To engage or not to engage in organisational citizenship behaviour: that is the question!

Abstract Organisational citizenship behaviour (OCB) can be defined as discretional, voluntary and useful informal behaviour that is not directly acknowledged by the organisation’s formal reward system. Such behaviour refers to actions that go beyond performing the tasks defined as part of one’s job. Previous studies have shown that organisations that promote OCB can notably improve their productivity and efficiency. It is therefore important to know what causes employees to engage in OCB rather than just limiting themselves to doing what is strictly expected at work. However, it is more important to know why they do not engage in OCB. Using a sample of public prison employees and the fsQCA method, this study examines how the combined effects of organisational characteristics, leadership behaviours and individual characteristics lead to the absence of OCB. The results indicate that the absence of affective commitment, or job satisfaction, or interactional justice is a necessary condition for the absence of OCB. Four conditions are identified as sufficient, and the absence of affective commitment and the presence of laissez-faire leadership are found to be the most relevant conditions for the absence of OCB. Managerial implications and directions for future studies are discussed at the end of the paper.


Introduction
In recent years, the scientific community has devoted considerable attention to organisational citizenship behaviour (OCB), which is defined as 'individual behavior that is discretionary, not directly or explicitly recognized by the formal reward system, and that in the aggregate promotes the effective functioning of the organization' (Organ, 1988, p. 4). 'These behaviors include any of those gestures (often taken for granted) that lubricate the social machinery of the organization but that do not directly inhere in the usual notion of task performance' (Bateman & Organ, 1983, p. 588).
But what leads employees to go beyond the call of duty at work? According to social exchange theory (Blau, 1964), when workers perceive good will from the organisation, they respond in kind, engaging in positive behaviours that yield benefits for both parties (Cropanzano & Mitchell, 2005). Such behaviours, which often go beyond the scope of the tasks defined to perform a certain job, include helping overloaded co-workers, voluntarily taking on additional responsibilities, developing additional skills that are beneficial to the organisation, working overtime, or publicly championing the organisation (Allen et al., 2011;Organ et al., 2006).
Scholars have defined OCB and have identified its main antecedents as well as the positive effects of OCB on organisations (LePine et al., 2002;Organ & Ryan, 1995;Organ et al., 2006;Podsakoff et al., 2000Podsakoff et al., , 2009. These positive effects range from improvements at the organisational level, such as greater productivity, efficiency and profitability as well as cost reductions (Podsakoff et al., 2009;Walz & Niehoff, 2000), to improvements at the individual level, such as strong performance evaluations (Allen & Rush, 1998;Podsakoff et al., 2009), reward recommendation decisions (Allen & Rush, 1998), and low or decreasing levels of turnover, turnover intentions (Chen, 2005) and absenteeism (Podsakoff et al., 2009).
In general, the study of OCB is a prominent theme in academic research, although not in the context of prisons. Prison staffs are civil servants in Spain. They have welldefined roles, but they are exposed to unique working conditions in that they have to interact on a daily basis with people who are held in prison against their will (Brough & Williams, 2007). Thus, in recent decades, scholars have undertaken major efforts to determine the positive and negative effects of these working conditions on prison staff (Lambert & Hogan, 2013). Few studies have examined what leads prison staff to engage in OCB or, more importantly, what causes employees not to engage in OCB and instead limit their actions to what is strictly expected of them at work. These studies have tended to focus on workers' attitudes. Under the assumption that OCB has a major voluntary component, researchers have focused on analysing the individual attitudes that encourage prison staff to adopt OCB. These attitudes include the worker's commitment to the organisation (Allen et al., 2011;Frank et al., 2020;Lambert et al., 2008Lambert et al., , 2017, job satisfaction, (Frank et al., 2020;Ingrams, 2020;Lambert, 2010) and even the perception of interactional justice (Lambert, 2003;Lambert & Hogan, 2013).
However, few studies have examined how variables that are not related to workers' attitudes (e.g., an unsuitable leadership style or bullying at work) might be responsible for preventing prison staff from engaging in OCB. And even fewer studies have examined the combined effect of these antecedents on the absence of OCB; almost all studies have focused on investigating which individual variables lead to the presence (rather than the absence) of OCB. Thus, the use of fuzzy-set qualitative comparative analysis (fsQCA) can help provide a better understanding of the factors that lead prison staff not to engage in OCB, offering an alternative approach that complements conventional practice (Ragin, 2008).
In light of the above, this study examined three of the principal antecedents of OCB (Podsakoff et al., 2000) to observe how the combined effects of individual characteristics (absence of affective commitment, perceptions of interactional justice and job satisfaction), leadership behaviours (presence of laissez-faire leadership) and organisational characteristics (presence of bullying at work) can lead to the absence of OCB among prison staff. The results contribute to the advancement of knowledge of OCB in two ways. First, they provide a more complete overview of this phenomenon by considering the attitudinal perspectives not only of workers but also of superiors and the organisation itself. Second, they provide insight from the analysis of the absence of OCB. The results thus shed light on the reasons why prison staff do not engage in this type of behaviour. The results of this study thereby provide prison managers with information on strategies that can be adopted to encourage OCB through the management and control of work-related variables that might hinder such behaviour.
The article is divided into several sections. Section 2 presents a review of the general literature with a focus on the links between OCB and the five antecedents used to explain why prison staff do not engage in OCB. Section 2 also states the propositions tested in this study. Section 3 describes the data, method, measurement scales and procedure followed in this study. Section 4 presents the results of the analysis. Section 5 discusses these results, framing them in the existing literature. Section 6 presents some of the limitations of the study as well as offering suggestions for future research. Section 7 briefly outlines the conclusions of the study.

OCB and affective commitment
The study of the positive effects of workers' OCB has elicited growing interest from the scientific community as a means of understanding what leads workers to perform tasks that go beyond any formal requirements (LePine et al., 2002). A pillar of the OCB literature is that OCB is driven by positive work attitudes such as organisational commitment (Frank et al., 2020;Organ & Ryan, 1995;Shahjehan et al., 2019). Organisational commitment is the emotional bond that leads employees to accept the organisation's goals (Porter et al., 1974). It is considered an attitudinal indicator of the degree to which workers perceive a high-quality social exchange relationship with the organisation (Cropanzano & Mitchell, 2005).
The organisational commitment literature discusses three forms of commitment: affective, normative and continuance. The present study focuses on affective commitment, which refers to the emotional bond, identification and involvement of the individual with the organisation (Allen & Meyer, 1990). Affective commitment is more strongly related to OCB than either normative or continuance commitment is (Chênevert et al., 2015;Meyer et al., 2002). When the emotional bond that links an individual to the organisation is strong, the worker becomes fully committed to the goals of the organisation (Allen et al., 2011) and engages in behaviours that, without being explicitly stated by the organisation, actually benefit it (Meyer & Herscovitch, 2001).
Furthermore, affective commitment is the most widely studied form of commitment in the context of prisons because this type of commitment best reflects the exchange relationship between a worker and the organisation in this context. Although prison employment is stable, it also entails considerable complexity, danger and the obligation to work in a crowded, inefficient system (Dur an, 2007). In this context, employees who develop emotional bonds with the organisation tend to go beyond the call of duty, displaying a greater level of OCB (Allen et al., 2011;Lambert et al., 2008Lambert et al., , 2017 because they are likely to feel a relational obligation to engage in behaviours that have positive consequences for both parties (Cropanzano & Rupp, 2008). But what happens when prison staff does not develop affective ties with the organisation? By the same logic, these employees are unlikely to engage in OCB because they perceive a need to meet only the formal requirements of their jobs. Thus, the following proposition was tested: P1: The absence of affective commitment leads to the absence of OCB.

OCB and interactional justice
Another antecedent of OCB is the perception of organisational justice (Colquitt et al., 2001;Podsakoff et al., 2000). Organisational justice refers to an individual's perception of and reactions to, how the organisation treats that individual (L opez-Cabarcos et al., 2016). As Greenberg notes (1993, p. 250), 'people will behave altruistically toward the organization in which they work if they believe they have been fairly treated by that organization'. According to social exchange theory, if employees deem that fair treatment by the organisation is beneficial, they will be more motivated to reciprocate, adopting positive attitudes and behaviours (Moorman & Byrne, 2005).
Organisational justice has three components: distributive, procedural and interactional justice. This study focuses on interactional justice, which relates to the importance of the perceived quality of interpersonal relationships (i.e., treating employees with respect, truthfulness and propriety and offering suitable justifications) when the organisation implements procedures (Bies & Moag, 1986). Some scholars have argued that the perception of interactional justice triggers more intense emotional and behavioural responses in workers than the perception of other forms of justice (Bies & Moag, 1986). Furthermore, when the outcome is the intention to engage in OCB, interactional justice is the only type of justice capable of predicting that intentionality (Williams et al., 2002). For this reason, the scientific community also requires for more research on interactive justice in the prison context .
In a context such as the prison system, where the working conditions are established in many countries by the state, the perception of fair, equal and respectful treatment by superiors and colleagues is likely to be especially important and is likely to lead to OCB that enriches the immediate working environment and the organisation as a whole (Chiaburu, 2007;Masterson et al., 2000;Moorman & Byrne, 2005). Conversely, however, unfair treatment may force employees into the economic exchange relationships stipulated in contracts, making the presence of OCB less likely (Moorman & Byrne, 2005). Thus, the following proposition was tested: P2: Absence of perceptions of interactional justice leads to the absence of OCB.

OCB and job satisfaction
Early research on OCB examined job satisfaction as one of OCB's first antecedents (Williams & Anderson, 1991). Job satisfaction is defined as an individual's positive emotional reaction to work (Oshagbemi, 2000). Most meta-analyses have shown the existence of a direct positive relationship between job satisfaction and OCB (Bateman & Organ, 1983;Ingrams, 2020;LePine et al., 2002;Organ & Ryan, 1995;Podsakoff et al., 2000). This relationship can also be explained using social exchange theory as follows: To the extent that employee satisfaction derives from the efforts of the organisation and these efforts are perceived as volitional rather than manipulative, the worker will tend to reciprocate these efforts (Bateman & Organ, 1983).
In the context of the prison system job satisfaction is a possible antecedent of OCB (Lambert, 2010), although studies examining this relationship are scarce. As discussed earlier, workers with high job satisfaction tend to interact reciprocally with the organisation by engaging in OCB. But what happens when employees are dissatisfied? Is this dissatisfaction enough to prevent OCB from flourishing? Thus, the following proposition was tested: P3: The absence of job satisfaction leads to the absence of OCB.

OCB and Laissez-Faire leadership
According to Podsakoff et al. (2000), the next group of antecedents of OCB consists of leadership behaviours. Leadership is defined as the art or process of influencing people so that they voluntarily and enthusiastically meet their goals (Koontz & Weihrich, 1998). It refers to the interactions among the members of a group, where the leaders are the agents of change. In other words, leaders are people whose actions affect the other members of the group (Bass, 1990).
Leadership studies have traditionally focused on analysing the factors associated with effective management, implicitly assuming that leadership ineffectiveness simply reflects an absence of leadership (Ashforth, 1997). However, scholars have recently drawn attention to the effects of negative leadership on workers' attitudes and behaviours and on organisational performance (Skogstad et al., 2007).
One of the negative forms of leadership is laissez-faire (passive) leadership, which has been described by Bass and Avolio (1995, p. 20) as 'the absence of leadership, the avoidance of intervention, or both. With laissez faire leadership, there are generally neither transactions nor agreements with followers. Decisions are often delayed; feedback, rewards and involvement are absent; and there is no attempt to motivate followers or to recognize and satisfy their needs'. This type of leadership entails a lack of effective leadership. Moreover, when superiors adopt a passive leadership style, they are unable to mobilise all the resources that the organisation places at employees' disposal to forge relationships that encourage organisational commitment and, ultimately, OCB (Chênevert et al., 2015). Thus, workers who feel isolated and ignored by those above them in the hierarchyand, consequently, by the organisation itselfmay not feel the obligation to reciprocate that arises when the organisation looks after workers' interests and well-being. This situation may undermine any chances of developing OCB (Al-Khasawneh & Futa, 2013). Thus, the following proposition was tested: P4: The presence of laissez-faire leadership leads to the absence of OCB.

OCB and bullying at work
The third group of antecedents of OCB relates to organisational characteristics (Podsakoff et al., 2000). One element of this group of antecedents is perceived organisational support, which vanishes when the organisation allows, or at least fails to avoid, bullying at work (L opez-Cabarcos et al., 2017). Bullying at work means, 'harassing, offending, or socially excluding someone or negatively affecting someone's work. It is an escalating process in the course of which the person confronted ends up in an inferior position and becomes the target of systematic negative social acts' (Einarsen et al., 2011, p. 22).
Some scholars have suggested that social exchange theory can provide a useful theoretical framework for investigating bullying at work in the context of worker-organisation exchange relationships because such a framework can help explain how intimidation affects the attitudes and organisational behaviour of victims and witnesses (Parzefall & Salin, 2010). Intimidated workers might perceive a psychological contract breach, which might negatively influence their attitudes and behaviours towards the organisation in general by leading them to the conclusion that the organisation does not look after their safety or well-being, especially when bullying continues even after the organisation is made aware of it (Parzefall & Salin, 2010). Thus, the following proposition was tested: P5: The presence of bullying at work leads to the absence of OCB.

Method
Qualitative comparative analysis (QCA) refers to a series of techniques that combine a qualitative approach such as the use of case studies with a quantitative, variablebased approach (Ragin, 2008;Ragin & Fiss, 2008). QCA makes it possible to establish asymmetrical configurations (causal conditions and/or combinations of them), which means that the fact that a certain cause leads to a certain outcome does not mean that the presence of the outcome implies the presence of the cause (Ordanini et al., 2014). It also applies the principle of equifinality, which means that different combinations of conditions can lead to the same outcome (Roig-Tierno et al., 2016). Of the different types of QCA, fuzzy-set QCA (fsQCA) enables various degrees of categorisation of the conditions and is a suitable method when studying causally complex social phenomena that can be articulated as groups and explained in terms of necessity and sufficiency (Woodside, 2016). This method is particularly suitable for studying small and medium-sized samples (Fiss, 2011) but is equally valid for larger samples (Vis, 2012), such as the one in this study. The fsQCA method was applied following the steps proposed by Schneider and Wagemann (2010) that is, calibration, analysis of necessity and analysis of sufficiency.

Participants and instruments
Data were obtained through a self-administered questionnaire to employees at all levels of five Spanish prison facilities. A questionnaire together with an explanatory cover letter was distributed to 1,545 employees, yielding 486 (31.4% response rate) useable questionnaires. Of these employees, 82.9% were men and 17.1% women, and 80.5% were aged between 35 and 49 years.
The 25-item Coyle-Shapiro scale (2002) was used to measure OCB (advocacy participation, helping, functional participation, loyalty and obedience) on a 5-point scale (1 ¼ not at all; 5 ¼ to a very great extent). Based on reviews of the nature of the OCB construct (LePine et al., 2002;Organ, 1997), the measurement used in this study was consistent with that of an aggregate construct, with dimensions that each added to the overall value of OCB (a ¼ 0.90).
The 7-item Allen and Meyer scale (1990) measured affective commitment (AC a ¼ 0.88) on a 5-point Likert scale (1 ¼ strongly disagree; 5 ¼ strongly agree). The 4item scale designed by Moliner et al. (2003) was used to measure interactional justice (IJ a ¼ 0.87), and the 15-item scale designed by Cellucci and DeVries (1978) was used to measure job satisfaction (WS a ¼ 0.72). Both instruments used a 5-point Likert scale (1 ¼ strongly disagree; 5 ¼ strongly agree). The 4-item scale designed by Bass and Avolio (1995) from MLQ-5X (short form) was used to measure laissez-faire leadership (LF a ¼ 0.87) on a 5-point scale (1 ¼ never; 5 ¼ always). The 29-item version of the Hoel and Cooper (2000) Negative Acts Questionnaire was used to measure bullying at work (BULL a ¼ 0.89). This instrument used a 5-point scale (0 ¼ never; 4 ¼ daily).

Procedure
After evaluating the psychometric properties of the scales, the values of all items in the scale were multiplied to obtain the value of each construct (Palacios-Marqu es et al., 2017). A process of calibration was then applied to transform the data into fuzzy sets. Three anchors were established for calibration: full membership, maximum ambiguity, and full non-membership (Ragin, 2008). The thresholds were set at the 90th, 50th and 10th percentiles, respectively (Misangyi & Acharya, 2014).

Results
The outcome of the model was the absence of OCB. The five conditions were affective commitment, interactional justice, job satisfaction, laissez-faire leadership and bullying at work. To conduct a more thorough analysis and provide greater insight into the proposed model, the analysis was conducted considering both the absence and the presence of OCB. Tables 1 and 2 present the results of the analysis of necessary conditions for the absence (and presence) of OCB among prison employees.
The results indicate that none of the five conditions considered herein on its own leads to the absence of OCB. Although the presence of one of the two negative conditions (BULL þ LF) was found not to be a necessary condition, the absence of one of the three positive conditions of the model (IJþACþWS) was a necessary condition for the absence of OCB. The absence of negative conditions (BULLþLF) and the presence of positive conditions (IJ þ AC þ WS) were quasinecessary for the presence of OCB because their consistency value was close to .90 .
Analysis of sufficient conditions was also conducted to identify causal configurations that lead to the outcome, again considering the absence and presence of OCB. The following models were studied: Intermediate solutions are presented for both models in Table 3. Four solutions lead to the absence of OCB among prison employees, which explain more than 51%   0.7721 0.8795 Note: ᭹ denotes the presence of the condition and ᭺ denotes the absence of the condition. Large circles denote core conditions, and small circles denote peripheral conditions (Fiss, 2011). Consistency cut-off: 0.79 and 0.90. Frequency cut-off: 5.00. Vector of expected directions (1,1,1,0,0) and (0,0,0,1,1) (Ragin & Sean, 2016). Source: Elaborated by the authors.
of the analysed cases. The results also reveal three solutions that lead to the presence of OCB, which explain almost 60% of the cases in this study.

Discussion
The results of this study help identify the key variables to avoid the absence of OCB in the context of the prison system. The results indicate that no individual condition was necessary for the absence of OCB. The joint analysis of the conditions that relate to negative characteristics of the organisation (i.e., laissez-faire leadership and bullying at work) did not reveal that these conditions were necessary for the absence of OCB. However, when the conditions related to individual characteristics were considered together, the absence of interactional justice or affective commitment or job satisfaction was observed to be a necessary condition for the absence of OCB. Studies have shown the key role of these three attitudes in the presence of OCB (Allen et al., 2011;Chiaburu, 2007;Frank et al., 2020;Lambert et al., 2017;Masterson et al., 2000;Moorman & Byrne, 2005). It is therefore reasonable to expect the absence of these employee attitudes to lead to the absence of OCB. Employees who do not develop an emotional bond with the organisation either because they lack job satisfaction or because they feel that they are not treated with equality, respect and dignity are unlikely to feel that they have an obligation or duty to the organisation. In such cases, the most likely outcome is that these employees simply comply with their formal contracts, not engaging in other forms of OCB that could potentially benefit the organisation (Moorman & Byrne, 2005).
To compare the results of the analysis of necessary conditions for the absence of OCB, analysis of necessary conditions for the presence of OCB was also performed. Once again, the results show that none of the five causal conditions has to be present or absent for the presence of OCB. In this case, both the joint presence of the conditions that relate to the worker's attitude (affective commitment, interactional justice and job satisfaction) and the absence of the conditions with negative connotations (laissez-faire leadership and bullying at work) are quasi-necessary conditions for the presence of OCB. This finding corroborates the results for the model explaining the absence of OCB and supports the findings of previous studies (Chênevert et al., 2015;Chiaburu, 2007;Frank et al., 2020;Lambert et al., 2017;LePine et al., 2002).
Continuing with the analysis of the absence of OCB, the results indicate that combinations of several conditions are sufficient for the absence of OCB. The first configuration that explains the absence of OCB is the lack of an affective bond between workers and the organisation, which is linked to a lack of job satisfaction, even if workers do not feel they are victims of bullying at work.
The second configuration also refers to the absence of this affective bond between the worker and the organisation, which is associated with the presence of a negative leadership style, once again even if workers do not feel they are victims of bullying at work. Thus, workers who do not have an emotional bond with the organisation and, consequently, feel isolated, ignored, or unsupported by their superiors, may not feel an obligation to reciprocate, which undermines any chance of developing OCB (Al-Khasawneh & Futa, 2013), even if they are not victims of bullying at work.
The third configuration coincides with the first sufficient condition. Thus, the combination of the absence of affective commitment and the absence of job satisfaction leads to the absence of OCB, regardless of whether employees feel they are treated with equality, respect, and fairness. Finally, the fourth configuration refers to the absence of an affective bond between the worker and the organisation, the absence of interactional justice and the presence of a negative leadership style, regardless of job satisfaction. As mentioned earlier, the prison work environment entails complex, dangerous work carried out in an inefficient system (Dur an, 2007). It is therefore unlikely for affective bonds to form between worker and organisation. The absence of these bonds, coupled with a lack of job satisfaction, leads prison staff to meet the obligations stipulated in their contracts without going beyond these formal duties. This is the case even when employees are not victims of bullying at work.
The results show that the most important condition for the absence of OCB among prison staff is the lack of an affective bond between the worker and the organisation. Given that OCB is driven by positive work attitudes (Organ & Ryan, 1995), the fact that the organisation is unable to build the type of affective bonds with workers that lead to high-quality social exchanges between the two parties (Cropanzano & Mitchell, 2005) reduces the likelihood that employees engage in OCB. Comparing the model for the absence of OCB with the model for the presence of OCB shows that affective commitment is present in the three necessary conditions as the most relevant condition explaining workers' OCB. This observation corroborates the results of the earlier analysis and the findings of previous studies (Chênevert et al., 2015;Frank et al., 2020;Meyer et al., 2002).
Another important condition that is always linked to the absence of OCB is the presence of a laissez-faire leadership style. Most prison staff carry out their work alone and have direct contact only with prisoners. Therefore, interactions with superiors are rare. In such a context, it is common for superiors not to motivate workers or take an interest in their needs (Bass & Avolio, 1995). This situation reduces workers' moral obligation to act reciprocally in this exchange by engaging in OCB. Comparing these results with those for the sufficient conditions that explain the presence of OCB among prison staff shows that the absence of a laissez-faire leadership style is present in two of the three configurations identified during the analysis.
Unlike with the absence of affective commitment or the presence of laissez-faire leadership, the absence of interactional justice or job satisfaction seems to play a less central role in explaining the absence of OCB among prison staff. These secondary conditions are nonetheless present in some configurations in conjunction with other conditions. This secondary importance may owe to the fact that both are antecedents of affective commitment (L opez-Cabarcos et al., 2016). Thus, if, as indicated by the results, the absence of affective commitment plays a key role in the absence of OCB, it seems reasonable to expect that the role of interactional justice and satisfaction would be secondary in explaining the absence of OCB. This conclusion is corroborated by the observation that these are peripheral conditions in some paths.
Finally, the absence of bullying at work is not a decisive condition for the absence or presence of OCB among prison staff. The explanation for this finding may be rooted in the very nature of bullying. Bullying may include subtle actions such as social isolation, persistent work criticism, gossiping, or undervaluing a person's work (Notelaers et al., 2011), which are common in military organisations or structures and are accepted or regarded as usual by their members (Archer, 1999).
In summary, the results of the empirical analysis support propositions P1 and P4. Propositions P2 and P3 are partially supported. Only proposition P5 is not supported by the data because in all cases, the presence rather than the absence of bullying at work leads to the absence of OCB among prison staff.

Conclusions
An issue related to the intention to develop OCBs in the workplace and, of course, in a prison setting is that managers are not in a position to require employees to participate in OCBs, as these behaviours are voluntary and discretionary. However, organisations could learn about the issues they should address if they wish to mould the attitudes and behaviours of their employees. In this sense, this study examined the factors that prevent workers from becoming involved with the organisation beyond the duties that are formally stipulated in their contracts. Managers should acknowledge that, in general, employees' attitudes towards the organisation influence their prosocial behaviours. This study suggests that affective commitment is a core condition which leads to both the absence and the presence of OCB. Therefore, prison managers should develop ways to provide prison staff with more opportunities to identify with and build affective relationships with the organisation. One way to create these affective bonds is by socialising employees, trying to get prison staff to find meaning in their jobs and to accept and share organisational values (Shim & Faerman, 2017). Lambert et al. (2020) recommend enhancing, among other aspects, job variety, instrumental communication, supervision and management/supervisory support because 'these variables have the ability to instil a sense of fairness and empowerment in the officers and are most likely to boost their affective commitment as tangible steps toward these goals are taken by the organization' (p. 16). It is therefore fundamental to create a supportive climate with free-flowing communication to encourage employees to develop strong affective bonds with the organisation.
Managers should also be aware that the wrong leadership style could shape employees' behaviour, reducing the likelihood that employees engage in OCB. Therefore, leadership styles that are deemed negative should be avoided. Collaborative and participative leadership styles should be embraced instead. Therefore, it could be advantageous for prison managers to design, structure and organise the work environment in such a way as to create a culture able to promote the development of fairness, trust and high-quality leader-follower relationships (Lee et al., 2020). Designing strategies aimed at increasing job satisfaction, nurturing fair, respectful, honest interpersonal relationships and tackling bullying at work can also help promote OCB among employees.

Limitations and future research
As in any empirical study, this study has some limitations. QCA with one main outcome variable (absence of OCB) and five potential antecedents might seem simple, so the opposite model (presence of OCB) was also tested to corroborate the results for the first model. Other limitations are that the sample only included employees of Spanish prisons and that the results are based on self-reported data. Other methods such as in-depth interviews could provide valuable information to complement the findings.
The results are consistent with previous research and create opportunities for further analysis. Although the research setting and variables considered in this study are important, including other variables could provide interesting findings. Some researchers state that the role of work climate, work design and leadership styles need to be investigated in the context of government organisations (Shim & Faerman, 2017). In this study, a passive leadership style was included, however it could be interesting to analyse the role of positive leadership styles in explaining the presence of OCB, mainly authentic leadership (Pastor-Alvarez et al., 2019) or servant leadership (Luu, 2017).
Since this study has focused on examining the combined effects of organisational characteristics, leadership behaviours and individual characteristics on the absence of OCB, a future study may focus on analysing the mediating and moderating role of other individual, organisational or even cultural variables. Some of these variables, such as perceived organisational support (Detnakarin & Rurkkhum, 2019), resilience (Ahmad et al., 2019), perceived self-efficacy (Testa et al., 2020) or tenure (Liu et al., 2020), have seldom been studied in the field of OCB and the prison context. Future research should analyse other sectors with similar characteristics to highlight their idiosyncrasies and generalise the findings of this study.

Disclosure statement
No potential conflict of interest was reported by the authors.