Expressing opportunities or objections: The influence of LMX
differentiation and the reciprocity principle on the voice of employees
Master’s thesis, specialization Human Resource Management University of Groningen, Faculty of Economics and Business
Expressing opportunities or objections: The influence of LMX differentiation and the reciprocity principle on the voice of employees
ABSTRACT
The purpose of this study is to examine the influence of LMX (differentiation) and reciprocity on prohibitive and promotive voice. Due to the recent development of voice in prohibitive and promotive voice (Liang et al., 2012), there is uncertainty in the effect of leaders on both dimension of voice in employees. A mediated moderation model is proposed, in which the motivation to reciprocate, emanated from social exchange (Blau 1964; Gouldner, 1960) explains the relationship between LMX and both dimensions of voice. Furthermore, LMX differentiation in the teams is taken into account to show the effect different exchange relationships might have on the reciprocity principle in employees. Questionnaires were conducted in teams of at least two employees from various organizations in the Netherlands and Germany (N = 180, 51.7% female, 48.3% male). For the analyses, regressions were performed using PROCESS model 7. None of the hypotheses are supported by the results. However, findings indicate a positive relationship between LMX and both prohibitive and promotive voice. Besides, a negative effect of negative reciprocity on promotive voice was found. Implications and possibilities for future research are highlighted.
Keywords: LMX, LMX differentiation, Reciprocity, Prohibitive voice, Promotive voice.
Expressing opportunities or objections: The influence of LMX differentiation and the reciprocity principle on the voice of employees
INTRODUCTION
In this fast changing and dynamic world, organizations have to adapt and innovate continuously in order to remain competitive in the market. Employees’ contribution to this survival is essential. Not only by doing their tasks sufficiently, but by behaving proactive and initiating ideas for processes in the organization as well (Liang, Farh, & Farh, 2012). The most obvious way for an employee to share their ideas through their voice, is to turn to their leader. The leader evaluates these ideas and decides if their idea is good enough to be
implemented (DeNisi & Murphy, 2017). Previous research developed the concept of voice to include the expression of employees (van Dyne, Ang, & Botero, 2003). As of today, the importance of voice is widely acknowledged. Studies showed that learning and performance in the workplace enhances when employees share their ideas, visions and opinions (Farh & Chen, 2014). In order to innovate and thrive, organizations need ideas and employees are the most obvious source to draw from. Therefore, it is interesting for organizations to know more about the context in which employees share their ideas and opinions with their leaders.
Although research has clearly demonstrated that leaders have a substantial influence on the extent to which employees express their opinions and thoughts (Chamberlin, Newton, & Lepine, 2017), there are several issues that remain unclear. The first unclarity, is the existence of different opinions about the underlying mechanisms of the leader-member relationship on the voice behavior of employees (Burris, Detert, & Chiaburu, 2008; Chamberlin et al., 2017). Previous research already acknowledges leader behavior as a key antecedent of voice, because of the direct impact of the leader on employee behavior (Chamberlin et al., 2017). The
relationship between a leader and employee can be explained by the Leader-Member
(low-high) relationships with their leader (Dienesch & Liden, 1986). The quality of the relationship between the leader and employee can influence to what extent the employee will feel safe enough to speak out. Hence, the argument that LMX has a great influence on voice is widely acknowledged, but the explanation of this relationship is still unclear. More insight in this relationship might be of great importance for leaders in order to understand their
influence on the voice of employees in general.
Secondly, there is a lot of ambiguity in the research on voice due to the recent transformation of voice as one construct into two separate dimensions: promotive and prohibitive voice (Liang et al., 2012). The majority of the voice research focused only on the promotive aspect of voice, however nowadays prohibitive voice is taking root in the literature as well (Liang et al., 2012). Given this transformation in the voice literature, established relationships from previous research, may be incomplete or incorrect (Chamberlin et al., 2017). Thus, research shows that leaders influence the voice in employees, but the effect of LMX on the different dimensions of voice is yet unknown. It could be the case that the influence is the same for both dimensions, but there are indications of differences in the antecedents of the dimensions as well (Liang et al., 2012; Wei, Zhang, & Chen, 2015). Hence, it is important to examine the mechanisms that lead employees to voice in a certain way. Moreover, the voice literature is not clear about the interpretation of promotive and prohibitive voice. In general, research looks at voice as a positive and beneficial construct (Liang et al., 2012). However, the partition of the two dimensions leaves room for further specification. Especially prohibitive voice has an ambiguous character and due to little research on this dimension it is uncertain if prohibitive voice always derives from a constructive motivation of the employee (Chamberlin et al., 2017).
(social exchange; Blau, 1964). Social exchanges can include favors and benefits or more negative behaviors (Blau, 1964). Thus, the quality of the exchange relationship (LMX) induces a motivation in the employee to reciprocate and share ideas with their leader or withhold them (Settoon, Bennet, & Liden, 1996; Wayne, Shore, & Liden, 1997). When examining the effect of the LMX relationship on voice, it is important to note that this relationship is relative due to the different exchange relationships a leader may have at the same time. These relationships could differ in quality and employees tend to compare their own relationship with other existing relationships. Due to quality differences in relationships (LMX differentiation), employees will be treated differently as well. These differences result in different perceptions in employees (Henderson, Liden, Glibkowski, & Chaudhry, 2009). For example, a comparison with a higher quality relationship than your own may cause perceived feelings of injustice. Accordingly, the various perceptions may influence the exchange relationships, and therefore the voice in employees as well.
The current study builds on the research of Liang et al. (2012), by looking at prohibitive and promotive voice in a leader-employee context. We contribute to the literature by
examining the motives that cause employees to use either promotive or prohibitive voice. Little is known about significant antecedents (Chamberlin et al., 2017) that cause this division, and the reciprocity principle (Gouldner, 1960) might contribute to this gap. Moreover, we will provide more insight in the ambiguous design of prohibitive voice.
influence this variety might have on the form of reciprocity. Hence, this research will contribute through providing more knowledge about the influence leaders may have on the voice of their employees in a team context.
For organizations to innovate and grow, the ideas and opinions of their employees are of great importance. Therefore, it is essential to know more about the ways in which you can promote and inhibit these kinds of behavior. Moreover, it is important to take into account that it is not always easy for employees to express themselves, depending for example on the nature and sensitivity of the ideas. As described above, leaders can influence the expression of the opinions and ideas of their employees to great extent. The current study can provide insights for the organization in the effectiveness of the different relationships leaders have with their employees. Furthermore, directions for best practices for leaders in terms of
motivating certain behaviors in their employees may be indicated. Hence, this research is very relevant for organizations in order to succeed in the dynamic and changing environment.
THEORY Leader-Member Exchange and Reciprocity
role-making. This dynamic process determines the quality of the exchange relationship of leaders and their employees (Dienesch & Liden, 1986; Dulebohn, Bommer, Liden, Brouer, & Gerris, 2012). These relationships are either based solely on economic exchanges (low-quality LMX relationships) or next to that on social exchanges (high-quality LMX relationships). Economic exchanges imply purely the tasks that are described in the job description of both parties. High quality relationships can be developed through social exchanges of contribution, loyalty, mutual affect and professional respect (Dienesch & Liden, 1986; Liden & Maslyn, 1998).
The motivation of the employee to make a contribution to the relationship (next to strict task behavior), is established in the reciprocity principle that follows from the exchange relationship (social exchange; Blau 1964; Gouldner, 1960). Thus, an urge to balance the relationship arises. Gouldner (1960) describes this norm of reciprocity: ‘people should help
those who have helped them, and people should not injure those who have helped them’.
Liden, 1997). Based on the above, this study proposes the following hypotheses:
Hypothesis 1a. LMX is negatively related to negative reciprocity.
Hypothesis 1b. LMX is positively related to positive reciprocity.
LMX differentiation
Next to the relationship of the employee itself, the perceived relationship of peers influence the reciprocity principle as well (Henderson et al., 2009). Often, a leader has more than one subordinate and engages in different types of exchanges with every subordinate. Accordingly, the subordinates develop various perceptions of the leaders’ behavior and their quality relationship (Cogliser & Schriesheim, 2000). The development of the different
relationships of a leader is referred to in the LMX research as the LMX differentiation process (Martin, Thomas, Legood, & Russo, 2015). LMX differentiation is defined as a set of
different, dynamic exchanges that emerges between a leaders and their subordinates. When looking at LMX (differentiation), the current study takes the level of the employee into consideration at all times. In line with the role-taking and role-making process of LMX, different quality relationships (low to high) are formed through engaging in different exchange patterns with subordinates (Henderson et al., 2009). Hence, LMX differentiation can create a context in which there is variability in the team concerning the (perception of) quality of relationships (Liden, Erdogan, Wayne, & Sparrowe, 2006). Due to these various relationships, comparison of LMX within teams will occur. Specifically, not only the
relationship employees hold with their leader influences the exchange of resources, but their relationship compared to those of their colleagues as well. A high quality relationship will imply unique advantages and obligations compared to employees in lower quality
feelings of endowment in high LMX employees (Blau 1964; Gouldner, 1960), when comparing to lower LMX colleagues. The feeling of endowment creates the motivation to positively reciprocate the benefits, in order to balance the relationship (Liden et al., 1997). These preferential treatments in a high quality LMX relationships can be perceived as unfair by lower LMX relationships and induces a negative affect (Seo & Lee, 2017). Moreover, research of Masterson, Lewis, Goldman, & Taylor (2000) states that these comparisons may evoke feelings of deprivation in low quality LMX employees. The higher the variability in the LMX relationships in the team, the stronger this effect (Henderson et al., 2009). This feeling of deprivation motivates employees to reciprocate negatively on their low quality LMX (Blau 1964; Gouldner, 1960; Sparrowe & Liden, 1997). By contrast, when the team variability is low, there are no additional benefits to gain from relative advantages and treatments (Erdogan & Liden, 2002). Consequentially, it can be said that the lower the variability, the less
important the comparison of LMX relationships is for the employee. In line with this, research of Henderson et al. (2008) found that the exchange relationship was strengthened when the LMX differentiation increased. The same study showed that when variability in the team is low, the comparison to group members might not be important in predicting the behavior in the relationship, as their individual exchanges were a stronger predictor for their behavior (reciprocity) (Henderson et al., 2008). Based on the literature described above, the current research proposes the following hypotheses:
Hypothesis 2a. The negative relationship between LMX and negative reciprocity is stronger
when LMX differentiation is high rather than low.
Hypothesis 2b. The positive relationship between LMX and positive reciprocity is stronger
when LMX differentiation is high rather than low.
Voice as means of reciprocation
resources in terms of behavior or attitudes that are exchanged (Graen & Uhl-Bien, 1995). In general, it is important to take into account that many of the consequences of reciprocity can be traced back to the extent to which leaders actually allow employees to reciprocate. A consequence of a negative exchange might include leaders who take credit for the work of their employees. The employee might reciprocate negatively by showing less positive and more harmful behavior by withholding crucial information for an important project, which leads to failure for the leader (Sparrow & Liden, 1997). If there is a high quality relationship, a leader may contribute to the development of the employee (i.e. training, promotion and coaching). To return this resource, the employee might increase in positive behavior through commitment and task responsibility in order to support the leader where possible (Graen & Scandura, 1987). Hence, it can be said that employees in a high LMX relationship have the urge to reciprocate positively, and therefore this will lead to more positive behavior and attitudes like organizational citizenship behavior (OCB; Settoon et al., 1996). Employees in a low LMX relationship, on the contrary, will be motivated to negatively reciprocate and therefore will engage in negative behavior like acting offensive (Caliendo et al., 2012). Voice
Literature on voice argues that there exist two dimensions: prohibitive and promotive voice (Liang et al., 2012). Both dimensions are constructive and aimed to benefit the
organization. However, they differ in their focus and goal-setting. Prohibitive voice entails the expression of concerns and comments about existing practices and behaviors that may harm the organization. It can be said that prohibitive voice is problem-focused in nature, because it emphasizes on the current harmful and wrong procedures or events (Liang et al., 2012). On the contrary, promotive voice entails the expression of ideas, ideals and suggestions to improve certain practices of their job or organization. Important in promotive voice is the future-oriented character of the initiatives (Liang et al., 2012). The majority of the voice literature only focuses on the promotive dimension (i.e. expression of ideas), but recent research acknowledges prohibitive voice (expression of potential harm) as an important dimension as well (Farh et al., 2004). This is due to the fact that prohibitive voice serves as an important detector of potential problems, incidents and process losses (Liang et al., 2012). Even more interesting is the ambiguous character of prohibitive voice. On the one hand prohibitive voice creates awareness for (potential) problematic practices, or addresses the violation of standards of justice and honesty (van Dyne et al., 1995). On the other hand, prohibitive voice can evoke negative emotions and lead to increased conflict. Furthermore, dissent (Graham, 1986) and whistle blowing (Miceli & Near, 1985) are considered prohibitive voice as well. For example, leaders may experience these expressions as annoying and
Even though there are indications that prohibitive voice may entail negative behavior, the existing literature states both voice dimension as constructive. Therefore it seems
plausible that they are related similarly to antecedents (Chamberlin et al., 2017). However, next to the general antecedents of voice included in research of Morrison (2014), there are studies that argue differences in antecedents between the two dimension of voice as well. For example, Liang et al. (2012) found a stronger relation between promotive voice and
obligation, than prohibitive voice; when employees feel obligated to contribute, they are more likely to engage in voice behavior (Liang et al., 2012). Leader behavior is an important antecedent of voice as well, because of the reciprocating relationship (Burris et al., 2008). It can be said that employees may be motivated to voice due to the reciprocity principle. Moreover, if the quality of the LMX relationship is high, employees are more likely to experience a safe and trusted relationship, and are therefore more likely to contribute through voice (Chamberlin et al., 2017). Besides, a safe and accepting relationship is especially important to stimulate an employee to express concerns or risky ideas (Zhao & Olivera, 2006). To conclude, the literature is not conclusive about the potential differences in antecedents of promotive and prohibitive voice (Chamberlin et al., 2017).
from the perceived LMX relationship. Hence, this study proposes two hypotheses (3b and 3c) for prohibitive voice to entail this ambiguous character, and we will derive further knowledge from the results. Hence, based on the above described literature, the current study proposes the following hypotheses:
Hypothesis 3a. Positive reciprocity is positively related to promotive voice.
Hypothesis 3b. Positive reciprocity is positively related to prohibitive voice.
Hypothesis 3c. Negative reciprocity is positively related to prohibitive voice.
Next to the influence the perception of the LMX relationship of employees might have on voice, the relative quality of their relationship can be of great importance as well. For example, when an employee has a relatively low LMX relationship, feelings of deprivation motivates an employee to reciprocate negatively (Blau 1964; Gouldner, 1960; Sparrowe & Liden, 1997). The higher the variability in LMX relationships in a team, the stronger the effect on the reciprocity (Henderson et a., 2009). For example, when there is high variability in the team, and an employee has a relatively high LMX relationship, this employee wants to retain or even improve this beneficial position more by expressing creative solutions and ideas where possible. Hence, LMX differentiation influences the motivation to reciprocate, and therefore the voice of employees. Based on the above, the current study proposes the following moderated-mediation hypotheses:
Hypothesis 4a. The indirect relationship between LMX and promotive voice as mediated by
positive reciprocity will be positive and stronger when LMX differentiation is high rather than
low.
Hypothesis 4b. The indirect relationship between LMX and prohibitive voice as mediated by
positive reciprocity will be positive and stronger when LMX differentiation is high rather than
Hypothesis 4c. The indirect relationship between LMX and prohibitive voice as mediated by
negative reciprocity will be positive and stronger when LMX differentiation is high rather
than low.
METHOD Procedure and Participants
The current study concerns a survey study. Questionnaires were conducted at teams of various firms in the Netherlands and Germany (51.7% female, 48.3% male). The teams were employed in divergent branches (i.e. Education, Government, Advertising, FMCG).
Two questionnaires were developed: one for the leader and one for the employees. The questionnaires were entered in Qualtrics and were made available in the Dutch, English and German language. It took approximately 30 minutes to complete the questionnaire. Teams were able to participate in the study by submitting their work mail addresses. Only work e-mail addresses were accepted, because this increased the chance of real teams applying for the study. Moreover, the chance the participants filled in the questionnaire at a controlled work environment, was larger. To inform the leaders of the teams about the study, an information letter was send by e-mail. The study took place from October 29th till November 19th 2018. Participation in the study was voluntary. At October 29th, 9:00 am, the participants received a link to the questionnaire. In November the participants received a reminder to complete the survey. After completing the survey an e-mail address was presented, in case the participants wanted to comment on the survey or ask a question. As a gratification for participating in the study, the leader received a general rapport of the overall scores of the instruments.
The teams existed of one leader and a minimum of two employees (M = 5.45
The response rate was 40 percent. Measures
Leader-member exchange (LMX). The LMX was measured in the employee by a questionnaire of Liden & Maslyn (1998). The questionnaire consists of 12 items. The items are being measured on a five-point Likert scale (1 = Totally disagree, 5 = Totally agree). An example of an item is: ‘I like my supervisor very much as a person’. In total the items form a reliable questionnaire (α = .87). For the analyses the relative score of the employees compared to the average score of their team was used.
LMX differentiation. To measure the LMX differentiation in the teams, the standard deviation of the LMX scores were used. The intrateam variation of the LMX are the LMX differentiation scores. The deviation in the relative LMX scores in the team result in the LMX differentiation scores.
Positive and negative reciprocity. Based on multiple studies, six items to measure reciprocity were composed (Caliendo, Fossen, & Kritikos, 2012; Egloff, Richter, &
Schmukle, 2013; Perugini, Gallucci, Presaghi, & Ercolani, 2003). Reciprocity was measured in the employees, not in the leader. The questionnaire consists of two dimensions: negative reciprocity (three items) and positive reciprocity (three items). The items were measured on a five-point Likert scale (1 = Totally disagree, 5 = Totally agree). An example of the dimension negative reciprocity is: ‘If my supervisor offends me, I will offend my supervisor back’ (α = .86). An example item of the dimension positive reciprocity is: ‘If my supervisor does me a
favor, I am prepared to return it’ (α = .72).
Promotive and prohibitive voice. In order to measure voice of the employees, a shortened version of the questionnaire of Liang, Farh, & Farh (2012) was used. The
disagree, 5 = Totally agree). An example item of the dimension promotive voice is:
‘Proactively suggest new projects which are beneficial to the work unit’ (α = .71). An example item of the dimension prohibitive voice is: ‘Advise other colleagues against
undesirable behaviors that would hamper job performance’ (α = .70).
Demographics. Participants filled in general questions about their demographics including the potential control variables age, gender and tenure. Questions about the seniority of the team and the relationship with their leader were asked as well. Based on previous research from Winslow (2005), expertise and therefore education may relate to creativity. Since voice overlaps with creative thinking, education has been measured as a potential covariate to control for next to age, gender and tenure of the leader-member relationship. Data analysis
For the current study, only the data of the questionnaire for employees was used. The data was analysed using SPSS. The hypotheses were tested with the aid of the PROCESS procedure developed by Hayes (2013). Within PROCESS, the regression was executed through model 7.
RESULTS
between LMX and positive reciprocity (r = .33, p < .01), however the correlation between LMX and negative reciprocity appears to be not significant (r = .00, p > .05).
Testing the Hypotheses
1a. LMX is negatively related to negative reciprocity.
Table 2 and 3 show that LMX has no significant relationship with negative reciprocity (b = .08, t (170) = .61, p > .05). Hence, the hypothesis is rejected.
1b. LMX is positively related to positive reciprocity.
Table 2 and 3 show that LMX has no relationship with positive reciprocity (b = .16, t (170) = 1.19, p > .05). Therefore, the hypothesis is rejected.
2a. The negative relationship between LMX and negative reciprocity is stronger when LMX
differentiation is high rather than low.
Table 2 and 3 show that the interaction variable LMX x LMX differentiation has no significant effect on negative reciprocity (b = - .01, t (170) = - .65, p > .05). Hence, the model does not show a moderation effect, therefore the hypothesis is rejected.
2b. The positive relationship between LMX and positive reciprocity is stronger when LMX
differentiation is high rather than low.
Table 2 and 3 show that the interaction variable LMX x LMX differentiation has no significant effect on positive reciprocity (b = - .00, t (170) = - .07, p > .05). Hence, the model does not show a moderation effect, therefore hypothesis is rejected.
Hypothesis 3a. Positive reciprocity is positively related to promotive voice.
Table 2 shows that the effect of positive reciprocity on promotive voice is not significant (b = .07, t (170) = .96, p > .05). Therefore, the hypothesis is rejected.
Hypothesis 3b. Positive reciprocity is positively related to prohibitive voice.
Hypothesis 3c. Negative reciprocity is positively related to prohibitive voice.
Table 3 shows that the direct effect of positive reciprocity on prohibitive voice is not significant (b = - .03, t (170) = - .33, p > .05). Therefore, the hypothesis is rejected.
4a. The indirect relationship between LMX and promotive voice as mediated by positive
reciprocity will be positive and stronger when LMX differentiation is high rather than low.
Table 2 shows that LMX, in the model without promotive voice, does not relate to positive reciprocity. The interaction variable LMX x LMX differentiation, does not relate to positive reciprocity as well. The whole model, including the outcome variable promotive voice, is shown in Table 2. The results show that LMX has a significant positive relationship with promotive voice (b = .08, t (170) = 2.66, p < .01). However, positive reciprocity has no significant relationship in this model. Hence, the relation does not show a mediated
moderation effect, therefore the hypothesis is rejected.
4b. The indirect relationship between LMX and prohibitive voice as mediated by positive
reciprocity will be positive and stronger when LMX differentiation is high rather than low.
Table 3 shows that LMX, in the model without prohibitive voice, do not relate to positive reciprocity. The interaction variable LMX x LMX differentiation, does not relate to positive reciprocity as well. The whole model, including the outcome variable prohibitive voice, is shown in Table 3. The results show a significant positive relationship between LMX and prohibitive voice (b = .07, t (170) = 2.07, p < .05). However, positive reciprocity has no significant effect in this model. Hence, the relation does not show a mediated moderation effect, therefore the hypothesis is rejected.
4c. The indirect relationship between LMX and prohibitive voice as mediated by negative
reciprocity will be positive and stronger when LMX differentiation is high rather than low.
shown in Table 3. The results show a significant positive relationship between LMX and prohibitive voice (b = .07, t (170) = 2.07, p < .05). However, negative reciprocity has no significant effect in this model. Thus, the relation does not show a mediated moderation effect, therefore the hypothesis is rejected.
Supplementary analyses
Due to the rejection of all the hypotheses, some supplementary analyses we executed. In order to look at the variance between and within the different teams, the intraclause correlation (ICC) was calculated. The ICC indicates the variance that is explained on team level (Sommet & Morselli, 2017). The ICC score of voice and reciprocity is .32. Hence, this shows the data is not independent, therefore multilevel analyses are needed.
For the multilevel analyses, mixed model analyses were executed. Four analyses were performed in order to test the conceptual model (Table 4). Results show no further effects. Next to the conceptual model, additional relations were tested (Table 5). First, a model including LMX, LMX differentiation and both dimensions of voice was tested. No further results came forward. Secondly, results do not support the proposition that LMX
differentiation shapes the reciprocity in an employee. Therefore we tested the moderating influence of LMX differentiation on the second path in the model, suggesting that LMX differentiation strengthens the motivation to act on the formed reciprocity. Results show a marginally significant effect for LMX differentiation x Positive Reciprocity on Prohibitive voice (b = .08, t (169) = 1.92, p < .06) (Table 5).
reciprocity principle. Therefore the correlations between the three items of the contribution dimension of LMX and the items of negative and positive reciprocity were measured (Table 6). Correlations indicate that the three contribution items relate to the three positive
reciprocity items. For example ‘I do work for my supervisor that goes beyond what is
specified in my job description’ is positively related to ‘I go out of my way to help my
supervisor who has been kind to me before’ (r = .32, p < .01). The contribution items of LMX
show no relationship with the items of negative reciprocity (Table 6). Therefore, we looked at other LMX dimensions that might be important in explaining negative reciprocity. Affect appears to relate significantly to negative reciprocity. For example, ‘If my supervisor puts me
in a difficult position, I will do the same to my supervisor’ shows a negative relation with ‘I like my supervisor very much as a person’ (r = - .23, p < .01) (Table 6).
DISCUSSION
The current study examined the influence of LMX (differentiation) and reciprocity on the both dimensions of voice. Moreover, the aim of this study was to clarify existing issues in this relationship. Firstly, the influence of a leader on the voice of an employee is widely acknowledged, however, the explanation of this relationship is not clear. The current study looked into the reciprocity principle as an explanation for the use of voice in employees. Secondly, in line with recent developments in the voice literature, the current study looked at both promotive and prohibitive voice in a leader-employee context. Attention was attributed to the motivation that leads employees to voice in either a promotive or prohibitive way. Based on the results, all the proposed hypotheses were rejected. Therefore the study suggests that both prohibitive and promotive voice are not explained by the proposed moderated mediation model including LMX (differentiation) and the reciprocity principle.
effects were found. First, a negative effect of negative reciprocity on promotive voice was found. This result suggests that negative reciprocity inhibits the promotive voice in
employees. Second, a positive direct effect of LMX on both promotive and prohibitive voice was found. These significant results suggest that an increase in the LMX relationship, causes employees to voice in a promotive and prohibitive way more. Lastly, a supplementary analysis showed a marginally significant interaction effect of positive reciprocity and LMX differentiation on prohibitive voice, which indicates that the negative relationship between positive reciprocity and prohibitive voice is stronger when LMX differentiation is low rather than high.
Theoretical implications
Although the study did not found support for the expected relationship between LMX and voice, it provides several important implications.
First, the results showed a significant positive relationship between LMX and both dimensions of voice. Even though the mediating effect of reciprocity was not found in this relationship, the findings strengthen previous studies that state the important influence of LMX on both the dimensions of voice (Chamberlin et al., 2017). It could be the that a high quality LMX is a boundary condition to voice, and the subsequent reciprocity principle does not explain the expression of the employees. Moreover, research of Chamberlin et al. (2017) argues that felt responsibility and the level of engagement of the employee mediates the relationship between LMX and voice in general. Most notable, is the positive effect of LMX on prohibitive voice, suggesting that a high quality LMX relationship induces prohibitive voice. This is consistent with research from Wei et al., 2015 that argues that a high quality relationship is needed, because of the high risk of voicing in a prohibitive way.
Second, the additional negative effect found between negative reciprocity and
We assumed in the current study that reciprocity leads to a certain kind of positive or negative behavior or attitude. However, this result indicates that reciprocity may not be that
straightforward as argued in the theory, and can also lead to an inhibition of ideas and
contributions. In line with this finding, research from Burris et al. (2008) argues that voice can be constrained when employees feel disconnected from their colleagues. Hence, this result contributes to the reciprocity literature by showing that reciprocity may also inhibit behavior. More specifically, next to negative reciprocity principle of harming those who have harmed them (Gouldner, 1960), it may be the case that the negative reciprocity motivates to act less positive.
of employees in a team with low variability might even be inhibited more, when acting on positive reciprocity, compared to a team with a high variability in LMX relationships. Limitations and future research
Despite various desirable features of the study (i.e. clear definitions of key variables, reliable measurements, heterogenous sample), there are several notable methodological limitations. First, the sample size of 33 teams consisting of 180 employees in total is quite small. Due to this sample size, caution must be applied in drawing conclusions based on the results of this study. Moreover, it declines the generalizability of the study. Second, the threshold of the teams was set on a minimum of two employees per team. To compare and examine variety within and between teams, two employees might be too little. Previous research argues to use teams of at least three persons (Henderson et al., 2008). The current study chose to use teams of two persons, because the sample size would decrease even more with a threshold of three persons. Thirdly, participants were collected through the use of personal network, therefore a lot of acquaintances participated in the study. This might have increased the possibility of social desirability when answering the questionnaires. Fourthly, the questionnaires were conducted in very divergent industries. In particular the construct voice may be influenced by the different fields. For example, a production environment is more focused on safety and, thus possible harm and wrongful events than for example education. This could imply that more prohibitive voice would occur in a plant compared to education. Therefore, type of industry, sample size, and team size are important factors to take into account regarding future research. Lastly, the current study used the employees’
perception to measure the LMX. As LMX reflects the dyadic relationship between leader and employee, both side’s perception should be measured to have a complete image of the
perception into account as well. Next to the dyadic aspect of LMX, the current study
described and measured LMX as an static construct. However, the leader-member exchange embodies a dynamic and ongoing relationship (Schyns & Day, 2010). Moreover, LMX differentiation includes the variability of dynamic relationships in the team. The possibility of sudden changes in these exchange relationships are not taken into account. Therefore, these static measures of LMX may be an explanation for the unsupported results of the interaction between LMX and LMX differentiation. Moreover, it might explain the unsupported
mediation effect of reciprocity, as reciprocity derives from the continuous exchange relationship (Blau 1964; Gouldner, 1960). Future research could measure the dynamic characteristic of LMX in a longitudinal design with frequent measurements. Next to the longitudinal design, future research should focus more on the four different dimensions of LMX described by Liden & Maslyn (1998). Table 6 provides indications that some dimensions relate more to either negative or positive reciprocity. For example, the affect dimension relates the most to negative reciprocity. Positive reciprocity, on the other hand, shows a positive relationship with the items of the contribution dimension (Table 6). These items contain the reciprocity principle (to help those who have helped them; Goulder, 1960) the most. Liden and Maslyn (1998) state that the dynamic levels of affect, contribution, professional respect and loyalty together determine the quality of the LMX relationship. Focusing more on the separate dimensions might help future research explain the proposed relationship between LMX and both positive and negative reciprocity.
dimensions. Before researching the both dimensions of voice more, new improved measures should be developed first. As of today, there is a large number of items available to create measures of prohibitive and promotive voice with greater differentiation (Chamberlin et al., 2017). Improved measures could increase the possibility of important differences between the dimensions to surface. Moreover, the process of improving the measures can help future research especially in revaluating prohibitive voice. As argued in the current study, it is not clear how prohibitive voice should be interpreted. The majority of the voice literature, regards prohibitive voice as a constructive construct (Chamberlin et al., 2017), however there are indication that it can be perceived negatively as well (Liang et al., 2012). Moreover, the current study shows a prudent negative relationship between positive reciprocity and
prohibitive voice that supports this inconsistency. Hence, future research is needed to examine prohibitive voice in depth. If future research confirms that there exists an inconsistency in the current construct of prohibitive voice, maybe an additional dimension of voice is needed in order to bring clarity.
Secondly, potential moderators could help explain the proposed relationship between LMX and both dimensions of voice. As argued above, there may exist a risky element in the expression of prohibitive voice (Wei et al., 2015). Accordingly, previous research of Morrison (2014) shows that psychological safety is important for employees to express themselves. Future research is needed to examine if psychological safety may help explain the relationship between reciprocity and both dimensions of voice. For example, a feeling of psychological safety may motivate the employee more to address the status quo or express creative ideas. A feeling of safety could be especially important for the expression of prohibitive voice, and may give the extra push to express risky opinions and comments.
Another possible moderator this study proposes is the regulatory focus of the
it might have in explaining the two dimensions of voice. There are several known factors that influence voice in general, but there is a missing link that explains why an employee chooses to voice in either a promotive or prohibitive way. The regulatory focus theory of Higgins (1997) might be the missing moderator that explains this difference (Chamberlin et al., 2017; Liang et al., 2012). The regulatory focus is a strategy that influences how a person approaches a desired end-state or goal, which consists of two foci: the promotion and the prevention focus (Higgins, 1997). The promotion focus supports the completion of tasks through an approach strategy and is concerned with attaining maximal goals. On the contrary, the prevention focus supports strict task behavior and it focuses on avoiding losses. From a motivational angle, employees with a promotion focus are eager to use different strategies to approach maximal goals. Employees with a prevention focus act vigilant in attaining goals and will avoid making mistakes (Higgins 1997). In line with this theory, a prevention focus can be related to
prevention focus and a promotion focus can be associated with a higher level of promotive voice (Lin & Johnson, 2015). Hence, the current study recommends to further research the influence of the regulatory focus in explaining the relationship between LMX and voice. Practical implications
prohibitive and promotive voice in employees. Leaders should continually invest in the relationships with their employees in order to benefit from contributions in terms of ideas and opinions. Therefore leaders should contribute to the relationship with their employees beyond their job description. This knowledge is of great importance for organizations, because the contributions of their employees help the organization innovate and thrive (Detert & Burris, 2007). Organizations may use this information by paying more attention to the quality of the relationships of their leaders and employees, and by making leaders aware of the influence they have on their employees behavior.
CONCLUSION
In order to survive in this dynamic and agile environment, organizations need ideas and opinions to develop. Employees are the most obvious source to draw from, and therefore it is interesting to know more about the context in which employees share their ideas and opinions with their leaders. Hence, the current study examined the influence of LMX (differentiation) and reciprocity on both the prohibitive and promotive voice of employees. The aim of this study was to clarify existing issues in this relationship. The proposed model was not supported by the results, however findings support the widely acknowledged importance of the LMX relationship on both prohibitive and promotive voice in employees. The additional effect found between negative reciprocity and promotive voice, provides more insight in the inhibiting effect reciprocity might have on voice behavior. Most importantly the current study addresses the importance to create more clarity about the dimension prohibitive voice. In order to further shape and develop the voice literature more attention to the
REFERENCES
Bernerth, J. B., Walker, H. J., & Harris, S. G. 2016. Rethinking the benefits and pitfalls of leader–member exchange: A reciprocity versus self-protection perspective. Human Relations, 69(3): 661–684.
Blau, P. M. 1964. Exchange and Power in Social Life. New York, NY: Wiley, Inc.
Burris, E. R., Detert, J. R., & Chiaburu, D. S. 2008. Quitting before leaving: The mediating effects of psychological attachment and detachment on voice. Journal of Applied Psychology, 93(4): 912–922.
Caliendo, M., Fossen, F., & Kritikos, A. 2012. Trust, positive reciprocity, and negative reciprocity: Do these traits impact entrepreneurial dynamics?. Journal of Economic Psychology, 33(2): 394-409.
Chamberlin, M., Newton, D. W., & Lepine, J. A. 2017. A Meta-Analysis of Voice and Its Promotive and Prohibitive Forms: Identification of Key Associations, Distinctions, and Future Research Directions. Personnel Psychology, 70(1): 11–71.
Cogliser, C. C., & Schriesheim, C. A. 2000. Exploring work unit context and leader-member exchange: a multi-level perspective. Journal of Organizational Behavior, 21(5): 487– 511.
DeNisi, A. S., & Murphy, K. R. 2017. Performance appraisal and performance management: 100 years of progress?. Journal of Applied Psychology, 102(3): 421.
Dienesch, R. M., & Liden, R. C. 1986. Leader-member exchange model of leadership: A critique and further development. Academy of management review, 11(3): 618-634.
Dulebohn, J. H., Bommer, W. H., Liden, R. C., Brouer, R. L., & Ferris, G. R. 2012. A Meta-Analysis of Antecedents and Consequences of Leader-Member Exchange. Journal of Management, 38(6): 1715–1759.
Dyne, L. Van, Ang, S., & Botero, I. C. 2003. Conceptualizing Employee Silence and Employee Voice as Multidimensional Constructs. Journal of Management Studies, 40(6): 1359–1392.
Egloff, B., Richter, D., & Schmukle, S. C. 2013. Need for conclusive evidence that positive and negative reciprocity are unrelated. Proceedings of the National Academy of Sciences, 110: 786.
Erdogan, B., & Liden, R. C. (2002). Social exchanges in the workplace. Leadership, 65-114.
Farh, C. I., & Chen, Z. 2014. Beyond the individual victim: Multilevel consequences of abusive supervision in teams. Journal of Applied Psychology, 99(6): 1074.
Farh, J. L., Zhong, C. B., & Organ, D. W. 2004. Organizational Citizenship Behavior in the People’s Republic of China. Organization Science, 15(2): 241–253.
Gouldner, A.W. 1960. The norm of reciprocity. American Sociological Review, 25, 165-167.
Graen, G., & Uhl-Bien, M. 1995. Relationship-based approach to leadership: Development of Leader–Member Exchange (LMX) theory of leadership over 25 years: Applying a multi-level multi-domain perspective. The leadership quarterly, 6(2): 219-247.
Graham, J. W. 1986. Principled organizational dissent: A theoretical essay. Research in Organizational Behavior, 8: 1–52.
Hayes, A. F. (2013). Introduction to Mediation, Moderation, and Conditional Process
Analysis: A Regression-Based Approach. New York, NY: The Guilford Press. Journal of Educational Measurement, 51(3): 335–337.
Henderson, D. J., Liden, R. C., Glibkowski, B. C., & Chaudhry, A. 2009. LMX differentiation: A multilevel review and examination of its antecedents and outcomes. The leadership quarterly, 20(4): 517-534.
Henderson, D. J., Wayne, S. J., Shore, L. M., Bommer, W. H., & Tetrick, L. E. 2008. Leader--member exchange, differentiation, and psychological contract fulfillment: A multilevel examination. Journal of applied psychology, 93(6): 1208.
Higgins, E. T. 1997. Beyond pleasure and pain. American Psychologist, 52: 1280-1300.
Liang, J., Farh, C. I. C., & Farh, J. L. 2012. Psychological Antecedents of Promotive and Prohibitive Voice: A Two-Wave Examination. Academy of Management Journal, 55(1): 71–92.
Liden, R. C., Erdogan, B., Wayne, S. J., & Sparrowe, R. T. 2006. Leader-member exchange, differentiation, and task interdependence: implications for individual and group
performance. Journal of Organizational Behavior, 27(6): 723–746.
Liden, R. C., & Maslyn, J. M. 1998. Multidimensionafity of Leader-Member Exchange: An Empirical Assessment through Scale Development. Journal of Management, 24(1): 43– 72.
Lin, S., & Johnson, R. E. 2015. A suggestion to improve a day keeps your depletion away: Examining promotive and prohibitive voice behaviors within a regulatory focus and ego depletion framework. Journal of Applied Psychology, 100(5): 1381-1397.
Martin, R., Thomas, G., Legood, A., & Dello Russo, S. 2018. Leader-member exchange (LMX) differentiation and work outcomes: Conceptual clarification and critical review. Journal of Organizational Behavior, 39(2): 151–168.
Masterson, S. S., Lewis, K., Goldman, B. M., & Taylor, M. S. 2000. Integrating Justice and Social Exchange: The Differing Effects of Fair Procedures and Treatment on Work Relationships. Academy of Management Journal, 43(4): 738–748.
Miceli, M. P., & Near, J. P. 1985. Characteristics of organizational climate and perceived wrongdoing associated with whistle blowing decisions. Personnel Psychology, 38(3): 525–544.
Morrison, E. W. 2014. Employee Voice and Silence. Annual Review of Organizational Psychology and Organizational Behavior, 1(1): 173–197.
Schyns, B., & Day, D. 2010. Critique and review of leader–member exchange theory: Issues of agreement, consensus, and excellence. European Journal of Work and
Organizational Psychology, 19(1): 1–29.
Seo, Y., & Lee, J. Y. 2017. Leader–member exchange level and differentiation: The roles of interpersonal justice climate and group affective tone. Social Behavior and Personality:
an international journal, 45(7): 1175-1186.
Settoon, R. P., Bennett, N., & Liden, R. C. 1996. Social exchange in organizations: Perceived organizational support, leader–member exchange, and employee reciprocity. Journal of Applied Psychology, 81: 219-227.
Sheppard, B., Lewicki, R., & Minton, J. 1992. Organizational justice: The search for fairness in the workplace. Lexington Books/Macmillan.
Sommet, N., & Morselli, D. 2017. Keep Calm and Learn Multilevel Logistic Modeling: A Simplified Three-Step Procedure Using Stata, R, Mplus, and SPSS. International Review of Social Psychology, 30(1): 203–218.
Sparrowe, R. T., & Liden, R. C. 1997. Process and Structure in Leader-Member Exchange. Academy of Management Review, 22(2): 522–552.
Uhl-Bien, M., & Maslyn, J. M. 2003. Reciprocity in Manager-Subordinate Relationships: Components, Configurations, and Outcomes. Journal of Management, 29(4): 511–532.
Wang, Q., Weng, Q., McElroy, J. C, & Ashkanasy, N. M., & Lievens, F. 2014. Organizational career growth and subsequent voice behavior: The role of affective commitment and gender. Journal of vocational behavior, 84(3): 431-441.
Wayne, S. J., Shore, L. M., & Liden, R. C. 1997. Perceived Organizational Support And Leader-Member Exchange: A Social Exchange Perspective. Academy of Management Journal, 40(1): 82–111.
Wei, X., Zhang, Z., & Chen, X. 2015. I will speak up if my voice is socially desirable: A moderated mediating process of promotive versus prohibitive voice. Journal of Applied Psychology, 100(5): 1641-1652.
Winslow, B. 2005. Developing creativity, motivation, and self-actualization with learning systems. International Journal of Human-Computer Studies, 63(4-5): 436-451.
APPENDIX A: QUESTIONNAIRE
Leader-Member Exchange (LMX)
1. I respect my supervisor’s knowledge of and competence on the job 2. I admire my supervisor’s professional skills.
3. I am impressed with my supervisor’s knowledge of his/her job.
4. My supervisor would defend me to others in the organization if I made an honest mistake. 5. My supervisor would come to my defense if I were “attacked” by others.
6. My supervisor defends (would defend) my work actions to a superior, even without complete knowledge of the issue in question.
7. My supervisor is the kind of person one would like to have as a friend. 8. I like my supervisor very much as a person.
9. My supervisor is a lot of fun to work with.
10. I do not mind working my hardest for my supervisor.
11. I do work for my supervisor that goes beyond what is expected of me in my job. 12. I am willing to apply extra efforts, beyond those normally required, to meet my supervisor’s work goals.
Negative reciprocity
1. If my supervisor causes me to suffer a serious wrong, I will take revenge as soon as possible, no matter what the cost.
2. If my supervisor puts me in a difficult position, I will do the same to my supervisor. 3. If my supervisor offends me, I will offend my supervisor back.
Positive reciprocity
1. If my supervisor does me a favor, I am prepared to return it.
3. I am ready to undergo personal costs to help my supervisor who helped me before.
Promotive voice
1. I proactively develop and make suggestions for issues that may influence the unit. 2. I proactively suggest new projects which are beneficial to the work unit.
3. I raise suggestions to improve the unit’s working procedure.
Prohibitive voice
1. I advise other colleagues against undesirable behaviors that would hamper job performance.
2. I speak up honestly with problems that might cause serious loss to the work unit, even when dissenting opinions exist.
3. I dare to voice out opinions on things that might affect efficiency in the work unit, even if that would embarrass others.
Demographics
1. How old are you? 2. What is your gender?
APPENDIX B
Table 1
Correlations, Means and Standard deviations (N = 180)
Table 2
Results of the PROCESS model for Promotive Voice (N= 178)
Effect on Negative Reciprocity
b SE t Covariates: Gender - .39 .31 -1.26 Age .00 .01 .27 Tenure Leader-Employee - .03 .06 - .54 Education .19 .21 .92 Independent Variables: LMX .08 .13 .61 LMX-D - .02 .09 - .16 Interaction LMX x LMX-D - .01 .02 - .65
Effect on Positive Reciprocity
b SE t Covariates: Gender .43 .32 1.36 Age - .03 .01 - 2.01 Tenure Leader-Employee - .01 .06 - .20 Education .36 .21 1.72 Independent Variables LMX .15 .13 1.19 LMX-D .00 .09 .01 Interaction LMX x LMX-D - .00 .02 - .07
Effect on Promotive voice
b SE t Covariates: Gender .05 .30 .16 Age .00 .01 .58 Tenure Leader-Employee .08 .05 1.16 Education .67** .20 3.42 Mediation: Negative Reciprocity - .17* .08 - 2.30 Positive Reciprocity .07 .07 .96 Independent Variables LMX .08** .03 2.66
Indirect effect of Reciprocity on Promotive voice
Independent variable: b SE Lower limit
Table 3
Results of the PROCESS model for Prohibitive Voice (N= 178)
Effect on Negative Reciprocity
b SE t Covariates: Gender - .39 .31 -1.26 Age .00 .01 .27 Tenure Leader-Employee - .03 .06 - .54 Education .19 .21 .92 Independent Variables: LMX .08 .13 .61 LMX-D - .02 .09 - .16 Interaction LMX x LMX-D - .01 .02 - .65
Effect on Positive Reciprocity
b SE t Covariates: Gender .43 .32 1.36 Age - .03 .01 - 2.01 Tenure Leader-Employee - .01 .06 - .20 Education .36 .21 1.72 Independent Variables LMX .15 .13 1.19 LMX-D .00 .09 .01 Interaction LMX x LMX-D - .00 .02 - .07
Effect on Prohibitive voice
b SE t Covariates: Gender - .30 .33 - .90 Age .01 .01 .36 Tenure Leader-Employee .14* .06 2.42 Education - .16 .21 - .76 Mediation: Negative Reciprocity - .03 .08 - .33 Positive Reciprocity .06 .08 .72 Independent Variables LMX .07* .03 2.07
Indirect effect of Reciprocity on Prohibitive voice
Independent variable: b SE Lower limit
Table 4
Results of the Mixed model analysis (subject: Leader) Effect on Negative Reciprocity
b SE t Covariates Gender - .35 .32 - 1.10 Age .00 .01 .14 Tenure - .03 .06 - .47 Education .19 .21 .86 Independent variables LMX .07 .13 .58 LMX-D - .02 .10 - .19 Interaction LMX x LMX-D - .01 .02 - .61
Effect on Positive Reciprocity
b SE t Covariates Gender .43 .32 1.36 Age - .03* .01 - 2.01 Tenure - .01 .06 - .20 Education .36 .21 1.72 Independent variables LMX .16 .13 1.19 LMX-D .00 .09 .01 Interaction LMX x LMX-D - .00 .02 - .07
Effect on Promotive Voice
b SE t Covariates Gender .05 .30 .16 Age .01 .01 .56 Tenure .08 .05 1.61 Education .67** .20 3.40 Mediation Positive reciprocity .07 .07 .96 Negative reciprocity - .17* .08 - 2.29 Independent variables LMX .08** .03 2.67
Effect on Prohibitive Voice
Table 5
Results of additional analyses through Mixed models (subject: Leader) Direct effect on Promotive voice
b SE t Covariates Gender .13 .31 .41 Age .00 .01 .26 Tenure .10 .06 1.69 Education .65** .21 3.15 Independent variables LMX .22 .12 1.78 LMX-D .05 .09 .49 Interaction LMX x LMX-D - .02 .02 - 1.05
Direct effect on Prohibitive voice
b SE t Covariates Gender - .26 .34 - .79 Age .01 .01 .37 Tenure .13* .06 2.07 Education - .02 .23 - .07 Independent variables LMX - .11 .13 - .83 LMX-D .04 .11 .38 Interaction LMX x LMX-D .03 .02 1.50
Moderation second path on Promotive Voice
Table 6
Additional correlations, Means and Standard deviations (N = 180)
Items: M SD 1 2 3 4 5 6 7 8 9 10 11 12
LMX (Contribution)
1. I do not mind working my hardest for my supervisor
3.86 .89 - 2. I do work for my supervisor that goes beyond
what is expected of me in my job.
3.40 1.09 .36** - 3. I am willing to apply extra efforts, beyond
those normally required, to meet my supervisor’s work goals.
3.48 .99 .56** .59** -
LMX (Affect)
4. I like my supervisor very much as a person. 4.14 .79 .27** - .01 .20** - 5. My supervisor is the kind of person one would
like to have as a friend.
3.79 .87 .33** .13 .26** .75** -
6. My supervisor is a lot of fun to work with. 4.13 .75 .37** .01 .30** .68** .61** -
Positive reciprocity
7. If my supervisor does me a favour, I am prepared to return it.
4.07 .76 .26** .24** .16* .09 .17* .11 - 8. I go out of my way to help my supervisor who
has been kind to me before.
3.34 .97 .22** .32** .27** .07 .14 .11 .48** - 9. I am ready to undergo personal costs to help
my supervisor who helped me before.
2.85 1.03 .25** .35** .32** .12 .18* .06 .46** .49** -
Negative reciprocity
10. If my supervisor causes me to suffer a serious wrong, I will take revenge as soon as possible, no matter what the cost.
1.49 .71 - .08 .12 .09 - .19* - .09 - .12 .02 .19* .21** - 11. If my supervisor puts me in a difficult
position, I will do the same to my supervisor.
1.63 .76 - .13 .03 .00 - .10 - .23** - .20** .06 .23** .19* .72** - 12. If my supervisor offends me, I will offend my
supervisor back.
