Antecedents of coworker trust: Leaders' blessings

Antecedents of coworker trust: Leaders' blessings
of 9
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  RESEARCH REPORTS Antecedents of Coworker Trust: Leaders’ Blessings Dora C. Lau Chinese University of Hong Kong Robert C. Liden University of Illinois at Chicago Although employee trust in leaders has garnered substantial empirical research, trust between coworkershas been virtually ignored. Extending the work of D. L. Ferrin, K. T. Dirks, and P. P. Shah (2006), theauthors examined the role of group leaders, an influential third party in the workplace, on coworker trustformation. The correlates of the extent to which coworkers trust one another were examined in aninvestigation of 146 members of 32 work groups representing 4 diverse organizations. In this study,which utilized full network data, coworker trust was operationalized as in-degree centrality in the trustnetwork. Controlling for relational demography and coworker helping behaviors, the authors found, ashypothesized, that coworkers tended to place more trust in fellow coworkers who were also trusted bythe teams’ formal leaders than in coworkers who were less trusted by leaders. In addition, consistent withthe social information processing theory, support was found for the hypothesis that the relationshipbetween leaders’ trust and coworker trust is stronger when group performance is poor. Keywords:  coworker trust, leaders’ influence, vertical dyads, social networks Interpersonal trust facilitates the occurrence of positive work consequences that may not be possible in its absence (Coleman,1990). For decades, many studies have documented that whenemployees trust their leaders, positive work outcomes result,among them, cooperation (Coleman, 1990), organizational citizen-ship behaviors (McAllister, 1995), reduced monitoring (Langfred,2004), and enhanced group perf ormance (Dirks, 2000) and orga-nizational performance (Davis, Schoorman, Mayer, & Tan, 2000).In general, this research has revealed that trust for leaders is relatedto supportive leadership styles and perceptions of fair treatment(Dirks & Ferrin, 2002). Despite the well-developed literature ontrust in leaders, relatively few studies have examined trust forcoworkers (Ferres, Connell, & Travaglione, 2004; Parker, Wil-liams, & Turner, 2006). Moreover, the few studies that haveexamined coworker trust have emphasized its role as an indepen-dent variable (Alge, Wiethoff, & Klein, 2003; Ladebo, 2006).The focus of the current investigation was on coworker trust asa dependent variable, that is, on the determinants of the extent towhich coworkers trust one another. Coworker trust is important forthree reasons. First, the prevalence of work teams and the inter-dependent nature of work tasks require employees to trust eachother and to collaborate and coordinate with one another foraccomplishment of organizational goals (Groysberg & Abrahams,2006). To this end, coworker trust is a major component of effective team decision making and proactive behaviors at work,both of which are necessary for the effective execution of inter-dependent work (Alge et al., 2003; Parker et al., 2006). Second,because it is common for tasks to be interdependent, reward andpenalty systems are often team oriented. Sometimes, individualemployees may not be rewarded in proportion to their own effortsbecause of coworkers’ mistakes or effort shirking. We contend thatwhen employees trust their coworkers to do their best, they aremore willing to work hard themselves, because they know thattheir efforts will be rewarded accordingly. Finally, interpersonaltrust facilitates social exchange relationships (Blau, 1964). Whencoworkers trust one another, they are more willing to help eachother, because they know that their peers are likely to reciprocatetheir help in the future (Gouldner, 1960).To begin examining coworker trust, we investigated why someemployees are trusted more by their coworkers than are others.Most interpersonal trust research has focused on the trust betweenmembers of dyads, and relatively few studies have examined theimportance of third persons. However, interpersonal trust does notoccur in a social vacuum, and significant others play a role indetermining trust levels (Burt & Knez, 1996; Ferrin, Dirks, &Shah, 2006). Recent work by Ferrin et al. (2006) demonstrated that“individuals . . . draw on third parties to inform their trust judg-ments” (p. 880). Similarly, Venkataramani and Dalal (2007) foundthat third parties influenced helping behaviors among sororitysisters. Among many types of third parties, we propose, the groupleader, through roles as the official performance appraiser, rewarddistributor, and, very often, mentor of subordinates, has the poten-tial to shape the immediate environment of the work group. Dora C. Lau, Department of Management, Chinese University of HongKong, Hong Kong, China; Robert C. Liden, Department of ManagerialStudies, University of Illinois at Chicago.This study was funded by a grant from the Center for Human ResourceManagement (CHRM) at the University of Illinois. Additional funding wasprovided by the Society for Human Resource Management (SHRM) Foun-dation. The interpretations, conclusions, and recommendations, however,are those of the authors and do not necessarily represent those of theCHRM or the SHRM Foundation. The authors gratefully acknowledge theresearch assistance of Sandy J. Wayne, Maria L. Kraimer, Raymond T.Sparrowe, and Timothy M. Franz.Correspondence concerning this article should be addressed to Dora C.Lau, Department of Management, Chinese University of Hong Kong,Shatin, New Territories, Hong Kong. E-mail:  Journal of Applied Psychology Copyright 2008 by the American Psychological Association2008, Vol. 93, No. 5, 1130–1138 0021-9010/08/$12.00 DOI: 10.1037/0021-9010.93.5.1130 1130  Through critical decisions that affect group members, includingpay raises, promotions and demotions, and training opportunities,group leaders set the tone for relationships within the work groupand influence the environment in which the work group is embed-ded. Specifically, we propose that leaders’ trust in a particularemployee influences the degree to which coworkers trust thisemployee. Our main purpose in this research was to investigate therole of leaders’ attitudes and beliefs in trust development amongcoworkers.In addition, we examined situations under which leaders’ influ-ence in coworker trust formation is strengthened or weakened.Drawing on social information processing theory (Salancik &Pfeffer, 1978), we identified two situations in which employeesrely heavily on third-party information, namely, third-party cred-ibility and situational uncertainty. In particular, we propose thatleaders’ trust is more critical when leaders are trustworthy andwhen the social context is uncertain.In summary, our goal in the current investigation was to con-tribute to the interpersonal trust literature in two ways. First, weextended our understanding of third-party influence on trust for-mation by focusing on group leaders, an important third party inthe workplace. Second, we examined boundary conditions thatmay accentuate the reliance on leaders as third parties in theformation of coworker trust perceptions. Leaders’ Trust and Coworker Trust There are many definitions of trust, but most contain commonelements, including uncertainty of dependability, vulnerability of dependency (Li, 2007), expectations that the trusted parties willnot harm the trustors (Gambetta, 1988), and willingness of trustorsto assume risk with the trusted parties (Mayer, Davis, & Schoor-man, 1995). To parsimoniously integrate these diverse compo-nents, Rousseau, Sitkin, Burt, and Camerer (1998) offered thefollowing definition of trust: “A psychological state comprisingthe intention to accept vulnerability based upon positive expecta-tions of the intentions or behaviors of another” (p. 395).The decision to trust invokes an evaluation of the trustworthi-ness of another party and the risk involved in the trusting behaviors(Mayer et al., 1995). Information from surrounding others is animportant source, especially when trustors do not know or havelimited knowledge of the potential trustees. Even when the trustorsand the trustees know each other, significant others may still playan important role in interpreting the meaning and the importanceof the trustees’ behaviors by drawing from past observations andinteractions, especially when ambiguity arises (Salancik & Pfeffer,1978). Such situations are more common than we expect. Trustorsmay simultaneously receive trustworthy and untrustworthy infor-mation from the trustees (Lewicki, McAllister, & Bies, 1998), orthey may encounter situations in which trustworthy trustees be-have in a seemingly dishonest manner (Robinson, 1996). In thesesituations, the need for social information increases.Ferrin et al. (2006) demonstrated the importance of third partiesin trust formation. They found that an employee’s trust for acoworker was positively related to the number of third parties whowere trusted by the employee and who simultaneously trusted thefocal coworker. They proposed that third parties transmit socialinformation about trustworthiness beliefs among employees andthat trust perceptions are affected via these third parties. Leaders’ Trust as an Antecedent of Coworker Trust As do coworkers, leaders may potentially serve as importanttrust information transmitters who indirectly influence trust devel-opment among coworkers. Social information theory (Salancik &Pfeffer, 1978) suggests that employee attitudes are shaped by thesocial context. Very often, the social cues emitted by coworkersplay an important role in affecting focal employees’ interpretationof what happens at work (O’Reilly & Caldwell, 1979; White &Mitchell, 1979). Among all individuals engaged in work settings,some are more influential than others, and we propose that leaders,in part due to their greater formal status, play a key role ininfluencing subordinates’ attitude formation. Leaders serve as rep-resentatives of organizations, and their communications are oftenespecially salient to employees (Levinson, 1965; Liden, Bauer, &Erdogan, 2004). Thus, we contend that the influence of leadershelps to shape perceptions of the work environment by subordi-nates (e.g., employee assessment, in particular, the trustworthinessof their colleagues).Research on leader–member exchange (LMX) also sheds lighton the relationship between leaders’ trust and coworker trust. LMXtheory addresses the relationship between leaders and their subor-dinates, and positive relationships between a leader and a memberare characterized by mutual respect, trust, and obligation towardeach other (Graen & Uhl-Bien, 1995). The quality of the dyadicrelationships between leaders and subordinates may have spillovereffects on coworker exchange (CMX) relationships (Sherony &Green, 2002; Sparrowe & Liden, 1997). When leaders have pos-itive LMX with two subordinates, the subordinates both belong tothe leaders’ in-group and probably share much information andpositive sentiments. Research on emotional contagion proposesthat positive sentiments may spread between two parties (Pugh,2001) or within teams (Barsade, 2002). Similarly, balance theory(Heider, 1958) suggests that if a leader trusts two coworkers, thesecoworkers will in turn trust each other, so that balance is main-tained. Paralleling our arguments, Sherony and Green (2002) con-firmed the proposition that CMX between two coworkers wasmore positive when the LMX relationships between each coworkerand the leader were similar.Leaders’ trust may be contagious, in that it enhances thetrustworthiness perceptions of trusted employees. Das and Teng(1998) argued that when a trustor has assumed risk in trustingothers, trusted others feel somewhat bounded and will in turnbehave in a trustworthy manner. Serva, Fuller, and Mayer(2005) demonstrated the reciprocal nature of trust among inter-acting teams. On the basis of their findings on dyadic trust, weextend their arguments to a broader level by proposing thatwhen a significant other, such as a leader, trusts an employee,coworkers will likely follow suit, for two reasons. First, whenleaders trust their subordinates, they likely share more confi-dential information, delegate more responsibility, and providesupport and advice to the trusted employees (Lau, Liu, & Fu,2007). Consequently, trusted subordinates possess more instru-mental and social resources, such that they perform well andbecome more capable in coworkers’ eyes. Second, trusted em-ployees may feel a sense of obligation and responsibility tobehave in a trustworthy manner (Deutsch, 1958). To continuereceiving the leaders’ trust, trusted employees consistently be-have according to the expectations of their leaders; these be- 1131 RESEARCH REPORTS  haviors are noticed by coworkers, who in turn will be promptedto engage in behaviors that reflect trust in their colleagues.  Hypothesis 1:  The extent to which employees are trusted bytheir coworkers is positively related to the trust their leadershave for them. Social Contextual Moderators Leaders’ trust is an important source of social information thatteam members may utilize in forming coworker trust. The extent of utilization may depend on the social context. We examined twopotential moderators: employee trust for leaders and group perfor-mance.  Employee Trust for Leaders Trust is transferable. Ferrin et al. (2006) suggested that trustmay be transferred via third parties by direct communication or bypassive observation. They further proposed that such third partiesenable employees to have “increased and triangulated evidencethat the coworker is indeed trustworthy” (Ferrin et al., 2006, p.875). When team members trust their leader and their leader trustsa particular employee, it is more likely that team members alsotrust this coworker. But whether the leaders’ trust of a certainemployee represents a signal to other coworkers that this employeeis trustworthy depends on the credibility of the source of theinformation. Team members’ trust for leaders implies leader trust-worthiness. According to the trust literature, common characteris-tics of trustworthy leaders include high integrity, concern, andbehavioral consistency (Mayer et al., 1995; Whitener, Brodt, Kors-gaard, & Werner, 1998). If an honest and caring leader identifiesan employee as being trustworthy, the signal is stronger and morepersuasive than is a recommendation from a leader with a dishon-est and self-interested reputation (Rosenbaum & Levin, 1968).  Hypothesis 2:  Trust for leaders moderates the relationshipbetween coworkers’ and leaders’ trust of them, such thatwhen team members trust their leaders, the relationship be-tween the extent to which employees are trusted by theircoworkers and leaders’ trust of them is strengthened. Group Performance Besides team members’ trust for leaders, another potential mod-erator is group performance. Social information processing theory(Goldman, 2001; Salancik & Pfeffer, 1978) suggests that socialinformation is given more weight when situations are ambiguousand complex. Social information helps construct subjective mean-ing, provides focus on relevant and critical perspectives (Bommer,Miles, & Grover, 2003), and serves to reduce information com-plexity as well as to avoid information overload (Shetzer, 1993).Specifically, according to social information theory (Salancik  &Pfeffer, 1978), when the group context is uncertain, ambiguous,and complex, team members will more likely rely on social cues toform their individual beliefs and attitudes (Larson & Callahan,1990).Literature on feedback intervention (e.g., Kluger & DeNisi,1996) suggests that negative group performance may add a newdimension of complexity and ambiguity to the group context.Negative feedback represents social information that may motivateindividuals to improve their performance by increasing their effortor by changing an existing routine or method (Kluger & DeNisi,1996). The latter possibility implies a challenge to individuals’implicit theories on what leads to team success (Staw, 1975). Forinstance, one such implicit theory is that successful groups aremore cohesive and that the group members help each other (Bach-rach, Bendoly, & Podsakoff, 2001). Establishing a new routine orimplicit theory is highly complex and ambiguous, and it is notalways successful (Kluger & DeNisi, 1996). The additional com-plexity and ambiguity further enhance the importance of socialinformation in attitude and judgment formation.When group performance suffers, members’ interests are atstake and the uncertainty triggers a greater need to process criticalsocial information. If the leader trusts the focal coworker, the othercoworkers will be especially trusting of the focal coworker. How-ever, if the leader lacks trust in a coworker, coworkers, faced witha difficult group climate of low performance, will exhibit a lack of trust for the focal coworker. Conversely, members of high-performing groups may feel relatively more independent and self-sufficient than do members of relatively lower performing groups;thus, they may not rely as much on those coworkers whom theleader trusts when they form their own assessments of whichcoworkers can be trusted.  Hypothesis 3:  Group performance moderates the relationshipbetween coworkers’ trust and leaders’ trust of group mem-bers, such that when groups perform worse, the relationshipbetween the extent to which employees are trusted by theircoworkers and leaders’ trust of them is strengthened. Method Sample Data were collected as part of a larger study from four organi-zations located in the United States. The organizations includednonacademic staff of a large university, two geographically sepa-rated manufacturing plants of a large consumer products company(consumer 1 and consumer 2), and a small organization engaged inthe production and distribution of building construction materials(construction). All 39 work groups (randomly selected from theseorganizations) that we invited to participate agreed to do so. Tomeet conditions necessary for meaningful analyses of full network data, we eliminated groups with missing data, inadequate groupsample size (  4), or inadequate response rates (  80%). Theresulting final sample of 32 groups (response rate 82%) included146 respondents out of 156 total group members (individual groupmember response rate 94%). We received 38, 39, 19, and 50responses from the above organizations, respectively, representing26%, 27%, 13%, and 34% of our sample. The 32 groups ranged insize from 4 to 8 members (  M   4.7;  SD  1.0). The response ratewas comparable with that of recent studies of similar design (Bono& Anderson, 2005; Bowler & Brass, 2006; Ferrin et al., 2006;Totterdell, Wall, Holman, Diamond, & Epitropaki, 2004; Venk at-aramani & Dalal, 2007) and met the 70% minimum requirementfor conducting social network analysis (Kossinets, 2006).Most respondents were female (68%), Caucasian (54%), highschool graduates (75%); 14% had completed bachelor’s or grad- 1132  RESEARCH REPORTS  uate degrees. On average, our respondents had 8 years of tenure inthe organization (range from newcomer to 40 years), and theiraverage age was 37 years (range 19–62 years). Procedure Employees completed a paper-and-pencil attitude survey andnetwork survey on site during regular work hours. Leaders com-pleted paper surveys but did so in separate rooms. The network surveys listed the names of employees to enhance respondentrecall.  Measures Coworker trust was operationalized as in-degree centrality onthe basis of responses to the social network item that asked teammembers to report the degree of trust in each employee. The itemwas “Do you talk to (name) about confidential work-related mat-ters?” Scales ranged from 1 (  Not at all ) to 7 ( Very much ). Con-sistent with our definition of trust, sharing confidential informationwith others has been commonly considered as a representativeform of trusting behavior. It is highly risky, and people will beunlikely to confide in those whom they do not trust. It alsoinvolves a positive expectation that the trustees will be concernedfor the benefits of the trustors and will not carelessly discuss theinformation with others (Burt & Knez, 1996). Dirks and Ferrin(2002) characterized this type of behavior as relationship-basedtrust, and Gillespie (2003) described it as disclosure trust.Given the number of people whom each member needs toconsider when responding to full network items, single-item mea-sures are commonly used, including single-item network measuresof trust (Ferrin et al., 2006; Sparrowe & Liden, 2005). For theanalyses, we calculated in-degree centrality of the extent to whicheach colleague would discuss confidential work-related matterswith the focal employee (Brass, 1995). In this way, every datapoint for coworker trust was based on multiple observations, whichgreatly reduced concerns about reliability with the use of a singleitem.Leader’s trust was measured with the social network survey thatasked leaders about their degree of trust in each employee and wasassessed with the item “Do you talk to (name) about confidentialwork-related matters?” The scale ranged from 1 (  Not at all ) to 7( Very much ). We used one item for leaders’ trust in order to beconsistent with our measurement of trust in fellow coworkers. Inaddition, there was concern for the length of the leader surveys,because leaders were asked to rate each group member on sevenperformance items and seven organizational citizenship behavioritems and to complete a group performance measure and demo-graphics.Trust for leader was measured by asking each member howmuch he or she trusted the team leader. We asked respondents toindicate the extent to which they confided in their leader using a7-point Likert scale ranging from 1 (  Not at all ) to 7 ( Very much ).These individual responses were aggregated to the group level,because we were interested in the reputation of team leaders in theeyes of their members. To verify whether the aggregation was justified empirically, we performed two tests (Bliese, 2000). Theintraclass correlation test measures the proportion of variance inthe selected variable between groups. Our analysis indicated sig-nificant between-group differences,   2 (31,  N     32)    59.12,  p   .01, and the proportion of between-group variance was 16.32%.The   2 test, which examines the proportion of between-groupvariance and total variance, was also conducted (Georgopolous,1986). An   2 of .34, which exceeds the .20 rule-of-thumb require-ment used in prior research (e.g., Dirks, 2000), further verified theappropriateness of data aggregation. As a result, we calculated theaverage of members’ trust for the leaders of each team as ourmeasure of trust for leader.Group performance was measured with a seven-item Likertscale completed by leaders (1    Very poor   to 7    Outstanding ).The scale captures a wide range of performance indicators, such asquality, productivity, response time, within-group cooperation, andinitiative (    .95; Liden, Wayne, & Bradway, 1997). Control Variables Trustors’ perceptions of coworkers are likely shaped by howcoworkers behave in the workplace. Coworkers are likely to beperceived as benevolent when they are willing to go beyond theirassigned work duties to help their colleagues (McAllister, 1995).Drawing on past results from Ferrin et al. (2006), we controlled f orteam members’ helping behaviors. Leaders completed a seven-item scale that assessed the helping behaviors of employees; itconsisted of six items from Smith, Organ, and Near (1983) and oneitem from Podsakoff, MacKenzie, Moorman, and Fetter (1990).An example item is “Helps others who have been absent” (   .91), with a scale ranging from 1 ( Strongly Disagree ) to 7( Strongly Agree ). We also controlled for potential organizationaldifferences in coworker trust by creating dummy variables for eachorganization (university, consumer 1, consumer 2, and construc-tion).Characteristics of the relationships between coworkers representpotential antecedents of the extent to which coworkers are trustedby others. The relational demography literature (Tsui, Egan, &O’Reilly, 1992; Tsui & O’Reilly, 1989) suggests that coworkerswho are similar to each tend to trust each other. As a result, wecontrolled for the relational demographic characteristics of respon-dents, including age, organization tenure, gender, highest educa-tion level attained, and race. Preliminary analyses indicated thatboth simple and relational demographic characteristics were notrelated to coworker trust. To enhance power in our analyses, wedid not include these controls in subsequent analyses.  Analyses The units of analyses are individual respondents who are nestedwithin their work groups, which, in turn, are nested within orga-nizations. To correctly partition the variance between individualsand groups, we used hierarchical linear modeling analyses (Bryk &Raudenbush, 1992; Hofmann, 1997), which estimates both within-group (Level 1) and between-group (Level 2) variability. Further-more, we grand mean centered all variables, because the absolutezero of Likert and dummy variables is meaningless. Results Descriptive statistics and zero-order correlations of all variablesare reported in Table 1. HLM analyses on the null model of  1133 RESEARCH REPORTS  coworker trust (no predictors) indicated significant group effects,  2 (31,  N     32)    74.90,  p    .001. Our analyses showed that23.18% of the variance of coworker trust was attributed to group-level effects; this result indicated the importance of testing ourhypotheses with multilevel analyses. We first entered the controlvariables (Model 1 in Table 2). Providing a replication of Ferrin etal.’s (2006) findings, results presented in Table 2 show that em-ployees who helped others were trusted more by other teammates(    .38,  p    .01). It was found that helping behaviors aloneexplained 7.09% of the Level 1 variance of coworker trust. Inaddition, a significant between-organization effect was found:Consumer 1 was found to have higher coworker trust (    .98,  p  .05).Supporting Hypothesis 1, leader’s trust was found to be posi-tively related to coworker trust (    .24,  p    .01). Beyond thatvariance explained by helping behaviors, 7.83% of the variance incoworker trust was explained at the individual level. Contrary toHypothesis 2 (tested in Model 3), the interaction between leaders’trust and team trust for leader was not significant (  .05,  p  .05). Consistent with Hypothesis 3 (Model 4), the interaction of group performance and leaders’ trust (illustrated in Figure 1) wasnegative and significant (  .16,  p  .01). Similar results werefound after controlling for group trust of the leader (Model 5).When groups had weak as compared with strong performance, therelationship between leaders’ trust and coworker trust was stron-ger. When we considered the total variance of coworker trust, theinteraction term explained about 1.07% (1.17%    .9178). Com-pared with the main effect of leaders’ trust, which was 5.90%(75.43%    .0783) of total variance, the interaction effect wasrelatively smaller. Discussion Despite the large and expanding literature on trust in leaders(Dirks & Ferrin, 2002), trust between coworkers has receivedrelatively little attention. This represents a serious omission, giventhe prevalence of organizations that base their operations on teamsand teamwork (Ilgen, 1999). The current investigation contributesto our understanding of the determinants of trusting relationshipsbetween coworkers in organizations. The results that demonstratethe association between leader trust in coworkers and the extent towhich coworkers trusted each other represent an especially salientfinding. The voluminous literature on leadership has clearly dem-onstrated the influence that formal leaders can have on followers(Bass, 1990; Yukl, 2006). Leaders have been shown to influencefollower job satisfaction and work-related behaviors (Liden &Graen, 1980). It has been documented that leaders often serve asrole models for subordinates (Schein, 1992) and engage in men-Table 1  Descriptive Statistics and Correlations Variable  M SD  1 2 3 4 5Level 1 variables1. Coworker trust 3.32 1.252. Leaders’ trust 2.01 1.48 .27 ** 3. Helping behaviors 4.79 1.10 .10 .40 *** Level 2 variables4. Trust for leader 4.08 1.285. Group performance 4.66 0.88   .006. University 0.28 0.46 .09 .287. Consumer 1 0.25 0.44 .21   .318. Consumer 2 0.16 0.37 .04 .47 ** 9. Construction 0.31 0.47   .32   .36 *  Note.  Correlations among dummy-coded organization variables are omit-ted, because they are mutually exclusive. Individual-level correlations arelisted in the upper left section, whereas group-level correlations are listedin the lower right section. Level 1  n  146, Level 2  n  32 (two-tailed  t  tests). *  p  .05.  **  p  .01.  ***  p  .001. Table 2  Hierarchical Linear Modeling Analysis on Coworker Trust  Intercept and variableModel1 2 3 4 5   SE     SE     SE     SE     SE  Intercept 3.33 *** .12 3.34 *** .12 3.36 *** .12 3.38 *** .12 3.41 *** .12Level 1 variablesHelping behaviors .38 ** .12 .21 * .09 .20 * .09 .21 * .09 .20 * .09Leaders’ trust .24 ** .07 .24 ** .07 .29 *** .05 .29 *** .05Level 2 variablesTrust for leader .16 .11 .16 .11Group performance .02 .23 .05 .24University   .13 .28   .50 .29   .60 * .28   .54 .30   .67 * .30Consumer 1 .98 * .37 .65 .40 .47 .39 .50 .42 .31 .41Consumer 2   .56 .37   .43 .40   .51 .41   .66 .50   .79 .52Cross-level interactionsLeaders’ Trust  Trust for Leader   .05 .04   .06 .04Leaders’ Trust  Group Performance   .16 ** .06   .17 ** .05% change in variance explained at Level 1 7.09 7.83% change in variance explained in leaders’ trust slope 0 91.87 95.83  Note.  *  p  .05.  **  p  .01.  ***  p  .001. 1134  RESEARCH REPORTS
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks