Ethiopian Public Universities Leadership Behavior and Instructor's Job Satisfaction: Evidence From Arba Minch University

The present study examines the link between leadership behavior and instructor's job satisfaction in four purposefully selected colleges and schools of Arba Minch University (AMU). The study participants were 167 randomly selected instructors of sampled colleges and schools. Of these, 149 were male instructors and the remaining18 were female ones. The data for the study were gathered via adapted questionnaire. The study employed quantitative correlational research design to examine the relationship between leadership behavior and instructor's job satisfaction. A Pearson's correlation coefficient (r) value was used to determine the correlation between the independent variable (leadership behavior) and the dependent variable (instructor's job satisfaction) and simple linear regression was employed to predict their relationship. The findings revealed that leadership behavior has statistically significant, positive and strong correlation with instructor's job satisfaction (r=.761, p<.049 two-tailed). The study further indicated that the value of R2 (.580) indicates 58 % of the job satisfaction is explained by leadership behavior. From these findings, it was concluded that there is a positive link between leadership behavior and instructor's job satisfaction. Therefore, it was recommended that through both formal and informal trainings, the university leaders should aware how their leadership behavior is correlated with instructor's job satisfaction and then they should adopt and practice different dimensions of leadership behavior to improve instructor's job satisfaction.


INTRODUCTION 1.1 Background of the Study
Leadership is considered as one of the key ingredients to the success of any organization (Jabbar & Hussein, 2017). In the educational institutions, it is crucial to attain educational goals and objectives (Ali & Dahie, 2015). Since instructional leaders are leading and working together with the key elements of education system like teachers and other stakeholders, they play an irreplaceable role in teachers' job satisfaction (Abbas & Asghar, 2010).
Job satisfaction is very important to improve organizational performance. As studied by Bakotić (2016) highly satisfied employees perform better results than dissatisfied employees. In the higher education institutions (HEIs), instructor's job satisfaction is associated with various organizational variables (Sharma, 2017). This study examines the correlation between leadership behavior and instructor's job satisfaction at AMU.

Problem Statement
Teachers are the most important element of the educational system (Ayalew, 2009). Thus, one of the most important objectives of an educational organization is making the teachers satisfied and fulfilled. Instructional leaders are expected to provide teachers with direction and psychological satisfaction to achieve school goals and objectives (Anyango, 2015). However, the most common weakness among higher education leaders in Ethiopia is their inability to satisfy employees (Aytaged, 2016).
In the literature, there are a number of studies (e.g. Ali & Dahie, 2015;Thomas, 2014) have been conducted to investigate the nexus between principals' leadership styles and teachers' job satisfaction and found positive link between them. Nevertheless, many of these studies were conducted abroad and focused on principals' leadership style and teachers' job satisfaction at primary and secondary school level.
In Ethiopia, a considerable amount of studies have been conducted on principals' leadership styles and their effects on teachers' performance (e.g., Ayene, 2016); teachers' job satisfaction and commitment (e.g., Teferi, Bekalu & Abebe,2016) at secondary and primary schools. Thus, little attention has been given to Ethiopian higher education leadership behavior and instructor's job satisfaction. This study is; therefore, intended to fill this research gaps in examining the link between leadership behavior and instructor's job satisfaction at AMU. The study seeks to answer the following three basic research questions.

Basic Research Questions
1. What is the relationship between leadership behavior and instructor's job satisfaction? 2. Which dimensions of leadership behavior is highly correlated with instructor's job satisfaction?
3. To what extent do leadership behavior affects instructor's job satisfaction?

Objectives of the Study
The main purpose of this study is to examine the nexus between leadership behavior and instructor's job satisfaction and the specific objectives are to address the basic research questions that included in the study.

RESEARCH DESIGN AND METHODOLOGY 2.1 Research Design
In this study, the quantitative correlational design was employed to examine the link between leadership behavior and instructor's job satisfaction as suggested by Gay &Mills (2012).

Population and Sampling Techniques
In the study, purposive and simple random sampling techniques were used. Among five colleges, two institutes and three schools of Arba Minch University, I purposefully selected two colleges and two schools, namely, College of Social Sciences and Humanities and Business and Economics, School of Pedagogical and Behavioral Sciences and Law as thought of Creswell (2012). Accordingly, from the total of 287 local instructors in the sampled colleges and schools, 167 instructors consisting of 149 male and 18 female were selected via simple random sampling technique at confidence interval of 95% according to the Yamane's (1967)  =167. Then, proportionate stratified random sampling technique was employed to take representative sample instructors from each department and sex.

Data Gathering Instruments
In the study, two sets of standardized questionnaire were adapted to survey leadership behavior and instructor's job satisfaction. The Leadership Behavior Description Questionnaire (LBDQ)-Form XII self (1962), developed by staff members of the Ohio State Leadership Studies consisting of one hundred (100) items was used to survey leadership behavior. On the other hand, forty five (45) items of Academic Job Satisfaction Questionnaire (AJSQ) designed by Al-Rubaish et al. (2014) was employed to assess instructor's job satisfaction. The response options for both LBDQ -Form XII self (1962) was gauged in a five point Likert scale that ranging from 5= always to 1= never. While AJSQ's response options were designed by using 5 points Likert scales ranging from 5=strongly agree to 1=strongly disagree. Before formal dissemination of the questionnaire, the instrument's reliability and validity was checked. To check the face validity of the questionnaire, I invited two psychology department staff members from Wolaita Sodo University who believed have ample experience in Educational Measurement and Evaluation. They reviewed the face validity of the questionnaire separately and jointly and reported the questionnaire as valid. Moreover, to check whether the questionnaire is reliable, I conducted pilot study on forty three (43) non-sampled teachers at Arba Minch Institute of Technology (AMIT). Then, the reliability of the twelve domains of the LBDQ and eight domains of the AJSQ including total reliability indices were computed at Cronbach's alpha level 0.5. The reliability results of LBDQ and AJSQ were judged according to George & Mallery (2003)

Data Collection Procedure
Permission to conduct the data was sought from the Dean of purposefully selected colleges and schools of Arba Minch University. Before administering tools for data collection, the study target groups were introduced about the purpose of the study and their informed consents were secured. Consequently, questionnaires were administered to randomly selected instructors in their offices and collected by me and additional facilitators of the study.

Data Analysis Method
The data was analyzed quantitatively using SPSS v.20. Both descriptive and inferential statistics was employed. Descriptive statistics such as frequency and percentage was used to describe respondents' demographics. On the other hand, Pearson's correlation coefficient (r) was employed to examine the link between overall leadership behavior and instructor's job satisfaction and to identify leadership variables that highly correlated with instructor's job satisfaction. The strength and direction of correlation coefficient or (r) value was judged according to Gay's & Mills's (2012) range: between +0.35 and -0.35 = weak or none, between +0.35 and +0.65 or between -0.35 and -0.65= moderate, between +0.65 and +1.00 or between -1.00 and -0.65= strong. Finally, simple linear regression analysis was employed to predict the relation between leadership behavior and instructor's job satisfaction.

Respondents' Demographics
In the first part of questionnaire, the respondents were asked about their general background information. The following  Concerning age composition, majority of them were in between 31-35 years old 93 (55.6%) followed by 26-30 years 59 (35.3%) and few of them were in between 36-40 years old 15 (8.9%). This finding reveals that many study participants were adults and matured enough to fill the questionnaire. With regard to study participants' work experience, many respondents were served 6-10 years 55 (32.9%), 11-15 years 50 (29.9%), and 16-20 years 42 (25.1 %) and few of them were served 21-25 years 20 (11.9%). This indicates that the study subjects have good work experiences to fill the questionnaire based on their past work experience. The results from the above table 3.1.1b show the respondents' demographics by educational background, academic rank and current work position. From the total of one hundred sixty seven (167) respondents, 159 (95.2 %) were M.A holders and the remaining few 8 (4.7%) were PhD holders. Hence, majority of them 159 (95.2%) were lecturers and few of them 8(4.7%) were assistant professors. Concerning the respondents' current work position, all of them 167 (100%) were instructors. This information reveals that the respondents have good educational background, academic rank and work position to respond the administered questionnaire in a proper manner.

Correlation Between Overall Leadership Behavior and Job Satisfaction
This part is intended to answer the first research question that describes the major purpose of this study. The following table summarizes correlation coefficient (r) value of overall leadership behavior and instructor's job  .05(2-tailed) The above table 3.2.1 shows the correlation between an overall leadership behavior and instructor's job satisfaction. According to the above table, there is statistically significant, strong and positive correlation between leadership behavior and instructor's job satisfaction (r=.761, p<.049). This finding is consistent with many other researchers' findings, for instance, Ali & Dahie (2015) and Thomas (2014).

Correlation Between Variables of Leadership Behavior and Overall Job Satisfaction
The intention of this part is to answer the second research question. The following table on page 10 shows the correlation between variables of leadership behavior and overall job satisfaction. Table 3.

Correlation Between Facets of Leadership Behavior and Overall Job Satisfaction (N=167)
Note: Correlation is significant at *0.05 (2-tailed) The above table 3.3.1 depicts the correlation between dimensions of leadership behaviors and instructor's job satisfaction. As we can see from the table, all leadership behavior variables have a positive correlation with job satisfaction. Nevertheless, representation (r=.762, p<.009), demand reconciliation (r=.951, p<.000), tolerance of uncertainty (r=.696, p<.000), persuasiveness (r=.698, p<.001), consideration (r=.958, p<.000), production emphasis (r=.760, p<.006), predictive accuracy (r=.871, p<.030 ), integration (r=.870, p<.029) and superior orientation (r=.667, p<.000) have strong and positive correlation with instructor's job satisfaction than other variables. On the other hand, initiation of structure (r=.582, p<.006), tolerance and freedom (r=.634, p<.023) and role assumption (r=.635, p<.023) were moderately associated with job satisfaction. The results suggest that if leaders do practice different leadership behaviors then the job satisfaction of teachers will increase, because leadership dimensions have strong, moderate and positive relationship with job satisfaction of teachers.

Simple Linear Regression Analysis
In this part, simple linear regression analysis was computed to answer the third basic research question. The following information shows the result of simple linear regression analysis. The model summary in Table 3.4.1 shows that the coefficient of determination (R 2 ) was 0.580. This means that leadership behavior explained 58% percent of the variations in instructor's job satisfaction. In other words, 42 percent of the variation in instructors' job satisfaction cannot be explained by leadership behavior. So, there must be other factors that are not incorporated in the model to explain job satisfaction of the instructor.

CONCLUSIONS AND IMPLICATIONS 4.1 Conclusions
Based on results and discussions of data, the following conclusions remarks made: As the finding of the overall leadership behavior and job satisfaction showed that there is a positive link between leadership behavior and instructor's job satisfaction. Furthermore, among twelve dimensions of leadership behaviors, nine variables are strongly and positively correlated with job satisfaction and the remaining three variables are moderately associated with job satisfaction. On the other hand, the independent variable (leadership behavior) makes strong and unique contribution to explaining dependent variable (job satisfaction) as the findings from simple regression analysis indicate.

Implications For AMU
There is a statistically positive link between leadership behavior and instructor's job satisfaction according to the findings of this study indicates. Hence, through both formal and informal trainings, the university leaders should aware how their leadership behavior is correlated with instructor's job satisfaction and then they should adopt and practice different dimensions of leadership behavior to improve teacher's job satisfaction.

Implications For Further Research
This study was confined to the purposefully selected colleges and schools from Arba Minch University and the findings may not be generalizable to other Ethiopian public universities. Thus, further studies can focus on the link between leadership behavior and instructor's job satisfaction by taking a representative sample from Ethiopian Public Universities.

ACKNOWLEDGEMENT
I am grateful to Professor Ayalew Shibeshi, Department of Educational Planning and Management, Addis Ababa University, for sharing all his endeavors and scholastic stamina in teaching two postgraduate courses which laid