University of Manchester



Ethical Leadership with Both “Moral Person” and “Moral Manager” Aspects: Scale Development and Cross-Cultural ValidationAbstractThe importance of ethical leadership in organizations has been increasingly recognized, especially as a shield against unethical employee behaviors and corporate misconducts. Ethical leadership has been theorized to include two aspects: “moral person” and “moral manager.” This conceptualization resonates well with Chinese teachings of Confucius on leadership and management—namely, xiuji (cultivating oneself) and anren (bringing the good to others). Based on the theoretical framework of ethical leadership, we develop and validate a new ethical leadership measure (ELM). Through qualitative studies (i.e., face-to-face interviews, open-ended surveys, and literature review) and five quantitative studies, we establish the reliability and convergent, discriminant, and predictive validities of the ELM in a Chinese context. In addition, using a U.S. sample, we find that the ELM has partial measurement invariance across Chinese and American contexts.Keywords: Ethical Leadership, Ethical Leadership Measure, Moral Person, Moral Manager, Xiuji Anren Ethical leadership has been widely studied in the past few decades, with researchers adopting the view that leaders play a significant role in shaping ethical business conduct. Brown, Trevino, and Harrison (2005) formally defined ethical leadership as “the demonstration of normatively appropriate conduct through personal actions and interpersonal relationships, and the promotion of such conduct to followers through two-way communication, reinforcement, and decision-making” (p. 120). Based on this conceptualization, great efforts have been devoted to empirically investigating the connections between ethical leadership and positive employee and organizational outcomes (e.g., Liu, Kwan, Fu, & Mao，2013；Schaubroeck et al., 2012). Nevertheless, significant gaps persist in the ethical leadership literature. One gap is particularly salient: the lack of formal conceptualization and measurement of ethical leadership in Eastern culture as well as cross-culturally. Trevi?o, Hartman, and Brown (2000) proposed that ethical leadership rests upon two essential pillars, “moral person” and “moral manager,” and developed the Ethical Leadership Scale (ELS) to measure these aspects. Kalshoven, Den Hartog, and De Hoogh (2011a) developed a scale called the Ethical Leadership at Work (ELW) questionnaire that includes seven dimensions of ethical leadership: fairness, integrity, ethical guidance, people orientation, power sharing, role clarification, and concern for sustainability. The ELW scale, however, has some limitations. First, it tends to overly broaden the conceptualization of ethical leadership. For example, several dimensions—such as the dimension of “role clarification”—could be considered neutral leadership behaviors and may not necessarily be related to ethicality. Second, the majority of the items of the ELW measure the “moral manager” dimension (e.g., “My leader pays attention to my personal needs (people orientation”). By comparison, only a few items of the ELW capture the “moral person” dimension. To further advance cross-cultural research on ethical leadership, it is essential to develop a more refined ethical leadership scale that draws upon both the philosophical foundations underlying ethical leadership and the overarching theoretical foundation established by Trevi?o et al. (2000), while simultaneously having cross-cultural validity. Ethical leadership appears to be particularly relevant and important for Chinese leadership, and is embedded in both traditional Chinese management and leadership (e.g., Chen, Eberly, Chiang, Farh, & Cheng, 2014). For example, the “Melamine Incident,” in which melamine-contaminated milk caused kidney stones in thousands of babies, shocked the public in China and elsewhere. As a result of this and other business scandals, many have begun to question the unethical practices and behaviors of business leaders in Chinese firms. However, the existing moral leadership dimension of paternalistic leadership (Cheng, Chou, Wu, Huang, & Farh, 2004) falls short of reflecting the full philosophical and theoretical richness of ethical leadership. Furthermore, understanding of cross-cultural similarities and differences in ethical leadership is invaluable to enhance effective leadership across cultures in today’s era of globalization (e.g., Chen, Leung, & Chen, 2009; Resick et al., 2011). The research presented in this paper has three main purposes: (1) from a theoretical standpoint, to explore aspects of ethical leadership in contemporary Eastern cultures—Chinese culture in particular; (2) to develop a new measure that captures the aspects of ethical leadership in China; and (3) to validate a measure that has cross-cultural validity. Following a standard scale development procedure (Hinkin, 1998), we develop an ethical leadership measure (ELM), which captures a second-order construct consisting of four first-order factors: moral characteristics, moral cognition, moral role modeling, and moral atmosphere creation. The first two factors address the “moral person” aspect of ethical leadership, whereas the latter two factors focus on the “moral manager” aspect. In this way, the ELM captures the richness and complexity of the ethical leadership domain. We also establish construct reliability and validity for the newly developed scale. In addition, using an American sample, we find that the proposed ELM has configural invariance and partial loading value invariance across Chinese and American contexts. This research contributes to the ethical leadership literature in important ways. Specifically, we integrate traditional Chinese philosophies (Confucius, 2006) with the contemporary Western conceptualization of ethical leadership (Trevi?o et al., 2000) to explore aspects of ethical leadership that are indigenous to the Chinese culture as well as applicable to the Western culture. Our research represents one of the first efforts to connect ethical leadership scholarship in different cultures. Our theory and findings suggest that ethical leadership—in particular, its two aspects of “moral person” and “moral manager”—has cultural universality. Thus, our work contributes to cross-cultural research on ethical leadership. Through inductive and deductive approaches and based on six samples and 2089 individuals across two cultures, we develop and validate a new measure of ethical leadership. In this way, we lay a strong foundation that will allow research to flourish in this area.Theoretical FoundationEthical Leadership: “Moral Person” and “Moral Manager” Trevi?o’s ethical leadership theory encompasses two aspects of such leadership: “moral person” and “moral manager.” According to Trevi?o and her colleagues, as a moral person, an ethical leader should possess personal moral characteristics and traits, such as honesty, integrity, and trustworthiness. In addition, an ethical leader needs to demonstrate “moral manager” behaviors to influence followers. For example, ethical leaders need to make decisions in an objective and fair way, adhere to a solid set of ethical values and principles, show thoughtful concern for both the local community and the larger society, and follow their ethical decision-related rules. Meanwhile, being a “moral manager” means that ethical leaders serve as a role model of ethical conduct and establish incentives to encourage their followers’ moral actions. In addition, ethical leaders issue sanctions to followers who fail to abide by ethical rules, and make ethics an explicit part of their leadership agenda by regularly communicating with employees about ethical standards, values, and principles (Trevi?o et al., 2000). Based on this “moral person” and “moral manager” theoretical framework, Brown et al. (2005) developed an ethical leadership scale based on factors found to influence employee and organizational outcomes. These outcomes include organizational citizenship behaviors (Avey, Palanski, & Walumbwa, 2011; Mayer, Kuenzi, Greenbaum, Bardes, & Salvador, 2009), employee voice (Zhu, He, Trevi?o, Chao, & Wang, 2015), recruitment (Ogunfowora, 2014), team effectiveness and subordinates’ optimism (De Hoogh & Den Hartog, 2008), unit unethical behavior and relationship conflict (Mayer, Aquino, Greenbaum, & Kuenzi, 2012), and followers’ ethical cognition, judgment, and behavior (Schaubroeck et al., 2012). Other scales measuring ethical leadership, such as the ELW developed by Kalshoven et al. (2011a) and the ethical leadership questionnaire developed by Yukl, Mahsud, Hassan, and Prussia (2013), have been empirically tested to a limited extent (e.g., Kalshoven, Den Hartog, & De Hoogh, 2011a, 2011b; Yukl et al., 2013).Xiuji and Anren: Corresponding Aspects of Ethical Leadership in Chinese Culture Traditional Chinese culture emphasizes moral and ethical values. For example, Bass and Steidlmeier (1999, p. 194) wrote: “In China, there seems to be some general agreement that among other things, the moral life rests upon foundations of individual virtue and that the individually virtuous person transforms others as well as the social environment.” xiuji and anren, two Chinese expressions originating from Chinese philosophies, which mean “cultivating oneself” and “bringing the good to others,” respectively, can be considered to correspond to Trevino et al.’s (2000) two pillars of “moral person” and “moral manager.” When addressing Chinese philosophy, people primarily refer to the dominant philosophical theme, Confucianism, which was originally proposed by Confucius and then expanded by Mencius. Confucianism considers benevolence and ethics to be the most fundamental values for individuals and leaders in Chinese culture. Confucian teachings pertinent to ethical leadership combine the concepts of xiuji and anren. Upon comparing the combined concept of xiuji anren with the Western ethical leadership conceptualizations, we observe a large overlap between Brown et al.’s (2005) conceptualizations of “moral person” and “moral manager” and xiuji anren. On the one hand, xiuji focuses on a leader’s personal cultivation and development into a good and moral person, which resonates with the “moral person” dimension proposed by Trevi?o et al. (2000). On the other hand, anren focuses on a leader’s ability to guide others’ behaviors in desirable ways, such as by developing followers’ full potential and well-being. Although anren might not exactly correspond to the “moral manager” aspect of Trevi?o et al.’s (2000) conceptualization, it does have a certain level of resonance with the notion of positively influencing followers. In our work, we endeavor to develop a more refined measure of ethical leadership by integrating the unique elements of an Eastern culture (China) with the “moral person” and “moral management” aspects of Brown et al.’s (2005) theoretical framework on ethical leadership. Ancient Chinese philosophers such as Confucius and Mencius, like their Western counterparts Socrates and Plato, produced seminal works on the significance and relevance of ethics in leadership. For example, Confucius said, “The power of a moral virtue is supreme, yet it is not commonly found among people anymore” (Analects: 6.29; Lau, 1970a), demonstrating the importance and value of ethics. According to Confucius, de (ethics) can be defined as both virtuous conduct (i.e., moral standards of behaviors) and virtuous character (i.e., the quality and character behind the behavior) (Chen, 2010). In traditional Chinese culture, de includes both meanings. To better understand the concept of ethics, Confucius’s notion of junzi (i.e., gentleman) is also highly relevant. A junzi is a morally holistic persona who expresses the good behaviors, ideals, values, and personality traits expected of an ethical individual (Chen, 2010). In Analects, the way (dao) and the virtues (de) of a gentleman are among the most prominent topics. Confucius, however, went beyond any particular virtue to emphasize the integrative nature of the virtuous conducts and characters that are associated with a junzi, thereby depicting the totality of human characteristics. In this sense, ethical leaders ostensibly satisfy both the “moral person” and “moral manager” criteria. In Confucianism, the Analects (14.42) records the conversations between Confucius and one of his disciples, Zilu. Zilu asked Confucius about how to become a junzi (gentleman, or ethical leader); Confucius replied, “Men have to cultivate selves with reverence.” Zilu again asked: “Would doing this be enough?” Confucius answered: “Men have to cultivate selves to bring the good to everyone else.” Confucius’s responses clearly show that “cultivating oneself” (xiuji) and “bringing the good and peace to others” (anren) are both essential components of being a junzi. Cultivating oneself reflects the fundamental belief in one’s need to become a good person and demonstrate anren—bringing the good and peace to others. As Confucius said: “Lead the people with governmental measures and regulate them with laws and punishment, and they will avoid wrongdoing but will have no sense of honor and shame. Lead them with virtue and regulate them by the rules of propriety, and they will have a sense of shame and, moreover, set themselves right” (Analects: 2.3). In the same vein, in Dao De Jing, the classical seminal work of Daoism (another dominant Chinese philosophy theme), Laozi described good leaders as those who focus on the betterment of the collective group instead of those who just build the hierarchy and structure to support their own positions and reward those subordinates who align with their own interests and obey their orders. Laozi used the term “sage” to describe the best or most ethical leader: “The sage has no fixed mind; he is aware of the needs of others and also treats those needs as his” (Dao De Jing: 49) and “The sage does not accumulate (for himself). The more that he expends for others, the more does he possess of his own; the more that he gives to others, the more does he have himself” (Dao De Jing: 81). Together, these messages convey the notion that a sage or ethical leader does not emphasize his or her own self-interests, but rather the interests of others around the leader and the collective group. Thus, both of the two prevailing Chinese philosophies—Confucianism and Daoism—emphasize the value and role of ethics and morality in leadership. This examination of the commonalities between the foundations of ethical leadership in the Chinese and Western cultures suggests that it might be theoretically meaningful to consider both Chinese philosophical teachings on leadership (i.e., xiuji anren) and Brown et al.’s (2005) operationalization of ethical leadership to develop a new measure for ethical leadership that would be applicable to both Chinese and Western cultures. The prior discussion has shown that traditional Chinese philosophical teachings on leadership resonate well with the modern Western conceptualization of ethical leadership developed by Trevi?o et al. (2000). This lays a solid theoretical foundation for our adoption of the “moral person”/“moral manager” concepts as the theoretical basis for developing a refined ethical leadership measure that may be utilized across cultures. Although Brown et al.’s (2005) ELS measure has the virtues of being parsimonious and having been applied in numerous empirical studies, there remains a need to develop a more refined measure for ethical leadership. Specifically, although the ELS includes items related to both the “moral person” and “moral manager” aspects, they are often empirically found to collapse to a single inseparable dimension (Kalshoven et al., 2011a). As an alternative, Kalshoven et al.’s (2011a) ELW questionnaire includes seven dimensions: fairness, integrity, ethical guidance, people orientation, power sharing, role clarification, and concern for sustainability. However, the ELW scale adopted a different conceptualization of ethical leadership from Trevi?o et al.’s (2000) original conceptualization and the Chinese philosophical teaching of leadership. To date, no research has been conducted to develop and validate an ethical leadership scale applicable in different cultural settings. Therefore, drawing upon Chinese philosophical teaching (Confucius, 2006) and Trevi?o et al.’s (2000) conceptualization, we develop a measure of ethical leadership and examine this scale’s applicability across cultures. Our ethical leadership conceptualization includes the two broad aspects of xiuji and anren.Methods Following the standard scale development procedure (Hinkin, 1998), we attempted to develop an instrument that (1) spanned the full domain of “moral person” and “moral manager,” (2) consisted of items that were understandable to working employees, and (3) could be applied in different cultural contexts. The measure was developed in six different studies. Studies 1?4 were conducted primarily to develop the items, examine the internal coherence, and test the convergent and discriminate validity of the new ethical leadership measure. Study 5 examined the predictive validity of the measure. Study 6 explored the cross-cultural measurement invariance of the scale across Chinese and American contexts.Study 1: Item Generation and Selection To increase the validity of the scale and to better cover the domain of ethical leadership, we utilized both inductive (i.e., open questionnaire, in-depth interviews) and deductive (i.e., literature review) methods to generate items to measure ethical leadership. Combining inductive and deductive methods can capitalize on the advantages of both methods (Hinkin, 1998). The first method was an open questionnaire. In total, 150 part-time MBA students and 30 middle and senior managers from a training program at a leading university in China completed the questionnaire. We provided a general working definition of ethical leadership (Brown et al., 2005) to the interviewees and asked them to describe a supervisor whom they considered to be an ethical leader. We then asked participants to list five or six personal traits or leadership and management behaviors associated with ethical leadership, which ultimately generated 810 descriptions (5.4 items, on average, per person). The second method relied on an inductive approach, as we conducted in-depth interviews with another 15 MBA students. We asked these participants to describe which traits or behaviors are important to effectively demonstrate ethical leadership. The interviewees mentioned several aspects of ethical leadership behaviors, including considering employee interests, listening to employees, setting a moral example, discussing business ethics or values with employees, and living one’s personal life in an ethical manner. Through these interviews, we collected 80 additional items. We then conducted content analysis on the 890 items collected from the open questionnaires and interviews. Specifically, we carried out a systematic category analysis by taking the following steps: (1) assuring every description or item had a clear meaning; (2) assuring that it was a leader behavior or characteristic; (3) combining similar items into one single item; and (4) deleting items that were mentioned twice or less frequently (e.g., Nunnally & Bernstein, 1994). To ensure face validity, we then consulted with five human resources professionals to evaluate each item and eliminate items beyond the domain of our ethical leadership conceptualization (xiuji anren) as well as those items that could be potentially confusing or redundant. After many rounds of identification, induction, and elimination (Hinkin, 1998; Nunnally & Bernstein, 1994), we retained 27 items in total. Finally, we reviewed the ethics-related literatures, including literatures dealing with ethical leadership scales and description of ethical leadership (Brown et al., 2005; Trevi?o et al., 2000), authentic transformational leadership (Bass & Steidlmeier, 1999; Zhu, Avolio, Riggio, & Sosik, 2011), moral identity traits (Aquino & Reed, 2002), justice (Alexander & Ruderman, 1987), and moral attentiveness (Reynolds, 2008). This exercise supplemented our item pool with an additional 60 items (Hinkin, 1998; Nunnally & Bernstein, 1994). Thus, the initial item pool included 87 (27 + 60) items.Study 2: Exploratory Factor Analysis We administered the 87-item survey to 250 business managers who attended Executive Developmental Program (EDP) training sessions at a leading university in China. These managers rated the extent to which they agreed or disagreed with the statements regarding their immediate supervisors’ ethical leadership on a 7-point scale (1 = strongly disagree, 2 = disagree, 3 = somewhat disagree, 4 = neither disagree nor agree, 5 = somewhat agree, 6 = agree, 7 = strongly agree). Out of 250 questionnaires administered, we received 199 valid responses, for a response rate of 79.6%. A sample size of about 200 is generally considered sufficient for exploratory factor analysis (EFA; MacCallum, Widaman, Zhang, & Hong, 1999). Our sample also exceeded the minimum requirement of a 2:1 subjects-to-items ratio necessary to conduct EFA (Kline, 1979). We conducted an EFA (principal axis factoring) with an oblique rotation (direct oblimin), allowing for correlations to be discerned among factors (Fabrigar, Wegener, MacCallum, & Strahan, 1999). Evaluation of the eigenvalues and scree plot were the principles of factor extraction. Four factors having eigenvalues greater than 1 accounted for 79.71% of the total variance, above the recommended level of 60% (Hinkin, 1998). After deleting items based on this criterion, 16 items were retained; their factor loadings were between .67 and .94. Table 1 shows the EFA results.--------------------------------------Insert Tables 1 to 10 about here---------------------------------------- The first factor, termed “moral characteristics,” refers to whether a leader possesses traits and characteristics relevant to ethics and morality. This factor includes four items, such as “is willing to help others.” This subdimension also somewhat overlaps with the people-orientation dimension of the ELW (Kalshoven et al., 2011a). The percentage of variance explained by the eigenvalues of this factor was 50.40%. The second factor, “moral cognition,” refers to a leader’s awareness and willingness to pay attention to a moral issue, and his or her capacity to make relevant judgments. This dimension includes four items, such as “often considers issues from an ethical perspective.” Interestingly, these four items were adapted from the moral attentiveness scale developed by Reynolds (2008). We call this dimension “moral cognition” because this term better reflects the broad cognition-related moral aspects of ethical leadership, including moral realization/attentiveness and moral judgment. The percentage of variance explained by the eigenvalues of this factor was 12.77%. The third factor, “moral role modeling,” refers to leaders’ adoption of a higher ethical standard for their own behaviors, such as practicing ethical teaching and setting a visible ethical example for followers. This subscale includes four items, such as “sets an example of how to do things in the ethically correct way.” The percentage of variance explained by the eigenvalues of this factor was 9.79%. The last factor, “moral atmosphere creation,” describes the important roles that ethical leaders play in building up a moral context through communication and by establishing standards and codes for their ethical behaviors. This factor has four items, such as “explains the values that guide his/her moral decisions to subordinates.” The percentage of variance explained by the eigenvalues of this factor was 6.76%. The first two factors (i.e., moral characteristics and moral cognition) are categorized under the “moral person” (i.e., xiuji) aspect of ethical leadership, because moral characteristics represent the nature of a person (i.e., being) and moral cognition represents the person’s moral thoughts (i.e., thinking). The latter two factors (i.e., moral role modeling and moral atmosphere creation) are categorized under the “moral manager” (i.e., anren) aspect of ethical leadership, because they show how managers can become moral role models for their followers and develop a moral atmosphere within which both leaders and followers operate. The four first-order factors are loaded on a higher-order general factor termed the ethical leadership measure (ELM), which has both “moral person” and “moral manager” aspects. Table 2 presents more detailed explanations of the four dimensions of the ELM.Higher-Order Nature and Multidimensionality of ELM In a reflective (superordinate) measurement model (as compared to a formative measurement model: Coltman, Devinney, Midgley, & Venaik, 2008; Jarvis, MacKenzie, & Podsakoff, 2003; MacKenzie, Podsakoff, & Jarvis, 2005), changes in the underlying construct are hypothesized to cause changes in the indicators (Bollen & Lennox, 1991; Fornell & Bookstein, 1982); the latent variable influences the indicators and accounts for their inter-correlations. The indicator items of a reflective measure are internally consistent because all of them are assumed to be equally valid for measuring the underlying construct. We approached the ELM as a superordinate/reflective multidimensional construct, indicated by four distinct dimensions that capture managers’ ethics-related behaviors. The ELM construct cannot be conceived separately from its specific dimensions (Johnson, Rosen, & Chang, 2011). The ELM construct, as a reflective and multidimensional construct, captures the intersections among its four subdimensions covering different aspects of ethical leadership, and represents a core underlying shared variance across those subdimensions. The reflective, higher-order, and multidimensional nature of the ELM can facilitate theory development in at least two ways. First, the higher-order ELM can better match the predictions of broadly defined outcomes that span various domains. Second, the higher-order ELM is more parsimonious than its individual dimensions for examining ethical leadership phenomena (El Akremi, Gond, Swaen, De Roeck, & Igalens, 2015; Johnson et al., 2011).Study 3: Confirmatory Factor Analysis The data set for the confirmatory factor analysis (CFA) was an independent sample of 900 business managers. We obtained 762 valid surveys, for a response rate of 84.67%. We employed Mplus 6.0 (Muthén & Muthén, 1998–2010) to conduct a series of CFAs to examine the factor structure of the scale. To examine the distinctiveness of the four dimensions of the ELM construct, we compared the baseline model against a series of alternative nested models, which merged two or more of the four ELM dimensions. In addition, we tested for overall discriminability by contrasting the four-factor baseline model with a single-factor model (El Akremi et al., 2015; Kinicki, Jacobson, Peterson, & Prussia, 2013). Following the path laid out in earlier works (e.g., Hannah, Jennings, Bluhm, Peng, & Schaubroeck, 2014), we provided empirical justification for including first-order dimensions based on three criteria: (1) the first-order dimensions have significant and substantive loadings on the second-order factor (cutoff of .70), (2) the second-order factor model exhibits a good fit, and (3) the second-order model has a high score of composite latent variable reliability (CLVR). The CFA results are presented in Table 3. We found support for the hypothesized four-factor baseline model. This model exhibits a good fit with the data: χ2 (df = 98) = 370.28, p < .001, CFI = .98, TLI = .97, RMSEA = .060, and SRMR = .023 (see Table 3). All first-order factor-loading values were statistically significant at the .01 level and reasonably large, ranging from .65 to .93. In alignment with the earlier studies (e.g., El Akremi et al., 2015), we found the average variance extracted (AVE) values for four first-order factors were .81, .70, .76, and .75, respectively; thus, they all exceeded the .50 criterion (Fornell & Larcker, 1981). The results also demonstrated support for the distinctiveness of the ELM dimensions. The single-factor model fit the data poorly: χ? (df = 104) = 1944.42, CFI = .86, TLI = .83, RMSEA = .15, and SRMR = .06. The four-factor model outperformed a series of more parsimonious models that merged different factors, thereby supporting the distinctiveness of the factors. For example, it compared favorably with the fit of three-factor models, including the model that merged moral role modeling and moral atmosphere creation: χ? (df = 101) = 883.64, CFI = .94, TLI = .93, RMSEA = .10, SRMR = .04, with Δχ2 (df = 3) = 513.36, p < .01. The difference between the three-factor model and the four-factor model met three criteria: significant chi-square difference tests, ΔCFI values greater than .01, and significant decreases in fit compared with the baseline model (Cheung & Rensvold, 2002). Finally, the proposed four-factor, second-order ELM model fit the data essentially as well as the first-order model: χ? (df =100) = 487.06, p < .01, CFI = .97, TLI = .96, RMSEA = .071, and SRMR = .034. According to Bollen (1989) and based on empirical testing (El Akremi et al., 2015; see also Linderbaum & Levy, 2010), a second-order model is preferable to a first-order model if it fits the data because it allows for covariation among first-order factors and accounts for the corrected errors that are common in first-order models. Therefore, we believe it is more appropriate to adopt a second-order model for the ELM. All the loading values of the second-order factor were statistically significant, ranging from .85 to .94, with the AVE for the second-order factor being 75.36%, which is well above the 0.50 criterion (Fornell & Larcker, 1981). Therefore, consistent with earlier work (e.g., Edwards, 2001, 2011; El Akremi et al., 2015), we conclude that ELM is a multidimensional, superordinate/reflective construct. The CFA results confirmed that the second-order ELM factor exists and significantly explains the relationships among the four first-order factors. Furthermore, these four first-order dimensions can be mapped into two aspects, such that “moral characteristics” and “moral cognition” belong to the “moral person” aspect, and “moral role modeling” and “moral atmosphere creation” belong to the “moral manager” aspect. The ELM demonstrated a high level of internal consistency, with an alpha score of .97. The alpha scores for the subdimensions (moral characteristics, moral cognition, moral role modeling, establishing moral context) were .94, .88, .93, and .92, respectively.Study 4: Convergent and Discriminant Validity To demonstrate the ELM’s convergent validity, we hypothesized that the construct of ethical leadership used in the ELM is positively related to, but empirically distinguishable from, transformational leadership (Bass & Riggio, 2006), authentic leadership (Avolio & Gardner, 2005; Walumbwa, Avolio, Gardner, Wernsing, & Peterson, 2008), servant leadership (Greenleaf, 1977; Liden, Wayne, Zhao, & Henderson, 2008), and the entity measured in the Leader Virtues Questionnaire (LVQ; Riggio, Zhu, Reina, & Maroosis, 2010). We investigated this hypothesis by surveying 500 full-time business executives. Ultimately, 393 (78.6%) valid responses were retained for analysis. Table 4 reports ethical leadership (ELM) was positively correlated with the idealized influence dimension of transformational leadership (r = .73, p < .01), the ELS (r = .85, p < .01), the “internalized moral perspective” dimension of authentic leadership (r = .63, p < .01), the “behaving ethically” dimension of servant leadership (r = .67, p < .01), and the LVQ (r = .73, p < .01). Conversely, it was negatively correlated with abusive supervision (r = –.26, p < .01) and leader narcissism (r = –.35, p < .01). Although the relationship between the ELM and the ELS was relatively strong, this outcome is not surprising because these two scales adopt the same theoretical framework of “moral person”/“moral manager” proposed by Trevi?o et al. (2000). We conducted a separate CFA to confirm the discriminant validity of the two measures. Discriminant validity means that a focal construct should be unrelated to dissimilar or non-overlapping constructs (Hinkin, 1998; Schwab, 1980). Specifically, the proposed two-factor model (i.e., ELM, as a second-order factor with four first-level factors, and ELS as a separate factor) fit the data [χ2 (df = 147) = 588.35, p < .01, CFI = .948, TLI = .940, RMSEA = .088, and SRMR = .053] better than [Δ χ2 (df = 2) = 99.85, p < .01] the one-factor model [i.e., ELM and ELS as one factor: χ2 (df = 149) = 688.15, p < .01, CFI = .937, TLI = .928, RMSEA = .096, and SRMR = .097]. Again, the results support that these two scales are distinct from each other. We found that the ELM was unrelated to the irrelevant constructs, such as the rater’s gender (r = –.09, ns), age (r = .03, ns), position (r = –.05, ns), education (r = –.05, ns), firm size (r = .07, ns), and number of years under the supervisor (r = .08, ns). Taken together, these results suggest that the ELM has good initial discriminant and divergent validity. To further establish the distinctiveness of the ELM from seven other relevant scales, we also conducted CFAs involving ELM and these scales. Due to the relatively small sample size, we followed the procedures adopted by previous research (e.g., Kovjanic, Schuh, Jonas, Van Quaquebeke, & Van Dick, 2012; Little, Cunningham, Shahar, & Widaman, 2002; Piccolo, Greenbaum, Den Hartog, & Folger, 2010) and created indicators from dimensional scores or item parcels based on the item-to-construct-balance method. We created four-item parcels for leader virtue, three-item parcels for ELS, two-item parcels for idealized influence of MLQ, four items for the “internalized moral perspective” of authentic leadership, four items for the “ethics” of servant leadership, four items for leader narcissism, and five items for abusive supervision. We found the eight-factor model exhibited a good fit: χ? (df = 828) = 2220.21, CFI = .92, TLI = .91, RMSEA = .065, and SRMR = .054. Again, this result shows that the ELM is distinct from the other relevant scales. The AVEs for the four dimensions and the ELM were 78.36%, 81.75%, 71.21%, 85.27%, and 85.34%, respectively. The AVE score of the ELM (i.e., 85.34%) was also larger than the squared correlation estimates (i.e., shared variance of 82.86%) between the ELM and the other seven relevant scales, which supported our contention that the ELM is a distinct measure (Hannah et al., 2014).Study 5: Predictive Validity Generally speaking, a good scale should have predictive validity. To determine whether the ELM satisfied this criterion, we designed a study to examine the predictive validity of the ELM above and beyond that of ELS (Brown et al., 2005) for a series of follower work-related outcomes, including follower organizational commitment, organizational citizenship behaviors (OCBs), job performance, and follower ethics-related outcomes (i.e., moral attentiveness). Ethical leaders instill higher trust, respect felt, and self-esteem in their followers, which collectively enhance followers’ work-related and ethics-related outcomes because ethical leaders set high ethical standards and act as ethical role models for their followers (Brown et al., 2005). When employees observe ethical leaders who stand up for doing what is right and righteous and who think ethically, those employees will be inclined to become more committed to their organization. Thus, ethical leadership encourages higher organizational commitment, OCBs, and job performance. Brown et al. (2005) argue that ethical leaders influence followers primarily through role modeling processes. Moral behavioral modeling is an important and effective means for leaders to transmit their moral values, attitudes, and behaviors to their followers. That is, followers learn about desirable moral traits and behaviors by observing their leaders, and then emulate those behaviors they find to be legitimate and credible. In this way, ethical leadership is positively related to followers’ moral attentiveness. Ethical leaders are also likely to take an interest in and engage with their followers’ moral growth and development (Schaubroeck et al., 2012). For example, they are more likely to provide ethical mentoring and feedback, with followers reciprocating by paying greater attention to moral issues, thereby increasing their moral attentiveness. To summarize these ideas, we propose four hypotheses:Hypotheses 1/2/3/4: Ethical leadership is positively related to higher levels of follower affective commitment/OCB/job performance/moral attentiveness. Data Collection and Sample Characteristics. We collected data from companies in the communication industry by administering a survey. We received 302 total responses (response rate of 75.5%) from the 400 participants. Of these surveys, 294 responses were found to be valid and were retained for analysis. Initially, employees reported their own affective organizational commitment and ethical leadership behaviors about their immediate supervisor using both the ELM and the ELS (Brown et al., 2005). One month later, supervisors rated their employees’ OCBs, job performance, and moral attentiveness. Measures. Affective organizational commitment was assessed using nine items from the measure developed by Mowday et al. (1979). Sample items included “I talk up this organization to my friends as a great organization to work for” and “I am proud to tell others that I am part of this organization.” OCBs were assessed by “organizational identification” and “altruism,” seen in the measure developed by Farh et al. (1997). Sample items included “willing to stand up to protect the reputation of the company” and “willingly gives his/her time to help others who have work-related problems.” Job performance was measured by a two-item scale developed for this study after consulting with the human resources manager of each firm being sampled. These two items were “the extent to which this employee is generally qualified to accomplish his or her job” and “the overall effectiveness of this employee in getting his or her job done.” Follower moral attentiveness was measured by the 12-item scale developed by Reynolds (2008). Sample items included “This employee frequently thinks about the moral implications of his/her actions” and “This employee often considers issues from an ethical perspective.” Results. As shown in Table 5—and not surprisingly—the correlation between the ELS and the ELM was relatively high (r = .80, p < .01), for the reasons noted earlier. However, the CFAs showed that the proposed two-factor model (i.e., ELM as a second-order factor with four first-order factors, and ELS as a separate/independent factor) fit the data [χ2 (df =147) = 456.32, p < .01, CFI = .95, TLI = .94, RMSEA = .085, and SRMR = .052] better than [Δ χ2 (df = 2) = 95.61, p < .01) the one-factor model [i.e., ELM and ELS as one factor, or the restrictive model: χ2 (df = 149) = 551.93, p < .01, CFI = .935, TLI = .925, RMSEA = .096, and SRMR = .093]. These CFA results support that these two scales are distinctive. We conducted hierarchical regression analysis to test the ELM’s predictive validity. Tables 6 to 9 present these results. After controlling for length of time working with the leader and the organizational/position level, which are important confounding variables for leadership effectiveness (Brown et al., 2005; Zhu et al., 2011), we found that (see Step 2a in the tables) the ELM had a positive effect on follower affective commitment (β = .36, p < .001), OCBs (β =.25, p < .001), job performance (β = .14, p < .05), and moral attentiveness (β = .16, p < .01). Thus, Hypotheses 1 to 4 were supported. To further assess the incremental validity (Kinicki et al., 2013) of the ELM as above and beyond that of the existing ELS measure, we conducted a usefulness analysis (Darlington, 1990; Johnson et al., 2011). We first entered the background variables, then the one-dimensional 10-item ELS scale (rated on a 5-point Likert scale) into the equation, followed by the ELM (see Step 2b and Step 3 in Tables 6 to 9). As expected, after controlling for the ELS, the ELM was still significantly related to these work-related and ethics-related outcomes: β = .25, p < .01 for affective organizational commitment; β = .17, p < .10 for OCBs; and β = .21, p < .05 for follower moral attentiveness. Meanwhile, the effects of the ELS were no longer significant. For follower job performance, after controlling for ELS, the ELM still marginally significantly predicted follower job performance (β = .14, p < .10, one-tailed test, Table 8, Step 3), but the effect of the ELS was negligible (β = .01, ns). More importantly, when judging these three models based on the adjusted R2 value, the model (Step 3, when ELS is added on top of ELM, adjusted R2 = 5.10%) became worse than the pure ELM model (Step 2a, adjusted R2 = 5.50%), yet was still better than the pure ELS model (Step 2b, adjusted R2 = 4.80%). These results offer support for the stronger predictive power of the ELM compared to the ELS. In addition, we performed another set of hierarchical regression analyses (see Step 2a and Step 3 in Tables 6 through 9) in which we added the new ELM first and then the ELS in the next step. The result showed that the ELS was no longer significantly related to the dependent variables when the ELM was controlled for. The ELM could still predict more variance after controlling for the effect of the ELS because the ELM captures more nuances and the broader complexity of the ethical leadership domain. For example, the ELM measures the “moral cognition” and “moral atmosphere creation” dimensions of ethical leadership, which are not captured by the ELS. Thus, the ELM can better capture the rich domain, including the depth and breadth, of ethical leadership in the Chinese context compared to the ELS. Study 6: Cross-Cultural Measurement Invariance of the ELM The primary purpose of this study was to examine the measurement equivalence of ELM across Chinese and American contexts. In doing so, we attempted to confirm the similarity of the ELM across these two contexts. When testing for measurement invariance across groups, three primary levels of potential invariance must be examined: (1) configural, (2) metric, and (3) scalar (Meredith, 1993; Vandenberg & Lance, 2000). Method and Samples. For the Chinese sample, we used the same sample on which the CFA (study 2) was conducted. To obtain the U.S. sample, we entrusted a research company () to solicit 1094 U.S. employees from a broad variety of industries. The research company sent the survey link to these potential participants on our behalf. We randomly selected 276 out of 1094 cases for the study. Analysis and Results. We conducted a multi-group confirmatory factor analysis using Mplus 6.12 (Muthén & Muthén, 1998–2010) and followed the instructions for the measurement invariance test (Billiet & McClendon, 2000; Cheung & Rensvold, 2002). Following the recommended steps (e.g., Byrne, 2012; van de Schoot, Lugtig, & Hox, 2012), we conducted an analysis of configural equivalence across two samples. In doing so, we specified the second-order hierarchical factor structure for the ELM for the Chinese and U.S. samples independently (Byrne, 2008, 2012). The fitness indexes for the Chinese context [χ2 (df = 100) = 487.06, p < .01, CFI = .97, TLI = .96, RMSEA = .071, and SRMR = .034, as shown in Table 3] and the U.S. context [χ? (df = 100) = 232.31, p < .01, CFI = .975, TLI = .970, RMSEA = .069, and SRMR = .028] were higher than the acceptable level of model fit (Chen, Sousa, & West, 2005; Hu & Bentler, 1999), as shown in Table 10, confirming that the form of the latent construct of the ELM held in each of the two cultures. Figure 1 graphically depicts the CFA results for these two samples.---------------------------------Insert Figure 1 about here--------------------------------- We also conducted a multi-group CFA to examine the measurement invariance. The results for the two-group configural baseline model are shown as Model 1 in Table 11 and were confirmed to be an acceptable fit: χ? (df = 200) = 719.37, CFI = .971, TLI = .965, RMSEA = .070, and SRMR = .034). This outcome highlighted the configural invariance across the Chinese and U.S. samples. Thus the second-order factor structure of the ELM held in both the Chinese and U.S. contexts. We continued testing the metric invariance of the ELM, which constrains factor loadings—including the item to four first-order factor loadings, and the first-order factor to second-order factor loadings—to be equivalent across these two samples. As shown in Model 2 in Table 11, constraining the item factor loadings to be invariant across these two countries did reduce the model fit (?χ? = 94.87**, df = 12), which means that some of the loading values for the items were not invariant across two samples. We subsequently used the Modification Indexes (critical value = 3.59, p < .05) to check those loading values that needed to be freed rather than constrained to be invariant. We found four items of the subdimension of “moral cognition” to be invariant across these two samples. In addition, among the loading values of nine items on three first-level factors that needed to be freed, three items (Q2, Q3, and Q4, in Table 1) were for the “moral characteristics” factor, three items (Q10, Q11, and Q12, in Table 1) were for “moral role modeling,” and three items (Q13, Q14, and Q15, in Table 1) were for the “moral atmosphere creation” factor. When we freed these nine loading values from the invariant condition, the fitness indexes of this partial loading value invariance [χ? (df = 203) = 720.22, CFI = .971, TLI = .966, RMSEA = .070, SRMR = .034] were not worse (?χ? = .85, df = 3) than the fitness indexes for the two-group configural baseline model (Model 1), indicating that the partial loading values invariance (or partial metric invariance) of the ELM was achieved. A full metric invariance is a precondition to testing the scalar equivalence (Byrne, 2012; Vandenberg & Lance, 2000), so we stopped the analysis and concluded that the ELM has configural invariance and partial loading values invariance, and the second-order factor ELM holds across the Chinese and U.S. cultures.DiscussionTheoretical Implications This research makes a number of theoretical contributions to the literature. First, we have advanced the field of ethical leadership research by examining a widely accepted theoretical model in terms of its cultural universality. Although the number of empirical studies in this area is increasing (e.g., Avey et al., 2011; Brown et al., 2005; Kacmar, Daniel, Kenneth, & Zivnuska, 2011; Mayer et al., 2012; Piccolo et al., 2010; Walumbwa & Schaubroeck, 2009), ethical leadership remains an under-researched leadership style from a cross-cultural perspective. The present study contributes to the ethical leadership field by examining the universality of ethical leadership structure in a cross-cultural context. This perspective is based on the alignment between the original conceptualization of ethical leadership proposed by Trevi?o et al. (2000) in a Western culture and the Chinese philosophical interpretation of ethical leadership (Confucius, 2006). In Eastern cultures such as China, ethical leadership structures consist of two aspects: “moral person” and “moral manager.” This conceptualization resonates with traditional Chinese culture, which proposes that xiuji and anren are two essential pillars of ethical leadership (Confucius, 2006). The ELS (Brown et al., 2005) was developed in a Western culture; in contrast, our scale was developed in an Eastern culture and tested in both an Eastern culture and a Western culture. Thus, the ELM’s development makes a significant contribution to cross-cultural management and leadership research. We found that the ELM has measurement invariance (configural invariance and partial loading value invariance) across the Chinese and U.S. contexts. In other words, the general second-order construct of the ELM holds across two cultures. The subdimension of “moral cognition” also achieved the full loading value invariance, indicating that both Chinese and American subjects attributed the same meaning to the latent construct of “moral cognition” (via similar relative relationships between each item and its latent construct). Nine out of 12 items for the three other subdimensions (moral characteristics, moral role modeling, and moral atmosphere creation) were not invariant across these two samples, however. Thus, Chinese and American employees appear to have rather different views and understanding of these items, the relative relationships among these items, and the latent construct of each subdimension. Interestingly, the loading values of the “moral characteristics” factor—in particular, the two items “My leader is benevolent” and “My leader often treats others mercifully”—differed markedly across the Chinese and U.S. samples. Chinese managers appear to need to exhibit a more “benevolent orientation,” which is an implicit model of Chinese (Eastern) culture (Confucius, 2006). This also aligns with the work of Resick and his colleagues (2006, 2011), who showed that in China, consideration and respect for others are highly important characteristics of ethical leaders. The ELM has several unique strengths. First, it successfully captures the more comprehensive and refined contents of the two broader aspects of leadership (“moral person” and “moral manager”). Second, the scale incorporates moral cognition as a subdimension of the “moral person” aspect. The inclusion of moral cognition into the model has the advantage of capturing the cognitive element of ethical leaders being moral persons. The new scale encompasses moral self-conception (i.e., moral characteristics: what type of person is the leader), moral cognition (i.e., how does the leader think ethically), and moral managerial behaviors (i.e., “moral role modeling” and “moral atmosphere creation”: what should the leader do ethically as a manager). Finally, the ELM explicitly stresses and measures ethical leaders’ role in developing a moral atmosphere among groups and followers, which is aligned with the original conceptualization that developing a moral atmosphere is an important role of ethical leaders (Trevi?o et al., 2000). As to the criteria-related validity, we found positive relationships between ethical leadership and follower work-related outcomes, including affective organizational commitment, OCBs, job performance, and follower moral attentiveness. We also established incremental predictive validity by showing that ELM-identified behaviors contributed to the explanation of variance in the dependent variables after controlling for the ELS.Practical Implications Our research has several practical implications. First, we found that in different cultures (specifically China and the United States), ethical leaders need to be both a “moral person” and a “moral manager.” Therefore, to be ethical leaders who can influence employee outcomes, leaders must be viewed as credible and legitimate moral role models. In particular, they need to fully exhibit their “moral person” traits, such as kind-heartedness, compassion, and an altruistic orientation. In addition, leaders need to demonstrate “moral manager” behaviors such as creating a fair work environment for employees while treating followers with respect and dignity. This has implications not only for leaders who aspire to be ethical in terms of how they lead, but also for leadership training and development programs in organizations, and for the managers who work now or will work in the future in cross-cultural contexts. Second, this research indicates that leaders should adhere to their inner moral values and make fair and ethical decisions in morally challenging situations to motivate their followers and to achieve sustainable development. Today’s society is full of unethical temptations and challenges in business contexts around the world. When leaders act morally and make ethical decisions with their subordinates’ best interests in mind, they help encourage and develop ethical behaviors within the organization, thereby combating those temptations and mitigating the challenges. It is important that ethical leaders exhibit “moral manager” behaviors such as shaping the ethical context of the organization. Given the important and positive impacts of ethical leadership, organizations would be well advised to promote ethical leadership across different levels of their organizations. First, ethical leadership may be used as a criterion for the recruitment and promotion of employees to managerial positions. Specifically, organizations can assess managerial candidates’ propensity to exhibit ethical leadership behaviors. During recruitment interviews, interviewers should clearly emphasize ethics and seriously take ethics into account when selecting candidates for managerial positions. Second, organizations can provide leaders with ethics-based training. Trevi?o (1992), for example, suggests that moral reasoning skills can be developed through appropriate training programs. Organizations might design training programs that include practice in moral role-taking or situational behaviors, group discussion, and case studies to improve managers’ ability to deal with complex ethical issues. Additionally, during the training process, managers might be encouraged to fully appreciate the importance of their behaviors and ability to affect followers’ moral outcomes. Finally, Chinese leaders and managers in Chinese organizations should adopt these Chinese philosophies in their work as leaders. For example, leaders should properly understand the Confucian teaching of xiuji anren and apply the teaching into their leadership practices through cultivating their moral character (i.e., xiuji), and then lead their followers to a more positive state of well-being (i.e., anren) in organizations. In the same vein, Chinese organizations should provide relevant training to their managers to help them better and more accurately understand the Confucian philosophy of xiuji anren. For their part, American organizations should provide relevant ethical leadership training based on the theoretical framework of “moral person” and “moral manager.”Limitations and Future Directions Our study has several limitations. First, when testing the predictive validity of the ELM, although we collected measures of independent variables and dependent variables at two points in time within a month to eliminate cross-sectional bias, it was still difficult to ascertain the true direction of causality between the independent (i.e., ethical leadership) and dependent (i.e., follower job-related and ethically related outcomes) variables. To resolve this issue, future studies might adopt experimental or longitudinal approaches. For example, researchers might randomly assign subjects to different ethical leadership intervention groups and examine the outcome variables to confirm the causal relationship. Second, this new scale was developed and tested in the contexts of China and the United States. Future research might be conducted in other cultural contexts as well. Although we demonstrated the configural invariance and partial metric invariance for the ELM across Chinese and American samples, we suggest that future research should further explore cross-cultural similarities and differences of ethical leadership in other cultures. It would also be interesting to examine (1) to what extent ethical leadership affects follower outcomes across different cultures, (2) whether cultural values influence ethical leadership and ethical values (Jackson, 2001), and (3) whether moral traits and integrity differ across different generations (e.g., Audi & Murphy, 2006; Bandsuch, 2009) and cultures. In addition, future research might compare the ELM with other relevant scales, including paternalistic leadership (Cheng et al., 2004) and the ELW questionnaire (Kalshoven et al., 2011a) to further establish the convergent and discriminant validity of the ELM, and to shed light on the complex cross-cultural dynamics in ethical leadership. Thanks to an anonymous reviewer’s insightful comment, we acknowledge that one limitation of our new scale is the inclusion of the word “ethical” in some of the items. One potential issue with the use of this word is that followers may have different notions about what is “ethical.” One way to help respondents interpret the meaning of “ethical” in similar ways is to share with them the definition of “ethical” before they respond to the survey items. As an example, when the ELM is used in a Chinese context, respondents could be informed that “being ethical” means xiuji anren in Confucian teaching. Future researchers might also ask both followers and leaders to rate leader moral characteristics on the ELM to better capture the construct of ethical leadership. In conclusion, we integrate traditional Chinese philosophies (Confucius, 2006) with the Western conceptualization of ethical leadership (Trevi?o et al., 2000) to explore aspects of ethical leadership that are indigenous to the Chinese culture and can be adapted and utilized in the Western culture. Through developing a reliable and valid scale, we contribute to cross-cultural research on ethical leadership and lay a solid foundation for the research to grow and flourish in the field. REFERENCESAlexander, S., & Ruderman, M. (1987). The role of procedural and distributive justice in organizational behavior. Social Justice Research, 1(2), 177–198.Aquino, K., & Reed, A. (2002). The self-importance of moral identity. Journal of Personality and Social Psychology, 83(6), 1939–1315.Audi, R., & Murphy, P. E. (2006). The many faces of integrity. Business Ethics Quarterly, 16, 3–21.Avey, J. B., Palanski, M. E., & Walumbwa, F. O. (2011). When leadership goes unnoticed: The moderating role of follower self-esteem on the relationship between ethical leadership and follower behavior. Journal of Business Ethics, 98(4), 573–582.Avolio, B. J., & Gardner, W. L. (2005). Authentic leadership development: Getting to the root of positive forms of leadership. Leadership Quarterly, 16(3), 315–338.Bandsuch, M. R. (2009). Understanding integrity across generations in China: Implications for personnel choices in Chinese corporations. Journal of International Business Ethics, 2 (2), 21–38. Bass, B. M., & Riggio, R. E. (2006). Transformational leadership. Mahwah, NJ: Lawrence Erlbaum Associates.Bass, B. M., & Steidlmeier, P. (1999). Ethics, character and authentic transformational leadership behavior. Leadership Quarterly, 10(2), 181?217.Billiet, J. B., & McClendon, M. J. (2000). Modeling acquiescence in measurement models for two balanced sets of items. Structural Equation Modeling, 7(4), 608–628.Bollen, K. A. (1989). Structural equations with latent variables. New York, NY: John Wiley.Bollen, K. A., & Lennox. R. (1991). Conventional wisdom on measurement: A structural equation perspective, Psychological Bulletin, 110(2), 305?314.Brown, M. E., Trevi?o, L. K., & Harrison, D. A. (2005). Ethical leadership: A social learning perspective for construct development and testing. Organizational Behavior and Human Decision Processes, 97(2), 117–134.Byrne, B. (2008). Testing for multigroup equivalence of a measuring instrument: A walk through the process. Psicothema, 20(4), 872–882.Byrne, B. M. (2012). Structural equation modeling with Mplus: Basic concepts, applications, and programming. New York, NY: Routledge.Chen, F., Sousa, K. H., & West, S. G. (2005). Testing measurement invariance of second-order factor models. Structural Equation Modeling, 12(3), 471–492.Chen, L. (2010). Virtue ethics and Confucian ethics. Dao, 9, 275–287.Chen, X. P., Eberly, M. B., Chiang, T. J., Farh, J. L., & Cheng, B. S. (2014). Affective trust in Chinese leaders: Linking paternalistic leadership to employee performance. Journal of Management, 40(3), 796–819.Chen, Y. R., Leung, K., & Chen, C. C. (2009). Bringing national culture to the table: Making a difference with cross-cultural differences and perspectives. Academy of Management Annals, 3, 217?249.Cheng, B. S., Chou, L. F., Wu, T. Y., Huang, M. P., & Farh, J. L. (2004). Paternalistic leadership and subordinate responses: Establishing a leadership model in Chinese organizations. Asian Journal of Social Psychology, 7(1), 89–117.Cheung, G. W., & Rensvold, R. B. (2002). Evaluating goodness of fit indexes for testing measurement invariance. Structural Equation Modeling, 9(2), 233–255.Coltman, T., Devinney, T. M., Midgley, D. F., & Venaik, S. (2008). Formative versus reflective measurement models: Two applications of formative measurement. Journal of Business Research, 61(12), 1250?1262.Confucius. (2006). Original Chinese title [Confucian analects] (Y. Zhang, Ed.). [Original work published during the Warring States period).] Beijing, China: Zhonghua Book. Darlington, R. B. (1990). Regression and linear models. New York, NY: McGraw-Hill.De Hoogh, A . H. B., & Den Hartog, D . N. (2008). Ethical and despotic leadership, relationships with leader’s social responsibility, top management team effectiveness and subordinates’ optimism: A multi-method study. Leadership Quarterly, 19 (3), 297–311.Edwards, J. R. (2001). Multidimensional constructs in organizational behavior research: An integrative analytical framework. Organizational Research Methods, 4, 144?192.Edwards, J. R. (2011). The fallacy of formative measurement. Organizational Research Methods, 14, 370?388.El Akremi, A., Gond, J. P., Swaen, V., De Roeck, K., & Igalens, J. (2015). How do employees perceive corporate responsibility? Development and validation of a multidimensional corporate stakeholder responsibility scale. Journal of Management.Fabrigar, L. R., Wegener, D. T., MacCallum, R. C., & Strahan, E. J. (1999). Evaluating the use of exploratory factor analysis in psychological research. Psychological Methods, 4(3), 272–299.Farh, J. L., Earley, P. C., & Lin, S. C. (1997). Impetus for action: A cultural analysis of justice and organizational citizenship behavior in Chinese society. Administrative Science Quarterly, 44(3), 421–444.Fornell, C., & Bookstein, F. L. (1982). Two structural equation models: LISREL and PLS applied to consumer exit-voice theory. Journal of Marketing Research, 19(4), 440–452.Fornell, C. R., & Larcker, D. F. (1981). Structural equation models with unobservable variables and measurement error. Journal of Marketing Research, 18(2), 39-50.Greenleaf, R. K. (1977). Servant leadership: A journey into the nature of legitimate power and greatness. Mahwah, NJ: Paulist Press.Hannah, S., Jennings, P. L., Bluhm, D., Peng, A. C., & Schaubroeck, J. M. (2014). Duty orientation: Theoretical development and preliminary construct testing. Organizational Behavior and Human Decision Processes, 123(2), 598?621.Hinkin, T. R. (1998). A brief tutorial on the development of measures for use in survey questionnaires. Organizational Research Methods, 1(1), 104–121.Hu, L., & Bentler, P. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6(1), 1–55. Jackson, T. (2001). Cultural values and management ethics: A 10-nation study. Human Relations, 54, 1267–1302.Jarvis, C. B., MacKenzie, S. B., & Podsakoff, P. M. (2003). A critical review of construct indicators and measurement model misspecification in marketing and consumer research. Journal of Consumer Research, 30(2), 199?218.Johnson, R. E., Rosen, C. C., & Chang, C. H. (2011). To aggregate or not to aggregate: Steps for developing and validating higher-order multidimensional constructs. Journal of Business and Psychology, 26, 241?248.Kacmar, M. K., Daniel, G. B., Kenneth, J. H., & Zivnuska, S. (2011). Fostering good citizenship through ethical leadership: Exploring the moderating role of gender and organizational politics. Journal of Applied Psychology, 96(3), 633–642.Kalshoven, K., Den Hartog, D. N., & De Hoogh, A. H. B. (2011a). Ethical Leadership at Work questionnaire (ELW): Development and validation of a multidimensional measure. Leadership Quarterly, 22(1), 51–60.Kalshoven, K., Den Hartog, D. N., & De Hoogh, A. H. (2011b). Ethical leader behavior and big five factors of personality. Journal of Business Ethics, 100(2), 349-366.Kinicki, A., Jacobson, K. J., Peterson, S. J., & Prussia, G. E. (2013). Development and validation of the performance management behavior questionnaire. Personnel Psychology, 66, 1?45.Kline, P. (1979). Psychometrics and psychology. London, UK: Academic Press.Kovjanic, S., Schuh, S. C., Jonas, K., Van Quaquebeke, N., & Van Dick, R. (2012). How do transformational leaders foster positive employee outcomes? A self-determination?based analysis of employees’ needs as mediating links. Journal of Organizational Behavior, 33, 1031?1052.Lau, D. C. (1970a). The analects. Harmondsworth, NY: Penguin Books.Liden, R. C., Wayne, S. J., Zhao, H., & Henderson, D. (2008). Servant leadership: Development of a multidimensional measure and multi-level assessment. Leadership Quarterly, 19(2), 161–177.Linderbaum, B. A., & Levy, P. E. (2010). The development and validation of the Feedback Orientation Scale (FOS). Journal of Management, 36(6), 1372?1405.Little, T. D., Cunningham, W. A., Shahar, G., & Widaman, K. F. (2002). To parcel or not to parcel: Exploring the question, weighing the merits. Structural Equation Modeling, 9, 151–173. Liu, J., Kwan, H. K., Fu, P. P., & Mao. Y. (2013). Ethical leadership and job performance in China: The roles of workplace friendships and traditionality. Journal of Occupational and Organizational Psychology, 86(4), 564-584.MacCallum, R. C., Widaman, K. F., Zhang, S., & Hong, S. (1999). Sample size in factor analysis. Psychological Methods, 4(1), 84–99.MacKenzie, S. B., Podsakoff, P. M., & Jarvis, C. B. (2005). The problem of measurement model misspecification in behavioral and organizational research and some recommended solutions. Journal Applied Psychology, 90(4), 710?730.Mayer, D. M., Aquino, K., Greenbaum, R. L., & Kuenzi, M. (2012). Who displays ethical leadership and why does it matter? An examination of the antecedents and consequences of ethical leadership. Academy of Management Journal, 55(1), 151–171.Mayer, D. M., Kuenzi, M., Greenbaum, R., Bardes, M., & Salvador, R. B. (2009). How low does ethical leadership flow? Test of a trickle-down model. Organizational Behavior and Human Decision Processes, 108(1), 1–13.Meredith, W. (1993). Measurement invariance, factor analysis and factorial invariance. Psychometrika, 58, 525–543.Mowday, R. T., Steers, R. M., & Porter, L. W. (1979). The measurement of organizational commitment. Journal of Vocational Behavior, 14(2), 224–247.Muthén, L. K., & Muthén, B. O. (1998–2010). Mplus user’s guide (5th ed.). Los Angeles, CA: Muthén & Muthén.Nunnally, J. C., & Bernstein, I. H. (1994). Psychometric theory (3rd ed.). New York, NY: McGraw-Hill.Ogunfowora, B. (2014). The impact of ethical leadership within the recruitment context: The roles of organizational reputation, applicant personality, and value congruence. Leadership Quarterly, 25(3), 528?543.Piccolo, R. F., Greenbaum, R., Den Hartog, D. N., & Folger, R. (2010). The relationship between ethical leadership and core job characteristics. Journal of Organizational Behavior, 31(2–3), 259–278.Resick, C., Hanges, P., Dickson, M., & Mitchelson, J. (2006). A cross-cultural examination of the endorsement of ethical leadership. Journal of Business Ethics, 63(4), 345?359.Resick, C., Martin, G. S., Keating, M., Dickson, M. W., Kwan, H. K., & Peng, A. C. (2011). What ethical leadership means to me: Asian, American, and European perspectives. Journal of Business Ethics, 101(3), 435–457.Reynolds, S. J. (2008). Moral attentiveness: Who pays attention to the moral aspects of life? Journal of Applied Psychology, 93(5), 1027–1041.Riggio, R. E., Zhu, W., Reina, C., & Maroosis, J. A. (2010). Virtue-based measurement of ethical leadership: The Leadership Virtues Questionnaire. Consulting Psychology Journal: Practice and Research, 62(4), 235–250.Schaubroeck, J. M., Hannah, S. T., Avolio, B. J., Kozlowski, S. W. J., Lord, R. G., Trevi?o, L. K., & Peng, A. C. (2012). Embedding ethical leadership within and across organization levels. Academy of Management Journal, 55(5), 1053–1078.Schwab, D. P. (1980). Construct validity in organizational behavior. In L. L. Cummings & B. M. Staw (Eds.), Research in organizational behavior (pp. 3?43). Greenwich, CT: JAI Press. Trevi?o, L. K. (1992). The social effects of punishment in organizations: A justice perspective. Academy of Management Review, 17(4), 647–676.Trevi?o, L. K., Hartman, L., & Brown, M. (2000). Moral person and moral manager: How executives develop a reputation for ethical leadership. California Management Review, 42(4), 128–142.Vandenberg, R. J., & Lance, C. E. (2000). A review and synthesis of the measurement invariance literature: Suggestions, practices and recommendations for organizational research. Organizational Research Methods, 3, 4?70.van de Schoot, R., Lugtig, P., & Hox, J. (2012). A checklist for testing measurement invariance. European Journal of Developmental Psychology, 9(4), 486–492.Walumbwa, F. O., Avolio, B. J., Gardner, W. L., Wernsing, T. S., & Peterson, S. J. (2008). Authentic leadership: Development and validation of a theory-based measure. Journal of Management, 34(1), 89–126.Walumbwa, F. O., & Schaubroeck, J. (2009). Leader personality traits and employee voice behavior: Mediating roles of ethical leadership and work group psychological safety. Journal of Applied Psychology, 94(5), 1275–1286.Yukl, G., Mahsud, R., Hassan, S., & Prussia, G. E. (2013). An improved measure of ethical leadership. Journal of Leadership & Organizational Studies, 20(1), 38?48.Zhu, W., Avolio, B. J., Riggio, R., & Sosik, J. (2011). The effect of authentic transformational leadership on follower and group ethics. Leadership Quarterly, 22(5), 801–817.Zhu, W., He, H., Trevi?o, L., Chao, M., & Wang, W. (2015). Ethical leadership and follower voice and performance: The role of follower identifications and entity morality beliefs. Leadership Quarterly, 26, 702–718.Table 1Exploratory Factor Analysis (Study 2)My immediate supervisor Factor 1Factor 2Factor 3Factor 4Sourcesis willing to help .71?.01?.07 .16Interview/open questionsis kind-hearted .82?.06 .09 .01Interview/open questionsis benevolent .77 .03 .17?.01Interview/open questionsoften treats others mercifully .79 .12 .04?.07Interview/open questionsthinks about the moral implications of his/her actions frequently .17 .80?.04?.04Adapted from Reynoldsoften considers issues from an ethical perspective .03 .92?.07 .04Adapted from Reynoldsreflects on the moral appropriateness of his/her decisions?.08 .91 .11 .01Adapted from Reynoldsfrequently thinks about ethical issues?.09 .74 .11 .16Adapted from Reynoldssets an example of how to do things in the ethically correct way .03 .04 .94 .01Adapted from Brown et al. (2005)often makes decisions consistent with his/her moral principles .05 .13 .77 .01Interview/open questionssets examples of ethical behaviors to subordinates .07 .00 .87?.01Interview/open questionspractices moral behaviors and serves as role models to subordinates in terms of behaving ethically .04 .03 .72 .16Interview/open questionsrequires subordinates to learn and understand the codes of ethics .02?.02?.03 .82Interview/open questionsprovides constructive feedback to subordinates regarding ethical conduct and standards .11 .00?.00 .83Interview/open questionsexplains the values that guide his/her moral decisions to subordinates ?.04?.01 .25 .67Interview/open questionsensures that proper procedures are available for subordinates to consult on moral issues .02 .11?.02 .68Interview/open questionsFactor nameMoral characteristics Moral cognitionMoral role modeling Establish moralcontext Cronbach’s alpha (subscales).94.88.93.92Cronbach’s alpha (ELM).97Average variance extracted (AVE)59.84%71.55%68.80%56.84%Table 2Definition and Description of Ethical Leadership DimensionsItems Factor 1Factor 2Factor 3Factor 4Factor name Moral characteristics Moral cognitionMoral role modelingCreating moral atmosphereDefinition Whether an individual possesses traits and characteristics that are relevant to ethics and morality, and demonstrates willingness and tendency to behave ethicallyAn individual’s awareness of and attention to a moral issue that has moral implications, and his or her capacity to make relevant judgmentsThe leader’s ability to adopt a higher ethical standard in his or her own behaviors, practice the ethical teaching he or she preaches, and set a visible ethical example for followersThe important roles a leader plays in building up a moral context through communications, setting standards, and developing codes/rules for enhancing subordinates’ ethical behaviorsWhat it intends to measureWhat a manager ethically isWhat a manager thinks ethicallyWhat a manager does to act as a moral role model to his or her subordinatesWhat a manager does to develop a moral context for the teamXiuji anren aspectsXiujiXiujiAnrenAnren “Moral person”/“moral Manager” categories Moral person Moral personMoral manager Moral manager Table 3Confirmatory Factor Analysis (Study 3)Modelχ?(df)?χ?(df)CFITLIRMSEASRMRProposed four-factor, first-order model 370.28(98).98.97.06.02Proposed four-factor, second-order model487.06(100)116.78**(2).97.96.07.03Alternative four-factor (merging moral role modeling and establishing moral contexts), first-order model 883.64(101) 513.36**(3).94.93.10.04Alternative three-factor (merging moral role modeling and establishing moral contexts), first-order model1370.06(101)999.78**(3).90.88.13.05Alternative one-factor model 1944.42(104)1574.14**(6).86.83.15.06*p < .05; ** p < .01.Table 4Descriptive Statistics and Correlations among Leadership Styles (Convergent and Divergent Validity, Study 4)MeanSD12345678910111213141. Moral characteristics6.50 .80 .932. Moral cognition5.65 1.22 .51** .963. Moral role modeling 6.49 .77 .86** .56** .954. Moral atmosphere creation 6.15 .89 .65** .67** .78** .905. ELM6.20 .79 .84** .84** .89** .89** .966. ELS4.45 .60 .74** .64** .82** .78** .85** .937. Idealized influence 4.36 .59 .64** .59** .69** .64** .73** .79** .858. Authentic leadership4.34 .57 .52** .50** .57** .59** .63** .65** .62** .709. Servant leadership 6.44 .84 .62** .47** .67** .61** .67** .72** .64** .51** .9010. Leader virtues 4.49 .63 .70** .51** .73** .66** .73** .76** .71** .58** .63** .8411. Leader narcissism 2.57 .71 –.39**–.20**–.42**–.27**–.35**–.38**–.36**–.23**–.37**–.39** .9312. Abusive supervision1.16 .45 –.31**–.11*–.32**–.22**–.26**–.30**–.25**–.21**–.29**–.22** .58** .9513. Age 2.60 .72 .02 .04?.01 .02 .03?.00?.04 .03 .01?.05 .01?.09 X14. Education 2.82 .50?.01?.07?.05?.03?.05?.01?.07?.00?.08?.02 .02 .01?.11* X15. Firm size 1.971.36 .07 .07 .06 .05 .07 .04 .05 .07 .08 .07?.02 .02?.02?.08Note: Numbers in the orthogonal are the Cronbach’s alpha values. The answer anchor of ELM and servant leadership ranges from 1 to 7; the answer anchor for the other scales ranges from 1 to 5. *p < .05; ** p < .01.Table 5Descriptive Statistics and Correlations among Variables (Predictive Validity, Study 5)Mean SD1234567891011121. Tenure under supervisor2.662.43 NA2. ELS3.96 .75–.10 .923. Moral characteristics5.921.14 –.01 .63**.944. Moral cognition5.18 1.37–.08 .70**.56** .955. Moral role modeling 5.80 1.19 .02 .75**.81** .62**.946. Moral atmosphere creation5.531.14–.03 .73**.68** .74**.75** .927. ELM5.611.06–.03 .80**.86** .85**.90** .90** .978. Organizational commitment 4.61 .94–.02 .35**.31** .31**.29** .35** .36** .909. OCB6.30 .68–.03 .24**.23** .23**.24** .19** .25** .05 .9110. Job performance 3.81 .75–.11 .12*.15** .13*.10 .13* .15* .07 .57**.9311. Follower moral attentiveness4.471.17 .03 .11.08 .18**.07 .23** .16** .10 .31**.30** .9112. Organizational level3.95 .25 .00?.06.01?.00.01?.02?.00?.17**?.03.18**?.04NANote: Numbers in the orthogonal are the Cronbach’s alpha values. The answer anchor of ELM ranges from 1 to 7; the answer anchor for the other scales ranges from 1 to 5.*p < .05; ** p < .01.Table 6Incremental Regression Analysis of Ethical Leadership Measurement versus Ethical Leadership Scale on Follower Affective Organizational Commitment (Predictive Validity, Study 5)Step 1Step 2a(Pure ELM)Step 2b(Pure ELS)Step 3ββββTenure under supervisor ?.02 ?.01.02 .00Organizational level?.17** ?.17**?.15**?.17**ELS——.34*** .14ELM—.36***— .25**R23.10% 15.80%14.30% 16.50%Adjusted R22.40% 15.00%13.40% 15.30%Step 2 △R2—12.80%**11.20%**— Step 3 △R2 compared to Step 2a——— .60%Step 3 △R2 compared to Step 2b——— 2.20%**F4.60*18.18***16.10***14.24***% increase in R2 compared to Step 2b 14.30%Note: Step 2a added ELM to Step 1; Step 2b added ELS to Step 1.*p < .05; **p < .01; *** p < .001.Table 7Incremental Regression Analysis of Ethical Leadership Measurement versus Ethical Leadership Scale on Follower OCBs (Predictive Validity, Study 5)Step 1Step 2a(Pure ELM)Step 2b(Pure ELS)Step 3 ββΒβTenure under supervisor?.03 ?.02 ?.00?.01Organizational level?.03 ?.03 ?.02?.02ELS——.24***.11ELM— .25***—.17?R2 .20%6.50% 6.00% 6.90%Adjusted R2 ?.50%5.60% 5.00% 5.60%Step 2 △R2— 6.30%** 5.80%***—Step 3 △R2 compared to Step 2a——— .40%Step 3 △R2 compared to Step 2b——— 1.00%?F.256.74*** 6.13*** 5.37***% increase in R2 compared to Step 2b 16.67%Note: Step 2a added ELM to Step 1; Step 2b added ELS to Step 1.? p < .10; *p < .05; **p < .01; *** p < .001.Table 8Incremental Regression Analysis of Ethical Leadership Measurement versus Ethical Leadership Scale on Follower Job Performance (Predictive Validity, Study 5)Step 1Step 2a(Pure ELM)Step 2b(Pure ELS)Step 3ββΒβTenure under supervisor?.11??.10? ?.10?.10?Organizational level .18** .18** .19*** .18**ELS——.12*.01ELM— .14*— .14?aR2 4.40% 6.40%5.80% 6.40%Adjusted R2 3.70% 5.50% 4.80% 5.10%Step 2 △R2 2.10%* 1.40%*—Step 3 △R2 compared to Step 2a——— .00%Step 3 △R2 compared to Step 2b——— .70%F 6.63** 6.63**5.91** 4.96***% increase in R2 compared to Step 2b12.07%Note: Step 2a added ELM to Step 1; Step 2b added ELS to Step 1.? p < .10; *p < .05; **p < .01; ***p < .001.a: Based on one-tailed test.Table 9Incremental Regression Analysis of Ethical Leadership Measurement versus Ethical Leadership Scale on Follower Moral Attentiveness (Predictive Validity, Study 5)Step 1Step 2a(Pure ELM)Step 2b(Pure ELS)Step 3ββΒβTenure under supervisor .03 .03 .04 .03Organizational level?.04?.04?.03?.04ELS—— .11??.06ELM— .16**— .21*R2 .20% 2.80% 1.40% 3.00%Adjusted R2 ?.50% 1.80% .40% 1.60%Step 2 △R2— 2.60%** 1.20%?—Step 3 △R2 compared to Step 2a——— .10%Step 3 △R2 compared to Step 2b——— 1.60%*F.302.81*1.362.20?% increase in R2 compared to Step 2b 114.29%Note: Step 2a added ELM to Step 1; Step 2b added ELS to Step 1.? p < .10; *p < .05; **p < .01; *** p < .001.Table 10Summary of Fit Statistics for Testing Measurement Invariance of Second-Order Factor Model of Ethical Leadership Measure (Measurement Invariance, Study 6)Modelχ?dfCFITLIRMSEASRMRModel Comparison ?χ??dfModel 1: Two-groupconfigural invariance719.37200.97.97.07.03Model 2:First- and second-order factor loadings invariant814.24212.97.96.07.072 vs. 194.87**12Model 3: Partial first- and second-order factor loadings invariant 720.22203.97.97.07.033 vs. 1.85 3Note: *p < .05; **p < .01; ***p < .001.Q1Q2Q3Q4Q5Q6Q7Q8Q9Q10Q11Q12Q13Q14Q15Q16Moral characteristicsMoral cognitionMoral role modelingMoral atmosphere creationEthical leadership measure (ELM).93 (.87).85 (.98).94 (.98).94 (.94).84 (.80).89 (.90).86 (.89).86 (.94).86 (.93).80 (.94).91 (.95).85 (.93).88 (.92).88 (.89).90 (.90).65 (.90).89 (.88).90 (.92).93 (.65).88 (.58)Q1Q2Q3Q4Q5Q6Q7Q8Q9Q10Q11Q12Q13Q14Q15Q16Moral characteristicsMoral cognitionMoral role modelingMoral atmosphere creationEthical leadership measure (ELM).93 (.87).85 (.98).94 (.98).94 (.94).84 (.80).89 (.90).86 (.89).86 (.94).86 (.93).80 (.94).91 (.95).85 (.93).88 (.92).88 (.89).90 (.90).65 (.90).89 (.88).90 (.92).93 (.65).88 (.58)Figure 1. Confirmatory factor analyses (CFA, Study 6)Notes: The values in the parentheses are for the U.S. sample. The average variance explained (AVE) indexes for moral characteristics, moral cognition, moral role modeling, establishing moral context, and ELM are .82 (.60), .70 (.81), .73 (.88), .73 (.88), and .74 (.78), respectively, for these two samples.Appendix 1: English Version of Ethical Leadership MeasureMy leader 1.is willing to help 2.is kind-hearted3.is benevolent4.often treats others mercifully5.thinks about the moral implications of his/her actions frequently 6.often considers issues from an ethical perspective7.reflects on the moral appropriateness of his/her decisions8.frequently thinks about ethical issues9.sets an example of how to do things in the ethically correct way10.often makes decisions consistent with his/her moral principles11.sets examples of ethical behaviors to subordinates12.practices moral behaviors and serves as role models to subordinates in terms of behaving ethically13.requires subordinates to learn and understand the codes of ethics14.provides constructive feedback to subordinates regarding ethical conduct and standards 15.explains the values that guide his/her moral decisions to subordinates16.ensures that proper procedures are available for subordinates to consult on moral issuesAppendix 2: Chinese Version of Ethical Leadership Measure我的领导：乐于助人心地善良是一个充满仁爱的人常常以慈悲之心待人每天都会想着自己行为是否合乎道德标准经常深刻思考道德问题会经常反思所做决定是否符合道德标准喜欢思考道德方面的事情在道德准则下如何行事方面为下属树立了榜样经常做出符合道德规则的决策在道德行为方面能为下属产生示范作用在行动中恪守道德价值观方面，给下属做出了表率要求员工确认已经阅读并理解了道德准则经常给员工提供有关道德行为和标准的建设性反馈会向员工阐明指导他/她道德决策和行为的价值观确信组织中已建立了相应程序，以使得员工能够询问道德准则的要求Note: The item order in the Chinese version is the same as that in the English version. ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download