When does job burnout not hurt employee behaviours?

Purpose – The purpose of this study was to describe and explain the relationship between job burnout (JB), organisational citizenship behaviours (OCBs) and counterproductive workplace behaviours (CWBs). The job demands-resources theorywas applied to test the moderating role of perceived organisational support (POS) in the relationship shared by JB, OCBs and CWBs. Design/methodology/approach – The study was conducted anonymously on a group of 253 telephone customer support services employees from companies operating in Poland. Moderation analyses for statistical verification were conducted with macro PROCESS version 3.3. Findings – The research confirmed a significant statistical relationship between JB and all the studied variables: POS, OCBs and CWBs. It may be concluded that JB increases the probability of CWBs and decreases employee readiness for OCBs.When employees experienced POS, a reduction in tendency for counterproductive behaviours was observed, while citizenship behaviours remained unchanged. Thus, it might be concluded that POS levels off the intensity of the influence of JB on employees’ organisational behaviours. Practical implications – Individuals who lack enough social resources to perform their job tasks limit their citizenship activity in the workplace within their behavioural strategy (helping, initiative, etc.). In order to protect their resources, they may also display strategies that are destructive for their organisational environment, e.g. incivility or production deviance. In the situations when the organisation and superiors provide employees with support and demonstrate concern for providing comfortable working conditions, such persons, even in the case of perceived emotional exhaustion, maintain a high level of job activity. Originality/value –The combination of variables presented in the researchmodel explains the significance of the chosen determinants of behaviours that are key from the perspective of the organisation’s effectiveness and market competitiveness. This research extends knowledge pertaining to the relationship between JB and organisational behaviours.


Introduction
In recent years, professional burnoutdefined as a specific psychological condition in which people experience emotional exhaustion, lack of personal accomplishment and a tendency to depersonalise others (Maslach et al., 2001) has been frequently raised as an issue both by researchers of organisations and representatives of international institutions (Schaufeli, 2018). This is because organisational stress and a low level of "psychological well-being", which are related to job burnout (JB), negatively impact job outcomes while generating external economic costs related to, among others, preventing this syndrome and treating the people whom it affects (Taris, 2006;Swider and Zmimmerman, 2010). For example, the European Agency for Safety and Health at Work (2014) notes that stress at the workplace and professional burnout generate annual costs amounting to approximately V620 billion for the European economy.
In the Polish context, recent analyses published in Work Safety in Poland 2019 (2020) show that as many as 24% of Poles experience JB. In addition, 58% are tired and exhausted, 41% report experiencing headaches and 31% feel negative emotions in professional situations. As a result, almost 27% of employees indicate that their malaise impacts the quality and effectiveness of their professional duties, and 25% indicate negative personal and family consequences (Praca Moc Energia w polskich firmach, 2017). Simultaneously, compared to the rest of Europe, employers in Poland are among the least active in tackling stress and burnout at work (Schaufeli, 2018). Therefore, perceived stress may contribute to lower effectiveness with regard to job task realisation and lead to a range of negative outcomes in the work environment.
Although many studies have already been conducted with regard to the relationships between JB and its outcomes, numerous questions remain unanswered. They are connected with specific mechanismsand thus the descriptions of the moderating variableswhich lead us to doubt whether JB decreases individual performance or increases the intensity of behaviours harmful for the organisation in some situations (absenteeism, CWBs, etc.) (Banks et al., 2012;Halbesleben and Leon, 2014;Zhang et al., 2014;Smoktunowicz et al., 2015;De Clercq and Belausteguigoitia, 2020).
According to the job demands-resources (JD-Rs) theory (Bakker and Demerouti, 2017), personal and organisational resources reduce the probability of burnout. In the organisational context, social and organisational support serve as the most important resources. Such support may take the form of actions implemented by the organisation and the employee's superiors, which are often referred to as perceived organisational support (POS). POS denotes employees' perceptions of how the organisation values their contributions and cares for their well-being (Eisenberger et al., 1986). Such perceptions are based on the frequency and intensity of organisational manifestations of approval via compliments and rewardsmaterial and socialfor the efforts undertaken by the employee. POS may eliminate the negative influence of JB on employee behaviours (Brink et al., 2016;Aronsson et al., 2017;Virgolino et al., 2017;Naseer et al., 2018).
The purpose of this paper is to describe how occupational burnout translates into employee behaviour and to explain the role of POS in this connection. The relationships conceptualised in this study are presented in Figure 1.
This study makes three primary contributions to the business literature. First, it tests how POS differentiates the relationship between burnout and organisational behaviour (CWBs and OCBs). Previous studies treat POS as a predictor of burnout, not as a moderator between burnout and employee behaviours. Therefore, this work presents a new perspective for studying these relationships. Second, this study focuses on two types of behaviours (CWBs and OCBs), which allows a more complete discussion of the studied phenomenon. Employees are likely to select and combine various behaviours in burnout situations rather than responding with only one behaviour (Petita and Vecchione, 2011). Thus, adopting a comprehensive perspective is crucial to understanding the responses of employees who experience burnout. Third, this study was conducted in the Polish context and with employees from the information and communication technologies (ICTs) sector, and thus, its results may be interesting for comparative research.
Theoretical framework and hypotheses JB JB is predominantly understood as a state of physical, emotional and mental exhaustion caused by prolonged engagement in situations which are emotionally draining (Schaufeli et al., 2009). This term was first used in the United States at the end of the 1970s as a metaphor to describe a career crisis among professionals working with people in some capacity (Schaufeli, 2017). According to Maslach et al. (2001), JB manifests in three dimensions: emotional exhaustion, depersonalisation and lack of personal accomplishment. Emotional exhaustion is characterised by a chronic sense of physical and emotional depletion and is manifested by physical fatigue and feelings of emotional drain. The most common symptoms of emotional exhaustion include lack of motivation, irritability, feelings of hopelessness and nervousness. Depersonalisation is the perception of detachment from work via emotional callousness. It manifests as negative, callous and cynical behaviours. Employees may interact with colleagues in an impersonal manner. Depersonalisation may be expressed as unprofessional comments directed towards co-workers or customers. Diminished personal accomplishment occurs when individuals feel that they are unable to perform their jobs adequately or are unable to have positive personal interactions. It is also manifested as generalised poor professional self-esteem as well as the tendency to evaluate one's work negatively and feel insufficient with regard to the ability to perform one's job (Maslach et al., 2001).
Although the three-factor concept of burnout serves as the gold standard to describe and assess burnout, some researchers criticise this approach on conceptual, practical and psychometrical grounds (Cole et al., 2012;De Beer et al., 2020). Moreover, given the latest research findings, it has been suggested that the theoretical structure of occupational burnout should be rethought and supplemented with new components, including cognitive impairment (De Beer et al., 2020).
Burnout can be explained by several theories and models (Schaufeli et al., 2009). However, the JD-R theory is predominantly utilised to explain burnout (Schaufeli and Taris, 2014;Bakker and Demerouti, 2017). Job demands refer to those physical, psychological, social or organisational aspects of the job that require either sustained physical or psychological (cognitive and emotional) effort or both and are, therefore, associated with certain physiological, and perhaps, psychological costs. Examples include high work pressure, role overload, emotional demands and poor environmental conditions. Job resources refer to those physical, psychological, social or organisational aspects of the job that are (1)  Job burnout and employee behaviours work goals, (2) reduce job demands and the associated physiological and psychological costs or (3) stimulate personal growth and development. Resources may be located at the organisational level (e.g. salary, career opportunities and job security), or they may take the form of interpersonal and social relations (e.g. supervisor and co-worker support and team climate), the organisation of work (e.g. role clarity and participation in decision making) and the level of the task (e.g. performance feedback, skill variety, task significance, task identity and autonomy) (Bakker and Demerouti, 2017). The JD-R theory indicates that in situations with high professional demands, the employee's mental and physical condition deteriorates, which initially results in lowered performance with regard to professional activities, and finally exhaustion, depersonalisation and cynicism (leading to the impairment of health). However, when employees are provided the appropriate resources, the process of mutual equalisation and buffering occurs, and thus, people do not experience negative psychological and behavioural consequences. Six independent meta-analyses confirm the general assumptions of the theory and indicate that high job demands increase the risk of burnout and lead to negative outcomes, such as health complaints or turnover intentions. On the other hand, job resources play a motivational role, stimulate work engagement and foster positive organisational outcomes, such as performance or organisational commitment (Crawford et al., 2010;Halbesleben, 2010;Alarcon, 2011;Nahrgang et al., 2011;Lesener et al., 2019).

JB and CWBs
Employees with burnout syndrome can withdraw from professional activities and performed roles in several ways. They may take the form of absenteeism, abandonment of performed job activities or intention to leave the organisation (Swider and Zimmerman, 2010). Even if the employee remains at the organisation, those affected by burnout are inefficient, and the quality of activities they perform is low. The meta-analysis performed by Shoji et al. (2016) shows that JB exerts an important influence on self-efficacy, which, in turn, may lead to a decrease in task performance. Moreover, employees with burnout syndrome stop identifying themselves with the organisation and other employees, which, in effect, may induce interpersonal conflicts (Leiter and Maslach, 1988). Additionally, the experience of stress triggers impulsive actions, which result in increased irritability, hostility and aggression (Spector and Fox, 2005). In this context, it is not surprising that the results of numerous empirical studies indicate that JB is a predictor of CWBs (Bolton et al., 2012;Banks et al., 2012;Zhang et al., 2014;Smoktunowicz et al., 2015;Shkoler and Tziner, 2017;Cohen and Diamant, 2019;Baka, 2020;Yun et al., 2020).
CWBs (also termed workplace deviance) are usually defined as employees' intentional activities which comprise a breach of organisational norms and may include sabotaging activities of other employees, in effect leading to decreased labour productivity for the whole organisation (Spector and Fox, 2005;Mercado et al., 2018). The most common manifestations of CWBs include abusing others (e.g. ignoring or influencing others negatively), sabotage (e.g. destroying organisational property), theft, production deviation (e.g. intentional activities that decrease labour effectiveness) and withdrawal (e.g. tardiness or taking longer breaks) (Spector and Fox, 2005).
The JD-R theory explains the relationship between JB and CWBs. Resources include objects, conditions and personal characteristics. Loss of these resources or a threat of such a loss may result in the experience of stress. The framework assumes that people strive to retain, protect and build upon what they value (their resources). Negative outcomes (emotional exhaustion, depersonalisation and lack of personal accomplishment) occur when these valued resources are threatened or lost, are deemed inadequate to deal with demands or do not yield anticipated returns. Following burnout, individuals may engage in organisationally deviant behaviours in an effort to protect limited resources (Bolton et al., 2012). As emphasised by Banks et al. (2012, p. 201), drained or depleted cognitive resources decrease the ability of an individual to selfregulate both emotionally and cognitively, which finally leads to counterproductive behaviours. These may include ignoring clients or customers, refusing to fulfil work demands or taking longer breaks than permitted. Shoshan and Sonnentag's (2020) analytical study shows that employee depersonalisation has a negative effect on customer perceptions of the employees and the service offered by the organisation.
Therefore, it might be assumed that employees' intentional behaviours harmful to the organisation and other members of the organisation are a reaction to situations experienced within the work environment. Bolton et al. (2012) argue that emotional exhaustion leads to depersonalisation and organisational misidentification, which result in CWBs of employees. The feeling of exhaustion leads to a strong cognitive or affective association between the organisation and the negative personal experience, such as the demanding situation responsible for the exhaustion. Emotionally exhausted individuals may feel a sense of value incongruence with the organisation, blaming it for their exhaustion (Cohen and Diamant, 2019). Therefore, CWBs result from an emotional response with the intention to attack the situation and possibly to "passively and indirectly cope with the situation" (Spector and Fox, 2005). Based on the above arguments, the following hypothesis was formulated.
H1. Burnout is positively related to CWBs.
OCBs are understood as behaviours that are voluntary and intentional (although often not fully recognised by the formal system of employee operations), directed at the welfare of an organisation or employee group. They are undefined in job roles, and thus, are not covered by the remuneration system. Moreover, they contribute to the effectiveness and productivity of the organisation as a whole (Organ et al., 2006). They comprise activities such as helping, sportsmanship, organisational loyalty, organisational compliance, individual initiative, civic virtue and self-development. Despite the fact that various authors divide OCBs into specific factors (helping, sportsmanship, etc.), it is possible to refer to one general construct, namely the tendency to cooperate and help in organisational settings (LePine et al., 2002). Such behaviours, sometimes also described as extra-role behaviours or contextual performance, are relevantly connected with organisational effectiveness and financial results (Podsakoff et al., 2009).
The relationships between JB and OCBs may be explained by referring to the JD-R framework discussed above. The following can be surmised logically with regard to the indicated dimensions of JB, namely emotional exhaustion, depersonalisation and lack of personal accomplishment: Employees who feel exhausted by job demands and role overload will presumably seek to protect their remaining psychological and physical resources. In addition to undermining people's ability to engage in OCBs, job stress may diminish their motivation to consider such discretionary behaviours (De Clercq and Belausteguigoitia, 2020). Similarly, when employees feel distressed by their job situation and suffer emotionally or physically as a consequence, their ruminations about their organisational functioning might distract them enough to worry about their ability to complete their regular job tasks (McCarthy et al., 2016). Therefore, they will not become engaged in extra-role job behaviours (Cohen and Abedallah, 2015). It can also be assumed that emotionally exhausted employees will feel more tired, expend less effort at work and be unwilling to help others. Cropanzano et al. (2003) report mixed results concerning this relationship. They found that emotional exhaustion predicted OCBs towards the organisation but not with regard to the supervisor. A probable explanation of these results was that it might be easier for emotionally exhausted individuals to withhold citizenship behaviours beneficial to the organisation rather than to a single person with whom they were likely to frequently interact (Cropanzano et al., 2003).

Job burnout and employee behaviours
In terms of depersonalisation, the distance between oneself and others and treating people like objects may inhibit pro-social activity and withdrawal from OCBs. For instance, Huarng (2001) shows that 36% of the software developers who participated in that study indicated that they depersonalised others regularly to cope with the demands of the people they encountered on the job.
Some authors conclude that depersonalisation may be a coping strategy (Bolton et al., 2012) that helps emotionally exhausted employees distance themselves from work demands to prevent further resource depletion (Hobfoll et al., 2018). Consistent with the JD-R theory, under conditions of extreme resource depletion, individuals are most likely to prevent further resource loss and withdraw from social interaction to protect their remaining resources. This conclusion can be confirmed by the results of the meta-analysis carried out by Nielsen et al. (2017), who show that the lack of resources leads to a decrease in well-being and employee performance.
Reduced personal accomplishment is prompted by a work situation with chronic, overwhelming demands that erode one's sense of effectiveness. Feelings of personal accomplishment may elicit a sense of obligation to help others, resulting in OCBs. Conversely, employees with little self-efficacy will also feel unable to help others.
Further to the above, the following initial hypothesis was formulated.

Moderating role of POS between JB and CWBs, and JB and OCBs
The relationship between burnout and behaviours is analysed by accounting for intermediate mechanisms.
The JD-R theory assumes that moderating factors, namely personal resources (e.g. optimism and self-efficacy) and organisational resources (e.g. social support and leadership support), are present. The meta-analyses conducted by Luchman and Gonz alez-Morales (2013) demonstrate that although job requirements influence JB the most, social support is another very important element in differentiating employee reactions. In another meta-analysis, Kim et al. (2018) show that social support has a significant negative relationship with personal burnout. Moreover, when individuals employed in an organisation sense that they can rely on their superiors (Zhang et al., 2014) or that the organisation provides them with the necessary support and resources for performing their job duties, their reaction to experienced stress will be less negative. Thus, it might be inferred that the relationship of JB with behaviours in the workplace will differ for persons provided with organisational support. POS may indicate that an employee feels that assistance and resources will be available when needed, and that continued effort will be rewarded and social and emotional needs will be met (Eisenberger et al., 1986). Consequently, it might be stated that POS provides employees with resources indispensable for dealing with difficult organisational conditions, and thereby, it may help maintain the current level of job task performance. Pow et al. (2017) show that perceived support not only translates into behaviours but also reduces occupational stress and has a positive impact on the sleep quality of employees.
As per a meta-analysis conducted by Ahmed et al. (2015), POS explains both the attitudes of employees (engagement, satisfaction, commitment and turnover intention) and their behaviours, specifically OCBs. The standardised correlation for the relationship between POS and OCB is estimated using 38 different studies and reported to equal 0.48. Other empirical studies also indicate that JB translates more strongly into counterproductive behaviours when social and organisational support is lacking (Nielsen, 2014;Smoktunowicz et al., 2015;Brink et al., 2016;Mercado et al., 2018;Baka, 2020), and POS balances the influence of JB on OCBs (Brown and Roloff, 2015;Naseer et al., 2018).
In accordance with the above discussion, the following two hypotheses were proposed.
H3a. POS moderates the relationship between experienced burnout and CWBs.
H3b. POS moderates the relationship between experienced burnout and OCBs.

Participants and procedure
The study was conducted on 253 employees from the telephone customer support divisions in ICT, insurance and information technology organisations operating in Poland. The study lasted from May 2016 to March 2018. The selection of the research sample was dictated by the fact that interactions with customers are conducted by employees in customer, social or medical services, which also report relatively higher rates of JB (Scherer et al., 2020). Moreover, this group of employees is highly exposed to consumer incivility, which negatively affects their well-being and can lead to JB (Arnold and Walsh, 2015). A homogeneous group was selected for the study as various occupational groups react differently to the experienced burnout (Restrepo et al., 2015). The socio-demographic characteristics of the sample are described in Table 1. The subjects worked at corporations hiring over 1,000 employees, wherein 38, 38 and 24% belonged to largesized (251-1,000 employees) and middle-sized (51-250 employees) companies, respectively. The sample was dominated by individuals with university education (75%). Females comprised 49% of the sample. The majority of the sample was composed of respondents aged 26-35 years (51%), followed by those aged 36-45 years (28%). In terms of job seniority, 64 and 27% of the subjects reported having worked for over 5 years and 1-5 years, respectively.

Measures
The survey was conducted with the following set of variables.
JB was measured with the 14-item Maslach Burnout Inventory (MBI), which was validated in the Polish cultural context by Chirkowska-Smolak and Kleka (2011). The instrument consists of three subscales: exhaustion (four items, including I feel tired when I get up in the morning and have to face another day on the job), cynicism (four items, including I have Respondents answer the question of how often they feel a particular way with reference to a 7-point (0-6) Likert scale, where 0 indicates "never" and 6 denotes "daily". The internal consistencies of the instrument was estimated to be 0.86. CWBs were measured with the 8-statement instrument based on Bennett and Robinson's (2000) questionnaire adopted for the Polish context by Turek et al. (2014) (examples of its items include Violating workplace health and safety rules or principles and Intentionally disturbing other employees in the performance of their work). The reliability of the tool was estimated to be 0.79. The study participants filled out the questionnaire using a 5-point Likert scale, where 1 and 5 denoted "never" and "always", respectively.
OCBs were measured with the 16-statement instrument adapted and validated for the Polish context by Wojtczuk- Turek and Turek (2016). The reliability of the tool was estimated to be 0.92. The following are the examples of the questionnaire statements: Helping people who have many job duties; When the situation demanded it, staying after hours or performing work at home; and Caring for the atmosphere at work. The study participants filled out the questionnaire using a 5point Likert scale, where 1 and 5 denoted "never" and "always", respectively.
POS was measured with the questionnaire developed by Eisenberger et al. (1986). Among the 36 statements originally developed by the authors, 6 statements that had been used by Brink et al. (2016) and other researchers were selected. The adaptation and validation of the questionnaire for the Polish context were carried out by Turek (2019). The reliability of the tool was estimated to be 0.89. The following are the examples of the questionnaire statements: Help is available from the organisation when I have a problem and The organisation really cares about my well-being. The study participants filled out the questionnaire using a 5-point Likert scale, where 1 denotes "strongly disagree" and 5 signifies "strongly agree".
While selecting the control variables, we followed the recommendations offered by Becker et al. (2016). Burnout is expected to be higher among younger workers and those in their early career stages. The evidence for gender (i.e. whether males or females are more prone to burnout) is mixed (Maslach et al., 2001). Therefore, the following control variables were considered: sex, age, education and tenure.

Treatment of common method bias
As all the questionnaires were completed via self-report, the risk of common method bias stands (Podsakoff et al., 2003). To avoid this bias, two methods, Harman's single factor test and common latent factor (CLF), were used. The results of Harman's single factor test showed that the single factor of each variable explained less than the suggested threshold (i.e. 50% of variance). The results of the CLF test showed that the regression weights for models with and without CLFs provided delta values much lower than 0.20, a commonly used threshold. Neither test revealed any common method bias.
In general, the use of self-report forms as research tools is empirically justified. Metaanalyses of studies conducted with self-assessment questionnaires for CWBs, OCBs and external measures of behaviour (e.g. assessments of colleagues and superiors) indicate that large coherence exists between different sources of data (Berry et al., 2012;Carpenter et al., 2014;Mercado et al., 2018). Moreover, after examining the psychological and attitudinal measures reported in the industrial/organisational literature, Spector (2019) points to the benefits of using self-report questionnaires and conducting cross-sectional research.

Measurement models
The postulated model (Figure 1) suggests the existence of direct relationships between JB and OCBs as well as JB and CWBs. POS serves as the moderator of these relationships. For testing the research hypotheses, a series of nested models were studied. AMOS software (ver. 25) was used to verify the research models. The results are shown in Table 3. This study used a baseline (four-factor) model and estimated all the theorised relationships between the stated constructs. The measurement model was assessed through confirmatory factor analysis (Kline, 2011), which comprised four latent variables. Different indices, namely, the root-meansquare error of approximation (RMSEA), comparative fit index (CFI), Tucker-Lewis Index (TLI), standardised root mean residual (SRMR) and χ 2 were employed to calculate the model fit (Hu and Bentler, 1999;Kline, 2011). The values of these fit indices (χ 2 5 335.978, df 5 173; p < 0.173; RMSEA 5 0.068; CFI 5 0.954; TLI 5 0.942; SRMR 5 0.066) indicated that the measurement model provided the best fit to the data.
The construct validity was evaluated with composite reliability (CR) and convergent and discriminant validity. The consistency reliability was tested through Cronbach's α and CR. Table 2 indicates that all values of Cronbach's α exceeded 0.70, suggesting that they were valid for the analysis. Similarly, all the CR results ranged from 0.711 to 0.822, higher than the threshold value of 0.6, thus confirming internal consistency reliability (Fornell and Larcker, 1981). Convergent validity was assessed using the ratio of the average variance extracted (AVE). The AVE of four constructs ranged between 0.621 and 0.766, which exceeded the standard threshold of 0.5 for convergent validity. The discriminant validity of the measurement model was evaluated with the construct correlation values. In this respect,  (Henseler et al., 2015) and consists of drawing a comparison between the AVE value in each construct and the square of the correlation between the construct and each variable. Therefore, if AVE exceeds the squared correlation, it can be accepted that each construct relates more intensely to its own measurements than to those of other variables. Table 4 presents the square root of AVE in the diagonal and the correlations estimated for each pair of constructs in the elements outside the diagonal. This information confirms the existence of discriminant validity in constructs.

Hypotheses testing
Regression-based analysis in SPSS macro PROCESS (ver. 3.3) using the recommended 5,000 bootstrap sampling with 95% confidence intervals (CIs) (Hayes, 2018) was employed to test the proposed relationships. The results are depicted in Table 5. Hypotheses 1 and 2 indicated that JB is significantly and positively related to CWBs, and significantly and negatively related to OCBs. The results revealed that JCB is significantly related to CWBs (β 5 0.245, p < 0.05) and OCBs (β 5 À0.212, p < 0.01), thus supporting Hypotheses 1 and 2. For Hypotheses 3a and 3b, it was assumed that POS is the moderator of the relationship between JB and CWBs and between JB and OCBs. The model of linear regression was used to test the effect of interaction. The results presented in Table 5 show that although JB is significantly related to CWBs and OCBs, in the presence of POS, the value of β is small. On the other hand, a combination of the two variables significantly increases the explained variance in the results, both for CWBs and OCBs. Finally, the regression model for CWBs explains 41% of the variance in the results, while the corresponding value for OCBs is 33%.
Graphic representation of the effect of interaction for CWBs ( Figure 2) shows that at low levels of burnout, POS lowers the CWBs output level insignificantly, whereas, with high JB, POS reduces employees' tendency to engage in behaviours harmful to the organisation. The interaction effect of JB and POS on CWBs is significant, as indicated by a bootstrapping CI that does not comprise zero (β 5 À0.159, lower 95% CI 5 À0.306, upper 95% CI 5 À0.012; p < 0.05).
A similar situation is observed for the variable OCBs (Figure 3). For a low level of JB, perceived support increases (by a small extent) the incidence of behaviours exceeding the job role, whereas, with a high level of JB, organisational support causes OCBs to persist at a relatively high level (β 5 0.121, lower 95% CI 5 0.195, upper 95% CI 5 0.046; p < 0.05).
The above results apply to the postulated research Hypotheses 3a and 3b. The hypotheses stated that POS reduces the influence of JB on employee organisational behaviours. As per the results of the presented analyses, the hypotheses proved to be correct. Organisational support causes employees to react differently to the symptoms of JB, and by the same token -     Table 5.

Effect on CWBs and OCBs
Job burnout and employee behaviours which can be inferred indirectlycauses organisations to incur lower losses related to JB. This finding can be explained as follows: employees who perceive high levels of POS do not increase their counterproductive behaviours and do not reduce citizenship behaviours due to the experienced JB. They maintain a level of behaviour similar to those who do not experience burnout.

Discussion
Drawing on the JD-R theory (Bakker and Demerouti, 2017), this study details how employees' experiences of JB might translate into their CWBs and OCBs, and how their access to perceived social support might modify such behavioural responses. To some extent, prior research already shows that the energy-depleting effects of JB experienced by employees can be countered by their access to pertinent resources (Virgolino et al., 2017;Naseer et al., 2018;  De Clercq and Belausteguigoitia, 2020). This evidence has evoked the need for studies that explicate when JB influences counterproductive and citizenship behaviours.

Theoretical implications
The following theoretical implications can be drawn from this study. First, JB translates into the "psychological well-being" of individuals and their health-related consequences and is also strongly related to work results (Taris, 2006;Swider and Zmimmerman, 2010;Crawford et al., 2010;Halbesleben, 2010;Nahrgang et al., 2011;Demerouti et al., 2014). The studies reviewed in this paper confirm this, as JB significantly and positively correlates with employees' readiness to act in a manner detrimental to the organisation (i.e. via CWBs). This notion is compatible with other empirical studies in which CWBs served as a dependent variable for JB (Marcus and Schuler, 2004;Cohen and Abedallah, 2015;Smoktunowicz et al., 2015;Raman et al., 2016;Baka, 2020). Thus, it can be concluded that exhaustion, depersonalisation and a sense of ineffectiveness caused by the state of JB increase the probability of employees ignoring their job duties, performing them unconscientiously or using various forms of deviance in a workplace. This aspect is especially observable in professions that require contacting clients and responding to their needs and problems (Scherer et al., 2020). However, the mechanism is more complex. For example, Banks et al. (2012) find that JB leads to decreased organisational commitment, which results in CWBs. Another study demonstrates the mediation of the JB-CWB relationship via depersonalisation and disidentification for individuals with low and moderate levels of self-control (Bolton et al., 2012). In general, JB results in depletion of personal resources (personality and psychological capital), leading to psychological withdrawal via depersonalisation, disidentification and consequent CWBs. Thus, mediating variables, such as attitudes (e.g. commitment and job satisfaction) and employee emotions, as well as individual (e.g. personality) and contextual variables (e.g. culture and climate of the organisation, job characteristics and organisational justice) differentiate the relationships between JB and employee behaviours (Mercado et al., 2018).
Another important conclusionwhich also finds support in recent academic publicationsfollows from results which indicate that JB lowers employees' readiness to support their organisation (i.e. JB leads to OCBs) by undertaking initiatives, helping co-workers, observing occupational virtues and rules or taking care of one's personal and professional development (Swider and Zmimmerman, 2010). In general, the experience of job stress drains employees' energy levels, which undermines their ability and motivation to devote substantial energy to voluntary efforts (De Clercq and Belausteguigoitia, 2020). When employees' jobs affect them negatively in emotional or physical terms, they do not retain the discretionary energy needed to be good corporate citizens. Moreover, employees may interpret their suffering from job stress as a signal that their employer does not respect them, and they may experience this suffering as offensive, reducing their willingness to perform any activities that benefit the organisation (McCarthy et al., 2016). However, this mechanism can be buffered by adjusting the employee to the work environment. Kasekende et al. (2020) show that the relationship between emotional exhaustion and OCBs can be mediated through a person-organisation fit.
First, as the presented studies show, the relationship between JB and employees' behaviours differs depending on the context in which the individuals function. It is, therefore, impossible to state conclusively and categorically that JB influences all employees negatively. When such individuals have organisational support or at least perceive that they can rely on their superiors or co-workers, strong relationships of JB with CWBs and OCBs are not observed. This means that POS allows the employees to maintain the current level of effectiveness in performing their job duties. According to Eisenberger et al. (2001), perceiving Job burnout and employee behaviours the organisation as caring for employees' needs, resources and their "psychological wellbeing" triggers the rules of reciprocity, which manifest in maintaining or increasing the level of individuals' engagement in the performed work. Studies on relationships of POS with individuals' behaviour demonstrate that perceived support forms a significant component which levels out the influence of JB on the behaviours of individuals in a workplace (Brink et al., 2016;Ahmed et al., 2015;Brown and Roloff, 2015;Naseer et al., 2018). By focusing explicitly on the buffering effects of POS on the incremental role of job stress in reducing OCBs and increasing CWBs, this study provides insights regarding when the sense of being overburdened is less likely to hinder voluntary efforts. Overall, this study provides extended insights into when energy depletion due to JB prevents employees from taking on additional responsibilities, thereby explicating the interactive roles of job stress and pertinent resources in spurring workplace behaviours. Second, the findings indicate that Polish employees from customer support divisions experience a relatively high level of burnout. This burnout is accompanied by only a moderate level of organisational support. In terms of the intensification of stress at work, customer support roles in the ICT sector are particularly demanding, as they require constant contact with customers and solving their problems (Scherer et al., 2020). If employees performing professional tasks feel overwhelmed and cannot count on support from the organisation, the quality of services will ultimately deteriorate, and consumer dissatisfaction will rise (Shoshan and Sonnentag, 2020).
Third, this study demonstrates that the JD-R theory is a useful framework for explaining the mechanism in which JB translates into employees' behaviours. Individuals who feel they lack the resources to perform their job tasks limit their citizenship activities (e.g. helping and initiatives) in the organisation. In order to protect their resources, they may also display strategies that are destructive for their organisational environment via incivility or production deviance. When the organisation and its leaders offer the employees support and demonstrate concern by providing comfortable working conditions, employees maintain a satisfactory level of job activity despite perceived emotional exhaustion.

Practical implications
The conducted analyses allow us to draw several important practical implications. First, the lack of concern for the "psychological well-being" of employees on the part of the management will negatively influence the results of employees' work and lead to lower organisational effectiveness. This applies to employees who have direct contact with clients or work in emotionally demanding jobs (Aziz et al., 2018).
Second, when employees struggle with difficulties, stress and burnout, it is important to increase their sense of control over their performed duties and provide them with social and organisational support. In such situations, it is helpful to use activities related to job crafting, which may help such individuals to rediscover the interest in their work and level out JB (Bakker and Demerouti, 2017).
Third, it is helpful to create a closer relationship between employees and the management and to form psychological contracts, which will help maintain an appropriate amount of employee engagement and level out the tendency for counterproductive behaviours (Jensen et al., 2010). Thus, organisations must take proactive steps to detect signals of excessive stress, for instance, by creating an open culture in which employees do not worry about voicing their concerns about the stress-inducing aspects of their jobs. Organisations that wish to prevent their employees from engaging in CWBs should also ensure that work demands do not overload the employees and deplete their resources, which in itself may cause higher levels of CWBs as a result of increasing their perception about lack of support because of the extraneous job duties assigned to them by the organisation. Although it might be difficult, if not impossible, to eliminate all job demands, organisations could seek to reduce these taxing job characteristics as much as possible.
Fourth, practitioners may also consider incorporating short work-appropriate breaks to encourage appropriate resource recovery and replenishment. These aspects have been shown to reduce occupational stress and increase workers' well-being and productivity (Largo-Wight et al., 2017).
Fifth, HR departments and line managers should conduct anti-stress training for employees; recent meta-analyses show that such efforts significantly reduce job stress and anxiety as well as improve general well-being (Slemp et al., 2019). Recent research has also demonstrated the importance of mentoring as an important factor leading to lower levels of burnout (Jyoti and Ryan, 2019). In essence, effective and regular mentoring could enable employees to perform their job with less stress and anxiety, thereby promoting psychological detachment from work at home.
Finally, organisations are encouraged to rethink how performance is assessed and compensated (e.g. via contingent rewards), as such practices may promote low-quality work motivation in employees (Tr epanier et al., 2020) and lead to negative effects over time and thus increasing CWBs and decreasing OCBs.
In the context of the services sector in Poland, the focus of this study, taking care of the psychological well-being of employees and providing them with indispensable job resources, will minimise the level of perceived stress, which will consequently translate into increased job efficiency. This finding is consistent with the results of the meta-analysis conducted by Dreison et al. (2018), who show that person-directed interventions are important for reducing employee burnout.

Conclusions and limitations
To conclude, although the studies presented in this paper fill the gaps in the current knowledge on the phenomenon of JB in the context of employees from customer support divisions in Poland's ICT sector, they still suffer from certain limitations that are worth mentioning.
First, the research was of a cross-sectional character, which directly precludes observations as to the causality in the interrelations between the analysed variables. The relationships assumed from the research model can be verified conclusively only via experimental research or longitudinal studies or both. Although a longitudinal design is desirable for empirical analysis of workplace performance outcomes, the presented predictions are grounded in theory, which allows a useful comparison with the existing evidence. Moreover, as pointed out by Spector (2019), the causality of longitudinal design is typically overstated, and longitudinal design offers limited advantages over cross-sectional design.
Second, only one source of data (self-report questionnaires) was used in the study. Thus, the results might be affected by common method bias. Moreover, the risk of endogeneity remains. Future research might reduce the risks posed by common method bias and endogeneity by collecting data simultaneously from multiple sources (co-workers and supervisors). However, a few meta-analyses (Berry et al., 2012;Carpenter et al., 2014;Mercado et al., 2018) show large coherence between different sources of data. Thus, studies conducted with multiple sources of data do not offer significantly more precise data, particularly with regard to employee behaviours in the organisational environment.
Also, the analytical procedure did not distinguish sufficiently between OCBs, CWBs and JB to facilitate a better understanding of the job situation. However, as emphasised by some authors, such distinctions for certain theoretical constructs do not always possess an empirical justification (LePine et al., 2002).
Future research should examine the role of burnout in relation to CWBs and OCBs in more detail. For example, it is worth explaining the co-existence of OCBs and CWBs in the presence of JB more precisely. Some analyses show that employees may demonstrate productive and Job burnout and employee behaviours counterproductive behaviours simultaneously. Else, in certain situations, their productive behaviours may lead to CWBs and vice versa (Fox et al., 2012). Therefore, a more detailed study is required to analyse how JB translates into such functioning of employees. Moreover, it is important to investigate the boundary conditions of the burnout-outcome relationships.
Exploring additional individual-level factors, such as personality variables and positive psychological capacities (e.g. self-efficacy and resilience), would improve our understanding of human resources, organisational behaviour and job burnout. Finally, future research should extend the findings of this work by exploring a stressordetachment model (Sonnentag and Fritz, 2015) incorporating JB and POS and examining the mediating role of psychological detachment in broader JD-R model relationships, including the effects of resources on employees. Yun, M., Roach, K.N., Do, N. and Beehr, T.A. (2020), "It's not how you say it, but what you say: communication valence in the workplace and employees' reactions", Occupational Health Science. doi: 10.1007/s41542-020-00070-5.