Perceptions of Organizational Politics: A Meta-Analysis of Theoretical Antecedents

Beginning of article

In the last twenty years, researchers and practitioners alike have become increasingly interested in political behavior within organizations. Although political behavior can have both positive and negative outcomes, most of the research has focused upon "dark side" (Ferris and King, 1991) political behaviors, which are behaviors not sanctioned by the organization (i.e., illegitimate) and characterized by self-interest (e.g., taking credit for others' work, coalition building, personal attacks upon others competing for the same resources, and decisions based upon favoritism). This type of political behavior tends to create a harmful and divisive work environment, reduces organizational efficiency and effectiveness, and has exceedingly harmful effects upon workers (Kacmar et al., 1999; Mintzberg, 1983).

Research on organizational politics has largely focused upon perceptions of organizational politics because of Lewin's (1936, Principles of Topological Psychology) suggestion that individuals act upon their perceptions of reality rather than objective reality (Ferris et al., 2002). Indeed, there is empirical evidence suggesting that perceived reality is the most important factor in determining workers' attitudes and behavior (Breaux et al., 2009). Perceived organizational politics "involves an individual's attribution of behaviors of self-serving intent and is defined as an individual's subjective evaluation about the extent to which the work environment is characterized by co-workers and supervisors who demonstrate such self-serving behavior" (Ferris et al., 2000: 90). Perceived organizational politics is sometimes cast as a "hindrance" or threatening form of stressor in that it constrains an individual's belief in their ability to achieve personal and professional goals (Chang et al., 2009; Lepine et al., 2005). Recent meta-analytic reviews provide strong evidence supporting the view that perceptions of organizational politics are related to reduced job satisfaction, organizational commitment, citizenship behavior, task performance and increased psychological strain (Chang et al., 2009; Miller et al., 2008).

Given the strong evidence linking perceptions of organizational politics (POP) with a variety of negative outcomes for individuals and their employers, it is critically important for organizations to attend to those factors that contribute to the development of these perceptions (Chang et al., 2009). In their extensive review of the literature examining antecedents of politics perceptions, Ferris et al. (2002) conclude that most of the antecedents have been substantiated by research. However, many of the antecedents included in the Ferris et al. (2002) review had only been examined a few times and half of the variables examined had only been investigated one time (see Table 3, pp. 196-197). Further, Ferris et al.'s (2002) literature review indicates inconsistency in findings in some of the cases where there is more than one study, including: centralization, formalization, hierarchical level, job autonomy, feedback, and a variety of personal influences (i.e., age, gender, organization tenure, minority status, and locus of control). Others also note these inconsistencies (e.g., Adams et al., 2008; Kacmar and Baron, 1999; Miller and Nichols, 2008). These types of inconsistencies may be due to low statistical power or may occur because there truly is no relationship between the antecedents and perceptions of organizational politics (Parker et al., 1995). Therefore, because of the small number of studies available for Ferris et al. (2002) to review and because Ferris et al. (2002) use a vote counting procedure to integrate the empirical literature (i.e., counting statistically significant and nonsignificant findings), it is likely that Ferris et al. (2002) may have reached the wrong conclusions in some cases (Hunter and Schmidt, 2004). Further, Hunter and Schmidt state that even "where the voting method correctly leads to the conclusion that an effect exists, the critical question of the size of the effect is still left unanswered" (2004: 446). Indeed, Vigoda notes that the question of which antecedents of perceived organizational politics "are the most significant remains an open one" (2003: 36).

One way to address this shortcoming in the literature is to conduct a comprehensive meta-analytic examination of the relationships proposed in the antecedent model of perceived organizational politics. Because there has been a significant amount of research performed in recent years, a meta-analytic integration of the literature would be based upon substantially more studies than those reviewed by Ferris et al. (2002). Furthermore, because a quantitative integration of the extant empirical research would provide an estimate of the true population effect size for each predictor of organizational politics, the results would indicate the relative strength and consistency of the different antecedents and categories of antecedents proposed in the theoretical model (i.e., organizational influences, job/work environment influences, and personal influences). Additionally, a meta-analytic integration of the extant literature would provide a quantitative review of the new antecedents Ferris et al. (2002) added to their Revised Model of Organizational Politics Perceptions (e.g., career development opportunities, participation/involvement, person-organization fit, positive affect, negative affect), as well as other antecedents which had only been examined once or twice at the time of Ferris et al.'s (2002) literature review (e.g., job autonomy, skill variety, Machiavellianism, self-monitoring, locus of control, met expectations). In short, a meta-analytic integration of the literature examining antecedents of perceptions of politics would not only update Ferris et al.'s (2002) review of the literature, but also provide new insights into those factors that are thought to influence the development of perceptions of organizational politics.

ANTECEDENTS OF PERCEPTIONS OF ORGANIZATIONAL POLITICS

The variables included in this quantitative review of the literature are those antecedents identified in the general theoretical model presented in early POP research efforts as well as variables identified in later research (e.g., Ferris et al., 1989; Ferris and Kacmar, 1992; Ferris et al., 2002). As noted by Parker et al. (1995), the antecedent model is largely based upon theoretical models of conditions thought to contribute to political behavior (e.g., ambiguity, uncertainty, the belief that political behavior is instrumental in gaining advantage or rewards, and workplace norms for political behavior) (Parker et al., 1995).

Organizational Influences

According to early theoretical models (e.g., Ferris et al., 1989; Ferris and Kacmar, 1992; Parker et al., 1995), there are a variety of organizational influences upon perceptions of organizational politics including centralization, formalization, hierarchical level, and span of control. Centralization, the extent to which power and control is concentrated in the upper echelons of the organization, is thought to influence perceptions of organizational politics because it may foster perceived lack of control and high levels of political behavior aimed at influencing key decision-makers. Research tends to support this view (e.g., Andrews and Kacmar, 2001; Chang, 2008; Muhammad, 2007). Formalization refers to the extent to which instructions, rules, and standards are written and clearly expressed to employees. Organizations with a high degree of formalization tend to have employees who have high role clarity and increased knowledge and control over their environment which should lead to lower perceptions of politics (Ferris and Kacmar, 1992). Research supports this hypothesized negative relationship (e.g., Andrews and Kacmar, 2001; Aryee et al., 2004; Chang, 2008; Liu, 2002; Muhammad, 2007; Vigoda, 2001; Yang, 2009). Hierarchical level within the organization is also thought to influence perceptions of organizational politics because political behavior is traditionally perceived to be an upper management phenomenon, or even part of the job for high level managers (Drory, 1993). Although research has produced mixed results for this hypothesis, with correlation estimates ranging from positive (e.g., Chang, 2008; Miller and Nichols, 2008) to negative (e.g., Valle and Perrewe, 2000), Ferris et al. (2002) concluded that hierarchical level should remain in the model as an organizational influence upon politics perceptions. Ferris and Kacmar (1992) also hypothesized that span of control would be positively related to perceptions of organizational politics. The assumption is that as the number of employees reporting to a supervisor increase, the supervisor will devote less time to each individual, which may increase employees' sense of uncertainty and ambiguity. However, research has found no relationship between span of control and politics perceptions (e.g., Ferris and Kacmar, 1992; Valle and Perrewe, 2000).

Procedural justice, or the perceived fairness of the procedures to allocate resources, has also been discussed as an organizational influence upon perceptions of politics (e.g., Doucet et al., 2009). When the procedures used to make decisions are perceived to be "fair" (i.e., consistent, free of bias, correctable, and ethical in nature), individuals tend to believe that there is less room for ad hoc decision-making due to the presence of decision-making guidelines and increased transparency. Consequently, when individuals perceive there to be procedural justice, they tend to feel they have a greater degree of control (Thibaut and Walker, 1975), which should reduce the perception of politics within the organization (Doucet et al., 2009). Research consistently reports a negative relationship between procedural justice and perceptions of organizational politics (e.g., Andrews and Kacmar, 2001; Aryee et al., 2004; Muhammad, 2007).

Job/Work Environment Influences

Some early research efforts show that job-related factors such as autonomy, feedback, interaction with supervisor/coworkers, and opportunity for advancement are more strongly related to perceptions of organizational politics than either organizational or personal influences (e.g., Ferris and Kacmar, 1992). Ferris and colleagues (e.g., Ferris et al., 1989; Ferris and Kacmar, 1992) originally proposed several job design characteristics as effective ways to reduce uncertainty in the workplace and, consequently, perceptions of politics. A lack of autonomy or skill variety would imply that an …