The Effects of LMX and Feeling Trusted on Job Performance and Workplace Ostracism Among Salespeople

Abstract

Purpose - The purpose of this study is to examine the impact of leader-member exchange (LMX) on employees' behaviors as well as to test the mediating roles of feeling trusted in the relationships. Research design, data, and methodology - Survey data were collected and analyzed from 285 sales people who work for various organizations in eastern China. The corresponding research models were verified using SPSS 22 and Amos 22. Results - The empirical results show that LMX plays a positive role in job performance and is positively related with subordinate's feelings of trust. However, the study was shown to have negative correlations between LMX and workplace ostracism. Feeling trusted by superiors mediated the relationship between LMX and job performance, resulting in employees being more motivated to do better, creating a positive synergetic effect. However, this relationship did not show to have significant mediating effects on the relationship between LMX and workplace ostracism. Conclusions - This study further enriches the relevant theory of LMX and has significance for management practices such as distribution sectors, guiding leaders to create a suitable team atmosphere. Leaders need to pay attention to build a high quality LMX with employees based on the empirical results of this study.

1. Introduction

Leader-Member Exchange (LMX) has been gaining increasing attention in the field of leadership theory in recent years. The current theory postulates that the relationship between leaders and members can be cultivated in a set period of time through a series of methods such as wait-and-see, temptation, and interaction as well as negotiation (Graen & Scandura, 1987). This in turn results in a vastly different dynamic and relationship between the leaders and subordinates. With this, it can be assumed that a leader with 10 subordinates will need to deal with 10 different LMX (Bauer & Green, 1996).

Empirical studies have indicated that the closer the relationship between leaders and subordinates, the better the job performance evaluation results made by the subordinates (Graen, Novak, & Sommerkamp, 1982), and the more opportunities for promotion that can be achieved by them (Ren & Wang, 2005). Thus, the quality of LMX is essential for the members to ensure and achieve their career development goals.

Considering this aspect, people can perceive that the LMX may not only stimulate members’ career success (Breland, Treadway, Duke, & Adams, 2007), but also lessen the possibility of turnovers (Graen, Liden, & Hoel, 1982). Alongside that, LMX can foster more harmonious and compatible atmospheres between the leaders and subordinates such as mutual reliance and respect, as well as appreciation (Dansereau, Graen, & Haga, 1975). In the long run, this will lead to more beneficial results for them as well as their enterprises (Graen, 1976; Graen & Cashman, 1975).

The different relationships between the leaders and the subordinates can convey trust to subordinates in LMX. Feelings of trust is a big driving force for behavior for subordinates, and it is also consistent with the philosophical statement that "trusting others is a virtue." In light of this truth, research on “trust” and “feeling trusted” has explored a number of positive effects on its impact (Brown, Goetzmann, Liang, & Schwarz, 2009; Lau, Lam, & Wen, 2014). For example, when a subordinate recognizes the trust of the leader, (s)he may meet the leader's expectations, (s)he is willing to make extra efforts to complete the task, and protect the interests of the leader (Sun, Long, & Li, 2018). The cases about “more trust may lead to more efforts” are pretty common in the enterprises. For instance, Wang and Zhang (2016) has stated that under the influence of the traditional culture of China, the subordinates think that they need to do more work to repay their leaders when they are trusted by them.

However, do LMX and trust only have positive impacts? Previous studies argued that low LMX can easily lead to some negative effects, such as a decline in employee professional satisfaction (Erdogan & Enders, 2007), resulting in reduced employee career successes (Wayne, Liden, Kraimer, & Graf, 1999). Unfriendly relationships between leaders and members will lead to a decline in job performance (Jaramillo, 2011), rarely mentioned in the negative effects of high LMX. But we expect that LMX and feeling trusted are not only positive but also negative as well. Some recent studies have pointed out that LMX has a negative side (Scandura, 1999), and high level LMX may be a double-edged sword, which not only gives members the motivation to work, but also creates a sense of work stress. Thus, in this paper we further explore the negative effects of when leaders are closely related to members and the members feel a high level of trust.

2. Literature Review and Hypothesis

2.1. LMX and Job Performance

Graen and Dansereau first proposed the LMX theory in 1972. Since then, this theory has attracted many researchers to conduct theoretical and empirical studies related to this as well. Before Graen and Dansereau proposed the LMX, the theory of leadership was almost always based on the assumption that the leader treated all his or her subordinates in the same way. However, Green, Anderson, and Liden (1996) argued that the key point of leadership behavior research should be on the personal relationship between the leaders and the subordinates. In particular, the leaders and different subordinates will have different exchange relationships, depending on the dynamics of the relationship. They pointed out that due to limited time and energy, the leaders must adopt different management styles and methods to treat their subordinates in their work, and then establish different types of relationships with different subordinates. If the leaders and some of the subordinates have established some special relationships, these subordinates will receive more trust and attention, and they may enjoy certain privileges, such as more autonomy, flexibility and more opportunities for promotion and rewards. These subordinates are called as the “in-group members“; in exchange for this relationship, the leader will be supported, respected and appreciated by the subordinates. Other subordinates will be regarded as the “out-group members”, who will take up less time from leaders and obtain fewer opportunities to receive rewards, and their leader-member relationship will be limited to merely a formal working relationship (Graen & Uhl-Bien, 1995).

The LMX is related with job performance（Bauer & Green, 1996; Kraimer, Wayne, & Jaworski, 2001; Liden & Graen, 1980). The members will be able to repay their leaders if they receive assistance, reliance, and some implicit and tangible benefits from their leaders. In this situation, the members in the high level of the LMX are willing to work much harder to increase their job performance in order to be rewarded the appreciation of their leaders. Scandura and Graen (1984), Howell and Hall-merenda (1999) indicated that there was a significant positive correlation between the LMX and job performance. After that Gerstner and Day (1997) verified this hypothesis through meta-analysis. The results demonstrated that the LMX had a significant correlation with job performance and goal achievement. It can be surmised that a close relationship exists for the LMX, members’ job performance and goal achievement.

When employees interact well with supervisors and perceive more supervisor support, they are more willing to invest more time to achieve their work goals (Wayne, Shore, & Liden, 1997). High LMX promote employees' cooperation with peers, thus increase employees' work performance (Rhoades & Eisenberger, 2002); On the other hands, low LMX members hide their efforts or unwillingness to cooperate, which has a negative impact on performance (Liden, Erdogan, Wayne, & Sparrowe, 2006). Wang and Liu (2005) examined the impact of the LMX on job performance and organizational commitment in the China’s cultural environment. The results portrayed that the LMX had a significant positive impact on job performance and organizational commitment. Zhou and Ye (2006) also found that the LMX could significantly affect the job performance and play a role in organizing member behavior.

In sum, many studies have pointed to the possibility of high levels of LMX and improvement on members’ job performance, so in this research we speculate that the LMX will have a significant and positive impact on job performance.

H1: LMX is positively related to job performance.

2.2. LMX and Workplace Ostracism

Workplace ostracism refers to the exclusion, neglect, and rejection from other members perceived in the workplace (Ferris, Brown, Berry, & Lian, 2008). The typical exclusions include the neglect and ignorance, no contact with others and our own needs and feelings being treated unfairly. Workplace ostracism is the excluded members’ perception, feeling and experience. Whether the exclusion comes from individuals or groups, it will make members suffer psychologically. Compared with other negative behaviors in the organization (such as violence and theft), the workplace ostracism (such as indifference and avoidance) is relatively indirect, concealed and vague, so it is less involved in the bodily attacks and verbal offenses.

A survey of more than 5,000 US employees showed that 13% of employees had experienced different degrees of ostracism in their work in the past six months (Hitlan, Clifton, & DeSoto, 2006). Another survey of 262 employees indicated that 66% of them had been neglected with silence in the past five years; 29% of employees had experienced that the others in their presence will leave once they entered the room; 18% of employees expressed that they were gradually being isolated (Fox & Stallworth, 2005). Kim, O’Neill, and Cho (2010) introduced the mediator - workplace envy - to explore the impact of LMX on organizational citizenship behavior. The results showed that the difference in the exchange and the LMX between one leader and many different subordinates leads to workplace envy from members with a low LMX to those with a higher LMX in the workplace. This will lead to members with higher workplace envy to reduce their standards to support others, such as organizational citizenship behavior. Worse, they may injure other members by improper methods. Thus, we expect that LMX can lead to negative member mentality and behavior.

According to the social categorization theory, the categorization is often accompanied by deviations in cognition and attitude. Some insiders will attract each other and have more positive emotional interactions. As for the outsiders, there are often negative comments such as distrust and difficulty in cooperation (Hogg & Terry, 2000), so the friction and conflict are easily generated in the interpersonal interactions. As Opotow (1990), Hafer and Olson (2003) pointed out, the individuals in conflict will develop two different results, and a fair treatment should be adopted to the insider in the same circle based on the moral rules. The outsider will be excluded from the scope of justice, which becomes the object of exclusion. In summary, it is not difficult to infer that when subordinates perceive the difference of LMX, it may cause conflicts, and it even leads to a series of exclusion behaviors. Based on the previous studies and above discussions, we propose the following hypothesis:

H2: LMX is positively related with workplace ostracism

2.3. Mediating Role of Feeling Trusted

In an organizational environment, trust is an important factor for the individual, group, and organizational productivity (Dirks & Ferrin, 2002). Liden (1993) found that members in the high-quality LMX will receive more support and trust from their leaders as well as more freedom in the work, thus demonstrating higher levels of job performance and organizational citizenship behavior. Dirks and Ferrin (2002) also found that if the relationship between the leader and the subordinates was harmonious, the subordinates will trust the leader (Whitener, Brodt, Korsgaard, & Werner, 1998) and build or maintain a compatible relationship with the leader based on this trust. Finally, positive work responses will have an impact on those such as work attitude, behavioral intentions, performance, and career development. When employees gain the trust from their leaders, they will believe that the leaders will give them fair treatment, respect, and even more resources and opportunities. These will meet the material and spiritual requirements of employees, making them more satisfied with the leaders. Thus, they will reward their leaders for the close interaction with them and in the end they will be more loyal to the leaders and achieve their expectations for job performance.

The formation of a good relationship with trust between leaders and subordinates will enable the members to develop an overall trust for organization from their mentality. Members in the organization often make a comprehensive judgment on the atmosphere of trust in their environment. If members and leaders can maintain a high-quality LMX based on trust and loyalty, it can help them improve job performance and stimulate organizational citizenship behavior (Dirks & Ferrin, 2002; Mayer & Gavin, 2005). It can be seen that trust may affect the members’ job performance. According to Dirks and Ferrin (2002), the theoretical framework of “Leader Behavior – Trust – Member Response” was proposed, and a high-quality LMX will enable leaders and subordinates to trust and care about each other and encourage the subordinates to show positive work responses. Therefore, the following hypothesis is proposed:

H3: Feeling trusted positively mediates the relationship between LMX and job performance.

As mentioned, a good relationship with the leader often means that the member will achieve better career development and other favorable returns (Wei, Liu, Chen, & Wu, 2010). To maintain a good relationship with the leaders will generate more trust between the leaders and members (Dansereau et al., 1975). The outsiders often cannot feel the same trust from the leaders with those of high LMX, so their perceived gap in their low LMX will trigger a series of negative emotions, leading to strained international communication with more negative emotions as well. In the absence of a foundation of trust, the subordinates will not have a mutually beneficial relationship with the leaders, so it is difficult to achieve the positive work response. It may even lead to ostracism for the insiders.

On the other hand, the occurrence of workplace rejection may be generated by defensive motives which is avoiding negative interaction with the others through exclusive behavior to protect oneself (Robinson, O’Reilly, & Wang,2013). Therefore, when there are deep differences, it may be expected that negative interpersonal interaction with the colleague will result in some behaviors that exclude colleagues such as intercepting key information, the silent treatment and so on.

Robinson et al. (2013) argued that the organizational environment tends to ignore others and that greater differences between employees’ values will affect workplace rejection. For example, when employees feel trust from their supervisor and try more to repay, it will be much easier to neglect other colleagues which leads to rejection. When employees have great differences in values or working attitudes, they subconsciously classify such subordinates as the kind of people who is outside of the group. At the same time, negative attitudes and perceptions such as indifference, prejudice and even threats will be aroused. Interpersonal attraction is premised on similarity. Not having the same values means that members have different understandings and perceptions of interpersonal behaviors such as rational behavior and social participation. Feeling trusted unequally caused by differences in exchange relationships among leaders will lead to relationship conflicts. At this time, both outsiders and insiders will have stronger motivations to hurt and defend, which will lead to rejection. Based on the above discussion, we propose the following hypothesis:

H4: Feeling trusted positively mediates the relationship between LMX and workplace ostracism.

3. Methodology

3.1. Research Model

From the above analysis, the model (Figure 1) attempts to illustrate the effects of LMX for members by linking LMX with the job performance and workplace ostracism, and mediating role of feeling trusted on the relationship between LMX and job performance and between LMX and workplace ostracism.

Figure 1: Research Model

3.2. Sample

The data used in this paper was collected from January to March 2018. The survey method of this study was based on a paper questionnaire and on-site distribution. We surveyed 4 department stores in the Shandong Province (eastern part of China). 400 questionnaires were distributed, and 300 were returned, of which 285 were valid (63 groups, 4 or 5 people with a leader in each group).

3.3. Measures

(1) LMX: LMX adopted the Liden and Maslyn (1998) scale, which had 12 questions and 4 dimensions. These four dimensions were: Affect, loyalty, contribution, and professional respect, which were used to describe the degree of LMX (1, “strongly disagree” to 5, “strongly agree”). The previous studies have confirmed the validity of the scale (Greguras & Ford, 2006; Liden & Maslyn, 1998).

(2) Job Performance: We used a self-assessment scale from subordinates and it was developed by the Wayne, Shore, and Liden (1997). We used the performance ratings with six items. All items were coded on a five-point scale.

(3) Workplace Ostracism: The measurement is based on the 10 item scale developed by Ferris et al. (2008). We used the 5-point system, which is 1, “never”, 5, “always”. The high scores represented the high workplace ostracism, and the low scores represented the low workplace ostracism.

(4) Feeling Trusted: This study used Gillespie (2003) ten-item trust measurement scale. This scale had been proved to be effective by many studies and had good psychometric characteristics. According to the original theory, the perceived trust is defined as a will perception that the other party will be more likely to take risks. We used 5-point scale (1, “strongly disagree” to 5, “strongly agree”).

3.4. Method: Analytical Strategy

Confirmatory factor analysis (CFA) was used to assess the factor structure of the study’s variables and structural equation modeling to test our hypotheses. We reduced the number of parameters to be estimated following the partial aggregation method (Bagozzi & Edwards, 1998; Little, Cunningham, & Shahar, 2002). This method involves averaging the responses of the subsets of items measuring a construct. Because all the variables except LMX were unidimensional constructs (LMX has 4 dimensions), we followed the procedure recommended by Little, Cunningham, and Shahar (2002) to create three parcels of randomly selected items to serve as indicators for these variables. The goodness-of-fit index should not be sensitive to the sample size, should fit the data, and should preferably contribute to the simplicity of the model (Hong, 2000). In this study, we evaluated the model’s goodness of fit using the Tucker–Lewis Index (TLI), the comparative fit index (CFI), and the root mean square error of approximation (RMSEA).

4. Analysis

4.1. Validity and Reliability

We conducted CFA using maximum likelihood estimation procedures to test the construct validity. After deleting 1 item from Feeling Trusted and 1 from Job Performance (those items are not good for the general criteria for factor loading), we evaluated the model fit. Finally, we used LMX 12 items, Feeling Trusted 9 items, Job Performance 5 items, and Workplace Ostracism 10 items.

We chose the combination of fit indices following Hu and Bentler (1998). The model fit was acceptable (X²=1724.798, df=588, X²/df=2.933, TLI=.858, CFI=.875, RMSEA=.081). All the factor loadings are statistically significant (ranging from .711 to .991, see Table 1).

Table 1: Confirmatory Factor Analysis Results

4.1.1. Validity

We conducted CFA to test construct validity of the study variables. As shown on the Table 1, component reliability index of all the factors were above .90 (above the recommended level of .70 by Hairr, Anderson, Tatham, and Black (1992)). Discriminant validity was tested by Fornell and Larcker (1982). According to them, AVE (average variance extracted) for each construct should be larger than any squared correlations with any other constructs to ensure discriminant validity (Fornell & Larcker, 1981). Out of 12 possible comparisons (each variable can have 3 correlations with any other variables), the AVE of all the variables was higher than its squared correlations with another variable (see Table 2). Thus, we concluded that all the variables used in the study have construct validity.

Table 2: Correlations and Reliability Estimates

Note : **<0.01, The first entry inside of the parentheses is Cronbach's index of internal consistency reliability (alpha) and the second one is AVE (Fornell & Larcker, 1981)

4.1.2. Reliability

We evaluated construct reliability using component reliability and cronbach’s alpha. The result of CFA shows that all the component reliability (C.R) index of each latent factor was above the recommended level(0.7) (see Table 1). As shown in Table 2, cronbach’s alpha values of all the study variables are above .885 (LMX 12 items, Cronbach’s alpha = .959; Feeling Trusted 9 items, Cronbach’s alpha = .928; Job performance 5 items, Cronbach’s alpha = .885; Workplace ostracism 10 items, Cronbach’s alpha = .949). Taken together, C.R index and cronbach’s alpha provided the evidence that all the measurements have reliability.

4.2. Common Method Variance Test

According to Podsakoff, MacKenzie, Lee, and Podsakoff (2003), the relationships between constructs can be inflated or deflated by the common method bias. The common method bias was assessed via a post hoc analysis using Harman’s single-factor test (Harman, 1967) for all items. No single factor emerged in the results, and there was no general factor that accounted for the majority of the variance. An unrotated factor analysis extracted four distinct factors that accounted for 65.5% of the total variance. The largest factor explained 13.449% of the variance. These results provide additional evidence that common method bias is not likely to be a significant problem in this analysis.

4.3. Correlation Analysis

Before testing the hypothesis, we analyzed the correlations of the study variables as shown in the Table 2. The correlation coefficients between variables were significant as expected. In detail, LMX was significantly and positively correlated with feeling trusted, job performance but negatively correlated with workplace ostracism. Feeling trusted was significantly and positively correlated with job performance, but negatively correlated with workplace ostracism as also expected. The results were shown in Table 2.

4.4. Hypothesis Tests

4.4.1. Effects of LMX

The fit index of the structural equation model was good (χ2/df=2.866, TLI=.953, CFI=.965, RMSEA=.080) as shown in Table 3. Thus, we examined the hypotheses by significance of the path coefficients. The results of path analysis are in Table 3.

Table 3: Path Analysis

Note: WO: Workplace Ostracism

Hypothesis 1 is that LMX is positively related with job performance. The path coefficient was .304(p<0.001) which means H1 was supported. The path estimate from LMX to workplace ostracism was significant (B=-.491, p<.001). And the path from LMX to feeling trusted (B=.835, p<.001), from feeling trusted to job performance (B=.293, p<.001) were all significant, however, the path from feeling trusted to workplace ostracism was not significant. The empirical results show that LMX affects feeling trusted and job performance positively, however, LMX affects workplace ostracism negatively opposed to our expectation. Thus, H1 was supported but H2 was not supported. Feeling trusted appears to mediate the relationship between LMX and job performance, but more extensive test was needed in the next section. The individual relationships are shown in Figure 2.

Figure 2: Summary of Findings

4.4.2. Mediating effects of feeling trusted

We examined the total effect, direct effect, and indirect effect of each path and conducted bootstrapping test to ensure the mediating effects of feeling trusted. We set the number of bootstrap samples as 2,000(95% confidence level). The results are presented in the Table 4.

Table 4: Total Effect Analysis

Note) JP: Job Performance, WO: Workplace Ostracism

* < 0.05, ** < 0.01 (the results of bootstrapping test)

Table 4 shows that the indirect effect of LMX on job performance is significant (B=.245, p<.05). However, the mediating role of feeling trusted between LMX and workplace ostracism was not significant because the path estimates from feeling trusted to workplace ostracism was not significant. Thus, Hypothesis 3 was supported but Hypothesis 4 was not supported.

5. Conclusion

5.1. Discussion

This study focuses on the influence of LMX on employees' behaviors, examines the LMX and job performance and workplace ostracism; and explores the mediating role of feeling trusted. As expected, LMX is significantly and positively correlated with job performance. Higher LMX relationship does promote employees' feelings of being trusted and consequently motivates them to achieve higher job performance. However, unlike expectations, LMX is negatively correlated with workplace ostracism and the mediating effect of feeling trusted between LMX and workplace ostracism is not significant.

The results show that LMX is positively correlated with job performance. Graen and Cashman (1975) found that high-quality LMX relationships can improve the performance of the whole enterprise. In addition, Gerstner and Day (1997) found through meta-analysis that LMX was positively correlated with employee subjective performance evaluations. When establishing a high-quality LMX relationship with leaders, the employees become insiders which enable them to get more trust, care and beneficial resources from leaders. At this time, employees will be more enthusiastic about the work, and hope to complete the tasks assigned by the leader in better quality which will make the LMX relationship more harmonious, and thus improve the performance overall. A leader who pays enough attention to the inner activities of employees and communicates with them frequently is more likely to establish trust relationship with employees and get their respect.

This study proves the mediating effects of feeling trusted between LMX and job performance. High-quality LMX creates a sense of trust for supervisors, which increases their job performance. The perspective of feeling trusted has been relatively ignored in recent trust studies. Only a few studies have analyzed the importance of feeling trusted, although feeling trusted does have an impact on job performance. Only an in-depth understanding of the intrinsic mechanism of how feeling trusted affects job performance and understanding the circumstances under which feeling trusted has a positive impact on performance, can it be better conducive to trust management in enterprise practices.

LMX was shown to have a negative impact on workplace ostracism. The hypothesis in this paper is contrary to the result. Nevertheless, we have filled some gaps in the study on the causes of workplace ostracism. The employees who have good relationship with their leaders will rarely be actively marginalized by their colleagues, less likely to feel neglected by others, that is to say, employees in the organization will be more valued and colleagues will pay more attention to their feelings. Having a good relationship with leaders not only reduces the negative experience in the workplace (workplace ostracism), but also increases the positive experience (performance improvement). Liu, Liao, and Loi (2012) found that the relationship between leaders and employees was negatively correlated with the marginalization of employees, which was further developed in this paper. Marginalization denotes the phenomena in which individuals are placed in an isolated situation, or others refuse to communicate with them (Blackhart, Nelson, Knowles, & Baumeister, 2009), which is one kind of initiative behavior with a certain purpose and motivation. Workplace ostracism is felt by employees individually. Sometimes colleagues do not have negative motivation or actively take negative actions, but make individuals feel isolated and excluded (Robinson et al., 2013).

The larger difference of LMX is, the more resources that can be grasped by the employees who are in good relationship with the leaders, the more performance improvements the employees can make with the help of the relationship they built with their leaders. This is consistent with the main logic previously discussed. In China, especially in organizations with a more obvious hierarchy structure, the more important LMX proved to be. This study provides a research idea for the future research on LMX differential order atmosphere and even the localization of LMX in China.

In the Chinese cultural context, leaders are at the core of the circle. Employees can establish a good LMX relationship to improve job performance and reduce workplace ostracism. Because of a strong tendency of "rule by man" in China, whether it is the division of "inside" and "outside" the circle of leaders, the operation of organizational rules and regulations, or the HR mechanism of the organization, the evaluation and promotion of employees are all affected by "relationships". The distance of "relationship" has produced "internal and external differences" (Wang, 2012). High-quality LMX helps to create a friendly working environment and push the interpersonal interaction to a positive light. The LMX is decisive for employees' experience in the workplace.

5.2. Limitations and Directions for Future Research

Obviously, the employees in this study cannot represent all the employees. Particularly, the sampling subjects are relatively random, and there are certain limitations in the geographical and industrial ranges. So the conclusions may not be generalized to employees in all different kinds of industries. From the theoretical point of view, the LMX, feeling trusted and workplace ostracism should span industries, regions and occupational categories, which is subject to further empirical research.

Research can focus on the following issues in the future. First, in an organization, some employees regard feeling trusted caused by LMX differences as a burden on their work, and these costs may also result in higher performance and leadership appreciation. Therefore, how to balance the cost and benefits of trust is very important for employees to deal with the relationship between supervisors and employees and peer relations, which can be explored further in the future. Second, this study considers LMX under the background of Chinese culture, LMX is deeply influenced by Confucianism. The "reciprocal consensus" between supervisors and employees makes employees think that it is necessary to repay the trust of supervisors, which strengthens the sense of reward of employees on one hand, increases their responsibility on the other hand. However, the personality of employees is also a very important regulatory factor. If employees are not very demanding on themselves and lack goals for their own development in the organization, they may ignore the trust of their supervisors. For those employees who have a strong sense of self-esteem, ambition and image, they will choose to repay the trust from their supervisors no matter their career development prospects or their image in the minds of their superiors. Therefore, this constitutes a direction for future research.