value of their team as a social exchange partner, leading team members to engage in even more active and desirable responses toward the team (Cropanzano et al., 2017), like sharing the team’s vision. I therefore hypothesize:
Team autonomy moderates the positive relationship between team members’ OCBO expectations of their teammates and sharing their team’s vision such that the positive relationship is stronger when team autonomy is higher and weaker when team autonomy is lower. Just as with positive discretionary social exchanges, prior research has found that higher levels of autonomy can lead employees to engage in more negative discretionary social exchanges, like deviant behaviors. For example, Arthur (2011) found that a higher level of team autonomy was associated with a higher level of deviant behaviors directed toward individuals. However, autonomy has been found to be associated with organizational deviance too. Wilson et al. (2015) found that autonomy was positively associated with one form of organizational deviance: production deviance (Robinson & Bennett, 1995). Although team members are more likely to observe team members engaging in interpersonal deviance rather than OD (Carpenter et al., 2017), this research suggests that managers who provide their teams more autonomy can lead team members to engage in more OD and, subsequently, more opportunities for team members to observe and form even stronger expectations of their team members to engage in OD. This is problematic because team members perceive their teams as less socially valuable exchange partners when they perceive their teams engaging in more OD (Schabram et al., 2018) and may lead team members to (in)actively engage in undesirable responses toward the team (Cropanzano et al., 2017), such as less sharing of their team’s vision. In contrast, managers who grant their teams less autonomy can result in team members having fewer expectations of their teammates engaging in OD, causing team members to perceive their teams as more valuable social exchange partners and therefore more likely to share their team’s vision to a stronger degree. I therefore hypothesize:
Team autonomy moderates the negative relationship between team members’ OD expectations of their teammates and sharing their team’s vision such that the negative relationship is stronger when team autonomy is higher and weaker when team autonomy is lower. Figure 1 summarizes the hypothesized theoretical model.

Hypothesized theoretical model of team members’ OCBO and OD expectations of their teammates on team members sharing their team’s vision and the moderating effects of trust and team autonomy.
Method
Data and Sample
Many theoretical and empirical decisions must be made when determining the appropriate data and sample to test hypotheses. This is especially true for hypotheses concerning teams and their leaders because obtaining such data is relatively more difficult considering the sample size requirements for team studies (Kirkman & Chen, 2011). The increased difficulty of obtaining appropriate sample sizes for team studies can influence the degree to which team researchers can equally focus on establishing both internal and external validity. Spector (2019) argued that establishing internal validity is less of a concern under certain conditions, such as not knowing whether independent and dependent variables covary or when investigating a naturally occurring independent variable. Cruz (2021) further argued that external validity should be established before internal validity, and this sometimes necessitates a sacrifice of internal validity. Aligning with these arguments, cross-sectional data from team members and team managers across 13 organizations were collected in order to (1) provide an initial test of the aforementioned hypotheses and (2) ensure the results of these initial tests had a relatively large degree of external validity. The 13 organizations ranged in size from 9 to 5,000 employees (M = 526). 1 Two of the organizations were in manufacturing industries and the remaining 11 organizations were in services industries.
Representatives for these organizations (e.g., Chief Operating Officer, Director of Human Resources) were provided Cohen and Bailey’s (1997: 241) definition of a team, which 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 (e.g., business unit or the corporation), and who manage their relationships across organizational boundaries,” to identify teams in their respective organizations. Using this definition, organizational representatives granted access to 374 team members across 69 teams, and the respective managers of the 69 teams. Each team member was a member of only one team (i.e., team members were not cross-classified).
Paper surveys and pre-addressed postage-paid envelopes were mailed to organizational representatives to distribute to the aforementioned team members and managers. Organizational representatives were provided instructions in the survey package to ensure that the paper surveys were distributed to the correct team members and managers, and numerical coding was utilized to ensure surveys could be matched after they were mailed back to the principal investigator. Team members and managers were provided no financial incentives for participation in the study, but they could request the same aggregate results as their organizations were receiving from the principal investigator. To reduce potential effects of common method bias, Podsakoff et al.’s (2012) procedural remedies were followed, such as obtaining one of the moderating variables from a different source than the dependent variable and taking steps to increase psychological separation (e.g., including the measurement of other constructs in the survey). Team members and managers mailed their completed surveys directly back to the principal investigator in the pre-addressed postage-paid envelopes.
Responses were received from 246 team members (65.78% response rate) and 56 team managers (81.20% response rate). Team members who did not provide responses for gender (5), age (6), education (2), or team tenure (5) were removed from the sample. Missing values for multi-item scales followed the procedure used by Cruz and Pinto (2019). Following this procedure, one team member was removed for team trust and one for shared vision, one team member for shared vision had his/her missing value(s) replaced, seven team members for OCBO expectations had their missing values replaced, and five team members for OD expectations had their missing values replaced. A further 19 team members from 6 teams were removed because the managers of those teams did not return their surveys. Lastly, 10 teams were removed from the final sample because they did not have at least two respondents. As a result, the final sample consists of 197 team members across 46 teams and 13 organizations. The teams range in size from 2 to 19 members (M = 5), with a mean within-team response rate of 78.94%.
Team Member Reported Measures
Following Cruz’s (2022) decision tree of when to use self- and other-reports, shared vision, OCBO expectations, OD expectations, and trust were all measured using self-reports of team members because (1) the variables are intra-individual and (2) the theory for each of these variables was written from the perspective of team members. Team member self-reports were also used to measure gender, age, education, team tenure, and team identification.
Shared Vision
Given the theoretical focus of this study to be as generalizable to as many teams as possible and the corresponding empirical strategy to collect data from teams in numerous organizations, a measure of shared vision had to be broad enough to apply to many teams. Shared vision was therefore measured by asking team members the degree to which they shared their teams’ visions using an adapted version of Leana and Pil’s (2006) six-item scale of shared vision (1 = strongly disagree to 5 = strongly agree). The adaptations were changing the wording from the third-person to the first-person and changing the referent from the school to the team. Sample items are “I share the same ambitions and vision for my team as my team members” and “I am in total agreement on my team’s vision.” Cronbach’s alpha for this scale is .85.
OCBO Expectations
OCBO expectations were measured by asking team members to indicate, based on prior experience with their team members, the extent to which they expected their team members to engage in organizational citizenship behaviors directed toward the organization in the future using an adapted version of Lee and Allen’s (2002) eight-item scale of OCBO (1 = never to 5 = all of the time). The adaptations were using a five-point scale rather than a seven-point scale, changing the referent from a focal individual to all team members to correspond to the theoretical focus of the team as a whole (following the approach of others, such as Aubé & Rousseau, 2011; Schabram et al., 2018), and instructing team members to use their prior experience with their team members as a basis for their expectations of their team members’ future behaviors. Sample items are “Attend functions that are not required but that help the organizational image” and “Defend the organization when other employees criticize it”. Cronbach’s alpha for this scale is .88.
OD Expectations
OD expectations were measured by asking team members to indicate, based on prior experience with their team members, the extent to which they expected their team members to engage in deviant behaviors directed toward the organization in the future using an adapted version of Bennett and Robinson’s (2000) 12 item scale of organizational deviance (1 = never to 5 = all of the time). The adaptations were using a five-point scale rather than a seven-point scale, changing the time referent of the scale to be symmetric with the OCBO expectations scale, changing the referent from a focal individual to all team members to correspond to the theoretical focus of the team as a whole (following the approach of others, such as Aubé & Rousseau, 2011; Schabram et al., 2018), and instructing team members to use their prior experience with their team members as a basis for their expectations of their team members’ future behaviors. Sample items are “Take additional or longer breaks than is acceptable at your workplace” and “Neglect to follow the boss’s instructions”. Cronbach’s alpha for this scale is .79. Because the responses did not exhibit normality, Cruz and Pinto’s (2019) transformation of subtracting the reciprocal responses from one was followed to make the responses as close to normal as possible, while maintaining the property of lower values indicating lower OD expectations and higher values indicating higher OD expectations.
Trust
Trust was measured using an adapted version of Jehn and Mannix’s (2001) three-item scale of trust (1 = strongly disagree to 5 = strongly agree). The adaptations were changing the Likert scale from a frequency scale to an agreement scale, using a five-point scale rather than a seven-point scale, changing the questions from question to statement form, and changing the referent from “group members” to “team members.” A sample item is “I trust my team members.” Cronbach’s alpha for this scale is .89.
Control Variables
Gender (0 = male, 1 = female), age (in years; 1 = less than 20 to 6 = 60+), education (1 = some high school to 6 = doctoral degree or equivalent), and team tenure (1 = less than 1 month to 6 = more than 2 years) were controlled for because prior research suggests that at least some demographic variables may be related to a shared vision (e.g., Tijunaitis et al., 2019). 2 Likewise, research suggests that the degree to which individuals identify with others may affect the degree to which individuals share the same cognitions as others (e.g., Postmes et al., 2005; Swaab et al., 2007). Identification was therefore measured using an adapted version of Van Der et al. (2003) four-item scale of team identification (1 = strongly disagree to 5 = strongly agree). A sample item is “I strongly identify with the other members of my team.” Cronbach’s alpha for this scale is .75.
Team Manager Reported Measure
Like with the team member reported measures, Cruz’s (2022) decision tree of when to use self- and other-reports was used to determine that team autonomy should be measured using team manager self-reports because although team autonomy can theoretically be observed by others (e.g., team members), the theory for team autonomy was written from the perspective of team managers. Thus, measuring team autonomy via team manager self-reports ensures the method matches the theory.
Team Autonomy
Similar to Langfred (2005), team autonomy was measured by asking team managers the degree to which they provided their teams autonomy using an adapted version of Breaugh’s (1985) nine-item work autonomy scale (1 = strongly disagree to 5 = strongly agree). The adaptations were using a five-point scale rather than a seven-point scale and changing the referent from the individual self to the team that is managed. Sample items are “I allow the team to decide how to go about getting its work done” and “I allow the team to have control over the sequencing of its activities.” Cronbach’s alpha for this scale is .78.
Organizational Representative Reported Measure
Control Variable
At the team-level, team size was controlled for using the aforementioned team lists provided by organizational representatives because research suggests that larger teams can have a negative impact on teams and their members (e.g., Aubé et al., 2011; Cruz et al., 2020). Like OD expectations, team size was not normal and team size was therefore transformed using its natural log.
Data Analysis
Hierarchical linear modeling was used to test the hypotheses because the data were nested (Raudenbush & Bryk, 2002; Snijders & Bosker, 2012). A random effects analysis of variance using a two-level model (i.e., individuals nested within teams) 3 found that there were significant differences between teams for shared vision, p = .007, with an ICC(1) value of .13. Following the recommendations of others for testing within-level and cross-level interactions in multilevel models (e.g., Aguinis et al., 2013; Enders & Tofighi, 2007; Hofmann & Gavin, 1998; Nezlek, 2012), group-mean centering was used for the individual-level predictors, except for the dichotomous gender variable and the interaction variables, and grand-mean centering was used for the team-level predictors.
Results
Descriptive Statistics and Correlations.
Notes. *p ≤ .05. **p ≤ .01. OD expectations and team size are transformed variables. Cronbach’s alphas are reported in parentheses.
Hierarchical Linear Modeling Results of Team Members’ OCBO Expectations, OD Expectations, Trust, and Team Autonomy, on Shared Vision.
Note. Individual-level N = 197. Team-level N = 46.
*p ≤ .05. **p ≤ .01.
Following the best practice recommendation of Aguinis et al. (2013) to test and report all cross-level interactions in the same model, Model 3 tests Hypotheses 3 – 6. Hypotheses 3 and 4 are concerned with the potential moderating effects of trust and trust is not found to significantly moderate the relationship between team members’ OCBO expectations of their teammates and team members sharing their team’s vision, γ = 0.15, SE = 0.14, p = .267. Thus, Hypothesis 3 is not supported. However, trust is found to significantly moderate the relationship between team members’ OD expectations of their teammates and team members sharing their team’s vision, γ = 0.96, SE = 0.26, p < .001. Hypothesis 4 is therefore supported. Using +/− one standard deviation for OD expectations and trust (Dawson, 2014), Figure 2 shows the form of this interaction using the full shared vision scale and Figure 3 shows the form of this interaction using the upper half of the shared vision scale. Preacher et al.’s (2006) computational tool for probing within-level interactions in a multilevel model was confirmed as an appropriate tool to further probe the significant within-level interaction in Model 3 (K. Preacher, personal communication, August 3, 2024). The simple slope for OD expectations is significant at one standard deviation below the mean for trust, γ = −0.52, SE = 0.20, p = .009, as well as at one standard deviation above the mean for trust, γ = 0.61, SE = 0.25, p = .012. Interaction between team members’ OD expectations of their teammates and trust on team members sharing their team’s vision using the minimum and maximum of the shared vision scale. Interaction between team members’ OD expectations of their teammates and trust on team members sharing their team’s vision using the midpoint and maximum of the shared vision scale.

Hypotheses 5 and 6 are concerned with the potential moderating effects of team autonomy and team autonomy is not found to significantly moderate the relationship between team members’ OCBO expectations of their teammates and team members sharing their team’s vision, γ = 0.11, SE = 0.10, p = .287. Hypothesis 5 is therefore not supported.
4
However, team autonomy is found to significantly moderate the relationship between team members’ OD expectations of their teammates and team members sharing their team’s vision, γ = −1.32, SE = 0.44, p = .003. Thus, Hypothesis 6 is supported. Using +/− one standard deviation for OD expectations and team autonomy (Dawson, 2014), Figure 4 shows the form of this interaction using the full shared vision scale and Figure 5 shows the form of this interaction using the upper half of the shared vision scale. Preacher et al.’s (2006) computational tool for probing cross-level interactions in a multilevel model was confirmed as an appropriate tool to further probe the significant cross-level interaction in Model 3 (K. Preacher, personal communication, August 3, 2024). The simple slope for OD expectations is significant at one standard deviation below the mean for team autonomy, γ = 0.81, SE = 0.31, p = .009, as well as one standard deviation above the mean for team autonomy, γ = −0.71, SE = 0.29, p = .014. Interaction between team members’ OD expectations of their teammates and team autonomy on team members sharing their team’s vision using the minimum and maximum of the shared vision scale. Interaction between team members’ OD expectations of their teammates and team autonomy on team members sharing their team’s vision using the midpoint and maximum of the shared vision scale.

Discussion
This study adds to our very limited knowledge base about shared vision (Matsuo, 2023), and the antecedents of sharing a vision in particular (Boyatzis et al., 2015). Moving beyond the adoption of a relatively narrow focus in examining the antecedents of sharing a vision (e.g., Preston & Karahanna, 2009; Zasa et al., 2022), I adopted a more generalizable approach to examining the antecedents of a shared vision by focusing on three characteristics common across most teams: expectations (Salas et al., 2007), trust (Salas et al., 2005), and autonomy (Hollenbeck et al., 2012). Following the basic tenets of social information processing theory (Salancik & Pfeffer, 1978) and theories of social exchange (Cropanzano et al., 2017), I found that team members’ OCBO expectations of their teammates are indeed positively associated with sharing their team’s vision. However, I found that team members’ OD expectations of their teammates was not associated with sharing their team’s vision, lending credence to the arguments that theories of social exchange are more precise when both individual and environmental factors are considered (Cropanzano et al., 2017) and this is especially true within teams (Salas et al., 2007). Although trust and autonomy were not found to be important for understanding the relationship between team members’ OCBO expectations of their teammates and sharing their team’s vision, they were found to be important for understanding the relationship between team members’ OD expectations of their teammates and sharing their team’s vision. The results suggest that the benefits of trust (Rempel et al., 2001; Simpson, 2007) indeed “buffer” the negative effect of team members’ OD expectations of their teammates on sharing their team’s vision. In contrast, the results build on prior work that suggests teams who engage in more OD are perceived as less valuable social exchange partners (e.g., Schabram et al., 2018) by demonstrating that higher team autonomy “amplifies” the negative effect of team members’ OD expectations of their teammates on sharing their team’s vision. Accordingly, I believe these results have a variety of important theoretical and practical implications.
Theoretical Implications
Many group development models that have been presented in the literature are not very generalizable and often focus on limited aspects of teams (Chang et al., 2006). The best theoretical models of groups and teams are those that can apply to as many groups and teams as possible. Shared cognition can be considered an emergent state within teams (Kozlowski & Chao, 2012), but not all forms of shared cognition need to necessarily develop within teams. For example, teams with low skill differentiation, or teams in which team members are easily substitutable (Hollenbeck et al., 2012), do not necessarily need to develop transactive memory systems, or systems that allow team members to know who knows what within a team (Wegner, 1986), because individual team members have largely the same knowledge as each other. In contrast, shared vision has been identified as an important component of effectiveness in virtually all teams (Salas et al., 2007), and as previously mentioned, has been associated with many different team outcomes. Likewise, behavioral expectations exist in virtually all teams and are also associated with team effectiveness (Salas et al., 2007), trust is critical in virtually all teams (Costa et al., 2018), and decisions must be made for most teams about how autonomy should be provided to teams (Hollenbeck et al., 2012). This study demonstrates that there are important opportunities to develop more generalizable models of group development than currently exists in the literature. Rather than focusing on specific types of teams, or variables that are germane to those specific types of teams, a more fruitful approach to significantly moving theory about groups and teams forward is to take a theoretical and empirical approach that places much more value on generalizability (Cruz, 2021). Thus, researchers should broaden their theoretical focus on groups and teams moving forward, similar to the approach used in this study, in order to increase our theoretical understanding of all groups and teams that exist in organizational settings.
Although researchers should broaden their focus when investigating organizational groups and teams, the results of this study suggest that not all discretionary behavioral expectations of teammates that exists within teams are directly associated with sharing a team’s vision. Unlike prior research that has placed a greater focus on how individuals visually perceive and judge other individuals (Phillips et al., 2014), this study adopted an approach similar to others by focusing on how individual team members perceived their teams as a whole rather than specific individuals within those teams (e.g., Aubé & Rousseau, 2011; Schabram et al., 2018). Adopting this “people perception” approach (Phillips et al., 2014) provided an opportunity to theorize and test whether team members’ expectations of their teammates positive or negative discretionary behaviors of the team as a whole played a significant role in whether they shared their team’s vision, which is also about the team as a whole, rather than about specific members of the team. In contrast to Baumeister et al.’s (2001) argument that negatives are more salient than positives, the results of this study suggest that there are times in which a “people approach” leads to positives being more salient than negatives. These results are more support for questioning the utility of social exchange (Cropanzano et al., 2017) because social exchange would argue that team members should share their team’s vision less because of expecting team members to engage in negative behaviors. Similar to Cruz et al.’s (2022) different findings regarding interpersonal citizenship and deviance leading them to speculate that different theoretical processes were operating for citizenship and deviance, the results of this study suggest that social exchange either (1) cannot be used as a universal theoretical explanation for the relationship between discretionary behavioral expectations and sharing a vision and/or (2) that other factors indeed impact the accuracy of social exchange predictions (Cropanzano et al., 2017). In other words, theories of social exchange are not sufficient alone for explaining at least some phenomena within teams when a “people approach” is adopted.
Rather, the results of this study lead credence to the arguments of Cropanzano et al. (2017) that other factors need to be considered in conjunction within social exchange when attempting to better predict the outcomes of social exchange. Within a team environment, trust should not be ignored because it is foundational to both teams (Costa et al., 2018) and social exchange (Cropanzano et al., 2017). Interestingly, the results of this study suggest that a team members’ levels of trust in their teammates have no bearing on the relationship between OCBO expectations and sharing a team’s vision but does have a bearing on the relationship between OD expectations and sharing a team’s vision. These findings suggest that trust is more important when considering negative behaviors than positive behaviors and their subsequent expectations, which makes sense because positive behaviors and the subsequent expectations should not lead individuals to question how vulnerable they should make themselves to those engaging in positive behaviors or are expected to engage in positive behaviors. However, these results do lend support to Brown et al.’s (2014) argument that trust is a “double-edged sword” when it comes to negative behaviors and their subsequent expectations. In this study, trust is positive in the sense that it potentially allows team members to reframe their OD expectations of their teammates in such a way that their OD expectations are not as viewed as negatively. The problem is that at the same time, from the organization’s perspective, this same exact phenomenon can be viewed negatively because many organizations would not want their employees to reframe their negative perspectives of negative behaviors and their subsequent expectations. Thus, trust has both a “bright” and “dark” side when it comes to behavioral expectations and future trust research should therefore pay closer attention to how trust is a double-edged sword, rather than the traditional approach of treating trust as universally positive (Gargiulo & Ertug, 2006).
Trust is not the only variable that is potentially a “double-edged sword” when it comes to more precisely predicting the outcomes of social exchange. Although Ryu et al.’s (2022) meta-analysis of team autonomy found that team autonomy was beneficial in reducing negative team member behaviors, the result of this study suggests that team autonomy is not beneficial in terms of team members’ OD expectations of their teammates. Rather, the result of this study suggests that team members focus even more on these negative discretionary behavioral expectations and do not focus on positive discretionary behavioral expectations (at least in terms of OCBO). This study therefore adds to the “dark” side of autonomy that has been found in prior research. For example, Langfred (2004) found that a combination of high trust and high individual autonomy was associated with lower team performance. This study suggests that higher autonomy at the team-level has detrimental effects as well. I hope future research continues to build upon this burgeoning line of research on the “dark” side of autonomy so that there is a more theoretically balanced perspective of team autonomy. This is especially true since Ryu et al.’s (2022) meta-analysis identified only four studies that examined team autonomy and team negative behaviors. An increased focus on fleshing out the potential “dark” side of team autonomy can provide a stronger theoretical foundation of when autonomy may be beneficial or detrimental to various parties involved in grating and receiving autonomy.
Practical Implications
The results of this study also suggest a variety of practical implications for organizations. Since OCBO expectations were found to be positively associated with sharing a team’s vision, organizations should focus on motivating employees to engage in OCBO so that employees’ team members have more opportunities to observe and form stronger expectations of their teammates to engage in OCBO. Prior research suggests that organizations should focus specifically on the positive and negative social exchange relationships team members perceive to have with their organizations since employees typically target their behaviors towards the source of social exchanges (Lavelle et al., 2007). Kurtessis et al. (2017) found in their meta-analysis that perceived organizational support, or degree to which employees believe that their organizations care about their contributions and their well-being (Eisenberger et al., 1986), was positively associated OCBO. In contrast, Zhao et al. (2007) found in their meta-analysis that psychological contract breach, or the degree to which employees believe that their organizations have broken perceived promises to them (Rousseau, 1989), is negatively associated with organizational citizenship behaviors. Organizations can therefore encourage employees to engage in OCBO by having more positive and fewer social exchanges with their employees, thereby providing team members more opportunities to observe their teammates’ OCBO and subsequently form stronger expectations of their teammates’ future OCBO.
In contrast, OD expectations were found to be significantly related to sharing a team’s vision only under certain conditions. Assuming deviant behaviors continue as they have in the past (Pearson & Porath, 2009; Sutton, 2007), it is likely that team members will have at least a minimal level of OD expectations of their teammates. Although organizations should take actions that can potentially decrease deviant behaviors, such as setting zero-tolerance expectations (Pearson & Porath, 2005), the results of this study suggest that a more fruitful path forward for organizations is to focus more on trust and team autonomy when it comes to sharing a team’s vision because they do not influence the relationship between OCBO expectations and sharing a team’s vision, but they do influence the relationship between OD expectations and sharing a team’s vision. Research suggests that trust is generally beneficial for teams (Costa et al., 2018) and this holds true for how OD expectations influence sharing a team’s vision. Organizations should therefore focus on those factors that encourage trust at both the team (e.g., composition)- and individual (e.g., propensity to trust)- levels (Costa et al., 2018). Likewise, the results of this study suggest that organizations should take actions to reduce team autonomy, which is in line with prior research supporting providing teams lower autonomy (e.g., Cruz & Pil, 2011). This could be accomplished by requiring team managers to set work schedules, requiring team managers to prescribe how tasks will be accomplished, and requiring team managers making more decisions for the team (Ryu et al., 2022).
Limitations and Directions for Future Research
The theoretical and practical implications of this study should be considered in light of the theoretical and empirical limitations of this study. Although shared cognition can emerge within teams (Kozlowski & Chao, 2012), shared vision was not theoretically or empirically treated as an emergent construct within this study. Rather, it was viewed as already existing within teams and empirically treated as an outcome. While this is in line with the fact that constructs can be treated as inputs, mediators, or outcomes because of the cyclical nature of teams and the fact that the connections between inputs, mediators, and outcomes may be nonlinear and/or conditional (Ilgen et al., 2005), this treatment of shared vision ignores the fact that these teams may be in different stages of development, which would affect the degree to which a vision exists within teams. For example, teams in early stages of development may not have a clear vision (Wheelan, 2013). Defining and measuring team cognition is fraught with problems (Mohammed et al., 2021). However, future research should attempt to add to our limited knowledge base about the development of shared vision (Boyatzis et al., 2015) by theoretically and empirically focusing on its emergent processes.
Similarly, it may take some time for team members to develop expectations of their teammates’ discretionary behaviors (Feldman, 1984), yet team members were assumed to have already developed these expectations. While the results not being substantively different when including team tenure as a control variable may help reduce this as a concern, it does not negate the possibility that, like with shared vision, an important boundary condition of these results may be the team’s state of group development. Teams who are further along in the development process are very likely to have stronger expectations of their teammates’ discretionary behaviors than teams who are in the early stages of the development process. Likewise, other elements of the team and its members are likely to impact the development of emergent states (Mohammed et al., 2010), such as shared vision. Examples include features of the team, such as skill differentiation, temporal stability, and virtual presence (Hollenbeck et al., 2012) or features of the individual team members, such as self-construal since it can affect behavior and cognitive conformity (Täuber & Sassenberg, 2012). I therefore encourage future research to build upon the foundational model presented in this study now that a relatively large degree of external validity has been established (Cruz, 2021).
An important direction to build upon this foundational model is to focus on behavioral expectations beyond just those toward the organization. Team members can develop expectations of their team members’ discretionary behaviors directed toward individuals. Importantly, the degree to which individual team members expect their teammates to engage in positive and negative discretionary behaviors toward their own team members versus individuals outside of the team may play an important role in the degree to which team members share the team’s vision given what we know about how individuals perceive in-(out-)group normative and deviant members (e.g., Marques et al., 1998). For example, the degree to which team members share the team’s vision may be very different when team members expect their teammates to help fellow team members who have been absent from work but not employees outside of the team who have been absent from work, vs. when team members expect their teammates to help fellow team members who have been absent from work but also publicly embarrass employees outside of the team at work. I therefore encourage future research to enhance the robustness of the theoretical model presented in this paper by exploring expectations of behavior beyond just those directed at the organization.
Footnotes
Acknowledgements
The publication of this manuscript represents the official closing (finally!) of a specific chapter in my Muppet man life and I therefore need to thank a few individuals and pets that were instrumental in this chapter of my Muppet man life. I thank my wife, Kristin, for always being there by my side through the many ups and downs of this chapter. I also thank our pets, Marty (our cat) and Jack and Timone (our dogs), for always being a respite during this chapter. I owe more thanks than I can give to John Hulland because my career simply would not exist without him. Finally, I thank Marc Street for his constant support from my days as an undergraduate student all the way up to his way too early passing. This manuscript is dedicated to the memory of Marc.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This manuscript was supported by grants from University of Pittsburgh’s Katz Graduate School of Business and Georgia Southern University’s Parker College of Business.
