Employees’ behaviours at work could either be positive or negative (Dahling & Gutworth 2017). A positive workplace behaviour would yield a productive result and enhance the work environment in the organisation. Negative workplace behaviours are also known as WDBs (Robinson & Bennett 1995). Workplace deviant behaviour is an intentional and voluntary act that violates organisational norms, and is capable of causing harm to the organisation, its staff or both (Robinson & Bennett 1995; Tuzun & Kalemci 2018). These acts include, but are not limited to, absenteeism, sabotage, cyberloafing, favouritism, withholding or withdrawing effort, sexual harassment, intentionally working slow, taking long breaks, theft, gossiping, blaming others, verbal abuse, physical abuse, receiving a bribe, and being corrupt (Bennett & Robinson 2000). Robinson and Bennett (1995) proposed a typology of WDB, by dividing workplace deviant acts into two categories based on the target, namely organisational deviance and interpersonal deviance. Organisational deviance refers to deviant acts exhibited by employees, which are directed towards the organisation, while interpersonal deviance refers to deviant acts targeted at co-workers, supervisors and subordinates in the workplace. Due to the differences in the target of each of these forms of deviant acts, this study examines both these two forms of WDB.

The principles of OJ have been recognised as a bedrock for the effective running of organisation as well as ensuring that employee’s personal needs are met (Greenberg 1990). Cropanzano et al. (2001) suggested various reasons why employees join an organisation, one of which may be the desire to satisfy their economic needs while for some it could be to satisfy their socio-emotional needs. Decisions relating to these needs and employees’ experience during these processes, arouse the question of fairness (Colquitt 2001). Some of these decisions may concern salaries, social environment of the organisation and the job or role performed by individuals (Colquitt 2001). Greenberg (1990) defined organisational justice as the extent an employee perceives the employee–organisational relationship to be fair, ethical and equitable (Greenberg 1990; O’Connor & Crowley-Henry 2019). Organisational justice emanates from attempts to examine the role of justice in the effective operation of the organisation (Greenberg 1987, 1990). There are three major forms of OJ, namely, distributive justice, procedural justice, and interactional justice, while interactional justice is classified into two components, namely, interpersonal justice and informational justice. This classification gave rise to further classification of the forms of justice into four categories, namely, distributive, procedural, interpersonal and informational justice by some researchers such as Colquitt et al. (2001). Procedural justice describes how organisational procedures are perceived to be fairly designed and how fairly these processes are used in the determination of outcome allocation (De Lara & Verano-Tacoronte 2007). Distributive justice describes the perception of fairness of distribution of rewards (Wu & Wang 2008). Interactional justice describes how employees perceive they are being fairly treated by the decision makers with respect and dignity. Informational justice defines how employees perceive they are fairly informed of procedures that affect them. It refers to explanations of why certain procedures are used and why certain rewards are distributed the way they were. How employees perceive these explanations to be honest and adequate leads to different forms of behavioural outcomes (Colquitt et al. 2001).

Organisational justice has been linked with several employee behaviours and attitudes such as organisational citizenship behaviour (Jnaneswar & Ranjit 2022), turnover intentions (for instance, Alyahya, Elshaer & Sobaih 2021; Zhang & Agarwal 2009) and WDB. It represents a situational based antecedent of WDB (Henle 2005) and has been recognised as an important predictor of WDB (Abbasi et al. 2020; Bennett & Robinson 2000; Dora & Azim 2019). Oluwole, Aderibigbe & Mjoli 2020, among several researchers proposed that one of the ways employees react to inequity or failure of the employers to fulfil their obligation to the employees is by involving in WDB. By this they believe they will be able to get even with the organisation. Additionally, employees appraise the fairness of their outcome, procedures, interactions within the organisation and what they perceive to be better done in other organisations which is yet to be embraced in their organisation (Osibanjo et al. 2015). If they perceive inequality or inequity, the feeling of frustration, anger, job dissatisfaction and a lack of trust often develop, which may result into employees engaging in acts that are harmful to the organisation and their co-workers (Dora & Azim 2019; Oluwole et al. 2020). The principle of retributive justice helps to understand the link between injustice and WDB. This principle explains that an individual who perceives injustice seeks to get back and inflict harm on the offender, as a means of getting even (Khattak et al. 2019).

Distributive justice refers to perceived fairness of the distribution of the outcome received (Adams 1965). It explains the level to which employees perceive the distribution of outcomes such as rewards and promotion to be fair (Colquitt 2001; Sabokro, Tavakoli & Malmiri 2020). This type of justice was founded on the assumption that outcome or reward stirs the reaction and behaviour that individuals exhibit in the social world. Cropanzano and Ambrose (2001) suggested that equity theory is not only useful in explaining distributive justice but is also useful in explaining all the other forms of justice. Based on argument in literature, it has become necessary to link the forms of justice to specific forms of deviance (Masterson et al. 2000). That is, because distributive justice has to do with outcomes, which is decided by the organisation, harming the organisation in an attempt to seek equity would bring satisfaction to such victims of injustice (Hasting & Finegan 2011; Tziner, Fein & Vasiliu 2020). Berry, Ones and Sackett (2007) found an insignificant positive correlation between distributive justice and organisational deviance. On the contrary, other researchers have found that distributive justice impacts not only deviance targeted at the organisation but also deviance targeted at individuals in the organisation. For instance, Henle (2005) found that distributive justice impacted the two forms of WDB, that is, organisational and interpersonal deviance. This was because in seeking equity, the employee seeks to harm the source of the inequity, either an organisation or individual. Additionally, Greenberg and Barling (1999) found that distributive justice also relates with supervisor’s directed aggression. Based on this argument in literature, this study examines the impact of distributive form of justice on both organisational and interpersonal deviance in Nigerian tertiary institutions, and postulate the following hypotheses:

Procedural justice refers to the fairness of the procedures used in allocating employee’s outcome (Cropanzano, Fortin & Kirk 2015). It has been revealed that the fairness of the process of allocation is, often, of more significance than the distribution of the outcome itself (Cropanzano et al. 2015). When rules guiding employees’ judgement of fair procedures are not followed, they tend to engage in acts that are harmful to the organisation. Skarlicki and Folger (1997) found that procedural injustice has a significant relation with retaliatory behaviours, while Cohen-Charash and Spector (2001) found that the best predictor of WDB is procedural justice. Retaliatory behaviour in this sense does not necessarily mean being against the organisation; the employee may also retaliate on the individuals in the organisation. On the other hand, Berry et al. (2007) found a negative but insignificant correlation between procedural justice and organisational deviance. Abbasi et al. (2020) also found an insignificant relationship between procedural justice and WDB. Despite this argument for and against procedural justice as the best predictor of WDB, researchers, such as (Henle 2005; Tuzun & Kalemci 2018) suggested that all the forms of justice could be linked with WDB. Hence, the study formulates the following hypotheses:

Interactional justice refers to the treatment employees receive while procedures leading to their outcome are being enacted (Bies & Moag 1986). Specifically, an individual perceives fair interpersonal treatment by the degree to which they are being treated with dignity, respect, sensitivity, truthfulness, politeness, and how well explanations are provided for decisions relating to them (Alyahya et al. 2021; Colquitt 2001; Greenberg 1990). Interactional justice also entails the manner in which information is shared and communicated and it is divided into two forms of justice, which are interpersonal and informational justice (Colquitt 2001). Interpersonal justice relates to the respect and dignity received from superiors, while informational justice relates to how honestly and promptly are employees informed on matters affecting them (O’Connor & Crowley-Henry 2019).

Interpersonal justice is mostly important in shaping employee’s behaviour, due to the day-to-day interaction among employees (Judge, Scott & Ilies 2006). Specifically, employees regard interpersonal justice to be more important when compared to other forms of justice (Leineweber et al. 2020). Social exchange theory (SET) and the norm of reciprocity explain that individuals reciprocate the treatment they receive, either good or bad. Therefore, when employees perceive they are being treated unfairly, they tend to reciprocate with WDB (Tziner et al. 2020). Thus, this study examined the two dimensions of interactional justice, namely, interpersonal and informational. As a result, the following hypotheses are formulated for the study:

Homans (1961) defines social exchange as a process in which a person’s actions impact on another either in the form of reward or punishment. The quality of the social exchange existing between employees and management can also be directly affected by the perception of fairness, that is, OJ (Henle 2005). When employees perceive unfair treatment from the supervisor or the organisation, they develop a feeling of violation of the social exchange and hence, compare the cost of remaining in the organisation with their rewards. If the cost outweighs the rewards, they seek means of revenge and hence, resort to acts that are harmful to the organisation or its individuals such as withdrawing of efforts, absenteeism and other deviant behaviours (Blakely, Andrew & Moorman 2005; Leineweber et al. 2020). Using SET to explain the relationship between OJ and turnover, Moon (2017) explains that employees believe that the values the organisation places on them can be judged by the fairness or unfairness of the treatment received (which includes their outcome, the outcome procedures and the interpersonal treatment received during the procedures) which to them is a benefit. When unfairness or injustice is perceived, according to the norm of reciprocity, they also seek to reciprocate by involving in deviant acts (Colquitt, Greenberg & Zapata-Phelan 2005). In using SET to explain OJ and WDB, Greenberg and Scott (1996) and Tuzun and Kalemci (2018) opine that employee’s perception of unfairness or injustice could lead them into involving in WDB. One of the ways this could be done is to change their input to restore equity. Social exchange theory is also used in explaining the fairness perception of employee’s relationship with the supervisor, as this could determine how well employees are committed to the supervisor, the organisation and other members of the organisation (Leineweber et al. 2020).

This theory, originally known as justice theory, suggests that individuals compare their input and outcome ratio to that of relevant referent; this comparison forms their perception of how fair the outcome they receive is (Miller, Konopaske & Byrne 2012). The theory better established the underlying processes and effects of OJ (Alexander et al. 2012). Specifically, equity theory explained that individuals rationally evaluate and compare the ratio difference between its input and outcome with that of another person. The theory maintained that inequity generates a psychological feeling of distress and tension that make individuals seek means of restoring balance (Adams 1965). Specifically, individuals are satisfied when there is balance in their comparison with others; but, when imbalance occurs, referred to as underpayment inequity, there is a feeling of anger and they tend to lower their contribution (Byrne & Cropanzano 2001) and resort into deviant acts. Colquitt et al. (2013) opined that to lessen the tendencies of employees engaging in deviant behaviour there is need for organisations to ensure that employees perceive their outcome and processes to be fair.

This quantitative design employed a positivism research paradigm in achieving its objectives. This approach focuses on scientific method of inquiry, intended to produce pure data and facts that are not influenced by the researcher’s interpretation or biasness; hence, the researcher maintains an objective stance (Saunder, Lewis & Thornbill 2019). Survey was conducted using open-ended questionnaire; 704 employees (academic and non-academic) were targeted, of which 572 provided usable data, thus yielding 81% response rate. The study employed simple random probability sampling to select samples of academic and non-academic employees. This gives equal chance of being selected to all the members of staff in the selected universities. The choice of the three institutions is also based on purposive non-probability sampling selection criteria (inclusion and exclusion criteria) and ease of data collection. In this context, employees of these universities have had at least one case of deviance reported on newspaper and social media (Folarin 2019; Sahara Reporter 2016, 2017). In addition, each of them must be in existence for 10 years or more.

To have a representative proportion of the sample, due to the largeness of the sample, the ‘Taro Model’ by Yamane (1967) was employed to derive sample size out of the captive population of the estimated 1955 academic staff and 5463 non-academic staff, while ratio scale analysis was used to delineate the sample:

n = the sample size

N = total population for the study

e = the acceptable sampling error at (0.05)

Relying on this model, the study sampled 704 respondents. The sample size from this model is found to be representative and guaranteed the required degree of precision and confidence. The sample size obtained from the model showed that 332 respondents of the sample size are academic while 372 are non-academic.

With the aid of ratio scale analysis, the sample 332 and 372 academic and non-academic staff members respectively from each of the three universities is presented in Table 1.

Quantitative data were collected using structured close ended questionnaire. The questionnaires were distributed with the aid of three research assistants. The research assistants were well oriented on the purpose of the data and processes of data collection. The data were collected within a 6-month period. The purpose of the study was explained, confidentiality of information supplied, and anonymity of respondents were guaranteed.

Workplace deviant behaviour measures of Bennett and Robinson (2000) were adopted. This scale consists of two forms of deviance, namely, organisational deviance which consists of 12 items with statements such as ‘falsified a receipt to get reimbursed for more money than you spent on office expenses’ and interpersonal deviance, consisting of 4 items with statements such as ‘said something hurtful to someone at work’. The respondents were asked to respond to how often they were involved in each of these items at work on a 4-point-Likert typing scale ranging from 4 (never) to 1 (every time). The reason for adopting this kind of scale is to ensure that the respondents can give their opinion of the subject matter without sitting on the fence. This is because the questions asked are about the daily experience of the respondents at work; therefore, the respondents cannot be neutral specifically as it relates to issues that concern their behaviour and perception. Also, this kind of scale aid ease interpretation (Hartley 2014). These measures were found to have reliability scores of 0.81 and 0.78 revealing that the measures adopted are reliable.

Organisational justice was measured using Colquitt (2001). This scale was used to measure procedural justice, distributive justice, interactional justice, and informational justice which consists of seven items, four items, four items and five items, respectively, with questions such as ‘have you been able to express your views and feelings during those procedures?’, ‘Does your (reward) reflect the effort you have put into your work?’, ‘Has (he or she) treated you in a polite manner?’, and ‘Has (he or she) explained the procedures thoroughly?’. The respondents were asked to respond to how often they experience each of these items at work on a 4-point Likert type scale ranging from 4 (never) to 1 (every time). The scales have reliability coefficients of 0.78, 0.79, 0.70, 0.92 respectively, revealing that the measures adopted are reliable.

The descriptive and inferential statistics were used to analyse the questionnaire collected for this study. Section A includes information on respondent’s demographics. These were analysed using descriptive statistics such as mean, standard deviation, frequency counts and simple percentages, which were presented in graphical and tabular forms using bar chart, pie chart and frequency tables. Data in Section B that includes information on WDB and OJ were analysed using inferential statistics such as Pearson correlational analyses and linear multiple regression analyses.

The demographic data collected reveals the following characteristics. Participants’ gender reveals 288 (50.6%) male, 279 (49.0%) female, and 2 (0.4%) preferred not to answer. Ninety-eight (17.1%) respondents have worked for 5 years and below, 138 (24.1%) have worked between 6and 10 years, 175 (30.6%) have worked between 11 and 15 years, 74 (12.9%) have worked between 16 and 20, and 84 (15.2%) have worked for 20 years and above. The educational qualification held by the respondents ranges from school certificates, national diploma, HND or BSc., MSc. and PhD, with percentages of 35 (6.1%), 74 (12.9%), 238 (41.6%), 159 (27.8%) and 66 (11.5%) respectively. Lastly, information on the age of the respondents reveals that 33 (5.8%), 62 (10.8%), 92 (16.1%), 134 (23.4%), and 251 (43.9%) of the respondents fall between the age of 21–25, 26–30, 31–35, 36–40, 41 and above respectively.

The objective of this study was to determine the link between OJ and WDB. The objective is achieved in two steps. Firstly, the study determines the extent of relationship between the dimensions of OJ (i.e. informational justice, distributive justice, procedural justice, interpersonal justice) and forms of WDB (organisational and interpersonal, and combined) by conducting Pearson Product-Moment correlation. Secondly, the study determines the influence of dimensions of OJ on forms of WDB using multiple linear regression.

The correlation results in Table 1 show that the dimensions of WDB have very strong, statistically significant correlation among them (690), suggesting high degree of relationship between interpersonal and organisational WDB. The relationships among the four dimensions of perceived OJ are positive and significant at 1% and 5% level of significance. The correlations between procedural and distributive (0.518), procedural and interpersonal (0.298), procedural and informational (0.317), are not high enough to pose multicollinearity problem in subsequent linear regression, albeit a high correlation between interpersonal and informational justice were identified.

Procedural and distributive justice have significant positive correlation with organisational deviance leading to the acceptance of null hypothesis, while interpersonal and informational justice have negative correlation with OJ, resulting in the rejection of null hypothesis. However, the latter is not significant. Likewise, there exists positive correlations between procedural and distributive justice vis-a-vis interpersonal deviance. However, the correlations are not significant. Furthermore, interpersonal and informational justice have significant negative correlations with interpersonal deviance.

In general, procedural and distributive justice have significant negative correlation with organisational deviance and insignificant positive correlation with interpersonal deviance. Conversely, informational and interpersonal justice have negative relationships with interpersonal deviance.

While the correlation measures the degree of relationship among the variables, it does not determine the effect of one on the other. Hence, the linear multiple regression is used to determine whether the negative (positive) relationship amounts to negative (positive) impact of OJ on the two forms of WDB with the results presented in Table 2. As depicted in the upper section of Table 3, the R square as shown in the regression model indicates 0.048 with adjusted R square of 0.042. This implies that the independent variables (informational justice, distributive justice, procedural justice, interpersonal justice) explain 4.8% variance in the organisational deviant behaviour. The standardised Beta value for each of the independent variables indicated: procedural justice (β = 0.150, p < 0.05); distributive justice (β = 0.063, p > 0.05); interpersonal justice (β = −0.198, p < 0.05); and informational justice (β = 0.022, p > 0.05). Based on this outcome, the acceptance or rejection of the null hypothesis (H1) which states that, OJ (informational justice, distributive justice, procedural justice, interpersonal justice) is positively related to organisational deviance, depends on the forms of justice. Specifically, the null hypothesis of positive relationship between procedural justice and interpersonal deviance is accepted, while the null hypothesis of positive relationship between other forms of justice and interpersonal justice is rejected. As opposed to the a priori, this finding suggests that procedural justice increases organisational deviance. In line with the a priori, interpersonal justice reduces organisational deviance. However, informational justice and distributive justice do not exert significant effect on organisational deviance.

With respect to the interpersonal justice, the regression result revealed that the independent variables (informational justice, distributive justice, procedural justice, interpersonal justice) explain 3.2% of its variation. Furthermore, the standardised Beta value for each of the independent variables indicated: procedural justice (β = 0.133, p < 0.05); distributive justice (β = −0.005, p > 0.05); interpersonal justice (β = −0.077, p > 0.05); and informational justice (β = 0.022, p > 0.05). Based on this outcome, the acceptance or rejection of the null hypothesis (H1) which states that, OJ (informational justice, distributive justice, procedural justice, interpersonal justice) is positively related to interpersonal deviance, depends on the forms of justice. Specifically, the null hypothesis of positive relationship between procedural justice and interpersonal deviance is accepted while the null hypothesis of positive relationship between other forms of justice and interpersonal justice is rejected. As opposed to the a priori, this finding suggests that procedural justice increases interpersonal deviance. In line with the a priori, interpersonal, informational justice and distributive justice reduce interpersonal deviance; however, the reducing effects are not statistically significant.

Overall, this study established through Pearson correlation and multiple linear regression that there exists significant positive relationship and influence of procedural justice on both forms of WDB, worsening WDB. This is not consistent with the a priori expectation. However, interpersonal justice has negative relationship with and influence on interpersonal deviance, ameliorating interpersonal deviance. This is consistent with the a priori expectation.

This study examines the effects of OJ on WDB by examining individual effect of each dimension of OJ on each dimension of WDB on the one hand, and by testing combined effects of all dimensions of OJ on aggregated WDB on the other hand.

Overall this study established, through Pearson correlation and multiple linear regression, that there is a significant positive relationship between procedural justice and both forms of WDB; indicating that procedural justice worsens WDB. Increasing effect of procedural justice on WDB is not consistent with the a priori expectation. It is also in contradiction to Faheem and Mahmud (2015), Dora and Azim (2019) and Abbasi et al. (2020) who found that procedural justice has insignificant negative effect on WDB among employees of Pakistan public hospital, and multimedia companies and small and medium enterprises (SMEs) in Malaysia, respectively. The finding also contradicts Baig and Ullah (2017) who concluded that procedural justice significantly reduces aggregate WDB. Under the procedural justice wave, the fairer the processes leading to the distribution of outcome, the lower the tendency to embark on deviant behaviour. The study finding contravenes this theory, suggesting that employees of the universities under consideration are negatively disposed to fairness of procedures and processes used in deciding their outcomes.

This study finds that distributive justice has insignificant positive impact on organisational deviance, and insignificant negative impact on interpersonal deviance. This finding suggests that employees of the studied public university are literally indifferent as to the fairness of distribution of outcome (Deutsch 1985). This study finding differs from Baig and Ullah (2017), Dora and Azim (2019) who found that distributive justice posed significant negative effect on WDB as well as Butt and Atif (2014) and Sabokro et al. (2020) who concluded that distributive injustice increases WDB significantly, although these studies examine aggregate deviant behaviour. On the other hand, Khan, Quratulain and Bell (2014) suggest that fairness may not be enough, that is, fairness does not often lead to employees exhibiting good behaviours. The authors argued that envy could make an employee involve in WDB even when the processes are fair. This could be as a result of employees comparing themselves with other employees who are enjoying higher fairness of distribution justice.

Extant literature (Baig & Ullah 2017; Dora & Azim 2019; Faheem & Mahmud 2015; Sabokro et al. 2020) has concentrated on interactional justice effect as opposed to its two subsets, namely the interpersonal and informational justice which are examined in this study. Although its negative effect on interpersonal deviance is not significant, the study finds that interpersonal justice has significant negative influence on organisational deviance, reducing deviant behaviour. This is consistent with the a priori expectation. With interpersonal justice being a subset of interactional justice, the finding of this study is in consonance with Sabokro et al. (2020) who submitted that interactional injustice spurred deviant behaviour, even though insignificantly. This finding also supports Azim and Dora (2016) and Abbasi et al. (2020) who concluded that interactional justice significantly reduces deviant behaviour. Unlike most extant literature, the current study shows the individual effects of interpersonal and informational components of interactional justice. Interpersonal justice is achieved when the interpersonal relationship of authorities with subordinates is guided by rules such as respect (need for treatment with sincerity, dignity and refraining from attacking or being rude to others) and propriety (avoidance of questions that could bring bias, such as sex, religion, age, race, among others) (Colquitt et al. 2005). The study finding suggests that higher level of interpersonal justice in the form of respectful treatment and propriety in the public university is accompanied by lowering WDB, especially the organisational deviance. In other words, respect and dignity received from superiors go a long way in reducing WDB.

This finding corresponds to the widely held view (Alfes et al. 2013; Leineweber et al. 2020; Moon 2007) that employees regard interpersonal justice to be more important when compared to other forms of justice as well as SET, and the norm of reciprocity which explains that individuals reciprocate the treatment they receive, either good or bad. As opposed to most studies who linked interpersonal justice to interpersonal deviance, this study shows that interpersonal justice predicts organisational deviance more than it predicts interpersonal deviance.

Furthermore, this study revealed an insignificant negative impact of informational justice on both organisational and interpersonal deviance. Informational justice ensued when the interpersonal relationship of authorities with subordinates is guided by rules such as truthfulness (honesty and candidness in communication, while implementing procedures) and justification (adequate explanation for the outcomes received by employees) (Colquitt et al. 2005). The study finding is consistent with Sabokro et al. (2020) who found that interactional injustice has insignificant positive effect on deviant behaviour, and with Faheem and Mahmud (2015) who found that interactional justice has significant negative effect on deviant behaviour. Although there is scant literature on informational justice and WDB, El Akremi, Vandenberghe and Camerman (2010) found that informational injustice triggers supervisor-directed deviance. It can be concluded that how honestly and promptly employees are informed on matters affecting them has no significant reducing effect on deviant behaviour in public universities. It can also be deduced that interpersonal justice is relatively more important than its counterpart (informational justice), as far as prediction of WDB is concerned.

The present study is not without its limitations. Firstly, cross-sectional design was adopted for the study, thereby making it difficult for causal inference to be made on the result of the findings. Future study could consider adopting longitudinal designs so as to examine the effects over time. Secondly, self-report measure was used which could lead to a common method bias; however, the study sample consists of employees from different universities and groups. This has been established to have reducing effect on common method bias (Alfes et al. 2013). Future studies could consider collecting data from multiple sources. Furthermore, for OJ to have a desirable effect on WDB, intervening variables may be required to mediate the relationship. For example, it has been suggested that job satisfaction, personality traits, inter alia, mediate the relationship between justice and WDB in recent literature. This is beyond the scope of this study; however, it presents a researchable gap for future researcher to fill in the Nigerian context. Regardless of these limitations, this study has been able to provide an understanding of the OJ and WDB relationship.

If the stakeholders have to choose among the four dimensions of justice, interpersonal justice must be accorded higher priority in dealing with organisational deviance in public universities in Nigeria. It must be ensured that superiors treat their subordinates with respect and dignity, as this would impact on the feeling of injustice which has been affirmed to lead to deviant behaviour. In addition, the study recommends that future studies examine one of the two groups of staff, that is, future studies should focus on either academic staff or non-academic staff due to the different deviant acts exhibited by them.