Retention preferences and the relationship between total rewards , perceived organisational support and perceived supervisor support

Some of the main challenges that organisations, especially South African organisations, are confronted with are skill scarcity and reducing turnover of talented employees (Ashton & Morton, 2005; Ready & Conger, 2007; Robert & Börjesson, 2006). Kerr-Phillips and Thomas (2009, p. 1) record that South Africa is undergoing an overall skill difficulty with regard to the recruitment of its leading talent or ‘knowledge’ employees. These knowledge employees could be perceived to be the key talent or upcoming talent in the organisation. The search for proficient employees, talent and intellectual capital is even more significant for strategic organisational success and increase in the organisation’s overall financial performance (Lawler, 2008; Michaels, Handfield-Jones & Axelrod, 2001; Reindl, 2007; Van Rooyen, Du Toit, Botha & Rothmann, 2010). Orientation: Currently there is much debate whether modifying traditional reward packages to focus on the preferences of multi-generations would be essential in attracting, motivating and retaining talent. Total reward factors, perceived organisational support and perceived supervisor support are distinct but related concepts, all of which appear to influence an employee’s decision to stay at an organisation.


Introduction Key focus
'The right total reward strategy can deliver the right amount to the right people at the right time, for the right reasons' (Gross & Friedman, 2004, p. 10).The adapted quote illustrates that the most effective reward strategy could provide valuable factors that ultimately motivate and retain talented employees.Some of the main challenges that organisations, especially South African organisations, are confronted with are skill scarcity and reducing turnover of talented employees (Ashton & Morton, 2005;Ready & Conger, 2007;Robert & Börjesson, 2006).Kerr-Phillips and Thomas (2009, p. 1) record that South Africa is undergoing an overall skill difficulty with regard to the recruitment of its leading talent or 'knowledge' employees.These knowledge employees could be perceived to be the key talent or upcoming talent in the organisation.The search for proficient employees, talent and intellectual capital is even more significant for strategic organisational success and increase in the organisation's overall financial performance (Lawler, 2008;Michaels, Handfield-Jones & Axelrod, 2001;Reindl, 2007;Van Rooyen, Du Toit, Botha & Rothmann, 2010).
The focus of this study was to understand the total rewards options for the multi-generation workforce in South African organisations in order to identify whether different generations prefer different total reward components.The link between perceived organisational support (POS) and perceived supervisor support (PSS) and total reward was also investigated.An enriched understanding of retention preferences that influences organisational commitment may benefit the organisation that wants to retain its valuable talent.Lawler (2008) explains that the definition of talent offers a characterisation of what it can preserve, a declaration of what is missing and a proposal of actions needed to be implemented.Numerous nations and organisations are now engaged in the war for best talent (Ferguson & Brohaugh, 2009;Frase, 2007;Human Capital Institute & Vurv Technology, 2008;Kontoghiorghes & Frangou, 2009;Lawler, 2008).Retention factors such as POS and PSS impact employees' job satisfaction, job performance, commitment and turnover (Allen, Shore & Griffith, 2003;Dawley, Andrews & Bucklew, 2008;Du Plessis, 2010;Riggle, Edmondson & Hansen, 2009;Tuzun & Kalemci, 2011).In other words, POS and PSS are negatively associated with judgements to quit the organisation (Hui, Wong & Tjosvold, 2007;Van Schalkwyk, Els & Rothmann, 2011).Longenecker and Scazzero (2003) indicate that intent to leave correlates with job opportunity elsewhere, more money, a poor supervisor, lack of appreciation or inability to get time off from work.Therefore, effective consideration of POS and PSS components can assist in attracting, motivating and retaining employees.

Trends from literature
Previous research has emphasised that the impact that generational differences have on, for example, communication styles, equipment needs, professional growth preferences, remuneration, benefits, desired leadership and effective reward and recognitions systems can have a significant and profound effect in the work environment (Haeberle, Herzberg & Hobbs, 2009;Nienaber, Bussin & Henn, 2011).
However, if organisations do not cater for the generational preferences, especially for Generation X and Generation Y, it may lead to the downfall of the organisation, as these two generations can be seen as the future leaders within any organisation.Generation X and Generation Y will lead the workplace in the upcoming years and management needs to consider the preferences and motivation of these generations to ensure an engaged and motivated workforce.Karp, Fuller and Sirias (2002) add that policies are generally formulated by Baby Boomers and consequently neglect the needs and preferences of Generation X and Generation Y.This may lead to reward policies that are based on Baby Boomers' preferences, neglecting any other generational preferences.The focus of relevant retention research has thus far been focused on: • The relationship between total reward components amongst different generations (Haeberle et al., 2009).• The employee's perception of supervisor and organisational support (Eisenberger, Singlhamber, Vandenberghe, Sucharski & Rhoades, 2002).
In recent years South Africa has focused much on organisational diversity that resulted from employee and cultural diversity; however, less attention has been given to generational differences within organisations.As can be expected, for many years only one generation was predominant; presently, there are four discrete generations prevalent in the working place, namely Veterans, Baby Boomers, Generation X and Generation Y.The understanding of these different generations has become increasingly important for organisations (Dawn, 2004;Du Plessis, 2010;Glass, 2007;Lancaster & Stillman, 2002;McGuire, Todnem & Hutchings, 2007;Reynolds, 2005;Sherman, 2006;Smola & Sutton, 2002;Zemke, Raines & Filipczak, 2002).

Research objectives
The primary objective is to determine whether, in South African organisations, generations prefer different total rewards options and what the relationship is between total reward, POS and PSS.More specifically, two research aims were compiled, namely to determine which reward factors employees rate as most important and to determine which factors are currently used to retain employees.In addition, the following six propositions were formulated from the literature: • Proposition 1: There is a significant difference in level of importance of total rewards components between managers and employees.• Proposition 2: Different generation groups prefer different total reward factors.• Proposition 3: Age, gender, race, qualification, industry, job level, years with company and years remaining at company play a moderating role in the relationship between total rewards and POS, as well as that between total reward and PSS.• Proposition 4: A direct positive relationship exists between the employee's total reward components and POS.• Proposition 5: A direct positive relationship exists between the employee's total reward components and PSS.• Proposition 6: A direct positive relationship exists between the POS and PSS.

Contribution to the field
South Africa has a very high unemployment rate, which is currently established to be 23.9% (Statistics South Africa, 2012), and one would expect that organisations would be enthusiastically recruiting employees.However, this is not the case, due to limited skills available, employee mobility and the pending retirement of Baby Boomers (Hall & Sandelands, 2009;Masibigiri & Nienaber, 2011).Talent management has the potential to be the reason for an organisation demise or the reason for its continuous success.
An organisation's ability to retain talent holds economic benefits for the organisation, both through cost containment (decreasing replacement costs) and income generation (through efficient application of talent).This study provides significant value in the form of pointers for organisations and human resource (HR) departments in the process of attracting, recruiting and most of all retaining high-calibre talent.More specifically, this study is helpful for managers who are concerned with retaining personnel with scarce skills and provides noteworthy suggestions for effective HR policies and retention strategies in order to retain a competent workforce.
In addition, research on total reward and retention preferences for multi-generation groups is limited in the South African workplace.In addition to its practical relevance, this study can also greatly contribute on an academic level.

What will follow
In the subsequent section of this article, the researchers outline the research design, followed by a presentation of the results.Finally they discuss the results, highlight the implications of the study, outline limitations and make suggestions for areas of future research.

Retention and reward
Total rewards is a term used to explain the support reward elements in the work environment.This can include financial, non-financial, indirect as well as direct, intrinsic and extrinsic rewards that are offered to an employee (Armstrong, 2006;Bussin, 2012;Cable & Judge, 2002;Costello, 2010;Hiles, 2009;WorldatWork, 2010).According to the Corporate Leadership Council Advisory Board, approximately 25% of the decision whether to stay at an organisation relates to remuneration (Bussin, 2012).This raises the question of whether higher compensation will in fact motivate talented employees to stay at the organisation.
Tang and West (1998) add that when employees receive adequate remuneration, they will concentrate more on inherent needs such as acknowledgement, appreciation, training and development and achievement.Reward models are used by many organisations as a starting point from which to design their own unique total reward model (Bussin, 2012;Nienaber et al., 2011).Such models include: The WorldatWork (2007) total rewards model, the Armstrong (2006) total reward model, the Zingheim and Schuster (2008)

Generations
Members of a generation group share a special history and these shared incidents or experience shape a generation's attitudes, perceptions and behaviours as they grow up together (Codrington & Grant-Marshall, 2004;Du Plessis, 2010;Sullivan, Forret, Carraher & Mainiero, 2009;Yang & Guy, 2006).However, Giancola (2008) argues that there are no distinct generations but rather that differences are related to individual life cycle and career stages.For this purpose of the study, the guidelines for multi-generations were based on the dates given by Lancaster and Stillman (2002), Reynolds (2005) and Zemke et al. (2002); these were identified as: Veterans (born 1922-1945), Baby Boomers (born 1946-1964), Generation X (born 1965-1980) and Generation Y (Born 1981(Born -1999)).The GI Generation (born 1900-1921) was excluded, as most people who were born before 1920 do not participate actively in the workplace any more.
Each generation has its own features and values and attitudes toward work-based events; more specifically, employees' requests and preferences change over time (Dwyer, 2009;Gross & Friedman, 2004;Jenkins, 2008;Kapoor & Solomon, 2011;Meyer & Kirsten, 2012;Netswera, Rankhumise & Mavundla, 2005;Orciani, 2009;Smola & Sutton, 2002;Snelgar et al., 2013;Van der Merwe, 2012).On a practical note it is impossible to compile unique reward profiles for large to medium-sized organisations as it would be difficult to manage and administer these profiles, as indicated by Nienaber et al. (2011).A resolution to this issue is, for instance, propositions of diverse rewards for different employee groups, based on characteristics such as job level, business unit, life cycle, geographic location, generation, age, family size, occupation, education level or religion (Du Toit, Erasmus & Strydom, 2007;Mercer, 2008;Snelgar et al., 2013).

POS:
Employees develop an outlook or a worldwide belief about the degree to which their organisation appreciates their contribution and cares about their general well-being, which is known as their POS, as defined by Eisenberger et al. (2002).Thus, it can be argued that individuals who perceive that organisational support is very high are less likely to seek and undertake alternative employment (Allen et al., 2003, Armstrong-Stassen & Ursel, 2009;Dawley et al., 2008;Du Plessis, 2010;Harris, Harris & Harvey, 2007;Hui et al., 2007;Jawahar & Hemmasi, 2006;Riggle et al. 2009;Van Schalkwyk et al., 2011;Van Vuuren, 2006).

PSS:
Employees develop overall insights about the degree to which their supervisors appreciate their contributions and are concerned about their well-being (Eisenberger et al., 2002).An employee's satisfaction with their direct supervisor and their perception of their supervisor's willingness to care for them has been shown to reduce intended turnover and improve the employee's commitment (Dawley et al., 2008;Du Plessis, 2010).Various studies have investigated the positive relationship between POS and PSS.However, very few studies have investigated the direction of the correlation between POS and PSS (Du Plessis, 2010;Eisenberger et al., 2002) or the mechanisms responsible for this association.It could be argued that PSS is a foundation or precursor of POS (Allen et al., 2003;Dawley et al., 2008;Rhoades & Eisenberger, 2002;Shanock & Eisenberger, 2006;Zagenczyk, Scott, Gibney, Murrell & Thatcher, 2010).
The proposed path of the relationship between these constructs is demonstrated in Figure 1 and provides the contextual framework for the study.

Research approach
A quantitative research approach was used for this study.
Quantitative research looks at numbers and statistical interpretation of the data gathered from questionnaires as opposed to looking at processes and meanings as in qualitative research (Creswell, 2007).Quantitative research is concerned with the facts or responses of participants.Moreover, a non-experimental design was used.This is the most widely used method and aims to describe and explore certain research topics (Maree, 2010;O'Neil, 2010).A cross-sectional design was used to gather the data in order to achieve the specific aims of this study.This study was conducted using a quantitative exploration paradigm, enabling the researcher to generate statistical analysis to investigate the research objectives of this study.
Various other studies have also opted to follow a quantitative survey research design approach during their exploration of possible relationships between an employee's POS and PSS (Chew & Wong, 2008;Dawley et al., 2008, DeConinck, 2010;Eisenberger et al., 2002;Jawahar & Hemmasi, 2006;Shanock & Eisenberger, 2006).This study was conducted electronically, distributing a selfadministered questionnaire through a Web survey tool, namely the SurveyMonkey website.

Research participants
Population refers to the complete set of cases or members (Saunders, Lewis & Thornhill, 2009).The chairperson of 21st Century Pay Solutions Group (Pty) Ltd made their database available, which enabled the use of the sample population of 5000 organisations and individuals in South Africa.
A total of 318 responses were received from the population of 5000.After cleaning the data, only 303 responses could be

Measuring instruments
Three constructs were consequently measured, namely: total reward components (which included the biographical questions), perceived organisational support and perceived supervisor support.

Total Reward Questionnaire
The Total Reward Questionnaire was compiled from the seven Retention Factor Scale and the WorldatWork reward model's five main components.The total reward instrument used by Moore (2010) showed a Cronbach's alpha of 0.821.Döckel (2003) developed the Retention Factor Scale (RFS) in order to measure participants' satisfaction with the following seven retention factors: compensation, job characteristics, opportunities for training and development, supervisor support career opportunities, work-life balance and commitment to the organisation.This RFS reported a Cronbach's alpha of 0.80 for the instrument's reliability.The Total Rewards Model Questionnaire contains the following components: • compensation (five items) • benefits (12 items) • work-life balance (eight items) • performance management and recognition (five items) • development and career opportunities (seven items).
Thus, the Total Rewards Questionnaire consists of 37 closedended questions.Two separate scales were used to determine current utilisation and level of importance.Current utilisation was indicated on a five-point scale ranging from 1 (no extent) to 5 (very great extent), whilst the level of importance was indicated on a six-point Likert-type scale ranging from 1 (not important at all) to 5 (extremely important), with a not applicable option (6).Question 6B of the Total Reward Questionnaire consisted of a ranking question where the participant could rank the five main components in order of importance according to their preference.

Survey of Perceived Organisational Support
The Survey of Perceived Organizational Support (SPOS) measures the employee's perception of the organisation's attitude towards them (Shore & Tetrick, 1991).As a result one can conclude that the SPOS aims to measure the POS of the employee.The shortened version of the SPOS consists of six items that require the respondent to indicate the extent of their agreement with each statement on a five-point Likerttype scale ranging from 1 (strongly disagree) to 5 (strongly agree).Du Plessis's (2010) study indicated a 0.863 Cronbach's alpha coefficient, which indicates that the SPOS is highly reliable and can consistently measure POS in a recruitment organisation.

Survey of Perceived Supervisor Support
This instrument aims to evaluate the employees' perception that their supervisor values their input and is concerned about their welfare.As a result one can conclude that the Survey of Perceived Supervisor Support (SoPSS) aims to measure the PSS of the employee.In order to assess this, it was adapted from the SPOS in the same manner as used by Eisenberger et al. (2002) and Shanock and Eisenberger (2006).They replaced the word organisation with the term supervisor throughout the SoPSS in order to determine the employees' PSS.The questionnaire consequently also consists of eight items and requires the respondents to score their answers on a seven point Likert-type scale ranging from 1 (strongly disagree) to 5 (strongly agree).Du Plessis's (2010) study indicated a 0.886 Cronbach's alpha coefficient, which indicates that the SoPSS is highly reliable.

Research procedure
A draft copy of the questionnaire was sent to five respondents randomly chosen for a pilot study.The final questionnaire was emailed to the potential respondents only after essential alterations indicated in the pilot study had been implemented.The researcher obtained ethical clearance from the chairperson of 21st Century Pay Solutions Group to make use of the database available.An orderly and systematic manner was followed to ensure that objectivity was maintained at all times and no comments or answers were imposed on the respondents.The primary researcher was available if any questions or concerns arose and data was recorded on the researcher's personal laptop as well as on a backup flash drive.No financial or non-financial incentives were used to encourage participation.

Statistical analysis
In this case, the data were extracted from the SurveyMonkey website and transcribed into an Excel file.This format made it possible to upload the data to a statistical software package.Analysis of the data was performed using the Statistical Programme for the Social Sciences (SPSS) program.Frequency tables represent the easiest kind of data analysis according to Hill and Lewicki (2007).This type of analysis was used due to descriptive nature of the study.
Factor analysis refers to a technique used to take a large set of variables and reduce or summarise the data using a smaller set of factors or components (Pallant, 2011).In this study factor analysis was used to determine how each http://www.sajhrm.org.zadoi:10.4102/sajhrm.v13i1.665 sub-component of the Total Reward Questionnaire loaded onto the five factors identified by WorldatWork (2007).The most general measure of reliability analysis is the Cronbach's alpha coefficient (Saunders et al., 2009).A cut-off point of 0.7 was used as a guideline for acceptable reliability (Field, 2009).
Pearson product-moment correlation was calculated as the analysis of the relationship between measured variables was obtained (Field, 2009).Multiple regression analysis was conducted to analyse the relationship between the total reward components and the variables, namely POS, PSS and biographical variables.In this case the t-test analysis was conducted to determine the significant differences in mean scores for gender group in terms of total reward preferences, POS and PSS.ANOVA calculations were used to determine if significant differences existed between the different generations' total reward and biographical variables.A confidence interval level of 95% was used to determine statistical significance where p was less than 0.05 (Pallant, 2011).

Results
The results and discussion thereof are reported according to the stated research aims and propositions.
Table 1 shows that the five total reward components are ranked or preferred as follows in order of importance: compensation, work-life balance, benefits, career development and opportunities and lastly performance management and recognition.
Aim 1: To determine the reward factors that are currently being used the most to retain employees The results of the gap analysis, which compared the means for current utilisation as well as level of importance of each factor under total reward, showed unique results.The study investigated seven main categories for total reward, namely compensation, communication work enabler, life convenience, benefits and safety, work-life balance, performance management and recognition as well as career development and opportunities.
More specifically, the study highlighted the following sub-points that are currently utilised to retain employees: base salary, provident or pension fund, comfortable work environment, performance support and training opportunities.

Aim 2: To determine the most important total reward factors to retain employees
In addition to current utilisation, the study also investigated the level of importance of the five main categories of total reward, namely compensation, benefits, work-life balance, performance management and recognition and career and development opportunities.The overall mean statistics for all the respondents were also analysed.
The results showed the total rewards in order of importance are, according to the study: compensation, work-life balance, benefits, development and career opportunities and performance management and recognition.More specifically, under compensation the sub-component market-related salary was rated the highest in terms of importance; under benefits provident or pension fund was rated highest, under work-life balance the highestranked component was community contribution and under performance management and recognition it was leadership style.Lastly, under development and career opportunities the sub-component organisational climate was rated the highest in terms of level of importance for participants.

Proposition 1: There is a significant difference in preference for total reward factors between managers and employees
To investigate this proposition, the mean differences and ANOVA analysis for the three main job level groups were analysed.The job level groups were general or executive managers, managers and staff.The three most important factors rated for general and executive managers (Group 1) were compensation, performance management and development and career opportunities.Junior and senior managers (Group 2) rated compensation, performance management and benefits and safety as most important to retain them, whilst staff, specialists, technicians and other, (Group 3) rated performance management, development opportunities and benefits and safety as the top three factors to retain them.
These results indicate that there are differences in retention preference factors between manager and employees.

Proposition 2: Different generation groups prefer different total reward factors
To investigate this proposition, the mean differences for the three main generation groups were analysed.The generation groups were determined according to age: Generation Y (aged 31 and younger), Generation X (32-47 years old) and Baby Boomers and Veterans (47 and older) (Lancaster & Stillman, 2002;Reynolds, 2005;Zemke et al., 2002).
A mean represents the arithmetic average of a group of scores; Table 2 shows that for Generation Y the top three most important total reward components are performance management, development and benefits and safety.For Generation X, performance management, development opportunities and compensation are important to retain them, whilst for Baby Boomers and Veterans the top three total reward components are compensation, performance management and benefits and safety.Overall, the different generation groups prefer more or less the same total reward components; therefore this proposition was rejected.
Proposition 3: Age, gender, race, qualification industry, job level, years with company and years remaining at company play a mediating role in the relationship between total reward and POS as well as that between total reward and PSS In order to test this proposition, the study used hierarchical regression and ANOVA analysis and the t-test.From the regression analysis only age, gender, race industry and job level showed significant results, indicating that only those biographical variables play a mediating role between total reward and POS as well as between total reward and PSS.

Proposition 4: A direct positive relationship exists between the employee's total reward and POS
This study investigated the Pearson product-moment analysis for this proposition; it revealed a strong significantly positive relationship between total reward and POS.This supports findings by Dawley et al. (2008), Du Plessis (2010), Harris et al. (2007) Jawahar andHemmasi (2006) and Riggle et al. (2009) that an employee's total reward are an originator or foundation for POS.This could be explained by the fact that an employee's perception of retention factors is supportive of the perception of how the organisation appreciates their input and is concerned about their general happiness.
The findings from Table 3 indicate that there is a strong practically significant positive correlation [r (df = 237, p > 0.001) = 0.298, medium effect] between total reward components and POS: high levels of total reward components are associated with an increase in POS.Table 3 further indicates that there is a strong practical significant positive relationship [r (df = 233, p > 0.001) = 0.250, medium effect] between total reward and PSS: high levels of total reward components are associated with high levels of PSS.
Table 3 portrays a large practically significant positive correlation [r (df = 230, p > 0.001) = 0.662, large effect] between POS and PSS in this study.This indicates that an increased perception of organisational support can be associated with an increased perception of supervisor support.More specifically, the correlation between total reward's seven subscales and POS as well as these seven subscales and PSS is illustrated in Table 4.
From Table 4, one can derive that each of the total reward subscales correlates significantly and positively with POS and PSS.However, between life convenience and POS there was no significant correlation nor was there correlation between life convenience and PSS.This is also true for benefits and safety and PSS, which shows a non-significant positive relationship.Thus, proposition 4, 5 and 6 were accepted.

Proposition 5: A direct positive relationship exist between the employee's total reward and PSS
The results of the Pearson product-moment correlation analysis displayed a medium practically significant positive correlation between employees' total reward and POS and between total reward and PSS.

Proposition 6: A direct positive relationship exists between POS and PSS
Pearson product-moment analysis was conducted and revealed a strong practically significant relationship between POS and PSS: high levels of POS are related with high levels of PSS.
The confirmation and strength of the various relationships are illustrated in Figure 2.
TABLE 2: Generation group differences and the level of importance for the seven total reward categories.

Discussion
The objective of this study was to investigate the relationship between total reward, POS and PSS.More specifically, the researchers sought to determine the reward factors that would retain employees in each of the four generations.

Current reward factors used in retention of employees
These results are in accordance with the literature (Nienaber et al., 2011), which shows performance and career management, work-life balance and variable pay are currently used to retain talent.The results are also in agreement with other literature of total reward models (Aon Hewitt, 2013;Armstrong, 2006;Armstrong & Thompson, 2002;Corporate Leadership Council, 2007;Towers Perrin, 2011;WorldatWork, 2007;Zingheim & Schuster, 2008) which explain financial and non-financial aspects of total rewards.All of these models are starting points from which organisations can design total rewards packages specific to their organisation's needs.
On the other hand, these results are is in contrast with other research (Chew & Chan, 2008;Mahaney & Lederer, 2006;Rynes, Gerhart & Minette, 2004;Tang, Luna-Aroca, Sutarso & Tang, 2004), which recognises that pay is known to be a possible precursor to lack of organisational commitment and intention to stay, but that payment alone will not be adequate to counter this.For instance, low payment might motivate an employee to leave, but high payment might not necessarily retain them.Thus, intrinsic rewards are equally important as pay; these include achievement, performance, challenging work, variety, responsibility and professional growth.
The results confirm previous studies that employees still regard compensation and remuneration as the most important factor to retain them at their respective organisations (Bussin, 2012;Eskom, 2009;Muteswa & Ortlepp, 2011;Parker & Wright, 2001).This contradicts numerous researchers who argue that promotion, career development and non-physical and intrinsic motivation factors are more important for employees than just compensation or remuneration (Deloitte, 2011;Mahaney & Lederer, 2006;SHRM, 2009;Tang et al. 2004;Towers Perrin, 2011).Metha et al. (2000) highlight that younger employees tend to prefer more physical rewards such as medical aid, maternity or paternity leave whilst older employees may value standard options or contributions to their retirement plans.
Proposition 1: There is a significant difference in preference for total reward factors between managers and employees It can be deduced that managers prefer compensation and performance management as they are already established in

Proposition 2: Different generation groups prefer different total reward factors
This research shows that there are some differences amongst generation preferences, especially total reward, which supports and corresponds to previous studies (Bussin, 2012;Deloitte, 2011;Reynolds, 2005;Zingheim & Schuster, 2008).
As shown by Jorgens (2003), Noble and Schewe (2003) and in South Africa by Moore (2010) as well as Nienaber et al. (2011), different generations do not have different reward preferences.Thus these previous studies differed from this study's results, which revealed that there are small differences in generational retention preferences.Moreover, these results are in contrast with research done by Giancola (2008), which states that there are no distinct generational differences and that differences in preferences correspond rather to individual life cycles and career stages.
It can be deducted that Baby Boomers and Veterans would choose compensation, work-life balance and benefits as the top three components that would convince them to stay at their respective organisations, as they place high value on hard work, obeying rules, dedication and military principles (Colon, 2005;Gesell, 2010;Hahn, 2011;Jacobson, 2007;Orciani, 2009;Zemke et al., 2001).The rating of benefits as the third most important was surprising as it was expected that preference for this reward would increase as the employee gets older.
Employees from Generation X are willing to develop their skills and take on challenges and are perceived to be very adaptive in the changing business world.Therefore, worklife balance, benefits, development and career opportunities as well as performance management and recognition are important to them in level of importance after compensation.
They are excellent at multi-tasking and working on projects simultaneously and place high value on work-life balance (Gursoy, Maier & Chi, 2008).Jacobson (2007) agrees with this statement and adds that Generation X employees view their jobs as temporary and themselves as free agents.
In contrast, Generation Y employees placed emphasis on compensation, work-life balance, and development and career opportunities as the top three total reward components.Generation Y employees favour teamwork and choose to follow orders; they prefer to work flexible hours in order to successfully complete their tasks in their own way (Dawn, 2004;Dwyer, 2009;Gesell, 2010;Gursoy et al., 2008;Iyer & Reisenwitz, 2009).Employees in this generation use the information channels that exist to familiarise themselves with the environment and display a constant need for knowledge.In addition, it can be believed that Generation Y employees seek challenges and learning and development opportunities in order to grow in their respective careers as they are open-minded and goal-orientated.
Proposition 3: Age, gender, race, qualification industry, job level, years with company and years remaining at company play a mediating role in the relationship between total reward and POS as well as between total reward and PSS The results in terms of these relationships are in accordance with previous studies by Twenge, Campbell, Hoffman and Lance (2010), Lawton and Chernyshenko (2008), Chiang and Birtch (2006), Gorman (2000), Hedge, Borman and Lammlein (2006), Nienaber et al. (2011) and Snelgar et al. (2013), which showed that reward preferences may differ according number of children, age, race job level, qualifications, years of service, marital status and gender.
In terms of age, this is explained in the discussion on proposition 2. There was a statistical significant difference identified for gender, with different scores for men and women in the work-life balance, performance management and development and career opportunities reward categories.This could mean that women have a higher preference for broader total rewards packages, as opposed to packages consisting of financial components only, which is in accordance with the literature (Chiang & Birtch, 2006;Nienaber et al., 2011;Snelgar et al., 2013), which found that women have stronger preferences for remuneration and benefits as well as a conducive work environment.
For race groups, there were statistical significant results mostly for Black, Indian and White employees.This supports previous research that culture and background play a role in preferences (Nienaber et al., 2011).A history marked by being ignored could explain why Black and Indian employees have different total reward preferences.These feelings could potentially extend to employment practices.
In terms of industry group, there was also statistical significant difference between all the industry groups, especially for communication work enabler and work-life balance total reward components.This could be due to organisational differences and preferences.For job level, statistical significant differences were explained in the discussion on proposition 1.

Proposition 4: A direct positive relationship exists between the employee's total reward and POS
This confirmation of this proposition is an important finding, since most evidence in the relevant literature shows that the impact of the employee's perception of retention factors and POS correlate.Similarly, findings from research shows that POS is harmfully connected to judgements of employees exiting the organisation (Hui et al., 2007;Van Schalkwyk, Els & Rothmann, 2011).Equally, POS was also positively linked to remaining with the organisation.

Proposition 5: A direct positive relationship exists between the employee's total reward and PSS
One would expect a correlation between these two concepts due to the unique total reward and support from supervisors.Supervisors are leader implementers of total reward practices and consequently they are associated with the employees' perception of the organisation's total reward practices.The employee's perception that the organisation's total reward practices improve talent retention will result in the perception that their supervisor values their input and cares about their general welfare.
This does affirm discoveries by Allen et al. (2003) and Knight-Turvey and Neal (2003) that supportive HR practices that demonstrate an organisation's willingness to invest in its talent will enhance talent retention.If organisations comprehend total reward practices, dedication of resources to these practices will lead to the acquisition and retention of top talent, as suggested by Giancola (2007), Heneman and Judge (2003), Nienaber et al. (2011) and Sung and Todd (2004).It can therefore be concluded that high levels of POS can be linked to an increased decision to stay the organisation concerned.This confirms the positive relationship between an employee's perception that a supervisor is willing to care for them and support them and the employee's intention to leave the organisation, as established by Dawley et al. (2008).
However, the direction of the relationship between PSS and POS is still debated, given that both directions have been confirmed in previous explorations.The extent to which employees perceive that their supervisors are concerned about their well-being and support them will therefore influence their perception that the organisation also cares about and supports them.This implies that a supervisor is a representative of the organisation through total reward practices and accordingly becomes a personification of total reward practices.On the other hand, one can argue that an employee's perception that the organisation values their contribution and cares for their overall well-being will direct them to believe that their supervisors are also positively inclined towards them.

Conclusion
The study of total reward and, specifically, different generation groups' perceptions regarding this field is ever changing and transformation still exists in the essence of knowledge.Talent management is a comprehensive, multidimensional concept with a myriad of perceptions that influence its effectiveness.It holds the potential to influence talent retention amongst different generations of employees.This potential ability of talent management to unlock capital has been the driving force behind it becoming a popular field of study.
The effect of work prospects and preferences on turnover intentions changes consistently, which implies that enriching employees' expectations of a 'brighter tomorrow' in their jobs can improve the probability of retention.So too does the present study propose that there is still much more to discover about the dynamic complex relationship between multi-generations' reward preferences and their perceived organisational or supervisor support.The key to attracting and most of all retaining excellent employees is based on an improved diverse total reward model that is a vital foundation for the employee value proposition.
(University of Johannesburg) helped with the data collection and contributed to the preparation of the manuscript.W.S. (University of Pretoria) was responsible for the research and fieldwork and wrote this article.
. This is a response rate of 4.3%.Of the 303 responses, the age group that provided the largest response was between the 48-66 years group (44.2%), 51.5% of the responses were from women, 72.9% of the responses came from White respondents and respondents with an honours degree provided 34.3% of the responses; the majority of the respondents were from the producer services sector (36.05%) and 33.7% of the respondents are in general or executive management positions.Respondents were grouped according to the length of time that they wished to remain at their current organisation.The group from which the largest number of responses came was intent to remain at the organisation for more than 5 years (55.8%). used

TABLE 1 :
Mean statistic of total reward components in order of importance for sample.

TABLE 3 :
Correlation table between total reward, perceived organisational support and perceived supervisor support.

TABLE 4 :
Correlation table between total reward subscales, perceived organisational support and perceived supervisor support.
Despite this study providing relevant insights into total reward amongst different generation within South Africa organisations, it is expected that this study has the usual limitations of questionnaire research.The sample of this study was a sample of convenience.The literature reviewed the perceptions and preferences of the generations regarding reward or remuneration packages.An effort should be made to investigate aspects such as organisational maturity or organisational life cycle.In terms of this study, a few methodological suggestions can be made.The six-point Likert current utilisation scale applied in the Total Reward Questionnaire should be further defined to improve the reliability and validity of responses.Another recommendation is that the age groups could be divided into intervals of 10 years.It is recommended that qualitative approaches and methods such as interviews or focus groups should be applied to complement the questionnaire.Organisations should encourage developing a culture of research and this must be supported by top management in order to encourage employees to participate in research studies.Future research should either focus on organisations in a bigger variety of industries or merely on different organisations in one specific industry to ensure that results can be easily generalised.Possible gaps for future research opportunities to explore include total reward, POS and PSS and their relation to performance of the organisation or talent management practices.