Sad Machiavellians: A New Perspective
on Job Dissatisfaction
THESIS
MIHNEA ALEXANDRU BIRZAN
International Master Programme in Business Studies
Faculty of Economics and Business
Student number: 6092209
1
stSupervisor: F. Belschak
2
ndSupervisor:
Acknowledgments
This Masters Thesis is the outcome of my interest in both organizational behavior and psychology. It is also a significant moment in my life as it announces the end of my studies. Thus, I would like to express my deepest gratitude and say thank you to all the people who inspired me throughout my studies in Romania and in the Netherlands, and provided me with knowledge, ideas and backing.
I would like to thank my supervisor, Dr. Belschak, for his guidance, useful recommendations and most importantly: patience. I would also like to thank my family and friends who always believed in me, supported me and had an encouraging and optimistic answer to all my concerns and doubts.
Last but not least, I am gratefully thankful for all the respondents who took time to fill out my questionnaire and provided the much needed data.
Table of contents Acknowledgments ... 2 Table of contents ... 3 ABSTRACT: ... 4 I Introduction ... 5 II Literature review. ... 8 2.1 Machiavellianism ... 8 2.2 Satisfaction ... 12 2.3 Trust ... 14 2.4 Perceived politics ... 16
2.5 Understanding and Perceived Control ... 18
III Hypothesis ... 19
IV Research Methodology ... 28
4.1 Methodology ... 28
4.2 Research Strategy and Instrument ... 29
4.3 Questionnaire design ... 31 4.4 Data collection ... 33 V Results ... 34 5.1 Descriptive ... 34 5.2 Reliability of scales ... 36 5.3 Correlations ... 37 5.4 Regression analysis ... 39 VI Discussion ... 45
VII Limitations and future research... 52
VIII Concluding remarks ... 54
References ... 55
Appendixes ... 65
Appendix 1 ... 66
ABSTRACT:
The thesis is an attempt to fill a gap in the organizational behavior research. More specifically, it attempts to explain why a personality trait such as Machiavellianism is a frequent determinant of job dissatisfaction. For this task, a conceptual model was created that brings together streams of research relevant in organizational behavior such as: personality constructs, organizational politics and job attitudinal outcomes. After a thorough analysis of results from 87 respondents, all of which were full-time business professionals from Romania, it was confirmed that Machiavellianism leads to job dissatisfaction. Moreover, trust and perceived politics play a fundamental part in this relationship and they are the key to understanding the dynamic.
I Introduction
In the last decade, employee satisfaction has taken center stage - especially in the Western, developed countries. Enterprises have a keen interest in preserving the well being of their employees, due to its close connection with performance indicators. But even though the management teams are struggling around the clock to find and implement people strategies to tackle key problems that often impair organizational output, some of the challenges they face seem too difficult to be handled effectively. One of them is low employee satisfaction. Even though extensive research has been done on job satisfaction, not much attention has been paid to traits that might impact this variable.
One of these traits, and also the focal point of this paper, is Machiavellianism. Research investigating the effects of the Machiavellianism trait suggests that it can often be a significant variable influencing career choice and behavior in the workplace. However, even though the Machiavellian variable has enjoyed some popularity in the organizational behavior research, many questions remain unsolved. The attraction to Machiavellianism as a subject of research, as well as the theories related to it, has created a vast and confusing literature. In that direction, attempts to organize the literature by relating it to theories outside the boundaries of organizational behavior (e.g.: evolution and game theory) have brought a new, fresh and mature look to the concept (Wilson et al., 1996).
One of the big question marks still remaining is around the positive correlation, identified by multiple studies, between Machiavellianism and job satisfaction, or to be more precise
Fig.1
This thesis is anchored in the relationship between Machiavellianism and job satisfaction. There are a few reasons for this.
Firstly, the relationship in itself is a very interesting one. It raises an entire series of important questions as to why Machiavellian individuals might be or are more prone to be dissatisfied in their work environments. These questions, in turn, pose just as important implications for both employees, as well as the businesses that employ them. Many of the implications around the relationship between Machiavellianism and satisfaction are dealt with throughout the body of the thesis, and are touched upon in the thesis's primary quantitative study.
Secondly, the author seeks to test the positive correlation between Machiavellianism and job dissatisfaction in the Romanian marketplace and culture. This will bring new insights into the local work environment, whilst at the same time test the relationship against a background of substantially different culture to those probed by previous studies, carried out in such parts of the world as North America or Western Europe. Romania is particularly interesting because it has a residual socialist culture, dating back to the Communist days, as well as a fairly recent history of corruption within the public and private sectors. All these aspects will surely interact with the Machiavellian - satisfaction dynamic and provide for a very interesting case study.
Machiavellianism
Thirdly, this research study will check the correlation between Machiavellianism, job satisfaction and a couple of other key variables such as: trust, perceived politics, understanding and perceived control in a totally new environment and set of conditions. The purpose is that of checking the findings of previous studies as well as attempting to come up with potential plausible explanations as to why Machiavellianism and job satisfaction are correlated. There is even an element of novelty in testing the relationship between trust (in others) and the other variables involved in the study, an area that has not received much attention to date, at least not in a consistent and substantial manner. Consequently, although modest, the results of this study will prove a net new contribution to the body of theory and knowledge of the filed of organizational behavior.
Conceptual model
The examination of the most important studies on organizational politics (Drory, 1993; Ferris et al., 1989, 1993; Ferris & Kacmar, 1992) revealed several conceptual models that shared the same cause (i.e. Machiavellianism) and the same outcome (i.e. job dissatisfaction). These models, together with the element of trust and its intrinsic negative or positive influence on human nature, were used in the development of the general framework and the selection of variables of our conceptual model (Fig.2):
Fig. 2
Perceived politics Satisfaction Machiavellianism Understanding Perceived control Trust
II Literature review.
2.1 Machiavellianism
Machiavellianism was conceptualized into an essentially unitary construct by the sociologists Chriestie and Geis in 1970. A well-known writer, diplomat, politician and psychologist of Italy in the 1500s, Niccolo Machiavelli, inspired their concept. With his famous work “The Prince” he encountered a lot of criticism from the church representatives and moralists. However, there are some scientists who believe that his teachings are a great addition to anyone that tries to ‘climb up the ladder’ (Jay, 1967).
Definition
Machiavellianism has been defined by Wilson et al. (1996) as a strategy of social conduct that involves manipulating others for personal gain, which is often against the other’s self-interest.
Characteristics of Machiavellianism
High Machiavellians are being perceived by their peers as more intelligent and attractive than their counterparts (Cherulnik et al., 1981). They initiate structure by taking center stage and adopting leadership roles in small group situation (Bochner, di Salvo & Jonas, 1975; Okanes & Stinson, 1974). These findings appear to be at odds, at least from some angles, with more recent research. For example, Machiavellianism has been positively correlated with features and dysfunctions of personality that are antisocial, such as: neuroticism (Ramanaich, Byravan & Detwiler, 1994), psychopathy and narcissism (Paulhus & Williams, 2002; Lee & Ashton, 2005; McHoskey et al., 1998), and interpersonal problems (Grutman, 1992). Further negative correlations were found also with agreeableness, conscientiousness (Paulhus & Williams, 2002) and the tendency to cooperate (Paal & Bereczkei, 2007).
One possible explanation is that some of the research was performed using social experiments while the other studies were performed in real life settings. This could explain
also why high Machs consistently tended to win in the interpersonal situations involving relatively unstructured settings (e.g. bargaining and alliance forming situations) (Chriestie and Geis, 1970).
The fact that Machiavellians tend to win in experimental settings does not necessarily apply to real world settings. Researchers on this issue have found no correlation between Machiavellianism and success in sales jobs (Turnbull, 1976), marketing professions (Hunt and Chonko, 1984), or college professors (Hollon, 1975). This comes as no surprise considering that high Machs have low job identification (Mudrack, 1989), are more dishonest (Ashton, Lee, & Son, 2000), and care little about the values of others (Yurtsever, 2003).
There is also evidence that high Machs are very good liars. Even though both high as well as low Machs do lie, cheat and perform unethical acts in experiments, at least two studies have shown that high-Machs are more believable liars than the lows (Exline et al., 1970, Chriestie and Geis, 1970). In those studies, high Machs maintained greater eye contact, confessed less often and were rated as more plausible liars by independent judges.
Concerning morality, research on the subject suggests that high Machs will often view unethical behavior as being acceptable (Hegarty, 1995; Mudrack, Mason and Stepanski, 1999). However, when the problem of winning is concerned, high Machs will undoubtedly be task driven and not personally involved in the situation, or with the other players. They appear as being guided by cognition dealing with the situation as a game, where certain game rules apply, as well as a general strategy (Christie & Geis, p.190).
Do Machiavellian individuals have emotions? Some believe that they don’t have, arguing that the concepts of Machiavellianism and Psychopathy are identical (Lee & Ashton, 2005; McHoskey et al., 1998) or that the Machiavellian individual is alexithymic (Wastell & Booth, 2003). According to Chriestie and Geis (1970), high Machs are detached (naming it the ‘cool syndrome’) and not emotionally involved in interactions with other individuals, which
basically means a lack of interpersonal affect in interpersonal relationships. For them, other individuals are considered ‘means to an end’. Even though there is no correlation with empathy (Wastell & Booth, 2003), it does not mean that Machiavellians cannot experience emotions. Other researchers suggest that although Machiavellianism and Psychopathy are overlapping, they are not identical constructs (Paulhus & Williams, 2002; Williams & Paulhus, 2004). It seems that one of the particularities the former personality trait has is a linkage with heightened anxiety (McHoskey et al., 1998; Ali et al., 2009) and envy (Vecchio, 2000; 2005).
If they manipulate, how do they do it? Reimers and Barbuto (2002) suggested that Machiavellians would firstly try to influence using charm, friendliness and tact, followed by exhibiting influence behaviors considered unethical, forceful and antagonistic, in case the first attempt fails. Nonetheless, Chriestie and Geis (1970) suggested that successful manipulation is very context sensitive. They considered that high Machs have the advantage in certain situations, where a direct contact with the partner was possible, in a relatively unstructured setting, leaving sufficient leeway for manipulation and dealing with a partner preoccupied with emotions.
Concerning group-level performance, Wilson et al. (1996) suggested that Machiavellian group dynamics are expected to be complex. They suggest that low-Machs will outperform high-Machs in any group situations that require coordinated action, subordination of individual interests to shared goals, etc. However, when a group versus group situation arises, high-Machs are expected to bring the winning edge.
Outcomes of Machiavellian behavior
As we can observe from the existent research, all of these personality characteristics refer to a rather regressive, non-integrated personality structure where the Machiavellian individual remains subject to angst and desires, and to rather antisocial attitudes and behavior. Therefore
the outcomes of such behavior are not surprising. Many studies have found a correlation between Machiavellinism and external locus of control (Paulhus, 1983; Richford and Fortune, 1984), job strain (Gemmill and Heisler, 1972), job disatisfaction (Corzine and Buntzman, 1999; Gemmill and Heisler, 1972; Gable and Topol, 1987, Ferris and Kacmar, 1992, Hunt and Chonko, 1984; Walter et. al, 2005, Wakefield, 2008), and low organizational, supervisor and team commitment (Zettler, Friedrich and Hilbig, 2011). Moreover, Machiavellian employees exhibit less caring and responsiveness (Teven and Winters, 2007) and are not desired candidates for managerial posts (Teven, McCroskey & Richmond, 2006). This furthermore explains the negative correlation between Machiavellianism and upward mobility (Heisler and Gemmill, 1972).
Obviously, the amount of research on Machiavellianism points that job dissatisfaction is one of the predominant outcomes of the variable. Taking this into account, and considering the importance of job satisfaction research in organizational behavior, one would expect that these issues had been solved by now. This is not the case however. Even though tentative explanations where offered for this relationship, it is still shrouded in mystery as Hollon (1983) explains: “whether dissatisfied, non-job-involved, stressed, managers who have perceived ambiguity in their roles and low participation in decision making become Machiavellian or managerial Machiavellianism leads to such work attitudes and perceptions constitutes an unresolved issue” (p.434).
Several studies found a negative correlation between Machiavellianism and job satisfaction (Gemmill & Heisler, 1972; Hollon, 1983; Gable and Topol, 1987, Ferris and Kacmar, 1992, Hunt and Chonko, 1984; Walter et. al, 2005, Wakefield, 2008), but not one of them has succeeded in finding a holistic and clear explanation (Fehr et al., 1992). In most of the research, tentative interpretations were given to this issue as a way of interpreting the results. For example, Gemmill and Heisler (1972) suggested that high Machiavellian
individuals tend to be more cynical and distrustful of others and therefore perceive less openness and opportunity for formal control. Or, that by engaging in salient Machiavellian actions, individuals face the actions of distressed colleagues or supervisors who, resentful, provide them with less information, less authority and less opportunity for formal control (Gemmill and Heisler, 1972). So, why is it that high Machiavellians perceive less opportunity for formal control? Why is it that they are so “sensitive” towards political issues (Ferris et. al, 1989)? And why are they felling nervous and have no trust in their colleagues or supervisors?
In the end, all these questions, if answered in any meaningful way, will perhaps contribute to a better understanding of the relationship between Machiavellianism and job dissatisfaction. And, as the latter is the end variable of our conceptual model, a closer look at it and its place in the scheme of things is required and offered in the following chapter.
2.2 Satisfaction
Definition
Traditionally, employee satisfaction functions as a barometer that reflects positive employee attitudes and role behaviors (Allen, Gotcher, & Seibert, 1993). This concept was defined by McShane and Von Glinow (2010) as an appraisal for the perceived job characteristics, work environments, and emotional experiences at work. Therefore, satisfied employees tend to have a favorable evaluation of their jobs, based on their personal observations and emotional experiences.
Characteristics of job satisfaction
Job satisfaction is quite a broad concept, covering many aspects of the job such as: pay, work conditions, relations with colleagues and bosses, career prospects, the intrinsic aspects of the job itself, etc. Actually, more than 30 constructs, in thousands of studies, have been analyzed empirically in relation to job satisfaction as antecedents or consequences. In their
meta-analytic review of job satisfaction of salespersons, Brown and Peterson (1993) classified them in four categories: work outcomes, individual differences, role perceptions and two types of organizational variables: supervisory behaviors and job/task characteristics.
For work outcomes, extensive research has tackled the issue of causality between satisfaction and performance (Schwab and Cummings, 1970). Notwithstanding which one causes the other, Brown and Peterson (1993) suggested that the conclusion of all this research is a modest but consistent, positive, correlation between the two variables. Another two well-analyzed variables studied in relation with job satisfaction are job commitment and the propensity to leave or turnover. For these two, a positive correlation has been found with the first one (Bateman and Strasser, 1984) and a negative correlation with the later (Johnston et
al., 1990). Furthermore, both negative correlations have been found with role conflict and role
ambiguity (Behrman and Perreault, 1984; Ford, Walker and Churchill, 1975) and positive with role clarity (Kohli, 1985).
Other studies have found that greater amounts of consideration and feedback (Teas, 1981), communication, close supervision and contingent approving behavior are associated with greater job satisfaction (Kohli, 1985, Brown and Peterson, 1993). While the quality of interaction with supervisors impacts the level of satisfaction felt by the employee, other aspects of the job itself play an equal role in the equation. Therefore, greater job satisfaction appears to be associated with more significant, challenging, and varied tasks, with greater amounts of participation and involvement, with shared values and higher pay (Brown and Peterson, 1993).
Regarding individual differences, with the exception of work motivation and task-specific and generalized self-esteem (Hafer and McCuen, 1985; Kohli, 1985), a relatively small amount of dispositional variables have been studied in connection with job satisfaction (Brown and Peterson, 1993). Past theories of job affect have viewed job satisfaction as a work
attitude influenced by external sources (Staw & Ross, 1985) or as a person/work role interaction (Locke, 1976). Even though some studies have correlated job satisfaction with scores on personality scales (Lockes, 1976), some researchers believe that this has not improved the fundamental theoretical development of job satisfaction by much (Judge and Hulin, 1993).
Arnold and Randall et al. (2010) suggested that an individual’s job satisfaction is a function of: 1) a person’s general personality or disposition; 2) the opinions of other people in the person’s workplace; or 3) the features of a person’s job. Judge and Hulin’s (1993) findings suggest that job satisfaction is determined to a significant extent by the individual’s general level of happiness and his or her way of looking at the world.
Equally fundamental to the individual’s outlook on the world, be it negatively or positively inclined, is the person's general level as well as quality of trust. As an essential variable in the conceptual model, trust is looked at in more detail in the following chapter.
2.3 Trust
Definition
Trust, was defined by Das and Teng (2001) as a psychological subjective state of positive expectation regarding another person’s goodwill in a risky situation.
Characteristics
Trust has been proved to increase people’s willingness to engage in spontaneous sociability such as cooperative and altruistic behavior (Kramer, 1999). On the same note, Dirks and Ferrin (2002) found that trust in leadership demonstrated substantial positive relationships with attitudinal outcomes such as job satisfaction and organizational commitment.
Parker et al. (1995) believe that trust is a significant indicator of the quality of one’s relationships at work. Their statement was based upon the idea that individuals who believe their coworkers are trustworthy are less likely to consider acting in ways that would be disadvantageous to others (Poon, 2006). Contrarily, individuals who believe their coworkers are untrustworthy are more likely to consider their coworker’s behavior as politically inclined. Research studies on this matter appear to generally support a statistically significant negative relationship between trust and perceptions of organizational politics (Parker et al., 1995; Poon, 2006). Moreover, Farrell and Peterson (1982) suggest that trust in not only helpful in predicting the overall level of political behaviors, but also it dictates the form of such behavior. Their assumption is based on the idea that high levels of trust should be associated with the exercise of legitimate political behavior because the alternative would be to risk backlash from authorities expected to produce desired outcomes. This could possibly explain why sometimes perceived politics can be associated with also positive outcomes such as organizational citizenship behavior (OCB) (Chang et al., 2012).
In a meta-analytical study, Dirke and Ferrin (2002) found that trust in leadership was analyzed using two theoretical perspectives: the character-based perspective and the relationship-based perspective. The character-based perspective implies that employees will most likely engage in risky behaviors (e.g. sharing information) when they perceive that their supervisors have capability, benevolence, and integrity and thus can be trusted. The latter perspective however, is based on the principles of social exchange and reciprocity (Blau, 1964 as cited by Poon, 2006). This means that employees who trust their supervisor will feel obligated to reciprocate with behaviors that will benefit their supervisor.
Of significant importance in the way the relationships described above pan out is individual’s perception of politics; And because of this reason this is an area probed into next.
2.4 Perceived politics
Definition
Kacmar and Ferris (1991) defined perceptions of politics as an individual’s observations of others’ self-interested behaviors and, namely, the suppression of competing entities and selective manipulation of organizational policies.
Characteristics
The predominant idea in organizational behavior research is that organizational politics has a dysfunctional role and the potential to disrupt organizational efficiency and effectiveness (Kacmar et al., 1999). It is not hard to understand why it has this negative outcome. Employees tend to associate organizational politics with self-serving behaviors, which promote personal objectives, usually at the detriment of others. These behaviors tend to be: discretionary; associated with manipulation, defamation, subversiveness and abuse of power; and contrary to organizational goals (Vigoda, 2000). Examples range from discrediting one’s rival at work in order to get ahead, not sharing useful information with other employees in order to exploit this at a later stage, doing favors for the boss to secure more benefits (Poon, 2003).
Why focus on a subjective measurement of politics in organizations?
Whether an individual perceives his work environment as political or not doesn’t necessarily mean that this reflects reality. Political behaviors in organizations may prove to be highly obscured, symbolic, and subject to differences in perceptions (Gandz and Murray, 1980; Ferris el al., 1989). James and James (1989) proposed that the choosing between personally beneficial or personally detrimental as a vehicle for understanding employee behavior is useful to comprehend how employees perceive managerial behavior. As a consequence, if an individual perceives a behavior as political in nature, this is likely to be a
function of the sum characteristics of the individual, the perceiver, the situation, and the ensuing cognitive evaluations formulated by the individual (Ferris, Fedor and King, 1994).
The literature on organizational politics, being quite vast, has been concentrated around two streams of research: antecedents and outcomes of perception of politics. The antecedents are based upon theoretical models of conditions thought to contribute to political behavior, while the outcomes are the consequences of it. Due to the high number of antecedents in the literature, Atinc et al. (2010) classified them in a meta-analysis study in: organizational influences, job/work environment influences and personal influences. From the pool of influence variables that have been tested empirically and found to be significant in connection with perception of politics, the following can be named: centralization and formalization (Andrews and Kacmar, 2001; Vigoda, 2000) for organizational influences, opportunity for advancement (Kacmar et al., 1999; Valle & Perrewe, 2000), trust (Parker et al., 1995; Poon, 2006), leader-member exchange (Kacmar et al., 1999), participation in decision-making (Ferris et al., 2002), opportunities for carrier development (Parker et al., 1995) for job/work environment influences and Machiavellianism (O’Connor and Morrison, 2001; Valle and Perrewe, 2000), external locus of control (Ferris et al., 1993), positive and negative affect (Ferris et al., 2002) for personal influences.
Among the outcomes of perceived politics that have been found in the research literature the following can be named: job stress (Ferris et al., 1996; Kacmar et al., 1999), employee attitudes such as job satisfaction (Gandz & Murray, 1980; Kacmar et al., 1999) and organizational commitment (Witt, 1998), turnover intention (Kacmar et al., 1999), neglect (Vigoda, 2000), self-reported performance (Kacmar et al, 1999), and organizational citizenship behavior (Vigoda, 2000).
Above all, personality characteristics have been quite overlooked as antecedents of organizational politics, probably due to the fact that personality problems cannot be solved
that easily and they are less susceptible to management intervention (Poon, 2003). However, other researchers believe that personality characteristics may account for substantially more variance in perception of politics than demographic variables (Adams et al., 2008; Attinc et
al., 2010). Moreover, they are thought to be important because they influence how individuals
perceive and interpret organizational politics (Adams et al., 2008).
Equally influenced by personality characteristics are ‘understanding’ and ‘perceived control’, key variables to our research and looked at in the following chapter.
2.5 Understanding and Perceived Control
Definition
Understanding was described by Ferris et al. (1996) as the extent to which one comprehends why and how things happen the way they do in the organization. Personal control on the other hand, is an individual belief in his or her ability to exercise influence or change over their general work environment (Ferris et al., 1996).
Characteristics.
McGrath (1976) argued that exposure to any given situation produces familiarity and, thus, a better understanding of the situation, consequently resulting in a reduction of subjectively experienced strain. In that regard, experience in the field and, implicitly, job tenure become very important, increasing the exposure to more events in the organization and, therefore, increasing the overall understanding of the organizational life. A good, well-positioned, placement in the hierarchy of an organization can provide a greater sense of control over apparently unpredictable events that can often characterize life in large organizations (Zaleznik, 1989). To the same extent, Miller (1981) suggested that control reduces stress for the reason that the person who is in control would expect less unpleasant outcomes than a person who possesses less control.
Perceived control, as well as understanding, are derived from stress related research and, in particular, from the Demand-Control Model (Karesek, 1979). It was emphasized by Arnold and Randall et al. (2010) that control is tremendously important not only as positive role in employee well-being but as well as providing employees with opportunities to deal better with demands they are facing, as well as fulfilling a basic human need for control. Empirical research on the subject attests that control moderated the relationship between job demands or stressors and anxiety (Perrewe & Ganster, 1989), and between job demands and satisfaction (Fox, Dwyer & Ganster, 1993). This confirmed Ganster’s (1989) suppositions that control attenuates the impact of stressors, or, more explicitly, it moderates the relationship between stressors and strains.
Even though it can be observed from the definitions that perceived control and understanding are overlapping to some extent, they are complementary. Thus, if one has understanding over his environment, they can exercise control over it and not the other way around. To the extent that an organizational member can understand how and why events happen and control the outcomes desired by effectively influencing the events, or others in the work environment, less strain will result from perceived organizational politics in the work environment (Sutton and Khan, 1986).
Having looked at the variables in the make-up of out conceptual model, hypotheses are now presented and a detailed process of their formulation, as well as rationale, is covered by the next chapter.
III Hypotheses
Based on previous empirical results, Judge and Hulin (1990) considered that the link between personality variables and organizational relevant attitudes has been, in general, disappointing. Moreover, they considered personality literature as being impaired by modest
results, revealing only tangential relations to a well-rounded theoretical framework for the above-mentioned relationship. This is also the case for the Machiavellianism – Job Satisfaction relationship. Even though a lot of studies have found that these two variables are linked to some extent (Corzine and Buntzman, 1999; Gemmill and Heisler, 1972; Gable and Topol, 1987, Ferris and Kacmar, 1992, Hunt and Chonko, 1984; Walter et al., 2005, Wakefield, 2008)), few managed to find an explanation for it. For example, Corzine and Buntzman (1999) reflected on this relationship and suggested that Machiavellians might be prone to job dissatisfaction based on their distrust and cynicism, along with an underlying suspiciousness. Others, such as Gable and Topol (1987), suggested that organizational settings such as fewer opportunities for advancement might be the cause.
As a consequence, this study proposes to shed light on this literature gap by proving that personality traits play a critical role in explaining organizational attitudes. Arnold and Randall
et al. (2010, p.471) suggested that individual differences play an important role in “how
people perceive their work environment and their psychological and physical well-being“. Moreover, those differences can play a significant role in ascertaining how strong the link is between the two. On this note, personality may have some impact on the person’s perceptions of what his or her job is like.
Contrary to the contentions of Davis-Blake and Pfeffer (1989), individuals' job satisfaction is not merely a function of external events. Job satisfaction is determined to a significant extent by the individual's general level of happiness and his or her way of looking at the world. This does not suggest that job conditions are unimportant determinants of job satisfaction or that they don’t contribute to the overall perceived level of satisfaction. However, there is evidence which shows that perceived reality holds a very strong sway over the individual's thinking and actions within their environment.
Sometimes, in work situations, competition for status and resources is an ongoing dynamic. This dynamic, is often experienced in jobs that can be viewed as examples of long-term structured relationships with little potential for exploitation (Wilson et al., 1996). As we mentioned in the literature, high Machs need sufficient leeway to deploy their manipulation tactics. Jobs settings, usually, don’t provide this. Thus, high Machiavellians, who use their skills to manipulate and deceive others in order to get ahead, would probably feel uncomfortable to work in such structure. Often lacking the stable power base (Machiavellians without managerial responsibilities) to deploy strong influence tactics (e.g. carrot and stick) will render them looking ridiculous in the attempt. It would be expected that, in a group setting, Machiavellian practices to be totally unacceptable. Why? As is mentioned in the literature, Machiavellianism has been associated with dysfunctions of personality that are antisocial (e.g. neuroticism, narcissism, interpersonal problems). It is also known that high Machiavellians are mostly self-centered and rarely consider the impact of their actions on the people around them. Moreover, it is expected that high-Machiavellians will be less likely to help others in ways that do not coincide with their own interest (Wilson et al., 1996). Having low empathy (Barnett & Thompson, 1985) and being less likely to help others in simulated emergency situations (Wolfson, 1981) means that cooperation is severely restricted if not absent altogether. This, as well, can only be regarded as dysfunctional and unacceptable in a regular work environment.
As a consequence, this implies that high Machs will not only fail to create friendly ties with their superiors or coworkers but - more likely - animosity, avoidance and even repulsion. As Wilson et al. (1996) mentioned: “Machiavellianism has obvious advantages, but it also has costs, mostly in the form of retaliation and avoidance by others, that allow less manipulative strategies to prevail in at least some situations.” (p. 295). On the same note, one of the
predictions that Wilson et al. (1996) have made is that high Machs will fail in long-term interactions and will succeed only by becoming phenotypically similar to low Machs.
While the previous argument assumed that Machiavellians engage in manipulative and deceitful actions to obtain their desired ends and suffer the retaliation of their colleagues or supervisor, what would happen if they didn't? Firstly, as an employee in the low ranks of organizations, it would be rare to engage in any kind of political action, because job tasks are pretty clear and one would first have to prove himself/herself. That is why politics is often displayed in the higher ranks of the organizations and not on the lowest ranks (Drory, 1993). Moreover, it is viewed as salient, normal and even mandatory in the higher echelons of the organization on the one hand, and dangerous and detrimental in the lowest ranks (Pfeffer, 2010). This fact, coupled with the nature of Machiavellians may result in even worst outcomes on commitment and job satisfaction. Knowing that high Machiavellians are anxious, sensitive to political issues and that they don’t trust others can only result in job dissatisfaction. Moreover, I believe that Machiavellianism as a variable would act as a magnifying glass over perceived politics, amplifying its negative outcomes. Therefore I propose the following hypothesis:
Hypothesis 1.: Machiavellianism is negatively correlated with job satisfaction.
The perceptual process begins when environmental stimuli are received through our senses, screened out, organized and interpreted. Once the input is in, the selective attention process occurs, where some information is being attended and the rest ignored. One important note is that the selective attention is influenced by the characteristics of the perceiver, sometimes even without the perceiver’s awareness (McShane and Von Glinow, 2010). The researchers also inform us that once the information is interpreted, certain perceptual processes and biases occur such as: self-fulfilling prophecy, attribution, and stereotyping.
In organizations, decisions such as resource and reward allocation, which are governed by political considerations, are usually viewed as being unfair. If their employees perceive decisions to be unfair, they will most likely be dissatisfied (Poon, 2003), regardless of whether they are Machiavellian or not. Moreover, the stress that is associated with working in a politically charged environment, or perceived as being so, will add to employees’ dissatisfaction with their job. It could be hypothesized,, however, that Machiavellians experience all this, but with an increased intensity due to their nature. Their way of decoding the information from their environment as a continuous threat has been noticed by Drory (1993), who believes that Machiavellians are quite sensitive to politics. Or, to put it another way, Machiavellians tend to have a more biased view of reality when it comes to politics.
To draw a parallel with popular wisdom, where we encounter the old adage ‘beauty is in the eye of the beholder’, Machiavellians appear to see reflected in others, and especially in their environments, aspects of their own, negative, worldview. Evidence is put forth in the form of negative correlations with agreeableness, conscientiousness (Paulhus & Williams, 2002) as well as positive correlations with neuroticism (Ramanaich, Byravan & Detwiler, 1994), psychopathy and narcissism (Paulhus & Williams, 2002; Lee & Ashton, 2005; McHoskey et al., 1998). Therefore the following hypothesis is put forth:
Hypothesis 2.: Machiavellianism is positively correlated with perceived politics.
Cropanzano et al. (1997) suggested that the role of politics in an organizational setting that is a politically charged environment can lead to employees being uncertain of whether their work will be rewarded appropriately or whether they will not be put at risk by the decisions and actions of their coworkers and superiors. He also highlighted the fact that risk, unpredictability and a threatening work environment can lead to increased levels of experienced job stress. There is also past evidence showing that organizational politics needs
only be perceived in order to generate a stressful environment in the work place (Cropanzano
et al., 1997; Ferris et al., 1996; Kacmar et al., 1999). Cropanzano et al. (1997) found clear
examples of employees reporting higher levels of stress when they perceived their work environment as being political.
Ferris et al. (1999) shape up the negative picture further by highlighting the association of perceived politics in the organization and decreases in job satisfaction. A potential explanation for this is that decisions having to do with resource allocation that are governed by political considerations, or perceived as such, are usually considered to be unfair. This unfairness will make employees unhappy. The unhappiness, coupled with the stress of working in a politically charged environment, will only add to the employees' dissatisfaction with their job.
Past studies employing both conventional regression techniques (e.g. Cropanzano et al., 1997; Ferris and Kacmar, 1992), as well as structural equation modeling techniques (e.g. Kacmar et al., 1999) have found perceived politics in the organization to be a negative predictor of job satisfaction.
Based on empirical evidence coming out from these studies, as well as for the purpose of helping to prove and explain the link between Machiavellianism and satisfaction, the following hypotheses are offered:
Hypothesis 3.: Perceived politics is negatively correlated with job satisfaction.
Individuals who present an external locus of control (LOC) may be more likely than those with an internal LOC to perceive the workplace as political (Vredenburgh & Maurer, 1984). The researchers haven't clearly outlined why this relationship exists but offered a possible explanation that individuals with an external LOC view themselves as unable to manipulate
their work environment, and, concomitantly, see themselves as vulnerable in the face of political actions of their co-workers.
To Machiavellians, who are known to be externals (LOC), those feelings of powerlessness may translate in frustration and ultimately into job dissatisfaction. They will find ways to blame any external influences for the bad situation they might find themselves in. Taking all this into consideration it is proposed that:
Hypothesis 4.: Perceived politics would act as a mediator between Machiavellianism and
Job satisfaction.
It is important to note that trust, like the majority of the variables involved in this study (e.g. perceived politics, understanding, perceived control), is a belief or perception held by every individual. It is not a property of the relationship between individuals and their colleagues and superiors, neither their characteristic. This belief will often be a trust-related concern about the characters of the colleagues and superiors. These concerns do play an important role because managers, for example, may have authority to make decisions that have a significant impact on the follower and the follower's ability to achieve his or her goals (e.g., promotions, pay, work assignments) (Dirks & Ferrin, 2002). From this perspective, every individual will try to gather information about their colleagues and superiors characteristics such as integrity, dependability, fairness and ability, which consequently will have an impact on their behaviour and attitudes. However, research on cognitive consistency and attitude change (Eagly and Chaiken (1993) has identified individuals' tendency to perceive and interpret their immediate environment in a way that would reinforce their prior knowledge and beliefs. So, everyone has a 'baggage', be it from school, home or previous work settings, that he or she has to work with. For Machiavellians the situation is no different. Knowing that their view of human kind is a cynical one, it would be expected from them to
have a negative expectation regarding the motives, intentions and prospective actions of other members on whom they depend, which will ultimately translate in a negative trust climate.
Farrell and Petersen (1982) implied that employees who have a high level of trust do not necessarily perceive a need for political action and are, consequently, less likely to engage in the political game than those with lower level of trust. Moreover, when participants with high level of trust do engage in politics, they are more likely to engage in legitimate and constructive political behavior such as organizational citizenship behavior (Parker et al., 1995). This however, is a kind of behavior that would be considered normal for any employee, and it will not be perceived as a threat.
Employees with greater Machiavellian orientations, on the other hand, are cynical and distrustful in cooperating with their colleagues, in evaluating superiors’ actions or the information that they provide (Grutman, 1992; Paulhus & Williams, 2002). With such low trust in others, high Machs may be more likely to view any normal event as political. This would indicate that they might be bothered by the impression that someone else may get the job for which they are directly in line, or by not knowing their opportunities for advancement, as well as what their superior thinks of them, etc. Therefore, it is proposed that:
Hypothesis 5.: Trust will function as a mediator between Machiavellianism and
perceived politics.
One of the aspects that make managers effective leaders is a thorough analysis of the complex web of power relationships and influence processes found in all organizations and their attitude towards them (Yukl, 2010). In the political process, qualities such as competence, intelligence, and innovativeness don’t necessarily prevail (Pfeffer, 2010). What can make the difference though, is the understanding of the political process. Even though understanding should not make the difference between success and failure in the political
playing field, it should mitigate the satisfaction felt by the individual. To some extent, job tenure should function as a proxy for understanding since, the more time one spends in an organization, the more time it has to observe all the important actors in the political arena and know how to react to them, although it is not a guarantee. That is why Schein (1977) suggested that one of the organizational challenges for new employees is to understand and effectively deal with political behavior.
Kackmar et al. (1999) presumed that an individual who has a clear understanding of who is responsible for making decisions and what was the criteria behind their election would have a better understanding of how and why things happen the way they do than someone who does not understand the decision-making process in organization. Moreover, researchers as Ferris et al. (1989) suggested that understanding should explain the extent to which individuals perceive politics as an opportunity or as a threat. They explained that, when an individual perceives his organization as political and his understanding is low, political actions are likely to be viewed as threat and the consequent outcomes should be negative (high anxiety, stress, increased turnover). On the other hand, however, if the individual perceives his organization as political and his understanding is high, political actions are more likely to be viewed as an opportunity, and subsequently the outcomes should be less negative. Therefore the following hypothesis can be put forth:
Hypothesis 6.: Understanding will moderate the relationship between perceived
politics and job satisfaction.
Reactions of individuals to perceptions of organizational politics may be moderated by the degree of control they experience, as well as the level of understanding they have. For example, if Machiavellians see organizational politics as an opportunity to get ahead because they can exercise control over their work environment, their job satisfaction would be higher
than Machiavellians who perceive organizational politics as a threat due to possessing little or no control over the situation. This is almost the same case with understanding. Besides that, as I mentioned before, Machiavellians have a desire for control and a pursuit of status which, if not met, can result in their dissatisfaction.
A feature of envy is the diminution of self-worth, which implies also the loss of social standing as a result of social comparisons (Mumford, 1983). This has been confirmed by Vecchio (2000). He suggested that such an emotional state is activated by the self-perceived inability to control events. In essence, it is a defense reaction to social threats. When a person perceives that he or she is unable to deal with or manage a social threat, certain feelings of powerlessness or helplessness will appear.
Thus, it can be argued that the presence of perceptions of organizational politics can lead to dissatisfaction when high Machiavellians feel that they do not have the control over the process. Therefore the following hypothesis is proposed:
Hypothesis 7.: Perceived control will moderate the relationship between perceived politics
and job satisfaction.
IV Research Methodology
4.1 Methodology
Explanatory analysis, by way of the quantitative approach, makes up the primary research method employed in this thesis. The causal and correlative relationships between the various variables in the conceptual model are explained in the analytical process and by means of the chosen research method (Saunders et al., 2009). Primary sources form the bases of the quantitative data employed, and are tapped into by the survey, which is the main vehicle of research strategy for this thesis. Statistical tools and techniques are further employed in order
to make sense of the raw data generated by the survey. This process presents a better understanding of relationships, which in turn can lead to relevant and meaningful conclusions to enhance the existing body of research. Common in science, the deductive approach is then made use of in order to move from theory to raw data interpretation. This is expected to yield a better and more accurate account of the relationships studied throughout the course of the empirical research (Saunders et al., 2009).
The methods employed in the analysis of the data are as follows: descriptive statistics in order to see the distribution and the frequencies of the data; correlation analysis in order to verify the strength and direction of the relationships described in the conceptual model (hypotheses 1 to 3); ordinary least squares linear regression analysis together with two Sobel tests (http://quantpsy.org/sobel/sobel.htm) in order to verify the mediator effect of trust between Machiavellianism and perceived politics, as well as that of perceived politics between Machiavellianism and job satisfaction (hypotheses 4 and 5); ordinary least squares nonlinear regression analysis with interaction term in order to verify the moderator effect of understanding and perceived control on the relationship between perceived politics and job satisfaction. In order to prepare the data for the latter operation, the variables Understanding, Perceived Control, Perceived Politics and Job Satisfaction were z-standardized in order to make their variation comparable and to exclude therefore any numeric effects resulted from differences of magnitude.
In conclusion, the methodology was selected to sustain a coherent and logical explanatory analytical process based on the literature review, raw data generated by the survey, statistical analysis and interpretation of results as well as a final drawing of conclusions around the Machiavellianism – job dissatisfaction dynamic, which is the primary focus of this thesis.
The survey makes up the core of the research strategy employed in this thesis. The reason for its selection is twofold: on the one hand it fits in well with the combination of deductive and quantitative methods; and on the other, it is in line with the strategy of most organizational behavior studies referenced in the thesis. In order to increase reliability and replicate the study easily, a structured methodology was also selected, in order to complement the deductive approach (Saunders et al., 2009; p. 118).
It is Saunders et al. (2009) again who assert the essential role of surveys in the structured collection of data from a statistically relevant number of respondents. The survey thus is a strategic tool of first resort to be applied within the context of a structured interview, observation or questionnaire. For the purpose of this thesis the strategy revolves around the questionnaire, whereby all the respondents involved in the study are asked to answer the same questions in a predetermined order (Saunders et al., 2009).
The questionnaire for this thesis employs closed questions that have been coded for easy processing and analysis, and it is of the self-administered type, whereby an interviewer is not required to be present in order for the questionnaire to be filled out by respondents.
The thesis makes use of a self-administered Internet-based questionnaire available at
www.thesistools.com. The responses are automatically collected by the web service’s
underlying software and become available as a coded Excel spreadsheet.
Making use of an Internet-based questionnaire is convenient in terms of both creation and distribution, with the added benefit of significantly reduced processing times. The electronic questionnaire is also cross-sectional, in that it allows the researcher to tap into the current situation directly, so as to avoid the longitudinal changes over time effects of a long-term research project (Saunders et al., 2009).
Saunders el al. (2009) promotes the questionnaire approach as an ideal strategy for making sense of the relationships between research variables. In their view, the highly unified
and standardized make up of the questionnaire saves the researcher considerable time and provides for a simple data collection process. There is also the added benefit of sound internal validity by means of an easy comparison of the structured data as well as prevalence detection.
Although the questionnaire is well adapted to quick big sample data collection, where the researcher doesn’t have to be present in the interview, there are certain weaknesses in this approach and they are mainly centered around respondent’s risk of misinterpreting or admitting in response to some of the questions. In turn, this would have an impact on the reliability and validity of the raw data. Therefore Pallant (2010, p.11) advises that the structure and phrasing of the questionnaire be as little ambiguous as possible, with clear-cut questions and answers.
Additionally, some respondents might find frustration with overly standardized questions and answers, or a lack of leeway in expressing their own opinions and take on the subject matter.
4.3 Questionnaire design
The questionnaire for the thesis research was designed to be a self administered tool, deployed electronically in a web environment and hosted at www.thesistools.com. The language chosen for the questionnaire was English, even though the respondents were Romanian individuals. There were a few reasons for this action. English gave the study the appearance of an international research project, which appealed more to the respondents and generated grater interest among them. The move was also encouraged by the fact that most of the companies involved in the study view English as the official business language, especially in regards to written communication.
Fifty-one questions make up the questionnaire. A few of them are multiple choice, whilst the majority of questions employ scales. They are easy to code as well as highly compatible with SPSS analysis. As Pallant (2010, p.9) advises, the purpose of the correlation analysis is very well served by the method of scales questions and the continuous stream of data they can generate. The scales employed in the questionnaire are 5-point Likert, with 1 being “strongly disagree” and 5 its positive counterpart, that is “strongly agree”. The complete questionnaire is attached under Appendix 1.
Firstly, the questionnaire starts with a brief introduction, which is followed by a few demographic information questions. The respondents are asked to specify their gender, age, job tenure and whether they have a supervisor/manager position or not.
Secondly, the respondents are asked to answer questions about their personality. More specifically, questions 1-14 are used to measure the level of Machiavellian personality of each respondent. This scale is a slightly modified version from the original Christie and Geis’s (1970, p. 17-18) 20-item scale on Machiavellianism (also known as Mach IV). Some of the key sample items deployed with the questionnaire are: “It is hard to get ahead without cutting corners here and there.” and “The best way to handle people is to tell them what they want to hear.”
Next, questions 15-24 are aimed at evaluating the level of the items which build up the job satisfaction. Here, nice items from the original scales developed by Macdonald and MacIntyre (1997) were used to measure respondents’ job satisfaction. This, together with the general item “I am satisfied with my job” formed the overall scale for job satisfaction.
For trust, four items (questions 25-28) from the scales developed by Poon (2003) were used for this study. One sample item from the scale is “There is a high level of trust throughout my organization”.
Furthermore, both “understanding” as well as “perceived control” were measured with a 12-item scale (6 for understanding and 6 for perceived control) developed by Tetrick & LaRocco, (1987). Sample items include: “To what extent do you know why others at work act as they do?” and “To what extent do you have influence over the things that affect you on the job?”
Perceptions of organizational politics were assessed using Kacmar and Ferris’s (1991) 12-item perceptions of organizational politics scale. From this scale just 11 12-items were used (questions 41-51). Sample items include: “One group always gets their way” as well as reverse-coded “Encouraged to speak out”.
4.4 Data collection
The questionnaire in its final format was deployed in English via the web service
www.thesistools.com. The respondents targeted for this study were Romanian professionals
working for Romanian as well as multinational companies operating within the Romanian economy. Most respondents were employees of Oracle Romania, a multinational IT company, Ubisoft, a gaming company, National TV, a national TV channel, Avicola S.A., a Romanian poultry producer and exporter, as well as Konika Minolta, another multinational company. Most of these individuals had graduate or postgraduate degrees, were Romanian natives, but held good command of the English language.
In order to widen the reach of the questionnaire, a snowball sampling technique was made use of, whereby individual respondent were encouraged to forward the questionnaire link onto their colleagues as well as friends and acquaintances that held similarly highly qualified roles. Not deviating from the minimum criteria of having a qualified job, the respondents where also encouraged to make use of social media outlets such as Linkedin and Twitter in order to propagate the questionnaire in more efficient manner. This yielded great results.
The final number of respondents came out at 87, which is enough to lend the study statistical significance, especially when the high quality of respondents is factored into the equation.
V Results
The web-hosted software www.ThesisTools.com as well as an electronic questionnaire was employed for data collection. Prior to formatting it for SPSS analysis, the data was first imported into an Excel spreadsheet. With the SPSS file in place, the data was analyzed for outliers and errors as well as tested with the purpose of determining various relationships between the variables employed in the study. No outliers or errors were found. All analysis was conducted using the SPSS and Excel software.
5.1 Descriptive
The size of the sample used for this study was 87 respondents. In the sample, the distribution among genders was around 2:1 in the favor of female respondents, where the number of males was 27 and the number of females was 60. The graph of percentage of male and female respondents is as follows in Fig. 3:
Fig. 3 Gender distribution
Regarding the age distribution, the sample again is not very homogeneous. Almost 80% percent of the respondents where situated in the 25-37 age group, whilst the other age groups share an almost even distribution of the last 20% of the sample as shown in Fig.4.
Fig.4 Age distribution
As far as the current position and job tenure are concerned, the respondents are clustered around some attributes. Around 80% of them confirmed that they do not hold any manager or supervisor positions with their current job while 36% of the respondents have 5 or more years of experience (see Appendix 2.).
5.2 Reliability of scales
Some of the variables used in the study have to be aggregated in order to correctly analyze correlation, regression and variance between them. The reliability of scales was looked at first in order to verify whether scales could be generated.
Because not all items were unidirectional, the coded items scores were reversed to obtain the proper measure for the scales. The reliability of scales principle was used to determine whether items covered by the same scale could be analyzed as a single variable. The items covered by the questionnaire scales were: Machiavellianism (14 items), Trust (4 items), Perceived Politics (11 items), Understanding (6 items), Perceived Control (6 items), and Satisfaction (10 items).
The 5-point Likert was applied to all the scales mentioned above. Cronbach’s alpha coefficient can be used in order to check the reliability of scales for internal consistency. When Cronbach’s alpha coefficient equals or exceeds .7 the scale can be considered reliable (Pallant, 2010). It is preferred that the coefficient should equal or exceed .8.
As it can be seen in the table bellow, all scales turned out to be reliable, most of them coming close or higher then the required minimum. These were successfully computed into new variables. This was done by way of computing means from the sets of items above.
Variable Cronbach’s Alpha N of items
Machiavellianism 0.694 14 Trust 0.87 4 Perceived Politics 0.674 11 Perceived Control 0.89 6 Understanding 0.713 6 Satisfaction 0.87 10
5.3 Correlations
A correlation analysis was also carried out in order to gain a deeper understanding of the relationships between variables as well as their strength and direction. And because the correlation analysis showed continuous variables, the best measure to be used was the Pearson product-moment correlation coefficient (r). The values the Pearson coefficient can assume ranged between -1 and 1, whereas a value of 0 indicates no correlation whatsoever. A value of 1 would indicate an absolutely positive correlation, and a value of -1 an absolutely negative correlation. In the case of positive correlations, when one variable is increased so is the other, whilst with negative correlations when one variable is increased the other decreases (Pallant, 2010).
The results of the SPSS correlation analysis yielded the correlation matrix (see table bellow). The matrix clearly indicates all correlations were statistically significant.
Correlations matrix Satisfaction Machiavelli- anism Trust Perceived Control Understanding Perceived Politics Satisfaction Pearson Correlation 1 N 87 Machiavelli- anism Pearson Correlation -0.325 1 Sig. (2-tailed) 0.002 N 87 87 Trust Pearson Correlation 0.732 -0.46 1 Sig. (2-tailed) 0 0 N 87 87 87 Perceived Control Pearson Correlation 0.507 -0.275 0.517 1 Sig. (2-tailed) 0 0.01 0 N 87 87 87 87 Understanding Pearson Correlation 0.393 -0.275 0.279 0.52 1 Sig. (2-tailed) 0 0.01 0.009 0 N 87 87 87 87 87 Perceived Politics Pearson Correlation -0.487 0.534 -0.649 -0.428 -0.124 1 Sig. (2-tailed) 0 0 0 0 0.251 N 87 87 87 87 87 87
According to Pallant (2010, p. 134) the strength of the correlation can be assessed on 3 levels: weak, medium and strong. Most of the correlations in this research belong to the medium and strong category.
The medium strength correlations are between r=0.3 to 0.49 mark (Pallant, 2010; p. 134). Here lie most of the correlations found in this study that pose interest to us. The matrix presented above shows for example that the correlation between Machiavellianism and Satisfaction can be included in this category: r=-0.325; p<0.05. This means that Machiavellianism explains 10.5% of variance in Satisfaction. Although this negative correlation is not as strong as the others, at r=0.325 it is strong enough to confirm the first hypothesis.
Additionally, there are another three negative correlations worth mentioning. Firstly, there is the correlation between Machiavellianism and Trust with r=-0.460; p<0.05. This indicates that Machiavellianism should explain 21.1% of the variance in Trust. Secondly there is the correlation between Perceived Politics and Satisfaction with r=-0.487; p<0.05. And thirdly there is the correlation between Perceived Control and Perceived Politics with r=-0.428; p<0.05. Here, the coefficients of determination were calculated at 23.7% and 18.3% respectively. This implies that as Perceived Politics increases Satisfaction as well as Perceived Control decreases confirming also the third hypothesis.
The correlations in the matrix where r =0.50 to 1 belong to the strong category. This implies that the value of one variable can determine the value of the second variable to a very high extent (Pallant, 2010; p.134). Here, the positive correlation between Machiavellianism and Perceived Politics with r=0.534; p<0.05 fold in this strong category. This confirms our second hypothesis meaning that as Machiavellianism increases, Perceived Politics also increases.
The strongest correlations in this study were found between Trust and Perceived Politics and between Trust and Satisfaction. The first correlation (r=-0.649; p<0.05) indicates that there is a 42% of explanation of variance in Perceived Politics when taking the Trust into consideration. More specifically, this negative correlation indicates that, as Trust increases Perceived Politics decreases. The latter, the strongest correlation found (r=0.732; p<0.05) means that the more the Trust variable increases, so does the Satisfaction variable. This variance in Satisfaction is explained in 53.5% by Trust.
5.4 Regression analysis
The relationships between independent and dependent variables, as well as the inter-correlations between these, were investigated by means of the multiple regression analysis method. This method of analysis was employed in order to determine how a set of independent variables would predict the dependent variable (Pallant, 2010).
In order to test Hypothesis 4 and 5, the method proposed in Baron and Kenny (1986) was employed. Thus, as in the case of Hypothesis 4, the mediator effect of Perceived Politics between Machiavellianism and Satisfaction was studied by carrying out three regressions. The first regression was of the dependent variable (Satisfaction) on the predictor (Machiavellianism), which yielded a coefficient of -0.516, significant at the 1% level.
Model Unstandardized Coefficients Standardized Coefficients t Sig. B Std. Error Beta 1 (Constant) 4.839 0.449 10.774 0 Machiavellianism -0.516 0.163 -0.325 -3.172 0.002 a. Dependent Variable: Satisfaction
The second step involved regressing the proposed mediator (Perceived Politics) on the independent variable (Machiavellianism), which yielded a coefficient of 0.659, significant at the 1% level.
Model Unstandardized Coefficients Standardized Coefficients t Sig. B Std. Error Beta 1 (Constant) 1.066 0.312 3.412 0.001 Machiavellianism 0.659 0.113 0.534 5.827 0
a. Dependent Variable: Perceived Politics
The last step involved regressing the dependent variable (Satisfaction) on both the predictor (Machiavellianism) and the mediator (Perceived Politics) in order to observe the decrease, in absolute value, of the coefficient of the predictor as a result of the introduction of the mediator. The result obtained is in line with this expectation, as the coefficient of Machiavellianism becomes -0.144, a significant reduction from -0.516, resulted from adding the mediator Perceived Politics. Furthermore, the coefficient of the mediator, Perceived Politics, is significant at the 1% level. From these results, according to the method proposed by Baron and Kenny (1986), it can be concluded that Perceived Politics functions as a mediator between Machiavellianism and Satisfaction.
Model Unstandardized Coefficients Standardized Coefficients t Sig. B Std. Error Beta 1 (Constant) 5.44 0.443 12.276 0 Machiavellianism -0.144 0.178 -0.091 -0.81 0.42 Perceived Politics -0.564 0.144 -0.439 -3.908 0
a. Dependent Variable: Satisfaction
A similar analysis is carried out in order to test Hypothesis 5. The role of Trust as a mediator between Machiavellianism and Perceived Politics is examined. First, the dependent variable (Perceived Politics) is regressed on the predictor (Machiavellianism), yielding a coefficient of 0.659, significant at the 1% level.
Model Unstandardized Coefficients Standardized Coefficients t Sig. B Std. Error Beta 1 (Constant) 1.066 0.312 3.412 0.001 Machiavellianism 0.659 0.113 0.534 5.827 0
a. Dependent Variable: Perceived Politics
The second step is to regress the mediator (Trust) on the predictor (Machiavellianism), obtaining a coefficient of -0.909, significant at the 1% level.
Model Unstandardized Coefficients Standardized Coefficients t Sig. B Std. Error Beta 1 (Constant) 5.408 0.529 10.22 0 Machiavellianism -0.909 0.193 -0.455 -4.707 0 a. Dependent Variable: Trust
The last step is to regress the dependent variable (Perceived Politics) on both the mediator (Trust) and the predictor (Machiavellianism) and to observe the decrease in absolute value of the coefficient of the predictor as a result of the introduction of the mediator. Indeed, the coefficient of Machiavellianism drops to 0.369, from 0.659 observed prior to the introduction of the moderator. The coefficient of the mediator (Trust) is significant at the 1% level. From these it can be concluded, according to Baron and Kenny (1986) that Trust acts as a mediator between Machiavellianism and Perceived Politics.
Model Unstandardized Coefficients Standardized Coefficients t Sig. B Std. Error Beta 1 (Constant) 2.684 0.383 7.005 0 Machiavellianism 0.369 0.108 0.299 3.415 0.001 Trust -0.28 0.048 -0.512 -5.846 0 a. Dependent Variable: Perceived Politics
