<psyRA JAPP 2012 5>
<Bottom-Line Mentality as an Antecedent of Social Undermining and the Moderating Roles of Core Self-Evaluations and> <Conscientiousness>
<Rebecca L. Greenbaum Mary Bardes Mawritz Gabi Eissa>
Abstract
We propose that an employees bottom-line mentality may have an important effect on social undermining behavior in organizations. Bottom-line mentality is defined as 1-dimensional thinking that revolves around securing bottom-line outcomes to the neglect of competing priorities. Across a series of studies, we establish an initial nomological network for bottom-line mentality. We also develop and evaluate a 4-item measure of bottom-line mentality. In terms of our theoretical model, we draw on social-cognitive theory (Bandura, 1977, 1986) to propose that supervisor bottom-line mentality is positively related to employee bottom-line mentality (Hypothesis 1). On the basis of conceptual arguments pertaining to bottom-line mentality (Callahan, 2004; Wolfe, 1988), we hypothesize that employee bottom-line mentality is positively related to social undermining (Hypothesis 2). We further predict a moderated-mediation model whereby the indirect effect of supervisor bottom-line mentality on social undermining, through employee bottom-line mentality, is moderated by employee core selfevaluations and conscientiousness (Hypothesis 3). We collected multisource field data to test our theoretical model (i.e., focalsupervisor coworker triads; N 113). Results from moderated-mediation analyses provide general support for our hypotheses. Theoretical and practical implications of bottom linementality and social undermining are discussed, and areas for future research are identified.
Social undermining in the workplace is defined as behavior intended to hinder, over time, the ability to establish and maintain positive interpersonal relationships, work-related success, and favorable reputation ( Duffy, Ganster, & Pagon, 2002, p. 332). Examples of social undermining behaviors include (a) delaying the work of coworkers to make them look bad or slow them down, (b) competing with coworkers for status and recognition, and (c) giving coworkers incorrect or misleading information about the job. Extant research has shown that social undermining is related to unfavorable outcomes on the part of victims and the organization. Employees who are socially undermined are more likely to experience somatic complaints, depression, and reduced self-efficacy ( Duffy et al., 2002; Duffy, Ganster, Shaw, Johnson, & Pagon, 2006). Victims are also more likely to engage in behaviors that are undesirable for the organization (e.g., increased counterproductive work behaviors, reduced job performance). Despite mounting evidence of the potentially damaging effects of social undermining, little research has examined why employees might intentionally try to hinder the success of coworkers (for an exception, see Duffy, Shaw, Scott, & Tepper, 2006). By further investigating this line of inquiry, organizations will have a better understanding of how to prevent social undermining in the workplace.
Employees may engage in social undermining because of the way they view priorities at work. Employees who adopt a one-dimensional frame of mind that revolves around bottom-line outcomes are apt to neglect competing organizational priorities, a phenomenon that we refer to as bottom-line mentality (BLM; Wolfe, 1988). These employees tend to treat every situation as if the bottom line is the only relevant outcome. In turn, employees with BLMs run the risk of approaching the bottom line with a high level of competivenessas if it were a game to be won ( Callahan, 2004; Wolfe, 1998). When employees operate with a one-dimensional, winloss mentality that revolves around bottom-line outcomes, they are inclined to view anyone as an opponent. The only way for them to win is to see others lose; thus, they may feel threatened by the success of coworkers. Accordingly, employees with BLMs may become so eager to secure bottom-line outcomes that they engage in social undermining as a way of helping coworkers fail.
Despite conceptual work that has alluded to the potentially dysfunctional nature of bottom-line thinking ( Barsky, 2008; Callahan, 2004; Kerr, 1975; Shah, Friedman, & Kruglanski, 2002; Shah & Kruglanski, 2002; Sims, 1992; Sims & Brinkman, 2002; Wolfe, 1988), the idea of a BLM has not been fully integrated into the organizational behavior literature. Although some empirical research has provided tangential evidence of an association between bottom-line thinking and unethical behavior ( Schweitzer, Ordez, & Douma, 2004; Tenbrunsel & Messick, 1999), BLM, as a construct of interest, has not been fully conceptualized, nor is there an established measure to capture this phenomenon. Additionally, it is unclear why employees might adopt a BLM in the first place and whether having such a mentality will lead to undesirable behavior, such as social undermining.
To address these gaps in the literature, we first provide a conceptualization and definition of BLM. Second, we draw on social-cognitive theory ( Bandura, 1977, 1986) to argue that a supervisor's BLM may prompt employees to mimic this frame of mind by adopting a BLM themselves. Third, we rely on conceptual work pertaining to BLMs ( Callahan, 2004; Sims, 1992; Sims & Brinkman, 2002; Wolfe, 1988) to argue that employees' BLMs may foster competitive, gamelike thinking that encourages adversarial relationships among coworkers. Accordingly, employees may engage in social undermining as a way of hindering the success of coworkers. Fourth, we argue that the effects of BLMs on social undermining may vary depending on employees' personality characteristics (viz., core self-evaluations [CSEs], conscientiousness). Fifth and last, we further develop the BLM concept by comparing it to related constructs and by developing and assessing a BLM measure.
Theoretical Background and Hypotheses
Bottom-Line Mentality: Definition and Conceptualization
The term bottom line is informally defined as profits or losses, as of a business or the basic or most important factor, consideration, meaning, etc. ( Webster's New World College Dictionary, 2000 , p. 172). Wolfe (1988), too, noted that the bottom line is often referred to in terms of financial consequences, but in a more general sense, it refers to whatever is worth paying attention to while everything else is discarded (p. 145). Although the interpretation of the bottom line may be construed as profits, the business, or other bottom-line outcomes, the primary tenet of a BLM aligns most closely with Wolfe's description of employees almost exclusively focusing on a particular factor or consideration that is identified as being most important, while the importance of everything else is minimized. In line with this notion, we specifically define BLM as one-dimensional thinking that revolves around securing bottom-line outcomes to the neglect of competing priorities.
Securing certain bottom-line outcomes is normally considered beneficial to organizational profitability ( Brenner & Molander, 1977; Trevio, Brown, & Hartman, 2003; Trevio, Hartman, & Brown, 2000) and is often a prerequisite for employee rewards ( Crotts, Dickson, & Ford, 2005; Drucker, 1963; Latham & Locke, 2007; Pringle & Longenecker, 1982). However, a focus on the bottom line may become problematic when employees treat it as if it is the only objective to strive for ( Wolfe, 1988). This is particularly true given that organizations are multivalent systems. Organizations have an interest in upholding multiple objectives and values that have different effects on different stakeholders. Yet the adoption of a BLM encourages simplistic thinking whereby employees treat every situation as if only one objective were relevant. In the sole pursuit of a single outcome, employees pay little attention to whether their actions have an effect on competing organizational priorities. Even the priorities of producing high-quality work and/or treating others properly may become muted as employees fixate on bottom-line outcomes. For example, employees might become so focused on meeting bottom-line productivity requirements that they cut corners without considering the quality of their work or the ethical consequences of their behaviors, including whether their behavior could harm another individual (e.g., potentially releasing dangerous products to customers).
Furthermore, when employees adopt a BLM in multivalent situations, they run the risk of approaching the bottom line as if it were a game to be won ( Wolfe, 1988). The only way they can win is by securing the bottom line; any other outcome is considered a loss. In turn, these employees approach the bottom line as if their actions have limited consequencesthe only relevant outcome is winning the bottom line, while the real consequences of their actions or the real people who may be harmed remain psychologically removed from their minds. By approaching the bottom line with a one-dimensional, winlose mentality, employees are apt to see just about anyone as an opponent who could interfere with bottom-line attainment. Thus, an employee's BLM may foster adversarial relationships among members of the organization. As noted by Levinson (1970), when people place substantial emphasis on quantifiable outcomes, they tend to exclude more subtle, but equally important, initiatives such as team cohesiveness.
Supervisor Bottom-Line Mentality and Employee Bottom-Line Mentality
An employee's adoption of a BLM may stem from his or her supervisor's BLM. This notion is supported by Bandura's (1977, 1986) social-cognitive theory. Social-cognitive theory explains human behavior in terms of observational learning and emphasizes that people learn through the observation and modeling of behaviors and attitudes of others ( Bandura, 1977). The theory further suggests that individuals are more likely to adopt a modeled behavior if it results in outcomes they value and/or if the role model is considered credible.
In the context of organizations, social-cognitive theory implies that subordinates learn what is expected of them through vicarious means ( Bandura, 1977, 1986). They observe the attitudes and behaviors of other organizational members (e.g., supervisors, coworkers) and use this information to construct their own realities. Social influences serve as a catalyst for subordinates to develop and modify their beliefs and values. Because of their assigned roles, supervisors usually have a considerable amount of influence over the attitudes and behaviors of their subordinates through this process of observational learning. This power to influence subordinates stems from the supervisors' visibility within the organization (i.e., subordinates' awareness of the supervisor's power and position) and their ability to control subordinates' resources (e.g., rewards, promotions, favorable work assignments; Hickson, Hinings, Lee, Schneck, & Pennings, 1971; Salancik & Pfeffer, 1978). These contingencies of power force subordinates to attend to their supervisors' attitudes and behaviors and look to their supervisors for cues on organizational attitudinal and behavioral norms ( Berscheid, Graziano, Monson, & Dermer, 1976). Additionally, supervisors are often deemed as credible role models within organizations ( Brown, Trevio, & Harrison, 2005), which further adds to the likelihood that supervisors will gain their subordinates' attention. Thus, employees direct their attention to their supervisors' attitudes and behaviors and acquire knowledge regarding acceptable work-related values, beliefs, and behaviors.
The notion that employees imitate their supervisors is consistent with a growing body of research on trickle-down models that link the attitudes and behaviors of higher level managers to employees' attitudes and behaviors through immediate supervisors. This research has mainly focused on how positive aspects of management, such as charismatic leadership and ethical leadership, trickle down from higher levels of management to lower level employees (e.g., Bass, Waldman, Avolio, & Bebb, 1987; Mayer, Kuenzi, Greenbaum, Bardes, & Salvador, 2009), but has also recently begun to examine how negative aspects of leadership, such as organizational injustice and the violations of psychological contracts, have similar effects (e.g., Aryee, Chen, Sun, & Debrah, 2007; Hoobler & Brass, 2006; Tepper, Duffy, Henle, & Lambert, 2006). Much of this work draws on social-cognitive theory, as we do here, suggesting that employees tend to engage in observational learning and mimic their supervisors' attitudes and behaviors.
When supervisors have BLMs, they give off cues suggesting that bottom-line attainment is more important than competing organizational priorities ( Callahan, 2004; Sims, 1992; Wolfe, 1988). They also behave in ways that support the bottom line. They reinforce bottom-line thinking by rewarding and punishing employees accordingly. In turn, supervisors' attitudes and behaviors concerning the bottom line serve as information for subordinates to create their own rules and strategies for dealing with business situations. Subordinates come to believe that a sole focus on the bottom line is an appropriate attitude to possess. As noted by Sims and Brinkman (2002), when supervisors approach the bottom line as if it were the only objective worth achieving, it is not long before employees think in this way as well. Thus, we expect subordinates to role-model their supervisors by adopting their supervisors' BLMs. On the basis of these arguments, we hypothesize the following:
Hypothesis 1: Supervisor BLM is positively related to employee BLM.
Employee Bottom-Line Mentality and Social Undermining
If employees with BLMs operate with one-dimensional thinking that revolves around bottom-line outcomes ( Wolfe, 1988), then it is possible that they attempt to secure the bottom line with a high level of competitiveness. Employees with BLMs relentlessly approach the bottom line as if it were the only objective worth attaining. In turn, these employees may be willing to engage in any action that brings them closer to securing the bottom line, with little consideration given to the residual effects that may ensue. In such instances, securing the bottom line is treated as if it were a game to be won, while any other outcome is considered a loss.
One problem with approaching the bottom line in terms of winning versus losing is that any other person may be construed as a potential opponent capable of impeding bottom-line success ( Callahan, 2004). Indeed, Wolfe (1988) noted that BLMs may foster significant rivalry among organizational members. Employees with BLMs are apt to believe that the organization must have winners and losers. These employees tend to believe that another's success puts them in the losing category. For them to win, they must attain bottom-line outcomes while others fail to do sotheir success and others' success cannot coexist.
Consequently, employees with BLMs may approach their work with a winner takes all mentality ( Callahan, 2004). Rather than cooperating with colleagues to ensure bottom-line success, these employees tend to believe that whoever contributes most to the bottom line is the winner, thus destroying the notion that everyone is in this together. When employees compete, rather than cooperate, with coworkers in a race to secure bottom-line outcomes, their competitiveness may drive them to take actions that move beyond an initial desire to see coworkers fail ( Wolfe, 1988). As noted by Wolfe (1988), employees who adopt one-dimensional thinking that revolves around bottom-line outcomes are a short step away from wanting others to fail to actually helping them fail by undermining their efforts, withholding information, or making them look bad (p. 149). Thus, an employee's sole pursuit of bottom-line outcomes may come at the expense of others within the organization ( Sims, 1992; Sims & Brinkman, 2002). Given that the success of coworkers, in particular, may be viewed as an obstacle to overcome in pursuit of one's own bottom-line achievement, we expect employees with BLMs to intentionally try to hinder the success of coworkers by engaging in social undermining (e.g., providing coworkers with incorrect or misleading information, delaying work to make coworkers look bad or slow them down, competing with coworkers for status and recognition). On the basis of these arguments, we hypothesize the following:
Hypothesis 2: Employee BLM is positively related to social undermining.
The Moderating Role of Personality
In accordance with social-cognitive theory ( Bandura, 1977, 1986), the effect of supervisor BLM on employee social undermining is expected to occur through role-modeling processes that encourage employees to adopt their supervisors' BLM. Yet the role of employee BLM in the relationship between supervisor BLM and social undermining may vary depending on employees' personality characteristics that dictate how they approach the bottom line. In particular, employees who are high in CSEs and conscientiousness may have the confidence and work ethic to rely on their own merit in securing bottom-line outcomes, whereas those who score low on these factors may have stronger tendencies to engage in social undermining as a way of making themselves look better in terms of bottom-line success.
First, CSE is a higher order personality trait indicated by four individual characteristics including (a) self-esteem, (b) locus of control, (c) generalized self-efficacy, and (d) emotional stability ( Judge, 2009; Judge, Erez, Bono, & Thoresen, 2003; Judge, Locke, & Durham, 1997). CSE, as a broad latent construct, unifies the psychological mechanisms inherent in each trait to capture a person's general self-regard ( Judge et al., 2003). It serves as a basic assessment of one's worthiness, effectiveness, and confidence across situations. People high in CSE tend to think highly of themselves, believe in their ability to complete tasks, and have a strong sense of personal control over their lives. Conversely, people who are low in CSE tend to see themselves as unworthy compared to others ( Judge & Hurst, 2007). They tend to focus on their failures and shortcomings and believe they have little control over their lives.
As a result of variations in self-confidence, an employee's level of CSE may strengthen or weaken the relationship between employee BLM and social undermining. Those high in CSE should believe that they are worthy and capable of securing the bottom line based on their own merit. Extant research suggests that people who are high in CSE tend to be confident in their own agency in handling life's challenges ( Judge et al., 2003; Judge, Locke, Durham, & Kluger, 1998). Their positive sense of self-worth leads them to believe that they will be successful regardless of impeding obstacles ( Kammeyer-Mueller, Judge, & Scott, 2009). In turn, those high in CSE should feel less threatened by the success of coworkers and should be less concerned that coworkers will hinder their own bottom-line success. Because they are confident in their own agency in attaining bottom-line outcomes, they are less likely to find it necessary to hinder a coworker's success for the purpose of making their own bottom line look better. Thus, high levels of CSE are expected to weaken the positive relationship between employee BLM and social undermining.
In contrast, those low in CSE tend to doubt their performance and feel powerless in high-pressure situations ( Judge et al., 1997). A one-dimensional focus on the bottom line may pose as a high-pressure situation whose outcome they have little confidence in obtaining by solely relying on their own merit. This may be particularly true given that they see themselves as victims of their environments ( Judge, Erez, & Bono, 1998; Judge & Hurst, 2007). Rather than having confidence that they can overcome obstacles pertaining to bottom-line achievement, they are likely to conclude that their success depends on external factors, such as coworkers failing. Extant research has also demonstrated that people low in CSE tend to assume the worst when confronted with challenges ( Judge & Hurst, 2007). Accordingly, they may be more likely to conclude that coworkers will oust them from the winning position of attaining bottom-line success. As a result of low confidence, powerlessness, and pessimistic attitudes, when focused on the bottom line, those low in CSE may be more likely than those high in CSE to believe that the bottom line can only be achieved by making others look bad. Thus, low levels of CSE are expected to strengthen the relationship between employee BLM and social undermining.
Second, conscientiousness is a personality trait that captures a person's tendency to be hardworking, efficient, organized, self-disciplined, and dutiful ( Costa & McCrae, 1992; Goldberg, 1990). Extant research has shown that a person's level of conscientiousness affects how he or she approaches his or her job ( Barrick, Mount, & Strauss, 1993; Demerouti, 2006). Those high in conscientiousness are harder workers ( Mount & Barrick, 1998). They tend to be persistent when faced with challenges. Research also suggests that they consider it their duty to do the right thing and abide by ethical principles ( Colquitt, Scott, Judge, & Shaw, 2006; McFerran, Aquino, & Duffy, 2010; Moon, 2001). However, those low in conscientiousness tend to approach their work in a less scrupulous manner ( Moon, 2001). Extant research has shown that they tend to procrastinate and have weaker self-discipline when it comes to fulfilling work responsibilities ( Renn, Allen, & Huning, 2011; Steel, 2007). They are also less concerned about whether their work adheres to ideal standards ( Moon, 2001).
As a result of an employee's level of conscientiousness influencing his or her approach to the bottom line, conscientiousness may moderate the relationship between employee BLM and social undermining. Employees high in conscientiousness are expected to think carefully about how to achieve bottom-line outcomes. Research suggests that their high level of self-discipline allows them to work smarter and harder in attaining desired outcomes ( Mount & Barrick, 1998). In turn, employees high in conscientiousness are expected to stay the course when it comes to bottom-line attainment. Their sense of deliberation, efficiency, and hard work should allow them to achieve bottom-line success without having to resort to social undermining to make themselves look better. Additionally, extant research has shown that those high in conscientiousness approach their work with integrity and attention to ethical principles (J. Hogan & Ones, 1997; Horn, Nelson, & Brannick, 2004; McFerran et al., 2010; Murphy & Lee, 1994). They also consider it their duty to care about the welfare of others ( Moon, 2001). Thus, when focused on the bottom line, employees high in conscientiousness may be less likely to hinder another's success for the purpose of enhancing their own bottom-line outcomes. In line with this notion, we expect high levels of conscientiousness to weaken the positive relationship between employee BLM and social undermining.
In contrast, those low in conscientiousness should be less likely to approach the bottom line with an action plan in place. Despite being focused on bottom-line outcomes, their lack of self-discipline and tendency to be inefficient may make it difficult for them to achieve the bottom line based on their own merit. Furthermore, research has shown that those low in conscientiousness do not work as hard and are more likely to procrastinate (Mount, 1998; Renn et al., 2011; Steel, 2007). A person who is unorganized, undisciplined, and inefficient and who waits until the last minute may respond to the pressure of bottom-line attainment by seeking alternative means to make him- or herself look better. In particular, such a person may have stronger tendencies to conclude that the only way to appear successful is by hindering the success of coworkers. This may be particularly true given that those low in conscientiousness are not as likely to act in accordance with their conscience ( Costa & McCrae, 1992; Goldberg, 1992; R. Hogan & Blake, 1996). Research suggests that they are not as attentive to moral duties ( Colquitt et al., 2006; McFerran et al., 2010; Moon, 2001). Accordingly, they may have stronger tendencies to handle bottom-line pressure by engaging in social undermining without giving much thought to the harm they are causing coworkers. Thus, we expect low levels of conscientiousness to strengthen the positive relationship between employee BLM and social undermining.
We have noted that employee BLM is expected to serve as the process through which supervisor BLM is related to employee social undermining (through observational learning; Bandura, 1977, 1986). Assuming that low (high) levels of CSE and conscientiousness strengthen (weaken) the positive relationship between employee BLM and social undermining, we expect these personality characteristics to conditionally influence the strength of the indirect relationship between supervisor BLM and social undermining. In effect, we expect low (high) levels of CSE and conscientiousness to strengthen (weaken) the likelihood that employee BLM explains the relationship between supervisor BLM and social undermining. Thus, to complete our theoretical model, we predict a pattern of moderated mediation among our study variables as shown in Figure 1. On the basis of the theoretical arguments provided in support of Hypotheses 1 and 2 and the moderator arguments provided above, we specifically hypothesize:
Hypothesis 3: Employee CSEs and conscientiousness moderate the strength of the indirect relationship between supervisor BLM and social undermining via employee BLM, such that the mediated relationship is stronger (weaker) when employee CSEs and conscientiousness are low (high).
Method
Sample and Procedure
We collected data from focal employees, their coworkers, and their immediate supervisors from various organizations located in the southeastern United States in industries including hospitality, retail, accounting, education, manufacturing, banking, and food service. We administered surveys via the Internet. Students were asked to serve as organizational contacts in exchange for extra credit. These students recruited a working adult (defined as working 20 hr per week or more) who was willing to serve as a participant (focal employee). The focal employee then asked his or her supervisor and a coworker to fill out the supervisor and coworker surveys, respectively. A number of researchers have used similar approaches when collecting data (e.g., Grant & Mayer, 2009; Lee & Allen, 2002; Mayer et al., 2009; Morgeson & Humphrey, 2006; Piccolo, Greenbaum, Den Hartog, & Folger, 2010; Skarlicki & Folger, 1997).
Similar to Judge, Scott, and Ilies (2006), we took a number of steps to ensure that the surveys were completed by the correct sources. First, when introducing the study, we emphasized the importance of integrity in the scientific process. We told the participants that it was essential for the focal employee, coworker, and supervisor respondents to fill out the correct surveys. Second, when participants submitted their on-line surveys, time stamps and IP addresses were recorded. We examined these data to ensure that the surveys were submitted at different times and with different IP addresses. We invited 395 students to serve as organizational contacts. We received responses from 147 focal employees, 142 coworkers, and 135 supervisors. After matching data across all three sources, we had usable data from 113 focalcoworkersupervisor triads, for an overall response rate of 29%.
Focal employee respondents were 48.3% male and 50.3% female, and 1.4% did not indicate their sex. Focal employees were 6.8% African American, 8.8% Asian American, 58.5% Caucasian, 15.6% Latino/a, 6.1% Hispanic, 1.4% Native American, 1.4% biracial, and 1.4% other. Focal employees were 59% employed part-time, were 41% employed full-time, had an average age of 25.37 years ( SD = 8.17), and had an average of 3 years of experience with their organization ( SD = 4.15).
Coworker respondents were 46.4% male and 53.6% female. Coworkers were 5.8% African American, 5.1% Asian American, 64.5% Caucasian, 9.4% Latino/a, 8.0% Hispanic, 1.4% Native American, 3.6% biracial, and 2.2% other. Coworker respondents were 30.4% employed part-time, were 69.6% employed full-time, had an average age of 30.79 years ( SD = 11.77), and had an average of 4.39 years of organizational tenure ( SD = 5.93).
Supervisor respondents were 59.1% male and 40.9% female. Supervisors were 7.3% African American, 5.8% Asian American, 65.0% Caucasian, 12.4% Latino/a, 4.4% Hispanic, 1.5% Native American, 0.7% biracial, and 2.9% other. Supervisor respondents were 2.9% employed part-time, were 97.1% employed full-time, had an average age of 38.68 years ( SD = 10.13), and had an average of 9.41 years of organizational tenure ( SD = 8.42).
The focal employee survey contained measures of the focal respondent's BLM, CSEs, conscientiousness, demographics, and negative affectivity and agreeableness as controls. The coworker survey contained measures of the focal employee's social undermining and demographics. The supervisor survey contained measures of the supervisor's BLM, demographics, and the focal employee's job performance as a control.
Measures
Supervisor bottom-line mentality
A series of studies was undertaken to examine the nomological network for BLM and to develop and evaluate a brief, four-item measure (see the Appendix). Accordingly, supervisor BLM was measured using the items created for this purpose. Supervisor respondents rated their own BLM by indicating how much they agreed with the BLM items as shown in the Appendix (1 = strongly disagree, 7 = strongly agree; a = .92).
Employee bottom-line mentality
Employee BLM was also measured using the four-item BLM measure that was developed and evaluated as described in the Appendix. Focal employees rated their own BLM by indicating how much they agreed with the BLM statements (1 = strongly disagree, 7 = strongly agree; a = .86).
Core self-evaluations
CSEs were assessed by the focal employees with 12 items from Judge et al.'s (2003) scale. Sample items include I am confident I get success I deserve in life, When I try, I generally succeed, and I am capable of coping with most of my problems. Responses for these items were made on a 7-point response scale from 1 = strongly disagree to 7 = strongly agree (a = .87).
Conscientiousness
Conscientiousness was assessed by the focal respondents with eight items ( Saucier, 1994) from Goldberg's (1992) measure of the Big Five personality traits. Focal respondents indicated how accurately a number of characteristics described them, including systematic, practical, and efficient. Responses to these items were made on a 7-point response scale from 1 = extremely inaccurate to 7 = extremely accurate (a = .81).
Social undermining
Coworkers rated how strongly they agreed that the focal employee engages in social undermining toward coworkers (1 = strongly disagree, 7 = strongly agree). We slightly adapted Duffy et al.'s (2002) 13-item social undermining scale for this purpose. Sample items include How much do you agree that the employee intentionally (a) delays work to make coworkers look bad or to slow them down, (b) does not give coworkers as much help as he or she promised, and (c) gives coworkers incorrect or misleading information about the job (a = .96).
Control variables
To establish incremental validity of BLM, we included a number of control variables when testing our theoretical model. The negative valence associated with a BLM could be partially capturing an employee's negative predisposition. Thus, we included measures of negative affectivity and (low) agreeableness in our model to decrease the likelihood that the relationship between employee BLM and social undermining is confounded by these individual differences. This is consistent with other deviance-based research that has included personality measures and mood predispositions as control variables because of the influence they could have on social interactions at work and employee outcomes (e.g., Duffy et al., 2002; Mitchell & Ambrose, 2007; Thau, Bennett, Mitchell, & Marrs, 2009; Zellars, Tepper, & Duffy, 2002).
Focal employees rated their negative affectivity by responding to 10 items from Watson and Clark's (1984) negative affectivity scale. They responded to items that included feeling guilt, hostility, and upset (1 = very slightly or not at all, 7 = very much; a = .87). Focal employees also rated their agreeableness by responding to eight items ( Saucier, 1994) from Goldberg's (1992) measure of the Big Five personality. Focal respondents indicated how accurately a number of characteristics described them, including cooperative, rude (reverse-coded), and sympathetic (1 = extremely inaccurate, 7 = extremely accurate; a = .81).
Because extant research has demonstrated that CSE and conscientiousness are related to job performance ( Barrick & Mount, 1991; Judge & Bono, 2001), the variance associated with job performance and each of the personality variables may confound the moderating roles of CSE and conscientiousness in our theoretical model. To account for this potential confounding effect, we included interactions between employee job performance and CSE and between employee job performance and conscientiousness as controls in each of the respective moderated-mediation analyses. Supervisors rated the focal employee's job performance by responding to six items adapted from Alper, Tjosvold, and Law's (2000) performance scale. Sample items include Is concerned about the quality of his/her work and Puts considerable effort into his/her job (1 = strongly disagree, 7 = strongly agree; a = .94).
We also included the focal employee's organizational position as a control variable. An employee's bottom line may be different depending on whether he or she is a nonmanagement employee, supervisor, middle manager, or senior manager. For example, nonmanagement employees might be more likely to think about the bottom line in terms of their own success, whereas senior-level managers might be more likely to think about the bottom line in terms of the entire organization's success. Thus, we controlled for organizational position because it could affect whether an employee's bottom-line success depends on the success of coworkers, which could affect the likelihood of engaging in social undermining. Focal employees indicated whether they were nonmanagement, line management, middle management, senior management, or other. Sixty-eight percent were nonmanagement, 14% line management, 10% middle management, 3% senior management, and 5% other.
Results
Measurement Model Results
To examine the distinctiveness of the study variables, we conducted confirmatory factor analyses with maximum-likelihood estimation in LISREL 8.8 ( Jreskog & Srbom, 2006). The measurement model consisted of eight factors: supervisor BLM, employee BLM, social undermining, CSE, conscientiousness, negative affectivity, agreeableness, and job performance. We randomly combined items for the social undermining, CSE, conscientiousness, negative affectivity, and agreeableness measures to create parcels to maintain a favorable indicator-to-sample size ratio (e.g., Bagozzi & Edwards, 1998; Bagozzi & Heatherton, 1994). The results indicated the eight-factor model provided a good fit of the data (see Table 1).
We compared the eight-factor model to seven alternative models. The first seven-factor model had the items of supervisor and employee BLM loading on the same factor. The second seven-factor model had the items of supervisor BLM and employee job performance (both rated by the supervisor) loading on the same factor. The six-factor model had the items of employee BLM, CSE, and conscientiousness (all rated by the focal employee) loading on the same factor. The four-factor model had the items of employee BLM, CSE, and conscientiousness and the employee-rated control variables (negative affectivity, agreeableness) loading on the same factor. The three-factor model was the same as the four-factor model except that the items of supervisor BLM and employee job performance (rated by the supervisor) were set to load on one factor. The two-factor model was the same as the three-factor model except that social undermining was added to the supervisor BLM and job performance factor. The one-factor model had all items loading onto one factor. A change in the chi-square test indicated the eight-factor model produced a significant improvement in chi-squares over the alternative models (see Table 1).
Descriptive Statistics
The means, standard deviations, and intercorrelations among the study variables are presented in Table 2.
Tests of Hypotheses
We tested the entire moderated-mediation model using a method described by Preacher, Rucker, and Hayes (2007). We utilized an SPSS macro created by Preacher et al. (2007) to run regression equations to estimate mediator variable and dependent variable models. The mediator variable models regressed the mediator (employee BLM) onto the independent variable (supervisor BLM) and the controls. The dependent variable models regressed the dependent variable (social undermining) onto the independent variable, the mediator, the controls, the moderators (CSEs, conscientiousness), and the interactions of the moderators with the mediator variable. We ran these analyses two times, once for each moderator.
The results for all hypotheses are presented in Tables 3 and 4. Consistent with Hypotheses 1 and 2, supervisor BLM was positively related to employee BLM ( B = 0.22, p < .01), and employee BLM was positively related to social undermining for the model with CSEs ( B = 1.10, p < .01) and conscientiousness ( B = 1.28, p < .001). To find support for Hypothesis 3, we first examined the moderating role of CSE ( B = -0.19, p < .01) and conscientiousness ( B = -0.22, p < .01) on the relationship between employee BLM and social undermining. The results suggest that CSE and conscientiousness moderate the relationship between employee BLM and social undermining. Next, we examined the conditional indirect effects of supervisor BLM on social undermining through employee BLM at three values of employee CSE and conscientiousness (one standard deviation below the mean, the mean, and one standard deviation above the mean). As shown in Tables 3 and 4, the conditional indirect effects were significantly different from zero at one standard deviation below the mean and became progressively weaker and nonsignificant at the mean and one standard deviation above the mean. These results indicate support for Hypothesis 3, such that the indirect effects of supervisor BLM on social undermining through employee BLM were stronger at low levels of employee CSE and conscientiousness but were weaker and not statistically significant when the moderators were high. The regions of significance for the indirect effect of supervisor BLM on social undermining at the levels of the moderators are shown in Figures 2 and 3.
To provide further support for Hypothesis 3, we examined the form of the interactions to confirm that they are consistent with the hypothesized pattern. We plotted the interactions with low values at one standard deviation below the mean, values at the mean, and high values at one standard deviation above the mean. As shown in Figures 4 and 5, the role of employee BLM onto social undermining is stronger (weaker) at low (high) levels of CSE and conscientiousness. The simple slopes of social undermining onto employee BLM with low employee CSE, t(109) = 3.39, p < .01, and conscientiousness, t(109) = 3.95, p < .01, were positive and statistically significant, whereas the slopes of social undermining onto employee BLM with high employee CSE, t(109) = -0.65, ns, and conscientiousness, t(109) = -1.34, ns, were not statistically significant.
Discussion
The present research offers a novel theoretical perspective by introducing BLM as a potential antecedent of social undermining in organizations. In particular, we have drawn on social-cognitive theory ( Bandura, 1977, 1986) and arguments pertaining to BLM ( Callahan, 2004; Wolfe, 1988) to explain how supervisor and employee BLMs may trigger employees to engage in social undermining behavior. Furthermore, we have contributed to the social undermining literature by explaining when BLM may be related to social undermining by examining the role of personality in mitigating/strengthening this relationship. More specifically, we have identified CSEs and conscientiousness as personality traits that may alter the way employees approach the bottom line and thus moderate the effect of BLM on employees' social undermining behavior. In introducing our model, we have also contributed to the literature by further refining the BLM construct. Additionally, we have introduced an initial nomological network for BLM and have developed and evaluated a brief BLM measure.
Implications for Theory
Social-cognitive theory ( Bandura, 1977, 1986) suggests that employees may alter their values and beliefs to align with those of a credible role model. By the nature of their assigned role as boss, supervisors are often deemed credible role models within organizations ( Brown et al., 2005). Thus, employees look to their supervisors to learn what is expected of them in terms of both attitudes and behaviors. Supervisors who are preoccupied with the bottom line may communicate this frame of mind to subordinates. Supervisors may engage in behaviors that serve to maintain the bottom line, such as rewarding and punishing employee behavior in favor of maintaining the bottom line. In turn, employees may develop their own BLMs.
Conceptual arguments pertaining to BLM provided the theoretical basis for explaining the positive and significant relationship between employee BLM and social undermining. Wolfe (1988) noted that BLMs create a sense of tunnel vision in which the bottom line is the only value to be considered. When employees approach the bottom line as if it were the only objective to strive for, they run the risk of treating bottom-line attainment as if it were a game to be won. By thinking with a one-dimensional, winlose mentality, employees may become psychologically removed from the real consequences of their behaviors. They may pay little attention to whether their actions harm other people; instead, their primary focus is on winning. Such a mentality is apt to breed a high level of competiveness whereby just about anyone is seen as a potential opponent ( Callahan, 2004; Sims, 1992; Sims & Brinkman, 2002; Wolfe, 1988). For an employee with a BLM to win, the employee must see others lose. Accordingly, the employee may view the success of coworkers as threatening to his or her own bottom-line success. Thus, employees with BLMs may intentionally try to hinder the success of coworkers by engaging in social undermining.
We have further contributed to the literature by determining how and when supervisor BLM is related to social undermining. This relationship is explained by role-modeling processes whereby employees adopt their supervisor's BLM and an employee's BLM subsequently encourages him or her to engage in social undermining. However, an employee's BLM may be less likely to mediate this relationship in some cases. In particular, employees' level of CSEs and conscientiousness may capture variations in how employees approach the bottom line. Those high in CSE and conscientiousness are expected to approach the bottom line by relying on their own agencythose high in CSE tend to be more confident, and those high in conscientiousness tend to work harder and uphold integritythus making them less likely to engage in social undermining as a way of improving the appearance of their own bottom line. Conversely, as a result of low confidence, powerlessness, and pessimism (i.e., low CSE) and a tendency to be unorganized, to be undisciplined, and to procrastinate (i.e., low conscientiousness), those focused on the bottom line and low in these characteristics are expected to have stronger tendencies to engage in social undermining as a way of making themselves appear more successful. In this way, we have contributed to the literature by demonstrating that not all employees who adopt a supervisor's BLM will pursue the bottom line by engaging in social undermining. Rather, the role of employee BLM in explaining the relationship between supervisor BLM and social undermining is stronger (weaker) when levels of CSE and conscientiousness are low (high).
Furthermore, we have contributed to the social undermining literature by examining relevant antecedents. Although some research has examined the consequences of social undermining ( Crossley, 2009; Duffy et al., 2002; Duffy, Ganster, et al., 2006), comparatively less research has focused on the antecedents of social undermining (for an exception, see Duffy, Shaw, et al., 2006). Extant research has shown that social undermining can have dire consequences in terms of employees' psychological health and well-being and organizational productivity ( Duffy et al., 2002). Thus, it is important to understand factors that may be associated with employees engaging in undermining behavior. Our research suggests that BLM, a construct that has received little empirical attention in the literature, may serve as an antecedent of social undermining.
Implications for Practice
Organizations should be aware that employees may emulate their supervisors' BLMs. Although this may seem beneficial in terms of attaining desired outcomes (e.g., profit maximization), it can also be quite dysfunctional. It is particularly important for supervisors to emphasize that appropriate behavior is a necessary bottom-line expectation. This goes along with initiatives related to a triple bottom line (i.e., a concern for social, environmental, and financial performance) that emphasizes managing responsibly as well as profitably ( Waddock, Bodwell, & Graves, 2002).
Organizations should also be aware that employees with BLMs may engage in social undermining. Social undermining can have dire consequences for those involved ( Duffy et al., 2002) and can be quite costly to organizations ( Tepper, 2007). Therefore, organizations should communicate to employees that even though the bottom line is important, so is the way in which it is achieved. Ethical leaders, for example, care about maintaining the bottom line ( Trevio et al., 2000), yet they define success not just by results but also by the way results are obtained ( Brown et al., 2005). Thus, organizations may be able to avoid negative outcomes associated with BLMs by employing leaders who enforce ethical compliance while also working to attain bottom-line outcomes.
Limitations and Future Directions
One limitation of our study is that it was cross-sectional, and thus, we could not determine the direction of causality among the variables. Our conceptualization of the relationships of interest is consistent with research on trickle-down models of leadership (e.g., Aryee et al., 2007; Mayer et al., 2009). In addition, research has suggested that due to formal power and authority being associated with higher organizational positions, the likely causal direction is that supervisor attitudes predict employee attitudes (e.g., Yukl, 1998). However, it is possible that attitudes in organizations may also flow in an upward direction, with employee attitudes predicting supervisor attitudes ( Yukl & Falbe, 1990, 1991). Future research that examines the relationship between supervisor and employee BLMs longitudinally is needed to explore the causal direction of this relationship.
Additionally, another alternative explanation for the relationship between supervisor and employee BLMs lies in the attractionselectionattrition (ASA) process ( Schneider, 1987; Schneider, Goldstein, & Smith, 1995). The ASA framework suggests that (a) potential employees are attracted to organizations with cultures that match their personalities, (b) organizations tend to select job applicants whose personalities fit with the organizational culture, and (c) employees who do not fit with the culture will voluntarily or involuntarily leave the organization. Hence, this process suggests that supervisors with BLMs may be more likely to select applicants who already possess orientations toward the bottom line. Our understanding of the relationship between supervisor and employee BLMs would benefit from longitudinal studies that account for the ASA process by assessing the BLMs of new employees before prolonged exposure to their supervisors.
Another limitation of our study is that we did not explicitly measure the mechanisms through which supervisor BLM is related to employee BLM. We have drawn on social-cognitive theory and trickle-down models of leadership to suggest that employees imitate the attitudes of their supervisors. However, we did not measure any role-modeling mediators to confirm that this is the underlying process of this relationship. Furthermore, there could be alternative mechanisms that account for the positive relationship between supervisor and employee BLM. Perhaps, when supervisors have BLMs, these mentalities contribute to an organizational environment in which the bottom line is seen as the most important priority. Thus, subordinates perceive that their organization has a strong bottom-line orientation. They recognize that other organizational members are preoccupied with the bottom line and tailor their attitudes accordingly, adopting a BLM themselves. Our understanding of the trickle-down effects of supervisor BLMs would benefit from future research that examines relevant mediators.
Additionally, future research would benefit from an examination of moderators of the proposed trickle-down relationship. For example, employees may be more likely to adopt a supervisor's BLM if they strongly identify with the supervisor, experience moral disengagement in the workplace, and do not internalize morality as part of their identities. Moreover, employees may be less likely to adopt a supervisor's BLM if they can easily change supervisors or leave the organization. Future research would benefit from exploring these potential moderators.
Future research would also benefit from further examining the nomological network of BLM. In addition to examining our theoretical model, we began to establish the nomological network of BLM by examining its association with elements of goal setting, performance orientation (see the Appendix), and individual difference variables (viz., negative affectivity, agreeableness). However, future research should investigate the association between BLM, rewards ( Kerr, 1975), and a focus on business decisions ( Tenbrunsel & Messick, 1999). Kerr (1975) discussed the folly of rewarding one behavior while hoping for another. By solely rewarding bottom-line outcomes and merely hoping for ethical compliance, organizations may inadvertently encourage unethical behavior. Certain business decisions may be associated with employees focusing on what is best for the business without considering ethical implications ( Tenbrunsel & Messick, 1999; Tenbrunsel & Smith-Crowe, 2008). Additionally, future research should examine additional consequences of BLM. It could be that BLM is associated with lying, cheating, financial fraud, and producing low-quality, dangerous products.
Wolfe (1988) conceptualized BLM in terms of one's own assigned bottom line (p. 149). Given that employees' assigned bottom lines are expected to vary, there may be differences in how employees approach the bottom line. Employees in management positions, for example, may be more likely to have assigned bottom lines that are contingent on their subordinates' performance. In turn, they may be less likely to treat employees poorly because they depend on them for rewards. Employees may also think about the bottom line in terms of individual, group, or organizational success depending on reward interdependence and individualistic versus collectivistic cultures. It would be interesting to see how team members treat one another when they collectively possess a BLM. It could be that team members do not perceive intrateam social undermining but that they do perceive interteam social undermining.
Furthermore, future research may benefit from identifying a person's specific bottom line when examining the effects of a BLM. In our research, we remained nonspecific regarding a person's specific bottom line because we were primarily interested in the underlying mentality of the respondents (as captured by a BLM). However, depending on the particular research question, there may be value in specifically identifying the most important factor that employees pay attention to in terms of the bottom line, whether it is profits, the business, or their own bottom-line outcomes. Thus, future research would benefit from further investigating the measurement of BLM, with a focus on assessing a variety of BLM items and asking participants to describe how they understand each of the items.
Conclusion
Business scholars have argued that there could be problems with simplistic thinking that revolves around bottom-line outcomes ( Callahan, 2004; Jones, 1991; Kerr, 1975; Levinson, 1970; Sims, 1992; Sims & Brinkman, 2002; Tenbrunsel & Messick, 1999; Wolfe, 1988). In an effort to fully understand this phenomenon, we examined the nomological network of a BLM. Our results support the notion that in some cases, a BLM can be dysfunctional in that it may be an antecedent of social undermining. In light of the potentially high costs associated with social undermining, we find this research particularly important for both theory and practice. 