Understanding the impact of resilience and perceived organisational support (POS) on employee engagement in a competitive sales environment.

Although employee engagement has been identified as a significant driver of revenue and financial performance in modern organisations (Bakker & Demerouti, 2014; Verbeke, Dietz & Verwaal, 2011), limited research has focused on the unique context of employee engagement as a construct in professional sales (Medhurst & Albrecht, 2016). Existing sales literature on employee engagement focuses mainly on drivers such as role conflict, role ambiguity, adaptive selling behaviour (Miao & Evans, 2013), trust, psychological contract and interactional justice (Agarwal, 2014).

Various studies indicate that there is a relationship between resilience and engagement (Bakker, Albrecht, & Leiter, 2011; Bande et al., 2015; King et al., 2015; Mache et al., 2014; Shin et al., 2012). Resilience appears to be closely linked to vigour, one of the three dimensions of employee engagement (Schaufeli, Salanova, González-Romá, & Bakker 2002). Medhurst and Albrecht (2011) conclude similarly that resilience positively influences sales performance through vigour and also the investment of high levels of energy when faced with challenging situations. Bande et al. (2015) argue that resilience can lead to a subjective assessment of well-being that includes engagement. In the South African context, Simons and Buitendach (2013) observe a practical and statistically significant relationship between total employment engagement scores and resilience among call centre employees. The call centre environment is comparable to a competitive sales environment with respect to the negative effect of emotional exhaustion on employee well-being (Johnson et al., 2005). There appears to be no published studies in South Africa investigating the relationship between employee engagement and resilience in a sales environment.

Perceived organisational support as a construct is positively associated with employee engagement (Kurtessis et al., 2015; Mathumbu & Dodd, 2013). Research by Rothmann and Rothmann (2010) indicates that organisational support is a strong predictor of employee engagement in various South African industries. In the context of sales, this association is underpinned by the stressful nature of sales positions accompanied by emotional exhaustion, which ultimately has important implications for levels of engagement and the need for organisational support (Kurtessis et al., 2015). Salespeople with a subdued perception may also experience diminished job involvement and identification with the organisation, which can lead to less dedication and absorption (DeConinck, DeConinck, & Lockwood, 2015).

The purpose of this study was to contribute to the theoretical understanding of resilience and POS as potential drivers of employee engagement in a competitive sales environment. Resilience as a personal resource is pliable and open to development (Krush et al., 2013). An improved understanding of resilience can make a contribution to human resource management practices in terms of training and employee development. Studying resilience as a potential driver of employee engagement will make a contribution towards the academic field of positive psychology and positive organisational behaviour. Exploring salespeople’s perceptions of organisational support offers valuable insights into the employee–organisation relationship in terms of concepts such as feeling valued as an employee (Gupta & Sharma, 2016), levels of supervisor support (DeConinck et al. 2015), fairness (Morelli & Braganza, 2012) and human resource management practices related to training, performance appraisal and compensation systems (Zampetakis, 2014). These insights can be utilised to create a more supportive environment with improved employee engagement and the ability to cope in a stressful profession (DeConinck et al. 2015). Studying the relationship between employee engagement and POS will make a contribution to the academic fields of positive organisational behaviour and organisational support theory.

In the next section, a synthesis and critical evaluation of the literature dealing with the constructs will be provided, followed by an explanation of the research design, reporting of results and a detailed discussion of the results.

Ethical clearance was obtained from the University of Pretoria’s Gordon Institute of Business Science Research Ethics Committee. All respondents were assured of confidentiality and anonymity, with participation being voluntary.

The emergence of engagement research in the 21st century is related to the developing importance of human capital, involvement of employees in organisations (Bailey, Madden, Alfes, & Fletcher, 2015; Schaufeli, 2013) and the need for businesses to maximise the inputs of employees (Rothmann & Rothmann, 2010). The first theoretical foundation of engagement stems from the work of Kahn (1990, p. 694), who views personal role engagement as the individual’s emotional, cognitive and physical expression of their authentic self at work. To be fully engaged, individuals must display their full selves within their work role (Bailey et al., 2015). Building on the work of Kahn (1990, p. 694), Rich, Levine and Crawford (2010) state that individuals who are engaged invest their head, heart and hands in their performance. The second theoretical view of engagement is an activated positive state of mind directed towards work tasks. This view is founded on the concept of engagement and burnout being opposites of each other. This theoretical view defines engagement as ‘a positive, fulfilling, work-related state of mind that is characterised by vigour, dedication and absorption towards work activities’ (Schaufeli et al., 2002, p. 71). Vigour assists individuals to foster a more proactive work style (Coetzee, Schreuder, & Tladinyane, 2014) and is characterised by high levels of energy and mental resilience. In addition, vigour is associated with persistent, conscientious efforts to devote oneself to work and when facing difficulties (Coetzee et al., 2014). Dedication involves being strongly involved in one’s work with a sense of pride and enthusiasm. Absorption involves being completely concentrated on and happily engrossed in one’s work, whereby time passes quickly and it is difficult to detach from work (Bailey et al., 2015; Schaufeli, 2013).

Employee engagement has been identified as a significant driver of salesperson performance in modern sales organisations (Albrecht & Medhurst, 2011; Rapp, Bachrach, Panagopoulos, & Ogilvie, 2014). According to Albrecht et al. (2015), the direct influence of employee engagement on sales performance assists organisations to achieve a competitive advantage through improved customer satisfaction, increased selling intentions and a more favourable attitude towards products and resources. The impact of employee engagement in sales is also related to sales force turnover. Sales positions are subject to high turnover rates and are often challenging positions to fill (Bande et al., 2015). A high level of employee engagement is associated with 37% lower absenteeism (Harter, Schmidt, Agrawal, & Plowman, 2013) and has the ability to reduce the negative impact of sales force turnover on the business results of sales organisations (Bande et al., 2015).

A study by Medhurst and Albrecht (2016) interprets engagement among salespeople as an aroused psychological state of focus and energy, aimed at addressing the situational and task-related opportunities and demands encountered in sales positions. According to these authors, salesperson engagement shares similar characteristics with the existing conceptualisation of engagement in terms of vigour, dedication and absorption (Schaufeli et al., 2002).

Bardoel, Pettit, De Cieri and McMillan (2014) propose that one of the theoretical approaches to resilience applied in the workplace is based on positive psychology and positive organisational behaviour. In positive psychology, resilience is described as a response where an individual adapts positively to exposure to a subjectively significant threat, risk or difficulty, without losing the ability to function normally (Bardoel et al., 2014; Bonanno, 2004). Recent applications of resilience in occupational literature focus on occupations associated with a high risk for experiencing acute stress and trauma, such as police officers and fire fighters (Freedman, 2004; Peres et al., 2011; Vanhove, Herian, Peres, Harms & Lester, 2015). However, Vanhove et al. (2015) propose that resilience may also be of significance in an employment context where stress can accumulate over time because of influences such as work overload, work relationships, lack of resources and support, emotional and physical exhaustion and work–life conflict. Johnson et al. (2005) identify teachers, ambulance workers, customer and social service workers, call centre staff and prison officers as examples of jobs where the above-mentioned influences and accumulated stress can have a negative effect on employee well-being and organisational functioning (Vanhove et al., 2015). Although no reference is made to sales employees, the nature of sales positions suggests that the sales environment can also be viewed as an employment context where resilience is of significance. Krush et al. (2013) support this view with the observation that resilience enables positive responses towards adversity, while simultaneously inhibiting negative responses.

Resilience in the workplace is also defined as the ‘positive psychological capacity to “bounce back” from adversity, uncertainty, conflict, failure, or even positive change, progress and increased responsibility’ (Kotzé & Nel, 2013; Luthans, 2002). According to the literature, a key component of resilience is whether an individual demonstrates simultaneous growth and positive change following a stressful event. Although some definitions refer to positive change, most simply require successful adaptation to adversity (Bande et al., 2015). In competitive sales, adaptation unaccompanied by growth or positive change will not enable a sales force to successfully meet sales targets or deal with rejection and competitors.

Employee engagement is a positive organisational outcome associated with resilience (King et al., 2015; Mache et al., 2014; Shin et al., 2012). Bande et al. (2015) report that resilience can lead to a subjective assessment of well-being that includes engagement. However, this prediction includes all the dimensions of self-efficacy, optimism, hope and resilience. This raises questions as to whether this position can be proposed by taking into account only the dimension of resilience. Luthans, Vogelgesang and Lester (2006) provide a possible explanation by arguing that resilient people may take a more pragmatic approach to dealing with stress than those with a high level of optimism or hope. Medhurst and Albrecht (2011) propose that resilience positively influences sales performance through vigour and the investment of high levels of energy when faced with challenging situations. Vigour as a dimension of employee engagement could therefore be positioned as a driver of employee engagement among sales employees (Schaufeli, 2013). This research study formulated the following hypothesis:

Organisational support theory has attracted considerable attention because of its potential ability to explain the employee–employer relationship. Various studies indicate that POS is related to behavioural outcomes for both employees and organisations (Caesens, Marique, Hanin, & Stinglhamber, 2016). Beneficial organisational behavioural outcomes include organisational commitment (Baran, Shanock, & Miller, 2012; Connelly, Gallagher, & Gilley, 2007), organisational citizenship behaviour (Baran et al., 2012; Coyle-Shapiro & Morrow, 2006), performance (Webster & Adams, 2010) and job satisfaction (Kurtessis et al., 2015). Beneficial employee behavioural outcomes related to POS include increased trust in the organisation, job satisfaction, positive psychological well-being and increased in-role performance. Unfavourable behavioural outcomes include job stress, burnout and withdrawal behaviour. Under the umbrella of organisational support theory, fairness, human resource practices and supervisor support are important antecedents of POS (Baran et al., 2012; Kurtessis et al., 2015).

Perceived organisational support as a construct is positively associated with employee engagement (Kurtessis et al., 2015; Mathumbu & Dodd, 2013). According to organisational support theory, employees also develop beliefs regarding the extent to which the organisation cares about their well-being and appreciates their contribution to achieving business goals. Feeling appreciated is a key driver of positive organisational outcomes such as engagement (Gupta & Sharma, 2016).

The perception of organisational support can be viewed from an organisational or employee perspective. Tavares, van Knippenberg and van Dick (2015) argue that the perception of organisational support can be viewed by the organisation as a social currency offered with the aim of increasing the quality of the employer–employee relationship. This relationship is discretionary in nature, which means that employees can determine the extent to which they engage in extra-role efforts to benefit the organisation and its employees. From the viewpoint of the employee, this relationship is reciprocal in nature as the employee feels obligated to support their employer with the expectation that increased performance will be noticed and rewarded. According to social exchange theory, employees with high POS will engage in greater job-related efforts (Kurtessis et al., 2015).

In a competitive and results-orientated milieu, sales representatives are more likely to succeed when they feel supported rather than directed (Adamson, Dixon, & Tolman, 2013). Sales representatives often work alone in different geographical locations and are detached from the organisation (Morelli & Braganza, 2012). Working outside the boundaries of the organisation in a complex sales environment may accentuate the need for organisational support. DeConinck et al. (2015) are of the opinion that with a subdued perception of organisational support, salespeople may also experience diminished job involvement and identification with the organisation, which can ultimately lead to less dedication and absorption. Dedication and absorption are critical dimensions of employee engagement based on the theoretical foundation of engagement as a positive, work-related state of mind (Schaufeli & Bakker, 2004). Based on the aforementioned, the following hypothesis is presented:

Given these research findings, the researchers proposed a positive relationship between resilience and POS and employee engagement, respectively.

Most existing empirical research on employee engagement makes use of either the Utrecht Work Engagement Scale (UWES) (Gupta & Sharma, 2016; Schaufeli & Bakker, 2003) or the Job Engagement Scale (JES) (Gupta & Sharma, 2016; Rich & Levine, 2010). The JES is grounded in theories of self-expression at work and based on Kahn’s (1990, p. 694) definition of engagement which includes physical, cognitive and affective dimensions (Gupta & Sharma, 2016). The UWES is grounded in burnout literature (Gupta & Sharma, 2016; Maslach & Leiter, 1997) and defines engagement as a positive, work-related state of mind with a strong sense of vigour towards, dedication to and absorption in work activities. According to Saks and Gruman (2014), the dimensions of the UWES are inconsistent with Kahn’s (1990) original assertion that engagement is an indication of bringing one’s true self to the performance of one’s role (Saks & Gruman, 2014). The authors argue that bringing one’s true self to the performance of one’s role involves a much deeper and more authentic level of engagement than just devoting energy and dedication to the performance of work activities. However, a systematic review involving 214 studies found the UWES to be the most widely adopted measure of engagement (Schaufeli & Bakker, 2003).

The challenges of developing an operational definition for resilience have led to different approaches to measuring the construct (Windle et al., 2011). There is currently no gold standard in terms of measuring resilience. Most existing measures have concentrated on examining resources or protective factors that might facilitate a resilient outcome. In contrast, the Brief Resilience Scale (BRS) is designed to measure resilience as an outcome measure defined as the ability to recover from stress. The BRS is the only measure that assesses resilience based on its original and most basic meaning as relating to the ability to bounce back from stress (Smith, Tooley, Christopher, & Kay, 2010). In the South African context, the BRS was used in a small mixed method study by Edwards, Edwards and Highley (2015), with no references made to internal consistency scales using Cronbach’s alpha. Additional measurement scales used in the local context include the Resilience Scale (Koen, van Eeden, & Wissing, 2011) and the Adult Resilience Indicator (Kotzé & Nel, 2013). A criticism from Jowkar, Friborg and Hjemdal (2010) is that most resilience measurement scales ignore the family and social aspects of resilience. An additional concern is that most of the resilience measurement instruments have been developed and utilised in the Western world, with concerns regarding their validity in non-western population (Abiola & Udofia, 2011; Dageid & Gronlie, 2015). Recent empirical research has demonstrated the BRS’s reliability and validity in emerging countries such as Malaysia (Amat et al, 2014) and China (Lai & Yue, 2014).

The original 36-item POS Scale developed by Eisenberger (1986) has gained considerable interest because of the benefit of understanding the employee–employer relationship. The scale measures employees’ perceptions regarding the extent to which employers value the contribution of employees and care about their well-being. Shorter versions of the scale are now available. Correlations among factor scores of POS Scale scores suggest that both the 8-item and 16-item version are just as effective as the original 36-item version, but are more efficient (Worley, Fuqua, & Hellman, 2009). A 6-item version of the POS Scale was used in a multi-industry, multi-generational and multicultural study with a high level of reliability (Smit, Stanz, & Bussin, 2015). In the South African context, the 8-item POS Scale has been used in a study of nurses and had acceptable levels of reliability and validity (Mathumbu & Dodd, 2013).

Quantitative research methods were used to test the predetermined hypotheses regarding the relationship between the variables of resilience, organisational support and employee engagement within a cross-sectional time frame. Factor analysis, multiple regression and analysis of variance tests were applied to determine whether a predictive relationship of significance exists among the variables. Factor validity and reliability were determined for all measurement scales.

The study made use of a convenience sampling approach. A survey was conducted on sales representatives from a Johannesburg Stock Exchange listed pharmaceutical company (N = 220) in South Africa. A response rate of 56.8% yielded a sample of 125 participants. The Monte Carlo simulation was used to determine the minimum sample size needed for multiple regression or prediction (Knofczynski & Mundfrom, 2007). From this method, it was inferred that for the purpose of this study, the minimum sample size needed for two predictor variables was 90 cases. The demographic characteristics of the sample indicate that almost 70% of the sales force were younger than or equal to 40 years of age, while 25% of the sales force had been employed by the organisation for 1 year or less. Female participants accounted for 69% of the sample, the majority of the respondents were white people (45%), followed by Indians (24%) and black people (17%), with the minority from the mixed race population (14%).

A letter explaining the objective of the study and requesting participation in the survey was emailed to sales employees within the organisation. The measuring instruments were pretested with selected sales representatives from a division not participating in the study prior to being distributed to the entire sample, with no changes made after the testing phase.

Surveys were distributed to sales employees in paper-based form and via a link to an electronic format in Typeform™. Convenience sampling was used to select participants nationally from a sample frame consisting of 220 employees from which 125 responses were obtained for final analysis. The first section of the survey collected demographic information, whereas the second section consisted of a questionnaire combining the three measuring instruments.

Coded data were analysed with the IBM SPSS (Version 22) statistical software tool. Descriptive statistics were calculated to describe the mean, median, standard deviation, skewness and kurtosis of the various constructs. Factor analysis was used to determine the construct validity of all measurement scales (Zikmund, Babin, Carr, & Griffin, 2010). First, the Kaiser–Meyer–Olkin (KMO) index and Bartlett’s test of sphericity were inspected to determine whether factor analysis was a feasible option. Principle component analysis and a varimax rotation strategy were used to explore the factor structure of the various scales used within this study (Yong & Pearce, 2013). Cronbach’s alpha was employed to determine reliability and internal consistency with a coefficient alpha between 0.70 and 0.80 considered to be reflective of good reliability (Zikmund et al., 2010). The corrected item-total correlation was used as an additional measure of reliability by correlating the individual item score to the sum of all scores. A correlation coefficient greater than 0.3 indicated adequate item-total correlation (Field, 2013).

Multiple regression determined whether resilience and POS hold predictive value for employee engagement. All required assumptions for multiple regression analyses were met before the analyses were conducted. The predictive power and relative contribution of the independent variables were tested with hierarchical multiple regression by entering the variables in a predetermined order. Demographic variables were controlled for and forced into the first step to determine whether the independent variables were still able to explain the remaining variance of the dependent variable.

All the scales were subjected to a factor and reliability analysis. Single factor structures were obtained for all scales, whereas acceptable reliabilities were obtained for all subscales and overall scales. The KMO index measure for sampling adequacy for employee engagement, resilience and POS exceeded the minimum accepted level of 0.6 for a good factor analysis (Pallant, 2005).

For the UWES, high factor loadings were observed in all nine items, and therefore, all items were included in the scale. A one-factor solution also seemed to best fit the data. All correlation coefficients were greater than 0.30, indicating adequate item-total correlation. With an alpha of 0.93, the UWES Scale was considered to be reliable with a high degree of consistency between multiple measurements of the scale. This level of reliability is within close range of a South African nurses study reporting an alpha coefficient of 0.79 (Mathumbu & Dodd, 2013) and a comparable South African study of call centre employees that reported an alpha of 0.95 (Simons & Buitendach, 2013).

For the BRS, the loadings of each of the items on the one component with factor loadings ranged from 0.43 to 0.81 exceeding the acceptable factor loading of 0.40. Therefore, all six scale items were included in the scale. All correlation coefficients were greater than 0.30, indicating adequate item-total correlation. The initial Cronbach’s alpha for the BRS was 0.73. Item R5 (‘I usually come through difficulty times with little trouble’) lowered the scale’s reliability and validity with a low correlation coefficient (< 0.30) and factor loading (0.434), and was subsequently removed from further analysis. There are no comparable South African studies reporting mean resilience scores based on the BRS.

For the POS Scale, high factor loadings were observed; therefore, all eight items were included in the scale. A one-factor solution also seemed to best fit the data. All correlation coefficients were greater than 0.30, indicating adequate item-total correlation. With a Cronbach’s alpha of 0.90, the POS Scale was considered to be reliable with a high degree of consistency between multiple measurements of the scale. These results are consistent with previous South African research indicating an acceptable level of reliability and validity for this scale with Cronbach’s alpha values ranging from 0.71 (Mathumbu & Dodd, 2013) to 0.89 (Smit et al., 2015).

The descriptive statistics for all constructs are presented in Table 1. For employee engagement, a mean score of 5.77 indicated a high level of employee engagement. On the 7-point Likert scale, this indicated a high level of engagement experienced often (a few times a week). A standard deviation dispersion of 0.87 showed that the individual responses did not deviate much from the main score; the responses were therefore not polarised. Resilience had a mean score of 3.70 on a 5-point Likert scale presenting a high perceived level of resilience close to ‘agree’. A standard deviation dispersion of 0.63 indicated that the individual responses deviated slightly from the main score. For POS, a mean score of 4.87 indicated a response close to ‘slightly agree’. A standard deviation dispersion of 1.31 indicated that there was a large amount of variation in the individual responses.

In conclusion, descriptive statistics indicated a high perceived level of resilience and employee engagement, and a ‘slightly agree’ perception of POS.

In Table 2, demographic variables (gender, age, length of service) predicted approximately 4% of the variance in employee engagement. Resilience predicted just more than 5% of the variance in employee engagement. Neither the demographic variables (F = 1.63, p > 0.05) nor resilience (F = 1.50, p > 0.05) was statistically significant predictor of employee engagement. Perceived organisational support predicted approximately 23% of the variance in employee engagement (F = 27.20, p < 0.05). In Table 3, the regression coefficient of POS was statistically significant (β = 0.44, p < 0.05).

Statistical power was increased with a large sample size (N = 125) than the required sample size of 90. The sample size was sufficient with a medium effect size (β ≥ 0.2) and power of 80%. The increase in power decreased the probability of a Type II error (false negative). With a critical α ≤ 0.05, there was only a 5% chance of a Type I error, or incorrectly rejecting the null hypothesis.

In conclusion, there is a statistically significant relationship between POS and employee engagement, but no statistically significant relationship between resilience and employee engagement.

Although resilience explained almost 5% of the variance in employee engagement, no statistically significant relationship was found with employee engagement. The results of this analysis contradict prior academic research indicating that employee engagement is a positive organisational outcome associated with resilience (King et al., 2015; Mache et al., 2014; Shin et al., 2012). In the South African context, Simons and Buitendach (2013) observed a practical and statistically significant relationship between total employee engagement scores and resilience among call centre employees. The call centre environment is similar to the sales environment in terms of emotional exhaustion and burnout.

Findings from Bakker et al. (2011) also suggest that resilience is an important predictor of work engagement. However, this prediction included additional dimensions of self-efficacy, optimism, hope and resilience. It is likely that the predictive power of resilience for engagement could be stronger if it was defined and quantified in conjunction with all the dimensions of psychological capital.

It is plausible that the relationship between resilience and employee engagement is more complex than what can be explained by a simple prediction model. Resilience may be a moderating variable between an independent variable and the outcome variable employee engagement. For example, the possibility of the moderating effects of resilience as a personal resource in the relationship between job resources and work engagement has been recognised in existing empirical research (Mäkikangas, Feldt, Kinnunen, & Mauno, 2013; Xanthopoulou, Bakker, Demerouti & Schaufeli, 2007). It is also possible that resilience as a personal resource may act as a mediating variable between an independent variable and the outcome variable of employee engagement (Bakker & Demerouti, 2008; Chen, Westman, & Hobfoll, 2015). Using the job-resources or demands model, Cooke, Cooper, Bartram, Wang and Mei (2016) found that employee resilience mediates the relationship between high-performance work systems and employee engagement within the context of Chinese banking.

In a sales context, resilience may act as a moderating variable between employee engagement and a different organisational outcome. Medhurst and Albrecht (2011) found that resilience influences a salesperson’s performance through engagement and the investment of high levels of energy when faced with challenging situations. In addition, although resilience did not predict employee engagement for the population under study, this may be different in other populations.

Perceived organisational support not only predicted approximately 23% of the variance in employee engagement (F = 27.20, p < 0.05) but also proved the regression coefficient to be statistically significant (β = 0.44, p < 0.05). This outcome is consistent with the results of earlier studies based on similar theory (Saks, 2006). According to social exchange theory, employees with high POS will engage in greater job-related efforts (Kurtessis et al., 2015). In more recent international and South African studies, positive organisational support was positively associated with a high level of engagement (Mathumbu & Dodd, 2013; Trinchero, Brunetto & Borgonovi, 2013).

It is also plausible that the relationship between the two constructs can be defined beyond prediction. Shantz, Alfes and Latham (2014) suggest that POS moderates the relationship between work engagement and behavioural outcomes such as turnover intentions and deviant behaviours. In a study of nurses, affective commitment mediated the relationship between POS and work engagement (Gupta, Agarwal, & Khatri, 2016).

While formulating the hypothesis, it was not assumed that employee engagement as an outcome would be entirely explained by the predictive power of the independent variable. Perceived organisational support predicted nearly a quarter (23%) of the outcome, which means that the majority, or almost three quarters of the predictive power of engagement, is still explained by different independent variables. It is also important to take into account that different predictors or drivers of employee engagement are interrelated and do not function separately. Examples of predictors of engagement reported in sales literature include role conflict, role ambiguity, activity and capability control, and adaptive selling behaviour (Miao & Evans, 2013), trust, psychological contract and interactional justice (Agarwal, 2014).

Employee engagement is a continuous process in need of integration across all facets of the employer–employee relationship and employee life cycle (Albrecht et al., 2015). Direct supervisors and sales managers should treat employee engagement as a priority and actively take responsibility for driving the process. Although there are many stakeholders involved in creating a culture of engagement, line managers and direct supervisors ought to be the ultimate owners of engagement (Aon Hewitt, 2015).

Resilience is pliable and open to development (Krush et al., 2013) and resilience should thus be viewed as a set of skills that can be developed as part of strategic human resource management. Resilience-building programmes have proven to be effective in the workplace with a positive effect on health, well-being and performance (Robertson, Cooper, Sarkar, Curran, & Cooper, 2015; Vanhove et al., 2015). This can be of particular importance in the stressful nature of a competitive sales environment.

With supervisor support being identified as an antecedent of POS (Morelli & Braganza, 2012), sales managers should maintain a reflective outlook on the quality and level of their supervision and support towards sales staff. A high level of supervisor support has the ability to reduce burnout and ultimately staff turnover (DeConinck et al. 2015). With fairness being one of the precursors of POS (Morelli & Braganza, 2012), managers responsible for the allocation of sales territories, structuring of sales targets and performance packages play an important role in making decisions that are perceived by sales staff as fair. Human resource practices is the third antecedent of POS (Morelli & Braganza, 2012). A study by Zampetakis (2014) indicated that human resource management practices such as training, performance appraisal and compensation systems play an important role in creating supportive environments for sales employees. Individually or collectively, the above-mentioned factors may assist in creating an engaged sales force.

The limitations of this study are mainly linked to shortcomings related to impact and generalisation. Firstly, the sample in this study was derived from a single pharmaceutical organisation with sales roles predominantly focused on doctor detailing and business-to-business selling. The results may consequently not be generalisable to sales employees working in different industries. Secondly, with pharmaceutical sales representatives working in a knowledge-intensive and science-based economy, the results may not be generalisable to sales employees working in other environments (e.g. telesales or retail sales). Thirdly, a cross-sectional research design provided only a snapshot of the specific time frame used. A different time frame containing dynamics such as restructuring or organisational change may have produced different results.

To validate and generalise the results of this study, future research should replicate the predictive relationship between POS and employee engagement in different sales organisations across different industries. Employee engagement has not been studied extensively in the South African pharmaceutical context. More research focusing on this sales environment can provide valuable insight on the constructs in question on an industry level. This research did not focus on sales performance as an outcome of employee engagement. An interesting area of research could be to explore how POS mediates the relationship between employee engagement and sales performance. The competitive nature of the pharmaceutical sales environment is bound to intensify with an increased focus on industry profit margins, and gaining and protecting market share. Demand for pharmaceuticals in South Africa is likely to rise in the future, reflecting demographic and epidemiological trends as well as government commitments to improved healthcare. Yet, the South African pharmaceutical industry will become more constrained (IMS Health, 2014) with squeezed industry margins due to increased drug pricing, intensified regulation of drug prices and the impact of inflation and exchange rate fluctuations on manufacturers. As a result, management will increasingly scrutinise sales performance as a driver of revenue.

Research measuring resilience as a single construct in a different context using the BRS would make a valuable contribution towards demonstrating the reliability and validity of the scale in the South African context as an emerging country. The positive association between employee engagement efforts and staff retention is well documented (Bande et al. 2015). Taking into consideration concerns related to high sales force turnover (Johnson, Friend, & Rutherford, 2016), future research can explore the moderating effect of POS and employee engagement on intention of sales staff to leave the organisation.