Positive organisation: The role of leader behaviour in work engagement and retention

How to cite this article: Mendes, F., & Stander, M.W. (2011). Positive organisation: The role of leader behaviour in work engagement and retention. SA Journal of Industrial Psychology/SA Tydskrif vir Bedryfsielkunde, 37(1), Art. #900, 13 pages. doi:10.4102/sajip.v37i1.900 Orientation: The positive organisation creates a framework in which its elements can be investigated in relation to the retention of talent.


Introduction
The current business environment has become unstable and is characterised by continuous change and renewal (Malone, 2004;Sellgren, Ekvall & Tomson, 2007). Globalisation has increased the workforce's mobility and employees are now expecting more from their employers than they previously did (Burke & Cooper, 2009). In order to be dominant in a global economy, organisations are forced to take an interest in more than mere profitability. These interests include the attraction, development and retention of talent (Boninelli & Meyer, 2004). The focus on talent is forcing organisations to adapt their business model in order to empower and engage their employees (Bakker & Demerouti, 2008).
within the organisation that lead to the development of human strength, foster vitality and fiourishing employees, make possible resihence and restoration and cultivate extraordinary individual and organisational performance (Cameron, Dutton & Quirm, 2003). Positive organisational behaviour fosters engaged employees and this is the key to ensuring high performance and overall Wellness for both the organisation and its employees, whilst increasing the commitment of employees, thereby lowering the risk of losing talent (McHugh, 2001).
The main outcome of a healthy, positive organisation is the retention of talent (Davenport & Harris, 2007;Ulrich, Brockbank, Johnson, Sandholtz & Younger, 2008). These organisations focus on their employees with as much passion and enthusiasm as they do on new processes and products (Bryan & Joyce, 2007). Talent management and the effective management of employee turnover is a central issue that is managed in order to avoid negative implications, such as high economic costs and disrupted social and communicative structures (Bergiel, Nguyen, Clenney & Taylor, 2009). An increase in profits, employees' happiness and productivity and customer satisfaction results from the retention of employees (Kaye & Jordan-Evans, 2000). The ability of leaders to empower their employees will have an impact on the organisation's labour retention (Kreisman, 2002;Taplin & Winterton, 2007).
According to Snyder and Lopez (20D2), leaders within an organisation play a vital role in designing a healthy work envirorunent that encoLirages the talent of the organisation to stay. Furthermore, leadership behavioLirs have a strong influence on employee and organisational outcomes (Chen & Silverthome, 2005), including work engagement and turnover intention. Engaged employees are aware of the organisational context and work with others to improve performance within their roles for the benefit of the organisation (Devi, 2009). When employees are engaged, they become less likely to leave the organisation. Baskin (2007) reports similar findings stating that an employee who is not engaged is more likely to leave the organisation.
Whilst it is expected that empowering behaviour by leaders wiU impact on psychological empowerment, work engagement and turnover intention, it is not clear what the effect of role clarity would be. There was no evidence in the literature exploring the possible mediating or moderating effects of role clarity on the relationships between leader empowering behaviour, psychological empowerment and turnover intention. The aim of this study was to investigate the relationship between leader empowering behaviour, role clarity, psychological empowerment and work engagement.
Leader empowering behaviour, role clarity and psychological empowerment Konczak, Stelly and Trusty (2000) identified six dimensions of leader empowering behaviour, namely: • the delegation of authority • the leader's ability to emphasise accountability • encouragement of self-directed decision-making • the leader's ability to share information • development of skills • coaching to promote innovation.
A leader's abihty to demonstrate these behaviours wul influence how employees perceive the tasks presented to them by tiieir leader (Wilson et ai, 2004). According to Greco, Laschinger and Wong (2006), employees will be empowered if a leader enhances the meaningfulness of work, allows participation in decision-making, facilitates the accomplishment of tasks, commimicates confidence in high performance and provides autonomy. A leader that utilises empowerment creates benefits for both the organisation and the employees, as empowerment improves the economic performance of an organisation and reduces role conflict and role ambiguity amongst employees (Greasley et al., 2008).
According to Mardanov, Heischmidt and Henson (2008), employee behaviour depends on the relationship between an employee and the leader, as experienced by the employee. Every employee within an organisation should have a specified set of roles and these allow the leaders of an organisation to hold the employee accountable for performance (Rizzo, House & Lirtzman, 1970). Mukherjee and Malhotra (2006) foimd that when a leader offers clarity in terms of these roles, a positive relationship results. Leader empowering behaviours influence employees' perceived role clarity in a positive way (Hong, Nahm & Doll, 2004;Nielsen, Randall, Yarker & Brenner, 2008). Leaders who provide guidance in terms of the tasks presented to employees create less uncertainty (Hong et al., 2004). Nielsen et al. (2008) support this in their findings that a positive relationship exists between supervisory consideration and perceived role clarity. Klidas, Van den Berg and Wilderom (2006) found that employees who indicated disempowerment due to leadership behaviours were experiencing low role clarity.
Role clarity consists of two concepts, namely role conflict and role ambiguity (Rizzo et ai, 1970), Role conflict occurs when two or more conflicting job requirements arise, so that complying with one would make doing the other more difficult (Rizzo et ai, 1970;Teh, Ooi & Yong, 2008). Role ambiguity refers to the lack of clarity and predictability of the outcomes of one's behaviour (Rizzo et ai, 1970;Slatten, 2008). Employees who feel empowered report low levels of role conflict and ambiguity (therefore higher levels of role clarity) in their roles because they are able to control their own environment (Greasley et al., 2008).
Psychological empowerment is a motivational construct manifested in four cognitions (Spreitzer, 1995): • meaning, which refers to the value of a work goal or purpose, judged in relation to one's own ideals or standards • competence, which is an individual's belief in his or her capability to perform activities with skill • self-determination, which indicates the individual's sense of choice in initiating and regulating action • impact, which is the degree to which an individual can influence strategic, administrative or operating outcomes at work. This motivational approach stresses psychological enabling as the main reason for an individual's feelings of empowerment (Conger & Kanungo, 1988;Menon, 2001;Spreitzer, 1995). Organisations which empower employees through greater meaning, competence, self-determination and Impact in their work experience positive outcomes (Thomas & Velthouse, 1990). Employees who are empowered offer the benefit of responding more quickly to environmental changes and stakeholder demands in comparison with their disempowered counterparts (Carson & King, 2005). According to Greasley et al. (2008), organisations with higher levels of empowerment have demonstrated improvements in various economic performance areas, global competition, the constantly changing business environment and the ability to deal with pressures to improve efficiency and performance. When employees experience empowerment they also demonstrate higher levels of engagement (Greco et al., 2006). Psychological empowerment is directed by the six dimensions of leader empowering behaviour (Konczak et ah, 2000). Leader empowering behaviour is correlated with greater feelings of empowerment (Greco et ah, 2006). According to Avey, Hughes, Norman and Luthans (2008) leadership style and psychological empowerment are significantly related to feeHngs of empowerment. Sauer (2003) found that leader empowering behaviour is significantly correlated to the degree of psychological empowerment that employees experience. There are two aspects of empowerment: empowerment as behaviour of a supervisor who empowers his or her subordinates and the psychological state of a subordinate resulting from his or her supervisor's ability to empower (Avey et al, 2008).
Eor the empowerment of employees to be successful, it is necessary to investigate the role of the leader because he or she has a substantial impact on the employee's perception of empowerment. It is the leader's responsibility to assess the employee's perceived feelings of empowerment (Greasley et al, 2004). An empowered organisation is one in which a leader encourages employees by involving them in decisionmaking and assigning responsibility to them (Malone, 2004). When leaders are effective in using empowering behaviours, employees are aware of the expectations placed upon them (role clarity) and they feel confident (empowered) in achieving them; consequently employees experience higher levels of engagement (Greco et al, 2006).

Work engagement and turnover intention
Work engagement is defined as a positive, fulfilling, work-related state of mind that is characterised by vigour, dedication and absorption (Schaufeli, Salanova, González-Romá & Bakker, 2002). Role clarity has been found to play a role in work engagement (Harter, Schmidt & Hayes, 2002;Rüssel, 2008;Saks, 2006;Steele & Fullagar, 2009). When expectations are not clarified employees are not engaged, as is evident in the expression of negative emotions like boredom and resentment (Harter et al, 2002). Increasing clarity of expectations was found to increase positive emotions that led to engagement of employees (Rüssel, 2008). When roles are not clearly defined, the likelihood of an employee's intentions to leave that job wül increase because of the lack of role engagement (Steele & FuUagar, 2009). Employees who perceive their environments as predictable and consistent are more engaged in their work (Saks, 2006). According to Konrad (2006), the more transparent managers are in terms of the organisation's operations, the more engaged employees wül be. Coffman (2002) endorses this, stating that the best way in which to engage a workforce is through offering employees clarity in terms of the desired expectations of their roles. This is confirmed in the findings of Prieto, Salanova, Martinez and Schaufeli (2008) that role stress are negatively related to work engagement, with role ambiguity or clarity being a strong predictor of dedication, a sub-construct of work engagement.
Intention to leave is the strength of an individual's viewpoint that he or she does not want to stay with his or her employer (Kahumuza & Schlechter, 2008). Intention to leave refers to the ultimate cognitive stage in the decision-making process of an employee, where quitting and searching for alternative employment occurs actively (Park & Kim, 2009). According to Park and Kim (2009), an employee's intentions to leave an organisation consist of both thoughts and statements, but these intentions can differ from the employee's actual behaviour. Intention to quit has been found to be a strong predictor of actual turnover and may be the most important antecedent of employee turnover (Kahumuza & Schlechter, 2008;Park & Kim, 2009).
An employee's intention to leave is a signal valid indicator of quitting (Weisberg, 1994). According to Grobler, Warnich, Carrell, Elbert and Hatfield (2006), a high turnover in employees costs South Africa several millions of rands a year through decreased productivity, increased accidents and quality problems. Tapün and Winterton (2007) found that orgarüsations with a low record of turnover encompassed leaders who perceived turnover as a costly expense and so took a pro-active approach in dealing with the problem. Turnover is an outcome that a positive, healthy organisation avoids because it is disruptive and consequently costly (Grobler et al, 2006). Greco et al (2008) found that nurses who experienced psychological empowerment were more engaged m their work. Empowered employees demonstrate the characteristics of an engaged employee (Avey et al, 2008;Dvir, Eden, Avolio & Shamir, 2002;Greasley et al, 2008;Reynders, 2005). Avey et al. (2008) found that empowerment is sub-sequential to engagement. This is substantiated in the findings of Stander and Rothmann (2010) who foimd that psychological empowerment was a statistically significant predictor of employee engagement. Competence and meaning are two sub-scales that encompass psychological empowerment and have been found to lead to work engagement. When employees experience meaning in their work they experience engagement (Luthans & Peterson, 2001;May & Harter, 2004). Engaged employees view themselves as competent in dealing with their job demands; they have positive self-efficacy (Schaufeli et ai, 2002). Reynders (2005) found in her study of employees in a government institution, that higher levels of psychological empowerment encompass increased levels of work engagement.
Work engagement is related to attitudes, intentions and behaviours of employees (Saks, 2006). Engagement can be utilised as a tool to reduce employees' intentions to leave the organisation (Baskin, 2007). According to Bhatnagar (2007), employee engagement is the most effective way in which to retain talent. Nurses' intention to leave was studied by Karlowicz and Temus (2007) and it was foimd that the lack of engagement was one of the most important issues contributing to intention to quit. Simpson (2009) found very similar results in her research with nurses, where higher turnover cognitions correlated directly with low work engagement levels. According to Saks (2006), employees who are more engaged are more tirusting of their employer and therefore report more positive attitudes and intentions towards the organisation. It was also found that job and organisation engagement predicted an employee's intention to quit (Saks, 2006).
Based on the review of the literature, the hypothesised conceptual model of this study is graphically depicted in Figure 1 below: Based on the aforementioned discussion, the following hypotheses are formulated: • Hypothesis 1: Leader empowering behaviour, role clarity and psychological empowerment predict work engagement within the business unit. • Hypothesis 2: Role clarity moderates the relationship between leader empowering behaviour, psychological empowerment and work engagement within the business unit. • Hypothesis 3: Work engagement predicts intention to leave within the business unit.

Research approach
The objectives set out for this research wül be achieved through a survey design (Shaughnessy & Zechmeister, 1997). Information regarding respondents' experience of leader empowering behaviour, role clarity, psychological empowerment, work engagement and intention to leave was obtained through the administration of surveys.

Research method Participants
This business unit consisted of 240 employees holding various positions such as managers, specialists, supervisors and administrative staff. Of the total questionnaires distributed 179 (75%) were retiirned. Table 1 illustirates the characteristics of these participants.

Measuring instruments
The Leader Empowering Behaviour Questionnaire (LEBQ) was developed by Konczak et al. (2000) and is aimed at providing leaders with feedback with regard to employees' behaviour that relates to employee empowerment. The original instrument consists of 17 items and is scored on a 7-point Likert scale ranging from 'strongly disagree' (1) to 'strongly agree' (7). A typical item is 'My manager gives me the authority I need to make decisions that improve work processes and procedures' (Konczak et ai, 2000, p. 307) and a high score signifies high leadership empowering behaviour. Two items were added from Arnold, Arad, Rhoades and Drasgow (2000) with the aim of increasing the number of items that demonstrated the 'information sharing' dimension. These items are 'My manager explains his/her decisions and actions to my work group' and 'My manager explains company goals to my work group'. In previous research (Konczak et ai, 2000) the interfactor correlations ranged from 0.40 to 0.88 whilst a Cronbach alpha coefficient of 0.95 for reliability was found (Dwyer, 2001). Mare (2007) found Cronbach alpha coefficients ranging from 0.57 to 0.78 in a large sample within a gold mining industry in South Africa. Mare (2007) fotmd a one factor sti-ucture for LEBQ. Tjeku (2006) and Dwyer (2001) fotmd that a 3-factor model was best in a study done within a steel manufacturing organisation. For the purpose of this study a simple principal components analysis was carried out on leader empowering behaviour items (as measured by thé LEBQ). An analysis of the Eigenvalues ( > 1.00) and scree plot indicated that three factors could be extracted, which explained 69.7% of the total variance. These were named Development, Accotmtability and Authority. The

Measures of Role Conflict and Ambiguity Questionnaire
(MRCAQ) was developed by Rizzo et al (1970) and is aimed at identifying role conflict and role ambiguity within complex organisations. The original instrument consists of 30 items, 15 of which deal with role ambiguity (even numbers) and 15 with role conflict (odd ntimbers). The measure is scored on a 7-point scale ranging from 'very false' (1) to 'very true' (7). Mukherjee and Malhotra (2006) fotmd that the role ambiguity items of this measure should be renamed 'role clarity' and this is substantiated in other research (Beehr, Glazer, Fischer, Linton, & Hansen, 2009;Bray & Brawley, 2002) and tiierefore, for this study these items were utilised because role clarity is of interest. A typical item is 'explanation is clear of what has to be done' (Rizzo et al, 1970, p. 156) and a high score indicates high role clarity (or low role ambiguity). In previous research (Koustelios, Theodorakis & GouUmaris, 2004) the reliability was fotind to be adequate with Cronbach alpha coefficients for this instrument ranging from 0.85 for role ambiguity and 0.86 for role conflict. In Mukherjee and MaUioti-a's (2006) study conducted on 342 call centre employees, a Cronbach alpha coefficient of 0.85 was obtained. A simple principal components analysis was carried out on Role clarity items (as measured by the MRCAQ). An analysis of the Eigenvalues ( > 1.00) and scree plot indicated that two factors could be extracted, which explained 44.95% of the total variance. Since this study is focusing on role clarity, only the items loading on this factor were utilised.
The Measuring Empowerment Questionnaire (MEQ) was developed by Spreitzer (1995) and is aimed at measuring the participants' psychological empowerment. The instrtiment consists of 12 items and is scored on a 7-point scale ranging from 'strongly disagree' (1) to 'strongly agree' (7). The measure has four sub-dimensions, namely (Spreitzer, 1995): Typical items within these sub-dimensions are, for:  (2009) fotmd a 4-factor structure for the MEQ, which is in line with previous research (Spreitzer, 1995). Confirmatory factor analyses on MEQ which were conducted with AMOS for the purposes of this study showed that a 4-factor model of psychological empowerment (consisting of Meaning, Competence, Self-determination and Impact) fiti:ed the data best (x^/df = 3.13; CFI > 0.90; RMSEA < 0.08).
The Utrecht Work Engagement Scale (UWES) was developed by Schaufeli et al (2002) and is aimed at measuring the participants' work engagement. The instrument consists of 17 items and is scored on a 7-point frequency scale, ranging from 'never' (0) to 'daily' (6  Storm and Rothman (2003). Research in various countries including South Africa showed that the fit of the hypothesised 3-factor structure to the data was superior to that of alternative factor models (Seppälä et al, 2008;Storm & Rothmarm, 2003). Confirmatory factor analyses on the UWES which were conducted with AMOS for the purposes of this study showed that a 3-factor model of work engagement (consisting of Vigour, Dedication and Absorption) fitted the data best (xV#= 3.53; CFI > 0.90; RMSEA < 0.08).
The Intention to Leave Scale (ILS) was developed by Firth, MeUor, Moore and Loquet (2004) and is aimed at measuring the strength of participants' intentions to leave. The instrument consists of two items and is scored on a 5-point scale ranging from 'very often' (1) to 'rarely or never' (5). A typical item is 'How often do you think of leaving your present job?' (Firth et al, 2004, p. 187). A high score reflects a low intention to leave. Firth et al (2004) found a Cronbach alpha coefficient of 0.75, which is an adequate reliability score. The ILS consists of only two items and therefore a factor analysis investigating the loadings of items was not necessary.

Research procedure
An entire business unit (A' = 240) in a chemical organisation was approached by means of a convenience sample. Permission was granted from the management team as well as the employees. Participation in this study was voluntary and all information was treated anonymously and with high regard to respondents' confidentiality. The raw data captured was converted by means of the SPSS program for further analyses.

Statistical analysis
In order to answer the research questions the SPSS programme (SPSS Inc, 2007) and AMOS program (Arbuckle, 2006) were used. Exploratory factor analysis was performed for this study to investigate the factor structure of the Leader Empowering Behaviour Questionnaire (LEBQ) and Measures of Role Clarity and Ambiguity Questionnaire (MRCAQ). Initially a principal components analysis was conducted on the constructs so that the Eigenvalues and scree plot could be investigated and so the number of factors could be extracted. Thereafter a principal axis factor analysis with a direct obUmin rotation was performed in order to determine factor loadings. Structural equation modelling, as implemented in AMOS (Arbuckle, 2006), was used to test the factorial models of the MEQ and the UWES, by using the maximum likelihood analyses. The following indexes produced by AMOS were used in this study: the Chi-square statistic {f), which is the test of absolute fit of the model, the Comparative Fit Index (CFI) and the Root-Means-Square Error of Approximation (RMSEA).
Descriptive statistics were used to analyse the distribution of scores. Pearson correlation coefficients were used to determine the relationships between the variables. The statistical significance was set at p < 0.05. Effect sizes were used to assess the practical significance of the correlation coefficients (Steyn, 2005). A cut-off point of 0.30 (medium effect) was set for practical significance of correlation coefficients (Cohen, 1988).
Canorucal analyses were used to determine the relationships between sets of constructs (Tabachnick & Fidell, 2001). Hierarchical multiple regression analyses were conducted with the variables in their continuous form. In the first step, the scales of the LEBQ were entered into the regression equation. In the second step the role ambiguity items of the MRCAQ were entered, whilst in the third step the scales of the MEQ were entered.
To investigate the moderating effects of role clarity, the predictors (i.e. leadership empowerment behaviour and psychological empowerment) and moderator (i.e. role clarity) were entered into a hierarchical multiple regression equation (with work engagement as dependent variable), followed by their interaction in the second step. The interaction term is represented by the product of two main effects (Aiken & West, 1991). Also, in line with the procedure suggested by these authors, the independent variables and the moderator were centred before testing for the significance of the interaction term. To centre a variable, scores are put into deviation score form by subtracting the sample mean from all individuals' scores on the variable, thus producing a revised sample mean of zero.

Results
Descriptive statistics, Cronbach alpha coefficients and correlation coefficients for all of the constructs which were measured are illustrated in Table 2. Table 2 shows that the Cronbach alpha coefficients for leader empowering behaviour, role clarity, psychological empowerment, engagement and intention to leave are all acceptable according to Foxcroft and Roodt (2005) who state that a Cronbach alpha coefficient of 0.65 or higher is acceptable. Table 2 Shows that Development is positively related to Role clarity. Impact and Self-determination (practically significant, large effect). Development also correlated positively with Meaning, Vigour, Dedication, Absorption and Intention to Leave (practically significant, medium effect). Accountability correlated positively with Competence and Self-determination (practically significant, medium effect), whilst Authority is positively related to Selfdetermination (practically significant, large effect). Authority correlated positively with Role clarity. Meaning, Impact and Intention to Leave (practically significant, medium effect). Role clarity related positively to Competence, Impact, Selfdetermination, Vigour, Dedication and Intention to Leave (practically significant, medium effect). Role clarity also correlated with Meaning (practically significant, large effect). Meaning correlated positively with Vigour, Dedication and Absorption (practically significant, large effect). Meaning also related positively to Intention to Leave (practically significant, medium effect). Impact related positively with Dedication (practically significant, large effect); and with Vigour, Absorption and Intention to Leave (all practically significant, medium effect). Self-determination correlated positively with Vigour and Dedicatiori (both practically significant, medium effect). Vigour and Absorption are positively related to Intention to Leave (practically significant, medium effect). Dedication relates positively with Intention to Leave (practically significant, large effect).
The multiple regression analysis shows that leader empowering behaviour, role clarity and psychological empowerment predicts a large percentage of the variance in engagement. More specifically, leader empowering behaviour, role clarity and psychological empowerment explains 43% of the variance in Vigour, 61% of the variance in Dedication and 38% of the variance in Absorption.
Based on the aforementioned statistical analysis hypothesis 2 is accepted. Leader empowering behaviour, role clarity and psychological empowerment predict work engagement within the business unit.

Moderation effects of role clarity:
The moderation effects of role clarity on the relationship between leadership empowering behaviour as (measured by the LEBQ) and psychological empowerment (as measured by the MEQ) and Work Engagement (Dedication and Absorption) were tested with hierarchical regression procedures. In an attempt to test the possibility of interaction effects, the centred predictors and moderators were entered first into the regression equation followed by their interactions in the second step to predict facets of work engagement. The results of the hierarchical regressions are reported in Table 5. Table 5 shows that the interaction terms amongst leader empowering behaviour, psychological empowerment and role clarity to predict Dedication and Absorption yielded statisticaUy significant effects lFa^¡^^^¡Jl5, 163) = 19.60, AR' = 0.03, p < 0.05 and F^^sorpHon 0^^' ^63) = 8.25, AR' = 0.05, p < 0.05]. Although small, the significant interaction effects were plotted as indicated by Figures 2, 3 and 4. Figure 1 shows that at a high level of Role Clarity (compared with a low level); high Competence was associated with a lower level of Dedication. Figure 2 shows that at a high level of Role Clarity (compared with a low level), high Meaning had a stronger effect on Dedication. Figure 3 shows that at a low level of Role Clarity (compared with a high level); high Development had a stronger effect on Absorption. Therefore, hypothesis 2 is accepted. Role clarity moderates the relationship between leader empowering behaviour, psychological empowerment and work engagement within the business unit.
Finally, a regression analysis was computed to determine if engagement and its three sub-scales predict intention to leave. The results are reported in Table 6. Table 6 summarises the regression analysis with work engagement as a predictor of Intention to Leave. The regression analysis produced a statisticaUy significant model  3,175) = 24.25;/? = 0.00), accounting for approximately 29% of the variance. More specifically, it seems that Dedication (ß = 0.43, t = 4.51, p = 0.00) predicts Intention to Leave. This result provides support for hypothesis 3.

Discussion
The aim of this study was to determine firstly, whether leader empowering behaviour, role clarity and psychological empowerment predict employee engagement, secondly if role clarity moderates the relationship between leader empowering behaviotir, psychological empowerment and work engagement and finally to determine if work engagement predicts intention to leave within the business unit.
The analyses showed that a leader's behaviour is related to employees' experiences of the work environment. A higher level of development was related to higher role clarity. Therefore, when a leader focuses on the development of employee, they are more aware of the expectations that are placed upon them. Higher levels of development relate to higher levels of impact; therefore, an empowering leader enstires that employees feel they can influence their work (Nielsen et al, 2008). Furthermore, higher levels of authority correlated with higher levels of self-determination, indicating that when a leader delegates authority appropriately, employees will experience autonomy in determining how to fulfil the expectations placed upon them. It is clear that leader empowering behaviour has a strong relationship with role clarity and psychological empowerment.
High levels of meaning correlated with high levels of vigour, dedication and absorption. Therefore, when employees experience their work as meaningful they will concurrently experience higher levels of energy in doing their work, be more enthusiastic in completing work-related tasks and demonstrate high levels of focus in their work. High levels of impact are related to elevated levels of dedication, which indicates that when employees feel they have control over their work environment they wül react with increased eagerness in doing their work. This is supported in the findings of Stander and Rothmann (2010).
When employees experience high levels of dedication they WÜ1 be less likely to have intentions of leaving. Employees who are enthusiastic about their work are more likely to have positive emotions about their work environment and as a result will be less likely to think about leaving their organisation. Hence, the more engaged an employee is the less likely he or she wül be to have cognitions of leaving (see Simpson, 2009).
The canonical analysis showed that leader empowering behaviour (development, accotmtabuity and authority), role clarity and psychological empowerment (competence, meaning, impact and self-determination) are strongly related to the three categories of work engagement (vigour, dedication and absorption) and intention to leave.
The results of the multiple regression analysis showed that vigotir was predicted by development, role clarity and meaning. Therefore, when leaders provide employees with frequent opportunities to develop their skills and ensure that expectations are clearly stated and when employees experience their work as meaningful, they will respond with positive affective responses like physical strength, emotional energy and cognitive liveliness. Rothmann and Jordaan (2006) found similar results in their study within a higher education institution.
Dedication, according to the multiple regression analysis, was predicted by development, role clarity, authority, competence, meaning and impact. This implies that when a leader is perceived as supportive, when information is shared and when authority is appropriately delegated, employees wul derive a sense of significance from their work. Furthermore, when employees feel competent in their work, experience this work as meaningful and feel they have the ability to influence their work environment, the result will be employees who feel proud, enthusiastic and inspired. These findings are corroborated in previous research (Greco et al, 2006).
The multiple regression analysis also showed that absorption was predicted by development, authority, competence and meaning. Employees who are totally immersed in their work are led by managers who make development a priority and share their authority (Quesada, González & Kent, 2008). Employees who believe in their capabilities and who experience alignment of their work roles, beliefs, values and behaviours are more likely to be irruriersed in their work.
With reference to the interaction effects, the following was found. Firstly, role clarity interacted with competence to affect employees' dedication. When role clarity is high, employees with a low competence experience higher levels of dedication. In other words, employees who experienced a low level of competence were more dedicated when role clarity was high. So role clarity is less of an issue for employees who feel competent, especially as far as their dedication is concerned.
Secondly, role clarity interacted with meaning to affect employees' dedication. Employees who experienced a high level of meaning were more dedicated when they experienced high role clarity. Finally, role clarity interacted with the developing of employees (as a facet of leader empowering behaviour) to affect absorption. Employees who experienced low role clarity were more absorbed in their work when they experienced that their leaders developed them.
Lastly, a regression analysis showed that dedication (a subconstruct of engagement) predicts an employee's intention to leave. Therefore, when employees feel inspired by their work and experience their work as challenging they wul be less likely to have thoughts of leaving the organisation or their current position (Karlowicz & Ternus, 2007).
In conclusion, the research found statistically significant relationships between leader empowering behaviour, role clarity, psychological empowerment, work engagement and intention to leave. It was also evident that development and meaningful work plays an extremely important role in the retention of talent. The retention of talent is a vital element in creating a positive organisation (Davenport & Harris, 2007;Ulrich et al, 2008). This research is beneficial in highlighting the importance of employee development and empowerment in creating a positive organisation and, consequently, ensuring the retention of talent.
Based on the results various recommendations can be made. Organisations that want to be market leaders need to recognise the importance of focusing on overaU Wellness for both the organisation and its employees. The interest in applying positive psychology principles to the workplace is proving to be extremely beneficial, resulting in lower absenteeism, lower turnover, decreased stress levels and diminished alcohol and tobacco usage (Wilson et al, 2004). It is recommended that interventions focusing on the aspects promoting overall Wellness be implemented in the business unit. Therefore, it is essential that the business unit understand the elements encompassed by a healthy organisation.
If the business unit wishes to benefit in terms of building a positive organisation and, consequently, talent retention, it should adopt empowerment behaviour in its management style. According to Nedd (2006), a leader has the strongest impact on an employee's intention to stay. Therefore the importance of leader behaviour in talent retention is vital. Development was found to be one of the most important aspects of leader empowering behaviour relating to retention (i.e. lower intentions of leaving). This is substantiated in previous research that foimd professional development to be the most important aspect leading to the retention of employees (Loeb & DarUng-Hammond, 2005;Rosser & Townsend, 2006). Taplin and Winterton (2007) found that a proactive approach to avoiding the costs involved in employee turnover is best and that serious investment in training may be the answer for many organisations. Stander and Rothmann (2009) reiterate this by identifying the development of employees as a key competence for managers. They point out that in order to be a good people developer, managers should be coached and developed to delegate authority, hold employees accountable for outcomes, lead by example, encourage subordinates, show concern for others' feelings, allow participative decisionmaking, share information and coach and mentor people. It is therefore advisable that the business unit apply leader empowering behaviour practically. The foUowing can act as a guide: • Delegation of authority: leaders within the business unit must create an environment that encourages employees to be involved in decision-making. • Accountability: it is the leaders' responsibility to ensure that all employees within the business unit are held accountable for the work they are assigned to, for performance and results and for customer satisfaction. • Self-directed decision-making: leaders within the business unit must allow the employees to utilise their skills in formulating solutions independently, thereby allowing them to make decisions that affect their work. • Information sharing: employees within the business unit must be given all the necessary information by the leader so that they are able to ensure high quality work performance within their assigned roles.
• Skill development: leaders must make continuous learning, skills development and employee problemsolving a priority within the business unit.
It is also important to ensure that the employees' roles are clarified through the provision of the necessary information regarding expectations placed upon them. The extent to which information is successfully received and understood is also important. Tasks must be communicated to employees in such a way that their fit and function within the organisation is comprehensively understood. Leaders must ensure that employees have clear career paths, detailed job models and a structured process to consult when clarification of expectations is needed.
Interventions employed within the business unit should also take engagement into account as engagement contributes to the enhancement of work-Ufe and promotes the well-being of employees (Bakker & Demerouti, 2008). Through the utilisation of engagement, employees wiU become happier in their work environment and be less likely to think of leaving. Wildermuth and Pauken (2008) state that the first step that leaders need to take in engaging em.ployees is to ensure that they themselves are engaged. Adopting this approach will ensure that employees are enthusiastic, energetic and focused on their work, which will essentially contribute to organisational success. Engaged employees promise higher productivity, improved customer satisfaction, increased profits and good safety records (Saks, 2006). The role of meaningnil work is extremely important with regard to engagement efforts. The business unit should conduct stayin interviews, group discussions and meetings in order to establish whether employees experience their work as meaningful. Interventions can also focus on increasing the meaningfulness of work in order to increase the engagement of employees within the business unit (Dychtwald & Morison, 2006).
The following limitations have been noted in terms of this study. The research design was cross-sectional and this limits the determination of cause-and-effect relationships; also, the participants' opinions, attitudes and feelings are representative at only one point in time. The sampling technique involved targeting an entire business unit. Although a larger sample would have been more beneficial, the sample size of 179 is a reasonable and representative sample size. The measures that were administered were in English and this may have been a limitation in the way in which items were understood by participants who were not English-speaking. Eurthermore, these measures were selfreport measures and this may lead to 'method variance'.
Longitudinal studies should be employed to establish the causal relationships amongst the variables. To enhance external validity, the sample size of 179 should be expanded, both to reach a larger sample size as well as to obtain the involvement of more organisations. It would be beneficial to investigate the effect of leader empowering behavioLU", role clarity, psychological empowerment and work engagement on absenteeism, general health and employee Wellness (e.g. alcohol and substance abuse and stress).