ArticlePDF Available

Setting the Tone for the Team: A Multi‐Level Analysis of Managerial Control, Peer Control, and their Consequences for Job Satisfaction and Team Performance



In this study, we develop a multi‐level theoretical framework linking antecedents and outcomes of peer control, defined as team members at the same hierarchical level noticing and responding to their peers’ behavior or performance. Analyzing multi‐level data from 356 volunteers and 58 regional teams in a non‐profit organization, we examine top‐down managerial controls as antecedents of lateral peer control, both directly (i.e., monitoring and responding directly to peers) and indirectly (i.e., gossiping about and avoiding underperforming peers), and peer control’s effects on individual‐ and team‐level outcomes. In line with our predictions, we find formal managerial control and clan control to be antecedents of peer control, albeit with differential effects on direct and indirect peer control. We also find a significant association between peer control and both individual‐level job satisfaction and team‐level performance, but again, with crucial differences between the two types of peer controls and the two outcomes. Our study contributes to the development of a better theoretical understanding of peer control, sheds light on inconsistent findings across prior studies, provides novel insights into how team leaders can influence team members’ individual satisfaction and team‐level performance via peer control, and reveals important trade‐offs with regards to peer control’s influence on individual‐ and team‐level outcomes.
Jorge Walter *
Department of Strategic Management and Public Policy
School of Business
The George Washington University
2201 G Street, NW, Funger Hall 615
Washington, DC 20052, USA
Phone: +1 (202) 994–7908, Fax: +1 (202) 994–8113
Markus Kreutzer
Management Group
EBS Business School
Gustav-Stresemann-Ring 3
65189 Wiesbaden, Germany
Phone: +49 (611) 7102–1413; Fax: +49 (611) 7102–10–1413
Karin Kreutzer
Management Group
EBS Business School
Gustav-Stresemann-Ring 3
65189 Wiesbaden, Germany
Phone: +49 (611) 7102–1412; Fax: +49 (611) 7102–10–1412
September 9, 2020
Forthcoming in the Journal of Management Studies
* All three authors contributed equally and are listed in reverse alphabetical order.
Acknowledgments: We thank the children’s rights non-profit organization for their support and all organi-
zational members for participating in our study. We also thank research seminar participants at the EBS
Business School (especially Marjo-Riitta Diehl, Christian Landau, and Klaus Uhlenbruck), Erasmus Uni-
versity Rotterdam, University of Bamberg, Technical University of Dortmund, Technical University of
Munich, and City University London, as well as Herman Aguinis for their helpful comments on earlier
versions of this paper. All remaining errors are our own.
In this study, we develop a multi-level theoretical framework linking antecedents and outcomes
of peer control, defined as team members at the same hierarchical level noticing and responding to their
peers’ behavior or performance. Analyzing multi-level data from 356 volunteers and 58 regional teams in
a non-profit organization, we examine top-down managerial controls as antecedents of lateral peer con-
trol, both directly (i.e., monitoring and responding directly to peers) and indirectly (i.e., gossiping about
and avoiding underperforming peers), and peer control’s effects on individual- and team-level outcomes.
In line with our predictions, we find formal managerial control and clan control to be antecedents of peer
control, albeit with differential effects on direct and indirect peer control. We also find a significant asso-
ciation between peer control and both individual-level job satisfaction and team-level performance, but
again, with crucial differences between the two types of peer controls and the two outcomes. Our study
contributes to the development of a better theoretical understanding of peer control, sheds light on incon-
sistent findings across prior studies, provides novel insights into how team leaders can influence team
members’ individual satisfaction and team-level performance via peer control, and reveals important
trade-offs with regards to peer control’s influence on individual- and team-level outcomes.
Key Words: Organizational control, peer control, formal and informal (clan) control, job satisfaction,
team performance, multi-level analysis, volunteers, non-profit organizations.
Organizational control addresses the fundamental managerial problem of aligning employee ac-
tivities with organizational goals (De Jong et al., 2014; Sitkin et al., 2010). Recent trends in organizations
toward flatter hierarchies and the increasing use of teams, electronic communication, and social media
(Colbert et al., 2016; Kirsch et al., 2010; Loughry, 2010) have brought organizational control back into
the spotlight (Cardinal et al., 2017). Contemporary organizational control work falls broadly into two lit-
erature streams. The first stream is focused on examining combinations of different types of top-down
managerial control, both formal (i.e., measuring and rewarding team members’ behavior or the outcomes
of this behavior) and informal (i.e., establishing shared norms, values, and vision to guide behaviors), and
their influence on performance outcomes (e.g., Cardinal et al., 2004; Cardinal et al., 2010; Kreutzer et al.,
2016; Kreutzer et al., 2015; Sihag and Rijsdijk, 2019). The second stream is focused on lateral peer con-
trol among people who are at the same organizational level and who thus have no formal authority over
one another (e.g., De Jong et al., 2014; De Klepper et al., 2016; Lange, 2008; Loughry, 2010; Loughry
and Tosi, 2008; Stewart et al., 2012).
Despite the important insights these studies have generated, how managerial and peer controls
influence each other, and what their individual- and team-level outcomes are, remain open questions for
which little theoretical or empirical guidance exists. While some studies have found that managerial and
peer controls can serve as substitutes for each other (e.g., Loughry and Tosi, 2008), others suggest that
managerial control choices serve as antecedents of peer control (De Jong et al., 2014; Kirsch et al., 2010;
Lange, 2008; Loughry, 2010). Moreover, peer control’s effect on team-level outcomes remains ambigu-
ous, with prior studies finding positive effects (Loughry and Tosi, 2008; Stewart et al., 2012), non-signifi-
cant effects (Loughry and Tosi, 2008), and contingent effects (Kohli and Jaworski, 1994; Welbourne and
Ferrante, 2008) of peer control on team performance. Peer control’s effects on individual-level outcomes,
such as job satisfaction, are similarly ambiguous (see, e.g., Kohli and Jaworski, 1994). Lastly, with one
notable exception (Stewart et al., 2012), prior peer control studies have focused on either the individual or
the team level, despite recurring calls for “more research on how control functions at different levels of
analysis and how its determinants and effects cross levels” (Sitkin et al., 2010, p. 13).
Complementing and extending both literature streams, we focus on peer controls and their ante-
cedents and consequences at the individual and team level. We examine formal managerial control and
clan control as antecedents of direct (i.e., monitoring and responding directly to peers) and indirect peer
control (i.e., gossiping about and avoiding underperforming peers). We further examine how peer control
affects individual satisfaction and team performance. We offer three contributions: First, by providing
novel insights into both its antecedents and outcomes, we contribute to a better theoretical understanding
of peer control, which is of increasing relevance in team-based and knowledge-intensive work environ-
ments where “a much greater emphasis is placed on control within rather than from outside the group”
(Manz and Sims, 1987, p. 107), and where “evaluation processes will need to evolve to include more
peer- and project-based reviews, as opposed to the lines of traditional reporting” (Beardsley et al., 2006,
p. 61).
Second, acknowledging Sewell’s (1998, p. 410) conclusion that “vertical surveillance […] cannot
be separated from the horizontal surveillance of peer group scrutiny,” this is one of the first studies that
simultaneously examine peer and managerial controls (see, Loughry and Tosi, 2008, for an exception). In
our theorizing, however, we complement and extend this study’s finding of peer and formal managerial
control as substitutes for each other in their performance effects by analyzing the role of managerial con-
trols as antecedents of peer control. We address the question of whether and to what extent organizational
leaders can influence peer control by appropriately designing managerial controls. We thus address “the
need for traditional control theory to be significantly—even radicallyrethought to account for new or-
ganizational forms, nonhierarchical sources of control; and the use of multifaceted control choices by
managers, peers, and others” (Sitkin et al., 2010, p. 13).
Third, by examining the effects of organizational control on both individual-level job satisfaction
and team-level performance, this is one of very few multi-level organizational control studies (see Stewart
et al., 2012, for a notable exception). We not only answer recurring calls to include individual-level
outcomes into organizational control research (Loughry and Tosi, 2008; Sitkin et al., 2010), our finding
that peer control has differential effects at the individual and team level further highlights the need to ex-
amine organizational control at different levels of analysis. Combining these findings with the antecedent
role of managerial control reveals important insights into how team leaders can influence their team mem-
bers’ job satisfaction and team performance via peer control, but also highlights a potential trade-off be-
tween achieving positive outcomes at the individual versus the team level: While team leaders have the
power to influence peer control within their teams with their choice of managerial controls, our findings
suggest that a focus on enhancing job satisfaction among team members can come at the expense of team
performance, and vice versa. Our multilevel theory and findings therefore help “illuminate the steps or-
ganizational actors may take, individually and collectively, to yield organizational benefits(Klein et al.,
1999, p. 243), and the trade-offs we uncover between levels add an important nuance to both individual-
level and team-level studies of organizational control.
We focus on team member-designed and -implemented peer control, which occurs laterally be-
tween employees with no formal authority over one another (De Jong et al., 2014; Loughry, 2010;
Loughry and Tosi, 2008; Stewart et al., 2012). Loughry and Tosi (2008, pp. 885-886) have shown that
peer control comprises two distinct dimensions: It can be direct when it “involves noticing peers’ behav-
ior or results and responding directly and openly, such as praising team members when they do a good
job, correcting team members when they make mistakes, reporting dishonest team members, and discuss-
ing how everyone does the job,” or indirect, “when workers gossip about or avoid poorly performing
peers.” Like other forms of control, peer control serves both a monitoring function by giving organiza-
tions better information about team members’ behavior and performance—thereby reducing opportunities
to engage in dysfunctional behavior—and an incentive function by motivating employees to engage in
behaviors beneficial for the organization (Loughry and Tosi, 2008).
Peer control is increasingly common across all types of organizations (Kirkman and Rosen, 1999;
Loughry, 2010). This proliferation can be attributed to the greater use of teams and self-managed work
groups, which has increased the importance of peers when it comes to directing and motivating work in
organizations (Loughry, 2010). The increasingly complex nature of team work, however, makes it more
difficult for team leaders to identify members’ individual contributions (Kirsch, 2004). Peers, on the other
hand, enjoy information advantages over their team leaders and can bring these to bear in a peer control
regime (Fama and Jensen, 1983; Welbourne and Ferrante, 2008). Moreover, email and social media pro-
vide more opportunities to influence large numbers of team members efficiently and in a timely manner,
thus further expanding peer control’s reach (Kirsch et al., 2010). And lastly, whereas economic and com-
petitive conditions might constrain organizations’ abilities to offer substantive monetary incentives, infor-
mal rewards and sanctions among team members are not only free but exert a meaningful influence on
team members’ motivations (Loughry, 2010).
In spite of peer control’s increasing importance in organizational practice and the recent surge in
academic interest (e.g., De Jong et al., 2014; Stewart et al., 2012), our theoretical understanding of this
topic is still limited, especially when it comes to antecedents and consequences. In the following section,
we will develop and test a theoretical framework linking direct and indirect peer control to its antecedents
and outcomes. For antecedents, we follow the traditional view of organizational controlrooted in organ-
ization and agency theory—which has long distinguished between two main types of control. Formal
managerial control relies either on the direct surveillance of employee behavior or on measuring the out-
comes of employee behavior, coupled with rewards for good and sanctions for inacceptable behavior and
performance (Eisenhardt, 1985; Ouchi, 1979; Ouchi and Maguire, 1975). Informal managerial control or
clan control emphasizes the role of shared norms, values, and vision in guiding and influencing behavior,
such that individual objectives become congruent with the organization’s goals (Kirsch, 1996; Ouchi,
1979). While adherence to such shared norms and values is enforced by both team leaders and peers
(Kirsch, 2004), many important parts of clan control, such as socialization using rituals and ceremonies,
are under the control of hierarchical management in most organizations (Loughry, 2010). That is, team
leaders facilitate and institutionalize an organization’s clan control (Kirsch et al., 2010; Turner and
Makhija, 2006), which is why we subsumed it under top-down managerial control for our study. Our
focus on the two most prominent managerial control types in the literature (see Cardinal et al., 2017;
Sihag and Rijsdijk, 2019, for recent reviews) allows us to examine the extent to which team leaders can
influence peer control by designing their organization’s managerial control regimes. Each type represents
a different focus of organizational control—i.e., surveillance of behavior/outcomes versus socialization,
each has its own advantages and disadvantages (e.g., Kreutzer et al., 2015), and each has been deemed
more or less appropriate for different organizational settings (e.g., Eisenhardt, 1985). These differences
would suggest different theoretical logics and mechanisms responsible for their effects on peer control.
Regarding the outcomes of peer control, we acknowledge the multi-level nature of organizational
control (Cardinal et al., 2017; Loughry and Tosi, 2008; Sitkin et al., 2010; Stewart et al., 2012). At the
individual level, we focus on job satisfaction, defined as “a pleasurable or positive emotional state result-
ing from the appraisal of one’s job or job experiences” (Locke, 1976, p. 1304). Besides being a crucial
outcome in and of itself, job satisfaction is also an important determinant of people’s intentions to remain
at their current organization and has been identified as particularly relevant in the context of peer control
(Boezeman and Ellemers, 2009; Loughry, 2010). At the team level, we follow recent work on peer control
(De Jong et al., 2014; Loughry and Tosi, 2008; Stewart et al., 2012) and focus on team performance, de-
fined as the extent to which a team’s productive output meets, or exceeds, the performance standards of
those who review and/or receive the output (Hackman, 1987).
Managerial Control as Antecedent of Peer Control
Classic work on the antecedents of organizational control has focused on the observability of con-
trolees’ behaviors—i.e., understanding the process through which inputs are transformed into outputs—
and on the measurability of the outcomes of this process (Eisenhardt, 1985; Ouchi, 1979). As means-ends
relationships become more ambiguous, process or behavior control becomes less effective; and as the reli-
ability and validity of outcome measures decrease, outcome control is deemed infeasible. If neither be-
havior nor outcomes are observable, informal or clan control becomes the only feasible control choice
(Ouchi, 1979). As a result, these early scholars have generally advocated for one type of control over the
other, dependent on the characteristics of the tasks people perform. More recent work has argued,
however, that this focus on task characteristics only partially explains the use and effectiveness of differ-
ent control types (Cardinal et al., 2017). Control is an inherently social or team phenomenon, and thus the
teams’ social context plays a role. While peer control is ultimately exercised by the social interactions be-
tween team members, team leaders can influence it (e.g., Barker, 1993; Lange, 2008; Loughry, 2010):
They can “initiate peer controls by prescribing norms of appropriate behavior for their teams, authorize
peer control by delegating control responsibilities to their teams, or facilitate peer control opportunities by
restructuring team work patterns” (De Jong et al., 2014, p. 1704, emphases in original). Peer control is
therefore an outcome of managerial choices, and “[t]he role of team members primarily lies in the subse-
quent stages and involves accepting, taking part in, and maintaining peer control once it is in place” (De
Jong et al., 2014, p. 1704). Building on this line of reasoning, we examine whether and to what extent
peer control is influenced by the organizational and social context set by formal managerial and clan con-
Formal Managerial Control. By examining formal managerial control as an antecedent of peer
control, we deviate from prior research’s predominant conceptualization of peer and managerial controls
as substitutes for each other (e.g., Fama and Jensen, 1983). In line with this substitution logic, Loughry
and Tosi (2008) found support for their hypothesis that peer and formal managerial controls are equifinal
in their effects on problem-free performance. According to this logic, combining formal managerial con-
trol with peer control would offer few additional benefits for providing constraints, deterring opportun-
ism, and offering information.
There is an equally plausible alternative that prior empirical research has largely overlooked,
however, namely formal managerial control acting as an antecedent of peer control. In support of this al-
ternative view, Loughry (2010) argues that “supervisors’ attitudes are likely to influence the types and
amounts of peer control that emerge in work groups,” and prior research has found a strong association
between managerial and peer control (Loughry and Tosi, 2008). Other work has shown that team mem-
bers tend to mimic their team leader’s behaviors, at least to a certain extent. In particular, building on so-
cial learning theory (Badura, 1977), we expect that team leaders exercising formal managerial control will
be regarded as role models by team members and will influence whether and how they engage in peer
With regards to direct peer control, when team leaders specify procedures, work assignments, and
concrete outcome goals, they also focus team members’ attention on these issues (Ocasio and
Wohlgezogen, 2010), thereby inducing the team members to follow their example. This focus likely trig-
gers discussions among team members about how work gets done within the team, and how to improve
these processes, thereby serving as a catalyst for direct peer control. In support of this argument, Hughes
(2004, p. 285) found in his study of new public sector workers that “while they were not directly involved
in the participants’ learning, the supervisors’ influence on learning was nevertheless extensive. It was
through the routine formal managerial functions of delegating tasks, setting expectations, requiring ac-
countability and providing feedback on work performed.” Similarly, in their study of two Fortune 500
companies’ salespeople, Kohli and colleagues (1998) found that formal managerial controls, by devoting
increased attention to and providing feedback on task-oriented and performance-oriented behaviors, en-
courage employees to engage in learning and improvement efforts. The increased learning orientation, in
turn, provides a fertile ground for peers to interact, observe each other’s behaviors and their success or
failure, and openly discuss how work gets done—in other words, for direct peer control.
Moreover, formal managerial control tends to entail a high level of transparency; the procedures
to be followed and outcomes to be achieved are explicitly defined, communicated to each team member,
and subsequently monitored and incentivized. Such transparency helps reduce ambiguity around team
leaders’ procedural and outcome expectations for the team and promote the same values throughout the
team. In a corporate context, for example, Kownatzki et al. (2013) found that more transparent decision
processes aligned corporate and strategic business unit-level interests and fast-tracked decision processes.
Since team leaders serve as role models, team members may abstain from indirect peer control, which, by
definition, is not transparent and open, but happens behind people’s backs (Loughry and Tosi, 2008). We
therefore propose:
Hypothesis 1: Formal managerial control is (a) positively related to direct peer control and (b)
negatively related to indirect peer control.
Clan Control. While the link between clan control and peer control remains ambiguous in prior
work, we expect that the more a common vision, norms, and goals guide team members’ behaviors, and
the more committed the individual members are to the team as a result of high levels of clan control
(Kirsch, 1996), the more they will monitor whether their peers act according to these norms, the more
they will praise them for aligned behavior, and the more they will correct them in case they deviate. Simi-
larly, higher levels of clan control help cultivate a shared understanding and language among team mem-
bers (Kirsch, 1996) and therefore facilitate more productive discussions about how work gets done—and
should be done—within a team (Sihag and Rijsdijk, 2019).
We further expect clan control to have a negative association with indirect peer control. In partic-
ular, the establishment of a common vision, norms, and goals among team members should be associated
with higher mutual trust (Choudhury and Sabherwal, 2003; Huang et al., 2005), stronger commitment
(Kirsch, 1996), reciprocity, and social cohesion within the team (Ouchi, 1979). Rituals and ceremonies
further serve to identify and reinforce acceptable behaviors among team members, and individuals are re-
warded for acting in accordance with the team’s values (Kirsch, 1996). Such socialization and reinforce-
ment of acceptable behaviors are targeted at effectively eliminating goal incongruence between team
members (Ouchi, 1979) and, at the very least, will lead to team members becoming more homogenous in
terms of goals, norms, and ways of operating. In sum, then, trust, commitment, and cohesion as a result of
clan control should reduce the need for team members to resort to indirect peer control, i.e., gossiping
about peers’ non-performance and about peers with different approaches, as well as avoiding deviant
peers. Formally:
Hypothesis 2: Clan control is (a) positively related to direct peer control and (b) negatively re-
lated to indirect peer control.
Outcomes of Peer Control
Job Satisfaction. Theoretical support for a relationship between direct peer control and job satis-
faction can be derived from the job characteristics model (JCM) (e.g., Hackman and Lawler, 1971;
Hackman and Oldham, 1975). Hackman and colleagues proposed that positive work outcomes—such as
job satisfaction—are dependent on intrinsically motivating jobs, characterized by five dimensions: Task
identity, task significance, skill variety, autonomy, and feedback. We expect direct peer control to posi-
tively affect all five core job dimensions, thereby having a positive influence on job satisfaction. First,
while some authors have argued that very high levels of alternative work practices, such as team responsi-
bility for a task, can have a negative impact on job satisfaction (Godard, 2001), the heedful interrelating,
enhanced coordination of tasks, and greater agreement about goals as a result of direct peer control
(Loughry, 2010) allow team members to understand how their contribution fits into the big picture of the
overall job, which enhances task identity. Second, the opportunities for peer recognition and praise that
direct peer control provides increase team members’ feelings of task significance and make them feel
more appreciated (Loughry, 2010). Third, Godard (2001) has argued that at very high levels, multi-skil-
ling can negatively influence job satisfaction as it can increase workload. Despite this potential downside,
direct peer control provides team members with cues about which tasks are considered relevant, and, even
more importantly, how well these tasks are performed. These cues, in turn, provide team members with
opportunities to appropriately apply and further improve their skills and talents (Bijlsma-Frankema and
Costa, 2010), enhancing skill variety. Fourth, some authors have cautioned that excessive peer control
may induce conformity to team norms (Merchant, 1985), which may undermine team members’ auton-
omy. Taken to its extreme, peer control could result in a more subtle, but also more difficult-to-resist,
concertive control regime with potentially oppressive tendencies (Barker, 1993; Wright and Barker,
2000). Even these authors acknowledge, however, that such dysfunctional aspects of peer control only
arise when the control system is too constraining, “paralleling the disadvantages of external control mech-
anisms such as supervision” (Wright and Barker, 2000, p. 347). The more recent literature has further
maintained that, compared to other forms of control, peer control tends to be perceived as leaving team
members with more choice, or autonomy, in their work (Loughry, 2010; Weibel, 2010). Moreover, with
direct peer control, any restrictions on individual autonomy are not the result of rules and structures that
the organizational hierarchy has promulgated, but of social obligation and normative pressures emerging
from team members’ agreed-upon core values (Barker, 1993). As a result, “workers willingly subjugate
themselves to a pressure that rationalizes work and ensures controlled collective action but that, unlike
bureaucratization, emerges socially rather than from formal organizational structure” (Lange, 2008, p.
721). Accordingly, we expect direct peer control to increase the perceived autonomy associated with a
given job. Fifth, direct peer control by definition involves direct and open responses to peers’ behavior
and results, such as praising team members for jobs well done, correcting team members when mistakes
were made, and discussing everyone’s job behavior and performance (Loughry and Tosi, 2008). Direct
peer control further enjoys distinct advantages over other types of control in that team members fre-
quently interact with each other, which provides ample opportunity for feedback in the form of recogni-
tion and praise from peers (Loughry, 2010). Thanks to their intimate knowledge about each other as a by-
product of their frequent interactions, peers are often the best judges of the appropriateness of their team
members’ behaviors (Loughry, 2010). Consequently, direct peer control creates “a tight link between the
coworker’s behavior and the peer administered consequences” (Loughry and Tosi, 2008, p. 885), allow-
ing for timelier and higher-quality feedback. We therefore propose:
Hypothesis 3a: Direct peer control is positively related to job satisfaction.
In contrast to direct peer control’s positive effect, we expect that indirect peer control is not con-
ducive to these core job dimensions, but rather detracts from at least two of themautonomy and feed-
back—and thereby negatively affects job satisfaction. In particular, rather than actively engaging team
members as is common in direct peer control, gossiping about and avoiding a particular team member
represent acts of disengaging oneself from that peer. This disengagement suggests that indirect peer con-
trol is more about protecting one’s own interests than working towards other team members’ or the team’s
interests (Loughry and Tosi, 2008). Such politically motivated control efforts are likely seen as unwar-
ranted and illegitimate attempts to undermine team members’ autonomy, and thereby diminish job
satisfaction (cf., Hackman and Lawler, 1971; Hackman and Oldham, 1975). Moreover, gossiping about
and avoiding poorly performing team members are in direct contrast to the JCM’s conceptualization of
feedback as “obtaining direct and clear information about the effectiveness of his or her performance”
(Hackman and Oldham, 1975, p. 161). The target of gossip is, by definition, not present and often not
even aware, and team members may actively shun this peer (Loughry and Tosi, 2008). There are thus few
(if any) tangible opportunities to learn from, and potentially change, one’s behavior. Indirect peer control
therefore is a comparatively poor feedback mechanism.
In addition, some authors have also cautioned that indirect peer pressure might subject peers to
unnecessary stress, and thereby lead to an unpleasant work environment (Loughry, 2010). In particular,
while acknowledging the possible benefits of gossip for enforcing team norms and values (Dunbar, 2004;
Gluckman, 1963), the literature on gossip suggests people learning that their team members are exchang-
ing negative, work-related information about them behind their backs likely causes them distress. This
distress can trigger interpersonal conflict and hostile relations within the team (Grosser et al., 2010).
Moreover, gossip can exacerbate negative relationships between team members and perpetuate the social
marginalization of organizational “outcasts,” who are often the targets of gossip (Grosser et al., 2012). On
the flipside, the prevalence of indirect peer control is a signal that even team members conforming with
expectations—i.e., those that resort to gossip about other, non-compliant team membersare unhappy
with their (non-compliant) team members and will thus likely experience lower levels of job satisfaction.
Consequently, especially negative gossip can create a hostile work environment for the targets of gossip
and its broader audience, including gossipers themselves, with detrimental effects on job satisfaction. We
therefore propose:
Hypothesis 3b: Indirect peer control is negatively related to job satisfaction.
Team Performance. The effect of peer control on team performance remains a subject of consid-
erable debate in the literature, and the empirical evidence is mixed (De Jong et al., 2014; Loughry and
Tosi, 2008; Stewart et al., 2012; Welbourne and Ferrante, 2008). On the one side of this debate, some au-
thors have argued that peer control could deduct time from team members’ assigned duties, upset team
leaders who might see peer control as team members’ encroachment on their formal managerial roles, and
contribute to performance problems if attempts at peer control are misinformed (Loughry, 2010;
Welbourne and Ferrante, 2008). These negative effects seem to diminish, however, with the flatter hierar-
chies characterizing modern organizations as well as team members’ increasing familiarity with—and, as
a result, more goal-oriented use ofpeer controls (e.g., De Jong et al., 2014; Kirsch et al., 2010; Loughry
and Tosi, 2008). Other authors have cautioned that excessive levels of team norm conformity due to peer
control may stifle creative approaches to work problems, and therefore reduce a team’s ability to adapt to
a changing environment (Merchant, 1985), which would suggest a boundary condition on the effective-
ness of very high levels of peer controls in settings where creativity and adaptation are paramount.
On the other side of this debate are three sets of arguments building on motivational, agency, and
learning theories. First, motivational theory work has extended the JCM from the individual to the team
level and found that more empowered jobs are associated with higher team performance (Kirkman and
Rosen, 1999). In particular, when team members take on greater responsibility in the form of peer control,
they are likely to perceive themselves as having more autonomy, a greater impact on their peers’ develop-
ment and rewards, more meaningful and a wider variety of skills, and a more accurate understanding of
their team’s potency (Kirkman and Rosen, 1999).
Second, compared to other control regimes, peer control enjoys distinct advantages with respect
to agency theory’s two key mitigating mechanisms, monitoring and incentive alignment (Loughry and
Tosi, 2008). More frequent interaction and a more intimate understanding between peers compared to
those with their team leaders allows peers to detect behaviors that other forms of monitoring might miss
and provides more opportunities to influence each other (Loughry, 2010; Welbourne and Ferrante, 2008).
Peer control further takes advantage of peers’ self-interests, which leads them to evaluate their team mem-
bers’ behavior with respect to its impact on their own performance and, by extension, that of the team as a
whole (Kandel and Lazear, 1992), and penalize the pursuit of individual goals at the expense of team
goals. Not surprisingly, prior research found it critical for reducing free-riding in teams and increasing
team effectiveness (Loughry and Tosi, 2008; Manz and Sims, 1987).
Third, direct peer control provides coordination and learning benefits for teams. By providing
constant feedback on how team members behave and perform their jobs, direct peer control offers guid-
ance, clarifies roles, and reduces team members’ perceived role ambiguity (Loughry, 2010), which in turn
facilitates coordination between team members (Loughry, 2010; Manz and Sims, 1987). Moreover, ob-
serving how peers perform specific tasks, discussing their approach, and correcting it if it is perceived as
suboptimal can be a catalyst for team learning (Kirkman and Rosen, 1999; LePine and van Dyne, 2001).
In line with these arguments, we propose:
Hypothesis 4a: Direct peer control is positively related to team performance.
There are similarly divergent views on the effects of indirect peer control on team performance.
One point of view builds on the individual-level effects discussed above that indirect peer control causes
team members distress. At the team level, this could lead to an unpleasant work environment marked by
interpersonal conflict and hostile relations within the team and, ultimately, to its fragmentation (Grosser et
al., 2010; Grosser et al., 2012). Grosser et al. (2012, p. 52) further acknowledge, however, that gossip
can be a tricky organizational phenomenon in that it can be both positive and negative at the same time;
this often depends on whether one is viewing the gossip from the employee’s perspective or the organiza-
tion’s perspective.”
Given this section’s focus on team-level performance, we subscribe to the alternative view that
indirect peer control in general (Barker, 1993; Sewell, 1998) and particularly the gossip and avoidance it
entails (Dunbar, 2004; Gluckman, 1963; Grosser et al., 2012) can be an effective and efficient means of
maintaining conformity and control over team members. Indirect peer control enjoys the monitoring and
incentive advantages peers have over team leaders (Loughry and Tosi, 2008). In comparison to direct peer
control, gossiping about and avoiding underperforming peers also carries significantly less costs and risks
for the person monitoring, as it avoids any open confrontation between team members (Loughry and Tosi,
2008). Nevertheless, prior work has identified gossip and avoidance as powerful means of mitigating ego-
istic behaviors and securing the subsequent cooperation of ostracized team members (e.g., Feinberg et al.,
2014). Moreover, based on the information obtained via gossip, the team may decide to avoid certain
individuals when subsequent tasks are assigned, allowing the team to achieve better outcomes (Grosser et
al., 2012). Finally, gossip may also signal trust within teams. The literature has long acknowledged the
exchange of gossip as a way to form and maintain relationships within an organization. It has also pointed
out that gossip can even bring individuals closer together (Dunbar, 2004; Grosser et al., 2012), with bene-
ficial consequences for team effectiveness and performance. In line with this second view, we propose:
Hypothesis 4b: Indirect peer control is positively related to team performance.
To test these hypotheses, we gathered data on regional coordinators, team leaders, and volunteer
teams working for the German country organization of a worldwide children’s rights non-profit organiza-
tion (NPO). In Germany, this NPO has around 1,800 volunteers, organized in 90 geographic teams. (The
approximately 6,000 occasional volunteers who help out at special events were not included in our study.)
Each team is headed by a volunteer team leader, and grouped together into five geographical regions,
each under the responsibility of a salaried regional coordinator. The average team consists of about 19
members plus a team leader, which is in line with prior work on peer control (e.g., De Jong et al., 2014;
Stewart et al., 2012).
As part of a multinational NPO, the German country organization we are studying exhibits very
high levels of professionalism and has implemented a professional system to manage volunteers. Unlike
other volunteer or grassroots associations, our NPO’s activities are defined by its German headquarters.
Rules and procedures regarding volunteer work are stipulated by the head of the volunteer coordination
division, who is also a member of the executive board, and who previously worked for a leading consult-
ing firm. A year before our study took place, the NPO took additional steps to homogenize its teams’
identities by establishing and promoting a common mission statement (Jacobs et al., forthcoming). Addi-
tional guidelines, activity handbooks, and procedures are outlined in the organization’s intranet, which is
used by all volunteer teams. Each team leader is further invited to participate in one of four annual
regional conferences, which feature team-building activities, volunteer training workshops, and best-prac-
tice exchanges.
Team leaders are selected by the regional coordinators, who screen all volunteer members of a
team and approach members who, beyond a high commitment, also have prior leadership experience and
show strong leadership skills, to become team leaders. Team leaders receive extensive training by the
NPO headquarters. Almost all team leaders possess a higher-education degree and, on average, have
around 15 years of professional experience. Team leaders specify concrete goals for the team’s activities,
staff team members on specific activities, and chair the biweekly team meetings, for which they also set
the agenda.
Team members engage in office management, administration, accounting, event management,
greeting card sales, fundraising, public relations, and media design tasks on a daily basis. These tasks are
highly interdependent, and none of them can be done by a single team member. Despite the common mis-
sion statement and other rules and procedures stipulated by headquarters, each team operates largely au-
tonomously in its geographical context, and the teams do not interact directly. We restricted our sample
population to those 83 teams that had worked together for more than a year.
Our setting represents a suitable context for testing our theoretical framework. On the one hand,
the teams have flat hierarchies and thus a wide span of control—with only one team leader responsible for
each team, and with all team members considered peers—but also require extensive interactions among
team members to accomplish team tasks, and so we would expect a certain degree of lateral peer control
to be present in such teams (cf., De Jong et al., 2014; Loughry and Tosi, 2008). On the other hand, the ex-
istence of experienced team leaders, supported by a strong mission statement, rules, and procedures estab-
lished by headquarters, also suggest top-down managerial control. The clear delineation of teams respon-
sible for their own regions that are overseen by regional coordinators allows us to gather reliable and
comparable performance data on the teams. And lastly, with teams being responsible for—and having
discretion over how to manage—their unique geographical contexts, we expect sufficient variance in both
our independent and dependent variables.
Our sample organization is testament to large, multinational NPOs becoming increasingly profes-
sionalized in their organizational and management approaches (e.g., Hwang and Powell, 2009), further
blurring the demarcation line between for-profit and not-for-profit organizations (e.g., Bromley and
Meyer, 2017). And even outside the NPO setting, our team-based organizational context exhibits many
features typical of 21st-century organizations—such as (voluntary) communities of practice (e.g., Wenger
and Snyder, 2000)—making our sample context comparable to the context of routine team tasks in for-
profit organizations.
We conducted several pre-tests of our survey with different audiences within our sample organi-
zation, in which participants had the opportunity to raise questions and were asked to identify any ambig-
uous or unclear items. Based on their suggestions, we slightly changed the wording of several items as
outlined below to better fit our empirical context. The survey was distributed in German. We followed the
common guidelines for translating our English questionnaire into a different language with a forward and
backward translation.
The NPO’s headquarters distributed the final eight-page questionnaire to all team leaders, who
then distributed the surveys to their team members. In addition, the NPO headquarters posted a note on
the organization’s intranet with a link to the survey website. We guaranteed all participants confidentiality
and sent out three e-mail reminders. Excluding two teams with only one respondent, we received usable
responses from 356 volunteers in 58 teams, or from about 20% of the volunteers and 70% of the teams.
To examine any potential for response bias, we compared responding and non-responding teams with re-
gards to the age of the teams and the team leaders’ ages and genders, which was the only information that
was made available to us for all teams. Results show no systematic differences between responding and
non-responding teams on all three variables.
Prior research has long established that “employee perceptions of their jobs have substantial con-
vergence with the assessments of objective job characteristics made by the researchers and by company
supervisors,” and that “results suggest, therefore, that employees’ perceptions of their jobs are of central
importance in affecting job attitudes and behaviors” (Hackman and Lawler, 1971, p. 275). In line with
both seminal and recent research on peer control (e.g., De Jong et al., 2014; Loughry and Tosi, 2008), we
therefore measured all organizational control variables by surveying team members. Our control for team
size was obtained from each team’s leader. To mitigate possible common method variance, we followed
Podsakoff and colleagues’ (2003; 2012) recommendations for survey design choices and used their latent
variable approach by adding an uncorrelated common method factor to the overall measurement model of
a confirmatory factor analysis, which enabled us to estimate the percentage of variance in responses due
to trait, method, and random error components. Partitioning the variance in this way revealed that
54.2% of the variance in Model 4 was accounted for by the trait factors, 35.3% by random errors, and
only 10.6% by the method factor. With the proportion of the variance accounted for by the method factor
being much less than that explained by the trait factors and the 24-25% typically found across studies in
the management field (Podsakoff et al., 2003), any potential bias would be minor and unlikely to affect
our results.
To avoid common method variance altogether in our team-level hypotheses, we measured team-
level performance with two distinct measures: We surveyed the regional coordinators who, on average,
evaluated a dozen teams each; and we used objective performance data to measure team-level
performance. Consequently, we collected data from four sources (team members, team leaders, regional
coordinators, and publicly available financial reports) and at two levels of analysis (individuals and
Independent Variables
Unless mentioned otherwise, all responses were recorded on a 1 = strongly disagree to
5 = strongly agree Likert-type scale. Please refer to the Appendix for a list of all survey items.
Formal Managerial Control. We adapted Bonner et al.’s (2002) measure, which captures pro-
cess and outcome control. As our pre-testers perceived that our third item made one item redundant (“up-
per management determined the team’s work process”), and another item did not fit our organizational
context (“upper management specified objectives for quality management and standards for this project”),
we deleted these two items. The resulting six-item measure of formal managerial control loaded on one
factor without any significant cross-loadings (AVE = 0.60; α = 0.88; CR = 0.90; rwg(j) = 0.63;
ICC(1) = 0.04; ICC(k) = 0.18).
Clan Control. We obtained the first two items for clan control from of a three-item scale used in
Kirsch et al. (2010). We dropped the original third item (“All project team members attempted to be ‘reg-
ular’ members of the project team”) after the pre-test as the pretest respondents did not understand it. We
replaced it with a new item: “All team members know [organization]’s vision and act accordingly
(AVE = 0.61; α = 0.85; CR = 0.83; rwg(j) = 0.82; ICC(1) = 0.05; ICC(k) = 0.23).
Direct Peer Control. We followed Loughry and Tosi’s (2008) distinction between direct and in-
direct peer control. After our pretest, we deleted six items of the original 14-item direct peer control scale
that were not applicable to our context as evident from the pre-tests. For instance, items such as “tell a su-
pervisor if a team member is stealing” of the subdimension report openly offended our respondents in the
pretests. This left us with eight items to measure direct peer control, comprising two items for each of
Loughry and Tosi’s (2008) four subdimensions, notice, praise, correct, and discuss. The answers were
indicated on a range from 1 = never to 5 = very often. An exploratory factor analysis showed, however,
that these four subdimensions loaded on two factors: notice/praise on the one hand, and correct/discuss on
the other hand. We performed a confirmatory factor analysis comparing our two-factor solution for direct
peer control (χ2(19) = 211.82; p < 0.001) with a one-factor solution combining the items for notice/praise
with correct/discuss (χ2(20) = 475.09; p < 0.001). The two-factor solution is clearly superior
(χ2diff(1) = 263.27; p < 0.001). Based on these factor analyses, we decided to use the two-factor solution:
direct peer control (notice & praise) (AVE = 0.41; α = 0.86; CR = 0.73; rwg(j) = 0.82; ICC(1) = 0.04;
ICC(k) = 0.19) and direct peer control (correct & discuss) (AVE = 0.55; α = 0.81; CR = 0.78;
rwg(j) = 0.61; ICC(1) = -0.03; ICC(k) = -0.17) and report the results for a one-factor solution in our Sup-
plemental Analyses section.
Indirect Peer Control. After deleting the original third item (“gossiping about coworkers’ per-
formance”) in Loughry and Tosi’s (2008) scale, because pre-tests indicated that it did not fit our non-
profit context, we ended up with four of the original five items to measure indirect peer control. Our final
measure comprises two items each for the sub-dimensions gossip and avoid and loaded on one factor
without any significant cross-loadings (AVE = 0.71; α = 0.89; CR = 0.91; rwg(j) = 0.60; ICC(1) = 0.07;
ICC(k) = 0.30).
Dependent Variables
Job Satisfaction. To measure job satisfaction, we used the three-item scale by Boezeman and El-
lemers (2009), which they had already adapted to a volunteer context (AVE = 0.78; α = 0.92; CR = 0.91).
Team Performance. We measured team performance in two different ways. First, in close inter-
action with the head of the volunteer coordination division, we developed three team performance items
aligned with the NPO’s existing internal team assessments on a scale ranging from 1 = very good to
5 = very poor. We then collected assessments from the five regional coordinators at headquarters who are
supervising all teams in their respective regions. We averaged these informants’ assessments across all
three items and reverse-coded the scale to allow for an easier interpretation of the results (AVE = 0.82;
α = 0.92; CR = 0.93). Second, in addition to the regional coordinators’ perceived performance measure,
we used the official profit data (i.e., the natural logarithm of the difference between revenues and costs)
the NPO collects for each team—and reports to German tax authorities—to measure team profits.
Revenues are mainly generated by fundraising efforts that include events the teams organize, such as
charity runs for children or concerts, but team members also activate their private social networks to so-
licit donations.
Control Variables
We controlled for several individual team member characteristics known to influence individual
satisfaction and/or team performance and the perceived level of peer control: gender (female) [1 for fe-
male and 0 for male team members]; age (ln) [natural logarithm of years old], which may influence job
satisfaction and the intention to remain; organizational tenure [on a 1 to 6 scale: 0-1 years; 1-5 years; 6-
10 years; 11-15 years; 16-20 years; > 20 years] as “[w]orkers in organizations and positions with low
turnover might be more likely to see the organization’s interest aligned with their own, and thus be will-
ing to engage in peer control that supports better performance” (Loughry, 2010, p. 350); respondents’
time commitment (ln) [natural logarithm of the average number of hours per month], as this may influence
satisfaction and intention to remain; respondents’ work experience in professional jobs (professional ex-
perience) [on the same scale as organizational tenure], as this may influence a respondent’s experience
with peer and managerial control in teams; whether a respondent is currently volunteering at another NPO
(other NPO engagement) [1 if that respondent currently volunteers at another NPO and 0 otherwise]; the
natural logarithm of the number of active team members in each team (team size (ln)); and team cohesive-
ness in line with prior work on the performance implications of peer control (Loughry and Tosi, 2008),
using three of the latter authors’ six items that pre-tests showed best captured our context (AVE = 0.48;
α = 0.79; CR = 0.73; rwg(j) = 0.84; ICC(1) = 0.11; ICC(k) = 0.44). At the team level of analysis (with
team performance/profits as dependent variables), we used the mean scores of these variables across team
members as controls.
Validity and Reliability
We performed exploratory factor analyses for all multi-item variables, and all items loaded on
their respective factors with no significant cross-loadings. The one exception was the described two-factor
solution for direct peer control. We also examined the average variance extracted (AVE) of each construct
and found that all AVEs were higher than the recommended minimum value of 0.50 (Fornell and Larcker,
1981), indicating adequate convergent validity. The exception was, again, the two-factor solution we
found for direct peer control and our control variable team cohesiveness. To ensure discriminant validity,
we ascertained that the AVEs for any two constructs were greater than the shared variance (i.e., squared
correlation) between the two constructs (Hair et al., 2009). Finally, to confirm scale reliability we calcu-
lated consistency reliabilities (Cronbach alphas) and composite reliabilities (CR) for each factor. All
measures’ alphas satisfied the generally recommended level of 0.70 (Nunnally, 1978), and all CRs were
higher than the recommended value of 0.70 (Hair et al., 2009).
Hypotheses Tests
As individual team members’ assessments are nested within their respective teams, and teams are
nested within their respective regions, we used a multilevel modeling (MLM) approach for all of our anal-
yses (Bryk and Raudenbush, 1992). For our individual-level analyses (H1-3 in Models 1-4), we tested our
hypotheses with a three-level MLM model, with individual team members nested within teams, and teams
nested within regions. For our team-level analyses (H4 in Models 5-6), we employed a direct-consensus
model using average team member responses to operationalize team-level scores (Chan, 1998). We exam-
ined whether an aggregation of the individual responses to the team level was warranted by calculating
the within-group agreement using rwg(j) statistics (James et al., 1993) as well as ICC(1) and ICC(k) indices
(Bliese, 2000). All variables had moderate (rwg(j) between 0.51 and 0.70) or strong interrater agreement
(rwg(j) between 0.71 and 0.90) (LeBreton and Senter, 2008). Several of our variables had ICC(1) values
of 0.05 or higher, providing “prima facie evidence of a group effect” (LeBreton and Senter, 2008, p. 838);
and all of the other constructs exhibited ICC(1) > 0.01, which is in line with Bliese’s (1998) threshold for
detecting group-level relationships not evident in the lower-level data, and which provides further evi-
dence that the aggregation to the team level was justified. For the models using team performance (Mod-
els 5 and 6), we tested our hypotheses with a two-level MLM model, with teams nested within regions.
Table I presents means, standard deviations, and bivariate correlations. In addition to an average
of 3.32 (on a five-point scale) for formal managerial control, providing evidence for team leaders actively
engaging in managerial control, and an average of 4.03 for clan control, we also see clear evidence of di-
rect peer control (average of 3.74 for notice & praise; average of 3.01 for correct & discuss) and indirect
peer control (average of 2.33).
Table II presents our MLM regression results. Models 1-3 show the individual-level relationships
between managerial and clan control and peer control; Model 4 shows the individual-level relationships
between managerial, clan, and peer controls and job satisfaction; and Models 5 and 6 show the team-level
relationships between managerial, clan, and peer control and, respectively, team performance and team
Our results suggest that the older team members are, the less likely they are to report direct peer
control among team members, and that the longer they are part of the organization, the less they report
evidence of noticing and praising among team members as part of direct peer control, which would sug-
gest a relatively higher likelihood of disengagement for older team members. In contrast, the more profes-
sional experience team members have, the more they report direct peer control in the form of noticing and
praising among team members. Also as expected, team cohesiveness is positively associated with direct
peer control and negatively with indirect peer control. And it makes sense—particularly for our volunteer
context—that team members reporting higher levels of job satisfaction also report a longer tenure at and a
higher time commitment to the NPO. Moreover, team size is positively related to team performance and
profits, which can be explained by the higher number of revenue-generating activities that are possible
with a higher number of active team members. Our finding that team cohesiveness is positively related to
team profits is in line with broad support for this relationship in the literature (e.g., Evans and Dion,
2012). Average team member age and organizational tenure have a positive and negative association, re-
spectively, with team performance, which may suggest that while the increased experience associated
with age seems to help improve team performance, increasing average tenure has detrimental effects on a
team’s performance. And lastly, the average team member’s engagement with other NPOs serves as an
indicator of the team’s experience with other organizations in the non-profit sector that can be leveraged
for the task at hand, explaining its positive association with team profits.
Providing partial support for Hypothesis 1a, formal managerial control is positively related to di-
rect peer control (notice & praise) (Model 1: b = 0.02, p = 0.76), and positively and significantly related
to direct peer control (correct & discuss) (Model 2: b = 0.17, p = 0.01). We find no support for H1b, how-
ever, as the relationship between formal managerial control and indirect peer control is non-significant
(Model 3: b = 0.03, p = 0.60). In support of Hypotheses 2a and 2b, clan control is positively and signifi-
cantly related to direct peer control (Model 1: b = 0.35, p < 0.001 for notice & praise, and Model 2:
b = 0.27, p < 0.01 for correct & discuss), whereas clan control is negatively and significantly related to
indirect peer control (Model 3: b = -0.28, p < 0.001). In line with Hypotheses 3a and 3b, direct peer con-
trol is positively and significantly related to job satisfaction (Model 4: b = 0.16, p < 0.05 for notice &
praise, and b = 0.06, p < 0.01 for correct & discuss), while indirect peer control is negatively and signifi-
cantly related to job satisfaction (Model 4: b = -0.13, p < 0.001).
In Models 5 and 6, we present our team-level findings on the relationship between direct peer
control (Hypothesis 4a) and indirect peer control (Hypothesis 4b) on team performance and team profits.
The findings for Hypothesis 4a are mixed: Contrary to our expectations, direct peer control is either not
significantly associated with team performance (Model 5: b = 0.32, p = 0.27 for notice & praise) and team
profits (Model 6: b = -0.51, p = 0.09 for correct & discuss); or it is negatively and significantly associated
with both team performance (Model 5: b = -0.66, p < 0.05 for correct & discuss) and team profits
(Model 6: b = -0.81, p < 0.001 for notice & praise). Our results support Hypothesis 4b, as we find positive
and significant associations between indirect peer control and team performance (Model 5: b = 0.71,
p < 0.05) as well as team profits (Model 6: b = 0.30, p < 0.05). Figure 1 provides an overview of our re-
Supplemental Analyses
We re-ran our models with a one-factor solution for direct peer control (Loughry and Tosi, 2008),
and the results are consistent with our two-factor findings: In Models 1-4, formal managerial control
(b = 0.09; p = 0.12) was not significantly associated with direct peer control while clan control was
(b = 0.23; p < 0.001), and direct peer control remained significantly associated with job satisfaction
(b = 0.17; p < 0.001). Direct peer control also remained negative and nonsignificant (b = -0.38; p = 0.21)
in Model 5 and negative and significant (b = -1.31; p < 0.001) in Model 6.
We also tested Loughry and Tosi’s (2008) hypothesized interaction effect between formal mana-
gerial control and peer control. Added to Models 5, none of the interaction effects between formal mana-
gerial control and direct peer control (notice & praise) (b = -0.50, p = 0.50), direct peer control (correct &
discuss) (b = 1.17, p = 0.22), and indirect peer control (b = -0.34, p = 0.68) attained significance. Added
to Model 6, none of the interaction effects between formal managerial control and direct peer control (no-
tice & praise) (b = 1.46, p = 0.10), direct peer control (correct & discuss) (b = 0.40, p = 0.77), and indirect
peer control (b = 0.06, p = 0.97) attained significance either. These results provide no evidence for an in-
teraction effect between formal managerial and peer control.
While not explicitly hypothesized, we also examined the indirect effects implied in our frame-
work. At the individual level, we estimated the indirect effects of formal managerial control and clan con-
trol on job satisfaction via peer control, following current recommendations for multilevel mediation anal-
yses (Aguinis et al., 2017). Our results provide support for three indirect effects: the indirect effect of clan
control via direct peer control (correct & discuss) (b = 0.02, p = 0.098; 95% CI: 0.000, 0.03); the indirect
effect of clan control via direct peer control (notice & praise) (b = 0.05, p = 0.02; 95% CI: 0.02, 0.08);
and the indirect effect of clan control via indirect peer control (b = 0.03, p = 0.01; 95% CI: 0.01, 0.05).
The indirect effects of formal managerial control do not attain statistical significance (b = 0.00, p = 0.78;
b = 0.01, p = 0.17; and b = 0.00, p = 0.63 for the two direct peer control dimensions and indirect peer con-
trol, respectively). These indirect effects map closely onto our main-effects results and provide additional
support for our theorizing of clan control as an antecedent of peer control.
At the team level, our sample size (n = 58 teams) is not large enough to provide enough statistical
power to detect mediation effects (Fritz and MacKinnon, 2007). The results nevertheless provide at least
some indication of a mediating effect at the team level. While none of the indirect effects for team perfor-
mance attained significance, for team profits, we obtained at least marginally significant indirect effects
for the effect of clan control via direct peer control (correct & discuss) (b = -0.18, p = 0.05; 95% CI: -
0.34, -0.03) and via indirect peer control (b = 0.11, p = 0.09; 95% CI: 0.004, 0.22).
To further examine whether our findings can be generalized to other professional contexts beyond
our volunteer sample, we collected data on six teams at our sample NPO’s headquarters that are com-
posed entirely of salaried employees. For all six headquarters teams (with a minimum of three and a max-
imum of nine members), the six team leaders and 29 out of 38 team members answered an almost identi-
cal survey to the one we sent out to our main sample; we only slightly adapted the wording of our items to
correspond to the professional context. While the small sample size did not allow us to replicate our anal-
yses with this sample alone, we can nevertheless use it to examine the generalizability of our findings.
First, we used t-tests to compare the means of all hypothesized variables at the individual level and found
that the only statistically significant difference was for clan control, which was higher among volunteers
(M = 4.03, SD = 0.78) than among salaried employees (M = 3.09, SD = 0.77; p < 0.001). This difference
likely reflects the previously mentioned efforts by the NPO’s corporate headquarters to homogenize the
volunteer teams’ identities by establishing and promoting a common mission statement a year before our
study took place. We also used Levene’s test for equality of variances, and found that the only statistically
significant difference between samples was for formal managerial control, which had a higher variance
among volunteers (σ² = 0.78) than among salaried employees (σ² = 0.27; p < 0.001), providing evidence
of the higher discretion volunteer team leaders enjoy.
Excluding two control variables that are not available for the headquarters teams—i.e., time com-
mitment and other NPO engagement—we re-ran our regression analyses for the individual-level models
(Models 1-4) with a pooled sample including both volunteer and salaried teams members (n = 385). All
hypothesized results remained the same with the exception of formal managerial control, which went
from marginally significant (b = 0.08, p = 0.09) to fully significant (b = 0.09, p = 0.04). As the salaried
headquarters teams had different goals and objectives than our volunteer teams, there was no comparable
dependent variable available, and so we had to restrict our supplemental analyses to Models 1-4. Despite
this limitation, these results provide at least tentative evidence that our findings apply equally to salaried
employees outside the volunteer context.
Our study focused on lateral peer control, its association with top-down managerial controls, and
its consequences at both the individual and team level. We find formal managerial control and clan con-
trol to be antecedents of peer control, albeit with differential effects on direct and indirect peer control.
We also find a significant effect of peer control on individual job satisfaction and team performance, but
again, with crucial differences between the two types of peer controls and the two outcomes.
Theoretical Implications
Our study has two main theoretical implications. First, our nuanced findings provide new avenues
for the burgeoning literature on configurations of different types of control in general (Cardinal et al.,
2004; Cardinal et al., 2010; Kreutzer et al., 2016; Kreutzer et al., 2015; Sihag and Rijsdijk, 2019), and for
prior studies suggesting an interactive (or substitution) effect of formal managerial and peer controls in
particular (Loughry and Tosi, 2008; Stewart et al., 2012; Welbourne and Ferrante, 2008). While Loughry
and Tosi (2008) have not tested a potential antecedent effect with their data, the positive and highly sig-
nificant correlation (r = 0.52) between formal managerial control and direct peer control they report
would suggest a positive association between those two control modes. On the flipside, we do not find
any evidence of their hypothesized substitute effect between formal managerial and peer control in our
supplemental analyses. These differences, however, could be an artefact of differences in the studies’ de-
pendent variables: while Loughry and Tosi (2008, p. 882) focused on problem-free performance, defined
as “the degree to which the work unit is free of employee behavior problems,” our study’s focus is on
more general team performance and team profits. Perhaps the difference in focus between avoiding the
negative (captured in problem-free performance) and enabling the positive (captured by our team
performance/profits) is responsible for some of the divergent findings between the two studies.
Complementing and extending prior studiessubstitution logic between different types of control,
our theorizing and findings corroborate managerial controls as important antecedents of peer controls.
Our study thus moves beyond prior work distinguishing between managerial controls as intentionally de-
signed, and lateral controls as resulting solely from the initiative and interactions among peers (e.g.,
Johnson and Gill, 1993) and instead provides empirical evidence for management’s ability to influence
the emergence of peer controls with the deliberate design of formal managerial control and clan control.
Moreover, our findings emphasize the importance of moving beyond task characteristics as key determi-
nants of organizational control regimes and of explicitly considering the social context in organizations,
such as the one set by top-down managerial controls.
As expected, clan control has a positive association with direct peer control and a negative associ-
ation with indirect peer control, suggesting that a common vision and goals guiding a team induce team
members to more closely monitor and control each other’s behaviors while, at the same time, reducing the
need for gossiping and/or avoiding non-compliant team members. The results pattern for formal manage-
rial control, however, is more complex. We found no support, for instance, for our argument that the
transparency characterizing formal managerial control might reduce indirect peer control. Instead, with
higher levels of formal managerial control, team leaders might actually exert more pressure on team
members to conform to behavioral expectations and to achieve certain performance outcomes. Such pres-
sure may, in turn, translate into a greater likelihood of team members avoiding less committed and relia-
ble peers and gossiping about any apparent or perceived shortcoming in their behavior or performance.
Our results also illustrate the need for a more differentiated understanding of direct peer control,
i.e., highlighting the distinction between notice & praise versus correct & discuss. This is evident from
our finding that only correct & discuss is positively associated with formal managerial control. By speci-
fying procedures, work assignments, and concrete goals, team leaders seem to be able to influence the
learning and improvement efforts within teams but to have less of an influence on team members noticing
and praising each other’s efforts.
Second, our findings establish peer control’s association with both individual job satisfaction and
team performance and thereby shed new light on inconsistencies found in prior work (Kohli and Jaworski,
1994; Loughry and Tosi, 2008; Stewart et al., 2012; Welbourne and Ferrante, 2008). In particular, our re-
sults illustrate the need to distinguish between direct and indirect peer control, which has contrary effects
on both outcomes, and to look at direct peer control in a more nuanced way. On the one hand, direct peer
control is positively related to job satisfaction, but has a negative effect on team performance (at least for
the two significant effects). These unexpected findings suggest that noticing peers’ work activities, prais-
ing and correcting them, and openly communicating and discussing work behavior may make team mem-
bers feel satisfied, but it detracts from team performance. A possible explanation is that direct peer control
does not always have to be consistent with accomplishing organizational objectives (Jaworski, 1988;
Loughry, 2010). On the contrary, our findings seem to suggest that direct peer control “could take time
away from workers’ assigned duties, upset supervisors, who might feel that control is a supervisory role,
or contribute to performance problems if the peer performing the control misunderstands what behavior is
appropriate” (Loughry, 2010, p. 341).
On the other hand, indirect peer control is negatively related to job satisfaction, but has a positive
effect on team performance. This differs from prior work that has deemed indirect peer control as a prob-
lematic behavior that is not in organizations’ interests (Loughry and Tosi, 2008). Our results provides em-
pirical corroboration for Grosser et al.’s (2012) observation that the impact of gossip in organizations de-
pends on whether one is taking the employee’s or the organization’s perspective. Our results suggest that
gossiping about peers within a team may provide relatively inexpensive, as well as relevant and timely,
information on which the rest of the team can act. While having a negative relationship with individuals’
job satisfaction, indirect peer control seems to unfold its disciplinary benefits for the team as a whole
(Barker, 1993; Feinberg et al., 2014; Sewell, 1998).
Our combined findings suggest both managerial discretion and an important trade-off when it
comes to peer control. While team leaders have the power to influence peer control within their teams
with their choice of managerial controls, a focus on enhancing job satisfaction among team members
comes at the expense of team performance, and vice versa. This trade-off poses an interesting dilemma
for team leaders, whose organizational control choices can benefit outcomes at the individual or team
level, but not both at the same time.
Limitations and Future Research Opportunities
Our findings are subject to several limitations. First and foremost is the empirical context of our
study—volunteers in an NPO. Our context is in line with recent work, however, that has used volunteer
samples to examine organizational and management theories (e.g., Florian et al., 2019). Moreover, the
typical daily tasks our respondent teams were engaged in as well as the increasing professionalization of
NPOs (Hwang and Powell, 2009) suggest that our sample context likely mirrors that of routine team tasks
in for-profit organizations. Even more importantly, the supplemental analysis of salaried headquarters
teams we report above provides no indication of any systematic differences between our volunteer and
salaried employee samples. This leaves us with few reasons to question the generalizability of our find-
ings. Nevertheless, future research comparing for-profit enterprises’ and NPOs’ organizational control
regimes should provide a more definite answer regarding the generalizability of these results.
A second and related limitation is our study’s German setting, which likely differs from other
peer control work that has focused on U.S. samples (Loughry and Tosi, 2008; Stewart et al., 2012). Key
differences of the German compared to the U.S. culture—e.g., its lower individualism, higher uncertainty
avoidance and long-term orientation, and lower indulgence (Hofstede, 2001)—could potentially influence
the extent to which teams use direct and indirect peer control and how this translates into individual satis-
faction and team performance. For instance, Germany’s lower level of individualism might imply that
team members feel more obliged to take care of each other than it would be expected in a more
individualistic culture, which could result in a higher likelihood and acceptance of direct peer control. In
addition, German culture is classified as relatively restrained, which implies that not much emphasis is
put on leisure time, and that people exert more control over the gratification of their desires (Hofstede,
2001). These social norms might make team members more likely to get upset with unreliable peers who
are shirking or slacking off, and to avoid and gossip about them. Future studies could replicate our find-
ings in similar settings in other countries or conduct multi-country studies to examine the influence of na-
tional culture on the antecedents and outcomes of peer control.
A third limitation is related to the cross-sectional nature of our data, which does not allow us to
make causal inferences. In line with recent recommendations in the literature (Aguinis et al., 2017), future
research should assess mediation for both the individual- and team-level model using longitudinal (and
preferably panel) data, which allow for an empirical comparison of alternative causal flows. Moreover,
while we did not find any evidence for this in our data—with team members’ average tenure being over
three yearsfuture research may examine potential differences in peer controls’ short- versus long-term
effects. For instance, direct peer control may unfold its performance benefits only over time, i.e., when
teams are working together long enough to translate the results of their discussion, correction, and prais-
ing activities into behaviors that increase team performance (cf., Barker, 1993). Indirect peer control, on
the other hand, could lead to turnover—which would entail a loss of knowledge—and health problems
(e.g., stress, burnout, etc.), and may thus have a negative effect on team performance in the long run.
We also did not control for task interdependence in our volunteer teams. As our sample teams all
engage in similar tasks that are outlined above and which are highly interdependent, we did not expect
significant differences in the teams’ task interdependencies. Loughry and Tosi (2008), however, found
that the link between direct peer control and work-unit performance was positively moderated by task in-
terdependence, while the link between indirect peer control and work-unit performance was not. Future
research may thus want to clarify any task interdependence-related contingency effects.
In conclusion, despite the proliferation and increasing importance of peer control in today’s in-
creasingly team-based, knowledge-intensive work environment, we still have a relatively shallow
understanding of its antecedents and organizational consequences. Our multilevel analysis of both ante-
cedents and outcomes of peer control represents a step towards the development of a better theoretical
foundation for this important phenomenon and reveals important trade-offs with regards to peer control’s
influence on individual- and team-level outcomes. Our study’s differential results highlight that a simulta-
neous examination of the antecedents and outcomes of peer control (and possibly other forms of organiza-
tional control) has the potential to generate new theoretical and empirical insights.
Aguinis, H., Edwards, J. R., and Bradley, K. J. (2017). 'Improving our understanding of
moderation and mediation in strategic management research'. Organizational Research
Methods, 20, 665-685.
Badura, A. (1977). Social learning theory, Englewood Cliffs, NJ: Prentice-Hall.
Barker, J. R. (1993). 'Tightening the iron cage: Concertive control in self-managed teams'.
Administrative Science Quarterly, 38, 408-437.
Beardsley, S. C., Johnson, B. C., and Manyika, J. M. (2006). 'Competitive advantage from better
interactions'. McKinsey Quarterly, 2, 53-63.
Bijlsma-Frankema, K. M. and Costa, A. C. (2010). 'Consequences and antecedents of managerial
and employee legitimacy interpretations of control: A natural open systems approach'. In
Sitkin, S. B., Cardinal, L. B., and Bijlsma-Frankema, K. M. (Eds), Organizational
control, Cambridge, U.K.: Cambridge University Press, 396-433.
Bliese, P. D. (1998). 'Group size, ICC values, and group-level correlations: A simulation'.
Organizational Research Methods, 1, 355-373.
Bliese, P. D. (2000). 'Within-group agreement, non-independence, and reliability: Implications
for data aggregation and analysis'. In Klein, K. K. and Kozlowski, S. W. J. (Eds),
Multilevel theory, research, and methods in organizations: Foundations, extensions and
new directions, San Francisco: Jossey-Bass, 349-381.
Boezeman, E. J. and Ellemers, N. (2009). 'Intrinsic need satisfaction and the job attitudes of
volunteers versus employees working in a charitable volunteer organization'. Journal of
Occupational and Organizational Psychology, 82, 897–914.
Bonner, J. M., Ruekert, R. W., and Walker, O. C. (2002). 'Upper management control of new
product development projects and project performance'. Journal of Product Innovation
Management, 19, 233–245.
Bromley, P. and Meyer, J. W. (2017). '“They are all organizations”: The cultural roots of
blurring between the nonprofit, business, and government sectors'. Administration &
Society, 49, 939-966.
Bryk, A. S. and Raudenbush, S. W. (1992). Hierarchical linear models, Newbury Park, CA:
Cardinal, L. B., Kreutzer, M., and Miller, C. C. (2017). 'An aspirational view of organizational
control research: Re-invigorating empirical work to better meet the challenges of 21st
century organizations'. Academy of Management Annals, 11, 559-592.
Cardinal, L. B., Sitkin, S. B., and Long, C. P. (2004). 'Balancing and rebalancing in the creation
and evolution of organizational control'. Organization Science, 15, 411–431.
Cardinal, L. B., Sitkin, S. B., and Long, C. P. (2010). 'A configurational theory of control'. In
Sitkin, S. B., Cardinal, L. B., and Bijlsma-Frankema, K. M. (Eds), Organizational
control, Cambridge, U.K.: Cambridge University Press, 51-79.
Chan, D. (1998). 'Functional relations among constructs in the same content domain at different
levels of analysis: A typology of composition models'. Journal of Applied Psychology,
83, 234-246.
Choudhury, V. and Sabherwal, R. (2003). 'Portfolios of control in outsourced software
development projects'. Information Systems Research, 14, 291-314.
Colbert, A., Yee, N., and George, G. (2016). 'The digital workforce and the workplace of the
future'. Academy of Management Journal, 59, 731-739.
De Jong, B., Bijlsma-Frankema, K. M., and Cardinal, L. B. (2014). 'Stronger than the sum of its
parts? The performance implications of peer control combinations in teams'.
Organization Science, 25, 1703-1721.
De Klepper, M. C., Labianca, G., Sleebos, E., and Agneessens, F. (2016). 'Sociometric status and
peer control attempts: A multiple status hierarchies approach'. Journal of Management
Studies, 54, 1-31.
Dunbar, R. I. M. (2004). 'Gossip in evolutionary perspective'. Review of General Psychology, 8,
Eisenhardt, K. M. (1985). 'Control: Organizational and economic approaches'. Management
Science, 31, 134-149.
Evans, C. R. and Dion, K. L. (2012). 'Group cohesion and performance: A meta-analysis'. Small
Group Research, 43, 690-701.
Fama, E. F. and Jensen, M. C. (1983). 'Separation of ownership and control'. Journal of Law &
Economics, 26, 301-325.
Feinberg, M., Willer, R., and Schultz, M. (2014). 'Gossip and ostracism promote cooperation in
groups'. Psychological Science, 25, 656-664.
Florian, M., Costas, J., and Kärreman, D. (2019). 'Struggling with meaningfulness when context
shifts: Volunteer work in a German refugee shelter'. Journal of Management Studies, 56,
Fornell, C. and Larcker, D. F. (1981). 'Evaluating structural equation models with unobservable
variables and measurement error'. Journal of Marketing Research, 18, 39.
Fritz, M. S. and MacKinnon, D. P. (2007). 'Required sample size to detect the mediated effect'.
Psychological Science, 18, 233-239.
Gluckman, M. (1963). 'Gossip and scandal'. Current Anthropology, 4, 307-316.
Godard, J. (2001). 'High performance and the transformation of work? The implications of
alternative work practices for the experience and outcomes of work'. Industrial and
Labor Relations Review, 54, 776-805.
Grosser, T. J., Lopez-Kidwell, V., and Labianca, G. (2010). 'A social network analysis of
positive and negative gossip in organizational life'. Group & Organization Management,
35, 177-212.
Grosser, T. J., Lopez-Kidwell, V., Labianca, G., and Ellwardt, L. (2012). 'Hearing it through the
grapevine: Positive and negative workplace gossip'. Organizational Dynamics, 41, 52-61.
Hackman, J. R. (1987). 'The design of work teams'. In Lorsch, J. (Ed), Handbook of
organizational behavior, Englewood Cliffs, NJ: Prentice-Hall, 315-342.
Hackman, J. R. and Lawler, E. E. (1971). 'Employee reactions to job characteristics'. Journal of
Applied Psychology, 55, 259-286.
Hackman, J. R. and Oldham, G. R. (1975). 'Development of the job diagnostic survey'. Journal
of Applied Psychology, 60, 159-170.
Hair, J. F., Black, W. C., Babin, B. J., and Anderson, R. E. (2009). Multivariate data analysis
(7th ed.), Upper Saddle River, NJ: Prentice Hall.
Hofstede, G. H. (2001). Culture's consequences: Comparing values, behaviors, institutions, and
organizations across nations (2nd ed.), Thousand Oaks, CA: Sage Publications.
Huang, G. Q., Zhang, X. Y., and Liang, L. (2005). 'Towards integrated optimal configuration of
platform products, manufacturing processes, and supply chains'. Journal of Operations
Management, 23, 267-290.
Hughes, C. (2004). 'The supervisor's influence on workplace learning'. Studies in Continuing
Education, 26, 275-287.
Hwang, H. and Powell, W. W. (2009). 'The rationalization of charity: The influences of
professionalism in the nonprofit sector'. Administrative Science Quarterly, 54, 268-298.
Jacobs, C. D., Kreutzer, K., and Vaara, E. (forthcoming). 'Political dynamics in organizational
identity breach and reconstruction: Findings from the crisis in UNICEF Germany'.
Academy of Management Journal.
James, L. R., Demaree, R. G., and Wolf, G. (1993). 'rwg: An assessment of within-group
interrater agreement'. Journal of Applied Psychology, 78, 306-309.
Jaworski, B. J. (1988). 'Toward a theory of marketing control: Environmental context, control
types, and consequences'. Journal of Marketing, 52, 23-39.
Johnson, P. and Gill, J. (1993). Management control and organizational behaviour, London,
U.K.: Sage Publications.
Kandel, E. and Lazear, E. P. (1992). 'Peer pressure and partnerships'. Journal of Political
Economy, 100, 801-817.
Kirkman, B. L. and Rosen, B. (1999). 'Beyond self-management: Antecedents and consequences
of team empowerment'. Academy of Management Journal, 42, 58-74.
Kirsch, L. J. (1996). 'The management of complex tasks in organizations: Controlling the
systems development process'. Organization Science, 7, 1-21.
Kirsch, L. J. (2004). 'Deploying common systems globally: The dynamics of control'.
Information Systems Research, 15, 374-395.
Kirsch, L. J., Ko, D.-G., and Haney, M. H. (2010). 'Investigating the antecedents of team-based
clan control: Adding social capital as a predictor'. Organization Science, 21, 469-489.
Klein, K. J., Tsoi, H., and Cannella, A. A. (1999). 'Multilevel theory building: Benefits, barriers,
and new developments'. Academy of Management Review, 24, 243-248.
Kohli, A. K. and Jaworski, B. J. (1994). 'The influence of coworker feedback on salespeople'.
Journal of Marketing, 58, 82-94.
Kohli, A. K., Shervani, T. A., and Challagalla, G. N. (1998). 'Learning and performance
orientation of salespeople: The role of supervisors'. Journal of Marketing Research, 35,
Kownatzki, M., Walter, J., Floyd, S. W., and Lechner, C. (2013). 'Corporate control and the
speed of SBU-level decision making'. Academy of Management Journal, 56, 1295-1324.
Kreutzer, M., Cardinal, L. B., Walter, J., and Lechner, C. (2016). 'Formal and informal controls
as complements or substitutes? The role of the task environment'. Strategy Science, 1,
Kreutzer, M., Walter, J., and Cardinal, L. B. (2015). 'Organizational control as antidote to
politics in the pursuit of strategic initiatives'. Strategic Management Journal, 36, 1317-
Lange, D. (2008). 'A multidimensional conceptualization of organizational corruption control'.
Academy of Management Review, 33, 710-729.
LeBreton, J. M. and Senter, J. L. (2008). 'Answers to 20 questions about interrater reliability and
interrater agreement'. Organizational Research Methods, 11, 815-852.
LePine, J. A. and van Dyne, L. (2001). 'Peer responses to low performers: An attributional model
of helping in the context of groups'. Academy of Management Review, 26, 67-84.
Locke, E. A. (1976). 'The nature and causes of job satisfaction'. In Dunnette, M. D. (Ed),
Handbook of industrial and organizational psychology, Chicago: Rand McNally, 1297-
Loughry, M. L. (2010). 'Peer control in organizations'. In Sitkin, S. B., Cardinal, L. B., and
Bijlsma-Frankema, K. M. (Eds), Organizational control, Cambridge, U.K.: Cambridge
University Press, 324-361.
Loughry, M. L. and Tosi, H. L. (2008). 'Performance implications of peer monitoring'.
Organization Science, 19, 876-890.
Manz, C. C. and Sims, H. P. (1987). 'Leading workers to lead themselves: The external
leadership of self-managing work teams'. Administrative Science Quarterly, 32, 106-129.
Merchant, K. A. (1985). Control in business organizations, Marshfield, MA: Pitman.
Nunnally, J. C. (1978). Psychometric Theory (2 ed.), New York: McGraw-Hill.
Ocasio, W. and Wohlgezogen, F. (2010). 'Attention and control'. In Sitkin, S. B., Cardinal, L. B.,
and Bijlsma-Frankema, K. M. (Eds), Organizational control, Cambridge, U.K.:
Cambridge University Press, 191-221.
Ouchi, W. G. (1979). 'A conceptual framework for the design of organizational control
mechanisms'. Management Science, 25, 833-848.
Ouchi, W. G. and Maguire, M. A. (1975). 'Organizational control: Two functions'.
Administrative Science Quarterly, 20, 559–569.
Podsakoff, P. M., MacKenzie, S. B., Lee, J. Y., and Podsakoff, N. P. (2003). 'Common method
biases in behavioral research: A critical review of the literature and recommended
remedies'. Journal of Applied Psychology, 88, 879–903.
Podsakoff, P. M., MacKenzie, S. B., and Podsakoff, N. P. (2012). 'Sources of method bias in
social science research and recommendations on how to control it'. In Fiske, S. T.,
Schacter, D. L., and Taylor, S. E. (Eds), Annual Review of Psychology, Vol 63, Vol. 63:
Sewell, G. (1998). 'The discipline of teams: The control of team-based industrial work through
electronic and peer surveillance'. Administrative Science Quarterly, 43, 397-428.
Sihag, V. and Rijsdijk, S. A. (2019). 'Organizational controls and performance outcomes: A
meta-analytic assessment and extension'. Journal of Management Studies, 56, 91-133.
Sitkin, S. B., Cardinal, L. B., and Bijlsma-Frankema, K. M. (2010). 'Control is fundamental'. In
Sitkin, S. B., Cardinal, L. B., and Bijlsma-Frankema, K. M. (Eds), Organizational
control, Cambridge, U.K.: Cambridge University Press, 3-15.
Snijders, T. A. B. and Bosker, R. J. (2012). Multilevel analysis: An introduction to basic and
advanced multilevel modeling, Thousand Oaks, CA: Sage.
Stewart, G. L., Courtright, S. H., and Barrick, M. R. (2012). 'Peer-based control in self-managed
teams: Linking rational and normative influence with individual and group performance'.
Journal of Applied Psychology, 97, 435-447.
Turner, K. L. and Makhija, M. V. (2006). 'The role of organizational controls in managing
knowledge'. Academy of Management Review, 31, 197–217.
Weibel, A. (2010). 'Managerial objectives of formal control: High motivation control
mechanisms'. In Sitkin, S. B., Cardinal, L. B., and Bijlsma-Frankema, K. M. (Eds),
Organizatioanl control, Cambridge, U.K.: Cambridge University Press, 434-462.
Welbourne, T. M. and Ferrante, C. J. (2008). 'To monitor or not to monitor: A study of
individual outcomes from monitoring one's peers under gainsharing and merit pay'.
Group & Organization Management, 33, 139-162.
Wenger, E. C. and Snyder, W. M. (2000). 'Communities of practice: The organizational frontier'.
Harvard Business Review, 139-146.
Wright, B. M. and Barker, J. R. (2000). 'Assessing concertive control in the team environment'.
Journal of Occupational and Organizational Psychology, 73, 345-361.
TABLE Ia: Means, Standard Deviations, and Bivariate Correlations of Individual-Level Variables
1. Female
2. Age (ln)
3. Organizational Tenure 3.18 1.33 0.17 0.37
4. Time Commitment (ln)
5. Professional Experience
6. Other NPO Engagement 0.39 0.49 -0.16 0.05 0.12 -0.10 -0.01
7. Team Size (ln)
8. Team Cohesiveness 3.94 0.71 -0.03 -0.02 0.08 0.06 -0.07 -0.06 0.08 0.791
9. Formal Manag. Control
10. Clan Control
11. Direct Peer Control (Notice & Praise) 3.74 0.77 -0.05 -0.10 -0.09 0.03 -0.07 0.06 0.09 0.52 0.22
12. Direct Peer Control (Correct & Discuss)
13. Indirect Peer Control 2.33 1.06 -0.05 -0.01 -0.05 0.03 -0.00 -0.04 0.04 -0.27 -0.10
14. Job Satisfaction
10. Clan Control
11. Direct Peer Control (Notice & Praise) 0.51 0.861
12. Direct Peer Control (Correct & Discuss)
13. Indirect Peer Control -0.29 -0.11 0.18 0.891
14. Job Satisfaction
N = 356; two-tailed tests; correlations with absolute value equal to or greater than 0.10 are significant at the p < 0.05 level.
1 Consistency reliability (Cronbach alphas).
Coding of control variables: female: dummy variable, which assumes a value of 1 for female members and 0 for male members; age: natural logarithm of the team
members’ age in years; organizational tenure and professional experience: scale from 1 to 6 (1 = 0-1 years; 2 = 1-5 years; 3 = 6-10 years; 4 = 11-15 years; 5 = 16-
20 years; 6 = > 20 years); time commitment: natural logarithm of the average number of hours invested by members per month; other NPO engagement: dummy
variable, which assumes a value of 1 if that respondent currently volunteers at another NPO; team size: natural logarithm of the number of active team members in
each team; team cohesiveness: Likert-type scale with 1 = strongly disagree to 5 = strongly agree.
TABLE Ib: Means, Standard Deviations, and Bivariate Correlations of Team-Level Variables
1. Female
2. Age (ln)
3. Organizational Tenure 3.23 0.76 0.17 0.49
4. Time Commitment (ln)
5. Professional Experience
6. Other NPO Engagement 0.42 0.20 0.02 0.22 0.21 0.01 0.15
7. Team Size (ln)
8. Team Cohesiveness 3.90 0.37 0.16 0.26 0.34 -0.29 0.14 -0.23 0.14
9. Formal Manag. Control
10. Clan Control
11. Direct Peer Control (Notice & Praise) 3.77 0.33 -0.02 0.29 0.16 -0.02 0.14 -0.17 0.04 0.42 0.19
12. Direct Peer Control (Correct & Discuss)
13. Indirect Peer Control 2.22 0.49 -0.19 -0.40 -0.15 0.05 -0.23 0.08 0.18 -0.10 0.05
14. Team Performance
15. Team Profits (ln)
10. Clan Control
11. Direct Peer Control (Notice & Praise)
12. Direct Peer Control (Correct & Discuss) 0.24 0.08
13. Indirect Peer Control
14. Team Performance
15. Team Profits (ln) -0.06 -0.20 -0.14 0.22 0.46
N = 58; two-tailed tests; correlations with absolute value equal to or greater than 0.26 are significant at the p < 0.05 level.
1 Consistency reliability (Cronbach alpha). 2 Means and standard deviations for team-level variables.
Coding of control variables: female: dummy variable, which assumes a value of 1 for female members and 0 for male members; age: natural logarithm of the team
members’ age in years; organizational tenure and professional experience: scale from 1 to 6 (1 = 0-1 years; 2 = 1-5 years; 3 = 6-10 years; 4 = 11-15 years; 5 = 16-
20 years; 6 = > 20 years); time commitment: natural logarithm of the average number of hours invested by members per month; other NPO engagement: dummy
variable, which assumes a value of 1 if that respondent currently volunteers at another NPO; team size: natural logarithm of the number of active team members in
each team; team cohesiveness: Likert-type scale with 1 = strongly disagree to 5 = strongly agree.
TABLE II: Multilevel Modeling (MLM) Regression Results
Model 1: Direct
Peer Control (No-
tice & Praise)
Model 2: Direct
Peer Control (Cor-
rect & Discuss)
Model 3: Indirect
Peer Control
Model 4: Job
Model 5: Team Per-
formance 1
Model 6: Team
Profits 1
Control Variables
Age (ln)
Organizational Tenure
Time Commitment (ln)
Professional Experience
Other NPO Engagement
Team Size (ln)
Team Cohesiveness
Independent Variables
Formal Manag. Control
Clan Control
Direct Peer Control
(Notice & Praise)
Direct Peer Control
(Correct & Discuss)
Indirect Peer Control
(Level-1 pseudo) R 2
(Level-1) N
Notes: Unstandardized coefficients shown, with robust standard errors (S.E.) in parentheses, based on random coefficient regression models using multilevel modeling (MLM).
Variance explained calculated as pseudo R 2 = 1 (level-1 restricted error + level-2 restricted error)/(level-1 unrestricted error + level-2 unrestricted error) (Snijders and Bosker,
p < 0.10; * p < .05; ** p < .01; *** p < .001.
1 Control and independent variables at the team level.
FIGURE 1: Multilevel Modeling (MLM) Regression Results
Appendix: Measurement Items
Adapted from
Please indicate your level of agreement with the following statements:
1. Team members liked the work that the team does.
2. Members of the team got along well.
3. Members of the team enjoyed spending time together.
Loughry and
Tosi (2008)
Please indicate your level of agreement with the following statements:
1. The team leader specified the processes or procedures which the team
had to follow during their activities.
2. The team leader specified the procedures (e.g., how to raise donations)
used by the team.
3. The team leader determined work assignments for individual team mem-
4. There were clear, planned goals and objectives set for this team by the
team leader.
5. The team leader specified concrete goals for specific activities (e.g., the
number of Christmas markets covered).
6. The team leader specified the product quality objectives for specific ac-
tivities within this team (e.g., the quality of the content of talks in the in-
formation work).
Bonner, Ruekert,
and Walker
Clan Control
Please indicate your level of agreement with the following statements:
1. The common vision of helping children influenced how team members
2. All team members know [organization]’s vision and act accordingly.
3. Shared norms and values based on [organization]’s vision influenced
team behaviors.
Kirsch et al.
Direct Peer
How often did members of your team…
1. see what other members did at their job? [notice]
2. notice what other members were doing at their job? [notice]
3. let others know that a team member had done good work? [praise]
4. tell team members that they did a good job? [praise]
5. take action if a team member had done the job incorrectly? [correct]
6. let team members know if they were doing something wrong? [correct]
7. talk about how team members did their job? [discuss]
8. discuss how everyone had performed at their jobs? [discuss]
Loughry and
Tosi (2008)
Indirect Peer
How often did members of your team…
1. get angry with unreliable members and gossip about them with their
friends on the team? [gossip]
2. gossip about it with their friends on the team if a team member has re-
peatedly let others down? [gossip]
3. refuse to socialize with team members who are unreliable? [avoid]
4. avoid team members who let others down repeatedly? [avoid]
Loughry and
Tosi (2008)
Job Satisfaction
Please indicate your agreement with the following statements:
1. All in all, I am satisfied with my volunteer job at [organization].
2. In general, I like my volunteer job at [organization].
3. In general, I like working as a volunteer at [organization].
Boezeman and
Ellemers (2009)
1. How advanced is this team’s (organizational and procedural) way of
2. How do you assess this team’s quality of information and media work?
3. How do you assess this team’s overall performance?
... On the one hand, direct peer monitoring involves noticing peers' results and behaviours and responding directly and openly to them. That is, managers examine each other's actions and their achievements or failures and openly discuss how work gets done (Lye et al., 2021;Walter et al., 2021). Direct peer monitoring also comprises the discussion of how peers do their job, praising peers whose performance is above expectations and correcting underperforming peers. ...
... The consequence of peer monitoring on goal commitment is still an open question in the literature and the prior empirical evidence is mixed (Stewart et al., 2012;De Jong et al., 2014;Walter et al., 2021). On one hand, prior work argues that peer monitoring could reduce task response time, increasing workload and, also, curbing creative approaches to problems. ...
... Hence, peer monitoring could lead to hostile relations, creating an unpleasant work environment. On the other hand, peer control is associated with more autonomy and empowerment, also enabling peers to detect dishonest behaviours, reducing free-riding and providing more opportunities to influence each other (Walter et al., 2021). Peer monitoring influences goal commitment by conveying normative information, persuading, highlighting role models or generating competition (Latham and Locke, 1991). ...
Full-text available
We examine whether the quality of performance metrics affects informal peer monitoring and, in turn, goal commitment. By fostering performance‐oriented behaviours, performance metrics drive managers to involve themselves in learning and improvement efforts, building a fertile atmosphere for informal peer monitoring. We argue that the quality of performance metrics is positively associated with direct peer monitoring and negatively linked to indirect peer monitoring. Subsequently, we postulate that direct (indirect) peer monitoring is positively (negatively) associated with goal commitment. We use partial least squares (PLS) to analyse survey data from store managers in a large retail firm. Results provide overall support for our hypotheses.
... Yang et al. (2010) find that the extent to which supervisors display transformational leadership behaviour directly influences subordinates' transformational leadership behaviour. Walter et al. (2021) find that when team managers use more formal controls, members in their team engage in higher levels of peer control towards their team peers. Mayer et al. (2009) find a significant relationship between top management and supervisory ethical leadership. ...
... This cascading effect helps in understanding how power flows throughout hierarchical organizations (Grant, 1996;Garicano, 2000;Dobrajska et al., 2015). Using social learning theory (Bandura, 1986;Walter et al., 2021), we posit that the authority which is received by a manager is passed on to their subordinates. We find a strong positive relationship between decentralization and autonomy. ...
Full-text available
We investigate how managers trade off the benefits of delegating authority to their employees with the risk of loss of control. Organizational economics theory identifies specific knowledge of subordinates and monitoring possibilities for the manager as determinants of delegation. Social learning theory predicts that when unit managers are themselves granted more authority, they will pass this on to their employees. This cascading of authority reduces the fear of loss of control associated with delegation. Using a survey among 215 unit managers in professional services firms, we find that managers delegate more authority to employees in their unit when those employees have more specific knowledge, when there are more exceptions in employee tasks, and when monitoring costs are lower. We also find support for the cascading effect: decentralization to the manager is positively related to autonomy granted to employees, while it moderates the effects of specific knowledge and monitoring costs.
... Even if we consider the positive connotations of peer monitoring, such as reduced information asymmetry because of enhanced transparency (Palanski et al. 2011;Walter et al. 2021), the reinforcing impact of peer monitoring on employees with psychological ownership is unnecessary because the intrinsic motivation from psychological ownership already enhances their likelihood of engaging in organizational citizenship behaviors. Although we expect only a marginally positive effect of psychological ownership on organizational citizenship behavior under high peer monitoring, it diminishes relative to the same effect under low peer monitoring; high psychological ownership already encourages high organizational citizenship behavior. ...
Full-text available
Organizational citizenship behavior is a highly sought-after outcome. We integrate insight from the psychological ownership perspective and agency theory to examine how the juxtaposition of informal psychological mechanisms (i.e., ownership feelings toward an organization) and formal and informal governance mechanisms (i.e., employee share ownership, agency monitoring, and peer monitoring) influences employees' organizational citizenship behaviors. Our empirical results show that psychological ownership has a positive effect on organizational citizenship behavior. Contrary to the common belief that informal and formal mechanisms complement each other, we find that the positive influence of psychological ownership on organizational citizenship behavior is more pronounced when employee share ownership and agency monitoring is low compared to high. Implications for theory and future research are discussed.
... It comes along with competitive advantages like cost and time to market reductions for organizations (Wieland et al., 2016;Yu et al., 2017) or a greater flexibility in terms of working time and place, therefore more autonomy for workers (Feldman, 1989;Gregg, 2011;Symon and Pritchard, 2015). However, the implementation of digital technologies also enables control of workers (Ball, 2010;Fairweather, 1999;Kallinikos, 2011;Walter et al., 2021) due to the collection, analysis and distribution of performance data. This has already been investigated in previous studies (Curchod et al., 2020;Heiland, 2021a, New modes of control in digital work contexts 2021b; Ivanova et al., 2018;Shapiro, 2017). ...
Purpose Digital technologies comprehensively change work processes and working conditions. However, the use of digital technologies and the modes of collaboration between technologies and human workers differ in terms of specific work organization and automatization. Referring to the job demands-resources model (JD-R), this paper investigates job demands and resources from the workers' perspectives and develops a digital work typology according to dimensions of digitalization and forms of human–computer interaction (HCI). Design/methodology/approach The authors conducted a qualitative-empirical study with 49 interviews in four German production and logistics organizations, emphasizing different job demands and job resources for five digital work types identified. Findings The results indicate that job demands and resources are to be differentiated in relation to specific work contexts. In this sense, this paper presents an analysis of dimensions of technology use and the impact of technology use on working conditions through empirically analyzing job demands and resources in digital work settings. Originality/value The contribution of this paper is to empirically analyze job demands and resources in digital work settings from the workers' perspectives and to develop a digital work typology based on the dimensions of digitalization and form of HCI. This typology can set the basis for further research insights as well as management practice measures in human resources management (HRM).
... In the last 3 decades, rule breaking conduct has been a significant and remarkable research subject. Morrison (2006) distinct rule breaking as "any instance where an employee intentionally violates a formal organizational policy, regulation, or ban with the primary purpose of promoting the welfare of organization or one of its stakeholders" Its features are the absence of self-benefit and the punitive offence that they may experience if certain policies or procedures are not followed and if they behave in the benefit of the corporation (Walter et al., 2021). It is about a person who works far beyond policies that he/she feels are false and assumes personal responsibility to violate the rules for the organization's advantages. ...
Full-text available
Although all forms of resources are necessary for organizational success, the literature seems to agree that human capital and its management have a substantial influence on the performance of corporations. In today's dynamic world, where communication and technologies continually grow, the actions and behavior of employees in their organizations are gaining importance with regard to the workers’ performance, SHRM, and servant leadership positions. In this study, the impact of SHRM, servant leadership, and rule control on Job Satisfaction and rule-breaking behavior of employees have been analyzed. Our study consists of a sample population of 287 white collar employees in the public and private sectors of Pakistan. Analysis was carried out using SPSS and AMO programs. Hypotheses were tested by the structural equation model. Finding reveals that SHRM has a positive and significant influence on job satisfaction but an insignificant impact on the rule-breaking behavior of employees. Servant leadership positively influenced job satisfaction and significantly moderates the association between SHRM and job satisfaction indicating that the servant attitude of the manager increases the satisfaction of the personnel. Rules control plays a significant mediating role. The current study contributed that employee exhibit greater levels of persistence, competence, and competitiveness while they are motivated by servant leaders. Managing workers from different backgrounds and with different work ethics require proactive leadership that disciplines organizational life according to predetermined guidelines for job involvement.
Distributed teams are a reality for several companies nowadays, many authors covered their benefits and problems, and the rate of adoption of such team's structure by companies is growing fast. Since these teams are more present in companies, a performance measurement system must get adapted to fulfill the gap of not having a vast theory about the subject. To fill that gap, this paper brings results from previous steps in the research (Systematic Literature Review and Qualitative analysis of the data). It presents to a group of experts to reach a consensus on which capabilities are essential to managing/developing distributed teams’ performance. The experts were exposed to the information following a Delphi Panel format and provided output that reached consensus and refined the list. The experts indicated that a group of six capabilities (engagement, development of a culture of performance measurement, organizational learning, alignment between planning and execution, accurate information and consistency) are essential to have their performance measurement system working correctly and reaching all functions. The work also identified the success factors for virtual teams, providing directions for the adoption and the monitoring of this kind of team that gained importance during the COVID-19 pandemic.
Purpose Digitalization is changing organizations with positive and negative impacts such as increased autonomy on the one hand and increased surveillance and control on the other hand. In this context, new modes of control occur: in addition to managerial control, new modes of control are multi-directed, stemming from colleagues, customers and underlying algorithms. This paper investigates the interrelation of autonomy and new modes of control in digital work contexts from the workers’ perspectives. Design/methodology/approach Empirical data are based on a mixed-methods approach combining qualitative interviews with 25 and a quantitative questionnaire with 127 workers from urban food logistics organizations in Germany. Findings The results show that new modes of control are relevant for work engagement in digital work contexts: managerial and algorithm control are perceived as support. Peer and customer control are perceived as coercion. Originality/value Besides investigating the interrelation of autonomy and control and differentiating new modes of control, our study also makes important contributions to the perception of control as support and coercion.
We examine how the complementarity of control formalization and control flexibility influences organizational performance across contexts of varying competitive turbulence. We build contingency arguments anchored in the efficiency logic of control theory and investigate both the restrictive and facilitative views of control formalization. Our empirical evidence is based on a survey of top executives from 536 organizations across the United States, Australia, China, and Israel. We find that control formalization and control flexibility are complementary in environments of low competitive turbulence. With increasing turbulence, the complementarity diminishes and shifts toward substitutive effects. JEL Classification: L2
We investigate the efficacy of organizational control interactions in contexts with varying levels of external environmental uncertainty. Specifically, we examine the contingent effect of external environmental uncertainty on the interactions of behavior and outcome control, behavior and clan control, and outcome and clan control on organizational performance. The empirical evidence draws from two temporally sequenced surveys of top executives in 203 firms. We theorize and find that these control combinations are complementary when the level of environmental uncertainty is low, and that this complementarity diminishes as environmental uncertainty increases. Further, in highly uncertain external environments, our findings reveal that higher performance is achieved through a high level of one organizational control type and a low level of the other, indicating a shift toward substitutive effects. Consequently, this study informs the complement-substitute debate in organizational control theory by explicitly investigating external environmental uncertainty as a contextual contingency.
Full-text available
This paper elucidates the political dynamics of organizational identity breach and reconstruction. Drawing on the revealing case of UNICEF Germany, we develop a process model of four phases of identity breach and reconstruction: build-up of identity tensions, revelation of identity breach, identity reconstruction, and enactment of the reconstructed identity. Our analysis explains how each phase was characterized by specific political strategies employed by the managerialist and idealist organizational members, the effects of these strategies on the power balance and consensus/conflict over organizational identity, and how particular triggers explain shifts from one phase to another. By so doing, our analysis advances our understanding of organizational identity breach and reconstruction as political processes and paves the way for new studies of political identity dynamics in other contexts.
Full-text available
Managing employees and external partners effectively has been a primary concern for organizations and their managers. Many studies have investigated the effectiveness of organizational controls in a wide variety of contexts. Using organizational controls literature that discriminates among outcome, behavior, and clan control, this study synthesizes the research on the effectiveness of these controls. In particular, the study examines 23,839 organizational controls–performance relationships from 120 independent samples, and tests several new hypotheses using advanced meta‐analytic methods. The results indicate that outcome, behavior, and clan controls generally enhance performance, with each control having a distinct performance effect. Our analysis also demonstrates that controls function as complements to one another. This finding indicates that one form of control increases the effectiveness of other forms of control. We also examine the organizational controls–performance relationships across various contexts, and our results show that they vary according to the type of task. The paper concludes with a discussion on the theoretical and managerial implications of these findings. This article is protected by copyright. All rights reserved.
Full-text available
The traditional view of control in organizations largely implies an “either-or” substitution logic, as opposed to the complementarity logic implied in the more recent view of control. This study examines whether formal and informal controls complement or substitute each other in their influence on performance outcomes, and whether such an interaction differs for more or less exploratory tasks. Our findings from an analysis of 184 strategic initiative teams in a cross-industry multicountry sample of firms support the complementary view. More specifically, we find support for our hypotheses that the combined use of formal and informal control has a positive impact on the performance of initiative teams, and that this complementary effect is more pronounced when the degree of exploration is lower. Accordingly, our study contributes to the organizational control literature both theoretically—by providing an explicit theoretical rationale for the complementary view—and empirically—by virtue of providing an empirical test of the interactive effects of formal and informal control.
This article draws on an ethnographic study of volunteer work in a German refugee shelter to explore how individual experiences of meaningfulness are intertwined with shifting discursive and organisational contexts. At the beginning of the so‐called refugee crisis, societal discourses portrayed this volunteer work as extraordinarily meaningful – a state we capture through the metaphor of ‘overflow’. This ‘overflow’ mobilised volunteers and was an important point of reference for framing their work experiences as meaningful. Later, shifting discursive and organisational contexts challenged their framings. Instead of letting go, however, the ‘overflow’ triggered volunteers to reframe their experience in dysfunctional ways in order to sustain their sense of meaningfulness. This paper reveals how shifting societal discourses feed into individual experiences of meaningfulness, shows how individuals may respond to such shifts in problematic ways and theorises the nature of such shifts in drawing on Swidler's notion of settling contexts. This article is protected by copyright. All rights reserved.
Organizational control is a key managerial function, and the focus of a great deal of research in the management and organizations field. Our concern is that research has not kept pace with changes in contemporary organizations and the external environment. In response to this concern, we review extant empirical work on organizational control with an emphasis on the consequences of control (i.e., the control-outcome linkage). As part of our analytical process, we surface theories underlying existing control frameworks used in empirical research and identify key dimensions implied by the frameworks. The three dimensions of control formality, control coerciveness, and control singularity map onto traditional vs. more current issues in and around organizations, and therefore prove helpful in assessing the existing research stream. Based on our review, we show how control-outcome research has in fact reached a troublesome point in its evolution, particularly concerning quantitative research. Older frameworks, theories, and issues seem to have limited theorizing that better fits today's realities, and several empirical tactics appear to be negatively affecting quantitative work. We close with actionable suggestions for an area of scholarship that continues to have great potential.
The statistical tests used in the analysis of structural equation models with unobservable variables and measurement error are examined. A drawback of the commonly applied chi square test, in addition to the known problems related to sample size and power, is that it may indicate an increasing correspondence between the hypothesized model and the observed data as both the measurement properties and the relationship between constructs decline. Further, and contrary to common assertion, the risk of making a Type II error can be substantial even when the sample size is large. Moreover, the present testing methods are unable to assess a model's explanatory power. To overcome these problems, the authors develop and apply a testing system based on measures of shared variance within the structural model, measurement model, and overall model.
We study a population of first year midshipmen within an elite military academy to explore the relationship between individuals' sociometric status (e.g., status conferrals based on positive interpersonal affect and perceived competence, and status degradations based on negative interpersonal affect) and their attempts to directly control their peers' behavior over a year's time. Results show that multiple informal sociometric status hierarchies develop early in the organization's life and remain remarkably stable. Control attempts are driven by these status hierarchies: Lower competence status individuals and those who attract negative status degradations are targeted for control by more people early in the group's life, those relatively free of negative status degradations attempt to control greater numbers of others throughout the group's existence, while higher positive status is generally unrelated to control attempts. However, control attempts do not lead to higher future sociometric status, suggesting they are not status signals. Findings also show that individuals targeted for control by many others leave the organization entirely. This article is protected by copyright. All rights reserved.
It is the objective of this study to describe three fundamentally different mechanisms through which organizations can seek to cope with this problem of evaluation and control. The three will be referred to as markets, bureaucracies, and clans. In a fundamental sense, markets deal with the control problem through their ability to precisely measure and reward individual contributions; bureaucracies rely instead upon a mixture of close evaluation with a socialized acceptance of common objectives; and clans rely upon a relatively complete socialization process which effectively eliminates goal incongruence between individuals. This study explores the organizational manifestations of these three approaches to the problem of control.