annual report

This article was downloaded by: [Air University]On: 07 May 2015, At: 21:39Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954 Registeredoffice: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Click for updates

Work & Stress: An International Journalof Work, Health & OrganisationsPublication details, including instructions for authors andsubscription information:http://www.tandfonline.com/loi/twst20

Is avoidant leadership a root causeof subordinate stress? Longitudinalrelationships between laissez-faireleadership and role ambiguityAnders Skogstada, Jrn Hetlanda, Lars Glasab & Stle Einarsenaa Department of Psychosocial Science, University of Bergen,Bergen, Norwayb Department of Leadership and Organizational Behaviour, BINorwegian Business School, Oslo, NorwayPublished online: 25 Sep 2014.

To cite this article: Anders Skogstad, Jrn Hetland, Lars Glas & Stle Einarsen (2014) Is avoidantleadership a root cause of subordinate stress? Longitudinal relationships between laissez-faire leadership and role ambiguity, Work & Stress: An International Journal of Work, Health &Organisations, 28:4, 323-341, DOI: 10.1080/02678373.2014.957362

To link to this article: http://dx.doi.org/10.1080/02678373.2014.957362

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (theContent) contained in the publications on our platform. However, Taylor & Francis,our agents, and our licensors make no representations or warranties whatsoever as tothe accuracy, completeness, or suitability for any purpose of the Content. Any opinionsand views expressed in this publication are the opinions and views of the authors,and are not the views of or endorsed by Taylor & Francis. The accuracy of the Contentshould not be relied upon and should be independently verified with primary sourcesof information. Taylor and Francis shall not be liable for any losses, actions, claims,proceedings, demands, costs, expenses, damages, and other liabilities whatsoever orhowsoever caused arising directly or indirectly in connection with, in relation to or arisingout of the use of the Content.

This article may be used for research, teaching, and private study purposes. Anysubstantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &

Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

Dow

nloa

ded

by [A

ir Un

iversi

ty] at

21:39

07 M

ay 20

15

Is avoidant leadership a root cause of subordinate stress? Longitudinalrelationships between laissez-faire leadership and role ambiguity

Anders Skogstada*, Jrn Hetlanda, Lars Glasa,b and Stle Einarsena

aDepartment of Psychosocial Science, University of Bergen, Bergen, Norway; bDepartment ofLeadership and Organizational Behaviour, BI Norwegian Business School, Oslo, Norway

(Received 9 February 2012; final version accepted 16 December 2013)

Within a stressor-strain framework, the aim of this study is to examine the influence of perceivedlaissez-faire leadership a type of leadership characterized by the superiors avoidance and inactionwhen subordinates are experiencing a situational need for leadership on stress in the form ofexperiences of role ambiguity. This was tested within a prospective 3-wave research design withtime lags of 2 and 3 years respectively. A combination of 2 constructive types of leadership,initiating structure and showing consideration, was included as a control. In a sample of 1771employees drawn from the Norwegian working population, structural equation modelling supportedthe hypothesis that perceived laissez-faire leadership would be positively related to experiencingrole ambiguity at 3 consecutive measurement points. The findings did not support a potentialreciprocal relationship between the two across time. Initiation of structure and Consideration werenot related to role ambiguity, providing strong evidence of the importance of laissez-faire leadershipwithin a stressor-strain framework. Demonstrating that laissez-faire leadership is a root source ofsubordinate role ambiguity underlines the importance of superiors perceiving situations andcircumstances where subordinates experience a need for leadership and, accordingly, approachingthis need instead of avoiding it.

Keywords: leadership; laissez-faire; initiation of structure; consideration; role ambiguity; work-related stress

Introduction

As pointed out by Hinkin and Schriesheim (2008a) as well as Judge and Piccolo (2004),research on laissez-faire leadership is scarce compared to the abundant studies ontransformational and transactional forms of leadership. However, studies have shown thatthe prevalence of laissez-faire leadership in contemporary working life is strikingly high(Aasland, Skogstad, Notelaers, Nielsen, & Einarsen, 2010), with there being documentednegative associations of such leadership with subordinate satisfaction with the leader,evaluations of leader effectiveness and subordinates experiencing stress (Judge & Piccolo,2004; Skogstad, Einarsen, Torsheim, Aasland, & Hetland, 2007). Kelloway, Sivanathan,

*Corresponding author. Email: [email protected]

Work & Stress, 2014Vol. 28, No. 4, 323341, http://dx.doi.org/10.1080/02678373.2014.957362

2014 Taylor & Francis

Dow

nloa

ded

by [A

ir Un

iversi

ty] at

21:39

07 M

ay 20

15

Francis, and Barling (2005) therefore put forward the hypothesis that poor leadership,including laissez-faire, is a root cause of important workplace stressors such as roleambiguity. In line with this, it is reasonable to believe that laissez-faire leadership characterized by the avoidance of leadership behaviours, or the sheer lack of leaderpresence when subordinates are in need of leadership (Bass & Avolio, 1994, 2004) willbe closely related to subordinates experiencing stressful work situations characterized bya lack of clarity regarding duties and responsibilities within the organization. Suchsystematic relationships between laissez-faire leadership and role ambiguity were foundin a cross-sectional study (Skogstad et al., 2007). However, to our knowledge, no studieshave investigated these relationships in a longitudinal design to substantiate the directionof their relationships. The present study thus aims to contribute to a more nuancedunderstanding of leadership behaviour by adding to our scarce knowledge of the potentialnegative outcomes related to laissez-faire leadership, employing a stressor-strainframework (see, e.g. Skakon, Nielsen, Borg, & Guzman, 2010 for an overview), as analternative to the traditional satisfaction and effectiveness frameworks often used inleadership studies (see, e.g. Judge & Piccolo, 2004). Role stress in the form of roleambiguity is one of the most frequently studied occupational stressors in contemporaryworking life (Beehr, 1995; rtqvist & Wincent, 2006), which in turn has beendocumented to be related to a wide range of negative outcomes, such as reduced jobsatisfaction, dissatisfaction with the leader, reduced organizational commitment and joband task performance, burnout reactions and turnover intentions (Gilboa, Shirom, Fried, &Cooper, 2008; Podsakoff, Bommer, Podsakoff, & MacKenzie, 2006). The study oflaissez-faire leadership as an antecedent of role ambiguity is thus an importantcontribution to the literature.

In addition to looking at the relationship between laissez-faire leadership and roleambiguity within such a stressor-strain framework, there may be arguments for a circularrelationship between the two variables in that respondents high on role ambiguity maythen start to perceive their superiors as being laissez-faire or to attribute their ambiguouswork situation to the lack of leadership from their immediate superior. Employing alongitudinal research design with three measurement points, we will investigate whethersuch a stressor-strain relationship may also be identified by investigating a possiblereversed relationship, in line with the so called gloomy perception mechanism(de Lange, Taris, Kompier, Houtman, & Bongers, 2005).

In their Full Range of Leadership model (FRL), consisting of transactional,transformational and laissez-faire forms of leadership, Bass and Avolio (1994, p. 4)define laissez-faire leadership as: the avoidance or absence of leadership and is, bydefinition, the most inactive as well as the most ineffective according to almost allresearch on the style. As opposed to transactional leadership, laissez-faire represents anon-transaction. Transactional, together with transformational leadership, may thus beperceived as an antipole of laissez-faire leadership on an active-passive continuum forsuperior-subordinate transactions (see the active-passive dimension in the FRL model,Bass & Avolio, 1994, p. 6). However, a meta-study by Judge and Piccolo (2004) showedfar from perfect negative correlations (q^ .38 and .51, where q^ is the estimated truecorrelation) between laissez-faire leadership and contingent reward and management-by-exception active leadership, respectively, as did Piccolo et al.s (2012) meta-studyshowing correlations of .47 and .62 (in two separate studies) between laissez-faireleadership and transactional leadership. These findings substantiate that leaders use oftransactional forms of leadership does not exclude their use of laissez-faire leadership and

A. Skogstad et al.324

Dow

nloa

ded

by [A

ir Un

iversi

ty] at

21:39

07 M

ay 20

15

vice versa. In line with this, Bass (1990, p. 547) stated that Laissez-faire leadership doesnot seem to be the exact opposite of active leadership. Likewise, Bass and Riggio (2006,p. 9) stated that Fundamental to the FRL model is that every leader displays each style tosome amount. Laissez-faire leadership is probably different therefore from transactionalleadership forms in more ways than simply representing its antipole on an inactivity-activity continuum for effective leadership.

As various descriptive definitions of laissez-faire leadership have been presented (seeHinkin & Schriesheim, 2008a, for an overview), one feasible way of approaching theconcept of laissez-faire leadership is to examine its operational definition in theMultifactor Leadership Questionnaire (Form 5x-Short; Bass & Avolio, 2004, p. 98),which has been the predominant measure of laissez-faire leadership. The MLQoperationalizes laissez-faire leadership as a situational need for leadership in three ofthe four items, which point to situations where subordinates are in need of some sort ofassistance from their superiors that is then not given (e.g. absent when needed; Bass& Avolio, 2004, p. 98). These needs do not represent basic physiological, psychologicalor social needs but are so-called quasi-needs originating from situational demands andpressures (Lewin, 1936; Reeve, 2005). The imbalance between subordinate competenceand resources on one hand and the situational demands on the other, show that this is asituation where the subordinate experiences situational vulnerability (Lapidot, Kark, &Shamir, 2007) and even stress (see, e.g. Lazarus, 1999) with a high correspondingpotential for negative outcomes. This need, characteristic of laissez-faire leadership, iscomparable to de Vries, Roe, and Taillieus (2002, p. 122) concept of the need forleadership defined as the extent to which an employee wishes the leader to facilitatethe paths toward individual, group, and/or organizational goals, and further explained asThe need is contextual in the sense that it depends on the persons assessment of theparticular setting (p. 123). Laissez-faire leadership may therefore be perceived as a typeof situational lack of leadership when a subordinate experiences lack of competence orother resources, indicating that laissez-faire leadership may have strong negativeconsequences in such critical situations, whereas it may be far less influential insituations where subordinates are able to cope with the situation.

In accordance with Bass and Avolios (1994) above-cited definition, two of four itemsin the scale specify that the leader avoids the leader behaviour that is required. In a thirditem, the word delay is used similarly. Bass and Avolio thus seem to support theacknowledged approach-avoidance distinction, which is fundamental in many theoreticalapproaches in psychology (see, e.g. Higgins, 1997 for overviews) including stresstheories and research (see, e.g. Skinner, Edge, Altman, & Sherwood, 2003). Laissez-faireleader behaviours thus represent volitional activity with respect to avoiding subordinateswhen they need assistance, while at the same time these leader behaviours reflectinactivity with respect to approaching subordinates who require leader assistance.

In conclusion, laissez-faire leadership may be defined as a follower-centred form ofavoidance-based leadership by focusing on subordinates perceived situational need forleadership, and leader non-response to such needs, as the main source of variance inoutcomes. In contrast, transactional and transformational forms of leadership are leader-centred leadership forms because of their focus on superiors initiation of transactionaland transformational behaviours as the main source of variance in outcomes (see Shamir,2007, for an overview). Moreover, laissez-faire leadership is perceived as leadersvolitional and active avoidance of subordinates when they are in need of assistance, incontrast to transactional and transformational forms of leadership. The latter forms

Work & Stress 325

Dow

nloa

ded

by [A

ir Un

iversi

ty] at

21:39

07 M

ay 20

15

include contingent rewards and consideration for individuals, which may be perceived asactive leadership approaches to attaining organizational goals and attending to individualneeds. Laissez-faire leadership is thus clearly distinguished from transactional andtransformational forms of leadership and constitutes a unique form of leadership thatshould be studied in its own right.

As laissez-faire leadership is an integral part of the Full Range of Leadership (FRL)model, probably the most studied leadership model over the last decades, our first ideawas to contrast and compare this leadership style with the two other styles of leadership inthe model, namely transactional and transformational. However, on reflection we decidedit would be even more fruitful to contrast and compare laissez-faire leadership with thestructuring and supportive leadership behaviours of Initiating Structure and showingConsideration for subordinates, which may be perceived as the original true oppositesof laissez-faire leadership (Lewin, Lippit, & White, 1939). Such forms of leadership aremissing in the FRL model (Yukl, 1999) but were firmly revalidated in a meta-study byJudge, Piccolo, and Ilies (2004). In an exceptional meta-study, integrating leadershipforms from the FRL model with the classical leadership styles of Initiating Structureand showing Consideration, Piccolo et al. (2012) found negative correlations of .55 and.60 between laissez-faire leadership and Initiating Structure and Consideration,respectively. The two meta-studies (Judge & Piccolo, 2004; Piccolo et al., 2012)nonetheless showed comparable relationships between laissez-faire leadership andtransactional leadership and between laissez-faire leadership and Initiating Structure andConsideration, respectively, which in turn substantiates that laissez-faire leadership by nomeans represents the absence of other leadership forms but is a unique type of leadershipworth studying by itself.

The present study aims to investigate whether perceived laissez-faire leadership,representing perceived active avoidance of leader behaviours when subordinates are inneed of such behaviours, explains subsequent increases in role ambiguity over and abovethat of the constructive leadership styles of Initiating Structure and showing Considera-tion. According to path-goal theory, in order to be effective, leaders have to providestructure for subordinates by letting them know what they are expected to do and toprovide them with what is missing or needed in the situation (House, 1996). In his paper,House (1996) proposed that leader Initiating Structure is related to lower levels offollower role ambiguity, a point of view that received support in his 1971 study. In a fieldexperiment, Schaubroeck, Ganster, Sime, and Ditman (1993) found that after superiorshad been trained in a method of clarifying subordinates roles, a follow-up questionnaireshowed reduced subordinate role ambiguity. In line with this, Bandura (1997, p. 90),stated that clear roles are congruent with the cognitive representation processes inwhich individuals vicariously imagine or experience success in a task. Leadersclarification of expectations in subordinates work roles may thus be defined as a coreingredient of leadership. There is reason to believe that a perceived avoidance of such aclarification will probably result in increased role ambiguity, which in turn may beassociated with a broad range of negative outcomes (see, e.g. Chen & Bliese, 2002;Jackson, Schwab, & Schuler, 1986; MacKenzie, Podsakoff, & Rich, 2001). Most studieson role ambiguity, which has been described as the lack of clarity regarding duties,objectives and responsibilities involved in fulfilling ones role in the organization (Beehr &Glazer, 2005), have employed the Rizzo, House, and Lirtzman scale (1970) in which, inaddition to an unpredictability component, role ambiguity is defined as a lack of theexistence of clarity of behavioural requirements, often in terms of inputs from the

A. Skogstad et al.326

Dow

nloa

ded

by [A

ir Un

iversi

ty] at

21:39

07 M

ay 20

15

environment, which would serve to guide behaviour, and provide knowledge that thebehaviour is appropriate. (Rizzo et al., 1970, pp. 155156), again demonstrating its closerelationship with leadership.

Even though Kahn, Wolfe, Quinn, Snoek, and Rosenthal (1964), in their seminalstudies on role conflict and role ambiguity, found superiors to be the strongest and mostpowerful role senders in their interpersonal relationship with subordinates and, therefore,probably a primary source of subordinates role ambiguity when exhibiting laissez-fairebehaviour, only a limited number of studies have investigated the actual relationshipbetween leader behaviour and follower role ambiguity. In addition, most studies are ratherold and, to our knowledge, only two have employed a longitudinal research design. In alongitudinal study, Moyle (1998) showed that role ambiguity mediated the relationshipbetween managerial support and job satisfaction, while Nielsen, Randall, Yarker, andBrenner (2008) found that work characteristics, including role clarity, meaningfulness andopportunities for development, mediated the relationship between transformationalleadership and psychological well-being.

Reviewing cross-sectional evidence, Fisher and Gitelson (1983) found a meancorrelation of .37 between role ambiguity and satisfaction with the supervisor in theirmeta-analysis. In their frequently cited meta-analysis, Jackson and Schuler (1985) foundthat both leader Consideration and Initiating Structure were related to employees roleambiguity (r = .43 and r = .44, respectively) which confirmed similar findings bySchriesheim, House, and Kerr (1976). More recent studies have also confirmed a negativerelationship between showing Consideration and Initiating Structure, and measures of roleambiguity and role stress (Dale & Fox, 2008; ODriscoll & Beehr, 1994). With respect torelationships between leadership forms in the FRL model and role ambiguity, Stordeur,Dhoore, and Vandenberghe (2001) found transformational forms of leadership to beassociated with subordinates role ambiguity (r = .42 and r = .39, p < .001), whilecontingent reward showed a more moderate yet, still, significant correlation (r = .29,p < .001). MacKenzie, Podsakoff, and Rich (2001) found comparable correlations (.48)between core transformational leadership and role ambiguity. Hinkin and Schriesheim(2008a) and Jackson et al. (1986) found significant correlations (.39 and .43, respectively)between contingent reward and role clarity, measured by unreversed scores of roleambiguity. Based on the argument presented earlier, that Initiating structure and showingConsideration may be defined as opposites of laissez-faire leadership, and their relativestrong predictions of role ambiguity, these leadership styles will be examined in thepresent study.

However, the main aim of the present study is to investigate the effect of laissez-faireleadership, representing perceived active avoidance of leader behaviours when sub-ordinates are in need of such behaviours, and if such leadership explains subsequentincreases in role ambiguity over and above that of Initiating Structure and showingConsideration. In this regard, it is interesting to note that Rizzo and colleagues (1970) intheir seminal work have already examined this relationship, documenting significantcorrelations between leader role abdication (which may be seen as a proxy of laissez-faireleadership) and role ambiguity (r = .28 and r = .17, p < .05), substantiating asystematic negative relationship between the two variables. In a more recent cross-sectional study in a representative sample of the Norwegian working population(N = 2539), Hauge, Skogstad, and Einarsen (2007) found laissez-faire leadership tocorrelate even more strongly with role ambiguity (r = .44, p

lack of studies focusing on laissez-faire leadership as an influential leadership style per se,making it the stepchild in the FRL model (see, e.g. Hinkin & Schriesheim, 2008a), it istime to reverse the lenses and turn them from the bright sides of leadership to its darkersides, in this case in the form of laissez-faire leadership as an antecedent of role ambiguity.This reversal is further supported by the fact that laissez-faire leadership has shown to beassociated with a variety of negative attitudinal and behavioural consequences, in the formof negative correlations with follower job satisfaction, satisfaction with the leader andleader effectiveness (Judge & Piccolo, 2004), as well as positive associations with jobinsecurity (Aasland, 2012), health complaints (Einarsen, Skogstad, & Aasland, 2010),burnout reactions (Hetland, Sandal, & Johnsen, 2007), work withdrawal and intentions toleave (Skogstad, Notelaers, & Einarsen, 2009) and bullying (Hauge et al., 2007; Skogstadet al., 2007). Accordingly, in the present study laissez-faire will be the focal leadership styleas regards its effect on role ambiguity.

Kelloway and colleagues (2005, p. 97) state that researchers have argued that roleambiguity is the variable that might be most readily influenced by managers. It is thereforetheoretically sound to hypothesize that laissez-faire leadership functions as a social stressorwhich over time may cause elevated levels of subordinate role ambiguity, even over andabove that of limited Initiation of structure and lack of Consideration. According to theDemand-Control-Support model (Karasek & Theorell, 1990) and the Job Demands-Resources model (Bakker & Demerouti, 2007), employees experiences of low support(which is highly probable during exposure to laissez-faire leadership) may lead to increasedstress reactions; this is a notion that has been supported in other longitudinal studiesshowing the main effects of support on employees well-being and health outcomes (see deLange, Taris, Kompier, Houtman, & Bongers, 2003, for an overview). Hence, the presentstudy will focus on laissez-faire leadership as the main candidate for explaining roleambiguity. However, as leaders over time will show destructive laissez-faire as well asinstances of constructive leadership behaviours, we will control for the influence of theconstructive behaviour of Initiation of structure and Consideration.

On the basis of this background our main hypothesis (H1) is that:

Hypothesis 1. Laissez-faire leadership will be positively related to role ambiguity at threeconsecutive measurement points.

Role ambiguity tends to be investigated in conjunction with role conflict, often defined asthe existence of two or more sets of expectations on the focal person (as well as sent rolepressures) such that compliance with one makes compliance with the other more difficult(Beehr, 1995, p. 58). However, in relation to laissez-faire leadership, such expectationsare contradictory as laissez-faire leadership represents a lack of leader behaviour and, assuch, also a lack of role expectations. As a lack of role expectations may not be directlyrelated to role conflict, we regard it as theoretically unsound to study relationshipsbetween this form of leadership and role conflict. Another issue of concern is that theconcept of role conflict is related to the presence of various role-senders (e.g. intra-sender,inter-sender and person-role conflicts) both in its definition (Beehr, 1995) and in theoperationalization of the concept (Rizzo et al., 1970), which makes it difficult to interpretresults relating to the influence of leaders role expectations compared to other sources ofexpectations. Thus, it is the respondents overall impression of a number of sources ofrole messages that is measured most often (Beehr, 1995). Consequently, role conflict isnot included in the present study.

A. Skogstad et al.328

Dow

nloa

ded

by [A

ir Un

iversi

ty] at

21:39

07 M

ay 20

15

However, we cannot rule out the possibility of followers experiences of roleambiguity leading to subsequent increased experiences and reports of laissez-faireleadership. According to the gloomy perception mechanism, stressed employees mayevaluate their work environment ever more negatively, a notion that was supported by afour-phase prospective study by de Lange et al. (2005) showing that so called unhealthystayers reported significantly reduced supervisory support over time in a Dutch cohortstudy. It may therefore be the case that followers experiencing role ambiguity developmore gloomy perceptions of leader behaviour, in the present case expressed by therespondents reports of high levels of laissez-faire leadership. We will therefore test analternative hypothesis (H2) that there is a reciprocal relationship between laissez-faireleadership and role ambiguity over time:

Hypothesis 2. A reciprocal relationship will exist between laissez-faire leadership and roleambiguity in that laissez-faire leadership will be positively related to role ambiguity at thesame time as role ambiguity is positively related to laissez-faire leadership at threeconsecutive measurement points.

In testing our two hypotheses we will control for the potential effect of a leader merelybeing low on a compound variable consisting of both Initiation of Structure andConsideration.

Method

The present study is based on longitudinal survey data from a nationwide sample of theNorwegian workforce. The survey was approved by the Regional Committee for MedicalResearch Ethics in Western Norway. Data were collected at three time points, with a timelag of two years between T1 and T2, and three years between T2 and T3. In 2005 (T1), arandom sample of 4500 employees was drawn from the Norwegian Central EmployeeRegister (NCER) by Statistics Norway (SSB). The sampling criteria were adults between18 and 67 years of age who were registered in the NCER as being employed during thepreceding six months in a Norwegian enterprise with a staff of five or more, working a meanof at least 15 hours per week. Questionnaires were distributed through the Norwegian PostalService. A total of 2539 questionnaires were returned in the first wave (response rate 57%),securing a response rate slightly above the typical response rate in this kind of surveyresearch (Baruch & Holtom, 2008). This sample may be considered to be representative ofthe Norwegian working population, according to the criteria above, when controlling forparticipants age and location (county), but female participants were somewhat betterrepresented in the actual sample (52%) compared to the original sample drawn (47%)(Hstmark & Lagerstrm, 2006). Using the same procedure, the second and third waves ofdata were collected in 2007 (T2) and in 2010 (T3). The study sample consisted ofparticipants responding in waves 1, 2 and 3, or in wave 1 combined with waves 2 or 3,respectively; a total of 1771 employees, representing a response rate of 70% from thefirst wave.

Participants

The mean age of participants in the T1 sample was 43.8 years (SD = 11.5), with a rangefrom 19 to 66 years of age. The majority of the respondents were in full-time (77%) or

Work & Stress 329

Dow

nloa

ded

by [A

ir Un

iversi

ty] at

21:39

07 M

ay 20

15

part-time (13%) employment, while 9% were on sick leave or rehabilitation leave.Disabled pensioners represented 1% of the respondents. The mean number of actualworking hours per week was 37.5 (SD = 10.4). Among the respondents, 15% weremanagers with personnel responsibilities, and 13% were elected union representatives orpersonnel safety representatives. Attrition analysis (t-tests) did not reveal significantdifferences between respondents and non-respondents at the different measurement pointson the reporting of either laissez-faire leadership or role ambiguity.

Measures

Role ambiguity was measured using the scale of Rizzo et al. (1970) and consisted of sixitems, formulated as expressions of role clarity. According to the authors instructions anduse of the scale (see, e.g. a meta-study by Jackson & Schuler, 1985), reversed scores wereused to measure role ambiguity. An example of an item is I know exactly what myresponsibilities are). The scale had seven response categories ranging from very falseto very true (Cronbachs alphas in the range .85.87). In some studies, unreversedscores are used as a measure of role clarity (see, e.g. Hinkin & Schriesheim, 2008a).

Laissez-faire leadership behaviour was measured by four items from the MultifactorLeadership Questionnaire (Bass & Avolio, 2004; Hinkin & Schriesheim, 2008b), anexample being is absent when needed. (Cronbachs alphas are in the range of .68to .71).

Leaders initiating structure and showing consideration for subordinates, representedin this study by a compound indicator Initiating Structure and Consideration, weremeasured by three items from a short version of Ekvall and Arvonens (1991) leadershipquestionnaire. The three items are Sets clear goals, Defines and explains the workrequirements clearly and Shows appreciation for good work. (Cronbachs alphas are inthe range of .73 to .76.).

For all leadership scales, four response categories were employed (never,sometimes, quite often and very often/nearly always), and the respondents wereasked to report on leadership behaviour that they had experienced during the last sixmonths.

Control variables in the present study were gender, age and change of leader betweenfollow-ups. With respect to the latter, two dichotomous variables were constructed,namely, change of leader between 2005 and 2007, and between 2007 and 2010.

Statistical analysis

Our hypotheses were tested by Structural Equation Modelling (SEM) using Mplus 6.0(Muthn & Muthn, 2012). In order to adequately handle non-normality in studyvariables, maximum likelihood with robust standard errors (MLR) was applied. MLR isrobust to violations of non-normality (Brown, 2006). The goodness-of-fit of the SEMmodels was evaluated using a chi-square value, with a non-significant p-value indicatinga good fit (Kelloway, 1998), a root mean square error of approximation (RMSEA) withvalues of .05 or less indicating a good fit (Steiger, 1990) and a comparative fit index(CFI) and Tucker-Lewis Index (TLI) with values of 0.90 or above indicating a good fit(Hu & Bentler, 1999). Comparisons between nested models were performed using theSatorra-Bentler scaled chi-square difference test taking into account the MLR correction

A. Skogstad et al.330

Dow

nloa

ded

by [A

ir Un

iversi

ty] at

21:39

07 M

ay 20

15

factor (Satorra & Bentler, 2001). Missing data were estimated by full informationmaximum likelihood (FIML).

Results

Descriptive statistics

Means, standard deviations and correlations for the study variables are shown in Table 1.Laissez-faire leadership, Initiating Structure and Consideration, and role ambiguity wererepresented by summing the scores across the three measurement occasions.

Multivariate autoregressive cross-lagged models

Prior to running the autoregressive cross-lagged models, a measurement model with thethree latent study constructs (laissez-faire leadership, leaders Initiating Structure andConsideration, and role ambiguity) were examined separately at each measurement point.In the measurement model, the latent factors were modelled using their respectiveobserved indicators, and the latent factors were allowed to covary. The measurementmodel showed good fit to the data at all measurement points (2 = 413.1 (74), CFI = .95,TLI = .93, RMSEA = .05; 2 = 429.0 (74), CFI = .93, TLI = .91, RMSEA = .06; 2 =336.2 (74), CFI = .94, TLI = .92, RMSEA = .06), at measurement points 1, 2 and 3,respectively). In the measurement model, the factor loadings for the latent factormeasuring laissez-faire leadership were in the range .41.75 across the three measurementoccasions, while the factor loadings for leaders Initiating Structure and Considerationand role ambiguity were in the range .55.82, and .51.88, respectively.

To test for possible lagged effects between the two leadership styles (laissez-faire andleaders Initiating Structure and Consideration) and role ambiguity across the threemeasurements, a conditional multivariate autoregressive cross-lagged SEM model wasapplied. In order to prevent inflated estimates of stability (Marsh & Hau, 1996), weallowed the indicators of the latent factors at the different measurements to covariate withthe corresponding indicators at the other measurements. Finally, the latent factors wereallowed to covary within time points. Three nested models were tested. Initially, a cross-lagged model was modelled with lagged effects from laissez-faire leadership on roleambiguity at the following measurement point, controlling for the possible correspondinglagged effect from leaders Initiating Structure and Consideration.

In the model, all dependent variables were controlled by gender, age and change ofleader between measurement points. As shown in Table 2, the initial cross-lagged modelshowed a good fit to the data (CFI = .94, TLI = .93 and RMSEA = .03). In support ofHypothesis 1, significant lagged effects from laissez-faire leadership on role ambiguitywere both found between T1 and T2, and between T2 and T3. However, the correspondinglagged effects between leaders Initiating Structure and Consideration and role ambiguitywere not significant. In the second model, reciprocal lagged effects of role ambiguity on thetwo leadership styles were added to the model. However, the lagged paths from roleambiguity to the two leadership styles were not significant, and adding the paths did notresult in a significant improvement of fit (TRd (4) = 2.86, n.s.). Thus, Hypothesis 2 was notsupported. In the final model, all insignificant paths were removed from the model.Removing insignificant paths from the analysis in the final model did not result insignificant deterioration of fit (TRd (24) = 28.21, n.s.), and the final model showed good fit

Work & Stress 331

Dow

nloa

ded

by [A

ir Un

iversi

ty] at

21:39

07 M

ay 20

15

Table 1. Means, standard deviations and correlations between study variables (N = 1771).

Mean SD 1 2 3 4 5 6 7 8 9 10 11 12

1. Laissez faire T1 0.57 0.52 1.002. Constructive T1 1.47 0.70 .32** 1.003. Role ambiguity T1 2.18 0.93 .45** .31** 1.004. Laissez faire T2 0.54 0.53 .52** .16** .30** 1.005. Constructive T2 1.45 0.69 .19** .45** .17** .32** 1.006. Role ambiguity T2 2.20 0.95 .32** .18** .60** .44** .29** 1.007. Laissez faire T3 0.51 0.49 .39** .09** .22** .39** .11** .26** 1.008. Constructive T3 1.51 0.70 .14** .35** .07* .15** .37** .09** .33** 1.009. Role ambiguity T3 2.21 0.94 .28** .16** .49** .35** .16** .53** .43** .28** 1.0010. Gender 1.55 0.50 .01 .01 .04 .05* .03 .05* .07* .03 .07* 1.0011. Age 45.22 11.29 .01 .05* .09** .04 .07* .11** .04 .09** .13** .07** 1.0012. Change leader

200520070.19 0.39 .02 .05 .03 .02 .04 .04 .03 .05 .01 .01 .05 1.00

13. Change leader20072010

0.46 0.50 .09** .05 .10** .05 .01 .06* .01 .00 .15** .03 .12** .45**

Note: Laissez-faire and Constructive refer to leadership styles, where constructive was operationalized as a combined measure of leaders initiating structure and showingconsideration.*p < .05; **p < .01.

A.Skogstad

etal.

332

D

o

w

n

l

o

a

d

e

d

b

y

[

A

i

r

U

n

i

v

e

r

s

i

t

y

]

a

t

2

1

:

3

9

0

7

M

a

y

2

0

1

5

to the data (CFI = .94, TLI = .93 and RMSEA = .03). Parameter estimates from the finalmodel are shown in Figure 1.

As shown in Figure 1, Stability coefficients in the range .53.65 were found forlaissez-faire leadership, while the corresponding range of coefficients for InitiatingStructure and Consideration, and role ambiguity were .49.56 and .52.59, respectively.The within-measurement correlations between laissez-faire leadership and role ambiguityranged from .57 at the first measurement to .40 at the final measurement. The within-measurement correlations between Initiating Structure and Consideration and roleambiguity, were .36, .33 and .38 at T1, T2 and T3, respectively. The correspondingwithin-measurement correlations between laissez-faire leadership and Initiating Structureand Consideration ranged from .42 at T1 to .49 at T3. Only three significant pathswere found between the control variables and the study constructs. A significant negative

Table 2. Fit indices for conditional multivariate autoregressive cross-lagged models.

Model 2 df CFI TLI RMSEA

Initial cross-lagged model 2224.4 896 .94 .93 .03Reciprocal cross-lagged model 2223.0 892 .94 .93 .03Final cross-lagged model 2252.2 916 .94 .93 .03

Note: CFI, comparative fit index; TLI, Tucker-Lewis Index; RMSEA, root mean square error of approximation.

Laissez faireleadership T1

Laissez faireleadership T2

Laissez faireleadership T3

Roleambiguity T1

Roleambiguity T2

Roleambiguity T3

.65

.52

.59

.45

.38

.56

.09

.11.53

.56

.49

0.05

Change of leader

20072010

Age

Gender

.11

0.07

Change of leader

20052007

0.14

0.45

0.07

0.36

Constructiveleadership T1

Constructive leadership T2

Constructive leadership T3

0.38

0.47

0.42

0.49

0.31

Figure 1. Parameter estimates in nal conditional cross-lagged model. For simplication,observed indicators of latent variables, residuals and residual co-variances are not included.

Work & Stress 333

Dow

nloa

ded

by [A

ir Un

iversi

ty] at

21:39

07 M

ay 20

15

path was found from gender to laissez-faire leadership at T2 ( = .05, p < .05),indicating lower reporting of laissez- faire leadership by female respondents. Correspond-ingly, age was negatively related to role ambiguity at T2 ( = .07, p < .01), and changeof leader between T2 and T2 was positively related to role ambiguity at T3 ( = .11,p < .01). Finally, the lagged paths from laissez faire leadership to role ambiguity were .09(p < .01) from T1 to T2, and .11 (p < .05) from T2 to T3.

Discussion

The effects of laissez-faire leadership on subordinates have rarely been studied. In thisthree-wave study we tested the effect of such leadership on subordinates stress in theform of experiences of role ambiguity.

The results support Hypothesis 1, that laissez-faire leadership is a precursor ofsubsequent role ambiguity, while Hypothesis 2, positing a bidirectional relationship, wasnot supported. Prior exposure to laissez-faire leadership was found to significantlyaccount for subsequent variation in role ambiguity over both time lags investigated. Thepaths from laissez-faire leadership at T1 to role ambiguity at T2, and from laissez-faireleadership at T2 to role ambiguity at T3, were significant when the possible lagged effectof leaders Initiating Structure and Consideration were controlled for. Whereas age andgender at T1 were associated with role ambiguity and laissez-faire leadership at T2, andchange of leader between the first and second follow-up was related to role ambiguity atT3, the relationships between laissez-faire leadership and role ambiguity remainedsignificant across all the three measurement points. It is also interesting to note that thelevels of role ambiguity, instead of increasing, remained relatively stable across the threemeasurement points. There may be further explanations for this pattern of stability.Podsakoff, MacKenzie, and Bommers (1996) meta-analysis showing relatively strongnegative correlations (.49, .37 and .48) between role ambiguity and leadershipsubstitutes in the form of task feedback, intrinsically satisfying tasks and organizationalformalization, indicating that subordinates may find compensation for leaders laissez-faire leadership in a number of ways. In this respect, it is reasonable to assume that moreproximal substitutes, such as support from colleagues and others at work (Viswesvaran,Sanchez, & Fisher, 1999) and the development of relevant coping strategies (Skinneret al., 2003) may, to an even greater extent, help subordinates cope with role ambiguity atwork. With respect to the longitudinal effects of negative social exchanges, represented inour case by a leadership style of laissez-faire avoidance and non-transactions, it may alsobe the case that negative social exchanges have a greater initial impact than do positiveexchanges such as those connected with transactional leadership, but that the effects ofnegative exchanges subside more over time (Rook, 1998).

The present results yield significant support to the assumption that poor leadership inthe form of laissez-faire leadership may be a root cause of workplace stress (Kellowayet al., 2005) through role ambiguity. The findings support Hypothesis 1 that is, theyindicate that the immediate superior is highly influential with respect to clarifyingsubordinates role expectations related to duties, objectives and responsibilities; this is apoint of view that has been stated but not tested by various scholars (e.g. Kellowayet al., 2005). The importance of such work role clarification is emphasized by all thenegative consequences of role ambiguity, which have been richly documented. Leaders andmanagers may act as a significant source of role stress when they do not fulfil basic

A. Skogstad et al.334

Dow

nloa

ded

by [A

ir Un

iversi

ty] at

21:39

07 M

ay 20

15

obligations and requirements in the leadership role, and especially when subordinates findthemselves in need of leadership. According to Rizzo and colleagues (1970) and Beehr andGlazer (2005), the lack of information and clarity of behavioural requirements are essentialin creating role ambiguity. There is reason to believe that role clarity and its opposite, roleambiguity may be influenced by various persons in a given role in the work arena).However, according to prevailing theoretical perspectives, such as the Full Range ofLeadership model, path goal theory and Leader-Member Exchange Theory, as well asaccording to empirical findings from meta-studies, there is reason to believe that the actions and the avoidance of actions of leaders are most influential with respect to subordinatesrole ambiguity.

The present longitudinal study, consisting of three waves, confirms theoretical modelsas well as findings from earlier cross-sectional studies, and provides support for therebeing a significant influence of immediate superiors laissez-faire leadership behaviour onsubordinates role ambiguity.

This study also supports the notion that laissez-faire leadership is a unique type ofleadership by being a significant predictor of role ambiguity over time, while by contrastthe classical constructive leadership styles of leaders Initiating Structure and showingConsideration (as a compound variable in the present study), appeared to have aninsignificant influence on role ambiguity. The avoidance of leadership responsibilities inleader-follower relationships, in the form of laissez-faire leadership, may thus be decisivein explaining subordinates experience of work-related stress. In line with the bad isstronger than good assumption, which has received strong support in a variety of humaninteraction arenas (Baumeister, Bratslavsky, Finkenauer, & Vohs, 2001), there are strongarguments for leadership researchers focusing primarily on the negative aspects ofleadership behaviour, such as laissez-faire leadership, especially when studying leadershipwithin a stressor-strain framework. Moreover, when role stress is studied as a mediator inthe leader behaviour-outcome chain (see Moyle, 1998; Nielsen et al., 2008), laissez-faireleadership is a strong candidate for inclusion. Leadership scholars may therefore benefitfrom including laissez-faire leadership in their studies more often. For example, whenstudying the unique effects of various forms of leadership on subordinate attitudes,behaviour and health in times of context uncertainty, laissez-faire leadership may be ofprimary interest (see, e.g. Balogun, 2003).

Study strengths and limitations

Whereas this study, in line with the majority of leadership studies, has its limitations, italso has notable strengths. The employment of a longitudinal design with threemeasurement points makes it possible to substantiate a plausible causal direction betweenlaissez-faire leadership and role ambiguity. As such, the present study contributesimportant knowledge to the leadership field, which, to our knowledge, has only examinedthis relationship by employing cross-sectional studies. Furthermore, the sample in thepresent study is valuable as it is systematically drawn from the Norwegian workforce(Hstmark & Lagerstrm, 2006), and, as such, reduces the probability of the results beinginfluenced by factors that may occur in a convenience sample.

Despite its strengths, the present study also has its limitations. Even though the three-wave study provides strong support for laissez-faire leadership being a predictor of roleambiguity, and not the other way around, the longitudinal design employed does not fully

Work & Stress 335

Dow

nloa

ded

by [A

ir Un

iversi

ty] at

21:39

07 M

ay 20

15

prove a causal relationship as such (see, e.g. Antonakis, Bendaham, Jacquart, & Lalive,2010). Furthermore, the predictor and the criterion variables were both collected fromsubordinates. Thus, the present study is based on subordinates rating both their perceptionof their immediate superiors behaviour and their own role ambiguity. However,employing a longitudinal design will minimize the likelihood of the relationship betweenthe two study variables being strictly artefactual. Nevertheless, alternative observations ofleadership behaviour, e.g. from the immediate leaders superior, as well as restricting thestudy to the behaviours of only the same immediate superiors across all threemeasurement points, may have strengthened the validity of the study. However, becausea majority of the respondents (944 of 1771) changed their immediate leaders from T1 toT2 and/or from T2 to T3, only including those of the respondents having the same leaderacross the three measurements would substantially reduce the statistical power in theanalysis. Furthermore, as mentioned above, a major strength of the present study is theuse of a systematically drawn national sample. By including less than half of the sample,the generalizability of the study would be weakened. Accordingly, we chose to control therespective dependent variables at T2 and T3 for the effects of changing leader from T1 toT2 and/or from T2 to T3.

This study employed an autoregressive cross-lagged model. Growth curve modellingcould alternatively have been used (see, e.g. Curran & Bollen, 2002). With respect to timelags, the study used two and three years, respectively. Although such time intervals maybolster confidence in the results, it may be that subordinates experiences of theimmediate superiors laissez-faire leadership as well as their own role ambiguity vary overtime. Future studies of the relationships between leadership behaviour and role stressshould therefore also test alternative time intervals to explore how long it takes forlaissez-faire leadership to significantly influence role ambiguity (see, e.g. Moyle 1998,where six-month intervals were used when testing relationships between perceptions ofmanagerial support, role clarity and job satisfaction).

In general, in order for the present study to be comparable to the majority of recentleadership studies, it may have benefited from including transactional and transforma-tional forms of leadership as measured by the MLQ, even though classical types ofleadership such as Initiating Structure and Consideration may be more relevant in thepresent study (see also Yukl (1999) for a discussion of the limitations of the MLQ). Thereis a strong need for theoretically-driven studies on this type of leadership, with respect toits precursors and outcomes as well as moderating and mediating factors. It is thereforehighly important within a stressor-strain framework to gain nuanced knowledge aboutwhat organizational and individual factors influence the impact of laissez-faire leadershipon stress outcomes. In this respect, leader-follower structural, functional and psycholo-gical distance (Antonakis & Atwater, 2002), and leader and follower competenceuncertainty (Mayer, Thau, Workman, Van Dijke, & Cremer, 2012) are some interestingcandidates worth investigating in future studies. Furthermore, in future studies of avoidantand non-leadership forms of leadership, it will also be highly relevant to employalternative approaches and measures. For instance, in their study of non-leadership,Hinkin and Schriesheim (2008a) investigated the effects of leader reward omission andpunishment omission on good and poor subordinate performance. This alternativeapproach illustrates that the domain of avoidant and non-leadership behaviours isprobably populated with a variety of leader-avoidant and non-leader behaviours.

A. Skogstad et al.336

Dow

nloa

ded

by [A

ir Un

iversi

ty] at

21:39

07 M

ay 20

15

Conclusion

The study was conducted for theoretical, empirical, methodological and applied reasons.Firstly, we believe that it makes an important contribution to our understanding of laissez-faire leadership as an independent type of leadership, within a stressor-strain framework,compared to effectiveness and satisfaction frameworks (rtqvist & Wincent, 2006). It alsoprovides valuable empirical findings by testing the theoretical proposition that laissez-faireleadership is a root cause of workplace role stress, employing a three-wave longitudinaldesign. By demonstrating an exclusive effect of laissez-faire leadership on role ambiguity,and no effect of constructive leadership in the form of Initiating Structure and Considerationas a control, the study provides strong evidence for laissez-faire leadership being a highlyrelevant and influential type of leadership worth studying in itself.

Our findings indicate that laissez-faire leadership may be more important forsubordinates than is reflected in the majority of leadership studies, which emphasizetransformational and transactional forms of leadership and their relationships with attitudeand effectiveness correlates. Bass and Avolio (1994) stated that laissez-faire leadership isthe least effective type of leadership. However, with respect to the effects on subordinatesworking conditions and experience of stress, the opposite may also be true; laissez-faireleadership is highly effective in a negative sense, i.e. in creating a stressful workingenvironment. In line with the Bad is stronger than good assumption (Baumeister et al.,2001), the employment of laissez-faire leadership may in various circumstances be just asimportant, or even more important, for follower attitudes and behavioural outcomes as theemployment of transactional and transformational forms of leadership.

From an applied perspective, our findings show that leaders in general should avoidlaissez-faire leadership, which from a follower-centred perspective may be defined as notmeeting the legitimate expectations of the subordinates (Skogstad et al., 2007, p. 81). Thisis probably especially true when subordinates are in need of leadership. Organizations ingeneral, as well as leadership development programmes in particular, should thus addresslaissez-faire leadership behaviour and its negative effects just as much as they dotransactional and transformational forms of leadership. It may be of the utmost importancefor superiors to know when to act and not to act in their relationships with subordinates,and thus not confuse empowering leadership with laissez-faire leadership behaviours. Inline with the present results, superiors consistently not taking action when subordinates arein need of leadership can probably not therefore be perceived as a zero-type of leadership,but more as destructive leaders maintaining and even escalating the experience of a stressfulwork environment, with potentially very strong negative effects on employees satisfactionand effectiveness.

Acknowledgements

This is a collaborative project between the University of Bergen and Statistics Norway, whichcollected the data. We would like to thank Bengt Oscar Lagerstrm and Maria Hstmark ofStatistics Norway and Stig Berge Matthiesen of the Faculty of Psychology, University of Bergen,for their contribution to the data collection.

Work & Stress 337

Dow

nloa

ded

by [A

ir Un

iversi

ty] at

21:39

07 M

ay 20

15

Funding

The project was made possible by joint grants from two Norwegian employer associations (theConfederation of Norwegian Enterprise and the Norwegian Association of Local and RegionalAuthorities), and the Norwegian government (the National Insurance Administration) and theirFARVE programme.

ReferencesAasland, M. S. (2012). Destructive leadership: Conceptualization, measurement, prevalence andoutcomes (Doctoral thesis). University of Bergen, Bergen.

Aasland, M. S., Skogstad, A., Notelaers, G., Nielsen, M. B., & Einarsen, S. (2010). The prevalenceof destructive leadership behaviour. British Journal of Management, 21, 438452.

Antonakis, J., & Atwater, L. (2002). Leader distance: A review and proposed theory. TheLeadership Quarterly, 13, 673704. doi:10.1016/S1048-9843(02)00155-8

Antonakis, J., Bendaham, S., Jacquart, P., & Lalive, R. (2010). On making causal claims: A reviewand recommendations. The Leadership Quarterly, 21, 10861120. doi:10.1016/j.leaqua.2010.10.010

Bakker, A. B., & Demerouti, E. (2007). The job demands resources model: State of the art. Journalof Managerial Psychology, 22, 309328. doi:10.1108/02683940710733115

Balogun, J. (2003). From blaming the middle to harnessing its potential: Creating changeintermediaries. British Journal of Management, 14(1), 6983. doi:10.1111/1467-8551.00266

Bandura, A. (1997). Self-efficacy: The exercise of control. New York, NY: Freeman.Baruch, Y., & Holtom, B. C. (2008). Survey response rate levels and trends in organizationalresearch. Human Relations, 61, 11391160. doi:10.1177/0018726708094863

Bass, B. M. (1990). Bass and Stogdills handbook of leadership: Theory, research and managerialapplications (Vol. 3). New York, NY: Free Press.

Bass, B. M., & Avolio, B. J. (1994). Improving organizational effectiveness through transforma-tional leadership. Thousand Oaks, CA: Sage.

Bass, B. M., & Avolio, B. J. (2004). Multifactor leadership questionnaire (3rd ed.). Manual andSampler Set. Menlo Park, CA: Mind Garden.

Bass, B. M., & Riggio, R. E. (2006). Transformational leadership. Mahwah, NJ: LawrenceErlbaum Associates.

Baumeister, R. F., Bratslavsky, E., Finkenauer, C., & Vohs, K. D. (2001). Bad is stronger than good.Review of General Psychology, 5, 323370. doi:10.1037/1089-2680.5.4.323

Beehr, T. A. (1995). Psychological stress in the workplace. London: Routledge.Beehr, T. A., & Glazer, S. (2005). Organizational role stress. In J. Barling, E. K. Kelloway, &M. R. Frone (Eds.), Handbook of work stress (pp. 733). Thousand Oaks, CA: Sage.

Brown, T. A. (2006). Confirmatory factor analysis for applied research. New York, NY: GuilfordPress.

Chen, G., & Bliese, P. D. (2002). The role of different levels of leadership in predicting self- andcollective efficacy: Evidence of discontinuity. Journal of Applied Psychology, 87, 549556.doi:10.1037/0021-9010.87.3.549

Curran, P. J., & Bollen, K. A. (2002). The best of both worlds. Combining autoregressive and latentcurve models. In L. M. Collins & A. G. Sayer (Eds.), New methods for the analysis of change(pp. 105136). Washington, DC: American Psychological Association.

Dale, K., & Fox, M. L. (2008). Leadership style and organizational commitment: Mediating effectof role stress. Journal of Managerial Issues, 20(1), 109130.

de Lange, A. H., Taris, T. W., Kompier, M. A. J., Houtman, I. L. D, & Bongers, P. M. (2003). Thevery best of the millennium: Longitudinal research and the demand-control-(support) model.Journal of Occupational Health Psychology, 8, 282305. doi:10.1037/1076-8998.8.4.282

de Lange, A. H., Taris, T.W., Kompier, M. A. J., Houtman, I. L. D., & Bongers, P. M. (2005).Different mechanisms to explain the reversed effects of mental health on work characteristics.Scandinavian Journal of Work, Environment & Health, 31(1), 314. doi:10.5271/sjweh.843

A. Skogstad et al.338

Dow

nloa

ded

by [A

ir Un

iversi

ty] at

21:39

07 M

ay 20

15

de Vries, R. E., Roe, R. A., & Taillieu, T. C. B. (2002). Need for leadership as a moderator of therelationships between leadership and individual outcomes. The Leadership Quarterly, 13(2),121137. doi:10.1016/S1048-9843(02)00097-8

Einarsen, S., Skogstad, A., & Aasland, M. S. (2010). The nature, prevalence and outcomes ofdestructive leadership: A behavioral and conglomerate approach. In B. Schyns & T. Hansbrough(Eds.), When leadership goes wrong. Destructive leadership, mistakes and ethical failures(pp. 145171). Charlotte, NC: Information Age.

Ekvall, G., & Arvonen, J. (1991). Change-centered leadership: An extension of the two-dimensionalmodel. Scandinavian Journal of Management, 7(1), 1726. doi:10.1016/0956-5221(91)90024-U

Fisher, C., & Gitelson, R. (1983). A meta-analysis of the correlates of role conflict and ambiguity.Journal of Applied Psychology, 68, 320333. doi:10.1037/0021-9010.68.2.320

Gilboa, S., Shirom, A., Fried, Y., & Cooper, C. (2008). A meta-analysis of work demand stressorsand job performance: Examining main and moderating effects. Personnel Psychology, 61,227271. doi:10.1111/j.1744-6570.2008.00113.x

Hauge, L. J., Skogstad, A., & Einarsen, S. (2007). Relationships between stressful workenvironments and bullying: Results of a large representative study. Work & Stress, 21,220242. doi:10.1080/02678370701705810

Hetland, H., Sandal, G. M., & Johnsen, T. B. (2007). Burnout in the information technology sector:Does leadership matter? European Journal of Work and Organizational Leadership, 16(1),5875. doi:10.1080/13594320601084558

Higgins, E. T. (1997). Beyond pleasure and pain. American Psychologist, 52, 12801300.doi:10.1037/0003-066X.52.12.1280

Hinkin, T. R., & Schriesheim, C. A. (2008a). An examination of nonleadership: From laissez-faire leadership to leader reward omission and punishment omission. Journal of AppliedPsychology, 93, 12341248. doi:10.1037/a0012875

Hinkin, T. R., & Schriesheim, C. A. (2008b). A theoretical and empirical examination of thetransactional and non-leadership dimensions of the Multifactor Leadership Questionnaire (MLQ).The Leadership Quarterly, 19, 501513. doi:10.1016/j.leaqua.2008.07.001

House, R. J. (1996). Path-goal theory of leadership: Lessons, legacy, and a reformulated theory.The Leadership Quarterly, 7, 323352. doi:10.1016/S1048-9843(96)90024-7

Hstmark, M., & Lagerstrm, B. O. (2006). Underskelse om arbeidsmilj: destruktiv atferd iarbeidslivet: Destruktiv atferd i arbeidslivet [A study of work environment: Destructivebehaviours in working life]. Dokumentasjonsrapport [Documentation Report]. Oslo: Statistisksentralbyr [Statistics Norway].

Hu, L. T., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis:Conventional criteria versus new alternatives. Structural Equation Modeling, 6(1), 155.doi:10.1080/10705519909540118

Jackson, S. E., & Schuler, R. S. (1985). A meta-analysis and conceptual critique of research on roleambiguity and role conflict in work settings. Organizational Behavior and Human DecisionProcesses, 36(1), 1678. doi:10.1016/0749-5978(85)90020-2

Jackson, S. E., Schwab, R. L., & Schuler, R. S. (1986). Toward an understanding of the burnoutphenomenon. Journal of Applied Psychology, 71, 630640. doi:10.1037/0021-9010.71.4.630

Judge, T. A., & Piccolo, R. F. (2004). Transformational and transactional leadership: A meta-analytic test of their relative validity. Journal of Applied Psychology, 89, 755768. doi:10.1037/0021-9010.89.5.755

Judge, T. A., Piccolo, R. F., & Ilies, R. (2004). The forgotten ones? The validity of considerationand initiating structure in leadership research. Journal of Applied Psychology, 89(1), 3651.doi:10.1037/0021-9010.89.1.36

Kahn, R. L., Wolfe, D. M., Quinn, R. P., Snoek, J. D., & Rosenthal, R. A. (1964). Organizationalstress: Studies in role conflict and ambiguity. Oxford: John Wiley.

Karasek, R., & Theorell, T. (1990). Healthy work: Stress, productivity and the reconstruction ofworking life. New York, NY: Basic Books.

Kelloway, E. K. (1998). Using LISREL for structural equation modeling: A researchers guide.Thousand Oaks, CA: Sage.

Kelloway, E. K., Sivanathan, N., Francis, L., & Barling, J. (2005). Poor leadership. In J. Barling,E. K. Kelloway, & M. R. Frone (Eds.), Handbook of work stress (pp. 89112). Thousand Oaks,CA: Sage.

Work & Stress 339

Dow

nloa

ded

by [A

ir Un

iversi

ty] at

21:39

07 M

ay 20

15

Lapidot, Y., Kark, R., & Shamir, B. (2007). The impact of situational vulnerability on thedevelopment and erosion of followers trust in their leader. The Leadership Quarterly, 18(1),1634. doi:10.1016/j.leaqua.2006.11.004

Lazarus, R. S. (1999). Stress and emotion. A new synthesis. New York, NY: Springer.Lewin, K. (1936). Principles of topological psychology. New York, NY: McGraw-Hill.Lewin, K., Lippit, R., & White, R. K. (1939). Patterns of aggressive behavior in experimentallycreated social climates. The Journal of Social Psychology, 10, 269299. doi:10.1080/00224545.1939.9713366

MacKenzie, S. B., Podsakoff, P. H., & Rich, G. A. (2001). Transformational and transactionalleadership and salesperson performance. Journal of the Academy of Marketing Science, 29(2),115134. doi:10.1177/03079459994506

Marsh, H. W., & Hau, K.-T. (1996). Assessing goodness of fit: Is parsimony always desirable?Journal of Experimental Education, 64, 364390. doi:10.1080/00220973.1996.10806604

Mayer, D. M., Thau, S., Workman, K. M., Van Dijke, M., & Cremer, D. D. (2012). Leadermistreatment, employee hostility, and deviant behaviors: Integrating self-uncertainty and thwartedneeds perspectives on deviance. Organizational Behavior and Human Decision Processes,117(1), 2440. doi:10.1016/j.obhdp.2011.07.003

Moyle, P. (1998). Longitudinal influences of managerial support on employee well-being. Work &Stress, 12(1), 2949. doi:10.1080/02678379808256847

Muthn, L. K., & Muthn, B. O. (2012). Mplus 6.12. Los Angeles, CA: Muthen & Muthen.Nielsen, K., Randall, R., Yarker, Y., & Brenner, S. O. (2008). The effects of transformationalleadership on followers perceived work characteristics and psychological well-being:A longitudinal study. Work & Stress, 22(1), 1632. doi:10.1080/02678370801979430

ODriscoll, M. P., & Beehr, T. A. (1994). Supervisor behaviors, role stressors and uncertainty aspredictors of personal outcomes for subordinates. Journal of Organizational Behavior, 15(2),141155. doi:10.1002/job.4030150204

rtqvist, D., & Wincent, J. (2006). Prominent consequences of role stress: A meta-analytic review.International Journal of Stress Management, 13, 399422. doi:10.1037/1072-5245.13.4.399

Piccolo, R. F., Bono, J. E., Heinitz, K., Rowold, J., Duehr, E., & Judge, T. A. (2012). The relativeimpact of complementary leader behaviors: Which matters most? The Leadership Quarterly, 23,567581. doi:10.1016/j.leaqua.2011.12.008

Podsakoff, P. M., Bommer, W. H., Podsakoff, N. P., & MacKenzie, S. B. (2006). Relationshipsbetween leader reward and punishment behavior and subordinate attitudes, perceptions, andbehaviors: A meta-analytic review of existing and new research. Organizational Behavior andHuman Decision Processes, 99(2), 113142. doi:10.1016/j.obhdp.2005.09.002

Podsakoff, P. H., MacKenzie, S. B., & Bommer, W. H. (1996). Meta-analysis of the relationshipsbetween Kerr and Jermiers substitutes for leadership and employee job attitudes, roleperceptions, and performance. Journal of Applied Psychology, 81, 380399. doi:10.1037/0021-9010.81.4.380

Reeve, J. (2005). Understanding motivation and emotion. Hoboken: John Wiley & Sons.Rizzo, J. R., House, R. J., & Lirtzman, S. I. (1970). Role conflict and ambiguity in complexorganizations. Administration Sciences Quarterly, 15, 150163. doi:10.2307/2391486

Rook, K. S. (1998). Investigating the positive and negative sides of personal relationships: Througha lens darkly? In B. H. Spitzberg & W. R. Cupach (Eds.), The dark side of close relationships(pp. 369393). Mahwah, NJ: Lawrence Erlbaum.

Satorra, A., & Bentler, P. M. (2001). A scaled difference chi-square test statistic for momentstructure analysis. Psychometrica, 66, 507514. doi:10.1007/BF02296192

Schaubroeck, J., Ganster, D. C., Sime, W. E., & Ditman, D. (1993). A field experiment testingsupervisory role clarification. Personnel Psychology, 46(1), 125. doi:10.1111/j.1744-6570.1993.tb00865.x

Schriesheim, C. A., House, R. J., & Kerr, S. (1976). Leader initiating structure: A reconciliation ofdiscrepant research results and some empirical tests. Organizational Behavior and HumanPerformance, 15, 297321. doi:10.1016/0030-5073(76)90043-X

Shamir, B. (2007). From passive recipients to active co-producers. In B. Shamir, R. Pillai, M. C.Bligh, & M. Uhl-Bien (Eds.), Follower-centered perspectives on leadership. A tribute to thememory of James R. Meindl (pp. ixxxxix). Greenwich, CT: Information Age.

A. Skogstad et al.340

Dow

nloa

ded

by [A

ir Un

iversi

ty] at

21:39

07 M

ay 20

15

Skakon, J., Nielsen, K., Borg, V., & Guzman, J. (2010). Are leaders well-being, behaviours andstyle associated with the affective well-being of their employees? A systematic review of threedecades of research. Work & Stress, 24(2), 107139. doi:10.1080/02678373.2010.495262

Skinner, E. A., Edge, K., Altman, J., & Sherwood, H. (2003). Searching for the structure of coping:A review and critique of category systems for classifying ways of coping. Psychological Bulletin,129(2), 216269. doi:10.1037/0033-2909.129.2.216

Skogstad, A., Einarsen, S., Torsheim, T., Aasland, M. S., & Hetland, H. (2007). The destructivenessof laissez-faire leadership behavior. Journal of Occupational Health Psychology, 12(1), 8092.doi:10.1037/1076-8998.12.1.80

Skogstad, A., Notelaers, G., & Einarsen, S. (2009, May 1316). Exposure to destructive leadership:Relationships with job satisfaction, work-withdrawal and intentions to leave. Paper presented atthe 14th European Congress on Work and Organizational Psychology, Santiago de Compostela,Spain.

Steiger, J. H. (1990). Structural model evaluation and modification: An interval estimationapproach. Multivariate Behavioral Research, 25, 173180. doi:10.1207/s15327906mbr2502_4

Stordeur, S., DHoore, W., & Vandenberghe, C. (2001). Leadership, organizational stress, andemotional exhaustion among hospital nursing staff. Journal of Advanced Nursing, 35, 533542.doi:10.1046/j.1365-2648.2001.01885.x

Viswesvaran, C., Sanchez, J. I., & Fisher, J. (1999). The role of social support in the process ofwork stress: A meta-analysis. Journal of Vocational Behavior, 54, 314334. doi:10.1006/jvbe.1998.1661

Yukl, G. (1999). An evaluation of conceptual weaknesses in transformational and charismaticleadership theories. The Leadership Quarterly, 10, 285305. doi:10.1016/S1048-9843(99)00013-2

Work & Stress 341

Dow

nloa

ded

by [A

ir Un

iversi

ty] at

21:39

07 M

ay 20

15

AbstractIntroductionMethodParticipantsMeasuresStatistical analysis

ResultsDescriptive statisticsMultivariate autoregressive cross-lagged models

DiscussionStudy strengths and limitations

ConclusionAcknowledgementsFundingReferences