beyond self management

8/7/2019 Beyond Self Management

1/18

Beyond Self-Management: Antecedents and Consequences of Team EmpowermentAuthor(s): Bradley L. Kirkman and Benson Rosen

Source: The Academy of Management Journal, Vol. 42, No. 1 (Feb., 1999), pp. 58-74Published by: Academy of ManagementStable URL: http://www.jstor.org/stable/256874

Accessed: 17/10/2010 15:01

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at

http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless

you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you

may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at

http://www.jstor.org/action/showPublisher?publisherCode=aom.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed

page of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of

content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms

of scholarship. For more information about JSTOR, please contact [email protected].

Academy of Managementis collaborating with JSTOR to digitize, preserve and extend access to The Academy

of Management Journal.
http://www.jstor.org/action/showPublisher?publisherCode=aomhttp://www.jstor.org/stable/256874?origin=JSTOR-pdfhttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/action/showPublisher?publisherCode=aomhttp://www.jstor.org/action/showPublisher?publisherCode=aomhttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/stable/256874?origin=JSTOR-pdfhttp://www.jstor.org/action/showPublisher?publisherCode=aom


2/18

? Academy of Management Journal1999, Vol. 42, No. 1, 58-74.

BEYOND SELF-MANAGEMENT: ANTECEDENTS ANDCONSEQUENCES OF TEAM EMPOWERMENT

BRADLEY L. KIRKMANUniversity of North Carolina at Greensboro

BENSON ROSENUniversity of North Carolina at Chapel Hill

We examined the antecedents, consequences, and mediational role of team empower-mentusing 111 work teams in four organizations.Theresults indicated that the actionsof external leaders, the production/serviceresponsibilities given to teams, team-basedhuman resources policies, and the social structure of teams all worked to enhanceemployee team empowerment experiences. More empowered teams were also moreproductive and proactive than less empowered teams and had higher levels of cus-tomer service, job satisfaction, and organizational and team commitment.

The use of work teams continues to grow in theUnited States (Lawler, Mohrman, &Ledford, 1995;Osterman, 1994) and in the foreign affiliates of U.S.multinationals (Kirkman & Shapiro, 1997; Manz &Sims, 1993). A work team is a group of individualswho work interdependently to solve problems orcarry out work (Hackman, 1987; Manz & Sims,1993). As both the use of work teams in industryand the amount of research on teams has increased,scholars have paid more attention to employee em-powerment (Argyris, 1998; Conger & Kanungo,1988; Ford & Fottler, 1995; Hardy & Leiba-O'Sullivan, 1998; Spreitzer, 1995, 1996; Spreitzer,Kizilos, &Nason, 1997; Thomas &Velthouse, 1990;Thomas & Tymon, 1994). Empowerment is in-creased task motivation resulting from an individ-ual's positive orientation to his or her work role(Spreitzer, 1995). Interestingly, there has been littlescholarly attention to the interaction of empower-ment and work team membership-that is, to em-powerment at the team level of analysis (Hyatt andRuddy [1997] is an exception). Empowerment re-searchers have devoted their attention primarily toindividual-level models, and more scholarly re-search is needed on the empowerment of teams.Research on empowered teams has lagged behindthat on self-managing teams, or teams whose mem-bers manage themselves, assign jobs, plan and

We thank Susan Cohen, Cristina Gibson, RichardGuzzo, Charles Manz, Gretchen Spreitzer, and PaulTesluk for their helpful comments on draftsof this arti-cle. We would also like to acknowledge the CatoCenterforApplied Business Research at the Universityof NorthCarolinaat ChapelHill for the grantthat made this studypossible.

schedule work, make production- or service-relateddecisions, and take action on problems (Wellins et al.,1990). Self-managing teams have been associatedwith high productivity (Cohen & Ledford, 1994;Goodman, Devadas, & Griffith-Hughson, 1988; Trist,Susman, &Brown, 1977; Wellins et al., 1990), quality(Cohen & Ledford, 1994; Wellins et al., 1990), cus-tomer service (Wellins et al., 1990), safety (Cohen &Ledford, 1994; Goodman et al., 1988; Trist et al.,1977), job satisfaction (Cordery, Mueller, & Smith,1991; Wall, Kemp, Jackson, &Clegg, 1986), and organ-izational commitment (Cordery et al., 1991). Someresearchers use the terms empowered teams and self-managing teams synonymously (Fisher, 1993; Ford &Fottler, 1995; Manz & Sims, 1993), but others differ-entiate the concepts (Mohrman, Cohen, &Mohrman,1995). Regardless of the semantic confusion, therehas been very little empirical work on teams strictlydefined as empowered teams.

Recently, we developed a theoretical model ofempowered teams that distinguishes the conceptfrom self-managing teams (Kirkman & Rosen,1997). Our team-level model contains four dimen-sions that parallel the dimensions of empowermentthat have been specified at the individual level ofanalysis (Thomas & Velthouse, 1990), along withempowerment's antecedents and consequences.The purpose of the research reported here was totest hypotheses generated from our team empower-ment model. In this article, we first review the fourteam empowerment dimensions and differentiatethe concept from the construct of self-managingteams. Second, we review the antecedents and con-sequences of team empowerment. Third, we reporta factor analysis of the team empowerment con-struct and separate analyses of its antecedents and

58


3/18

Kirkman and Rosen

consequences. Fourth, we report tests of the directrelationships between team empowerment and itsantecedents and consequences. Fifth, we examineteam empowerment as a mediator between the an-tecedents and consequences. Finally, we discusstheoretical implications and future research, man-agerial implications, and limitations of our study.

THEORY AND HYPOTHESESTeam Empowerment Defined

In our earlier work, we defined team empower-ment as having four dimensions: potency, mean-ingfulness, autonomy, and impact (Kirkman &Rosen, 1997).Potency. Potency, which parallels the individual-level empowerment construct of competence, orself-efficacy (Conger & Kanungo, 1988; Thomas &Velthouse, 1990), is the collective belief of a teamthat it can be effective (Guzzo, Yost, Campbell, &Shea, 1993; Shea & Guzzo, 1987a). Potency is dif-ferent from self-efficacy in at least three ways: (1)self-efficacy refers to individual performance andpotency refers to team performance, (2) self-effi-cacy experiences are private but potency experi-ences develop collectively, and (3) self-efficacy re-lates to specific task performance but potencyrefers to generalized effectiveness (Guzzo et al.,1993).Meaningfulness. Meaningfulness, correspond-

ing to meaningfulness at the individual level ofanalysis (Thomas & Velthouse, 1990), refers to ateam's experiencing its tasks as important, valu-able, and worthwhile (Hackman, 1987; Hackman&Oldham, 1980). Team members collectively de-velop and share the meaningfulness of theirtasks. Thus, team members have direct effects onthe experiences of meaningfulness of other mem-bers.Autonomy. Autonomy parallels the individual-level empowerment construct of choice (Thomas &Velthouse, 1990) and is the degree to which teammembers experience substantial freedom, indepen-

dence, and discretion in their work (Hackman,1987; Susman, 1976). Important decisions are madeand executed by teams. Thus, high levels of teamautonomy may actually decrease individual auton-omy as important decision making is shared ratherthan carried out alone and responsibility is dif-fused rather than granted to a single individual(Uhl-Bien & Graen, 1998).Impact. Team members experience impact,which is similar to impact at the individual level ofanalysis (Thomas &Velthouse, 1990), when a teamproduces work that is significant and important for

an organization (Hackman, 1987). Team membersseek out, share, and collectively understand feed-back from other organization members. Team mem-ber interaction enables the gathering of more infor-mation on team impact from customers thanindividuals alone could gather (Ancona, 1990).Both self-managing teams and empowered teamsare autonomous, but the members of the latter alsoshare a sense of doing meaningful work that ad-vances organizational objectives; thus, team em-powerment is a much broader construct. Self-man-agement is most analogous to only one of ourempowerment dimensions-autonomy-and somescholars have even used "autonomous work teams"as a synonym for self-managing teams (e.g., Corderyet al., 1991; Pearson, 1992; Wall et al., 1986). Inaddition, measures of the two constructs are verysimilar. Self-management scales have includeditems that assess the extent to which team members

believe that they have high levels of decision-mak-ing latitude and responsibility (e.g., Campion, Med-sker, & Higgs, 1993; Campion, Papper, & Medsker,1996; Kirkman, Shapiro, Novelli, & Brett, 1996).Items measuring autonomy typically specify thedegree to which individuals have freedom to makeimportant decisions about how and what they do(Cordery et al., 1991; Gulowsen, 1972; Susman,1976).The multidimensionality of team empower-ment. We have argued that team empowermentconsists of four related (but independent) dimen-sions (Kirkman &Rosen, 1997). The dimensions arerelated because they are likely to be mutually rein-forcing (Spreitzer, 1995). For example, if a team'smembers experience impact (that is, talk to custom-ers about how the team's work affects them), theymay find their work more meaningful (Ancona,1990; Hackman, 1987). Research on psychologicalempowerment at the individual level has shownthat the four dimensions, although related, are stilldistinct components of empowerment (Spreitzer,1995; Tymon, 1988). Thus,Hypothesis 1. The four dimensions of teamempowerment-potency, meaningfulness, au-tonomy, and impact-are distinct but related.

Antecedents of Team EmpowermentFrom an extensive review of the work team, em-

powerment, and group motivation literatures, wetheoretically identified job and organizational char-acteristics that may act as antecedents to team em-powerment (Kirkman & Rosen, 1997). Our searchyielded antecedents in four thematic areas: externalleader behavior, production/service responsibili-

1999 59


4/18

Academy of Management Journalties, team-based human resources policies, and so-cial structure. We believed that most of the job andorganizational characteristics identified wouldlikely affect all four dimensions of team empower-ment (Kirkman & Rosen, 1997).External team leader behavior. External teamleaders have a supervisory role but are not mem-bers of the teams they lead (Manz & Sims, 1987).When team leaders delegate responsibility, ask forand use employee input, and enhance team mem-bers' senses of personal control, the team membersare more likely to experience meaning, impact(Hackman, 1987), and autonomy in their work be-cause they are taking on more responsibility (Sus-man, 1976; Thomas &Velthouse, 1990). When teamleaders actually use member ideas, membersshould become more confident in their abilities, orexperience more potency (Guzzo et al., 1993).External team leaders who allow teams to settheir own performance and output goals createmore autonomy experiences (Manz &Sims, 1987)and increase team potency as members decidewhich goals should be adjusted and how mucheffort is needed in relation to performance(Guzzo et al., 1993). Members will likely findthese goals more meaningful because they partic-ipate in their creation (Hackman, 1987; Hackman&Oldham, 1980). When leaders have high expec-tations, team members are more likely to com-plete challenging assignments, further strength-ening potency experiences (Burpitt & Bigoness,1997; Manz & Sims, 1987). Guzzo and colleagues(1993) argued that transformational leaders-those who energize, inspire, and communicatehigh performance expectations-directly influ-ence potency. For many team leaders, however,empowering their teams may translate directlyinto losing power (Hardy & Leiba-O'Sullivan,1998). Thus, if team leaders do not trust the ca-pabilities of team members, they will be lesslikely to empower them (Burke, 1986; Culbert &McDonough, 1986). Consequently,

Hypothesis 2. The more that an external teamleader exhibits encouraging leader behaviorsby, for instance, delegating responsibility to ateam, soliciting and using team input whenmaking decisions, enhancing team members'senses of personal control, encouraging teamgoal setting, self-evaluation, and high team ex-pectations, and trusting the team, the more theteam's members will experience team empow-erment.Production/service responsibilities. Whenteams set production schedules and standards,monitor customer feedback, develop and train for

quality improvement practices, and assume own-ership for the completion of finite units of work,they have high production/service responsibili-ties (Kirkman & Rosen, 1997). In practice, teammember participation in the day-to-day regula-tion of a team's work varies greatly (Cohen &Bailey, 1997). More participation in goal settingleads to greater intrinsic motivation (Hackman &Oldham, 1980) and a greater sense of empower-ment (Gulowsen, 1972), but effects on perfor-mance have been mixed (Locke, Shaw, Saari, &Latham, 1981). Perhaps empowerment mediatesthe relationship between participative goal set-ting and team performance (cf. Campion et al.,1993; Campion, Papper, &Medsker, 1996). At theindividual level, participative goal setting leadsto higher task comprehension (Latham & Saari,1979). A more complete understanding of taskscan, in turn, enhance meaningfulness (Thomas &Velthouse, 1990). Participative goal setting alsoincreases team member autonomy by transferringthe goal-setting responsibility from managementto employees (Susman, 1976).

Similarly, increased decision making in pro-duction scheduling and job assignments makesteam members a meaningful part of the produc-tion process (Hackman, 1987; Manz & Sims,1993) and creates more autonomy (Susman,1976). Team members who make job assignmentshave a demonstrable impact on the work flow ofother members (Hackman, 1987). Further discre-tion exists for teams created within a total qualitymanagement (TQM) environment (Lawler et al.,1995). Teams responsible for quality frequentlycollect data to measure discrepancies (Ishikawa,1985), which can allow teams to make adjust-ments in their work and lead to more potencyexperiences (Guzzo et al., 1993). More controlover product or service quality also creates moreautonomy (Hackman, 1987; Susman, 1976). Oneof the basic tenets of TQM is that employees whohave increased responsibility for quality will findtheir work more personally meaningful (Ish-ikawa, 1985). Team members constantly updatethe skills and knowledge necessary to achievehigh levels of production/service quality (Lawleret al., 1995), and their doing so should affect allfour dimensions of empowerment (Thomas &Velthouse, 1990).Related to quality and learning is a team's level ofcustomer contact (Ancona, 1990). Increased cus-tomer contact and feedback should make teammembers feel more potent (Guzzo et al., 1993),demonstrate that a team's work makes a difference

February0


5/18

Kirkman and Rosenfor customers (impact; Cummings, 1978),1 make theproduction or delivery of a service more personallymeaningful to team members (Manz &Sims, 1993),and allow team members to experience more free-dom (autonomy) in handling customer issues (Sus-man, 1976). Team members who provide customerswith a whole product or service use a variety ofskills that are likely to enhance meaningfulness(Hackman, 1987; Hackman & Oldham, 1980), con-fidence in their team's ability to perform (potency;Guzzo et al., 1993), and knowledge of how theirefforts affect the overall organization (impact; Grif-fin, 1991). Consequently,

Hypothesis 3. The higher the level of a team'sproduction/service responsibility-the more,for instance, that team members set their ownproduction or service goals, make importantdecisions such as scheduling and production/service assignments, monitor and train forquality, handle customer issues and com-plaints, and work with a whole, containedproduct or service-the more the team mem-bers will experience team empowerment.Team-based human resources policies. Humanresources policies for teams, including team-basedrewards, receiving or delivering cross-training, andmaking staffing decisions, should support and en-hance team empowerment. For example, Shea andGuzzo (1987b) found that when highly interdepen-dent teams received team pay, they were more

likely to experience potency. Team incentives pro-vide motivation that may enhance the meaningful-ness of a team's work (DeMatteo, Eby, & Sund-strom, 1998; Gibson &Kirkman, 1999; Mohrman etal., 1995). Some team members receive incentivesto cross-train for team jobs or the jobs of otherteams (Manz & Sims, 1993; Wellins et al., 1990).Cross-training results in higher team flexibility andbreadth of experience (meaningfulness; Hackman,1987), confidence in multiple job-related skills (po-tency; Guzzo et al., 1993), and a higher chance thatteam members will have a significant impact ontheir organization (Manz & Sims, 1993).

In addition to cross-training, team membersmight also train other team members or assist intheir selection, performance evaluation, discipline,and dismissal (Gibson & Kirkman, 1999). By dem-onstrating relevant skills and behaviors in the train-ing of new team members, members are more likelyto feel confident that their team can perform tasks

1Throughout the rest of this section, a dimensionidentified in parentheses after a statement should beunderstood to be exemplified by the statement.

(potency; Guzzo et al., 1993), find intrinsic interestin their work (meaningfulness), have a greater im-pact because other team members will be directlyaffected by the quality of their training (Hackman,1987), and feel more autonomy in carrying out awider variety of jobs (Susman, 1976). When assess-ing fellow members with peer evaluations (Saave-dra & Kwun, 1993), team members will experiencemore autonomy in the evaluation process (Susman,1976), have a greater impact on fellow members'development and rewards, use the wide variety ofskills that are required in a performance appraisalprocess (meaningfulness; Hackman, 1987), andmore accurately perceive how capable their team isby rating its performance (potency; Guzzo et al.,1993). Consequently,

Hypothesis 4. The more that organizations im-plement team-based human resources policies,whereby, for instance, team members are paidon the basis of team membership, at least inpart are cross-trained within and across teams,participate in the selection, training, disci-pline, and dismissal of fellow team members,and formally evaluate the performance of fel-low members, the more the team members willexperience team empowerment.Social structure. Spreitzer (1996) defined socio-political support as the endorsement, approval, andlegitimacy obtained from various constituencies inorganizational political networks. Belonging to a

support network increases an individual's interde-pendence with important organizational constitu-ents and, in turn, increases that individual's senseof personal power (Crozier, 1964). Increased per-sonal power will likely result in more competenceand impact at the individual level (Thomas & Velt-house, 1990). Manz (1990) argued that at the teamlevel of analysis, participation broadens team mem-bers' activities in organizational networks and thus,their sense of potency.With increased legitimacy and participation innetworks comes a higher degree of access to strate-gic organizational information, which in turn canhelp team members determine their particular im-pact on overall organizational performance (Spre-itzer, 1996), enable team members to experiencehigher levels of potency (Guzzo et al., 1993), andenhance the meaningfulness of team tasks (Hack-man, 1987). Similarly, access to important resourc-es-from other teams or departments or even fromoutside an organization-will likely enhance theexperience of empowerment (Spreitzer, 1996). Be-yond access, some teams provide resources to otherteams, departments, or external customers. Teammembers in charge of providing important informa-

1999 61


6/18

Academy of Management Journaltion or resources will likely utilize their full capa-bilities (potency; Guzzo et al., 1993), have a bettersense of how their actions affect other teams orcustomers (impact; Manz & Sims, 1993), heightentheir experience of autonomy (Susman, 1976), andcreate more intrinsic meaning in their work (Hack-man, 1987).Team members who develop their own rules andprocedures experience a greater sense of participa-tion in how their teams function on a day-to-daybasis than do team members who are not able todecide on rules and procedures (Manz & Sims,1993). Such teams will likely experience a highlevel of autonomy, because they have the authorityto design and enforce their own particular mannerof operation (Susman, 1976), a high level of intrin-sic caring about the work (meaningfulness; Hack-man, 1987), and a high level of impact, because ofthe effect of the rules and procedures on fellowmembers (Manz & Sims, 1993). Consequently,

Hypothesis 5. The more that a team's membersare embedded in a well-developed social struc-ture-for instance, have sociopolitical support,have access to strategic information and workunit resources, have a high degree of interteamcoordination and communication, and maketheir own rules and policies, the more they willexperience team empowerment.

Consequences of Team EmpowermentFrequently cited criteria of work team effective-ness include productivity (Banker, Field, Schroe-der, & Sinha, 1996; Cohen & Ledford, 1994; Glad-stein, 1984; Hackman, 1987; Pearce &Ravlin, 1987;Shea &Guzzo, 1987a; Trist et al., 1977; Wall et al.,

1986), quality (Banker et al., 1996; Cohen, Ledford,& Spreitzer, 1996; Hackman, 1987; Shea & Guzzo,1987a), low costs (Cohen et al., 1996; Trist et al.,1977), safety (Cohen et al., 1996; Goodman et al.,1988; Pearce &Ravlin, 1987; Trist et al., 1977), jobsatisfaction (Cordery et al., 1991; Wall et al., 1991),and organizational commitment (Cordery et al.,1991). We included productivity, proactivity, andcustomer service as performance outcomes and jobsatisfaction, organizational commitment, and teamcommitment as attitudinal outcomes. We made thedistinction between performance and attitudinaloutcomes to be consistent with previous research(Campion et al., 1993; Campion, Papper, & Med-sker, 1996; Gladstein, 1984; Hackman, 1987).Team productivity. Empowerment has been as-sociated with productivity at both the team (Hyatt& Ruddy, 1997; Tesluk, Brass, & Mathieu, 1996)and individual levels of analysis (Spreitzer, 1995;

Spreitzer et al., 1997; Thomas & Tymon, 1994; Ty-mon, 1988). At the individual level, managers havehigher levels of performance when they feel a senseof control (autonomy) on the job (Wood &Bandura,1989). Employees who have more complete knowl-edge of their jobs (impact) often make better job-related decisions (Miller &Monge, 1986). Gorn andKanungo (1980) found that employees were moreproductive when they actively participated in de-cision making and found their jobs meaningful.Conger and Kanungo (1988) conceptualized em-powerment at the individual level as self-efficacy,which has been linked to productivity (Frayne &Latham, 1991; Gist, Schwoerer, &Rosen, 1991). Atthe team level of analysis, Guzzo and colleagues(1991) found that more potent teams were alsomore productive than those with less potency.Thus,

Hypothesis 6. The more that a team's membersexperience team empowerment, the more pro-ductive the team will be.Proactivity. Working at the individual level of

analysis, Bateman and Crant (1993) defined proac-tive behavior as individuals' actions effecting en-vironmental change through their scanning foropportunities, showing initiative, taking actionon and solving problems, and persevering untilchanges are made. Spreitzer (1995) argued that em-powerment leads to a proactive orientation towardjobs, management, and organizations. High levelsof self-efficacy lead to more initiating behaviorsand persistence in the face of obstacles (Bandura,1997). Deci and Ryan (1985) found that the more anindividual perceived that he or she had autonomy,the more initiative that person took in work-relatedsituations. At the team level of analysis, teams areproactive when they seek continuous improve-ment, revise work processes, and seek innovativesolutions to work problems (Hyatt &Ruddy, 1997).Empowered teams have been found to frequentlytake action on problems and improve the quality oftheir work by initiating changes in the way work iscarried out (Wellins, Byham, & Wilson, 1991).Thus,Hypothesis 7. The more that a team's membersexperience team empowerment, the more pro-active their team will be.

Customer service. Company-reported evidencedemonstrates consistent links between the use ofwork teams and high levels of quality and customerservice (Lawler et al., 1995; Manz & Sims, 1993;Wellins et al., 1990). Empowered teams take re-sponsibility for handling customer complaints di-rectly and often diagnose their own quality prob-

62 February


7/18

Kirkman and Rosenlems and issues (Wellins et al., 1991). Guzzo andhis associates (Guzzo et al., 1991; Shea & Guzzo,1987b) found that more potent teams also providedhigher levels of internal and external customer ser-vice. Thus,

Hypothesis 8. The more that a team's membersexperience team empowerment, the higher willbe the team's level of customer service.Team job satisfaction. Tymon and his associates(Thomas &Tymon, 1994; Tymon, 1988) and Spreit-zer and her colleagues (1997) found associationsbetween empowerment and job satisfaction at theindividual level of analysis. In addition, employeesworking in teams have reported higher levels of jobsatisfaction than employees working in traditionalsettings in the same company (Cordery et al., 1991;Wall et al., 1986). Gorn and Kanungo (1980) foundthat the more meaningful an employee's job was,

the more satisfied the employee was with his or herjob. Typically, employees find more meaning intheir jobs when the scope of their activities is large(Griffin, 1991), which is often the case on empow-ered work teams (Wellins et al., 1991). Thus,Hypothesis 9. The more that a team's membersexperience team empowerment, the higher willbe the team's level of job satisfaction.Team organizational commitment. Steers(1977) found that work-related experiences andperceptions, rather than personal, job, or organiza-tional factors, were the most powerful predictors of

organizational commitment. Thus, an employee'sexperience of empowerment may account for morevariance in his or her commitment level than moreobjective job or organizational characteristics.

Steers also found that a positive attitude amongone's peers was one of the more important experi-ences affecting commitment. Empowered teams of-ten generate these positive peer experiences (Wel-lins et al., 1991). Finally, Steer's research showedthat if employees perceive that their organizationconsistently makes and keeps its commitments toemployees, they are more likely to be committed.The high level of support and trust inherent in anempowered team system will likely contribute tohigher commitment levels among team members(Manz & Sims, 1993; Wellins et al., 1991). At theteam level of analysis, Cordery and colleagues(1991) found that organizational commitment washigher for employees in autonomous teams than forthose traditionally organized in the same company.Hackman (1987) suggested that commitment ef-fects may be stronger for an employee's work teamthan for his or her organization. For example,Wall and colleagues (1986) found no effect of au-tonomous team membership on employee organi-zational commitment, a fact attributed to anoverwhelming team commitment effect. To be con-sistent with our theoretical model and these previ-ous arguments, we included both organizationaland team commitment. Thus,

Hypothesis 10. The more that a team's mem-bers experience team empowerment, thehigher will be the team's level of organizationalcommitment.Hypothesis 11. The more that a team's mem-bers experience team empowerment, thehigher will be the team's level of team commit-ment.

STAGE 1

Organizational andJob Characteristics

* External team leaderbehavior* Production/serviceresponsibilities* Team-based humanresources policies* Social structure

FIGURE 1A Model of Work Team EmpowermentaSTAGE 2

Team Empowerment

* Potency* Meaningfulness* Autonomy* Impact

STAGE 3

Work TeamEffectiveness

Performance Outcomes* Productivity* Proactivity* Customer serviceAttitudinal Outcomes* Job satisfaction* Organizationalcommitment* Team commitment

a Source: Kirkman and Rosen (1997).

1999 63


8/18

6Academy of Management Journal

Figure 1 summarizes the hypothesized relation-ships.Our model is consistent with previous models ofwork team effectiveness (e.g., Campion, Stevens, &Medsker, 1996; Hackman &Morris, 1975; McGrath,1964) as well as with models of job design framedat the individual level of analysis, such as Hack-man and Oldham's (1980) Job CharacteristicsModel. Classified as input-process-output models,these models separate objective job characteristicsfrom both effectiveness and internal responses tothese characteristics. All these models include athree-stage process in which organizational leaderstake action in stage one (inputs), those actions af-fect employee experiences in stage two (process),and important outcomes result from positive em-ployee orientations toward work in stage three (out-puts). Like our model of team empowerment, thesemodels imply that employee experiences in stagetwo will mediate managerial actions taken in stageone and outcomes realized in stage three. For ex-ample, Guzzo and colleagues (1991) found that po-tency mediated the relationships between bothteam composition and team effectiveness and goalsetting and team effectiveness. Consequently,

Hypothesis 12. Team empowerment will medi-ate the relationships between job and organi-zational characteristics (external leader behav-iors, production/service responsibilities, team-based human resources policies, and socialstructure) and team effectiveness outcomes(productivity, proactivity, customer service, jobsatisfaction, organizational commitment, andteam commitment).

METHODSSample

We used a field study to test the multidimen-sionality of team empowerment, the relation-ships between team empowerment and its ante-cedents and outcomes, and the mediational roleof team empowerment. We conducted the studyin four organizations (two Fortune 50 organiza-tions and two smaller companies) that had for-mally implemented work teams. The companies,all of which were located in the southeastern andsouthwestern United States, included two textilemanufacturers, a high-technology manufacturer,and an insurance company. Response rates, ageranges, race, sex, education levels, organizationaltenure, team size, and team tenure are shown inTable 1 for team members and team leaders over-all and for each organization separately. Table 1

also shows significant differences between orga-nizations.

Level-of-Analysis IssuesResearchers can measure group-level phenom-

ena using individual member data in at leastthree ways (Tesluk, Zaccaro, Marks, & Mathieu,1997). First, respondents can rate themselves ontheir individual attributes, and researchers canthen aggregate these data to the group level. Sec-ond, individuals can rate their groups or teams onparticular attributes (rather than rate their ownattributes as individuals), and these ratings canbe averaged to form a group score (see Campion etal. [1993], Campion, Papper, and Medsker [1996],and Hyatt and Ruddy [1997] for examples).Third, groups can provide consensus survey rat-ings. Following the recommendations of previousresearchers (Campion et al., 1993; Guzzo et al.,1993) we measured team empowerment by facil-itating group consensus on 26 team empower-ment items for each team. Thus, our team em-powerment scores captured a group-levelphenomenon without aggregation.We measured the antecedents of team empower-ment (job and organizational characteristics) by sur-veying external team leaders in order to avoid thesame-source bias that would have been present if wehad used team member data (Podsakoff & Organ,1986). We also obtained measures of the performanceoutcomes (productivity, proactivity, and customerservice) from external team leaders. Purely objectiveperformance data were not used because most of theteams in the study were assessed by their organiza-tions with their own measurement systems, so com-parability within and across organizations was lim-ited (Spreitzer, 1995, 1996; Spreitzer et al., 1997).However, in a cover letter accompanying the surveywe instructed the team leaders to review performancedata to help ensure that objective data were consid-ered in their ratings.We obtained data on the attitudinal outcomes(job satisfaction, organizational commitment,and team commitment) by aggregating team mem-ber data. Researchers typically use self-reportmeasures of satisfaction and commitment for di-rect measures of these internal states (Markoczy,1997). We used different sources and measure-ment methods to minimize common method vari-ance (Podsakoff & Organ, 1986). Consequently,our study design followed a two-source (externalteam leaders and team members), three-method(external team leader ratings, team member con-sensus interviews, and team member aggregatedratings) strategy.

February4


9/18

Kirkman and RosenTABLE 1

Demographic InformationaTeam Member Data Team Leader Data

Organization OrganizationVariable Overall A B C D Overall A B C D

Response rateSurveys 85% 93% 88% 72% 82% 91% 100% 75% 78% 100%Interviewsa 81 93 71 84 78

Age rangeUnder 20 1.88 1.67 0.42 3.46 2.08 0.00 0.00 0.00 0.00 0.0020-35 55.15 49.16 54.58 52.81 65.42 37.50 21.43 60.00 15.79 51.7236-50 35.64 38.46 38.75 36.36 28.33 48.96 67.86 30.00 47.37 44.83Over 50 7.33 10.71 6.25 7.37 4.17 13.54 10.71 10.00 36.84 3.45

Chi-square 31.31 (df = 12, p < .002) 22.35 (df = 6, p < .001)Race

Caucasian 54.72 67.38 57.02 26.73 63.14 77.08 100.00 70.00 63.16 68.97African American 37.49 26.24 42.10 69.13 17.36 19.79 0.00 30.00 36.84 20.69Asian American 0.42 0.00 0.44 0.46 0.85 0.00 0.00 0.00 0.00 0.00Hispanic American 6.54 5.67 0.44 1.84 17.80 3.13 0.00 0.00 0.00 10.34Other 0.83 0.71 0.00 1.84 0.85 0.00 0.00 0.00 0.00 0.00

Chi-square 210.01 (df = 12, p < .001) 19.09 (df = 6, p < .004)Sex

Female 58.62 39.19 63.03 52.47 83.82 44.79 17.86 70.00 10.53 75.86Chi-square 114.54 (df = 3, p < .001) 33.70 (df = 3, p < .001)Education

Some high school 8.87 15.31 5.48 10.71 2.53 2.11 3.70 0.00 5.26 0.00High school diploma 83.47 79.59 88.61 85.71 81.01 71.58 77.78 70.00 73.68 65.52Four-year-college degree 6.55 4.42 4.22 2.68 15.19 26.31 18.52 30.00 21.05 34.48Master's 1.11 0.68 1.69 0.89 1.27 0.00 0.00 0.00 0.00 0.00Ph.D. 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00

Chi-square 67.44 (df = 9, p < .001) 4.23 (df = 6, p < .645)Team characteristics

Average organizational tenure in yearsb 6.36d 4.00w 7.50X 9.17y 5.65 10.54g 5.95 x .82x 18.61z 10.72, yAverage team tenure in yearsb 2.05e 1.84X 3.12y 1.83, 1.43z 1.74h 2.81X 0.76y 1.66y 1.43yAverage team size bc 12.75f 13.83, 11.78y 14.48, 11.12y 13.03i 11.40X 12.00, 15.32y 10.75XAverage number of teams led by one 2.61i 3.31X 3.60x 2.53x,y 1.21zleaderba Percentage calculated from returned surveys.b Different subscripts indicate significant differences.c Number of members.d F3, 977 = 50.41, p < .001.e F3, 900 = 60.25, p < .001.f F3, 925 = 13.63, p < .001.gF3, 9 = 19.96, p < .001.hF3 88 = 9.96, p


10/18

Academy of Management JournalTABLE 2

Descriptive Statistics, Reliabilities, and Correlationsa, b,cVariable N Mean s.d. rwg r n 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

1. Potency 982. Meaningfulness 983. Autonomy 984. Impact 985. Team empowerment 986. Externalleader behavior 1017. Production/service 101

responsibilities8. Team-based human 101resources policies9. Social structure 101

10. Team-level productivity 10111. Team-level proactivity 10112. Team-level customer service 10113. Team-level job satisfaction 11114. Team-level organizational 111commitment15. Team-level team 111commitment

5.92 0.855.67 1.283.91 1.795.74 1.245.31 1.105.59 0.834.68 1.103.89 0.944.735.364.965.604.054.56

1.121.020.960.880.86 0.870.86 0.85

(.82).69.48.79.81

0.27 .230.42 .05

(.92).60.80.89.32.16

(.94).53.82.42.41

(.93).88 (.96).33 .40 (.91).10 .25 .58 (.86)0.39 .09 .16 .22 .12 .22 .48 .53 (.74)0.460.470.480.38

4.32 1.07 0.83

.14 .29 .28 .18 .27 .55 .59

.60 .55 .45 .63 .64 .59 .30

.39 .48 .38 .43 .49 .68 .40

.40 .27 .28 .40 .38 .51 .36

.29 .36 .54 .36 .48 .32 .15

.44 .44 .55 .44 .56 .45 .24

.47 (.90).21 .33

.41 .56

.25 .44

.23 .30

.27 .37

(.91).70.73.29.43

a N indicates the number of teams and varies with the data source and method used; for team member consensus ratings, N = 98; fordata from external team leaders, N = 101; for team-member-aggregated data, N = 111.b For correlations greater than .20, p < .05; for r > .26, p < .01; for r > .33, p < .001.c The statistic r,g represents the reliability within groups averaged across all teams (James et al., 1984, 1996); rbetweens the correlationbetween team leader and team member data. Reliabilities are in parentheses.Measures

All of the items described below were measuredon a seven-point Likert-type scale, with 1 for"strongly disagree" and 7 for "strongly agree."Antecedents of team empowerment. The itemsmaking up the antecedents were based on our lit-erature review (Kirkman & Rosen, 1997). The exter-nal leader behavior group construct was measuredwith a 14-item scale. Examples of the items includethe extent to which a team leader agreed or dis-agreed that he or she gave a team many responsi-bilities, asked the team for advice when makingdecisions, controlled much of the activity of theteam (reverse-coded), allowed the team to set itsown goals, stayed out of the way when the teamworked on its performance problems, told the teamto expect a lot from itself, and trusted the team.

The production/service responsibilities groupconstruct was measured with a 12-item scale. Ex-amples include the extent to which a team leaderagreed or disagreed that the team he or she led setits own goals, had the responsibility to decide pro-duction/service scheduling, and had the responsi-bility for measuring the quality of its products.Team-based human resources policies (a groupconstruct) was measured with a 9-item scale. Ex-amples include the extent to which a team leaderagreed or disagreed that the team he or she leddecided who could be a member, was rewarded as

a group, was cross-trained to do different jobs, andformally evaluated the performance of its ownmembers.

The social structure group construct was mea-sured with an 11-item scale. Examples include theextent to which a team leader agreed or disagreedthat the team he or she led had access to otherteams' resources, got support from other groups inthe company when it needed it, had access to im-portant information, depended on other teams forresources or information, and had the responsibil-ity to make its own rules.Team empowerment. Potency was assessed withGuzzo and colleagues' (1993) 8-item team-levelmeasure. The items assessed the extent to whichteam members agreed or disagreed that their teamhad confidence in itself, believed it could be ex-tremely good at producing high-quality work, ex-pected to be known as a high-performing team, feltit could solve any problem, believed it could bevery productive, could get a lot done when itworked hard, believed that no job was too tough,and expected to have influence.Team meaningfulness was assessed with Thomasand Tymon's (1993) 6-item individual-level mea-sure adapted for the team level. The items assessedthe extent to which team members agreed or dis-agreed that their team cared about what it did,believed that its work was valuable, believed that

(.90).61 (.91).23 .15 (.85).40 .34 .77 (.87).48 .44 .52 .47 .57 .33 .16 .17 .32 .35 .33 .28 .65 .70 (.95)

February6


11/18


12/18

Academy of Management Journalinterviewed owing to absenteeism or work-relatedduties. The interviews, which were conducted on-site and on company time in conference rooms,took place no earlier than 3 weeks and no later than6 weeks after survey completion; the mean timebetween survey and interview was 5.2 weeks. Eachmember of the team was given a copy of the 26-itemteam empowerment measure. The researcher in-formed each team's members that they must reachconsensus on where their team stood on each of the26 items and that one team member should recordthe team's answers. Responses were made on aone-to-seven scale (1 = strongly disagree and 7 =strongly agree). The researcher read each item andthen instructed the team to discuss the item, re-solve differences of opinion, and reach consensus.The researcher did not clarify any of the items,instead instructing team members to form theirown opinions about the meaning of each item. Atotal of 868 team members participated (81 percentof all the team members who completed surveys).

RESULTSWe conducted several sets of analyses on the data,including (1) three factor analyses, one for the job andorganizational characteristics, one for the team em-powerment items, and one for the team performanceoutcome items,2 (2) reliability checks on the factoredscales, (3) interrater agreement checks on the aggre-gated scales, (4) correlation and multiple regression

analyses, to test hypotheses regarding direct relation-ships, (5) a hierarchical regression analysis, to test forthe increased explanatory power of team empower-ment beyond autonomy on the team effectivenessoutcomes, and (6) a hierarchical regression analysis,to test for the mediating effects of team empowerment(Baron &Kenny, 1986).Factor Analysis and Correlation Matrix

There were too many job and organizationalitems (46 in all) to analyze using confirmatory fac-tor analysis (Bentler &Chou, 1987). In addition, theitems were based on a literature review and wereexploratory (Kirkman & Rosen, 1997). Thus, weused exploratory factor analysis to assess whetherthe four antecedents of team empowerment shouldbe kept separate in subsequent analyses (Nunnally& Bernstein, 1994). Four factors emerged from a"varimax" rotation with eigenvalues greater than

2 We did not factor-analyze the attitudinal outcomeitems because all three of the scales had been previouslyfactor-analyzed in earlier studies.

1.0 and explaining a total of 75 percent of thevariance. The four factors corresponded to our orig-inal theoretical specification (Kirkman & Rosen,1997). Six items (1 from external leader behavior, 3from production/service responsibilities, and 2from team-based human resources policies) failedto load on any factor and were dropped from sub-sequent analyses. Post hoc analyses of regressionsthat included these dropped variables demon-strated highly similar results.3The 26-item measure of team empowerment re-solved into four separate factors (under varimaxrotation) with eigenvalues greater than 1.0 and ex-plaining a total of 91 percent of the variance. Thefour factors corresponded to our original theoreti-cal specification. One potency item ("My team be-lieves that no job is too tough") did not load on anyfactor and was subsequently dropped from theanalysis. The three performance scales "factored"as expected (varimax rotation) with three factorswith eigenvalues greater than 1.0 explaining a totalof 92 percent of the variance.Table 2 shows the correlations and reliabilitiesfor the study's variables. Table 2 also shows accept-able interrater reliabilities for the aggregated vari-ables averaged across all of the teams (James et al.,1984, 1993).

Hypothesis TestingWe included organization-level variables in all

regression analyses. The results are shown withthese variables included.The multidimensionality of team empower-ment. The results of the team empowerment factoranalysis supported Hypothesis 1, which states thatteam empowerment consists of four distinct dimen-sions. For further support, we examined the corre-lations between the dimensions, which, contrary toour expectations, were exceedingly high (for mean-ingfulness and impact, r = .80; for potency andimpact, r = .79; and for potency and meaningful-ness, r = .69). Thus, given the inability of an ex-ploratory factor analysis to provide more solidevidence for multidimensionality (as might be ob-tained with confirmatory factor analysis) and thevery high correlations between the dimensions, weused a composite measure of team empowerment.Spreitzer (1995) used a similar strategy to measureempowerment at the individual level of analysisand also noted that the high correlations between

3 Tables depicting the results of each of the factoranalyses are not included owing to space limitations butare available from the first author upon request.

68 February


13/18

Kirkman and Rosenthe dimensions did not provide sufficient evidencefor discriminant validity. Thus, overall, Hypothesis1 received only mixed support.Job and organizational characteristics. To testthe relationship between the job and organizationalcharacteristics and team empowerment, we enteredall four characteristics simultaneously into a re-gression equation as predictors. To check for mul-ticollinearity, we included measures of tolerance(Darlington, 1990) that did not indicate the pres-ence of multicollinearity. The results showed thatteam empowerment was significantly related to ex-ternal team leader behaviors (/3 = .37, p < .01),production/service responsibilities (8/ = .26, p


14/18

Academy of Management JournalSecond, although we found high correlations be-tween the four empowerment dimensions, the teamempowerment composite explained variance inteam effectiveness that went beyond what auton-omy explained. This pattern underscores the im-portance of not treating self-management and em-

powerment as synonymous concepts. In addition,Table 2 shows that the correlations between auton-omy and the other dimensions were much lowerthan the correlations those dimensions had witheach other. Clearly, autonomy makes up a veryimportant part of team empowerment. A team withno autonomy that compensates with high potency,meaningfulness, and impact is probably not trulyempowered. However, our results suggest that inorder for teams to be highly effective, they must beautonomous and their members must experiencepotency, meaningfulness, and impact (cf. Alper,Tjosvold, & Law, 1998), even though autonomyalone did explain some unique variance in teameffectiveness. In fact, self-managing teams have amuch greater productivity impact on complextasks, which are likely to generate potency, mean-ingfulness, and impact experiences, than on rou-tine takes (Cordery, Wall, & Wright, 1997). Thus,autonomy is most likely a necessary, but not asufficient, condition for team empowerment.Third, we found that external leader behavior in-fluenced team empowerment experiences. Thesefindings support previous theoretical arguments forthese relationships (Culbert & McDonough, 1986;Cummings, 1978; Denison, 1982) and empirical stud-ies of supervisory empowerment (Burpitt &Bigoness,1997) and self-management (Manz & Sims, 1987).Similarly, we found a relationship between produc-tion/service responsibilities and team empowermentexperiences, a finding that supports previous theoret-ical work in this area (Cummings, 1978; Hackman,1987; Manz &Sims, 1993).The link between team-based human resourcespolicies and team empowerment supports previousempirical findings (Blackburn & Rosen, 1993;Frayne & Latham, 1987; Gist et al., 1991) and the-oretical work on teams (Hackman, 1987; Manz &Sims, 1993). In addition, the fact that the mean wereport here for team-based human resources poli-cies is lower than the other antecedents' means (seeTable 2) may indicate that organizations adoptteam pay and peer evaluations at a much slowerrate than other team-related changes or perhaps donot adopt them at all (cf. Mohrman et al., 1995).Changing existing pay and evaluation systems iscomplex and can be emotional for change recipi-ents (DeMatteo et al., 1998; Gibson & Kirkman,1999; Kirkman et al., 1996). Researchers shouldcontinue to examine team-based human resources

policies because our findings indicate that they areintegral drivers of team empowerment and, conse-quently, of team effectiveness.With regard to social structure, we validatedSpreitzer's (1996) individual-level findings at theteam level of analysis and supported previous empir-ical (Gladstein, 1984) and theoretical work (Hack-man, 1987) on structure at the team level of analysis.Specifically, work units that provided sociopoliticalsupport and access to important information and re-sources enhanced team empowerment. In summary,our findings highlight the importance of the organi-zational context in creating team empowerment ex-periences (cf. Wageman, 1997). Organizations shouldattend to context at multiple levels of analysis (fromleader behavior to organization-wide policies) tomore fully realize the benefits of team empowerment(Campbell &Martinko, 1998).Our fourth implication concerns Spreitzer's (1995)suggestion that empowerment researchers expandtheir research by broadening outcome measures toinclude organizational commitment, organizationaleffectiveness, and total quality management. Follow-ing Spreitzer's recommendations proved useful; teamempowerment was positively associated with a broadrange of positive employee and organizational out-comes such as commitment, proactivity, and cus-tomer service. Researchers may have understated thepayoffs for empowered work teams.Perhaps what is needed most now in the teameffectiveness literature is research that examines em-powerment at the individual and team levels simul-taneously (see Tesluk et al. [1996] for an example).Manz (1993) suggested that granting teams more em-powerment might, in fact, detract from individuallevels of empowerment, in that an individual mayactually feel less autonomy on a team where decisionmaking and responsibilities have to be shared amongteam members. Such research could help determinethe specific impact of empowerment at multiple lev-els of analysis and thus identify optimal levels ofempowerment at both the individual and team levels(cf. Uhl-Bien &Graen, 1998). We urge more research-ers to use the team consensus technique in thesefuture studies. Perhaps future research will also helpdetermine when it is most appropriate to use teamconsensus and when it might be suitable to use team-member-aggregated ratings.Team empowerment research should also be con-ducted on various types of teams, including man-agement teams, project teams, and virtual teams, todetermine if the results of our findings with perma-nent work teams are generalizable (Cohen &Bailey,1997). Obviously, empowered teams will not beright for every task or work situation. The challengefor work team and empowerment researchers will

70 February


15/18

Kirkman and Rosenbe to (1) continue moving beyond anecdotal evi-dence of team success to more rigorous researchdesigns that effectively test both the causes andpayoffs of work team empowerment and (2) con-tinue to move beyond self-managing work teammodels to broaden the conceptualization of workteam effectiveness.Managerial Implications

First, we recommend that managers use the teamempowerment scale developed for this research pro-gram to assess the teams they lead. Our findings sug-gest that it is imperative that managers identify thoseteams with low levels of empowerment and engage inactivities designed to raise their levels of empower-ment. Managers should take actions in multiple con-textual arenas and at multiple levels in their organi-zations to (1) ensure that team leaders receive trainingto exhibit appropriate behaviors, such as encouragingteams to solve their own problems and setting highteam expectations, (2) increase the production/ser-vice responsibilities of teams, such as the productionof whole products or the delivery of integrated ser-vices, and allow team members to set their own goals,(3) alter human resource policies, for example, in-creasing the amount of cross-training and trainingteam members to hire and discipline fellow teams,and (4) modify social structures to increase teammember access to resources and information and es-tablish more communication and coordination acrossteams.Second, there is evidence that the team consensus

technique meets more of the theoretical requirementsfor obtaining team-level data than does the aggrega-tion method and that the former is a superior pre-dictor of team-level outcomes (Kirkman, Tesluk, &Rosen, 1998). Managers may want to consider theteam consensus technique as a viable alternative tocollecting team-level data and as a way to overcomethe limitations of the aggregation technique.Limitations

To preclude the questions of causality oftenbegged by cross-sectional studies, in our design weonly included organizations in which the relevantjob and organizational characteristics (the team em-powerment antecedents) were present at the site ororganizational level-not team-by-team. As a par-tial statistical check on whether empowermentpractices were uniform within organizations, ananalysis of variance was conducted, with organiza-tion as the predictor variable and the four job andorganizational characteristics as criterion variables.The test statistic (F) was significant for all four of

the characteristics, which indicated lower variancewithin organizations than between organizations.Thus, the empowerment of the teams in our samplecannot have been a function of their relative effec-tiveness.A cross-sectional design cannot speak to the pos-sibility that team empowerment experiences arecaused by team effectiveness, rather than vice versa.In fact, researchers have suggested that effectivenessand empowerment may be reciprocally related(Spreitzer, 1995) and self-reinforcing (Thomas &Velthouse, 1990) at the individual level of analysis.Lindsley, Brass, and Thomas (1995) made the sameargument for effectiveness and potency at the teamlevel of analysis. Longitudinal studies are needed toexplore these reciprocal effects over time.Conclusion

Our study adds to the growing body of research thathas isolated the effects of work team implementationon organizational effectiveness. We have tried to takea first step with a relatively new construct, team em-powerment. In view of our findings, we hope thatmanagers will attempt to create empowerment expe-riences for their work teams. We also hope that re-searchers will continue to examine team empower-ment and use the team consensus technique to assessimportant team constructs. More generally, we hopethat researchers will keep identifying and assessingthe conditions necessary to make work teams opti-mally effective in organizations.

REFERENCESAlper, S., Tjosvold, D., & Law, K. S. 1998. Interdepen-dence and controversy in group decision making:Antecedents to effectiveself-managing eams.Organ-izational Behavior and Human Decision Processes,74: 33-52.Ancona, D. G. 1990. Outwardbound: Strategiesfor teamsurvival in an organization.Academy of Manage-ment Journal, 33: 334-365.Argyris, C. 1998. Empowerment: The emperor's new

clothes. Harvard Business Review, 76(3):98-105.Bandura,A. 1997. Self-efficacy: The exercise of control.New York:Freeman.Banker,R.D., Field, J.M., Schroeder,R.G.,&Sinha, K. K.1996. Impact of work teams on manufacturing per-formance: A longitudinal field study. Academy ofManagement Journal, 39: 867-890.Baron,R. M., &Kenny,D. A. 1986. The moderator-medi-ator variable distinction in social psychological re-search:Conceptual,strategic,and statistical consid-erations. Journal of Personality and SocietyPsychology, 51: 1173-1182.

1999 71


16/18

Academy of Management JournalBateman, T. S., & Grant, J. M. 1993. The proactive com-

ponent of organizational behavior: A measure andcorrelates. Journal of Organizational Behavior, 14:103-118.Bentler, P. M., & Chou, C. P. 1987. Practical issues instructural modeling. Sociological Methods & Re-

search, 15: 78-117.Blackburn, R. B., & Rosen, B. 1993. Total quality andhuman resources management: Lessons learned from

Baldridge Award-winning companies. Academy ofManagement Executive, 7: 49-60.

Burke, W. 1986. Leadership as empowering others. In S.Srivastva (Ed.), Executive power: 51-77. San Fran-cisco: Jossey-Bass.Burpitt, W. J., & Bigoness, W. J. 1997. Leadership andinnovation among teams: The impact of empower-ment. Small Group Research, 28: 414-423.Campbell, C. R., & Martinko, M. J. 1998. An integrative

attributional perspective of empowerment andlearned helplessness: A multimethod field study.Journal of Management, 24: 173-200.Campion, M. A., Medsker, G. J., & Higgs, A. C. 1993.Relations between work group characteristics andeffectiveness: Implications for designing effectivework groups. Personnel Psychology, 46: 823-8550.Campion, M. A., Papper, E. M., & Medsker, G. J. 1996.Relations between work group characteristics andeffectiveness: A replication and extension. Person-nel Psychology, 49: 429-452.Campion, M. A., Stevens, M. J., &Medsker, G. J. 1996. An

input-process-output (IPO) model of work team ef-fectiveness. Paper presented at the 11th annual con-ference of the Society for Industrial and Organiza-tional Psychology, San Diego.

Cohen, S. G., & Bailey, D. E. 1997. What makes teamswork: Group effectiveness research from the shopfloor to the executive suite. Journal of Management,23: 239-290.Cohen, S. G., &Ledford, G. E., Jr. 1994. The effectivenessof self-managing teams: A quasi-experiment. HumanRelations, 47: 13-43.Cohen, S. G., Ledford, G. E., Jr., & Spreitzer, G. M. 1996.A predictive model of self-managing work team ef-fectiveness. Human Relations, 49: 643-676.Conger, J., & Kanungo, R. 1988. The empowerment pro-cess: Integrating theory and practice. Academy of

Management Review, 13: 471-482.Cordery, J. L., Mueller, W. S., & Smith, L. M. 1991.Attitudinal and behavioral effects of autonomous

group working: A longitudinal field study. Academyof Management Journal, 34: 464-476.

Cordery, J.L., Wall, T. D., &Wright, B. M. 1997. Towards amore comprehensive and integrated approach towork design: Production uncertainty and self-man-aging work team performance. Paper presented at the

12th annual conference of the Society for Industrialand Organizational Psychology, St. Louis.Cotton, J. L., Vollrath, D. A., Froggatt, K. L., Lengnick-Hall, M. L., &Jennings, K. R. 1988 Employee partic-ipation: Diverse forms and different outcomes.

Academy of Management Review, 13: 8-22.Crozier, M. 1964. The bureaucratic phenomenon. Chi-

cago: University of Chicago Press.Culbert, S., & McDonough, J. 1986. The politics of trustand organization empowerment. Public Administra-tion Quarterly, 10: 171-188.Cummings, T. 1978. Self-regulating work groups: A socio-technical synthesis. Academy of Management Re-view, 3: 625-634.Darlington, R. B. 1990. Regression and linear models.New York: McGraw-Hill.Deci, E. L., &Ryan, R. M. 1985. Intrinsic motivation and

self-determination in human behavior. New York:Plenum.

DeMatteo, J. S., Eby, L. T., & Sundstrom, E. 1998. Team-based reward systems: Currentempirical evidence anddirections for future research. In B. M. Staw & L. L.Cummings (Eds.), Research in organizational behav-ior, vol. 20: 141-183. Greenwich, CT:JAIPress.

Denison, D. R. 1982. Sociotechnical design and self-man-aging work groups: The impact on control. Journalof Occupational Behaviour, 3: 297-314.

Fisher, K. 1993. Leading self-directed work teams: Aguide to developing new team leadership skills.New York: McGraw-Hill.

Ford, R. C., & Fottler, M. D. 1995. Empowerment: Amatter of degree. Academy of Management Execu-tive, 9(3): 21-31.Frayne, C. A., & Latham, G. P. 1987. The application ofsocial learning theory to employee self-managementand attendance. Journal of Applied Psychology, 72:387-392.Gibson, C. B., & Kirkman, B. L. 1999. Our past, present,and future in teams: The role of human resources

professionals in managing team performance. InA. L. Kraut & A. K. Korman (Eds.), Evolving prac-tices in human resource management: Responsesto a changing world of work: Forthcoming. SanFrancisco: Jossey-Bass.Gist, M. E., Schwoerer, C., &Rosen, B. 1991. The effectsof alternative training methods on self-efficacy andperformance in computer software training. Journalof Applied Psychology, 74: 884-891.

Gladstein, D. 1984. Groups in context: A model of taskgroup effectiveness. Administrative Science Quar-terly, 29: 499-517.

Goodman, P. S., Devadas, R., & Griffith-Hughson, T. L.1988. Groups and productivity: Analyzing the effec-tiveness of self-managing teams. In J. P. Campbell &R. J. Campbell (Eds.), Productivity in organizations:

February2


17/18

Kirkman and RosenNew perspectives from industrial and organiza-tional psychology: 295-327. San Francisco: Jossey-Bass.

Gorn, G. J., &Kanungo, R. N. 1980. Job involvement andmotivation: Are intrinsically motivated managersmore job involved? Organizational Behavior andHuman Performance, 26: 265-277.

Griffin, R. W. 1991. Effects of work redesign on employeeperceptions, attitudes, and behaviors: A long-terminvestigation. Academy of Management Journal,34: 425-435.

Gulowsen, J. 1972. A measure of work group autonomy.In L. Davis & J. Taylor (Eds.), Design of jobs: 374-390. Harmondsworth, U.K.: Penguin.Guzzo, R. A., Campbell, R. J., Moses, J. L., Ritchie, R. R.,Schneider, B., Shaff, K., Wheeler, J., & Gufstason,P. W. 1991. What makes high-performing teams

effective? Unpublished manuscript, University ofMaryland, College Park.Guzzo, R. A., Yost, P. R., Campbell, R. J., & Shea, G. P.1993. Potency in groups: Articulating a construct.British Journal of Social Psychology, 32: 87-106.Hackman, J. R. 1987. The design of work teams. In J.Lorsch (Ed.), Handbook of organizational behav-ior: 315-342. Englewood Cliffs, NJ: Prentice-Hall.Hackman, J. R., &Morris, C. G. 1975. Group tasks, groupinteraction process, and group performance effec-tiveness: A review and proposed integration. In L. L.Berkowitz (Ed.), Advances in experimental social

psychology, vol. 8: 47-100. New York: Academic.Hackman, J. R., & Oldham, G. R. 1980. Work redesign.

Reading, MA: Addison-Wesley.Hardy, C., & Leiba-O'Sullivan, S. 1998. The power be-hind empowerment: Implications for research and

practice. Human Relations, 51: 451-483.Hyatt, D. E., &Ruddy, T. M. 1997. An examination of the

relationship between work group characteristics andperformance: Once more into the breech. PersonnelPsychology: 50: 553-585.

Ishikawa, K. 1985. What is total quality control, theJapanese way. Englewood Cliffs, NJ: Prentice-Hall.

James, L. R., Demaree, R. G., &Wolf, G. 1984. Estimatingwithin-group interrater reliability with and withoutresponse bias. Journal of Applied Psychology, 69:85-98.

James, L. R., Demaree, R. G., & Wolf, G. 1993. rwg: Anassessment of within-group interrater agreement.Journal of Applied Psychology, 78: 306-309.

Kirkman, B. L., & Rosen, B. 1997. A model of work teamempowerment. In R. W. Woodman & W. A. Pasmore(Eds.),Research in organizational change and devel-opment, vol. 10: 131-167. Greenwich, CT:JAIPress.

Kirkman, B. L., & Shapiro, D. L. 1997. The impact ofcultural values on employee resistance to teams: To-ward a model of globalized self-managing work team

effectiveness. Academy of Management Review,22: 730-757.Kirkman, B. L., Shapiro, D. L., Novelli, L., Jr., & Brett,J. M. 1996. Employee concerns regarding self-man-

aging work teams: A multidimensional justice per-spective. Social Justice Research, 9: 47-67.

Kirkman, B. L., Tesluk, P. E., &Rosen, B. 1998. Compar-ing the aggregation of individual responses versusteam consensus ratings in measuring team leveldata: If you want team level data, shouldn't youask the team? Paper presented at the annual meetingof the Academy of Management, San Diego.

Latham, G. P., & Saari, L. M. 1979. Importance of sup-portive relationships in goal setting. Journal of Ap-plied Psychology, 64: 151-156.

Lawler, E. E., III, Mohrman, S. A., & Ledford, G. E., Jr.1995. Creating high performance organizations:Practices and results of employee involvement andtotal quality management in Fortune 1000 compa-nies. San Francisco: Jossey-Bass.

Lindsley, D. H., Brass, D. J., & Thomas, J. B. 1995. Effi-cacy-performance spirals: A multilevel perspective.Academy of Management Review, 3: 645-678.

Locke, E. A., Shaw, K. N., Saari, L. M., Latham, G. P.1981. Goal setting and task performance: 1969-1980.Psychological Bulletin, 90: 125-152.

Manz, C. C. 1990. Beyond self-managing work teams:Toward self-leading teams in the workplace. In R. W.Woodman & W. A. Pasmore (Eds.), Research in or-ganizational change and development, vol. 4: 273-299. Greenwich, CT: JAI Press.

Manz, C. C. 1993. Teams and empowerment. Paper pre-sented at the annual meeting of the Academy ofManagement, Atlanta.

Manz, C. C., & Sims, H. P., Jr. 1987. Leading workers tolead themselves: The external leadership of self-managing work teams. Administrative ScienceQuarterly, 32: 106-128.

Manz, C. C., & Sims, H. P., Jr. 1993. Business withoutbosses: How self-managing teams are buildinghigh-performance companies. New York: Wiley.

Markoczy, L. 1997. Measuring beliefs: Accept no substi-tutes. Academy of Management Journal, 40: 1228-1242.McGrath, J. E. 1964. Social psychology: A brief intro-duction. New York: Holt, Rinehart & Winston.Miller, K., & Monge, P. 1986. Participation, satisfaction,and productivity: A meta-analytic review. Academy

of Management Journal, 29: 727-753.Mohrman, S. A., Cohen, S. G., &Mohrman, A. M., Jr.1995.Designing team-based organizations: New forms for

knowledge work. San Francisco: Jossey-Bass.Nunnally, J. C., & Bernstein, I. H. 1994. Psychometrictheory (3rd ed.). New York: McGraw-Hill.Osterman, P. 1994. How common is workplace transfor-

1999 73


18/18

beyond self management

Documents