TRANSFORMATIONAL LEADERSHIP AND JOB BEHAVIORS · PDF filetransformational leadership and job behaviors: the mediating role of core job characteristics ronald f. piccolo university

TRANSFORMATIONAL LEADERSHIP AND JOB BEHAVIORS:THE MEDIATING ROLE OF CORE JOB CHARACTERISTICS

RONALD F. PICCOLOUniversity of Central Florida

JASON A. COLQUITTUniversity of Florida

Although the effects of transformational leadership on task performance and organi-zational citizenship behavior (OCB) are well-documented, the mechanisms that ex-plain those effects remain unclear. We propose that transformational leadership isassociated with the way followers view their jobs, in terms of Hackman and Oldham’s(1976) core job characteristics. Results of our study support a structural modelwhereby indirect effects supplement the direct effects of transformational leadershipon task performance and OCB through the mechanisms of job characteristics, intrinsicmotivation, and goal commitment. Additional analyses revealed that transformationalleadership relationships were significantly stronger for followers who perceived high-quality leader-member exchange.

Over the past two decades, transformationalleadership has emerged as one of the most popularapproaches to understanding leader effectiveness.Transformational leadership theory rests on the as-sertion that certain leader behaviors can arouse fol-lowers to a higher level of thinking (Bass, 1985;Burns, 1978). By appealing to followers’ ideals andvalues, transformational leaders enhance commit-ment to a well-articulated vision and inspire fol-lowers to develop new ways of thinking aboutproblems. Indeed, the positive association betweentransformational leadership and follower behaviorsis well documented (Fuller, Patterson, Hester, &Stringer, 1996; Judge & Piccolo, 2004; Lowe, Kro-eck, & Sivasubramaniam, 1996), and studies havebegun to examine the process by which those ef-fects are ultimately realized (e.g., Bono & Judge,2003; Dvir, Eden, Avolio, & Shamir, 2002; Kark,Shamir, & Chen, 2003).

A central tenet of the transformational approachis that such effects are transmitted through followerreactions to a leader. Early studies of the transfor-mational process, therefore, tended to emphasizethe mediating role of followers’ attitudes towardleaders, such as trust, satisfaction, personal identi-fication, and perceived fairness (e.g., Kark et al.,

2003; Pillai, Schreisheim, & Williams, 1999; Pod-sakoff, MacKenzie, Moorman, & Fetter, 1990).Other studies have suggested that transformationaleffects are explained by how followers come to feelabout themselves or their group, in terms of self-efficacy or group potency (Bono & Judge, 2003;Shamir, House, & Arthur, 1993; Sosik, Avolio, &Kahai, 1997).

We propose a different mechanism for explainingthe effects of transformational leaders—one rootednot in perceptions of leader or self, but rather,rooted in the job. One of the more powerful influ-ences a leader can have on followers is in the “man-agement of meaning” (Smircich & Morgan, 1982),as leaders define and shape the “reality” in whichfollowers work. Hackman and Oldham’s (1976,1980) five core job characteristics—variety, iden-tity, significance, autonomy, and feedback—offerone means of capturing key facets of that reality.We therefore tested the model shown in Figure 1, inwhich the direct effects of transformational leader-ship on task performance and organizational citi-zenship behavior (OCB) are supplemented by ef-fects on followers’ perceptions of core jobcharacteristics. Those perceptions then go on topredict intrinsic motivation and goal commit-ment, which are themselves related to the twooutcomes. This study therefore offers a uniqueintegration of two of the more visible literaturesin the organizational behavior domain: transfor-mational leadership and job characteristicstheory.

An earlier version of this article was presented at the2004 annual meeting of the Society for Industrial andOrganizational Psychology, Chicago. We would like tothank Dov Eden and three anonymous reviewers for theirguidance during the review process.

� Academy of Management Journal2006, Vol. 49, No. 2, 327–340.

327

EXPLAINING TRANSFORMATIONALLEADERSHIP EFFECTS

Since its introduction by Burns (1978) and Bass(1985), transformational leadership theory hasevolved to describe four dimensions of leader be-havior. Idealized influence is the degree to whichleaders behave in charismatic ways that cause fol-lowers to identify with them. Inspirational motiva-tion is the degree to which leaders articulate vi-sions that are appealing to followers. Intellectualstimulation is the degree to which leaders chal-lenge assumptions, take risks, and solicit followers’ideas. Individualized consideration is the degree towhich leaders attend to followers’ needs, act asmentors or coaches, and listen to followers’concerns.

Of all the transformational leadership effects,perhaps the most often studied are its associationswith beneficial job behaviors. Transformationalleaders have the ability to raise follower task per-formance while also encouraging OCB—those “ex-tra-role” behaviors that are discretionary and notdirectly recognized by an organization’s formal re-

ward system, and that help improve organizationalfunctioning (Smith, Organ, & Near, 1983). As threeseparate meta-analytic reviews have summarized(Fuller et al., 1996; Judge & Piccolo, 2004; Lowe etal., 1996), the transformational leadership dimen-sions have displayed strong and consistent correla-tions with task performance and OCB acrossorganizations.

According to Bass (1985), transformational lead-ers provide constructive feedback to their follow-ers, convince followers to exhibit extra effort, andencourage followers to think creatively about com-plex problems. As a result, followers tend to behavein ways that facilitate high levels of task perfor-mance. In addition, transformational leaders maketheir organizations’ missions salient and persuadefollowers to forgo personal interests for the sake ofthe collective. When followers equate their ownsuccess with that of their organizations’ and iden-tify with the organizations’ values and goals, theybecome more willing to cooperate in order to makea positive contribution to the work context (Podsa-koff et al., 1990).

FIGURE 1Proposed Model of Transformational Leadership and Core Job Characteristic Effects

328 AprilAcademy of Management Journal

Hypothesis 1. Transformational leadership ispositively related to follower task performanceand OCB.

Transformational Leadership and CoreJob Characteristics

Of course, the contribution of the present studylies not in testing Hypothesis 1, but rather in ex-ploring whether core job characteristics provideone mechanism for explaining those effects. Hack-man and Oldham (1976) introduced job character-istics theory to explain conditions in which em-ployees would be intrinsically motivated whenperforming a job. According to the theory, organi-zations can encourage positive work attitudes andincreased work quality by enhancing jobs alongfive dimensions. These include variety (the degreeto which a job requires the use of a number ofdifferent skills and talents); identity (the degree towhich the job requires completion of a “whole”piece of work, or doing a task from beginning to endwith a visible outcome); significance (the degree towhich the job has a substantial impact on the livesof other people); autonomy (the degree to which thejob provides substantial freedom); and feedback(the degree to which the job provides clear infor-mation about performance levels).

Although perceptions of core job characteristicsare clearly dependent on structural aspects of one’sformal job description, transformational leaderscan foster such perceptions through their own ac-tions. Smircich and Morgan suggested that leadersinfluence followers by “mobilizing meaning, artic-ulating and defining what has previously remainedimplicit or unsaid, by inventing images and mean-ings that provide a focus for new attention, and byconsolidating, confronting, or changing prevailingwisdom” (1982: 258). Leaders “frame” or “bracket”followers’ work experiences to create a new pointof reference for understanding the day-to-day flowof work (Goffman, 1974; Schutz, 1967; Smircich &Morgan, 1982).

This “management of meaning” perspective onleadership is similar to the social information pro-cessing model introduced by Salancik and Pfeffer(1978). These authors suggested that individualsrely on informational cues from their social con-texts when making assessments about work envi-ronments. Leaders, for example, as central charac-teristics of a work context, are relevant informationpoints when followers make judgments about theirjobs. Both Salancik and Pfeffer (1978) and Griffin,Bateman, Wayne, and Head (1987) argued that jobperceptions do not depend exclusively on objectivecharacteristics of actual jobs, but instead on social

constructions of the information available to work-ers at the time they make judgments.

Griffin (1981) was among the first to test thenotion that leaders can influence job perceptionswithout making any adjustments to objective jobcharacteristics. He argued that individual task per-ceptions stem from five basic sources of informa-tion: (1) technology, (2) organizational structure, (3)coworkers, (4) characteristics of a job incumbentand—most relevant in the current study—(5) anindividual’s immediate supervisor. In Griffin’sstudy of leader behaviors and job characteristics,managers reported the extent to which they exhib-ited behaviors intended to influence job percep-tions. Three months later, subordinates in an ex-perimental group reported higher ratings of core jobcharacteristics, even though no tangible changes totheir actual jobs had been made. Griffin (1981) ex-plained these results by suggesting that informa-tional cues from supervisors may have caused em-ployees to perceive their tasks differently.

Transformational leaders may play a particularlystrong role in the management of meaning and so-cial information. Shamir and his coauthors (1993)suggested that leaders who exhibit transforma-tional behaviors can influence how followers judgea work environment by using verbal persuasionand by clearly communicating the value of an or-ganization’s mission. Similarly, Bono and Judge(2003) suggested that transformational leaders helpfollowers view work goals as congruent with theirown values. There are reasons to expect that thesame management of meaning processes may beused to influence job perceptions.

In addition, many of the behaviors subsumed bythe transformational pattern have direct implica-tions for levels of core characteristics. Leaders whoutilize intellectual stimulation by seeking new per-spectives and developing new ways to perform jobtasks may enhance follower perceptions of varietyand autonomy. Leaders who engage in individual-ized consideration by coaching and teachingshould have followers who see more autonomy andfeedback in their jobs. When leaders engage in ide-alized influence (by emphasizing the moral andethical consequences of work decisions) or inspira-tional motivation (by articulating a compelling vi-sion of the future), followers may view their jobs asmore significant. Shamir and colleagues (1993) pro-vided indirect support for these assertions by sug-gesting that leaders who appeal to ideological val-ues and engage in intellection stimulation interjectmeaningfulness into their organization and theirfollowers’ work.

2006 329Piccolo and Colquitt

Hypothesis 2. Transformational leadership ispositively related to follower perceptions ofcore job characteristics.

Effects of Core Job Characteristics

One of the central predictions of job characteris-tics theory is that enhanced core job characteristicsare associated with higher levels of what Hackmanand Oldham termed “internal motivation,” de-scribed as a “self-perpetuating cycle of positivework motivation driven by self-generated (ratherthan external) rewards for good work” (1980: 72).The authors likened internal motivation to Csik-szentmihalyi’s (1975) “flow” concept and Deci’sdiscussion of “intrinsic motivation,” described asinvolvement in “an ongoing process of seeking andconquering challenges” (1975: 131). Our modeluses Deci’s intrinsic motivation terminology,which has come to be the more common label inthe larger motivation literature (e.g., Kanfer, 1991).

Comprehensive summaries of the literature onjob characteristics theory have provided supportfor the notion that jobs regarded as challenging,important, and autonomous are more intrinsicallymotivating. For example, Fried and Ferris (1987)meta-analyzed over 200 studies and reported cor-rected correlations ranging from .22 to .52 betweenthe five core characteristics and intrinsic motiva-tion. Path analysis studies have suggested that theintrinsic motivation effects are driven by multiplecharacteristics and are particularly reliant on thecharacteristics that create perceived meaningful-ness in one’s job (Johns, Xie, & Fang, 1992).

Hypothesis 3. Follower perceptions of core jobcharacteristics are positively related to fol-lower intrinsic motivation.

Hackman and Oldham (1980) argued that em-ployees who are intrinsically motivated engage inhigher levels of task performance because perform-ing well creates positive affect. Staw (1977) made asimilar argument, suggesting that intrinsically mo-tivated individuals derive satisfaction from task ac-complishment and therefore work harder to excel.Although these arguments apply most obviously towork quality, Hackman and Oldham (1980) alsoprovided conceptual arguments for the effects ofintrinsic motivation on work quantity. Specifically,the authors argued that intrinsic motivation shouldreduce the forms of task withdrawal (e.g., day-dreaming, breaks, socializing) that slow work ef-forts. Kanfer (1991) made a similar point, notingthat one of the key mechanisms for explaining theperformance effects of intrinsic motivation is con-sistency of task engagement. Intrinsically moti-

vated individuals are more likely to choose to workon particular tasks at a given moment. Indeed, thatengagement not only increases work quantity overtime, but can also improve the acquisition of task-related skills, thereby affecting work quality (Kan-fer, 1991).

Unfortunately, these types of arguments have re-ceived very little empirical testing. Most of theresearch on intrinsic motivation in the larger moti-vation literature has treated it as a dependent vari-able and used either job characteristics or the pres-ence of extrinsic rewards as predictors of motivatedbehavior (see Fried and Ferris [1987] and Deci,Koestner, and Ryan [1999] for meta-analytic re-views). The most recent studies linking intrinsicmotivation with task performance have come out ofthe related area of psychological empowerment(e.g., Thomas & Velthouse, 1990). For example,Spreitzer (1995) linked the empowerment versionof intrinsic motivation to subordinate perceptions ofmanagerial task performance in an industrial firm.

Research relating intrinsic motivation to OCB issurprisingly rare, yet there appear to be naturallinks between the two constructs. Individualslikely execute discretionary behaviors that go be-yond the formal requirements of a job to satisfysome higher-order individual need or to align workbehavior with individual values. Because such be-haviors are less likely to be formally rewarded thanare required job behaviors, they are presumablyperformed for self-generated, intrinsic reasons.Some support for this assertion was found by Leeand Allen (2002), who linked “intrinsic cognitions”to some forms of OCB, and by Rioux and Penner(2001), who linked the outcome to an organization-al concern motive that captured, among otherthings, having a genuine interest in work.

Hypothesis 4. Follower intrinsic motivation ispositively related to follower task performanceand OCB.

Our model also draws a link between job charac-teristics and goal commitment in explicating thereasons for transformational effects. FollowingLocke, Shaw, Saari, and Latham (1981), Hollen-beck, Williams, and Klein defined goal commit-ment as “the determination to try for a goal and thepersistence in pursuing it over time” (1989: 18). Weare not aware of any research linking the five corejob characteristics to goal commitment, but thereare conceptual reasons to expect a relationship.Klein, Wesson, Hollenbeck, and Alge (1999), forexample, identified antecedents of goal commit-ment beyond expectancy and goal attainment, thetwo most proximal drivers of commitment. Amongthose antecedents was volition, which overlaps


substantially with autonomy, and performancefeedback, which is a central aspect of job charac-teristics theory. Both volition and feedback wereproposed to have an influence on goal commitmentthrough their effects on expectations of goal attain-ment. Thus, by shaping expectancies for accom-plishment of work-related goals, core characteris-tics appear to be linked to expressions of goalcommitment.

Hypothesis 5. Follower perceptions of core jobcharacteristics are positively related to fol-lower goal commitment.

In most studies of goal setting theory, goal com-mitment has been regarded as a moderator of therelationship between goal difficulty and perfor-mance. However, Locke, Latham, and Erez (1988)suggested that goal commitment would also have adirect link to performance and introduced a modelof job performance with goal commitment as themost proximal predictor. Drawing on earlier workby Salancik (1977), the authors noted that commit-ment is the binding of individuals to specific be-havioral acts. Thus, individuals who are more com-mitted to goals should try harder to achieve themand persist in that effort longer (Locke et al., 1988).Such persistence is one of the most powerful driv-ers of task performance (Locke & Latham, 2002). Insupport of this view, two meta-analyses concludedthat goal commitment does indeed have a positive,moderate relationship with task performance(Klein et al., 1999; Wofford, Goodwin, & Premack,1992).

In contrast, the relationship between goal com-mitment and OCB remains unclear and untested.On the one hand, many individual goals do notinclude a citizenship component, and extra-rolebehaviors that distract from goal-based dutiesmight hinder the achievement of such goals. On theother hand, group- or organization-level goals in-herently encourage cooperation and helping behav-iors among employees, and a commitment to thosegoals should facilitate altruistic behaviors that arenot directly rooted in an individual’s job descrip-tion. Indeed, a leader’s ability to foster acceptanceof group goals is a strong predictor of OCB (Podsa-koff, MacKenzie, Paine, & Bachrach, 2000).

Hypothesis 6. Follower goal commitment ispositively related to follower task performanceand OCB.

Alternative Models

In summary, we propose that the relationshipbetween transformational leadership and beneficial

job behaviors (task performance and OCB) is par-tially mediated by perceptions of core job charac-teristics, which are related to intrinsic motivationand goal commitment. The model in Figure 1 there-fore offers a new, relatively unexplored set ofmechanisms for transformational effects on job be-haviors. Of course, one could envision alternatestructures for Figure 1 that could also have merit.For example, it may be that the job characteristicsmechanisms fully mediate transformational effects,suggesting that direct paths to task performanceand OCB should be dropped. On the other hand, itmay be that Figure 1 understates the direct effectsof transformational leadership, which may havedirect links with goal commitment and intrinsicmotivation (Shin & Zhou, 2003). We thereforetested two alternate models that provide differentrepresentations of transformational effects. Figure 2depicts these alternate models.

Potential Boundary Condition

In addition, we explored a potential “boundarycondition” for transformational effects. In describ-ing leaders as managers of meaning, Smircich andMorgan noted that some followers can choose toresist such management, writing that leadership“involves a complicity or process of negotiationthrough which certain individuals, implicitly orexplicitly, surrender their power to define the na-ture of their experience to others. Indeed, leader-ship depends on the existence of individuals will-ing, as a result of inclination or pressure, tosurrender, at least in part, the powers to shape anddefine their own reality” (Smircich & Morgan,1982: 258). This reasoning implies that some fol-lowers are resistant to transformational behaviors.One construct that could capture such resistance isleader-member exchange (LMX) quality (Graen &Uhl-Bien, 1995). Followers in high-quality LMXrelationships report high levels of trust in leadersand commitment to their visions, so they may bemore responsive to transformational behaviors. Incontrast, followers in low-quality LMX relation-ships have formal, impersonal communication pat-terns with leaders that could prove insufficient fortransmitting changes in job perceptions.

METHODS

Respondents

Respondents included 283 individuals from abroad cross-section of job types. The most commonjob categories were administration/support (11%),Web design and computer networking/technology


(10%), and education/training (9%); but as partic-ipants were drawn from multiple organizations inseveral industries, we cannot assert that the sampleis representative of any definable population. For-ty-four percent of the respondents were men, 56percent were women, and 78 percent were Cauca-sian. Their average age was 34 years, and theiraverage tenure in their current jobs was 7 years. Werecruited respondents from the StudyResponse ser-vice (Stanton & Weiss, 2002), a nonprofit academicservice that attempts to match researchers in needof samples with individuals willing to completesurveys. In exchange for this service, the StudyRe-sponse researchers examine the relationship be-

tween study characteristics (e.g., survey length)and survey effectiveness (e.g., response rate, miss-ing data rates). As of the most recent update (Au-gust 10, 2005), 95,574 individuals had registeredwith the StudyResponse service.

For the present study, recruits were limited tofull-time employees who reported to a supervisor.A random sample of 1,491 such employees wasgenerated. The StudyResponse staff sent out re-cruitment e-mails with links to an online survey. Inaccordance with our Institutional Review Board’sprotocols, participants were told that the researchwas voluntary and that the study pertained to “therelationship between job attitudes and job behav-

FIGURE 2Alternate Models of Transformational Leadership and Core Job Characteristic Effects


iors.” Respondents were further told that theywould receive a $10 Amazon.com gift certificate ifthey filled out the survey and their supervisorsfilled out a shorter set of questions. Respondentssigned on to the online survey using their StudyRe-sponse ID number, which was the only identifierincluded with their data. Once the participants hadfilled out their survey, they e-mailed their supervi-sors a link to the supervisory survey, so the super-visor data were identified with the same ID num-bers. A total of 283 individuals completed the self-survey, resulting in a response rate of 19 percent.Of these, 217 had supervisors who filled out thesupervisor survey; thus, the overall response ratewas 15 percent.

Measures

Unless otherwise indicated, all measures used aresponse scale in which 1 was “strongly disagree”and 5 was “strongly agree.”

Transformational leadership. The four dimen-sions of transformational leadership were mea-sured with items from the Multifactor LeadershipQuestionnaire (MLQ Form 5X; Bass & Avolio,1995). Four items were used to measure intellectualstimulation (e.g., “My supervisor . . . seeks differ-ing perspectives when solving problems”), inspira-tional motivation (e.g., “. . . articulates a compel-ling vision of the future”), and individualizedconsideration (e.g., “. . . treats me as an individualrather than just a member of a group”). Eight itemswere used to measure idealized influence (e.g.,“. . . instills pride in me for being associated withhim/her”). Transformational leadership was mea-sured at the individual level because the level oftheory—dictated by the outcome variables—was atthe individual level (Rousseau, 1985).

Job characteristics. The ten Likert items fromthe revised form of the Job Diagnostic Survey(Idaszak & Drasgow, 1987; see Hackman & Oldham,1974) were used. On a seven-point scale (1, “veryinaccurate,” to 7, “very accurate”), participants in-dicated the accuracy of statements such as, “Thejob requires me to use a number of complex high-level skills” (variety), “The job provides me thechance to completely finish the pieces of work Ibegin” (identity), “The job is very significant andimportant in the broader scheme of things” (signif-icance), “The job gives me considerable opportu-nity for independence and freedom in how I do thework” (autonomy), and “After I finish a job, I knowwhether I have performed well” (feedback).

Intrinsic motivation. This variable was mea-sured with four items developed by Hackman andOldham (1974). Sample items are, “My opinion of

myself goes up when I do this job well” and “I feelbad and unhappy when I discover that I have per-formed poorly on this job.”

Goal commitment. The five items validated byKlein, Wesson, Hollenbeck, Wright, and DeShon(2001), based on the earlier work of Hollenbeck,Klein, O’Leary, and Wright (1989), were used. Sam-ple items (both reverse-coded) are, “It’s hard to takemy work goals seriously” and “Quite frankly, Idon’t care if I achieve my work goals or not.”

LMX. This was assessed using Graen and Uhl-Bien’s (1995) seven-item scale. Sample items are,“How would you characterize your working rela-tionship with your supervisor?” (1, “extremely in-effective,” to 5, “extremely effective”) and “Howwell does your supervisor recognize your poten-tial?” (1, “not at all,” to 5, “fully”).

Task performance. Supervisors were asked tocomplete the seven-item scale developed by Wil-liams and Anderson (1991). Supervisors indicatedthe extent to which they agreed with statementsabout their subordinates’ performance, such as,“This employee . . . adequately completes assignedduties” and “. . . fulfills responsibilities specifiedin his/her job description.”

OCB. Supervisors also completed the 16-itemmeasure of OCB published by Lee and Allen (2002),indicating the extent to which they agreed withstatements about their subordinates’ behavior.Items included, “This employee . . . helps otherswho have been absent,” “. . . assists others withtheir duties,” “. . . attends functions that are notrequired but that help the organizational image,”and “. . . offers ideas to improve the functioning ofthe organization.”

Measurement model. Confirmatory factor analy-ses revealed the existence of coding factors inscales on which minorities of items were worded inan opposite direction (i.e., were reverse-coded; seeSchmitt and Stults [1985] for a discussion of thiscommon problem). These items exhibited lowerfactor loadings and/or highly correlated errorterms. We therefore dropped the negatively wordeditem in the intrinsic motivation scale, both posi-tively worded items in the goal commitment scale,and both negatively worded items in the task per-formance scale. Our analyses also revealed an un-acceptably low loading for a task performance itemwith vague item content (“Engages in activities thatwill directly affect his/her performance”), so thisitem was also dropped. The four facet scores wereused as manifest indicators of the latent transfor-mational leadership factor, and the same techniquewas used to represent job characteristics. Given thelength of the OCB scale, we used four 4-item par-cels as manifest indicators of the latent variable.


The resulting measurement model provided an ad-equate fit to the data (�2[215] � 467.22; �2/df �2.17; CFI � .93; SRMR � .05; RMSEA � .08).

RESULTS

Table 1 provides the descriptive statistics andcorrelations for the study variables. We assessedthe fit of the structural model in Figure 1 by addingthe predicted paths to the measurement model. Weallowed the disturbance terms for task performanceand OCB to covary in order to provide a noncausalassociation between the two. The structural modelprovided an adequate fit to the data (�2[219] �481.93; �2/df � 2.20; CFI � .92; SRMR � .07;RMSEA � .08). The path coefficients in Figure 1supported all of the hypotheses except for Hy-pothesis 6 (goal commitment and OCB). Transfor-mational leadership was significantly related totask performance and OCB and, more importantto this study, transformational leadership wasalso significantly related to perceptions of corejob characteristics, which were related to intrin-sic motivation and goal commitment. Intrinsicmotivation was then related to both outcomes,while goal commitment was only related to taskperformance.

We compared the fit of our hypothesized modelto the two alternate models in Figure 2. The firstmodel predicted complete mediation of transfor-mational leadership effects, with no direct pathsbetween leadership and the outcomes included.This model exhibited a moderately poorer fit to thedata (�2[221] � 522.66; �2/df � 2.36; CFI � .91;SRMR � .12; RMSEA � .08), and that difference infit was statistically significant (�2[2] � 40.73, p �.001). The second model added two other transfor-mational paths, to intrinsic motivation and goalcommitment. The fit of this model was almost iden-tical to that of Figure 1 (�2[217] � 476.76; �2/df �2.20; CFI � .92; SRMR � .06; RMSEA � .08), withthe difference in fit nonsignificant (�2[2] � 5.17,

n.s.). Figure 1 therefore displays a more parsimoni-ous model that achieves the same fit level.

As noted in the Introduction, we exploredwhether LMX moderated the transformational rela-tionships in Figure 1 using moderated regression.Transformational leadership explained 10 percentof the variance in core characteristic perceptions(p � .001), with LMX explaining an incremental 1percent (p � .10). The product of the two explainedan additional 2 percent (p � .05). Transformationalleadership explained 9 percent of the variance intask performance (p � .001), with LMX explainingan incremental 2 percent (p � .05). The productterm explained an incremental 8 percent (p � .001).Transformational leadership explained 21 percentof the variance in OCB (p � .001), with LMX notcontributing incremental variance. However, theproduct term explained an additional 9 percent ofthe variance (p � .05). The plots of these interac-tions are shown in Figure 3, with each illustratingthat transformational leadership relationships arestronger when LMX is high rather than low.

DISCUSSION

In this study, we introduced and tested a modelthat used job characteristics theory as a tool forexplaining the relationship between transforma-tional leadership and beneficial job behaviors.What stands out most from our results is that fol-lowers of truly exceptional leaders regarded theirjobs as more challenging and important. That is,followers of leaders who engaged in inspirationalmotivation, idealized influence, intellectual stimu-lation, and individualized consideration behaviorsperceived higher levels of Hackman and Oldham’s(1976, 1980) five core job characteristics (variety,identity, significance, autonomy, and feedback).We had reasoned that transformational leaderscould “manage meaning” by using language andimagery to frame followers’ job experiences (Salan-cik & Pfeffer, 1978; Smircich & Morgan, 1982). This

TABLE 1Descriptive Statistics and Zero-Order Correlationsa

Variable � Mean s.d. 1 2 3 4 5 6

1. Transformational leadership .96 3.43 0.872. Core characteristic perceptions .90 5.20 1.07 .32*3. Intrinsic motivation .68 3.84 0.65 .33* .25*4. Goal commitment .86 3.96 0.83 .27* .35* .52*5. Task performance .93 4.31 0.57 .30* .36* .30* .42*6. Organizational citizenship behavior .94 4.07 0.64 .46* .41* .35* .30* .64*7. Leader-member exchange .93 3.59 0.94 .70* .32* .40* .33* .31* .38*

a n � 202 after listwise deletion of missing data.*p � .05.


management of meaning provides one explanationfor the association observed in our data.

The relationship between transformational lead-ership and core job characteristics is important, asthose characteristics were significantly related tointrinsic motivation. Hackman and Oldham (1976,

1980) pointed to the importance of intrinsic moti-vation, suggesting that individuals whose actionswere powered by self-generating, internal rewardswould prove to be more effective employees. Ourresults support this assertion, as intrinsic motiva-tion was significantly associated with task perfor-

FIGURE 3aLeader-Member Exchange as a Boundary Condition of Transformational Leadership Effects

on Core Characteristic Perceptions

FIGURE 3bLeader-Member Exchange as a Boundary Condition of Transformational Leadership Effects

on Task Performance


mance. Moreover, our study is among the first tolink intrinsic motivation to OCB, as individualsdriven by self-generating rewards were more likelyto perform discretionary behaviors that are rarelyassociated with external rewards.

The relationship between transformational lead-ership and job characteristics is important for an-other reason, as job characteristics were signifi-cantly related to goal commitment. Goalcommitment, which is characterized by a determi-nation to try for goals and to persist in that pursuitover time, has been linked to task performanceacross a wide variety of contexts (Klein et al., 1999;Wofford et al., 1992). Yet this is perhaps the firststudy to link job characteristics to goal commit-ment—a significant predictor of task performancein our sample. We are also among the first to ex-plore the relationship between goal commitmentand OCB, which was near zero in our sample. Goal-committed individuals may regard extra-role be-haviors as a means to support team-level goals, butthey may also view extra-role behaviors as a dis-traction from individual goal achievement. Thus,the link between goal commitment and OCB re-mains unclear.

The marriage of transformational leadership andjob characteristics theory opens up a new domainof potential mediators of transformational effects.Our focus on core job characteristics complementsthe content of other research that is beginning toexplore the process whereby transformational lead-

ership is associated with beneficial job behaviors.That is, our results point to a potential job-basedmediator that adds to the leader-based and self-based mechanisms explored in past research, suchas trust, satisfaction, identification, perceived fair-ness, and efficacy (e.g., Bono & Judge, 2003; Dvir etal., 2002; Kark et al., 2003; Pillai et al., 1999; Pod-sakoff et al., 1990; Shamir et al., 1993). In line withthat other research, our tests of alternate modelsshowed that the core job characteristics only par-tially mediated transformational relationships; thecompletely mediated model in Figure 2 provided asignificantly worse fit to the data.

We should also note that our LMX results pointto potential boundary conditions for transforma-tional relationships. Although the results should beinterpreted with the requisite caution, given thatthey were not predicted a priori, transformationalrelationships were significantly stronger for follow-ers who perceived high-quality exchange relation-ships with their supervisors. Smircich and Morgan(1982) noted that leaders could not manage orframe the realities of their followers if those indi-viduals did not grant the leaders such power. Itmay be that followers in high-quality LMX relation-ships are more open to the social influence of trans-formational leaders, given that the leaders haveearned their trust and commitment. It may also bethat the iterative role-taking/role-making processthat has occurred in high-quality LMX relation-ships (Graen & Scandura, 1987) makes followers

FIGURE 3cLeader-Member Exchange as a Boundary Condition of Transformational Leadership Effects

on Organizational Citizenship Behaviors


more open to being “stretched” by behaviors suchas intellectual stimulation or individualized con-sideration. In contrast, the occurrence of transfor-mational behaviors in relationships with low-qual-ity LMX could be met with resistance, becauseleaders may not have engaged in the more routineactions needed to foster effective working relation-ships. At the very least, this result points to theneed for more integration of leadership perspec-tives in future research, as many studies in theliterature focus on a single theory or approach.

Limitations

The theoretical contributions discussed aboveshould be interpreted in light of this study’s limi-tations. The data that were collected are cross-sec-tional, so alternative explanations for observed re-sults may exist. There may be ambiguity in causaldirection, for example, such that highly enrichedjobs provide a medium for the emergence of trans-formational leadership, and jobs regarded as lowalong the core characteristics constrain transforma-tional behaviors and ultimately make transforma-tional leadership impractical. Individual character-istics of followers may also have an important rolein the transformational leadership process (Dvir &Shamir, 2003) in such a way that some traits (e.g.,extraversion, self-esteem) both facilitate assign-ments to jobs rich in core characteristics and alsomake employees more receptive to transforma-tional leadership. It may also be that followers whoengage in high levels of task performance and OCBenable leaders to act more transformationally, ac-counting for the observed associations.

Another possibility is that the observed relation-ship between transformational leadership and per-ceived job characteristics is spurious, in that anunmodeled variable (e.g., affective disposition) ac-counts for the association between the two. Somelinks in the model did include variables measuredfrom a common source, raising concerns about ef-fect size inflation due to same-source bias. How-ever, we collected the two outcome variables (taskand citizenship performance) from supervisors;thus, their source differs. All variables were col-lected with survey measures, and are therefore sub-ject to monomethod bias. We should also note thatour survey measure of goal commitment was notreferenced to explicitly assigned or identifiedgoals—a departure from common practice in thatliterature.

The sample used in the study also offers someimportant limitations. Because of our use of theStudyResponse service, we have little knowledgeof who respondents were or why they chose to

participate. For example, the individuals whosigned up for the service may be more likely toengage in OCB on their jobs, or be more likely toearn high ratings of task performance. We cantherefore make no claims as to the generalizabilityor representativeness of our results. Our samplingstrategy also prevented us from collecting data frommultiple followers per leader and aggregating datato the unit level of analysis, something that hasbecome a common approach in the transforma-tional leadership literature (e.g., Bass, Avolio, Jung,& Berson, 2003).

Practical Implications

If the association between transformational lead-ership and core job characteristics does indeed rep-resent a causal connection between the two, thenour results have important implications for jobsthat are low on those characteristics. Traditionally,the job design literature has looked to potentiallyexpensive and time-consuming initiatives like jobenlargement or enrichment to boost core character-istic levels. Our results suggest that leaders couldinfluence perceived core characteristic levels bychanging the language, imagery, and symbols usedto communicate meaning on the job. This is not toassert that the manipulation of imagery should re-place well-grounded enrichment of objective jobcharacteristics when feasible, only that task percep-tions can be influenced by how leaders communi-cate with followers. Indeed, both tangible, objectivecharacteristics of a job and an individual’s socialconstruction of information provided by a workcontext determine task perceptions (Griffin et al.,1987). Alternatively, day-to-day job assignmentsand interactions could be altered with the goal ofusing transformational actions to stretch followersin such a way that perceptions of the core charac-teristics are fostered.

Of course, such efforts would require specificinitiatives geared toward increasing leaders’ use oftransformational language and imagery. Transfor-mational behaviors could be incorporated into thetraining courses that new leaders are often requiredto complete. Dvir and coauthors (2002) recentlydemonstrated that transformational leadershiptraining could improve follower outcomes and thattransformational training yielded better resultsthan did eclectic leadership training (see alsoBarling, Weber, & Kelloway, 1996). Such trainingprograms often occur after a detailed needs assess-ment that identifies which units are most in need ofsuch training. Assessments of core job characteris-tics could be used as one tool in such a needsassessment. However, we would also suggest build-


ing a transformational component into the yearlydevelopmental assessments (e.g., managerial skillssurveys, 360-degree feedback instruments) thatleaders fill out, to make the improvement of trans-formational behaviors more continuous.

Suggestions for Future Research

This study offers some suggestions for future re-search. First and foremost, experimental and longi-tudinal research in the laboratory and in the field isneeded to ascertain the causal nature of the rela-tionship between transformational leadership andthe five core characteristics. Such work could alsobring a finer-grained approach to the topic by clar-ifying the mediators of the relationship or by iden-tifying which aspects of the transformational pat-tern account for favorable task assessments. Forexample, does intellectual stimulation have a moreor less powerful effect on job perceptions than in-dividualized consideration or idealized influence?Moreover, future research should broaden the job-focused concepts connected to transformationalleadership, as leaders could manage meaning inother respects. For example, transformational lead-ers could alter the way stressors are perceived onthe job, potentially framing stressors as “chal-lenges” rather than “hindrances” (LePine, LePine,& Jackson, 2004). In addition, future studies couldexamine the mediating role of goal commitmentwith studies that manipulate or measure specificgoals, rather than the general goals referenced byour measure.

Future research should also explore boundaryconditions for transformational relationships, bothwith job characteristics and with beneficial job be-haviors like task performance or OCB. If such re-search replicated the LMX moderation effect foundhere, and depicted in Figure 3, then a potentiallypowerful boundary condition will have been iden-tified. Of course, the quality of leader-member ex-change is only one means of classifying follower-authority relationships (e.g., Shore & Coyle-Shapiro, 2003). Other examples include assessmentof psychological contract breach and of the amountof support offered by the authority. These variablescould have moderating effects that mimic the LMXresults observed in our data. If so, such effectswould shed more light on when transformationalleaders can and cannot be expected to have themost significant effects on follower reactions.

REFERENCES

Barling, J., Weber, T., & Kelloway, E. K. 1996. Effects oftransformational leadership training on attitudinal

and financial outcomes: A field experiment. Journalof Applied Psychology, 81: 827–832.

Bass, B. M. 1985. Leadership and performance beyondexpectations. New York: Free Press.

Bass, B. M., & Avolio, B. J. 1995. MLQ Multifactor Lead-ership Questionnaire for research. Redwood City,CA: Mind Garden.

Bass, B. M., & Avolio, B. J., Jung, D. I., & Berson, Y. 2003.Predicting unit performance by assessing transfor-mational and transactional leadership. Journal ofApplied Psychology, 88: 207–218.

Bono, J. E., & Judge, T. A. 2003. Self-concordance atwork: Toward understanding the motivational ef-fects of transformational leaders. Academy of Man-agement Journal, 46: 554–571.

Burns, J. M. 1978. Leadership. New York: Harper & Row.

Csikszentmihalyi, M. 1975. Beyond boredom and anxi-ety. San Francisco: Jossey-Bass.

Deci, E. L. 1975. Notes on the theory and metatheory ofintrinsic motivation. Organizational Behavior andHuman Performance, 15: 130–145.

Deci, E. L., Koestner, R., & Ryan, R. M. 1999. A meta-analytic review of experiments examining the effectsof extrinsic rewards on intrinsic motivation. Psycho-logical Bulletin, 125: 627–668.

Dvir, T., Eden, D., Avolio, B. J., & Shamir, B. 2002. Impactof transformational leadership on follower develop-ment and performance: A field experiment. Acad-emy of Management Journal, 45: 735–744.

Dvir, T., & Shamir, B. 2003. Follower developmentalcharacteristics as predicting transformational leader-ship: A longitudinal field study. Leadership Quar-terly, 14: 327–344.

Fried, Y., & Ferris, G. R. 1987. The validity of the jobcharacteristics model: A review and meta-analysis.Personnel Psychology, 40: 287–322.

Fuller, J. B., Patterson, C. E. P., Hester, K., & Stringer,D. Y. 1996. A quantitative review of research oncharismatic leadership. Psychological Reports, 78:271–287.

Goffman, E. 1974. Frame analysis. New York: HarperColophon Books.

Graen, G. B., & Scandura, T. 1987. Toward a psychologyof dyadic organizing. In L. L. Cummings & B. M.Staw (Eds.), Research in organizational behavior,vol. 9: 175–208. Greenwich, CT: JAI Press.

Graen, G. B., & Uhl-Bien, M. 1995. Relationship-basedapproach to leadership: Development of leader-member exchange (LMX) theory of leadership over25 years. Leadership Quarterly, 6: 219–247.

Griffin, R. W. 1981. Supervisory behaviour as a source ofperceived task scope. Journal of Occupational Psy-chology, 54: 175–182.

Griffin, R. W., Bateman, T. S., Wayne, S. J., & Head, T. C.


1987. Objective and social factors as determinants oftask perceptions and responses: An integrated per-spective and empirical investigation. Academy ofManagement Journal, 30: 501–523.

Hackman, J. R., & Oldham, G. R. 1974. The Job DiagnosticSurvey: An instrument for the diagnosis of jobs andthe evaluation of job redesign projects. Catalog ofSelected Documents in Psychology, 4: 148–149.

Hackman, J. R., & Oldham, G. R. 1976. Motivationthrough the design of work: Test of a theory. Orga-nizational Behavior and Human Performance, 16:250–279.

Hackman, J. R., & Oldham, G. R. 1980. Work redesign.Reading, MA: Addison-Wesley.

Hollenbeck, J. R., Klein, H. J., O’Leary, A. M., & Wright,P. M. 1989. Investigation of the construct validity ofa self-report measure of goal commitment. Journal ofApplied Psychology, 74: 951–956.

Hollenbeck, J. R., Williams, C. R., & Klein, H. J. 1989. Anempirical examination of the antecedents of commit-ment to difficult goals. Journal of Applied Psychol-ogy, 74: 18–23.

Idaszak, J. R., & Drasgow, F. 1987. A revision of the JobDiagnostic Survey: Elimination of a measurementartifact. Journal of Applied Psychology, 72: 69–74.

Johns, G., Xie, J. L., & Fang, Y. 1992. Mediating andmoderating effects of job design. Journal of Manage-ment, 18: 657–676.

Judge, T. A., & Piccolo, R. F. 2004. Transformational andtransactional leadership: A meta-analytic test oftheir relative validity. Journal of Applied Psychol-ogy, 89: 755–768.

Kanfer, R. 1991. Motivation theory and industrial andorganizational psychology. In M. D. Dunnette & L. M.Hough (Eds.), Handbook of industrial and organi-zational psychology, vol. 1: 75–170. Palo Alto, CA:Consulting Psychologists Press.

Kark, R., Shamir, B., & Chen, G. 2003. The two faces oftransformational leadership: Empowerment and de-pendency. Journal of Applied Psychology, 88: 246–255.

Klein, H. J., Wesson, M. J., Hollenbeck, J. R., & Alge, B. J.1999. Goal commitment and the goal-setting process:Conceptual clarification and empirical synthesis.Journal of Applied Psychology, 84: 885–896.

Klein, H. J., Wesson, M. J., Hollenbeck, J. R., Wright,P. M., & DeShon, R. P. 2001. The assessment of goalcommitment: A measurement model meta-analysis.Organizational Behavior and Human DecisionProcesses, 85: 32–55.

Lee, K., & Allen, N. J. 2002. Organizational citizenshipbehavior and workplace deviance: The role of affectand cognitions. Journal of Applied Psychology, 87:131–142.

LePine, J. A., LePine, M. A., & Jackson, C. L. 2004. Chal-lenge and hindrance stress: Relationships with ex-

haustion, motivation to learn, and learning perfor-mance. Journal of Applied Psychology, 89: 883–891.

Locke, E. A., & Latham, G. P. 2002. Building a practicallyuseful theory of goal setting and task motivation.American Psychologist, 57: 705–717.

Locke, E. A., Latham, G. P., & Erez, M. 1988. The deter-minants of goal commitment. Academy of Manage-ment Review, 13: 23–39.

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.

Lowe, K. B., Kroeck, K. G., & Sivasubramaniam, N. 1996.Effectiveness correlates of transformational leader-ship and transactional leadership: A meta-analyticreview of the MLQ literature. Leadership Quarterly,7: 385–425.

Pillai, R., Schreisheim, C. A., & Williams, E. A. 1999.Fairness perceptions and trust as mediators for trans-formational leadership and transactional leadership:A two-sample study. Journal of Management, 25:897–933.

Podsakoff, P. M., MacKenzie, S. B., Moorman, R. H., &Fetter, R. 1990. Transformational leader behaviorsand their effects on followers’ trust in leader, satis-faction, and organizational citizenship behaviors.Leadership Quarterly, 1: 107–142.

Podsakoff, P. M., MacKenzie, S. B., Paine, J. B., & Bach-rach, D. G. 2000. Organizational citizenship behav-iors: A critical review of the theoretical and empiri-cal literature and suggestions for future research.Journal of Management, 26: 513–563.

Rioux, S. M., & Penner, L. A. 2001. The causes of organ-izational citizenship behavior: A motivational analy-sis. Journal of Applied Psychology, 86: 1306–1314.

Rousseau, D. M. 1985. Issues of level in organizationalresearch: Multi-level and cross-level perspectives. InL. L. Cummings & B. M. Staw (Eds.), Research inorganizational behavior, vol. 7: 1–37. Greenwich,CT: JAI Press.

Salancik, G. R. 1977. Commitment and the control oforganizational behavior and belief. In B. M. Staw &G. Salancik (Eds.), New directions in organizationalbehavior: 1–54. Chicago: St. Clair.

Salancik, G. R., & Pfeffer, J. 1978. A social informationprocessing approach to job attitudes and job design.Administrative Science Quarterly, 23: 224–254.

Schmitt, N., & Stults, D. M. 1985. Factors defined bynegatively keyed items: The result of careless re-spondents? Applied Psychological Measurement,9: 367–373.

Schutz, A. 1967. Collected papers I: The problem ofsocial reality. The Hague: Martinus Nijhoff.

Shamir, B., House, R. J., & Arthur, M. A. 1993. Themotivational effects of charismatic leadership: A


self-concept based theory. Organization Science, 4:577–594.

Shin, S. J., & Zhou, J. 2003. Transformational leadership,conservation, and creativity: Evidence from Korea.Academy of Management Journal, 46: 703–714.

Shore, L. M., & Coyle-Shapiro, J. M. 2003. New develop-ments in the employee-organization relationship.Journal of Organizational Behavior, 24: 443–450.

Smircich, L., & Morgan, G. 1982. Leadership: The man-agement of meaning. Journal of Applied BehavioralScience, 18: 257–273.

Smith, C. A., Organ, D. W., & Near, J. P. 1983. Organiza-tional citizenship behavior: Its nature and anteced-ents. Journal of Applied Psychology, 68: 653–663.

Sosik, J. J., Avolio, B. J., & Kahai, S. S. 1997. Effects ofleadership style and anonymity on group potencyand effectiveness in a group decision support systemenvironment. Journal of Applied Psychology, 82:89–103.

Spreitzer, G. M. 1995. Psychological empowerment inthe workplace: Dimensions, measurement, and vali-dation. Academy of Management Journal, 38:1442–1465.

Stanton, J. M., & Weiss, E. M. 2002. Online panels forsocial science research: An introduction to theStudyResponse project (Technical report no. 13001;www.studyresponse.com). Syracuse, NY: SyracuseUniversity School of Information Studies.

Staw, B. 1977. Motivation in organizations: Synthesisand redirection. In B. Staw & G. Salancik (Eds.), Newdirections in organizational behavior: 55–95. Chi-cago: St. Clair.

Thomas, K. W., & Velthouse, B. A. 1990. Cognitive ele-ments of empowerment: An “interpretive” model ofintrinsic motivation. Academy of Management Re-view, 15: 666–681.

Williams, L. J., & Anderson, S. E. 1991. Job satisfactionand organizational commitment as predictors of or-ganizational citizenship behaviors. Journal of Man-agement, 17: 601–617.

Wofford, J. C., Goodwin, V. L., & Premack, S. 1992. Meta-analysis of the antecedents of personal goal level andof the antecedents and consequences of goal commit-ment. Journal of Management, 18: 595–615.

Ronald F. Piccolo ([email protected]) is an assistantprofessor in management at the University of CentralFlorida. He received a doctorate from the University ofFlorida. His research interests include personality, lead-ership, cross-cultural studies in organizational behaviorand human resources, emotion in the workplace, and riskbehavior.

Jason A. Colquitt is an associate professor in the man-agement department at the University of Florida’s War-rington College of Business. He received his Ph.D. fromMichigan State University’s Eli Broad Graduate School ofManagement. His research interests include organization-al justice, trust, team effectiveness, and personality in-fluences on task and learning performance.
