The Effect of Perceived Similarity on Leader-member Exchange and Deviant Workplace Behavior

비교이론이 리더-구성원 교환이론과 직장 내 일탈행위에 미치는 영향

Abstract

Using a sample of two-hundred thirty-eight employees in the service industry, this study examined the relationships between perceived similarity, leader-member exchange (LMX), envy, and deviant workplace behavior (DWB) drawing on social comparison theory. Specifically, the current study predicts that an individual who perceives similarity to co-workers whose LMX is higher is more likely to report lower levels of LMX by focusing on upward social comparison. In addition, the study examines the moderating effect of envy in the relationship between perceived similarity and LMX. The results of the study supported the hypotheses, implying that, first, envious feelings coming from lower levels of LMX positively predict DWB. Second, employee envy moderated the relationship between perceived similarity and LMX such that envious people who report being similar to their colleagues also report lower levels of LMX with their supervisors. This means that employees who reported lower levels of LMX with their supervisors, reported committing more DWB than their counterparts. Finally, LMX mediates the relationship between perceived similarity and DWB, indicating that individuals who view themselves similarly to their colleagues report lower levels of LMX with their supervisors and in turn commit more DWB. We believe that it highlights the need for employees to feel special in order to have unique attention by, and relationships with their supervisors. These results have implications and suggest the need for the further study of employee's negative emotions, LMX theory, and DWB.

본 연구는 서비스업에 종사하는 238명의 종업원을 대상으로 조직 내 인식된 유사함, 리더-구성원 교환이론, 시기심과 직장 내 일탈행위와의 관계를 사회비교이론을 토대로 실증적으로 분석하였다. 상향비교에 초점을 맞춘 연구분석 결과, 상향비교에 초점을 맞추어, 리더-구성원 교환이 상위에 있는 동료와 본인을 비교하는 종업원은 리더-구성원 교환이 하위에 머물 수 있다는 결과가 나왔고, 이들의 시기심은 비교이론와 리더-구성원 교환이론에 영향을 미치는 것으로 드러났다. 연구 결과를 보면 시기심은 리더-구성원 교환이론과 직장 내 일탈행위에 영향을 주고, 또한 시기심은 하위 리더-구성원 교환이론과 관계가 있는 것으로 나타났다. 조직 내 부정적인 감정이 조직에 미치는 영향에 관한 연구가 미비한 실정을 감안하면 본 연구는 감정과 리더십, 성과분야의 연구에 긍정적인 기여를 할 것으로 기대된다.

I.Introduction

The service industry has become one of the largest industries in today’s economy. Therefore, researchers have been paying greater attention to this sector, including the behavior of employees in this sector. In the current study, we examine the prevalence and antecedents of deviant workplace behavior(DWB). DWB is defined as behaviors that harm the organization or its members[1]. We believe that certain variables such as leader-member exchange, (LMX), perceived similarity, and envy will affect the prevalence of DWB.

When Festinger(1954) used the term social comparison, operationalized by perceived similarity in this study, he assumed that there exists a motive to know that one’s opinions are correct and to understand accurately what one is capable or incapable of doing[2]. To the extent that a more successful person is seen as being similar to one, there exists a potential for invidious social comparisons and particularly comparisons with others regardless of fair or unfair situations. Perceiving similarity in another person is associated with the expectation that the other ought to experience similar outcomes. Such similarity, the preference for comparing with similar others, is revealed when one evaluates her or his similarity with the more successful other[3]. One of the social comparison predictions is upward social comparison (superior other), a comparison against those faring better than the self.

Even though the majority of LMX research has investigated the positive side of LMX, the current study had a focus on the negative side of LMX, believing that the negative side of LMXrelates to destructive consequences such as DWB in any organizations. As such, based on the perceived similarity and social comparison theory, this study is to investigate the antecedents of DWB while examining the relationship between perceived similarity, LMX, Envy and DWB.

II. Theoretical Framework and ResearchHypotheses

1. Perceived similarity and LMX

Using upward social comparison, this study used perceived similarity in the following way: perceived similarity in another person is associated with the expectation that the other ought to experience similar outcomes, for instance, relationships in leader-member exchange (LMX). LMX theory represents that leaders develop different relationships with followers in the same work group[4-6]. In other words, a person X expects that another person Y, who is similar to the X, ought to have a similar relationship in LMX. Since we focused on the lower quality LMX employees, the person X should be one of the lower quality LMX employees and the expected similar relationship could be a lower LMX relationship. Therefore we hypothesize the following:

Hypothesis 1: Perceived similarity will negatively influence LMX.

2. Envy and LMX

Because of our focus on lower quality LMX than their counterparts (i.e. higher quality LMX employees), such upward comparisons, instead of downward comparisons, were of prime concern. Using balance theory[7], we expected when a person perceives that another person who is similar to oneself obtains a valuable outcome (for instance, a higher LMX relationship) the first person failed to obtain, the first person may feel a violation of self-balance. This upward comparison and violation of self-balance should prompt a negative affect such as envy and may include shame for inferiority and hostility[8][9]. If a person is doing an upward comparison, which results in undesirable outcomes for self and desirable outcomes for another, one primarily would be experiencing envy[10]. Envy is a complex and dually focused emotion, which means focusing on what the self lacks and what the other possesses[11]. In this study, envy was used to individual difference as a moderator in the relationship between perceived similarity and LMX. We postulated that if an employee perceives a higher envy, s/he may have a lower level of LMX with superior. Therefore, we hypothesize the following as shown in [Figure 1]:

Figure 1. Hypothesis 1

Note. LMX=leader-member exchange, PS=perceived similarity

Hypothesis 2: There will be an interaction between perceived similarity and envy such that similarity negatively relates to LMX when envy is high.

3. LMX and DWB

Although several studies have suggested that unfavorable supervisor treatment may result in DWB[12][13], to our knowledge there has been very few published studies to prove such relationship, i.e, the negative relationship between LMX and DWB. By its nature, LMX focus on the dyadic relationship between leaders and subordinates and therefore each relationship may be different. Since leaders develop unique and different relationship with followers, the quality of the relationships can affect the followers’ behaviors[14]. Whereas high quality relationships are supportive and personal, low quality relationships can easily be impersonal and economic exchanges[15]. Because they may want to defend themselves against harsh treatment from supervisors and if the treatment from supervisors is important for them, lower level LMX employees may be more shown aggressive or abusive behaviors at work. Therefore, with all the reasons shown above, we hypothesize the following as shown in [Figure 2]:

Figure 2. Hypothesis 2

Note. LMX=leader-member exchange, PS=perceived similarity, DWB=deviant work behavior

Hypothesis 3: LMX will mediate the relationship between perceived similarity and the outcome variable, DWB.

III. Methodology

1. Samples and procedures

In the current study, a sample of 238 front-line hotel employees (i.e. food and beverage, front desk, and housekeeping employees) in 6 hotels (Western and Central Pennsylvania) completed a self-administered questionnaire. The author had a permission to attend big group meetings in those hotels with the help of the general managers of each hotel in July and August 2018. The response rate was about 79% (i.e. 238 usable data from 300 total survey questionnaires). Employees’ ages ranged from 18 to 64, and an average age was 38. Male employees comprised 41.5% of the study population, while female employees were 58.4%. Most of the population was full-time employees compared to part-time employees (94.9% vs. 5.0%, respectively). The numbers of years employees worked in their current jobs ranged from less than one year to 37 years with an average of 5.3 years. The employees worked with their immediate supervisor from less than one year to 17 years. The average length of this relationship was 3.6 years. The number of years working in the hotel industry ranged from less than one year to 50 years with an average of 7.4 years. [Table 1] shows the composition of the study participants.

Table 1. Participants’ characteristics

IV. Results

Based on the result from regression analysis shown in [Table 3], Hypothesis 1 was supported; perceived similarity was negative related to LMX. As shown in [Figure 3], hypothesis 2 was supported. Envy moderates the relationship between perceived similarity and LMX such that very envious people who report being similar to their colleagues also report lower levels of LMX with their supervisor.

Table 2. Means, standard deviations and reliability of main variable scales

Note. LMX=leader-member exchange, PS=perceived similarity, DWB=deviant work behavior

Table 3. Regression analysis

Note. LMX=leader-member exchange, PS=perceived similarity, DWB=deviant work behavior,** p<.01.

Figure 3. Interaction of similarity with envy on LMX

Hypothesis 3 is also supported in that LMX mediates the relationship between perceived similarity and DWB. We conducted the Soble test to measure the mediating effect of LMX on the relationship between similarity and DWB. That is, conducting the Sobel test is to measure if the indirect effect of similarity on DWB via LMX is significantly different from zero. The results of the Sobel test supported Hypothesis 3 (z=3.7775, p <.001). Such mediating effect of LMX on the relationship between perceived similarity and DWB means that due to individual's different perceived similarity, employees with lower LMX may more often report DWB.

V. Limitations and conclusions

Even though the current study suggests an interesting point of LMX and DWB that sheds light on the negative side of LMX and a related finding (i.e. DWB), there are some limitations. The results of the study could have confounding effects, coming from a self-report survey of the participant employees. To counteract this result is to survey employees’ supervisors to match the responses or to interview them. Surveying employees’ supervisors also could reduce the common-method variance problem for the future. Interest in LMX and DWB continues to grow; however, to date there have been very few studies to focus on low quality LMX [16] and the relationship between LMX and DWB. With upward social comparison theory and balance theory, this study examined the relationships between envy, perceived similarity, leader-member exchange (LMX), and deviant work behavior (DWB). Specifically, the current study predicted that an individual who perceives similarity to another is more likely to report low levels of LMX by focusing on upward social comparison. In addition, the study examined the potential moderating effect of envy in the relationship between perceived similarity and LMX. Our study found support for the following findings. First, LMX negatively predicts DWB and envious feelings positively predict DWB. Second, employee envy moderated the relationship between perceived similarity and LMX such that envious people who report being similar to their colleagues also report lower levels of LMX with their supervisors. This means that employees who reported lower levels of LMX with their supervisors, reported committing more DWB than their counterparts. Third, LMX mediates the relationship between perceived similarity and DWB, indicating that people who view themselves similarly to their colleagues report lower levels of LMX with their supervisors and in turn commit more DWB. While on the surface this is a paradoxical finding, we believe that it highlights the need for employees to feel special in order to have unique attention by, and relationships with their supervisors.

The results of this study have both practical and theoretical implications for LMX, employee emotion (envy) and as well as DWB. For LMX theory, there have been very few studies focusing on the lower quality of LMX relationships, whereas a lot of attentions have given to the higher quality of LMX relationships. For the practical side, managers pay more attention to the lower quality of LMX employees, since we found that lower quality LMX relationships were one of the antecedents of DWB. Managers need to find solutions to reduce such counterproductive work behaviors. Finally, our study has a contribution to the body of discrete emotions. Compared to the area of general mood or trait affect (e.g., positive or negative affectivity), the area of discrete emotions such as envy in organizational behavior is at a premature stage [17].