flood victims

Justice, social support and stressors 1 Running head: JUSTICE, SOCIAL SUPPORT AND STRESSORS

Why does interactional justice promote organizational loyalty, job performance and prevent mental impairment? The role of social support and social stressors

Kathleen Otto1 & Nihal Mamatoglu2 1 2

University of Leipzig, Germany

Abant İzzet Baysal University of Bolu, Turkey

This is an Accepted Manuscript of an article published by Taylor & Francis in The Journal of Psychology on 10 Mar 2014, available online: http://www.tandfonline.com/doi/abs/10.1080/00223980.2013.866535”

Author’s Note We would like to thank Alexandra Budjanovcanin and Jule Epp for their valuable comments on previous versions of the present manuscript, as well as Patrick Großa and Sabine Schmidt for support in programming the online questionnaire. Please address correspondence concerning this article to Kathleen Otto now at: Philipps University of Marburg, Faculty of Psychology, Work and Organizational Psychology, Gutenbergstr. 18, D-35032 Marburg, Germany; Phone: +49 6421/28-23781; Fax: +49 6421/28-24281; Email: [email protected]

Justice, social support and stressors 2 Abstract Using social exchange theory as a conceptual framework, we investigated the relationship between interactional justice and the outcomes organizational loyalty (affective commitment, turnover intentions), perceived job performance (self-rated performance, personal accomplishment), and mental impairment (cognitive irritation, emotional exhaustion) in an online survey of 218 employees working in the field of computer technology. Specifically, we predicted that interactional justice would heighten the quality of social exchange relationships and therefore expected perceived social support (POS) and bullying to mediate the proposed relationships. We tested our hypotheses applying a latent structural equation model. Our findings revealed that POS mediated the relationship between interactional justice and organizational loyalty, while bullying mediated the relationship between interactional justice and mental impairment. Practical implications are discussed concerning how to foster interactional justice and POS and how to weaken bullying behavior.

Key words: organizational justice, POS, bullying, mobbing, social exchange, SEM

Justice, social support and stressors 3 Why does interactional justice promote organizational loyalty, job performance and prevent mental impairment? The role of social support and social stressors Although a debate is still taking place concerning the “right” number of dimensions for organizational justice, four distinct dimensions have most often been distinguished from each other (for a review, see Bies, 2005): Employees perceive their organization to be just if: allocations of resources, such as tenures, are carried out in correspondence to the employee’s performance (distributive justice; Adams, 1965); employees have influence on organizational decisions and the implementation of changes (procedural justice); employees receive all necessary information (informational justice); and employees experience the social interaction with superiors as respectful and empathetic (interpersonal justice). The two latter constructs are usually summarized as interactional justice (Bies & Moag, 1986; Greenberg, 1993). Colquitt, Conlon, Wesson, Porter, and Ng state in their meta-analysis “… that [the consequences of] interpersonal and informational justice have received less attention than [those of] distributive and procedural justice.” (2001, p. 438; italics added by the authors). Accordingly, in this study we aimed to extend research on the meaning of interactional justice for work-related outcomes (organizational loyalty, perceived job performance) and mental impairment, by exploring social exchange as a mechanism through which perceived interactional justice is important for these outcomes. As research so far has failed to provide a clear picture of the psychological mechanisms explaining the particular role of interactional justice, we seek to fill this gap in the literature by introducing social support and social stressors as mediators explaining its impact on organizational loyalty (high affective commitment, low turnover intentions), perceived job performance (short-term self-rated performance, long-term personal accomplishment) and mental impairment (cognitive irritation, emotional exhaustion). We assume that interactional justice heightens the quality of social exchange relationships

Justice, social support and stressors 4 between the employee and his or her organization or supervisor, leading to higher perceived social support (POS) and a lower occurrence of social stressors (bullying in the workplace). This positive form of reciprocity increases employees’ loyalty to their organization, as well as their (perceived) efficiency in their jobs and leads to reports of better mental health. To the best of our knowledge, only a single study has already considered POS as a mediator in the relationship between interactional justice and organizational outcomes (i.e. organizational identification; Cheung & Law; 2008), whereas no prior research has focused on bullying and its potential mediating role in explaining the consequences of interactional justice. In sum, different from other studies on organizational justice in general, and to interactional justice in particular, we considered two opposing social factors in work settings and their potential mediating roles. In the following sections, we will first outline the central role of social exchange to our research, and then guided by this theoretical framework, we derive our hypotheses. The theoretical model can be found in Figure 1. – insert Figure 1 about here – Social Exchange as a Theoretical Framework Employees assign human characteristics to their organizations. For example, they perceive the caring nature of an organization. They then develop a set of beliefs about the extent to which their organization values their contributions and cares about their well-being. These beliefs have been labeled perceived organizational support (POS). More concretely, POS is described as the “quality” of social exchange that takes place between an employee and an employer (Setton, Bennett, & Linden, 1996; Wayne, Shore, & Liden, 1997). Social reciprocity lies at the core of the employment relationship. Obligations are carried out by an employee in exchange for equitable rewards. In other words, one person does another a favor with the expectation of receiving some future return (Gouldner, 1960). POS is closely related to fairness perceptions, it facilitates trust, and it increases loyalty, which leads to, for example, a greater willingness to remain with the organization and a reduced intention to leave

Justice, social support and stressors 5 (Eisenberger, Hunnington, Hutchinson, & Sowa, 1986). Overall, employees tend to take a long-term approach to social exchange relationships, with the pattern of reciprocity over time determining the perceived balance in exchanges (Blau, 1964). In contrast to contractual obligations, which demand repayment, social exchange creates unspecified reciprocal obligations that are enforced through cultural and normative standards of behavior (Cropanzano & Mitchell, 2005). As a result, individuals are more likely to increase their job performance as well as their extra-role behaviour and commitment (Niehoff & Moorman, 1993; Wayne, Shore, Bommer, & Tetrick, 2002). It is known that supervisors are often considered to be acting on behalf of the organization. Therefore, POS is increased when employees perceive their supervisors to be supportive. We argue that supportive supervisors are those who demonstrate interactional justice in their actions (Bies & Moag, 1986; Greenberg, 1993) such as by giving their subordinates all necessary information (informational justice) and treating them respectfully and empathetically (interpersonal justice). Research suggests that interactional justice in particular facilitates the formation of social exchange relationships (Moorman, Blakely, & Niehoff, 1998; Pillai, Schriescheim, & Williams, 1999; van Knippenberg, De Cremer, & van Knippenberg, 2007). Although the social setting at work might provide support for employees, it can also constitute a major source of stress (van Dierendonck, Haynes, Borril, & Stride, 2004). This is particularly so when negative social behavior such as bullying or mobbing, occurs (Davenport, Distler-Schwartz, & Pursell-Elliott, 1999). Leymann (1996) defined mobbing as “psychological terror (…) in working life [that] involves hostile and unethical communication which is directed in a systematic way by one or a few individuals mainly towards one individual who (…) is pushed into a helpless and defenseless position” (p. 168; italics added by the authors). It has to be argued that different terms and concepts, as for instance psychological terror, scapegoating, health-endangering leadership, work abuse, bullying,

Justice, social support and stressors 6 harassment, or petty tyranny (for a review, see Einarsen, 2000) have been used to reflect the phenomenon where either superiors, colleagues or subordinates are involved in “harassing, offending, socially excluding someone or negatively affecting someone’s work tasks” (Einarsen, Hoel, Zapf, & Cooper, 2003, p. 15). Though in the literature “bullying” and “mobbing” have been mostly employed interchangeably (e.g., Escartín, RodríguezCarballeira, Zapf, Porrúa, & Martín-Peña, 2009), some authors question “whether the word utilized in Scandinavia, ‘mobbing’, covers the same range of meanings as the English term ‘bullying’… [that] is perceived as covering a wider phenomenon” (Agervold, 2007, p. 171; italics added by the authors). In the following thus the term bullying will be used, except when citations from other sources are made. In particular, leadership behavior has been found to be a main precursor to bullying in the workplace (Skogstad, Torsheim, Einarsen, & Hauge, 2011). Leaders who are especially authoritarian or abusive, and who have little respect and tolerance for their employees, are considered to contribute to bullying. This kind of supervisor can psychologically harm their employees with their discourteous behavior. Accordingly, abusive leadership was shown to be negatively associated with interactional justice (Burton, & Hoobler, 2011). Interactional Justice and Outcomes Meta-analytic findings emphasize the role of organizational justice for work-related outcomes (see Cohen-Charash, & Spector, 2001; Colquitt et al., 2001). Furthermore, there are comprehensive reviews which suggest that organizational justice buffers against work-related stress (see Greenberg, 2004; Vermunt & Steensma, 2005). In the following, we present findings relevant to our hypotheses (outlined below) focusing on the relationship between interactional justice and outcomes only. First, however, we argue that the experience of interactional justice promotes organizational loyalty. From a theoretical standpoint it can be assumed that supportive, empathetic, information-sharing supervisors arouse positive feelings and thoughts in

Justice, social support and stressors 7 employees. As a result employees feel highly affectively committed to their organization and do not wish to work anywhere else (low turnover intentions), i.e., they remain loyal. Research to date indicates that interactional justice leads to stronger commitment (e.g., Leung, Wang, & Smith, 2001) and in particular to affective commitment (Barling, & Phillips, 1993; CohenCharash, & Spector, 2001). Withdrawal has also been extensively studied because of its negative consequences for the efficient functioning of organizations and the high costs for employers (Sagie, & Birati, 2002). Specifically, the experience of interactional justice at work has been found to be related to less workplace absenteeism (Gellatly, 1995) and lower turnover intentions (Sparr, & Sonnentag, 2008). Second, we also predict that the perception of interactional justice sustains (perceived) job performance. In accordance with Viswesvaran and Ones (2000), we define job performance as actions, behavior and outcomes that an employee contributes to fulfilling organizational goals. It can be argued that a good relationship with one’s supervisor, involving respectful treatment and the provision of adequate information and fair feedback, might be a helpful condition for performing well in one’s job (cf. Williams, 1999). In line with this thinking, different studies have found leader interactional justice to be a predictor of employee task performance (Masterson, Lewis, Goldman, & Taylor, 2000; Wang, Liao, Xia, & Chang, 2010) and to explain organizational citizenship behavior (Byrne, 2005). Metaanalytic evidence can also be found for the relationship between interactional justice and performance (Cohen-Charash, & Spector, 2001), and for informational justice in particular (Colquitt et al., 2001). Third, we believe that interactional justice might be seen as a vital resource preventing mental impairment. Supportive supervisors are those who give their subordinates all necessary information (informational justice) and treat them respectfully and empathetically (interpersonal justice), which should foster positive feelings among their subordinates and accordingly affect their psychological health positively. Confirming this assumption, research

Justice, social support and stressors 8 has found that the perception of interactional justice in the workplace is related to lower job depression (Sparr, & Sonnentag, 2008), lower reported anxiety (Spell, & Arnold, 2007), and fewer general health problems (Kivimäki, Vahtera, Elovainio, Virtanen, & Siegrist, 2007). Experimental studies found interactional injustice to arouse negative emotions in the short run (Stecher & Rosse, 2005), which we believe might result in the draining of emotional resources of employees (Michinov, 2005) in the long run. Moreover, interactional injustice might not only have adverse effects on the affective dimension of mental health but also on its cognitive dimension. Hence, in addition to emotional exhaustion (Cole, Bernerth, Walter, & Holt, 2010) we focus on cognitive irritation, which describes “a state of mind in which individuals cannot restrain themselves from ongoing regulatory cognitive processes” (Mohr & Wolfram, 2010, p. 171). Irritation was shown to be a precursor to further severe mental impairment (Mohr, Müller, & Rigotti, 2005) such as depressive reactions (Dormann & Zapf, 2002), for example. Based on these theoretical assumptions and former research, we arrive at the following hypothesis: Hypothesis 1: Interactional justice will be associated with (H1a) increased organizational loyalty, (H1b) higher perceived job performance, and (H1c) lower mental impairment. Perceived Organizational Support as Mediator POS can be conceptualized as an organization-based resource associated with employees’ general conviction that their organization values their contribution and cares about their well-being (Kinnunen, Feldt, & Mäkikangas, 2008). McFarlane, Shore, and Shore stated that POS “focuses on the role of employer commitment in exchanges with employees, [whereas] organizational justice focuses on the fairness of exchanges between employee and employer” (1995, p. 149; italics added by the authors). Eisenberger, Fasola, and DavisLamastro (1990) argued that the perception of POS creates a feeling of obligation in

Justice, social support and stressors 9 employees to repay the organization with the desired work-related attitudes and behavior. (Interactional) justice perceptions are, for the most part, more immediate, whereas POS seems to involve long-term development, reflecting the history of how the organization and its representatives have dealt with their employees to date. In line with this assumption, a handful of recent studies underlined that POS functions as a mediator in the relationship between (distributive) justice and organizational citizenship behaviors (Cho & Kessler, 2008), as well as between (interpersonal and informational) justice and organizational identification (Cheung & Law, 2008). As already considered, when employees believe that their supervisor behaves supportively and respectfully, and shares organizational information with them, or in another words, when employees perceive high interactional justice, POS is also increased. Then because of social exchange principles employees repay this favorable behavior with positive attitudes and behaviors. As a result, the norms of social reciprocity, individuals are more likely to increase their loyalty towards their organization and also put more effort into their job (i.e., higher job perceived performance). This line of thinking leads to the following hypothesis: Hypothesis 2: POS will fully mediate the relationship between interactional justice and (H2a) organizational loyalty as well as (H2b) between interactional justice and perceived job performance. Workplace Bullying as Mediator We also expect that where interactional justice is demonstrated by the supervisor, the occurrence of social stressors, reflected in the present study by bullying in the workplace, is less likely. Firstly, empirical evidence supports a negative association between interactional justice and bullying (Tsuno, Kawakami, Inoue, & Abe, 2011). From a theoretical perspective, it seems logical to argue that when supervisors treat their subordinates with little respect (i.e., low interpersonal justice), this might also legitimize coworkers treating others in an unkind

Justice, social support and stressors 10 and abusive way, free from the threat of punishment by other employees. It has also been argued that those who hold power may abuse this by bullying others who are less powerful (Hodson, Roscigno, & Lopez, 2006). Former research has demonstrated that, specifically in the context of psychological and job-related well-being, bullying (or mobbing) can be regarded as an extreme social stressor, (e.g., Davenport et al., 1999; Rodriguez-Monuz, Baillien, De Witte, Moreno-Jeminez, & Pastor, 2009) that can lead to harm and long-term general ill health for the affected individual (e.g., Zapf & Leymann, 1996). In additional to its consequences for mental impairment, permanent encounters with social stressors will also have a detrimental impact on employees’ (perceived) job performance (Nagami, Tsutsumi, Tsuchiya, & Morimoto, 2010). Working in a social setting where one is confronted with insults or dishonesty by others might affect the trust in one’s own abilities directly or, alternatively, the constant ruminations about the reasons of this kind of negative treatment might distract from performing one’s job well as could be derived from experimental findings on dealing with stereotype threat (e.g., Smith, & Postmes, 2011; Steele, & Aronson, 1995). From a social exchange theoretical point of view, the expected negative consequence of bullying on (perceived) job performance could also be traced back to assumptions made by the Effort-Reward Imbalance (ERI) Model (Siegrist, 1996). The model argues that social reciprocity lies at the core of an employment contract, where obligations are performed in exchange for equitable rewards. We believe that if employees perceive an imbalance caused by high effort (job performance) and low reward (being the target of workplace bullying), they might try to restore this kind of non-reciprocity by putting less effort into performing their job well. Umphress, Labianca, Brass, Kass, and Scholten (2003) noted that “…justice perceptions are not formed in isolation; rather, they are subject to the influences of those with whom we interact” (p. 739). Caused by social contagion, we argue that when a supervisor cares for his or her subordinates within his or her team, an atmosphere of mutual

Justice, social support and stressors 11 responsibility results.Thus, based on a positive social exchange quality, we assume that when supervisors treat their subordinates respectfully and share relevant information in a timely manner with all of them, a supportive social climate results within the team where every team member feels responsible for the well-being of his or her coworkers. This supportive climate, indicated by a low probability of bullying experiences, helps employees to do their job efficiently and to stay healthy. Thus, drawing on the above literature in relation to interactional justice, perceived job performance and mental impairment, as well as considering the mediating role of bullying as a social stressor, we wish to test this final hypothesis: Hypothesis 3: Bullying in the workplace will fully mediate the relationship that interactional justice has with (H3a) perceived job performance, as well as with (H3b) mental impairment. Method Sample and Procedure Data were collected in several organizations in Germany whose business is related to computer science or information technology. Organizations were targeted whose core concern is in the field of technology (e.g., developing software or hardware) and/or in the field of service related to technology (e.g., sales of information technology products, client support). Initially, an employee, usually from the Human Resource (HR) department of the selected organizations, was approached and asked to participate in an online survey about justice in the workplace. In total, six organizations agreed to take part in the study. The web link to the study was placed on the university’s website to emphasize that this investigation was not a survey being undertaken by HR in the respective organizations. The contact in each participating organization was then instructed to share the link with all employees in the organization that have internet access. A password was provided together with the link to make sure that only employees from the approached organizations could participate and

Justice, social support and stressors 12 complete the questionnaire. All subjects were reassured that participation would be confidential, and in particular, that their superiors would not be able to identify their individual data. In total, 218 employees (51.8% female) took part in the study. The age of participants varied between 20 and 57 years (M=33.48; SD=9.00). Eighteen participants worked as unskilled blue collar workers, 28 as skilled blue collar workers, 93 as lower level white collar workers, 67 as intermediate white collar workers or supervisors of white collar workers, 3 as upper white collar workers, middle management or executive staff, and 9 in management positions or as directors. The participants had worked in their job for up to 36 years (M=7.32; SD=7.30). Research Instruments Interactional justice. We assessed both dimensions of interactional justice using the German version of Colquitt’s (2001) measure which has proved to be valid and reliable in previous studies (see Streicher et al., 2008). The response options of both scales varied between 1 (to a small extent) and 5 (to a large extent). Informational justice was assessed with 5 items (Cronbach’s =.90; e.g., "Has he/she [the superior] seemed to tailor his/her communications to individuals’ specific needs?"). Interpersonal justice was measured with 4 items (Cronbach’s =.86; e.g., "Has he/she [the superior] treated you in a polite manner?"). Social support and stressors. Data concerning POS was gathered using a shortened version of Eisenberger et al.’s (1990) 36-item scale (4 items; Cronbach’s =.79; e.g., "My organization really cares about my well-being"), as has been applied in former research (Rigotti, Mohr, Clinton, & Guest, 2010). The scale ranged from 1 (strongly disagree) to 5 (strongly agree). To assess bullying in the workplace, we applied the Work Harassment Scale (cf. Björkqvist, Österman, & Hjelt-Bäck, 1992). For reasons of parsimony, we shortened what was the origin 24 item-measure. As according to a recent model by Rodríguez-Carballeira, Escartín, Visauta, Porrúa, and Martín-Peña (2010) workplace bullying could take six different

Justice, social support and stressors 13 forms (e.g., isolation, control and manipulation of information, devaluation of the professional role), we selected 6 items covering the various facets based on face validity. The participants were questioned: "How often have you been exposed to some of the following actions during the last 12 months?" The answering mode ranged from 0 (never) to 4 (very often; Cronbach’s =.81; e.g., "Lies about you told to others?"). Organizational loyalty. Affective commitment and turnover intentions were measured to reflect the concept of organizational loyalty. Affective commitment was measured with a German scale ranging from 1 (strongly disagree) to 5 (strongly agree; see Felfe, Schmook, Schyns, & Six, 2008; 4 items; Cronbach’s =.87; e.g., "I would be very happy to spend the rest of my career with this organization"). Turnover intentions were examined with a modified scale from Sjoberg and Sverke (2000; see Rigotti et al., 2010, 3 items; Cronbach’s =.88; e.g., "Despite the obligations I have made to this organization, I want to quit my job as soon as possible") and had to be answered on a scale from 1 (strongly disagree) to 5 (strongly agree). Perceived job performance. Perceived job performance was measured using two scales which differed with respect to time frame. First, following Abramis (1994), (shortterm) self-rated performance was assessed via self-report by asking employees the following: "In your own judgment, how well have you fulfilled the following tasks during the last week?" (Cronbach’s 6 items, =.76; e.g., "performed without mistakes?"). The scale ranged from 1 (very badly) to 5 (very well). Moreover, (long-term) personal accomplishment was measured. This scale was taken from the Maslach Burnout Inventory (MBI; Maslach & Jackson, 1981; German version: Büssing & Perrar, 1992; 8 items; Cronbach’s =.77; e.g., "I have accomplished many worthwhile things in this job") and answered on response categories varying between 1 (never) and 6 (very often). Mental impairment. In order to adequately reflect mental impairment in the survey, both a cognitive and an emotional dimension were assessed. Cognitive irritation, as an

Justice, social support and stressors 14 indicator of work-related strain, was measured with a scale developed by Mohr et al. (2005; 3 items; Cronbach’s =.91; e.g., "Even on holiday I think about my problems at work"), which could be rated on seven levels from 1 (does not apply at all) to 7 (applies nearly completely). Emotional exhaustion (9 items; Cronbach’s =.83; e.g., "I feel fatigued when I have to get up in the morning to face another day on the job") was assessed with the MBI (Büssing & Perrar, 1992; Maslach & Jackson, 1981) on a scale from 1 (never) to 6 (very often). Control variables. As there is continuing debate in organizational research about the adequate use of control variables (e.g., Becker, 2005; Spector & Brannick, 2011), authors are encouraged to provide explanations for their selection of controls rather than just considering the "usual" ones such as sex, age, and tenure. We therefore give theoretical or empirical reasons for the inclusion of our four selected controls in hypothesis testing, namely sex, age, job tenure and position, to prevent biased results. Sex was measured as a dichotomous variable and coded 0 for male and 1 for female. Age and job tenure were assessed in years. Position was evaluated on levels from 1 (unskilled blue collar worker) to 6 (management position or director). First, regarding organizational loyalty, various meta-analytic findings emphasize that sex, age and tenure (Mathieu & Zajac, 1990; Ng & Feldman, 2011; Meyer, Stanley, Herscovitch & Topolnytsky, 2002) are related to commitment. Correspondingly, Viscusi (1980) showed sex, age and tenure to be relevant predictors of turnover intentions. Moreover, in relation to job performance, a recent meta-analysis of Ng and Feldman (2010) reported that tenure would be favorably related to this outcome. In our study, besides job tenure (r=.14, p