It was expected that peer control mediates the positive relation between intrateam trust and team performance

(1)

The role of Intrateam Trust and Peer Control on Team Performance moderated by Team Learning Orientation and Team Identification

Esther Kamminga S3134679

esther.kamminga@student.rug.nl

Supervisor: dr. K. M. Bijlsma-Frankema Second assessor: dr. Y. Yuan

Master thesis MSc Human Resource Management Faculty of Economics and Business, University of Groningen

14-06-2020

(2)

Abstract

The aim of this research is to deepen the insight on the relation between trust and control by testing specific hypotheses about the nature of this relation. It was expected that peer control mediates the positive relation between intrateam trust and team performance. We proposed that team learning orientation moderates the relationship between intrateam trust and peer control, such that if team learning orientation is high, intrateam trust promotes peer control. We also proposed that team identification moderates the relationship between peer control and team performance, such that if team identification is high, peer control will promote team performance. To test our hypotheses, data was gathered among 34 teams of several Dutch organizations. The results suggest a negative mediation effect between intrateam trust and team performance, mediated by peer control. No moderation effect for team learning orientation was found. For team identification, the results showed an effect, however this relation was found to be negative.

Keywords: Intra-team Trust, Peer Control, Team Performance, Team Learning Goals, Team Identification

(3)

The role of Intrateam Trust and Peer Control on Team Performance moderated by Team Learning Orientation and Team Identification

The idea that trust and control are two related mechanisms of governance has caught attention after extensive scholarly focus on control as a governing mechanism for organizational relations (Bijlsma-Frankema & Costa, 2005). Whereas in the past, control used to dominate organizational relations, increasing flexibility of organizational relations have been related to the growing importance of trust besides control (e. g.

Zucker, 1986; Kramer, 1996; Lewicki & Bunker, 1996). The hierarchical relationships that used to dominate the framing of work relations have made room for more lateral work relations (Bachmann, Knights, & Sydow, 2001; Sheppard & Tuchinsky, 1996), decreasing the effectiveness of conventional control systems (Singh, 2008).

Given the changed organizational environment, it is not surprising that research on the effectiveness of conventional organizational control has brought mixed results.

Positive outcomes of organizational control for the functioning of teams and organizations are found, for example employee compliance with rules (Tyler & Blader, 2005), teams reaching organizational goals (Korsgaard, Meglino & Jeong, 2010) and enrichment of knowledge exchange (Mayer & Argyres, 2004). Conversely, harmful effects of organizational control have been found, such as lower organizational commitment (Mechanic, 1962), resistance to change (Ezzamel & Willmott, 1998) and depersonalization towards the organization (Sundaramurthy & Lewis, 2003). Cardinal, Sitkin and Long (2004) explain these harmful effects of control as a result of overly authoritative control mechanisms that breed distrust and hostility among employees.

(4)

Peer control, a voluntary form of organizational control has become a highly relevant form of control in organizational context. In increased dynamic organizations, peer control is advocated to be more effective than conventional forms of organizational control (Henderson & Lee, 1992; Loughry, 2010). Peer control has been found to have many positive effects on performance (Langfred, 2004), enhanced feedback and coordination processes between team members (Marks & Panzer, 2004) and effort to reach team goals (Brown, Topping, Hennington & Skinner, 1999). Although peer control has shown to have many positive effects, negative consequences are also shown.

Research showed that members of organizations sometimes intentionally influence peers to work against the organization’s interest (Taylor, 1916; Roethlisberger & Dickson, 1939). For example, Westphal and Khanna (2003) showed that members of corporate boards used peer control to encourage fellow board members to act against the interests of the organization. Thus, despite the managerial interest in peer control, there is relatively little research into how, why and when peer controls impact organizational outcomes (De Jong, Bijlsma-Frankema & Cardinal, 2014; Loughry & Tosi, 2008).

Therefore, we will investigate conditions under peer control is exerted and will be received positively.

Intrateam trust is such a condition under which we expect peer control to be promoted. Intrateam trust has become increasingly important in work-relations (Costa &

Bijlsma-Frankema, 2007) and implies an attitude of vulnerability (Rousseau, Sitkin, Burt,

& Camerer, 1998). Dirks and Ferrin (2001) showed that intrateam trust, for instance, implies an increasing willingness to take risks (for example through information sharing).

In a high intrateam trust situation, peer control increases, because no negative

(5)

consequences from other team members are expected. However, intrateam trust has not always shown to be effective by itself: as control, trust can have harmful consequences since higher levels of trust can lead to deception (Lane, 2001) and may encourage parties to suspend judgment of others (Langfred, 2004), implying that team members take a less critical perspective, which may lengthen the period before mistakes are corrected.

Although the importance of trust and control for organizational governance is acknowledged, scholars have not yet found solid theoretical frameworks that can help to understand how trust and control in general relate (Long & Sitkin, 2006). The Journal of Management Studies has recently brought attention to this topic by pleading for more research on the relationship between trust and control (and its dynamics) (Lumineau, 2017; Long & Sitkin, 2006). Therefore, the aim of this research is to contribute to scholarly understanding of the trust-control relation by proposing and testing a specific relation between these two variables.

We will examine the effect of intrateam trust on peer control and, through peer control, on team performance. We examine two boundary conditions under which we expect the intrateam trust – peer control – team performance relation to be positively influenced: team learning orientation and team identification. We propose that team learning orientation (TLO) moderates the positive relationship between intrateam trust and peer control in such way that if TLO is high, this will enlarge the positive effect of intrateam trust on peer control. According to Bunderson and Sutcliffe (2003), teams high in TLO pursue shared goals. Shared team learning goals help team members to detect behavior of others which does not contribute to realization of these goals, and thus may enhance the promotion of peer control in high trust conditions. In this study we propose

(6)

that, given a certain level of intrateam trust, if a TLO is highly present in teams, members will exert peer control more and peer control will be more readily accepted. Thus, if TLO is high, the positive relationship between intrateam trust and peer control will be stronger.

We also expect that through peer control, behaviors to reach team goals will be promoted, with a positive effect on team performance.

The second boundary condition under which we expect a positive trust – peer control – team performance relation is team identification. In line with social identity theory (Tajfel, 1981), we propose that team identification moderates the relationship between peer control and team performance in such a way that if team identification is high, there will be a stronger positive relationship between peer control and team performance. Team identification is a strong motivational source, which can provoke extra-role behavior (Tyler & Blader, 2005). Extra-role behavior involves doing things that are not required, but that help the group (Tyler & Blader, 2005). If team identification is high, we expect that team members will exert extra effort for the team to succeed, because the teams’ success influences the team member’s self-esteem (Ellemers, De Gilder & Haslam, 2004). If team identification is low, we expect that peer control will also promote team performance, but the effect will be less strong.

Following from this, the research questions of this research are:

(1) Does intrateam trust through peer control promote team performance?

(2) Does team learning orientation moderate the relationship between intrateam trust and peer control?

(3) Does team identification moderate the relationship between peer control and team performance?

(7)

Theoretical Framework Intrateam Trust

Despite multiple definitions of trust, reported by diverse scholars (Costa &

Bijlsma-Frankema, 2007), most academics agree that trust is in essence a psychological state that demonstrates itself in the behavior towards others (Kramer, 1999). Trust is defined as “a psychological state comprising the intention to accept vulnerability based upon positive expectations of the intentions or behaviours of another” (Rousseau, Sitkin, Burt, & Camerer, 1998, p. 395). In this research, the concept intrateam trust is used to describe the level of trust that is present between team members. The term intrateam trust refers to the aggregated levels of trust that team members have in their fellow teammates (Langfred, 2004). Intrateam trust is therefore conceptualized as the “shared generalized perceptions of trust that team members have in their fellow teammates” (De Jong &

Elfring, 2010, p. 536).

In the definition of trust, positive expectations and intention to accept vulnerability are important to point out. Positive expectations refer to a belief in the trustee’s competence to perform well, a trustee’s goodwill towards the trustor and a trustees’ willingness to fulfill commitments to the trustor (Mayer, Davis, & Schoorman, 1995). Möllering (2005) states that intrateam trust implies the suspension of doubt about the possibility that another’s action will be based on self-interest, assuming that the other will reciprocate the signaled us-rationality. A willingness to accept vulnerability indicates the intention to depend on others (McKnight, Cummings & Chervany, 1998).

Accepting vulnerability means that something of significance can be lost, because the other has the power to take something valuable away. Therefore, the willingness to be

(8)

vulnerable is a risk-taking psychological state. Subsequently, trust is “not taking risk per se, but rather trust is a willingness to take a risk” (Mayer, Davis & Schoorman, 1995, p.

712).

Peer Control

Peer control is part of the broader construct organizational control. Organizational control is defined as “any process whereby managers direct attention, motivate, and encourage organizational members to act in ways desirable to achieving the organization’s objectives” (Cardinal, Sitkin & Long, 2010, p. 56). Peer control is a form of organizational control that appears when team members exert control over team members over whom they have no formal control (Loughry, 2010). Leifer and Mills (1996) define peer control as “a mechanism that is used to direct individual behavior toward organizational goals by using social influence, which points out the informal nature of this form of control. In this research, peer control is conceptualized as follows:

“Peer control occurs when work is organized in a fashion that encourages co-workers to monitor and influence each other’s level of productivity” (Howard, 1993: 241). Peer control is a voluntary behavior; peers can choose to exert peer control or refrain from it, because of the negative consequences peer control may have (Loughry, 2010). Negative consequences of peer control include employees influencing their team members to work against the organization’s interest (Taylor, 1916; Roethlisberger & Dickson, 1939). Also, peer control may be received as overstepping the controlee’s boundaries of authority (Welbourne & Ferrante, 2008). Furthermore, Westphal and Khanna (2003) showed the controlling team members risks unacceptance by those addressed.

Intrateam Trust, Peer Control and Team Performance

(9)

We propose that peer control is an extra-role behavior that can be very beneficial for teams and organizations. Trust has been found to promote several extra-role behaviors such as investing effort to reach team goals (De Jong et al., 2014), knowledge sharing (Rosendaal & Bijlsma-Frankema, 2013), cooperation with teams (Costa, 2003) and an increased willingness to take risks (Dirks & Ferrin, 2001). Similar to these behaviors, peer control (1) is a discretionary choice of employees, (2) cannot be enforced by command and control, and (3) has been shown to contribute to the performance of teams (De Jong et al., 2014). However, researchers have also found negative consequences of intrateam trust. For example, Lane (2001) found that intrateam trust could have harmful consequences, such as deception. According to Langfred (2004), trust can encourage team members to suspend judgment of others, which may lengthen the time to correct mistakes. Although negative consequences of intrateam trust have been found, in this research we will focus on the positive outcomes of intrateam trust. In a high intrateam trust situation, we expect the trustor to exert more peer control because the others will not perceive the peer control as harmful. However, in a low trust condition, employees choose to be silent instead of speaking up. Therefore, we expect that intrateam trust will promote peer control.

Hypothesis 1. Intratream trust promotes peer control.

Subsequently, we propose that peer control mediates positively between intrateam trust and team performance. Team performance is conceptualized as “the extent to which the productive output of a team meets or exceeds the performance standards of those who review and/or receive the output (De Jong, Bijlsma-Frankema & Cardinal, 2014, p.

1707). We expect that peer control promotes team performance, because peer control

(10)

allows team members to encourage their fellow members to perform well and prevent inappropriate behavior by increasing the chance that this behavior will be detected (Loughry & Tosi, 2008). In accordance with De Jong, Bijlsma-Frankema and Cardinal (2014), who argue that peer control is fundamental to the functioning of teams and organizations, we propose that peer control promotes team performance. This expectation results in hypothesis 2 and combined with hypothesis 1 in hypothesis 3.

Hypothesis 2. Peer control promotes team performance.

Hypothesis 3. Peer control mediates between intrateam trust and team performance.

The Moderating Effect of Team Learning Orientation

Edmondson (1999) defines team-learning behavior as “an ongoing process of reflection and action, characterized by asking questions, seeking feedback, experimenting, reflecting on results, and discussing errors or unexpected outcomes of actions” (p. 353). Bunderson and Sutcliffe (2003) define team learning orientation (TLO) as “a shared perception of team goals related to learning and competence development;

goals that guide the extent, scope and magnitude of learning behaviors pursued within a team” (Bunderson & Sutcliffe, 2003, p. 553). To cover team learning at the team level, the definition of Bunderson and Sutcliffe (2003) will be followed in this research.

As was previously mentioned, we expect that intrateam trust will promote peer control. We also propose that TLO will strengthen this positive relationship. TLO offers a shared perception of the appropriate way to behave (Chadwick & Raver, 2015), encourages proactive learning (Bunderson & Sutcliffe, 2003) and may help detect behavior that is not in line with realization of these goals. In other words, TLO provides

(11)

direction to peer control. Through TLO it becomes clear which behaviors are not in line with the goals and when TLO is high, we expect that peer control will be more effective in the realization of team goals and thus will have a positive effect on team performance.

If TLO is low, the relationship between intrateam trust and peer control will be weaker.

This leads to hypothesis 4.

Hypothesis 4. Team learning orientation strengthens the positive relationship between intrateam trust and peer control.

The Moderating Effect of Team Identification

Team identification is in this research conceptualized as “the emotional significance that members of a given group attach to their membership in that group”

(Van der Vegt & Bunderson, 2005, p. 533). Ellemers, Kortekaas and Ouwekerk (1999) state that team identification regulates the tendency of team members to behave in terms of their group membership. Tajfel’s (1981) social identity theory supposes that the self- concept consists of both a personal and a social identity. If team identification is high, part of the individual’s self is explained by the identity of the team. We expect that in teams with high team identification, the positive relation between peer control and team performance will be stronger. High team identification implies that team members will put in extra effort for the team to succeed, because the teams’ success will have a positive influence on the team member’s self-esteem (Ellemers, De Gilder & Haslam, 2004). The success of the team will subsequently have a positive effect on the self-esteem of the individual employee, leading him or her to put in extra effort. The positive relationship between intrateam trust and team performance, through peer control will therefore be even stronger when team identification is high. This proposition leads to hypothesis 5.

(12)

Hypothesis 5. Team identification strengthens the positive relationship between peer control and team performance

The conceptual model contains the content and relationship of the sections above (see figure 1). The model proposes that intrateam trust promotes team performance through peer control. The relationship between intrateam trust and peer control is proposed to be moderated by team learning goals. The relationship between peer control and team performance is moderated by team identification. This conceptual model helps to explain how and under which circumstances intrateam trust is expected to lead to higher team performance. Subsequently, the conceptual model is used to further test the hypotheses.

Figure 1. Conceptual Model

Methodology Procedure and Participants

For this study, online questionnaires were distributed among 55 teams across several Dutch organizations. Invitations to participate in the questionnaire were distributed by e-mail to 417 team members and to 52 direct supervisors of these teams.

(13)

Team members were contacted with the request to participate in a study on team development. They received a link to an online questionnaire containing measures of several variables, including the variables in our model, e. g. intrateam trust, peer control, team learning orientation and team identification. 302 team members responded to the questionnaire; a response rate of 72,4%. After verifying the dataset, 30 participants were excluded because they did not fully complete the questionnaire. Only teams of which at least 60% of its members responded were included in this research. This means 81 team members of teams with a score less than 60% were removed from the dataset. The final dataset consisted of 34 teams, with a total of 191 team members. The sizes of the teams ranged from 3 to 13 employees and the average team size was 6.50 (SD = 2.61). Ages of the team members ranged from 18 to 64 years (M = 42.03, SD = 12.24) and 51,3% of the team members were female (SD = 0.51).

To reduce common-source biases, we measured the dependent variable team performance using a different team leader survey (Podsakoff, MacKenzie, Lee, &

Podsakoff, 2003). Team leaders were contacted with the request to evaluate the team performance of their team members regarding the quality and quantity of their work. To prevent that the validity of any conclusions about the relationship between the independent variable, moderators and the dependent variable would suffer from common source bias, the dependent variable was thus drawn from a different source of data than the other constructs. 45 team leaders responded, which resulted in a response rate of 86,54%. Ages of the team leaders ranged from 23 to 65 years (M = 43.91, SD = 10.20) and 48,6% of the team leaders were female.

Measures

(14)

Intrateam trust. Intrateam trust was measured using a 5-item Likert scale (1 =

‘completely disagree’, to 7 = ‘completely agree’), developed by De Jong and Elfring (2010). Example items: ‘I am able to count on my team members for help if I have difficulties with my job’; ‘I am confident that my team members will take my interests into account when making work-related decisions’ (see appendix A for all the items). The Cronbach’s alpha of the 5-item scale was .87. For a reliably measurement, the Cronbach’s alpha should be above .80. Thus, the measurement of intrateam trust can be said to reliable. The five items were combined to one intrateam trust variable (M = 5.71, SD = .83).

Team learning orientation. Team learning orientation was measured using a 4- item Likert scale (1 = ‘completely disagree’; 7 = ‘completely agree’), developed by Bunderson and Sutcliffe (2003). Example items: ‘Our team looks for opportunities to develop new skills and knowledge’; ‘Our team likes challenging and difficult assignments that teach new things’. The Cronbach’s alpha of the 4-item scale was .86 and so formed a reliable scale of team learning orientation. The four items were combined into one team learning orientation variable (M = 5.46, SD = .96)

Team identification. Team identification was measured using a 4-item Likert scale (1 = ‘completely disagree’, to 7 = ‘completely agree’), from Allen and Meyer’s (1990) affective commitment scale as selected and translated by Van der Vegt, Van der Vliert and Oosterhof (2003). Example items: ‘I think it is great to be part of my team’; ‘I would like to continue working with my team’. The Cronbach’s alpha of the 4 items was .87. The scale formed a reliable representation of team identification. The four items were combined into one team identification variable (M = 5.46, SD = 1.07)

(15)

Peer control. Peer control was measured using a 7-item Likert scale (1 =

‘completely disagree’, to 7 = ‘completely agree’) with one item on a binary scale developed by G. Ockels, based on the scale in De Jong, Bijlsma-Frankema and Cardinal (2014). Every team member answered the question: “If this team member does something that I perceive as not right, I will speak up.” for every other member of the team. After data collection, for every team an average peer control score was calculated. (M = 5.87, SD = .54)

Team performance. Following De Jong, Bijlsma-Frankema & Cardinal (2014),

team performance was measured using a 11-item Likert scale. Supervisors were asked to give their team members a grade from 0 to 10 for the quantity and for the quality of their team members’ work (1 = ‘0”, to 11 = ‘10’). After data collection, for every team a score was calculated for quality of the work (M = 8.89, SD = .72) and quantity of the work (M

= 8.95, SD = .76). Quality and quantity of the work combined formed a team performance variable (M = 8.92, SD = .70).

(16)

Results Factor Analysis

Table 1: Factor Analysis with Varimax Kaiser rotation for 13 items Intrateam trust Team learning

orientation

Team identification

Intrateam trust 1 .77 .29

Intrateam trust 4 .84

Team learning orientation 1 .75 .24

Team learning orientation 2 .82

Team identification 1 .26 .28 .81

Team identification 2 .29 .23 .79

Team identification 3 .26 .85

Team identification 4 .31 .74

Note. Factor loadings < .2 were suppressed

To evaluate whether the intrateam trust, peer control and team performance scales were sufficiently discriminating, we conducted an exploratory factor analysis, a principal components analysis (PCA), on 13 items with Kaiser Varimax rotation. Coefficients with a value below .2 were suppressed. The analysis produced 4 components, with loadings between .74 and .85 for each measured construct and cross-loads below .31. The

(17)

eigenvalues of all four factors extracted explained 72% of the variance. Based on this factor analysis, all items were maintained.

Test for Multicollinearity

Table 2. Multicollinearity analysis

Tolerance VIF

Intrateam trust .58 1.73

Peer control .96 1.04

Team learning orientation .80 1.25

Team identification .55 1.81

Note: N = 191 team members, Dependent variable: Team performance

To reassure whether multicollinearity between predictors in the regression analysis did not affect our results, Variance Inflation Factors (VIF) were calculated.

Table 2 indicates that, with tolerance scores all above .5, there is no concern for multicollinearity.

(18)

Test for Normality

Table 3: Shapiro-Wilk test for normality

Descriptives Shapiro-Wilk test Skewness Kurtosis Statistic

Intrateam trust -1.30 2,30 .89**

Peer control -.45 -.39 .96**

Team performance -.01 -.39 .97**

Team learning orientation -.82 .41 .94**

Team identification -1.07 .98 .92**

Note. N = 182, ^**p <.001

To test whether the data used in this research was normally distributed, a Shapiro- Wilk test was conducted. The results of this test are shown in table 3. For all variables, the results were significant, indicating that the data for these variables is not normally distributed. Although the variables are not distributed normally, we choose to test our hypotheses using bias-corrected bootstrapping analysis (Hayes, 2013). Bootstrapping is a way to avoid the issue of non-normal sample distributions, by estimating the sample distribution based on the gathered data (Field, 2018). However, we need to take a critical perspective on the outcomes of the tests.

(19)

Descriptive Statistics

Table 4: Pearson Intercorrelations

Variables (1) (2) (3) (4) (5) (6) (7) (8)

(1) Quality of team performance - .81** -.04 -.22** .00 -.02 -.08 -.10 (2) Quantity of team performance - -.16* -.32** .00 -.13 -.01 .08

(3) Intrateam trust - .17* .36** .65** .11 -.09

(4) Peer control - -.02 .12 .14 -.24**

(5) Team learning orientation - .44** .10 .02

(6) Team identification - -.01 -.06

(7) Age - .11

(8) Gender -

Notes. N = 182, ^*p < .05, ^**p < .01

In table 4, the Pearson zero-order correlations of the variables are shown.

Correlations are used to show how variables are associated. The significant relation of peer control with intrateam trust (r = .17, p < .01) and both quality of team performance (r = -.22, p < .01) and quantity of team performance (r = -.32, p < .01) indicates the possibility of mediation. Furthermore, the correlations between independent variable intrateam trust and the moderator team learning orientation (r = .36, p < .01), intrateam trust and team identification (r = .65, p < .01) and team learning orientation and team identification (r = .44, p < .01) are positive and significant.

The correlation analysis also shows whether we need to take control variables into account for further analysis. Age and gender were considered as control variables. Table 4 shows that age does not significantly correlate with any main variable in the conceptual

(20)

model. Therefore, age is excluded for further analysis. Gender correlates significant negative to peer control (r = -.24, p < .01) but not with the dependent variable team performance for both quality and quantity and is therefore excluded for further analysis.

Multiple Regression Analysis

Before using the Hayes (2013) method to test our hypotheses, a multiple regression was performed in order to investigate a possible mediation effect (table 5 and table 6).

Table 5: Regression analyses results for Team performance (Quantity)

Regression I (DV = Quantity of team performance)

Regression II (DV = Peer control)

Regression III (DV = Quantity of team performance)

Regression IV (DV = Quantity of team performance

B SE t p B SE t p B SE t p B SE t p

Intrateam trust -.16 .06 -2.22 .03* .17 .07 2.30 .02* - - - - -.11 .06 -1.58 .12

Peer control - - - - - - - - -.32 .05 -4.50*** .00 -.30 .05 -4.19*** .00

R² .03 .03 .10 .11

N=182 team members; p ^ < .10, p* < .05, P** < .01; p*** < .001; Standardized regression coefficients are presented

Table 6: Regression analyses results for Team performance (Quality)

Regression I (DV = Quality of team performance)

Regression II (DV = Peer control)

Regression III (DV = Quality of team performance)

Regression IV (DV = Quality of team performance

B SE t p B SE t p B SE t p B SE t p

Intrateam trust -.04 .05 -.54 .59 .17 .07 2.30 .02* - - - - -.00 .05 -.05 .96

Peer control - - - - - - - - -.22 .05 -2.99** .03 -.22 .05 -2.93** .00

R² .00 .03 .05 .05

We did find a significant negative effect for intrateam trust on the quantity of team performance (B = -.16, p = .03). However, for quality of team performance, we did

(21)

not find a significant effect of intrateam trust (B = -.04, p = .59). We found a significant positive effect for the A-path in our mediation model: intrateam trust on peer control (B = .17, p = .02). Also, a significant negative effect was found for the B-path of our mediation model: peer control on both quantity of team performance (B = -.32, p < .001) and quality of team performance (B = -.22, p = .03).

Baron & Kenny (1986) provided four requirements that must be met to form a mediation relationship (hypothesis 3). First of all, there should be a significant relation between the independent variable (intrateam trust) and the dependent variable (team performance). Secondly, the relation between the independent variable (intrateam trust) and the mediator (peer control) should be significant. Third, also the relation between the mediator (peer control) and the dependent variable (team performance) should be significant. And fourth, the significant relation between the independent variable (intrateam trust) and the dependent variable (team performance) should be less significant (partial mediation) or not significant (full mediation), when the mediator (peer control) is added to the model.

In our research, the relationship between intrateam trust (IV) and quantity of team performance (DV) is significant. However, the relationship between intrateam trust (IV) and quality of team performance (DV) is not significant. The relation between intrateam trust (IV) and peer control (M) was also found to be significant. Furthermore, the relation between peer control (M) and team performance (DV) was significant. And finally, the relation between intrateam trust (IV) and team performance (DV) is less significant when peer control (M) is added to the model. Thus, the rules for full mediation are met for quantity of team performance, but not for quality of team performance. In other words,

(22)

with the Barron & Kenny method, a full mediation effect was found for quantity of team performance. For quality of team performance, no mediation effect was found using the Baron & Kenny method. In spite of that, we will use bias corrected bootstrapping (Hayes, 2013; model 4) to further study these effects and to check whether these results can be replicated.

Hypothesis Testing

In order to test the mediation and moderation effects in the conceptual model, the SPSS program Process (Hayes, 2013) was used. First, a mediation (model 4) and a moderated mediation (model 21) analysis were conducted. Furthermore, an additional moderation analysis is performed (model 1). Team performance was split up in quantity of team performance and quality of team performance, in order to obtain more detailed information about the way in which the constructs in our model relate to the different components of team performance. Findings of these analyses are shown in table 7, 8, 9 and 10.

(23)

Table 7. Regression results for the mediation effect for quantity of team performance (Hayes model 4)

Dependent variable model DV = Peer control R²= .03, p = .02*

Predictor B SE t p

Intrateam trust .17 .07 2.30 .02

Dependent variable model DV = Team performance (Quantity) R² = .11, p = .00***

Predictor B SE t p

Intrateam trust -.09 .06 -1.58 .12

Peer control -.23 .05 -4.19 .00***

Total effect model DV = Team performance (Quantity)

Effect SE t p

Direct effect of IV on DV -.09 .06 -1.58 .12 Indirect effects (mediation model)

Effect Boot SE BootLLCI BootUCLI

IV > M > DV -.04 .02 -.07 -.01

Table 7 shows a positive significant effect for intrateam trust on peer control (B = .17, SE = .07, p = .02). Furthermore, a significant negative effect for peer control on quantity of team performance is shown (B = -.23, SE = .05, p < .001). In addition, table 7 shows a significant mediation effect for the intrateam trust – peer control – quantity of team performance relation, displayed by the absence of zero in the 95% confidence interval [-.07; 0.01].

(24)

Table 8. Regression results for the mediation effect for quality of team performance (Hayes model 4)

Dependent variable model DV = Team performance (Quality) R² = .05, p = .01

Predictor B SE t p

Intrateam trust -.00 .05 -.05 .96

Peer control -.16 .05 -2.93 .00**

Total effect model DV = Team performance (Quality)

Effect SE t p

Direct effect of IV on DV -.00 .05 -.05 .96 Indirect effects (mediation model)

Effect Boot SE BootLLCI BootUCLI

IV > M > DV -.03 .01 -.06 -.00

Table 8 shows a significant negative effect for peer control on quality of team performance (B = -.16, SE = .05, p < .01). In addition, Table 8 shows a significant mediation effect for the intrateam trust – peer control – quantity of team performance relation, displayed by the absence of zero in the 95% confidence interval [-.06; -.00].

(25)

Table 9. Regression for the total model for quantity of team performance (Hayes model 21)

DV = Peer control, R² = ,04, p = .06

Predictor B SE t p LLCI ULCI

ITT .23 .08 2.71 .01* .06 .39

TLO -.08 .08 -1.07 .29 -.24 .07

ITT*TLO .06 .07 .87 .38 -.08 .21

DV = Team performance (Quantity), R² = .13, p = .00***

Predictor B SE t p LLCI ULCI

ITT -.10 .07 -1.34 .18 -.24 .05

PC -.23 .05 -4.29 .00*** -.34 -.13

TI -.04 .07 -.56 .57 -.18 .10

PC*TI -.11 .05 -2.00 .05* -.21 -.00

Index of moderated mediation

Mediator Index BootSE BootLL

CI

BootULCI

Peer control -.01 .01 -.03 .01

Conditional effects for peer control on team performance (quantity) at values of team identification

Mediator Effect SE T LLCI ULCI

-.90 low -.14 .07 -1.93 -.28 .003

-.27 medium -.26 .06 -4.61*** -.37 -.15

.97 high -.34 .08 -4.39*** -.49 -.18

Conditional indirect effects of intrateam trust on team performance at values of the moderators

Mediator TLO TI Effect BootSE BootLLCI BootULCI

Peer control -1.26 -.90 -.02 .02 -.06 .01

-1.26 .27 -.04 .02 -.09 .01

-1.26 .97 -.05 .03 -.11 .01

.31 -.90 -.03 .02 -.08 .00

.31 .27 -.06 .03 -.12 -.02

.31 .97 -.08 .03 -.16 -.02

.83 -.90 -.04 .03 -.10 .00

.83 .27 -.07 .03 -.15 -.01

.83 .97 -.09 .04 -.19 -.02

N=182 team members; p* < .05, P** < .01; p*** < .001; Standardized regression coefficients are presented

Table 9 shows a significant positive effect for intrateam trust on peer control (B = .23, SE = .07, p = .01). Also, table 9 shows that peer control is negatively related to

(26)

quantity of team performance (B = -.23, SE = .05, p < .001). Table 9 shows no result for a significance main effect for team learning orientation and neither for a significant interaction effect for intrateam trust and team learning orientation on peer control. A negative effect for the interaction of peer control and team identification on quantity of team performance is found (B = -.11, SE = .05, p = .05). The conditional effects of peer control on quantity of team performance are found to be significant on medium [-.37; - .15] and high [-.49; -.18] levels of team identification (95% confidence interval).

Furthermore, a significant negative mediation effect is found for the following levels of team learning orientation and team identification: medium TLO and medium team identification, medium TLO and high team identification, high TLO and medium team identification and high TLO and high team identification. For low TLO and low team identification, no significant mediation effect is displayed.

(27)

Table 10. Regression for the total model for quality of team performance (Hayes model 21)

DV = Team performance (Quality), R² = .07, p = .01

Predictor B SE t p LLCI ULCI

ITT -.03 .07 -.50 .62 -.17 .10

PC -.16 .05 -3.05 .00** -.26 -.06

TI -.01 .07 -.10 .92 -.14 .13

PC*TI -.11 .05 -2.17 .03* -.22 -.01

Index of moderated mediation

Mediator Index BootSE BootLLCI BootULCI

Peer control -.01 .01 -.03 .01

Conditional effects for peer control on team performance at values of team identification

Mediator Effect SE T LLCI ULCI

-.90 low -.06 .07 -.87 -.20 .08

-.27 medium -.19 .06 -3.46*** -.30 -.08

.97 high -.27 .07 -3.62*** -.42 -.12

Conditional indirect effects of intrateam trust on team performance (quality) at values of the moderators

Mediator TLO TI Effect BootSE BootLLCI BootULCI

Peer control -1.26 -.90 -.01 .01 -.03 .01

-1.26 .27 -.03 .02 -.06 .01

-1.26 .97 -.04 .02 -.09 .01

.31 -.90 -.01 .02 -.05 .01

.31 .27 -.05 .02 -.09 -.01

.31 .97 -.07 .03 -.13 -.02

.83 -.90 -.02 .02 -.07 .01

.83 .27 -.05 .03 -.11 -.01

.83 .97 -.08 .04 -.16 -.01

N=182 team members; R²=.10^***; p ^ < .1, p* < .05, P** < .01; p*** < .001;

Standardized regression coefficients are presented

Table 10 shows a negative relation between peer control and quality of team performance (B = -.16, SE = .05, p < .01). Furthermore, table 10 shows no result for a significance main effect for team learning orientation and neither for a significant interaction effect for intrateam trust and team learning orientation on peer control. A negative effect for the interaction of peer control and team identification on quality of

(28)

team performance is found (B = -.11, SE = .05, p = .03). The conditional effects of peer control on quantity of team performance are found to be significant on medium [-.30; - .08] and high [-.42; -.12] levels of team identification (95% confidence interval).

Furthermore, a significant negative mediation effect is found for the following levels of team learning orientation and team identification: medium TLO and medium team identification, medium TLO and high team identification, high TLO and medium team identification and high TLO and high team identification. For low TLO and low team identification, no significant mediation effect is displayed.

Hypothesis 1

Hypothesis 1 stated that intrateam trust promotes peer control. Zero-order Pearson correlations reported in table 4 gave an indication of the presence of a main effect between intrateam trust and peer control (r = .17, p < .05). Using model 4 of Hayes (2013), we found a significant positive effect of intrateam trust on peer control (B = .17, SE = .07, p < .05) (see table 7). Therefore, we can conclude that intrateam trust promotes peer control and we can accept hypothesis 1.

Hypothesis 2

Hypothesis 2 stated that peer control promotes team performance. Pearson correlations reported in table 4 gave an indication of a significant negative relation between peer control and team performance (quality) (r = -.22, p < .01) and team performance (quantity) (r = -.32, p < .01). Regression analysis (model 4) revealed that peer control decreases team performance, for both quality (B = -.16, SE = .05, p < .001) and quantity (B = -.23, SE = .05, p < .001) of team performance (see table 7 and 8).

(29)

Thus, a significance effect was found, but in the opposite direction of what was expected.

Therefore, hypothesis 2 was rejected.

Hypothesis 3

Hypothesis 3 stated that peer control positively mediates the relation between intrateam trust and team performance. The results show that, with the absence of zero in the 95% confidence interval, the indirect effect of intrateam trust on team performance mediated by peer control was found to be significant for both quality of team performance [ -.06; -.00] and quantity of team performance [-.08; -.01] (see table 7 and 8). A significant mediation was found, but in the opposite direction of what was expected. Hypothesis 3 was therefore rejected.

Hypothesis 4

Hypothesis 4 stated that team learning orientation would strengthen the positive relation between intrateam trust and peer control. The results failed to support hypothesis 4 because first no significant interaction effect was found for team learning orientation on the relation between intrateam trust and peer control (B = .06, SE = .07, p = .29) (Hayes model 21, see table 9). This was also shown by the presence of zero within the confidence interval CI95% [-.08; .21]. Therefore, hypothesis 4 was rejected.

Hypothesis 5

Hypothesis 5 stated that team identification would strengthen the positive relation between peer control and team performance. As indicated, a significant negative effect for the relation between peer control and team performance was found (7 and 8). A significant moderation effect was found for team identification on the relationship between peer control and team performance, for both quality of performance (B = -.11,

(30)

SE = .05, p < .05) and quantity of performance (B = -.11, SE = .05, p < .05). The confidence interval CI95% [-.22; -.00] also showed a significant effect, with the absence of zero in the interval (table 9 and 10). However, the effect is significantly negative.

Therefore, hypothesis 5 was rejected.

Discussion

The aim of this study was to contribute to scholarly understanding of trust and control as contemporary mechanisms of governance by proposing and testing a specific relation between these two variables. Due to developments in the organizational context of recent years, the role of control as an effective mechanism of governance has diminished and trust is increasingly seen as an important factor for organizational effectiveness. This research aimed to create a better understanding of the control trust nexus by investigating the effect of the moderator team learning orientation on the relationship between intrateam trust and peer control and the moderator team identification on the relationship between peer control and team performance.

The results confirmed the positive relation between intrateam trust and peer control (hypothesis 1), meaning that when intrateam trust is high, more peer control is exerted. This expectation was based on Dirks and Ferrin (2001), who argued that intrateam trust implies an increased willingness for risk-taking behaviors (such as peer control) and peer control will be exerted because no negative consequences from other team members are expected. Also, Rousseau et al. (1999) stated that trust would comprise positive expectations and the willingness to become vulnerable to actions of others. Thus, peer control would be exerted more, because no negative expectations of the control were to be expected.

(31)

The results did not confirm a positive relation between peer control and team performance (hypothesis 2). A significant relation was found, however, we found this relationship to be negative. We argued that peer control would promote team performance because peer control allows team members whose interests are aligned with those of the team to encourage their fellow-members to perform well and prevent inappropriate behavior by increasing the chance that it would be detected (Loughry &

Tosi, 2008). However, our findings suggest that the less peer control is exerted, the higher the team performance is. Thus, maybe the performance in these teams is higher if peer control is limited. It might be the case that team members experience a restricted form of peer control as appropriate. Barker (1993) argued that self-management could increase the team member’s stress levels, suggesting that employees who experience too much peer control might suffer from it individually. Another explanation of this negative effect for peer control may be that in a high trust situation, team members are comfortable with addressing each other directly, but the focus of these confrontations is not necessarily at improving team performance which may lead to norms regarding interactions that exclude peer control as appropriate behavior towards team performance.

The hypothesized positive mediating role of peer control between intrateam trust and team performance was not confirmed (hypothesis 3). First, intrateam trust positively related to peer control (hypothesis 1). However, peer control related negatively to team performance (hypothesis 2). Contrary to what was predicted, a positive mediation effect was not found. Although we could not confirm hypothesis 3, we did find a significant mediation effect for the intrateam trust – peer control – team performance relationship.

(32)

We found that intrateam trust promotes peer control, and peer control decreases team performance. This implicates that lower peer control will increase team performance.

The expectation that intrateam trust would promote more peer control when team learning orientation is high (hypothesis 4), was not supported by our results. We expected that team learning orientation would encourage proactive learning (Bunderson &

Sutcliffe, 2003) and would help detect behavior that is not in line with the team goals with the aim of improving team performance (Edmondson, 1999). Furthermore, we expected that team learning orientation would give direction to peer control and together with peer control, teams would be more effective in the realization of goals. However, we did not find a significant effect for team learning orientation as a moderator on the relationship between intrateam trust and peer control nor did we find a significant main effect for team learning orientation on peer control. An explanation could be that teams compromise team performance by overemphasizing on team learning, because learning effort consume resources and direct attention from existing primary tasks that contribute directly to team performance (Bunderson & Sutcliffe, 2003).

The expectation that team identification would moderate the positive relation between peer control and team performance (hypothesis 5) was not supported by the findings of this study. We based this expectation on the argument that team members higher in team identification would put in extra effort for the team to succeed, because the team’s success would have a positive influence on the team member’s self-esteem (Ellemers, De Gilder & Haslam, 2004). In our study, a significant negative effect for the moderation of team identification was found. A possible explanation for this negative effect of team identification could be that identifying with the team might not be a key

(33)

factor into being successful in the team. It might be that team members identify with the team on a social level, but do not link this identification to team performance.

Additional Research

Since we did not find a significant effect for the moderator team learning orientation (hypothesis 4) and we aimed to achieve better understanding into the trust – control – performance relationship, we decided to test for the effect of team learning orientation as a moderator on the relation between peer control and team performance.

We expected that team learning orientation would have a positive effect on the relation between intrateam trust and peer control because a team learning orientation can encourage adaptive behaviors that lead to improved performance (Bunderson & Sutcliffe, 2003). If team learning orientation is high, peer control will be understood as a means aimed to improve team performance.

Table 11. Moderation of team learning orientation on the relationship between peer control and quality of team performance (model 1)

Dependent variable model DV = Team performance (Quality) R² = .05, p = .03*

Predictor B SE t p

Peer control -.16 .05 -3.02 .00**

Team learning orientation -.01 .05 -.10 .92

PC*TLO .03 .05 .57 .57

(34)

Table 12. Moderation of team learning orientation on the relationship between peer control and quantity of team performance (model 1)

Dependent variable model DV = Team performance (Quantity) R² = .10, p = .00***

Predictor B SE t p

Peer control -.24 .05 -4.42 .00***

Team learning orientation -.00 .05 -.07 .95

PC*TLO .02 .06 .27 .79

No significant effect for team learning orientation as a moderator on the relation between peer control and quality of team performance (B = .03, SE = .05, p = .57) and quantity of team performance (B = .02, SE = .06, p = .79) was found (see table 10 and 11). Surprisingly, in this study we didn’t find a significant effect for the moderator team learning orientation in the model.

Practical Implications

An implication of this study is that peer control may decrease team performance.

Thus, teams may perform better if peer control is rather restricted. Low peer control may be enough to stimulate team performance, more peer control may show to have a counterproductive effect. Even though prior research identified possible positive effects of peer control (Langfred, 2004; Marks & Panzer, 2004; Brown, Topping, Hennington &

Skinner, 1999), our findings suggest negative effects of peer control. Multiple scholars have taken a critical perspective towards the supposed positive effects of peer control (e.

g. Taylor, 1916; Roethlisberger & Dickson, 1939; Welbourne & Ferrante, 2008;

Westphal & Khanna, 2003). For example, Barker (1993) addresses that team members felt more closely monitored by peers than they felt monitored by their supervisors. Barker

(35)

(1993) addresses that team members punished other team members who showed “bad attitudes”, made them feel wrong, and pressured them to conform, tolerating less undesirable behavior than the supervisors would allow.

As stated before, we introduced the moderator team learning orientation as a moderator that would have a positive effect on the relation between intrateam trust and peer control. However, a non-significant yet negative impact of team learning orientation was found in our study. An explanation for this negative effect of team learning orientation may be that in high-trust teams with a high team learning orientation, team members are less likely to control each other because they assume their team members will strive for learning too and have the same learning goals as they have themselves.

Different specific circumstances of peer control are likely to trigger different outcomes.

For example, other attributes that help to clarify the mixed results of peer control could be the appropriateness of peer control, or team culture, regulating which behavior is appropriate and which behavior is not. Thus, our findings suggest that organizations should carefully consider the ways control is exerted in teams to increase the benefits for organizational goals.

Furthermore, this study shows that intrateam trust promotes peer control.

Organizations, which aim to create an open environment in which organizational members freely approach each other, can considering doing so by encouraging intrateam trust. However, organizations should strive for forms of peer control that encourage organizational outcomes and take into account that peer control can have negative consequences.

(36)

Another finding of this research is that, against our expectations, peer control moderated by team identification decreases team performance. An explanation might be that team members identified with the team on a social level but did not link performance of the team to themselves on a performance level. This way, it might be possible that someone identifies with the group, but negative performance of the team does not influence the self-esteem of the team member. A mechanism that might explain the underlying effect is labeled as ‘group think’; “a defective decision-making process afflicting highly cohesive and conforming groups” (Janus, 1972). Manz and Sims (1982) argue that cohesive groups might be particularly vulnerable to groupthink, because the way these team strive for unanimity overrides the motivation to realistically appraise alternative courses of action. Especially in teams high in team-identification, it might be the case that team members strive for unanimity rather than team performance. In doing so, the chances that non-effective behavior is corrected becomes smaller and team performance is not likely to increase. Although many researchers found that team

identification lies at the heart of many social behaviors (Haslam, 2004), for supervisors it is therefore important to focus not only on team bonding, but also on other factors that might help to improve performance.

Strengths, Limitations and Future Research

Since this study did not support all hypotheses, addressing limitations of this study is paramount. In calling attention to these limitations, we simultaneously propose directions for future research. First of all, this study included only 34 teams with a total of 191 respondents. For future research, it would be interesting to test the hypotheses on a larger and random sampled set of teams, with the aim of increasing the statistical power

(37)

of the study. In addition, all data was gathered at the same moment. This may have led to an amplification of the correlational effects, because team members filled in all constructs (except for team performance) once and may have given similar scores to different constructs. Also, the self-report scales in this study could have been subject to common method bias, which could have had an effect on the outcomes (Podsakoff, Mackenzie, Lee & Podsakoff, 2013).

Secondly, we found that the data in this research was not normally distributed.

Although corrected-bootstrapping analysis was used, this non-normal distribution was a concern. Specifically, we found that most of our data was right skewed, meaning that most participants answered in the upper right corners of the Likert scales. One explanation of these high scales might be that in this study, confidential and sensitive information was asked from participants. Although we ensured anonymity, some participants might have given more socially accepted answers. Furthermore, 30 participants were removed from this study because they did not fully complete the questionnaire. It might have been that participants who feel less positive about their team members refused to fill in the questions and that only participants who felt positive about their team took the time to fill in the survey. This selected drop out could have caused a selection effect, with an overly positive number of teams with a positive attitude towards their team. Future research could focus on ensuring anonymization. For example, by doing an experimental study in which candidates have to assess imaginary team members. Because a virtual situation might be more private and less personal than a real life-situation, and thus less threatening, perhaps more team members in non-cooperative teams would be encouraged to participate in the study.

(38)

Third, team performance was only measured by input of the supervisor and thus may not have represented an objective view of the actual team performance. Again, for team performance, the data was right skewed, revealing that most supervisors reported a highly optimistic view of the team performance within the team. Future research might include more objective measures of team performance, such the number of successful projects over a certain period of time.

Conclusion

To conclude, the results of this study add a new perspective to a growing body of research on trust and peer control in teams. We found that intrateam trust has a positive effect on peer control such that more peer control is exerted in teams with a higher level of intrateam trust. A very counterintuitive finding is that the more peer control is exerted in a team, the lower the team’s performance is. Also, against our expectations, we did not find significant positive effects of for team learning orientation and team identification on the trust – peer control – performance relation. These surprising findings point at the need for further examination of the variables studied, which might hopefully lead to enriched understanding of peer control in teams.

(39)

References

Allen, N. J., & Meyer, J. P. (1990). The measurement and antecedents of affective, continuance and normative commitment to the organization. Journal of Occupational Psychology, 63, 1-18.

Bachmann, R., Knights, D., & Sydow, J. (2001). Trust and control in organizational relations. Organization Studies, 22(2), V-VIII.

Barker, J. R. (1993). Tightening the iron cage: concertive control in self-managing teams. Administrative Science Quarterly, 38(3), 408–437.

Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research: conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51(6), 1173–82.

Bijlsma-Frankema, K., & Costa, A. C. (2005). Understanding the trust-control nexus.

International Sociology, 20, 97-118.

Brown, C., Topping, K., Henington, C., & Skinner, C. (1999). Peer monitoring of learning behaviour: The case of “Checking Chums”. Educational Psychology in Practice, 15(3), 174-182. doi:10.1080/0266736990150304

Bunderson, J. S., & Sutcliffe, K. M. (2003). Management team learning orientation and business unit performance. Journal of Applied Psychology, 88(3), 552-560. Doi:

10.1037/0021-9010.88.3.552

Cardinal, L. B., Sitkin, S. B., & Long, C. P. (2004). Balancing and rebalancing in the creation and evolution of organizational control. Organizational Science, 15, 411- 431.

(40)

Cardinal, L. B., Sitkin, S. B., Long, C. P. (2010). A configurational theory of control.

Sitkin, S. B., Cardinal, L. B., & Bijlsma-Frankema, K. M. (Eds.) Control in Organizations, 51-79, Cambridge University Press; UK

Chadwick, I., & Raver, J. (2015). Motivating organizations to learn: Goal orientation and its influence on organizational learning. Journal of Management, 41(3), 957-986.

doi:10.1177/0149206312443558

Costa, A. (2003). Work team trust and effectiveness. Personnel Review, 32(5), 605-622.

doi:10.1108/00483480310488360

Costa, A. C., & Bijlsma-Frankema, K. (2007). Trust and control interrelations: New perspectives on the trust-control nexus. Group & Organization Management, 32(4), 392-406. Doi: 10.1177/1059601106293871.

De Jong, B. A., Bijlsma-Frankema, K. M., & Cardinal, L. B. (2014). Stronger than the sum of its parts? The performance implications of peer control combinations in teams. Organization Science, 25(6), 1703-1721. Doi: 10.1287/orsc.2014.0926 De Jong, B. A., & Elfring, T. (2010). How does trust affect performance of ongoing

teams? The mediating role of reflexivity, monitoring and effort. Academy of Management Journal, 53, 535-549. Doi: 10.5465/AMJ.2010.51468649.

Dirks, K. T., & Ferrin, D. L. (2001). The role of trust in organizational settings.

Organization Science, 12, 450–467. http://dx.doi.org/10.1287/

orsc.12.4.450.10640

Edmondson, A. (1999). Psychological safety and learning behavior in work teams.

Administrative Science Quarterly, 44: 350–383.

(41)

Ellemers, N., De Gilder, D., & Haslam, S. A., (2004). Motivating individuals and groups at work: a social identity perspective on leadership and group performance.

Academy of Management Review, 29(3), 459-478.

Ellemers, N., Kortekaas, P., & Ouwekerk, J. (1999). Self-categorisation, commitment to the group and group self-esteem as related but distinct aspects of social identity.

European Journal of Social Psychology, 29(2/3), 371-389.

Ezzamel, M., & Willmott, H. (1998). Accounting for teamwork: A critical study of group-based systems of organizational control. Administrative Science Quarterly, 43(2), 358-396.

Field, A. P. (2018). Discovering statistics using ibm spss statistics (5th ed., Ser. Sage edge). SAGE Publications.

Haslam, S. A. (2004). Psychology in organizations: the social identity approach (2nd ed.).

Sage Publications.

Henderson, J., & Lee, S. (1992). Managing i/s design teams: A control theories perspective. Management Science, 38(6), 757-777. doi:10.1287/mnsc.38.6.757 Howard, J. (1993). Peer control in the industrial workplace. Sociological Focus, 26(3),

241-256.

Janis, I. L. (1982). Groupthink: psychological studies of policy decisions and fiascoes (2nd ed.). Houghton Mifflin.

Korsgaard, M., Meglino, B., & Jeong, S. (2010). The role of motivational orientations in formal and informal control. In S. Sitkin, L. Cardinal, & K. Bijlsma-Frankema (Eds.), Organizational Control (Cambridge companions to Management, pp. 222-