CEO TRANSFORMATIONAL LEADERSHIP: THE ROLE OF GOAL … _et_al.pdf · 2019-03-26 · CEO TRANSFORMATIONAL LEADERSHIP: THE ROLE OF GOAL IMPORTANCE CONGRUENCE IN TOP MANAGEMENT TEAMS

CEO TRANSFORMATIONAL LEADERSHIP: THE ROLE OFGOAL IMPORTANCE CONGRUENCE IN TOP

MANAGEMENT TEAMS

AMY E. COLBERTAMY L. KRISTOF-BROWN

BRET H. BRADLEYUniversity of Iowa

MURRAY R. BARRICKTexas A&M University

Using data from 94 top management teams, we found that dyadic goal importancecongruence between CEOs and vice presidents (VPs) partially mediated the relation-ship of CEO transformational leadership with individual VPs’ attitudes, but not theirperformance. However, finer-grained analyses suggested it may be higher VP percep-tions of goal importance, rather than the exact correspondence between CEO and VPgoal importance ratings, that are associated with both CEO transformational leader-ship and VP attitudes. At the organizational level, CEO transformational leadershipwas positively related to within-team goal importance congruence, which in turn waspositively related to organizational performance.

In his seminal book Leadership, Pulitzer Prize–winning author James McGregor Burns wrote, “Thefunction of leadership is to engage followers, notmerely to activate them, to commingle needs andaspirations and goals in a common enterprise, andin the process to make better citizens of both lead-ers and followers” (1978: 461). A large body oftheoretical and empirical research emanating fromBurns’s perspective has supported the relationshipof such transformational leadership with followerattitudes, motivation, and individual, group, andorganizational performance (Judge & Piccolo,2004). Recently, empirical research has begun toidentify intervening mechanisms that help explainthe link between transformational leadership andthese positive outcomes, especially at the individ-ual level. For example, followers of transforma-tional leaders have been found to have higher lev-els of trust in their leaders (Podsakoff, MacKenzie,Moorman, & Fetter, 1990) as well as higher levels ofself-efficacy (Kirkpatrick & Locke, 1996). Yet, de-spite Burns’s emphasis on the “commingling” ofaspirations and goals, there has been surprisinglylittle research on the relationship between transfor-mational leadership and shared perspectives aboutorganizational goals.

One study, however, provides preliminary evi-dence of a relationship between transformationalleadership and shared perceptions about organiza-tional goals. Berson and Avolio (2004) found thattransformational leaders were rated by followers as

more effective at communicating organizationalgoals. Their exploratory analysis of 11 leader-fol-lower dyads also showed that leaders’ and follow-ers’ qualitative reports of organizational goalstended to be more similar when a leader was ratedas more transformational. Thus, there is some evi-dence that transformational leadership and sharedperceptions about organizational goals are posi-tively related.

The purpose of the current research was to betterunderstand this relationship between transforma-tional leadership and shared perceptions about or-ganizational goals, and its association with criticaloutcomes—including individual attitudes and per-formance, as well as organizational performance.By focusing on transformational leadership withintop management teams (TMTs), we were able toexamine CEO transformational leadership, the atti-tudes and performance of individual members ofTMTs (typically vice presidents [VPs]), and organ-izational performance in the upper echelons of amultiorganization sample.

Thus, this study contributes to existing researchin three key ways. First, we examine dyadic goalimportance congruence between CEOs and VPs asan intervening mechanism by which CEO transfor-mational leadership is related to VP attitudes andperformance. We define dyadic goal importancecongruence as the similarity between CEO and in-dividual VP perceptions about the importance ofspecific goals to the organization. Unlike Berson

� Academy of Management Journal2008, Vol. 51, No. 1, 81–96.

81

Copyright of the Academy of Management, all rights reserved. Contents may not be copied, emailed, posted to a listserv, or otherwise transmitted without the copyright holder’s expresswritten permission. Users may print, download or email articles for individual use only.

and Avolio (2004), who collected qualitative ac-counts of the goals of a small sample of leaders andfollowers, we quantitatively assessed dyadic goalimportance congruence in multiple organizations.Specifically, we calculated dyadic goal importancecongruence by comparing CEO and VP ratings ofthe importance of seven specific organizationalgoals, identified from a presurvey of CEOs as rele-vant for their industry. This approach is consistentwith previous research on goal congruence (e.g.,Vancouver & Schmitt, 1991; Witt, 1998), which hasbeen described as an important type of person-organization fit (Kristof-Brown, Zimmerman, &Johnson, 2005). Thus, a second contribution of ourresearch comes from investigating transformationalleadership as a possible antecedent to goal congru-ence. As Kristof-Brown et al. (2005) noted, little isknown about antecedents of person-organization fitin general. Our study is an attempt to fill this gap.

Finally, we extend previous research on transfor-mational leadership to the organizational level ofanalysis. Although research has begun to examine atransformational leadership–organizational perfor-mance link (e.g., Waldman, Ramirez, House, &Puranam, 2001), little is known about the interven-ing mechanisms that help explain this relationship.To fill this gap, research must move beyond thedyadic leader-follower relationship to examinehow transformational CEOs relate to their top man-agement teams as collectives (Dess, 1987; Katzen-bach, 1997). As such, in this study, we examinewithin-team goal importance congruence as a pos-sible link between CEO transformational leader-ship and firm performance. We define within-teamgoal importance congruence as the similarity ofperceptions about the importance of specific organ-izational goals among all members of a TMT (in-cluding the CEO). A combination of detailed surveydata from upper echelon leaders and objective fi-nancial indicators of performance allowed us tobetter understand the relationship between trans-formational CEOs and firm performance.

TRANSFORMATIONAL LEADERSHIP ANDINDIVIDUAL OUTCOMES

A key role of all business leaders is definingstrategic goals for their organizations and aligningthe efforts of all organization members with thesegoals (e.g., Messick, 2005). Strategic leadership the-ory suggests that the values, experiences, andknowledge of leaders in the upper echelons of or-ganizations impact the strategic decisions made bythese leaders, ultimately influencing organizationalperformance (Finkelstein & Hambrick, 1996). Re-cent reviews of the strategic leadership literature

suggest that theories of transformational leadershipshould be integrated with strategic leadership per-spectives to increase understanding of the pro-cesses by which leaders communicate and imple-ment organizational strategies (Boal & Hooijberg,2001; Cannella & Monroe, 1997).

According to Bass (1985), transformational lead-ership comprises four primary behaviors. First,transformational leaders exhibit idealized influ-ence, behaving consistently with their promisesand gaining the trust of others. Second, transforma-tional leaders communicate compelling visions ofthe future and emphasize to others how their workcontributes to the achievement of the vision, a be-havior referred to as inspirational motivation.Third, through the behavior of intellectual stimula-tion, transformational leaders provide a safe envi-ronment in which others can think creatively andchallenge the status quo. Finally, transformationalleaders exhibit the behavior of individualized con-sideration by recognizing the developmental needsof others and providing support to their followers.

We propose that CEO transformational leader-ship is associated with higher levels of dyadic goalimportance congruence between CEOs and individ-ual VPs. Several of the behaviors characteristic oftransformational leaders support this association.First, as noted previously, a transformational leadercommunicates a vision of the future of an organi-zation that may serve as the beginning of the goal-setting process, with the vision describing “a clearstrategic direction in overarching terms for the or-ganization” (Conger & Kanungo, 1998: 158). Thus,the transformational leader has a clear understand-ing of what goals are important to the organization,which he or she communicates through a vision forthe organization’s future.

Second, in communicating organizations’ vi-sions, transformational leaders use rhetorical strat-egies and modeling to ensure that followers attendto, understand, and remember the strategic direc-tions embodied in the visions (Emrich, Brower,Feldman, & Garland, 2001). For example, transfor-mational leaders’ use of image-based rhetoric maybe associated with higher levels of dyadic goal im-portance congruence because the images capturefollowers’ attention and increase their comprehen-sion of a leader’s message (Emrich et al., 2001).Additionally, transformational leaders model be-haviors that are consistent with organizationalgoals in their day-to-day activities (Bass, 1985;Shamir, House, & Arthur, 1993). Through the be-havior of idealized influence, transformationalleaders serve as influential role models, makingdecisions and exhibiting behaviors that supporttheir words (Bass, 1985). Thus, goals communi-

82 FebruaryAcademy of Management Journal

cated and modeled by transformational leadersshould be attended to, understood, and remem-bered more fully. Therefore, we hypothesize:

Hypothesis 1. CEO transformational leader-ship is positively related to dyadic goal impor-tance congruence.

One reason that dyadic goal importance congru-ence is important is its relationship with followerattitudes, such as job satisfaction, organizationalcommitment, and satisfaction with one’s leader.Extensive research has supported the positive rela-tionship between demographic similarity withinleader-follower dyads and follower attitudes (e.g.,Tsui, Xin, & Egan, 1996); however, Harrison, Price,and Bell (1998) emphasized the importance of“deep-level” similarity, including similarity in at-titudes, beliefs, and values. Although judgmentsabout similarity on surface-level demographic vari-ables may be made more quickly, similarity ondeep-level variables is argued to have a strongerimpact over time. Similarly, in the TMT literature,Priem, Lyon, and Dess (1999) questioned the use ofdemographic variables as proxies for underlyingattitudes and beliefs. Thus, research on deep-levelsimilarity within leader-follower dyads is neededto supplement the current literature on demo-graphic similarity. Because of the central impor-tance of goals in organizations, and especiallyamong members of TMTs, we suggest dyadic goalimportance congruence as a key example of deep-level similarity between CEOs and their VPs.

Dyadic goal importance congruence may be asso-ciated with follower attitudes for at least two rea-sons. First, when leaders and followers have simi-lar perceptions about the importance of specificorganizational goals, the followers are likely to re-ceive higher levels of positive reinforcement fortheir work (Boswell, 2000), because their efforts aredirected toward organizationally endorsed goals,which may, in turn, be positively associated withtheir job satisfaction. Additionally, leaders mayhave higher-quality relationships with followerswho share their perceptions about the importanceof specific organizational goals (Vancouver &Schmitt, 1991). Research on leader-member ex-change has shown that high-quality exchange rela-tionships are positively associated with followerjob satisfaction, organizational commitment, andsatisfaction with one’s leader (Gerstner & Day,1997).

Past empirical research has supported the posi-tive relationship of dyadic goal importance congru-ence with follower attitudes (Vancouver & Schmitt,1991; Witt, 1998). For example, in their study ofschool principals and teachers, Vancouver and

Schmitt found that principal-teacher goal impor-tance congruence was positively related to teach-ers’ job satisfaction and organizational commit-ment and negatively related to their intent to quittheir jobs. Similarly, in their meta-analysis of theperson-environment fit literature, Kristof-Brown etal. (2005) found a corrected correlation betweengoal congruence and job satisfaction of .31 (k � 3).None of these studies investigated leadership styleas an antecedent to goal congruence, however, orpredicted follower performance as a possibleoutcome.

Given past research findings and the importanceof goals within TMTs, we hypothesize:

Hypothesis 2. Dyadic goal importance congru-ence is positively related to individual VP atti-tudes (job satisfaction, organizational commit-ment, and satisfaction with leader).

Finally, we expect dyadic goal importance con-gruence to be positively associated with individualVPs’ job performance. Goals are an important mo-tivational construct that help employees choose theactivities on which they should expend effort(Locke & Latham, 2002). As Jauch, Osborn, andTerpening noted, dyadic goal importance congru-ence improves the likelihood that employees “willdirect their efforts toward those goals most highlyprized by top management” (1980: 544). A sharedunderstanding of the importance of specific goalsby CEOs and their VPs reduces ambiguity abouteffort allocation and helps ensure that VPs’ activi-ties in their own divisions directly contribute totheir organizations’ overarching goals. These ef-forts, in turn, should be viewed positively by theCEOs and be associated with higher performanceratings for the VPs. Therefore, we hypothesize:

Hypothesis 3. Dyadic goal importance congru-ence is positively related to CEOs’ ratings ofindividual VP performance.

As noted earlier, past research has demonstratedthe relationship of transformational leadershipwith follower attitudes and performance (Judge &Piccolo, 2004). Given these findings and the pre-ceding hypotheses, we propose that dyadic goalimportance congruence provides one explanationfor the relationship between CEO transformationalleadership and VP attitudes and performance.However, other processes have also been shownto mediate the relationship of transformationalleadership with follower attitudes and performance(e.g., trust in a leader [Podsakoff et al., 1990];follower self-efficacy [Kirkpatrick & Locke, 1996]).Thus, we propose two partial mediationhypotheses:

2008 83Colbert, Kristof-Brown, Bradley, and Barrick

Hypothesis 4. The relationship between CEOtransformational leadership and VP attitudesis partially mediated by dyadic goal impor-tance congruence.

Hypothesis 5. The relationship between CEOtransformational leadership and VP perfor-mance is partially mediated by dyadic goalimportance congruence.

TRANSFORMATIONAL LEADERSHIP ANDORGANIZATIONAL PERFORMANCE

We now move from a focus on leader-followerdyads and individual attitudes and performance toa focus on the relationship of CEO transformationalleadership with a TMT as a whole and organiza-tional performance. Theory suggests that within-team goal importance congruence, or the similarityof perceptions about the importance of specificgoals to their organization among all members of aTMT (including the CEO), may help explain therelationship between CEO transformational leader-ship and organizational performance. As we notedabove, transformational leaders communicate aconsistent sense of their organizations’ strategicgoals through both words and actions (Bass, 1985;Shamir et al., 1993). As a result, followers of trans-formational leaders should have a common under-standing of the importance of specific organization-al goals and thus have higher levels of within-teamgoal importance congruence. Additionally, trans-formational leadership theory suggests that trans-formational leaders emphasize collective, ratherthan individual interests (Shamir et al., 1993;Waldman & Yammarino, 1999). Because of this col-lective focus, followers of transformational leadersare likely to view the goals of their organizations astheir own. This heightened salience of organization-al goals within a highly cohesive group is likely tobe related to higher levels of within-team goal im-portance congruence. On the basis of these trans-formational leadership behaviors, we hypothesize:

Hypothesis 6. CEO transformational leader-ship is positively related to within-team goalimportance congruence.

Furthermore, we propose that within-team goalimportance congruence is positively related to or-ganizational performance. As Dess noted, “a highlevel of consensus in strategy-making is consideredto be critical in promoting a unified direction forthe firm and enhancing the successful implemen-tation of a given strategy” (1987: 260). The strategicmanagement literature has long highlighted the im-

portance of agreement about an organization’s goalsor objectives (i.e., ends) and the methods by whichthese objectives are reached (i.e., means) as a part ofstrategy formulation (e.g., Ansoff, 1965; Bower &Doz, 1979). A shared understanding of the impor-tance of specific organizational goals among TMTmembers is a likely first step toward successfulorganizational performance on those goals.

Empirical research has generally shown thatwithin-team goal importance congruence is posi-tively related to organizational performance. Forexample, in a study of 12 nondiversified publiccorporations (Bourgeois, 1980), the similarity ofTMT member perceptions about the importance ofspecific corporate objectives and the means to ac-complish them was positively related to a compos-ite of objective economic performance indicators.In a similar study, Dess (1987) focused on the rela-tionship between agreement about ends and meansand organizational performance within a single in-dustry. The similarity of TMT member perceptionsabout the importance of either specific overall or-ganizational objectives or the methods for achiev-ing those objectives was found to be positivelyrelated to organizational performance. Thus, wedraw from the strategic management literature tohypothesize:

Hypothesis 7. Within-team goal importancecongruence is positively related to organiza-tional performance.

As noted earlier, past research has supported therelationship of transformational leadership with or-ganizational performance (Judge & Piccolo, 2004;Waldman et al., 2001). Given these findings and thepreceding hypotheses, we propose that one reasonthat transformational leadership is positively asso-ciated with organizational performance is its rela-tionship with within-team goal importance congru-ence. However, because other mechanisms mayalso mediate the relationship of transformationalleadership with organizational performance, wepropose a partial mediation hypothesis:

Hypothesis 8. The relationship between CEOtransformational leadership and organization-al performance is partially mediated by within-team goal importance congruence.

METHODS

Participants and Procedures

We surveyed members of the top managementteams of 96 credit unions across the United States,including each credit union’s CEO and other mem-


bers identified by the CEO. Because only TMTswith at least 3 team members responding were in-cluded in our sample, data from 2 credit unionswere dropped from the study. For the remaining 94credit unions, TMT size ranged from 4 to 14 mem-bers and averaged 6.4 members. Of the 601 totalteam members surveyed, 517 provided usable re-sponses, yielding a response rate of 86 percent. Allof this information was included in our organiza-tional level analyses (Hypotheses 6–8). Ratings ofindividual VP performance, provided by the CEOs,were available for only 323 VPs. Hence, completedata were available for 323 CEO-VP dyads and 94credit unions. Of the 517 team members respond-ing, 54 percent were male and 93 percent wereCaucasian. Seventy percent of the sample heldbachelor’s or graduate degrees. The average teammember had been with the current credit union for11.5 years and had been a member of the TMT for7.1 years.

Survey data were collected at two points in time.An initial survey asked the CEO of each creditunion to provide a written list of up to five goalsthat the credit union held in each of ten goal con-tent categories identified by Bateman, O’Neill, andKenworthy-U’Ren (2002). These categories in-cluded personal goals (not work-related, and there-fore, not included in our analyses), financial goals,customer goals, market goals, operations goals,product goals, organizational goals, people goals,competitive goals, and strategy-making goals. TheCEO was also asked to provide a list of all TMTmembers. Categorical coding of CEO responses gen-erated a list of seven broad organizational goalsendorsed in most credit unions: improving cus-tomer service, improving the credit union’s imagein the community, improving the efficiency of in-ternal operations, improving relationships withemployees, introducing new products and services,improving financial performance, and growing thenumber of members served. Approximately fourmonths later, the CEO and all other TMT memberscompleted an electronic survey. All TMT membersprovided ratings of their perceptions of the impor-tance of these seven goals to their organization. VPsalso provided ratings of the CEO’s transformationalleadership behaviors and of their own overall jobsatisfaction, satisfaction with the CEO, and organi-zational commitment. In a separate survey, the CEOrated the performance of each VP. Organizationalperformance, assessed as return on assets (ROA),was collected from the National Credit Union Ad-ministration’s Web site (ncua.gov) approximatelyone year after all other data were collected.

Measures

Transformational leadership. We used the Mul-tifactor Leadership Questionnaire (MLQ Form 5X;Bass & Avolio, 1995) to measure the four compo-nents of transformational leadership: idealized in-fluence, inspirational motivation, intellectual stim-ulation, and individualized consideration. TheMLQ measures idealized influence with an eight-item scale and the remaining three behaviors withfour-item scales. Respondents indicated the fre-quency with which each item described their CEOusing a five-point Likert response scale (1 � “not atall,” 5 � “frequently, if not always”). An average of4.5 (s.d. � 1.9) TMT members rated each CEO. Inour data set, confirmatory factor analysis revealedthat a higher-order transformational leadership fac-tor explained the common variance among the fourleadership components (�50

2 � 279.31, p � .01;SRMR � .05, NNFI � .91, IFI � .93); therefore, weaggregated over dimensions (� � .94). Moreover,following past research (e.g., Bono & Judge, 2003),we averaged over raters to obtain an aggregate lead-ership rating for each leader (F95, 433 � 2.41, p �.01; ICC[1] � .24; ICC[2] � .58; rwg � .85, assuminga slight negative skew in the data).

Goal importance. We developed three-itemscales to assess the importance of each of the sevenCEO-identified credit union goals (21 items over-all). Sample items included, “Improving customerservice is an important goal to my credit union,”“Improving the efficiency of internal operations is agoal that my credit union is actively pursuing,” and“The CEO and senior management team spend a lotof time planning and setting objectives about im-proving relationships with employees.” Each itemwas rated by all members of a TMT, including theCEO (1 � “strongly disagree,” 5 � “stronglyagree”). High ratings on the goal importance scalesindicated that a respondent believed the specificgoal was an important one to which the organiza-tion was devoting time and effort. We used a nor-mative (i.e., rating), rather than an ipsative (i.e.,ranking) format for evaluating goal importance fortwo reasons: (1) qualitative results from the prelim-inary CEO survey suggested that multiple goalsmay be viewed as equally important, and (2) ipsa-tive measures provide no information regarding themagnitude of differences between goals, which iscritical in congruence research (Edwards, 1993;Nunnally, 1978). Examining the mean level andstandard deviation of goal importance ratingsacross all goals suggested that no single type of goal(e.g., financial) was consistently rated as most im-portant. Thus, although some goals may have beenpursued to help accomplish others, that did not


appear to unduly influence the magnitude of theimportance ratings. Consistently with prior re-search on goal congruence, we did not ask TMTmembers to compare the relative importance ofthese seven goals or other potential organizationalgoals, or to evaluate the consistency of the organi-zational goals with their own goal priorities (Van-couver & Schmitt, 1991), but only asked them torate the importance of each goal to the credit union.A confirmatory factor analysis of the 21 itemsshowed that seven goal factors captured the vari-ance among the items (�168

2 � 775.20, p � .01;SRMR � .04, NNFI � .89, IFI � .91).

Dyadic goal importance congruence. We calcu-lated a dyadic goal importance congruence scorefor each CEO-VP dyad using the D-statistic (Cron-bach & Gleser, 1953), computed as the square rootof the sum of the squared differences between VPgoal importance ratings and CEO goal importanceratings. The D-statistic was multiplied by �1 sothat a high score indicates high levels of similarityabout the importance of specific organizationalgoals between a CEO and an individual VP.

Within-team goal importance congruence. Fol-lowing previous empirical research on strategicconsensus (e.g., Bourgeois, 1980; Dess, 1987), weoperationalized within-team goal importance con-gruence by first calculating a standard deviationfrom the goal importance ratings provided by allTMT members (including the CEO) for each of theseven goals. The standard deviations were thensummed, and the sum was multiplied by –1. A highscore on this variable implies high goal importancecongruence among all TMT members.

VP attitudes. We included measures of three VPattitudes in our survey. A five-item response scale(1 � “strongly disagree,” 5 � “strongly agree”) wasused for all items. Overall job satisfaction was mea-sured using Brayfield and Rothe’s (1951) five-itemmeasure (e.g., “I feel fairly satisfied with mypresent job”). Each VP also completed the nine-item short form of the Organizational CommitmentQuestionnaire (OCQ; Mowday, Porter, & Steers,1979; e.g., “I am proud to tell others that I am partof this organization”). Finally, VPs completed threeitems from the Job Diagnostic Survey (JDS; Hack-man & Oldham, 1980) that assessed their satisfac-tion with their CEO (e.g., “I am satisfied with theoverall quality of supervision I receive in mywork”). A confirmatory factor analysis of the threeattitudes scales revealed that the corrected correla-tions among the three factors ranged from .61 to .70.A higher-order factor analysis showed that a gen-eral attitudes factor explained the variance amongjob satisfaction, organizational commitment, andsatisfaction with leader (�24

2 � 58.35, p � .01;

SRMR � .03, NNFI � .97, IFI � .98). The compositeattitudes scale exhibited high reliability (� � .94).

VP performance. CEOs rated the performance ofeach VP using an 11-item scale adapted from exist-ing measures by Barrick and colleagues (Barrick,Parks, & Mount, 2005; Barrick, Stewart, Neubert, &Mount, 1998) and following Campbell’s (1990) per-formance taxonomy. Items assessed four primaryaspects of performance: task performance, commu-nication, teamwork, and leadership. These itemsare included in the Appendix. A higher-order fac-tor analysis revealed that a single overall perfor-mance factor explained the variance among thefour performance factors (�40

2 � 207.80, p � .01;SRMR � .04, NNFI � .91, IFI � .94). The overallperformance scale exhibited high reliability (� � .93).

Organizational performance. In the prelimi-nary survey, the credit union CEOs identified ROAas the most commonly used measure of organiza-tional performance in the industry. Archival mea-sures of ROA for 2004 were collected from theNational Credit Union Administration one year af-ter the second survey administration.

Control variables. Because organizational sizehas been found to have a positive relationship withtransformational leadership and performance (e.g.,Waldman et al., 2001), credit union size (number ofmembers of the credit union) was included as acontrol variable. In addition, in keeping with pastresearch on TMTs and organizational performance(Simons, Pelled, & Smith, 1999), credit union ageand TMT size were also included as controlvariables.

Analysis

To test the individual-level hypotheses (Hypoth-eses 1–5), we estimated structural equation modelsusing LISREL 8 (Joreskog & Sorbom, 1996). Fortransformational leadership, VP attitudes, and VPperformance, the lower-order factors served as in-dicators of the higher-order latent variables. Be-cause there is debate in the literature regarding thecausal direction of relationships between attitudesand performance (Judge, Thoresen, Bono, & Patton,2001), we tested these outcomes in two separatemodels, to avoid the possible confound of an atti-tudes-performance relationship. Criteria articu-lated in James, Mulaik, and Brett (2006) suggestedthat support for the partial mediation hypothesesrequired that (1) CEO transformational leadershipbe related to dyadic goal importance congruence,(2) CEO transformational leadership be related toVP attitudes or performance, with dyadic goal im-portance congruence controlled for, and (3) dyadicgoal importance congruence be related to VP atti-


tudes or performance, with CEO transformationalleadership controlled.

As noted above, we used a D-statistic to assessdyadic goal importance congruence. This approachallowed us to test the mediation hypotheses usingstructural equation modeling. However, the D-sta-tistic is prone to overly restrictive constraints andmay disguise deviations from a relationship of “ex-act correspondence,” which is what is traditionallyassumed when the term “congruence” is used (Ed-wards, Cable, Williamson, Lambert, & Shipp,2006). Specifically, the statistical significance of aD-statistic may be due primarily to a strong maineffect of one of the predictors. Moreover, there maybe mean-level differences when congruence occurson a high level of the predictor (versus a low level),or the incongruence in one direction (i.e., predictorX � predictor Y) may have different effects thanincongruence in the opposite direction (X � Y)(Edwards et al., 2006). Therefore, to more preciselyexamine the exact nature of dyadic goal importancecongruence on specific goals, we used techniquesrecommended by Edwards (1993, 1995).

First, to examine dyadic goal importance congru-ence as a dependent variable, we conducted tworegression analyses for each of the seven goals (Ed-wards, 1995). We regressed CEO ratings of goalimportance on CEO transformational leadership,and VP ratings of goal importance on CEO transfor-mational leadership. Both regression lines weregraphed together and the pattern of convergence ofthe two lines was examined. Second, we used poly-nomial regression analysis and three-dimensionalsurface plots to determine the precise nature of therelationship between CEO and VP ratings of theimportance of the seven goals and VP attitudes andperformance (Edwards, 1993). Five terms wereused to represent the relationship between a CEO’sand a VP’s goal importance ratings: the CEO’s rat-ing of the importance of a goal (X), the VP’s ratingof the importance of that same goal (Y), the CEO’simportance rating squared (X2), the interaction be-tween the CEO’s and VP’s ratings (XY), and theVP’s importance rating squared (Y2). To reducemulticollinearity and facilitate interpretation of thegraphs, all predictor variables were scale-centered.A separate polynomial regression was run for eachof the seven goals and each of the two dependentvariables (attitudes and performance).

To test the organization-level hypotheses (Hy-potheses 6–8), we again used LISREL 8 (Joreskog &Sorbom, 1996), with the covariance matrix as inputand procedures for testing partial mediation sug-gested by James et al. (2006). Polynomial regressionanalyses were not necessary at the organizationallevel of analysis, because the standard deviation in

goal importance ratings, rather than a differencescore, was used to assess the overall level of within-team goal importance congruence.

The direction of the relationships in all of ourstructural equation models is based on transforma-tional leadership theory. However, as we describemore fully in our discussion, our data did not allowus to firmly establish the causal direction of therelationships we tested. Equivalent models with areverse causal direction could be posited thatwould yield equivalent fit indexes (Henley, Shook,& Peterson, 2006). Nevertheless, we tested the di-rection of relationships most consistent with trans-formational leadership theory and supported byexperimental research on transformational leader-ship (e.g., Bono & Judge, 2003).

For all analyses, we screened the data for outliersusing leverage, studentized residuals, and Cook’sD-statistic (Belsley, Kuh, & Welsch, 1980; Fox,1991). In only one of the analyses did observationsexceeded the minimum cutoff for all three criteria(Bollen & Jackman, 1990) and impact the interpre-tation of the results. In this analysis (the polyno-mial regression relating financial goal importanceratings and VP attitudes), the three outlying caseswere removed from further analysis.

RESULTS

Individual Analysis

Table 1 shows descriptive statistics for the vari-ables at the individual level of analysis.

To test the hypotheses regarding the relation-ships between CEO transformational leadership,dyadic goal importance congruence, and VP atti-tudes, we estimated a partially mediated modelwith dyadic goal importance congruence as the me-diator. Table 2 includes the path coefficients forthis model, which was a good fit to the data. Sup-porting Hypothesis 1, these results showed thatVPs who worked with transformational CEOs hadhigher levels of dyadic goal importance congruencethan VPs who worked with less transformationalCEOs. Supporting Hypothesis 2, dyadic goal impor-tance congruence was positively related to individ-ual VP attitudes. Consistently with partial media-tion (Hypothesis 4), CEO transformationalleadership was positively related to VP attitudeseven when the mediator variable was included.

For VP performance, the partially mediatedmodel also provided a good overall fit to the data;however, testing the hypotheses required examin-ing specific parameter estimates. In keeping withHypothesis 1, CEO transformational leadership waspositively related to dyadic goal importance con-


TA

BL

E1

Mea

ns,

Sta

nd

ard

Dev

iati

ons,

and

Cor

rela

tion

sof

Ind

ivid

ual

-Lev

elV

aria

bles

a

Var

iabl

eM

ean

s.d

.1

23

45

67

89

1011

1213

1415

1617

18

1.C

EO

tran

sfor

mat

ion

alle

ader

ship

3.91

.40

.94

2.C

EO

goal

imp

orta

nce

—C

ust

omer

serv

ice

4.49

.54

�.1

8**

.76

3.C

EO

goal

imp

orta

nce

—Im

pro

vin

gim

age

3.73

.88

.10

.23*

*.9

0

4.C

EO

goal

imp

orta

nce

—E

ffic

ien

top

erat

ion

s

4.17

.71

�.0

4.2

7**

.04

.90

5.C

EO

goal

imp

orta

nce

—E

mp

loye

ere

lati

ons

3.82

.78

�.0

2.4

1**

.50*

*.4

1**

.84

6.C

EO

goal

imp

orta

nce

—N

ewp

rod

uct

s4.

12.7

2.1

2*.2

6**

.42*

*.0

2.3

2**

.86

7.C

EO

goal

imp

orta

nce

—F

inan

cial

4.32

.58

�.2

2**

.08

.02

.28*

*.2

9**

�.0

6.8

5

8.C

EO

goal

imp

orta

nce

—G

row

ing

mem

bers

4.17

.66

.09

.24*

*.1

0.1

4*.1

7**

.11*

.13*

.89

9.V

Pgo

alim

por

tan

ce—

Cu

stom

erse

rvic

e4.

28.6

6.2

1**

.08

.07

.01

.07

�.0

5�

.06

.03

.77

10.

VP

goal

imp

orta

nce

—Im

pro

vin

gim

age

3.53

.88

.28*

*�

.13*

.33*

�.0

8�

.01

.05

�.0

8�

.02

.38*

*.9

0

11.

VP

goal

imp

orta

nce

—E

ffic

ien

top

erat

ion

s3.

95.6

8.1

2*.0

2.0

1.1

7**

.10

�.0

7.0

8.0

1.3

2**

.20*

*.8

5

12.

VP

goal

imp

orta

nce

—E

mp

loye

ere

lati

ons

3.62

.86

.16*

*.0

4.0

7.0

4.2

8**

.04

.07

.05

.45*

*.2

9**

.30*

*.9

1

13.

VP

goal

imp

orta

nce

—N

ewp

rod

uct

s3.

95.7

3.2

6**

.00

.14*

�.0

2.0

5.1

6**

�.1

7**

.08

.35*

*.2

3**

.23*

*.2

8**

.87

14.

VP

goal

imp

orta

nce

—F

inan

cial

4.39

.62

.03

�.0

9.0

0.0

4.0

2�

.13*

.15*

*.0

4.3

2**

.25*

*.3

6**

.14*

.18*

*.8

915

.V

Pgo

alim

por

tan

ce—

Gro

win

gm

embe

rs3.

93.9

0.2

0**

�.0

5.0

3.0

2�

.06

�.0

1�

.08

.33

.30*

*.3

7**

.12*

.14*

.30*

*.2

0**

.92

16.

Dya

dic

goal

imp

orta

nce

con

gru

ence

�2.

30.8

8.1

5**

�.0

4.0

5.0

3.1

4*�

.04

.04

.01

.32*

*.2

5**

.28*

*.3

3**

.21*

*.1

4*.3

3**

17.

VP

atti

tud

es4.

19.5

4.3

5**

�.0

2.0

6�

.02

.00

�.0

1�

.06

.09

.26*

*.2

4**

.33*

*.3

1**

.29*

*.1

9**

.22*

*.2

2**

.94

18.

VP

per

form

ance

3.62

.76

.12*

.04

�.1

0.1

0.0

5�

.08

.11

.05

.05

�.0

3.1

8**

.15*

*.0

5.0

7.0

4.0

9.3

7**

.93

an

�32

3.V

alu

esin

bold

onth

ed

iago

nal

are

alp

ha

reli

abil

ity

coef

fici

ents

.*

p�

.05

**p

�.0

1

gruence. However, contrary to Hypothesis 3, dy-adic goal importance congruence was not related toVP performance. Thus, the results were not consis-tent with partial mediation (Hypothesis 5). Rather,CEO transformational leadership was directly re-lated to both dyadic goal importance congruenceand VP performance. As we would expect fromthese parameter estimates, the fit of an alternativemodel, with no link between dyadic goal impor-tance congruence and performance, was not signif-icantly different from the fit of the partially medi-ated model in terms of the chi-square test ofdifferences (�26

2 � 94.00, p � .05; SRMR � .04;NNFI � .94; IFI � .96; ��2 � 2.00, p � .05).

To better understand the precise relationship be-tween transformational leadership and dyadic goalimportance congruence, we followed Edwards’s(1995) procedures and regressed CEO and VP rat-ings of goal importance on CEO transformational

leadership. For all goals except financial, the twolines converged at higher levels of CEO transforma-tional leadership (one standard deviation above themean). An example of this pattern of results isshown in Figure 1 for the goal of growing the num-ber of members served. The graph demonstratesthat this convergence occurred primarily becauseof the positive relationship between CEO transfor-mational leadership and VP ratings of goalimportance.

We next used polynomial regression analysis andthree-dimensional surface plots to determine theprecise relationship between CEO and VP impor-tance ratings for the seven goals and VP attitudesand performance. Table 3 presents the results of thepolynomial regression analyses, and Figure 2shows selected surface plots. For all seven goals,the amount of variance explained in VP attitudesby the combination of five goal importance terms

TABLE 2Fit Indexes and Standardized Path Coefficients for Structural Equation Models

Model �2 (df) SRMR NNFI IFI

TransformationalLeadership 3

Goal ImportanceCongruence

Goal ImportanceCongruence 3

Outcome

TransformationalLeadership 3

Outcome

VP attitudes: Partiallymediated (H4)

70.32 (18)* .03 .94 .96 .15* (H1) .19* (H2) .37*

VP performance: Partiallymediated (H5)

92.00 (25)* .04 .94 .96 .15* (H1) .08 (H3) .12*

Organizational performance:Partially mediated (H8)

44.82 (20)* .05 .86 .93 .19* (H6) .20* (H7) .09

* p � .05

FIGURE 1Relationship of CEO Transformational Leadership with CEO and VP Importance

Ratings of Growing the Number of Members Served


was statistically significant (R2 � .05–.14). For VPperformance, the combination of goal terms was onlysignificant for two of the seven goals (efficiency andrelationships with employees). Examining the sur-face plots for VP attitudes, we found that VP attitudeswere highest when VPs and CEOs both rated a goal ashighly important to the credit union for five of theseven goals (shown by the highest point in the farback corner of Figure 2a). For six of the seven goals,attitudes were lowest when the CEO rated a goal ashighly important, but the VP rated the goal as notimportant (shown in the far right corner of Figure 2a).However, when the VP rated a goal as important andthe CEO rated it as less important, attitudes were stillhigh (see the far left corner of Figure 2a). This patternof relationships suggests that “exact correspondence”between CEO and VP goals is not necessary for pos-itive VP attitudes for six of the seven goals examined.

The relationship between CEO and VP ratings offinancial goals and VP attitudes showed a differentpattern (Figure 2b). For financial goals, attitudeswere highest when either the VP or the CEO (orboth) reported financial goals as important. If bothreported that financial goals were unimportant, at-titudes were at their lowest point. Again, theseresults do not suggest that exact correspondence in

CEO and VP goals is associated with the highest VPattitudes. However, because there may be problemswith extrapolating response surface graphs topoints where few or no data occur (Atkins & Wood,2002), caution needs to be taken in interpretingpoints on the financial goals graph below the mid-point of the importance scale. Only 1.8 percent ofVP ratings and 2.8 percent of CEO ratings of finan-cial goals fell below the midpoint of the scale.

Only for the goal of efficient operations was theresupport for exact correspondence. For efficiencygoals, the interaction between CEO and VP goalimportance ratings was statistically significant, andthe graph demonstrates that any incongruence onthe importance of this goal was associated withlower attitudes (see the low points on the left andright sides of Figure 2c). The same relationship wasexhibited for efficiency goals and VP performance.

Organizational Analysis

Table 4 shows descriptive statistics for the vari-ables at the organizational level.

To examine the relationships of CEO transforma-tional leadership, within-team goal importance con-gruence, and organizational performance, we esti-mated a partially mediated model. In this model,which was an overall good fit to the data (see Table 2),credit union size was the only control variable thatwas positively related to organizational performance(� � .24). CEO transformational leadership was pos-itively related to within-team goal importance con-gruence, supporting Hypothesis 6. Further, as pre-dicted in Hypothesis 7, higher levels of within-teamgoal importance congruence were positively relatedto organizational performance, as assessed by ROA.Because CEO transformational leadership and organ-izational performance were not significantly related,even when within-team goal importance congruencewas excluded from the model, the results were incon-sistent with a partially mediated relationship as pre-dicted in Hypothesis 8. Instead, the results were con-sistent with an indirect relationship between thevariables (Mathieu & Taylor, 2006). As we expectedin view of the parameter estimates, a chi-square test ofdifference showed that the fit of a model with nodirect link between CEO transformational leadershipand organizational performance was not significantlydifferent from the fit of the partially mediated model(�2

21 � 45.46, p � .05; SRMR � .05, NNFI � .87, IFI �.93; ��2 � 0.64, p � .05).

DISCUSSION

This study adds to the growing body of researchthat examines the relationship of transformational

TABLE 3Results of Polynomial Regression Analyses for

Vice Presidents’ and CEOs’ Ratings of GoalImportancea

Variables X(b1) Y(b2) X2(b3) XY(b4) Y2

(b5) R2

VP attitudesGoal 1 �.20 .08 .04 .07 .02 .07*Goal 2 �.01 .11* �.01 .01 .05 .06*Goal 3 �.15 .09 �.04 .21* �.05 .14*Goal 4 �.08 .15* �.02 .07 .01 .11*Goal 5 �.06 .19* �.04 .08 �.03 .10*Goal 6 .18 .16 �.06 �.11 .05 .05*Goal 7 .00 .02 �.01 .04 .07* .08*

VP performanceGoal 1 �.19 �.30 .03 .13 .09 .02Goal 2 �.13* �.01 .03 .04 �.02 .01Goal 3 �.10 �.19 �.04 .30* .01 .06*Goal 4 �.06 .12 .08 .00 .01 .03*Goal 5 �.25* �.05 .03 .13 �.01 .02Goal 6 .33 �.16 �.15 .10 .05 .03Goal 7 �.06 �.12 .01 .10 .03 .02

a Values are unstandardized regression coefficients. X � CEOgoal importance rating; Y � VP goal importance rating; goal 1 �improving customer service; goal 2 � improving image in thecommunity; goal 3 � improving efficiency of internal opera-tions; goal 4 � improving employee relationships; goal 5 �introducing new products and services; goal 6 � improvingfinancial performance; goal 7 � growing number of membersserved.

*p � .05


leadership with follower attitudes, follower perfor-mance, and organizational performance, and makesa unique contribution to the literature by examin-ing goal importance congruence as one mechanism

that helps explain these relationships. Within topmanagement teams, we used survey data and ob-jective indicators of organizational performance toexplore dyadic goal importance congruence as a

FIGURE 2Surface Graphs of the Relationship of CEO and VP Goal Importance Ratings with VP Attitudes

TABLE 4Means, Standard Deviations, and Correlations of Organization-Level Variablesa

Variable Mean s.d. 1 2 3 4 5 6

1. Credit union size 50.01 49.502. Credit union age 51.63 18.34 �.21*3. TMT size 6.37 1.90 .30** �.084. CEO transformational leadership 3.88 0.45 .20 �.01 .09 .945. Within-team goal importance congruence �5.20 1.17 .08 �.13 �.15 .23*6. Credit union financial performance 0.88 0.42 .27** �.14 .10 .16 .22*

a n � 94; credit union size is in thousands of members. Bold value is an alpha reliability coefficient.* p � .05

** p � .01


mediator of the relationship of CEO transforma-tional leadership with vice presidents’ attitudesand performance, and within-team goal importancecongruence as a mediator of the CEO transforma-tional leadership and organizational performancerelationship. Results from structural equation mod-els suggested that CEO transformational leadershipwas related to both dyadic and within-team goalimportance congruence, which in turn were relatedto individual VP attitudes and organizational per-formance, respectively. However, finer-grainedanalysis at the individual level suggested a morecomplex pattern of results.

Theoretical Implications

First, our results have implications for bothtransformational leadership theory and person-environment fit research. Our structural equationmodels supported the hypothesis that transforma-tional leadership is associated with higher dyadicgoal importance congruence; however, follow-upanalyses showed a convergence of CEO and VPratings of goal importance occurred when CEOswere more transformational primarily because VPswho worked with transformational CEOs ratedgoals as more important than did VPs who did notwork with such CEOs. Further, polynomial regres-sion results showed that exact correspondence be-tween CEO and VP goal importance ratings wasassociated with the highest VP attitude levels onlyin the case of efficiency goals. For most goals, lowerVP attitudes appeared only when the VPs rated thegoals as less important than did their CEOs, but notwhen both CEOs and VPs rated goals as highlyimportant or when the CEOs rated them as lessimportant than did the VPs.

Taken together, these individual-level resultssuggest it may be higher goal importance percep-tions among VPs, rather than the exact correspon-dence between CEO and VP goal importance rat-ings, that are associated with both higher levels ofCEO transformational leadership and more positiveVP attitudes. For all the sampled CEOs, the meangoal importance level is higher than the mean goalimportance level as rated by the VPs for six of theseven specific goals (excluding financial goals).However, in transformational CEO-VP dyads, thisdifference is smaller because mean VP goal impor-tance perceptions are higher than the mean in dy-ads with less transformational CEOs. It is possiblethat transformational CEOs communicate such abroad and compelling vision for their organizationsthat VPs perceive all organizational goals as beingof high importance. Alternatively, it is possible thatin transformational CEO-VP dyads, empowered

VPs had more voice in determining the direction oftheir credit unions and, through their increasedinvolvement, perceived all of the specific goals ashighly important. Future research is needed to ex-plore these possibilities, yet it is clear that goalimportance perceptions play an important rolein the story told by our individual-level results.These results and the unique pattern of findingsfor different goals would have been masked if wehad only used the D-statistic (Cronbach & Gleser,1953) to operationalize dyadic goal importancecongruence.

In our organization-level analysis, the structuralequation modeling results showed an indirect rela-tionship between CEO transformational leadershipand organizational performance, with within-teamgoal importance congruence as an intervening vari-able. These results suggest the utility of combiningtransformational leadership and strategic leader-ship perspectives to understand the relationshipbetween leadership at the upper echelons and or-ganizational performance. Although in strategicleadership theory (Finkelstein & Hambrick, 1996)leader values, experiences, and knowledge are pro-posed as impacting strategic choices and ultimatelyorganizational performance, our results suggestedthat goal importance congruence among TMTmembers also impacts the relationship betweenCEO leadership and organizational performance.

Managerial Implications

Combining our results with conclusions fromprevious research on goal congruence, goal setting,and transformational leadership offers some com-pelling implications for leaders and organizations.First, although our findings supported a positiverelationship between dyadic goal importance con-gruence and VP attitudes, they also suggested thatattitudes were lowest when a leader rated a goal ashighly important and followers rated it as low inimportance for six of the seven goals. Thus, organ-izations should focus on reducing this type of dis-agreement. If a goal is indeed an important part ofan organization’s overall strategy, the leader mayneed to use multiple techniques to persuade fol-lowers of the importance of the goal. Given thepositive relationship between transformationalleadership and VP goal importance ratings found inour study, transformational leadership behaviors—such as communicating a compelling vision of thefuture—may prove to be effective for increasingfollower goal importance ratings. However, it isalso important to note that for efficiency goals, VPattitudes were at their highest levels only whenCEO and VP goal importance ratings showed exact


correspondence. Thus, our results suggested thatorganizations must focus on reducing all types ofdisagreement for this type of goal.

Our study also highlights an association betweenwithin-team goal importance congruence and or-ganizational performance. Despite the often criti-cized lack of teamwork at the top levels in organi-zations (Katzenbach, 1997), our results suggest anassociation between higher agreement about theimportance of specific organizational goals amongTMT members and organizational performance. Al-though other research has suggested that TMTs canfunction effectively even as loosely coupled work-ing groups (Barrick, Bradley, Kristof-Brown, & Col-bert, 2007; Katzenbach, 1997), our current resultsunderscore the importance of shared objectives. Or-ganizations are encouraged to devote time and re-sources to actively promoting a shared understand-ing of the importance of specific organizationalgoals within their TMTs. Although our study sug-gests that CEO leadership behaviors may helpachieve this objective, within-team goal impor-tance congruence may ultimately be influenced byall team members.

Strengths and Limitations

As do all studies, this one has strengths andlimitations. First, because the majority of the vari-ables were collected simultaneously, we were notable to unambiguously specify the causal directionof the relationships in our models. The positedcausal direction from transformational leadershipthrough goal importance congruence to individualand organizational outcomes is based on transfor-mational leadership theory and past experimentalresearch (e.g., Kirkpatrick & Locke, 1996); however,equivalent models with reverse causal directionsare also plausible (Agle, Nagarajan, Sonnenfeld, &Srinivasan, 2006). When followers share percep-tions about the importance of specific organization-al goals with their leader, they may assess leader-ship behaviors positively, irrespective of actualleader behavior. We attempted to minimize thisbias by aggregating follower ratings of leadership.However, future research should utilize experi-mental and longitudinal designs to empirically testcausal ordering among these variables.

Second, although the collection of survey datafrom all TMT members is a strength of the study,the use of survey measures does present some lim-itations. For example, the goal importance scaleasked TMT members to assess the importance ofseven goals to their organization. These goals werebased on qualitative data collected from CEOs;however, this list may have omitted some goals that

specific credit unions were pursuing. Additionally,CEOs provided ratings of VP performance, whichcould have been influenced by a “similar-to-me” bias.

Third, although our model includes goal impor-tance congruence as a mediator of the relationshipbetween transformational leadership and out-comes, we did not measure additional mecha-nisms, such as social identification, that wouldhelp explain why transformational leadership isrelated to goal importance congruence. Future re-search that explicitly measures these mechanismswill further understanding of the relationship oftransformational leadership with dyadic and with-in-team goal importance congruence. Such re-search should also address the possibility that goalimportance congruence is influenced not only byleader behaviors, but also by follower proactivity ininfluencing goal direction and seeking information.

Future Research

Taken as a whole, these results provide compel-ling evidence of important links between leader-ship, goal importance congruence, and outcomes.However, this is only a first step. David McClellandwrote, “Whatever the source of a leader’s ideas, hecannot inspire his people unless he expresses vividgoals which in some sense they want” (1975: 260).Yet most research on goal importance congruence,including the current study, focuses on whetherpeople have similar perceptions about the impor-tance of specific goals to an organization, ratherthan whether these goals are congruent with theirpersonal goals and values. It is likely that organi-zational goals have an even stronger associationwith individual and organizational outcomes whenthey are not only commonly held, but also areviewed as motivating and are influences on morespecific individual goals. The ability of TMT mem-bers to implement an organization’s goals, the effortexpended to do so, and the appropriateness of thesegoals to the organization’s environment also likelyinfluence performance. Thus, future research ef-forts incorporating ability, goal striving (Locke &Latham, 2002), and goal appropriateness would beimportant extensions of this research. Addition-ally, given the differences in the patterns of resultsamong the different goal types, future researchshould continue to explore how and why differentgoal types differentially impact important individ-ual and organizational outcomes.

REFERENCES

Agle, B. R., Nagarajan, N. J., Sonnenfeld, J. A., & Sriniva-san, D. 2006. Does CEO charisma matter? An empir-


ical analysis of the relationships among organization-al performance, environmental uncertainty, and topmanagement team perceptions of CEO charisma.Academy of Management Journal, 49: 161–174.

Ansoff, H. I. 1965. Corporate strategy: An analyticalapproach to business policy for growth and expan-sion. New York: McGraw-Hill.

Atkins, P. W. B., & Wood, R. E. 2002. Self- versus others’ratings as predictors of assessment center ratings:Validation evidence for 360–degree feedback pro-grams. Personnel Psychology, 55: 871–904.

Barrick, M. R., Bradley, B. H., Kristof-Brown, A. L., &Colbert, A. E. 2007. The moderating role of top man-agement team interdependence: Implications for realteams and working groups. Academy of Manage-ment Journal, 50: 544–577.

Barrick, M. R., Parks, L., & Mount, M. K. 2005. Self-monitoring as a moderator of the relationships be-tween personality traits and performance. PersonnelPsychology, 58: 745–768.

Barrick, M. R., Stewart, G. L., Neubert, M., & Mount,M. K. 1998. Relating member ability and personalityto work team processes and team effectiveness. Jour-nal of Applied Psychology, 83: 377–391.

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

Bass, B. M., & Avolio, B. J. 1995. Multifactor leadershipquestionnaire (2nd ed.). Redwood City, CA: MindGarden.

Bateman, T. S., O’Neill, H., & Kenworthy-U’Ren, A. 2002.A hierarchical taxonomy of top managers’ goals.Journal of Applied Psychology, 87: 1134–1148.

Belsley, D. A., Kuh, E., & Welsch, R. E. 1980. Regressiondiagnostics: Identifying influential data andsources of collinearity. New York: Wiley.

Berson, Y., & Avolio, B. J. 2004. Transformational lead-ership and the dissemination of organizational goals:A case study of a telecommunication firm. Leader-ship Quarterly, 15: 625–646.

Boal, K. B., & Hooijberg, R. 2001. Strategic leadershipresearch: Moving on. Leadership Quarterly, 11:515–549.

Bollen, K. A., & Jackman, R. W. 1990. Regression diag-nostics: An expository treatment of outliers and in-fluential cases. In J. Fox & J. S. Long (Eds.), Modernmethods of data analysis: 257–291. Newbury Park,CA: Sage.

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.

Boswell, W. R. 2000. Employee alignment and the role of“line of sight.” Human Resource Planning, 23(4):48–49.

Bourgeois, L. J. 1980. Performance and consensus. Stra-tegic Management Journal, 1: 227–248.

Bower, J., & Doz, Y. 1979. Strategy formulation: A socialand political process. In D. Schendel & C. Hofer(Eds.), Strategic management: 152–166. Boston:Little, Brown.

Brayfield, A. H., & Rothe, H. F. 1951. An index of jobsatisfaction. Journal of Applied Psychology, 35:307–311.

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

Campbell, J. P. 1990. Modeling the performance predic-tion problem in industrial and organizational psy-chology. In M. D. Dunnette & L. Hough (Eds.), Hand-book of industrial and organizational psychology,vol. 2: 687–732. Palo Alto, CA: Consulting Psychol-ogists Press.

Cannella, A. A., Jr., & Monroe, M. J. 1997. Contrastingperspectives on strategic leaders: Toward a morerealistic view of top managers. Journal of Manage-ment, 23: 213–237.

Conger, J. A., & Kanungo, R. N. 1998. Charismatic lead-ership in organizations. Thousand Oaks, CA: Sage.

Cronbach, L. J., & Gleser, G. C. 1953. Assessing similaritybetween profiles. Psychological Bulletin, 50: 456–473.

Dess, G. G. 1987. Consensus on strategy formulation andorganizational performance: Competitors in a frag-mented industry. Strategic Management Journal, 8:259–277.

Edwards, J. R. 1993. Problems with the use of profilesimilarity indices in the study of congruence in or-ganizational research. Personnel Psychology, 46:641–665.

Edwards, J. R. 1995. Alternatives to difference scores asdependent variables in the study of congruence inorganizational research. Organizational Behaviorand Human Decision Processes, 64: 307–324.

Edwards, J. R., Cable, D. M., Williamson, I. O., Lambert,L. S., & Shipp, A. J. 2006. The phenomenology of fit:Linking the person and environment to the subjec-tive experience of person-environment fit. Journalof Applied Psychology, 91: 802–827.

Emrich, C. G., Brower, H. H., Feldman, J. M., & Garland,H. 2001. Images in words: Presidential rhetoric, cha-risma, and greatness. Administrative Science Quar-terly, 46: 527–557.

Finkelstein, S., & Hambrick, D. C. 1996. Strategic lead-ership: Top executives and their effects on organ-izations. St. Paul: West.

Fox, J. 1991. Regression diagnostics: An introduction.Newbury Park, CA: Sage.

Gerstner, C. R., & Day, D. V. 1997. Meta-analytic reviewof leader-member exchange theory: Correlates andconstruct issues. Journal of Applied Psychology, 82:827–844.


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

Harrison, D. A., Price, K. H., & Bell, M. P. 1998. Beyondrelational demography: Time and the effects of sur-face- and deep-level diversity on work group cohe-sion. Academy of Management Journal, 41: 96–107.

Henley, A. B., Shook, C. L., & Peterson, M. 2006. Thepresence of equivalent models in strategic manage-ment research using structural equation modeling.Organizational Research Methods, 9: 516–535.

James, L. R., Mulaik, S. A., & Brett, J. M. 2006. A tale oftwo methods. Organizational Research Methods,9: 233–245.

Jauch, L. R., Osborn, R. N., & Terpening, W. D. 1980. Goalcongruence and employee orientations: The substi-tution effect. Academy of Management Journal, 23:544–550.

Joreskog, K. G., & Sorbom, D. 1996. LISREL 8: User’sreference guide. Chicago: Scientific Software Inter-national.

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.

Judge, T. A., Thoresen, C. J., Bono, J. E., & Patton, G. K.2001. The job satisfaction–job performance relation-ship: A qualitative and quantitative review. Psycho-logical Bulletin, 127: 376–407.

Katzenbach, J. R. 1997. Teams at the top: Unleashingthe potential of both teams and individual leaders.Boston: Harvard Business School Press.

Kirkpatrick, S. A., & Locke, E. A. 1996. Direct and indi-rect effects of three core charismatic leadership com-ponents on performance and attitudes. Journal ofApplied Psychology, 81: 36–51.

Kristof-Brown, A. L., Zimmerman, R. D., & Johnson, E. C.2005. Consequences of individuals’ fit at work: Ameta-analysis of person-job, person-organization,person-group, and person-supervisor fit. PersonnelPsychology, 58: 281–342.

Locke, E. A., & Latham, G. P. 2002. Building a practicallyuseful theory of goal setting and task motivation: A35-year odyssey. American Psychologist, 57: 705–717.

Mathieu, J. E., & Taylor, S. R. 2006. Clarifying conditionsand decision points for mediational type inferencesin organizational behavior. Journal of Organiza-tional Behavior, 27: 1031–1056.

McClelland, D. C. 1975. The achievement motive. NewYork: Irvington Publishers.

Messick, D. M. 2005. On the psychological exchangebetween leaders and followers. In D. M. Messick &R. M. Kramer (Eds.), The psychology of leadership:New perspectives and research: 81–96. Mahwah,NJ: Erlbaum.

Mowday, R., Porter, L., & Steers, R. 1979. The measure-ment of organizational commitment. Journal of Vo-cational Behavior, 14: 224–247.

Nunnally, J. C. 1978. Psychometric theory. New York:McGraw-Hill.

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.

Priem, R. L., Lyon, D. W., & Dess, G. G. 1999. Inherentlimitations of demographic proxies in top manage-ment team heterogeneity research. Journal of Man-agement, 25: 935–953.

Shamir, B., House, R. J., & Arthur, M. B. 1993. The mo-tivational effects of charismatic leadership: A self-concept based theory. Organization Science, 4:577–594.

Simons, T., Pelled, L. H., & Smith, K. A. 1999. Makinguse of difference: Diversity, debate, and decisioncomprehensiveness in top management teams.Academy of Management Journal, 42: 662–673.

Tsui, A. S., Xin, K. R., & Egan, T. D. 1996. Relationaldemography: The missing link in vertical dyad link-age. In S. E. Jackson & M. N. Ruderman (Eds.), Di-versity in work teams: 97–129. Washington, D.C.:American Psychological Association.

Vancouver, J. B., & Schmitt, N. W. 1991. An exploratoryexamination of person-organization fit: Organiza-tional goal congruence. Personnel Psychology, 44:333–352.

Waldman, D. A., Ramirez, G. G., House, R. J., & Puranam,P. 2001. Does leadership matter? CEO leadershipattributes and profitability under conditions of per-ceived environmental uncertainty. Academy ofManagement Journal, 44: 134–143.

Waldman, D. A., & Yammarino, F. J. 1999. CEO charis-matic leadership: Levels-of-management and levels-of-analysis effects. Academy of Management Re-view, 24: 266–285.

Witt, L. A. 1998. Enhancing goal congruence: A solutionof organizational politics. Journal of Applied Psy-chology, 83: 666–674.

APPENDIX

Performance Items

Task Performance

1. Job knowledge: Understands work responsibilities,scope of job tasks, and routines to be performed.

2. Quality of work: Completes work thoroughly, accu-rately, and according to specifications.

3. Adherence to rules: Acts with integrity; avoids law orrules infractions, excessive absenteeism, or other be-haviors that may have a negative impact on the or-ganization or other employees.


Communication

4. Written communication: Clearly and appropriatelycommunicates information in writing.

5. Oral communication: Clearly and appropriately com-municates information orally.

Teamwork

6. Teamwork: Contributes to the top management teamby supporting other team members, resolving con-flict between members, and contributing to generalteam functioning.

7. Helping others: Supports peers and performs coop-erative, considerate, and helpful acts that assist co-workers’ performance.

Leadership

8. Management and administration: Forms goals, allo-cates resources to meet them, and monitors progresstoward them.

9. Supervision and leadership: Influences the perfor-mance of others in achieving the goals of the organ-ization. Includes communicating goals to others,modeling appropriate behaviors, coaching others tohelp them attain goals, and providing reinforcementupon the attainment of goals.

10. Adapting to change: Overcomes natural resistance toorganizational change; strives to behave in ways thatare consistent with the change goals and corporatestrategy.

11. Managing change: Effectively manages the transitionperiod while organizational changes are being imple-mented. This involves dealing with the rate at whichchange is introduced and the processes used to in-troduce change.

Amy E. Colbert ([email protected]) is an assistantprofessor in the Department of Management and Organi-zations at the University of Iowa. She received her Ph.D.in organizational behavior and human resource manage-ment from the University of Iowa. Her research interestsinclude leadership, individual differences, and the con-nections that employees form with their work, their co-workers, and their organizations.

Amy L. Kristof-Brown ([email protected])is the Weissman/Sinicropi Research Fellow at the HenryB. Tippie College of Business, University of Iowa. Sheearned her Ph.D. in organizational behavior and humanresource management from the University of Maryland.Her research interests include assessing the factors thatcontribute to individuals’ fit in jobs, teams, organiza-tions, and expatriate assignments.

Bret H. Bradley ([email protected]) is a doctoralcandidate in organizational behavior at the Henry B.Tippie College of Business, University of Iowa. He holdsa masters degree in accounting from Brigham YoungUniversity. His research interests include team composi-tion and dynamics, ethics and misbehavior, and the pos-itive and negative effects of leadership.

Murray R. Barrick ([email protected]) is thehead of the Management Department and the Paul M. andRosalie Robertson Chair in Business at the Mays BusinessSchool, Texas A&M University. He earned his Ph.D. inindustrial/organizational psychology from the Universityof Akron. His research interests include assessing theimpact individual differences in behavior and personal-ity have on job and team performance, and methods ofmeasuring and predicting such differences.


Documents

CEO TRANSFORMATIONAL LEADERSHIP: THE ROLE OF GOAL … _et_al.pdf · 2019-03-26 · CEO TRANSFORMATIONAL LEADERSHIP: THE ROLE OF GOAL IMPORTANCE CONGRUENCE IN TOP MANAGEMENT TEAMS