Opportunity, job pressure and deviant workplace behaviour: does neutralisation mediate the relationship? A study of faculty members in public universities in Nigeria

PurposeThe purpose of the present study was twofold: (1) to examine the direct effect of the dimensions of opportunity (i.e. ethical climate and institutional policy) and dimensions of job pressure (i.e. workload and work pressure) on workplace deviance (i.e. organisational and interpersonal deviance) and (2) to assess the mediation of neutralisation in the relationship between the dimensions of opportunity, job pressure and workplace deviance.Design/methodology/approachThe present study drew from the fraud triangle theory (FTT; Cressey, 1950) and the theory of neutralisation (Sykes and Matza, 1957) to achieve the research objectives. Survey data from 356 full-time faculty members in Nigerian public universities were collected. Partial least square-structural equation modelling (PLS-SEM) was employed to analyse the data.FindingsThe results indicated that opportunity and job pressure significantly affected workplace deviance. As expected, neutralisation was found to mediate the negative relationship between ethical climate and interpersonal deviance and the positive relationship between workload, work pressure and interpersonal deviance. Contrary to expectation, neutralisation did not mediate the relationship between opportunity, pressure and organisational deviance.Research limitations/implicationsThe sample was drawn from academics in public universities and the cross-sectional nature of this study means that the findings have limited generalisations.Practical implicationsThis study offers insights into the management of Nigerian public universities on the need to curb workplace deviance amongst faculty members. This study recommends that the management improve the work environment by enhancing the ethical climate and institutional policies and reviewing the existing workload that may constitute pressure to the faculty members.Originality/valueThe present study provides empirical support for the fraud triangle theory and theory of neutralisation to explain workplace deviance.


Introduction
The literature on workplace deviance or counterproductive behaviour is growing, which suggests that such a phenomenon, if not managed well, is likely to affect the organisation and its members negatively (Bennett and Marasi, 2015). Workplace deviance can be understood as acts that are against the norms and expectations of the organisation that could harm its well-being and stakeholders (Bennett et al., 2018). It was reported that US businesses lost an average of US$113m in 2016 to employee theft. According to the 2017 Hiscox Embezzlement Study (Security Newswire, 2017), US businesses also suffered more than US$120bn a year from workplace violence, according to estimates by the National Institute for Occupational Safety and Health (Neckerman Insurance Services, 2012). Not only companies suffer monetary losses because of workplace deviance but also such an occurrence could also result in emotional costs to the employees. For instance, when employees experience workplace bullying or aggression at work, society could suffer when companies have to stop operating as a result of such incidence (Bennett et al., 2018). Others pointed out that reduced employee productivity and increased staff turnover are some of the adverse consequences organisations face as a result of workplace deviance (Tuna et al., 2016), affecting the organisational financial capacity.
Because workplace deviance could harm organisational effectiveness, many scholarly attempts have been made to theoretically understand the phenomenon and recommend practical measures to prevent such an occurrence. Towards this end, different theoretical perspectives have been employed. Some of these include the theory of reciprocity (Gouldner, 1960), justice theory (Adams, 1965), social exchange theory (Blau, 1964), the theory of strain (Agnew, 1985(Agnew, , 1992 and the theory of social learning (Bandura, 1963), amongst others. The present study is an attempt to add further to the literature on workplace deviance by offering a different theoretical lens. Specifically, it seeks to validate the applicability of the fraud triangle theory (FTT; Cressey, 1950) as a potential theoretical landscape. Although FTT is a theory that deals with fraud, some scholars argue that FTT has a theoretical value in the study of workplace deviance because fraud is a form of workplace deviance, albeit a serious one (Lawrence and Robinson, 2007;Dorminey et al., 2010;Abdullahi and Mansor, 2015). FTT postulates that unethical acts do not occur randomly; instead, it is a conscious act of individuals (Tuna et al., 2016), which is made possible by three factors: opportunity, pressure and rationalisation (Cressey, 1950). Consistent with FTT, the present study sought to investigate to what extent these factors could also explain different facets (organisational vs interpersonal deviance) and degrees of workplace deviance (minor vs major deviance). Specifically, it investigated the influence of ethical climate and institutional policy (facets of opportunity) and pressure (workload and work pressure) on workplace deviance.
However, the present study deviates from the premises of FTT in one significant way. According to FTT, individuals also justify the legitimacy of their fraudulent act by engaging in rationalisation (Cressey, 1950). That is, rationalisation is postulated to be a direct predictor of fraudulent behaviour. The present study argues that such theoretical postulation is somewhat problematic. The rational choice theory asserts that individuals are rational human beings who make rational choices based on the understanding of the consequences of his or her behaviour (Harsanyi, 1977). According to Vardi and Wiener (1996), employee deviance is motivational and goal-directed behaviour. Employees are likely to demonstrate harmful behaviour at work as a response to a stimulus (or stimuli) in the organisation (Skarlicki and Folger, 1997;Skarlicki et al., 1998). In this regard, the present study proposes that neutralisation could not directly predict workplace deviance; instead, employees use underlying mechanisms (i.e. neutralisation) to exhibit questionable behaviours (Lim, 2002). Hence, the present study theorised rationalisation or neutralisation as a psychological mechanism that explains why individuals engage in workplace deviance as a result of opportunity and pressure at work. The mediation of neutralisation was also investigated in past studies on workplace deviance (e.g. Lim, 2002). The present study, hence, adds to the literature because past studies that considered the underlying mechanism of workplace deviance are limited.
Another contribution of the present study is that it offers empirical evidence of workplace deviance in higher educational institutions (HEIs). Even though workplace deviance is said to occur in many organisational settings, limited attention was given to HEIs, especially public universities in Nigeria. Public universities in Nigeria present an intriguing case because deviance in these institutions is a national problem as attested to by President Muhammadu Buhari (Babachir, 2015). Further, employee deviance is said to be higher in public universities than in private universities (Geidam et al., 2011). For instance, Delta State University, Abraka, Nigeria (DELSU) sacked 14 faculty members and 17 non-academic staff members for unethical acts, such as plagiarism, sexual harassment, property theft, financial extortion from students, alteration of students' scores and absenteeism, amongst others, in 2017 (Dike, 2017). Consistent with the mandate of President Muhammadu Buhari to confront negative deviance in all facets of Nigeria, the present study is a scholarly attempt towards responding to a national call. Also, it is an exercise to test the validity of the FTT theory in such an organisational context since the theory tended to be examined in financial-related institutions and occupations (e.g. Brown et al., 2016).
Towards this end, this paper is organised as follows: a review of the relevant literature on workplace deviance, the factors purported to influence fraudulent behaviour according to FTT and the theoretical role of neutralisation in explaining deviance as expounded by the theory of neutralisation is offered towards the development of the research hypotheses. Then, a description of the methodology, analysis and results are presented, followed by a discussion of the findings and their implications. Finally, this paper ends with some concluding remarks.

Conceptual framework
The negative consequences of workplace deviance have attracted much scholarly attention and research. The literature is exponentially growing, partly reflected by the various terminologies used to define the concept, enhancing our theoretical understanding of the concept. Some of the terminologies used to describe the phenomenon include organisational misbehaviour (Ackroyd and Thompson, 1999;Vardi and Weitz, 2004), antisocial behaviour of employees (Robinson and O'Leary-Kelly, 1998), dysfunctional work behaviour (Griffin et al., 1998) and workplace incivility (Lim and Lee, 2011;Morrow et al., 2011;Taylor et al., 2012).
Despite the various terminologies used, the growing literature on workplace deviance has benefitted much from the seminal works of Robinson and Bennett (1995), who defined deviant workplace behaviour as "voluntary behaviour that violates significant organisational norms and in so doing threatens the well-being of an organisation, its members, or both" (p. 556). Extending the earlier works of Hollinger and Clark (1982), Robinson and Bennett developed a typology of workplace deviance by using a technique of multidimensional scaling. Based on the target of the deviant acts, they found that workplace deviance could be grouped into two: organisational deviance and interpersonal deviance. While sabotage, theft of organisational property, coming in to work late and sharing company secrets with outsiders are some of the examples of organisational deviance, gossiping about co-workers or physically or verbally abusing co-workers or customers are examples of interpersonal deviance (Bennett et al., 2018). While the former refers to acts targeted at the organisation, the former targets at members of the organisation. Their work has been widely used in the literature ever since the publication of their seminal work (e.g. Chen et al., 2018;Yasir and Rasli, 2018).
Because of the adverse effects of workplace deviance on organisations, researchers tend to focus on identifying contributing factors or antecedents using a variety of theoretical lenses. While it is beyond the scope of the paper to elaborate on the existing studies, it is safe to conclude that some factors seem to play a significant role in contributing to workplace deviance. By using social exchange theory, researchers found such poor working conditions, such as abusive supervision (Thau et al., 2009) and psychological contract breach (Bordia et al., 2008) were significantly related to workplace deviance. Justice theory has also been used to explain the link between perceived injustice at work and workplace deviance (Holtz and Harold, 2013;. Individual factors, such as personality, have also been considered by examining different models of personality structure. For instance, in comparing the validity of the Big Five and HEXACO model, Pletzer et al. (2019) found that the HEXACO domains explained more variance in workplace deviance that the Big Five domains (i.e. 31.97 vs 19.05%) in their meta-analytic study. Other studies also suggested that future researchers go beyond the Big Five model to understand the phenomenon (O'Neill and Hastings, 2011).
The present study contributes to the growing literature on workplace deviance by adopting a different theoretical lens, i.e. the FTT. The following discusses how this theory was hypothesised to explain workplace deviance.

Opportunity and workplace deviance
As mentioned earlier, in explaining workplace deviance behaviour, the present study drew from the FTT developed by Cressey (1950). Consistent with this theory, the present study postulates that opportunity at the workplace is likely to facilitate the engagement of deviant acts by employees. According to the FTT, an opportunity is created by ineffective governance system, absence of ethical climate, lack of internal control systems and poor policy implementation (Thanasak, 2013). That is, opportunity is created when there is a weak internal control system in the organisation or when such system does not exist (Cressey, 1950). This study specifically focused on two aspects of opportunity, i.e. ethical climate and institutional policy, as these have been found to affect employee behaviour significantly (Appelbaum et al., 2005;Taştan, 2019). Also, the present study was an attempt to respond to the call that more studies are needed to diagnose the relationship between ethical climate and deviance, which has remained mostly under-researched (Simha and Cullen, 2012).
Ethical climate reflects the organisational procedures, policies and practices with moral consequences (Martin and Cullen, 2006). Based on this definition, it is reasonable to theoretically speculate that when an organisation's work climate is ethical, employees are less likely to demonstrate undesirable or negative behaviour. Empirical evidence to suggest such a link is extant (Vardi, 2001;Peterson, 2002;Chen et al., 2013). Consistent with the FTT, employees are likely to take an opportunity to engage in deviant acts to achieve their goals when the organisation has a weak ethical climate. In such a climate, the organisation may have a bottom-line mentality in that the organisation is more concerned about financial success at the expense of other values (Appelbaum et al., 2005) and turns a blind eye to such act or maybe the organisation does not have relevant policies to deal with such behaviour. Weak internal control and unfavourable working conditions have been argued to facilitate employee deviance at work (Sauser, 2007). Hence, based on the theoretical argument and the empirical evidence, the following hypotheses were developed: H1. There is a negative relationship between ethical climate and interpersonal deviance.
H2. There is a negative relationship between ethical climate and organisational deviance.
Institutional policy represents the facet of opportunity in FTT. A policy serves as a guide to the running of an organisation. It states the boundary within which organisational activities must be performed and gives directions to deterrence measures (Trevino et al., 2005(Trevino et al., , 2006. Also, it may contain statements on reward and punishment to create the desired work climate. When the organisation does not have an explicit institutional policy regarding negative behaviour and reward and punishment, an opportunity could be created for employees to engage in negative or deviant behaviour at work (Sauser, 2007). According to the general deterrence theory (GDT; Gibbs, 1975), when the institutional policy that prescribes punishment for distasteful acts and such punishment is assured and severe, employees may be discouraged from such acts because of the pains that accompany such reprimand.
Despite limited empirical evidence, past studies suggest a significant influence of institutional policy on workplace deviance. For instance, the perceived severity of formal sanction was found to significantly relate to information systems security violation behaviours (Cheng et al., 2013). Past research also showed that the severity of the penalty and punishment discouraged employees from engaging in deviance targeted at the organisation (D' Arcy and Hovav, 2009;Kura et al., 2015). The theoretical arguments and evidence thus led to the formulation of the following hypothesis: H3. There is a negative relationship between institutional policy and interpersonal deviance.
H4. There is a negative relationship between institutional policy and organisational deviance.

Pressure and workplace deviance
According to FTT, pressure is the second factor that enables individuals to engage in fraud. Pressure refers to the motivation or needs to engage in fraud (Cressey, 1950). In the present study, pressure referred to job pressure, which is considered as having two dimensions, namely, academic workload and work pressure. Academic workload was operationalised as the professional efforts a faculty member devotes to activities such as teaching, research, publications, administration, community services and other academic-related tasks (Burke, 2011) while work pressure is conceptualised as the degree to which an academic must work fast and hard, has a great deal to do but with too little time (Karasek and Theorell, 1990). The literature appears to support the key premise of FTT in that the higher the workload and work pressure, the higher the possibility that undesirable behaviour will be exhibited. The general strain theory (GST; Agnew, 1992) seems to have a similar postulation in that strain causes undesirable reactions, which generate inspiration for deviance as a surviving tactic. Past studies have shown that job pressure has a significant effect on employee behaviour and work-related outcomes (Jones et al., 2010;Kayatasha and Kayatasha, 2012;Yadav, 2017), such as bullying in organisations (Yeh, 2015), and dysfunctional behaviours and lower audit quality (L opez and Peters, 2012). Hence, the following hypotheses were developed: H5. Work pressure is positively related to interpersonal deviance.
H6. Work pressure is positively related to organisational deviance.
H7. Workload is positively related to interpersonal deviance.
H8. Workload is positively related to organisational deviance.

The mediation of neutralisation
The theory of neutralisation postulates that when individuals engage in undesirable behaviour, they are likely to redefine their behaviour to make it acceptable because they are aware that immoral behaviour is against the societal norms and expectations (Sykes and Matza, 1957;Yu, 2013). By implication, this theory suggests that those who do behave morally do not create an excuse or do not have to justify their immoral action. As workplace deviance is purported to be triggered by events in the environment (Skarlicki and Folger, 1997;Skarlicki et al., 1998;Vardi and Wiener, 1996), employees do not engage in deviant behaviour randomly. They are aware of the organisational expectations and the implications of their behaviour. Consistent with the theory of neutralisation, when employees engage in deviant acts at work as a result of a stimulus in the work environment, they will rationalise and justify their behaviour to make it acceptable. The present study theorised that opportunity (poor ethical climate and ineffective institutional policy) and job pressure (high workload and work pressure) are used to justify employee engagement in deviant behaviour at work.
Past studies provided support for the theory of neutralisation to explain undesirable behaviour at work. In their study on personal use of the internet while at work, Cheng et al. (2013) concluded that employees justified the personal use through the lens of cost-benefit analysis in that the behaviour outweighed the cost of getting detected. In a qualitative study involving 44 restaurant workers, Shigihara (2013) observed that the workers justified restaurant theft by indicating that there was excess food and that no one cared if they took food home without permission, suggesting poor institutional policy implementation. Hence, the following hypotheses were formulated: H9. Neutralisation mediates the negative relationship between ethical climate and interpersonal deviance.
H10. Neutralisation mediates the negative relationship between ethical climate and organisational deviance.
H11. Neutralisation mediates the negative relationship between institutional policy and interpersonal deviance.
H12. Neutralisation mediates the negative relationship between institutional policy and organisational deviance.
Workplace stress may result from excessive workload, work pressure and role ambiguity, or social factors, such as poor leadership and feeling undervalued (Salami, 2010;Ogunsanya and Olorunfemi, 2012). Based on the theory of reciprocity (Gouldner, 1960) and social exchange theory (Blau, 1964), individuals who feel they have been short-changed in an employment relationship may invoke neutralisation to reinstate the impression of fairness. The reward deficit may result in employees being dissatisfied, justifying their deviant act at work (Parker, 2014). Perceived injustice and opportunities for neutralisation may help employees to dissipate internalised norms and social censure, allowing them to engage in undesirable acts without the feeling of guilt and shame (Warkentin et al., 2011;Ogungbamila, 2017). Hence, the following hypotheses were developed: H13. Neutralisation mediates the positive relationship between work pressure and interpersonal deviance.
H14. Neutralisation mediates the positive relationship between work pressure and organisational deviance.
H15. Neutralisation mediates the positive relationship between workload and interpersonal deviance.
H16. Neutralisation mediates the positive relationship between workload and organisational deviance. Figure 1 shows the proposed research model of the present study that illustrates the hypothesised link between the key constructs consistent with the FTT.

Methodology
Survey data were collected from 356 faculty members recruited from public universities across Nigeria. Because the incidence of employee deviance was reported to be higher in public universities than in private universities, the choice of surveying faculty members in public universities was justified (Geidam et al., 2011;Babachir, 2015;Nigerian Feminist Forum, 2015). As with any research that involves human subjects, ethical concerns need to be addressed. One of the ethical issues was the protection of the participants. The participants were made clear before they agreed to take part in the study that their participation was voluntary, and they could stop participating at any time during the survey. They were also assured that they would not be exposed to any harm financially, physically, mentally or socially, that their identity would remain anonymous, and that their responses were kept confidential. The sample profile was as follows: most of the participants were male (77%), married (83.7%), had been working in academia for more than six years (70%), and close to half of them had a master's degree (45.2%). Concerning age, most of them were relatively young aged between 31 and 50 years. The sample also consisted of faculty members of various ranks, from professors to assistant lecturers. Established instruments with good psychometric properties were used to measure the variables of interest. Bennett and Robinson's (2000) DWB scale was used to measure deviant workplace behaviour. The DWB scale was reported to have sound psychometric properties with internal reliability of 0.81 and 0.78 for organisational and interpersonal deviance, respectively (Bennett and Robinson, 2000). To measure DWB, 28 items were scored on a 5-point scale (1 5 Never; 2 5 Rarely; 3 5 Sometimes; 4 5 Often; 5 5 Always). The items in the original scale were later validated by six subject matter experts to suit the academic setting. The items went through several validation exercises to ensure that the modified items were valid and relevant. Sample items include "I do not complete the required syllabus in a semester" and "I raise tempers at colleagues/students".
Ethical climate was assessed by seven items (α 5 0.79; Schwepker and Hartline, 2005) from the ethical climate scale of Schwepker and Hartline (2005). Participants specified their perceptions of ethical climate on a 5-point scale (1 5 Mostly false; 2 5 Somewhat false; 3 5 Somewhat true; 4 5 Mostly true; 5 5 Completely true). Sample items include "The climate in this institution allows lecturers to do some unethical things at work" and "Top management does not support ethical behaviour in this institution." Institutional policy was assessed by five items (α 5 0.73-0.82; Comer et al., 1989) adapted from Comer et al.'s (1989) measure of company policy. Items were rated on a 5-point Likert scale (1 5 Strongly disagree; 2 5 Disagree; 3 5 Neither disagree nor agree; 4 5 Agree; 5 5 Strongly agree). Some of the items asked were, "This institution operates efficiently and smoothly because of effective policies." Workload was assessed by eight items (α 5 0.74 to 0.78; Houston et al., 2006), which were taken from the job demands scale of Houston et al. (2006). A 5-point Likert scale was used. Participants were asked questions, such as "I often need to work after working hours to meet my work requirements."

Study of faculty members in Nigeria
Five items from Karasek and Theorell's (1990) job pressure scale were used to measure work pressure (α 5 0.73 -0.85; Karasek and Theorell, 1990). A 5-point Likert scale was also used. Neutralisation centres on the cognitive and/or socially interactive stage before individuals exhibit a norm-contradicting behaviour (Sykes and Matza, 1957). Neutralisation was measured by six items (α 5 0.861; Rogers and Buffalo, 1974) adapted from Rogers and Buffalo's (1974) neutralisation scale. All participants indicated their level of agreement on a 5-point Likert scale (1 5 Strongly disagree to 5 5 Strongly agree).
Once the data were collected, they were screened for missing values, outliers, nonresponse bias, common method variance, normality and multicollinearity (Tabachnick and Fidell, 2007;Hair et al., 2010). No significant issues were found, allowing the data to be analysed using partial least square-structural equation modelling (PLS-SEM) on SmartPLS-SEM 3.2.7. PLS-SEM is useful to test a mediating effect where Preacher and Hayes's (2008) bootstrapping technique of estimating indirect effects could be employed (Hair et al., 2013).

Analysis and results
The PLS-SEM analysis involves two stages of assessment: the measurement model and the structural model (Henseler et al., 2009;Hair et al., 2013). While the first assessment is to ensure the goodness of the measures used, the second assessment is carried out to test the research hypotheses (Hair et al., 2013). In the first stage, the internal consistency of the model was assessed using the composite reliability index (Hair et al., 2012). Table 1 shows that the composite reliability of each construct ranged from 0.883 to 0.975, exceeding the minimum acceptable level of 0.70 (Hair et al., 2014(Hair et al., , 2017. Hair et al.'s (2017) recommendations were followed in that loadings of 0.60 and above were retained since the average variance extracted (AVE) values for all constructs were higher than 0.50 (refer Table 1). All the parameters for measurement of the model showed that the model had adequate internal reliability.
Convergent validity, discriminant validity and item reliability were also ascertained (Hair et al., 2017). Convergent validity means the degree to which two or more measures of the same theoretical construct assessed by different methods agree. Convergent validity was assessed by examining the AVE for each latent construct. According to Hair et al. (2017), the AVE for each latent construct should be 0.50 or more. Table 1 shows that the AVE for each latent construct was greater than 0.50, indicating adequate convergent validity.
Discriminant validity refers to the degree to which one theoretical construct differs from another (Hair et al., 2017). Discriminant validity was ascertained by the Fornell-Larcker criterion and heterotrait-monotrait ratio-HTMT (Fornell and Larcker, 1981;Henseler et al., 2015). Table 2 compares the square root of the AVE for each latent construct with the correlations amongst the latent constructs. The table also demonstrates that adequate discriminant validity had been established in the present study because the square roots of AVEs were greater than the correlations between constructs (Fornell and Larcker, 1981;Hair et al., 2017). Table 3 shows that discriminant validity was achieved because the highest correlation found between workload and work pressure was 0.828, which was within the conventional yardsticks of 0.85 (Clark and Watson, 1995;Henseler et al., 2015). Figure 2 shows the retained items and the beta values of the constructs. It presents evidence that the individual item reliability was found to be acceptable.
After the goodness of measures of the model was ascertained, the next analysis was to assess the structural model, i.e. evaluating the predictive abilities and the interrelationships (paths) between the latent constructs (Hair et al., 2014). The structural model was evaluated based on the following criteria: the significance of the structural path coefficients, coefficient of determination (R 2 ), the effect size (f 2 ) and predictive relevance of PLS estimates at the construct level (Q 2 ) (Chin, 1998(Chin, , 2010

Study of faculty members in Nigeria
In assessing the structural model, direct and indirect effects were analysed. The recommendation of Hair et al. (2014) and Henseler et al. (2009) was followed. A bootstrapping procedure with 5000 bootstrapped samples and 356 cases were used to evaluate the significance of the path coefficients to generate beta values, standard errors, tvalues and p-values of the estimate to determine the precision of the model. The direct effect model amongst the latent variables without including a mediator was computed to assess hypotheses 1-8. As shown in Table 4 and Figure 3, the result of the structural path coefficients revealed that only HI, H4, H5 and H7 were statistically significant while H2, H3, H6 and H8 were not. Specifically, a significant negative relationship between perceived ethical climate and interpersonal deviance was found (β 5 À0.255; t 5 3.559; p < 0.01), supporting H1. A negative relationship between institutional policy and organisational deviance was also supported (H4) (β 5 À0.447; t 5 9.240; p < 0.01). Similarly, workload was found to be significantly and positively related to interpersonal deviance (β 5 0.133; t 5 1.568; p < 0.1), supporting H5. Likewise, a positive relationship between work pressure and interpersonal deviance was observed (β 5 0.341; t 5 3.854; p < 0.01), offering support for H7.
Next, the coefficient of determination and predictive relevance of the model were assessed simultaneously by running the blindfolding procedure (Hair et al., 2017). Table 4 shows that the direct effect model explained 40% of the total variance in interpersonal deviance and 22% of the total variance in organisational deviance. Table 4 also indicates that the Q 2 value for interpersonal deviance was 0.30 and 0.11 for organisational deviance. Both values exceeded zero, which suggests satisfactory predictive relevance of the model (Henseler et al., 2009). The standard root mean square residual (SRMR) value of 0.07 also shows that the model had a good fit.
The indirect effect of neutralisation was then analysed. As shown in Table 5, the coefficient of determination (R 2 ) was 36.8% (0.368), which suggests that the indirect effect model explained 36.8% of the total variance in interpersonal deviance and 49.2% of neutralisation. After running the blindfolding procedure, the results showed that the Q 2 value for interpersonal deviance was 0.280, organisational deviance was 0.011, and neutralisation 0.394. Statistically speaking, all values were greater than zero, signifying acceptable predictive relevance of the indirect model (Preacher and Hayes, 2008).
Furthermore, standard bootstrapping procedure with 5000 bootstrap samples and 356 cases were applied to assess the significance of the path coefficients (Henseler et al., 2015). Figure 4 lends credence to indirect effect results. Meanwhile, Table 5 indicates that the indirect effect of ethical climate on interpersonal deviance via neutralisation (mediator) was found to be significant (β 5 À0.090; t 5 3.420; p < 0.01) and predicted interpersonal deviance in a negative direction, thus supporting H9. In keeping with the classical approach, the direct effect of EC→ED was significant (t 5 3.559) without the mediator variable. When the mediator variable was introduced, the indirect effect remained significant (t 5 3.420) but lower than when the mediator variable was not incorporated (t 5 2.553). As such, Similarly, the relationship between perceived institutional policy and interpersonal deviance via neutralisation remained statistically significant (β 5 À0.038; t 5 2.157; p < 0.05), suggesting that faculty members did not use institutional policy as a justification for interpersonal deviance. Hence, there was statistical support for H11. Neutralisation was also found to mediate the relationship between workload and interpersonal deviance in a positive direction (β 5 0.091; t 5 3.335; p < 0.01), giving support to H13. Similarly, neutralisation was

Discussion
The primary objective of the present study was to validate the application of FTT in explaining deviant workplace behaviour (i.e. organisational and interpersonal deviance). Specifically, our result indicated that the dimensions of opportunity (i.e. ethical climate and institutional policy) and job pressure (workload and work pressure) were significantly linked to workplace deviance. When the ethical climate and institutional policy are perceived to be weak, employees are likely to engage in workplace deviance. Such a finding corroborates the literature (Appelbaum et al., 2005;Lister, 2007;Sauser, 2007;Hooper and Pornelli, 2010;Taştan, 2019). Similarly, high workload and work pressure were also found to increase the likelihood of employee engagement in workplace deviance, supporting past studies (Salami, 2010;Appelbaum et al., 2012). The result also appeared to support the proposition that neutralisation is a psychological mechanism that explains why employees engage in deviant behaviour at work in contrast to the postulation of the FTT that neutralisation or rationalisation is a factor that predicts directly undesirable behaviour. That is, faculty staff members justified their engagement in workplace deviance as a result of the poor work conditions. The significant role of neutralisation in mediating the effects of opportunity and job pressure on workplace deviance is in line with the literature (Dabney, 1995;Lim, 2002;Fritsche, 2005). However, when examining the direct and indirect effects on a specific type of workplace deviance, the result was somewhat unexpected. The faculty members in the present study appeared to engage in interpersonal deviance and not organisational deviance when  opportunity and job pressure were perceived to exist. The result is contrary to the model of and Robinson and Bennett (1995) that employees target their deviant acts at the organisation as a result of the poor treatment they receive. A probable explanation could be contextual. We speculate that despite the weak ethical climate and institutional policy, faculty members did not take the opportunity to engage in organisational deviance because doing so could jeopardise their job. Nigeria is facing a high unemployment rate of 23.1%, an increase from 18.1% in 2017 (Carsey, 2018). In this context, job security becomes a concern for many people, including those working in public universities. The need to retain the job might explain why faculty members did not target the organisation when the work environment was perceived to be poor and unconducive. The literature indicates that employees are likely to be abusive towards their colleagues as a way to vent their anger at the management or the management (De Cuyper et al., 2009;Harold et al., 2016). A survey of 992 employees from employee assistance programme and occupational health provider Health Assured reported that nearly 9 out of 10 (86%) workers regularly vent their anger and frustration at co-workers (Frith, 2018). In the context of the present study, faculty members vented the frustration or anger from the organisation to the colleagues and/or students because the latter was in a more vulnerable position than the former (i.e. colleagues) due to the power gap. Despite the possibility of explaining the result, the job insecurity explanation should be validated further. The findings of this study provide a theoretical insight into FTT, an alternative framework to understand deviant workplace behaviour. Especially through the mechanism of neutralisation, how opportunity (in this case, poor ethical climate and institutional policy) and pressure (high workload and work pressure) could be used by employees to justify and rationalise their engagement in deviant behaviour at work is better understood. In the literature of workplace deviance, the mechanism to explain why engages in deviant behaviour at work is less explored; only a handful of studies considered the role of the psychological or cognitive process as mentioned earlier. The present study provides empirical evidence that such a process is important to enhance our theoretical understanding of the negative phenomenon at work. While the mediation role of neutralisation is observed in this study, more research is needed to further confirm its significance or other psychological or affective states as a result of the perceived negative work environment. Also, future studies may want to explore the neutralisation techniques used by employees in the context of poor work conditions or environment. Moreover, contrary to expectation, the significant effect of opportunity and job pressure on interpersonal deviance and not on organisational deviance deserves more attention.
The findings benefit policymakers and management of public higher education institutions in Nigeria, particularly in managing workplace deviance. The result points out the need to strengthen the implementation of institutional policies and enhance the ethical climate of the institutions in curbing workplace deviance. Weak enforcement of institutional policy and ethical climate provides an avenue for employees to engage in workplace deviance because of the perception that such behaviour may not bring any adverse consequences to the perpetrator. Much is known about the high level of workload and work pressure faculty members in public universities in Nigeria are facing (Ofoegbu and Nwadiani, 2006;Ogunsanya and Olorunfemi, 2012;NEEDS Report, 2012;Ikonne, 2015). Even though the present study did not explore the antecedents of workload and pressure because such investigation is beyond the scope of the study, it could be speculated that inadequate government funding could explain the poor work conditions of the sampled institutions (Okiy, 2005;Ikonne, 2015). In short, as the finding indicates that poor work environment can be used to justify workplace deviance, the management of public universities should take concerted effort to make the work environment more conducive to mitigate the opportunity to justify the deviant behaviour.
While the results of the present study are insightful, they need to be interpreted by considering the following limitations. Firstly, the cross-sectional nature of the present study makes it impossible for causal inferences. Hence, this study needs to be replicated using longitudinal research or experimental design. Secondly, the sample was drawn from faculty members in public universities, which may limit the generalisation of the findings. In Nigeria, private universities tend to have better financial standing, which translates to better facilities and infrastructure for the faculty members (Akpotu and Akpochafo, 2009;Ajadi, 2010). Hence, if our speculation is correct, in such institutions, issues related to workload and pressure may be less likely (Akpotu and Akpochafo, 2009;Ajadi, 2010). However, to what extent this is likely to affect workplace deviance needs to be examined. By considering the faculty members in private universities, a better generalisation of the phenomenon under study could be achieved. Also, because the present study was conducted in Nigeria, we recommend future studies to consider cross-country investigations to compare the present findings with the results from other countries.
Various opportunities exist for future research in addition to those mentioned above. One of them is the possibility of investigating a boundary condition that will either mitigate or reinforce the neutralisation process. Personality traits, such as self-control or personal values, may be able to help us understand when one is likely to strengthen his/her justification to engage in deviant behaviour at work in a particular work environment. Secondly, since the dimensions of opportunity and pressure are widely defined in FTT, future researchers may consider predictors or variables that fall within each domain. In these two instances, the theory is likely to be further refined and improved. Thirdly, for a meaningful comparison, future researchers may wish to carry out the same survey outside Nigeria and across organisations. By doing so, it is possible to gauge the level of the pervasiveness of workplace deviance and demonstrate the need to address it because of the harm it could bring.

Conclusion
To ignore deviant workplace behaviour is to allow the erosion of organisational standards, regulations and norms, which are likely to lead to organisational deterioration. Although deviant workplace behaviour is an international phenomenon, the present study focussed on public universities in Nigeria because such a phenomenon has been personally acknowledged as a national problem by President Muhammadu Buhari (Babachir, 2015). As the present study has demonstrated, deviant behaviour may not be exhibited when the organisation is perceived to have a good work environment, characterised by ethical climate, proper enforcement of institutional policy, and reasonable workload and work pressure. However, more research needs to be carried out to support the finding further and by considering other institutional and work-related factors so that a better understanding of the phenomenon and the implementation of preventive measures can be undertaken.