€¦ · Web viewWithin the existing leadership literature, the role of context for shaping the effectiveness of leadership is yet to be fully understood. One type of context that

Running head: CONTEXTUALIZING LEADERSHIP

Contextualizing leadership: Transformational leadership and Management-By-Exception-Active in safety-critical contexts

Sara Willis*, Sharon Clarke, and Elinor O’Connor Alliance Manchester Business School, The University of Manchester, UK

*Correspondence should be addressed to Sara Willis, Alliance Manchester Business School, Booth Street West, Manchester, M15 6PB, UK (e-mail: [email protected]).

Accepted for publication in Journal of Occupational and Organizational Psychology on 13 February 2017

This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record.Please cite this article as Willis, Clarke, O’Connor (in press). Contextualizing leadership: Transformational leadership and Management-By-Exception-Active in safety-critical contexts. Journal of Occupational and Organizational Psychology. doi: 10.1111/joop.12172

This article is protected by copyright. All rights reserved.

mailto:[email protected])

CONTEXTUALIZING LEADERSHIP

2

Contextualizing Leadership: Transformational leadership and Management-By-Exception-

Active in safety-critical contexts

Abstract

Within the existing leadership literature, the role of context for shaping the effectiveness of

leadership is yet to be fully understood. One type of context that poses particular challenges

for leaders is an environment where safety is highly critical (i.e., high exposure to risk and

likelihood of an accident). We hypothesize that such environments call for specific

transformational and transactional leadership behaviors, which differ from those behaviors

most effective in less safety-critical contexts. We tested for moderating effects of perceptions

of hazard exposure and accident likelihood on the relationship between transformational

leadership and Management-By-Exception-Active with safety and job performance

outcomes. The moderation effects of accident likelihood on the link between transformational

/ MBEA leadership and subordinate performance were supported, demonstrating variation in

the effectiveness of leader behaviors dependent on followers’ perceptions about the

likelihood for an accident. MBEA leadership was found to be more strongly linked to

contextual performance and safety participation if accident likelihood was high, but not under

low accident likelihood conditions. Transformational leadership was found to be less strongly

related to these performance outcomes in contexts where safety was perceived as highly

critical. Our findings have important theoretical and practical implications, and call into

question the universality of the transformational-transactional leadership framework.

Practical considerations focus on the implications for managers and supervisors who operate

in safety-critical contexts.

Keywords: Transformational Leadership; Transactional Leadership; Management-By-

Exception-Active; Occupational Safety.


3

Practitioner Points

Safety-critical contexts pose particular challenges to leaders. If safety is perceived as

highly critical, leaders and/or followers may hold different expectations about

leadership and different leadership styles could be required compared to contexts

where safety is not critical.

Perceived effectiveness of transformational leadership and Management-By-

Exception-Active for employees’ safety participation and contextual performance is

influenced by employees’ perceptions of the risk for an accident within their work

context.

Management-By-Exception-Active is effective for enhancing team members’ extra

effort for safety and contextual performance if the perceived risk of an accident is

high, but less effective if perceptions of accident likelihood are low.

Managers and supervisors should therefore pay attention to employees’ perceptions of

risk of an accident and the factors that determine how employees perceive their

context.

Contextualizing Leadership: Transformational Leadership and Management-By-

Exception-Active in Safety-Critical Contexts


4

The transformational-transactional leadership framework (Bass, 1985) continues to

dominate the leadership literature and studies have linked transformational leadership to a

range of outcome variables, including workplace safety (e.g., Barling, Loughlin, & Kelloway,

2002; Clarke, 2013; Mullen & Kelloway, 2009; O’Dea & Flin, 2001). However, criticism has

been growing that current leadership thinking needs to consider further the context in which

the leadership process takes place (Antonakis, Avolio, & Sivasubramaniam, 2003; Liden &

Antonakis, 2009; Rowold, 2011; Schriesheim, Wu, & Scandura, 2009). Leadership styles that

are valued and effective in one work context might be less desired and less effective in

another context (Antonakis et al., 2003). Concerns have been raised that an approach that

advocates a universal way to lead oversimplifies the leadership process (Rowold, 2011).

While we recognize that stable personality characteristics (Bono & Judge, 2004; Lim &

Ployhart, 2004) and development of a personal style of leading (Johnson, Venus, Lanaj, Mao,

& Chang, 2012) will bring a degree of consistency, there is evidence to suggest that

leadership behaviors are adapted to the context (Denison, Hooijberg, & Quinn, 1995;

Hannah, Balthazard, Waldman, Jennings, & Thatcher, 2013; Vroom & Jago, 2007), and that

followers’ expectations about leadership vary in dependence of the context (Lord & Emrich,

2001; Smothers, Bing, White, Troccchia, & Absher, 2011).

The present study contributes to the occupational safety leadership literature by

investigating whether the degree to which safety is perceived as critical within a work


5

environment impacts on the effectiveness of transformational leadership (TFL) and

Management-By-Exception-Active (MBEA) as a dimension of transactional leadership. TFL

is concerned with inspiring followers, providing a positive vision and challenging existing

standards, and a leader practicing MBEA vigilantly monitors performance and corrects

irregularities as soon as these occur (Bass, 1985; Bass & Riggio, 2006). The present study

also makes contributions to the wider leadership literature by examining leadership in a

prevention-focused setting, by exploring contextual boundaries of transformational-

transactional leadership and by combining behavioral and contingency approaches to

leadership.

Contexts where safety is highly critical, such as oil and gas installations, mining or

construction sites, are inherently hazardous and errors, deviations from standards and other

failures can have severe consequences. In addition, safety-critical contexts, have been

described as ambiguous, given that safety often competes with other goals, such as cost

(Zohar, 2010). Yet, our understanding of leadership in contexts where the risk for accidents is

pervasive and employees face dangerous hazards, is less developed than that of leadership in

more common workplace contexts were safety is not a strong concern (Baran & Scott, 2010).

Unstable, high-risk situations engender particular expectations about leadership (Lord &

Emrich, 2001) and require emphasis on particular actions in the leadership process (Baron &

Scott, 2010). Results by Antonakis et al. (2003) for example indicate that the relationships


6

between dimensions of the transformational-transactional leadership model vary when

comparing samples from higher and lower risk environments. Thus, contexts where safety is

highly critical might trigger different expectations about leadership amongst followers and/or

can bring about an adjustment in leaders’ behavior. We propose that in safety-critical

conditions, leader practices such as correcting errors and active monitoring of safety behavior

(MBEA) will be more strongly associated with performance whilst inspirational leader

behaviors (TFL) will be less strongly related to performance. We use the term safety-critical

to refer to employees’ perceptions of the level of exposure to hazards and the degree of

likelihood for experiencing an accident in their work environment. Accident likelihood refers

to more subjective perceptions about the risk to be involved in an accident, and the possibility

to be injured at work. Hazard exposure relates to perceptions about the physical presence of

hazards in one’s work environment. By choosing these two characteristics, the level of

hazards present within a work environment as well as the degree of risk for an accident

stemming from these, which can differ depending on how hazards are handled, are

represented.

We empirically test the proposition that perceptions of accident likelihood and hazard

exposure will moderate the effectiveness of TFL and MBEA using leaders’ ratings of

subordinate job performance (i.e., task performance, and contextual performance) and safety

performance (i.e., safety compliance and safety participation). Figure 1 illustrates a summary


7

of the study’s hypotheses and the conceptual integration of contingency and behavioral

leadership perspective.

Theory Integration: Merging Contingency and Behavioral Leadership

Conceptualizations

DeRue, Nahrgang, Wellman, and Humphrey (2011) warn that proliferation of

leadership theories has hampered research in developing an understanding of leadership that

matches the complexities of reality where different approaches to leadership, such as

contingency and behavioral elements, take effect concurrently.

We borrow tenets from substitutes for leadership theory (Kerr & Jermier, 1978) and

integrate these with the transformational-transactional leadership framework. Kerr and

Jermier (1978; Howell, Dorfman, & Kerr, 1986) suggested that certain conditions diminish or

amplify the effect of leadership behaviors. It was suggested that empirical evidence for

traditional contingency theories (e.g., Fiedler, 1964; Kerr & Jermier, 1978) failed to flourish

because the theories’ propositions were too rigid, disorganized and centered on contingency

factors that were too abstract and lacking in real-world value (Arvonen & Ekvall, 1999; Villa,

Howell, Dorfman, & Daniel, 2003). In addition, whilst traditional contingency frameworks

identify context factors, the leader behavior categories of substitutes for leadership theory

have been criticized as too crude and incomprehensive (Yukl, 2010). To overcome some of

the above shortcomings of traditional contingency theories, we examine perceptions of


8

hazard exposure and accident likelihood as contingency factors that have applied value and

test these as moderators of transformational-transactional leadership as a framework with

established behavioral dimensions.

Leadership in Safety-Critical Contexts

There is considerable evidence that leaders play a pivotal role in shaping workplace

safety and can influence employee safety behaviors and safety attitudes (Barling et al., 2002

Clarke, 2013, 2010; Mullen & Kelloway, 2009; Zohar, 2002), which in turn relate to bottom-

line safety criteria such as accident and incident rates (Clarke, 2013; Kelloway, Mullen, &

Francis, 2006; Zohar, 2002). However, the type of leadership behaviors that are effective in

contexts where safety is perceived as critical might differ from leadership practices that are

effective if perceptions of risk and hazards are lower.

Work context refers to the ‘backdrop’ that entails a variety of moderating factors that

impact on job role requirements and work behaviors (Dierdorff, Rubin, & Morgeson, 2009).

Johns (2006, p. 386) defines context as “situational opportunities and constraints that affect

the occurrence and meaning of organizational behavior as well as functional relationships

between variables.” It has been shown that contextual factors can influence cognitive

tendencies such as regulatory-focus orientation (Kark & van Dijk, 2007). Regulatory focus

theory (Crowe & Higgins, 1997) postulates that actions are guided by a prevention focus,

which is oriented towards avoidance of loss, or a promotion focus, which regulates behavior


9

towards gain and accretion. Leaders’ behavior is influenced through their inherent regulatory

focus and their situational regulatory focus, with the former being relatively stable, but the

latter being altered by contextual cues (Crowe & Higgins, 1997; Kark & van Dijk, 2007).

Thus, even if a leader’s inherent regulatory focus is more promotion-oriented, contextual

factors can elicit this to shift towards a prevention-oriented focus.

The heightened importance of reliability and avoidance of failure in highly safety-

critical contexts, might act as a trigger to activate a more prevention focus compared to less

safety-critical contexts. In support of this argument, Wallace, Johnson, and Frazier (2009)

have linked a prevention regulatory focus to increased safety performance. In contexts where

errors do not have the potential for catastrophic outcomes, the focus might be more

persistently on promoting gain and a prevention focus might be less activated (Rodriguez &

Griffin, 2009). Moreover, Wallace et al.’s (2009) results showed that in a model where

workplace-situational regulatory focus and inherent regulatory focus were entered

simultaneously, only workplace-situational focus was significantly related to safety

performance. This indicates that regulatory focus orientations that were prompted through

contextual factors such as work surroundings are most relevant in predicting behavior

(Wallace et al., 2009).

Transformational leadership has been associated with a promotion focus and

transactional leadership with a prevention focus (Kark & van Dijk, 2007). If promotion and


10

prevention regulatory focus are differently balanced depending on how critical safety is

perceived within a work context, then this implies that the effectiveness of transformational

and transactional leadership will vary accordingly. Based on the foregoing conceptualization

that work context influences regulatory focus, our research investigates whether the level to

which safety is perceived as critical is a context factor that impacts on the relationship

between leadership style and employee safety and job performance.

Call for Context: Transactional Leadership in Safety-Critical Contexts

Transactional leadership has received considerably less attention from researchers

than transformational, with studies that have examined the effectiveness of transactional

leadership producing mixed evidence. In a meta-analytic review, Judge and Piccolo (2004)

note that positive, neutral as well as negative relationships between transactional leadership

and leader effectiveness criteria have been reported. They suggest that these inconsistencies

on the effectiveness of transactional leadership could be explained by context differences. For

example, Antonakis et al. (2003) argued that MBEA is unnecessarily controlling in some

contexts, whilst in other contexts such practices are beneficial for performance (Antonakis et

al., 2003). If a prevention-oriented focus is activated in contexts where safety is perceived as

highly critical, then leadership practices related to MBEA might be more relevant given that

MBEA has been associated with a prevention-oriented focus (Antonakis et al., 2003;

Rodriguez & Griffin, 2009). In line with this, research has demonstrated that fit between


11

followers’ prevalent regulatory focus and leadership style leads to beneficial outcomes

(Hamstra, Van Yperen, Wisse, & Sassenberg, 2015).


12

In Bass’ (1985) original framework, transactional leadership consisted of MBEA (i.e.,

monitoring for problems as they arise and correcting these proactively) as well as

Management-By-Exception-Passive (i.e., not taking action until issues occur, waiting for

things to go wrong before interfering) and Contingent Reward (clarifying what followers can

expect in exchange for their efforts and rewarding achievements). However, examinations of

the structure of transactional leadership have repeatedly identified MBEA as a single factor,

with Contingent Reward representing a dimension of transformational leadership and

Management-By-Exception-Passive representing a passive leadership factor together with

laissez-faire leadership (Den Hartog, Van Muijen, & Koopman, 1997; Hinkin & Schriesheim,

2008). Hence, there is evidence that MBEA is the only dimension of transactional leadership

and we therefore focus on MBEA. We do not include passive leadership in our study as

empirical evidence has consistently shown that passive leadership is negatively related to job

performance and workplace safety outcomes (Judge & Piccolo, 2004; Kelloway et al., 2006).

The purpose of our study is to examine whether different active leadership styles are more or

less effective in dependence of the extent to which safety is perceived as critical within a

work context.


13

Cognitive assumptions about leadership are dynamic with contextual attributes

influencing leaders’ and followers’ schemas of effective leadership (Shondrick & Lord,

2010). Implicit leadership theories postulate that a change in context can shift followers’ as

well as leaders’ implicit expectations about aspects of leadership (Derler & Weibler, 2013;

Lord & Emrich, 2001; Smothers et al., 2011). As a consequence, leaders might adjust their

behavior and followers can become more or less receptive towards certain leadership

practices (Lord, Hannah, & Jennings, 2001). Therefore, in safety-critical contexts, the

effectiveness of MBEA might be enhanced because: 1) leaders perceive MBEA behaviors as

more apt and adjust their use of MBEA accordingly, and/ or 2) followers increase their

receptiveness towards MBEA. Thus, high hazard and high accident risk conditions may

activate a prevention focus and in conjunction sway leaders’ and/or followers’ implicit

expectations of MBEA, leading to a stronger positive relationship with safety and job

performance.


14

There is evidence from leadership research in crisis and disaster contexts, which share

characteristics with safety-critical contexts, that leadership styles akin to MBEA have

increased relevance under such conditions. For example, DeChurch et al. (2001) discussed

the importance for leaders to structure work and to provide coordination in disaster

conditions. Others have discussed facilitation of sensemaking (Baran & Scott, 2010;

Mumford, Friedrich, Caughron, & Byrne, 2007) and pragmatic leadership (Yammarino,

Mumford, Connelly, & Dionne, 2010) as key for effective leadership in dangerous contexts.

MBEA behaviors such as clarifying responsibilities are likely to aid sensemaking, and by

monitoring task execution for problems leaders can be pragmatic and put solutions into place.

Drawing on this body of research in dangerous contexts, it can be asserted that MBEA

becomes relevant for effective leadership if safety is perceived as highly critical, as it

addresses issues of sensemaking and pragmatic problem solving.


15

Consistent with this proposition, Clarke (2013) in her meta-analysis showed that

active transactional leadership was positively related to safety compliance (ρ = .41) and

safety participation (ρ = .36). Griffin and Talati (2011) showed in their meta-analysis that the

mean level of MBEA within a sample moderated the correlation between MBEA and

employees’ extra effort and satisfaction with leadership. The authors explained that high

mean levels are a proximal indicator for MBEA practices being favorably perceived within

the work context, and therefore have a stronger impact on leadership outcomes (Griffin &

Talati, 2011). This is in line with our proposition that the effectiveness of MBEA varies

across different work settings. Yet, the understanding of the influence of specific context

factors on relationships between variables is little developed (Johns, 2006). Thus, we further

advance Griffin and Talati’s (2011) work by investigating exposure to hazards and likelihood

for an accident as concrete contextual factors that may impact on the relationship between

MBEA and performance outcomes.


16

Whilst we argue that MBEA plays a particularly important role if safety is perceived

as highly critical, it can be suggested that these leadership tactics have no effect or might

even be detrimental for performance if safety is perceived as less critical as the need for

continuous correction of errors is reduced (Antonakis et al., 2003). As argued above, context

can alter followers’ and leaders’ evaluations of what constitutes effective leadership (Emrich,

1999; Lord et al., 2011). Just as fit between leadership style and followers’ prominent

regulatory focus fosters a positive impact by leaders, mismatch between leadership behavior

and followers’ prevention-promotion focus prohibits such desirable effects (Hamstra et al.,

2015). Therefore, if employees perceive lower hazard-levels and assess little threat for an

accident, a consequently more promotion-oriented regulatory focus could clash with MBEA

leader behaviors, so that these might be less positively related to safety and job performance.


17

We expect that followers’ perceptions of hazard exposure and accident likelihood will

moderate the relationship between MBEA with both safety performance and job

performance, as both performance types take place under the same contextual constraints.

Dierdorff et al. (2009) explain that role enactment (e.g., of a managerial role or team member

role) is shaped by contextual factors, so that the role in its entirety interplays with the context

in which it is performed. In their roles, employees commonly perform safety-specific aspects

at the same time and intertwined with general performance elements (e.g., performing a

general job task whilst wearing protective equipment correctly), so that a regulatory focus or

implicit expectations about leadership are less likely to be determined by the specific

performance element, but the overall context in which job performance and safety

performance takes place. We therefore expect perceptions of hazard exposure and accident

likelihood to have the same moderation effects for job performance as well as safety

performance.


18

Job performance and safety performance are multidimensional outcomes and in our

study we investigate the individual dimensions, i.e. task performance, contextual performance

and safety participation, safety compliance separately. Most taxonomies of work performance

incorporate a distinction between formal, within-role behaviors (task performance) and extra-

role, discretionary behaviors (contextual performance; e.g., Borman & Motowidlo, 1993).

Based on this widely accepted partitioning of work performance, Griffin and Neal (2000)

developed an analogy of safety performance. They proposed safety compliance with rules

and regulations as a correspondent to task performance, and safety participation, which

describes extra-role behaviors for safety, as a correspondent to contextual performance.

Transactional leadership has been more strongly related to task performance and safety

compliance, and transformational leadership more strongly related to contextual performance

and safety participation (e.g., Clarke, 2013). Accordingly, including overall measures of job

and safety performance could obscure differences in the strength of relationship between the

two leadership styles with these outcome dimensions. Therefore, we investigate individual

performance dimensions (i.e., safety compliance/participation and task/contextual

performance), and propose separate hypotheses for each of the performance outcomes (Figure

1):

Hypothesis 1(a, b, c, d): Perceived accident likelihood moderates the relationship of

MBEA to safety compliance (hypothesis 1a), safety participation (hypothesis 1b), task


19

performance (hypothesis 1c), and contextual performance (hypothesis 1d). It is

expected that MBEA will be more positively related to safety performance and job

performance outcomes if perceived accident likelihood is high.

Hypothesis 2(a, b, c, d): Perceived hazard exposure moderates the relationship of

MBEA to safety compliance (hypothesis 2a), safety participation (hypothesis 2b), task


expected that MBEA will be more positively related to safety performance and job

performance outcomes if perceived hazard exposure is high.

Call for Context: Transformational Leadership in Safety-Critical Contexts

The lack of consideration of contextual factors does not just apply to transactional

leadership practices, but also to investigations of TFL (Liden & Antonakis, 2009; Rowold,

2011). Empirical studies have demonstrated that TFL practices are related to workplace

safety and have associated TFL with outcomes such as employee safety behavior (e.g.,

Barling et al., 2002) and safety incident rates (e.g., Kelloway et al., 2006). However, there is

little research aimed at understanding the degree of effectiveness of TFL practices under

different contextual conditions (Rowold, 2011). Such research is essential to be able to offer

precise guidance to leaders who are operating in safety-critical contexts. We propose that

TFL is less strongly related to performance in contexts where safety is perceived as highly

critical compared to contexts where safety is perceived as less critical.


20

As argued above safety-critical contexts activate a prevention-oriented focus and

trigger expectations about leadership that focus on creating structure and problem solving

(Antonakis et al., 2003; Rodriguez & Griffin, 2009: Wallace et al., 2009). TFL, which is

associated with a promotion-focus (Kark & van Dijk, 2007), therefore would mismatch the

dominant regulatory focus and not be consistent with context-induced expectations of

leadership. Consequently, it can be expected that leaders will be less reliant on TFL in safety-

critical contexts and/or followers could be less responsive towards TFL practices. In line with

this, Dóci and Hofmans (2015) showed that leaders engage less frequently in TFL if task

complexity is high, explaining that in such circumstances “there is no space for stimulating,

inspiring behaviors, or soliciting followers’ ideas but it is more beneficial to give clear

directions” (p. 9). Similarly, followers who perceive their work context as hazardous and

evaluate the risk for an accident as high, might have less capacity to respond to any TFL

behaviors that their leader adopts, so that these are less influential on follower performance.

Research has indicated that threats of danger can increase complexity in a work environment,

which could then lead to increased cognitive demands (Baran & Scott, 2010; Hannah, Uhl-

Bien, Avolio, & Cavaretta, 2009). Furthermore, there has been discussion whether TFL

practices, such as intellectual stimulation, provide employees with greater latitude in how

they interpret and implement safety rules and regulations, which could add to existing


21

ambiguity and be detrimental for performance if safety is highly critical (Clarke, 2013;

Inness, Turner, Barling, & Stride, 2010).

As noted above there is evidence that TFL is more strongly related to extra-role

behaviors (i.e., contextual performance, safety participation) than in-role performance (i.e.,

task performance, safety compliance). We therefore expect that the direction of the

relationship between TFL with within-role and extra-role performance criteria is in the same

direction, but that the interaction effect between TFL with accident likelihood and hazard

exposure is more pronounced for contextual performance/safety participation than for task

performance/safety compliance. To be able to examine whether the moderation effect differs

in strength for within-role and extra-role outcomes, we present separate hypotheses for each

of the performance outcome dimensions. Based on the above argument that TFL is less

relevant if hazards and risk are perceived as high, we propose the following hypotheses:

Hypothesis 3(a, b, c, d): Perceived accident likelihood moderates the relationship of

TFL to safety compliance (hypothesis 3a), safety participation (hypothesis 3b), task


expected that TFL will be less positively related to safety performance and job

performance outcomes if perceived accident likelihood is high.

Hypothesis 4(a, b, c, d): Perceived hazard exposure moderates the relationship of TFL

to safety compliance (hypothesis 4a), safety participation (hypothesis 4b), task


22


expected that TFL will be less positively related to safety performance and job

performance outcomes if perceived hazard exposure is high.

Method

We conducted a two-source study with responses from both leaders and team

members to test our hypotheses. Responses had a team-based structure as members from the

same team rated their communal leader. Thus, we investigated whether hazard exposure and

accident likelihood represent team-level perceptions. Members of the same team are exposed

to similar contextual characteristics as they work in the same environment. Safety

performance behaviors are important outcomes in occupational safety as these have been

demonstrated to predict error-making and accident involvement (Christian, Bradley, Wallace,

& Burke, 2009; Clarke, 2010, 2013). Leadership was investigated as an individual-level

variable, as we did not want to assess a leaders’ general approach, but the effect of specific

leadership behaviors as experienced by individual team members. It is well-acknowledged

within leader-member-exchange theory for example, that leaders vary in their behavior with

different team members (Graen & Uhl-Bien, 1995), which could not be captured by

aggregating leadership to the team-level. Focusing on individual-level leadership also

matches our examination of safety performance and job performance as individual-level

outcomes.


23

Procedure and Participants

We distributed questionnaires to employees of two organizations, an oil and gas

services company and a food manufacturing company. We selected these two sectors to

compile a sample of participants, who all work in contexts where safety plays a role, but with

variation in the levels of hazard exposure and likelihood for an accident. In the oil and gas

service company, the employees who were invited to participate worked on offshore rigs.

Offshore installations pose major hazards such as handling explosive materials,

pressurization control and risk for hydrocarbon release. In addition, participants’ work

activities pose risks from handling and lifting activities (e.g., operating pulley systems), being

struck by objects, and working at heights. The participants from the food manufacturing

company were based in a factory setting, with the majority of participants working in the

production process. Common risks associated with this work setting are colliding with

machinery, handling hot goods and chemicals as well as slips, trips and falls.

A total of 239 team members and their team leaders from twenty-four teams were

invited to take part. In the oil and gas company, for one team none of the team members

choose to participate in the study and for two further teams only one team member

participated respectively. It was therefore decided to omit these cases prior to the analysis.

Outlier analysis led to deletion of four cases. The final sample included in the analysis

consisted of 160 team members from twenty-one teams (oil and gas company: 69 team


24

members from ten teams; manufacturing company: 91 team members from eleven teams).

The mean age amongst team members was 36.35 years, SD = 11.59. A total of 119 team

members were male, twenty-seven female and fourteen team members did not indicate their

gender. Team leaders provided ratings on their team members’ safety and job performance

(Mage = 38.38, SD = 8.65; seventeen team leaders were male). Team sizes ranged from 4 to 17

members with a mean size of 10.76 members (SD = 3.5; oil and gas M team size = 11, SD = 3.62,

manufacturing M team size = 10.54, SD = 3.56). On average teams had 7.62 team members who

were included in the analysis (SD = 2.67; oil and gas services company M = 6.9, SD = 2.42,

manufacturing company M = 8.27, SD = 2.82). The average percentage of members per team

who were included in the analysis was 71.66% (oil and gas average rate per team 63.88%,

ranging from 40% to 83.3%; manufacturing average rate per team 79.44%, ranging from

64.29% to 100%).

Measures

Transformational leadership and MBEA. Team members rated their team leader’s

leadership style on the Multifactor Leadership Questionnaire (MLQ version 5x-Short, Avolio

& Bass, 1995). CFA showed acceptable fit for a factor structure with MBEA as a single

factor and TFL consisting of its sub-dimensions of Idealized Influence, Individual

Consideration, Inspirational Motivation and Intellectual Stimulation (χ2 (342) = 486.10, CFI

= .93, TLI = .92, RMSEA = .05). This factor structure showed better fit compared to a model


25

where Contingent Reward loaded together with MBEA on a transactional leadership factor

(χ2 (342) = 498.22, CFI = .92, TLI = .91, RMSEA = .05). The results from CFA are in line

with other research on the MLQ dimensionality (Den Hartog et al., 1997; Hinkin &

Schriesheim, 2008). The transformational leadership scale (α = .90) and MBEA scale (α

= .76) showed good internal consistency.

Accident likelihood. Team members reported their accident likelihood perceptions on

two items, adapted from Cox and Cheyne (2000), “I am sure it is only a matter of time before

I am involved in an accident” and “In my job the chances of being involved in an accident are

quite large”. The response scale ranged from 1 (strongly disagree) to 5 (strongly agree). The

original scale included a third item (‘I am rarely worried about being injured on the job’).

However, CFA showed that this item had a non-significant loading onto the common factor

(-.07). The problematic item was reverse worded and it has been shown that reversed items

can have lower item-scale correlations, resulting in poorer psychometric properties for the

measure (e.g., Barnette, 2000; Schriesheim & Eisenbach, 1995; Spector, Van Katwyk, &

Brannick, 1997). Moreover, the item’s content focuses on injuries whilst the other two items

are concerned with accidents which could have also contributed to its lack of coherence with

the other two items. Thus, it was decided to exclude this third item in the analysis. The

remaining two items both showed significant factor loadings of .59 and .76. Cronbach’s alpha


26

for the two-item scale was .56, which although below the commonly accepted value of .70, is

acceptable given the small number of items (Eisinga, Te Grotenhuis, & Pelzer, 2013).

Hazard exposure. Team members rated the level of hazard exposure in their job on

one item from DeJoy, Schaffer, Wilson, Vandenberg, and Butts (2004): “Rate your current

work environment in terms of your exposure to safety hazards”. The response scale ranged

from 1 = very high hazard level (very unsafe) to 5 = very low hazard level (very safe). We

reverse coded hazard exposure scores prior to the analysis, so that high values represent

higher levels of hazard exposure. DeJoy et al. (2004) demonstrated that this single-item

measure is significantly linked to related constructs such as environmental safety conditions.

It has been demonstrated that if a construct is concrete or uniform, single-item measurement

is acceptable and performs comparably to multiple-item scales (Nagy, 2002; Wanous,

Reichers, & Hudy, 1997). Similar single-item scales have been successfully used in previous

safety research (e.g., Dedobbeleer & Béland, 1991; Morrow & Crum, 1998). We conducted a

CFA to test whether the two accident likelihood items and single hazard exposure item could

be combined into an overall factor representing how safety-critical a context is overall. The

hazard exposure item did not load significantly onto the common factor (-.01) and only the

two accident likelihood items showed significant factor loadings. Therefore, accident

likelihood and hazard exposure were tested as separate moderator variables in the subsequent

analysis.


27

Safety compliance and safety participation. Team leaders rated each of their team

members’ safety performance on six items by Griffin and Neal (2000). Three items addressed

compliance with safety rules and procedures (safety compliance) and three items addressed

extra effort for safety (safety participation). An example item is “This person uses all the

necessary safety equipment to do his/her job” with a response scale from 1 (strongly

disagree) to 5 (strongly agree). Both measures had good internal consistency (safety

compliance α = .86 and safety participation α = .93).

Task performance and contextual performance. Team leaders rated their team

members’ job performance on six items. In line with Borman and Motowidlo’s (1993) model

of different performance behaviors, we assessed task performance (the extent to which an

employee meets in-role expectations) and contextual performance (the extent to which an

employee goes beyond prescribed responsibilities) as separate dimensions of overall work

performance. Three items by Griffin, Neal and Parker (2007) addressed task performance and

three items addressed contextual performance. These were modified from Griffin and Neal’s

(2000) safety performance scale to represent non-domain specific extra effort. An example

item is “This person makes suggestions to improve the overall effectiveness of the company”

with a response scale from 1 (strongly disagree) to 5 (strongly agree). Both measures had

good internal consistency (task performance α = .87 and contextual performance α = .90).


28

We conducted a CFA for the four performance outcomes (i.e., safety compliance,

safety participation, task performance, and contextual performance). All item loadings were

significant and the model showed good fit (CFI = .97, TLI = .95, RMSEA = .08).

As we obtained several variables from the same source (i.e., team members rated

leadership style and perception of hazard exposure and accident likelihood; leaders rated

safety compliance, safety participation, task performance, contextual performance), it is

important to establish discriminant validity for latent variables of the same source. For this,

we followed Fornell and Larcker (1981) by calculating the square root of the average

variance explained (AVE) for each construct. Table 2 shows that the square root of the AVE

for each construct exceeds the correlation between the construct and any other construct that

was rated by the same source, indicating that variables obtained from the same source have

adequate discriminant validity.

Multilevel Analysis

The data had a nested structure as participants were clustered within teams. Failing to

address dependency amongst observations is likely to underestimate the standard errors of

model parameters and increase the risk for a Type I error (Raudenbush & Bryk, 2002). Table

1 shows intraclass correlation coefficients (ICC1 and ICC2) and one-way analysis of variance

(ANOVA) for differences in variable means across the twenty-one teams. We expected

perceptions of accident likelihood and hazard exposure to be team-level variables as these are


29

contextual attributes that are shared across all members of a team. For hazard exposure, the

ICC1 and ICC2 value and ANOVA supported aggregation to the team-level. However, for

accident likelihood the ICC1 and ICC2 values were low and ANOVA was non-significant.

Small ICC values indicate that most of the variation is within teams rather than between

teams and hence do not support aggregation to team-level (Bliese, 2000). Thus, accident

likelihood was not aggregated and assessed as an individual-level moderator in the analysis.

One’s perceptions of risk are not only constructed by objective facts within the physical

context but also determined through subjective components (Hansson, 2010), which could

explain why there is variation in perceptions of accident likelihood amongst members of the

same team. For example, an individual’s assessment of risk is impacted by their judgement of

their own ability to handle potential hazards, which can be over-or underestimated depending

on individual factors (Deery, 2000; Ulleberg & Rundmo, 2003). Thus, perceptions of

accident likelihood are not independent from individual characteristics and subjective

experiences (Deery, 2000; Ulleberg & Rundmo, 2003), which can explain why accident

likelihood perceptions did not converge at the team-level.

Whilst there is no hard rule for a minimum proportion of between-group variance to

indicate non-independence, it has been stressed that even low ICC1 values of .05 can lead to

bias (Aguinis, Gottfredson, & Culpepper, 2013). Values for the remaining study variables

suggested that between-group variance should be controlled to avoid bias of the analysis.


30

Thus, when testing individual-level relationships between-group variance was accounted for

by modeling variables on the within- and between-level using Mplus software (Muthén &

Muthén, 2010). To test the cross-level moderation effect of team hazard-exposure on the

relationship of MBEA and TFL on safety and job performance at the individual-level, we

specified a random slope model as described below.

---Insert Table 1 here---

Results

Table 2 reports descriptive statistics and intercorrelations for the study variables. As

expected, participants who worked in the oil and gas services company reported higher levels

of accident likelihood and more exposure to hazards compared to the food manufacturing

sub-sample (for accident likelihood: M oil and gas = 3.06; M manufacturing = 2.56; t = 3.56, p < .001;

for team-level hazard exposure: M oil and gas = 3.90; M manufacturing = 3.24; t = 12.12, p < .001). A

series of multilevel path models were specified to test whether accident likelihood and team-

level hazard exposure moderate the relationships of TFL and MBEA to safety compliance,

safety participation, task performance and contextual performance. At the team-level, we

controlled for sub-sample membership. To avoid model non-convergence, interactions were

tested in separate models for the outcome variables (Table 3 for safety compliance and safety

participation, Table 4 for task performance and contextual performance). We used the MPlus

TYPE = TWO LEVEL RANDOM command with Maximum-Likelihood-Robust (MLR)


31

estimation (Muthén & Muthén, 2010). To test the cross-level interaction between team-level

hazard exposure and individual-level leadership style (i.e., TFL and MBEA), we followed

guidelines by Aguinis et al. (2013) and Preacher, Zyphur, and Zhang (2010). Hazard

exposure at level-2 was tested as a moderator of the relationship between TFL and MBEA

with performance outcome variables at level-1. Random slopes were specified for the within-

level component of the respective leadership variables and outcome variables, and team-level

hazard exposure was regressed on the random slopes. For accident likelihood, we specified

the interaction between accident likelihood and leadership at level 1, and included the same

relationships between leadership and performance outcomes as level 2. Log-likelihood based

chi-square difference test with scaling correction factors were used to test whether adding the

cross-level interaction term improved model fit (Muthén & Muthén, 2010).




In partial support of hypothesis 1, accident likelihood moderated the relationship

between MBEA and safety participation (1b) and contextual performance (1d). We used

Preacher, Curran, and Bauer’s (2006) computational tool for plotting interaction effects in

multilevel models. We drew interaction plots for safety participation (Figure 2) and

contextual performance (Figure 3) to depict the moderation effect of accident likelihood at


32

one standard deviation above and below the mean and at the mean value1. In line with our

hypothesis, the direction of the moderation effect indicates that engaging in MBEA

leadership may have a positive effect on subordinates’ performance if the risk for an accident

is perceived as high. However, if the risk for an accident is perceived as low, adopting

MBEA leadership practices could actually be detrimental for employees’ safety participation

and contextual performance. Moreover, in support of hypothesis 3d, accident likelihood also

moderated the relationship between TFL and contextual performance. The direction of this

moderation effect was in line with our hypothesis, showing that TFL was positively

associated with contextual performance if accident likelihood was low, but that the

relationship was negative if accident likelihood was high (Figure 4).

Hypotheses 2 and 4 were not supported, as team-level hazard exposure did not

moderate any of the relationships between MBEA and TFL with any of the job and safety

performance outcomes. In addition, there were no significant main effects of either MBEA

or TFL on safety and job performance.

---Insert Figure 2 here---



1 Dawson (2014) discusses that values above/below one standard deviation are relatively arbitrary examples of high and low levels of the moderator variable. We depicted the moderation effect at these commonly used values as perceptions of accident likelihood slide along a continuum without a definitive value that represents attested levels of high or low perceptions of accident likelihood. However, the plots should be interpreted with Dawson’s (2014) notion in mind, illustrating the direction of the effect rather than the relationships at levels of the moderator that do not necessarily reflect a relevant threshold.


33

Discussion

We explored whether perceptions of accident likelihood and hazard exposure impact

on the effectiveness of MBEA and TFL for safety and job performance outcomes. In partial

support of our hypotheses, individual-level perceptions of accident likelihood moderated the

relationship between MBEA and safety participation and contextual performance. If

employees perceived the risk to be involved in an accident as high, MBEA was positively

related to safety participation and contextual performance, but showed a negative relationship

if accident likelihood was rated as low. These findings suggest that the effectiveness of

MBEA is not entirely independent from individuals’ perceptions of context. However, no

evidence was found for hazard exposure as a higher-level context factor as moderator.

Our finding on the moderating role of accident likelihood perceptions is in line with

research on crisis in work environments. It has been argued that under crisis conditions

employees are more receptive to directive forms of leadership (Dóci & Hofmans, 2015;

Morgeson, 2005) which potentially is due to enhanced complexity within the situation (Dóci

& Hofmans, 2015; Mumford et al., 2007). Drawing on complexity leadership theory (Uhl-

Bien, Marion, & McKelvey, 2007), Baran and Scott (2010) suggest that reducing ambiguity

and sensemaking through behaviors such as direction setting and situational awareness is core

to effective leadership under extreme conditions. Our results indicate that if the likelihood for

an accident is perceived as high, instructive leader tactics, which pre-empt error making and


34

clarify the priority of safety, can have a positive influence on certain performance outcomes

such as safety participation and contextual performance. Dahl and Olsen (2013) showed in a

sample of offshore petroleum workers that leadership involvement (they did not test MBEA

specifically) enhances safety compliance through increasing role clarity. However, if the risk

for an accident is perceived as lower then this level of vigilance and close leader intervention

can be perceived as too controlling and therefore is associated with lower levels of

performance (Antonakis et al., 2003; Griffin & Talati, 2011). Whilst perceptions of high

accident likelihood are likely to activate a prevention focus, lower accident likelihood can

shift the focus towards a promotion focus which does not match up with MBEA behaviors.

Such imparity between the activated regulatory focus and a leader’s adopted style is

detrimental and leads to reduced performance (Hamstra et al., 2015). We did not find support

for team-level hazard exposure as moderator of the relationship between leadership style and

performance. Perceptions of risk for an actual accident might elicit an enhanced focus on

prevention and need for structure, but the mere presence of hazards possibly does not

engender such shift in focus to the same degree.

In the introduction, we proposed that an enhanced effect of MBEA on performance

under safety-critical conditions, could be due to leaders adjusting their leadership style,

and/or due to followers being more attuned to MBEA. Although our study did not set out to

formally test these two mechanisms, correlation results indicated that the moderation effect,


35

cannot be explained through leaders’ behavioral adjustment and instead could be due to

followers’ being more receptive towards MBEA if the likelihood for an accident is perceived

as high. Zero-order correlations showed that leaders’ level of MBEA was negatively related

to followers’ perceptions of accident likelihood. Thus, if followers perceived the level of

accident likelihood as high, leaders were actually rated lower on MBEA. This is in contrast to

the proposition that leaders adjust their leadership style to include more MBEA if safety is

highly critical and a prevention-oriented regulatory focus dominates. Instead, followers being

more receptive towards their leaders’ MBEA might explain the positive relationship between

MBEA with safety participation and contextual performance. However, we did not assess

followers’ expectations of MBEA and future research should further examine the

mechanisms that underlie the variation of effectiveness of MBEA in situations with different

levels of accident likelihood.

Interestingly, accident likelihood only emerged as a moderator of the link between

MBEA with extra-role performance measures (i.e., safety participation and contextual

performance), but not for safety compliance and task performance as outcomes. Existing

research has reported stronger links between transactional leadership and compliance than

with extra-role performance; whereas transformational leadership has been more strongly

linked to extra-role behaviors such as safety participation (Clarke, 2013; Griffin & Hu, 2013;

Mullen & Kelloway, 2009). The current results appear plausible given that the relationships


36

were moderated by perceptions of the context. If team members perceive the threat for an

accident within their work context as high, it is likely that they are intrinsically motivated to

comply with safety rules and regulations regardless of their leader’s behavior. Thus, accident

likelihood perceptions did not influence the link between MBEA and safety compliance. This

interpretation is in line with the leadership substitute approach, which postulates that certain

contextual factors can reduce the need for leadership (De Vries, Roe, & Tallieu, 2002; Kerr &

Jermier, 1978). However, MBEA practices such as vigilantly checking for mistakes and

clarifying the priority of safety, may encourage employees beyond their self-propelled efforts

to comply with safety rules and to direct extra attention and resources to safety and their job

objectives. Research has generally argued that transactional leadership is less effective for

influencing extra-role behaviors (Clarke, 2013). However, the present results showed that this

conclusion may not hold under conditions where threat for an accident or harm is perceived

as high. Thus, our findings further underpin the importance of considering perceptions of

contextual characteristics when investigating the effectiveness of leadership.

The moderating effect of accident likelihood on the relationship between MBEA with

safety participation and contextual performance could be explained through fit between

leaders’ behavior and followers’ situational regulatory focus. The direction of one’s self-

regulatory focus can be shaped through situational factors (Crowe &Higgins, 1997).

Perceiving a high threat for an accident is likely to orient followers’ self-regulatory focus


37

more towards prevention and the avoidance of losses (Rodriguez & Griffin, 2009), and

therefore make followers more sensitive towards their leaders’ MBEA behavior.

Nevertheless, it must be acknowledged that variables other than situational factors, such as

someone’s inherent regulatory orientation can also influence whether a promotion or

prevention orientation dominates (Crowe & Higgins, 1997; Kark & van Dijk, 2007).

Moreover, it can be argued that followers to some extent create their work context through

their own performance. As individuals interact with their context and the people within this

context, they themselves effectuate contextual aspects such as risk and hazard-levels. For

example, the safety climate in an organization has been identified as a leading as well as

lagging indicator of accidents (Payne, Bergman, Beus, Rodríguez, & Henning, 2009). Thus,

perceptions of safety-related contextual factors could, to some degree, be the outcome of

employees’ safety-related behavior within the context. These alternative interpretations mean

that leaders are not always able to adapt their style to a given context, as followers’

perceptions of the context are determined by a range of factors including followers’ inherent

characteristics. The cross-sectional nature of our data prevents us from pinpointing the source

of followers’ context perceptions. Longitudinal research is needed to separate the direction of

the relationships between leadership style, context perceptions, and follower safety and job

performance.


38

Nevertheless, our results suggest that MBEA is not universally ineffective, but that its

relationship with performance varies for different conditions. MBEA has often been

discussed as a negative, controlling form of leadership (Griffin & Talati, 2011). The present

results challenge this unrefined, negative view and call for a more differentiated, condition-

specific view of MBEA. Clarke (2013) provided meta-analytical evidence that active

transactional leadership is positively linked with safety compliance and safety participation.

In addition, literature on high-reliability organizations has emphasized the need for

monitoring of procedural deviations as critical actions to control intrinsic hazards (Roberts,

1993). Together with the present findings, this makes MBEA of interest to future research in

the area of safety and highlights the importance of merging behavioral leadership approaches

with contingency views.

As hypothesized, accident likelihood moderated the relationship between TFL and

contextual performance. In support of our hypothesis, the direction of this effect showed that

TFL is less effective for contextual performance if individuals perceive the risk for an

accident as high. If accident likelihood is perceived as high, TFL might be mismatched with

followers’ intensified focus on prevention of loss and failure and is therefore less effective in

influencing follower performance (Hamstra et al., 2015). Cognitive demands could also play

a role in our findings that TFL is less effective if accident likelihood is perceived as high.

Dóci and Hofmans (2015) explain that when a task becomes overly demanding,


39

psychological resources are diminished which reduces a leader’s ability to engage in TFL

behaviors. Perceptions of high risk for an accident are likely to add complexity to a work

situation (Baran & Scott, 2010). In such conditions, not only the leader’s but also followers’

psychological resources are reduced and followers may be less receptive towards TFL

behaviors. In an early study, Seltzer, Numerof and Bass (1989) demonstrated that TFL under

high stress conditions further exacerbates stress levels. Similarly, if perceptions of high risk

for an accident make a situation more cognitively challenging, then TFL behaviors that ask

followers to question the status quo or buy into a longer-term vision may only further mount

demands and strain cognitive resources and therefore become counterproductive. Moreover,

Inness et al. (2010) posit that TFL provide subordinates with greater latitude in how they

implement safety rules and procedures. For example, intellectual stimulation could indirectly

increase deviations from safety procedures as it encourages employees to try out new

approaches to their work (Clarke, 2013). TFL is associated with a promotion focus which in

turn has been associated with greater risk taking (Friedman & Förster, 2001). This

interpretation suggests that it might not be the context per se but the leaders’ own behavior

that shapes follower regulatory focus and their context perceptions. Thus, our findings further

add to this discussion that TFL may not be equally effective across all work contexts and

future research is needed to investigate cognitive demands and regulatory focus as

mechanisms of this effect.


40

Neither TFL nor MBEA had significant main effects with any of the performance

outcomes. A strength of our study was the use of two-source data with supervisory

performance ratings. The lack of support for a relationship between TFL and MBEA to safety

and job performance calls into question whether the transformational-transactional leadership

framework sufficiently captures leader practices that impact upon employee job and safety

behaviors. Others have argued that research needs to look beyond the transformational-

transactional framework for a more comprehensive, real-life understanding of effective

leadership (van Knippenberg & Sitkin, 2013). Within the occupational safety literature,

studies that have found support for a link between transformational leadership and safety

outcomes have often used safety-specific operationalizations of leadership (e.g., Barling et

al., 2002). Kapp (2012) showed that non-domain specific transformational leadership was

only related to safety behavior if it was coupled with a positive safety climate. Together with

our findings, this criticism challenges researchers in the leadership field to not only pay

attention to perceptions of contextual factors, but also to extend conceptualizations of

leadership beyond the transformational-transactional leadership model.

The present research made an important, overarching contribution by integrating the

contingency and behavioral perspectives on leadership. Although we only found support for

perceptions of accident likelihood as moderator, the findings reinforce requests that the


41

transformational-transactional leadership framework needs to be refined by integrating

contingency views into its postulations (Avolio & Yammarino, 2013).

Practical Contributions

The research identified three moderation effects which lends some support that the

perceptions of risk for an accident interact with leadership styles’ effectiveness. These

findings have important practical implications for managers and supervisors. The degree to

which safety is perceived as critical within subordinates’ work environments is often not a

constant, but changes depending on the type of work-task or project employees are currently

performing. During some work periods employees work in settings where safety is highly

critical and on other work days perceptions of the risk for an accident is more moderate or

low. For example, one of the host organizations was an oil and gas service provider where

participants alternate between working on an offshore installation and working onshore at the

company’s base. Similarly, despite working in the same location, perceptions of context

might vary between teams as different job groups interact differently with their work

environment. Our results indicate that their supervisors should engage in MBEA when

working offshore in situations when employees are likely to perceive the risk for an accident

as higher, but that MBEA practices when working on tasks where perceived risk for accidents

is lower, could be detrimental for performance. This practical implication can be transferred


42

to a range of other work settings where subordinates oscillate between conditions with

differing degrees of risk for an accident. However, as discussed above it is likely that

followers’ perceptions of their context are influenced by variables other than the objective

context itself (e.g., individual differences, leader’s own behavior), which were not assessed in

the present study. Moreover, team-level hazard exposure did not interact with leadership

style. Even if leaders have a repertoire of different behaviors, the key to using these

effectively may be in attuning leaders to their followers’ perceptions of the context and

understanding what factors, including leaders themselves, shape how followers perceive

context.

Limitations and Future Directions

Whilst our study had some key strengths such as the use of two-source data to counter

common source bias, the study exhibits several limitations. As noted above, one limitation is

the cross-sectional study design, which restricts conclusions about the direction of causality

for the links between leadership behaviors, perceptions of context and employee

performance. We investigated perceptions of accident likelihood and hazard exposure as

moderators, however these are also possible outcomes of leadership and follower behavior.

Both TFL and MBEA were negatively related to perceptions of accident likelihood, so that

followers might perceive a higher risk for an accident because of their leaders’ lack of active

leadership. Due to the cross-sectional nature of our data and because we assessed perceptions


43

of contextual aspects rather than a direct assessment of work context, we cannot make any

firm conclusions whether perceptions of accident likelihood impact on the effectiveness of

TFL and MBEA, or whether the level of TFL and MBEA determines followers’ perceptions

of accident likelihood. Future research should contrast perception of context to more

objective measures of hazards and risk, such as safety audits or external risk assessments, to

examine whether these overlap or interact differently with leader behavior. Moreover, future

research using a longitudinal design is needed to shed further light on the degree to which

followers’ perceptions of their context are influenced by external factors such as their

leader’s behavior as well as their own safety behavior and inherent characteristics (e.g.,

chronic regulatory-focus).

Our sample size at the team-level was modest and the number of cases within teams

was small. Aguinis et al. (2013) discuss that multilevel research faces a particular dilemma

of balancing out the chances for making a Type I and Type II error. We chose to employ

multilevel modeling to take the nested data structures into account and avoid overestimating

significance levels; however, small sample sizes at the higher-unit reduce statistical power,

increasing the risk for Type II error (Aguinis et al., 2013). Although, this means that

estimates are likely to be conservative and the moderator effect that did emerge as significant

might have actually been underestimated (Mathieu, Aguinis, Culpepper, & Chen, 2012).

There is a complex combination of factors influencing power in models with random slopes


44

and clear guidelines for estimating power in cross-level interaction models are still to be

developed (Scherbaum & Ferreter, 2009; Mathieu & Chen, 2011). In a simulation study,

Mathieu et al. (2012) reported that effect size of the cross-level interaction, Level 1 sample

size, Level 2 sample size, and standard deviation of the slopes are critical factors for

influencing power in cross-level interaction models and that these factors interact with each

other in determining power.

Conclusion

Our research made a principal contribution by merging transformational-transactional

leadership as a behavioral leadership approach with contingency views of leadership by

investigating the level at which safety is critical as a contextual characteristic. Results showed

that perceptions of the level of accident likelihood in subordinates’ work context present a

constraint of the effectiveness of TFL and MBEA for team members’ safety participation and

contextual performance.


45

References

Aguinis, H., Gottfredson, R. K., & Culpepper, S. A. (2013). Best-practice recommendations

for estimating cross-level interaction effects using multilevel modeling. Journal of

Management, 39(6), 1490-1528. doi: 10.1177/0149206313478188

Antonakis, J., Avolio, B. J., & Sivasubramaniam, N. (2003). Context and leadership: An

examination of the nine-factor full-range leadership theory using the Multifactor

Leadership Questionnaire. The Leadership Quarterly, 14(3), 261-295. doi:

10.1016/S1048-9843(03)00030-4

Arvonen, J., & Ekvall, G. (1999). Effective leadership style: Both universal and contingent?

Creativity and Innovation Management, 8(4), 242-250. doi: 10.1111/1467-

8691.00143

Avolio, B. J., & Bass, B. M. (1995). MLQ Multifactor leadership questionnaire. Redwood

City, CA: Mind Garden.

Avolio, B. J., & Yammarino, F. J. (2013). Introduction to, and overview of, transformational

and charismatic leadership. In B. J. Avolio & F. J. Yammarino (Eds.),

Transformational and Charismatic Leadership: The road ahead (pp. xxvii-xxxiii).

Bingley, UK: Emerald.

Baran, B. E., & Scott, C. W. (2010). Organizing ambiguity: A grounded theory of leadership

and sensemaking within dangerous contexts. Military Psychology, 22(S1), S42 – S69.

doi: http://dx.doi.org/10.1080/08995601003644262

http://dx.doi.org/10.1080/08995601003644262


46

Barling, J., Loughlin, C., & Kelloway, E. K. (2002). Development and test of a model linking

safety-specific transformational leadership and occupational safety. Journal of

Applied Psychology, 87(3), 488-496. doi: 10.1037/0021-9010.87.3.488Bass, B.

(1985). Leadership and performance beyond exceptions. New York, NY: Free Press.

Barnette, J. J. (2000). Effects of stem and Likert response option reversals on survey internal

consistency: If you feel the need, there is a better alternative to using those negatively

worded stems. Educational and Psychological Measurement, 60(3), 361–370. doi:

10.1177/00131640021970592

Bass, B. (1985). Leadership and performance beyond exceptions: New York, NY: Free Press.

Bass, B. M., & Riggio, R. E. (2006). Transformational leadership. Mawah, NJ: Erlbaum.

Bliese, P. (2000). Within-group agreement, non-independence, and reliability. In K.

Klein & S. Kozlowski (Eds.), Multi-level theory, research, and methods in

organizations (pp. 349-381). San Francisco, CA: Jossey-Bass.

Bono, J. E., & Judge, T. A. (2004). Personality and transformational and transactional

leadership: A meta-analysis. Journal of Applied Psychology, 89(5), 901-910. doi:

10.1037/0021-9010.89.5.901

Borman, W. C., & Motowidlo, S. J. (1997). Task performance and contextual performance:

The meaning for personnel selection research. Human Performance, 10(2), 99-109.

doi: 10.1207/s15327043hup1002_3

http://psycnet.apa.org/doi/10.1037/0021-9010.89.5.901


47

Christian, M. S., Bradley, J. C., Wallace, J. C., & Burke, M. J. (2009). Workplace safety: A

meta-analysis of the roles of person and situation factors. Journal of Applied

Psychology, 94(5), 1103-1127.doi: 10.1037/a0016172

Clarke, S. (2010). An integrative model of safety climate: Linking psychological climate and

work attitudes to individual safety outcomes using meta-analysis. Journal of

Occupational and Organizational Psychology, 83(3), 553-578. doi:

10.1348/096317909X452122

Clarke, S. (2013). Safety leadership: A meta‐analytic review of transformational and

transactional leadership styles as antecedents of safety behaviours. Journal of


10.1111/j.20448325.2012.02064.x

Cox, S., & Cheyne, A. (2000). Assessing safety culture in offshore environments. Safety

Science, 34(1-3), 111-129. doi: 10.1016/S0925-7535(00)00009-6

Crowe, E., & Higgins, E.T. (1997). Regulatory focus and strategic inclinations: promotion

and prevention in decision-making. Organizational Behavior and Human Decision

Processes, 69(2), 117-132. doi: 10.1006/obhd.1996.2675

Dahl, Ø., & Olsen, E. (2013). Safety compliance on offshore platforms: A multi- sample

survey on the role of perceived leadership involvement and work climate. Safety

Science, 54, 17-26. doi: 10.1016/j.ssci.2012.11.003

Dawson, J. F. (2014). Moderation in management research: What, why, when, and how.


48

Journal of Business and Psychology, 29(1), 1-19. doi: 10.1007/s10869-013-9308-7

DeChurch, L. A., Burke, C.S., Shuffler, M.L., Lyons, R., Doty, D., & Salas, E. (2011). A

historiometric analysis of leadership in mission critical multiteam environments. The

Leadership Quarterly, 22(1), 152-169. doi: 10.1016/j.leaqua.2010.12.013

Dedobbeleer, N., & Béland, F. (1991). A safety climate measure for construction sites.

Journal of Safety Research, 22(2), 97-103. doi: 10.1016/0022-4375(91)90017-P

Deery, H. A. (2000). Hazard and risk perception among young novice drivers. Journal of

Safety Research, 30(4), 225-236. doi: 10.1016/S0022-4375(99)00018-3

DeJoy, D. M., Schaffer, B. S., Wilson, M. G., Vandenberg, R. J., & Butts, M. M. (2004).

Creating safer workplaces: Assessing the determinants and role of safety climate.

Journal of Safety Research, 35(1), 81-90. doi: 10.1016/j.jsr.2003.09.018

Den Hartog, D. N., Van Muijen, J. J., & Koopman, P. L. (1997). Transactional versus

transformational leadership: An analysis of the MLQ. Journal of Occupational

and Organizational Psychology, 70(1), 19-34.doi: 10.1111/j.2044-

8325.1997.tb00628.x

Denison, D. R., Hooijberg, R., & Quinn, R. E. (1995). Paradox and performance: Toward a

theory of behavioral complexity in managerial leadership. Organization Science, 6(5),

524-540. doi: 10.1287/orsc.6.5.524


49

Derler, A., & Weibler, J. (2014). The ideal employee: context and leaders’ implicit follower

theories. Leadership & Organization Development Journal,35(5), 386-409. doi:

http://dx.doi.org/10.1108/LODJ-12-2012-0158

DeRue, D. S., Nahrgang, J. D., Wellman, N., & Humphrey, S. E. (2011). Trait and behavioral

theories of leadership: An integration and meta‐analytic test of their relative validity.

Personnel Psychology, 64(1), 7-52. doi: 10.1111/j.1744-6570.2010.01201.x

De Vries, R. E., Roe, R. A., & Taillieu, T. C. (2002). Need for leadership as a moderator of

the relationships between leadership and individual outcomes. The Leadership

Quarterly, 13(2), 121-137. doi: 10.1016/S1048-9843(02)00097-8

Dierdorff, E. C., Rubin, R. S., & Morgeson, F. P. (2009). The milieu of managerial work: an

integrative framework linking work context to role requirements. Journal of Applied

Psychology, 94(4), 972 – 988. doi: 10.1037/a0015456

Dóci, E., & Hofmans, J. (2015). Task complexity and transformational leadership: The

mediating role of leaders' state core self-evaluations. The Leadership

Quarterly, 26(3), 436-447. doi: 10.1016/j.leaqua.2015.02.008

Eisinga, R., Grotenhuis, M. T., & Pelzer, B. (2013). The reliability of a two-item scale:

Pearson, Cronbach, or Spearman-Brown?. International Journal of Public Health, 58,

637-642. doi: 10.1007/s00038-012-0416-3

Emrich, C. G. (1999). Context effects in leadership perception. Personality and Social

Psychology Bulletin, 25(8), 991-1006. doi: 10.1177/01461672992511007

http://dx.doi.org/10.1016/j.leaqua.2015.02.008


50

Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models with

unobservable variables and measurement error. Journal of Marketing Research, 18(1),

39-50. doi: 10.2307/3151312

Fiedler, F. E. (1964). A contingency model of leadership effectiveness. In L. Berkowitz (Ed.),

Advances in experimental social psychology (pp. 149-190). New York, NY:

Academic Press.

Friedman, R. S., & Förster, J. (2001). The effects of promotion and prevention cues on

creativity. Journal of Personality and Social Psychology, 81(6), 1001-1013. doi:

10.1037/0022-3514.81.6.1001

Graen, G. B., & Uhl-Bien, M. (1995). Relationship-based approach to leadership:

Development of leader-member exchange (LMX) theory of leadership over 25 years:

Applying a multi-level multi-domain perspective. The Leadership Quarterly, 6(2),

219-247.doi: 10.1016/1048-9843(95)90036-5

Griffin, M. A., & Hu, X. (2013). How leaders differentially motivate safety compliance and

safety participation: The role of monitoring, inspiring, and learning. Safety Science,

60, 196-202. doi: 10.1016/j.ssci.2013.07.019

Griffin, M. A., & Neal, A. (2000). Perceptions of safety at work: A framework for linking

safety climate to safety performance, knowledge, and motivation. Journal of

Occupational Health Psychology, 5(3), 347-358. doi: 10.1037/1076-8998.5.3.347

Griffin, M.A., Neal, A., & Parker, S.K. (2007). A new model of work role performance:


51

Positive Behaviour in uncertain and interdependent contexts. Academy of

Management Journal, 50(2), 327-347. doi: 10.5465/AMJ.2007.24634438

Griffin, M. A., & Talati, Z. (2011, August). Meta-analysis of active management-by-

exception-active variability across leadership contexts. Paper presented at the

Academy of Management annual conference, San Antonio, TX.

Hamstra, M. R., Van Yperen, N. W., Wisse, B., & Sassenberg, K. (2014). Transformational

and transactional leadership and followers’ achievement goals. Journal of Business

and Psychology, 29(3), 413-425. doi: 10.1007/s10869-013-9322-9

Hannah, S. T., Balthazard, P. A., Waldman, D. A., Jennings, P. L., & Thatcher, R. W. (2013).

The psychological and neurological bases of leader self-complexity and effects on

adaptive decision-making. Journal of Applied Psychology, 98(3), 393-411. doi:

10.1037/a0032257

Hannah, S. T., Uhl-Bien, M., Avolio, B., & Cavarretta, F. (2009). A framework for

examining leadership in extreme contexts. Leadership Quarterly, 20, 897-919.

doi:10.1016/j. leaqua.2009.09.006

Hansson, S. O. (2010). Risk: objective or subjective, facts or values. Journal of Risk

Research, 13(2), 231-238. doi: 10.1080/13669870903126226

Hinkin, T. R., & Schriesheim, C. A. (2008). A theoretical and empirical examination of the


52

transactional and non-leadership dimensions of the Multifactor Leadership

Questionnaire (MLQ). The Leadership Quarterly, 19(5), 501-513. doi:

10.1016/j.leaqua.2008.07.001

Howell, J. P., Dorfman, P. W., & Kerr, S. (1986). Moderator variables in leadership research.

Academy of Management Review, 11(1), 88-102. doi: 10.5465/AMR.1986.4282632

Inness, M., Turner, N., Barling, J., & Stride, C. B. (2010). Transformational leadership and

employee safety performance: A within-person, between-jobs design. Journal of

Occupational Health Psychology, 15(3), 279-290. doi: 10.1037/a0019380

Johns, G. (2006). The essential impact of context on organizational behavior. Academy of

Management Review, 31(2), 386-408. doi: 10.5465/AMR.2006.20208687

Johnson, R. E., Venus, M., Lanaj, K., Mao, C., & Chang, C. H. (2012). Leader identity as an

antecedent of the frequency and consistency of transformational, consideration, and

abusive leadership behaviors. Journal of Applied Psychology, 97(6), 1262-1272. doi:

10.1037/a0029043

Judge, T. A., & Piccolo, R. F. (2004). Transformational and transactional leadership: A meta-

analytic test of their relative validity. Journal of Applied Psychology, 89(5), 755-768.

doi: 10.1037/0021-9010.89.5.755

Kapp, E. (2012). The influence of supervisor leadership practices and perceived group safety

climate on employee safety performance. Safety Science, 50(4), 1119- 1124. doi:

10.1016/j.ssci.2011.11.011

http://psycnet.apa.org/doi/10.1037/0021-9010.89.5.755


53

Kark, R., & Van Dijk, D. (2007). Motivation to lead, motivation to follow: The role of the

self-regulatory focus in leadership processes. Academy of Management Review, 32(2),

500-528. doi: 10.5465/AMR.2007.24351846

Kelloway, E. K., Mullen, J., & Francis, L. (2006). Divergent effects of transformational and

passive leadership on employee safety. Journal of Occupational Health Psychology,

11(1), 76-86. doi: 10.1037/1076-8998.11.1.76

Kerr, S., & Jermier, J. M. (1978). Substitutes for leadership: Their meaning and

measurement. Organizational Behavior and Human Performance, 22(3), 375-403.

doi: 10.1016/0030-5073(78)90023-5

Liden, R. C., & Antonakis, J. (2009). Considering context in psychological leadership

research. Human Relations, 62(11), 1587-1605. doi: 10.1177/0018726709346374

Lim, B. C., & Ployhart, R. E. (2004). Transformational leadership: relations to the five-factor

model and team performance in typical and maximum contexts. Journal of Applied

Psychology, 89(4), 610-621. doi: 10.1037/0021-9010.89.4.610

Lord, R. G., & Emrich, C. G. (2001). Thinking outside the box by looking inside the box:

Extending the cognitive revolution in leadership research. The Leadership

Quarterly, 11(4), 551-579. doi: 10.1016/S1048-9843(00)00060-6

Lord, R. G., Hannah, S. T., & Jennings, P. L. (2011). A framework for understanding

leadership and individual requisite complexity. Organizational Psychology Review,

1(2), 104-127. doi: 10.1177/2041386610384757

Mathieu, J. E., Aguinis, H., Culpepper, S. A., & Chen, G. (2012). Understanding and


54

estimating the power to detect cross-level interaction effects in multilevel modeling.

Journal of Applied Psychology, 97(5), 951-966. doi: 10.1037/a0028380

Mathieu, J. E., & Chen, G. (2011). The etiology of the multilevel paradigm in management

research. Journal of Management, 37(2), 610-641. doi: 10.1177/0149206310364663

Morgeson, F. P. (2005). The external leadership of self-managing teams: Intervening in the

context of novel and disruptive events. Journal of Applied Psychology, 90(3), 497-

508. 10.1037/0021-9010.90.3.497

Morrow, P. C., & Crum, M. R. (1998). The effects of perceived and objective safety risk on

employee outcomes. Journal of Vocational Behavior, 53(2), 300-313. doi:

10.1006/jvbe.1997.1620

Mullen, J. E., & Kelloway, E. K. (2009). Safety leadership: A longitudinal study of

the effects of transformational leadership on safety outcomes. Journal of


10.1348/096317908X325313

Mumford, M. D., Friedrich, T. L., Caughron, J. J., & Byrne, C. L. (2007). Leader cognition in

real-world settings: How do leaders think about crises? The Leadership

Quarterly, 18(6), 515-543. doi:10.1016/j.leaqua.2007.09.002

Muthén, L. K., & Muthén, B. O. (2010) Mplus user’s guide (7th ed). Los Angeles, CA:

Muthén & Muthén.

Nagy, M. S. (2002). Using a single-item approach to measure facet job satisfaction. Journal


55

of Occupational and Organizational Psychology, 75(1), 77-86. doi:

10.1348/096317902167658

O'Dea, A., & Flin, R. (2001). Site managers and safety leadership in the offshore oil and gas

industry. Safety Science, 37(1), 39-57. doi: 10.1016/S0925-7535(00)00049-7

Payne, S. C., Bergman, M. E., Beus, J. M., Rodríguez, J. M., & Henning, J. B. (2009). Safety

climate: Leading or lagging indicator of safety outcomes?. Journal of Loss Prevention

in the Process Industries, 22(6), 735-739. doi: 10.1016/j.jlp.2009.07.017

Preacher, K. J., Curran, P. J., & Bauer, D. J. (2006). Computational tools for probing

interactions in multiple linear regression, multilevel modeling, and latent curve

analysis. Journal of Educational and Behavioral Statistics, 31(4), 437-448. doi:

10.3102/10769986031004437

Preacher, K. J., Zyphur, M. J., & Zhang, Z. (2010). A general multilevel SEM framework for

assessing multilevel mediation. Psychological Methods, 15(3), 209-233. doi:

10.1037/a0020141

Raudenbush, S. W., & Bryk, A. S. (2002). Hierarchical linear models: Applications and data

analysis methods. London, UK: Sage.

Roberts, K. H. (1993). Cultural characteristics of reliability enhancing organizations. Journal

of Managerial Issues, 5(2), 165-181. doi: 40603976

Rodriguez, M. A., & Griffin, M. A. (2009). From error prevention to error learning: The role


56

of error management in global leadership. Advances in Global Leadership, 5, 93-112.

doi: 10.1108/S1535-1203(2009)0000005008

Rowold, J. (2011). Relationship between leadership behaviors and performance: The

moderating role of a work team's level of age, gender, and cultural heterogeneity.

Leadership & Organization Development Journal, 32(6), 628-647. doi:

10.1108/01437731111161094

Scherbaum, C. A., & Ferreter, J. M. (2009). Estimating statistical power and required sample

sizes for organizational research using multilevel modeling. Organizational Research

Methods, 12(2), 347-367. doi: 10.1177/1094428107308906

Schriesheim, C. A., & Eisenbach, R. J. (1995). An exploratory and confirmatory factor-

analytic investigation of item wording effects on the obtained factor structures of

survey questionnaire measures. Journal of Management, 21(6), 1177–1193. doi:

10.1177/014920639502100609

Schriesheim, C. A., Wu, J. B., & Scandura, T. A. (2009). A meso measure? Examination of

the levels of analysis of the Multifactor Leadership Questionnaire (MLQ). The

Leadership Quarterly, 20(4), 604-616. doi: 10.1016/j.leaqua.2009.04.005

Seltzer, J., Numerof, R.E., & Bass, B.M. (1989). Transformational leadership: Is it a source

of more burnout and stress?. Journal of Health and Human Resources Administration,

12, 174-185. doi: http://www.jstor.org/stable/25780396

Shondrick, S.J., & Lord, R.G. (2010). Implicit leadership and followership theories: Dynamic

http://www.jstor.org/stable/25780396


57

structures for leadership perceptions, memory, leader-follower processes. In G.P.

Hodgkinson & J.K. Ford (Eds.), International Review of Industrial and

Organizational Psychology, Vol. 25 (pp. 1 – 33). Chichester, UK: John Wiley &

Sons.

Smothers, J., Bing, M. N., White, D., Trocchia, P. J., & Absher, K. (2011). From the

follower’s viewpoint: a configurational approach to the ideal academic

leader. Journal of Leadership & Organizational Studies, 18(3), 293 -307. doi:

10.1177/1548051811404420

Spector, P. E., Van Katwyk, P. T., Brannick, M. T., & Chen, P. Y. (1997). When two factors

don’t reflect two constructs: How item characteristics can produce artifactual

factors. Journal of Management, 23, 659–677. doi: http://dx.doi.org/10.1016/S0149-

2063(97)90020-9

Uhl-Bien, M., Marion, R., & McKelvey, B. (2007). Complexity leadership theory: Shifting

leadership from the industrial age to the knowledge era. Leadership Quarterly, 18(4),

298-318. doi: 10.1016/j.leaqua.2007.04.002

Ulleberg, P., & Rundmo, T. (2003). Personality, attitudes and risk perception as predictors of

risky driving behaviour among young drivers. Safety Science,41(5), 427-443. doi:

10.1016/S0925-7535(01)00077-7


58

Van Knippenberg, D., & Sitkin, S. B. (2013). A Critical Assessment of Charismatic—

Transformational Leadership Research: Back to the Drawing Board? The Academy of

Management Annals, 7(1), 1-60.

Villa, J. R., Howell, J. P., Dorfman, P. W., & Daniel, D. L. (2003). Problems with detecting

moderators in leadership research using moderated multiple regression. The

Leadership Quarterly, 14(1), 3-23. doi: 10.1016/S1048-9843(02)00184-4

Vroom, V. H., & Jago, A. G. (2007). The role of the situation in leadership. American

Psychologist, 62(1), 17-24. doi: 10.1037/0003-066X.62.1.17

Wallace, J. C., Johnson, P. D., & Frazier, M. L. (2009). An examination of the factorial,

construct, and predictive validity and utility of the regulatory focus at work scale.

Journal of Organizational Behavior, 30(6), 805-831. doi: 10.1002/job.572

Wanous, J. P., Reichers, A. E., & Hudy, M. J. (1997). Overall job satisfaction: How good are

single-item measures?. Journal of Applied Psychology, 82(2), 247-252. doi:

10.1037/0021-9010.82.2.247

Yammarino, F. J., Mumford, M. D., Connelly, M. S., & Dionne, S. D. (2010). Leadership and

team dynamics for dangerous military contexts. Military Psychology, 22(S1), 15-41.

doi: 10.1080/08995601003644221

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

Pearson.


59

Zohar, D. (2002). The effects of leadership dimensions, safety climate, and assigned priorities

on minor injuries in work groups. Journal of Organizational Behavior, 23(1), 75-92.

doi: 10.1002/job.130

Zohar, D. (2010). Thirty years of safety climate research: Reflections and future directions.

Accident Analysis & Prevention, 42(5), 1517-1522. doi: 10.1016/j.aap.2009.12.019

Table 1

Intra-Class-Coefficients and ANOVA for Study Variables

ICC1 ICC2 ANOVA F

Leadership variables

Transformational leadership .12 .50 2.17 **

Management-By-Exception-Active .09 .46 1.61*

Safety-critical context variables

Accident likelihood .05 .33 1.34

Hazard exposure .31 .77 4.44 ***

Outcome variables

Safety compliance .24 .68 3.15***

Safety participation .42 .82 5.57***

Task performance .16 .56 2.29**

Contextual performance .34 .76 4.14***

Notes. *p < .05, **p < .01, ***p < .001.


60

Table 2

Means, Standard Deviations, and Intercorrelations for Study Variables

Descriptive Statistics Correlations

M SD α 1. 2. 3. 4. 5. 6. 7. 8.

1.Transformational leadership 3.51 .71 .90 .96 .69*** -.28** .05 .03 -.01 .07 .02

2. MBEA 3.94 .68 .76 .79 -.30*** -.01 -.03 -.05 -.04 -.04

3. Accident likelihood 2.76 .90 .56 .64 .15 -.20* -.03 -.17 -.09

4. Team-level hazard exposure 3.52 .46 - - .19* .38*** .18* .38***

5. Safety compliance 3.58 .65 .86 .90 .51*** .56*** .52***

6. Safety participation 3.00 .81 .93 .90 .59*** .81***

7. Task performance 3.98 .65 .87 .85 .59***

8. Contextual performance 3.18 .91 .90 .88


61

Notes. *p < .05, **p < .01, ***p < .001, MBEA = Management-By-Exception-Active. Values along the diagonal in italic are

square root of AVE.


62


63

Table 3

Testing the Moderating Effects of Team-Level Hazard Exposure and Accident likelihood on the Association between

Transformational Leadership and Management-By-Exception-Active to Safety Compliance and Safety Participation.

Variables a. Safety Compliance b. Safety Participation

SE SE SE SE

Intercept 3.72 3.78 4.08 4.08

Sample -.07 .18 -.07 .21 -.58 .40 -.53 .43

Transformational leadership .15 .11 .15 .10 -.04 .09 .02 .09

MBEA -.14 .12 -.12 .08 .07 .14 -.03 .11

Accident likelihood -.17** .06 -.20** .07 -.08 .08 -.12 .08

Team-level hazard exposure .19 .17 .20 .19 -.05 .38 .06 .45

Accident likelihood × TFL .16 .12 -.18 .11

Accident likelihood × MBEA -.02 .16 .62*** .17

Team-level hazard exposure × TFL -.19 .17 -.18 .20


64

Team-level hazard exposure × MBEA .35 .32 .49 .30

χ2 Log Likelihood -603.897 -567.289 -626.082 -579.712

∆χ2 LL (∆df =6)a 133.69*** 169.62***

Notes. *p < .05 **p < .01 ***p < .001, acompared to model without interactions, TFL = Transformational leadership, MBEA =

Management-By-Exception-Active.


65


66

Table 4

Testing the Moderating Effects of Team-Level Hazard Exposure and Accident likelihood on the Association between

Transformational Leadership and Management-By-Exception-Active to Task Performance and Contextual Performance (N = 160)

Variables c. Task Performance d. Contextual Performance

SE SE SE SE

Intercept 3.82 3.81 4.44 4.37

Sample .14 .19 .20 .21 -.70* .34 -.62 .35

Transformational leadership .16 .09 .27** .11 -.01 .13 .04 .13

MBEA -.14 .12 -.21 .12 .01 .12 -.07 .09

Accident likelihood -.11 .06 -.12* .06 -.17** .06 -.19** .06

Team-level hazard exposure .30* .14 .39* .18 .10 .34 .15 .40

Accident likelihood × TFL .03 .15 -.41* .17

Accident likelihood × MBEA .29 .19 .88*** .26

Team-level hazard exposure × TFL .19 .21 .03 .26


67

Team-level hazard exposure × MBEA .21 .29 .50 .23

χ2 Log Likelihood -613.09 -571.67 -644.31 -601.20

∆χ2 LL (∆df =6)a 155.22*** 214.62***

Notes. *p < .05 **p < .01, ***p < .01 acompared to model without interactions, TFL = Transformational leadership, MBEA = Management-By-

Exception-Active.


68

Figure 1. Summary of hypotheses and conceptual integration of behavioral and contingency

perspectives of leadership.


69

Figure 2. Accident likelihood as moderator of the relationship between MBEA and safety

participation for low accident likelihood (-1 SD below the mean), medium (mean) and for

high accident likelihood (+1 SD above the mean). MBEA = Management-By-Exception-

Active.


70

Figure 3. Accident likelihood as moderator of the relationship between MBEA and

contextual performance for low accident likelihood (-1 SD below the mean), medium (mean)

and for high accident likelihood (+1 SD above the mean). MBEA = Management-By-

Exception-Active.


71

Figure 4. Accident likelihood as moderator of the relationship between TFL and contextual

performance for low accident likelihood (-1 SD below the mean), medium (mean) and for

high accident likelihood (+1 SD above the mean). TFL = Transformational Leadership.

Documents

€¦ · Web viewWithin the existing leadership literature, the role of context for shaping the effectiveness of leadership is yet to be fully understood. One type of context that