Figure 1. A model of subjective quality of life homeostasis (adopted from Mellor et al., 2003)
Abstract: The aim of this study was to examine factors that affect the subjective wellbeing (SWB) of university teachers, at Mizan – Tepi University (MTU), Ethiopia. Accordingly, the study employed the homeostasis theory of SWB as a theoretical framework and tested two major hypotheses. First, it was hypothesized that the homeostatic model of SWB (extraversion, emotional stability, experiential input, perceived control, self-esteem and optimism) would significantly predict the SWB of teachers. The second hypothesis was that Job- and organization-related factors (job satisfaction and organizational culture), as additional aspects of experiential input, would contribute significant unique variance in SWB above homeostatic model predictors. A sample of 162 teachers participated in this study by responding to questionnaires that included measures of the abovementioned variables. Hierarchical linear regression was used to test the hypotheses. According to the results, the homeostatic model of SWB was a significant predictor of teachers’ SWB. In this case, perceived control and optimism made unique contributions while extraversion, emotional stability and self-esteem did not make unique contributions but shared variance with one another. Besides, Job- and organization-related factors were found to be significant predictors of teachers’ SWB. Both job satisfaction and organizational culture contributed significant unique variance above the homeostatic model predictors. Based on the results of the study, it can be concluded that job- and organization- related factors join forces with homeostatic model factors in contributing to or affecting the SWB of teachers.
Keywords: Homeostatic Model; Job Satisfaction; Organizational Culture; Subjective Wellbeing; University Teachers
It is crystal clear that proper education is a major instrument to bring about development in a country. Understanding this, the current government of Ethiopia has been making effort to expand its education sector. Within the education sector, higher education is one of the most emphasized areas by the government, particularly in recent years. Accordingly, the number of public Universities in Ethiopia has reached 34 (MoE, 2013). In addition, some 11 new universities were about to be opened in different parts of the country, during the Second Growth and Transformation Plan period (FBC, January 13, 2015). There are also more than 60 non-governmental/ private higher education institutions in the country (Tefera, 2015). This expansion of the higher education system in Ethiopia is also accompanied by high demand for and employment of university teachers, making university teachers crucial parts of the higher education system of Ethiopia.
1 Corresponding author : michael.emru7@gmail.com
University teachers' role in producing competent graduates, research, and community service is well understood in Ethiopia. Evidence for this can be found in the National Education Policy of Ethiopia (Federal Democratic Republic of Ethiopia (FDRE), 1994); the Higher Education Proclamation (FDRE, 2003); the Growth and Transformation Plan I (FDRE, 2010); and the Growth and Transformation Plan II (FDRE, n.d.). Nevertheless, it seems that enough attention has not been given to university teachers in Ethiopia, especially in relation to their wellbeing/ subjective wellbeing (SWB). One indicator for this could be the absence of studies regarding the SWB of university teachers in the country. Therefore, given university teachers are instrumental for the development and effectiveness of higher education in Ethiopia, the present study investigated factors that affect their SWB.
The concept of SWB should be seen and discussed as a dimension or an aspect of the broader concept of Wellbeing. Defining wellbeing has been difficult and controversial for many. This resulted from the multidimensional nature of the concept which covers physical, mental, social, material, and spiritual aspects of people's lives. Another reason could be the fact that different disciplines (such as philosophy, psychology, sociology, economics and social policy) and even different scholars within a single discipline define it differently. Nevertheless, all may agree that simply defined, wellbeing could mean 'good life' or 'quality life'. Yet, the difficult question that remains is: "what do we mean by 'good life' or 'quality life'?" In this case, wellbeing or 'good life' can be conceptualized by identifying its different dimensions. Accordingly, the concept of wellbeing could be classified into two broad categories, i.e., objective wellbeing and SWB (Camfield et al., 2008; CDC n.d.; Kobau et al., 2010; Schwannauer et al., 2013).
Objective wellbeing is about the presence of external factors (material and financial factors like housing and income) that are essential for a person's life. Indicators of objective wellbeing could involve health status, employment status, educational attainment, housing status, etc (Schwannauer et al., 2013).
SWB is regarding people's subjective perception and evaluation of their feelings (negative and positive affect) and different aspects of their lives. Indicators of SWB include feeling, experience, evaluation of life, etc (Camfield et al., 2008; Schwannauer et al., 2013). The concept of SWB is also further classified by different authors differently. Some authors (e.g., Ryff, 1989; Ryff & Singer, 2008), following ancient Greece philosophers' distinct traditions, argue that SWB has two types or traditions, i.e., hedonism (including affect and life evaluation/satisfaction) and eudemonism (psychological wellbeing). Other authors classify SWB into three categories as: life evaluations/satisfaction, affect (hedonia), and eudaimonia (psychological wellbeing). In this case, some authors (Deci & Ryan, 2006; O'Donnell et al., 2014) assert that life evaluation/ satisfaction is not a hedonic concept since it has cognitive dimensions (unlike hedonia which is primarily emotional in nature).
The hedonism tradition (associated with the work of Aristippus of Cyrene) perceives wellbeing as happiness or pleasure. Here, wellbeing or happiness is considered to be an end rather than a process. In this case, the source of happiness is not given attention (whether it is right or wrong/moral or immoral). Hedonism has influenced other philosophies and theories like the utilitarian philosophy (e.g., Bentham). The eudaimonism perspective (which is Aristotelian tradition), on the other hand, defines wellbeing in relation to virtue, meaning, goal-directedness, purpose, or self-realization. In other words, eudaimonia is about knowing oneself (one’s strengths and limitations) and choosing oneself or becoming oneself, within the context of the challenges of life. Here, wellbeing or eudaimonia is considered as more of a process than an end or outcome. It is asserted that having positive affect and satisfaction with life does not guarantee psychological wellbeing, unless the source of happiness is eudaimonia. It is believed that leading eudaimonic life ultimately results in happiness. Eudaimonism is also evident in current-day theories like humanistic psychology (e.g., Maslow's concept of self-actualization) and Ryff's work of the dimensions of psychological wellbeing (Deci & Ryan, 2006; Garcia, 2006; Olsson et al., 2012; Ryff & Singer, 2008; Watermana et al., 2010).
Hedonism and eudaimonism have their foundations on different understandings of human nature. For hedonism, people are born tabula rasa and then influenced by their environment. On the contrary, for eudaimonism, people are born with contents and their job should be identifying and inculcating that content - self-realization (Deci & Ryan, 2006).
Another explanation for SWB is that of the set-point theories of wellbeing. The underlying idea of this approach is that every person has a genetic- and personality- based or determined set-point of wellbeing (happiness or life satisfaction). In this case, it is argued that negative life events (e.g., as loss of a job, serious injury) and positive life events (e.g., higher income, getting married) may decrease or increase one's level of happiness for the time being. But, the person's level of happiness will return to the initial set-point, after some time, as a result of adaption. This means, both improvements and deteriorations in our lives have short-term impact on our SWB because we adapt to changes quickly (Easterlin, 2003).
One variety in the set-point approach to SWB, which is emphasized in this study because of its comprehensiveness, is called the Homeostasis Theory. For this theory, every individual, irrespective of age, has a genetically determined level of SWB, in the positive direction. The set-point of SWB (life satisfaction as measured by Personal Wellbeing Index (PWI) is hypothesized to have an average of 75% points. Like bodily functions (e.g., body temperature) are maintained, this set-point is protected (maintained and defended) by homeostasis or stabilizing forces such as adaptation, positive affectivity, and a system of cognitive protections (optimism, perceived control and self-esteem). This theory acknowledges that significant changes of life events, that are stronger than stabilizing forces can affect SWB (Cummins 1995, 2010, cited in Tomyn, et al., 2011; Cummins 2002, cited in Thomas n.d.; Tomyn & Cummins 2010).
In this study, efforts were made to test the homeostasis theory and its model by considering job- and organization- related factors (job satisfaction and organizational culture) as significant life events that can affect the SWB of teachers.
Job satisfaction can be understood as pleasurable or positive emotional state of employees resulting from the evaluation of their job or job experience (Lock, 1976). Job satisfaction is one of the most widely discussed topics by scholars concerned with organizational studies. This topic has been studied as cause, correlate and consequence of various work- and nonwork- related variables (Bowling & Hammond, 2008). In this case, its effect on employees' wellbeing/SWB has also been noted (Dolbier et al., 2005; Lai & Cummins 2013).
A related concept to job satisfaction is organizational culture. Organizational culture can be understood as systems of value and assumptions that guide the way an organization runs its business (Schhneider & Reicher, 1983). Organizational culture is what differentiates one organization from another and what makes employees be attracted to one organization instead of another (Smith, 2003). It could also determine degree of creativity and innovation in an organization (Tesluk et al. 1997). It is believed that organizational culture influences job satisfaction, behaviors, and attitudes of employees (Hebb, 1949; Morse, 1953).
Based on the above arguments, the general objective of this study was to examine factors that affect the SWB of university teachers at MTU, Ethiopia. Accordingly, the study primarily focused on examining the homeostatic model and job- and organization- related factors as predictors of teachers' SWB. Specific research questions and hypotheses of the study are presented below:
In line with the research questions above, the following two hypotheses are proposed
Hypothesis 1. The homeostatic model of SWB significantly predicts the SWB of teachers of MTU.
The homeostatic model of SWB (the original one) asserts that SWB of people can be predicted by the interaction of experiential input (positive or negative life experiences or major life event); personality factors (extraversion and neuroticism/emotional stability); and cognitive buffer factors (perceived control, self-esteem, and optimism) (Mellor, Cummins, Karlinski & Storer, 2003). Here, it is important to note that this model has been revised and in this case personality factors have been replaced by HPMood (Core Affect). Explanation for not using this revised model is provided in the 'Limitation' section of this paper.
Figure 1. A model of subjective quality of life homeostasis (adopted
from Mellor et al., 2003)
Hypothesis 2. Controlling for predictors of the homeostatic model of SWB, job- and organization- related factors (job satisfaction and organizational culture) significantly predict the SWB of teachers of MTU.
Job- and organization- related factors (job satisfaction and organizational culture) could be considered as additional aspects of experiential inputs (besides major life events) within the homeostatic model of SWB. And, in this study, it is hypothesized that they would add to the variance accounted for by the homeostatic model predictors of SWB.
This study used a cross-sectional survey design. Data for this study were collected from November 2015 up to February 2016.
This study was conducted in Mizan – Tepi University. Mizan – Tepi University is one of the 34 public universities in Ethiopia. According to MTU’s official website, Mizan – Tepi University was established in the year 2006 in Mizan - Aman and Tepi towns of the Southern Nations, Nationalities, and People's Region of Ethiopia. The university has two campuses, i.e., Mizan campus and Tepi campus. Each campus has different colleges and departments. The Mizan campus includes college of business and economics (CBE), college of agriculture and natural resources (CANR), college of humanities and social sciences (CHSS), college of health science (CHS) and school of law (SL). The Tepi campus also has college of engineering and technology (CET), college of natural and computational sciences (CNCS), and school of computing and informatics (SCI) (MTU, 2016).
Currently, according to MTU's human resource office, Mizan – Tepi University had some 1065 academic staff members (personal communication (16/02/2015) and a total of 8, 802 regular students (MTU, 2016). It is also noted that some of the academic staff members of the university are pursuing their second and third degrees, being sponsored by the university. According to MTU's human resource office, there were 214 (males = 195 and females = 19) teachers on study leave (personal communication, 16/02/2015). The proportion of female teachers in the university is around 11% (MTU, 2016).
The population of this study was all Ethiopian teachers of MTU who had first degree and above educational qualification and who were teaching at the university at the time of data collection. Those who were studying their second and third degree outside of MTU were excluded from the study because the study emphasized the role of job-and organization related factors on subjective wellbeing. The total number of these teachers, according to MTU's Human Resource Office, was 617 (Personal communication, 16/02/2015).
This study was mainly concerned with investigating factors that affect the SWB of University teachers using hierarchical multiple regressions. Cohen (1992), a frequently cited author in the field of power analysis, asserts that sample size for studies which test hypotheses (like multiple regressions) should be determined on the basis of power analysis. Doing so enables researchers to acquire sufficient sample size and to minimize statistical errors. He also argues that though power analysis is very crucial in statistical studies, it is "neglected" mainly because of "the inaccessibility of a meager and mathematically difficult literature" (p. 155). In order to deal with this problem, Cohen (1992) presents a handbook that has a calculated sample size table indicating sufficient sample size for different statistical tests, by specifying particular values for a (.01, .05 and .10), power (.80), and effect size (small, medium and large). Besides, in the case of multiple regressions, the number of independent variables is taken into consideration.
The sample size for this study was determined based on Cohen's (1992) advice and expectation of some dropout/attrition rate. Hence, the following specifications were made in this study. The sufficient/minimum sample size required for this study was known by specifying a two-tailed test with the conventional a (.05), power (.80), effect size for multiple regression (medium, .15), and number of independent variables (8). Accordingly, the minimum sample size required for the study was 107. In the case of attrition rate, since no previous study was available, the researcher subjectively expected a 50% attrition rate. Hence, the final sample size for this study was determined by computing the expected attrition rate (.5) with that of the minimum sample size determined by the power analysis (107), using the following formula: 107/ (1 - .50) = 214. This means, if 50% of the final sample size (214) dropped out or failed to give adequate data for the study, 107 participants would remain in the study providing sufficient amount of data.
Study participants were selected using proportionate stratified sampling technique from eight colleges in the University. The researcher assumed that these colleges could have unique characteristics as they were found in different campuses located in different towns. Accordingly, the eight colleges/schools were considered as strata from which respondents were selected randomly. Out of the planned 214 participants, data is obtained from 162 of them. This represented 75.7% return rate.
Among the 162 respondents, while 146 (90.1%) were males, the rest 16 (9.9%) were females. In terms of academic rank, 35 (21.9%) were Graduate Assistants & Assistant Lecturers, 122 (76.3%) were lecturers, and 3 (1.9%) were Assistant Professors. Two participants did not report their rank. The age range of the participants was between 21 and 48 years with mean age of 28.24. In terms of years of teaching experience the range is between one month and nine years with mean of 3.79 years. Regarding the participants marital status, 54 (34%) are married, 104 (65.4%) are single, 1 (0.6%) is separated and the remaining 3 (1.8%) did not report.
The main source of data for this study was primary data. Accordingly, Self-administered questionnaire was used as a tool of data collection. This questionnaire had both close- and open-ended forms. It included, among others, questions on background information of respondents and measures of SWB, personality, internal buffers, major life events, job satisfaction, and organizational culture. The questionnaire was prepared and administered in the English language. Before the main study was conducted, the questionnaire and the measures in it were Pilot tested on 54 teachers found in other colleges outside of MTU (Aman Health Science College and Mizan Agriculture College). Overall, the questionnaire and the measures were found to be good. In addition, based on the pilot test, improvements were also made on the questionnaire.
The major focus of this study was examining factors that affect the SWB of university teachers. In this case, there were one dependent variable and eight independent variables.
SWB could be defined as "a normally positive state of mind that involves the whole life experience" (Tomyn et al., 2011a, para. 2). In this study, teachers' SWB was measured using a scale called the Personal Well-being Index-Adult (PWI-A) originated in Australia. The scale has eight items of satisfaction or domains of quality of life that include standard of living, health, achieving in life, relationships, safety, community-connectedness, future security and spirituality. The eight domains are thought to be theoretically embedded, as representing the first level deconstruction of the global question: ‘How satisfied are you with your life as a whole?' (International Wellbeing Group (IWbG), 2013).
The PWI-A asks respondents how satisfied they are with the different domains of their quality of life on an 11- point satisfaction scale ranging from "no satisfaction at all" (0) to "completely satisfied" (10). The PWI-A can be scored and analyzed by summing and averaging the scores of all the eight domains to get a score of SWB or PWI-A. (All the other scales in this study were scored and analyzed the same way). In this case, some standardization of the scores should be performed by converting the scores that ranged from 0-10 into scores that range from 0-100. This can be done easily by bringing the decimal point to the right. For example, a score of 6 is converted into 60 points and a mean score of 6.95 is changed into 69.5 points (IWbG, 2013). This kind of standardization was also used for all of the independent variables discussed in the subsequent sub-sections so that direct comparisons can be made.
The scale has demonstrated good psychometric properties. The reliability record of the scale was good. According to Lau and Cummins (2005) cited in IWbG (2013), a Cronbach's alpha of 0.70 – 0.85 was found in Australia and other countries. In addition, a moderate interdomain correlation (0.30 - 0.55) as well as item-total correlation (0.50) were found. Testretest reliability (1-2 week interval) was also found to be satisfactory (0.84 intra-class correlation coefficient). In this study a Cronbach's alpha of .85 was found for this scale.
Regarding construct validity, 40 – 60% of the variance in ‘Satisfaction with Life as a Whole' was explained by the combination of both unique and shared variance of the domains. Besides, the domains were found to have formed a single stable factor consistently, accounting for about 50% of the variance, in Australia and other countries (IWbG, 2013). This was also confirmed in this study.
When it comes to convergent validity, this scale was found to have a strong correlation with satisfaction with life scale. In this case, a correlation of 0.78 was found (Thomas, 2005, cited in IWbG, 2013).
Personality. Personality could be defined as stable patterns of thoughts, attitudes and behavior that, with the exception of minor changes over time, are relatively stable throughout adulthood (Costa & McCrae, 1992). This study focused on two personality dimensions, i.e., extraversion and neuroticism (emotional stability), as prescribed in the homeostatic model of SWB. Each dimension was measured using a positively-worded single item from the Ten Item Personality Inventory (Gosling, Rentfrow and Swann, 2003) which was designed to measure the Big Five personality dimensions. In the case of extraversion, the item used was "I see myself as extraverted and enthusiastic." And, for emotional stability, the item selected was: "I see myself as calm and emotionally stable." This measure was adapted to an 11 point end-defined scale ranging from "strongly disagree" (0) to "strongly agree" (10).
Cognitive Buffers. There are three cognitive/internal buffer factors according to the Homeostatic Model of SWB, i.e., perceived control, self-esteem and optimism. All these variables were measured on separate scales with an 11 point end-defined scale ranging from 'Disagree Completely' (0) to 'Agree Completely' (10).
Perceived control. Control could be conceptually understood as an individual's overall belief that he/she can influence important outcomes in his/her life (Wallston, 2005). Control was measured in this study by using the Perceived Control Scale (Chambers et al., 2003). The scale has three dimensions, i.e., primary, secondary and relinquished control, each dimension having three items. For the purpose of this study, only items of primary and secondary control were used. According to Davern (2005) cited in O'Connor (2005) and O'Connor (2005) it is advisable to omit the relinquished control subscale as it did not load onto the perceived control factor. Participants were told to rate how much they agree or disagree with statements about their way of coping when something bad happens to them. For instance, the item "I ask others for help or advice" indicates primary control, while the item "I remember that the situation will improve if I am patient" shows secondary control. In this study, all the six items were used together to form the perceived control scale and a Cronbach's alpha of .77 was found.
Self-esteem. Self-esteem can be conceptually defined as a favorable or unfavorable attitude toward the self (Rosenberg, 1965, 1989). Self-esteem was measured using the Rosenberg Self-Esteem Scale (RSES) (Rosenberg, 1979). Though the scale has ten items, only five of the positively-worded items were used in this study. Previous studies based on five-point scales demonstrated reliabilities that ranged from .72 to .90 for adult samples (Gary-Little, Williams & Hancock, 1997; Robins, Henden & Trzesniewski, 2001). In this study the scale was adopted to an 11- point end-defined scale. Participants were asked to rate how much they agree or disagree with statements such as "I take a positive attitude toward myself". This study also found a Cronbach's alpha of .79 for the scale.
Optimism. Optimism could be conceptualized as an individual's expectation that the future will be advantageous or beneficial, and a belief that contingencies in life can be successfully dealt with (Peterson, 2000). Optimism was measured using the Life Orientation Test-Revised (LOT-R). Three positively-worded items from the LOT-R scale (Scheier, Carver, & Bridges, 1994) were used in this study. Example of items include: "In uncertain times I usually expect the best". In this case, participants rated their level of agreement or disagreement on an 11point end defined scale. A Cronbach's alpha of .68 was found in this study for the scale.
Experiential input Life event (LE). Experiential input life event (LE) was measured in this study by using the item: "Has anything happened to you recently causing you to feel happier or sadder than normal?" This item was rated using an 11 point end-defined scale ranging from 'Yes Happier' (0) to 'Yes, Sadder' (10). Reverse scoring was done in this case.
Job satisfaction. Job satisfaction could be defined as employees' pleasurable or positive emotional state resulting from the evaluation of their job or job experience (Lock, 1976). Job satisfaction in this study was measured using the three-item Job Satisfaction Subscale (JSS) of the Michigan Organizational Assessment Questionnaire (MOAQ) (Cammann, Fichman, Jenkins, & Klesh, 1983). Though the original version of the MOAQJSS used a 7-point agree–disagree scale, for the purpose of this study, an 11 point enddefined scale ranging from "Disagree Completely" (0) to "Agree Completely" (10) was used. An example of the items include: "In general, I don't like my job". This item was the only reverse-scored item of the scale. In this study, a Cronbach's alpha of .66 was found for the scale.
Organizational Culture. Organizational culture could be defined as systems of value and assumptions that guide the way an organization runs its business (Schhneider & Reicher, 1983). Organizational culture was measured in this study using 21 items that have been adapted from Yang (2003) by Sabri, Ilyas and Amjad (2011), in Pakistan. In this case, organizational culture has two components i.e. organizational culture related to managers and leaders (OCM), with 16 items and organizational culture related to employees (OCE), with five items. An example of OCM items is: "My organization supports employees who take calculated risks." An example of OCE is: "In my organization, people give open and honest feedback to each other". For the purpose of this study, all the 21 items were used to get the organizational culture score. In addition, this scale was adapted to an 11-point end-defined scale ranging from "Disagree Completely" (0) to "Agree Completely". A Cronbach's alpha of .96 was found for the scale in this study.
After the data were collected through questionnaires, data cleaning and management was carried out. Every questionnaire was checked for problems such as inappropriate responses, response sets and no response. Such problems were found in some of the questionnaires and treated as missing values. The next step was data entry into SPSS Version 20 and checking that the data were properly entered. In this case, descriptive statistics were used to see the plausibility of means, standard deviations and ranges of values.
Then, the data were checked using SPSS for meeting/violating fundamental assumptions relevant for regression analysis including miss-specification, normality, outliers, multicollinearity, independent errors, homoscedasticity, and linearity. Other assumptions such as random samples and interval/ratio level of variables were also met. One of the eight independent variables, i.e., major life event/ experiential input (single item), was excluded from the model after preliminary analysis showed that its T-statistic was not significant at the 10% level. This might be attributed to error in measurement (Gupta, 1999). The researcher is convinced that the way the variable was rated confused many of the respondents.
After checking the assumptions, descriptive statistics were used to summarize the samples characteristics and results on different variables. Then, estimation of teachers' level of SWB was done. Both point and interval estimations were done at 95% confidence level. In this case, all the eight domains of SWB as measured by the PWI-A were included. Finally, hierarchical linear regressions were carried out to examine the contribution of different factors (successive regression models) in predicting the SWB of teachers at MTU.
In this research, the researcher made sure that the ethics of professional research was maintained. In this case, the following essential ethical considerations were emphasized: informed consent and permission, anonymity of participants, confidentiality of information given by participants, and avoidance of possible harm to participants as a result of participating in the study. Permission and consent to carry out this research was sought at different levels. Firstly, MTU's Research and Community Development Support Directorate gave its permission to conduct the study after reviewing the proposal of the study. Secondly, each respondent of the study gave his/her informed consent during data collection.
In this section, results of the study concerned with determining the SWB level of teachers at MTU are presented. In this case, both mean point and interval estimations were done at 95% confidence level for the variable SWB (as measured by PWI-A) and its eight domains, i.e., satisfaction with standard of living, satisfaction with health, satisfaction with achievement in life, satisfaction with personal relationship, satisfaction with safety feeling, satisfaction with community connectedness, satisfaction with future security, and satisfaction with spirituality/religion. In addition, some normative ranges of the PWI-A that are calculated from survey mean scores (30 surveys conducted between April 2001 and August 2013) (Australian Center on Quality of Life, 2016) are provided here for the purpose of comparison.
Table 1 Point and Interval Estimation of Teachers' Level of SWB (Mean Points
of PWI-A) (N=146)
As indicated in Table 1, teachers of MTU had a low level (mean points) of SWB as measured by the PWI-A, i.e., 61.66 (95% CI: 58.57 to 64.74). In addition, their mean points on most of the different domains (five of the eight) of the scale were very low. These domains include: standard of living = 45.68 (95% CI: 41.43 to 49.94); future security = 49.66 (95% CI: 44.94 to 54.38); achieving in life = 54.86 (95% CI: 50.51 to 59.21); community connectedness = 55.61(95% CI: 50.72 to 60.51); and safety = 58.08 (95% CI: 53.45 to 62.71).
Yet, high mean points were also observed on the following three other domains: spirituality/religion =78.84 (95% CI: 74.70 to 82.97); health = 74.93 (95% CI: 70.05 to 79.81); and personal relationships = 72.19 (95% CI: 67.87 to 76.51). Here it is important to note that teachers' mean point on spirituality/religiosity is the highest and even greater than the normative range. Besides, their score on health is comparable with that of the normative range while their scores on the rest of the domains are lower than the corresponding normative ranges.
Predictors of Teachers' SWB: Homeostatic Model Factors and Job- and Organization- Related Factors
This study had two major hypotheses. The first was concerning the role of homeostatic model of SWB as a predictor of SWB of teachers at MTU. And, the second was about the role of job- and organization- related factors in predicting the SWB of teachers at MTU.
The homeostatic model of SWB has five predictor variables which include: major/recent life event; extraversion; neuroticism/emotional stability; perceived control; self-esteem; and optimism (Mellor, Cummins, Karlinski & Storer, 2003). In this study, as discussed earlier, there was no significant correlation between the dependent variable (SWB) and one of the independent variables, i.e., major/recent life event (p = .858). Therefore, this variable was removed from further analysis.
Job- and organization- related factors also include two variables, i.e., job satisfaction and organizational culture. Descriptive statistics, bivariate correlation, and hierarchical regressions results are provided in this section.
Table 2 Intercorrelations, Means, and Standard Deviations among Variables in
the Study
*p<.05, **p<.01, p<.001 Ex=Extraversion, ES=Emotional Stability, SE=Self
Esteem, OP=Optimism, PC=Personal Control, JS=Job satisfaction, OC=Organizational
Culture
Table 2 shows means and standard deviations for each variable in the models. Accordingly, teachers’ level of SWB was found to be low (M= 61.66, SD = 18.8).Teachers’ scores on organizational culture (M= 40.59, SD= 20.24) and job satisfaction (M= 59.37, SD= 24.79) were also found to be low. Scores on extraversion (M= 60.19, SD = 26.54) was low, too. Scores on the rest of the homeostatic predictors were higher: emotional stability (M=65.66, SD= 28.02); perceived control (M= 73.90, SD= 16.64); optimism (M= 77.13, SD= 19.14); and self-esteem (M= 82.05, SD= 15.62).
Table 2 also presents bivariate correlations between variables of the study, with significance levels. As expected, the dependent variable, i.e., SWB (PWI-A) had significant positive correlations with all the seven independent variables. These correlations, in their order of strength, are with: job satisfaction (r= .494, P< .001); organizational culture (r= .478, P< .001); perceived control (r= .464, P< .001); Self-esteem (r= .429, P< .001); optimism (r= .426, P< .001); emotional Stability (r= .287, P< .001); and extraversion (r= .194, p< .05).
Other interesting correlations were also observed between job- and organization- related factors and some of the homeostatic predictors. In this case, job satisfaction was correlated with: organizational culture (r= .581, P< .001), perceived control (r= .323, P< .001), optimism (r= .256, p< .01), and Self-esteem (r= .242, P< .01). AS it can be seen, there was a
Table 3 Hierarchical Regression Analyses for Factors of Homeostatic Model and Job- and Organization- related Factors on SWB of Teachers
Model 1: Total explained unique variance = .111 ; Total explained shared variance = .202 Model 2: Total explained unique variance = .124 ; Total explained shared variance = .338 * p< .05. ** p< .01. *** p< .001; sr² = amount of unique variance contributed Ex=Extraversion, ES=Emotional Stability, SE=Self Esteem, OP=Optimism, PC=Personal Control, JS=Job satisfaction, OC=Organizational Culture
As presented in Table 3, a hierarchical regression was carried out with SWB (PWI-SC) as a dependent variable. The homeostatic model factors of extraversion, emotional stability, selfesteem, optimism, and perceived control were entered into model one while job- and organization- related factors, i.e., job satisfaction and organizational culture, were entered into model two.
Model one demonstrated that the homeostatic model factors accounted for a significant 31.3 % of the variance in SWB (PWI-A), F(5, 126) = 11.47, p< .001, with optimism (?= .216) and Perceived Control (?= .295) being the strongest predictors, accounting for significant unique variability (sr² =.032 (3.2%) and .062 (6.2%), respectively). In this model, the total explained unique variance is 11.1 % while the total explained shared variance is 20.2 %. The variables, extraversion, emotional stability and self-esteem, did not make significant unique variance, but shared variance.
The addition of job- and organization- related factors (job satisfaction and organizational culture) in model two added significant prediction to the variance in SWB (PWI-A); an additional 15% of the variance was accounted for (R2 change = .150), F(2, 124) = 17.25, p< .001. Here, both variables added significant unique variance (job satisfaction = 2.4% and organizational culture = 4.5%). Of the homeostatic model factors, optimism and Perceived control continued to provide unique variance (2% (with a reduction of 1.2%) and 2.1% (with a reduction of 4.1%), respectively). In model two, the total explained unique variance is 12.4% while the total explained shared variance is 33.8%.
According to these results, the four strongest and significant predictors of SWB (PWI-A) of teachers at MTU are: organizational culture (ß= .262); job satisfaction (ß= .200); Perceived control (ß=.176); and optimism (ß= .174). These results provide support for the two major hypotheses of this study.
This study examined factors that affect the SWB of teachers of MTU. In so doing, the study determined teachers' level of SWB and investigated the role of various variables (homeostatic model factors and job - and organization - related factors) in predicting the SWB of teachers.
The results of the study showed that teachers of MTU had a low level of SWB as measured by the PWI-A, i.e., 61.66 (95%CI: 58.57 to 64.74). This conclusion is reached by comparing the current result with normative data/ranges for different groups in different parts of the world. According to IWbG (2013), a group's mean scores of the PWI-A (and other parallel forms) can be interpreted and referenced by comparing it with the normal distribution of group means. In this case, the normative range of Western means is between 70-80 points and Australian mean is between 73.4 and 76.4 points. Lau (2013) also reported that the normal distribution of group means of non-Western countries, specifically, Asians (China and East Asia) has a mean of 65 points.
In addition, within the Ethiopian context, there is one study that examined adolescents' SWB in Addis Ababa. This study used the Personal Wellbeing Index – School Children (PWI-SC), one of the parallel forms of PWI-A. Results of this study showed two different scores for two different groups, i.e., adolescents who practiced taekwondo (81.95) and adolescents who did not practice taekwondo (71.64) (Tadesse, 2015). Based on such comparisons, it can be said that MTU's teachers' level of SWB was low.
Similarly, MTU's teachers' mean points on most of the different domains of the PWI-A scale were very low, when compared to the normative ranges (Australian Center on Quality of life, 2016): standard of living (45.68 vs 77.84); future security (49.66 vs 71.07); achieving in life (54.86 vs 73.58); community connectedness (55.61 vs 71.04); and safety (58.08 vs 79.06). Meanwhile, some high mean points of teachers were observed on three other domains: spirituality/religion (78.84 vs 73.82); health (74.93 vs 74.58); and personal relationships (72.19 vs 79.46).
From the above comparisons, two interesting but unsurprising findings can be noted. Teachers' level of satisfaction with their standard of living was the lowest while their score on spirituality/ religion was the highest (even greater than the mean of the normative range). These results could be explained by linking them to the economic and religious nature of the country where the teachers lived. Economically, according to United Nations Statistics Division (2015), Ethiopia is one of the least developed countries in the world having Gross Domestic Product (GDP) Value of $46.17 billion and a GDP per capita of $489. As a result, seeing low level of satisfaction with standard of living among teachers can be expected. In the case of religion, according to a recent study by Pew Research Center (n.d.), most of the population in Ethiopia can be described as highly religious, Christianity and Islam being the dominant religions. Therefore, it may not be surprising to observe high level of satisfaction with spirituality/ religion among teachers of MTU.
This study tested two major hypotheses using the homeostatic model of SWB as a theoretical framework/model. The first hypothesis was that the homeostatic model of SWB significantly predicts the SWB of teachers of MTU. According to this model, SWB of people is determined by the interplay among personality factors (extraversion and neuroticism/emotional stability); cognitive buffer factors (perceived control, self-esteem, and optimism); and experiential inputs (major life event) (Mellor, Cummins, Karlinski & Storer, 2003). The second hypothesis is that job- and organization- related factors (job satisfaction and organizational culture) add significant prediction to the SWB of teachers of MTU, above and beyond the homeostatic model factors. Here, job - and organization - related factors (job satisfaction and organizational culture), are integrated into the homeostatic model of SWB, as additional aspects of experiential inputs (besides major life events). In this case, according to Cummins, Gullone and Lau (2002), the cognitive buffers are supposed to mediate the relationships between personality factors and SWB, as well as experiential inputs (including job- and organization- related factors) and SWB. It is also believed that each of the cognitive buffers moderates the effects of personality factors and experiential inputs (including job- and organization- related factors) on SWB.
Results of this study supported both hypotheses, indicating that there is a room in the homeostatic model of SWB for another dimension of experiential input, i.e., job-and organization related factors (job satisfaction and organizational culture).
The correlation analysis, as expected, showed significant positive correlations between the dependent variable (SWB) and the seven independent variables, i.e., job satisfaction, organizational culture, perceived control, self-esteem, optimism, emotional stability, and extraversion. Similarly, there was a strong positive correlation between the two job- and organization- related factors, i.e. job satisfaction and organizational culture (r = .581, P < .001), indicating that these two variables are indeed related to each other and can be grouped together. In addition, a significant positive correlation was observed between the three cognitive buffer factors (ranging from r = .411 to r =.533). This result is consistent with the idea that cognitive buffer factors form a combined buffering system for the purpose of SWB output (Cummins, Gullone & Lau, 2002).
Results from the hierarchical regression also showed that the homeostatic model factors accounted for a significant 31.3 % of the variance in SWB. In the same vein, job- and organization- related factors (job satisfaction and organizational culture) added a significant prediction to the variance in SWB, by 15%. This means, 46.3% of the variance in teachers’ SWB was explained by the combination of both groups of factors. Of the homeostatic model factors, optimism and perceived control provided significant unique variance while the rest of them (extraversion, emotional stability and self-esteem) contributed only shared variance to the prediction of teachers’ SWB. Both of the job- and organization- related factors, i.e., job satisfaction and organizational culture, also provided unique variance. Accordingly, the four strongest and significant predictors of teachers’ SWB were found to be: organizational culture (<span class="polytonic" style="font-family: 'SBL BibLit', 'SBL Greek', Athena, 'EB Garamond 12', 'Foulis Greek', Cardo, 'Gentium Plus', Gentium, Garamond, 'Palatino Linotype', 'DejaVu Sans', 'DejaVu Serif', FreeSerif, FreeSans, 'Arial Unicode MS', 'Lucida Sans Unicode', 'Lucida Grande', Code2000, sans-serif;">ß</span> = .262); job satisfaction (<span class="polytonic" style="font-family: 'SBL BibLit', 'SBL Greek', Athena, 'EB Garamond 12', 'Foulis Greek', Cardo, 'Gentium Plus', Gentium, Garamond, 'Palatino Linotype', 'DejaVu Sans', 'DejaVu Serif', FreeSerif, FreeSans, 'Arial Unicode MS', 'Lucida Sans Unicode', 'Lucida Grande', Code2000, sans-serif;">ß</span> = .200); Perceived control (<span class="polytonic" style="font-family: 'SBL BibLit', 'SBL Greek', Athena, 'EB Garamond 12', 'Foulis Greek', Cardo, 'Gentium Plus', Gentium, Garamond, 'Palatino Linotype', 'DejaVu Sans', 'DejaVu Serif', FreeSerif, FreeSans, 'Arial Unicode MS', 'Lucida Sans Unicode', 'Lucida Grande', Code2000, sans-serif;">ß</span> =.176); and optimism (<span class="polytonic" style="font-family: 'SBL BibLit', 'SBL Greek', Athena, 'EB Garamond 12', 'Foulis Greek', Cardo, 'Gentium Plus', Gentium, Garamond, 'Palatino Linotype', 'DejaVu Sans', 'DejaVu Serif', FreeSerif, FreeSans, 'Arial Unicode MS', 'Lucida Sans Unicode', 'Lucida Grande', Code2000, sans-serif;">ß</span> = .174).
The findings of this study demonstrated that job - and organization- related factors (job satisfaction and organizational culture), as environmental experiences, interact with homeostatic model factors to regulate teachers’ level of SWB. This means that if a teacher experiences poor organizational culture and/or low level of job satisfaction, his/her level of SWB will be negatively affected (lowered), depending on the severity of the experience. Similarly, Lai and Cummins (2013) found in their study that job satisfaction can have effect on people’s level of SWB.
The low levels of SWB, job satisfaction, and organizational culture in MTU could be attributed to the various problems that existed in the University. Findings of a recent investigation by MTU’s quality audit team, for instance, highlighted the severity of the situation at the university. The quality audit identified problems with regard to (1) good governance (e.g., accountability problems, problem with decision making and fairness, poor human/ financial/ material resource management, problem with implementation of policies/ guidelines); (2) infrastructure (e.g., lack of office and office appliances, poor toilet facilities, lack of recreational facilities, poor transportation service, sever ICT-related problems); and (3) treatment of academic staff (e.g., lack of incentive mechanisms and problems with recruitment/ promotion/ transfer). This report also underlined the fact that there was a very high rate of academic staff turnover and absence of staff retention mechanisms (MTU, 2016).
This study made effort to investigate factors that affect the SWB of teachers of MTU. In so doing, two limitations were observed. The first limitation was associated with the model of SWB used in this study. This study used the original homeostatic model of SWB. However, this model was recently revised after three studies demonstrated that the model could be more effective if personality factors were replaced by HPMood (Core Affect). HPMood (Core Affect) is measured by three affective adjectives as 'Happy', 'Content' and 'Alert'. In this study, HPMood (Core Affect) was not used because the researcher noticed a possible language and/or cultural barrier for respondents to understand and differentiate terms/concepts like 'happy', 'content' and 'satisfaction.' For most Ethiopians (study participants), these three words mean the same thing. Only, psychologists and few other professionals (such as psychiatrists and social workers) could tell the difference between these concepts.
The other limitation was related to a possible measurement error that led to the exclusion of one of the homeostatic model variables from analysis. As previously discussed, this variable is called major life event/ experiential input. The researcher believes that the proper measurement and inclusion of this variable might have improved the prediction power of the model.
In spite of the abovementioned limitations, the researcher believes that the findings of the current study could make important contributions in terms of testing the homeostatic model of SWB and highlighting the role of job- and organization- related factors in affecting SWB of teachers.
This study examined the SWB of teachers of MTU in relation to factors that affect their SWB. In this case, the homeostatic model of SWB was employed and especial attention was given to Job- and organization- related factors (job satisfaction and organizational culture) as predictors of SWB. It was found that teachers of MTU had not only low level of SWB but also low level of job satisfaction as well as poor rating of their organization's culture. It was also found, as hypothesized, that job- and organization- related factors (job satisfaction and organizational culture) joined forces with homeostatic model factors in affecting the SWB of teachers of MTU.
Based on the results of this study, the researcher urges MTU and other concerned government bodies (such as the Ministry of Education and the regional government) to strive for improving the organizational culture of the university and the job satisfaction of teachers. Doing so will help to improve the SWB of the teachers.
This research was made possible with the help and support of a number of organizations and a lot of people. First, I would like to thank Mizan – Tepi University for providing me with the funding for this study. Second, I really appreciate participants of this study (in both the pilot and main studies) for sacrificing their time and energy to participate in the study. I also thank very much those individuals, departments and colleges/organizations that helped me during data collection. Finally, I would like to express my deep gratitude for those individuals who commented on the research and who provided me with important information and materials.