The affective commitment of academics in a university in Botswana

1 Department of Sociology, University of Botswana, Private Bag 00705 Gaborone, Botswana. 2 Botswana Accountancy College, University of Derby in Botswana, School of Business and Leisure Gaborone, Botswana. 3 School of Nursing, Austin Peay State University, Clarksville, TN, USA. 4 Department of Statistics, University of Botswana, Private Bag 00705, Gaborone, Botswana. 5 Optimal Solutions in Healthcare and International Development, Institute of Sociology, University of Valparaiso, Chile.


INTRODUCTION
There exists no single human factor that has impacted organisational outcomes more than organisational commitment (Alsiewi and Agil, 2014).Commitment in the workplace has the potential to influence organizational effectiveness and employee well-being (Meyer and Herscovitch, 2001).As a result, organisational commitment is one of the most important constructs for organisational researchers and managers of organisations.It has been a major focus of research since the 1990s (Meyer et al., 2002).It is particularly relevant in an era when resource scarcity and cutback management have become the norm, and when managers of public institutions across countries are addressing challenges of attraction and retention of highquality employees as well as their morale, motivation and performance (Chordiya et al., 2017).
Most research on organisational commitment has been undertaken in business organisations (Lovakov, 2016).Although academic employees are a university"s key asset that determines the institution"s success (Lovakov, 2016), and their attitudes influence their performance and willingness to do more than what is formally described in their contracts (Jing and Zhang, 2014), scholars have been slow to explore how the tertiary education environment influences their commitment (Chughtai, 2013).It is important to understand the special psychological bond between academics and their university (Lovakov, 2016).

Conceptual framework
Commitment has been conceptualized as a stabilizing or obliging force that gives direction to behaviour by restricting freedom and binding the individual to persist in a course of action even in the face of conflicting motives and attitudes (Meyer and Herscovitch, 2001;Scholl, 1981).A committed individual persists in a given course of action even under circumstances where they would otherwise be tempted to change their course of action or behaviour (Meyer and Herscovitch, 2001).It is a state of being in which an individual becomes bound by his actions and beliefs that sustain the activities of his or her own involvement (Salancik, 1977).
Commitment is a binding force that is experienced as a mind-set (that is, a frame of mind or psychological state that compels an individual towards a course of action) that has three distinguishable themes: perceived cost of leaving, obligation to remain, and affective attachment to the organization (Meyer and Herscovitch, 2001).The three mind-sets of a committed individual have been labelled as affective commitment (desire or emotional attachment to the organization), continuance commitment (perceived cost of leaving the organization), and normative commitment (obligation to remain in the organization (Allen and Meyer, 1990;Meyer and Allen, 1984, 1991, 1997).
Affective commitment is an employee"s emotional attachment to, identification with and involvement in the organisation (Meyer and Allen, 1997).
It involves cohesion or attachment to social relationships which absorb the individual"s fund of affectivity and communion or becoming part of a whole through the mingling of the self with a group (Kanter, 1968).It is an attitude towards an organization which links or attaches the identity of the employee to the organisation (Sheldon, 1971).
Affective Commitment manifests as a psychological bond to the organisation (Buchanan, 1974) as well as positive feelings for and social attachment to the organisation (Still, 1983).It is an attitudinal type of Nkhukhu-Orlando et al. 13 commitment (Meyer and Allen, 1991), which involves "identification" or the alignment of individual values with values of the organisation, "involvement" or feelings of care for the organisation, devotion and dedication to the organisation, pride in the organisation, willingness to put forth extra effort into the organisation, making sacrifices for the good of the organisation, and loyalty or a strong desire to maintain membership in the organization (Gbadamosi et al., 2007;Liou, 2008;Mowday et al.,1979).The desire of the individual to remain in the organisation is the basis for Attitudinal Commitment Theory (Mercurio, 2015).Following an integrative literature review of high impact journal articles, Mercurio (2015) noted that although many conceptualizations since the 1960 and 1970s have depicted commitment as equally weighted components (affective, normative, continuance), the affective construct of commitment has remained central and constant through a wide diversity of the theorizing and multidimensional conceptualizations of organizational commitment.
Affective, or emotional and attitudinal attachment to the organization was demonstrably an important core essence of the organizational commitment construct.It was the most influential, enduring, indispensable, and central characteristic of organisational commitment that seems to serve as an historical and theoretical base for organisational commitment theories (Mercurio, 2015).It is the centre core that most strongly affects work behaviours and feelings and shapes individual perceptions than other components or proposed forms of commitment.

The impact of affective commitment
An affective bond with an organization represents an emotional involvement, identification, and value congruence with the organization.It contributes to a mind-set that involves a cognitive recognition that there is an important purpose in what one is doing in an organization characterised by desire to follow a course of action and exert effort to achieve organizational goals (Meyer and Herscovitch, 2001).
Several practical and theoretical implications of affective commitment have been highlighted in the research literature (Chordiya et al., 2017;Lam and Liu, 2014;Meyer and Allen, 1991;Schoemmel and Jønsson, 2014;Wang et al., 2010).
In addition to positive associations with employee physical and psychological well-being, and with job satisfaction (Lovakov, 2016), affective commitment works to strengthen employees" identification with the organization (Allen and Meyer, 1990).It also enhances emotional, mental and physical investment in the organisation, attachment to the goals of the organization and its vision (Moon et al., 2014).
This study focuses attention on affective commitment on the understanding that an individual with a strong affective commitment to an organization might be willing to go the extra mile, above and beyond the call of duty, to achieve organizational goals compared to one whose commitment is based primarily on a sense of obligation to the organization or a recognition of lack of alternatives or the costs of leaving the organization.The logical choice of affective commitment is also derived from the fact that continuance commitment and normative commitment have a weaker impact on behaviour than affective commitment.

Purpose of the study
Although it is one of the most predictive factors of employees" behaviour in university settings and corporate contexts (Morin et al., 2011), there is limited research output on organisational commitment of academic employees in the African continent (Fako, et al., 2018).The few empirical studies on organisational commitment and related constructs in Botswana were not focused on academic employees (Fako and Forcheh, 2000;Fako et al., 2009;Ongori, 2007).One study that focused on academic employees used a single indicator "employees" desire to work for the University for ever" (Fako et al., 2014) to study affective commitment.Another study investigated organizational commitment and not affective commitment (Fako et al., 2018).This study investigates affective commitment using eight well established items adapted from Allen and Meyer (1990).
The study aims to explore the nature of associations between affective commitment and several variables that have been reported as possible antecedents, correlates and consequences of component measures of commitment in the extant literature (Fako et al., 2014;Lam and Liu, 2014;Meyer and Allen, 1991;Meyer et al., 2002;Schoemmel and Jønsson, 2014;Wang et al., 2010).From the list of variables that were found to be associated with affective commitment, further analyses were done to identify a parsimonious set of factors that predict affective commitment among academic employees.Scholars and management practitioners have suggested that predictors of affective commitment should be integrated into human resource processes (Chughtai, 2013;Meyer and Allen, 1997).A predictive model of affective commitment should be useful in attempts to identify factors that tend to lead employees to developing a mind-set of desire to be highly involved and to exert effort to achieve organizational goals due to a belief in the value of what they are doing (that is, affective commitment).
It was hypothesized that employees with affective commitment are more likely to consider the best interests of the organization than those whose mind-set is characterised by remaining in the organization to avoid costs (continuance commitment) or out of a sense of obligation (normative commitment) (Meyer and Herscovitch, 2001).

Research design
The study used a cross-sectional survey design.Respondents were selected from among a population of academic employees with office space at the University of Botswana.The study population was stratified into units with an average size of nine academic employees each.Large departments such as English, Mathematics and Economics that had about 25 or more academic staff were divided into three enumeration units each.Two or three academic staff members were selected from each enumeration unit.A selfadministered questionnaire was distributed over a period of two weeks.Informed consent was obtained from each respondent.The study adopted a quantitative non-experimental design using a survey instrument.Both the research protocol and the study instruments were approved by the University of Bostwana's Human Subjects Review Board.

Measurement of variables
Affective commitment was measured using eight items adapted from Allen and Meyer (1990) and phrased to be meaningful in the context.For each of the eight items, a five-point Likert-type scale, which ranged from strongly disagree (SD=1) to strongly agree (SA=5) was used.The reliability of the scale was reasonable (Cronbach"s alpha = 0.79).The research protocol and instruments were approved by the University of Botswana"s Human Subjects Review Board.
Demographic and personal characteristics were measured by requesting participants to check appropriate categories.These included gender, age-group, marital status, citizenship status, academic faculty in which the respondents worked, highest academic qualification, academic rank, salary scale, etc. Attitudinal and other variables were measured using questions on a five-point Likert scale that ranged from 1 (low) to 5 (high), indicating the presence of "very little" or small amounts of the measured attitude to "a lot" or large amounts.Attitudinal and other variables included: (1) extent of contribution to policy making, (2) frequency of responding to emails, (3) perception of the status of academics, (4) satisfaction with the university management, (5) perception of opportunities for research and scholarship, (6) perception of the working habits of students, (7) personal efficiency, (8) satisfaction with the job, (9) sense of accomplishment, (10) acceptance of the mission statement, (11) internal locus of control.
The scales used to measure attitudes and other variables demonstrated a good reliability (Cronbach"s alpha (α) ranged between 0.83 and 0.89).

Data analysis
During analyses, the research variable, affective commitment was converted to a dichotomous variable coded into "committed" versus "not committed" (whereby 1= affectively committed, and 0=not affectively committed).Respondents who "agreed" with positively worded statements were treated as having affective commitment, while those who "disagreed" were treated as not having affective commitment.Independent variables considered for analysis were Felt there were inadequate opportunities for research 50.9 also recoded as categorical variables.Cross tabulations were done to determine variables that were significantly associated (p≤0.05) with affective commitment using chi-square tests of association.
In order to predict the affective commitment of academic employees, variables that had a statistically significant relationship with affective commitment were included in a hierarchical multiple logistic regression procedure.The percentage of employees that were correctly reclassified by the fitted model was used as a measure of goodness of fit.This statistical technique facilitated the use of step-wise multiple-regression which showed the relative importance of the independent variables in predicting affective commitment.
Most respondents (71.5%) were males.The majority (81%) were aged 40 years or older.Some 42.7% were employed on two-to-five-year renewable fixed-term contracts that were typically held by non-citizens.Almost a quarter (24.5%) had worked for the university for five (5) years or less, while 16 % had worked for the university for over 20 years.Table 1 shows the distribution of respondents by background and other attributes.

Extent of affective commitment
Analyses started with a determination of the extent to which respondents had affective commitment to the organisation.The results indicate that, only one third (34.1%) of the academic employees sampled had affective commitment to the university.There were no significant differences in affective commitment by antecedent demographic variables and personal characteristics such as gender, marital status, citizenship status, academic rank and number of years employed at the university.This lack of significant differences in affective commitment by demographic variables reflects the homogeneity of the sample.There do not appear to be great differences in marital status, citizen status, academic rank, and years employed by the university which would produce significant differences in affective commitment.

Factors associated with affective commitment
The results show that the following nine factors were associated (p≤0.05) with affective commitment: (1) extent  4) satisfaction with the university management, (5) perception of opportunities for research, (6) perception of the working habits of students, (7) sense of personal efficiency, (8) satisfaction with the job, and (9) sense of accomplishment (Table 2).
Table 3 shows that respondents who were not happy with many aspects of their university life were more likely to have affective commitment than those who were generally happy.For instance, those who were not satisfied with the university management were more likely ( 2 =5.493, p=0.019) to have affective commitment than those who were satisfied with the management.Those who had not contributed to policy making, were more likely (p=0.006) to have affective commitment than those who had contributed to policy making.
Those who perceived that opportunities for research at the university were not good were more likely ( 2 =4.810, p=0.028) to have affective commitment than those who felt that opportunities for research were good.Those who felt that they were not efficient at work were more likely ( 2 =4.118, p=0.042) to have affective commitment than those who felt that they were efficient at work.Those who felt that students had poor working habits were more likely ( 2 =4.311, p=0.038) to have affective commitment than those who felt that students had good working habits.
Those who did not feel a sense of accomplishment at work were more likely ( 2 =3.845, p=0.050) to have affective commitment than those who felt a sense of accomplishment.Similarly, those who did not like their job more than others were more likely ( 2 =3.963, p=0.047) to have affective commitment than those who liked their job more than others.Two factors had a positive association with affective commitment: (1) responding to e-mail communication and (2) perceptions of the status of academic staff.Those who "always responded" to e-mail communication were more likely ( 2 =5.847, p=0.016) to have affective commitment than those who "did not always" respond to e-mails.Those who indicated that the status of academic staff was good were more likely ( 2 =5.716, p=0.017) to have affective commitment to the university than those who believed the status of academic staff was good.
An additional eight (8) factors were marginally associated with affective commitment (0.05<p<0.1).These included: (1) acceptance of the university mission statement, (2) locus of control, (3) perception of the quality of management, (4) acceptance of student evaluation of courses and teaching, (5) support from administrative staff, (6) perception of the quality of the teaching and learning environment, (7) perception of the ability of the students and (8) desire for a managerial job.The marginally significant factors were excluded from further analyses on the basis of the rejection criterion ( 0.05)

 
suggested by Hair et al. (2006).However, they are reported here to aid further exploration by subsequent studies since they could have statistically significant relationships with affective commitment in different and larger samples.

Predictive model for affective commitment
The nine factors associated with affective commitment (a) contributing to policy making, (b) responding to emails, and (c) satisfaction with management.Table 4 indicates that the improvement in chi-squared was significant (P<0.01)when a factor was added at each step.The percentage of case that was correctly classified as having affective commitment or not was 65.9% when satisfaction with the management was the only predictor in the model.This percentage increased to 67.1% when contribution to policy making was added as a factor, and to a further 68.3% when responding to emails was added as an additional factor.
A model for predicting affective commitment among the academic employees is shown in Table 5.Overall, the Hosmer and Lemeshow goodness of fit chi-squared test statistics was found to be 2.77, with six degrees of freedom, which gave a p-value of 0.837.This implies that the final model fits the data well.
The model predicts that an academic who is not satisfied with the management is 2.39 times more likely to have affective commitment as one who is satisfied with the management.An academic who does not contribute to policy making is 2.32 times more likely to have affective commitment as one who contributes to policy making.An academic who always responds to e-mail communication is 2.47 times more likely to have affective commitment than one who does not always respond to emails.

DISCUSSION
This study sought to establish whether or not academic employees had affective commitment to the university and whether there were predictors of this affective commitment.The study investigated factors that have been identified in the literature as antecedents, correlates and consequences of commitment.In addition, the study identified variables that can predict affective commitment among academic employees.

Demographic variables
Contrary to studies that found associations between commitment and antecedent background factors such as   and Zafar, 2013;Pathardikar and Sahu, 2011;Yucel and Bektas, 2012), this study found no association between affective commitment and demographic variables.

Other variables
Logistic regression analysis identified three predictors of affective commitment among academic employees.They were satisfaction with management, contribution to policy making, and responding to emails.It could be argued that academic employees with a sense of self efficacy, who were emotionally attached to the university had some issues with the university management about the introduction of a major policy to reorganise the academic organisational structure of the university.Restructuring of the academic organisational structure was a management policy initiate that had generated considerable debate at the time of data gathering.Many academic employees (57%) felt that the proposed restructuring was not a good idea.Many felt that they had been excluded from meaningful participation or had not been adequately consulted as the policy went through various stages of approval.As a result, many (55.2%)indicated that they had not made a useful contribution to new policy initiates at the university.

Conclusions
Employee affective commitment remains an essential factor in key university organization outcomes such as effective teaching and research productivity.When professionals work in bureaucratic organizations, they tend to feel conflicting goals and are compelled to choose one loyalty over another (Setyowati and Suharnomo, 2017).They may have an attachment to the organization, including some of its goals, for reasons that are not related to the achievement of goals articulated by management.
Committed academic employees, effective university teachers and productive researchers may not necessarily agree with or believe in some of the policy initiates that the management may initiate and promote.The positive relationship between dissatisfaction with management and affective commitment suggests alienation of employees with a mind-set focused on what they believe is the reason for the existence of the university.More research is required to understand the different loyalties that academic professionals may have to their students, colleagues, superiors, unions, management, research collaborators, performance targets, institutional policies and the organization in general.

LIMITATIONS AND FURTHER RESEARCH
Like most research on organizational commitment, this study was cross-sectional and focused on statistical associations, which makes it difficult to ascertain the direction of causality.Research using experimental or longitudinal designs is needed to verify causal effects.Another limitation of the study is that the sample was collected from only one institution.It would be useful to undertake a much more comprehensive study that involves several public and private institutions.

Table 1 .
Distribution of respondents by attributes.

Table 2 .
Associations between affective commitment and independent factors.

Table 3 .
Association between affective commitment and predictors.

Table 5 .
Logistic regression model for predicting affective commitment.