Work satisfaction of municipality employees while performing duties under the conditions of limited financial resources ”

The purpose of the study on which this article reports is to use affective events theory as a theoretical framework to develop a model that could be used by municipal supervisors to create an environment conducive to working. The study comprised 227 non-management, middle managers and senior managers in a municipality. Confirmatory factor analysis (CFA) and structural equation modelling (SEM) were used to analyze the data. There was a low negative correlation between supervisory support and job satisfaction. It was also found that positive emotions mediated the relationship between supervisory support, participation and job satisfaction.


Introduction
One of the challenges facing municipal supervisors is to understand environmental determinants leading to job satisfaction of employees in the municipality.If supervisors provide guidance and support and treat employees fairly, employees tend to be satisfied (Gregory, Osmonbekov & Gregory, 2009).Supervisors who empower employees to participate in decision-making (Buitendach & Rothmann, 2009), and who give employees autonomy to make decisions create a working environment that is conducive to creating job satisfaction for employees (Al-Hamdan, Bawadi, Bawadi & Mrayyan, 2013).In a study done by Markham and Van Zyl (2013), on average, 14.67% of 40 respondents in a South African municipality were dissatisfied, because they believed that their supervisors did not care for them.Employees working in a municipality in Iran responded that work overload was the highest predictor of job satisfaction ( = -0.36;p = 0.01) (Bemana, Moradi, Ghasemi, Taghavi & Ghayoor, 2013).Another challenge was that in a Cape Town municipality, 90% of the respondents were disgruntled, because there were no training opportunities and they were not placed in jobs that matched their competencies (Koketso & Rust, 2012).To enhance employees' competency and satisfaction, Dywili (2015) recommended that the Nkondobe municipality in Eastern Cape gives employees different tasks.Wegge 2006) used linear regression and found that environmental determinants (i.e., autonomy, supervisory support, participation in decision-making and work overload) related to job satisfaction, and that emotions mediated the relationship between environmental determinants and job satisfaction.Conversely, international research by Yu (2016) illustrated the relationship between environmental factors and job satisfaction using advanced statistical tests such as SEM.The authors of this paper align themselves with the research using SEM, because "SEM really means extending … repertoire of data analysis skills to the next level, one that offers even more flexibility than the GLM" (Kline, 2011, p. 13).
In the study on which this article is based, the research questions were: what is the relationship between environmental determinants and job satisfaction?and do positive emotions mediate the relationship between environmental determinants and job satisfaction?
1. Literature review 1.1.Conceptual model.Affective event theory (AET) was deemed to be relevant to develop a model for South African municipalities regarding the environmental determinants to job satisfaction.AET was developed by Weiss and Cropanzano (1996).Work with a high workload produces emotional reactions (i.e., positive or negative emotions) (Ashkanasy & Duas, 2002).Depending on the emotional reactions of the employees, they will be either satisfied or dissatisfied, and they might be committed to the organization or not (Robbins & Judge, 2015).
Job satisfaction is defined as a pleasurable or emotional state of relations to superiors, co-workers and the work environment (Locke, 1976).Within the AET theoretical framework, job satisfaction has to do with how employees evaluate an event (Robbins & Judge, 2015).When relating job satisfaction to environmental determinants in the call centre industry in the United Kingdom, Wegge et al. (2006) found that environmental determinants (autonomy, supervisory support, participation in decision-making, work overload) related either positively or negatively to job satisfaction.

Autonomy and job satisfaction.
In their research, Butler, Grzywacz, Bass and Linney (2005) found a positive relationship between autonomy and job satisfaction.In terms of the relationship, literature showed that there was a positive relationship between autonomy and job satisfaction (Butler et al., 2005).Factors that made autonomous employees satisfied were that they were given freedom to make decisions on their own, and they were given permission to control their work schedules (Ahuja & Thatcher, 2005).Furthermore, employees who are given autonomy are satisfied (Saragih, 2011).In light of this, the following hypothesis was stated: Hypothesis 1: There is a positive relationship between autonomy and job satisfaction.

Supervisory support and job satisfaction.
Supervisory support means that a supervisor listens to, cares for and supports employees when they are experiencing challenges (Cieslak, Korczynska, Strelau & Kaczmerek, 2008).Supervisors influence employees' job satisfaction when they treat employees fairly and implement organizational policies consistently (Rodwell & Munro, 2013), foster team cohesiveness, support employees who are going through difficult times (Basim & Begenirbas, 2012) and give employees opportunities to develop their careers (Fisher, 2010).The majority (65%) of the respondents at the Cape Town municipality responded that they never had a meeting with their supervisors to discuss their career development (Koketso & Rust, 2012).That been said, the following hypothesis was postulated: Hypothesis 2: There is a positive relationship between supervisory support and job satisfaction.

Participation in decision-making and job satisfaction.
Participation in decision-making has to do with whether employees have control over their work schedule and are trusted to perform work activities without asking permission from their supervisors (Patterson, West, Shackleton, Dawson, Lawthom, Maitlis & Robinson, 2005).Griffin and Moorhead (2011) established that there was a positive correlation between employee participation and job satisfaction.In another study, employees were satisfied because they were trusted to make decisions without asking permission (Lephalala, Ehlers & Oosthuizen, 2008).Similarly, Dywili (2015) reported that educated municipal workers were satisfied, because they had the opportunity to participate in planning work activities.Loyalty due to job satisfaction was also found at a municipality in Nkonkobe in Eastern Cape (Dywili, 2015).The hypothesis that was tested in this context was: Hypothesis 3: There is a positive relationship between participation in decision-making and job satisfaction.

Work overload and job satisfaction.
Unlike the other determinants discussed above, it had been found that there is a negative relationship between work overload and job satisfaction (Yavas, Karatepe & Babakus, 2013).Giving employees difficult tasks and excessive work had a psychological effect (i.e., stress and burnout) on them (Beham & Drobnic, 2010).Furthermore, overloaded employees did not meet their targets and some of them intended to leave their or had left their organizations (Kalisch, Lee & Rochman, 2010).This led to the following hypothesis: Hypothesis 4: There is a negative relationship between work overload and job satisfaction.

Positive emotions as a mediator.
Whereas the above discussion was based on the direct relationship between environmental determinants (i.e., autonomy, supervisory support, participation in decision-making and work overload) and job satisfaction, Wegge et al. (2006) found that positive emotions mediated or had an indirect effect between environmental determinants and job satisfaction.Wegge et al. (2006) calculated a Sobel test and found positive emotions to be a mediator between environmental effects and job satisfaction.Based on this finding, the following hypothesis was developed: Hypothesis 5: Positive emotions mediate the relationship between environmental determinants and job satisfaction.
Based on the above literature, a conceptual model, shown in Figure 1, was proposed.In the next section, the methods followed to address the aim of the present study, namely, to develop a model of environmental determinants on job satisfaction for a South African municipality, are discussed.

Research design.
The study used a survey and followed a cross-sectional research design.The survey design was influenced by positivism.The authors took Babbie's (2013) advice, namely that through positivism, numbers and hypotheses could be used to address the study hypotheses.

Participants and procedure.
Between 2011 and 2015, the municipality under study was under administration, and at the time when this study was conducted, it was unclear which environmental determinants affected employees' job satisfaction.Municipal employees work in an environment where they interact with disgruntled and rude customers and they are expected to perform even when their municipalities do not have resources to assist them to perform (Du Plessis & Fouche, 2015).This creates an environment which is not conducive to working and, ultimately, affects municipal workers' job satisfaction.
There were 1 495 workers at the South African municipality.Convenience sampling was used to select participants based on the availability of the respondents (Leedy & Ormrod, 2015).Using the Abraxas Energy Consulting tool at margin of error of 6% and confidence level of 95%, the sample size on this study was to be 227.However, the author who collected the data, also gave two more respondents questionnaires to complete; hence, the sample sizewas 229.A sample size of 200 is sufficient to enable the researchers to conduct advanced statistical tests, such as CFA and SEM (Kline, 2011).
The demographic characteristics (refer to Table 1) were as follows: there were more females (54%) than males (46%) in the sample.The majority of respondents were not in management positions (79%).Hundred and eighteen respondents (52%) had a diploma, degree or post-graduate qualification.However, the authors did not ask the respondents which type of qualification they had.

Measuring process.
Respondents were given consent forms prior to completing the questionnaire.They were informed that their participation was voluntary and anonymous.In addition, the respondents were informed that the results of the study would be communicated to management before publication.Prior to data collection, the researchers were given permission by the municipal administrator to conduct the study.

Data analysis and results.
To determine whether respondents agreed or disagreed with the items, 4 was used as mean (M) on a scale where 1 meant strongly disagree and 7 meant strongly agree.
On the autonomy scale, three items had mean scores that were lower than 4. Supervisory support responses were all above 4.For example, "My supervisor listens to what I have to say" had a mean of M=4.44.Participation responses were above 4, except for this item: "I am trusted to make decisions without having to first get permission" (M = 3.55).Workload responses were all below 4: "The job that I am doing is too difficult for me" (M=2.59) and "There are tasks that I have to perform that I cannot do well" (M = 2.94).These were the job satisfaction items that had a mean of above 4: "I like the kind of work that I am doing" (M=5.05) and "My work gives me a sense of personal accomplishment" (M=5.10).For positive emotions, some mean score was above 4: "I feel proud of the work that I do" (M=5.32) and "I am happy to come to work" (M = 4.92).For all the scales, the standard deviation (SD) ranged from 1.90 to 2.41, which suggested a high variation of responses.

Structural results.
For the authors to develop the SEM model of environmental determinants on job satisfaction for a South African municipality, factor loadings below 0.4 were not retained, as can be observed in Table 2.The CFA model confirmed that items related to constructs were 2 =251.128;RMSEA=0.079;CFI=0.928.It can also be observed from Table 2 that all the Cronbach's alphas were within the 0.5 to 0.9 range, as suggested by Maree (2016).This suggests that the questionnaire used was reliable and convergent validity was achieved.
To test the conceptual model (see Figure 1) and hypotheses, SEM was used.To determine an absolute goodness-of-fit index, a 2 was calculated to "evaluate the differences between the sample covariance and index covariance of the suggested model" (Zacharewicz & Martines-Inigo, 2015, p. 169).Since 2 is affected by the sample size (Kline, 2011), comparative indices were also used.As can be observed in Table 3, the revised model is in line with the SEM goodness-of-fit model.Based on the factor loadings (refer to Table 2), direct and indirect coefficients (refer to hypotheses testing section below), the structural graph below was developed.The structural path is shown graphically in Figure 2 (see Appendix).

Hypotheses testing
The only significant relationship was between supervisory support and job satisfaction.Since there was no significant relationship between autonomy, participation and work overload, the authors did not interpret hypotheses 1, 3 and 4. Earlier in the article, the hypothesis 2 relating to supervisory support and job satisfaction was stated as follows: There is a positive relationship between supervisory support and job satisfaction.The data showed that there was a low negative correlation between supervisory support and job satisfaction ( = -0.28,p = 0.05*).The results, therefore, did not support hypothesis 2, but the results were significant.Positive emotions had significant indirect effects on results relating to supervisory support ( = 0.34, p = 0.05*) and participation ( = 0.70, p = 0.05*).Therefore, hypothesis 5, positive emotions mediate the rela-tionship between environmental determinants and job satisfaction was partially supported.

Discussion
The aim of the present study was to develop a model for a South African municipality.Employees responded that they were not given autonomy to make decisions, they were not encouraged to give feedback about the activities, and they were not involved in the changes that were made in the organization.The data also showed that there was a positive relationship, albeit not significant, between autonomy and job satisfaction.In the literature, it was also argued that there was a positive relationship between autonomy and job satisfaction (Butler et al., 2005).
The mean scores of the relationship between supervisory support items and job satisfaction were above 4, suggesting that employees agreed that their supervisors listened to them and supported them with work-related challenges.In terms of the latter, Basim and Begenirbas (2012) are of the opinion that employees are satisfied when their supervisors support them during challenging times.Consistent with the literature (Griffin & Moorhead, 2011), the data of this study showed that there was a positive relationship between participation in decision-making and job satisfaction, but this relationship was not significant.Unlike in Lephalala et al.'s (2008) study, the present study showed that respondents moderately disagreed (M3.55) that they were trusted to make decisions on the own.The other results of the present study showed that there was a negative relationship between work overload and job satisfaction, but not significantly so.Similarly, Yavas et al. (2013) also found that work overload related negatively with job satisfaction.Unlike the findings of Wegge et al. (2006), the present study showed that positive emotions only mediated the relationship between supervisory support ( = 0.34; p = 0.05*) and participation ( = 0.70; p = 0.05*) and job satisfaction.
There was other noteworthy positive emotions finding.For example, it was found that the mean score of the item "I am happy to come to work" (M= 4.92) was above 4. Earlier it was stated that municipal workers interacted with disgruntled, rude customers (Mafini et al., 2011) and they were expected to perform even though, at the time of the study, their municipalities did not have resources to assist them to perform (Du Plessis & Fouche, 2015).

Limitations, contribution and recommendations
This study showed a few limitations.A selfcompleted questionnaire was used to collect data, the research followed a cross-sectional design and a non-probability sampling technique was used.Owing to the sampling technique, the results cannot be generalized for the target population (Bryman, 2012).Despite these limitations, the present study contributed a model (Figure 2, see Appendix), which can be used by municipal managers or supervisors to enhance job satisfaction.is recommended that in future, similar research should be conducted using a probability technique and/or mixed method design.The latter will enable researchers to gather rich data (Bry-man, 2012).In terms of the municipality management, the following is recommended: Give employees opportunity to develop their skills; Consult employees before decisions are made; Involve employees about the changes that are made in the department; Give educated and skilled workers autonomy to make decisions on their own; and Encourage supervisors to have interpersonal skills (i.e., listening to and support of employees).

Implications for management
Employees are satisfied if they are given an opportunity to make decisions and they should be consulted on any major decisions that are made.In terms of the latter, if employees are not involved in the decision-making, they do not support the changes.Engaged employees were found to be good performers, they were committed and they stayed loyal to the organizations.Not training municipal workers in emotional intelligence, for example, might have unintended consequences, since the workers might not know how to handle difficult and despondent customers.
In conclusion, both the data and the literature showed that municipal supervisors might not be able to create a workplace conducive to working if they do not understand how autonomy, supervisory support, participation and work overload relate to job satisfaction and how positive emotions mediate the relationship between environmental determinants and job satisfaction.It is envisaged that through the model developed by this study, municipal supervisors would be empowered to create a working environment where employees experience job satisfaction.
, Van Dick, Fisher, West and Dawson (2006) mention that autonomy, supervisory support, participation in decision-making and work overload are classified as environmental determinants.Molefe Jonathan Maleka, Thulani Bhuti Skosana, Josiah Patrick Lekgothoana, 2016.Molefe Jonathan Maleka, Dr., Department of People Management and Development, Tshwane University of Technology, South Africa.Thulani Bhuti Skosana, Mr., Department of People Management and Development, Tshwane University of Technology, South Africa.Josiah Patrick Lekgothoana, Mr., Department of People Management and Development, Tshwane University of Technology, South Africa.

Table 1 .
Demographical characteristics of respondents

Table 3 .
Fit indices of environmental determinants on job satisfaction