EXPLORING INTERVENING INFLUENCE OF INTERACTIONAL JUSTICE BETWEEN PROCEDURAL JUSTICE AND JOB PERFORMANCE : EVIDENCE FROM SOUTH ASIAN COUNTRIES

The paper investigates the intervening influence of interactional justice between procedural justice and job performance (task, contextual and adaptive performance) of the faculty members of Karachi (Pakistan) and Dhaka (Bangladesh) based government colleges by using Structural Equation Modelling (SEM). Data, for this study, has been collected through pre-designed close-ended questionnaire. The intervening variable fully mediated the relationship between procedural justice and job performance. The result of this study indicates that the performance of government college faculty members can be improved by ensuring fair procedures and dignified treatment of faculty members in the working environment. It can be concluded that teachers can accommodate harsh procedures, subject to courteously and fairly communicated. Significance of this study is that it has investigated the least researched areas in Pakistan and Bangladesh. Its findings can be helpful to the government and college administration while making and implementing policies for college education development in both countries.


Introduction
Despite, human being may share similar features from the physiological perspective; individuals vary in perceiving the same phenomenon.On the basis of their perceptions, they act and react distinctively.Varying perceptions may be an outcome of environmental effects, cultural compulsions, and numerous behavioural reactions influenced by the human psychology and context.To measure the unique outcomes of perception in organizational management context, researchers have conducted significant amount of researches.Greenberg (1987) coined the term organizational justice and observed distinct behaviours' of people in working environment while judging the actions and procedures of organization (Infante, Smirnova, 2016).Leventhal (1980) found the influence of perception as an outcome of procedural fairness among people engaged in several activities.Bies and Moag (1986) have discussed the ways of promoting staff's favourable perceptions due to the organizational decisions.They believe, it may not be important to the people what they get as an outcome of their efforts, but the accumulated significance lies in their dignified treatment.As natural instinct, employees often nourish perception, retain it and carry with them at and outside their work place which influences their performances.Ripley et al. (2006), Komendat and Didona (2016) found that an employee's perception of the work environment influences behaviour and predict their performance.The perception among employees is an outcome of procedures being executed, hence the loyalty and motivation to work is the reflection of fairness (Xu, Loi and Ngo, 2016).Procedural justice as a dimension of organizational justice has been studied both negatively and positively (Vermunt and Steensma, 2016).There is significant amount of research on procedural justice and performance in academic institutions but there is no considerable research has been done in Pakistan and Bangladesh.
Quality contribution of teachers may result in production of quality graduates.There are several factors, which may influence the outcome of quality graduate, and procedural fairness is one such factor (Graso et al., 2014).Kalay (2016) studied the impact of procedural justice on performance of teacher's of Karachi and Dhaka but did not find any significant relationship.
Employees and particularly teachers expect fair treatment with reference to their contributions to the work and their performance.The work environment is said to be fair which provide opportunity of effective voice, sense of security, dignity and esteem (Fair Work Convention, 2016).
Hence, the problem statement of this research is that there is a general perception that faculty working in the government colleges are not being rewarded in accordance with the value they are adding.Such perception may lead to counter work behaviours.Such as taking other paid assignments during college hours, and poor job performance.
Research gap.This study is an empirical attempt to diagnose the influence of procedural justice in predicting the job performance of public sector college faculty of Karachi and Dhaka, largest industrial cities of Pakistan and Bangladesh.Findings of this research, certainly, will fill the existing knowledge gap in the procedural fairness and job performance of intervening influence of interactional justice in these parts of the subcontinent.

Theoretical review and framework
The working environment in any organization often constituted on several factors, changes in macro and micro conditions forces the organizations and institutions to rejuvenate their policies and procedures regularly (Nadim et al., 2012).Whatever the profession may be, perception of people play significant role in grasping the deviation in behaviour (Tuytens and Devos, 2012).The teaching profession is probably more prone to diversified perception.This may be because all professions seek their nourishment from teaching profession (Nadim et al., 2012).The findings of Menga et al. (2016) have confirmed the relationship between perception and teachers' performance.Colquitt et al. (2005) is of the opinion that decision about distribution, procedures are perceived to be fairly communicated, it may lead towards organizational productivity.Both in Pakistan and Bangladesh the significant relationship between HR practices and perceived fairness has also been found (Shahzad et al., 2008).A value addition to the research literature regarding justice perception and healthcare employees' service behaviour has been added in Bangladeshi context (Nuruzzaman and Talukder, 2015).However the perception of government college faculty with reference to procedural justice and their reactions have not yet drawn considerable attention of researchers.
The fairness of organizational practices and policies, particularly in execution of HR practices, develop the portal of information for people's perceptions of procedural justice (Kuvaas, 2008), this state of affairs help them to shape their cognizance of several managerial, collective, or personal rewards (Bowen and Ostroff, 2004).Procedural justice has a significant influence on behavioural exhibitions of employees, such as: dedication to work, emotional intelligence, compliance of policies, teamwork and task performance (Colquitt et al., 2005).Greenberg (1990) recommended that procedural justice is compound of structural and interpersonal dimensions.Number of studies conducted by Moorman and his peers documented significant support for the association between Organizational Citizenship Behavior (OCB) and dimension of procedural justice.By applying Structural Equation Modelling (SEM) method (using LISREL), Moorman (1991) established significant association between interactional justice and OCB.
When it comes to the employees performing in knowledge-based institutions, there is strong significance of behaviours shaped as a result of various procedures and policies.Despite such significance of procedural justice in predicting employees' performance, still there is paucity of research on investigating the teachers' perception of the procedural justice they are getting in the institutions and their job performance.
Well established and generally agreed upon consensus can be observed in several studies on organizational justice, that it can be viewed as two distinct but interlinked segments namely distributive justice and procedural justice (Danaeifar et al., 2016).Distributive justice focuses the fairness in outcome and procedural justices focuses the fairness in process of distribution.People can tolerate certain harsh decision about distribution and procedure if they perceive the process was fair.The third type of organizational justices is termed as interactional justice.It refers to fair treatment (Bies and Moag, 1986).The interactional justice has been studied and intervening variable in this study.
In Pakistani context the indirect influence of faculty trust between organizational justice and organizational citizen behaviour documented significant impact (Tahseen and Akhtar, 2016).Campbell (1990) explains that job performance refers to an action taken by an individual.Campbell considers the job performance as an individual-level construct.This notion makes it different from managerial performance or public performance.Job performance, in the context of individual or collective concept has been considered as yardstick for testing the extent of managerial achievement and frequently studied as endogenous variable in several models (Grant, 2008).Usually, such model formulation carries determinants, moderators, mediators, predictors, identifiers or dimensions of job performance (Lang et al., 2007).
To grasp the intervening influence on the job performance of government colleges' faculty, the plethora of literary evidences showed that, there is significant influence of interactional justice on various employee behaviours.Such as Adam's (1963) equity theory model, selfinterest model developed by Leventhal (1976), Group value model of Lind and Tyler (1988).People in working environment perceive the interaction between input and output relationship as even handed, they react positively, and become productive for the organization.Contrary to that the inequitable perception will force them to exhibit negative performance (Adams, 1963;Allen and White, 2002).In addition to that, self-interest model having six peculiar justice rules developed by Leventhal (1976) are recommended to be considered while making decision in order to ensure the fairness in interactional justice: decision being supported with authenticated intelligence, execute persistent allocation procedures, restrain biasness, accurate decisions, follow moral & ethical principles, and that the distribution process is intact with recipients' anticipation and needs.The proper execution of Leventhal's six identified justice rules can be a result oriented mechanism to promote the sense of fair treatment among employees (Ismail and Shariff, 2008), people tends to react or pro-act by avoiding or exhibiting indifference to unfairness and attempt to restore justice through different behaviours (Greenberg, 1987).Masterson (2001) have studied the trickle down effect of all three components of organizational justice among instructors and students and found, that instructors having positive perception about the distribution, process and treatment put more energy towards imparting the education.People may compromise injustice in distribution and process but strive to restore the dignified treatment (interactional justice), thus the distribution if supported with respect and procedures appended with dignity can cause positive change in working behaviour (Al Afari and Abu Elanain, 2014).In order to empirically test if the notion is significant antecedent of job performance among public sector colleges faculty we have proposed the following conceptual framework/model and hypotheses on the basis of the identified variables in Figure no. 1.
Variables.The above model shows the variables being investigated in this study.The procedural justice is predicting the task, contextual and adaptive performance, therefore procedural justice is the independent variable, and task, contextual and adaptive performance are the dependent variables.Whereas the interactional justice is being tested as mediating variable.On the basis of conceptual and given variables following hypotheses have developed for the study.

Figure no. 1: Conceptual framework
Hypotheses formulation.The mediating effect of interactional justice between procedural justice and dimensions of job performance has been tested as main hypothesis for this study.Management scholars have inked the mixed empirical findings about the direct and indirect influence of several work related behaviours among teachers and education managers.Such as Trivellas and Santouridis (2016) have found full intervening role of job satisfaction between 'Higher Education Service Quality' and 'Organizational Commitment', studied in Greece.A meta-analysis on justice in organization conducted by Charash and Spector (2001) and found the counter-work behaviours and job performance as resultant of perceived fairness, which were predominantly associated with procedural justice.Justice as mediator has been studied by Niehoff and Moorman (1993) and narrated that although monitoring method negatively influence on organizational citizenship behaviour but showed positive and significant through perception of fairness.The interactional justice was studied as intervening variable between influence of pay and job satisfaction and it was found that it fully mediates the effect between said variables (Ismail et al., 2011).In Pakistan, Tahseen and Akhtar (2016) have studied the mediating effect of faculty trust between organizational justice and organizational citizenship behaviour.Whereas in Bangladeshi context the relationship between organizational justice and organization directed reaction were studied and found strong relationship between among three types of justice and three types of reactions there respondents were US Faculty, Managers and Bangladeshi faculty and managers (Rahim et al., 2001).The ample literary evidences have provided the academic grounds to formulate the following hypothesis for this study: H1: The Significant association between procedural justice and job performance is mediated by interactional justice.
The perception about procedures applied in decision-making, resource allocation have drawn significant amount of attention of researchers.For example the study suggested through empirical investigation, that procedural justice was more significant criterion in decision about allocation of resources (Howard et al., 1986).From the psychological Interactional Justice Procedural Justice Adaptive Performance Task Performance Contextual Performance perspective procedural justice has been predicted as strong regulator of basic autonomy needs (Prooijen and Willem, 2009).Procedural justice has been studied as determinant of task performance; however it was also found that it was partially intervened by intrinsic motivation (Phelan et al., 2009).These significant evidences have nurtured the necessity of following hypothesis for this study: H1(a): Procedural justice is positively associated with task performance.Leung et al. (2001) have concluded that perceived interactional justice is significant predictor of job attitudes and perceived fairness in salary fairness.In South Asian perspective, particularly in China, the study concluded by Wu et al. (2012) showed full mediation of interactional justice on trust in supervisor, however it did not mediate the effect between authoritarian leadership and trust in supervisor.In Pakistani context the influence of procedural justice on employee engagement studied by Osman et al. (2016).They studied the negative influence of organizational justice and turnover intention among professional workers.However the study was conducted making banking sector as subject of interest.This study has therefore developed a following hypothesis to test the mediating effect of interactional justice between procedural justice and task performance: The relationship between procedural justice and task performance is mediated by interactional justice.
Procedural justice being the perception about fairness in procedures and its ability to predict the task performance among employees was concluded as significant (Phelan et al., 2009).The influence of procedural justice on mutual commitment was studied in India and Ghana in a single study and found significant predictability of procedural justice (Acquaaha and Padhyeb, 2015).In Pakistani context the relationship between procedural justice and psychological ownership over employees was found significant and positive (Butt and Atif, 2015).Rahim et al. (2001) documented significant and negative influence of organizational justice on turnover intention between Bangladeshi and U.S. faculty.This study has therefore formulated following hypothesis to test:

H2(a): Procedural justice is positively associated with contextual performance.
People may afford discontentment with pay, but will not compromise on dignified treatment, therefore may react in several counterproductive work behaviours (Gieter et al., 2012).It can be assumed that compromised procedural fairness can usher the fair perception if interpersonal and informational justice prevailed, hence following hypothesis is developed for testing:

H2(b): The relationship between procedural justice and Contextual performance is mediated by interactional justice.
Concerns about the fairness in organizations may be universal, however execution of justice procedures is highly particularistic (Greenberg, 2001).Thus the perception about the procedures may produce unique outcomes in various employee behaviour.In order to test the phenomenon we have developed the following hypothesis and studied in Pakistani and Bangladeshi context:

H3(a): Procedural justice is positively associated with adaptive performance.
The educational institutes around the globe in general and in Pakistan and Bangladesh in particular facing stiff competition pertaining to retaining the talent.In order to response the confounding competition, the managers of educational institutes doing their utmost to single out the causes of inflexible performance among the faculty.

H3(b):
The relationship between procedural justice and adaptive performance is mediated by interactional justice.

Research methodology and design
This study has causal research design to investigate the causal influence of procedural justice (as independent variable) on Job performance (as dependent variable) and interactional justice (mediating variable).Besides these identified variables numerous other variables cause influence the job performance of the government college faculty, therefore we have used the Nomothetic Causal Explanation (Bachman and Schutt, 2006) method of causation for this study.Karachi and Dhaka based government college faculty members is the universe of population for this study.The education and literacy department Govt of Sindh revealed that there are 4000 faculties engaged in 118 public sector colleges of Karachi.The population frame in the shape of seniority list duly furnished by the department has been uploaded on the official website of the department (Govt. of Sindh, 2015).The official sources of the Bangladesh govt.revealed that there 4153 engaged in several intermediate and degree level colleges (Ministry of Education, 2015).
The population of our interest is segmented into four unique classes: namely Lectures, Assistant professors, Associate Professors and Professors.For each identified category separate population frame was available.The faculty having experience of 5 years and above has been targeted and were requested to become the respondents for this study.While screening the seniority list, there were 200 such respondents whose names were provided, but the place of posting was not identified.This constraint has forced us to rely on 3400 each fully known respondent.Hence our accessible population available to us for this study was 6800.In order to compute the representative sample size for the study we have used Yamane (1967) formula.Calculation of the sample size is given below: Where: n = required sample size, N = population, e = precision.
In our case, we have assumed 95% confidence level and precision = 0.05.Thus we have equation for this study was: As noted above the population for this study is segmented into four unique classes, therefore we have allocated the sample as per the proportion of each category (Table no.1).The procedural justice being exogenous variable of this study is measured through 6 items developed by Niehoff and Mooraman (1993) and 9 items, proposed for measuring interactional justice as mediating variable.The scale developed by Coole (2003) for measuring the task performance based 5 items was used to accomplish the purpose of this study.Contextual performance was measured by using scale developed by Borman (1993).The five item based scale for measuring adaptive performance was also used (Plamondon, 2000).

Data analysis and results
SEM is considered as suitable approach to investigate mediation paths (Baron and Kenny, 1986).To evaluate the fitness of both measurement and structural model, we have applied Root Mean Squire of Approximation (RMSEA) along with Comparative Fit Index (CFI).
The reason behind choosing RMSEA because it parsimoniously adjusts the index (Hooper et al., 2008).The value for RMSEA lies between 0.08 and 0.10 is considered as mediocre, and less than 0.08 is deemed good fit (MacCallum et al., 1996).Comparative Fit Index (CFI) as incremental fit index was also used to judge the fitness of the model of this study.Hu and Bentler (1999) proposed the rules of thumb for CFI cut-off as CFI value >0.95.
The correlation analysis along with means and standard deviations for the measures of interest are shown in Table no.2. The table shows that there was significant positive correlation among all measures except adaptive performance.This also shows Chronbach's alpha reliability diagonally into parentheses.Structural Equation Model Analysis.Confirmatory factor analysis (CFA) has been executed to assess the measurement abilities of the variables.All insignificant items were eliminated by dropping those elements, which had loading ≥0.60 (Cooper and Schindler, 2003).The preliminary conditions for mediation were authenticated by conducting significant direct influence of procedural justice on Task, Contextual and Adaptive performance.With introduction of intervening variable, interactional justice the impact of exogenous variable on endogenous were reduced.The path coefficient of procedural justice and task performance with the introduction of mediating variable interactional justice reduced to insignificant, thus the evidence of full mediation in case of this study was witnessed (Baron and Kenny, 1986).The direct effect of procedural justice on contextual performance was significant, however with intervening effect of interactional justice turns that significance into insignificant, hence there were full mediation witnessed in this study.The direct effect of procedural justice on adaptive performance was significant, with intervening effect of interactional justice the effect turned into insignificant, therefore interactional justice fully mediated between procedural justice and adaptive performance.Testing of hypothesis.For testing the hypotheses developed for this study, we have conducted structural equation modelling.To accomplish the mediating objective of this study we have constructed mediation model for H1(b), H2(b) and H3(b).The latent constructs namely procedural justice was independent variable; task, contextual and adaptive performance were dependent variables, whereas interactional justice was mediating variable for this study.For conducting mediation analysis, it was obligatory to test the direct effect between independent and dependent variables (Baron and Kenny, 1986).Thus we have formulated H1(a), H2(a), and H3(a) as shown in Table no.4.  (Baron and Kenny, 1986).

H1(a) is retained
 the Indirect effect 0.35 ×0.47=0.164 The product indirect effect is higher than direct effect; therefore the evidence of full mediation was there (Baron and Kenny, 1986;Sobel, 1982). Procedural justice to interactional justice was significant,  Interactional justice to task performance was also significant,  The effect of procedural justice through mediating variable was also significant as per Sobel's (1982)  Procedural Justice and Contextual Performance was Significant. Thus the constraint of mediation was achieved (Baron and Kenny, 1986).

H2(a) is retained
 The Indirect effect 0.35 ×0.46=0.161 The product indirect effect is 0.161, which is more than the value of direct effect .09.  Therefore the evidence of full mediation is confirmed (Baron and Kenny, 1986;Sobel, 1982). Procedural justice to interactional justice was significant,  Interactional justice to Contextual performance was also significant,  The effect of procedural justice through mediating variable was also significant as per Sobel's (1982)  Procedural Justice and Adaptive Performance was Significant. Thus the precondition of mediation was substantiated (Baron and Kenny, 1986).H3(a) is retained  The Indirect effect 0.35 ×0.19=0.066 The resultant of indirect effect is 0.066, which is higher than the value of direct effect .03.  Therefore the evidence of mediation is authenticated (Baron and Kenny, 1986;Sobel, 1982). Procedural justice to interactional justice was significant,  Interactional justice to Adaptive performance was also significant,  The effect of procedural justice through mediating variable was also significant as per Sobel's (1982) test. Direct effect .03, insignificant  Full Mediation H3(b) is retained

Discussions, conclusions and recommendations
Discussion.This study has investigated the relationship of procedural justice with three types of job performance namely task performance, contextual performance and adaptive performance.The interactional justice has been tested as intervening variable.The findings of this study are in support with the earlier studied conducted by various authors in other countries.Authors have found that procedural justice has influence on wok performance of employees (Wang et al., 2010;Early and Lind, 1987).The interactional justice has been concluded as facilitator to the development of social exchange relationship (Cropanzano et al., 2002).Bies and Moag (1986) predicted that a person's judgement for fairness is an outcome of quality treatment received while execution of procedures in working environment.In congruence of these findings we found that interactional justice has significant influence on procedural justice, task, contextual and adaptive performance.Similar findings have also been documented by (Maserson et al., 2000;Wang et al., 2010).
We also formulated that interactional justice would behave as intervening variable between task, contextual and adaptive performance.Regarding the mediating influence, this study found that interactional justice fully intervened the relationship between procedural justice and three types of job performance.Justice scholars have discussed distinction between procedural justice and interactional justice.Procedural justice relies social exchange theory, with the perspective of individual to organizational relationship, whereas interactional justice is more inclined to the exchange of individual and supervisor relationship.Hence the findings of this study regarding fully mediating effect of interactional justice seem to be logical.

Conclusions and recommendations.
The study concluded that the significance of teachers' perception translates into task, contextual and adaptive performance.The perception regarding fairness in procedures significantly predicted the performance of teachers.Although in Pakistan and Bangladesh, Govt.Colleges the unified and predefined procedures are existed to execute the decisions, some time, unified procedures may not work and this study have found such evidences.Despite, there are voices in favour of the procedural fairness; there is more demand of dignified treatment in the executions of procedures.It is therefore recommended in the light of this study that not only procedures should be revised from time to time, but also their execution needs to be monitored more seriously.
test  Direct effect .09, insignificant  Full Mediation.H1(b) is retained H2(a): Without Mediation H2(b): With Mediation  The Direct Effect between test  Direct effect .09, insignificant  Full Mediation H2(b) is retained H3(a) Without Mediation H3(a) With Mediation  The Direct Effect between

Table no 2: Means, standard deviations, reliability coefficients, and correlations
All fit indices are given left corner of the model (Figure no.2).The assessments of hypotheses have been given in Table no.3, and the model was tested for direct and indirect effect shown in Table no.4.