비교이론을 조절변수로 한 리더-종사원 교환이론이 시기심과 직장 내 일탈 행위에 미치는 영향

The Effect of Leader-member Exchange on Envy and Counterproductive Behaviors Moderated by Similarity

Abstract

본 연구의 목적은 리더-종사원 이론이 시기심과 직장 내 일탈 행위에 미치는 영향을 탐구/조사하고, 이때 인식된 유사함은 조절변수로 사용되었다. 특히, 리더-종사원 이론은 부정적인 측면에 초점을 맞추고 연구를 진행하였는데, 리더-종사원 이론과 시기심과의 관계, 그리고 시기심이 리더-종사원 이론과 직장 내 일탈 행위 사이의 매개 변수로 어떤 역할을 하고 있는지 등에 관해 연구하였다. 더 나아가, 종사원들의 인식된 유사함이 조절변수로서 시기심과 직장 내 일탈 행위에 미치는 영향 또한 분석하였다. 직장 내 일탈 행위가 조직에 매우 부정적인 영향을 끼치는 것을 고려하면, 본 연구는 왜 이 행위를 하는지에 대한 연구를 수행함으로써 의미가 있다고 볼 수 있다. 총 238명의 서비스 종사원이 본 연구에 참여하였고, 연구의 결과는 모두 연구 가설을 모두 채택하였다. 본 연구의 결과는 리더십의 실무적인 측면과 이론적인 측면에 모두 이바지할 것이고, 특히 매니저들이 그들의 종사원을 어떻게 대해야 하는지에 대해 중요한 의미를 부여할 것이다.

The purpose of this study is to investigate the effect of leader-member exchange (LMX) on envy and counterproductive behaviors (CWB) moderated by similarity. Specifically, we focused on the negative side of LMX to examine the relationship between LMX and envy, and the mediating role of envy on the relationship between LMX and CWB. Further, we also examined the moderating role of similarity on the relationship between envy and CWB. Given that CWB can be harmful to any organizational, it was worthwhile to find possible antecedents of CWB, envy and LMX. A total number of 238 employees participated in this study and the results supported our hypotheses. The results of this study can have managerial implications, showing the important role of manager's personalized treatment for each of his/her subordinates.

I. Introduction

Counterproductive work behaviors (CWB) against norms and well-being of the organization and its members[1] have resulted in negative organizational consequence such as low morale, job dissatisfaction and negative customer services among employees[2]. Consequently, understanding what causes such behaviors has become an important research topic over the past few years[3]. Employee’s adaptation to a frustrating situation at work is likely related to CWBs, and not having a quality working relationship with a supervisor can easily create a frustrating situation where the employee is tempted to exhibit CWBs.

An employee’s working relationship with his or her supervisor is known as leader-member exchange (LMX)[4]. In LMX, a supervisor develops different levels of working relationships with each of his or her employees (e.g. high or low) due to time or resource constraints[5]. Higher quality LMX relationships provide employees with greater resources or better job assignments while lower quality LMX employees do not receive the same or similar support from the same supervisors[6]. As a result, lower quality LMX relationships are more likely to produce unfavorable consequence such as CWBs on employees and organizational outcomes[7].

As such, lower quality LMX relationships may lead to CWBs when employees experience envy at work. As the definition of CWBs indicates, employees voluntarily engage in such behaviors, implying that there could be antecedent conditions triggering employees to engage in such behaviors. One condition that might stimulate counterproductive behaviors is when employees feel envy. Envy is defined as an unpleasant emotion driven by upward social comparison with others and focuses on what one has compared to what the others have[8]. LMX can be a facilitator that might provoke employee envy since supervisors often control valuable resources that are critical for their success[9]. As a reaction of envy, harmful behaviors (i.e. CWBs) would be emerged by lower quality LMX employees toward their counterparts (i.e. higher quality LMX employees)[10]. Such harmful behaviors can hurt collaborating behaviors to achieve quality customer service and decrease helping behaviors among frontline hotel employees. The purposes of this study are to develop and to test a research model on LMX and CWBs within an envy and similarity framework among hotel employees.

II. Theoretical Background and Hypotheses

1. LMX and CWB

Leader-member exchange (LMX) explores different levels of dyadic working relationships between an employee and his or her immediate supervisor. Due to constraints of time and resources, supervisors develop close relationships with only a few key employees (i.e. higher quality LMX), and maintain their distance from the other employees (i.e. lower quality LMX)[11]. The greater the valued tangible and intangible resources, information, and support exchanges, the higher the quality of LMX relationships. Simultaneously, employees in lower quality LMX relationships seem to have less of valued tangible and intangible resources, information, and support exchanges[12]. Consequently, lower quality LMX employees may have higher possibility to report counterproductive behaviors or low performance.

Employee counterproductive work behaviors (CWBs) can be defined as employees’ intentional and harmful behaviors that are opposed to the legitimate interests of members in the organization and are detrimental to the effectiveness of the organization[13]. CWBs can be exhibited towards other individuals or organization and has been categorized in two dimensions in terms of targets: behavior directed to the organization (CBO) and behavior directed to people (CWBP)[14]. We focused on CWBP, because lower quality LMX may generate discretionary harmful behaviors towards other employees. CWBP at work include insulting others, spreading rumors, emotional and/or physical, and various types of violence[15].

A number of researchers have argued that LMX and CWBs have a negative relationship [16]. One important antecedent of CWBs be supervisory support, which can be LMX relationships. Rotundo and Sackett argued that when superiors evaluate their employees, they value employee CWBs as the most important indicator over task performance or organizational citizenship behaviors[17]. Given that high-quality LMX employees have better evaluations from their supervisors, it can be expected that LMX will have a negative relationship with CWBs. Using relative deprivation theory and reactions of relative deprivation, Bolino and Turnley recently discussed CWBs as a negative reaction of low-quality LMX relationships. Relative deprivation is defined as a person’s discrepancy between the real and the ideal[18]. From the low-quality LMX employees’ perspective, the ideal situation can be high-quality LMX and the real situation can be low-quality LMX. Relative deprivation can be acted as a facilitator on where low-quality LMX employees engage in CWBs as a reaction of relative deprivation toward high-quality LMX employees.

2. LMX, Envy, and CWB

Work envy might be an important variable that explains how low quality LMX triggers employee deviant behaviors. Envy can occur when a person learns that he or she does not own another’s superior quality, achievement, or possession, and wants to have those superiorities[19]. To the extent that a more successful person is seen as being similar to one, there exists potentially unpleasant social comparison[20].

As the definition of CWBs indicates, employees voluntarily engage in deviant behaviors. This suggests that it is important to find antecedent conditions that trigger employees to begin engaging in deviant behaviors. One condition that might provoke deviant behaviors is when employees experience envy. Envy is an unpleasant and hostile emotion driven by social comparison with others and focuses on what one has compared to what the others have. Specifically, when employees compare what they have with what their coworkers have, the comparison can easily be upward social comparison, as envy theory suggests[21]. Thus, when employees perceive that their relationships with a supervisor is low quality (i.e. low LMX), they are very likely to engage a social comparison process with their colleagues who appear to have higher quality LMX relationships. Therefore, we hypothesize the following:

Hypothesis 1: Similarity moderates the relation between envy and CWB such that the relation is stronger when similarity is high.

Hypothesis 2: Envy mediates the relationship between LMX and CWB.

Figure 1. Conceptual Model

III. Methodology

1. Respondents and procedures

For the purpose of this study, we sampled 238 hotel service employees, the average age was 38 and female employees were 58.4%. The number of years working their current job is 5.3 years and the number of years working in the hotel industry is 7.4 years. For the respondents, there was not much difference among the variables.

Table 1. Respondents' descriptive statistics

2. Measure

For the LMX measure, we used the seven-item LMX 7[5]. The example scales of LMX were “My supervisor and I get along well together, ” or “My working relationship with my supervisor is effective.” For the envy measure, Payne’s (2001) scale was used in this current study. Two sample scales for envy is “The bitter truth is that I generally fell inferior to the people I work with, ” or “Feelings of envy constantly torment me when I’m at work.”[22]. For the CWB measure, Bennett and Robinson’s (2000) scale was used for this study. Two samples of this measure was “Said something hurtful to someone at work abuse, ” or “Insulted someone about their job performance”[2]. Schaubroeck and Lam’s (2004) perceived similarity scale was used to measure similarity. A sample item for similarity was “my coworkers and I have similar experiences at work”[23].

IV. Results

Means, standard deviations (SDs), and Cronbach’s alpha for all variables are presented in [Table 1]. Cronbach’s alpha was examined to see variables’ reliabilities.

Table 2. Means, SDs and Cronbach‘s Alpha (N = 238)

For hypothesis 1, perceived similarity moderated the relation between envy and CWB such that the relation is strong when perceived similarity is high, was accepted as shown in [Figure 2]. That is, similarity moderates the relationship between envy and CWB such that people report higher similarity with higher envy also report higher CWB. In this moderated relationship, similarity was divided by low and high, and the value of the low moderator (e.g., low similarity) was 9.29 for low variable 1 and 8.08 for high variable 1. These are the predicted criterion values. For the value of the high moderator (e.g., high similarity) was 11.38 for low variable 1 and 12.18 for high variable 1 at varying levels of variables 1 and the moderator variable.

Figure 2. Interaction of envy with similarity on CWB

For hypothesis 2, path analysis was used to test direct and indirect effects of variables. The result of path analysis confirmed the significant relationship from LMX to CWB through envy, showing -.809*** from LMX to envy, .465*** from envy to CWB, -.603*** from LMX to CWB (*** p<0.001). That is, envy with lower LMX relationships are more likely to perceive envy, and in turn, such employees are more likely to exhibit CWB.

Figure 3. Path analysis for H2

V. Discussion

It was not our intention to suggest that LMX does not have benefits in the workplace; rather our intention was to explore a possible negative effect of LMX at some points under which LMX can be harmful in service organizations, maybe further in all organizations. The finding that LMX finally can lead to a negative effect is a reverse outcome in that LMX has been regarded as a benefit to the organizations. The findings, however, is fruitful since there is little research regarding a dark side of LMX, the reverse relationship between LMX and envy, and finally LMX and CWB. Therefore, this study can give some answers why employees sometimes harm their coworkers and peers, given that harmful behaviors are critically destructive in the service setting. Therefore, if employees perceive envy due to different working relationships with their immediate supervisors, managers should consider it carefully. In advance, managers may try to prevent it before it happens. Once envy comes out from any employees, harmful behaviors can also be emerged.

However, this study does have limitation as every research has. We did not examine situational factors such as work group norms, climate of the organizations, things that can influence the overall relationships. Another limitation was that this study did not consider the time factor; that means all variables were assessed at the same time. Therefore, correlations among variables can be inflated.

Despite of such limitations, this study can contribute to the practical and theoretical leadership field in that little research regarding a negative side of LMX, the relationship between LMX, envy, and CWB has examined. Much research has devoted to the positive side of LMX and the results instead.