osh assist

Journal of risk research 2024, Vol. 27, no. 1, 85–107

Improving workplace safety through mindful organizing: participative safety self-efficacy as a mediational link between collective mindfulness and employees’ safety citizenship

Matteo Curcurutoa, Michelle Renecleb, Francisco Graciab, James I. Morganc and Ines Tomasb

aDepartment of human sciences, european university of rome, rome, italy; bresearch institute on Personnel Psychology, organizational Development, and Quality of Working life (iDocal), university of Valencia, Valencia, spain; cleeds school of social sciences, leeds Beckett university, leeds, uk

ABSTRACT Mindful organizing is a team-level capability that allows teams in high-risk environments to anticipate when something can potentially go wrong and adapt their operations just in time to protect the organizational system from negative consequences. This study aimed to extend our understanding of how mindful organizing affects employees’ propensity to engage in a broad range of safety citizenship behaviours through the mediation of participative safety self-efficacy. Participative safety self-efficacy is a psychological state that enables individuals to have confidence in their capability to engage in constructive behaviours that go beyond the formal requirements of their job description. A multilevel mediation model was tested using data collected from a large sample of chemical workers (N = 443) operating in fifty work teams. The findings showed that mindful organizing on a team level fosters both individual safety citizenship (helping; voice; initiative) and prescribed safety com- pliance through enhancing individual participative self-efficacy. This mediation relationship is significantly stronger for safety citizenship than for safety compliance.

1.  Introduction

High-reliability Organizations (acronym: HROs) are ‘organizations in which errors can have cat- astrophic consequences but which consistently seem to avoid such errors’ (Roberts et  al. 2005, 216) in an environment where accidents can be expected due to risk factors and complexity (Perrow 1984). Examples of such organizations are nuclear power plants and air traffic control centres. Although there are some well-known classical models that describe what these orga- nizations do to be reliable (Bierly III & Spender, 1995; LaPorte and Consolini 1991; Roberts 1993, 1990; Roberts and Bea 2001; Roberts and Rousseau 1989), during the last two decades, the HRO literature has focused on mindful organizing as being responsible for almost error-free operations (Sutcliffe, Vogus, and Dane 2016; Vogus and Sutcliffe 2012; Weick, Sutcliffe, and

© 2023 informa uk limited, trading as Taylor & francis Group

CONTACT Matteo curcuruto [email protected] Department of human sciences, european university of rome, Via degli aldobrandeschi, 190, 00163, rome, italy.

ARTICLE HISTORY Received 13 June 2022 Accepted 1 December 2023

KEYWORDS Mindful organizing; high-reliability organizations; self-efficacy; safety citizenship behaviour; multi-level analysis

https://doi.org/10.1080/13669877.2023.2293043

86 M. CURCURUTO ET AL.

Obstfeld 1999; Weick and Sutcliffe 2007). Mindful organizing refers to a team’s capability to discern discriminatory details about emerging risks and threats and act swiftly in response to these details (Weick, Sutcliffe, and Obstfeld 1999). In its essence, mindful organising is seen in the actions and interactions of teams, where team members collectively anticipate potential threats and work together to quickly recover from these threats (Sutcliffe, Vogus, and Dane 2016). Studies conducted in other HROs argue that the absence of appropriate levels of mindful organizing can be associated with severe negative consequences for organizations and their stakeholders, such as death as a consequence of medical errors (Weick and Sutcliffe, 2007) or high-profile disasters in the aerospace industry (Weick and Sutcliffe, 2015).

The existing studies in the literature provide evidence about the relevance of mindful organizing to the creation of safer organization. Firstly, previous studies have found significant associations between mindful organizing and individual safety behaviours. In the chemical industry, Renecle et  al. (2021) found a positive association between mindful organizing and safety citizenship behaviours (voice, initiative, and helping). These authors also found a positive relationship between mindful organizing and individual safety compliance, and a negative relationship with safety violations. In addition, longitudinal and multilevel studies conducted in the nuclear sector by Gracia et  al. (2020) found that mindful organizing positively affects safety compliance and safety participation. In another study conducted in the same industry, Renecle et  al. (2020) extended these results showing that mindful organizing was able to predict safety compliance and safety participation above and beyond other team safety-related variables, such as safety culture, team safety climate, and team learning. Secondly, individual safety behaviours are considered to be immediate antecedents of safety outcomes for teams and organizations (e.g. accidents, incidents, etc.) (Christian et  al. 2009, Griffin and Neal 2000). Furthermore, there is some empirical evidence of a negative association between mindful organizing and safety outcomes in the healthcare sector, such as medication errors and patient falls (Ausserhofer et  al. 2013, Vogus and Sutcliffe 2007a, Vogus and Sutcliffe 2007b). Finally, other studies have focused on the role of mindful organizing as a mediator in the relationship between other group safety-related variables and individual safety behaviours. Particularly, empirical evidence exists about the mediator role of mindful organizing in the relationship between team safety climate and safety behaviours (Renecle et  al. 2021), and in the relation- ship between team empowering leadership and safety behaviours (Gracia et  al. 2020). All together these studies are contributing to extending the nomological network of mindful organizing, providing quantitative empirical evidence that was absent only a few years ago (Sutcliffe, Vogus, and Dane 2016). Team safety climate and team empowering leadership are predictors of mindful organizing, and mindful organizing contributes to individual safety behaviours and, eventually to safety outcomes.

However, although the existing studies offer us an insightful framework of the multileveled factors at play that support the overall reliability of organizational systems, to our best under- standing, there is currently a general lack of studies that take into account the psychological mechanisms through which mindful organizing affects individual safety -behaviour. Very little is known about how and why a collective phenomenon such as mindful organizing ends up affecting individual safety behaviours. This is a significant deficiency in the existing literature, considering that if we only consider the contextual antecedents of individual behaviour (e.g. safety climate, empowering leadership), we end up treating the individual as a passive agent within the system, wholly influenced by the social expectations and desired behavioural models of their organization (Parker, Bindl, and Strauss 2010). On the other hand, there is a great body of research that studies the individual as an active element of the system, able to initiate changes and drive improvement, development and resilience (Curcuruto, Mearns, and Mariani 2016; Hollnagel 2014). This research stream shows that multiple psychological mechanisms that drive individuals to act as proactive agents for the promotion of safety in their organization.

JOURNAL OF RISk RESEARCH 87

In our study we will shed some light on the psychological mechanisms responsible for the association found in previous studies between mindful organizing and individual safety behaviours. Drawing on proactive motivation theory (Parker, Bindl, and Strauss 2010) we introduce the construct of participative safety self-efficacy, that refers to employees perceived capability of carrying out a broader and more proactive, interpersonal and integrative set of work tasks and goals to do with safety beyond individual prescribed requirements (Curcuruto, Mearns, and Mariani 2016). We aim to investigate how mindful organizing affects participative safety self-efficacy, introduced in our study as a psychological condition that could motivate individuals to engage in constructive behaviours of relevance for safety critical contexts, with special attention to the proactive forms of individual contribution to the promotion of safety in the workplace, like safety citizenship behaviours (acronym: SCBs), such as: (1) personal initiatives for the improvement of workplace safety, (2) helping coworkers with safety related responsibilities included in their job, or (3) voicing personal safety concerns about workplace issues that can represent (or create) potential threats for the safety of individuals teams and their organizational system. Specifically, we develop and test a model where participative safety self-efficacy is proposed as a mediating variable in the relationship between mindful organizing and individual safety behaviours (see Figure 1). These proposals will be studied by conducting a multilevel structural equation model using data from 50 teams and 443 chemical plant workers.

There are at least two main contributions of our study to the advancement of literature on SCBs. Traditionally, the emergence of safety citizenship is explored in relation to constructs like safety climate, organizational support, leader-member-exchange, constructs that refer to the existing vertical relationships between the employees and their superiors, and/or between the employees and the organization itself (Curcuruto and Griffin, 2018). Mindful organizing refers to a set of teamwork processes that are developed at a group level of analysis and that are developed through daily peer-to-peer social interactions among coworkers. Investigating SCBs as the outcome of within-group interactions between colleagues is something relatively new in safety literature (Curcuruto et al. 2019a; 2019b). By investigating the role of mindful organizing in the emergence of safety citizenship, we aim to extend the research on safety citizenship to incorporate the analysis of group processes that go beyond the ones usually explored in liter- ature. For example, organizational rules and norms (i.e. safety climate), or social exchange processes between their employees with their supervisors or the overall organization (i.e. leader-member-exchange and organizational support).

Figure 1. research model.

88 M. CURCURUTO ET AL.

We believe that in literature the investigation of the psychological mediators that are usually analysed by the researchers to explain the emergence of safety citizenship behaviour is usually limited to the examination of the role of constructs like safety knowledge and safety motivation. Consequently, by investigating the role of participative safety self-efficacy we aim to enlarge the focus of the research on the psychological mediators facilitating the emergence of safety citizen- ship. This contribution appears particularly relevant because employees’ proactive role in safety promotion is currently well recognised in literature as a reliable predictor of positive risk man- agement in organizations (Curcuruto et al. 2019a; Hollnagel 2014). Therefore, we aim to contribute to filling this gap by exploring how the teamwork processes of mindful organizing influence the emergence of safety citizenship through the mediation of employees’ participative safety self-efficacy.

In the next sections, we will present a review of the conceptual foundations of mindful orga- nizing, and how it is supposed to facilitate individual and team reliability and commitment in workplace safety management. Then, a set of research hypotheses will be discussed for the advancement of our understanding of how mindful organizing affects a broad range of individual work behaviours of relevance for the maintenance of safety in daily operations and for the con- stant improvement of the organizational safety system. Thereafter, we present our empirical study where we run a multilevel structural equation model to examine the relationship between mindful organizing self-efficacy and safety outcomes in a sample of fifty teams operating in a large chem- ical plant. We then discuss the implications of our study results for research advancement and the practical implications of these findings for decision makers in high-risk industries.

2.  Conceptual background: mindful organizing and workplace safety

The concept of mindful organizing is linked to the work of karl Weick and colleagues, and their research into how HROs managed to achieve almost error-free performance under such trying conditions (Weick and Roberts 1993; Weick, Sutcliffe, and Obstfeld 1999; Weick and Sutcliffe 2007). In the next lines, we will summarize the main contributions of this work. These authors observed that HROs had a different social and relational infrastructure to other kinds of orga- nizations. They discovered that teams in effective HROs engaged in ‘heedful interrelating’. This ‘heedful interrelating’ meant that teams were highly attentive in their actions and interactions with one another. Further research into these highly attentive actions and interactions showed that it allowed teams to have an expanded understanding of the system in which they operated. This expanded understanding of the system was also linked to a wider range of possible responses to novel or unexpected situations. This meant teams were able to manage the unex- pected and contain errors far more effectively than teams operating in other high-risk environ- ments. They called this team phenomenon mindful organizing. Mindful organizing was then defined as the collective capability to detect discriminatory details about emerging issues and act swiftly to respond to such details. The detection of discriminatory detail about emerging issues allowed teams on the front line to anticipate potential errors, anomalies, or unexpected events. The ability to act swiftly in responding to these errors, anomalies, or unexpected events allowed these teams to recover from, or contain, these possibly problematic events. These definitions appear to conceptualise mindful organizing as a two-factor variable, with the ability to anticipate errors, anomalies, and unexpected events as the first factor and the ability to act swiftly to contain these events as the second factor. However, the analysis of this collective capability through case study analyses of effective HROs showed that mindful organizing was enacted by five interrelated practices and attitudes. They are: (1) a preoccupation with error, (2) a reluctance to simplify interpretations, (3) a sensitivity to operations, (4) a commitment to resilience and (5) deference to expertise. It appeared that the first three processes underpinned a team’s capability for anticipation and the last two processes underpinned a team’s capability for containment and recovery.

JOURNAL OF RISk RESEARCH 89

2.1.  The five characteristic processes of mindful organizing

In this section, we conceptually delve into each one of the five processes that constitute mindful organizing. Although mindful organizing first appeared in the article by Weick, Sutcliffe, and Obstfeld (1999), the most elaborated description of the five characteristic processes of mindful organizing comes from the three editions of the book ‘Managing the Unexpected’ written by Weick and Sutcliffe (2001, 2007, 2015). These dimensions are explained below.

2.1.1.  Preoccupation with error Teams that engage in mindful organizing are preoccupied with errors. This means that teams are always concerned about potential or actual mistakes that they can generate. This concern is manifested through observable activities enacted by the team members, such as spending time and effort trying to anticipate everything that could go wrong, or emphasizing the impor- tance of detecting and reporting errors (Rochlin, 1993)), or taking any error or near-error very seriously as it could indicate any larger problem underlying the surface of work operations. Overall, this sort of chronic concern with errors is an essential practice for anticipating potential threats and unexpected events within a system, and strongly influence both safety attitudes and behaviours of team members, leading the team to remain cautious and attentive at all times (Schulman 1993), always treating small deviations and mistakes seriously, as they could potentially mean a bigger problem elsewhere in the system (LaPorte and Consolini 1991).

2.1.2.  Reluctance to simplify interpretations This concept means that the team tries to actively avoid simple analyses of complex phenomena as it could lead to incorrect conclusions. Mindful organizing encompasses team activities such as: refraining from making assumptions or drawing conclusions too quickly when interpreting and diagnosing what is happening in their environment (Schulman 1993); paying attention to new evidence or information that a situation has changed, rather than relying on old explana- tions when making sense of something new or unexpected at work; encouraging rich exchanges of points of view to be able to have a more complete picture of the situation; or reinforcing a questioning attitude in all the members of the team when interpreting what is happening in their workplace (Rochlin, 1993). Overall, this component of mindful organizing helps teams to gain as much information about what is going on in their work, especially regarding unex- pected events or errors (Weick and Sutcliffe 2007). This safeguards teams, to a certain extent, from coming to incorrect conclusions about the causes or consequences of unexpected events that can lead to wrong decisions, errors and mistakes with potentially catastrophic consequences

2.1.3.  Sensitivity to operations Teams that organize mindfully are also sensitive to operations. This means that teams and leaders strive to remain aware of the reality of what is happening in their work operations at any given moment (Rochlin, 1993). In showing sensitivity to operations, teams constantly engage to be updated on the details of current operations and the big picture status of their work, constantly communicating with the higher organizational management levels about the intri- cacies of current operations. At the same time, sensitivity to operations is also sustained by leaders’ actions, where leaders are committed to remaining in touch with the reality of operations happening on the front-line. This has similarities with the concept of ‘work as done’ versus ‘work as imagined’ discussed by Hollnagel (2014) as workers will constantly update management on the realities of how work is actually done. Thanks to the efforts deployed by both the team members and their leader, sensitivity to operations allows teams to remain aware of the import- ant intricacies of operations within the system that affect their work (Weick, Sutcliffe, and

90 M. CURCURUTO ET AL.

Obstfeld 1999). The connectedness of the team with others in the system coupled with an awareness of what is happening elsewhere, allows team members and leaders to quickly detect and communicate any important information as it happens (Weick and Sutcliffe 2007). Sensitivity to operations is made observable by team communication practices and entails regular contact and communication exchanges with the team leader.

2.1.4.  Commitment to resilience Teams that engage in mindful organizing are committed to resilience. Resilience means being able to bounce back from adverse events and continue to operate normally. This is seen in teams being able to quickly recover and maintain the stability of the system through flexibly using a wide range of responses. Therefore, commitment to resilience has to do with essential actions and practices that help teams in recovering from mishaps, errors or unwanted surprises (Weick, Sutcliffe, and Obstfeld 1999). Among these practices, there is a further distinction between ‘preparing for resilience’ and ‘acting resiliently’. Work practices aimed to ‘prepare for resilience’ include training, simulations and learning from errors. These practices are carried out to expand team members’ knowledge, skills and capabilities to better deal with unexpected events so that they are better equipped to correct and contain these events before they desta- bilize the system (Weick and Sutcliffe 2007). On the other side, ‘acting resiliently’ has to do with teams having the capability to deploy adequate resources and flexible strategies that allow them to recover from mistakes and unexpected events as they arise, assuring the maintenance of the stability within the system (Weick and Sutcliffe 2007). The concept of commitment to resilient action in safety-critical industries and HROs has been well documented and has been a central feature of both the engineering and human resources discourse in high-risk industries.

2.1.5.  Deference to expertise Engaging in mindful organizing means that teams defer to expertise. This entails that when facing unexpected events, decision-making migrates to those in the team with the best exper- tise rather than those with the highest rank. Deference to expertise is primarily developed through the production of mutual knowledge among the members of the team of each member’s knowledge and capabilities, so they know who to call on to help make decisions when facing an unexpected event or novel situation. This entails that when these situations happen, ‘experts’ within the system are called upon to help make decisions, independently from the role in the organizational hierarchy. In other words, deference to expertise refers to the practice of decisions migrating to those with the best expertise, rather than the highest rank, in the face of unexpected events or crises, empowering them to make decisions during unexpected events (Roberts, Stout, and Halpern 1994; Weick, Sutcliffe, and Obstfeld 1999). In practice, sometimes the workers who are closest to the potential problem take on the respon- sibility of the decision-making (e.g. air-traffic controllers), interpreting and managing the unexpected event in reason of their first-hand knowledge and local understanding of the causes and implications of the problem. Some other times, expert decision-making is driven by networks of people with a diversity of expertise making decisions together. This expertise could come in from the previous experience and educational backgrounds of the team mem- bers, or even pooling of various capabilities in networks, allowing the team to make better decisions.

Since its inception, these five processes of mindful organizing have been validated and applied in various studies across different sectors. The model has been explored in theoretical articles (e.g. Gajda 2018; Gebauer 2013; Martínez-Córcoles and Vogus 2020; Vogus 2011; Vogus and Sutcliffe 2012) and empirical studies (e.g. Dernbecher, Risius, and Beck 2014; Ndubisi and Al‐Shuridah 2019; Renecle et  al. 2020; Vogus and Sutcliffe, 2007a). For example, Gebauer (2013) explored how the principles of mindful organizing could be used in management development

JOURNAL OF RISk RESEARCH 91

programs to encourage self-observation and high reliability seeking. Gajda (2018) proposed a theoretical framework in which mindful organizing (directly) and organizational mindfulness (indirectly) enhance individual talent management outcomes(e.g. motivation to work, organi- zational commitment and extra-role behaviours) resulting in better company performance. Examples of empirical research on mindful organizing include the study conducted by Dernbecher, Risius, and Beck (2014), who define mindful organizing as a bottom-up construct emerging from the employees and organizational mindfulness as a top-down strategic process enacted by top management. When operationalising these definitions according to hierarchical job role, they found a significant positive influence of a differentiated effect of both, mindful organizing and organizational mindfulness, as well as a highly significant positive effect of the combination of both on the job performance of workers in a mobile work environment. In a later study by Ndubisi and Al‐Shuridah (2019), they also defined mindful organizing and organizational mind- fulness as two separate constructs. Their analysis of data collected from 92 Saudi firms within the oil and gas industry suggested that mindful organizing is significantly related to environ- mental and resources sustainability, and it fully or partially mediates in the relationship between some of the dimensions of organizational mindfulness and these sustainability outcomes. Other empirical work has focused on the adaptation and validation of measurement scales to oper- ationalise mindful organizing to specific industrial and national contexts. For example, Renecle et  al. (2020) validated a unidimensional Spanish version of the Mindful Organizing Scale utilising nuclear power plant workers.

However, recently, Martínez-Córcoles and Vogus (2020) provide a contemporary overview of the topic area noting criticisms concerning the mixed views on what distinguishes mindful organizing, conceptually from the related concept of organizational mindfulness, and the con- sequent difficulties that derive from this conceptual ambiguity in creating and sustaining it in practice. This specific conceptual aspect is addressed in the next two subsections of conceptual background.

2.2.  Mindful organizing, individual mindfulness and organizational mindfulness

Mindful organizing is different from individual mindfulness and from organizational mindfulness. Vogus and Sutcliffe (2012) stress the importance of distinguishing mindful organizing from related mindfulness concepts such as organizational mindfulness and individual mindfulness, as they may seem similar but are theoretically and operationally different.

Individual mindfulness is the most widely studied and best understood of all the mindfulness constructs. It refers to a state of consciousness where attention is focused on events occurring in the present moment: both internally and externally (Dane, 2011). It is a mental activity or a state of concentration that occurs in one’s mind. However, the term ‘mindful’ in mindful orga- nizing follows Langer’s (1989) conceptualisation of mindfulness on an individual level. Langer (1989) posits that a mindful state comes from actively differentiating and clarifying existing categories and distinctions which creates new disconnected categories out of the connected series of events that happen in one’s work or life. From this, a more nuanced appreciation of context and alternative ways of dealing with one’s context arises. This conceptualization of mindfulness argues that mindfulness is just as much about what we do with what we notice in our ‘state of concentration’ as it is about the act of noticing itself. Mindful organizing found in HROs is characterised by noticing weak signals before critically analysing and reframing such signals leading to an enlarged understanding of what is noticed (Weick, Sutcliffe, and Obstfeld 1999). This enlarged understanding of what is noticed is closely linked to a repertoire of action capabilities which is a defining feature of what makes HROs effective (Westrum, 1988).

The key difference between mindful organizing and individual mindfulness is that mindful organizing is not an intra-psychic process that occurs in the minds of individuals (Morgeson

92 M. CURCURUTO ET AL.

and Hofmann 1999); rather, it is an emergent, collective process that is seen in the actions and interactions of team members (Vogus and Sutcliffe, 2007a). Mindful organizing is a social process of organizing in such a way that sustains attention to salient stimuli that may pose a threat to the operation of the organization, sparking corrective action (Vogus and Sutcliffe 2012). It can be seen and recorded in the conversations, interactions, and actions of team members. Mindful organizing is also different from organizational mindfulness (Vogus and Sutcliffe 2012).

Organizational mindfulness is more similar to mindful organizing than individual mindfulness as it is also a collective capability to anticipate and recover from unexpected events. However, organizational mindfulness is a strategic top-down construct which is more enduring in an organization as it is brought about through the practices, strategies, and structures put in place by top management (Vogus and Sutcliffe 2012). In contrast, mindful organizing is a bottom-up collective process enacted mainly but not only by those on the front line; it is fragile and needs constant reinforcement (Weick and Sutcliffe 2007).

In our paper, we focus on mindful organizing (not on individual or organizational mindful- ness). As it is a team process, and a collective construct, it is valuable to study mindful orga- nizing at the team level, rather than at the individual level (Vogus and Sutcliffe 2007b). However, it is important to understand how mindful organizing emerges from individual properties and their implications for operationalization.

2.3.  The nature of the emergence process and operationalization of mindful organizing

Multilevel models in organizational and social sciences frequently involve higher-level (e.g. team) constructs that have their origin in lower-level (e.g. individual) properties. To fully understand the nature of higher-level constructs (i.e. mindful organizing), it is of utmost importance to explain the processes through ‘which lower-level properties emerge to form collective phenom- ena’ (kozlowski and klein 2000, 15).

Mindful organizing is a shared unit property, meaning that it1) represents phenomena that span two or more levels, 2) originates at lower levels (i.e. individuals) but are manifest as higher-level phenomena (i.e. team), 3) emerges from the characteristics, behaviours, or cognitions of unit members, and their interactions-to characterize the unit as a whole, and 4) is, essentially, similar across levels (that is, isomorphic), representing composition forms of emergence.

The literature on mindful organizing suggests it only exists to the extent that it is collectively enacted (Levinthal and Rerup 2006; Vogus and Sutcliffe 2007a, 2007b; Weick and Sutcliffe 2007). One way to assess the extent to which a set of behaviours is customarily enacted is whether there are shared perceptions regarding the prevalence of the behaviours (Morgeson and Hofmann 1999). Vogus and Sutcliffe (2012) argue that behaviours and perceptions about mindful orga- nizing are likely to converge and coalesce among team members for at least two reasons. First, bottom-up attraction–selection–attrition processes (Schneider 1987) can improve the similarity in members’ mindful organizing by favouring the attraction, selection and retention of new members that express similar attitudes and behaviours to those exhibited by the older members. Second, task interdependence and time working together can increase the homogenizing effects of social influence and social learning by creating continual opportunities for work-related interactions.

In the operationalization of mindful organizing, we have followed the two general recom- mendations for the measurement of shared unit properties, that is, to focus respondents on description as opposed to evaluation of their feelings and, to use items that reference the higher level, not the level of measurement. Therefore, the subject and content of all mindful organizing scale items refer to team level practices and behaviours but they are rated by indi- viduals. Because mindful organizing is conceptualised as a shared unit property, an essential part of creating empirical evidence to back up the theoretical understanding of mindful orga- nizing is to show that individual team member’s mindful organizing scores can be aggregated

JOURNAL OF RISk RESEARCH 93

to the group level (Sutcliffe, Vogus, and Dane 2016). Aggregating individual responses about team level practices and behaviours to create a team score is meaningful provided that adequate consensus is found between individual scores. We will provide empirical evidence about this issue in the method section.

3.  Research hypotheses: self-efficacy as a mediational link between mindful organizing and individual safety behaviour

Recent studies (Gracia et  al. 2020; Renecle et  al. 2020, 2021) showed how mindful organizing positively influences employees’ engagement in safety participation and extra-role behaviours supporting workplace safety. These studies provide evidence that mindful organizing serves as a teamwork level mechanism that enables the team to translate managerial safety values and priorities into observable safety behaviours (Renecle et  al. 2021). On the other hand, what is still relatively under-investigated is the nature of the psychological mechanisms that translate team mindful organizing in these extra-role behaviours. In order to contribute to filling this conceptual gap in the safety research literature, we referred to the theory of proactive motiva- tion (Parker, Bindl, and Strauss 2010). This conceptual framework explains which kinds of psy- chological states support individuals’ propensity to engage in proactive behaviours which are also known as safety citizenship behaviours (Conchie 2013), and that are not part of the employees’ formal job description (Griffin and Curcuruto 2016),

According to proactive motivation theory, a prominent psychological driver of proactive behaviour is an individual’s perceived capability to achieve short term, proactive goals. In high-risk contexts rife with unexpected events, it can be daunting to engage in safety citizen- ship behaviours such as initiating changes, voicing concerns or taking the lead in managing safety by helping or guiding others to be safer in the moment. Believing in one’s own ability to be able to successfully carry out these daunting activities is likely to be a powerful moti- vator for engaging in these activities. Therefore, the present study wanted to examine whether individual capability drivers such as self-efficacy played a role in facilitating individual safety citizenship behaviours in a context where teams engage in mindful organizing. In particular, we wanted to examine whether self-efficacy played an important role in mediating the rela- tionship between team mindful organizing and individual safety behaviours.

Participative safety self-efficacy refers to ‘employees perceived capability of carrying out a broader and more proactive, interpersonal and integrative set of work tasks and goals to do with safety beyond prescribed requirements (Curcuruto, Mearns, and Mariani 2016). An important distinction to make is that this safety-specific form of self-efficacy does not merely refer to an individual’s capability, knowledge and skills to comply with the safety prescrip- tions in place in the organization. Rather, it refers to an individual’s confidence to perform extra-role behaviours such as analysing safety issues to propose solutions, coming up with new methods to improve safety, helping to facilitate safety goals in team, or discussing with others how to improve safety conditions in the workplace (Curcuruto et al. 2019a).

Engaging in the five processes of mindful organizing boosts a team’s ability to understand and diagnose the risks they face (through the anticipation processes) as well as enhances a team’s ability to successfully navigate unexpected events and contain errors (through the con- tainment processes) (Vogus 2011). We believe that individuals who form part of a team that is able to collectively manage unexpected events and small errors effectively are likely to develop more confidence in their individual ability to fulfil their extra-role tasks to enhance safety. This increased participative safety self-efficacy is likely to lead to higher proactivity to carry out safer practices in the organization such as engaging in helping, voice and initiative.

We posit that the anticipation processes of mindful organizing (preoccupation with error, reluctance to simplify and sensitivity to operations) will lead to higher participative safety

94 M. CURCURUTO ET AL.

self-efficacy to voice safety concerns to others. Preoccupation with error entails teams contin- uously searching for, detecting and voicing concerns about potential errors and anomalies (Weick and Sutcliffe 2007). Reluctance to simplify entails challenging assumptions and trying to uncover blind spots in operations through rich discussions about possible categories and labels (Schulman 1993). Sensitivity to operations means teams make sure to be aware of the realities of operations on the front line and communicate these challenges and realities to one another and leaders (Weick and Sutcliffe 2015). These three actions and activities increase the range of situations that each individual team member becomes more self-assured to address and discuss, increasing their confidence to correctly identify, and voice, a wide range of safety issues. This increased participative safety self-efficacy is likely to motivate these team members to engage in voicing safety concerns to others on their own accord, over and above mindful organizing and what is required by their formal job description. Therefore, the following is hypothesized:

Hypothesis 1: Participative safety self-efficacy mediates the relationship between mindful organizing and voice so that the relationship is positive and significant.

We argue that the containment processes of mindful organizing (commitment to resilience and deference to expertise) will lead to an increased individual safety self-efficacy to start safety related initiatives on an individual level, like initiating changes to ensure safer practices. On one side, commitment to resilience has to do with growing team capabilities to quickly recover from unexpected events so teams can act swiftly and make changes to bounce back from errors (Weick and Sutcliffe 2015). This group capability may stimulate employees’ self-confidence to engage in initiatives to improve the current work practices to make them safer. On the other side, deference to expertise has to do with the shared knowledge in the workgroup about the expertise of each member of the team, which ensures that the best expertise available in the team is utilised to cope with problems that may threaten safety within the workplace (Roberts, Stout, and Halpern 1994). We hypothesized that when such team dynamics exist, where the members of the group feel their expertise is valued by their peers and superiors, employees will develop a stronger sense of participative safety self-efficacy. Through this empowered self-confidence, they will be more motivated to engage in personal initiatives to improve the safety conditions in the workplace, like proposing suggestions to the organizations to improve the work practices and the work procedures to achieve better management of safety problems. This might be particularly relevant for organizations, because work operators are those who see most of the reality of operations and are the closest to the potential sources of problems for workplace safety, therefore, they are the ones with the best expertise in the matter (Weick and Sutcliffe 2015). In summary, we believe that mindful orga- nizing, through its containment processes of commitment to resilience and deference to expertise is likely to increase an individual’s confidence in their own ability to initiate changes to ensure a safer workplace, and this increased confidence in their capability to initiate these actions, will then lead to them engaging in initiating changes to increase safety. Therefore the following is hypothesized:

Hypothesis 2: Participative safety self-efficacy mediates the relationship between mindful organizing and initiative so that the relationship is positive and significant.

Mindful organizing creates a broader awareness of the work and knowledge of others in a team (through sensitivity to operations, commitment to resilience and deference to expertise), which is likely to enhance each individual’s understanding of which team members are likely to need support or help with safety protocol and practices. This, coupled with the knowledge and experience in managing safety that comes from engaging in mindful organizing continu- ously as a team is likely to build individuals perceived confidence in successfully helping the less experienced to follow and achieve safety goals. The enhanced participative safety self-efficacy

JOURNAL OF RISk RESEARCH 95

to engage in extra role helping will increase an individual’s propensity to actually reach out to less experienced or knowledgeable colleagues to assist them with safety related matters. Therefore, the following is hypothesized:

Hypothesis 3: Participative safety self-efficacy mediates the relationship between mindful organizing and helping so that the relationship is positive and significant.

The anticipation processes entailed by mindful organizing (i.e. preoccupation with failure; reluctance to simplify interpretations; sensitivity to operations) support the collective capability of the workgroup to anticipate unexpected events that can jeopardize employees’ health and safety. As discussed above, mindful organizing enforces the confidence of team members to engage in a course of actions that promote a safer workplace, through the mediation of safety self-efficacy. While at the individual level of analysis this mediational influence is expressed by the emergence of safety citizenship behaviours (i.e. helping, voice, initiative), we propose that at the group level this mediational influence will result in a higher compliance with safety standards and a minor level of violations. Assuming the anticipatory nature of mindful orga- nizing, we expect that groups characterised by high levels of mindful organizing will be char- acterised by a stronger awareness of the risks associated with the lack of compliance with safety standards and procedures, like accidents or injuries. Part of the construct of participative safety self-efficacy refers to individual self-confidence to support the workgroup in achieving the safety goals of the team. We expect that workgroups characterised by high levels of mindful organizing will be characterised as well by higher levels of safety compliance and lower levels of safety related violations. We hypothesize that these relationships will be mediated by the employees’ feelings of safety-specific self-efficacy, as we expect that employees presenting higher levels of participative safety self-efficacy will be more motivated to contribute to achieving the team goal of reducing the accident rates in the work activities. In other words, in a group context where its members develop high participative safety self-efficacy from engaging in mindful organizing, individuals will be highly committed to upholding safety procedures and rules. We therefore hypothesize the following:

Hypothesis 4: The relationship between mindful organizing and safety compliance is mediated by partici- pative safety self-efficacy, and this mediated relationship is positive and significant.

Hypothesis 5: The relationship between mindful organizing and safety violations is mediated by participative safety self-efficacy, and this mediated relationship is negative and significant.

4.  Method

4.1.  Sample and procedure

The data used in this research was collected within a large sample of Ukraine-based chemical plant workers (N = 443) identifying 50 teams. All participants were employed in a single large chemical industrial facility deputed to the manufacturing, treatment, refinement and storage of vegetable fibres. A significant part of the production processes in this kind of facility is auto- mated, and the functioning of the machinery and manufacturing lines contemplated a design of the work activities allocated to work teams composed of a variable number of employees, with many teams working simultaneously at different points of the manufacturing lines, and under a periodic shift rotation schedule. The members of each workgroup reported to a single team leader, who in turn reported directly to a middle manager of the department division. In terms of risks for health and safety of the workforce, different sources of hazards include per- sonal exposure to biological agents (bacteria, viruses, parasites), exposure to chemical agents (nicotine, ammonia, dehydrogenated alcohol), fire risk and exposure to flammable products, as well as injury risks in the usage of the machinery.

96 M. CURCURUTO ET AL.

Participation was voluntary and all workers were informed that the data would be used for scientific research and to gain insight into safety culture improvements in each plant. The majority of participants (60%) had been working in the company for more than 10 years, 33% had been working in the company for 5 to 10 years, 3% had been working in the company for 2 to 5 years, 2% had been in the company for less than 5 years and 2% did not indicate their tenure in the company. Participants were employed in primary and secondary production (30%), the filter production workshop (12%), the warehousing department (15%), quality assurance department (13%), the engineering department (8%) and 22% came from other departments. The questionnaire was administered in Russian using the same scales created and translated through back-translation.

4.2.  Measures

4.2.1.  Mindful organizing Mindful organizing is a team’s collective capability to anticipate and contain errors and unex- pected events. Mindful organizing was measured using a nine-item scale (α = .94) taken from Vogus and Sutcliffe (2007b). Participants were asked to report their personal agreement with a set of statements referring to complimentary team-working aspects supporting the five mindful organizing processes. Responses were collected on a five-point Likert scale (1 = strongly disagree; 5 = strongly agree). Example items are ‘We talk about mistakes and ways to learn from them’, ‘We spend time identifying activities we do not want to go wrong’, ‘When attempting to resolve a problem, we take advantage of the unique skills of our colleagues’, ‘We have a good “map” of each other’s talents and skills’, ‘We discuss alternatives as to how to go about our normal work activities’. The suitability of the content of the items with the group activities performed in the plant was verified with a group of workers’ representativeness before the administration of the survey.

4.2.2.  Participative safety self-efficacy Participative safety self-efficacy is the confidence individuals have in their own ability to carry out a more participative and broader set of safety tasks beyond formalised role requirements. In the present study, it was measured using a 5-item safety-specific scale of role breadth self-efficacy (α = .93) adapted to safety specific contents by Curcuruto, Mearns, and Mariani (2016) from the original scale developed by Parker (1998). Participants were asked to report their personal judgement about the extent they perceived themselves confident with engaging in a set of extra-role actions supporting the promotion of workplace safety. Responses were collected on a five-point Likert scale (1 = not confident at all; 5 = highly confident). Examples of the content of the items are ‘Feeling confident in devising new methods to improve safety in my work area’, ‘Feeling confident in setting up and achieving safety objectives of my group’, and ‘Feeling confident in analysing recurring problems regarding safety in order to suggest solutions’.

4.2.3.  Safety citizenship behaviours Safety citizenship behaviours (SCBs) are discretionary and prosocial activities essential for man- aging risk in safety critical industries (Curcurutoet al 2019b). For the present study, we analysed three SCBs, namely: voice, initiative and helping. These forms of safety citizenship were assessed with three scales originally created by Hofmann, Morgeson, and Gerras (2003). Participants were asked to rate the frequency to which they engaged in these three forms of safety citizenship. Responses were collected on a five-point Likert scale (0 = never; 4 = very frequently). More spe- cifically, voice was measured using a 4-item scale (α = .92). An example of item is ‘voluntarily

JOURNAL OF RISk RESEARCH 97

raising safety concerns in planning sessions’. Initiative was measured using a 4-item scale (α = .87), and an example of an item is ‘voluntarily trying to make policies and procedures safer’. Finally, helping was measured using a 6-item scale (α = .90). An example item is ‘voluntarily helping with teaching safety procedures to newest crew members’.

4.2.4.  Safety compliance and safety violations Safety compliance is the degree to which an individual complies with the safety protocol of the chemical plant. Safety violation refers to the extent to which an individual deliberately violates safety protocol. Both scales were taken from Hansez and Chmiel (2010), and participants were asked to report the frequency they had recently engaged in examples of safety compliance and violation of safety standards. Responses were collected on a five-point Likert scale (0 = never; 4 = very frequently). More specifically, safety compliance was measured using a 5-item scale (α = .82). An example item is ‘using protection devices, even if it is hard to find them’. Safety violation was measured using a 5-item scale (α = .94). An example item is ‘neglecting some safety rules when performing familiar or routine work’.

4.3.  Analyses

To test our proposed model, we ran a multilevel structural equation model (MSEM). Mindful organizing was analysed on the team level while participative safety self-efficacy, safety com- pliance, safety violation and the SCBs were analysed on the individual level. First, confirmatory factor analyses (CFA) of the seven scales (mindful organizing, participative safety self-efficacy, safety compliance, safety violation, voice, initiative, and helping) were carried out in order to gain evidence of the discriminant validity of these measures. A seven-factor model with all the items loading onto seven separate factors using individual level data was run with Mplus (Muthén and Muthén 2017). Thereafter, four alternative CFA models were conducted, and the fit of these models was compared with the seven-factor model. The alternative models are: (1) a one factor model with all the items of the seven scales loading onto one single factor, (2) a six factor model with mindful organizing and role breadth self-efficacy both loading onto the same single factor and all the other items loading onto their corresponding factors, (3) a five factor model with the three SCBs (voice, initiative, and helping) loading onto the same single factor and all the other items loading onto their corresponding factors, (4) a six factor model with safety compliance and safety violation both loading onto the same single factor and all the other items loading onto their corresponding factors and (5) a four factor model with the three SCBs (voice, initiative, and helping) loading onto the same single factor, safety compliance and violation loading on to the same factor and mindful organizing and participative safety self-efficacy loading onto their corresponding factors. Model fit was evaluated by calculating the chi-square statistic, the root mean square error of approximation (RMSEA; Steiger 1990), the comparative fit index (CFI; Bentler, 1990) and the Tucker Lewis index (TLI; Tucker and Lewis 1973). RMSEA values below .05 indicate good fit, values of between .08 and .05 show a rea- sonable error of approximation and values of .10 or more indicate poor fit, (Browne and Cudeck 1993; Browne and Du Toit 1992). For the CFI values, values above .90 are considered acceptable fit and values close to 1 indicate good fit (Hu and Bentler 1999). TLI values near 1 indicate good fit, with the conventional cut off being .90 for acceptable fit (Tucker and Lewis 1973). When comparing alternative models, we used the following criteria: (1) whether the differences between TLI and CFI values of the competing models were larger than .01 (Cheung and Rensvold 2002), and (2) whether the differences between RMSEA values were larger than .015 (Chen et al. 2008). These criteria indicate whether there is a notable disparity between the models and when these differences in practical fit indices are detected, the model showing better fit will be selected. Additionally, the difference in chi-square statistics along with the difference in

98 M. CURCURUTO ET AL.

degrees of freedom was also used as a criterion to check for statistically significant differences among competing models. If the difference is significant, the model with the smaller chi-square value is argued to have better fit to data.

Second, the aggregation indices (average deviation indices (ADIs), Rwg values, intraclass correlation coefficient ICC(1)) and ANOVAs, were calculated for mindful organizing to evaluate the within group agreement and between group discrimination, respectively.

Third, we ran a multilevel structural equation model to assess our proposed mediation model and the pathways between our variables. Monte Carlo (MC) confidence intervals were used for testing the significance of the indirect effects, as it is argued to be a more viable and robust method for calculating confidence intervals for complex and simple indirect effects when working with a multilevel model.

5.  Results

Descriptive statistics and the correlations between the measures of the study variables can be found in Table 1. As expected, the measure of participative safety self-efficacy presented sig- nificant relationships with all the three forms of safety citizenship (voice, initiative and helping), a moderate, but positive, correlation with safety compliance, and finally a moderate negative correlation with safety violation. Following previous research conducted by Curcuruto et  al. (2019b) across various multi-national samples using these same measures, we kept them as separated indicators of distinct forms of safety citizenship. Finally, in the present sample, the measure of safety compliance showed significant correlations with the three forms of safety citizenship, and as expected, a substantial negative correlation with safety violation.

5.1.  Confirmatory factor analysis

Before testing our research hypotheses, confirmatory factor analysis (CFA) were carried out to evaluate the goodness of our measurement factor model in the present sample. Table 2 shows the goodness of fit indices for alternative models tested in our analyses. We examined the distinctiveness of the seven study variables through a seven-factor model (with all seven vari- ables in the study loading onto seven separate factors) and compared the fit of this model with five alternative models.

The differences between the theorised seven-factor model and the alternative model 1 (ΔRMSEA = .07, ΔCFI = .39, ΔTLI = .41), alternative model 2 (ΔRMSEA = .02, ΔCFI = .08, ΔTLI = .09), and alternative model 4 (ΔRMSEA = .01, ΔCFI = .04, ΔTLI = .04) were notable, indicating that the seven-factor model had a better fit to the data. The differences between the theorised seven-factor model and alternative model 3 (where initiative, voice and helping loaded onto a

Table 1. Descriptive statistics and correlations among observed variables (N = 488).

Variable M SD 1 2 3 4 5 6

1. Mindful organizing

4.01 .66 --

2. safety self-efficacy

4.10 .70 .61** --

3. safety compliance

4.69 .48 .37** .39** --

4. safety Violation 1.36 .73 −0.24** −0.20** −0.48** -- 5. Voice (scB) 3.36 .96 .54** .59** .27** −0.10* -- 6. initiative (scB) 3.29 .93 .50** .55** .26** −0.04 .78** -- 7. helping (scB) 3.52 .96 .59** .56** .30** −0.15* .81** .72**

note. * p < .05, **p < .001.

JOURNAL OF RISk RESEARCH 99

single factor) were notable for the CFI and TLI values (ΔCFI = .02, ΔTLI = .02), however, there were no relevant differences in the RMSEA values (.06). Therefore, we examined the difference in chi-square values for the theorised seven-factor model and the alternative model 3, and we found a statistically significant difference (Δχ2 = 153.96, Δdf = 11, p < .001). Given that the theorised seven-factor model had a smaller chi-square value, we concluded that it was the best fitting model. Thus, the evidence above supported the discriminant validity of the seven scales.

5.2.  Aggregation indices

The results of the within-team agreement and inter-rater reliability analyses for mindful orga- nizing provided adequate justification for aggregating the data at the team level. The average ADI value was .50 (SD = .19), which is below the .83 cut off for a 5-point Likert-type scale (Burke and Dunlap 2002). The rwg(J) value was .94, indicating strong within team agreement (LeBreton and Senter 2008). The ICC(1) value was .09, which is above the recommended .05 cut-off (Bliese 2000). Additionally, ANOVA results for mindful organizing (F (49,379) = 1.80, p < .05) indicated adequate between-team discrimination.

5.3.  Multilevel analysis of the study model

The results of the MSEM analysis indicated that the hypothesized multilevel mediation model showed a satisfactory fit (χ2 = 0.61, df = 5, p >.05; RMSEA = 0.00; CFI = 1.00; TLI = 1.00; SRMR-within = .001; SRMR-between = .015). All hypothesized pathways were significant (see Figure 2).

Regarding the multilevel mediation, at the team level (between level), mindful organizing had a positive statistically significant indirect effect (IE) on voice (IE = 0.84, p < .001, MC CI = 0.09, 2.14), initiative (IE = 0.68, p < .001, MC CI = 0.16, 1.18) helping (IE = 1.00, p < .001, MC CI = 0.20, 2.31) and safety compliance (IE = 0.31, p < .001, MC CI = 0.11, 0.55) through partic- ipative safety self-efficacy. As expected, the indirect between relationship from mindful organizing to safety violation through self-efficacy was negative and significant (IE = −0.65, p < .001 MC CI = −1.09, −0.17).

To further examine full vs partial mediation, we tested an alternative model that included the direct paths from mindful organizing to the five outcomes. The extra paths were not

Table 2. confirmative factor analysis: hypothesized and alternative factor solutions (N = 488).

Model description χ2 (df ) p rMsea cfi Tli srMr

Hypothesized seven-factor model: seven variables loading onto seven separate factors

1226.57 (506) .000 .06 .92 .91 .04

Alternative model 1 (method bias): seven variables loading onto a single factor

4691.99 (527) .000 .13 .53 .50 .14

Alternative model 2: six factor model with mindful organizing organizing and participative safety self-efficacy loading onto the same single factor, and with initiative, helping, voice, safety compliance and safety violation each loading onto separate factors

1938.84 (512) .000 .08 .84 .82 .06

Alternative model 3: five factor model with the SCBs (initiative, helping, voice) loading onto the same single factor and mindful organizing organizing, participative safety self-efficacy, safety compliance and safety violation each loading onto separate factors

1380.53 (517) .000 .06 .90 .89 .04

Alternative model 4: six factor model with safety compliance and safety violation loading onto the same single factor and mindful organizing organizing, participative safety self-efficacy, initiative, helping and voice each loading onto separate factors

1581.56 (512) .000 .07 .88 .87 .08

Alternative model 5: four factor model with the three SCBs (voice, initiative, and helping) loading onto the same single factor, and safety compliance and violation loading on to another single factor. Mindful organizing and participative safety self-efficacy loading onto their corresponding separated factors

1728.95(521) .000 .07 .86 .85 .09

100 M. CURCURUTO ET AL.

statistically significant (p > .05). The partial mediation model was a complete model (with no degrees of freedom) that showed satisfactory fit (χ2 = 0.45, df = 0, p <.01; RMSEA = 0.00; CFI = 1.00; TLI = 1.00; SRMR-within = .000; SRMR-between = .006). However, the difference between the chi-square statistics provided by the hypothesized full mediation model and the partial mediation model was not statistically significant (Δχ2 = 0.16, Δdf = 5, p > .05). Considering all together, and according to the parsimony principle, the full mediation model was selected against the alternative partial mediation model. These results confirmed that participative safety self-efficacy fully mediated the relationship between mindful organizing and SCBs and individual safety behaviours.

At the within (individual) level, participative safety self-efficacy showed a positive and sig- nificant relationship with voice (b = .76, p < .001), initiative (b = .70, p < .001), helping (b = .71, p < .001) and safety compliance (b = .26, p < .001). However, at the individual level, self-efficacy was not related to safety violation (b = −0.15, p > .05).

6.  General discussion

This study aimed to investigate the influence of mindful organizing on a broad range of safety behaviours in the context of a safety-critical work environment. Furthermore, we intend to explore the mediational role of a safety-specific form of self-efficacy in translating the positive influence of mindful organizing into a range of desired behaviours with a positive impact on the promotion of workplace safety. The construct of participative safety self-efficacy presented in the article was derived from the concept of role-breadth self-efficacy, originally proposed by Parker (1998) to describe the feeling of self-confidence experienced by employees when under- taking initiatives in the workplace that are not formally contemplated in their formal job description. In the context of workplace safety, we primarily proposed that this kind of participation-oriented self-efficacy motivates employees to take on the responsibility of engaging in discretional forms of safety citizenship behaviours (SCB) that can contribute to the creation of a safer workplace. Furthermore, we also proposed that, at a group level of analysis, partici- pative safety self-efficacy can also mediate the influence of mindful organizing on safety com- pliance and safety violations. From this perspective, mindful organizing would stimulate the individual feeling of confidence in being able to contribute to the achievement of the safety

Figure 2. Parameter estimates for the hypothesized model. *p <.05, **p <.001.

JOURNAL OF RISk RESEARCH 101

goals of the team (i.e. reduction of accident rates) by complying with the safety standards and safety procedures, and by reducing violations of these safety standards as much as possible.

Our statistical analyses provided general support for a model where mindful organizing was proposed as a predictor of participative safety self-efficacy, which in turn would result in pos- itively stimulating safety behaviours expected by the organization (safety citizenship and safety compliance), and reducing undesirable behaviours (safety violations). Overall, statistical support was obtained for all our research hypotheses. Furthermore, the magnitude of the resulting statistical effects led us to conclude that mindful organizing presents a stronger influence on safety citizenship behaviours, such as initiative, voice and helping, which are voluntary in nature, rather than expected aspects of safety-critical roles, like safety compliance.

6.1.  Conceptual contributions for literature advancement

Overall, our study contributes to the advancement of safety research literature in several ways, and they are of particular relevance to understanding the positive influence of mindful orga- nizing on workplace safety from a multilevel perspective of analysis. In particular, our research is one of the few studies in the literature that analyses the relationship between mindful orga- nizing and self-efficacy in the domain of workplace safety. This contribution is significant for several reasons.

First, to our best understanding, this was the first study to introduce the construct of par- ticipative safety self-efficacy in safety research. We derived this construct from the more general concept of role breadth self-efficacy (RBSE) originally introduced by Parker (1998). In her seminal work, the author intended to explain why employees choose to engage in behaviours that are not prescribed by their job description, focusing on how individuals develop a specific psy- chological experience of self-efficacy in undertaking extra-role behaviours. Variables such as co-worker support and job enrichment are two examples of facilitating contextual factors identified by the author that support the development of this kind of self-efficacy (Parker, Bindl, and Strauss 2010). Previous studies had already showed how mindful organizing can be associated with a broad range of safety behaviours, including safety compliance and discretional safety actions like safety citizenship behaviours (Gracia et  al. 2020; Renecle et  al. 2020; Renecle et  al. 2021). In addition to the existing literature on mindful organizing, our study offered new insights about one of the psychological mechanisms that can positively affect the relationship between mindful organizing and individual safety related work conduct.

Second, even if the positive influence of mindful organizing on safety behaviour is currently well established in safety research literature, our study helped to understand which kind of safety behaviour is most likely affected by mindful organizing. Said differently, our study tried to investigate if the positive influence of mindful organizing mediated by participative safety self-efficacy affects in the same way different forms of safety related behaviours. In accordance with our expectations, the mediated effect was significant and positive for both safety compli- ance, and extra-role safety citizenship behaviours (i.e. safety voice, safety helping, and safety initiative). However, the relationship was significantly higher for the three kinds of extra-role safety behaviours, rather than safety compliance. This result is particularly indicative of how the mechanisms expected by mindful organizing affect dimensions of individual behaviour at work that go beyond the normative management of workplace safety through safety compliance, by embracing a broader and more flexible approach for ‘managing the unexpected’ (Weick and Sutcliffe 2007) Given that these actions require going beyond one’s ‘comfort zone’ and what is usually expected, the higher mediation effect of participative safety self-efficacy on safety cit- izenship behaviour seems to reiterate how mindful organizing plays an integrative function for the management of risks that exceed merely complying with the organizational procedures and protocol.

102 M. CURCURUTO ET AL.

However, while this evidence seems to suggest that mindful organizing and participative safety self-efficacy are particularly crucial for the emergence of extra-role safety behaviours (rather than safety compliance), we need to recognize that other mechanisms not included in the present investigation need to be taken in account in order to clarify the conditions for which higher levels of mindful organizing are associated with higher levels of safety compliance (i.e. alternative mediation variables – like safety knowledge or safety training - or moderation variables related to job design, team composition or the typology of risks and hazards).

A third contribution offered by our research concerns the level of analysis of safety behaviour. Our study results help to understand if mindful organizing and participative safety self-efficacy affect safety behaviours in the same way when these behaviours are analysed at the group level, rather than the individual level of analysis. Our findings revealed two importance differences. On the one hand, all the positive behavioural outcomes examined in the present study (safety com- pliance and safety citizenship behaviours) emerged from mindful organising through self-efficacy both at the individual and group levels. However, the relationship showed a more differentiated and articulated trend at group level, rather than at the individual level of analysis, where the regression indices resulted quite similar for all the three forms of safety citizenship. This particular result seems to suggest the importance of studying safety citizenship behaviours as the expression of collective dynamics that mainly occur at the group level of analysis (rather than individual), in order to identify and explain the drivers of safety initiative, safety voice and helping behaviours.

In addition, safety violations appeared to be statistically influenced by mindful organizing and participative safety self-efficacy only at the group level of analysis, and in a negative direc- tion. Conversely, there was no relationship between these two variables and safety violation was verified at the individual level of analysis. These results confirmed our research hypotheses, and they also suggest the relevance of investigating the beneficial influence of mindful orga- nizing at a group level of analysis, in order to understand how mindful organizing contributes to the reduction of unsafe behaviour at work (i.e. safety violations) that can be more difficult to explore and explain at an individual level of analysis.

6.2.  Limitations and future research avenues

This study presents several strengths, such as the inclusion of a broad range of safety-specific behavioural indicators, and the usage of sophisticated multi-level mediation analysis. However, like all the studies, there are notable limitations to the present research. In this section, these limits will be addressed, together with suggestions for future replications and/or extensions of the present study.

First, we introduced the concept of mindful organizing as a group level multidimensional process comprehensive of five distinct team-working processes. However, the measurement of mindful organizing was provided only at a general holistic level, and it was not possible to take into account the specific influence of the five single mechanisms. Unfortunately, a well validated, general multidimensional questionnaire assessing the five dimensions of mindful organizing is not yet available in the literature. Therefore, it is not currently possible to examine the specific influence of each one of the five mechanisms of mindful organizing on the various safety behaviours included in our study, nor it is possible to examine the mediation effect of participative safety self-efficacy for each of the five dimensions. Future research should look to address this gap in literature, by providing a general multi-dimensional questionnaire with sound psychometric properties so researchers can investigate the sub-measures of each component of mindful organizing in more depth.

Second, even if this study relies on the usage of a broad set of safety behavioural indicators, all the variables investigated in the present study were assessed with self-reported measures, and given the sensitive topic of safety, the results could be affected by social desirability bias.

JOURNAL OF RISk RESEARCH 103

However, the confidentiality of the scores was clearly communicated to participants and strictly adhered to in order to promote honesty. Furthermore, pre-existing studies have shown evidence of external validity of the behavioural assessment of safety behaviour at work through the usage of self-report measures in various safety critical industries (Curcuruto et  al. 2015).

Third, the present research design is cross sectional. A longitudinal study would have been preferable as it would have allowed for a more robust study of the nature and direction of our study variables. In addition, it would have offset the potential for common method bias to inflate the relationships between the variables studied. Future replications of the present study should adopt a longitudinal research design allowing the assessment of all the independent and dependent variables at the different times of data collection, in order to compare the alternative hypotheses about causal relationships among research variables.

Fourth, the present study did not include any objective safety outcomes (like accident or injury rates, or like near miss indices). However, past studies conducted with the same measures of safety citizenship behaviours found significant associations of these SCB measures with rel- evant objective safety outcomes collected at a later time (Curcuruto et  al. 2015, 2019a). Future replications of the present study should consider including alternative, more objective safety indicators (e.g. organizational data, manager ratings, department key performance indicators for safety) as part of the research model included in this study to further validate this model.

6.3.  Practical implications for managerial programs

This study shows the importance of mindful organising as a starting point for safer operations. For this reason, besides traditional safety training aimed to enable individual to safely perform their individual tasks in the workplace, complimentary learning and development initiatives could focus on enhancing the five processes of mindful organizing: preoccupation with error, reluctance to simplify, sensitivity to operations, commitment to resilience, deference to expertise. To do so, organizations could implement several strategies.

First, organizations could design safety training programs that focus on the five team-working processes of mindful organizing (rather than only individual compliance with safety standards and rules). To train teams in ‘sensitivity to operations’ organizations could develop team members to have a broader awareness of different operations that are proceeding in parallel with their work, and to understand the details of the interdependence of their work, and how an error or change in another area may impact them.

Second, organizations can consider designing training programs that enable managers and team leaders to facilitate and support teams to speak up and empower them to take ownership of important decisions where they are closest to the information or problem. This will help to stimulate ‘deference to expertise’ - where leaders’ willingness to delegate responsibilities to their subordinates is essential - but also in relation to ‘reluctance to simply the operations’, a dimen- sion that can strongly benefit from leaders’ willingness to listen to their subordinates, and from leaders’ ability to stimulate and integrate different elements that can emerge from the group discussion about the activities of the team.

Third, managers could consider designing post-accident investigation activities involving all the group members after an accident or a near-miss event has occurred in the company, even if the critical event happened in a different department of the organization, and it did not involve members of their department. This kind of activity can foster mindful organizing dimen- sions such as ‘preoccupation with failure’ and ‘commitment to resilience’. Post-accident analyses can enable the members of the group to recognize and identify those circumstances in the future that caused a critical event in the past. In addition, this kind of post-accident analysis can enable group members to identify alternative ways to carry on their activities in a safer manner.

104 M. CURCURUTO ET AL.

All the practical intervention strategies listed above can also support the development of higher safetyself efficacy, the psychological mediator at the centre of our investigation, and through it, contributing to the expression of positive work behaviour contributing to the promotion of safety. For instance, by fostering work-team potential to engage in the anticipation processes of mindful organizing (preoccupation with error, reluctance to simplify and sensitivity to operations), the range of situations that team members become more self-assured to address and discuss is increased, growing their confidence to correctly identify, and voice, a wide range of safety issues. This, in turn, makes them more likely to perform the SCB of voicing safety concerns on their own. Furthermore, engaging in the containment processes of mindful organizing boosts an individual’s confidence in their own ability to initiate changes in the moment to quickly act to ensure a safer workplace. This increased confidence in their capability to initiate these actions, will then lead to them engaging in initiating changes to increase safety. The processes of sensitivity to operations and deference to expertise will lead team members to identify colleagues who may need support or assis- tance with safety protocols and practices. This, coupled with knowledge and experience of how to manage safety that comes from engaging in mindful organizing, is likely to build team members’ perceived confidence in successfully helping less experienced colleagues achieve safety goals. This belief in their ability to mentor or assist others is likely to lead these team members to reach out to their colleagues that need help with safety related issues when the situation arises.

7.  Conclusions

This study aimed to investigate the role of mindful organizing on the psychological state of participative safety self-efficacy, a motivational capability that supports personal engagement in a broad range of safety related work conducts, with a special focus on safety citizenship behaviours, described in literature like constructive and discretional actions undertaken by the employees to improve safety in the workplace. Adopting a multi-level statistical approach, the results of our study showed a significant function of participative safety self-efficacy in mediating the beneficial influence of mindful organizing on the behavioural safety criteria stemming from the existing literature. The study advocates for the importance of investigating the link between the teamwork processes contemplated by mindful organizing and the psychological experience of individuals, in order to better understand the factors that facilitate the emergence of con- structive safety behaviours that can help organizations to improve the flexibility and the reliability of their safety management.

Disclosure statement

No potential conflict of interest was reported by the authors.

References

Ausserhofer, D., M. Schubert, M. Desmedt, M. A. Blegen, S. De Geest, and R. Schwendimann. 2013. “The Association of Patient Safety Climate and Nurse-Related Organizational Factors with Selected Patient Outcomes: A Cross-Sectional Survey.” International Journal of Nursing Studies 50 (2): 240–252. https://doi.org/10.1016/j. ijnurstu.2012.04.007

Bentler, P. M. 1990. “Comparative fit indexes in structural models.” Psychological Bulletin 107: 238–246. Bierly, P. E., III, and J. C. Spender. 1995. “Culture and High Reliability Organizations: The Case of the Nuclear

Submarine.” Journal of Management 21 (4): 639–656. https://doi.org/10.1177/014920639502100403 Bliese, P. D. 2000. “Within-Group Agreement, Non-Independence, and Reliability: Implications for Data Aggregation

and Analysis.” In k. J. klein & S. W. J. kozlowski (Eds.), Multilevel Theory, Research, and Methods in Organizations: Foundations, Extensions, and New Directions, 349–381. San Francisco, CA: Jossey-Bass.

JOURNAL OF RISk RESEARCH 105

Browne, M. W., and R. Cudeck. 1993. “Alternative Ways of Assessing Model Fit.” In k.A. Bollen & J.S. Long (Eds.), Testing Structural Equation Models, 136–162. Newbury Park, CA: Sage.

Browne, M. W., and S. H. C. Du Toit. 1992. “Automated Fitting on Nonstandard Models.” Multivariate Behavioral Research 27 (2): 269–300. https://doi.org/10.1207/s15327906mbr2702_13

Burke, M. J., and W. P. Dunlap. 2002. “Estimating Inter-Rater Agreement with the Average Deviation Index: A User’s Guide.” Organizational Research Methods 5 (2): 159–172. https://doi.org/10.1177/1094428102005002002

Chen, F., P. J. Curran, k. A. Bollen, J. kirby, and P. Paxton. 2008. “An empirical evaluation of the use of fixed cut- off points in RMSEA test statistic in structural equation models.” Sociological Methods & Research 36 (4): 462–494. https://doi.org/10.1177/0049124108314720

Cheung, G. W., and R. B. Rensvold. 2002. “Evaluating Goodness-of-Fit Indexes for Testing Measurement Invariance.” Structural Equation Modeling: A Multidisciplinary Journal 9 (2): 233–255. https://doi.org/10.1207/S15328007SEM0902_5

Christian, Michael S., Jill C. Bradley, J. Craig Wallace, and Michael J. Burke. 2009. “Workplace Safety: A Meta-Analysis of the Roles of Person and Situation Factors.” The Journal of Applied Psychology 94 (5): 1103–1127. https://doi. org/10.1037/a0016172

Conchie, S. M. 2013. “Transformational Leadership, Intrinsic Motivation, and Trust: A Moderated-Mediated Model of Workplace Safety.” Journal of Occupational Health Psychology 18 (2): 198–210. https://doi.org/10.1037/a0031805

Curcuruto, M., S. M. Conchie, and M. A. Griffin. 2019b. “Safety Citizenship Behaviour (SCB) in the Workplace: A Stable Construct? Analysis of Psychometric Invariance across Four European Countries.” Accident; Analysis and Prevention 129: 190–201. https://doi.org/10.1016/j.aap.2019.05.023

Curcuruto, M., S. M. Conchie, M. G. Mariani, and F. S. Violante. 2015. “The Role of Prosocial and Proactive Safety Behaviours in Predicting Safety Performance.” Safety Science 80: 317–323. https://doi.org/10.1016/j.ssci.2015.07.032

Curcuruto, M., k. J. Mearns, and M. G. Mariani. 2016. “Proactive Role-Orientation toward Workplace Safety: Psychological Dimensions, Nomological Network and External Validity.” Safety Science 87: 144–155. https://doi. org/10.1016/j.ssci.2016.03.007

Curcuruto, M., and M. A. Griffin. 2018. “Prosocial and proactive “safety citizenship behaviour” (SCB): The mediating role of affective commitment and psychological ownership.” Safety Science 104: 29–38. https://doi.org/10.1016/j. ssci.2017.12.010

Curcuruto, M., S. k. Parker, and M. A. Griffin. 2019a. “Proactivity towards Workplace Safety Improvement: An Investigation of Its Motivational Drivers and Organizational Outcomes.” European Journal of Work and Organizational Psychology 28 (2): 221–238. https://doi.org/10.1080/1359432X.2019.1572115

Dane, E. 2011. “Paying Attention to Mindfulness and Its Effects on Task Performance in the Workplace.” Journal of Management 37 (4): 997–1018. https://doi.org/10.1177/0149206310367948.

Dernbecher, S., M. Risius, and R. Beck. 2014. “Bridging the Gap-Organizational Mindfulness and Mindful Organizing in Mobile Work Environments.” Proceedings of the European Conference on Information Systems (ECIS), Tel Aviv, Israel, June 9-11, 2014, ISBN 978-0-9915567-0-0

Gajda, D. 2018. “Talent Management in the Context of Mindful Organizing and Organizational Mindfulness.” Journal of Positive Management 8 (3): 42–57. https://doi.org/10.12775/JPM.2017.124

Gebauer, A. 2013. “Mindful Organizing as a Paradigm to Develop Managers.” Journal of Management Education 37 (2): 203–228. https://doi.org/10.1177/1052562912458573

Gracia, F. J., I. Tomás, M. Martínez-Córcoles, and J. M. Peiró. 2020. “Empowering Leadership, Mindful Organizing and Safety Performance in a Nuclear Power Plant: A Multilevel Structural Equation Model.” Safety Science 123: 104542. https://doi.org/10.1016/j.ssci.2019.104542

Griffin, M. A., and M. Curcuruto. 2016. “Safety Climate in Organizations.” Annual Review of Organizational Psychology and Organizational Behavior 3 (1): 191–212. https://doi.org/10.1146/annurev-orgpsych-041015-062414

Griffin, M. A., and A. Neal. 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. https://doi. org/10.1037//1076-8998.5.3.347

Hansez, I., and N. Chmiel. 2010. “Safety Behaviour: Job Demands, Job Resources, and Perceived Management Commitment to Safety.” Journal of Occupational Health Psychology 15 (3): 267–278. https://doi.org/10.1037/a0019528

Hofmann, D. A., F. P. Morgeson, and S. J. Gerras. 2003. “Climate as a Moderator of the Relationship between Leader-Member Exchange and Content Specific Citizenship: Safety Climate as an Exemplar.” The Journal of Applied Psychology 88 (1): 170–178. https://doi.org/10.1037/0021-9010.88.1.170

Hollnagel, E. 2014. Safety-I and Safety-II the past and Future of Safety Management. Farnham, Uk: Ashgate Hu, L. T., and P. M. Bentler. 1999. “Cutoff Criteria for Fit Indexes in Covariance Structure Analysis: Conventional

Criteria versus New Alternatives.” Structural Equation Modeling: A Multidisciplinary Journal 6 (1): 1–55. https:// doi.org/10.1080/10705519909540118

kozlowski, S. W. J., and k. J. klein. 2000. “A Multilevel Approach to Theory and Research in Organizations. Contextual, Temporal, and Emergent Processes.” In k. J. klein & S. W. J. kozlowski (Eds.), Multilevel Theory, Research, and Methods in Organizations, 3–90. San Francisco, CA: Jossey-Bass.

LaPorte, T. R., and P. M. Consolini. 1991. “Working in Practice but Not in Theory: Theoretical Challenges of" High-Reliability Organizations.” Journal of Public Administration Research and Theory: J-PART 1: 19–48.

106 M. CURCURUTO ET AL.

LeBreton, J. M., and J. L. Senter. 2008. “Answers to 20 Questions about Interrater Reliability and Interrater Agreement.” Organizational Research Methods 11 (4): 815–852. https://doi.org/10.1177/1094428106296642

Levinthal, D., and C. Rerup. 2006. “Crossing an Apparent Chasm: Bridging Mindful and Less-Mindful Perspectives on Organizational Learning.” Organization Science 17 (4): 502–513. https://doi.org/10.1287/orsc.1060.0197

Martínez-Córcoles, M., and T. J. Vogus. 2020. “Mindful Organizing for Safety.” Safety Science 124 (1): 104614. https:// doi.org/10.1016/j.ssci.2020.104614

Morgeson, F. P., and D. A. Hofmann. 1999. “The Structure of Collective Constructs: Implications for Multilevel Research and Theory Development.” The Academy of Management Review 24 (2): 249–265. https://doi. org/10.2307/259081

Muthén, L. k., and B. O. Muthén. 2017. Mplus: Statistical Analysis with Latent Variables: User’s Guide (Version 8). Los Angeles, CA: Authors.

Ndubisi, N. O., and O. Al‐Shuridah. 2019. “Organizational Mindfulness, Mindful Organizing, and Environmental and Resource Sustainability.” Business Strategy and the Environment 28 (3): 436–446. https://doi.org/10.1002/bse.2219

Parker, S. k. 1998. “Enhancing Role Breadth Self-Efficacy: The Roles of Job Enrichment and Other Organizational Interventions.” The Journal of Applied Psychology 83 (6): 835–852. https://doi.org/10.1037/0021-9010.83.6.835

Parker, S. k., U. k. Bindl, and k. Strauss. 2010. “Making Things Happen: A Model of Proac- Tive Motivation.” Journal of Management 36 (4): 827–856. https://doi.org/10.1177/0149206310363732

Perrow, C. 1984. Normal Accidents: Living with High-Risk Technologies. New York, NY: Basic Books. Renecle, M., M. Curcuruto, F. J. Gracia Lerín, and I. Tomás Marco. 2021. “Enhancing Safety in High-Risk Operations:

A Multilevel Analysis of the Role of Mindful Organizing Organizing in Translating Safety Climate into Individual Safety Behaviours.” Safety Science 138: 105197. https://doi.org/10.1016/j.ssci.2021.105197

Renecle, M., I. Tomás, F. J. Gracia, and J. M. Peiró. 2020. “Spanish Validation of the Mindful Organizing Scale: A Questionnaire for the Assessment of Collective Mindfulness.” Accident; Analysis and Prevention 134: 105351. https://doi.org/10.1016/j.aap.2019.105351

Roberts, k. H. 1993. “Cultural Characteristics of Reliability Enhancing Organizations.” Journal of Management Issues, 5 (2): 165–181. https://doi.org/10.2307/40603976

Roberts, k. H. 1990. “Some Characteristics of One Type of High Reliability Organization.” Organization Science 1 (2): 160–176. https://doi.org/10.1287/orsc.1.2.160

Roberts, k. H., and R. Bea. 2001. “Must Accidents Happen? Lessons from High-Reliability Organizations [and Executive Commentary].” Academy of Management Perspectives 15 (3): 70–78. https://doi.org/10.2307/4165761

Roberts, k. H., and D. M. Rousseau. 1989. “Research in Nearly Failure-Free, High-Reliability Organizations: Having the Bubble.” IEEE Transactions on Engineering Management 36 (2): 132–139. https://doi.org/10.1109/17.18830

Roberts, k. H., P. Madsen, V. Desai, and D. Van Stralen. 2005. “A Case of the Birth and Death of a High Reliability Healthcare Organisation.” Quality & Safety in Health Care 14 (3): 216–220. https://doi.org/10.1136/qshc.2003.009589

Roberts, k. H., S. k. Stout, and J. J. Halpern. 1994. “Decision Dynamics in Two High Reliability Military Organizations.” Management Science 40 (5): 614–624. https://doi.org/10.1287/mnsc.40.5.614

Rochlin, G. I. 1993. “Defining “high reliability” organizations in practice: A taxonomic prologue.” In k. H. Roberts (Ed.), New challenges to understanding organizations, 11–32. New York: Macmillan.

Schneider, B. 1987. “The People Make the Place.” Personnel Psychology 40 (3): 437–453. https://doi. org/10.1111/j.1744-6570.1987.tb00609.x

Schulman, P. R. 1993. “The Negotiated Order of Organizational Reliability.” Administration & Society 25 (3): 353–372. https://doi.org/10.1177/009539979302500305

Steiger, J. H. 1990. “Structural Model Evaluation and Modification: An Interval Estimation Approach.” Multivariate Behavioral Research 25 (2): 173–180. https://doi.org/10.1207/s15327906mbr2502_4

Sutcliffe, k. M., T. J. Vogus, and E. Dane. 2016. “Mindfulness in Organizations: A Cross-Level Review.” Annual Review of Organizational Psychology and Organizational Behavior 3 (1): 55–81. https://doi.org/10.1146/ annurev-orgpsych-041015-062531

Tucker, L. R., and C. Lewis. 1973. “A Reliability Coefficient for Maximum Likelihood Factor Analysis.” Psychometrika 38 (1): 1–10. https://doi.org/10.1007/BF02291170

Vogus, T. J. 2011. “Mindful Organizing: Establishing and Extending the Foundations of Highly Reliable Performance.” In G. M. Spreitzer & k. S. Cameron (Eds.), The Oxford Handbook of Positive Organizational Scholarship, 664–676. Oxford (Uk): Oxford Handbooks Online.

Vogus, T. J., and k. M. Sutcliffe. 2007a. “The Impact of Safety Organizing, Trusted Leadership, and Care Pathways on Reported Medication Errors in Hospital Nursing Units.” Medical Care 45 (10): 997–1002. https://doi.org/10.1097/ MLR.0b013e318053674f

Vogus, T. J., and k. M. Sutcliffe. 2007b. “The Safety Organizing Scale.” Medical Care 45 (1): 46–54. https://doi. org/10.1097/01.mlr.0000244635.61178.7a

Vogus, T. J., and k. M. Sutcliffe. 2012. “Organizational Mindfulness and Mindful Organizing: A Reconciliation and Path Forward.” Academy of Management Learning & Education 11 (4): 722–735. https://doi.org/10.5465/am- le.2011.0002c

JOURNAL OF RISk RESEARCH 107

Weick, k. E., and k. H. Roberts. 1993. “Collective Mind in Organizations: Heedful Interrelating on Flight Decks.” Administrative Science Quarterly 38 (3): 357–381. https://doi.org/10.2307/2393372

Weick, k. E., k. M. Sutcliffe, and D. Obstfeld. 1999. “Organizing for High Reliability: Processes of Mindful Organizing.” In B. M. Staw & L. L. Cummings, (Eds.), Research in Organizational Behaviour, 21, 81–123. Oxford (Uk): Elsevier.

Weick, k. E., and k. M. Sutcliffe. 2007. Managing the Unexpected: Assuring High Performance in an Age of Complexity (1st ed.). San Francisco, CA: Jossey-Bass.

Weick, k. E., and k. M. Sutcliffe. 2015. Managing the Unexpected: Sustained Performance in a Complex World (3rd ed.). San Francisco, CA: Jossey-Bass.

Copyright of Journal of Risk Research is the property of Routledge and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.