In this study, job performance is considered to be those actions, behaviours and outcomes that individual employees engage in or bring about that contribute to the goals of the organisation (Rotundo & Sackett 2002). Job performance is influenced by three principal factors (Farh, Seo & Tesluk 2012; Kacmar et al. 2009), namely: declarative knowledge (knowledge about facts, principles and objects), procedural knowledge and skill (ability to implement declarative knowledge), and motivation (choice to expend effort, level of effort and persistence). In addition, general mental ability has been found to be one of the most important predictors of job performance (Schmidt & Hunter 1998). Earlier research by Welbourne, Johnson and Erez (1998) had led to the development of the Role-Based Performance Scale, which is widely acknowledged by various authors (e.g. Bray & Brawley 2002; Jawahar & Raghavendra 2011; Waddar & Aminabhavi 2012) as an authentic instrument for measuring the aforementioned predictors of job performance in the workplace. With reference to its outcomes, job performance has connections with a number of organisational outcomes that include customer service and product quality (Blignaut 2011), manager-employee relationships and employee turnover (Dalal & Hulin 2008), as well as job satisfaction (Bono & Judge 2003). Therefore, job performance appears to be multifaceted with regard to its antecedents and has a wide spectrum of applications in terms of its outcomes.

For the purposes of this study, organisational citizenship behaviour is perceived as ‘individual behaviour that is discretionary, not directly or explicitly recognised by the formal reward system and that, in the aggregate, promotes the effective functioning of the organisation’ (Organ 1988:4). Organ’s (1988) five-factor structure identifies five dimensions − specifically, altruism, conscientiousness, gamesmanship (not complaining about trivial matters), courtesy and civic virtue (maintaining an awareness of matters that affect the organisation) – and is one of the most well-known organisational citizenship taxonomy amongst researchers. The elements listed in that taxonomy were translated by Spector, Bauer and Fox (2010) into a checklist of measuring organisational citizenship behaviours in the workplace, which is known as the ‘Organisational Citizenship Behaviour Checklist’. On the one end, predictors of organisational citizenship behaviour include job satisfaction (Podsakoff et al. 2000), perceptions of organisational justice (Organ, Podsakoff & MacKenzie 2006), organisational commitment and personality characteristics (Organ et al. 2006), task characteristics (Todd & Kent 2006) and leadership behaviour (LePine, Erez & Johnson 2002). On the other end, organisational citizenship behaviour has an influence on a number of organisational factors that include sales performance as well as operating efficiency (Nielsen, Hrivnak & Shaw 2009), customer satisfaction and loyalty (Hoseini, Karim & Soodi 2013), service and product quality (Bel & Simon 2002), recruitment and selection (Podsakoff et al. 2011), task performance and overall performance (Dalal 2005). Its overarching influence gives the impression that it is imperative for managers to cultivate organisational citizenship cultures in organisations if the positive outcomes are to be realised.

There is no single universally accepted definition for quality of work life. However, in this study, quality of work life is defined as an individual's evaluative reactions to and satisfaction with his or her work conditions and the total working environment (Penny & Joanne 2013). Sirgy et al. (2001) suggest that quality of work life is multifaceted and is influenced by a collection of job-related factors. These include job satisfaction, job involvement, work role ambiguity, work role conflict, work role overload, job stress, workload, work-family balance, shift work, involvement in decision-making, professional isolation, recognition, relationships with peers and supervisors, and opportunities to learn new skills (Ellis & Pompli 2002). A popular instrument for measuring these factors is the Work-Related Quality of Life (WRQoL) scale developed by Van Laar, Edwards and Easton (2007). In terms of the outcomes of quality work life in an organisation, it is considered important to provide employees with a healthy quality of work life, since this is linked with a number of desirable outcomes, such as the ability to increase task performance, lower absenteeism and turnover rate, lower tardiness frequency, and increased organisational effectiveness and organisational commitment (Rathi 2009; Wilson et al. 2004).

Transformational leadership is a form of leadership that creates positive change in the followers by enabling them to take care of each other's interests and act in the interests of the group as a whole (Yukl 2002). Unlike transactional leadership, in which the leader manipulates the followers by offering rewards in return for satisfying the demands of the leader, the transformational leader enhances the motivation, moral and performance of his follower group (Dionne et al. 2004). According to Bass and Aviolio (1994), the characteristics of a transformational leader can be summarised through four critical components, namely: charisma or idealised influence, inspirational motivation, intellectual stimulation and personal and individual attention. These scholars further developed the Multifactor Leadership Questionnaire (MLQ), which has since been widely used to measure transformational leadership, as reported by Alsayed, Motaghi and Osman (2012). Another study by Rafferty and Griffin (2004) identified five sub-dimensions of transformational leadership, which are vision, inspirational communication, intellectual stimulation, supportive leadership and personal recognition. Such leadership has been shown to enhance both follower performance (Jorg & Schyns 2004; McColl & Anderson 2002; Yu, Leithwood & Jantzi 2002) as well as organisational performance (Boerner, Eisenbeiss & Griesser 2007; Waldman, Javidan & Varella 2004; Zhu, Chew & Spangler 2005). On the overall, by virtue of its impact on personal and organisational outcomes, transformational leadership is generally highly regarded by most management and organisational researchers and practitioners (Tucker & Russell 2004).

Previous literature seems to link transformational leadership to organisational citizenship behaviour. For instance, as suggested by Rafferty and Griffin (2004), transformational leaders tend to display a number of developmentally-oriented behaviours, such as coaching followers, identifying appropriate training courses for followers to undertake and encouraging followers to develop their job-related skills and abilities in order to foster their self-confidence on undertaking a wide range of proactive tasks. Emanating from these efforts will be employees who are prepared to voluntarily exercise organisational citizenship behaviours (Ehrhart 2004). The inspiration provided by transformational leaders tends to motivate subordinates to articulate the future, providing meaning and challenge to their works, which in turn activates organisational citizenship behaviours (Bass & Riggio 2006). The results of a study conducted by Humphrey (2012) reveal that transformational leadership predicts organisational citizenship, but the relationship was mediated by organisational identification. Transformational leadership also enables followers to exceed what is expected of them, thereby positively impacting on extra-role performance through organisational citizenship behaviours (Vega-Vázquez, Cossio & Martin-Ruiz 2012). Consequently, the transformational leader is able to alter the priorities and aspirations of employees by directing them towards becoming more committed towards the organisation (Vila, Perez, & Morillas 2012). Additionally, a number of previous studies (Pillai, Schriesheim & Williams 1999; Podsakoff, MacKenzie & Bommer 1996; Wang et al. 2005) attest to the existence of a positive relationship between individualised considerations given by transformative leaders and organisational citizenship behaviours. A fairly recent study conducted by López-Domínguez et al. (2013) shows that there are mediated relations between transformational leadership and other dimensions of organisational citizenship behaviour. These empirical results lead to the following hypothesis:

H₁: Transformational leadership has a positive influence on organisational citizenship behaviour amongst supply management professionals.

An examination of the transformational leadership-quality of work life nexus appears to suggest the existence of a positive linkage between the two factors. Yair and Linton (2003) found that both transformational leadership and transactional leadership are related to the establishment of a quality environment. Goh and Geoffrey (2005) concluded that transformational leadership is positively linked to a number of organisational outcomes, such as job performance and ethical conduct, which are also related to quality of work life. Omolayo (2007) concluded that workers under transformational leadership styles experience less job-related tension than workers under who operate under undemocratic leadership styles. In a study by Helmiatin et al. (2012) it was observed that transformational leadership has a positive effect on both organisational citizenship behaviour and quality of work life. Transformational leadership is also directly associated with staff work satisfaction, organisational commitment, quality of work life and work performance (Fang, Chang & Chen 2009). In a more recent study by Gillet et al. (2013), the existence of a positive relationship between transformational leadership and quality of work life found support, although the relationship was mediated by organisational justice. Based on the foregoing discourse, it is sound to presuppose that a positive linkage between transformational leadership and quality of work life in this study also exists. The following hypothesis is therefore proposed:

H₂: Transformational leadership has a positive influence on quality of work life amongst supply management professionals.

According to Wang and Wong (2011), there has been an exponential increase of scholarly concern in the relationship between organisational citizenship behaviour and job performance over the past two decades. Spector and Fox (2002) advocate that organisational citizenship behaviours enable employees to perform better by improving their perception and willingness to help their co-workers or support their organisation. An investigation by Ilies, Scott and Judge (2006), which focused on investigating the relationship between personality traits, positive affects and citizenship behaviours, reveals that personality traits and positive affects correlated with organisational citizenship, which in turn correlated with job performance. Avey, Wernsing and Luthans (2008) concluded that organisational citizenship behaviour is directly associated with organisational commitment, which fundamentally increases both employee and organisational performance. Cicei’s (2012) study highlights the importance of fostering organisational citizenship behaviours with a view of attaining extra-role performance by employees. Organisational citizenship behaviour is also positively correlated with job satisfaction, which in turn positively impacts on employee performance (Tsai & Wu 2010). Moreover, a study by Ozer (2011) indicates that co-worker relations mediate the positive relationship between organisational citizenship behaviours and job performance. As such, sufficient evidence is available to presume a positive interconnection between organisational citizenship behaviour and job performance in this study. The following hypothesis is therefore put forward:

H₃: Organisational citizenship behaviour has a positive influence on the job performance of supply management professionals.

Mainstream research testifies to the existence of either direct or indirect relationships between quality of work life and job performance. A study by Huang, Lawler and Lei (2007) reveals that quality of work life has a significant impact on organisational commitment and intention to quit, both of which are interrelated with job performance. Consistently, a study by Lee, Singhapakdi and Sirgy (2007) revealed that quality of work life generally has a positive influence on job performance related outcomes, such as job satisfaction, organisational commitment and camaraderie. Koonmee et al. (2010) found that the implicit form of ethics institutionalisation as well as quality of work life both have positive impacts on the three employee job-related outcomes: job satisfaction, organisational commitment and team spirit. In other studies by Normala (2010) and Ma et al. (2011) quality of work life was found to have a significant relationship with organisational commitment and employee performance. Furthermore, Probst (2003) found a positive connection between quality of work life and improved job performance through job security, whilst Cuyper and Witte (2010) opine that quality of work life mediates the relationship between job involvement and job performance. Based on the empirical findings discussed, a positive interplay between quality of work life and job performance in the present study can be envisaged. Therefore, the following hypothesis was formulated:

H₄: Quality of work life exerts a positive influence on the job performance of supply management professionals.

This study was part of a larger study conducted amongst supply management professionals employed in the South African government. This being the case, research data were collected from six government departments in Gauteng Province, South Africa. Two postgraduate students from a South African university of technology were recruited and trained to distribute and collect the questionnaires after appointments with targeted government departments were made by telephone. Out of 500 questionnaires that were initially distributed, a total of 264 usable questionnaires were retrieved for the final data analysis, representing a response rate of approximately 53%. This response rate was acceptable because it is in line with the suggestion by Sturgis, Smith and Hughes (2006) that response rates between 50% and 60% are sufficient to presume sample representativeness in cross-sectional surveys. In addition, Hatcher (1994) recommended that the minimum number of respondents should be five times larger than the number of variables, which implies that the minimum sample size in the present study should be 200, since there were four variables that were being tested. The sample size used in the current study is still regarded as adequate when other older and authoritative prescriptions, such as the Rule of 200 (Guilford 1954) and the Rule of 250 (Cattell 1978), are taken into consideration. Thus, using these recommendations as a point of reference, the actual sample size of 264 used in this study is considered to be adequate.

Nonresponse bias, which was considered to be the degree to which a researcher does not succeed in obtaining the responses from all potential respondents included in the sample (Atif, Richards & Bilgin 2012), was assessed using the Linear Extrapolation Technique suggested by Armstrong and Overton (1977). This involved analysing the first quartile of responses (66) against the last quartile of responses (66). The comparison yielded no significant differences at the 95% confidence interval (sig = 0.05), including age group, respondent work experience, highest academic qualification, occupational area, type of employment and gender. This result suggests a minimal level of nonresponse bias in the study. The profile of participants is reported in Table 1.

An analysis of Table 1 discloses that approximately 70% of respondents were aged 40 years and younger. Approximately 62% of the respondents had work experiences ranging between 2 and 9 years. A majority of the respondents (61percent) were holders of either a diploma or a first degree. Approximately 70% of the respondents were employed either in customer/client services or procurement. A further 64% of the respondents were employed on a permanent basis. The analysis also reveals that 57% of the respondents were male whilst 43% were females.

Measurement scales used in this study were operationalised primarily on the basis of previous research studies. Minor adaptations were made in order to fit the current research context and purpose. Transformational leadership was measured using five items adapted from the MLQ developed by Bass and Avolio (1994). This questionnaire has been validated by a number of previous studies (e.g. Hemsworth, Muterera & Baregheh 2013; Kirkbride 2006; Muenjohn & Armstrong 2008; Ozaralli 2003), which confirms that it is an appropriate instrument for adequately capturing the constructs of transformational leadership. Organisational citizenship was measured using six items adapted from the ‘Organisational Citizenship Behaviour Checklist’ developed by Spector et al. (2010) and validated by Gualandri (2012), as well as Uzonwanne (2014). Quality of work life was measured using five items adapted from the WRQoL scale developed by Van Laar et al. (2007). Previous validation of the WRQoL scale was conducted by several scholars (e.g. Chaiea 2014; Sirisawasd et al. 2014). Job performance was measured using six items adapted from the ‘Role-Based Performance Scale’ (Welbourne et al. 1998), which was validated by Walumbwa et al. (2008). Response options were configured on 7-point point Likert-type scales that were anchored by 1 = strongly disagree and 7 = strongly agree. The selection of a 7-point Likert-scale configuration ahead of the 5-point Likert-scale type was premised on the recommendation by Finstad (2010) that in terms of reliability, scales with more points are considered more reliable because they prevent subjects from being too neutral in their responses.

For the purposes of this study, the two-step protocol proposed by Anderson and Gerbing (1988), which entails conducting a Confirmatory Factor Analysis (CFA) prior to the testing of hypothesis, was adopted. Overall acceptable model fit are indicated by Goodness-of-Fit Index (GFI) values ³ 0.80, Adjusted Goodness-of-Fit Index (AGFI) values ³ 0.80, Root Mean Square Error of Approximation (RMSEA) values £ 0.08, Incremental Index of Fit (IFI), Tucker-Lewis Index (TLI) and Comparative Fit Index (CFI) values ³ 0.90, and Chi-square degrees of freedom ratio (CMIN/DF) value < 3 (Blunch 2008). Recommended statistics for the final overall model assessment showed acceptable fit of the measurement model to the data: CMIN/DF= 2.173, GFI = 0.809, AGFI = 0.863, IFI = 0.917, TLI = 0.934, CFI = 0. 942, and RMSEA = 0.076. These results signify that there was no signal of model misspecification in this study. The loadings of individual items on their respective factors are reported in Table 2.

To assess scale reliability, the Composite Reliabilities (CR) and Average Variance Extracted (AVE) were computed using formulas proposed by Fornell and Larcker (1981). CR was calculated using the following formulae:

(CR): CRη= (Σλyi)²/[(Σλyi)²+(Σεi)],

or stated in full: Composite Reliability=(square of the summation of the factor loadings)/[ (square of the summation of the factor loadings)+(summation of error variances) ].

AVE was calculated using the following formulae:

Vη = Σλyi²/(Σλyi² + Σεi),

or stated in full: Average Variance Extracted=(summation of the squared of factor loadings)/[ (summation of the squared of factor loadings)+(summation of error variances) ]

As further revealed in Table 2, the CR values were all above the recommended value of 0.7, suggested by Hulland (1999), thereby indicating acceptable internal consistency and reliability of the respective measures. All AVE values were above 0.5, which is regarded as acceptable by Fraering and Minor (2006). These results provided evidence for acceptable levels of research scale reliability. In addition, Cronbach alpha values (0.756 − 0.811), for all scales were also beyond the recommended 0.7, which attests to the acceptable levels of internal consistency, in accordance with the prescription by Malhotra (2010).

The individual item loadings ranged between 0.527 and 0.913, which satisfies the 0.5 minimum threshold recommended by Anderson and Gerbing (1988). By implication, this result indicated acceptable individual item convergent validity as more than 50% of each item's variance was shared with its respective construct. This evidence supported the convergent validity of all scale items. Moreover, the CR value was above the recommended threshold of 0.6, therefore further confirming the existence of convergent validity. Discriminant validity was established by testing (Fornell & Larcker 1981) whether the AVE value was greater than the highest Shared Variance (SV), as well as the chi-square difference in all two-factor (i.e. any paired latent constructs) CFA tests, which restricted the factor intercorrelations to unity (Anderson & Gerbing 1988). All pairs of the constructs and the two-factor CFA tests results revealed an adequate level of discriminant validity.

In this study, structural equation modelling (SEM) was used to estimate the interconnection amongst the factors as proposed on the theoretical framework illustrated in Figure 1. In accordance with Anderson and Gerbing (1988), a two-step model structure approach was utilised, with the testing of the structural model being preceded by an assessment of measurement models. Acceptable model fit is indicated through satisfaction of the following statistical parameters: CMIN/DF value < 3, both GFI and AGFI values ³ 0.80, RMSEA values £ 0.080, and IFI, TLI and CFI values ³ 0.90 (Blunch 2008). The results of the present study reveal the following: CMIN/DF = 2.165, GFI = 0.891, AGFI = 0.826, IFI = 0.971, TLI = 0.946, CFI = 0.970, and RMSEA = 0.042, as shown in Table 3. This constellation of results denote that the model converged well (acceptable model fit) since it satisfied the recommended numerical thresholds (Bentler 1990; Browne & Cudeck 1993; Marsh, Balla & Hau 1996). For testing the hypotheses, the Maximum Likelihood Estimation (MLE) method was employed. This was prompted by the fact that MLE has desirable asymptotic properties and is scale-free, yielding minimum variance and unbiasedness (Joreskog & Sorbom 1993). The results of the SEM are reported in Table 3.

The results in Table 3 provide empirical support for the quartet of hypotheses proposed in the present study. The first hypothesis (H₁) proposed that the influence of transformational leadership on organisational citizenship behaviour is positive. Consistent with this hypothesis, the results of the study reveal that the influence of transformational leadership on organisational citizenship behaviour is positive, moderate and significant (r = 0.349***). This demonstrates that higher levels of transformational leadership are associated with moderately higher levels of organisational citizenship behaviours amongst supply management professionals in government. The second suggested hypothesis (H₂) proposed that the influence of transformational leadership on quality of work life is positive. In line with this proposition, a strong positive and significant relationship (r = 0.834***) was observed between the two factors. This finding implies that higher levels of transformational leadership are likely to trigger stronger positive perceptions of quality of work life amongst supply management professionals in the government.

The third hypothesis (H₃) suggested that organisational citizenship behaviour positively influences employee productivity. The results of the study endorsed this hypothesis by showing a positive, moderate and significant (r = 0.482***) relationship between the two factors. This finding suggests that greater organisational citizenship behaviours by supply management practitioners are expected to stimulate moderately higher employee performance levels amongst them. The final hypothesis (H₄) proposed that quality of work life exerts a positive influence on job performance. In support of H₄, the study revealed a strong positive and significant relationship (r = 0.755***) between these two factors. Thus, higher levels of quality of work life are positively associated with higher levels of job performance amongst supply management professionals.