Tracking the employee satisfaction-life satisfaction binary: The case of South African academics

The South African higher education landscape has experienced extensive transformation since the emergence of democracy in 1994. The creation of a three-tier university system consisting of traditional universities, new generation universities and universities of technology, the creation of new regulatory bodies, the changing nature of student and staff distribution, the changing models of instructional delivery and intensified research emphases are some of the hallmark undercurrents characterising these dynamics (Jansen, 2004). The transformations were intended to enable higher education institutions in the country to produce outcomes that are in line with the dynamic needs and demands of the contemporary society (Zohreh, Nadergholi & Alia, 2011). Be that as it may, the transformation process remains hamstrung by a cocktail of challenges, one of which is the shortage of academics at universities (Council on Higher Education, 2011). This shortage may be attributed, in part, to high attrition rates triggered by the failure to satisfy psychological contracts of such academics (Pienaar & Bester, 2006). High turnover of academics obviously has severe repercussions for the South African higher education sector, which is currently in its growth stage (Muller, 2013). To address this challenge, most academic institutions in South Africa have had to contend with the momentous task of recruiting and retaining their academics (Samuel & Chipunza, 2013). A possible solution to this challenge, though, could be to provide employment conditions that facilitate the satisfaction of the needs of university academics, with a view to retaining them.


Introduction
The South African higher education landscape has experienced extensive transformation since the emergence of democracy in 1994.The creation of a three-tier university system consisting of traditional universities, new generation universities and universities of technology, the creation of new regulatory bodies, the changing nature of student and staff distribution, the changing models of instructional delivery and intensified research emphases are some of the hallmark undercurrents characterising these dynamics (Jansen, 2004).The transformations were intended to enable higher education institutions in the country to produce outcomes that are in line with the dynamic needs and demands of the contemporary society (Zohreh, Nadergholi & Alia, 2011).Be that as it may, the transformation process remains hamstrung by a cocktail of challenges, one of which is the shortage of academics at universities (Council on Higher Education, 2011).This shortage may be attributed, in part, to high attrition rates triggered by the failure to satisfy psychological contracts of such academics (Pienaar & Bester, 2006).High turnover of academics obviously has severe repercussions for the South African higher education sector, which is currently in its growth stage (Muller, 2013).To address this challenge, most academic institutions in South Africa have had to contend with the momentous task of recruiting and retaining their academics (Samuel & Chipunza, 2013).A possible solution to this challenge, though, could be to provide employment conditions that facilitate the satisfaction of the needs of university academics, with a view to retaining them.
The purpose of the current study is to investigate the relationship between employee satisfaction and life satisfaction amongst university academics in South Africa.To achieve this, the study measures the impact of five employee satisfaction factors, specifically workplace flexibility, skills utilisation, teamwork, remuneration and autonomy, which have been given emphasis in previous higher education research (e.g.Filiz, 2013;Ignat & Clipa, 2012;Jusoff, Hussein, Ju & Din, 2009), on life satisfaction amongst a cohort of South African university academics.There are several motivations for conducting this study.Firstly, although research interests focusing on the employee satisfaction of academics are evidently growing, they are dominated by evidence from either Western European or North American environments.Consequently, research is needed in order to understand how employee satisfaction relates to other factors in the context of academics in developing countries such as South Africa.Secondly, a number of previous studies that have examined the subject of satisfying the needs of academics in South Africa (e.g.Nieuwenhuizen, 2011: Rasool & Botha, 2011;Schulze, 2006) have excluded the dimensions that are examined in this study.As such, the conceptual framework that is put forward in this study is yet to be tested within the South African higher education environment.Thirdly, as suggested by Mokoditoa (2011), the attraction and retention of academics involves, amongst other things, paying particular attention to specific elements of organisational behaviour that are linked to employee motivation.Employee satisfaction and life satisfaction occupy an important niche in organisational behaviour.This presents a need to conduct research that generates knowledge on the behaviour of academics, with the aim of retaining them within the South African academic sector.In sum, it is necessary to continue generating new knowledge on any topic within any discipline of organisational behaviour, in order to update organisational theory (Taylor & Hansen, 2005).

Research theory
There are three dominant theoretical explanations accounting for the relationship between employee satisfaction and life satisfaction.The segmentation theory (Loscocco & Roschelle, 1991) postulates that there is no relationship between employee satisfaction and life satisfaction.The theory is premised on the assumption that job-related variables experienced by employees tend to be kept separate from other non-work activities.The second model is the compensatory theory (Bamundo & Kopelman, 1980), which assumes that the two constructs are negatively related such that employees who are dissatisfied at work are more likely to compensate for this dissatisfaction by engaging in more satisfying nonwork activities.The third theory is the spillover model, which maintains that there is a positive correlation between employee satisfaction and life satisfaction (Westman, 2006).According to Ilies, Wilson and Wagner (2009), most contemporary research tends to give more support to the spillover theory than to the other two theories.However, despite this empirical support, a contentious and unresolved issue associated with the spillover model relates to the direction of the relationship between employee satisfaction and life satisfaction.Notwithstanding this fact, the spillover model was adopted as the nominal anchor in the current study.

Employee satisfaction
Scholars have conceptualised employee satisfaction from different viewpoints, culminating in a number of definitions.Bernhardt, Donthu and Kennett (2000) define employee satisfaction as a personal evaluation of the present conditions of the job as well as outcomes that arise as a result of having a job.Koys (2003) conceptualises employee satisfaction in terms of the employees' perception and evaluation of the job.As defined by Wangenheim, Evanschitzky and Wunderlich (2007), employee satisfaction is a function of the perceived relationship between what one wants from one's job and what one perceives it as offering.Chi and Gursoy (2009) define employee satisfaction in terms of the extent to which employees like their work.The most prevailing precursors to employee satisfaction include environmental antecedents and personal factors (Matzler & Renzl, 2007).Environmental antecedents include all factors that are associated with the job itself as well as the work atmosphere, whereas personal factors give emphasis to individual attributes and characteristics of employees (Judge, Thoresen, Bono & Patton, 2001).
It can be argued that the importance of employee satisfaction lies in the fact that organisational success is determined, to a large extent, by the efforts of all employees.Employee satisfaction has been linked to a number of desirable outcomes in the organisation, such as increased quality of outputs and productivity (Faragher, Cass & Cooper, 2005), enhanced employee confidence, commitment and loyalty (Yee, Yeung & Cheng, 2008), decreased turnover (Goslin, 2005) and improved organisational citizenship behaviours (Shields, 2006).As such, organisational prosperity is likely to be greater when employee satisfaction is higher.In terms of its measurement, the Minnesota Satisfaction Questionnaire (MSQ) (Spector, 1997) has emerged as arguably the most popular tool for measuring the employee satisfaction.This instrument measures employee satisfaction on 20 five-item scales.Accordingly, this instrument was adopted for the purposes of measuring employee satisfaction in the current study.

Life satisfaction
There is a general lack of unanimity on the exact definition of life satisfaction.Heller, Judge and Watson (2002) define life satisfaction as an emotional reaction of an individual to life, consisting of work time, spare time and time after work, as well as expressing the individual's satisfaction about life.Rode (2004) conceptualises life satisfaction in terms of the cognitive element of the subjective well-being of people.As defined by Graves, Ohlott and Ruderman (2007), life satisfaction may be perceived as a measure of one's appraisal of life in general.Ye, Yu and Li (2012) define life satisfaction as an individual's evaluation about their quality of life.Rode's definition was adopted for the present study since it focuses primarily on employees.
Primary antecedents to life satisfaction include personality, genetic and social-cognitive factors such as goaldirected activity, self-efficacy, outcome expectations and environmental support (Buetell, 2006).Other conditions that influence life satisfaction include demographic, http://www.sajip.co.zaOriginal Research environmental and interpersonal factors (Koohsar & Bonab, 2011).Furthermore, taking pleasure in life, finding life meaningful, consistency on the matter of reaching goal satisfaction, positive individual identity, physical fitness, economic security and social relationships are all important life satisfaction indicators (Judge et al., 2001;Lent, Taveira, Sheu & Singley, 2009;Saari & Judge, 2004).
Life satisfaction is positively related to a number of personal and work-related factors that include inter alia selfmotivation, optimism and resilience (Palmer, Donaldson & Stough, 2002), commitment and loyalty (Nickerson & Nagle, 2005), high productivity (Guney, Kalafat & Boysan, 2010), high employee satisfaction (Bastug & Duman, 2010), quality of work life (Ozer & Saçke, 2011) and improved family and peer relations (Zhang & Howell, 2011).In view of this, it is an important possibility that organisations stand to reap immense benefits by cultivating high levels of life satisfaction amongst their human resources.In terms of its measurement, the Satisfaction with Life Scale (SWLS) developed by Diener, Emmons, Larsen and Griffin (1985) has stood out as the most popular tool for measuring life satisfaction.This instrument was therefore adopted for the purposes of measuring life satisfaction in the current study.

Employee satisfaction and life satisfaction
The relationship between employee satisfaction and life satisfaction has been a subject of prominence for a host of research practitioners (e.g.Cummings & Worley, 2005;Eskildsen & Dahlgaard, 2000;Georgellis, Lange & Tabvuma, 2012;Mafini, Surujlal & Dhurup, 2012).In general, employee satisfaction is regarded as an antecedent to the overall life satisfaction of employees to the extent that employees who are not happy in their non-work activities will not be happy in their work activities (Ghiselli, La Lopa & Bai, 2001;Zhao, Qu & Ghiselli 2011).The same results have been noted within the realm of academia.For instance, Jusoff et al. (2009) found a positive association between the employee satisfaction and life satisfaction of both academic and non-academic staff in a Malaysian higher education institution.A positive interconnection between employee and life satisfaction was also observed in a study conducted amongst academic staff at a Turkish university (Feliz, 2013).Moreover, the results of a study conducted by Ignat and Clipa (2012) that sampled Romanian academics confirms the existence of a positive relationship between employee satisfaction and life satisfaction.Therefore, employees in academia and beyond who have a positive perception of their jobs typically evaluate their overall well-being in life better and are likely to be more satisfied with their life in general.

Employee satisfaction factors and life satisfaction
As has been highlighted before, employee satisfaction can be measured through at least 20 sub-factors.However, there currently exists a wide body of scholarly research that appears to place major emphasis on workplace flexibility, skills utilisation, teamwork, remuneration and autonomy as central employee satisfaction factors that determine life satisfaction in the higher education environment.For example, studies by Cech and Blair-Loy (2014), Pienaar and Bester (2006) and Phillips and Connell (2003) point to workplace flexibility as a factor of significant importance for many faculty members in that it has a strong impact on life satisfaction, job loyalty and organisational citizenship, amongst other factors.With reference to skills utilisation, some studies focusing on employee satisfaction (e.g.Johnsrud, 2002;Volkwein & Zhou, 2003) classify skills utilisation amongst intrinsic motivators that are essential ingredients for satisfaction in other areas of an academic's life.Schulze (2006) likewise identifies high skills utilisation as one of the numerous factors that can have a positive influence on the life satisfaction of academic staff members.Additionally, Holub (2003) concludes that the level of life satisfaction that is experienced by university lecturers is greatly influenced by various determinants, amongst them skills utilisation.
In terms of remuneration, a number of studies (Kipkebut, 2010;Rosser, 2004;Shoaib, Noor, Tirmizi & Bashir, 2009;Tettey, 2006) conducted within various academic environments have established that remuneration is associated with other employee-related factors such as life satisfaction, productivity, job security and general well-being.In respect of teamwork, the results of studies conducted by a number of scholars (Abdullah, Ahsan, Yong Gun Fie & Shah Alam, 2009;Riley, 2009) reveal that academic staff members are likely to experience stress, low employee and life satisfaction as a result of a lack of a camaraderie in the workplace.In previous research carried out by Weiss (2002), the lack of teamwork was found to be one of the factors contributing to the deep feeling of pessimism that is heavily pronounced amongst academics at South African higher education institutions.With regard to autonomy, the findings of several studies (Daly & Dee, 2006;Dockel, 2003;Ngethe, Iravo & Namusone, 2012) attest that autonomy is positively linked to life satisfaction of academics, through its stimulus effect on the satisfaction of higher order needs for achievement and accomplishment.
The literature reviewed in the foregoing sections presupposes the existence of affirmative associations between employee satisfaction factors and life satisfaction amongst academics.These presuppositions make it logical to anticipate significant positive relationships between life satisfaction and the specified employee satisfaction factors.Therefore, the following hypotheses within the present study are put forward.
Hypothesis 0: Life satisfaction is not significantly and positively related to the five employee satisfaction factors, namely workplace flexibility, skills utilisation, teamwork, remuneration and autonomy, amongst academics.Hypothesis 1: There is a significant positive relationship between workplace flexibility and life satisfaction amongst academics. doi:10.4102/sajip.v40i2.1181http://www.sajip.co.za

Original Research
Hypothesis 2: There is a significant positive relationship between skills utilisation and life satisfaction amongst academics.Hypothesis 3: There is a significant positive relationship between teamwork and life satisfaction amongst academics.Hypothesis 4: There is a significant positive relationship between remuneration and life satisfaction amongst academics.Hypothesis 5: There is a significant positive relationship between autonomy and life satisfaction amongst academics.

Conceptual framework
The conceptual framework illustrated in Figure 1 is proposed, which reveals the linkage between the five employee satisfaction dimensions and life satisfaction.

Research design
In the current study, a quantitative design using a crosssectional survey method was adopted.The survey method was utilised because of its empirical nature, the ability to produce a large amount of data in a short time at a low cost and the fact that it allows for the generalisation of results since it has a broader breadth of coverage of people or events (Story, Ranong & Jones, 2011).

Participants and sampling
The sample used in this study was made up of 273 academics drawn from three universities (two traditional universities and a university of technology) located in the Gauteng province of South Africa.Respondents were selected using the non-probability convenience sampling technique.The convenience sampling technique was deemed appropriate for this study as it was difficult to locate the academics at the same place and time due to their busy work schedules.The sample size was determined using Green's (1991) rule of thumb, which specifies that a minimum of 50 respondents are sufficient for a correlation or regression, with the number increasing as the number of independent variables (IVs) increases.The implication of Green's rule in the current study is that a minimum of 250 respondents was required since five independent variables were used.Based on this prescription, the eventual sample size of 273 was considered satisfactory for the purposes of the study.The demographic characteristics of the respondents are illustrated in Table 1.
An analysis of the demographic profile of respondents (Table 1) reveals that approximately 59% (n = 162) of the respondents were male whereas approximately 41% (n = 111) were female.After collapsing the respondents' ages, it emerged that the largest group of respondents (81%; n = 220) were aged between 26 and 45 years.Additionally, approximately 47% (n = 127) of the respondents were in possession of a master's degree.In terms of racial group, black people (53%; n = 144) were the highest number.With regard to the position occupied, the largest group of respondents (45%; n = 123) was lecturers, followed by junior lecturers (23%; n = 62), with full professors (3%; n = 7) being the smallest number.

Data collection procedures and measurement instruments
Data were collected through the use of a three-section selfadministered questionnaire.Section-A of the questionnaire requested the respondents to provide their demographic information.In Section-B, respondents were requested to indicate their level of employee satisfaction on a five-point Likert scale which ranged from 1 (very dissatisfied) to 5 (very satisfied).An adapted version of the validated MSQ (Spector, 1997) was used in this section.Section-C used the SWLS (Diener et al., 1985) to elicit responses regarding the respondents' perceived level of well-being and happiness.Since life satisfaction is a complex construct that has to be assessed within a specific context (Chang & Sanna, 2001), the  work environment was used for this study.Likert scales were used in Sections-B and Section-C because they offer a number of advantages such as simplicity, versatility, universal applicability, greater accuracy and ease of administration (Carifio & Perla, 2007).
Before its administration, the items in the questionnaire were reviewed by three academics from a South African university of technology who are experts in the field of industrial psychology.In addition, a pretest was conducted using a conveniently selected sample of 20 academics.
The feedback obtained from both the panel of experts and the pretest sample was used as the basis for adapting the questionnaire in order to ensure that it would validly capture the required information, in accordance with Olson's (2010) recommendation that questionnaire adaptation ought to be based on the input from both expert reviewers and the pretest sample.The amendments included substantive adjustments to such aspects as language used, grammar, the level of difficulty in each question and reduction in the number of items used.More to this, only items that are relevant to the constructs examined in this study were used.Eventually, Section-A of the questionnaire consisted of six questions (Table 1) whilst Section-B consisted of 25 questions (Table 2).The final distribution of questions in Section-B was as follows: four items for workplace flexibility, five items for skills utilisation, four items for teamwork, four items for remuneration, three items for autonomy and five items for life satisfaction.
Data collection was conducted between January 2013 and March 2013.Prior to the administration of the questionnaire, permission to conduct the study was sought and granted by the ethical committees of the three institutions involved.Two trained postgraduate students at the Vaal University of Technology were employed as research assistants.The students were paid using funds availed through the Research Stimulation Fund at the university of technology.Contacts established at the three institutions were also instrumental in assisting in the distribution of the questionnaires.
Respondents participated in the study on a voluntary basis and were assured that the information they provided would be used solely for academic purposes and that their identities would remain anonymous at all times.Out of the 450 questionnaires that were initially distributed, 284 questionnaires were returned.However, 11 questionnaires were eliminated in the screening process because they were either incomplete or had more than one response on a number of questions and were thus spoiled.Ultimately, 273 questionnaires were used in the data analysis, giving a response rate of approximately 61%, which satisfies Fincham's (2008) recommendation that a minimum response rate of 60% is necessary to ensure credible results in a survey.

Analysis of the results
Data analysis was conducted using a combination of the Statistical Packages of the Social Sciences (SPSS version 21.0) and AMOS 15 software.The results obtained in the current study are presented in the following configuration: the confirmatory factor analysis is presented first, followed by the psychometric properties of the constructs.This is succeeded by a section detailing the results of the correlation analysis and regression analysis.

Confirmatory factor analysis
Since the scales used in this study were validated in other contexts, confirmatory factor analysis (CFA) using AMOS 15.0 software was performed in order to ensure that these scales were in fact relevant and reliable within a South African context.The CFA and the model fit of the constructs with multiple indicators (workplace flexibility, skills utilisation, teamwork, remuneration and autonomy) were determined to be acceptable.A number of indices suggested by Bagozzi and Yi (1998) were adopted for the purposes of assessing the model fit.These included CMIN/DF (< 0.3), goodness-offit index (GFI ≥ 0.90), comparative fit index (CFI ≥ 0.90), Tucker and Lewis index (TLI ≥ 0.90) and root mean square error of approximation (RMSEA ≤ 0.08).The CMIN/DF was 1.810, GFI 0.920, CFI 0.961, TLI 0.948 and RMSEA 0.06, which indicates satisfactory fit of the model (Hulland, 1999).

Psychometric properties of constructs
The means, standard deviations, inter-construct correlations, square root of average variance extracted (AVE) scores, standardised factor loadings and composite reliability (CR) estimates are reported in Table 2.
As shown in Table 2, Cronbach's alpha values for all constructs were satisfactory, ranging from 0.736 to 0.864.Furthermore, CR estimates ranged from 0.836 to 0.945, which indicates acceptable construct reliability.All items loaded strongly (i.e.all standardised loadings > 0.50) and significantly (p < 0.01) on respective constructs.Thus, adequate construct validity can be assumed in the current study.Clark and Watson (1995) suggest that a valid measure of a construct requires inter-item correlations in the range of 0.40 to 0.50.The inter-item correlations ranged from 0.511 to 0.749 for all the constructs, thereby providing evidence of construct validity.A measure is said to possess convergent validity if the instrument correlates highly with other variables with which it should theoretically correlate (Leo, Bennet & Hartel, 2005).Convergent validity was assessed using the average variance extracted from each construct, which were well above 0.5 (Chin, 1998), thereby affirming convergent validity.Content and face validities were ascertained in the pretest of the questionnaires.Predictive validity was assessed through regression analysis (Table 4), in which three of the five employee satisfaction factors were statistically significant, thereby reflecting sufficient levels of predictive validity in the scale.
A further analysis of Table 2 reveals that mean scores ranging between 3.189 and 3.974 (out of 5.0) were computed for all the constructs examined in the study.These scores depict a collective inclination towards both the 'agree' and 'strongly agree' positions on the Likert scale, which demonstrates that the academics had satisfactory levels of both employee satisfaction and life satisfaction.Skills utilisation (x = 3.974; P < 0.05) emerged as the highest-ranking factor in terms of the mean scores, followed by workplace flexibility (x = 3.812; P < 0.05), remuneration (x = 3.673; P < 0.05), autonomy (x = 3.547; P < 0.05) and teamwork (x = 3.189; P < 0.05).This result implies that academics were more satisfied with skills utilisation and least satisfied with teamwork amongst the five employee satisfaction factors extracted in the study.

Correlations: Employee satisfaction and life satisfaction
The Kendall's tau correlation coefficient (Agresti, 2010) was used to examine the strength and direction of the association between the five employee satisfaction domains and life satisfaction.The results are presented in Table 3.
An analysis of the correlation matrix (Table 3) reveals that there were positive correlations, anchored by a τ value of 0.26 (P < 0.900) and a τ value of 0.78 (P < 0.059), between life satisfaction and all five employee satisfaction factors.This result shows that an increase in any one employee satisfaction factor can actuate increases in life satisfaction amongst university academics.In contrast, any decrease in any employee satisfaction factor may lead to a decrease in life satisfaction.

Regression analysis: Employee satisfaction and life satisfaction
Since there were positive associations between the five employee satisfaction dimensions and life satisfaction, multiple regression analysis using the 'enter' method (Malhotra, 2010) was employed to ascertain whether the employee satisfaction factors predicted life satisfaction.A principal assumption that was made to justify the use of linear regression analysis for purposes of predicting the existence of relationships amongst the constructs is the existence of linearity between employee satisfaction factors and life satisfaction.V ariance inflation factor (VIF) values ranged between 1.156 and 2.210, which is within the maximum permissible threshold of 10, as prescribed by Field (2009).Additionally, tolerance values ranged between 0.509 and 0.795, which were acceptable since they were above the recommended minimum of 0.5 (Denis, 2011).Therefore, multicollinearity diagnosis did not signal a serious threat, which made a regression analysis both feasible and appropriate.The results of the regression analysis are reported in Table 4.   accounted for by other factors that were not examined in this study.

Discussion
Hypothesis 1 was supported and subsequently accepted in the present study.For confirmation, the results of the correlation analysis exhibited a strong positive correlation (τ = 0.61; P = 0.100) between workplace flexibility and life satisfaction.Complimentary to this, in the regression analysis, workplace flexibility (β = 0.324; t = 4.615; P < 0.001) emerged as statistically significant.The finding of the correlation analysis demonstrates that the life satisfaction of academics increases as workplace flexibility increases.Furthermore, the result of the regression analysis illustrates that workplace flexibility is an indicator of life satisfaction amongst academics.Flexible working is associated with a number of positive outcomes that tend to boost life satisfaction amongst academics.For instance, Goulden, Mason and Frasch (2011) found that flexible working was a positive factor in achieving better work-life balance amongst faculty staff.The majority of academics in a study conducted by Drago et al. (2006) reported that flexible working had a positive effect in reducing and managing stress levels, which tends to increase life satisfaction.Flexible employees were found to have higher levels of both employee satisfaction and life satisfaction in a study conducted by Turkyilmaz, Akman, Ozkan and Pastuszak (2011).Workplace flexibility has been associated with job security of academics as well as the opportunity to embark on personal and professional development programs (Bunton & Corrice, 2011).Mederer and Silver (2011) further advocate that workplace flexibility is a reliable instrument for the attraction, retention, engagement, wellness and productivity of top academic talent.It appears then that workplace flexibility occupies an important place as a tool for meeting the needs of academic faculty staff.
Hypothesis 2 found support and was accepted in the present study.This decision was premised on the fact that the correlation matrix indicated a strong positive association (τ = 0.78; P = 0.059) between skills utilisation and life satisfaction.Further to this, the scores in the regression analysis indicate that skills utilisation (β = 0.227; t = 2.427; P < 0.000) is statistically significant.The result of the correlation analysis implies that life satisfaction amongst academics increases with an increase in the extent to which the skills of these academics are used in the workplace.
The result of the regression analysis shows that the degree to which the competencies or skills of academics are utilised is an important reflector of the extent to which the same academics are satisfied with life.Consistent with this finding, a study by Chung et al. (2010) concludes that increased opportunities for faculty members to apply their abilities at work may result in higher levels of life satisfaction.
In an earlier study conducted by Zarafshani and Alibaygi (2008), skills utilisation was found to be a significant predictor of both employee satisfaction and life satisfaction of Iranian academics.Additionally, a study by Bender and Heywood (2006) that investigated satisfaction levels of PhD-level academics reveals a positive interconnection between skills utilisation and life satisfaction.In the African context, intrinsic facets of work, inclusive of skills utilisation were found to be indicators of the life satisfaction of Ugandan academics (SSesanga & Garret, 2005).It is a logical supposition then that in order to enhance the life satisfaction of academic employees, one has to positively adjust their degree of skills utilisation.
With regard to teamwork, Hypothesis 3 was not supported and was therefore rejected.For endorsement of this outcome, reference is given to the correlation matrix, which revealed a weak correlation (τ = 0.26; P = 0.900) between teamwork and life satisfaction.Consistently, teamwork (β = -0.243;t = -3.757;P < 0.156) was statistically insignificant in the regression analysis.The results of the correlation analysis suggest that when teamwork amongst academics increases, there will be very insignificant (immaterial) changes in their life satisfaction.The negative beta associated with teamwork in the regression model indicates that teamwork cannot be used as an indicator for life satisfaction amongst academics.It appears that teamwork does not seem to influence the life satisfaction of academics in the context of the present study.This result may be astonishing, since it contradicts the conventional view that teamwork within the organisation is an indispensable stimulant of both employee and organisational performance (Griffin, Patterson & West, 2001;Kalisch, Lee & Rochman, 2010;Turman, 2003).A plausible explanation for this result could be the existence of manifest conflict amongst academics who work in the same institution.It is well acknowledged that: debate and dissent are the lifeblood of universities and some of our greatest scholars pride themselves on being uncompromising, which makes conflict inevitable in such situations from time to time.(West, 2006, p. 187) Debates and controversies are commonplace within academic circles.Intellectual competition is also rife as academics jostle for intellectual honours associated with the profession (Hyland, 2009).As noted by Volpe and Chandler (2001), even collaborations between academics can be associated with a plethora of challenges as divergent opinions emerge.
To avoid conflict stemming from partnerships, some academics tend to prefer collaborating with academics from other institutions rather than with those from their own institutions (Clark, 2004).The existence of conflicting personalities and perspectives amongst academics compels some to resort to individual efforts, rather than teamwork (White & Lowenthal, 2011).Issues of compatibility with each other also emerge as the academic arena in South Africa has become very diverse in terms of age, race and even nationality (Mayekiso & Snodgrass, 2008).Such complexities tend to stifle teamwork amongst academics, resulting in a weak relationship between teamwork and life satisfaction amongst academics.Measures have to be implemented to ensure that teamwork or partnering occupies its rightful place as a promoter of employee satisfaction and life satisfaction in academic circles.
Hypothesis 4 received partial support in this study.As evidence for this outcome, a moderate positive correlation (τ = 0.43; P = 0.705) was observed between remuneration and life satisfaction.However, the regression matrix indicated that remuneration (β = 0.056; t = 1.326;P < 0.789) was not a statistically significant predictor of life satisfaction.The results of the correlation analysis denote that changes in the levels of remuneration are likely to trigger moderate (neither low nor high) changes in the life satisfaction of academics.Despite this positive association, the results of the regression analysis disclose that remuneration is not an indicator (predictor) of the life satisfaction of academics.The findings of the current study seem to confirm the respected traditional opinion by Herzberg (1966) that money is a hygiene (essential) factor but is not a motivator.In parallel, Amar (2004) concludes that the ability of money to motivate skilled employees is almost negligible.In a recent study by Samuel and Chipunza (2013), it was found that remuneration does not influence the turnover of South African academics.As such, the findings of the current study do not seem to be surprising.Notwithstanding this fact, it may be interesting to note that there is a disparity in the remuneration of lowerlevel and senior academics in South Africa.For instance, the entry-level remuneration for academics in South Africa remains depressed (Kgosana, 2011).On the other end, the remuneration of top-level academics in South Africa remains competitive when compared with countries such as the United Kingdom, United States, Canada and Australia (Jansen, 2013).It is possible then that this existent dichotomy places strain on the extent to which remuneration as an employee satisfaction factor is able to predict life satisfaction.For instance, the lowly paid entry-level academics may be less satisfied with life, in contrast with the highly paid toplevel academics, which may create a mismatch between the performances of the two cohorts.
In the present study, Hypothesis 5 found support and was accepted.To sustain this decision, a strong positive association between autonomy (τ = 0.66; P = 0.059) and life satisfaction was observed in the correlation analysis.Autonomy also emerged as statistically significant (β = 0.186; t = 1.219;P < 0.000) in the regression analysis.The results of the correlation test indicate that significant increases in the life satisfaction of academics can be observed when the degree of their autonomy at work is enhanced.The results of the regression analysis demonstrate that the level of autonomy amongst academics is an important indicator of the depth of life satisfaction amongst these academics.The findings of the current study are in line with previous research by a number of scholars (Finn, 2001;Martensen & Gronholdt, 2001;Turkyilmaz et al., 2011) who found that autonomy was a significant factor that contributed to both employee satisfaction and life satisfaction.Moreover, it is natural for academics to require high levels of autonomy.In South Africa, Section 16 of the Constitution (1996) offers specific protection to the autonomy of academics through the support of academic freedom in higher education.Moreover, as highlighted by Slabbert (2003), the lack of autonomy in higher education institutions is a worrisome issue because the intellectual life of academics is about the non-stop subversion of orthodoxy and dogmatism.As such, management practices that include the micromanaging of academics or subjecting them to successions of operational rules in their line of work would have a negative impact on their well-being at work (Oshagbemi, 1996).It is a normal expectation, then, that academics would feel satisfied in their jobs and with their lives when they enjoy a high degree of freedom in the execution of their duties.

Limitations and implications for further research
Although the findings of the present study constitute an up-to-date source of knowledge and information on the relationship between employee satisfaction and life satisfaction amongst university academics, two limitations associated with the study are worth mentioning.It is expected that the findings of the study cannot be generalised to other geo-spatial contexts, since the results are based on a limited sample of 273 academics that were based at three South African institutions only.Another limitation is the use of the convenience sampling technique, which increased the study's susceptibility to sampling bias.
Several implications for further research can be highlighted.The study could be replicated using amplified sample sizes drawn from other universities that were not included in this study.Future studies could also examine the relationship between individual employee satisfaction factors and life satisfaction.The relationship between specific demographic factors such as gender, employment period, type of employment, level of education, age group and either employee satisfaction or life satisfaction provides fertile ground for possible future studies.It would also be interesting to conduct longitudinal studies on the same subject, so that trends on the employee satisfaction-life satisfaction nexus within the South African academic environment can be captured at different periods of time.Future research endeavours could examine the relationship between employee satisfaction and life satisfaction amongst academics using other employee satisfaction factors that were not used in the current study.

Managerial implications
Managers in higher education institutions could refer to the findings of the study for information in solving problems related to labour relations, motivation and human performance within the rank and file of academics.Since both employee satisfaction and life satisfaction are positively associated with a multiplicity of desirable personal and organisational outcomes, the expediting of these factors is likely to have a stimulus effect on the outcomes.Academic institutions should reinvent their cultures by the adoption of contemporary best practice that includes such activities as the continued entrenchment of flexitime, reduced workloads (e.g.fewer students and teaching hours) and increased time off for academics to address personal and family issues, sick leave, paid time off to care for children, sabbaticals and virtual offices (working from home) in order to enhance workplace flexibility (Friedman, 2012).
Regular employee satisfaction surveys may be necessary to tap the views of academics on their perceptions on the extent to which their skills are being used in their environments (Riddell, Edward, Raffe, Tett & Weedon, 2008).To improve remuneration, salaries should be increased to attract and retain academics in all levels of the hierarchy.The stimulation of academic entrepreneurship could be adopted to provide third-stream income to all academics.Furthermore, funds earned from research efforts can be availed as income to the academics involved.To bolster teamwork amongst academics, interventions such as team building programs could be implemented on a regular basis to encourage more teamwork and to flush out the dysfunctional aspects associated with current teamwork efforts (Ogbonna & Harris, 2004).To enhance autonomy, it is necessary for managers to cultivate environments in which academic freedom is promoted amongst academics, so that academics can be free to apply their intellectual resources without fear of censure.Individual efforts and innovation could also be encouraged in academic institutions and those that excel in these endeavours should be rewarded openly.

Conclusions
The aim of this study was to investigate the relationship between employee satisfaction and life satisfaction.The study was conducted within the ambit of university academics in South Africa, using five employee satisfaction factors, namely workplace flexibility, skills utilisation, teamwork, remuneration and autonomy.Academics were most satisfied with skills utilisation and least satisfied with the level of teamwork.The results of the correlation analysis showed strong positive associations between life satisfaction and three employee satisfaction factors, namely workplace flexibility, skills utilisation and autonomy.A moderate positive correlation was observed between remuneration and life satisfaction and a weak positive correlation was observed between teamwork and life satisfaction.In the regression analysis, it emerged that workplace flexibility, skills utilisation and autonomy were significant predictors of life satisfaction amongst academics.Based on the foregoing results, two primary conclusions can be made that address the aim of the study.Firstly, the employee satisfaction of university academics is positively associated with their life satisfaction.As such, the degree of life satisfaction amongst university academics runs parallel to their existing levels of employee satisfaction.Secondly, employee satisfaction amongst university academics predicts, albeit partially, their life satisfaction.This means that some employee satisfaction factors (in this case workplace flexibility, skills utilisation and autonomy) may be used as indices for life satisfaction, whereas some factors (in this case teamwork and remuneration) may not serve as indicators of the life satisfaction of academics.

FIGURE 1 :
FIGURE 1: Conceptual framework for employee satisfaction and life satisfaction.The framework depicts that workplace flexibility, skills utilisation, teamwork, remuneration and autonomy are employee satisfaction dimensions, all of which are positively related to life satisfaction.

Table 4
reflects that the adjusted R 2 was 0.428, which denotes that employee satisfaction explained approximately 43% of the variance in life satisfaction.The remaining 57% is

TABLE 2 :
Composite reliability and accuracy analysis statistics.

TABLE 3 :
Kendall's tau correlation coefficients for association between employee satisfaction and life satisfaction.
a, Correlation is significant at the 0.01 level.

TABLE 4 :
Regression analysis: Employee satisfaction and life satisfaction.