1
ABUSIVE SUPERVISION: HOW DO SUBORDINATES RESPOND?
Master thesis, M.Sc., specialization Human Resource Management University of Groningen, Faculty of Economics and Business
February 28, 2016
ANNE DE KLEINE Studentnumber: 2844664
Heymanslaan 31a 9714 GG Groningen tel: +31(0)649967337
e-mail: a.m.de.kleine.1@student.rug.nl
Supervisor S. Täuber
University of Groningen
2 ABSTRACT
This research discusses what the main influence of both abusive supervision and distrust-triggering behaviors is on distrust. A new relation is researched because I expect distrust to be an antecedent of value incongruence and I expect distrust to result in diminished organizational citizenship behavior. This research is mainly based on a recent study written by Bijlsma-Frankema, Sitkin and Weibel (2015), who indicated once distrust is triggered it results in a self-amplifying cycle. I want to research whether abusive supervision puts this self-amplifying cycle into motion. Furthermore, the overall model is based on the social exchange theory (Blau, 1964; Gouldner, 1960), the need for revenge and avoidance of harm because of an increase in vulnerability.
The method consists of qualitative data by conducting eight interviews to retrieve a more in depth view of three hypotheses and several in-depth questions concerning the formation of and responses towards distrust.
The overall conclusion is that abusive supervisory behaviors and distrust-inducing behaviors lead to distrust and distrust leads to an increase in value incongruence. It appears that diminished organizational citizenship behavior (OCB) is not that negative as expected and I concluded a reason for that might be detachment.
Keywords: Distrust, values, abusive supervision, organizational citizenship behavior,
vulnerability
3
INTRODUCTION
Today’s dynamic market consists of competition regarding innovation, price and performance. According to scholars, leadership is the key to improving organizational performance (Zhu, Chew & Spangler, 2005) which makes researching leadership important for an organization. Hence, a crucial aspect of a successful organization is an effective leader.
Unfortunately, having an effective leader is not a matter of fact for all organizations. My research focuses on one of the aspects of negative leadership, namely abusive supervision.
Multiple scholars found negative consequences of abusive supervision, such as negative attitudes, turnover intention, lower job and life satisfaction and lower performance (Martinko, Harvey Brees & Mackey, 2013, Tepper, 2007). Tepper (2000:p178) defines abusive
supervision as “subordinates’ perceptions of the extent to which supervisors engage in the sustained display of hostile verbal and nonverbal behaviors, excluding physical contact”.
Tepper (2000) argued that the harm abusive supervision can do to subordinates should be a warning to organizations. Tepper and his co-authors calculated the cost of abusive supervision for U.S. organizations (e.g. health care costs, lost production and absence) to be 23.8 billion US dollars per year (Tepper, Duffy, Henle & Lambert, 2006), to support their argument that abusive supervision is a significant present-day problem in need of further scholarly exploration. Therefore, it is an organizational issue which should be addressed.
Furthermore, scholars have pointed out consequences of distrust for organizational relations. Research indicates that people who distrust, in this particular matter subordinates, tend to disbelieve (Kramer, 1994) those distrusted others, in this case supervisors, avoid interaction (Bijlsma-Frankema et al., 2015; March & Olsen, 1975) and are less willing to cooperate (Cho, 2006). In addition, those subordinates have a tendency to express hostility towards, (Chambers & Melnyk, 2006) and get in conflict (Fiol, Pratt, & O’Connor, 2009) with the supervisors. Given the negative consequences of both abusive supervision and distrust in supervisors, exploring the relationship between these two previously unconnected phenomena appears to be worthwhile.
Little research has been done concerning distrust. A reason for this is that there was no model available concerning a distrust cycle. Bijlsma-Frankema et al. (2015) proposed such a distrust cycle and I will research how this distrust cycle was set into motion. Therefore, I will partly address this gap in distrust literature.
I aim to investigate three matters related to abusive supervision and distrust. Firstly,
my research aims to investigate whether abusive supervisor behavior promotes distrust, and
with that, as Bijlsma-Frankema, Sitkin and Weibel (2015) argue, sets into motion a self-
4
amplifying cycle of distrust of which distrust, negative behavioral reciprocity, negative attributions and perceived value incongruence are the elements. In my study I want to focus on how distrust is engendered by testing whether distrust is promoted by abusive supervision.
Distrust can be defined in this context as subordinates’ unwillingness to accept vulnerability based on prevalent negative perceptions of the other person’s motives, intentions or behaviors (Bijlsma-Frankema et al., 2015). Unfortunately, there has been little research concerning which specific behaviors lead to distrust. Bijlsma-Frankema et al. (2015) indicate that distrust emerges when actors’ feelings of vulnerability are so high that they become unacceptable to the actor. Their study did not elaborate on the behaviors that may stir these feelings of unacceptable vulnerability. Abusive supervision, defined as the supervisor continuously showing adverse verbal and nonverbal behavior towards the subordinates (Tepper, 2000), may stir feelings of vulnerability that surpass the level an actor may be willing to accept, thus provoke distrust of the supervisor.
A further exploration of the relation between abusive supervision and distrust of the supervisor is relevant for organizations. Answering the question ‘Which of the aspects of abusive supervision is strongly related to distrust?’ will give a more concrete answer to how to prevent distrust, distrust development and its negative consequences, to arise.
A second matter regards the role of value incongruence perceptions (of subordinates regarding their values and those of the supervisor) about and its relation to distrust of the supervisor within the self-amplifying cycle of distrust development. So far, value
incongruence has been proposed in the distrust literature as antecedent of distrust (Sitkin and colleagues), but the cyclical model of Bijlsma-Frankema et al. (2015) suggests that value incongruence is also a consequence of distrust. In my study, I will go into depth concerning the matter whether value incongruence perceptions of subordinates are indeed a consequence of distrust. In accordance with the notion that negative cognitions, such as distrust, will be justified by subordinates by negative perceptions of or attributions to the supervisor,
including the notion that the supervisor acts from a quite different sets of values. I expect that if subordinates distrust their supervisor, they will put more emphasis on the difference in values between themselves and the supervisor.
Thirdly, I am going to research an expected consequence of distrust, namely
diminished organizational citizenship behavior, a consequence proposed based on the social exchange theory. The social exchange theory argues that when people are treated well by others, they feel a certain obligation towards that person to return the favor (Blau, 1964;
Gouldner, 1960). However, when treated in a bad way, they will react with bad behavior
5
(Gouldner, 1960). Hence, as a response to abusive supervision, increased defiant behavior towards the supervisor is expected (Mitchell & Ambrose, 2007). Tepper et al. (2002), based on the social exchange theory, tested and found that subordinates will react to abusive supervision in a negative situation by a decrease in organizational citizenship behavior (OCB), thus withholding behavior that benefits the supervisor and the organization. The social exchange theory proposes trust as a base of positive exchanges. In my study I will therefore propose distrust, if triggered by abusive supervision, will act as a mediator of the abusive supervision OCB relation. Hence, I research the direct relation between distrust and OCB.
My research model, in sum, proposes that abusive supervision increases subordinates’
distrust towards the supervisor and distrust leads to both diminished OCB towards the supervisor and more perceived value incongruence from the subordinates’ perspective (see Figure 1a for the full conceptual model). The overall research question is: What is the
influence of abusive supervision on subordinates’ distrust and how is distrust related to both value incongruence and OCB?
FIGURE 1a: Conceptual model
THEORETICAL FRAMEWORK
In the following sections, the constructs abusive supervision, distrust, value incongruence, and OCB will be discussed in detail and the hypotheses to be tested will be introduced and substantiated with relevant literature.
Abusive Supervision
This research focuses on abusive supervision and its consequences. Prior research mainly focused on how constructive leaders behave despite the common notion that
destructive leaders can have a negative impact by abusing their subordinates (Aryee, Chen, Abusive
Supervision
Distrust
Value Incongruence
OCB
6
Sun, & Debrah, 2007; Tepper, 2000, 2007; Tepper et al., 2006; Tepper, Duffy, & Shaw, 2001; Zellars, Tepper & Duffy, 2002). Abusive supervision is described as an active way of leading, it consists of e.g. yelling, ridiculing someone and letting the subordinates feel small (Skogstad, Einarsen, Torsheim, Aasland, & Hetland, 2007).
A definition of abusive supervision which I am going to use in my research is the definition given by Tepper (2000:p178), namely the “subordinates’ perceptions of the extent to which supervisors engage in the sustained display of hostile verbal and nonverbal
behaviors, excluding physical contact”. In other words, I will focus on the subordinates’ point of view concerning the amount of adverse verbal and nonverbal supervisor behavior. Tepper (2000) defined and measured abusive supervision by using fifteen items. Example items refer to subordinates’ perception of supervisors telling them that ‘their thoughts and feelings are stupid’, ‘give them the silent treatment’ or ‘invades their privacy’ (Tepper, 2000: p189).
These items are partially used to operationalize abusive supervision in this research and to analyze how abusive supervision and distrust relate.
Distrust
In my research, I expect abusive supervision to be an antecedent of distrust.
Therefore, distrust is discussed in this paragraph. For a long time, distrust was seen as equal to low trust (Hardin, 2004, Luhmann 1979, Mayer, Davis & Schoorman, 1995, Robinson 1996, Rotter 1980). I am going to distinguish trust and distrust as two separate entities similar to the research of Sitkin and Roth (1993). Recent research distinguished trust and distrust as two separate dimensions (Benamati, Serva, Fuller, 2006, Chang & Fang 2013, Cho 2006, Dimoka 2010, Komiak & Benbasat 2008, Lewicki, McAllister & Bies, 1998, Ou & Sia 2010, Sitkin & Roth 1993, Sitkin & Stickel 1996, Tomlinson & Lewicki, 2006).
An article by Sitkin and Roth (1993) was the first article to speak of trust and distrust as two separate dimensions, which led to more articles concerning distrust as a separate construct. Distrust can be defined as a subordinate’s unwillingness to accept vulnerability based on pervasive negative perceptions of the other person’s motives, intentions or behaviors (Bijlsma-Frankema et al., 2015). In the context of the subordinate – supervisor relation distrust of the supervisor may lead to doubt by the subordinates about whether their beliefs and values and those of their supervisor can co-exist. The following paragraph will discuss consequences of distrust.
Distrust has already been linked to several consequences. Bijlsma-Frankema et al.
(2015:p1) mention for instance lacking cooperation (Cho, 2006), unwillingness to share their
7
views and preferences (Bijlsma-Frankema, 2004, March & Olsen, 1975), avoiding influence (Sheppard & Tuchinsky, 1996) and/or interaction (Bies & Tripp, 1996), stigmatization (Sitkin
& Roth, 1993), paranoid behavior (Kramer, 2006), hostility towards others who are distrusted (Chambers & Melnyk, 2006), distortion and disbelief (Kramer, 1994), uncontrollable
intergroup conflicts (Fiol, Pratt & Connor, 2009; Tomlinson & Lewicki, 2006) and trying to avoid interaction (Bies & Tripp, 1996). In the following section I am going to discuss subordinates’ perceived value incongruence, as I expect it to be a consequence of subordinates’ distrust towards supervisors.
Value incongruence
Values guide behavioral choices (Jehn, Chadwick & Thatcher, 1997). They can be described as core beliefs of individuals regarding behavioral choices (Enz, 1988; Rokeach, 1973). According to Jehn (1994) and Pelled (1996), the degree to which a group has similar values within an organization influences the amount of conflict within that group. My
research focuses on the perceived difference in values between subordinates and supervisors, namely value incongruence. In its strongest form, it may concern the disbelief that values of the organization or the supervisor and values of the employees can co-exist (Kristof, 1996).
Bijlsma-Frankema et al. (2015) describe value incongruence as the belief that other individuals do not accept the individuals’ core values and stick to values that are seen as opposing these core values. Examples of value incongruence are that persons have different ideas concerning for example achievement, honesty, and helping others (Naus, Iterson &
Roe, 2007). According to Bijlsma-Frankema et al. (2015). Organization studies indicates that value incongruence can provoke conflict (Bijlsma-Frankema et al., 2015) between for
example professionals and organizations (Sorensen & Sorensen, 1974).
Organizational Citizenship Behavior (OCB)
I expect subordinates’ to respond to distrust with diminished OCB. In this paragraph I will discuss OCB. OCB can be defined as ‘individual behavior that is discretionary not directly or explicitly recognized by the formal reward system, and in the aggregate promotes the efficient and effective functioning of the organization’ (Organ, Podsakoff & Mackenzie, 2005: p 3). Hence, the subordinate adds something extra to the organization on top of the fundamental job description.
There are two subcategories of OCB which are important for this research, namely
OCBI (Organizational citizenship behavior – individual) and OCBO (organizational
8
citizenship behavior- organization). OCBI refers to the helping behaviors of the subordinate directed towards supervisors and OCBO are behaviors by which the subordinate helps the organization, though according to Wech (2002) both behaviors can, in the end, benefit the supervisor. Therefore, I expect OCB in both forms to decrease when abusive supervisory behaviors occur since this could be the subordinates’ way to take revenge towards both the organization and supervisor. In the following sections, I will substantiate why I am expecting a relation between the previously discussed concepts.
FIGURE 1b: Research model
The expectation that abusive supervision results into distrust and distrust results in diminished OCB is based on the social exchange theory. The social exchange theory explains that when people are treated in a good way by someone else, they will feel an obligation towards that person to return that favor (Blau, 1964; Gouldner, 1960). Hence, the social exchange theory substantiates that when a subordinate is not treated well, it is expected that the subordinate will not favor a supervisor and might treat the supervisor in a bad manner. I expect distrust to be an explanatory factor because both fairness and trust are critical aspects of the social exchange theory (Konovsky & Pugh, 1994). Therefore, two direct relations of this research are drawn from the social exchange theory, namely the relation between abusive supervision and distrust and, the relation between distrust and OCB.
Abusive supervision in relation to distrust
In this section, I will discuss the direct relation between abusive supervision and distrust. I want to examine how abusive supervision and distrust are related, and which abusive supervisory behaviors trigger distrust. Abusive supervision is often experienced as harmful, which possibly increases the subordinates’ feeling of vulnerability. According to Tomlinson and Lewicki (2006), distrust arises when the expectation of harm is higher than
Abusive Supervision
Distrust
Value Incongruence
OCB
9
the level of vulnerability acceptable to the subordinate. When this is the case, the subordinate will no longer accept the harmful behavior. It is proven that abusive supervision leads to harmful effects such as lower performance, psychological distress, aggression and deviance (Martinko et al., 2013; Tepper, 2007).
As discussed in the introduction, I will look at the specific negative behaviors that seem to trigger distrust and examine which abusive behaviors are most promising in the prediction of distrust. Distrust is said to develop in a self-amplifying cycle (Fox, 1974; Harvey, White, Hood & Sherif, 1961; Sitkin and Stickel, 1996; Zand, 1972) and I want to know which behaviors set this cycle into motion.
The abusive behaviors are measured by a scale designed by Tepper (2000) that covers multiple negative behaviors to see which behavior invokes distrust. I expect that abusive supervision will cause fear amongst subordinates and, as a reaction, those subordinates will distrust their supervisor. Hence, this research will examine which abusive supervisory behaviors trigger distrust. Therefore, the first hypothesis is:
Hypothesis 1: Abusive supervision by supervisors is positively related to subordinates’
distrust of supervisor.
In the following section I will discuss distrust in relation to OCB.
Distrust in relation to OCB
A response to distrust is that the subordinate can try to avoid interaction and in that way try to prevent greater vulnerability resulting from the supervisor’s actions and avoid unpleasant meetings (Bijlsma-Frankema et al., 2015). The social interactionist approach as described by Bijlsma-Frankema et al. (2015) indicates that when one group distrusts the other, they interpret their actions negatively, which justifies negative actions towards the other group. Hence, distrust developing negative reciprocal behaviors becomes more harmful.
In this context it would mean that when the supervisor is distrusted by the subordinate, it justifies the negative future actions the subordinate will carry out towards his supervisor, which results in negative reciprocity. Negative reciprocity between subordinate and
supervisor can be described as perceptions which become increasingly negative amongst one another. This results in accumulating negative actions and counteractions (Bijlsma-Frankema et al., 2015). In this case I expect it to lead to subordinates’ diminished OCB as a response to subordinates’ distrust of the supervisor.
Furthermore, according to the self-amplifying circle by Bijlsma-Frankema et al.
(2015) distrust causes diminished cooperation. Motowidlo (2000) and Organ (1997) provide a
10
modern definition of OCB and indicate that OCB might be redefined as being cooperative and helpful in and organization. Therefore, I believe that distrust leads to diminished OCB as well.
Hypothesis 2: subordinates’ distrust of supervisor is negatively related to subordinates’
OCB
Distrust in relation to value incongruence
Literature concerning distrust suggests (Sitkin & Roth 1993, Sitkin & Stickel 1996, Tomlinson & Lewicki 2006) and shows (Chambers & Melnyk 2006) that when there is perceived value incongruence, distrust arises. Furthermore, the article written by Bijlsma- Frankema et al. (2015) researches value incongruence as an antecedent of distrust. But as the model designed by Bijlsma-Frankema et al (2015) is a cyclical model, it suggests that value incongruence is also a consequence of distrust. As mentioned in the introduction, I will investigate the latter relation because I expect that when distrust occurs, the subordinates will have a negative idea concerning the supervisors’ intentions, motives and behaviors (Bijlsma- Frankema et al., 2015) and as a result I expect them to become more aware of the difference in values between them and the supervisor. Hence, I believe the subordinates’ perception of value incongruence will increase as a consequence of increasing distrust and therefore my research will focus on distrust as an antecedent of value incongruence.
Hypothesis 3: subordinates’ distrust of supervisor is positively related to perceived value incongruence by subordinates.
In the following section, I will discuss what I have done to research these hypotheses and what resulted from the interviews.
METHODOLOGY Method
The purpose of this study was to look at constructs of the overall research model in depth and test the hypotheses. Because there has not been a lot of research concerning distrust, I wanted to find new ideas which could contribute to future research. I used the following questions to create new ideas for future research.
The questions which I aimed to answer were:
- To what extent do respondents agree with the definition of distrust?
- How does distrust emerge?
- How do people respond to distrust?
11
- What are consequences of distrust? To further analyze the consequences, I asked whether the respondents experienced a difference between certain constructs before and after experiencing distrust.
Furthermore, the hypotheses tested were:
Hypothesis 1: Abusive supervision by supervisors is positively related to subordinates distrust of the supervisor.
Hypothesis 2: Subordinates’ distrust of supervisor is negatively related to subordinates organizational citizenship behavior.
Hypothesis 3: Subordinates’ distrust of supervisor is positively related to perceived value incongruence by subordinates.
Participants and Procedure
The research question was further explored by conducting eight interviews on distrust.
The interview questions can be found in Appendix B. The first part of the interviews was non-directive and open. I asked open questions and let the interviewees tell their story. The second part was semi-structured as I gave the interviewees a scheme which they had to fill out and then I asked questions afterwards (See Appendix C). The interviewees were told beforehand that the interview would be anonymous and names would be deleted from the transcripts. The interviewees were asked questions concerning their job (See Appendix B) and they were given a list of items on which they should indicate whether or not they
experienced these items (See Appendix C). The interviews were conducted with persons who all indicated that they experienced distrust towards their supervisor. The duration of each interview was about one hour to one and a half hour. Afterwards, the data was transcribed and coded into tables in order to select the main information from the story told by the
respondents. This was done because this way the answers could be structurally reported in the results section and answers could be compared.
Measures
Abusive supervision. I measured abusive supervision with items based on the scale
Tepper (2000) developed (See Appendix A) and with the items uncovered in a pilot study
which I conducted. Based on findings of the pilot study I conducted, I discovered that the
Tepper (2000) items were less useful than expected because they not meant for conceptual
purposes. Therefore I found items designed by Bies and Tripp (1996) which were meant as a
category system which is more appropriate for this research. The categories by Bies and
12
Tripp (1996) were described as ‘violation of trust/triggers of distrust’. Hence, the abusive behaviors designed by Tepper (2000) and the additional items found in the pilot study fit this category because those items also trigger distrust. Using these items made it easier to
structure the distrust-inducing items as found in the pilot study while, simultaneously, a relation between abusive supervision and distrust could be seen. The relation can be explained by linking the distrust-triggering items found in the pilot study to the categories designed by Bies and Tripp (1996). Those items of abusive supervision and additional items together form distrust-triggering items. When those distrust-triggering items fitted in the items of Bies and Tripp (1996), became triggers of distrust which therefore substantiates the relation between abusive supervision and distrust.
All of the categories of Bies and Tripp (1996) can be found in Appendix G. Not all categories were used to subdivide the items designed by Tepper (2000). The categories relevant for my research were:
- Broken Promises
- Disclosure of confidences and secrets - Abusive Authority – Intolerable Boss - Stealing of ideas or credit from others - Lying
- Insult to self or collective - Shirking of job responsibilities
- A damaged “identity” - Public criticism
A category which was added so the categories would cover all ‘Tepper (2000) items’ and the additions made in the pilot study was: ‘Unfair/unjust treatment of respondents or
respondent’s colleagues’ which was seen as a trigger of distrust. The respondents were asked to fill out a table in which they could indicate whether they experienced the distrust triggering items as found in the pilot study or not. When they indicated that they did, they were asked to explain or give examples.
Distrust. I measured distrust with eight items (See Appendix C) that were also put in a
table for the interviewees. What should be mentioned is something which was striking in the
beginning of the analysis, namely that the interviewees told almost similar stories when
answering questions about abusive supervision and distrust. It turned out that the
interviewees told the stories which resulted in distrust when answering questions about
distrust. Distrust is the fear that something happens in the future and the situations they
13
described were the reason they distrusted their supervisor. Hence, there was overlap between those stories which was adapted but should still be noted.
Value Incongruence. Furthermore, the variable value incongruence was measured by presenting the respondents a figure in which two circles were shown with a various distances between the circles (See Figure 3a). I asked the respondents how distanced their own values and their supervisor’s values were before and after they experienced distrust. The first row indicates the amount of value incongruence before distrust, where the circles are far apart it indicates value incongruence and where the circles are closest it indicates value congruence.
The second row refers to the period after they started experiencing distrust towards their supervisor.
OCB. In order to analyze OCB, I asked the respondents an open question. I asked them whether they did anything extra for the organization or the supervisor, on top of performing their usual tasks. Their answers were subdivided in two categories, OCBO (Organizational citizenship behavior – organization) and OCBI (Organizational citizenship behavior – individual).
FIGURE 3a: Value incongruence
RESULTS
In the following section the answers and the examples of respondents are discussed. In
order to write the results in a logical order, the sequence as indicated in the research model
(Figure 1b) was used.
14 Abusive Supervision
In this section, I will discuss the respondents’ descriptions of distrust-triggering behaviors.
Broken Promises
Six out of eight respondents indicated that the supervisor broke his/her 1 promises, referring to different specific situations, such as failing to keep a promise, for instance a respondent ending up with an extra workload despite the promise of the supervisor to hire extra personnel in order to prevent that, or recurrently failing to show up at appointments with the subordinate, refusing to ratify a promise of the owner of the firm to the respondent.
Besides breaking a promise, several supervisors tried to avoid taking responsibility for their behavior by saying they did not remember, by indicating that he was confronted with circumstances beyond his control (in case of hiring personnel), or by flatly denying the promise was made. One respondent, confronted with repeated denial of promises by her supervisor, suggested her supervisor probably tried to avoid the blame for broken promises:
“He keeps on denying because then you cannot be held accountable”. (Respondent 5, page 22).
Disclosure of confidences and secrets
For this category, six interviewees responded affirmatively. Several respondents indicated that they expected their supervisor to tell confidential information to others because the supervisor told the respondents confidential information about others as well. Therefore, they indicated that they knew they could not tell their supervisor confidential information about themselves. One respondent indicated that she doubted whether he would talk about her as well. Furthermore, one respondent mentioned he noticed that his colleague knew
something which the respondent told the supervisor privately. Therefore, the respondent was aware he could not tell his supervisor confidential information. Another respondent indicated that she thinks her supervisor does not tell personal things about her. However, she does not like what the supervisor tells her about others. Therefore, she sometimes wonders whether he tells something about her when she is not present. She said:
1