Role clarity, perceived cohesion and felt responsibility as antecedents of altruism and conscientiousness among college teachers in Kerala

The concept of Organizational Citizenship Behaviour (OCB) was first coined by Organ and his colleagues in 1983 (Podsakoff et al., 2000). Earlier, Katz and Kahn (1966) had proposed the concept of extra role behaviour as those behaviours of employees that go beyond formal requirements of job, which Bateman and Organ (1983) considered as a lubricant to the social mechanism of an organization. OCB is employee’s voluntary commitment with the organization where he puts in efforts beyond his contractual obligations for which there are no specified rewards (Smith et al., 1983). It is more of a discretionary behaviour (Organ, 1988) based on voluntary actions (Somech and Drach-Zahavy, 2004). These behaviours do not find place in the formal performance appraisal systems (Van Dyne and LePine, 1998) and these roles ultimately are beneficial to the organization (Organ et al., 2006). OCB has been found to have considerable effect on organizational effectiveness (Podsakoff et al., 2000) and is needed to develop a healthy culture in workplace. It can prevent unethical and illegal practices and raise productivity (Van Dyne and Ang, 1998), and can support existing processes to maintain existing working conditions (Pickford and Joy, 2016). Importantly, organizations would become more dependent on employees who contribute irrespective of formal job requirements (Somech and Drach-Zahavy, 2004). Moreover, Halbesleben and Bellairs (2015) argue that employees may be motivated to select behaviours that provide them the opportunity to achieve their workplace goals, which ultimately manifest as OCBs.

While Katz (1964) identified two dimensions of OCB (innovation and spontaneity), Smith et al. (1983) argued OCB has altruism and general compliance as its dimensions. Later, Organ (1988) expanded the dimensions of OCB to five types of behaviours – altruism, courtesy, conscientiousness, sportsmanship and civic virtue. Van Dyne et al. (1994) suggested that obedience, loyalty and participation were the three dimensions of OCB. Podsakoff and MacKenzie (1994) concluded that helping behaviour, civic virtue and sportsmanship were the three dimensions. Podsakoff et al. (2000) condensed behaviours into seven dimensions – helping behaviour, sportsmanship, organizational loyalty, organizational compliance, individual initiative, civic virtue and self-development.

Olitalia et al. (2013) suggested that altruism is a very important factor that defines the role of a teacher. The nature of the job requires a teacher to be a motivator and facilitator (Djamarah, 1997), and teachers tend to have a strong desire to serve others (Stinnett, 1968). Teachers are strongly motivated by a sense of altruism and commitment (Scott et al., 1999). Since teachers are responsible for educating students, who are a group of the youth group of the society, it is important that teachers exhibit altruistic behaviour (Li et al., 2022). A teacher cannot have a day at work without exhibiting altruistic behaviour; it motivates the teachers and permeates their profession (Olitalia et al., 2013). Conscientiousness among teachers has a positive correlation with student achievements (Rockoff et al., 2008). Cheng and Zamarror (2016) found causal effect of conscientiousness of teachers on their quality, ratings and classroom protocols. Salgado (1997) found that conscientiousness is a strong predictor of work-related performance. Corocoran and O'Flaherty (2016) reported that the importance of measuring teachers’ conscientiousness is being realized and is becoming an area of increasing interest in research. Consciousness in teachers, being a part of their personality, has an effect on the behaviour and performance of teachers (Klassen and Tze, 2014) and influences their relations with students and pedagogical approaches (Murray, 1972). Given the evidence of the significance of altruism and conscientiousness among college teachers, the present study focuses on these two behaviours only.

Altruism is a sacrificial action with the intent of helping another person (Pavenkov et al., 2015). It is a state of love directed to others where egoism and self-indulgence are thrown away (Enc, 1990), and is aimed at improving the welfare of others (Organ et al., 2006). Podsakoff et al. (2000) considered altruism as supporting co-workers who have work-related problems. Altruism has been found to be an important factor for a teacher’s moral and professional success (Scott et al., 1999). It can lend meaning to the life of a teacher, since the job of a teacher is motivated by the need to help others (Stinnett, 1968). An altruistic teacher works without selfish purposes. Altruism is the primary criterion used to describe a teacher, and is one of the elements that express a teacher strongly (Scott et al., 1999). It is altruism that motivates one to become a teacher, and researches on altruism among teachers are important. Conscientiousness is a pattern of going well beyond the minimum needed levels of workplace behaviour including attendance, punctuality, housekeeping, resource management and internal maintenance matters (Organ, 1988). Many researchers have identified a form of behaviour called individual initiative, which is similar to conscientiousness, where employees involve in creativity or innovation to enhance organization’s performance. They also persist with enthusiasm and take up extra responsibilities (Podsakoff et al., 2000). It indicates an individual’s inclination to plan, organize and execute tasks to be reliable (Costa and McCrae, 2006). Conscientiousness is the personality domain that is most strongly associated with academic performance, and is very analogous to significant drivers of success (Kim et al., 2016). Higher levels of conscientiousness among teachers have also been found to enhance work performance and ultimately the academic achievement of the organization (Shrestha and Dangol, 2020). Teachers’ conscientiousness measures the most important dimension of teacher quality (Cheng and Zamarror, 2016), and has been proven to be the most important predictor of workplace performance (Dudley et al., 2006). Thus, based on the literature, it is evidenced that altruism and conscientiousness are the most important dimensions of behaviours that are the most important among teachers. Hence, the present study proposes to examine the levels of altruism and conscientiousness exhibited by college teachers in Kerala, and to identify its major antecedents.

Researchers have extensively attempted to explore the antecedents of OCB, with focus on four major categories (Podsakoff et al., 2000) – individual characteristics, task characteristics, organizational characteristics and leadership behaviours. Job satisfaction, organizational commitment, task clarity, etc. are among the most commonly studied antecedents of OCB.

Employees may develop many perceptions regarding their job which ultimately may be a predictor of the level of OCB among employees – for example, studies have proven that role ambiguity has been found to be a negative predictor of OCB. Similarly, role clarity is positively related to OCB (Podsakoff et al., 2000) and can play an important role in reducing costs related to work overload and conflict (Salamon and Deutsch, 2006). Given the multiplicity of roles that an employee is expected to do in work environment, role clarity has become a very important variable (Yadav and Rangnekar, 2016). Organ (1997) identified role clarity as an antecedent of OCB. Role clarity is the level to which an employee is sure about how the job has to be done (Shoemaker, 1999). Lack of role clarity can have impact on the psychological and behavioural outcomes of an employee (Rhoads et al., 1994). When an employee is sure about his role, he would be more directed towards achieving it (Braxton, 2008), ultimately resulting in increased competencies (Armstrong and Baron, 1998).

Though the effects of work groups on employee attitudes and behaviours were initially ignored, the importance of work groups was recognized and researched upon later (Kidwell et al., 1997). Members in a work group can be mutually committed towards each other regarding their tasks, and such feelings can be more than just attraction (Mudrack, 1989). Group cohesiveness is a group member’s affinity to other members of the group, which ultimately results in their decision to remain in the group. Where there is cohesion among members, they are more sensitive to each other and tend to assist each other (Schachter et al., 1951). Cohesive groups tend to be stronger in intra-group communication and influence (Cartwright, 1968), and exhibit greater enthusiasm, engaging in more positive interconnections (Schriesheim and Schriesheim, 1980). Brief and Motowidlo (1986) suggested that prosocial behaviours can be significantly influenced by group cohesiveness. Group cohesion can have an effect on an individual’s intention to group and such members perceive things in a very positive way (George and Brief, 1992). Where the members put in efforts to maintain social exchange relations within the group, such efforts can lead to OCB (Kidwell et al., 1997) and such effort is an important antecedent to promotive behaviours (Van Dyne et al., 1995) similar to OCB.

Since in-role behaviours of employees are insisted upon by the organization, an employee’s cognitive or affective mindset may not be used to explain how he performs in his task (Mowday et al., 1982). The feel of performance is often reflected in how the employee engages in extra-role behaviour (Pearce and Gregersen, 1991), which is largely, OCB. Felt responsibility is the degree to which an employee feels accountable for the results of their work (Goranson and Berkowitz, 1966). Morrison and Phelps (1999) suggested that employees who feel more responsible would seek out avenues and actions to improve their environments. Employees engage in extra role performance not because they expect a return from the recipient but because they feel a responsibility (Pearce and Gregersen, 1991). If employees feel responsibility towards the organization or fellow workers, they are more likely to help others through extra roles (Pearce and Gregersen, 1991). Hackman and Oldham (1976) postulated that the sense of responsibility in employees would lead them towards quality. Schwartz and Howard (1982) argued that a sense of subjective responsibility to help is an essential antecedent of altruistic behaviours. Felt responsibility was found to essentially have a prosocial nature (Fuller et al., 2006). Based on its prosocial origins, felt responsibility may be considered as a predictor of behavioural outcomes (Cantor, 1990).

Success of an educational institution depends on the teachers’ commitment to its goals and values (Somech and Oplatka, 2014). Prescribed role descriptions alone are not sufficient, and teachers are expected to go beyond formal duty (Belogolovsky and Somech, 2010). Support from colleagues is often found to be important for teachers (Halbesleben, 2006). Locke et al. (1983) found that for a college teacher, role clarity ensures that the teacher knows what is expected of him, gets feedback on his performance and also allows participation in decisions. Since “role” can be considered as duties that are unrestricted in power in comparison with “job” (Yadav and Kumar, 2017), it is natural that employees consider almost all activities (including helping others) at workplace as in-role (Morrison, 1994). Unclear job roles can lead to conflicts among employees (Fields, 2002), and role conflicts and role ambiguity were found to be negatively related to altruism and conscientiousness (as components of OCB) by Podsakoff et al. (2000). Importantly, where employees feel unrestricted in their roles, work becomes interdependent, it leads to felt responsibility for others’ work (Kiggundu, 1983). Behaviours at work place to help others, and being concerned about others’ work, can also be a result of highly cohesive groups (Van Dyne et al., 1995). Based on the evidences from the literature, the following hypotheses were formulated:

As discussed in the previous section, the attempt of the study is to establish the relation between Role clarity, Perceived cohesion and Felt responsibility (as antecedents) and Altruism and Conscientiousness (as outcomes) among the college teachers in Kerala. These five formed the constructs under study. Perception of teachers on Role clarity was measured through a scale adapted from Rizzo et al. (1970). It is a five-item scale anchored on a seven-point Likert-type scale (ranging from 1 = strongly disagree to 7 = strongly agree). The items relate to having planned goals for job, exact expectations out of the employee, performance evaluation, and authority and responsibilities related to job. Perceived cohesion was measured using a scale based on Bollen and Hoyle (1990). It is a six-item scale anchored on a seven-point Likert-type scale. The items measure the level to which an employee feels part of the group, and his or her level of happiness and belongingness to the work group. Felt responsibility was measured using the scale adopted from Pearce and Gregersen (1991) and Bishop et al. (2005). It is a four-item scale anchored on a seven-point Likert-type scale. The items measure how responsible an employee feels towards the work to be done, and how the employee feels about the credit or blame to be taken for the work. Altruism was measured based on the scale developed by Podsakoff et al. (1990). It is a four-item scale anchored on a seven-point Likert-type scale. The items measure how an employee would help co-workers at the time of need. Conscientiousness was measured using the scale developed by Podsakoff et al. (1990). It is a five-item scale anchored on a seven-point Likert-type scale. The items measure the level to which employee engages in work without leave or breaks, obeying rules even without supervision.

The study was conducted among the undergraduate and postgraduate college teachers across Kerala. There are government-owned, government-aided and self-financing colleges in the State. To ensure representation, samples were drawn from all the three categories of colleges. It was also decided to limit teachers with at least three years of undergraduate- or postgraduate-level teaching experience to be the sample respondents. This was done to ensure that the respondents are well-versed with the job requirements, and have some level of affinity towards the job of teaching at undergraduate or postgraduate level, so that they understand the meaning of the items in the survey instrument against the context of their job. The survey instrument was designed to collect information regarding the demographic profiles of the respondents and the constructs under study. The district-wise lists of colleges in Kerala were collected from the government department and universities concerned. There are 14 revenue districts in Kerala, and 5 districts were randomly selected from three geographic regions (Southern, Central and Northern) of Kerala. The colleges in the five districts form the sampling frame for the study. It was decided to cover at least 20% of the colleges. Using the list of colleges, employing systematic sampling method and using a sampling interval, the sample colleges were selected. The survey instrument was distributed among the teachers with a request to complete the survey (limited to teachers with at least three years of teaching experience at college level). The survey was conducted from December 2020 to March 2021. A total of 386 respondents completed the survey, but only 354 responses were useable for final data analysis. Profile of the respondents showed that out of the final sample, 62% of respondents were female and 56% of the respondents had a teaching experience of more than ten years at college level. It was also revealed that 24% of the respondents belonged to government colleges, 35% belonged to aided colleges and the remaining to self-financing colleges. Majority of the respondents belonged to an age group of 30–45 years. The overall objective of the study was to examine the causal effect of Role Clarity, Perceived Cohesion and Felt Responsibility on Altruism and Conscientiousness of college teachers in Kerala. The cause–effect relationship was examined using Partial Least Square based Structural Equation Modeling, which is a causal predictive approach which develops statistical models to provide causal explanations (Sarstedt et al., 2017). It is a variance-based analysis (Rigdon et al., 2017) and its use is advocated where the model is composed of complex relations (Hair et al., 2017). The assessment was divided into three stages – model specification, measurement model and structural model (Chin, 2010). The assessment of the model was performed using SmartPLS Version 3.0 (Ringle et al., 2022).

The data collected were examined for missing values and outliers. On visual examination, it was found that there were some cases of missing entries. The missing values were deleted list-wise. Data were also examined for suspicious response patterns, referred to as straight lining (Hair et al., 2014), where a respondent would mark the same response for a very high proportion of questions. There were no such patterns in the data. A statistical power analysis (Sun and Fan, 2017) was performed for sample size estimation using G*Power software Version 3.1. The effect size in the study was set at 0.15, considered to be medium based on Cohen’s (1988) criteria. With an alpha value = 0.05 (type I error, probability of finding significance when there is none) and power = 0.80 (probability of finding true significance), the estimated sample size is 119. The actual sample size of 354 is more than three times the estimated sample size. Thus, the sample size for the study is proven to be adequate. To identify multivariate outliers in data (Ghorbani, 2019), Mahalanobis Distance and its p-value were computed using SPSS Version 20.0. Since all the p-values were greater than 0.001, it is confirmed that the data did not have outliers. For examining whether data conform to normal distribution characteristics, multivariate normality (Zhou and Shao, 2014) was also examined using Mahalanobis distance. Since for large samples, observed Mahalanobis distance approximates chi-square distribution, the maximum observed value (7.322) was compared with chi-square critical value (9.488) at degree of freedom = 4; significance level = 5%. Since maximum observed value is less than the chi-square critical value, it is concluded that multivariate normality exists. Possibility of common method bias (Podsakoff et al., 2012) in the survey instrument was examined through Harman Single Factor Test (Tehseen et al., 2017), using Principal axis factoring with extraction of one factor. The results showed that a single factor accounted for 26.58% of the total variance only, indicating that the data is free from bias. Existence of multicollinearity among predictor constructs was examined. Multicollinearity is the high correlation among the predictor constructs (Grewal et al., 2004). Mason and Perreault (1991) documented that correlation among predictors can lead to inaccurate estimates of path coefficients, standard errors and inference errors. Chin (1998) suggested using Variance Inflation Factor (VIF) to examine multicollinearity, where VIF values above 4.0 indicate its existence (Garson, 2016). Results showed that VIF computed values (Role Clarity = 1.641; Perceived Cohesion = 1.803 and Felt Responsibility = 1.676) were much lower than the recommended threshold of 4.0, proving that there exists no collinearity among the predictors.

The theoretical model is composed of three predictors (Role Clarity, Perceived Cohesion and Felt Responsibility) and two outcome (Altruism and Conscientiousness) constructs. As discussed in the methods section above, role clarity was measured using five indicators (items), perceived cohesion using six indicators and felt responsibility using four indicators. Altruism was measured using five indicators and conscientiousness using five indicators. All the five constructs were defined as reflective constructs, based on literature.

After the completion of the preliminary analysis of data, the evaluation of the measurement model was performed by confirmatory factor analysis. It involved the examination of reliability and validity of the measures of the construct (Ali et al., 2018). Tables 1 and 2 give the results of measurement model evaluation. Reliability measures the quality of the construct and the stability of the measurement (Sekaran and Bougie, 2010). Since the model is reflective in nature, reliability was measured using composite reliability (Janadari et al., 2016) and Cronbach’s alpha (Cronbach, 1951). Results show that for all constructs, the computed composite reliability scores – Role Clarity (0.925), Perceived Cohesion (0.919), Felt Responsibility (0.848), Altruism (0.900) and Conscientiousness (0.864) – lie between the minimum of 0.60 recommended by Henseler and Sarstedt (2013) and the maximum limit of 0.95 recommended by Drolet and Morrison (2001). Thus reliability of the measurement model is established.

Convergent validity signifies the accuracy of a measure, truly representing a concept under discussion (Zikmund et al., 2013). It is measured by item loadings and Average Variance Extracted (AVE) (Hair et al., 2014). Item loadings are coefficients in a factor structure matrix and represent the extent to which the indicators load on its respective construct which it measures and has to be above 0.708 recommended by Hair et al. (2017). The results show that all the indicators load highly on their respective construct – Role Clarity (0.847–0.967), Perceived Cohesion (0.719–0.963), Felt Responsibility (0.769–0.857), Altruism (0.792–0.898) and Conscientiousness (0.800–0.823). AVE is the grand mean value of the squared loadings of the indicators associated with the construct (Hair et al., 2014). AVE values above 0.50 are recommended by Hair et al. (2017) and Chin (1998). The results show that in all cases, the computed AVE values are above the recommended level – Role Clarity (0.835), Perceived Cohesion (0.751), Felt Responsibility (0.674), Altruism (0.715) and Conscientiousness (0.658). Hence, convergent validity is established.

Discriminant validity is the extent to which a construct is truly distinct from all other constructs and thus is a measure of its uniqueness (Hair et al., 2017). Presence of discriminant validity proves that a particular construct does not correlate high with other constructs (Campbell, 1960). It is measured using Fornell–Larcker criterion (Fornell and Larcker, 1981) which examines the square root of AVE values with latent variable correlations (Hair et al., 2014). It is prescribed that the values have to be greater than the correlations (Fornell and Larcker, 1981; Hair et al., 2014). The results show that the computed F–L criterion values of each construct satisfy the criteria recommended. Since the values are greater than the correlations with all other constructs, it is concluded that the model possesses discriminant validity. Since it was proven that the model satisfies all the requirements of reliability and validity, the study moves to assessment of structural model (Chin, 2010).

Structural model presents the cause–effect path relations between the predictor and dependent construct (Götz et al., 2010). As such, its assessment helps to understand the empirical model’s capability to predict the dependent construct (Hair et al., 2017), providing evidence to support the theoretical model developed (Chin, 2010). Structural model’s quality is based on path coefficients direction and its significance levels (Chin, 1998). Coefficient of determination (R²) of dependent construct and Stone–Geisser’s Q² value are also assessed (Chin, 1998). Tables 3 and 4 give the results of structural model assessment.

The causal relation (path coefficient) from Role clarity to Altruism is 0.072, a very low positive effect, which is also not statistically significant (t = 1.104; p-value > 0.05) at 5% level. The causal relation from Perceived cohesion to Altruism is 0.463, a high positive effect, which is statistically significant (t = 7.941; p-value < 0.001). The effect from Felt responsibility to Altruism is a very high positive 0.512, which is statistically significant (t = 14.165; p-value < 0.001). With Conscientiousness as the dependent construct, the cause–effect relation from Role clarity (path coefficient = 0.375; t = 6.476; p-value < 0.001), Perceived cohesion (path coefficient = 0.428; t = 7.332; p-value < 0.001) and Felt responsibility (path coefficient = 0.394; t = 6.909; p-value < 0.001) are all positive and statistically significant.

Coefficient of determination (R²) measures the model’s predictive accuracy and measures the sum of the effect of all predictor constructs on each of the outcome construct (Hair et al., 2014). Ranging from zero to one, as a rough rule of thumb, values of 0.75, 0.50 and 0.25 may be considered as substantial, moderate or weak effect (Henseler et al., 2009). The R² value of Altruism is 0.413, indicating that 41.3% of variation in Altruism is explained by its predictors, Role clarity, Perceived cohesion and Felt responsibility, and it is statistically significant (t = 8.179; p-value < 0.001). The R² value of Conscientiousness is 0.584, indicating that 58.4% of variation in Conscientiousness is explained by its predictors, Role clarity, Perceived cohesion and Felt responsibility, and it is statistically significant (t = 12.171; p-value < 0.001). Based on Henseler et al. (2009) recommendation, the predictive accuracy of Altruism is weak (between 0.25 and 0.50), while the predictive accuracy of Conscientiousness is moderate (between 0.50 and 0.75). Stone–Geisser’s Q² value (Geisser, 1974; Stone, 1974) is a measure of predictive relevance of the model, with values larger than zero for a particular outcome construct indicating path model’s predictive relevance for that construct (Hair et al., 2014). The computed value of predictive relevance (Q²) of Altruism is 0.305 and of Conscientiousness is 0.429, which implies that the model has good predictive relevance for both the outcome constructs.

Based on the results of the structural model assessment, the following are the results of the tests of hypotheses formulated. The hypotheses were tested on the basis of two results: (a) p-value of the path coefficients and (b) whether zero falls into the confidence interval (CI).

Table 5 gives the results of the hypotheses test. The effect of Role clarity on Altruism is positive, but not statistically significant (path coefficient = 0.072; T-static = 1.104; p-value = 0.237; CI [−0.056, 0.200], 0 ∈ CI). Thus, there is no support to the hypothesis that role clarity has a significant positive effect on altruism. The effect of Perceived cohesion on Altruism is positive, and statistically significant (path coefficient = 0.463; T-static = 7.941; p-value < 0.001; CI [0.349, 0.577], 0 ∉ CI), supporting evidence of significant effect of perceived cohesion on altruism. Felt responsibility is found to have significant positive effect on altruism, and is statistically significant (path coefficient = 0.512; T-static = 14.165; p-value < 0.001; CI [0.441, 0.583], 0 ∉ CI).

Table 5 also shows that the effect of Role clarity on Conscientiousness is positive, and statistically significant (path coefficient = 0.375; T-static = 6.476; p-value < 0.001; CI [0.262, 0.488], 0 ∉ CI). The effect of perceived cohesion on conscientiousness is positive and statistically significant (path coefficient = 0.428; T-static = 7.332; p-value < 0.001; CI [0.314, 0.542], 0 ∉ CI). Felt responsibility has a positive statistically significant effect on conscientiousness (path coefficient = 0.394; T-static = 6.909; p-value < 0.001; CI [0.282, 0.506], 0 ∉ CI).

Overall, the study was conducted to develop a structural model based on the cause–effect relation between three predictors (Role Clarity, Perceived Cohesion and Felt Responsibility) and two outcome (Altruism and Conscientiousness) constructs. The study had drawn on existing literature to develop the theoretical model for the cause–effect relations. Over the years, a number of researchers have attempted to establish the theoretical model on OCB and its antecedents. Evidences suggest that Altruism and Conscientiousness are two of the major components of OCB among employees. The present study attempted to measure altruism and conscientiousness among the college teachers in Kerala. Altruism indicates the readiness of the college teachers to assist colleagues who either have been absent and with pending responsibilities, or have extra and heavy workload. Teachers also exhibit a propensity to a kind of training or orientation to newly recruited teachers. In a nutshell, altruism among college teachers is simply being there to lend a helping hand at workplace. Conscientiousness among college teachers in the present study dealt with a feeling among them to express commitment to the organization by performing above the norm in terms of attendance, and being obedient to rules even without supervision. There exists a feeling of mutuality between the teachers and the organization, where the teacher gives something back to the college for the money that is taken as a day’s pay.

As antecedents or predictors of altruism and conscientiousness, from the existing literature, role clarity, felt responsibility and perceived cohesion were identified. Role clarity for college teachers consisted of having a feeling that teaching has pre-established, clear and planned goals to be achieved. These are not found to be strictly limited to formally prescribed job roles, but also included other activities within small or large subgroups or groups (for example, activity subgroups like career guidance committee, small groups like a particular department or large groups like college council). Felt responsibility is particularly important for college teachers, since any slight discrepancy in the discharge of official functions of a teacher can have serious consequences on the future of the organization, as well as the life of the students. Study found that the teachers are well aware of the personal responsibility that is bestowed upon them in the discharge of their duties and roles. It can span from classroom teaching to personal mentoring and counselling. Thus, it is important that the teachers understand that it is their own responsibility to take the blame for something that is not well done. Moreover, problems particularly pertaining to students need to be taken care of immediately, before it escalates. Ultimately, felt responsibility is simply about teachers recognizing and accepting that it is their personal responsibility whether the job is properly done or not. Perceived cohesion relates to the feeling of belongingness to the groups to which teachers may be assigned. Irrespective of whether the group is small or big, it is important that the teachers develop a sense of being a part of the group, easily able to identify themselves with the philosophy of the group. In the presence of cohesiveness among the members, individual initiatives, creativity, mutuality and performance become the identity of the group. The members develop a feeling that their group is the best anywhere, and inculcate the feeling of contentment and self-actualization.

The major objective of the study was to examine whether Role Clarity, Perceived Cohesion and Felt Responsibility have a causal effect on Altruism and Conscientiousness of college teachers. The causal effect was tested using partial least square based structural equation modelling. Measurement model was evaluated to ensure that the constructs exhibited acceptable levels of reliability and validity. All the indicators of the five constructs were highly correlated with the other indicators of their own respective construct, as exhibited by Cronbach’s alpha and high item loadings. Moreover, AVE was also more than 0.5, indicating high levels of discriminant validity. Once the reliability and validity of the measurement model were established, the study moved on to assess the structural model. The assessment of structural model involved computation of path coefficients, its significance, predictive accuracy and predictive relevance of the empirical model. The results proved that the causal effect of Role clarity on Altruism was very low and not statistically significant, but on Conscientiousness was positive and statistically significant. It evidences that the clarity with which the college teachers understand their job requirements, goals and expectations does not inculcate a feeling to help their colleagues; conversely, such clarity of role induces them to exhibit more diligence in the execution of their job, making them more obedient and truthful. The causal effect of Perceived cohesion on Altruism and Conscientiousness was found to be positive, high and statistically significant. It became evident that where college teachers felt being a part of the work group, and took pride in it, they were willing to help their colleagues at work. Thus, belongingness enhanced their levels of empathy for others, and usually, such proactive feelings towards others need not be limited to group members alone, but can percolate to colleagues in other groups, but within the same organization. Cohesiveness in the group also makes a teacher more oriented towards organizational goals, thereby contributing more towards its achievement, even in the absence of strict supervision and control. The effect of Felt responsibility on Altruism and Conscientiousness was also found to be positive, high and statistically significant. The study provided support to the theory that Felt responsibility has the highest and most significant effect on Altruism. In fact, it is proved that when a teacher feels that he or she is utmost responsible for the outcome of own actions, and that failure to execute it successfully can lead to very undesirable consequences, such feeling prompts them to help others in their job. This can be because of two reasons – one, to ensure that colleagues do not fail, and two, there is the realization that even when others fail, it can lead to outcomes that can affect everyone. It is also natural that the feeling of responsibility about one’s actions can lead a teacher to be more attentive at work, contributing more to the general cause beyond personal gains, and being a law-abiding employee.

The study contributed to the existing theory on antecedents of OCB, by providing evidences that its two components – Altruism and Conscientiousness – can be significantly affected by antecedents like Role clarity, Cohesiveness and Felt responsibility. It has to be noted that the study does not provide support to the theory of significant direct effect of Role clarity on Altruism, inconsistent to the findings of Podsakoff et al. (2000) and Yadav and Kumar (2017). In this context, Jahangir et al. (2004) suggested that employee satisfaction may be a significant mediator between Role clarity and Altruism. It is concluded through the study that the model has high levels of predictive accuracy, meaning that the predictors – Role clarity, Cohesiveness and Felt responsibility – are substantially capable of explaining the variations in the dependent constructs – Altruism and Conscientiousness.

Drawing on the idea of “willingness to cooperate” by Barnard (1938), the concept of OCB among employees is a topic that has been extensively researched upon, even leading to some conceptual confusion (Podsakoff et al., 2000). Though such behaviour is a multidimensional concept, the present study has limited itself to two dimensions only – Altruism and Conscientiousness – to understand how and why college teachers behave the way they do. In future, further research can adopt a more comprehensive view of such proactive behaviours. Moreover, Podsakoff et al. (2000) have identified that research has focused on four broad categories of antecedents of OCB – individual characteristics, task characteristics, organizational characteristics and leadership behaviours. The present study has analyzed a limited set of predictors of proactive behaviours of college teachers. Further studies can attempt to develop a more elaborate theory of antecedents of OCB among teachers. Moreover, the consequences of OCB can also be studied so as to develop a more rigorous theoretical framework on how OCB among college teachers can affect organizational performance and success.

Makesh Gopalakrishnan presently works as Assistant Professor in Commerce, TM Government College Tirur, Malappuram, Kerala, India. He is an approved guide for PhD in Commerce at MG University, Kottayam, Kerala. He was awarded PhD in Commerce from University of Kerala, Thiruvananthapuram, Kerala. He has numerous publications and has presented papers at international conferences. He has a teaching experience of over 15 years.

Ajish Abu presently works as Assistant Professor in Commerce, Government College Tripunithura, Ernakulam, Kerala, India. He is pursuing his doctoral studies in Commerce at MG University, Kottayam, and is Assistant Professor in Commerce, Government College Tripunithura, Ernakulam, Kerala. He has a teaching experience of over 15 years.