Shahidul Hassan is assistant professor
of public management in the John Glenn
School of Public Affairs at The Ohio State
University. His research focuses on how
leadership and managerial practices
in government organizations infl uence
motivation, commitment, and performance
Bradley E. Wright is associate profes-
sor of public management and policy in
the Andrew Young School of Policy Studies
at Georgia State University. His research
focuses on how employee attitudes and
behavior are infl uenced by the interaction
between characteristics of the employee
and the employee’s organizational work
Gary Yukl is professor of management
at the University of Albany, State University
of New York. He is a fellow of the American
Psychological Association, the American
Psychological Society, the Society for
Industrial-Organizational Psychology, and
the Academy of Management. He has pub-
lished many articles in professional journals
and is author of several books, including
Leadership in Organizations (8th
ed., Prentice Hall, 2013). He has received
several awards for his research, including
two lifetime achievement awards: the 2007
Walter Ulmer Applied Research Award from
the Center for Creative Leadership and the
2011 Eminent Leadership Scholar Award
from the Academy of Management Network
of Leadership Scholars.
Does Ethical Leadership Matter in Government? Effects on Organizational Commitment, Absenteeism, and Willingness to Report Ethical Problems 333
Public Administration Review,
Vol. 74, Iss. 3, pp. 333–343. © 2014 by
The American Society for Public Administration.
Shahidul Hassan The Ohio State University
Bradley E. Wright Georgia State University
Gary Yukl University at Albany
Recent ethical scandals involving managers in government organizations have highlighted the need for more research on ethical leadership in public sector organizations. To assess the consequences of ethical leader- ship, 161 managers in a large state government agency and 415 of their direct reports were surveyed, and person- nel records were obtained to measure absenteeism. Results indicate that after controlling for the eff ects of employee characteristics, perceptions of procedural fairness, and supportive leader behavior, ethical leadership reduced absenteeism and had a positive infl uence on organiza- tional commitment and willingness to report ethical problems. Implications of the fi ndings and suggestions for future research are presented.
Public administration scholars have long dis-cussed the importance of managerial ethics from a normative perspective by specifying what public managers should do or how they ought to behave (Cooper 1982; Hart 1974, 1984; Rohr 1989). Hart (1984), for example, proposed that public man- agers should be prudent, trustworthy, and considerate, and their actions should be consistent with public values and interests. Given this emphasis on the moral person, it is not surprising that so many public sector organiza- tions rely on their senior leader- ship to establish and support an ethical climate (Berman, West, and Cava 1994; Bruce 1994; West and Berman 2004). Empirical evidence also sup- ports the importance of ethics for good governance and democracy (Cowell, Downe, and Morgan 2014) because it predicts satisfaction with government services, trust in government, and the amount of citizen participation (Vigoda-Gadot 2007; Villoria, Van Ryzin, and Lavena 2013).
In the United States, recent ethical scandals involv- ing the targeting of conservative political groups for scrutiny, violations of the privacy of journalists and world leaders, and sexual harassment and assault
in the military reveal not only the importance of ethical behavior but also the diffi culty of ensuring it (Alexander and Stewart 2013; Dinan 2013; Horwitz 2013; Shane 2013). In a 2007 national survey of 744 randomly selected government employees in the United States (Ethics Resource Center 2008), 57 per- cent of government employees reported that they had witnessed a violation of ethical standards, policies, or law in their workplace during the previous 12-month period. While ethical violations can take many diff erent forms, commonly cited examples include misreporting hours worked, employment discrimina- tion, sexual harassment, and violations of privacy (Ethics Resource Center 2008; Kaptein et al. 2005). Of those who reported witnessing ethical violations, nearly one-third (30 percent) of government employ- ees in the United States did not report the violation (Ethics Resource Center 2008). Just as troubling is the opportunity and pressure for misconduct. Nearly half (48 percent) of the government employees surveyed reported experiencing situations that they felt invited misconduct, while 14 percent reported that they had been pressured to compromise ethical standards in the course of performing their jobs. Existing evidence
also suggests that the leader- ship in public organizations often fails to achieve the desired ethical standards. For example, the 2012 Federal Employee Viewpoint Survey found that nearly half (45 percent) of federal employees do not believe the leaders in their organiza-
tion maintain high standards of honesty and integrity, while well over one-third (38.5 percent) do not feel that they can disclose a suspected violation of laws or regulations without fear of reprisal (OPM 2012).
Given the importance and complexity of the topic, there have been surprisingly few empirical studies assessing the eff ects of ethical leadership in govern- ment organizations (Beeri et al. 2013; Hassan et al. 2013; Huberts, Kaptein, and Lasthuizen 2007;
Does Ethical Leadership Matter in Government? Eff ects on Organizational Commitment, Absenteeism,
and Willingness to Report Ethical Problems
Existing evidence also suggests that the leadership in public organizations often fails to achieve the desired ethical
334 Public Administration Review • May | June 2014
leaders take into account the moral implications of their choices. Th e last attribute captures the moral manager aspect of ethical leadership. Th at is, ethical leaders are not only honest, trustworthy, and principled decision makers, they also actively promote ethical behavior among their followers by clearly communicating ethical standards and expectations, providing ethical guidance, and holding followers accountable for ethical and unethical conduct (Brown and Treviño 2006; Treviño, Brown, and Hartman 2003).
Infl uence of Ethical Leadership on Employee Attitudes and Behaviors Leadership has long been considered important in discussions on managerial ethics (Barnard 1938). However, ethical leadership as a construct or distinct form of leadership has been proposed only in the last decade (Brown, Treviño, and Harrison 2005), and more needs to be learned about the eff ects on employee attitudes and behaviors that can infl uence organizational performance and integ- rity. Some recent research provides evidence that ethical leadership can infl uence the attitudes and behavior of subordinates. Managers who exhibit higher levels of ethical leadership not only decrease unethical behaviors and increase citizenship behaviors but also infl uence subordinate attitudes about such behaviors (Mayer et al. 2012; Resick et al. 2013; Walumbwa et al. 2011; Walumbwa and Schaubroeck 2009). Ethical leadership can increase follower satis- faction with the leader, the perception of leader eff ectiveness, the quality of the leader-member exchange relationship, organizational commitment, and prosocial behavior, as well as reduce deviant employee behavior (Hassan et al. 2013; Kalshoven, Den Hartog, and De Hoogh 2011; Mayer et al. 2012; Mayer et al. 2009; Resick et al. 2013; Walumbwa et al. 2011; Walumbwa and Schaubroeck 2009; Yukl et al. 2013).
Th e number of relevant studies, however, is still relatively small and conducted primarily in business organizations. While research on the consequences of ethical leadership in public organizations is lim- ited, the studies that do exist suggest that this can be a productive area of research. Miceli and Near (1984, 1985, 1988), for example, found that personal and organizational factors infl uence employee reports of unethical conduct in federal agencies. Consistent with public service motivation theory, Brewer and Selden (1998) showed that public employees with a higher regard for the public interest are more likely to report illegal or wasteful activity in their agen- cies. Moreover, several studies examined the prevalence of diff erent leadership practices and ethic codes and their eff ects on employee ethical behavior in public organizations (Berman, West, and Cava 1994; Bruce 1994; Cowell, Downe, and Morgan 2014; West and Berman 2004). More recently, public sector studies have begun to fi nd a connection between ethical leadership and integrity violations (Kaptein et. al 2007; Kolthoff , Erakovich, and Lasthuizen 2010), quality of work life (Beeri et al. 2013), and aff ective commitment (Hassan et al. 2013). In this article, we aim to contribute to research in public administration by examining how ethical leadership is related to employee willingness to report ethical problems, organiza- tional commitment, and frequency of absences from work.
Ethical leadership and willingness to report ethical problems. As noted earlier, recent national surveys suggest that many government employees who observe workplace misconduct fail to report it (Ethics Resource Center 2008; OPM 2012). When explaining their
Kolthoff , Erakovich, and Lasthuizen 2010). More research is needed to understand the potential eff ectiveness and importance of leader- ship in facilitating ethical behavior and preventing unethical con- duct in government organizations. In this article, we aim to extend the public administration literature by examining how ethical leadership can facilitate positive outcomes in government organi- zations. More specifi cally, we examine the relationships between ethical leadership and the organizational commitment of public sector employees, their absenteeism, and their willingness to report ethical problems. After a brief overview of ethical leadership theories and research fi ndings, we develop a set of testable hypotheses based on prior research and theory. Th en we describe the research meth- ods used in this article and present the fi ndings. We conclude by discussing the theoretical and practical implications of our research fi ndings and avenues for future research on ethical leadership in government organizations.
Literature Review and Hypotheses Ethical Leadership Integrity and honesty have long been considered important deter- minants of leadership eff ectiveness (Kirkpatrick and Locke 1991; Kouzes and Posner 1992; Posner and Schmidt 1992), with Barnard noting as early as 1938 that one function of the executive is to cre- ate a moral framework for the organization. Even so, organizational scholars have only recently begun to adopt a systematic approach to describe what ethical leadership means and to examine the anteced- ents and consequences of ethical leadership in organizations (Brown and Treviño 2006; Brown, Treviño, and Harrison 2005; Treviño, Brown, and Hartman 2003).
Ethical leadership has been described in many ways. Kanungo (2001) noted that ethical leaders engage in behaviors that benefi t others and, at the same time, refrain from behaviors that can cause harm to others. Khuntia and Suar (2004) suggested that ethical leaders incorporate moral principles into their values, beliefs, and actions. Brown, Treviño, and Harrison provided a more comprehen- sive defi nition conceptualizing ethical leadership as “the demonstra- tion of normatively appropriate conduct through personal actions and interpersonal relationships, and the promotion of such conduct through two-way communication, reinforcement and decision- making” (2005, 120).
According to these perspectives, three essential attributes or com- ponents of ethical leadership are (1) being an ethical role model to others, (2) treating people fairly, and (3) actively managing ethics in the organization (Brown and Treviño 2006; Brown, Treviño, and Harrison 2005; Treviño, Brown, and Hartman 2003). Th e fi rst two attributes capture the moral person aspect of ethical leadership. Ethical leaders demonstrate ethical values such as honesty, integ- rity, and altruism (e.g., sacrifi cing personal gains for the benefi t of others), and they conduct themselves in an ethical manner even in the midst of adversity, risks, or pressure (Brown and Treviño 2006; Brown, Treviño, and Harrison 2005; Treviño, Brown, and Hartman 2003). Ethical leaders also serve as ethical role models for other people. Th ese leaders establish themselves as credible and legitimate role models by demonstrating normatively appropriate behaviors and treating others with consideration and respect (Brown and Treviño 2006; Brown, Treviño, and Harrison 2005). While making important decisions that may aff ect the well-being of others, ethical
Does Ethical Leadership Matter in Government? Effects on Organizational Commitment, Absenteeism, and Willingness to Report Ethical Problems 335
the relationship between ethical leaders and followers is likely characterized by social rather than economic exchanges. Economic exchanges are largely impersonal, whereas social exchanges are based on mutual aff ec- tion, trust, and reciprocity (Blau 1964; Gouldner 1960). Because ethical leaders are trustworthy, care about the well-being of their followers, and are fair decision makers, they are likely to develop high-quality relationships
with their followers; this, in turn, is likely to infl uence followers to reciprocate by showing loyalty to the leader and commitment to the work group and organization that they represent. Consistent with these expectations, a recent study by Ko and Hur (2014) showed a positive correlation between manager trustworthiness and the intention of public employees to stay in their organization. A few recent studies also found a positive connection between ethical leadership and subordinate organizational commitment (Beeri et al. 2013; Hassan et al. 2013), but these studies relied on self-reported measures of organizational commitment. To further corroborate the relationship, we assessed manager perception of subordinate com- mitment as well as subordinate self-reported aff ective commitment to test the following hypothesis:
Hypothesis 2: Ethical leadership increases the organizational commitment of public sector employees.
Ethical leadership and absenteeism. Absenteeism is a costly personnel problem for organizations. In addition to its direct fi nancial costs (e.g., statutory sick pay, replacement costs, and overtime costs), absenteeism can lower individual productivity and group performance (Hacket 1989; Tharenou 1993; Viswesvaran 2002). While there is evidence that absenteeism may be higher in the public sector than in private sector organizations (Dibben, James, and Cunningham 2001; Klein 1986; Vandenheuvel 1994), few studies in public administration have investigated how to reduce absenteeism in government organizations (Dalton and Perry 1981; Garcia 1987; Perry and Angle 1980; Perry and Long 1984; Vigoda- Gadot and Meisler 2010; Wright and Pandey 2011).
March and Simon (1958) distinguished two categories of absences: voluntary absences over which employees have some control (e.g., uncertifi ed sickness and vacation) and involuntary absences (e.g., certifi ed sickness and family emergencies). Unfortunately, it is very diffi cult to determine whether an absence is voluntary or involun- tary (Darr and Johns 2008). While both types of absence can be costly to the organization, organizations in the United States put most of their attention on identifying and reducing the costs associ- ated with voluntary or unnecessary absences, especially given the employment protections (e.g., the Americans with Disabilities Act and the Family and Medical Leave Act) that aff ord employees many sources of involuntary absenteeism.
Ethical leadership may have an infl uence on both voluntary and involuntary absences in a number of ways. Unscheduled, voluntary absences from work are often caused by illness or family emergencies, but these absences may also be caused by mistreatment at work,
failure to report misconduct, government employees often noted doubt that management would take appropriate corrective action and a fear of retaliation. Ethical leaders can create a safe organizational climate in which employees feel comfortable discussing ethical issues and reporting ethical problems without fear of retaliation. When people are afraid to voice concerns about ethical problems in their organization, ethical leadership can reduce this fear (Walumbwa and Schaubroeck 2009). When employees have a leader who is honest, trustworthy, and fair, they are more likely to think that the leader will agree with or understand their concerns and respond to them appropriately. They will feel more comfortable discussing sensitive ethical issues and will be more likely to report ethical problems. Quickly discovering and resolving ethical problems can prevent or reduce negative consequences, such as damage to the organization’s reputation, costly lawsuits, and loss of public trust (Victor and Cullen 1988).
Social learning theory suggests that individuals learn about appro- priate behavior by observing the behavior of role models (Bandura 1977, 1986), and managers can serve as legitimate models for nor- mative behavior (Mayer et al. 2012; Mayer et al. 2009). By behav- ing in an ethical manner and holding others accountable for ethical and unethical actions, managers are able to positively infl uence subordinates to avoid unethical actions and to report ethical issues and problems to management. Walumbwa and Schaubroeck (2009) found that ethical leadership promoted employee voice by enhanc- ing the perception of psychological safety, and Brown, Treviño, and Harrison (2005) found that ethical leadership was signifi cantly related to followers’ willingness to report problems. All of these fi nd- ings suggest the following hypothesis:
Hypothesis 1: Ethical leadership increases the willingness of public sector employees to report ethical problems to management.
Ethical leadership and organizational commitment. In addition to increasing willingness to report unethical behavior, ethical leadership can infl uence employee attitudes in benefi cial ways. Brown, Treviño, and Harrison (2005) suggested that ethical leaders are likely to have a positive effect on a subordinate’s commitment to the organization. Organizational commitment refers to emotional attachment to, identifi cation with, and involvement in one’s organization (Meyer and Allen 1991; Porter et al. 1974), and it refl ects agreement with organizational values and goals as well as feelings of personal satisfaction derived from involvement in the organization (Meyer and Allen 1991). Organizational commitment has important implications in terms of decreasing turnover intentions and increasing job performance and organizational citizenship behavior (Cooper-Hakim and Viswesvaran 2005; Mathieu and Zajac 1990; Meyer et al. 2002).
Social exchange theory (Blau 1964; Gouldner 1960) provides insight into how ethical leadership may infl uence followers’ com- mitment to their work group and organiza- tion. Brown and Treviño (2006) noted that
Ethical leaders can create a safe organizational climate in
which employees feel comfort- able discussing ethical issues
and reporting ethical problems without fear of retaliation.
Ethical leadership may have an infl uence on both voluntary and
336 Public Administration Review • May | June 2014
of the managers in their organizations was inversely correlated with their self-reported absenteeism. In a study by Rosenblatt, Shapira- Lishchinsky, and Shirom (2010), an organizational climate empha- sizing a genuine interest in the welfare of others (both inside and outside the organization) was associated with lower absenteeism among teachers. Th ese fi ndings suggest that ethical leadership is an important factor in establishing an ethical climate in public sector organizations (Beeri et al. 2013; Bruce 1994). Consistent with these fi ndings, we test the following hypothesis:
Hypothesis 3: Ethical leadership reduces the frequency of absenteeism for public sector employees.
Research Methods Sample and Procedure Th e study used data collected from personnel records and from two separate surveys conducted in a large agency in state government in the Midwestern United States. Th e agency was responsible for providing procurement, human resources, information technology, and other administrative services to other agencies, commissions, and boards in the state government as well as to local govern- ment organizations. Th e subordinate survey and the supervisor survey were conducted in the summer of 2012 as part of a long- term project undertaken by the agency for leadership training and development. Th e questionnaires for both surveys were distributed and collected electronically. Th e subordinate and supervisor data were matched using a four-digit unique identifi er. During the data collection procedure, the participants were assured repeatedly that no identifying information would be shared by the researchers with anyone inside or outside the agency. Only aggregate results at the division level were included in the fi nal report shared with the agency head and division managers.
Th e subordinate questionnaire was designed to collect data regard- ing ethical leadership, subordinate willingness to report ethical problems to management, and aff ective commitment. First, the
questionnaire was pre-tested with a small group of employees (n = 9) working in the agency’s division of human resources to obtain feedback on the questionnaire as well as to test the electronic survey distribution tool. Next, the agency head and division managers sent an e-mail to all employees to communi- cate the purpose of the study and to explain that participation was voluntary and that all responses would remain confi dential. Th en, the research team contacted all 820 employees
(except the agency head) by sending an e-mail that repeated the explanation of the study and explained that the electronic survey administration tool allowed participants to complete the survey at a time convenient to them during their normal work hours. Th e sur- vey remained open for three weeks, during which time up to three e-mail reminders were sent to boost the response rate. Altogether, 477 usable responses were returned for an overall response rate of 59 percent.
Data regarding subordinate commitment were collected through a separate survey of 176 supervisors. Each supervisor rated the commit- ment of direct reports, who were the subordinates surveyed earlier.
low morale, stress, and a sense of entitlement (CCH 2006; Lach 1999). Ethical leaders exhibit normatively appropriate conduct in their interpersonal relationships as well as their concern for oth- ers (Brown, Treviño, and Harrison 2005). Consistent with social learning principles, the positive and supportive behaviors exhibited by ethical leaders encourage subordinates to behave similarly toward their coworkers (Mayer et al. 2009). As a result, they should be more satisfi ed at work and experience less stress and confl ict. Given that work-related stress can increase stress-induced illness (Nixon et al. 2011) and, to a lesser extent, the avoidance of work (Darr and Johns 2008), ethical leadership may decrease both voluntary and involuntary absenteeism.
Absenteeism and other work withdrawal behaviors (e.g., lateness) can refl ect a lack of job satisfaction and commitment to the organi- zation (Hanisch and Hulin 1991). Sagie (1998) noted that being voluntarily absent from work is consciously or unconsciously show- ing negative emotional attachment to the workgroup and organi- zation. Th e notion of reciprocity in social exchange theory (Blau 1964; Gouldner 1960) explains how ethical leadership can infl uence subordinate absences. Just as positive interactions often lead to posi- tive reciprocity, negative interactions can result in norms of negative reciprocity (Gouldner 1960). Unethical leader behavior such as manipulation, abusive treatment, breach of trust, and unfair treat- ment of subordinates may encourage counterproductive behavior such as lateness, absenteeism, and even turnover. In contrast, ethical leadership is likely to lower such behavior by creating a sense of fair- ness and positive reciprocity. Two recent studies provide empirical support for these claims, fi nding a signifi cant negative connection between ethical leadership and deviant or unethical behavior in work groups (Mayer et al. 2012; Mayer et al. 2009).
Ethical leadership can also reduce the improper use of sick leave (Rosenblatt, Shapira-Lishchinsky, and Shirom 2010). To the extent that the absence represents work avoidance or shirking behavior, it is likely to violate shared organizational or societal norms and standards of conduct. Taking time off from work unnecessarily represents a cost to the organization when it delays important projects, disrupts the work of colleagues, and reduces the quality of customer service. Furthermore, employees who use their sick leave to avoid work or extend their vacations often engage in acts of deceit or dishonesty to request or explain their absence. While not all absenteeism represents dishonest or selfi sh behavior, national surveys fi nd that 31 percent of employees admit that they have used sick leave when they were not sick (Lach 1999) and that personal illness may only account for 35 percent of unscheduled absences (CCH 2006). To the extent that absenteeism represents misconduct, ethical leaders can reduce subordinate absenteeism by clearly communicating and modeling ethical standards and by using the organization’s reward system to reinforce those standards.
Th e eff ect of ethical leadership on subordinate absenteeism has not been examined, but indirect support for a relationship is provided by two studies. In a national workforce survey, Prottas (2013) found that employee perception regarding the behavioral integrity
Taking time off from work unnecessarily represents a cost
to the organization when it delays important projects,
disrupts the work of colleagues, and reduces the quality of
Does Ethical Leadership Matter in Government? Effects on Organizational Commitment, Absenteeism, and Willingness to Report Ethical Problems 337
Th e internal reliability of the ethical leadership measure was high (Cronbach’s alpha = 0.96).
Willingness to report ethical problems was measured with a single item in the subordinate survey: “I feel comfortable reporting ethical problems to upper management.” Th e item had a six-point Likert- type response format (1 = strongly disagree, 6 = strongly agree). As in previous studies, we measured only the respondent’s intention to report (as opposed to actual reporting behavior) and did not pro- vide specifi c examples of ethical problems. Earlier studies (Brown, Treviño, and Harrison 2005; Caillier 2012) asked employees only about failures to report ethical problems out of fear of retaliation. However, because there are other possible reasons for failing to report ethical problems (e.g., do not think the problem is impor- tant, do not know how to report the problem, do not believe that management can or will do anything to resolve the problem), we did not specify a reason.
Aff ective commitment was measured with four items adapted from the Organizational Commitment Scale developed by Meyer and Allen (1991). Th ese four items measure an employee’s emotional attachment to the work unit, and each item has a six-point Likert- type response format (1 = strongly disagree, 6 = strongly agree). Sample items include the following: (1) “I feel a strong sense of belonging to my unit”; and (2) “I really feel as if my unit’s problems are my own.” Cronbach’s alpha for this measure was 0.83.
Overall employee commitment was measured with three items that asked supervisors to rate an employee’s overall commitment to the job, unit, and organization. Th ese three items were developed by the research team for the purpose of this study. Each item had a fi ve-point response format (1 = not at all committed, 5 = committed to a large extent). Th e three items were as follows: (1) “How would you rate this subordinate’s overall commitment toward his/her job?”; (2) “How much is this subordinate willing to contribute toward the success of the unit?”; and (3) “How much is this subordinate willing to contribute toward the success of the agency?” Cronbach’s alpha for this measure was 0.95.
To measure each employee’s absenteeism during the past nine months, data were obtained from personnel records of the state agency. Th e human resource department of the agency provided disaggregated absenteeism records (number of absence hours as well as the reason for those absences). Th ese records did not include employee vaca- tion hours during the nine-month period. Th e absenteeism records were matched with subordinate and supervisor survey responses with a four-digit identifi er. In calculating total absences, we excluded the number of work hours that employees were absent because of a hospital stay, disability, workers’ compensation, and wellness visits. We also excluded sick hours that were donated by employees. Our measure only included subordinate absences that were not attribut- able to unavoidable personal and family issues; it is comparable to the measure used by Vigoda-Gadot and Meisler (2010).
We measured supportive behavior by the leader and procedural fair- ness in order to control for their eff ects on the dependent variables. Research has found that employee willingness to report ethical prob- lems is infl uenced by perceptions of procedural fairness (cf. Victor, Treviño, and Shapiro 1993) and supervisor support (Caillier 2012).
For supervisors who had more than fi ve direct reports, the research team randomly selected fi ve subordinates for the supervisor to rate. Th is limitation resulted in a small reduction in the sample size from 477 to 443 respondents, a retention rate of approximately 93 percent. Th e supervisors were provided two weeks to complete the surveys, and they had the option to complete the surveys at diff erent times for diff erent subordinates. Similar to the subordinate survey, up to three individualized e-mail reminders were sent to the supervisors to boost the response rate. In all, 161 of the 176 supervisors (91 per- cent) returned completed surveys for their subordinates. Overall, we received both subordinate and supervisor data for a total of 415 out of 443 possible pairings. Absenteeism data were obtained from agency records for the nine-month period after the subordinate survey.
An overview of the demographic data for the sample is provided in table 1. A large majority of the sample (78 percent) identifi ed themselves as Caucasian, and nearly half of the respondents (45 per- cent) were women. A majority of the responding subordinates were technical/professional employees (56 percent), and the remainder included clerical/support employees (15 percent) and managers/ supervisors (29 percent). As shown in table 1, the average age of the subordinates was between 41 and 50 years. Tenure in current posi- tion ranged from 1 year to 35 years, with a mean of 6 years. Tenure in agency ranged from 1 year to 38 years, with a mean of 11 years.
Measures Ethical leadership was measured with nine items from the Ethical Leadership Questionnaire (ELQ) developed by Yukl et al. (2013). Th e ELQ items capture diff erent aspects of ethical leadership and include both values and behaviors. Respondents rate their leader’s honesty, integrity, fairness, accountability, integrity (consistency of actions with values), and ethical guidance. All ELQ items have a six-point Likert-style response format (1 = strongly disagree, 6 = strongly agree). Sample items include the following: (1) “Holds members accountable for using ethical practices in their work”; (2) “Communicates clear ethical standards for members”; (3) “Can be trusted to carry out commitments that he or she makes”; and (4) “Keeps actions consistent with stated values (‘walks the talk’).”
Table 1 Sample Characteristics
Ethnicity Asian 2.4% African American 16.4% Hispanic 1.0% Native American 0.4% Caucasian/white 78.0% Other 1.8%
Gender Female 45.2% Male 54.8%
Nature of Position Clerical/support 15.0% Professional/technical 55.7% Managerial 29.4%
Age < 20 years 0.2% 21–30 years 5.2% 31–40 years 11.9% 41–50 years 37.3% 51–60 years 32.1% > 60 years 9.2%
338 Public Administration Review • May | June 2014
recommendations provided by Hu and Bentler (1999), we exam- ined multiple indices to assess the fi t of the measurement model. According to Hu and Bentler (1999), a satisfactory model fi t can be inferred when the values for the comparative fi t index (CFI) and incremental fi t index (IFI) are 0.95 or higher and the value for the root mean square error of approximation (RMSEA) is 0.06 or lower.
Th e CFA results indicated that the measurement model had a satisfactory fi t to the data (χ2  = 835.80, IFI = 0.95, CFI = 0.95, RMSEA = 0.06). All scale items had statistically signifi cant factor loadings (p < .01) for their respective latent constructs. Standardized factor loadings (λs), as shown in table 2, ranged from 0.59 to 0.99, and only four out of the 26 items had a loading below 0.70. Given the strong bivariate relationships between our measures of ethical leadership, supervisor support, and procedural fairness, we also tested whether these measures captured the same variation and thus should be combined. We examined whether the proposed six-factor measurement model provided a better fi t to the data than an alternative fi ve-factor model in which the correlation between ethical leadership and supportive leader behavior was set to 1 (Δχ2 = 638.3, IFI = 0.89, CFI = 0.89, RMSEA = 0.09) and another model in which the correlation between ethical leadership and procedural fairness was set to 1 (Δχ2 = 866.2, IFI = 0.87, CFI = 0.86, RMSEA = 0.10). Th e CFA results indicated that the proposed six-factor model fi t the data better than the alternative fi ve-factor models, which indicated that employee perceptions of ethical leadership were empirically distinct from perceptions of supervisor support and procedural fairness (Williams and Anderson 1994).
Descriptive Statistics and Correlations among Variables Table 3 presents means, standard deviations, and correlation coeffi – cients of all the study measures. Although composite scale scores for
Employee organizational commitment is also infl uenced by proce- dural fairness and supervisor support (cf. Hassan 2013; Hassan and Rohrbaugh 2011; Meyer et al. 2002). As noted earlier, ethical lead- ers are likely to develop open and trusting relationships with their employees as a result of their honesty and normatively appropriate behavior. Ethical leaders are likely to be seen as supportive and fair (Kalshoven, Den Hartog, and De Hoogh 2011; Yukl et al. 2013), and procedural fairness also involves organizational policies. We wanted to avoid confl ating ethical leadership with these other deter- minants of our dependent variables.
To measure procedural fairness, the subordinate survey included a four-item measure adapted from a Procedural Justice Scale developed and validated by Colquitt (2001). Th ese items asked respondents to indicate their perception of the formal organizational procedures used in making job-related decisions, such as whether they are free of bias or favoritism or allow employees to express their views and concerns. Th e Cronbach’s alpha for this measure was 0.93. Supportive leader behavior was measured using fi ve items in the subordinate survey from the Managerial Practices Survey devel- oped by Yukl and colleagues (Kim and Yukl 1995; Yukl, Gordon, and Taber 2002). Th ese items asked the subordinate to assess how much their supervisor “is sympathetic and supportive when you are worried or upset” or “provides encouragement and support when there is a diffi cult and stressful task.” Cronbach’s alpha for the meas- ure was 0.92. Responses for items of both measures were collected using a fi ve-point response format, with an anchor for each choice indicating how much the behavior described by the item is used by the focal manager (1 = not at all, 5 = to a great extent).
We also controlled for the eff ects of individual employee characteris- tics, including age, gender, organization tenure, and managerial sta- tus in the analysis. Previous research has shown that such factors can infl uence absenteeism behavior (Hacket 1989; Sagie 1998; Porter and Steers 1973). Age was measured with a single item (“What is your current age?”) from the employee survey; the fi ve response choices were 1 = 21 to 30 years, 2 = 31 to 40 years, 3 = 41 to 50 years, 4 = 51 to 60 years, and 5 = over 60 years. We created separate dummy variables (1 = yes, 0 = no) to assess the unique eff ect of each age group. Th e base/reference category in the regression analyses was 41 to 50 years. Manager status and gender (i.e., female) was meas- ured each with a dummy variable (1 = yes, 0 = no). Organization tenure was measured with a single item from the employee survey (“How long in number of years have you worked in your current agency?”).
Results Psychometric Properties of the Measures Prior to testing the hypotheses, we conducted confi rmatory fac- tor analysis (CFA) to assess validity for the measures. Because willingness to report ethical problems was assessed using a single indicator, the square root of the reliability estimate of the indica- tor (estimated at .90) was used to fi x the measurement parameter (Williams and Hazer 1986). Th e error variance was consequently fi xed to 1 minus the reliability multiplied by the item variance. Th is procedure is common in covariance structure analysis (cf. Farkas and Tetrick 1989; Frone, Russell, and Cooper 1992; Williams and Hazer 1986), and resulting parameter estimates are accurate and unbiased (Netemeyer, Johnston, and Burton 1990). Following
Table 2 Standardized Factor Loadings (λ) from CFA
Ethical Leader- ship
Affective Com- mitment
Overall Commit- ment
Willingness to Report Ethical Problems
1 0.83 2 0.65 3 0.88 4 0.88 5 0.92 6 0.77 7 0.89 8 0.94 9 0.93 10 0.85 11 0.91 12 0.77 13 0.83 14 0.65 15 0.84 16 0.87 17 0.92 18 0.87 19 0.59 20 0.80 21 0.81 22 0.60 23 0.90 24 0.97 25 0.95 26 0.99
Does Ethical Leadership Matter in Government? Effects on Organizational Commitment, Absenteeism, and Willingness to Report Ethical Problems 339
for further analysis. Ethical leadership, as anticipated, was found to have signifi cant positive correlations with employee willingness to report ethical problems (r = 0.54, p < .01), self-reported aff ective commitment (r = 0.49, p < .01), and supervisor-rated subordinate commitment (r = 0.37, p < .01) and a signifi cant negative cor- relation with absences (r = –0.13, p < .01). Th ese results showed preliminary support for the three research hypotheses.
Tests of Hypotheses We tested the three hypotheses with a series of regression analyses; the results are reported in tables 4, 5, and 6. Because willingness to report ethical problems was measured with a single survey item, we conducted both ordered probit and ordinary least square (OLS) regression analyses to test hypothesis 1. Th e results, shown in table 4, were very similar. Th e regression results reported in tables 4, 5, and 6 are based on composite scale scores that were standardized before conducting regression analyses. Th e estimated variance infl a- tion factor scores were much lower than the typical threshold value of 4, indicating that multicollinearity did not adversely infl uence the regression results.
Hypothesis 1 proposed a positive relationship between ethical leader- ship and willingness to report ethical problems to management. As indicated in table 4, after controlling for subordinate characteristics, procedural fairness, and supportive leader behavior, ethical leadership was positively related to subordinate willingness to report ethical prob- lems. A positive change in ethical leadership by one standard deviation from the mean resulted in an increase employee willingness to report ethical problems to upper management by almost half a standard
ethical leadership, supportive leader behavior, procedural fairness, and self-reported and supervisor-rated subordinate commitment were skewed in a slightly negatively way, the diff erences between the average scores and the scale midpoints were not very large. Additionally, the standard deviations for each of the measures were relatively high, indicating adequate variability in the data suffi cient
Table 3 Means, Standard Deviations, and Correlation Coeffi cients of the Study Measures
Measures Mean SD 1 2 3 4 5 6 7 8 9 10
Age 4.29 1.00 — Gender 0.45 0.50 –0.10* — Organization tenure 10.94 9.05 0.28** 0.05 — Manager 0.30 0.46 0.06 –0.02 –0.02 — Supportive behavior 3.38 1.14 –0.09* 0.03 –0.14** 0.03 — Procedural fairness 3.28 1.20 0.01 –0.09* –0.12* 0.20** 0.63** — Ethical leadership 4.85 1.31 0.06 –0.07 –0.09 0.16** 0.72** 0.68** — Willingness to report ethical problems 4.67 1.62 –0.03 –0.08 –0.09 0.24** 0.42** 0.51** 0.54** — Supervisor-rated overall commitment 4.07 0.88 –0.07 0.04 –0.06 0.15** 0.39** 0.26** 0.37** 0.17** — Self-rated affective commitment 4.40 1.21 0.01 0.04 –0.02 0.27** 0.46** 0.50** 0.49** 0.43** 0.20** — Hours absent 33.22 34.07 –0.03 0.15** 0.10 –0.13** –0.04 –0.09* –0.13** –0.08 –0.15** –0.05
N = 415. *p < .05; **p < .01.
Table 6 OLS Regression Results: Hours Absent
Measures β t VIF 21–30 years –0.04 –0.66 1.19 31–40 years 0.02 0.37 1.21 51–60 years –0.01 –0.25 1.23 60 years and older 0.08 1.53 1.13 Manager –0.11 –2.11* 1.09 Female 0.17 3.20** 1.04 Organization tenure 0.12 2.13* 1.10 Supportive behavior 0.08 1.04 2.11 Procedural fairness 0.03 0.44 2.32 Ethical leadership –0.20 –2.51* 2.30 F-ratio 3.22** Adjusted R2 0.06
N = 415. *p < .05; **p < .01.
Table 4 Ordered Probit and OLS Regression Results: Employee Willingness to Report Ethical Problems
Ordered Probit Results OLS Results
β z β t VIF
21–30 years –0.13 –0.49 –0.02 –0.44 1.21 31–40 years –0.19 –1.01 –0.05 –1.17 1.22 51–60 years –0.32 –2.24* –0.15 –3.32** 1.24 60 years and older –0.26 –1.08 –0.08 –1.73 1.14 Manager 0.48 3.47** 0.13 3.01** 1.09 Female –0.27 –2.18* –0.09 –2.23* 1.04 Organization tenure –0.06 –0.86 –0.02 –0.42 1.12 Supportive behavior 0.03 0.34 –0.04 –0.60 2.32 Procedural fairness 0.32 3.74** 0.21 3.67** 2.06 Ethical leadership 0.52 5.43** 0.43 6.89** 2.33 χ2/F 160.49** 24.18** Pseudo R2/Adjusted R2 0.15 .38 Cut point 1 0.08 Cut point 2 –0.71 Cut point 3 –1.27 Cut point 4 –1.54 Cut point 5 –2.07
N = 415. *p < .05; **p < .01.
Table 5 OLS Regression Results: Self- and Supervisor-Rated Employee Commitment
Self-Rated Commitment Supervisor-Rated
β t VIF β t VIF
21–30 years –0.06 –1.29 1.21 –0.09 –1.40 1.27 31–40 years –0.02 –0.34 1.21 –0.05 –0.78 1.24 51–60 years –0.04 –0.83 1.24 –0.14 –2.31* 1.26 60 years and older 0.00 0.11 1.14 –0.05 –0.75 1.17 Manager 0.18 4.16** 1.09 0.12 2.12* 1.09 Female 0.07 1.57 1.05 0.02 0.42 1.06 Organization tenure 0.07 1.64 1.12 0.02 0.32 1.14 Supportive behavior 0.24 3.77** 2.06 0.33 4.08** 2.23 Procedural fairness 0.26 4.47** 2.32 –0.06 –0.83 1.87 Ethical leadership 0.16 2.59** 2.33 0.20 2.55* 2.10 F 23.93** 7.69** Adjusted R2 0.39 0.20
N = 415. *p < .05; **p < .01.
340 Public Administration Review • May | June 2014
and Berman 2004) and the potential impact of that behavior on democratic institutions (Vigoda-Gadot 2008), but few studies have directly investigated how ethical leaders can infl uence the attitudes and behavior of public sector employees (Beeri et al. 2013; Hassan et al. 2013).
Our fi ndings showed that ethical leadership is likely to increase subordinate willingness to report ethical problems. Employee will- ingness to report ethical concerns is a critical element in establishing an ethical climate in public sector organizations (Miceli and Near 1985). Not only will such reporting discourage misconduct, but it also refl ects employee confi dence that the organization’s leadership will take appropriate and corrective action.
Our fi nding that ethical leader behavior can increase organiza- tional commitment is important because commitment can help reduce turnover, improve job performance, and increase citizen- ship behavior (Cooper-Hakim and Viswesvaran 2005; Mathieu and Zajac 1990; Meyer et al. 2002). Our results for organizational commitment replicate fi ndings in recent studies of public sector agencies in the United States (Hassan et al. 2013) and Israel (Beeri et al. 2013), and we extended these earlier fi ndings by showing that
the relationship is signifi cant for supervisor ratings of subordinate commitment as well as for subordinate self-reports of commitment. Finally, our study is the fi rst to show that ethical leadership can reduce absenteeism. Because absences often represent a misuse or abuse of sick leave policies, this fi nding
also provides some evidence that ethical leadership can reduce the unethical behavior of subordinates.
Our study has some advantages not found in the majority of the earlier studies on eff ects of ethical leadership. We avoided the problem of confl ating ethical leadership behavior with other likely determinants of subordinate attitudes and behavior such as sup- portive leadership and procedural justice. We used diff erent sources of information about leadership behavior and two of the outcome variables. We controlled for the infl uence of respondent characteris- tics that are likely to infl uence the dependent variables, such as job type (manager versus nonmanager) and respondent age, gender, and organization tenure. While assessing the eff ects of these respondent attributes was not a research objective, we found several signifi cant results that provide additional evidence for accurate measurement of the dependent variables. For example, the level of organizational commitment and willingness to report ethical problems was greater for managers than for nonmanagers. Additionally, the frequency of absences, as anticipated, was higher for women than for men because of their family obligations.
Study Limitations Like most studies, this one has some limitations, and additional research would be useful to verify our fi ndings. One limitation is that only a single agency provided all of the data for the study, and the extent to which the results can be generalized to other types of public sector organizations is not clear. Another limitation is that the data for ethical leadership and subordinate willingness to report ethical problems were collected from the same source at the same time, and the results may be infl ated somewhat by same-source
deviation from the mean (β = 0.43, p < .01). Th e results also indicated that procedural fairness had a positive infl uence on the likelihood of reporting ethical problems (β = 0.21, p < .01), subordinate manag- ers were more likely to report ethical problems than nonmanagerial subordinates (β = 0.13, p < .01), and female subordinates were less likely to report problems than male subordinates (β = –0.09, p < .05). Moreover, subordinates between 51 and 60 years old were less inclined to report ethical problems than subordinates between 41 and 50 years old (β = –0.15, p < .01). Altogether, these fi ve factors explained 38 percent of the total variance in willingness to report ethical problems.
Hypothesis 2 proposed a positive relationship between ethical leader- ship and subordinate commitment, and the results support this hypothesis for both measures of commitment. As shown in table 5, ethical leadership had a signifi cant positive eff ect on both subordi- nate self-reports of aff ective commitment (β = 0.16, p < .01) and supervisor ratings of subordinate commitment (β = 0.20, p < .01). Supportive leader behavior also had a signifi cant positive infl uence on self-reported and supervisor-rated subordinate commitment (βs = 0.24 and 0.33, p < .01, respectively). Perceptions of procedural fairness had a positive infl uence on self-reported aff ective commit- ment (β = 0.26, p < .01), but the results were not signifi cant for supervisor ratings of subordinate commit- ment. Subordinates who were managers had signifi cantly more self-rated commitment than nonmanagerial subordinates (β = 0.18, p < .01) and signifi cantly higher boss ratings of subordinate commitment (β = 0.12, p < .05). Th e older subgroup of subordinates received a slightly lower commitment ratings from their supervisors (β = –0.15, p < .01). Taken together, these factors explained 39 percent and 20 percent of the total variance in self-reported and supervisor-rated subordinate commitment, respectively.
Hypothesis 3 proposed that ethical leadership reduces subordinate absences, and this hypothesis was also supported. As shown in table 6, ethical leadership had a signifi cant and sizeable negative infl uence on employee absences (β = –0.20, p < .05). Specifi cally, a positive change in ethical leadership by one standard deviation resulted in a decrease in absenteeism by almost seven hours. Moreover, several employee characteristics were found to have an infl uence on absenteeism. Th e results indicated that female employees had more absences than their male counterparts (β = 0.17, p < .01) and subordinate managers had fewer hours of absences than nonmana- gerial subordinates (β = –0.11, p < .05). Age had no infl uence on absences, but employees with longer tenure had more absences (β = 0.12, p < .05). Altogether, 6 percent of the total variance in absen- teeism was explained by the four factors.
Discussion Summary and Contributions In recent years, there has been growing interest in studying the determinants and consequences of ethical leadership, but there has not been much research on ethical leadership in public sector organizations. Our study was conducted to learn more about the potential benefi ts from ethical leadership in public sector agencies. Previous studies have highlighted the important role that leaders are expected to play in managing the ethical behavior of public sector employees (Bruce 1994; Berman, West, and Cava 1994; West
Our study is the fi rst to show that ethical leadership can
Does Ethical Leadership Matter in Government? Effects on Organizational Commitment, Absenteeism, and Willingness to Report Ethical Problems 341
attention include the infl uence of specifi c leadership behaviors on ethical behavior by subordinates, the eff ects of formal management programs to enhance ethical behavior, the infl uence of a strong ethical culture on leaders and subordinates, how ethical and unethi- cal practices are reported for elected offi cials, the consequences of ethical scandals for diff erent types of public sector organizations, and eff ects of cultural diff erences on the defi nition of ethical and unethical behavior.
References Alexander, David, and Phil Stewart. 2013. Sex Assault Scandals Put Unprecedented
Pressure on Pentagon. Reuters, May 15. http://www.reuters.com/arti- cle/2013/05/16/us-usa-defense-sexassault-idUSBRE94F00E20130516 [accessed March 23, 2014].
Bandura, Albert. 1977. Self-Effi cacy: Toward a Unifying Th eory of Behavioral Change. Psychological Review 84(2): 191–215.
———. 1986. Social Foundations of Th ought and Action: A Social Cognitive Th eory. Englewood Cliff s, NJ: Prentice Hall.
Barnard, Chester I. 1938. Th e Functions of the Executive. Cambridge, MA: Harvard University Press.
Beeri, Itai, Rachel Dayan, Eran Vigoda-Gadot, and Simcha B. Werner. 2013. Advancing Ethics in Public Organizations: Th e Impact of an Ethics Program on Employees’ Perceptions and Behaviors in a Regional Council. Journal of Business Ethics 112(1): 59–78.
Berman, Evan, Jonathan West, and Anita Cava. 1994. Ethics Management in Municipal Governments and Large Firms Exploring Similarities and Diff erences. Administration & Society 26(2): 185–203.
Blau, Peter M. 1964. Exchange and Power in Social Life. New York: Wiley. Brewer, Gene A., and Sally Coleman Selden. 1998. Whistle Blowers in the Federal
Civil Service: New Evidence of the Public Service Ethic. Journal of Public Administration Research and Th eory 8(3): 413–39.
Brown, Michael E., and Linda K. Treviño. 2006. Ethical Leadership: A Review and Future Directions. Leadership Quarterly 17(6): 595–616.
Brown, Michael E., Linda K. Treviño, and David A. Harrison. 2005. Ethical Leadership: A Social Learning Perspective for Construct Development and Testing. Organizational Behavior and Human Decision Processes 97(2): 117–34.
Bruce, Willa. 1994. Ethical People Are Productive People. Public Productivity and Management Review 17(3): 241–52.
Caillier, James Gerard. 2012. Agency Retaliation against Whistle-Blowers. Public Integrity 15(1): 29–50.
CCH Incorporated. 2006. CCH Unscheduled Absence Survey 2006. http://www. cch.com/absenteeism2006/ [accessed March 25, 2014].
Colquitt, Jason A. 2001. On the Dimensionality of Organizational Justice: A Construct Validation of a Measure. Journal of Applied Psychology 86(3): 386–400.
Cooper, Terry L. 1982. Th e Responsible Administrator: An Approach to the Ethics of the Administrative Role. Port Washington, NY: Kennikat Press.
Cooper-Hakim, Amy, and Chockalingam Viswesvaran. 2005. Th e Construct of Work Commitment: Testing an Integrative Framework. Psychological Bulletin 131(2): 241–59.
Cowell, Richard, James Downe, and Karen Morgan. 2014. Managing Politics? Ethics Regulation and Confl icting Conceptions of “Good Conduct.” Public Administration Review 74(1): 29–38.
Dalton, Dan R., and James L. Perry. 1981. Absenteeism and the Collective Bargaining Agreement: An Empirical Test. Academy of Management Journal 24(2): 425–31.
Darr, Wendy, and Gary Johns. 2008. Work Strain, Health, and Absenteeism: A Meta-Analysis. Journal of Occupational Health Psychology 13(4): 293–318.
Dibben, Pauline, Philip James, and Ian Cunningham. 2001. Absence Management in the Public Sector: An Integrative Model? Public Money and Management 21(4): 55–60.
variance. However, we did not fi nd evidence of a strong common source bias, and controlling for eff ects of supportive leadership probably eliminated any bias that was present.
Another concern is that only a single item from the subordinate survey was used to measure willingness to report ethical problems. Although single-item scales do not necessarily lack reliability and validity (Wanous, Reichers, and Hudy 1997), it is desirable to verify our fi ndings in a study with a multi-item measure of willing- ness to report diff erent types