Transformational and social processes of leadership as predictors of organisational outcomes

Leadership & Organization Development JournalEmerald Article: Transformational and social processes of leadership as predictors of organisational outcomesMichael K. Muchiri, Ray W. Cooksey, Fred O. Walumbwa

Article information:

To cite this document: Michael K. Muchiri, Ray W. Cooksey, Fred O. Walumbwa, (2012),"Transformational and social processes of leadership as predictors of organisational outcomes", Leadership & Organization Development Journal, Vol. 33 Iss: 7 pp. 662 - 683

Permanent link to this document: http://dx.doi.org/10.1108/01437731211265241

Downloaded on: 24-09-2012

References: This document contains references to 58 other documents

To copy this document: [email protected]

Access to this document was granted through an Emerald subscription provided by Emerald Author Access

For Authors: If you would like to write for this, or any other Emerald publication, then please use our Emerald for Authors service. Information about how to choose which publication to write for and submission guidelines are available for all. Please visit www.emeraldinsight.com/authors for more information.

About Emerald www.emeraldinsight.comWith over forty years' experience, Emerald Group Publishing is a leading independent publisher of global research with impact in business, society, public policy and education. In total, Emerald publishes over 275 journals and more than 130 book series, as well as an extensive range of online products and services. Emerald is both COUNTER 3 and TRANSFER compliant. The organization is a partner of the Committee on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive preservation.

*Related content and download information correct at time of download.

Transformational andsocial processes of

leadership as predictors oforganisational outcomes

Michael K. MuchiriSchool of Management and Marketing, CQUniversity, Rockhampton, Australia

Ray W. CookseySchool of Business Economics and Public Policy,

University of New England, Armidale, Australia, and

Fred O. WalumbwaW.P. Carey School of Business, Arizona State University, Tempe, Arizona, USA

Abstract

Purpose – The purpose of this paper is to examine the separate and combined effects oftransformational leadership behaviour and social processes of leadership on key organisationaloutcomes within Australian local councils.Design/methodology/approach – A survey research methodology was used to gather quantitativedata from employees from nine local councils. Data were analysed using Item clustering analysis forscale construction. Hierarchical multiple regression analysis was employed to test the proposedconceptual framework.Findings – It was found that transformational leadership predicted performance outcomes,collective efficacy/outcomes expectancies and organisational commitment. Social processes ofleadership predicted performance outcomes, collective efficacy/outcomes expectancies andorganisational citizenship behaviours.Practical implications – Results indicate that by practising aspects of transformational leadershipsuch as articulating clear standards and expectations for performance and showing recognitionto work unit members for specific task or goal achievements, work unit leaders may establisha foundation that later leads to higher performance outcomes. Furthermore, promoting aspects ofsocial processes of leadership such as communication, enhancing adaptability and resolvinguncertainties may lead to greater clarification and subsequent higher performance outcomes.Originality/value – The paper is one of the first to examine the separate and combined effectsof transformational leadership behaviour and social processes of leadership on key organisationaloutcomes in Australian local councils.

Keywords Australia, Local authorities, Transformational leadership, Social processes,Public sector organizations, Organizational commitment, Collective efficacy/outcomes expectancies,Item clustering

Paper type Research paper

The current issue and full text archive of this journal is available atwww.emeraldinsight.com/0143-7739.htm

Leadership & OrganizationDevelopment JournalVol. 33 No. 7, 2012pp. 662-683r Emerald Group Publishing Limited0143-7739DOI 10.1108/01437731211265241

This paper is based on the same dataset used in Muchiri, M.K. and Cooksey, R.W. (2011), “Examiningthe effects of substitutes for leadership on performance outcomes”, Leadership & OrganizationDevelopment Journal, Vol. 32 No. 8, pp. 817-836. An earlier version of this paper was presented at theAustralian and New Zealand Academy of Management (ANZAM) Annual Conference, Auckland,New Zealand, 2-5 December 2008. The title of that paper was “Leadership viewed through doublelenses: an examination of transformational leader behaviours and social processes of leadership andtheir impact on key organisational variables within Australian local councils”.

662

LODJ33,7

IntroductionThere is growing research evidence indicating that effective leadership is integral toorganisational effectiveness (Avolio et al., 2009; Lowe and Gardner, 2001; Muchiri et al.,2011; Walumbwa et al., 2011; Yukl, 2010). A review of the extant leadership literaturefurther suggests that a reliable indicator of leadership effectiveness is the performance ofthe workgroup or organisation, and therefore it is vital to explore the dynamics of theleader’s relationship with followers. However, there is limited empirical research onleadership within public-sector organisations (Andersen, 2009; Currie et al., 2009;Fernandez et al., 2010; Van Wart, 2003, 2005). Therefore, it is vital to examinerelationships among leader behaviours and followers’ intermediate attitudinal outcomesand organisational outcomes within public-sector organisations.

To enhance findings from ongoing leadership research and to advance the existingleadership theorising, researchers have proposed that the most beneficial way wouldbe to examine leadership through multiple lenses (Dansereau and Yammarino, 1998;Yammarino et al., 2001). This approach, termed integrative leadership theory testing inwhich several leadership theories are examined concurrently, is viewed as invaluablein the quest to better understand the direct and indirect contributions of specificleadership styles and behaviours to targeted organisational outcomes (Fernandez et al.,2010; Podsakoff et al., 1996).

Based on the above discussion that an integrative leadership theory testing will yieldmore accurate research findings as compared to a single-theory testing, the current studywas interested in examining two leadership perspectives. The first perspective wastransformational leadership as proposed by Bass and colleagues (Avolio and Bass, 2002,2004; Bass, 1985). As this theory was conceptualised in the USA, its efficacy to explainorganisational outcomes within Australian contexts has been debated (Parry and Sarros,1996). The second perspective, the social processes of leadership, was proposed by Parryand colleagues (Parry, 1998, 1999, 2004) and was conceptualised in an Australian andNew Zealand context.

These two perspectives were chosen to be representative of different ways ofconceptualising leadership: leadership theorised as a change-focused process andleadership theorised as a social process. While transformational leadership wastheorised in a positivistic a priori fashion, the social processes of leadership emergedfrom an interpretive grounded-theory approach. The researchers reasoned that such astudy would be fruitful considering that social processes of leadership have previouslybeen shown to impact organisational variables in a similar pattern to transformationalleadership but with differing points of focus (Parry, 2002, 2004). The investigators alsoreasoned that testing these two leadership theories side by side within an Australiancontext would unearth more findings than would a stand-alone test of a single theory.

Theoretical background and research modelA review of the extant literature uncovers research findings based on stand-aloneindividual leadership theories, and how individual leadership theories are projected toinfluence single or multiple organisational outcomes (Avolio and Bass, 2002; Lowe andGardner, 2001; Lowe et al., 1996; Parry, 1998; Walumbwa and Schaubroeck, 2009;Walumbwa et al., 2005, 2010, 2011; Yukl, 2010). For example, transformational leadershiphas been linked to employee attitudes and behaviours in a variety of settings acrosscultures ( Judge and Piccolo, 2004; Kirkman et al., 2009; Walumbwa et al., 2007), and hasbeen related to meaningful work (Arnold et al., 2007), organisational commitment and jobsatisfaction (Walumbwa et al., 2005), and follower work engagement (Zhu et al., 2009).

663

Social processesof leadership

The work by Bass and colleagues has helped clarify the transformational-transactional leadership theory (Avolio and Bass, 2002, 2004; Bass, 1985, 1997). Avolioand Bass propose that transformational leaders behave in ways which achievesuperior results by emphasising one or more of four dimensions of transformationalleadership. First, leadership is idealised when followers seek to identify with theirleaders and emulate them. Second, leadership inspires followers with challenges andpersuasion that provide meaning and understanding. Third, leadership is intellectuallystimulating, expanding the followers’ use of their abilities. Finally, leadership isindividually considerate, providing the followers with support, mentoring andcoaching. Transformational leaders motivate their associates, colleagues, followers,clients and even their bosses to go beyond their individual self-interests for the good ofthe group, organisation or society. Avolio and Bass explain that transactionalleadership occurs when the leader rewards or disciplines a follower on the basis ofadequacy of the follower’s performance. Furthermore, this leadership approachdepends on contingent reinforcement, either positive contingent reward or the morenegative or passive forms of management-by-exception (Avolio and Bass, 2002).

So far, researchers have shown that transformational leadership is an effectiveform of leadership ( Judge and Piccolo, 2004). However, to generalise thetransformational-transactional theory to different organisational and culturalcontexts, there is need for further investigations. Furthermore, to counter thecriticism that the transformational-transactional leadership model does not separateleadership from hierarchical rank and authority, researchers are turning to integrativeleadership research approaches to better understand the leadership phenomenon.These research approaches examine concurrently several leadership theories,including newer leadership theories that focus on the process of influence of self andothers, alongside the more established leadership theories (like transformationalleadership) in order to unpack the complex relationship between leadership and theattainment of organisational success.

Related to the above discussion, recent studies now conceptualise leadership as asocial process of influence (Parry, 1998, 1999, 2002; Parry and Proctor-Thomson, 2001),and a relationship between those who aspire to lead and those who choose to follow.Parry (2004, pp. 70-1) emphasises four aspects of the social process of leadership:optimising, resolving uncertainty, enhancing adaptability and promulgating a vision.Optimising refers to the effective leadership process of making the best of the situation,making the most effective or optimal use of all available and potential resources, andmoving beyond mediocrity towards excellence. Optimising is conceptually likened totransformational leadership, with leaders who work on the basis of contingent rewardand management-by-exception as survivors, whereas leaders who are transformers areclear examples of successful leaders. Parry also identifies two social processesoccurring in relation to leadership, change and following. These social processesinvolve resolving the uncertainty and enhancing the adaptability of followers and theleaders in their organisations. Where there was effective leadership, the uncertainty ofchange of both the followers and leaders could be resolved through strategies,behaviours and activities demonstrated by the leaders. In the absence of effectiveleadership, both leaders and followers became consumed with uncertainty, and theirknowledge, performance and morale deteriorated while inaccurate perceptions weremaintained. Hence, the resolution of uncertainty due to change and turbulence maydirectly reduce the degree to which the change becomes threatening to individuals inthe organisation.

664

LODJ33,7

Therefore, our study was designed to examine the above two leadershipconceptualisations, and was based on the premise that it would yield a richerunderstanding of effective organisational leadership than is the case of studying singletheories. Our study also sought to extend organisational leadership research findingswithin local councils and public-sector contexts.

Leadership research within the public-sector contextThe literature on leadership in the public sector highlights the ever-increasing economicand social pressures facing public-sector leaders as they seek to reform managerial andorganisational practices (Van Wart, 2003, 2005). In line with a framework of reinventingthe government (Osborne and Gaebler, 1992), there has been a sustained governmentalfocus on improving leadership in public-sector organisations. Indeed, policy makersin countries such as Australia, Canada, New Zealand, Sweden, the UK and the USAhave actively promoted leader behaviours which change the culture of public-sectorbureaucracies and improve the performance of public-sector organisations (Andersen,2009; Currie et al., 2009; Dunoon, 2002; Javidan and Waldman, 2003; Thach andThompson, 2006; Parry and Proctor-Thomson, 2001). Furthermore, public managers areexpected to find more efficient and effective methods of delivering services, implementingmanagerial reforms and executing strategies adopted by elected officials (Fernandez andRainey, 2006; Fernandez et al., 2010).

Ingraham and Getha-Taylor (2004) note that most public-sector organisations followthe “split leadership” model, where elected and appointed officials often lead topmanagement teams composed of career bureaucrats. As a consequence, it is possible thatcontextual factors unique to public-sector organisations may constrain the way leadershipis exercised within public organisations. Javidan and Waldman (2003) suggest thatpolitical changes and budget cuts may promote environmental uncertainty which mayimpact on leadership behaviours among managers in public-sector organisations.Therefore, it can be surmised that contextual and sector-based differences may impingeupon a public sector manager’s ability to exhibit charismatic and transformationalleadership behaviours as evidenced in private-sector organisations. Importantly,many extant leadership theories have been ostensibly developed from research onprivate-sector organisations. Thus, it is debatable whether they would be applicableto public-sector organisations. Nevertheless, the extant research suggests that managersin public-sector organisations do exhibit effective forms of change-oriented leadership(Bass, 1997). Furthermore, emerging research findings suggest that it would be beneficialto extend the knowledge gained from the vast research of leadership phenomena in theprivate sector to the management of public-sector organisations (Currie et al., 2009;Fernandez et al., 2010).

Relating leadership to organisational outcomesIn a bid to understand how leadership influences organisational outcomes, researchershave often studied the relationship between leadership and other intermediate attitudinalvariables such as organisational commitment (Bycio et al., 1995), collective efficacy(Walumbwa et al., 2005), organisational efficacy (Bohn, 2002) and organisationalcitizenship behaviours (OCBs) (Podsakoff et al., 1996). Recent research findings indicatethat social processes of leadership influence organisational outcomes in Australia(Muchiri et al., 2011; Parry, 1998, 2004). Other studies conducted in private, public andnon-profit organisations indicate that transformational leaders were more effective andsatisfying as leaders than were transactional leaders, although the best leaders frequentlyemployed a blend of transformational and transactional behaviours ( Judge and Piccolo,

665

Social processesof leadership

2004). Transformational leadership was also positively related to employee’s satisfaction,self-reported effort, job performance, organisational commitment and citizenshipbehaviours (Avolio and Bass, 2002).

In order to extend the literature on leadership in public-sector organisations, thecurrent study focused on local government councils in the state of New South Wales,Australia. These councils are administered through the division of local government inthe Department of Premier and Cabinet. Through the Department of Premier andCabinet, the Government of New South Wales constantly monitors key councilperformance indicators across a broad spectrum of services and activities throughcouncil financial reports, rating records, library circulation records, domestic wasteand recycling services, water supply services, sewerage services, planning anddevelopment services, environmental and health services, recreation and leisureservices, and community services (Department of Premier and Cabinet, 2010).Therefore, exploring leadership within local councils would help us understand theleadership roles played by the councils’ employees in these changing times, when manylocal councils are under pressure to increase organisational efficiency and effectiveness.In order to achieve their objectives and to offer services at a level that would assuretheir survival and competitive edge, councils have been forced to restructure, torationalise and appraise operations and policies, and to train their staff. How theyachieve their objectives may largely depend on the capability of the councils’ leadershipto motivate and energise staff when implementing potential changes, and on howwell the councils’ leadership positively addresses issues of employees’ citizenshipbehaviours, efficacy and commitment in the workplace. Therefore, as proposed in ourconceptual research model depicted in Figure 1, our goal was to investigate thepotential effects of transformational leader behaviours and social processes ofleadership, individually and collectively, on employees’ criterion variables(organisational commitment, collective efficacy beliefs and collective outcomesexpectancy, organisational efficacy, OCBs and performance outcomes). Based on ourearlier discussion, this study investigated the following propositions:

P1. Transformational leadership in the local councils would be positively relatedto organisational commitment, efficacy, OCBs, organisational efficacy andperformance outcomes.

Organisationalcommitment

Organisational citizenshipbehaviour

Collective efficacy beliefsand collective outcomes

expectancy

Organisational efficacy

General satisfaction

Extra effort

Overall effectivenesssatisfaction with unit,

supervisor and customers

Work unit morale,productivity and ability

Errors and complaints

Leadership approach Intermediate outcomes Performance outcomes

Transformational andtransactional leadership

Social processes ofleadership

Arrows indicate direct relationships between leadership and criterion variables

Arrow indicates an indirect relationship between leadership and criterion variable

Figure 1.Summary of studypredictions

666

LODJ33,7

P2. Transactional leadership in the local councils would be negatively relatedto organisational commitment, efficacy, OCBs, organisational efficacy andperformance outcomes.

P3. Social processes of leadership in the local councils would be positively related toorganisational commitment, efficacy, OCBs, organisational efficacy andperformance outcomes.

MethodSampleMeasures of leader behaviours and job attitudes for this study were collected from 177employees, representing a return rate of 59 per cent. Participants were drawn from nineAustralian local councils and represented white collar, managerial and professionalpositions. Respondents also represented multiple divisions, organisational levels andfrom councils of varying sizes. More specifically, the sample comprised a near equalgender spread (50.3 per cent males, 49.7 per cent females), mostly in non-managerialpositions (63.8 per cent non-management, 36.2 per cent management positions), welleducated (76.7 per cent post-secondary, 23.3 per cent primary/high school). Most ofthe respondents came from three divisions of Administration and Corporate Affairs(27.1 per cent), Planning and Community Services (23.2 per cent) and Engineering andWorks (23.7 per cent).

ProcedureFollowing clearance from the Human Research Ethics Committee, survey questionnaireswere administered to employees at various organisational levels in nine New South Waleslocal government councils. With the help of human resource departments of theparticipating councils, respondents filled out the questionnaires in their work settingsduring normal working hours, or respondents were allowed to take the survey home tocomplete if they so chose. The researchers had included a prepaid self-addressedenvelope, encouraging respondents to post back their surveys directly to the researchers.All participants had been assured of anonymity and their right to withdraw fromparticipation. The surveys assessed relationships between different aspects oforganisational leadership and a number of distinct organisational outcomes. Councilemployees from various departmental levels completed the cross-sectional survey, and inthe process evaluated leadership capabilities as well as helped get useful data explaininghow leaders impacted organisational performance. The study integrated instruments thatboth quantified leadership in terms of observable behaviours (Multifactor LeadershipQuestionnaire (MLQ)), as well as those defining leadership in terms of underlyingprocesses and mechanisms (social processes for leadership scale). The other partsof the survey had items on OCBs, organisational commitment, organisational efficacy,collective efficacy beliefs and collective outcome expectancy, satisfaction, productivityand overall effectiveness. The six scales used for most of the major variables, that is,transformational-transactional leadership, social processes of leadership, organisationalcommitment, citizenship behaviours, organisational efficacy, and collective efficacy beliefsand collective outcome expectancy were well-established scales whose validity andreliability had been widely reported in previous studies (Bass and Avolio, 2004;Bohn, 2002; Meyer et al., 1993; Podsakoff and MacKenzie, 1997; Parry, 2002, 2004;Riggs et al., 1994).

667

Social processesof leadership

Approach to data analysisData analyses were conducted in two major phases. First, we investigated the factorstructure and reliability of all scales using item clustering (Revelle, 1978, 1979). Then,we used hierarchical multiple regression (HMR) analyses to examine the aggregateeffects of the set of transformational leader behaviours and social processes ofleadership on the various criterion variables (Cohen et al., 2003).

Approach to exploratory and confirmatory scale constructionTo judge the dimensionality and internal homogeneity of the scales/instruments used inthis study, the ICLUST item-clustering procedure (Revelle, 1978, 1979) was implemented.Recent writings on alternative approaches to scale construction (Cooksey and Soutar,2006; Muchiri and Cooksey, 2010; Zinbarg et al., 2005) have shown the merits of usingICLUST analysis, especially for a small sample size as for this study (N¼ 177). Moreover,as the research used many scales developed in the USA but tested in Australia, ICLUSTanalysis was deemed the most appropriate procedure for scale construction. Theresearchers agreed with Cooksey and Soutar’s (2006) argument that ICLUST analyses are“relevant in cross-cultural studies or in studies that use the same measures in differentcircumstances, where contextual variations in construct measurement may requiredifferent scale compositions” (p. 81). Following on from the results of both the first-orderand second-order ICLUST analyses, the scale structures, reliabilities and validities of allinstruments used in this study were found to be comparable to those from originalresearches that had used the prevalent factor analysis. Appendices 1-4 show selectedhierarchical tree diagrams for transformational and social processes of leadership fromthe ICLUST analysis. The researchers were encouraged to emulate earlier studies that hadused second-order clusters for each scale as this would help reduce the large number ofvariables proposed.

HMR analysisHMR analysis (Cohen et al., 2003) was used to determine if the successive addition ofinformation regarding leadership styles and behaviours, social processes of leadership,organisational commitment, collective efficacy beliefs and outcome expectancy,organisational efficacy and citizenship behaviours would improve the prediction ofperformance outcomes beyond that afforded by demographic characteristics. Table Irepresents a logical development of a priori order of entry of predictor sets, where weworked from more stable individual characteristics such as demographic characteristics“outward” into more general and dynamic work context factors such as role aspects, workenvironment and culture. Working backwards, OCBs, organisational efficacy, collectiveefficacy/expectancies and organisational commitment would form bases for separateHMR analyses.

Predictor variablesTransformational and transactional leadership was measured using the 45-item MLQ(Form 5x-Short) developed by Bass and Avolio (2004). The leadership dimensionsmeasured in this survey were: transformational leadership, transactional leadership andlaissez-faire leadership, which is a non-leadership dimension. The factor structure for thisinstrument had been well established in non-Australian contexts (Avolio and Bass, 2004).Using ICLUST to evaluate the internal structure of the instrument in our Australiansample, we validated two second-order clusters, active individualised transformationalleadership and active management-by-exception with an overall goodness-of-fit of 0.96.

668

LODJ33,7

Hie

rarc

hic

alre

gre

ssio

nm

odel

pre

dic

tin

g

Ste

pIn

dep

end

ent

var

iab

lese

tP

erfo

rman

ceou

tcom

esO

rgan

isat

ion

alci

tize

nsh

ipb

ehav

iou

rsO

rgan

isat

ion

alef

fica

cyC

olle

ctiv

eef

fica

cy/

exp

ecta

nci

esA

ffec

tiv

e-n

orm

ativ

eco

mm

itm

ent

Con

tin

uan

ceco

mm

itm

ent

1D

emog

rap

hic

s|

||

||

|2

Lea

der

ship

||

||

||

3O

rgan

isat

ion

alco

mm

itm

ent

||

||

DV

/IV

DV

/IV

4C

olle

ctiv

eef

fica

cy/

exp

ecta

nci

es|

||

DV

DV

/IV

5O

rgan

isat

ion

alef

fica

cy|

|D

V6

Org

anis

atio

nal

citi

zen

ship

beh

avio

urs

|D

V

7P

erfo

rman

ceou

tcom

esD

V

Note

s:

DV

,dep

end

ent

var

iab

lefo

ra

spec

ific

anal

ysi

s;|

,IV

set

use

din

the

anal

ysi

s;D

V/I

V,t

wo

asp

ects

ofor

gan

isat

ion

alco

mm

itm

ent

(aff

ecti

ve/

nor

mat

ive

and

con

tin

uan

ce)

each

serv

edas

ad

epen

den

tv

aria

ble

wh

ile

the

oth

erse

rved

asan

ind

epen

den

tv

aria

ble

Table I.Pattern of hierarchical

regression model tests ofthe research framework

presented in Figure 1

669

Social processesof leadership

The social processes of leadership were measured using the 20-item social processes ofleadership scale (Parry, 2002, 2004). The specific social processes dimensions measuredwere optimising, resolving uncertainty and enhancing adaptability. The reliability ofthe factor structure of this scale had also been well established (Parry, 2002). However,as a check on the internal structure of the scale for our sample, ICLUST was againused. We were able to extract one second-order cluster called social processes ofleadership with an overall goodness-of-fit of 0.99.

Criterion variablesFor this study, OCBs were measured using the 24-item OCB scale (Podsakoff et al.,1993, pp. 14-15). The dimensions measured by the OCB scale included: altruism,conscientiousness, sportsmanship, courtesy and civic virtue. Using ICLUST, we extractedone second-order cluster called OCB with an overall goodness-of-fit of 0.99.

Organisational commitment in this survey was measured using the organisationalcommitment scale (Meyer et al., 1993, p. 544) and measured three dimensions ofaffective commitment, continuance commitment and normative commitment.Following the ICLUST – item clustering analysis, we extracted two second-orderclusters, called affective/normative commitment and continuance commitment with anoverall goodness-of-fit of 0.91.

Organisational efficacy was measured using Bohn’s (2002) 22-item organisationalefficacy scale. ICLUST identified one second-order cluster called organisationalefficacy with an overall goodness-of-fit of 0.99. Collective efficacy beliefs and collectiveoutcome expectancy were operationalised in accordance with work by Riggs andcolleagues. Studies by Riggs and Knight (1994) reported robust reliability and validityindices. ICLUST identified one second-order cluster called collective efficacy beliefsand outcomes expectancy with an overall goodness-of-fit of 0.94.

For this study, we developed a subjective measurement for performance outcomes.Items for this construct were developed from the existing literature which targetsfollowers’ satisfaction, productivity and overall effectiveness within a workplace (Richardet al., 2009). General satisfaction, productivity and overall effectiveness variables were allmeasured using: three items on general satisfaction; three items on morale andproductivity in the work unit, and the work unit’s ability to complete tasks on time; twoitems on number of work unit’s errors and customer complaints; and, one item askedparticipants to rate the overall effectiveness of their work units. ICLUST identified onesecond-order cluster called performance outcomes with an overall goodness-of-fit of 0.97.

ResultsLeadership explanation of performance outcomesHierarchical regression analysis was employed to test for the effects of all IV sets onpredicting performance outcomes. For purposes of this paper, we report only theportion of the results that were associated with the role of leadership within thehierarchy of predictive contributions. For all F-tests of significance, “Model II error”,where all predictor effects have been removed, was employed (see Cohen et al., 2003).The left-hand portion of Table II displays the leadership variable set when entered inthe sequential order described in Table I, the R2 change and its associated degrees offreedom, F change (partial F-test) and significant F change (p-values for the partialF-test). The right-hand portion of the table lists the significant individual IVcontributors to predictors at that step, showing the variables analysed, the part(semi-partial) correlation, partial F and significance (p-value for the partial F-test).

670

LODJ33,7

The table also displays the overall R2, the adjusted R2 after entry of all independentvariable sets and the overall model F and p-value. After the final step, with all IV sets inthe equation described in Table I, the overall regression model was significant(R2¼ 0.739; adjusted R2¼ 0.704, F(21, 155)¼ 20.93, po0.001). Thus, with all variablesets entered, nearly 74 per cent of the variance in performance outcomes wasexplained. The leadership set contributed significantly to the prediction of performanceoutcomes over and above what the demographic set could predict (R2 change¼ 0.512,F change (3, 155)¼ 101.35, po0.001). Among the leadership variables, activeindividualised transformational leadership (partial correlation¼ 0.444, partialF¼ 117.07, po0.001), active management-by-exception (partial correlation¼�0.093,partial F¼ 5.14, p¼ 0.002) and social processes of leadership (partial correlation¼ 0.277,partial F¼ 45.56, po0.001) were all significant in the explanation of the variance inperformance outcomes at this step. Therefore, those local council leaders perceived toscore higher on the measures of active individualised transformational leadership andsocial processes of leadership were associated with a significant increase in performanceoutcomes. Furthermore, those local council leaders perceived to score higher on activemanagement-by-exception were associated with a significant decrease in performanceoutcomes.

Leadership explanation of other organisational outcomesWith all variable sets entered, nearly 65 per cent of the variance in OCBs was explained(Table III). The leadership set contributed significantly to the prediction oforganisational citizenship behaviours over and above what the demographic set

Variableset

R2

change dfF

changeSignificant F

change VariablePartial

correlationPartial

F a Significance

Leadership 0.512 3 101.35 o0.001* Active individualisedtransformationalleadership 0.444 117.07 o0.001*Active managementby exception �0.093 5.14 0.002*Social processes ofleadership 0.277 45.56 o0.001*

Notes: Overall model: apartial F for F change and partial correlation calculated using Model II errorwith error df¼ 155. R2¼ 0.739; adjusted R2¼ 0.704; F (21, 155)¼ 20.93.

Table II.Effects of leadership onperformance outcomes

Variableset R2 change df F change

SignificantF change Variable

Partialcorrelation

PartialF a Significance

Leadership 0.349 3 52.00 o0.001* Socialprocessesof leadership

0.512 117.18 o0.001*

Notes: Overall model: apartial F for F change and partial correlation calculated using Model II errorwith error df¼ 156. R2¼ 0.651; adjusted R2¼ 0.606; F (20, 156)¼ 14.54. *p0.05 consideredsignificant; **0.05opp0.10 considered marginally significant

Table III.Effects of leadership

on organisationalcitizenship behaviours

671

Social processesof leadership

could predict (R2 change¼ 0.349, F change (3, 156)¼ 52.00, po0.001). Among theleadership variables, only social processes of leadership (partial correlation¼ 0.512,partial F¼ 117.18, po0.001) was significant in the explanation of the variance inOCBs at this step. Therefore, those local council leaders perceived to score higher onthe measure of social processes of leadership were associated with a significantincrease in OCBs.

After the entry of all variable sets nearly 60 per cent of the variance inorganisational efficacy was explained (Table IV). The leadership set contributedsignificantly to the prediction of organisational efficacy over and above what thedemographic set could predict (R2 change¼ 0.128, F change (3, 157)¼ 16.62, po0.001).Among the leadership variables, only social processes of leadership (partialcorrelation¼ 0.231, partial F¼ 20.79, po0.001) was significant in the explanation ofthe variance in organisational efficacy. Therefore, those local council leaders perceivedto score higher on the measures of social processes of leadership were associated with asignificant increase in organisational efficacy.

With all variable sets entered, nearly 46 per cent of the variance in collectiveefficacy/expectancies was explained (Table V). The leadership set contributedsignificantly to the prediction of collective efficacy/expectancies over and above whatthe demographic set could predict (R2 change¼ 0.309, F change (3, 158)¼ 30.14,po0.001). Among the leadership variables, active individualised transformationalleadership (partial correlation¼ 0.203, partial F¼ 12.06, po0.001) and social processesof leadership (partial correlation¼ 0.357, partial F¼ 37.29, po0.001) were allsignificant in the explanation of the variance in collective efficacy/expectancies at

Variableset

R2

change dfF

changeSignificantF change Variable

Partialcorrelation

PartialF a Significance

Leadership 0.128 3 16.62 o0.001* Socialprocessesof leadership

0.231 20.79 o0.001*

Notes: Overall model: apartial F for F change and partial correlation calculated using Model II errorwith error df¼ 157. R2¼ 0.597; adjusted R2¼ 0.548; F (19, 157)¼ 12.23. *p0.05 consideredsignificant; **0.05opp0.10 considered marginally significant

Table IV.Effects of leadership onorganisational efficacy

Variable setR2

change dfF

changeSignificantF change Variable

Partialcorrelation

PartialF a Significance

Leadership 0.309 3 30.14 o0.001* Activeindividualisedtransformationalleadership 0.203 12.06 o0.001*Social processesof leadership 0.357 37.29 o0.001*

Notes: Overall model: apartial F for F change and partial correlation calculated using Model II errorwith error df¼ 158. R2¼ 0.460; adjusted R2¼ 0.398; F (18, 158)¼ 7.47, *p0.05 consideredsignificant; **0.05opp0.10 considered marginally significant

Table V.Effects of leadershipon collective efficacy/expectancies

672

LODJ33,7

this step. Therefore, those local council leaders perceived to score higher on themeasures of active individualised transformational and social processes of leadershipwere associated with a significant increase in collective efficacy/expectancies.

After the entry of all variable sets, nearly 45 per cent of the variance in affective-normative commitment was explained (Table VI). The leadership set contributedsignificantly to the prediction of the affective-normative commitment over and abovewhat the demographic set could predict (R2 change¼ 0.112, F change (3, 159)¼ 10.75,po0.001). Among the leadership variables, only active individualised transformationalleadership (partial correlation¼ 0.221, partial F¼ 14.07, po0.001) was significant inthe explanation of the variance in the affective-normative commitment at this step.Therefore, those local council leaders perceived to score higher on the measures ofactive individualised transformational were associated with a significant increase inthe affective-normative commitment.

After the entry of all variable sets, nearly 28 per cent of the variance in continuancecommitment was explained (Table VII). The leadership set contributed significantly tothe prediction of the continuance commitment over and above what the demographicset could predict (R2 change¼ 0.059, F change (3, 159)¼ 4.18, p¼ 0.007). Among theleadership variables, only active individualised transformational leadership (partialcorrelation¼�0.144, partial F¼ 4.56, p¼ 0.004) was significant in the explanation ofthe variance in continuance commitment at this step. Therefore, those local councilleaders perceived to score higher on the measures of active individualisedtransformational leadership were associated with a significant decrease in continuancecommitment.

Variableset

R2

change dfF

changeSignificantF change Variable

Partialcorrelation

PartialF a Significance

Leadership 0.112 3 10.75 o0.001* Activeindividualisedtransformationalleadership

0.221 14.07 o0.001*

Notes: Overall model: apartial F for F change and partial correlation calculated using Model II errorwith error df¼ 159. R2¼ 0.448; adjusted R2¼ 0.388; F (17, 159)¼ 7.58, *p0.05 consideredsignificant; **0.05opp0.10 considered marginally significant

Table VI.Effects of leadership on

affective-normativecommitment

Variableset

R2

change dfF

changeSignificantF change Variable

Partialcorrelation

PartialF a Significance

Leadership 0.057 3 4.18 0.007* Activeindividualisedtransformationalleadership

�0.144 4.56 0.004*

Notes: Overall model: apartial F for F change and partial correlation calculated usingModel II error with error df¼ 159. R2¼ 0.277; adjusted R2¼ 0.200; F (17, 159)¼ 3.59, *p0.05considered significant; **0.05opp0.10 considered marginally significant

Table VII.Effects of leadership on

continuance commitment

673

Social processesof leadership

Conclusions and implicationsThis study was unique in that it was one of the first studies to examine at once the jointinfluence of two major leadership conceptualisations (i.e. transactional-transformationalleadership and social processes of leadership) on a range of organisational outcomeswithin the Australian public sector. In accordance with our three propositions, activeindividualised transformational leadership, active management-by-exception and socialprocesses of leadership in the local councils were predictive of employees’ organisationalcommitment, collective efficacy beliefs and outcomes expectancy, organisational efficacy,OCBs and performance outcomes. Further, in line with P1, P2 and P3, the impact of activeindividualised transformational leadership and social processes of leadership in the localcouncils was positive in these relationships whereas the impact of active management-by-exception was negative.

Our findings support the core thesis of transactional-transformational leadershiptheory, that active transformational (which here incorporates contingent reward)leader behaviour was positively related to performance outcomes whereas the passivedimension was negatively related to performance (Avolio and Bass, 2004). In thepresent study, active management-by-exception (part of transactional leadership) wasnegatively related to performance outcomes. Thus, by practising aspects oftransformational leadership such as articulating clear standards and expectationsfor performance and showing recognition to work unit members for specific task orgoal achievements, work unit leaders in local councils may establish a foundation thatlater leads to higher performance outcomes. The findings also extend the work byParry (2002, 2004) in that social processes of leadership were correlated with work unitoutcomes (morale, effectiveness and productivity). Thus, promoting aspects of socialprocesses of leadership like communication, enhancing adaptability and resolvinguncertainties may lead to greater clarification and subsequent higher performanceoutcomes in Australian local councils.

For this sample from public-sector organisations, only social processes of leadership(and not active individualised transformational leadership as proposed in extant USresearch) were predictive of citizenship behaviours and organisational efficacy. Theabsence of influence from active individualised transformational leadership to OCBsruns contrary to findings from a study in the USA (Podsakoff et al., 1996) wheretransformational leadership predicted OCBs. However, it could be speculated thatOCBs, at least in the context of Australian local councils, is a function of socialprocesses facilitated by the leader, rather than a function of the leader behavioursthemselves. The findings that social processes of leadership were predictive of bothOCBs and efficacy also lends credence to the fact that the social processes of leadership,originally conceptualised in an Australian and New Zealand context, were able tocapture cultural aspects specific to OCBs and organisational efficacy that the MLQcould not.

As in the Bycio et al. (1995) study, the present study found a negative correlationbetween active individualised transformational (incorporating contingent reward) andcontinuance commitment. It may be that the modus operandi of active individualisedtransformational leadership in Australian local councils decreases followers’continuance commitment through intellectual stimulation, vision definition,expectation clarification, charismatic behaviour and inspirational motivation. Thefinding that social processes of leadership did not predict organisational commitmentmay be construed to mean that, for this sample from the Australian public sector,organisational commitment may be more responsive to individualised council leader

674

LODJ33,7

behaviours (where a leader shows an individualised interest in the employee, ascaptured by active individualised transformational leadership) than to moregeneralised social processes facilitated by local council leaders.

There are a number of methodological, theoretical and managerial implications.First, in regard to research methodology, the study successfully used ICLUST analysisand hierarchical multiple regression analysis as alternative approaches toconfirmatory factor analysis. Therefore, researchers in the management disciplinecan successfully circumvent a failure to fit models to research data using confirmatoryfactor analysis and structural equation modelling. The method is especially suitablefor small sample sizes like that used for this research and for scale construction whichis careful to maintain cross-cultural (using US scales in an Australian context) andcross-organisational (using scales developed for private sector to public sector)equivalences.

Second, in regard to theoretical contribution, the study reinforced the importance ofcombining the transformational leader behaviours and social processes of leadership inthe same investigation if one wished to obtain a comprehensive understanding of theantecedents of the employee criterion variables examined in this study.

Third, the present study extended Parry and colleagues’ work (Parry, 2004; Parryand Proctor-Thomson, 2001), especially with respect to the proposition that socialprocesses of leadership are somewhat related to transformational leadership in regardto their potential impact on criterion variables. The findings that active individualisedtransformational leadership (part of transformational leadership in this study) andsocial processes of leadership both predicted performance outcomes and collectiveefficacy/expectancies confirmed that each was able to uniquely predict outcomes over andabove what the other predicted. In essence, both variables were getting at different driversof performance outcomes. If the relationship between the two constructs had been tooclose, then the inclusion of one would leave little for the other to uniquely contribute,which was not the case here. In the current study, active individualised transformationalleadership and social processes of leadership did different predictive work.

Fourth, during the scale construction, it became clear that the cluster compositionsof several scales indicated some differences in how the Australian sample perceiveseveral constructs as compared to findings from US samples. For example, for thetransformational leadership construct, our sample viewed leadership as pairingincreased active individualised transformational leadership with a reduced focus onactive management-by-exception. Therefore, leadership development in local councilsmay opt to develop those specific active individualised transformational leadershipand inhibit active management-by-exception aspects that were shown to be associatedwith employees’ criterion variables.

Consequently, the findings of the current study could be used to design appropriatetransformational leadership and social processes of leadership training programmesfor local council leaders and other public-sector organisations. Through understandinghow transformational leaders or leaders facilitating social processes motivate theirfollowers, managers may enhance those attributes associated with leadershipeffectiveness and thus increase performance outcomes among followers. For example,this could be done as suggested by Avolio and Bass (2004) where designated andpotential leaders in local councils undergo leadership development programmestargeting specific behavioural actions over time. These trainees could be provided withdevelopmental support including coaching on how to develop desired specificbehaviours (in this case transformational-transactional and social processes) and

675

Social processesof leadership

subsequent follow-up activities to augment specific leadership training programmes.Again, as suggested by Parry (2004), local councils could implement leadershiptraining through the emphasising of facets of the social processes of leadership. Parry(2004) advises trainers that key social processes dimensions can be “deconstructed intomore operational factors like personal efficacy and reciprocity [y] These lower-orderfactors can then be related to the many and varied manifestations of day-to-daybehaviours that reflect leadership” (p. 78). Through understanding how transformationalleaders or leaders facilitating social processes motivate their followers, managers couldenhance those attributes associated with leadership effectiveness and thus increaseperformance outcomes among followers in public-sector organisations.

Finally, given the generality and centrality of the leadership mechanism inthe fundamental structures governing diverse aspects of organisational functioning,programmes aimed at developing a resilient sense of leadership can yield significantdividends in performance accomplishments. Moreover, it is imperative for leaders inpublic-sector organisations like the local councils to understand, develop andpromote those active individualised leadership behaviours and social processes ofleadership that enhance employee performance outcomes, citizenship behaviours,organisational efficacy, collective efficacy/expectancies and affective-normativecommitment.

Limitations and future researchThe study had a few limitations. First, there was a relatively low response rate in thelocal councils’ samples. It was not easily established whether those who respondedwere different in any way from those who did not. The low response rate may havebeen due to the length of the survey and the fact that there were no obvious benefits forparticipants themselves. Second, there is a reduced capacity to draw causal inferencesinherent in the survey design of this study, especially as the data for testing theconceptual model were obtained on a single occasion from a single source. To allayconcerns for same-source bias, future research should examine the relationshipbetween the leadership variables and organisational outcomes variables by collectingthese data from different sources at separate points in time to the extent possible and,over a longer period of time. Third, this study relied heavily on self-reports which mayhave inflated relations among the research variables. However, the investigators tookthis limitation into consideration when conducting initial analysis for demographiccharacteristics and transformed data for variables found to be skewed. Duringsubsequent hierarchical multiple regression analyses, demographic characteristicswere statistically controlled for. Moreover, during the testing of the researchpropositions, the researchers employed second-order cluster compositions whosereliability and goodness-of fit measures were quite robust. Clearly, the emergingsecond-order clusters correlated with the primary measures in a manner generallyconsistent with the research propositions and extant theory. This enhances ourconfidence in the meaningfulness of the current findings. Fourth, this study examinedmembers of only one level of local government, namely local councils/shires in regionalNew South Wales. It is imperative that the current research findings are replicated inother occupations and contexts.

In terms of future research, we note that this study did not investigate how thepublic-sector context influenced the relationship between transformational and socialprocesses of leadership and organisational outcomes. Therefore, a future study couldexamine the direct and indirect effects of the public-sector context on the relationship

676

LODJ33,7

between leadership and organisational outcomes ( Javidan and Waldman, 2003).Additionally, this research did not investigate whether performance outcomes wouldvary as a function of the type of task, level of training and development provided to themanagers/leaders to work with employees/followers, and the nature of performancebeing assessed over time. However, findings from this study should stimulate moreresearch that will enhance our understanding of the relationship betweenorganisational leadership and employees’ criterion variables and otherorganisational performance outcomes in public-sector organisations. Furthermore,while this study employed a quantitative research approach, there is a need for futureresearch to make greater use of qualitative methods, especially embedded in mixed-methods leadership studies. Greater use of mixed-method approaches providesresearchers with grounded basis for the development of new leadership theory whilealso allowing for the integration of phenomena operating at multiple levels (Mumfordet al., 2009). Finally, future research may examine the assertion by Parry (2004)regarding transformational leadership and the social processes of leadershipmeasuring the same underlying construct through a more focused exploration of thetwo theories.

References

Andersen, J.A. (2009), “Public versus private managers: how public and private managers differin leadership behavior”, Public Administrative Review, Vol. 70 No. 1, pp. 131-41.

Arnold, K.A., Turner, N., Barling, J., Kelloway, E.K. and McKee, M.C. (2007), “Transformationalleadership and psychological well-being: the mediating role of meaningful work”, Journalof Occupational Health Psychology, Vol. 12 No. 3, pp. 193-203.

Avolio, B.J. and Bass, B.M. (2002), Developing Potential Across a Full Range of Leadership: Caseson Transactional and Transformational Leadership, Lawrence Erlbaum Associates,Mahwah, NJ.

Avolio, B.J. and Bass, B.M. (2004), Multifactor Leadership Questionnaire: Manual and SamplerSet, 3rd ed., Mind Garden, Inc, Redwood City.

Avolio, B.J., Walumbwa, F.O. and Weber, T.J. (2009), “Leadership: current theories, research, andfuture directions”, Annual Review of Psychology, Vol. 60 No. 1, pp. 421-49.

Bass, B.M. (1985), Leadership and Performance Beyond Expectations, Free Press, New York, NY.

Bass, B.M. (1997), “Does the transactional-transformational leadership paradigm transcendorganisational and national boundaries?”, American Psychologist, Vol. 52 No. 5,pp. 130-39.

Bass, B.M. and Avolio, B.J. (2004), Multifactor Leadership Questionnaire. Manual and SamplerSet, 3rd ed., Mind Garden, Inc, Redwood City, CA.

Bohn, J.G. (2002), “The relationship of perceived leadership behaviors to organizational efficacy”,Journal of Leadership and Organizational Studies, Vol. 9 No. 2, pp. 65-79.

Bycio, P., Allen, J.S. and Hackett, R.D. (1995), “Further assessments of Bass’s (1985)conceptualization of transactional and transformational leadership”, Journal of AppliedPsychology, Vol. 80 No. 4, pp. 468-78.

Cohen, J., Cohen, P., West, S.G. and Aiken, S.L. (2003), Applied Multiple Regression/CorrelationAnalysis for the Behavioural Sciences, 3rd ed., Lawrence Erlbaum Associates, Mahwah, NJ.

Cooksey, R.W. and Soutar, G.N. (2006), “Coefficient beta and hierarchical item clustering: ananalytical procedure for establishing and displaying the dimensionality and homogeneityof summated scales”, Organisational Research Methods, Vol. 9 No. 1, pp. 78-98.

677

Social processesof leadership

Currie, G., Lockett, A. and Suhomlinova, O. (2009), “Leadership and institutional change in thepublic sector: the case of secondary schools in England”, Leadership Quarterly, Vol. 20No. 44, pp. 664-79.

Dansereau, F. and Yammarino, F.J. (Eds) (1998), “A multiple-level mosaic: one way to putthe pieces together”, Leadership: The Multiple-Level Approaches, JAI Press, London,pp. 327-49.

Department of Premier and Cabinet (2010), “Annual report 2009-2010”, Division of LocalGovernment, Department of Premier and Cabinet, NSW, available at: www.dpc.nsw.gov.au/_data/assets/pdf_file/0016/109420/Annual_Report_0910.pdf (accessed 17 July 2011).

Dunoon, D. (2002), “Rethinking leadership for the public sector”, Australian Journal of PublicAdministration, Vol. 61 No. 3, pp. 3-18.

Fernandez, S. and Rainey, H.G. (2006), “Managing successful organisational change in the publicsector”, Public Administration Review, Vol. 66 No. 2, pp. 168-76.

Fernandez, S., Cho, Y.J. and Perry, J.L. (2010), “Exploring the link between integrated leadershipand public sector performance”, Leadership Quarterly, Vol. 21 No. 2, pp. 308-23.

Ingraham, P.W. and Getha-Taylor, H. (2004), “Leadership in the public sector: models andassumptions for leadership development in the federal government”, Review of PublicPersonnel Administration, Vol. 24 No. 2, pp. 95-112.

Javidan, M. and Waldman, D.A. (2003), “Exploring charismatic leadership in the public sector:measurement and consequences”, Public Administrative Review, Vol. 63 No. 2, pp. 229-42.

Judge, T.A. and Piccolo, R.F. (2004), “Transformational and transactional leadership:a meta-analytic test of their relative validity”, Journal of Applied Psychology, Vol. 89No. 5, pp. 755-68.

Kirkman, B.L., Chen, G., Farh, J.L., Chen, Z.X. and Lowe, K.B. (2009), “Individual power distanceorientation and follower reactions to transformational leaders: a cross-level, cross-culturalexamination”, Academy of Management Journal, Vol. 52 No. 4, pp. 744-64.

Lowe, K.B. and Gardner, W.L. (2001), “Ten years of the leadership quarterly: contributions andchallenges for the future”, Leadership Quarterly, Vol. 11 No. 4, pp. 459-514.

Lowe, K.B., Kroeck, K.G. and Sivasubramaniam, N. (1996), “Effectiveness correlates oftransformational and transactional leadership: a meta-analytic review of the MLQliterature”, Leadership Quarterly, Vol. 7 No. 3, pp. 385-425.

Meyer, J.P., Allen, N.J. and Smith, C.A. (1993), “Commitment to organizations and occupations:extension and test of a three-component conceptualization”, Journal of Applied Psychology,Vol. 78 No. 4, pp. 538-51.

Muchiri, M.K. and Cooksey, R.W. (2010), “Using hierarchical item clustering to establish thedimensionality of the Multifactor Leadership Questionnaire”, International Journal ofOrganisational Behaviour, Vol. 15 No. 1, pp. 1-14.

Muchiri, M.K., Cooksey, R.W., Di Milia, L.V. and Walumbwa, F.O. (2011), “Gender and manageriallevel perceptions of effective leadership”, Leadership & Organization Development Journal,Vol. 32 No. 5, pp. 462-92.

Mumford, M.D., Friedrich, T.L., Caughron, J.J. and Antes, A.L. (2009), “Leadership research:traditions, developments, and current directions”, in Buchanan, D.A. and Bryman, A.(Eds), The Sage Handbook of Organizational Research Methods, Sage, Los Angeles, CA,pp. 111-27.

Osborne, D. and Gaebler, T. (1992), Reinventing Government: How the Entrepreneurial Spirit isTransforming the Public Sector From Schoolhouse to State House, City to Pentagon,Addison-Wesley, Reading, MA.

Parry, K.W. (1998), “Grounded theory and social processes: a new direction for leadershipresearch”, Leadership Quarterly, Vol. 9 No. 1, pp. 85-105.

678

LODJ33,7

Parry, K.W. (1999), “Enhancing adaptability: leadership strategies to accommodate change inlocal government settings”, Journal of Organisational Change Management, Vol. 12 No. 2,pp. 134-56.

Parry, K.W. (2002), “Hierarchy of abstraction modelling (H.A.M) and the psychometric validationof grounded theory research”, International Journal of Organisational Behaviour, Vol. 5No. 5, pp. 180-94.

Parry, K.W. (2004), “Comparative modelling of the social processes of leadership in work units”,Journal of Australian and New Zealand Academy of Management, Vol. 10 No. 2, pp. 69-80.

Parry, K.W. and Proctor-Thomson, S.B. (2001), “Leadership, culture and performance: thecase study of the New Zealand public sector”, Journal of Change Management, Vol. 3 No. 4,pp. 376-99.

Parry, K.W. and Sarros, J.C. (1996), “An Australian perspective on transformational leadership”,in Parry, K.W. (Ed.), Leadership Research and Practice: Emerging Themes and NewChallenges, Pitman Publishing, Melbourne, pp. 105-11.

Podsakoff, P.M. and MacKenzie, S.B. (1997), “The impact of organisational citizenship behaviouron organisational performance: a review and suggestions for future research”, HumanPerformance, Vol. 10 No. 2, pp. 133-51.

Podsakoff, P.M., MacKenzie, S.B. and Bommer, W.H. (1996), “Transformational leader behavioursand substitutes for leadership as determinants of employee satisfaction, commitment,trust, and organisational citizenship behaviours”, Journal of Management, Vol. 22 No. 2,pp. 259-98.

Podsakoff, P.M., MacKenzie, S.B. and Fetter, R. (1993), “Substitutes for leadership and themanagement of professionals”, Leadership Quarterly, Vol. 4 No. 1, pp. 1-44.

Revelle, W. (1978), “ICLUST: a cluster analytic approach for exploratory and confirmatory scaleconstruction”, Behavior Research Methods & Instrumentation, Vol. 10 No. 5, pp. 739-42.

Revelle, W. (1979), “Hierarchical cluster analysis and the internal structure of tests”, MultivariateBehavioral Research, Vol. 14 No. 1, pp. 57-74.

Richard, P.J., Devinney, T.M., Yip, G.S. and Johnson, G. (2009), “Measuring organizationalperformance: towards methodological best practice”, Journal of Management, Vol. 35No. 3, pp. 718-804.

Riggs, M.L. and Knight, P.A. (1994), “The impact of perceived group success-failure onmotivational beliefs and attitudes: a causal model”, Journal of Applied Psychology, Vol. 79No. 5, pp. 755-66.

Riggs, M.L., Warka, J., Babasa, B., Betancourt, R. and Hooker, S. (1994), “Development andvalidation of self-efficacy and outcome expectancy scales for job-related applications”,Educational and Psychological Measurement, Vol. 54 No. 3, pp. 793-802.

Thach, E. and Thompson, K.J. (2006), “Trading places: examining leadership competencesbetween for-profit vs public and non-profit leaders”, Leadership & OrganizationDevelopment Journal, Vol. 28 No. 4, pp. 356-75.

Van Wart, M. (2003), “Public-sector leadership theory: an assessment”, Public AdministrationReview, Vol. 63 No. 2, pp. 214-28.

Van Wart, M. (2005), Dynamics of Leadership in Public Service: Theory and Practice, ME Sharpe,Armonk, NY.

Walumbwa, F.O. and Schaubroeck, J. (2009), “Leader personality traits and employee voicebehavior: mediating roles of ethical leadership and work group psychological safety”,Journal of Applied Psychology, Vol. 94 No. 5, pp. 1275-86.

Walumbwa, F.O., Hartnell, C.A. and Oke, A. (2010), “Servant leadership, procedural justiceclimate, service climate, employee attitudes, and organizational citizenship behavior:a cross-level investigation”, Journal of Applied Psychology, Vol. 95 No. 3, pp. 517-29.

679

Social processesof leadership

Walumbwa, F.O., Lawler, J.J. and Avolio, B.J. (2007), “Leadership, individual differences and workattitudes: a cross-culture investigation”, Applied Psychology: An International Review,Vol. 56 No. 2, pp. 212-30.

Walumbwa, F.O., Luthans, F., Avey, J.B. and Oke, A. (2011), “Authentically leading groups: themediating role of collective psychological capital and trust”, Journal of OrganizationalBehavior, Vol. 32 No. 1, pp. 4-24.

Walumbwa, F.O., Lawler, J.J., Avolio, B.J., Wang, P. and Shi, K. (2005), “Transformationalleadership and work-related attitudes: the moderating effects of collective and self-efficacyacross cultures”, Journal of Leadership and Organizational Studies, Vol. 11 No. 3, pp. 2-16.

Yammarino, F.J., Dansereau, F. and Kennedy, C.J. (2001), “A multiple-level multidimensionalapproach to leadership: viewing leadership through an elephant’s eye”, OrganisationalDynamics, Vol. 29 No. 3, pp. 149-63.

Yukl, G.A. (2010), Leadership in Organizations, 7th ed., Pearson, Upper Saddle River, NJ.

Zhu, W., Avolio, B.J. and Walumbwa, F.O. (2009), “Moderating role of follower characteristics withtransformational leadership and follower work engagement”, Group OrganizationManagement, Vol. 34 No. 5, pp. 590-619.

Zinbarg, R., Revelle, W., Yovel, I. and Li, W. (2005), “Cronbach’s a, Revelle’s b and McDonald’s oH:their relations with each other and two alternative conceptualizations of reliability”,Psychometrika, Vol. 70 No. 1, pp. 123-33.

Further reading

Arnold, K.A., Barling, J. and Kelloway, E.K. (2001), “Transformational leadership or the ironcage: which predicts trust, commitment and team efficacy?”, Leadership & OrganizationDevelopment Journal, Vol. 22 No. 7, pp. 315-20.

Feldman, M.S. (2005), “Management and public management”, Academy of Management Journal,Vol. 48 No. 6, pp. 958-60.

Fernandez, S. (2004), “Developing and testing an integrative framework of public sectorleadership: evidence from the public education arena”, Journal of Public AdministrationResearch and Theory, Vol. 15 No. 2, pp. 197-217.

About the authors

Michael K. Muchiri, PhD, is a Senior Lecturer in Management at the School of Management andMarketing, CQUniversity in Rockhampton, Australia. He lectures in Management and hiscurrent research is on positive leader behaviours. Michael K. Muchiri is the corresponding authorand can be contacted at: [email protected]

Ray W. Cooksey, PhD, is a Professor at the School of Business Economics and Public Policy,University of New England, Australia. He lectures in Organisational Behaviour and his researchis focused mainly in the areas of decision making and cognition, chaos and complexity theory,and organisational and behavioural systems theory.

Fred O. Walumbwa, PhD, is an Associate Professor of Management at the W.P. Carey Schoolof Business, Arizona State University, Arizona, USA. He lectures in Management and hisresearch is focused mainly in the areas of authentic and transformational leadership.

680

LODJ33,7

Appendix 1

Attention towardsfailures (27)

Concentrates onmistakes (22)

Tracks mistakes (24)

Variable � �

0.588

Size of final cluster Name of final cluster

0.612 3 Active managementby exception

0.547

0.588

Considers mydifferent needs (29)

Examines problemsfrom many angles (30)

0.605

0.667 4 Intellectualstimulation

0.599

0.605

Differing perspectives (8)

Re-examines criticalassumptions (2) 0.539 0.539

Fails to interfere tillproblems arise (3)Absent when needed (7)

Delays responding (33)

Waits for things to gowrong (12)Problems chronicbefore action (20)

Avoids decisions (28)

Avoids gettinginvolved (5)

0.833 70.710

0.695 0.695

0.746 0.746

0.607 0.607

0.825 0.761

0.781 0.743

Assistance in exchangefor efforts (1)Treats measanindividual (19)Talks values and beliefs (6)

Optimistic about future (9)

Enthusiastic onaccomplishments (13)Importance ofpurpose (14)

Compellingvision (26)

Importance ofcollective (34)

Instil spride (10)

Discussesresponsibilities

Expressessatisfaction (35)

Expressesconfidence (36)Clarifies rewardsafter goal (16)

Teaches andcoaches (15)

Beyond self-interest (18)

Builds myrespect (21)

Develop my strengths (31)Suggests newways (32)

0.952 180.860

Laissez-faire/passivemanagement by exception

0.951 0.871

0.949 0.779

0.590 0.590

0.896 0.862

0.754 0.754

0.949 0.886

0.806 0.806

0.943 0.857

0.765 0.765

0.939 0.884

0.772 0.772

0.930 0.887

0.882 0.882

0.8620.862

0.832 0.832

0.913 0.867

Notes: Item numbers are in parenthesis and bold; coefficient � are in italics, and coefficient�s are in bold; overall goodness-of-fit (VSS) = 0.95

Transformationalcontingent reward

Figure A1.First-order hierarchical

tree diagram fromICLUST analysis of theMultifactor Leadership

Questionnaire

681

Social processesof leadership

Appendix 2

Appendix 3

Active management by exception (1) Active management by exception

0.738 0.657 3

Laissez-Faire/passivemanagement byexception (3)

Transformationalcontingentreward (4)

Intellectualstimulation (2)

0.742 0.742

Variable � � Size of final clusters Name of cluster

Active individualisedtransformationalleadership

1

Notes: Cluster numbers are in parentheses and in bold; cluster 1 did not join other clusters; coefficient�s are in italics, and coefficient �s are in bold; overall goodness-of-fit (VSS) = 0.96

Figure A2.Second-order hierarchicaltree diagram fromICLUST analysis of theMultifactor LeadershipQuestionnaire

Variable � � Size of final clusters

Face-to-face or phonediscussions (7) 0.768 0.768Discuss organisationaldirection (8)

Rely on each otherfor info (15)

Reciprocity/interdependence0.829 0.829 2

Rely on each other for support (16)

Display perceptionof own leadership (13) Personal

leadership0.769 2Move work unitforward (14)

Personal responsibilityto educate (11) 0.846 0.799 Resolution

andpersonal

responsibility

4Display personalvalues (9)

0.825 0.795

Resolve membersuncertainty (10)

0.791 0.791Resolve personaluncertainty (12)

Rational explanations (5) 0.903

Optimising-activemanagement

Persist in efforts (2)

0.895 0.847Co-operate (3)

0.793 0.793

Share information (4) 0.885 0.845

Effectively planahead (6) 0.808

0.808

Sound managementpractice (17) 0.846 0.814

Distribute tasks(20)

0.726 0.726 Facilitateinfo flow (19)

Notes: Item numbers are in parentheses and in bold; coefficient �s are in italics, and coefficient �s are inbold; overall goodness-of-fit (VSS) = 0.96

Name of cluster

2 Communication

0.769

0.843 8

Figure A3.First-order hierarchicaltree diagram fromICLUST analysis of thesocial processes ofleadership scale

682

LODJ33,7

Appendix 4

Variable � � Size of final clusters

Reciprocity/interdependence (2)

Notes: Cluster numbers are in parentheses and in bold; coefficient �s are in italics, andcoefficient �s are in bold; overall goodness-of-fit (VSS) = 0.99

0.854 0.776 Socialprocesses ofleadership

5

Personal leadership (3)0.843 0.773

Communication (1) 0.830 0.801

Optimising- active management(5) 0.7910.791Resolution andpersonal reponsibility (4)

Name of cluster

Figure A4.Second-order hierarchical

tree diagram fromICLUST analysis of the

social processes ofleadership scale

To purchase reprints of this article please e-mail: [email protected] visit our web site for further details: www.emeraldinsight.com/reprints

683

Social processesof leadership