Rewards : A predictor of well-being and service quality of school principals in the North-West province

The South African school system continues to be under intense scrutiny, criticised for failing to deliver the required services (Department of Basic Education, 2015; South African Human Rights Commission, 2014; Taylor, Mabogoane & Akoobhai, 2011). Van der Berg, Taylor, Gustafsson, Spaull and Armstrong (2011) found that some of the shortcomings in school teaching practices are attributable to school leadership. The role of school principals in the quality of education in schools is therefore imperative (Ayeni, 2012; Clark, Martorell & Rockoff, 2009; Steyn, 2013; Trisno & Nurdianto, 2014).


Introduction
The South African school system continues to be under intense scrutiny, criticised for failing to deliver the required services (Department of Basic Education, 2015;South African Human Rights Commission, 2014;Taylor, Mabogoane & Akoobhai, 2011).Van der Berg, Taylor, Gustafsson, Spaull and Armstrong (2011) found that some of the shortcomings in school teaching practices are attributable to school leadership.The role of school principals in the quality of education in schools is therefore imperative (Ayeni, 2012;Clark, Martorell & Rockoff, 2009;Steyn, 2013;Trisno & Nurdianto, 2014).
The demanding role of school principals in maintaining quality education is well documented (Frederici & Skaalvik, 2011;Newton, 2015;Özer, 2013;Tanjung, 2015).School principals have, among others, the task of managing pupils, handling criticism from parents, and placating frustrated teachers (Boyland, 2011), while at the same time managing daily operations at a school and monitoring the school's infrastructure (plants) (Amanchukwu & Ololube, 2015).Furthermore, in the midst of the economic downturn and political upheaval, school principals are expected to play an important role in education reform (Bas & Yildirim, 2012).Therefore, school principals are expected to display the principles of effective and productive leadership, and to make a meaningful contribution to society (Bartlett, 2011;Boyland, 2011).
The South African education system may soon face the dilemma of a 'brain drain' of school principals if they are not managed and compensated well (Maforah & Schulze, 2012;Reddy et al., 2010;Steyn, 2013).Evidence suggests that the reward system of the Department of Education is not attracting more capable and talented principals (Arends, 2007;Armstrong, 2009;South African Council for Educators, 2010).A study by Maforah and Schulze (2012) showed that poor salaries discourage talented teachers to apply for promotion to the position of school principal.This problem is further exacerbated by the fact that remuneration of principals differs in terms of the Post Provision Model (PPM) of the school.According to this model, a school principal of a large school in terms of learners receives a higher salary than the principal of a smaller school (Smit & Oosthuizen, 2011).The model does not take into consideration the responsibilities or the work load of individual principals (Maarman, Steyn & Wolhuter, 2006).In large schools, the principal is responsible for management and administration only, and has the support of a complementary school management team, whereas, in small and medium schools, the principal manages and teaches at the same time, and has very few management team members.
School principals therefore have the difficult task of managing a school or classroom successfully, maintaining a healthy family life and preserving their own wellness (Bas & Yildirim, 2012).Increasing role demands without supportive measures can lead to stress and burnout among school principals (Boyland, 2011).Bakker, Gierveld and van Rijswijk (2006) found that social support from colleagues, together with rewards in the form of career development opportunities, enhances the work engagement (the presumed antidote to burnout) of school principals.Furthermore, a principal's health and well-being, in turn, have an impact on effective and quality service delivery in schools (Igwebuike, 2013).
Against this background, the main objective of this research was to determine whether rewards are a significant predictor of well-being (burnout versus work engagement) and service quality of school principals in the North-West province.Price (2009) postulated that the manner in which people are rewarded affects the quality of their work, their attitude towards customers and their willingness to be flexible or learn new skills and suggest innovations.Moreover, Andreyko (2010) pointed out that the demanding nature of the position of principal results in a shrinking pool of available school leaders.Eventually, teachers and school principals will look outside the Department of Education and beyond the borders of the country for better remuneration (Steyn, 2013).This will leave a vacuum that is very difficult to fill, and will hamper effective service delivery in schools, as envisaged in the Action Plan 2019 Towards the Realisation of Schooling (Department of Basic Education, 2011).
The next section of this article highlights the available literature on the relationships between rewards, well-being and service quality.Flowing from literature review the research method and results are presented, followed by a discussion of the results.This article concludes with recommendations for practice and further research.

Rewards
Rewards encompass all forms of financial returns and tangible services and benefits employees receive as part of an employment relationship (Milkovich, Newman & Gerhart, 2013).These include base pay, merit pay, short-and long-term incentives, perquisites, recognition awards, and retention awards (Van der Westhuizen & Wessels, 2011).For purposes of the present research, the term rewards represents a multidimensional construct consisting of basic salary, benefits, work-life balance, performance management and recognition, and training and development (World at Work Society, 2011).According to the Total Rewards model, compensation includes fixed or variable pay for services rendered.Benefits can include a variety of programmes, such as healthcare and retirement, to provide security for employees and their families.Performance management refers to the alignment between employee efforts and business goals, whereas career development opportunities can assist school principals in advancing their skills and competencies (Bas & Yildirim, 2012).
According to Chapman (2005), quality principalship cannot be maintained without a coherent, integrated and systematic approach to leadership development.Researchers continue to point out the challenges relating to the career development of school principals.A study by Mulkeen, Chapman, DeJaeghere and Leu (2007) highlighted the need for organised and systematic training to enhance the professionalisation of leadership in education.A study by Bartlett (2011) pointed out the lack of mentoring for school principals, concomitant with poor career development prospects.As regards compensation, a study by Maforah (2010) showed that principals felt that their salaries were not aligned with their workload, and that more effective reward systems were needed.This study furthermore highlighted the poor performance management systems of the Department of Education, as well as policy applications that are eroding the autonomy of principals and their ability to manage effectively.

Well-being
Well-being is a concept that consists of two opposite poles: burnout and work engagement (Schaufeli, 2003;Schaufeli & Salanova, 2007).Burnout is a prolonged pathological illbeing condition (Mäkikangas, Kinnunen, Feldt & Schaufeli, 2016), whereas the concept of work engagement emerged through the positive occupational health psychology movement to denote a positive, fulfilling work-related state of mind (Shimazu & Schaufeli, 2009).According Bakker and Demerouti (2008), burnout occurs as a result of excessive striving to reach unrealistic work goals, which leads to work overload, patterns of overcommitment, and exhaustion of physical and mental resources (Bakker & Demerouti, 2008).A person displaying such behaviours for a prolonged period eventually becomes listless, ineffective, inefficient and unproductive (Van der Westhuizen & Wessels, 2011).Burnout consists of three dimensions, namely exhaustion, mental distance, and lack of professional efficacy (Schaufeli, 2003).Exhaustion refers to a feeling of being overextended and depleted of one's emotional and physical resources (Leiter & Maslach, 2008).Schaufeli, Taris and Van Rhenen (2008) defined mental distance as a distant attitude towards one's job (cynicism) or people (depersonalisation).Lack of professional efficacy includes feelings of inefficiency, which are a result of a tendency to evaluate one's work negatively (Schaufeli & Taris, 2005).
Burnout is a frequently reported phenomenon in human services professions such as education (Brouwers & Tomic, 2014;Thomas, Kohli & Choi, 2014).School principals are particularly prone to emotional and physical burnout, as they have to manage an environment of which humans are the main input and output (Özer, 2013).Continuous intensive interpersonal relationships with problem students, complaining parents, and a criticising society, while having to take care of daily administrative functions, can result in school principals feeling exhausted, aloof, and deprecated (Bas & Yildirim, 2012;Friedman, 2002).A study by Whitehead, Ryba and O'Driscoll (2000) revealed high levels of emotional exhaustion among school principals.Özer (2013) maintained that school principals with a weak sense of efficacy where interpersonal relations are profound are likely to feel physically worn out and experience a sense of despair.According to Tomic and Tomic (2008), school principals who feel a sense of inefficacy may doubt whether they should remain in their job.
Work engagement: According to Scott and McMullen (2010), work engagement includes a high level of employee involvement, commitment to the organisation, and job satisfaction.Work engagement is a positive psychological state of mind and includes three dimensions: vigour, dedication and absorption (Schaufeli & Bakker, 2010).Vigour is the energy dimension of work engagement and is characterised by high levels of mental resistance and persistence in carrying out difficult tasks (Coetzer & Rothmann, 2013).Coetzee and De Villiers (2010) defined dedication as the willingness of people to spend considerable time and effort on doing something meaningful.Absorption is a state in which individuals are fully focused on their work and experience a high level of concentration while performing a task (Bakker, Schaufeli, Leiter & Taris, 2008).
Work engagement is the all-involved psychological commitment of an employee to the role he or she fulfils in the organisation, and enhances employees' motivation to try harder and invest extra effort (Nel et al., 2008).Work engagement in the context of school education is characterised by a sense of connecting with student learning and identification with the school as employer (Newton, 2015).Engagement of school principals is therefore important, as they are then more willing to take on extra tasks that are not part of their job description, invest more effort in their jobs, share information with other employees, and remain with the organisation longer than employees who are less engaged (Bakker & Demerouti, 2008;Scott & McMullen, 2010).

Service quality
Service quality is an attitude towards the service offered, resulting from a comparison of expectations with perceptions (Cüliberg & Rosjek, 2010).The service expectations of school principals are varied and extensive.Service quality in the context of student education is the ability of a school to transform students' knowledge base, ability, wisdom and character; to prepare them academically for higher education; and also shape them personally to become responsible and contributing members of society (Al-Refai, Al-Omar, Son & Hamdy, 2015).School principals further have the responsibility of effectively managing their staff.A study by Dworkin, Saha and Hill (2003) showed that principals who are non-authoritarian, supportive and collegial reduce the incidences of burnout among their staff members, which, in turn, facilitates effective service delivery.Amanchukwu and Ololube (2015) highlighted the importance of school principals managing and monitoring school plants such as buildings, equipment and essential structures for effective teaching and learning.
For purposes of the present research, we used The Five-factor SERVQUAL Model of Parasuraman, Zeithaml and Berry (1988) to assess school principals' service quality orientation.According to this model, service quality consists of five dimensions: Reliability, Responsiveness, Assurance, Empathy, and Tangibles (Sharabi, 2013).Reliability refers to the ability to perform the promised service dependably and accurately.Responsiveness includes the willingness to assist customers and to provide prompt service.Assurance refers to the knowledge and courtesy of employees and their ability to convey trust and confidence.Empathy refers to how a company cares and gives individualised attention to their customers, making the customers feel valued and special.Tangibles refers to the appearance of physical facilities, equipment, personnel and communication materials.Mosahab, Mahamad and Ramayah (2010) found that schools lacked quality service in the areas of responsiveness and tangibles.The importance of school principals in the total quality management of schools cannot be underestimated (Al-Refai et al., 2015;Cheng & Yau, 2011).

Reciprocity between rewards, well-being, and service quality
Rewards and burnout: The relationship between rewards and burnout is well documented.According to Leiter and Maslach (2008), insufficient rewards (whether financial, institutional or social) increase people's vulnerability to burnout.This implies that, if a reward is insufficient, it may lead to exhaustion, which will create a callous or excessively detached response to various aspects of the job.A study by Jackson and Rothmann (2006) showed that poor pay and benefits increase incidences of burnout among teachers.Coetzee and De Villiers (2010) further found that burnout in conjunction with poor remuneration result in lower levels of absorption and subsequent lower commitment to an organisation.Tomic and Tomic (2008) alluded to a study of Sari (2004), who found that lack of rewards and recognition, among other factors, is a source of burnout among school principals.Boyland (2011) found that opportunities for professional development and improved performance evaluation processes may reduce incidences of stress and related illnesses among school principals: • Hypothesis 1: Rewards would be negatively related to burnout.

Burnout and service quality:
The competitive nature of schools, concomitant with increased accountability and responsibilities, is resulting in significant physical and emotional costs for school principals (Andreyko, 2010).Ashtari, Farhady and Khodaee (2009) found that high levels of exhaustion and depersonalisation result in an inability to perform.A study by Abdullah and Yuen (2011) found that high levels of exhaustion result in poor performance among employees in human service organisations.Humborstad, Humborstad and Whitfield (2008) found that burnout reduces a service individual's willingness to deliver quality service.A study by Igwebuike (2013) showed that a principal's health and well-being have a positive influence on educational service delivery.Omanukwue (2013) furthermore found that burnout leads to school principals disengaging from their work tasks, as they find the tasks uninteresting and no longer challenging: • Hypothesis 2: Burnout would be negatively related to service quality.
Rewards and service quality: According to Armstrong (2012), reward makes a positive impact on performance when it contributes to the development of a high-performance culture.This signifies that, if school principals are rewarded well, their performance will increase, and they will provide a high quality service.According to Sigh (2007), incentives motivate employees to put forth their best efforts, and the average productivity level of labour increases.Reward systems inspire staff to provide their best attempts in performing assigned tasks (Gohari, Kamkar, Hosseinipour & Zohori, 2013).Demerouti, Bakker and Leiter (2014) found that employees who compensate themselves in terms of career development opportunities are less likely to experience burnout, and are able to perform better on the job.Several studies have shown that both financial and non-financial rewards (i.e.salary, bonus, incentives, allowances and fringe benefits) show a strong relationship with job performance (Ashraf, Bandiera & Jack, 2014;Osibanjo, Adeniji, Falola & Heirsmac, 2014).A study by (Billger, 2007) showed that the low pay of school principals was related to poor accountability and, subsequently, poor matric pass rates: • Hypothesis 3: Rewards would be positively related to service quality.
Rewards and work engagement: Jacobs, Renard and Snelgar (2014) found that intrinsic rewards, such as meaningfulness, choice, competence and progress, resulted in higher levels of work engagement of employees in the South African retail industry.Likewise, Abdelhadi and Drach-Zahavy (2012) found that financial rewards significantly enhance the work engagement of employees.Bakker et al. (2006) found that development opportunities enhance the work engagement of school principals.Bakker et al. (2008) found that performance feedback is the best predictor of work engagement.Bakker, Hakanen, Demerouti and Xhantopoulou (2007) found that job resources, such as appreciation, act as a buffer between pupils' misbehaviour and teachers' work engagement.Effort-reward imbalances can have a negative impact on work engagement (Inoue et al., 2014;Kinnunen, Feldt & Makikangas, 2008).Therefore, satisfaction with both financial and non-financial rewards is crucial in enhancing the work engagement of employees (Malik, 2013): • Hypothesis 4: Rewards would be positively related to work engagement.
Work engagement and service quality: According to Newton (2015), organisational science research on employee engagement follows a general trend of exploring human capital as a means for organisations to meet their goals.A study by Salanova et al. (2005) showed that organisational resources and work engagement enhance service climate, which, in turn, has a positive spillover effect on employee performance and customer loyalty.Barkhuizen, Mogwere and Schutte (2014) found that the work engagement of support staff in an academic institution enhances their service quality orientation.In essence, the long-term outcomes of work engagement include better quality products and services and more satisfied and loyal customers (Bakker & Demerouti, 2008).Koch, Binnewies and Dormann (2015) found that the work engagement of school principals is a motivational catalyst for enhanced innovation and creativity of school teachers.Furthermore, Tanjung (2015) found that the work engagement of school leadership also enhances the professionalism of school teachers: • Hypothesis 5: Work engagement would be positively related to service quality.

Research approach
A quantitative, cross-sectional research design was followed in the present study.Cross-sectional research occurs when data are collected from the research participants at a single point in time or during a single, relatively brief time period (Johnson & Christensen, 2008).Cross-sectional research is ideally suited to testing for correlations between variables (Field, 2009).

Measures Sampling
Public school principals in four districts of the North-West province were identified as the unit of analysis for this study.Of the 300 questionnaires that were distributed, only 155 (N = 155) were received and analysed, which gave a total response of 51.7%.Most of the respondents in this research were female (65.8%), aged between 30 and 39 years (43.2%), and representative of the black ethnic group (43.2%).Most of the respondents (76.1%) had worked for their current school for more than 16 years, and worked more than 40 hours per week (67.7%).
The Utrecht Work Engagement Scale (UWES) (Schaufeli & Bakker, 2003) was used to measure work engagement.The UWES consists of 11 items, and measures three dimensions, namely Vigour, Dedication and Absorption.Responses are provided on a six-point scale varying from 0 (Never occurs) to 6 (Occurs every day).The questionnaire has been validated in the South African context (Barkhuizen & Rothmann, 2006;Bell & Barkhuizen, 2011), with Cronbach alphas ranging from 0.78 to 0.85 for the three dimensions of Work engagement.
The researchers adapted the SERVQUAL scale (Parasuraman et al., 1998) to measure five dimensions of the respondents' orientation towards delivering quality service: Reliability, Tangibles, Responsiveness, Assurance and Empathy.
Responses were measured on a six-point Likert scale ranging from 1 (Strongly disagree) to 6 (Strongly agree).The questionnaire has been validated in the South African context (Barkhuizen et al., 2014).

Research procedure
Permission was obtained from the head of the Department of Education of the North-West province to use the school principals within his jurisdiction as participants.The surveys were distributed in hard copy format to the participants.Ethical clearance was obtained prior to the execution of the study.The confidentiality of the participants was ensured, as the researcher did not request or record the respondents' identity.

Data analysis
Statistical analysis was carried out using the SPSS program (SPSS Inc., 2015).The reliability and validity of the measuring instruments were determined by means of Cronbach alpha coefficients, as well as exploratory factor analysis.Linear regression analyses were applied to test for the significance of predictive relationships between the variables (see Field, 2009).

Results
Prior to the testing of the hypotheses, the measurements were subjected to a Kaiser-Meyer-Olkin (KMO) test of sample adequacy and sphericity for factor analyses.The measurements obtained the following KMO measure results: Total Rewards Questionnaire -0.823, MBI-GS -0.847, UWES -0.893 and SERVQUAL -0.809.The results verified that all the measurements were factorable (Hair, Anderson, Tatham & Black, 2010).A subsequent confirmatory factor analysis was performed on the four measurements.
Confirmatory factors analysis appeared to be the obvious choice, as these measurements have been standardised in the South African context.However, because of poor model fit, the researchers decided to use exploratory factor analysis to uncover the factor structure for the measurements for this sample.Factor analysis was deemed necessary, as it was the first time that these measurements were applied to a sample of school principals in South Africa.The results of the factor analysis are reported next.

Total Rewards Scale
An exploratory factor analysis, using principal component analysis, was performed on the 17 items of this measurement.
The initial analysis showed that four factors could be distinguished, based on the eigenvalues.However, closer inspection of the scree plots showed that three factors could be distinguished.Exploratory factor analysis, using sphericity rotation, was performed, specifying three factors.The factors were labelled Performance management (Factor 1), Benefits (Factor 2) and Career development (Factor 3); these explained 56.772% of the variance.Five items were deleted because of problematic loadings.

MBI-GS
An exploratory factor analysis, using the principle component factoring extraction method, was performed on the 21 items of the MBI-GS.The factor analysis resulted in three underlying factors.An exploratory factor analysis, using varimax rotation, was conducted on the MBI-GS, specifying three factors.Two items were deleted because of problematic loadings.The factors were labelled Exhaustion (Factor 1), Professional efficacy (Factor 2) and Mental distance (Factor 3).The three factors explained 50.892% of the variance.

UWES
An exploratory factor analysis, using the principle component factoring extraction method, was performed on the 17 items of the UWES.The items of the Absorption scale showed problematic loadings, and were omitted from further analyses.The research therefore only used the Vigour and Dedication scales of the UWES in the analysis.A subsequent factor analysis was done on the 11 items of the UWES.The factor analysis resulted in one underlying factor, which was labelled Work engagement.One item was deleted because of a problematic loading.The factor explained 46.312% of the variance.

SERVQUAL
An exploratory factor analysis, using the principle component factoring extraction method, was performed on the 22 items of the SERVQUAL measurement.The items of the Tangibles scale showed problematic loadings, and were omitted from further analyses.A subsequent factor analysis was done on the 18 items of the measurement.The factor analysis resulted in three underlying factors, which were labelled Empathy (Factor 1), Assurance (Factor 2) and Responsiveness (Factor 3).
The three factors explained 49.893% of the variance.Two items were deleted because of problematic loadings.
The descriptive statistics of the measurements are reported in Table 1.
The results in Table 1 show acceptable to excellent reliabilities for all the scales of the measurements, except for the Career Development Scale of the Total Rewards Questionnaire (see Field, 2009).This scale was omitted from further analyses.
The results showed that, on average, the respondents viewed rewards as important.The results further showed that the respondents showed a low level of professional efficacy and average levels of exhaustion and mental distance.The respondents further indicated average levels of work engagement and a high orientation towards quality service delivery relating to Empathy, Assurance and Responsiveness.
The next section reports the results in terms of the hypotheses.Linear regression analysis was applied to test for predictive relationships between the variables.The results of the interactive relationship between Rewards and Burnout are reported in Table 2.
The results in Table 2 show that Performance management is a significant negative predictor of the Mental distance dimension of Burnout: F (1155) = 4.646 and accounted for approximately 17.2% of the variance of Mental distance (R 2 = 0.029, adjusted R 2 = 0.023).The effect was small.Benefits is a significant positive predictor of Professional efficacy: F (1155) = 8.920 and accounted for approximately 23.5% of the variance of Professional efficacy (R 2 = 0.055, adjusted R 2 = 0.049).The effect was small.
Based on the above results, Hypothesis 1, that there is a significant negative relationship between rewards and burnout, is partially accepted.The results of the regression analysis of the relationship between Burnout and Service quality are reported in Table 3.
The results in Table 3 show that Professional efficacy is a significant positive predictor of the Reliability dimension of Service orientation: F (1154) = 5.015 and accounted for approximately 17.8% of the variance of Reliability (R 2 = 0.032, adjusted R 2 = 0.025).The effect was small.The results furthermore showed that Mental distance is a significant positive predictor of the Assurance dimension of Service orientation, and accounted for approximately 16.8% of the variance of Reliability (R 2 = 0.028, adjusted R 2 = 0.022).The effect was small.
Based on the above results, Hypothesis 2, that there is a significant negative relationship between burnout and service quality orientation, is partially accepted.The results of the regression analysis of the relationship between Rewards and Service quality orientation are reported in Table 4.
The results in Table 4 show that Performance management is a significant positive predictor of the Reliability dimension of Service orientation: F (1154) = 7.402 and accounted for approximately 21.5% of the variance of Reliability (R 2 = 0.046, adjusted R 2 = 0.040).The effect was small.The results furthermore showed that Benefits is a significant positive predictor of the Assurance dimension of Service orientation, and accounted for approximately 18.8% of the variance of Reliability (R 2 = 0.035, adjusted R 2 = 0.029).The effect was small.Based on the above results, Hypothesis 3, that rewards are significantly positively related to service quality, is partially accepted.The results of the regression analysis of the relationship between Rewards and Work engagement are reported in Table 5.
The results in Table 5 show that Performance management is a significant positive predictor of Work engagement: F (1154) = 3.925 and accounted for approximately 15.8% of the variance of Mental distance (R 2 = 0.025, adjusted R 2 = 0.019).The effect was small.Based on the above results, Hypothesis 4, that rewards are significant positively related to the work engagement, is partially accepted.The results of the regression analysis of the relationship between Work engagement and Service quality orientation are reported in Table 6.
The results in Table 6 show no significant relationship between Work engagement and Service quality.Based on the above results, hypothesis 5, that work engagement is positively related to service quality is rejected.

Discussion
The main objective of this research was to determine whether rewards are a significant predictor of well-being (i.e.burnout versus work engagement) and service quality of school principals.We hypothesised that the availability of rewards enhances the well-being of school principals, which, in turn, has a positive effect on the quality of their service delivery.The results showed that the availability of performance management systems decreases the mental distance of school principals.In line with the findings of Boyland (2011), these results suggest that adequate performance evaluation systems may reduce incidences of stress-related illnesses such as burnout.As a result, we would expect school principals to display less callous attitudes towards their job and the individuals (i.e.school pupils, teachers and parents) who are the recipients of their service (Schaufeli et al., 2008).
The results further show that the availability of benefits enhances the professional efficacy of school principals.A school system that provides adequate benefits will therefore enhance school principals' self-belief that they are capable of performing well in their job (Bas & Yildirim, 2012).
The results of this study confirmed that professional efficacy enhances reliable service delivery of school principals.School principals who have a strong professional self-belief will therefore be able to manage and maintain school plants and culture, so that services can be delivered in an accurate and timely manner to school pupils and the broader community (Amanchukwu & Ololube, 2015).The mental distance of school principals in this study was found to be positively related to the Assurance dimension of Service quality.The results can be explained by the fact that the school principals displayed average levels of mental distance, indicating that they are not experiencing a strong detachment from the people that are working with or their work tasks.As a result, these school principals might build effective interpersonal and trusting relationships with teaching staff, school pupils, and parents (Nel et al., 2008), while walking the extra mile to do meaningful work (Scott & McMullen, 2010).
In line with previous research, the results of this study also showed that the quality of service delivery of school principals is dependent on an effective rewards system (Ashraf et al., 2014;Demerouti et al., 2014).The availability of a performance management and subsequent acknowledgement system will increase the accurate and reliable service delivery of school principals (Billger, 2007).The availability of benefits enhanced the Assurance dimension of Service quality of school principals, which enables them to instil confidence and trust in teachers to deliver quality services through innovation and creativity (Koch et al., 2015).
Performance management was also found to be positively related to work engagement.This implies that job resources in the form of performance feedback and appreciation will increase the work engagement of school principals (Bakker et al., 2007(Bakker et al., , 2008)).In contrast with previous research, work engagement was not found to be significantly related to the quality of service of school principals (Bakker & Demerouti, 2008;Salanova et al., 2005).

Practical implications
School principals play an important role in the quality service delivery of schools.The results of this study highlighted the importance of an adequate school reward system that includes adequate performance management systems, performance evaluation, and performance feedback.As shown by the results, performance management plays an important role in decreasing job burnout, and enhances work engagement and subsequent service quality.Furthermore, the availability of benefits was found to increase school principals' sense of professional self-efficacy, which, in turn, enhances reliable service delivery and quality.Performance management, in conjunction with benefits, is therefore imperative for the effective functioning of school principals.

Limitations
The study had some limitations.Firstly, a cross-sectional research design was used, which limited the research in terms of making cause-and-effect inferences over the long term.The study used self-assessment as the method of data collection.This means that the results are based on the perceptions of the participants.Only questionnaires were used to collect the data; other methods such as observation and interviewing could have been used to obtain another perspective.

Recommendations
This study has implications for future research.Larger studies using longitudinal data representing all the principals in the country are needed to provide additional evidence in support of our main findings.It should be noted that the study findings can be relevant to all levels of management, from line-and middle-to top-management, in various sectors of government.From a methodological point of view, the items of the SERVQUAL measure should be revisited to present an accurate reflection of the five dimensions of Service quality.The analyses of the SERVQUAL measure in this study resulted in various cross-loadings, rendering factors such as Tangibles and Empathy unusable.Furthermore, the items of the Total Rewards Questionnaire should be expanded to give a more accurate reflection of the five dimensions of the Total Rewards Model that formed the foundation of this research.

Conclusion
In conclusion, schools need effective principals to succeed and deliver quality services.It is vital that the Department of Education look after school principals, to curb the brain drain.Researchers have revealed that many organisations, particularly in the public sector, fail to address the needs of their employees, which causes them to be victims of burnout and eventually quit.It is quite evident that many educators, including principals, quit because of poor remuneration, which has led to a shortage of skilled personnel in schools.It is of paramount importance that schools look closely at the impact of reward on the wellbeing of school principals and their service quality.
Schaufeli, Leiter, Maslach and Jackson (1996) school principals.The questionnaire consists of 17 statements that measure five components (i.e.Pay, Benefits, Performance management, Worklife balance and Career development) of the World at Work Model for Total Rewards in Organisations.Responses are measured on a five-point Likert scale ranging from 1 (Not important at all) to 5 (Extremely important).Moore and Bussin (2011)found an overall reliability of 0.821 for the measurement.The Maslach Burnout Inventory -General Survey (MBI-GS) ofSchaufeli, Leiter, Maslach and Jackson (1996)was used to measure the Exhaustion, Cynicism and Professional efficacy dimensions of Burnout.The Depersonalisation dimension of the Maslach Burnout Inventory Educator Survey (MBI-ES) was also included in the questionnaire.Responses to 21 items were given on a six-point scale varying from 0 (Never occurs) to 6 (Occurs every day).The questionnaire was validated in the South African context

TABLE 1 :
Descriptive statistics of the measurements.

TABLE 2 :
Regression analyses of rewards and burnout.

TABLE 3 :
Regression analyses of rewards and service quality orientation.

TABLE 5 :
Regression analyses of rewards and work engagement.

TABLE 4 :
Regression analyses of burnout and service quality orientation.

TABLE 6 :
Regression analyses of work engagement and service quality.