Blame It on Individual or Organization Environment: What Predicts Workplace Deviance More?

: Deviant workplace behavior is one of the widely present employee behaviors that create signiﬁcant organizational cost, create an unhealthy working environment, and lead to various social and psychological job- and non-job-related consequences. Although various personality, situational, and organizational factors have been analyzed as instigators of such behavior, literature calls for a more comprehensive approach that analyzes interaction and mutual e ﬀ ects of di ﬀ erent sources of deviant behavior. This paper explores organizational culture and individual personality as the antecedents of deviant workplace behavior. A multilevel perspective is applied in empirical research that was done on a sample of 251 employees from 11 organizations in Croatia. Results of our research and hierarchical linear modeling imply that individual-related factors, namely, age and gender, as well as personality traits, are greater predictors of both individual and organizational deviance as opposed to organizational culture.


Introduction
In the last decades, workplace deviant behavior (WDB) has drawn increasing interest among organizational behavior scholars. Studies have analyzed both constructive (positive and functional) and destructive (negative and dysfunctional) deviant workplace behavior, trying to provide deeper understanding of its main antecedents, as well as mediators and moderators leading to such behavior.
Positive deviance departs and varies from the organizational norms in a desirable way so it helps the firm to accomplish tasks and achieve its goals. While positive workplace deviance is defined as a behavior that violates significant organizational norms contributing to the well-being of an organization, its members, or both (Bodankin and Tziner 2009), negative deviance violates interests of the firm and is destructive toward the organization and/or its employees. One of the most often cited definitions of destructive workplace deviance is one by Robinson and Bennett (1995, p. 556) that defines it as "a voluntary behavior that violates significant organizational norms, and may be harmful to the organization or the people in the organization, including employees, clients or customers".
Negative deviant workplace behavior has many undesirable consequences for organizations and/or their employees. Many scholars emphasize that deviant workplace behavior is a disruptive and costly problem (Greenberg and Baron 2003) with negative economical, sociological, and psychological outcomes (Bodankin and Tziner 2009). Such workplace deviance weakens organizational citizenship behaviors, decreases productivity, and boosts phenomena such as employee absenteeism and withdrawals (Brooks 2012), to name just a few.
Many researchers provide classifications of deviant behaviors (e.g., Hollinger and Clark 1982;Gruys and Sackett 2003;Spector et al. 2006;Bowling and Gruys 2010). However, one of the most widely used classifications is the one by Robinson and Bennett (1995). Although developed nearly

Personality Traits as Individual Antecedent of Deviant Behavior
Individual-related factors of workplace deviance are seen as characteristics, emotions, and cognitions of the employee (O'Boyle et al. 2011). Not surprisingly, most research on workplace deviance has typically focused on individual predispositions for deviant behavior, such as demographic or personality variables. One of the areas of research in individual-related factors is the individual's mental health. Scholars have found roots of certain dysfunctional workplace behavior in brain structure and psychopathy (e.g., Arboleda-Florez et al. 1998). Furthermore, differences in attitudes, perceptions, motives, and values of the individual are likely to affect workplace deviance as well ( Van Fleet and Griffin 2006). Many scholars explained unethical attitudes, deviation in moral philosophy, and low level of general trust as individual-related factors. Bolin and Heatherly (2001) found that each of four attitude variables (such as theft approval, company contempt, intent to quit, and dissatisfaction) had a relationship with at least one of four types of deviance. O'Boyle et al. (2011) cited other scholars that add attributional variables such as organizational constrains, justice seeking, turnover intentions, and job burnout as individual-related factors that facilitate workplace deviance. Possessing some of these individual preconditions does not mean that the employee will definitely be involved in certain types of deviant behavior but presents a risk for it. Furthermore, O'Boyle et al. (2011) state that these attributional variables interact "with one another and personality variables in the prediction of different forms of work deviance".
Personality variables have been among the most often researched determinants of workplace deviance on the individual level (e.g., Henle 2005;Berry et al. 2007;Kozako et al. 2013). There have also been suggestions to implement personality screening as a tool to minimize deviant behaviors (Henle 2005;Berry et al. 2007;. For example, workplace deviance is related to integrity, risk taking, and seductiveness (O'Neill and Hastings 2011), impulsivity or acting with little forethought (Henle 2005), low consciousness (irresponsibility, untrustworthiness, dishonesty, destructivity) and low agreeableness, low emotional stability, neuroticism, low extraversion (Berry et al. 2007;Henle 2005;Bolton et al. 2010). Furthermore, workplace deviant behavior is related to negative affectivity that is connected with provocativeness and aggressiveness (Berkowitz 1998), narcissism, manipulative personality (Bennett and Robinson 2000;Paulhus and Williams 2002), low self-awareness, and so forth. However, meta-analysis by Berry et al. (2007) indicated that among the Big Five, neuroticism, agreeableness, and conscientiousness had the most important impact on workplace deviance. Organizational workplace deviant behavior was lower in the case of higher agreeableness, and lower in the case of higher neuroticism and openness to experience (Kozako et al. 2013). Individual workplace deviant behavior was impacted negatively by extraversion and agreeableness, but positively by neuroticism and openness to experience (Kozako et al. 2013). Still, research did not reveal a dominant deviant personality type and in fact showed that personality alone accounts for only a small proportion of workplace deviant behavior (Robinson and Greenberg 1998).
Although researchers found several personality traits to be connected with negative workplace deviance, the effect of personality cannot be captured with the Big Five model solely (O'Neill and Hastings 2011). Apart from individual differences in personality, some other individual factors such as ethical ideology  have been found to impact workplace deviance. According to Henle et al. (2005), employees lower in idealism (according to Forsyth 1981) have higher probability of workplace deviance.

Organizational Culture and Deviant Behavior
Individuals and groups commonly conform to social norms that encourage certain kinds of behavior (Robinson and Greenberg 1998;Choi 2018). In organizations, such social norms form part of organizational culture and members of the organization can see disrespect to these norms as deviant behavior. However, not all of these will actually be regarded as workplace deviance as defined by Robinson and Bennett (1995), that is, only those behaviors that actually harm the organization or its members are examples of workplace deviant behaviors.
Organizational culture and climate have been identified as organizational characteristics that support constructive deviance (Vadera et al. 2013). For example, constructive deviance was higher in the case of less bureaucratic cultures (Stamper and Dyne 2001). However, organizational culture, a powerful force that shapes employee behavior (Schein 2004), can also elicit dysfunctional behaviors ( Van Fleet and Griffin 2006;Treviño et al. 2006;Kalemci et al. 2019). Such dysfunctional behaviors by the whole organization often reflect dysfunctional behaviors by the organizational leader ( Van Fleet and Griffin 2006). There are examples of organizational cultures that support aggressive behaviors as a practical method of motivating employees, so, incivility and rude behavior may emerge if disrespectful behaviors persist (Valentine et al. 2018) and these represent workplace deviance.
Several studies emphasized the role of ethical organizational climate in shaping employee behavior, more specifically, the occurrence of ethical behavior and deviant workplace behavior (Appelbaum et al. 2005). Organizational climate is an integral part of organizational culture and these terms are often used interchangeably. Due to the ever-present confusion about the definition and differences between climate and culture construct, these authors emphasize climate as an antecedent of workplace deviant behavior but their explanations of climate impact can be also attributed to organizational culture (e.g., see Ehrhart and Raver 2014). Orientation to self-interest or emphasis on efficiency (Peterson 2002) is likely to increase incidence of workplace deviant behavior, unlike employee focus (Peterson 2002) that tends to decrease deviant workplace behavior. However, Treviño et al. (1998) showed that although both climate and culture constructs could elicit unethical behaviors, culture had a larger impact on unethical behaviors. Appelbaum et al. (2005) emphasize that developing strong and positive cultural environment with ethical values might be the solution to prevent workplace deviance.
The impact of culture on employee behavior is grounded in organizational values, defined by Rokeach (1973) as conditions that the person finds desirable, thus guiding one's behavioral actions and cognitive perceptions of appropriate behavior. An organization's value structure is usually unique for all organizations and shapes the features of organizational culture. Emphasis on materialism or extrinsic motivation as an organizational value is likely to increase interpersonal deviant behavior (Deckop et al. 2015). Individualism/collectivism orientation, as well as power distance, can also be related to workplace deviant behavior. As shown by Kalemci et al. (2019), workplace deviant behavior is lower in the case of widespread collectivism, lower power distance, and high paternalism values. In terms of types of cultures that have been defined as antecedents of dysfunctional behaviors, empirical research is scarce (Ehrhart and Raver 2014). Di Stefano et al. (2019) showed that clan and adhocracy culture are related to lower levels of workplace deviant behavior. Aleksić et al. (2019), however, found that emphasis on market and hierarchy culture can be related to fewer deviant behaviors.

Research Instrument
Similar to the study of Di Stefano et al. (2019), this paper employs a multilevel perspective taking into consideration personality traits at the individual level and organizational culture at the organizational level and how they relate to deviant workplace behavior. In contrast to the study of Di Stefano et al. (2019), this paper takes into consideration their mutual effects, together with the effects of personal, demographic variables. A quantitative, cross-sectional study design was used, with an anonymous self-reported survey questionnaire designed to assess both individual and organizational levels of analysis, based on previously well-defined measures.
For measuring deviant workplace behavior, a 19-item scale developed by Bennett and Robinson (2000) was used. This measure allows assessing the level of both individual and interpersonal deviant behavior and has been previously extensively applied by numerous researchers in the field. Respondents were asked to indicate how often in the last six months they have engaged in a certain type of behavior by using the Likert seven-point scale (1-never, 7-daily). In order to ensure questionnaire validity, questionnaire items that were originally in the English language were translated into the Croatian language, with special attention to translation equivalence. Experts in the field of organizational behavior and English language did the translation and back-translation into English.
Organizational culture was assessed using a FOCUS questionnaire (van Muijen et al. 1999). This 16-item measure is designed to assess organizational culture based on a competing values framework that depends on the orientation of internal or external focus and stability or flexibility structure, allowing the identification of a dominant type of organizational culture (adhocracy, clan, market, or hierarchy culture) (Cameron and Quinn 1999). By using a Likert five-point scale (1-not at all, 5-a lot), respondents were asked to assess their level of agreement on the presence of certain behaviors and values in their company. As this is an internationally developed questionnaire, the Croatian version has been previously developed and validated (. Sušanj 2005).
Several demographic variables were controlled for their potential impact on workplace deviance. Gender was controlled due to the fact that males more often showed examples of workplace deviant behavior (Appelbaum et al. 2005;Chernyak-Hai et al. 2018), and age because previous research showed that older employees tend to be less deviant in workplaces than younger employees (Hollinger 1985;Pletzer et al. 2017). Moreover, control variables were work experience and education level, as previous research has found that longer-tenured employees showed lower levels of workplace deviance, just like more educated employees (Appelbaum et al. 2005).

Sample and Procedure
Due to specific issues and requirements needed to measure organizational culture (Glick 1985) as well as difficulties in assessing deviant behavior (Etikan et al. 2016), when defining our sample, we used a nonprobability sampling technique. For the purpose of this study, we used a snowball sampling technique. Through our personal and professional contacts, we recruited research participants. We contacted a representative from selected organizations in Croatia and asked them to distribute the questionnaire to randomly chosen employees from their company directories. Each employee received an e-mail invitation to access an online survey together with the information on the study and its objectives. They could access the survey only once; their confidentiality was secured and they were given the option to opt out of the survey at any time.
In total, 11 organizations and 251 employees from selected organizations participated in the study. As suggested by literature (Glick 1985), organizations from the sample were heterogeneous, coming from different businesses; 57.8% were large organizations (with more than 250 employees), 25.9% medium (50 to 250 employees), and 16.3% were small organizations (less than 50 employees), while most of the organizations (68.9%) were ones with public ownership.
Respondents were mostly female (64.5%), with average age 40.6 years (SD = 10.3) and mostly highly educated (94.4% respondents had a bachelor's or master's university degree). Regarding their working experience, respondents on average were working for 16.3 years (SD = 10.1).

Statistical Methods
Several statistical methods were used for data analysis, with the help of SPSS 18.0 statistical software package. The descriptive statistics (means, standard deviations) and reliability data analysis were first done. Cronbach's alpha reliability coefficients were assessed in order to check for the instrument validity. All the variables in the study had the reliability coefficients higher than the cut-off value of 0.70 (interpersonal deviant behavior 0.73, organizational deviance 0.82, adhocracy culture 0.84, clan culture 0.90, hierarchy culture 0.88, and market culture 0.88), thus indicating internal consistency of the measurement items as suggested by Kim and Feldt (2008). Then, the nonparametric correlation analyses were done to check for negative or low correlation coefficients, as they could be indicators of potential problems in data due to validity (Bolin and Heatherly 2001). Finally, several linear hierarchical regression models were done to test for the significance of study variables in predicting both organizational and interpersonal deviant workplace behavior. Hierarchical regression can easily integrate heterogeneous variables at the aggregate level into one model, and their significances can be estimated (Chi and Voss 2005). In addition, as it is used to analyze variance in the outcome variables when the predictor variables are at varying hierarchical levels (Woltman et al. 2012), we decided on this as the most appropriate method regarding our research.

Research Results
Means, standard deviations, and correlation coefficients for all study variables are shown in Table 1. Age showed a negative statistically significant correlation with organization-related workplace deviance (p < 0.05), indicating that older employees might be less prone to exhibit workplace deviant behavior. Deviant workplace behavior was also statistically significantly correlated with gender (p < 0.01). Both organizational and interpersonal workplace deviant behavior were negatively and statistically significantly related to the personality traits agreeableness and consciousness (at p < 0.01), but positively related to neuroticism. Additionally, openness showed a low but positive statistically significant correlation with organizational workplace deviant behavior (p < 0.01). In terms of features of organizational culture, research results in Table 1 show a negative and statistically significant correlation between organizational and interpersonal workplace deviant behavior and hierarchy and market culture features (p < 0.05). Such findings indicate that organizational and interpersonal workplace deviant behavior might be reduced in cases where organizational culture features emphasized market and hierarchy culture. In total, six models were established to test the connection among our study variables: three to test the impact of study variables on organizational deviant workplace behavior and three to test the impact of study variables on interpersonal deviant workplace behavior. The study variables were introduced hierarchically into the equations.
The first model included only demographic variables: age, gender, education level, and tenure as predictors. The second model included demographic variables together with personality traits, while our third model combined the aforementioned variables with the added variables of organizational culture. Table 2 shows output of hierarchical linear regression for the impact of several variables on organizational workplace deviant behavior. Model 1 included only controlled variables and the R 2 coefficient of this model is low although age was a statistically significant predictor of workplace deviant behavior. Model 2 included personality traits as predictors of organizational workplace deviant behavior. Personality traits agreeableness, consciousness, and openness statistically significantly predicted workplace deviant behavior, with higher proportion of the variation in the dependent variable explained by Model 2 (∆R 2 0.235 at significance p < 0.001). The final model additionally included features of organizational culture as predictors of workplace deviant behavior. Results indicate that features of organizational culture do not individually predict organizational workplace deviant behavior. Furthermore, Model 3 did not significantly increase the proportion of the variation in the dependent variable explained by the model (∆R 2 0.004). Predictors of interpersonal workplace deviant behavior are shown in Table 3. Model 1 was based on demographic variables as predictors of workplace deviant behavior and it did not show that independent employee features had a statistically significant impact on interpersonal workplace deviant behavior (R 2 0.013). In the following step, personality traits were introduced into the model. Two personality traits showed impact on interpersonal workplace deviant behavior: agreeableness showed a negative and statistically significant impact (r = −0.328, p < 0.01) and neuroticism was positively and statistically significantly related to interpersonal workplace deviant behavior (r = −0.130, p < 0.05). The proportion of the variation of interpersonal workplace deviant behavior explained by Model 2 was still relatively low 0.199 (∆R 2 0.186; p < 0.001), so in the third model, features of organizational culture were included as additional predictors of workplace deviant behavior. Model 3 did not confirm that features of organizational culture impacted interpersonal workplace deviant behavior and the change in proportion of the variation of interpersonal workplace deviant behavior explained by the model (∆R 2 ) increased by only 0.016 with no statistical significance.

Discussion
Due to the negative effect of deviant behaviors on employees and the whole organization, several attempts have been made to control such behaviors and to prevent their occurrence. Deviant workplace behavior is a complex phenomenon whose antecedents are not solely individual-related but can be drawn both from interpersonal and organizational factors (Henle 2005). This paper contributes to the extant research for antecedents of such deviant workplace behaviors.
Initial results confirmed that some demographic features other than personality traits could affect workplace deviant behavior. For example, this research confirmed previous findings that WDB is more often shown by men (Appelbaum et al. 2005;Chernyak-Hai et al. 2018) and that its level decreases with increasing employee age (Hollinger 1985;Pletzer et al. 2017). However, unlike previous research, this research did not confirm that WDB shows a correlation with employee educational level or tenure. The research objective of this paper has been to examine the interaction of individual and organizational sources of WDB. Since previous research identified personality traits as a strong predictor of deviant behavior (Berry et al. 2007), personality was examined as individual-related antecedent of deviance. Features of organizational culture were examined in the context of organizational sources of WDB due to their impact on shaping employee behavior (Schein 2004). Hierarchical linear regression was used to analyze the interaction of these variables in explaining two types of deviant behaviors: interpersonal and organizational deviance. Several independent individual variables have been controlled due to their previously found or proven relationship with workplace deviance.
Hierarchical modeling for antecedents of organizational workplace deviant behavior showed that independent employee characteristics accounted for only 7% of variation in organizational deviance but combined with personality traits explained 31% of such deviance. When culture was included as an additional antecedent, the explained proportion of the variance did not change significantly. Such findings indicate that individual factors act as dominant antecedents of organizational workplace deviance and that organizational context has almost insignificant impact on organizational deviance. Employees high on agreeableness and consciousness will exhibit lower levels of deviant behaviors, unlike employees high on openness that might show higher levels of deviance. Such findings on the impact of personality traits generally confirm previous research conducted in different research settings (e.g., Kozako et al. 2013). Previous research (e.g., Di Stefano et al. 2019;Aleksić et al. 2019) identified that certain types of culture can be related to fewer deviant behaviors, but this research showed that culture has a minor impact when personality is taken into account. Values and norms surely guide employee behavior, but the decision to undertake organization-aimed deviance is related to individual factors. Organization-aimed workplace deviant behavior can occur in any type of organizational culture and in fact, if it becomes part of the cultural norms, culture stimulates such behaviors. An emphasis on features of market and hierarchy culture could possibly limit the overall orientation to organizational workplace deviant behavior, but a key factor that will determine its occurrence, among those analyzed by this research, is personality traits, especially agreeableness, consciousness, and openness.
Hierarchical regression applied for interpersonal deviant behavior revealed that independent employee characteristics (age, gender, tenure, and education) accounted for an even smaller proportion of variation in deviant behavior than in the case of organizational deviance. Controlled variables in combination with personality traits explained approximately 20% of variations in interpersonal deviance, and similarly to organizational workplace deviant behavior, introducing organizational culture to the model did not significantly increase variance. Overall, it can be seen that personality and culture have a relatively minor impact on individual workplace deviant behavior and some additional factors should be explored as its antecedents.

Conclusions
Due to their negative effects, deviant behaviors are of great concern for organizations. In order to be able to prevent such behaviors, it is necessary to identify and predict their causes. This paper contributed to the analysis of both individual-and organization-related factors as predictors of deviant workplace behavior. More specifically, we tested if organizational culture as an organization-related factor and personality traits as individual-related factors can be seen as antecedents and instigators of deviant behavior. Previous empirical research is missing a more comprehensive research on various mutual effects of different variables as possible instigators of deviant behavior, so this study can be seen as contributing to this field of research. Research results imply individual factors are mostly to be blamed for the occurrence of organizational deviance, while organizational culture with values typically found in market and hierarchy culture has a small impact in diminishing this kind of behavior. This impact is even smaller when it comes to individual deviant workplace behavior.
Overall, the results of the study suggest that an individual's demographic and personality make a greater contribution to their deviant behavior than environmental factors of the workplace. This finding is of great importance to human resources managers and all those engaged in workforce development as it means that in order to solve the problem of deviance, it is important to look for solutions that lie within individual workers rather than concentrate on environmental factors such as organizational culture. All this leads to conclude that in order to prevent the occurrence of deviance in organizational settings, a greater emphasis of managers and human resource practitioners should be put on individual selection, socialization, education, and training aimed at enhancing personal value systems and individual beliefs on the forms of acceptable behavior.
Moreover, in spite of relatively small effects of culture on deviant behavior, this research has shown that organizational culture also has to be considered and taken into account in managerial attempts to diminish deviance in organizations. Managers have to be aware that creating a certain organizational culture and the values it promotes can stimulate or inhibit deviance. Organizational culture, as a glue that connects people, creates additional value and enhances acceptable forms of behavior.
Still, our research results and study variables explain only a small variation in deviant behavior (31% variation in organizational deviance and 20% in interpersonal deviance), thus indicating a wider spectrum of variables need to further be analyzed, alongside personality traits and organizational culture. Those, for instance, may be personal ethical or more job-related attitudes, such as job satisfaction or organizational commitment, but also some work-related factors, such as job autonomy, task variety, task independence, or context factors, such as, for instance, national culture and the values it promotes.
In line with study limitations, our study also has some limitations related to the statistical method and research instrument used and our sample. Although heterogeneous, our sample is rather small and gender-biased. Moreover, by using the snowball sampling technique to recruit our participants, there is the possibility of a sampling bias so future research should incorporate a wider sample. A wider sample is also needed to provide more complete data necessary to make more generalizations based on hierarchical linear regression models as a method used for data analysis. This method did allow testing the relative influence of several predictor variables on deviant workplace behavior, but it allowed for only certain variables to be tested. The measures used are self-reported ones, coming from a single source, and thus with possible effects of personal biases. We have tried to overcome this by using already widely accepted and validated measures. Still, in order to test deviance more accurately, future research should also strive to incorporate additional research instruments, such as interviews or real-time observations, but also to include different measures for both deviant workplace behavior and organizational culture. Future research, as mentioned, should also include additional variables into analysis.