Extrinsic rewards in family businesses : Perspectives of nonfamily employees

The rewards employees receive in exchange for their labour play an important role in influencing their levels of job satisfaction and organisational commitment, and ultimately in retaining their services. The issues surrounding rewards are, however, complex in the field of family business. The primary objective of this paper is to investigate the influence of selected extrinsic rewards, namely, compensation, promotion opportunities and job security, on the levels of job satisfaction and ultimately on the levels of organisational commitment of nonfamily employees working in family businesses. Structural equation modelling was used to assess the hypothesised relationships, and the results show that compensation and job security are significantly and positively related to job satisfaction. Job satisfaction was found to be significantly and positively related to organisational commitment and also found to act as a mediator between these extrinsic rewards and organisational commitment. Establishing the perspectives that nonfamily employees have of the rewards given to them by the family businesses in which they work, provides family business owners with valuable insights into attracting and retaining this valuable stakeholder group.

The issues surrounding rewards are complex in the field of family business.According to Klein (2007: 27), the level of compensation and the fairness thereof are often criticised, especially by nonfamily employees working in family businesses.Employee compensation at all levels is often lower in family businesses than in nonfamily businesses (Ceja & Tápies, 2009: 6), and nonfamily employees are generally less satisfied than family members with their compensation and benefits (Poza, 2010: 235).Similarly, promotion opportunities for nonfamily employees working in family businesses are often perceived as limited (Tsao, Chen, Lin & Hyde, 2009: 320;Chua, Chrisman & Sharma, 2003: 97), with nonfamily employees expecting to miss out on these opportunities because they are reserved for family members (Davis, Allen & Hayes, 2010: 1110;Poza, 2010: 236;Shelton, 2010).This reservation of senior positions for family members contributes to the high levels of job insecurity often experienced among nonfamily employees working in family businesses (Fishman, 2009: 205).
Given that the majority of people employed in family businesses are nonfamily members (Sharma n.d.;Mitchell, Morse & Sharma, 2003: 534), developing and maintaining the job satisfaction and organisational commitment levels of these individuals is one of the highest priorities of family business owners (Sharma n.d.;Mitchell et al., 2003: 534).According to Fishman (2009: 226), ensuring that nonfamily employees remain satisfied is essential to the success of a family business.If nonfamily employees perceive that decision outcomes, processes and decision-makers in a family business are unfair, acquiring their commitment and support is likely to be difficult (Barnett & Kellermanns, 2006: 837).
Although compensation (rewards) has been investigated in several disciplines such as economics, psychology, sociology and finance (Ibrahim & Boerhaneoddin, 2010:43), as far as can be established, few studies (e.g.Gomez-Mejia, Larraza-Kintana & Makri, 2003;McConaughy, 2000) in the field of family business have investigated the perspectives of nonfamily employees concerning compensation.Klein (2007: 27) suggests that investigating the relationship between the compensation packages available in family businesses and job satisfaction is an avenue for research worth pursuing.Sharma (2004: 15) emphasises that a need exists to devote more attention to understanding the perspective of nonfamily employees in family businesses, as well as the issues that are important to them.The purpose of this study is to respond to this need.Against this background, the primary objective of this paper is to investigate the influence of selected extrinsic rewards on the levels of job satisfaction and ultimately on the organisational commitment of nonfamily employees working in family businesses.To achieve this objective the following research questions are posed: Do the rewards Compensation, Promotion opportunities and Job security influence the levels of Job satisfaction experienced by nonfamily employees working in family businesses?Does Job satisfaction mediate the relationship between the rewards investigated in this study and the Organisational commitment of nonfamily employees to the family businesses in which they are employed?Establishing the perspectives that nonfamily employees have of the extrinsic rewards given to them by the family businesses in which they work can provide family business owners with valuable insights into attracting and retaining this valuable stakeholder group.
The following concepts adopted in this study require clarification.A 'family business' is a business where a single family owns at least 51% of the equity of the business; where a single family is able to exercise considerable influence in the business; and where at least two family members are concerned with the senior management of the business (Eybers, 2010;Farrington, 2009).In addition, 'nonfamily employees' refers to executives, managers, supervisors and/or administrative staff working in a family business who are not related by blood, marriage or adoption to the family owners and/or family managers of the business.

Literature and hypothesis development
Various factors such as the nature of the work, co-workers, leadership and physical working conditions, amongst others, have been identified as influencing the levels of employee job satisfaction and organisational commitment.However, rewards are most commonly cited in human resources management theory as necessary for creating value for both employees and organisations (Ibrahim & Boerhaneoddin, 2010: 43), and thus deserve specific attention.Extrinsic rewards have been found to be significantly more correlated with job satisfaction than intrinsic rewards (Rafiq, Javed, Khan & Ahmed, 2012: 25).According to Linz and Semykina (2010: 7), extrinsic rewards directly influence an employee's financial situation and are often viewed as the more obvious correlates of job satisfaction.Similarly, Rafiq et al. (2012: 23) contend that rewards such as pay and promotion are most related to job satisfaction because when employees perceive that they are rewarded fairly, they experience greater satisfaction and overall employee wellbeing.Therefore the focus of this study is on the extrinsic rewards Compensation, Promotion opportunities and Job security.These extrinsic rewards serve as the independent variables, while Job satisfaction and Organisational commitment serve as the mediating and dependent variable respectively.
Job satisfaction and organisational commitment are constructs that have been the focus of discussion and research attention for many years (Lumley, Coetzee, Tladinyane & Ferreira, 2011: 101;Dirani, 2009: 192;Warsi, Fatima & Sahibzada, 2009: 400;Lok & Crawford, 2001: 594).According to Yiing and Ahmad (2009: 54), a substantial amount of research exists on the antecedents and outcomes of both job satisfaction and organisational commitment.Therefore the purpose here is not to provide an in-depth discussion of the literature relating to rewards, job satisfaction and organisational commitment, or the relationships between them, but rather to present sufficient evidence of the relationships to be empirically tested.
In general, job satisfaction is defined as a "positive attitude or emotional state resulting from the appraisal of one's job or job experience" (Byars & Rue, 2006: 406).According to Buitendach and Witte (2005: 28), "job satisfaction has to do with an individual's perception and evaluation of his job and this perception is influenced by the person's unique circumstances such as their needs, values and expectations".Warsi et al. (2009: 402) assert that organisational commitment has been defined and measured in several different ways, but Padala (2011: 18) maintains that the definition of Porter, Steers, Mowday and Boulian (1974), namely "the identification with an organisation and acceptance of its goals and values as one's own", is most common.Most researchers suggest that organisational commitment represents both an attitude that describes an individual's relationship with an organisation and a set of behaviours by which individuals manifest that relationship (Lumley et al., 2011: 105;Elizur & Koslowsky, 2001: 594).
Studies investigating the determinants of organisational commitment have consistently reported a significant and positive relationship between job satisfaction and organisational commitment (Lumley et al., 2011: 112;Warsi et al., 2009: 407;Buitendach & Witte, 2005: 33;Lok & Crawford, 2001: 607).However, according to Yiing and Ahmad (2009: 57-58), research on the antecedents of organisational commitment and the relationship between job satisfaction and organisational commitment lack consistency.Some authors report a negative relationship between job satisfaction and organisational commitment, whereas others report both positive and negative relationships, and some even report no relationship (Yiing & Ahmad, 2009: 76;Rayton, 2006: 139).Elizur and Koslowsky (2001: 594) are of the opinion that the relationship between job satisfaction and organisational commitment is complex, and it is not clear which is the precursor of the other.Against this background, the following hypothesis is subjected to further empirical testing: H 1a : There is a positive relationship between the level of Job satisfaction of nonfamily employees working in family businesses and their level of Organisational commitment to the family business.
In their study on the antecedents of organisational commitment, Lok and Crawford (2001: 607) found that job satisfaction has a significant positive influence on organisational commitment, and that job satisfaction is a mediator, or partial mediator, between organisational commitment and other variables influencing organisational commitment (Lok & Crawford, 2001: 597).According to Ibrahim and Boerhaneoddin (2010: 48), a review of the literature shows that the relationship between job satisfaction and organisational commitment has been widely researched, but not much has been done on researching the mediating role of job satisfaction between compensation structure and organisational commitment.Against this background, the following hypothesis is proposed.
H 1b : Job satisfaction mediates the relationship between the extrinsic rewards investigated in this study and the Organisational commitment of nonfamily employees to the family business in which they are employed.
For the purpose of this study, Job satisfaction refers to nonfamily employees being satisfied with their jobs and working relationships in the family business, and experiencing their involvement in the family business as enjoyable, rewarding and fulfilling.Although organisational commitment has been found to be a multi-dimensional construct, consisting of affective, continuance and normative commitment (Allen & Meyer, 1990), as in the case of most research undertaken in the area of organisational commitment (Yiing & Ahmad, 2009: 56), the present study focused on measuring affective commitment only.According to Yiing and Ahmad, (2009: 56), affective commitment relates to "the emotional attachment to, the identification with, and the involvement in the organisation" experienced by an employee.Therefore, for the purpose of this study, Organisational commitment refers to nonfamily employees having pride in the family business, having an emotional attachment and a sense of belonging to the family business, caring about the fate of the business, and being willing to put in a great deal of effort to ensure its success.

Key extrinsic rewards
Extrinsic rewards occur as a consequence of performing the job, and the benefits associated with these rewards are obvious to an employee (Linz & Semykina, 2010: 7).Extrinsic rewards are tangible and visible to others, and those most commonly cited include factors such as pay, promotions and security (Mahdi et al., 2012(Mahdi et al., : 1519;;Rafiq et al., 2012: 24;Khalid et al., 2011: 35;Linz & Semykina, 2010: 7).Therefore, the three extrinsic rewards identified for investigation in this study are Compensation, Promotion opportunities and Job security.
According to Byars and Rue (2006: 249), compensation and pay are not the same.'Pay' refers to the actual cash an employee receives in exchange for the work done, whereas 'compensation' is broader and includes pay as well as all bonuses and fringe benefits (Byars & Rue, 2006: 249;Yasir & Fawad, 2009: 7).Ample evidence exists to support a positive relationship between compensation and job satisfaction (Danish & Usman, 2010: 163;Yasir & Fawad, 2009: 19;Byars & Rue, 2006: 248;Rayton, 2006: 148).In their meta-analysis investigating the relationship between pay and job satisfaction, Judge, Piccolo, Podsakoff, Shaw and Rich (2010: 160) conclude that pay level is positively correlated with job satisfaction.In contrast, Byars and Rue (2006: 248) report that some studies show that actual pay and access to benefits have a relatively insignificant influence on employee satisfaction.
A lack of fairness in terms of workload and corresponding compensation is often cited as a reason why nonfamily members are hesitant to accept employment in family businesses (Maas, Van der Merwe & Venter, 2005: 132).Feelings of inequity arise among nonfamily employees when family members are favoured in terms of compensation (Gedajlovic, Carney, Chrisman & Kellermanns, 2011: 9) and serious problems can arise when family members are paid more than nonfamily members in similar positions (Veale, n.d.).According to Harvey-Jones (1999), most employees in family businesses do not expect to own shares in the family business, but are happy as long they are reasonably well paid.For the purpose of the present study, the extrinsic reward Compensation refers to a salary (pay and fringe benefits) that is competitive.The following hypothesised relationship is subjected to empirical testing: H 2 : There is a positive relationship between the Compensation received by nonfamily employees working in family businesses and their level of Job satisfaction experienced in the family business.
Promotion occurs when an employee moves in an upward direction in the hierarchy of an organisation (Dessler, 2008) and can lead to an increase in salary, authority and responsibility (Naveed, Usman & Bushra, 2011: 30).Several studies have empirically confirmed a positive relationship between the availability of promotion opportunities and job satisfaction (Naveed et al., 2011: 304; Career opportunities for family as well as nonfamily employees are important for promoting family business continuity (Poza, 2010: 235).However, nonfamily employees perceive that they will be excluded from senior positions (Tsao et al., 2009: 320;Chua et al., 2003: 97) because family members typically come first when promotion opportunities arise (Davis et al., 2010(Davis et al., : 1110;;Flemming, 2000: 278).Highly qualified nonfamily members are often reluctant to accept a job in a family business because of this perception (Maas et al., 2005: 132) and many leave their careers with the family business because they find that there are limitations in terms of promotional opportunities (Fishman, 2009: 200).Most nonfamily employees recognise that they are not likely to attain top leadership position in the business (Fishman, 2009: 201), but the promotion of unqualified family members can lead to dissatisfaction and a sense of inequity, especially among more qualified nonfamily employees (Fishman, 2009: 41;Gedajlovic et al., 2011: 9).For the purpose of this study, Promotion opportunities refer to employees having the opportunity for advancement and promotion in the family business.Based on the above discussion, the following relationship is hypothesised: H 3 : There is a positive relationship between the Promotion opportunities available to nonfamily employees working in family businesses and their level of Job satisfaction.
Job security or job insecurity relate to the feelings of fear that people have about the continued existence of their job, of losing their job, or of becoming unemployed (Buitendach & Witte, 2005: 28).Job insecurity leads to low levels of job satisfaction and organisational commitment, and to a lesser degree, to reduced physical and mental well-being (Klandermans, Hesselink & Van Vuuren, 2010: 4;Buitendach & Witte, 2005: 28).For the purpose of this study, Job security refers to an employee feeling safe, secure and unthreatened about the future of his/her job in the family business.
Because they are not part of the family, nonfamily employees in particular face a complex and uncertain situation in family businesses (Barnett & Kellermanns, 2006: 845).Feelings of insecurity about their future in the family business are one of the main reasons why nonfamily employees leave these businesses (Veale, n.d.).This state of insecurity is amplified when successors enter the family business, because the successors, who may or may not be competent, are ultimately responsible for the future success of the business (Fishman, 2009: 207).Against this background, the following hypothesis is formulated: H 4 : There is a positive relationship between the Job security experienced by nonfamily employees working in family businesses and their level of Job satisfaction.

Research design and methodology
In order to achieve the objective of this study a quantitative research approach was adopted.This approach was considered appropriate given that the focus of the study was on testing the proposed hypotheses, as well as on analysing the data collected quantitatively.Structural equation modelling (SEM) was the main statistical technique adopted to test the hypothesised relationships.

Sampling procedure and response rate
Convenience snowball sampling was used in this study as no existing database containing the contact details of nonfamily employees exists.Family business owners were contacted and requested to encourage their nonfamily employees to participate in the study.The database developed by previous researchers focusing on family businesses (Eybers, 2010;Farrington, 2009) served as the starting point for this process.Where possible, nonfamily employees were approached directly and requested to participate.In addition, research contacts and friends across South Africa were asked to identify nonfamily employees that met the criteria, who were working in a family business.Once identified, nonfamily employees working in family businesses throughout South Africa were asked to participate in the study.The sample consisted of 800 nonfamily employees.Of the 324 questionnaires returned, 280 were usable for further statistical analyses.An effective response rate of 35% was realised.Given the satisfactory response rate, nonresponse bias was not considered a problem in this study.

Development of scales and administration of measuring instrument
A survey making use of a structured, self-administered questionnaire was used to gather the primary data.The measuring instrument consisted of a covering letter and two sections.Section 1 consisted of 13 statements describing the rewards Compensation, Promotion opportunities and Job security, as well as 11 statements relating to Job satisfaction and Organisational commitment.Items were based on the literature, and several were sourced from previous studies (See Appendix A for items and sources of items).Those sourced from previous studies were rephrased to contextualise them for the present study.A 7-point Likerttype scale (1 = strongly disagree and 7 = strongly agree) was employed, and each respondent was asked to indicate their extent of agreement with each statement.In Section 2 of the questionnaire, demographic information relating to the respondent and the family businesses in which they worked was requested.The questionnaire was made available to respondents via postal mail, email or personal delivery.

Statistical analysis phases
The Statistica (Version 10) and AMOS (Version 19) computer programs were used to statistically analyse the data collected.This analysis was conducted in several phases:  Firstly, descriptive statistics were determined to profile the respondents;  Secondly, in order to measure the goodness-of-fit of the data to the model, structural equation modelling (SEM) was conducted.Five goodness-of-fit indices were used, namely the normed chi-square ( 2 /df), Root mean squared error of approximation (RMSEA), Comparative fit index (CFI), Tucker-Lewis index (TLI) and Parsimony goodness-of-fit index (PGFI);  Thirdly, the validity of the measuring instrument was established by assessing for construct validity.The factor loadings resulting from the confirmatory factor analysis (CFA) and the variance extracted (VE) estimates were used to verify the validity of the measuring instrument;  Fourthly, both construct reliability (CR) estimates and Cronbach's alpha coefficients were computed to establish the reliability of the measuring instrument;  During the fifth phase of the statistical analysis, the hypothesised relationships were tested by evaluating the point and interval estimates of the parameters resulting from the SEM analysis;  In the sixth phase the mediating role that Job satisfaction plays between the rewards identified in this study and Organisational commitment was investigated by means of a series of multiple regression analyses (as recommended by Baron andKenny, 1986: 1177); and  Lastly, descriptive statistics and correlation coefficients were calculated to summarise the sample data.

Demographic profile of respondents
Approximately the same number of males (47,14%) and females (52,86%) participated in the study.Most respondents were between the ages of 21 and 30 years (45,71%) or between the ages of 31 and 40 years (30,36%), and most were either White (45%) or Black (29,64%).The family businesses in which the respondents were employed were found to operate within a variety of industries, with most being in the retail, motor trade and repair services (24,29%), hospitality (16,79%) or finance and business services (12,14%) industries.Most respondents indicated between 11 and 25 (21,79%) or between 5 and 10 (21,43%) employees working in the family business.The majority (88,57%) of respondents indicated having worked in the family business for less than 10 years, with most being in managerial/supervisory (42,86%) or administrative/clerical (33,21%) positions.The vast majority (91,79%) of respondents also reported not possessing any shares in the family businesses.

Structural equation modelling results
The hypothesised relationships to be tested in this study were subjected to SEM.The SEM process revealed the existence of an insignificant relationship (p-value = 0,343) between the extrinsic reward Promotion opportunities and the mediating variable Job satisfaction (see Table 1).In addition, the goodness-of-fit indices of the hypothesised model (Model 1), namely the CMin/df or  2 /df (4,685), RMSEA (0,115), CFI (0,843), TLI (0,826) and PGFI (0,619) showed that the model could be adapted to fit the data better.According to Cooper and Schindler (2007: 584) as well as Garson (2006), significant relationships in poor-fitting models are not meaningful.Therefore to improve the model fit, the factor Promotion opportunity was removed from it, and a second model was subjected to the SEM process.The normalised Chi-square ( 2 /df) or CMin/df goodness-of-fit indices for the fitted Model 2 was 2,965.Given that the normed Chi-square of Model 2 was within the recommended value of 3 or less, the data can be described as having a reasonable fit with the model.The RMSEA of 0,084 for Model 2 was slightly above the recommended value of less than 0,08 but very close to the recommended cutoff value.The CFI and the TLI indices for Model 2 were above the recommended 0,9 level (0,931 and 0,921 respectively) and are therefore regarded as indications of a model with a good fit.Model 2 thus provided a satisfactory fit for the data as supported by the fit indices.The results of Model 2 showed that all the parameters were significant, all factors were significant at the 1 percent level of significance or less, all indices had improved, and the PGFI of 0,0664 for Model 2 was greater than that of Model 1.Based on these results, Model 2 was then adopted and used in the subsequent data analysis phases.

Validity and reliability
The validity of the measuring instrument was assessed for by means of a confirmatory factor analysis (CFA) in the SEM process.The factor loadings and the variance extracted (VE) estimates were used for this purpose.From Table 2 it can be seen that the factor loadings of each item exceeded the required minimum loading of 0,5 (Hair, Black, Babin, Anderson & Tatham, 2006: 777).As a result it was not necessary to remove any of the items to improve the validity of the measuring instrument.

Items & factors
Factor loadings Items & factors Factor loadings JobS1 <---Job security 0,865 JS5 <---Job satisfaction 0,801 JobS2 <---Job security 0,881 JS4 <---Job satisfaction 0,840 JobS3 <---Job security 0,888 JS3 <---Job satisfaction 0,852 JobS4 <---Job security 0,581 JS2 <---Job satisfaction 0,924 JobS5 <---Job security 0,659 JS1 <---Job satisfaction 0,876 OC1 <---Org.commitment 0,785 COMP4 <---Compensation 0,521 OC2 <---Org.commitment 0,843 COMP3 <---Compensation 0,854 OC3 <---Org.commitment 0,827 COMP2 <---Compensation 0,735 OC4 <---Org.commitment 0,870 COMP1 <---Compensation 0,824 OC5 <---Org.commitment 0,727 OC6 <---Org.commitment 0,600 In addition, in order to determine the discriminant validity of the measuring instrument, the average variance extracted (AVE) estimates for any two factors were compared with the square of the correlation estimates between the two factors.The decision criteria suggest that if the AVE is greater than the squared correlation, discriminant validity is good.From Tables 3 and 4 it can be seen that in three of the six cases the AVE was greater than the squared correlation indicating that the discriminant validity was good.In two of the other three cases at least one of the AVE scores was only marginally less than the squared correlation.Although not complying with the general interpretation that discriminant validity was good, the values were so close that disputing this claim would be debatable.In addition, all the AVE estimates for the factors under investigation were close to or above the recommended value of 0,5 (Hair et al., 2006:779).Therefore, as in the case of the CFA, evidence of the discriminant validity of the measuring instrument was satisfied.Both construct reliability (CR) estimates and Cronbach's alpha coefficients (CA) were computed to establish the reliability of the measuring instrument.These estimates are shown in Table 5.From Table 5 it can be seen that Job security, Job satisfaction and Organisational commitment all returned construct reliability estimates and Cronbach's apha coefficients exceeded the recommended value of 0,7 for both measures of reliability (Hair et al., 2006: 778;Nunnally & Bernstein, 1994).Satisfactory evidence of a reliable measuring instrument for these factors is thus provided.
Although the CA returned for Compensation was above the acceptable 0,7, the CR estimate of 0,5818 was marginally less than the minimum value of 0.6 as recommended by Hair et al. (2006).Given the importance of compensation in the family business literature, the satisfactory CA reported (0,808), as well as the close proximity of the CR estimate's value to 0,6, the items measuring Compensation were regarded as reliable, and the factor was retained for further statistical analysis in this study.Given the above discussion and the results presented in Tables 3 and 4, the validity and reliability of the measuring instrument as a whole was thus considered satisfactory.

Hypothesis testing
The next phase of the statistical analysis involved testing the hypothesised relationships.This was done by means of evaluating the point and interval estimates of the parameters resulting from the SEM analysis.The parameter estimates and p-values reported for Model 2 are summarised in Table 6.Given that Promotion opportunities was omitted during the SEM process and excluded from Model 2, H 3 was no longer subjected to empirical testing.

Mediating role of job satisfaction
In order to test whether Job satisfaction mediates the relationship between the rewards Job security and Compensation, and the dependent variable Organisational commitment, the three steps recommended by Baron andKenny (1986: 1177) to test for mediation were undertaken.This involved a series of regression analyses.Baron andKenny (1986: 1177) assert that the beta coefficient of several regression equations must be compared.First the mediator (Job satisfaction) must be regressed on the independent variables (Job security and Compensation), second the dependent variable (Organisational commitment) must be regressed on the independent variables (Job security and Compensation), and third the dependent variable must be regressed on both the mediator and the independent variables.According to Baron andKenny (1986: 1177), "to establish mediation, three conditions must exist, first the independent variables must affect the mediator in the first equation, second, the independent variable must affect the dependent variable in the second equation, and third the mediator must affect the dependent variable in the third equation.If these conditions all hold in the predicted direction, then the effect of the independent variables on the dependent variables must be less in the third equation than in the second.Perfect mediation holds if the independent variable has no effect when the mediator is controlled."

Descriptive statistics and correlations
The descriptive results for the factors under investigation as well as their correlation coefficients are reported in Table 8.For discussion purposes, the response categories on the 7point Likert-type scale were categorised as disagree (1≤x<4), neutral (4≤x<5) and agree (5≤x<8).Job satisfaction and Organisational commitment returned means of 5,106 and 5,289 respectively, with the majority of respondents agreeing that they felt satisfied (66%) in their jobs and committed (67%) to the family business in which they were employed.Job security ( x = 4,844) and Compensation ( x = 4,400) returned lower mean scores with only 54% and 43% agreeing that they experienced job S.Afr.J.Bus.Manage.2014,45(3)security or were paid competitive salaries.From Table 8 it can be seen that all the factors investigated are positively and significantly (p < 0,05) correlated with each other.According to Gravetter and Wallnau (2009), r values greater than 0,5 reflect strong positive associations, as is the case of the r values reported between all the factors in this study.

Discussion and recommendations
The objective of this study was to investigate the influence of Compensation, Promotion opportunities and Job security on the levels of Job satisfaction and Organisational commitment of nonfamily employees working in family businesses.Despite theoretical and empirical support for a positive relationship between Promotion opportunities and Job satisfaction, no relationship was reported in this study.However, the goodness-of-fit indices showed that the model (Model 1) could be adapted to fit the data better if the reward Promotion opportunities was omitted from the model being subjected to SEM.It is well documented that in family businesses, top management positions are reserved for family members (Davis et al., 2010(Davis et al., : 1110;;Flemming, 2000: 278) and a possible explanation for the findings relating to Promotion opportunities is that the participating respondents are aware of the practice of promoting family members in family businesses and thus have no expectations regarding advancement and promotion.Another possible explanation is that the scale measuring Promotion opportunities is not valid.
The results of the SEM analysis (Model 2) show a significant positive relationship between the independent variables, Job security and Compensation and the mediating variable Job satisfaction.In other words, the more unthreatened, safe and secure nonfamily employees felt about the future of their jobs, and the more salary packages were perceived as competitive, the more likely it was that nonfamily employees would be satisfied with their job and working relationships in the family business, as well as experiencing working in the family business as enjoyable, rewarding and fulfilling.
The findings of this study show that only 54% of the nonfamily employees participating in this study agreed that they felt secure about the future of their jobs, and only 43% perceived their salary packages to be competitive.This finding provides a possible explanation as to why only twothirds of respondents reported being satisfied in their jobs and committed to the family businesses in which they were employed.Given the influence of job security and compensation on job satisfaction, the influence of job satisfaction on organisational commitment, and ultimately on performance (Danish & Usman, 2010: 161), family business owners are advised to give increased attention to these aspects of their rewards policy.
In order to ensure that nonfamily employees feel safe and secure concerning the future of their jobs and are satisfied with their remuneration, several recommendations are proposed.Without divulging specific financial details, nonfamily employees should be kept up to date on the financial standing of the family business.Should the family business experience financial difficulties, employees should be informed so that they are aware of the threat to the future of their employment.The lack of succession planning and succession itself is also a source of insecurity for nonfamily employees.To reduce these levels of insecurity, succession planning should take place timeously and be clearly communicated to all stakeholders.Key nonfamily employees should even be involved when succession planning takes place, increasing their commitment to the future successor and the continuity of the family business.
Although the results of this study show that Job security has a greater influence on Job satisfaction than Compensation does, this does not mean that compensation is less important or less motivating.According to Judge et al. (2010: 163), the level of pay a person receives may have a limited ability to satisfy.However, they emphasise that employers should be mindful that this limited ability to satisfy does not mean that pay is not motivating.Furthermore, research has shown that people who earn more are not necessarily more satisfied with their jobs (Spector, 2008).What is important is the perception of fairness of the compensation received.According to Spector (2008), fairness plays a major role in the behaviours and reactions of people towards their work.Family business owners thus need to ensure that their compensation practices are fair to both family and nonfamily employees, and fair in terms of position, education and work experience.In addition, compensation packages should be market-related and fair in terms of packages paid elsewhere, provided that the family business can afford it.
A significant positive relationship is reported between the mediating variable Job satisfaction and the dependent variable Organisational commitment in this study, which implies that the more nonfamily employees experience job satisfaction, the more likely they are to have pride in the family business, have an emotional attachment and a sense of belonging to the family business, care about the fate of the family business, and be willing to put in a great deal of effort to ensure its success.In addition to confirming a positive relationship between Job satisfaction and Organisational commitment, the results also show that Job satisfaction mediates the relationship between the rewards, Job security and Compensation, and Organisational commitment.This finding supports that of Ibrahim and Boerhaneoddin (2010: 55) who report that job satisfaction plays a significant role in mediating the relationship between compensation and organisational commitment.Given the lack of consistency and even contradictory results arising from research on the relationship between job satisfaction and organisational commitment (Yiing & Ahmad, 2009: 76;Rayton, 2006: 139), the present study contributes to clarifying this relationship.Furthermore, job satisfaction is said to be a mediator, or partial mediator, between organisational commitment and 'other' variables influencing organisational commitment (Lok & Crawford, 2001: 597).This study confirms that Job security and Compensation are included in those 'other' variables.

Implications
The findings of this study have implications for both researchers and family business owners.For researchers, the findings of this study regarding Promotion opportunities raise several questions.Are promotion opportunities important in the context of a family business?How can promotion opportunities be made available to nonfamily employees without jeopardising the influence of the family?Researchers in the field of family business are hopefully challenged to answer these questions.
The results relating to Job security and Compensation show that existing theories on the factors influencing job satisfaction are also relevant among nonfamily employees in family businesses.This suggests that theories relating to the other antecedents of job satisfaction and organisational commitment are also applicable in the context of family businesses.Family business owners could find this useful because existing practices to ensure that employees are satisfied in their jobs will also be effective in family businesses.Less than half of respondents participating in this study perceived their salary packages as competitive, yet more than two-thirds agreed that they were satisfied in their jobs.These findings highlight the possibility that the job satisfaction of nonfamily employees is influenced less by compensation than by other rewards or conditions that exist in the family business.For example, it has been suggested that in the context of a family business, emotional and social compensation as well as psychological ownership may be more relevant than financial compensation (Astrachan & Adams, 2005).Family business owners should attempt to identify these non-financial aspects that motivate their nonfamily employees, so that these aspects can be monitored and nurtured.
Low compensation leads to job dissatisfaction, but good compensation does not necessarily lead to job satisfaction.According to Flemming (2000: 278), good compensation should be accompanied by several other conditions, including overall working conditions and fair treatment.'Fair treatment' is one of the most important requirements for nonfamily employee satisfaction (Maas et al., 2005: 133).In many cases, nonfamily employees choose to work in family businesses knowing that family members come first when it comes to rewards and promotion (Davis et al., 2010(Davis et al., : 1110)).Having this knowledge in not necessarily demotivating to nonfamily employees, as long as family business owners are upfront, fair and honest about these practices.

Limitations and future research
Several limitations to this study are highlighted.The use of a non-probability sample technique (convenience snowball sampling) results in one not being able to generalise the findings to the entire family business population.In addition, the data collected for this study relies on the perceptions of the respondents themselves, which could introduce a degree of bias into the findings of this study.
When eliminating the factor Promotion opportunities from the model, the results of the SEM analysis showed improved goodness-of-fit indices, suggesting that the scale measuring this factor could be invalid.Furthermore, the construct reliability estimate reported for Compensation was less than the required minimum.The findings of this study should be interpreted in light of these limitations, and future studies should attempt to improve the scales measuring these factors.
Job satisfaction and Organisational commitment were considered as one-dimensional constructs in this study, and the results should be interpreted in light of this.Job satisfaction is a heterogeneous phenomenon, and self-reports of job satisfaction may reflect satisfaction with different aspects of the job, depending on the individuals' perceptions of job satisfaction (Millán, Hessels, Thurik and Aguado, 2011: 2).Furthermore, according to Buitendach and Witte (2005: 29), the literature reveals that job satisfaction consists of two dimensions, one relating to extrinsic and the other to intrinsic factors.Organisational commitment is a multidimensional concept consisting of affective commitment, continuance and normative commitment (Landy & Conte, 2010: 416;Buitendach & Witte, 2005: 29;Allen & Meyer, 1990), with each aspect reflecting a distinct nature of organisational commitment (Lok & Crawford, 2001: 595).According to Landy and Conty (2010: 417), understanding of length, breadth and depth of the commitment construct is still in an early stage.Future research should account for the multidimensional nature of both these constructs.
A further limitation of the present study is that it focused on selected extrinsic rewards only, and did not consider the numerous other factors, such as fairness, interpersonal relationships, job involvement, the personal circumstance of nonfamily employees and the culture of the family business, amongst others, which could also potentially influence the job satisfaction and organisational commitment levels of nonfamily employees working in family businesses.Investigating these factors is an avenue for future research.

SATISFIED1
I am satisfied with my job in this family business (Dua, 1994).

SATISFIED5
I am satisfied with the way that we work together in this family business (Eybers, 2010;Farrington, 2009).ORGANISATIONAL COMMITMENT COMMIT1 I am proud to tell others that I am employed by this family business (Boshoff & Arnolds, 1995;Arnolds, 2005).

COMMIT2
I feel a strong sense of belonging to this family business (Liu, 2007).

COMMIT3
I feel emotionally attached to the business (Liu, 2007).

COMMIT4
This family business has a great deal of personal meaning for me (Liu, 2007).

COMMIT6
I am willing to put in a great deal of effort beyond that which is normally expected of me, in order to help this family business to succeed (Boshoff & Arnolds, 1995;Arnolds, 2005).COMPENSATION

COMP1
In this family business my salary is competitive with what I could earn in another business (Robbins, 2003).

COMP2
My compensation in this family business is on the same level as what others would get for similar work in other businesses (DeVaney & Chem, 2003).

COMP3
In this family business my salary is on the same level as what someone with a similar education and work background would get in another business (DeVaney & Chem, 2003).

COMP4
In this family business I have the same opportunities to earn non-cash rewards (e.g.travel opportunities, paid time off) as I would in another business (Kayoki & Lesaoana, 2011).

PROMO1
There are promotional opportunities for me in this family business (Dua, 1994).

PROMO2
I am able to achieve the position that I strive for in this family business (Dua, 1994).

PROMO3
My opportunities for advancement in this family business are unlimited (DeVaney & Chem, 2003).

PROMO4
In this family business I have a good chance of promotion (DeVaney & Chem, 2003).

SECURITY1
In this family business my job is secure (Dua, 1994).

SECURITY2
My future in this family business is safe (Emberland & Rundmo, 2010)

SECURITY3
In this family business my job provides me with a sense of security (DeVaney & Chem, 2003).

SECURITY4
My job in this family business is not threatened by incoming family members (Devries, 2007).

SECURITY5
My job in this family business is not threatened by a lack of succession planning (Harvey-Jones, 1999).

Table 6 : Parameter estimates and p-values
variable Job satisfaction.The results do, however, show that Job security had a greater influence on Job satisfaction than Compensation does.Against this background, support was found for hypotheses H 2 and H 4 .A significant positive relationship (estimate 0,951; p < 0,001) was also reported between the mediating variable Job satisfaction and the dependent variable Organisational commitment.Support is thus found for hypothesis H 1a .

Table 7 : Regression analyses testing for mediation Step 1 Mediator variable: Job satisfaction
From Table7it can be seen that Job satisfaction mediates the relationship between both the independent variables Job security and Compensation, and the dependent variable explanatory power of the model increases (as indicated by an increase in R² of 0,1269).Therefore Job satisfaction adds to explaining the variation in Organisational commitment.Support is thus found for hypothesis H 1b .