Demographic and job satisfaction variables influencing academic staffs’ turnover intention in Debre Berhan University, Ethiopia

Abstract Academic staff turnover intention is one of the problems facing different universities in Ethiopia. This problem has also become a serious one at Debre Berhan University. Thus, this study aimed to investigate the influence of demographic factors and job satisfaction variables on the academic staff turnover intention at Debre Berhan University. For this purpose, both descriptive and explanatory research designs were employed. The research followed a quantitative research approach. The study used primary data. The data was obtained from 265 academic staff selected from nine colleges using both stratified sampling and simple random sampling techniques. A binary logistic regression model was used to examine the influence of demographic factors and job satisfaction variables. Accordingly, it is found that pay and benefits, working environment, and ethnicity were found to be the significant factors that affect academic staff turnover intention. This study suggests that the management of the university and the Ministry of Education, in order to increase the staff’s salary based on the current labor market, develop different benefits packages such as building them residence houses, providing health insurance and facilitating credit services for different purposes by collaborating with banks and other financial institutions. Besides, the government should ensure political stability by minimizing ethnic-led conflict by avoiding ethnic-based federalism.


PUBLIC INTEREST STATEMENT
Employee turnover intention is influenced by demographic characteristics and job satisfaction both in developing and developed countries. This study intended to examine the effect of demographic factors and job satisfaction variables on the academic staff turnover intention at Debre Berhan University. It is found that career advancement and pay and benefits were found to be the significant factors that affect academic staffs' turnover intention in Debre Berhan University.

Introduction
Academic turnover intention is a major problem worldwide, especially in developing countries, particularly in Africa (Ahmad et al., 2012). Academic turnover intention is a challenging issue in higher academic institutions in Ethiopia. But it is also a key indicator of human resource development practice and the overall effectiveness of organizational leadership (Yarinbab and Mezgebu, 2019). High employee turnover rates may endanger efforts to attain organizational objectives. It indicates that turnover is one of the most expensive and difficult workforce challenges facing organizations. Employee turnover can be extremely devastating for any institution (Kemal, 2013).
The turnover rate of academic staff in Ethiopian universities has been increasing from time to time. The effect of employee turnover results in an extra workload on the remaining employees. The increased workload of employees leads to decreased employee morale and increased stress levels, which in turn increases employee absenteeism. Moreover, job satisfaction, job security, salary, allowance/incentive, career growth opportunity, training opportunity, good relationship with employees, and good relationship with the head were found to be the significant factors that affect staff turnover intention (Haileyesus et al., 2019).
Ibrahim Yimer et al. (2017) stated that the intention of turnover of academic staff in Mada Wellabu University was a bad work environment (lack of facilities like offices, chairs, internet, and toilets) was the most frequently cited reason for leaving, followed by poor management and leadership and an inadequate salary. Academic staff who had worked five or more years were 4.5 times more likely to leave their institution. This severe problem is also true at Debre Berhan University. The university's 2021 report also confirmed that 190 academic staff turned out from the university from 2018-2021. The turnover intention has a very significant effect on the organization's cost. There are many negative consequences due to turnover intentions for any organization. Having highly skilled employees leave organizations can have severe consequences, such as impaired organizational functioning, poor service delivery, a lack of systematic administration, or administrative task delays. Losing highly skilled employees may result in an increase in hiring costs and retraining employees (Alubelkassaw Belete, 2018). However, the reason why it was so has not been yet studied. According to Islam (2013), the organizational fit concept identifies confluent goals and recognizes the employee and organization's value as an important aspect of affective commitment. Research explained organizational fit as employee cooperation and willingness towards organization and proposed that requisite disposition to work together. According to Alubelkassaw Belete (2018), the employee selection process should also consider improving congruency between employee values and corporate culture. However, the effect of academic staff's personal values matching or mismatching with the university they are working at has not been studied yet. Furthermore, many other studies also revealed that demographic factors such as age, sex, work experience, level of education, and marital status had their own contribution to turn-over intention, but their findings have not been consistent yet. For instance, gender was not significantly related to turnover intention (Hayes, 2015;Muhangi, 2019). Regarding level of education, as the employees' level of education and marketability are increased turnover intentions (Islam et al., 2013;Stanley et al., 2013;Wren et al., 2014). On the contrary, there was no statistically significant relationship between teachers' level academic qualifications and turnover intentions (Assefa, 2011). Thus, it needs more investigation to add to the generalizability of whether age, sex, marital status, and level of education affect academic staff turnover intention or not. Therefore, given the major research and knowledge gap, this study was aimed to examine the influence of demographic factors and job satisfaction variables on the turnover intention of academic staff at Debre Berhan University, Ethiopia. Thus, the results of this study will have a practical contribution to the university by indicating the determinants of the intention to leave the university. If the study's recommendations are implemented by the university, the academic staff will benefit as well.
The rest of the paper is structured as follows; the next section provides "a literature review" on the factors influencing turnover, the intention of academic staff, and related issues, which supports the study from different countries. The "Methodology" section describes the data and econometric analysis used in the study. The findings and discussion are presented in 'Results and Discussion." The final section offers some "Conclusion and Recommendations."

Literature review
Turnover intention of employees refers to the likelihood of an employee leaving the current job he/ she is doing. It is considered a silent danger that exists in every organization, regardless of its type or its activities. An educational institution is one such organization that comes under threat from this danger when its employees and faculty members are exposed to thoughts of leaving their jobs (Ngamkroeckjoti et al., 2012). The importance of academic staff's retention and the cost of employees' quitting is well known in the literature. Quitting an employee means quitting tacit knowledge and social capital. Turnover increased operation costs and costs for induction and training (Ahmed et al., 2015). The available literature indicated various factors that why employees quit job. This study focuses on demographic and job satisfaction variables, which have been confirmed in literature as major determinants of academic staff turnover intention in different parts of the world. Hayes (2015) stated that employees' age, gender, and level of education are significant factors that influence turnover intentions. Almalki et al. (2012) explain that associations exist between turnover intentions and demographic variables of gender, age, education level, religious affiliation, and level of experience. Dachew et al. (2016) discovered that being a male subject, having more than ten years of teaching experience, job position dissatisfaction with pay and benefits, and dissatisfaction with autonomy and professional opportunities were the factors significantly associated with leaving. Yarinbab and Mezgebu (2019) also stated that sex, employees' feelings of an easy way to get a better job, loyalty to the organization, and the existence of alternative employment were significantly associated with job turnover intention in Mettu University. Minda (2015) found that economic, social, technical, personality, education, and environmental factors were identified in descending order as reasons for teacher turnover at college. Furthermore, in the study, it was found that older and more experienced male trainers were more dissatisfied than their colleagues.

Demographic factors and academic staff turnover intention
There are also some other studies being done to test whether demographic characteristics affect turnover intention or not. Gender was not significantly related to turnover intention (Alubelkassaw Belete, 2018). On the contrary, Hayes (2015) confirmed that gender has a statistically significant effect on turnover intention. Stanley et al. (2013) also found that level of education had a significant influence on turnover intentions. On the contrary, there was no statistically significant relationship between teachers' level of academic qualification with turnover intentions (Assefa, 2011).

Job satisfaction and academic staff turnover intention
Job satisfaction is defined as a pleasurable and emotional state that is the result of an assessment of one's job. It is an effective reason for one's job. It is the general attitude emanating from individual characteristics, factors specifically connected with the job, and relationships among groups outside the horizon of the job (Ahmed et al., 2015). Job satisfaction has a strong negative association with the intention to leave the organization (MacIntosh & Doherty, 2010).
Job satisfaction has a strong negative impact on leaving the company (MacIntosh & Doherty, 2010). Shah et al. (2010) conducted their research on the subject of measuring push, pull, and personal factors that influence fluctuating intent in the case of university teachers in Pakistan. It turns out that all factors (individual, pull, and push) contribute to an employee's intention to leave. The link between rewards and employee retention has been highlighted in research studies (Ahmed et al., 2015), and provides insights into what workers want to do, their words about the rewards, and their feelings about work and reward matters. Gustafson (2002) has stated that compensation and opportunities for better pay largely contribute to employee turnover intention. Feseha (2017) also states that displeasure with salary and pay robustly motivates employee turnover. The lack of different compensation packages like fringe benefits and incentive pay creates an impact on retaining employees or departing from the organization.
Ibrahim Yimer et al. (2017) recommend that pay alone is not sufficient to retain the employees. Ultimately, the workers stay in the organization due to other factors as well, i.e. work environment, co-workers' behaviour, supervisor support, etc., which compels the employee to remain in the organization.
Employees' career advancement is a phenomenon which is formalized, organized, and it is a planned effort to accomplish the balance between the requirements of the organization's workforce and individual career needs. An organization must spend on the development of employees in order to strengthen the bond with employees (Jiang et al., 2012). It creates promotion opportunities within the organization and provides training opportunities and skill development to improve their employees' employability in the internal and/or external labour market (MacIntosh & Doherty, 2010). Employee value match with a job means jobs matching with employees in terms of skill, knowledge, qualifications, ability, and other characteristics of employees that match and are suitable for the job (Ngamkroeckjoti et al., 2012). The sociological driver's previous work showed that the value system centred on the self and family, but now employees see it as a place of belonging. The psychological driver is self-care. In the past, the focus was more on money, but now the prime positive reinforcer is self-care. Personal value match with job is taken into account in the context of employee selection and is based on employee value beliefs that are congruent with the organizations, or person and organization goals (Alubelkassaw Belete, 2018).

Study area profile
Debre Berhan University (DBU) was established in 2007 GC. It is one of the applied Ethiopian higher education institutions located in the city of Debre Berhan, Amhara Region. Debre Berhan city is located in the North Shewa Zone of the Amhara Region, about 120 kilometres northeast of Addis Ababa, the capital city of Ethiopia. The town is located at 9°41′N, 39°32′E and has an elevation of 2,840 meters.

Research design and approach
In this research, both explanatory and descriptive research designs were employed based on a mainly cross-sectional design. The choice of the design is because descriptive design helps to expose and uncover the existing conditions based on the research-formulated research hypothesis. Explanatory research design, on the other hand, helps to determine the impact relationship of various independent variables with the dependent variables and allows the manipulation of one or more independent variables. A quantitative research approach was employed since the aim of the study was to test the effect of certain variables on academic staff turnover intention (Table 2). To analyse the quantitative data, self-administered questionnaires were used.

Sampling method and size
The following sample determination formula of Yamane (1967) as cited in (Fikire, 2021) was used.
Where N is total population; e is the error term (0.05) and n is sample size drawn from the total population n ¼ 1047 ð1 þ ð1047 � 0:0025ÞÞ ¼ 290 A total of 290 academic staff were included using sample respondents were taken from all the nine colleges using a stratified sampling technique. Stratified sampling is often used where there is a great deal of variation within a population. Its purpose is to ensure that every stratum is adequately represented (Taherdoost, 2016). In different colleges in Debre Berhan University, there are slightly different benefits that may be due to allowance, over time, or night work payment. That is why stratification based on college was utilized in the study. Then, a simple random sampling technique was employed to take the sample from each stratum proportionally, as a simple random sampling technique gives equal chance to the elements of the strata (Table 1).

Data collection methods
Data was collected from primary sources. Primary data were collected through self-administered pre-tested questionnaires and interviews of the questionnaire has been carried out, and then, depending on the results, some adjustments will be made to the final version of the questionnaire.

Methods of data analyses
The data was analysed using descriptive and inferential statistics. Descriptive statistics were employed to gain a better understanding of the magnitude of the turnover intention of the academic staff in the study area. Descriptive statistics such as frequency and percentage were employed in the study. Furthermore, the binary logistic regression model was used if the dependent variable is a dichotomous variable, which has a "yes" or "no" outcome. In this study, the dependent variable is the intention to quit, which is a dichotomous variable.

Model specification
In this study, the dependent variable is dichotomous, which takes a value of 1 = if is academic staffs 'have no turnover intention and the value 0 = if yes the academic staffs "have turnover intention. Here, we are interested in estimating the probability that academic staffs" turnover intention household, given the explanatory variables. Since the dependent variable is dichotomous we are interested in binary logistic regression allows us to estimate the relationship between dichotomous dependent variables and dichotomous, interval, and ratio independent variables. Logistic regression uses maximum likelihood estimation to generate estimates. Specifically, maximum likelihood uses the observed data and probability theory to find the most likely or the most probable population value given the sample data. In logistic regression, the formula that is used to determine the population value most likely to yield the sample data is given by the likelihood function: The likelihood function is an expression for the likelihood of observing the pattern of occurrences (y = 1) and non-occurrences (y = 0) of an event in a given sample. In other words, it tells us the probability of getting our sample data from a population with probabilities equal to Pi. The Y i above can be either 1 or 0, depending on the score of persons i on the dependent variable. The P i refers to predicted probabilities on the dependent variable (there is one for every person in the sample) given the person's scores on the independent variables (Pampel, 2020).

Introduction
This section outlines the results of the descriptive and regression analysis of the study. The first part provides the descriptive statistics, while the second part presents a regression analysis, which is an examination of the results based on the equation specified in chapter three. The descriptive analysis made use of tools such as frequency and percentage. Econometric analysis was also used to identify the major demographic and job satisfaction variables influencing academic staff's intention to turnover in Debre Berhan University, Ethiopia. Among 290 questionnaires dispatched to academic staff, only 265 were properly filled out and returned. To this extent, in this study, the sample size calculated was 290 but data was collected from 265 households due to the unwillingness of the selected unit to cooperate and provide some information requested. The rate of response is 91.37 percent. Therefore, this study was analysed by 265 sampled households in the study area. Table 3 shows that 51.7% of respondents were male academic staff, while the remaining sampled academic staff were female. From the sampled staff, about 53.21% were above 30 years old, while the rest of the sampled staff were below 30 years old. The information presented in Table 3 also shows that about 81.13% of the respondents had an academic rank of lecturer or below, while the remaining 18.87% were assistant lecturers and above. Regarding ethnicity, 61.89% are Amhara, and the remaining sampled staff are non-Amhara. Surprisingly, 73.58 sampled academic staff are single and the remaining staff are married. Finally, about 54.34% of respondents had more than five years of working experience.
In terms of intention to quit, respondents were asked if they planned to leave their current job. Table 4 shows that 140 (52.83%) of the 265 respondents intend to leave their current jobs. However, the remaining 125 (47.17%) of them have no intention of quitting their current job. Therefore, the majority of respondents had the intention of quitting their current job.

Econometric analysis
Before going into the basic steps of regression and model interpretation, it is imperative to check the multicollinearity, heteroskedasticity, non-normality, and goodness of fit of the specified model. Therefore, the study tests for multicollinearity for categorical independent variables by using the test of contingency coefficient or correlation coefficient among all categorical explanatory variables. The results of the test show that the coefficient correlation matrix is lower than 0.85. This implies that there is no multicollinearity problem in the specified model. Additionally, the goodness of fit was conducted by using Hosmer-Lemeshow. The result of the test suggested that the model has acceptable measures of fit. Finally, robust regression has been used to control the problem of heteroskedasticity and non-normality. As shown in Table 5, pay and benefits are positive and significant at a 1% level of significance. The estimated odd ratio of academic staff who are satisfied with pay and benefits is 106.048. This implies that the probability of having the intention to quit a job for a satisfied academic staff with his/her pay and benefits is 106.048 times higher than the counterfactual control for the other covariates in the model. Similar findings were also found by other scholars (Haileyesus et al., 2019;Kariuki, 2015) 1 .
The working environment was also positive and significant at the 5% level of significance. The estimated odd ratio of academic staff that are satisfied with the working environment in the university is 95.619. This implies that the probability of having the intention to quit a job for staff that were satisfied with the working environment in the university is 95.619 times higher than for employees who were satisfied with their academic staff, controlling for the other covariates in the model. This study is consistent with (Ibrahim Yimer et al., 2017;Yimer et al., 2017).  Ethnicity was also found to be statistically significant at 1%, and it was confirmed that ethnicity has a statistically significant negative effect on academic staff turnover intention. The estimated odd ratio of academic staff that are satisfied with the working environment in the university is 39.116. This implies that the probability of an academic staff member who had a non-Amhara ethnic background having the intention to quit his or her job is 39.116 times higher than an Amhara academic staff member, controlling for the other covariates in the model. This might be because the study area is found in the Amhara Region of Ethiopia, and ethnic conflict has become common in the country due to the country's political instability. Thus, non-Amhara academic staff have the intent to transfer to a university or any other institution in their own region.

Conclusions and recommendations
The turnover rate of academic staff in Ethiopian universities has been increasing from time to time, particularly at Debre Berhan University. The study revealed that pay and benefits, working environment, and ethnicity were the major factors affecting academic staff's turnover intention. Thus, the study recommended that it is important to improve academic staff's pay, benefits, and working environment by arranging discussion on how the university and Ministry of education on the issues such as making appropriate salary and house allowance increment; providing them with residence houses, health insurance, and credit services for different purposes by collaborating with banks and other financial institutions. Furthermore, to reduce academic staff turnover due to ethnicity, the federal government of Ethiopia needs to ensure political stability by officially avoiding ethnicbased federalism that led millions of Ethiopians to ethnic-led death and migration. The researchers also recommend that further study ought to be conducted by including other factors such as level of income and religion.

Limitations and areas for further research
As this study used cross-sectional data, it is limited to showing the time effect of demographic and job satisfaction variables influencing academic staff's turnover intention. Future research is recommended to examine the impact of academic staff's turnover intention on organizational efficiency by using panel data.