FUNDAÇÃO GETULIO VARGAS ESCOLA BRASILEIRA DE ADMINISTRAÇÃO PÚBLICA E DE EMPRESAS DOUTORADO EM ADMINISTRAÇÃO

IN THE (VALUE-LADEN) EYES OF THE BEHOLDER: LINKING FOLLOWER PERCEPTIONS TO LEADERSHIP EMERGENCE AND EFFECTIVENESS

TESE APRESENTADA À ESCOLA BRASILEIRA DE ADMINISTRAÇÃO PÚBLICA E DE EMPRESAS PARA OBTENÇÃO DO TÍTULO DE DOUTOR

GUSTAVO MOREIRA TAVARES Rio de Janeiro - 2019

FUNDAÇÃO GETÚLIO VARGAS ESCOLA BRASILEIRA DE ADMINISTRAÇÃO PÚBLICA E DE EMPRESAS

GUSTAVO MOREIRA TAVARES

IN THE (VALUE-LADEN) EYES OF THE BEHOLDER: LINKING FOLLOWER PERCEPTIONS TO LEADERSHIP EMERGENCE AND EFFECTIVENESS

Dissertação de Doutorado apresentada à Escola Brasileira de Administração Pública e de Empresas da Fundação Getulio Vargas como requisito parcial para a obtenção do título de Doutor em Administração.

Orientador: Prof. Dr. Filipe João Bera de Azevedo Sobral

RIO DE JANEIRO 2019

Dados Internacionais de Catalogação na Publicação (CIP) Ficha catalográfica elaborada pelo Sistema de Bibliotecas/FGV

Tavares, Gustavo Moreira In the (value-laden) eyes of the beholder : linking follower perceptions to leadership emergence and effectiveness / Gustavo Moreira Tavares. – 2019. 124 f.

Tese (doutorado) - Escola Brasileira de Administração Pública e de Empresas, Centro de Formação Acadêmica e Pesquisa. Orientador: Filipe Bera de Azevedo Sobral. Inclui bibliografia.

1. Liderança transformacional. 2. Motivação no trabalho. 3. Clima organiza- cional. 4. Administração pública. I. Sobral, Filipe. II. Escola Brasileira de Administração Pública e de Empresas. Centro de Formação Acadêmica e Pesquisa. III. Título.

CDD – 352.39

Elaborada por Márcia Nunes Bacha – CRB-7/4403

Dedico este trabalho a meus pais Maria Rita e Geraldo, minha esposa Luana e meu filho Pedro.

ACKNOWLEDGEMENTS

I am sincerely and heartily grateful to my advisor, Filipe Sobral, for the support, patience, and guidance throughout my PhD journey. Filipe was extremely competent in helping me become a researcher (since my Masters). The doors he opened to me helped me reach places I never imagined would be possible. I am also immensely thankful to

Bradley Wright, who generously welcomed and advised me during my stay at the

University of Georgia. Bradley helped me improve my work and research skills tremendously. It is a great honor to have him as a member of my defense’s committee. I would also like to express my immeasurable gratitude to James Perry, who provided valuable feedback to my work at its early stages and who gently welcomed me at the

University of Indiana. I am truly indebted to these gentlemen.

I am also very thankful to Ishani Aggarwal, Sergio Fernandez, and Liliane Furtado for kindly accepting being members of my defense’s committee (and qualifying exam).

Their constructive feedback was fundamental for the improvement of this work.

I would like to thank EBAPE-FGV for providing me with the best professors, staff, and structure—and free expresso coffee! EBAPE was my second home for 5 years, wherein I could grow both as a human being and scholar. I also thank IFRJ, my employer, for giving me all the support I needed during my PhD program.

I am also grateful to my colleagues who directly or indirectly contributed to my work (by providing advice, helping me design the studies and collect data, or just by being good friends). They are (in alphabetical order): Claudio, Duque, Fernanda C., Gustavo

C., Juliana C., Juliana K., Kelly, Martins, Miguel, Rafael V., and Ula. I could not forget to thank the 1,842 anonymous participants who voluntarily gave me an amount of the most valuable thing they have: their time.

Finally, I would like to thank my family. They were always by my side when I decided to quit my old job to start an academic career from scratch. My special thanks goes to Luana (my wife) and Pedro (my son). Luana is the strongest and most supportive person I have ever met. I would never have been able to make my career transition and succeed in my graduate studies without her love and help. Pedro is a lovely boy who could understand my absence and who made my life much happier during this process. I love you all very much.

RESUMO

Este trabalho compreende três artigos que exploram diferentes facetas da relação entre a percepção dos seguidores sobre seus (potenciais) líderes e a emergência e eficácia da liderança. Comum a esses artigos, é a ideia de que as percepções dos liderados sobre seus líderes (e as consequências dessas percepções) dependem dos valores articulados— ou sinalizados—pelos líderes e aqueles mantidos pelos seguidores. O primeiro artigo propõe que a autenticidade pessoal é um preditor-chave da emergência de líderes em grupos e que os efeitos da autenticidade são fortalecidos quando o indivíduo é percebido como alguém que incorpora e exemplifica os valores do grupo. O segundo artigo explora como o comprometimento dos líderes públicos com valores públicos afeta a emergência de formas carismáticas de liderança, o que, por sua vez, se traduz em níveis mais baixos de rotatividade de funcionários. Os efeitos moderadores de fatores contextuais, como o nível de violência e pobreza no entorno da organização, também são considerados. O terceiro artigo investiga como funcionários públicos respondem (em termos de satisfação no trabalho) à percepção de que suas organizações ignoram sistematicamente o desempenho insatisfatório de colegas de trabalho. Também analisamos se a motivação para o serviço público dos funcionários pode mudar a forma como eles reagem a essas percepções.

Palavras-chave: Autenticidade; Emergência de Liderança; Valores Públicos; Liderança

Carismática; Rotatividade de Pessoal; Medidas Disciplinares; Motivação para o Serviço

Público; Satisfação no Trabalho.

ABSTRACT

This work comprises three research articles that explore different facets of the relationship between followers’ perceptions of their (potential) leaders and leadership emergence and effectiveness. Common to these articles, is the idea that followers’ perceptions of leaders (and the consequences of these perceptions) depend both on the values articulated—or signaled—by leaders and those held by followers. The first article proposes that personal authenticity is a key predictor of leader emergence in groups, and that authenticity’s effects are enhanced when the target individual is perceived to embody the group’s identity-defining values. The second article explores how leaders’ commitment to public values—as perceived by followers—affects the emergence of charismatic forms of leadership in public organizations, which, in turn, translates into lower levels of employee voluntary turnover. The moderating effects of contextual factors, such as the level of violence and poverty around the organization, are also considered. The third article investigates public employees’ responses (in terms of job satisfaction) to the perception that their organizations systematically ignore poor- performing coworkers. We also investigate if employees’ public service motivation can change how they react to these perceptions.

Keywords: Authenticity; Leadership Emergence; Public Values; Charismatic Leadership;

Employee Turnover; Workplace Discipline; Public Service Motivation; Job Satisfaction.

SUMÁRIO

ARTIGO 1: AUTHENTICITY AND LEADER EMERGENCE: EXPLORING THE SOCIAL POWER OF BEING YOURSELF Introduction ...... 1 Authenticity as a Psychological Construct ...... 4 Authenticity in AL Research ...... 7 Hypotheses Development ...... 8 Overview of Studies ...... 17 Study 1 ...... 18 Study 2 ...... 25 General Discussion ...... 35 References ...... 44

ARTIGO 2: COMMITMENT TO PUBLIC VALUES AND CHARISMATIC LEADERSHIP IN STREET-LEVEL BUREAUCRACIES Introduction ...... 52 Public Values in Public Administration Research ...... 54 Charismatic Leadership in Public Organizations ...... 56 Hypotheses Development ...... 59 Methods ...... 66 Discussion ...... 80 Conclusion ...... 85 References ...... 86

ARTIGO 3: DISSATISFIED WHEN MY ORGANIZATION IGNORES POOR PERFORMERS: THE MODERATING ROLE OF PUBLIC SERVICE MOTIVATION Introduction ...... 94 Managing Poor Performance and its Social Effects ...... 96 Methods ...... 105 Discussion ...... 113 References ...... 118

GENERAL CONCLUSION ...... 124 1

ARTIGO 1

AUTHENTICITY AND LEADER EMERGENCE: EXPLORING THE SOCIAL

POWER OF BEING YOURSELF

“[L]eaders who are trusted make themselves known, make their positions clear.” —Warren Bennis

Introduction

Authenticity has become another buzzword “rolling off the tongues of politicians, celebrities, web gurus, college admissions advisers, reality television stars” (Rosenbloom,

2011). Besides its popularity in non-academic environments, the concept of personal authenticity has also received increased attention in recent psychological (Ryan & Ryan,

2019; Jongman-Sereno & Leary, 2018) and organizational research (Lehman, O'Connor,

Kovacs, & Newman, 2019; Cha et al., 2019). Authenticity was shown to be related to many variables indicative of psychological health, well-being, and relationship quality

(Brunell et al., 2010; Lenton, Slabu, & Sedikides, 2016; Ryan & Ryan 2019; Kernis &

Goldman, 2006). Specifically in the field of leadership studies, the concept of authenticity underpinned the development of one of today’s most popular theories of (ethical) leadership: Authentic Leadership (AL; George, 2003; Luthans & Avolio, 2003;

Walumbwa, Avolio, Gardner, Wernsing, & Peterson, 2008). Since the advent of AL, most leadership studies focusing on leader authenticity have been conducted under its theoretical umbrella. We argue, however, that studying the role of leader authenticity to leadership processes only through the lens of AL theory may be problematic for some reasons. 2

Existing measures of AL questionably carry a strong ethical component (Algera

& Lips-Wiersma, 2012; Alvesson & Sveningsson, 2013), making it overlap substantially with other positive forms of leadership, especially Ethical Leadership (Anderson & Sun,

2017; Brown & Treviño, 2006). Such overlap limits AL studies’ ability to make unique contributions to leadership theory (Treviño & Brown, 2007). Empirically, the ethical component of AL’s instruments confounds the net effects of leader authenticity. That is, because the “authentic leader” is always ethical and virtuous in other respects, we cannot know which proportion of AL’s effects are driven only by the leader’s authenticity.

Besides that, except for early AL studies that drew on Kernis’ (2003) work about the psychology of authenticity (e.g., Gardner, Avolio, Luthans, May, & Walumbwa, 2005),

AL has clearly evolved as a stand-alone literature, quite disconnected from the psychological research on authenticity. Consequently, more recent advancements related both to the conceptualization and measurement of authenticity have not been incorporated into AL research. Altogether, these issues suggest that, although we have seen a profusion of studies about AL in the last decades, our comprehension of the effects of authenticity— in its most precise meaning—on leadership processes and outcomes remains limited.

Another point we believe deserves more attention is the role of authenticity to informal power dynamics and emergent leadership processes within groups. AL research has concentrated on the effectiveness of individuals occupying formal management positions in organizations. Recent studies, however, acknowledges that “[a]uthenticity has the potential to generate impressive benefits for individuals in terms of their personal and social power” (Cha et al., 2019, p. 657; see also Gan, Heller, & Chen, 2018). Besides that, we believe it is time to explore the limits and potentials of authenticity; that is, the boundary conditions that can influence authenticity’s effects on leadership outcomes.

Although some individual characteristics were already proposed to enhance (or limit) the 3

benefits one may accrue from behaving authentically, especially in terms of social power

(e.g., Cha et al., 2019), these moderating effects remain to be empirically tested.

To help address these gaps, we draw on different theories of leadership emergence

(e.g., information processing, social identity, and self-schema activation; Acton, Foti,

Lord, & Gladfelter, 2019) to propose that authenticity can be a critical factor to explain the emergence of influencial group leaders. We also hypothesize that authenticity’s effect on leadership emergence will be moderated by the extent to which the group member is perceived by peers to embody the group’s identity-defining values and characteristics

(i.e., the extent to which that group member is perceived to be group-prototypical; Hogg,

2001; Van Knippenberg and Hogg, 2003; Van Knippenberg, 2011). This moderation effect means that being “true to the self” can be of less value—or even harmful—to one’s leadership potential when that self is not perceived to be congruent with an “idealized self” endorsed by the group (i.e., the group-prototype).

To test these hypotheses, we conducted two studies. Study 1 comprises an experiment in which we manipulated the authenticity and group-prototypicality levels of a stimulus group member and then measured participants’ endorsement for that member’s emergence as a group leader. In Study 2, we surveyed Air Force Cadets using a round- robin design (collection wave 1; Nestler, Grimm, & Schönbrodt, 2015) and social network analysis (collection wave 2; Hoppe & Reinelt, 2010). This second study sought to improve the external validity of Study 1 and to test the hypothesized effects of authenticity on group members’ actual influence over peers, proxied by their network centrality (Balkundi & Kilduff, 2006).

We believe this study contributes to the leadership and authenticity literatures in some ways. Differently from previous AL studies, we focus on more recent 4

conceptualizations of authenticity (which do not include positive ethical elements that might confound results) and explore the role of authenticity to emergent leadership processes and influence dynamics within groups. Moreover, we help shed light on the notion that authenticity’s effects on leadership outcomes should not be uniformly positive

(see Cha et al., 2019). We sought to balance leader- and follower-centric perspectives

(Uhl-Bien, Riggio, Lowe, & Carsten, 2014) to advance our understanding of how, why, and when individuals emerge as influential group leaders.

Authenticity as a Psychological Construct

Interest in personal authenticity is not new. The idea of “being true to the self” was explored by philosophers from different times and perspectives (Harter, 2002; Kernis

& Goldman, 2006; Martens, 2007). Within psychology, this topic gained attention in the mid-19th century by humanistic psychologists as Carl Rogers and Abraham Maslow

(Jongman-Sereno & Leary, 2018; Cha et al., 2019). Later, Self-Determination Theory studies (Sheldon, Ryan, Rawsthorne, & Ilardi, 1997) provided important insights about the relationship between authenticity, basic needs satisfaction (especially the need for autonomy), intrinsic motivation, and well-being (Ryan & Ryan, 2019). In the last 20 years, renewed interest in authenticity was motivated by the growth of positive

(organizational) psychology movements (Harter, 2002; Cameron, & Dutton, 2003). Since then, basic and applied psychological research, mainly from an individual-differences perspective, has been trying to better conceptualize and measure authenticity (Kernis &

Goldman, 2006; Knoll, Meyer, Kroemer, & Schröder-Abé, 2015; Wood, Linley, Maltby,

Baliousis, & Joseph, 2008), exploring its potential causes, consequences, and its position within extant models of personality (e.g., Maltby, Wood, Day, Pinto, 2012). More 5

recently studies started looking at authenticity as an important source of social power

(Cha et al., 2019; Gan et al., 2018): “When an actor is perceived as authentic by an observer, this increases the actor’s influence—defined as the ability to produce desired thoughts, feelings, or behavior in others” (Cha et al., 2019, p. 654).

Defining and Measuring Authenticity

Authenticity has been conceptualized in many ways, reflecting the plurality of literatures that studied this theme (Jongman-Sereno & Leary, 2018). However, despite the existing debates about the nature and possibility of authenticity (e.g., Ford & Harding,

2011), a common thread to most recent psychological conceptualizations of authenticity is that it entails the alignment of one’s outward behavior with one’s sense of self (Cha et al., 2019). That is, in the first place, authenticity requires the existence of a clearer sense of self, which explains why self-awareness is often referred to as necessary condition for authenticity (Kernis & Goldman, 2006; Knoll et al., 2015). In other words, only by having a clear sense of “who I am”—i.e., by being more deeply aware of one’s core values, preferences, beliefs, and identities—one can more resolutely manifest one’s “true self”1.

Authenticity implies autonomy, congruence, and genuineness (Ryan & Ryan,

2019). It happens because an authentic person’s actions and communications are more self-authored—i.e., they are willingly enacted and self-endorsed—and they reflect one’s core abiding values and beliefs. In this sense, when someone acts falsely or defensively, suppressing his/her “true self” merely to attain approval and rewards or to avoid punishment, we might say this individual is being inauthentic (Kernis & Goldman, 2006).

1 We do not imply that an immutable, transcendental “true self” either exists or can be known by an individual. Instead, by “true self” we mean the more conscious beliefs, values, preferences, and identities that are central to an individual’s self-concept (see Knoll et al., 2015; Jongman-Sereno & Leary, 2018). 6

Authors agree that authenticity requires courage (e.g., Ryan & Ryan, 2019), since

“behaving authentically runs the risk of prompting others’ scorn or ridicule, costs which can be very powerful inhibitors” (Kernis & Goldman, 2006, p. 340).

Some studies have concentrated on the development of instruments to measure authenticity. These studies generally treated authenticity as an individual trait (rather than a state). The first popular model was developed by Kernis and Goldman (2006). It included four factors: self-awareness (knowledge about one’s self), unbiased processing

(accurate processing of self-relevant information), behavior (behaving in accordance with one’s values, preferences, and beliefs), and relational orientation (valuing and achieving transparency in close relationships). As it will be further explained, a modified version of this model was used to develop the AL theory. Subsequently, Wood et al. (2008) developed an alternative measure comprised of three dimensions: self-alienation (i.e., self-awareness reverse-scored), authentic living (behaving in accordance with one’s values and beliefs), and acceptance of external influence (meaning that authentic individuals are less susceptible to external influence).

As both models were shown to have theoretical and psychometric limitations,

Knoll et al. (2015) proposed a revised model, which comprises only two factors: authentic self-awareness and authentic self-expression. According to the authors, these two factors are common to most previous definitions and measures of authenticity. In a series of studies with multiple samples, Knoll and colleagues provided evidence that their simpler model meets the standards of content, discriminant, and criterion validity, as well as of internal and temporal consistency. Building on this newer model, and on the commonalties of previous conceptualizations of authenticity, we advance a working definition of an authentic person: Someone who is more deeply aware of his/her core 7

values, beliefs, preferences, and identities (authentic self-awareness), and who consistently and resolutely expresses them in his/her actions, communications, and decisions (authentic self-expression).

Authenticity in AL Research

Authenticity gained increased attention in leadership research with the advent of the AL theory (Avolio, Gardner, Walumbwa, Luthans, & May, 2004; George, 2003;

Luthans & Avolio, 2003). Drawing from positive organizational psychology (Luthans &

Avolio, 2003), AL theory was advanced as a response to the growing unethical and socially irresponsible leadership practices in the corporate world (Gardner et al., 2011).

Building on the multidimensional conceptualization of authenticity proposed by Kernis

(2003), four underlying dimensions of AL were proposed: self-awareness, balanced processing, relational transparency, and internalized moral perspective (Walumbwa,

Avolio, Gardner, Wernsing, & Peterson, 2008).

Important to note, the “internalized moral perspective” factor of AL was not part of Kernis’ (2003) original conceptualization of authenticity (it substituted the “behavior” factor). For that reason, “[t]here has been some disagreement within the AL literature about the inclusion of ethics as a core component” (Gardner et al., 2011, p. 1129). That disagreement was motivated mainly by the notion that authenticity is not intrinsically ethical (Algera & Lips-Wiersma, 2012; Brown & Treviño, 2006). In this sense, one could be an “authentic jerk” (Zander, 2013; see also Cha et al., 2019; Jongman-Sereno & Leary,

2018). In empirical terms, the problem of combining ethical elements with elements of authenticity in a single measure (AL) is that it loses its capacity to reflect only a leader’s authenticity. It would not be a problem if AL instruments were used only to measure an 8

ethical form of leadership that builds on authenticity. The problem emerges when we use

AL instruments to draw conclusions about the effects of leader authenticity.

Of course, we do not mean that AL is not an interesting and powerful form of leadership. On the contrary, we strongly believe that when leaders behave in accordance with the prescriptions of the AL model, it should be very beneficial for organizations and society—and empirical evidence provides support to that (Garner et al., 2011). What we put into question is the capacity of AL instruments to serve as a valid measure of leader authenticity. Because of that, we develop our hypotheses and conduct our empirical tests drawing on more recent psychological conceptualizations of authenticity (e.g., Knoll et al., 2015).

Hypotheses Development

Leadership emergence is defined as a social process in which an individual without formal authority becomes a group leader (Durham, Knight, & Locke, 1997). It entails the attainment of social power and the exercise of greater influence over the group

(Guinote & Chen, 2016; Hollander, 1964, 1985; Pescosolido, 2002). Socio-cognitive theories of leadership assert that some individuals are more likely to emerge and establish themselves as group leaders because they display specific characteristics and behaviors that make them be recognized by peers as more effective, reliable leaders (Lord, De

Vader, & Alliger, 1986; Lord & Maher, 1991; Van Knippenberg, 2011). In other words, they are more likely to emerge as group leaders because their leadership is endorsed and supported by peers. It resonates with the notion that there is no leadership without followership—"a willingness to defer to another in some way” (Uhl-Bien et al., 2014, p.83). For that reason, in order to explore the role of authenticity to leadership emergence, 9

we begin by investigating why and when authenticity should lead to greater leadership endorsement of group members by peers.

Authenticity and Leadership Endorsement

We propose that perceived authenticity will typically lead to greater leadership endorsement for two main reasons, involving primarily observers’ cognitive (rather than affective) responses. First, we argue that authentic individuals have characteristics that people usually associate to the “leader category” (see Gan et al., 2018). These characteristics are known as people’s implicit leadership theories (ILTs; e.g., Lord, Foti,

& De Vader, 1984, Epitropaki & Martin, 2004). Individuals use their ILTs as a cognitive benchmark to categorize others as leaders (Tavares, Sobral, Goldszmidt, & Araújo, 2018).

Second, authentic individuals are likely to display greater behavioral integrity (BI;

Simons, 2002)—i.e., greater alignment between their words and deeds—which was shown to be an important predictor of trust in the leader (Palanski, Cullen, Gentry, &

Nichols, 2015; Palanski & Yammarino, 2011) and willingness to follow a leader

(Moorman, Darnold, and Priesemuth, 2013).

Arguably, people will be more willing to endorse and be vulnerable to a leader they trust in both in terms of leadership competence and consistency/predictability (Dirks,

2006). Next, we explore each of these explanations (ILTs congruence and perceived BI) more in detail.

ILTs Congruence. There is initial evidence that authentic individuals are perceived to be more powerful, more effective leaders, and better negotiators (Gan et al., 2018).

Gan et al., (2018) explains this result by arguing that an authentic individual’s behaviors fit observers’ lay theories and expectations of the powerful. This explanation is very 10

aligned with our view that authentic individuals display behaviors that are congruent with observers’ ILTs. Indeed, the categories powerful and leader have many commonalities, for leadership is “a special form of power” (Burns, 1978, p. 12).

Research shows that authentic individuals have greater self-esteem (Lenton et al.,

2016; Kernis, 2003; Wood et al. 2008), attachment security (Gillath, Sesko, Shaver, &

Chun, 2010), social self-efficacy (Satici, Kayis, & Akin, 2013), are more promotion- oriented (Kim, Che, Davis, Hicks, Schlegel, & 2019), and have higher levels of energy

(Lenton et al., 2016). These characteristics, coupled with the fact that they have a clearer sense of self and are more willing to stand by their core values and beliefs (Kernis &

Goldman, 2006), makes them likely to display (or signal) greater self-confidence, autonomy (Ryan & Ryan, 2019), decisiveness, courage, strength (Kernis & Goldman,

2006), honesty (Cooper, Sherman Rauthmann, Serfass, & Brown, 2018; Maltby, Wood,

Day, Pinto, 2012), negotiation ability (Gan et al., 2018), and extroversion (Wood et al.

2008; Fleeson & Wilt, 2010).

All these characteristics match many of people’s ILTs (see Epitropaki & Martin,

2004; Lord et al., 1984; Gerstner and Day, 1994; Offermann, Kennedy Jr, & Wirtz, 1994;

Den Hartog et al., 1999)—especially the more agentic components of their ILTs—thus signaling to observers leadership ability and power (Gan et al., 2018). These characteristics are also frequently associated to greater social status (e.g., Anderson, John,

Keltner, & Kring, 2001) and leadership emergence (Ensari, Riggio, Christian, & Carslaw,

2011; Lord et al., 1986; Riggio, Riggio, Salinas, & Cole, 2003). ILTs theorists (e.g.,

Lord, Brown, & Harvey, 2001) assert that observing few leadership cues in others’ behaviors would be enough for the activation of the whole “leadership schema” in people’s minds. We believe a perception of the “authentic behaviors” informed above can 11

activate such schemas. Besides that, experienced authenticity was also shown to increase an individual’s self-view as a leader (Gan et al., 2018), which encourages him/her to more actively engage in behaviors typical of leaders (Emery, Daniloski, & Hamby, 2011) and further increases their chances of being perceived as “leader-like” (Lord, Brown, &

Freiberg, 1999).

Perceived BI. According to Moorman et al. (2013, p. 427), the utility of leader BI in leadership research “stems from the importance of perceptions made by followers as they decide if and to what degree they will follow a leader” (see also Palanski &

Yammarino, 2007). The theoretical link between authenticity and perceived BI

(understood mainly as word-deed consistency; Simons, 2002) has been explored in previous works (e.g., Leroy, Palanski, & Simons, 2012). Indeed, some authors consider

BI and authenticity to be almost equivalent constructs (see Palanski & Yammarino, 2007).

A fundamental reason for why authentic individuals should display greater BI is their increased self-awareness and transparent communication of self-relevant values (Leroy et al. 2012) and their courage to consistently behave according to these values (e.g.,

Kernis & Goldman, 2006, Ryan & Ryan, 2018). Perceived leader BI affects many follower attitudinal outcomes, especially trust in the leader (Palanski et al., 2014; Simons

2002; Simons, Leroy, Collewaert, & Masschelein, 2015) and willingness to follow

(Moorman et al., 2013). According to Simons et al. (2015), perceived BI should be related to follower trust in leaders because it “support[s] the willingness of the follower to be vulnerable in that a leader who consistently adheres to espoused values and keeps promises becomes more predictable” (p. 833).

By combining both explanations (ILTs congruence and perceived BI), we conclude that perceived authenticity will increase peers’ trust in a group member’s ability 12

to be an effective, reliable group leader, which involves primarily the more cognitive aspects of trust (see Dirks & Skarlicki, 2004; McAllister, 1995). Therefore, we conclude that perceived authenticity should be translated into greater leadership endorsement.

Hypothesis 1: Group members perceived to be more authentic are more likely to be endorsed as group leaders by peers than group members perceived to be less authentic.

So far, we have concentrated on the cognitive effects of perceived authenticity.

However, we believe that one important moderating factor can enhance (or hinder) authenticity’s effects on leadership endorsement by triggering more affective responses in observers: group-prototypicality.

The Moderating Effect of Group-prototypicality

The concept of group-prototypicality is central to the social identity theory of leadership (Hogg, 2001). In social groups, a group prototype represents the shared, idealized notion of “who we are;” that is, it captures what is group-defining and group- normative (the group’s salient values, beliefs, characteristics, and identity; Van

Knippenberg, 2011). According to Van Knippenberg and Hogg (2003, p. 247), “[b]eing a prototypical group member means being ‘one of us,’ embodying the group’s identity, and representing what group members have in common and what differentiates the group from other groups.” Research shows that group-prototypicality elicits affective responses in in-group observers: group-prototypical members are more liked and admired by peers

(Van Knippenberg & Hogg, 2003). They are also seen as more charismatic (Platow, Van 13

Knippenberg, Haslam, Van Knippenberg, & Spears, 2006), have greater status, and peers trust more in these members’ intention to pursue the group’s best interests (Van

Knippenberg, 2011). For these reasons, group-prototypical members are more likely to be endorsed as group leaders than non-prototypical members (e.g., Platow & Van

Knippenberg, 2001).

We believe that group-prototypicality is the key to unleash the more affective effects of perceived authenticity, for two main reasons. First, authentic group members naturally have greater chances of having their values, beliefs, preferences, and identities recognized by peers, since they express their “true selves” more openly and consistently

(Knoll et al., 2015). Moreover, authenticity not only makes one’s values and beliefs more visible to others; it also signals one’s strong adherence to one’s values and beliefs (Kernis

& Goldman, 2006). Then, we expect that when a group member is highly group- prototypical, authenticity will signal his/her unwavering commitment to the group’s normative values; that is, he/she will likely be recognized as a distinguished, courageous representative of “us”—an “in-group champion” (Haslam, Reicher, & Platow, 2010); someone who not only stands by his/her personal values and interests, but by the group’s values and interests. In this sense, “being true to oneself” and “being one of us” combine to result in “being true to what we stand for” (see Steffens, Mols, Haslam, Okimoto,

2016). This way, we expect that “group-prototypical authenticity” will more strongly elicit in observers the affective responses associated to high group-prototypicality, such as personal identification, admiration, deference, social attraction, liking, attribution of charisma, respect, willingness to follow etc. (Hogg and Knippenberg, 2003; Platow et al.

2006; Van Knippenberg, 2011). 14

On the other hand, when a group-member is not group-prototypical, authenticity will arguably make their non-prototypicality more salient to peers. Indeed, studies have conceptually explored the idea that when one’s authenticity is not oriented by prosocial or shared values, it can backfire (Cha et al, 2019; Jogman-Sereno & Leary, 2018; Zander,

2013).

A conclusion similar to ours regarding the interactive effect between authenticity and group-prototypicality is offered by Tomlinson, Lewicki, and Ash (2014, p. 725), who found that the effects of perceived BI is moderated by perceived value congruence (which relates to perceived group-prototypicality): “When high values congruence accompanies high BI, this strengthens an observer’s conclusion that the actor has good character

(compared with high BI alone).”

Second, although we believe that authentic individuals will continue to be perceived as having greater leadership ability and BI even when they are not group- prototypical (i.e., arguably, they will still display “leader-like” behaviors and word-deed consistency), to our view, the valence of that leadership ability and BI will change in observers’ minds. If a group member is perceived to have greater leadership ability and word-deed consistency but is not seen as “one of us,” he/she may be recognized as a potential threat to the group’s best interests. Such group member will likely be perceived as a “competent deviant” of the group’s norms, who can more drastically—and negatively—affect the group. Then, in this case (low group-prototypicality), the perceived leadership ability and BI of authentic members should not be fully translated into leadership endorsement by peers. These arguments resonate with social cognition theories (Cuddy, Glick, & Beninger, 2011) that pose that once someone is perceived to have “low warmth” (i.e., not to be friendly and trustworthy)—which is more likely to be 15

the case when someone is not group-prototypical—his/her subsequent competence- related behaviors (e.g., leadership ability) can be interpreted as his/her capacity to inflict harm to others; that is, as a threat.

For these reasons, we expect that perceived authenticity’s effect on leadership endorsement will be enhanced to the extent to which the target group member is perceived to be group-prototypical.

Hypothesis 2: Perceived group-prototypicality moderates the effect of perceived authenticity on leadership endorsement, such that the effect of authenticity is stronger when group-prototypicality is higher.

Authenticity’s Effects on Actual Influence over the Group

So far, we have explored the link between perceived authenticity and leadership endorsement. However, what more clearly indicates leadership emergence within groups is a member’s actual level of influence over peers (Pescosolido, 2002, p. 585): “Emergent leaders can be defined as group members who exercise influence over the group” (see also Hollander, 1961, 1964, 1985). An individual’s ability to influence others is known as social power (Lammers, Stroker, & Stapel, 2009; Guinote & Chen, 2016). One primary reason why authenticity should lead to increased influence over the group concerns the increased leadership endorsement of authentic individuals (which was explored in H1 and

H2).

Mediation by leadership endorsement. An individual’s influence potential within a group will be greatly enhanced (or undermined) by the extent to which that individual 16

is recognized and endorsed as a leader by peers (Lord & Maher, 1991; Epitropaki &

Martin, 2004; Van Knippenberg, 2011) and, thus, seen as worthy of influence (Kenney,

Schwartz-Kenney, & Blascovich, 1996). That is, leadership endorsement indicates how much people will be willing to pay attention to, support, and follow an individual’s ideas and proposals and their openness to that individual’s influence attempts. Furthermore, when a group member perceive that peers endorse his/her leadership in the group, they will feel more encouraged to engage in influence attempts in the first place (see Howell

& Shamir, 2005, p. 106): “[F]ollowers can empower their leaders no less than leaders can empower their followers.” Therefore, because perceived authenticity leads to greater leadership endorsement (H1), which, in turn, should increase one’s likelihood of exerting influence over the group, we predict that:

Hypothesis 3: Group members perceived to be more authentic exert greater influence over the group than group members perceived to be less authentic.

Hypothesis 4: Perceived authenticity will have an indirect effect on influence over the group through the mediation of leadership endorsement.

Because perceived authenticity’s effect on leadership endorsement should be strengthened by perceived group-prototypicality for the reasons set forth in the development of H2, we end up with a moderated-mediation model (Figure 1; Hayes,

2013). Therefore, we expect that:

17

Hypothesis 4: The indirect effect of perceived authenticity on actual influence over the group (through the mediation of leadership endorsement) will be stronger when group-prototypicality is higher (compared to lower).

Figure 1. Full Theoretical Model

Overview of Studies

We designed two studies to test our hypotheses. In Study 1, we conducted a survey experiment to test H1 and H2 (effects on leadership endorsement) by manipulating the authenticity and group-prototypicality levels of a hypothetical group member. In other to create a more realistic sense of group membership for participants, we assigned them to hypothetical groups whose members had the same of their informed political orientations.

In Study 2, we sought to enhance the external validity of Study 1 and further test H3 to

H5 (effects on actual social influence) using a sample of Air Force Cadets. This study comprised two data collection waves. In wave 1, we used a round-robin design (Kenny,

1994; Nestler, Grimm, & Schönbrodt, 2015) to measure perceived authenticity, group- prototypicality, and leadership endorsement. In wave 2, we collected social network data

(Hoppe & Reinelt, 2010; Balkundi, & Kilduff, 2006) to measure group members’ actual level of influence over the group. 18

Study 1 – Experimental Effects of Authenticity on Leadership Endorsement

Sample

We recruited 539 Brazilian volunteers from diverse professional backgrounds via social media to participate in this study. Of them, 95 did not finish the online experiment and were excluded2, producing a final sample of 444 participants. Regarding the final sample, participants were 65% male, with an average age of 35.35 year (SD = 9.60), 90% had college degree, and 98% were employed at the time of the experiment. Of participants who informed being employed, 51% worked in the public sector (considering both the military and civil service). Participants were provided a term of informed consent explaining the study’s purpose, that their participation was anonymous, and that they could withdraw their participation at any time.

Procedures

Experimental scenario. In order to set up a more engaging hypothetical scenario for participants, we took advantage of the current moment of political polarization in

Brazil3 to create groups that participants would more likely identify themselves with.

Initially, among other demographic questions, we asked participants to indicate their general political orientation in a continuum ranging from political left to political right

2 ANOVA and Chi-squared tests revealed that participants who finished the experiment did not differ from those who did not finish in terms of age (F = 2.81, p = .09), political orientation (left vs. right; χ2 = 1.78, p = .19), education level (χ2 = .10, p = .94), and job experience (χ2 = 1.24, p = .26). We only found a small, but significant, difference in terms of gender (χ2 = 6.61, p < .05): female participants were more likely to abandon the experiment than male participants (19% versus 11%). 3We conducted this survey experiment shortly after the impeachment process of Brazil’s ex- President Dilma Rousseff in 2016, a moment of strong political polarization in the country (see https://www.huffpostbrasil.com/entry/brazil-corruption- scandal_n_56fbf5dae4b083f5c6063e80). 19

(adapted from Mirisola, Sibley, Boca, and Duckitt, 2007). Then, we told participants to imagine they were part of a group of volunteers who conducted independent auditing on government expenditures (an organization called Transparency Watch). After that, we provided participants with information about Transparency Watch’s members’ shared values and beliefs: depending on the participant’s informed political orientation, the

Transparency Watch’s members were said to share either left-wing (more liberal) or right- wing (more conservative) values and beliefs. With this procedure, we sought to increase participants’ feelings of value congruence and group membership in relation to the hypothetical group. After that, we randomly assigned participants to the experimental conditions. An example of an initial message presented to a right-wing participant was:

“Imagine you are a volunteer in a group called Transparency Watch. The group's mission is to conduct independent audit of government expenditures and to share relevant information with the public. The Transparency Watch team consists of 13 members (including you) and it is fairly homogeneous in terms of its members' values and beliefs. A key feature of this group is that its members are mostly supporters of right-wing parties. They tend to be more conservative and believe that government should not interfere too much in the economy.”

Manipulations. We conducted a 2 (authenticity: high vs low) x 2 (group- prototypicality: high vs low) experiment. To do that, after presenting the initial message shown above, we drew participants’ attention to a specific hypothetical group member:

Antonio Batista. To manipulate Antonio’s group-prototypicality, we used vignettes that informed his general political values and beliefs. In the high-group-prototypicality condition, Antonio had the same of the other Transparency Watch’s members’ (and probably of the participant’s) political values and beliefs. In the low-group- prototypicality condition, Antonio had values and beliefs mostly divergent from his peers.

The following vignette was presented to participants who informed being more aligned with the political right (low group-prototypicality in brackets): 20

“One of your colleagues at Transparency Watch is Antonio Batista. Like [Differently from] the other Transparency Watch’s members, Antonio is politically aligned with right-wing [left-wing] parties. Clearly, he shows a high [low] level of identification with the group. We might say that he is [not] a good representative of the values and beliefs shared by the group.”

To manipulate authenticity, we used vignettes that informed if Antonio displayed behaviors commonly associated either to authentic or inauthentic individuals. To build these vignettes, we used the characteristics and behaviors described in a more recent instrument of authenticity (Knoll et al., 2015). In the high-authenticity condition, we informed that Antonio displayed greater clarity of his values, stood by what he believed in even in face of social pressure, and was not easily influenced by the opinion of others.

In the low-authenticity condition, we provided the opposite information.

“Antonio Batista [not] always stand by what he believes in. Once he said, ‘When I have to choose, I prefer to be myself than to be popular [to be popular than to be myself].’ Since Antonio seems [not] to have strong conviction in his values and beliefs, he is not much [easily] influenced by the opinions of others. Antonio seems not to be [seems to be] afraid to expose his thoughts, even when [especially when] people disagree with him.”

Measures

Leadership endorsement. After the manipulations, we informed participants about upcoming elections for the new president of Transparency Watch, and that Antonio would be one of the candidates. Then, we asked participants to inform how likely it would be that they would vote for Antonio for president of the group on a 7-point Likert scale ranging from “extremely unlikely” to “extremely likely.” Intention to vote was used as a proxy to leadership endorsement in previous studies (e.g., Platow & Knippenberg, 2001). 21

Demographic variables. We collected information about participants’ age, gender, general political orientation (left-wing or right-wing), educational level, and employment status.

Pre-test

Before running this online experiment in Brazil, we conducted a pre-test with a sample of 210 MTurk workers living in the US. They were 51% female, had an average age of 36.30 years (SD = 13.11), and 98% had prior job experience. We followed the same procedures as in the main study conducted in Brazil.

Results

Pre-test results. We conducted our analysis with OLS regression. As expected, manipulated authenticity was positively related to leadership endorsement (b = 2.73, SE

= 0.22, p < .001) and there was an authenticity by group-prototypicality interaction (b =

1.62, SE = 0.42, p < .001). Manipulated group-prototypicality was also significantly related to leadership endorsement (b = 0.70, SE = 0.19, p < .01).

Main study results. Table 1 shows that participants’ characteristics were balanced across the experimental conditions.

22

Table 1. Participants’ Characteristics per Experimental Condition Experimental Condition Authenticiy: Low Low High High Group-Prototypicality: Low High Low High N: 100 112 110 112

Variables Mean Age 35.99 34.63 35.60 35.20 Gender 0.39 0.37 0.35 0.30 Education 3.23 3.05 3.12 3.04 Political Orientation 0.81 0.78 0.75 0.78 Notes. Gender: 0 = males; 1 = female. Education: 1 = Less than high-school; 2 = high- school; 3 = college degree; 4 = master’s degree or PhD. Political orientation: 0 = left- wing; 1 = right-wing.

Confirming H1, which predicted that perceived authenticity would lead to greater leadership endorsement, we found that manipulated authenticity was significantly related to participants’ intention to vote for the target group member (b = 1.90, SE = 0.15, p <

.001). Group-prototypicality also had a positive effect on leadership endorsement (b =

1.16, SE = 0.15, p < .001). Supporting H2, we found an authenticity by group- prototypicality interaction (b = 1.28, SE = 0.30, p < .001), meaning that authenticity’s effect on leadership endorsement is enhanced when group-prototypicality is high

(compared to low). Simple main slopes analyses revealed that the effect of authenticity on leadership endorsement was positive both when group-prototypicality was high (b =

2.54, SE = 0.21, p < .001) and low (b = 1.26, SE = 0.21, p < .001). Figure 2 shows the average level of leadership endorsement by experimental conditions, which reveals the moderation effect.

23

Figure 2. Interaction between Perceived Authenticity and Group-prototypicality

Although we expected that the effect of authenticity on leadership endorsement could be null or even negative when group-prototypicality was low (Cha et al., 2019), our results did not confirm that. That is, authenticity continued to be positively associated to leadership endorsement even when it was “anti-prototypical.” It might have happened because our experimental scenario is hypothetical and participants’ exposure to the experimental conditions was brief. Maybe in real-world settings, group members’ continued exposure to a colleague’s anti-prototypical authenticity could lead to different results.

Additional analysis. In order to explore the idea that more affective responses may explain the interaction effect between authenticity and group-prototypicality, we asked participants to rate two additional items (after they informed their intention to vote): one item measuring the extent to which they perceived the target group member as leader-like

(“Antonio is very typical of individuals who emerge as leaders”)—“leader typicality”— and another indicating their “willingness to follow” (“I would like to be led by someone 24

like Antonio”). Leader typicality relates to ILTs congruence, a more cognitive response

(Epitropaki & Martin, 2004), whereas, according to Burke, Sims, Lazzara, and Salas

(2007), willingness to follow is an affective response. These items were rated in a 7-points

Likert scale ranging from “completely disagree” to “completely agree.” Supporting our arguments, only when “willingness to follow” was the dependent variable (black lines;

Figure 3) the interaction between authenticity and prototypicality emerged (b = 1.39, SE

= 0.25, p < .001). When “leader typicality” was the dependent variable (grey lines; Figure

3), there was no interaction (b = -0.09, SE = 0.29, p > .05). These results provide initial support to the notion that more affective responses underlie the interactive effect between authenticity and group-prototypicality.

Figure 3. Cognitive vs Affective Effects of Perceived Authenticity by Group- prototypicality Level

Notes. DV = Dependent variable.

25

Given that this experiment has important limitations, mainly in terms of ecological/external validity, and because it cannot test authenticity’s effect on actual influence over the group (H3 to H5), we move on to Study 2.

Study 2 – Effects of Authenticity on Actual Social Influence

Sample

This study’s sample consists of cohorts of Air Force Cadets. We chose Cadets for some reasons: First, Cadets go through an intense socialization process based on specific values and codes of ethics. This is important because the effects of group-prototypicality depends on the existence of a more salient group identity (Van Knippenberg, 2011).

Second, Cadets interact daily with each other for years. Then, everyone in the group tend to know everyone else reasonably well, which is particularly important when using a round-robin survey design and social network analysis, designs in which everyone in the group is asked to rate all the other colleagues. Third, in social network studies, having high response rates (ideally 90% or more; Ferrin, Dirks & Shah, 2006) is critical to the reliability of the statistics. Because Cadets usually have high levels of discipline, they were expected to produce high response rates. Groups of Cadets were used in other leadership studies using social network analysis (e.g., Kwok, Hanig, Brown, & Shen,

2018)

Two entire classes of Air Cadets of the Brazilian Air Force Academy were invited to participate in this study: those in the 3rd and the 4th years of the course (Junior and

Senior Cadets). They totalized 164 Cadets (79 Juniors and 85 Seniors)4. Of the 3rd-year

4 The 1st- and 2nd-year Cadets (Freshmen and Sophomores) were not included because, as they spent less time together, they would be less likely to provide reliable information about all their colleagues. 26

Cadets, 97% participated in the first collection wave (round-robin survey), and 95% participated in the second one (social network analysis). Of the 4rd-year Cadets, 89% participated in the first collection wave, and 92% participated in the second one.

Participants were 96% male with and average age of 22.30 (SD = 1.18).

Procedures

We collected data in two waves, using online questionnaires. In the first wave, we collected data about Cadets’ perceptions of their peers’ group-prototypicality, authenticity, and the extent to which they would endorse their peers as group leaders using a round-robin survey design (Kenny, 1994). In the second wave we conducted a social network study to measure the actual level of influence each Cadet exerted over their groups. The social network analysis required participants to be identified. We provided them with a term of informed consent explaining that their participation was voluntary and that their names would not be disclosed, that the data would only be treated in the aggregate, and that they could withdraw the study and their consent at any moment.

Round-robin survey (wave 1). A round-robin survey design is a data collection method in which the members of a group evaluate one another in specific dimensions

(Kenny, 1994; Nestler, Grimm, & Schönbrodt, 2015). Usually, each individual member rate all the other members (and is rated by all the other members too). Among other possibilities, a round-robin survey provides an average estimate of how each group member is perceived by his/her peers. However, in larger groups, and when multiple variables are being measured, evaluating all colleagues can be a daunting task for participants and can lead to fatigue, especially if each variable is measured with multiple items (see Ferrin et al., 2006). For instance, in the present study, if each Cadet needed to 27

rate all their peers in the three dimensions (authenticity, group-prototypicality, and leadership endorsement) using a single item for each variable, they would have to provide about 240 ratings. Besides fatigue and low-quality responses, it would create other problems: common-source and consistency biases, which could intensify problems of spurious correlations between the study variables (Podsakoff, MacKenzie, Lee, &

Podsakoff, 2003).

To mitigate these problems, we used a split-sample design (Ostroff, Kinicki, &

Clark, 2002). In practical terms, we randomly divided each class of Cadets in thirds (of about 26 Cadets). Each of these thirds rated all their colleagues only in one dimension

(either authenticity, group-prototypicality, or leadership endorsement) using a single item. As such, the source of information for each dimension came from a different, random subgroup of about 26 Cadets (which served to address common-source bias).

Because these subgroups we randomly created, we expected them to be representative of their larger groups.

The use of single-item measures may be appropriate when situational constraints

(e.g. respondents’ fatigue) limit the use of multiple-items scales (Wanous, Reichers, &

Hudy, 1997). To make our single-item measures more reliable, we followed the procedures used by Ferrin et al. (2006). That is, before participants rated their classmates, they were provided with a text explaining the meaning of each variable. After reading the explanatory texts, they were given a roster with the names of all their classmates, whom they rated using a 4-points scale (see Measures section).

Social network analysis (wave 2). Two months after the first data collection, we collected social network data to build the influence network of each of the two classes of

Cadets. This network data allowed us to differentiate Cadets in terms of their actual 28

influence exerted over the group. Social network logic and metrics are being increasingly used in leadership studies (Bono and Anderson, 2005; Hoppe & Reinelt, 2010; Balkundi,

& Kilduff, 2006; Balkundi, Kilduff, & Harrison, 2011).

Measures

Perceived group-prototypicality. One third of the Cadets of each class (randomly selected) were provided with the following text. This text is based on the literature on group-prototypicality (e.g., Hogg, 2001; Van Knippenberg, 2011):

“In a social group, such as the group of Cadets, members usually share some characteristics, values, and beliefs. These characteristics define the group identity. However, not all members embody these group characteristics equally. In this question, we would like to know, according to your perception, how much each of your colleagues exemplify what the group of Cadets have in common and consider important. That is, indicate how much you believe the following colleagues are representative of the values, beliefs, and behaviors that characterize the group of Air Force Cadets.”

After that, participants were given a list with the names of all their classmates.

They rated each colleague using a 4-points Likert scale ranging from 0 (not at all representative) to 3 (extremely representative). After assessing the intra-class correlations

(ICCs) to verify agreement among raters, we averaged the ratings of each Cadet (around

26 ratings per target Cadet) to form his/her individual measure of group-prototypicality.

Perceived authenticity. We measured perceived authenticity following the same procedures mentioned above. This time, we explained to participants what it means to be an authentic individual based on Knoll et al.’s (2015) instrument:

“Authenticity is a trait that reflects how much an individual is aware of his/her personal values and acts transparently and consistently in accordance to these values. An authentic individual stands by what he/she believes in even in the face of social pressure and is not much influenced by the opinion of others. 29

Based on this description, please indicate how much you perceive that each of your classmates is an authentic individual.”

Participants rated all their colleagues using a 4-points Likert scale ranging from 0

(not at all authentic) to 3 (extremely authentic). As we did for group-prototypicality, we performed ICC tests and averaged the ratings.

Leadership endorsement. We asked Cadets how much they would endorse each of their colleagues as a group leader using the following text:

“In a group, there are members we see as more effective leaders and we like when they are in charge. On the other hand, there are some people we would prefer not be led by. Please, indicate to what extent you would provide support for each of the Cadets listed below to be your leader in a workgroup.”

Cadets rated their colleagues using a 4-points scale ranged from 0 (definitely would not [provide support]) to 3 (surely would [provide support]). This measure captures one’s intention to vote for a group member to be a group leader (as in Study 1). We performed ICC tests and averaged the ratings of each Cadet.

Actual social influence. Two months after the first data collection, we collected network data of social influence by asking participants “How often do you believe each of these colleagues listed below have an influence on your way of thinking, attitudes and behaviors?” Cadets rated all their colleagues using a 4-points Likert scale ranging from 0

(never) to 3 (very often). Similar influence network questions were used in previous leadership studies (see Bono & Anderson, 2005). We used the social network metric weighted in-degree centrality (also known as “node strength”) as our indicator of a group member’s level of influence over peers (Barrat, Barthelemy, Pastor-Satorras, &

Vespignani, 2004). It was calculated with the social network analysis software Gephi.

This measure results from the sum of the number of links directed to a Cadet (that is, the 30

number of times he/she was indicated by peers as someone who exerts at least some influence over them), weighted by the average intensity of each link, which derived from the 4-points scale (Barrat et al., 2004). Figure 4 shows the resulting influence network of the two groups of Cadets. The size of the circles (each circle representing one Cadet) is proportional to their in-degree centrality.

Figure 4. Cadets’ Social Influence Network

Control Variables. We controlled for: (i) Cadets’ rank in the class. Rank can be a confounder in this study because it relates to a Cadet’s status in the class. Moreover, high- ranked Cadets commonly take formal leadership roles within their classes. Then, controlling for rank can help mitigate halo effects that could create spurious correlations between the study variables. Because the best rank equals 1 and the worst equals 85, the correlations between rank and the studied variables are expected to be negative. (ii)

Cadets age and gender were included as controls because these variables may be part of participants’ ILTs (Lord et al., 1984), which may affect their leadership perceptions. (iii) 31

Finally, we controlled for the Cadet’s class (a dummy variable that differentiates 3rd- from

4th-year Cadets).

Results

We first assessed interrater agreement of the three independent variables by

calculating the ICCs (Woehr, Loignon, Schmidt, Loughry, & Ohland, 2015). We

followed the procedures outlined by Bliese (1998). Group-prototypicality had an ICC(1)

of .19, and a ICC(2) of .85; Authenticity had an ICC(1) of .08, and an ICC(2) of .67; and

leadership endorsement had an ICC(1) of .21, and a ICC(2) of .88. These ICCs indicates

that there is good agreement between participants’ perceptions of their peers (Woehr et

al., 2015), thus supporting the aggregation of the observations.

Our unit of analysis is the individual Cadet. Then, to combine the data of the two

groups of Cadets in a single database, we first standardized the study variables within

each group following the same procedure used by Kwok et al. (2018).

Table 2. Means, SD, and Correlations Meana SD 1 2 3 4 5 6 7 Study Variables: 1 Authenticity 2.07 0.32 - 2 Group-prototypicality 2.01 0.42 .67*** - Leadership 3 Endorsement 1.74 0.45 .70*** .80*** - b 4 Actual Influence 89.03 31.59 .74*** .85*** .82*** - Control Variables: c 5 Rank in the class 41.28 23.59 -.31*** -.64*** -.57*** -.58*** - 6 Age 22.30 1.18 -.08 -.15 -.14 -.14 .20** - 7 Genderd 0.04 0.18 -.17* -.11* -.16* -.11 -.11 .05 - 8 Classe 0.52 0.50 -.01 .00 .00 .00 .05 .46*** -.07 Notes. N = 164 (unit of analysis: individual Cadet). *p < .05 ** p < .01 ***p < .001. aMean and SD of the non- standardized variables; bweighed in-degree centrality; cthe better the rank, the closer to 1; d female = 1, male = 0. efourth-year Cadet = 1, third-year Cadet = 0.

32

Table 2 shows the bivariate correlations between all variables. It reveals that rank in the class is an important control as it correlates significantly with all study variables

(suggesting the existence of a halo effect). We conducted path analysis to test the hypotheses using structural equation modeling (SEM) in Stata 14 software (Figure 5).

Importantly, we tested all paths controlling for group-prototypicality in addition to the other control variables.

Figure 5. SEM Results

Notes. *p < .05 ** p < .01 ***p < .001. SE in parenthesis. Control variables are omitted. Standardized coefficients are shown.

Providing additional support to H1, we found that perceived authenticity was significantly related to leadership endorsement (β = .33, SE = 0.07, p < .001). Confirming

H2, which predicted that group-prototypicality would strengthen the effect of perceived authenticity on leadership endorsement, we found a significant authenticity by group- prototypicality interaction (β = .13, SE = 0.04, p < .01). Illustrating this interaction effect,

Figure 6 shows the marginal effects of perceived authenticity on leadership endorsement at different levels of group-prototypicality. Note that at very low levels of group- 33

prototypicality the effect of authenticity is not statistically significant (the 95% confidence interval includes zero).

Figure 6. Marginal effects of Perceived Authenticity on Leadership Endorsement

To test H3 (effect of authenticity on actual influence), H4 (mediation of authenticity’s effects on social influence by leadership endorsement) and H5 (stronger indirect effect of authenticity at higher levels of group-prototypicality), we implemented the procedures outlined by Hayes (2013) to conduct a conditional path analysis. More specifically, we used Hayes’ (2013) moderated-mediation model number 7.

Table 3 summarizes the results of the mediation analysis using bootstrapped standard errors (5000 replications). Supporting H3, we found a positive effect of authenticity on actual influence over the group (β = .34, SE = 0.05, p < .001). The average indirect effect of authenticity on actual influence through leadership endorsement was also significant (β = .08, SE = 0.03, p < .01), thus supporting to H3. We also found that at higher levels of group-prototypicality (1 SD above the mean) the indirect effect of 34

authenticity on influence is stronger than at lower levels of group-prototypicality (1 SD below the mean), thus confirming H5 (Δβ = .06, SE = 0.03, p < .05). We also found that the direct effect (c’) of authenticity on actual influence is positive and significant (β = .26,

SE = 0.05, p < .001), showing that authenticity’s effect on influence is only partially mediated by leadership endorsement.

Table 3. Moderated-Mediation Analysis’ Results Effect of Authenticity

on Social Influence

β SE p-value 95% CI Total effect (c) .34 0.05 < .001 [.24 .44] Average direct effect (c’) .26 0.05 < .001 [.15 .36] Average indirect effect (c - c’) .08 0.03 <.01 [.03 .14] Conditional Indirect Effects of

Authenticity on Social Influence Level of β SE p-value 95% CI Group-prototypicality: Low (1 SD below the mean) .05 0.02 < .05 [.01 .09] High (1 SD above the mean) .12 0.04 < .01 [.04 .19] Δ (β High – β Low) .06 0.03 <.05 [.01 .12] Notes. All effects were controlled by group-prototypicality (in addition to all other control variables).

It is worth noting that authenticity correlates strongly with group-prototypicality

(r = .67; Study 2). This result can be explained by the fact that group-prototypical members normally have greater latitude (i.e., they feel freer) to engage in a wider range of behaviors and to express themselves more authentically, without running the risk of losing peers’ trust and respect (Cha et al., 2019; Giessner, Van Knippenberg, & Sleebos,

2009). Moreover, because authenticity may signal leadership ability, and because leadership ability is considered an important individual characteristic in the military, 35

perceptions of authenticity may end up affecting perceptions of group-prototipicality in this setting.

General Discussion

Altogether, the results of the two studies provided support to the premise that authenticity can be an important source of social power (Gan et al., 2018; Cha et al.,

2019). We showed that both exogenously manipulated (Study 1) and measured authenticity (Study 2) have an impact on leadership endorsement and, ultimately, affect the level of influence a group member exert over his/her peers. Important to note, the effects of authenticity on both leadership endorsement and actual influence in Study 2 were significant even controlling for perceived group-prototypicality and rank in the class, two “strong” control variables clearly related to a Cadet’s status in the group. It gives us confidence that authenticity’s effects is not spurious. That is, perceived authenticity seems to explain unique variance of our dependent variables.

We also investigated when authenticity should be more (or less) beneficial to a group member’s leadership potential. Drawing from the notion that “individuals who choose to be authentic in opposition to contextual standards face the risk of negative social repercussions (i.e., reduced social power)” (Cha et al., 2019, p. 654), we predicted that the benefits a group member would accrue from being authentic—especially in terms of leadership endorsement—would be enhanced (or limited) to the extent that group member is perceived to be group-prototypical. A similar idea—of “antisocial authenticity”—was also explored theoretically by Jongman-Sereno and Leary (2018). We argued that perceived group-prototypicality would enhance authenticity’s effects on leadership endorsement mainly by triggering more affective responses is observers 36

(liking, identification, admiration, respect, willingness to follow, etc.). The additional analysis of Study 1 provided initial evidence for that.

Based on the arguments set forth above, we could expect authenticity to be negatively related to leadership endorsement when group-prototypicality was low (i.e., it could be interpreted as a threat to the group’s best interests). However, the results of both

Study 1 and 2 did not provide support for that. In Study 2, we found that authenticity’s effect can be null when group prototypicality was very low, but not negative. We believe it may happen because people attach great value to authenticity—i.e., authenticity seems to be a highly regarded virtue in our society, kind of a moral duty (see Jongman-Sereno and Leary, 2018). Then, people are usually penalized for being perceived as inauthentic, independently of their group-prototypicality.

Although we could not test that, we believe that the interaction effect between perceived authenticity and group-prototypicality can be enhanced in the presence of intergroup conflict and more salient social identities (“us versus them”), since in such contexts in-group members will arguably expect leaders to more authentically and decidedly stand by their group’s values and interests and be “in-group champions” (see

Haslam et al., 2010). For instance, this interaction effect may be more pronounced in moments of high political polarization (which is the case in many countries nowadays), helping explain the recent emergence of political leaders such as Donald Trump in the

USA, Boris Johnson in the UK, and Jair Bolsonaro in Brazil. These political leaders are apparently perceived by their supporters as individuals who more authentically embody and defend their groups’ political values and interests. They are perceived as the authentic prototypes of their groups—not only “one of us,” but “the one for us;” and that perception 37

seems to trigger strong affective responses in their supporters, like identification, admiration, liking, deference, willingness to follow, etc. (see development of H2).

Another noteworthy result relates to the partial mediation of leadership endorsement. Study 2 revealed a significant indirect effect of perceived authenticity on actual influence through leadership endorsement, especially when group-prototypicality was higher. However, our results evidenced that authenticity also has a direct effect on actual influence (c’), which was not transmitted through leadership endorsement. As this direct effect was relatively strong (see Table 3), we believe this result deserves a more detailed discussion.

Direct effect of authenticity on actual influence. The leadership endorsement mechanism explored by this work entails primarily observers’ responses to perceived authenticity; that is, it concerns people’s (potential followers’) perceptions of and attitudes towards a group member (the potential leader). However, it seems that

(in)authentic individuals’ own cognitions, motivations, and behaviors can have an effect on their level of influence over the group regardless of they being endorsed as leaders by peers. We believe it may happen because of authentic individuals’ greater extroversion and assertiveness (Wood et al., 2008; Fleeson & Wilt, 2010), and their increased self- view as leaders and sense of power (Cha et al., 2019; Gan et al., 2018).

As mentioned before, authentic individuals have a more secure sense of high self- esteem (Lenton et al., 2016; Kernis, 2003; Wood et al. 2008), greater social self-efficacy

(Satici, Kayis, Akin, 2013), and higher levels of energy (Lenton et al., 2016), characteristics that facilitate extroversion (Robins, Tracy, Trzesniewski, Potter, &

Gosling, 2001; Watson, Suls, & Haig, 2002). Extroversion entails being more talkative, articulate, and assertive (Goldberg, 1990). Indeed, studies confirm the positive 38

association between authenticity and extroversion (Wood et al., 2008; Fleeson & Wilt,

2010). We believe that the assertiveness of authentic individuals is further intensified by their greater clarity of their own values, beliefs, and preferences (Kernis & Goldman,

2006).

Authentic individuals’ increased extroversion and assertiveness will likely be translated into greater social influence. More extroverted and assertive group members are more likely to seek interactions and express their opinions in public (Goldberg, 1990), they also have higher speech rates (Feldstein & Sloan, 1984) and use greater vocal effort

(Scherer, 1978). All these factors correlate positively with persuasive power and interpersonal influence (Bonner, 2000). Moreover, authentic individuals’ greater self- view as a leader and sense of power (Gan et al., 2018) makes them more motivated to seek advantageous positions within the group (Kwok, Hanig, Brown, & Shen, 2018) and increase their effectiveness in persuading others (Emery et al., 2011; Guinote & Chen,

2016).

To our view, these mechanisms are not necessarily related to leadership endorsement by peers. That is, one may not be recognized and endorsed as a leader by peers and still exert greater influence over the group due to his/her increased extraversion and assertiveness, and more frequent influence attempts.

Authenticity vs. Self-monitoring. Extensive research has explored the positive association between self-monitoring and leadership emergence (see Day, Shleicher,

Unckless, & Hiller, 2002; Kudret, Erdogan, & Bauer, 2019). Because, at a first glance, authenticity and self-monitoring appear to be at the opposite extremes of a continuum, our hypotheses and findings may seem contradictory. However, empirical studies reveal that authenticity and self-monitoring are very weakly or not at all correlated (see Pillow, 39

Hale Jr, Crabtree, & Hinojosa, 2017), which means that being authentic—i.e. abiding to one’s core values and beliefs—does not imply being insensitive and unable to adapt to contextual demands. As asserted by Pillow et al. (2017) “the ability to restrain and express oneself, read and entertain others, and adapt one’s message to fit an audience does not necessitate abandoning one’s core values.” Moreover, as pointed out by Kernis and

Goldman (2006) authenticity “is not reflected in a compulsion to be one’s true-self, but rather in the free and natural expression of core feelings, motives, and inclinations” (p.

299; original emphasis). As such, we conclude that our theory and findings do not contradict existing evidence suggesting that self-monitoring relates to leadership emergence (see Gan et al., 2018).

Contributions

We identify five main contributions of this study. First, we contribute to the stream of literature initiated by AL studies, which explores the role of authenticity to leadership processes and outcomes. However, differently from AL studies, we sought to use a more recent psychological conceptualization of authenticity (Knoll et al., 2015) that do not include ethical elements that might confound results. We believe that by distancing leader authenticity from ethical leadership (Treviño & Brown, 2007), we were able to better test the net effects of authenticity and more uniquely contribute to the leadership theory.

Second, differently from AL works, which have concentrated on the effectiveness of individuals occupying formal managerial positions in organizations, this study contributes to the leadership emergence literature (see Acton et al., 2019), which focuses on how, why, and when individuals without formal authority become leaders. Although prior research has evidenced (or theorized about) the positive association between 40

authenticity and social power (Cha et al., 2019; Gan et al., 2018), authenticity’s effects have not been explored from an emergent leadership perspective. Three leadership emergence theories helped us investigate authenticity’s effects (and we believe these theories may also benefit from our conclusions): (i) ILTs (Epitropaki & Martin, 2004) underpinned the notion that authentic group members are more likely to be perceived as

“leader-like” by peers. We argued that authenticity combines many characteristics and behaviors people often associate to the leader category (Lord et al., 1984); (ii) leadership self-schema activation theories (see Acton et al., 2019) were used to explain that authentic individuals are more likely to develop a leader role identity (i.e., a self-view as a leader), which increases their chances of behaving like leaders and seek positions of influence; and (iii) the social identity theory of leadership (Hogg, 2001) helped us theorize about the limits and potentials of authenticity.

Third, this study helps shed light on the notion that authenticity’s effects on leadership outcomes should not be uniformly positive. Although we identified theoretical works (outside the mainstream AL literature) suggesting that some individual characteristics can limit (or even reverse) authenticity’s positive effects (e.g., Cha et al.,

2019), we did not find any leadership study testing that. Because AL theory has assumed that authentic leaders are necessarily ethical and virtuous in many ways, and that their values are congruent with their followers’ values (see Algera & Lips Wiersma, 2012), AL studies have overlooked the possibility of combining authenticity with other (less positive) leader characteristics (e.g., leaders’ low value congruence with followers or low group-prototypicality). Because we used a more “neutral” conceptualization of authenticity (Knoll et al., 2015), we were able to combine authenticity with different levels of group-prototypicality and test its contingent effects on leadership endorsement. 41

Our results confirmed that group-prototypicality functions as a boundary condition for authenticity’s effects.

Fourth, our research contributes the trait approach to leadership (Zaccaro, 2007), an approach that that investigates the personal characteristics most associated to leadership emergence and effectiveness. To our view, authenticity is a noteworthy leader trait because it combines many characteristics predictive of leadership perceptions and leadership emergence (see development of H1). Although the trait approach to leadership has been criticized for not accounting for other important determinants of leadership— especially the role of followers and the context/situation (see Day & Zaccaro, 2007)—we believe our research tries to overcome these limitations. Our theorization involves not only what leaders are or do, but how followers perceive and respond to these leaders, considering the social context they are embedded in (i.e., the group’s shared values, beliefs, and identity). That is, this study tries to balance leader- and follower-centric perspectives (Uhl-Bien et al., 2014) in order to produce a more comprehensive picture of the role of authenticity to emergent leadership processes.

Finally, we consider this paper may provide useful information for leadership development programs and for individuals who aspire to become leaders. As asserted by

Cha et al. (2019, p. 657), “[e]ach of us would be wise to consider the significant power implications of our own and others’ authenticity in the workplace”. Based on our results, our conclusion is that, when it comes to leadership emergence, personal authenticity pays off—and pays well—especially if your values align with your group’s shared values.

42

Limitations and Future Research

The combination of both an experimental and a field study improves our ability to draw conclusions about the role of authenticity to leadership emergence. However, we recognize that both studies have important limitations. Although Study 1 ensures the exogeneity of the independent variables, thus allowing us to investigate their causal effects, as most lab experiments, it lacks external and ecological validity (Reis & Judd,

2000). To a certain extent, we compensated for that limitation in the field study (Study

2). Study 2 uses a split-sample round-robin design (Ostroff et al., 2002; Kenny, 1994), social network analysis (Balkundi, & Kilduff, 2006), and two separate waves of data collection to mitigate the effects of consistency and common method biases (Podsakoff et al., 2003). However, we recognize that, even by adopting these procedures, endogeneity problems could not be fully ruled out. We believe that the replication of our tests with different methods and samples would help ascertain the robustness of our findings.

In the present study, we have assumed that the effect of authenticity on leadership emergence and actual influence over the group is linear. However, based on the idea that

“too much of a good thing” can be detrimental to leadership (Judge, Piccolo, & Kosalka,

2009), we believe future research can explore the non-linear relationship between authenticity and leadership outcomes. For instance, studies suggest that “authenticity may result in rigid behavior and undermine effectiveness” (Jongman-Sereno & Leary, 2018, p. 136). Another assumption of this work refers to the unidirectional effect of authenticity on social power. Although, this effect was supported by experimental research (Gan et al., 2018), studies also show that induced power can lead to greater experienced authenticity (Kraus, Chen, & Keltner, 2011). Gan et al. (2018) concludes, then, that the 43

effects of authenticity and power can be reciprocal. Although that reciprocity does not invalidate our hypotheses (since authenticity continues to be a predictor of social power and leadership emergence), it may have helped produce the high correlation between perceived authenticity and actual social influence in Study 2 (r = .74).

As mentioned before, another point to be explored by future research is the moderating role of intergroup conflict and social identity salience (Tajfel & Turner, 1979;

Van Knippenberg, 2011). To our view, these variables may strengthen the interactive effects of authenticity and group-prototypicality (a three-way interaction) and, perhaps, they may reveal the “dark side” of authenticity (i.e., its negative effects on leadership emergence). Finally, we believe that future research can further investigate the mechanisms we proposed to explain the effects of authenticity on leadership endorsement, namely: ILTs congruence, perceived BI, and the affective responses

(identification, admiration, liking, willingness to follow, etc.) that are expected to be triggered by authenticity when group-prototypicality is high.

44

References

Acton, B. P., Foti, R. J., Lord, R. G., & Gladfelter, J. A. (2019). Putting emergence back in leadership emergence: A dynamic, multilevel, process-oriented framework. The Leadership Quarterly, 30(1), 145-164. Alvesson, M., & Sveningsson, S. (2013). Essay: Authentic leadership critically reviewed. In D. Ladkin, & C. Spiller (Eds.). Authentic leadership: Clashes, convergences and coalescences (pp. 39-54). Cheltenham: Edward Elgar. Anderson, C., John, O. P., Keltner, D., & Kring, A. M. (2001). Who attains social status? Effects of personality and physical attractiveness in social groups. Journal of Personality and Social Psychology, 81(1), 116-132. Anderson, M. H., & Sun, P. Y. (2017). Reviewing leadership styles: Overlaps and the need for a new ‘fullrange’ theory. International Journal of Management Reviews, 19(1), 76-96. Bliese, P. D. (1998). Group size, ICC values, and group-level correlations: A simulation. Organizational Research Methods, 1(4), 355-373. Brunell, A. B., Kernis, M. H., Goldman, B. M., Heppner, W., Davis, P., Cascio, E. V., & Webster, G. D. (2010). Dispositional authenticity and romantic relationship functioning. Personality and Individual Differences, 48(8), 900-905. Cameron, K., & Dutton, J. (Eds.). (2003). Positive organizational scholarship: Foundations of a new discipline. San Francisco: Berrett-Koehler Publishers. Cha, S., Hewlin, P. F., Roberts, L. M., Buckman, B., Leroy, H., Steckler, E., ... & Cooper, D. (2019) Being Your True Self at Work: Integrating the Fragmented Research on Authenticity in Organizations. Academy of Management Annals, 13(1), 633-671. Cuddy, A. J., Glick, P., & Beninger, A. (2011). The dynamics of warmth and competence judgments, and their outcomes in organizations. Research in Organizational Behavior, 31, 73-98. Day, D. V., Shleicher, D. J., Unckless, A. L., & Hiller, N. J. (2002). Self-monitoring personality at work: A meta-analytic investigation of construct validity. Journal of Applied Psychology, 87(2), 390-401. Emery, C., Daniloski, K., & Hamby, A. (2011). The reciprocal effects of self-view as a leader and leadership emergence. Small Group Research, 42(2), 199-224. Algera, P. M., & Lips-Wiersma, M. (2012). Radical authentic leadership: Co-creating the conditions under which all members of the organization can be authentic. The Leadership Quarterly, 23(1), 118-131. Avolio, B. J., & Gardner, W. L. (2005). Authentic leadership development: Getting to the root of positive forms of leadership. The Leadership Quarterly, 16(3), 315-338. Avolio, B. J., Gardner, W. L., Walumbwa, F. O., Luthans, F., & May, D. R. (2004). Unlocking the mask: A look at the process by which authentic leaders impact follower attitudes and behaviors. The Leadership Quarterly, 15(6), 801-823. Balkundi, P., & Kilduff, M. (2006). The ties that lead: A social network approach to leadership. The Leadership Quarterly, 17(4), 419-439. 45

Balkundi, P., Kilduff, M., & Harrison, D. A. (2011). Centrality and charisma: Comparing how leader networks and attributions affect team performance. Journal of Applied Psychology, 96(6), 1209-1222. Barrat, A., Barthelemy, M., Pastor-Satorras, R., & Vespignani, A. (2004). The architecture of complex weighted networks. Proceedings of the National Academy of Sciences, 101(11), 3747-3752. Bonner, B. L. (2000). The effects of extroversion on influence in ambiguous group tasks. Small Group Research, 31(2), 225-244. Bono, J. E., & Anderson, M. H. (2005). The advice and influence networks of transformational leaders. Journal of Applied Psychology, 90(6), 1306-1314. Brown, M. E., & Treviño, L. K. (2006). Ethical leadership: A review and future directions. The Leadership Quarterly, 17(6), 595-616. Burke, C. S., Sims, D. E., Lazzara, E. H., & Salas, E. (2007). Trust in leadership: A multi- level review and integration. The Leadership Quarterly, 18(6), 606-632. Cooper, A. B., Sherman, R. A., Rauthmann, J. F., Serfass, D. G., & Brown, N. A. (2018). Feeling good and authentic: Experienced authenticity in daily life is predicted by positive feelings and situation characteristics, not trait-state consistency. Journal of Research in Personality, 77, 57-69. Day, D. V., & Zaccaro, S. J. (2007). Leadership: A critical historical analysis of the influence of leader traits. In L. L. Koppes (Ed.), Historical perspectives in industrial and organizational psychology (pp. 383–405). Mahwah, NJ: Erlbaum. Den Hartog, D. N., House, R. J., Hanges, P. J., Ruiz-Quintanilla, S. A., Dorfman, P. W., Abdalla, I. A., ... & Akande, B. E. (1999). Culture specific and cross-culturally generalizable implicit leadership theories: are attributes of charismatic/transformational leadership universally endorsed? The Leadership Quarterly, 10(2), 219-256. Dirks, K. T. (2006). Three fundamental questions regarding trust in leaders. In R. Bachmann and A. Zaheer (Eds.), Handbook of trust research (pp. 15-28). Cheltenham, UK: Edward Elgar. Dirks, K. T., & Skarlicki, D. P. (2004). Trust in leaders: Existing research and emerging issues. In R. M. Kramer & K. S. Cook (Eds.), Trust and distrust in organizations: Dilemmas and approaches (pp. 21-40). New York, NY: Russell Sage Foundation. Durham, C. C., Knight, D., & Locke, E. A. (1997). Effects of leader role, team-set goal difficulty, efficacy, and tactics on team effectiveness. Organizational Behavior and Human Decision Processes, 72(2), 203-231. Ensari, N., Riggio, R. E., Christian, J., & Carslaw, G. (2011). Who emerges as a leader? Meta-analyses of individual differences as predictors of leadership emergence. Personality and Individual Differences, 51(4), 532-536. Epitropaki, O., & Martin, R. (2004). Implicit leadership theories in applied settings: factor structure, generalizability, and stability over time. Journal of Applied Psychology, 89(2), 293-310. Feldstein, S., & Sloan, B. (1984). Actual and stereotyped speech tempos of extraverts and introverts. Journal of Personality, 52(2), 188-204. 46

Ferrin, D. L., Dirks, K. T., & Shah, P. P. (2006). Direct and indirect effects of third-party relationships on interpersonal trust. Journal of Applied Psychology, 91(4), 870- 883. Fleeson, W., & Wilt, J. (2010). The relevance of Big Five trait content in behavior to subjective authenticity: Do high levels of withinperson behavioral variability undermine or enable authenticity achievement?. Journal of Personality, 78(4), 1353-1382. Ford, J., & Harding, N. (2011). The impossibility of the ‘true self’of authentic leadership. Leadership, 7(4), 463-479. Gan, M., Heller, D., & Chen, S. (2018). The power in being yourself: Feeling authentic enhances the sense of power. Personality and Social Psychology Bulletin, 44(10), 1460-1472. Gardner, W. L., Avolio, B. J., Luthans, F., May, D. R., & Walumbwa, F. (2005). “Can you see the real me?” A self-based model of authentic leader and follower development. The Leadership Quarterly, 16(3), 343-372. Gerstner, C. R., & Day, D. V. (1994). Cross-cultural comparison of leadership prototypes. The Leadership Quarterly, 5(2), 121-134. Gardner, W. L., Cogliser, C. C., Davis, K. M., & Dickens, M. P. (2011). Authentic leadership: A review of the literature and research agenda. The Leadership Quarterly, 22(6), 1120-1145. George, B. (2003). Authentic leadership: Rediscovering the secrets to creating lasting value. San Francisco: Jossey-Bass. Giessner, S. R., van Knippenberg, D., & Sleebos, E. (2009). License to fail? How leader group prototypicality moderates the effects of leader performance on perceptions of leadership effectiveness. The Leadership Quarterly, 20(3), 434-451. Gillath, O., Sesko, A. K., Shaver, P. R., & Chun, D. S. (2010). Attachment, authenticity, and honesty: dispositional and experimentally induced security can reduce self- and other-deception. Journal of Personality and Social Psychology, 98(5), 841- 855. Goldberg, L. R. (1990). An alternative "description of personality": the big-five factor structure. Journal of Personality and Social Psychology, 59(6), 1216-1229. Guinote, A., & Chen, S. (2018). Power as active self: From acquisition to the expression and use of power. In K. Deaux & M. Snyder (Eds.), Oxford handbook of personality and social psychology (pp. 645-668). New York: Oxford University Press. Hayes, A. F. (2013). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach. New York: Guilford Publications. Harter, S. (2002). Authenticity. In C. R. Snyder & S. J. Lopez (Eds.), Handbook of positive psychology (pp. 382-394). New York: Oxford University Press. Haslam, S. A., Reicher, S. D., & Platow, M. J. (2010). The new psychology of leadership: Identity, influence and power. New York: Psychology Press. Hogg, M. A. (2001). A social identity theory of leadership. Personality and Social Psychology Review, 5(3), 184-200. 47

Hollander, E. P. (1961). Emergent leadership and social influence. In L. Petrullo, & B. Bass (Eds.), Leadership and interpersonal behavior (pp. 30-47). New York: Holt, Rinehart, and Winston. Hollander, E. P. (1964). Leaders, groups, and influence. New York: Oxford University Press. Hollander, E. P. (1985). Leadership and power. In G. Lindzey, & E. Aronson (Eds.), Handbook of social psychology (3rd ed.) (pp. 485 – 538). New York: Random House. Hoppe, B., & Reinelt, C. (2010). Social network analysis and the evaluation of leadership networks. The Leadership Quarterly, 21(4), 600-619. Howell, J. M., & Shamir, B. (2005). The role of followers in the charismatic leadership process: Relationships and their consequences. Academy of Management Review, 30(1), 96-112. Hosmer Jr, D. W., Lemeshow, S., & Sturdivant, R. X. (2013). Applied logistic regression (Vol. 398). New York: John Wiley & Sons. Jongman-Sereno, K. P., & Leary, M. R. (2018). The enigma of being yourself: A critical examination of the concept of authenticity. Review of General Psychology, 23(1), 133-142. Judge, T. A., Piccolo, R. F., & Kosalka, T. (2009). The bright and dark sides of leader traits: A review and theoretical extension of the leader trait paradigm. The Leadership Quarterly, 20(6), 855-875. Kennedy, F., & Kolb, D. G. (2016). The alchemy of authenticity. Organizational Dynamics, 4(45), 316-322. Kenney, R. A., Schwartz-Kenney, B. M., & Blascovich, J. (1996). Implicit leadership theories: Defining leaders described as worthy of influence. Personality and Social Psychology Bulletin, 22(11), 1128-1143. Kenny, D. A. (1994). Interpersonal perceptions: A social relations analysis. New York: Guilford Press. Kernis, M. H. (2003). Toward a conceptualization of optimal self-esteem. Psychological Inquiry, 14(1), 1-26. Kernis, M. H., & Goldman, B. M. (2006). A multicomponent conceptualization of authenticity: Theory and research. Advances in Experimental Social Psychology, 38, 283-357. Kim, J., Chen, K., Davis, W. E., Hicks, J. A., & Schlegel, R. J. (2019). Approaching the true self: Promotion focus predicts the experience of authenticity. Journal of Research in Personality, 78, 165-176. Knoll, M., Meyer, B., Kroemer, N. B., & Schröder-Abé, M. (2015). It takes two to be yourself. Journal of Individual Differences, 36, 38-53. Kraus, M. W., Chen, S., & Keltner, D. (2011). The power to be me: Power elevates self- concept consistency and authenticity. Journal of Experimental Social Psychology, 47(5), 974-980. 48

Kudret, S., Erdogan, B., & Bauer, T. N. (2019). Selfmonitoring personality trait at work: An integrative narrative review and future research directions. Journal of Organizational Behavior, 40(2), 193-208. Kwok, N., Hanig, S., Brown, D. J., & Shen, W. (2018). How leader role identity influences the process of leader emergence: A social network analysis. The Leadership Quarterly, 29(6), 648-662. Lammers, J., Stoker, J. I., & Stapel, D. A. (2009). Differentiating social and personal power: Opposite effects on stereotyping, but parallel effects on behavioral approach tendencies. Psychological Science, 20(12), 1543-1548. Lenton, A. P., Slabu, L., & Sedikides, C. (2016). State authenticity in everyday life. European Journal of Personality, 30(1), 64-82. Lehman, D., O'Connor, K., Kovacs, B., & Newman, G. (2019). Authenticity. Academy of Management Annals, 13(1), 1-42. Leroy, H., Palanski, M. E., & Simons, T. (2012). Authentic leadership and behavioral integrity as drivers of follower commitment and performance. Journal of Business Ethics, 107(3), 255-264. Lord, R. G., Brown, D. J., & Freiberg, S. J. (1999). Understanding the dynamics of leadership: The role of follower self-concepts in the leader/follower relationship. Organizational Behavior and Human Decision Processes, 78(3), 167-203. Lord, R. G., Brown, D. J., & Harvey, J. L. (2001). System constraints on leadership perceptions, behavior and influence: An example of connectionist level processes. In M. A. Hogg & R. S. Tindale (Eds.), Blackwell handbook of social psychology: Group processes (pp. 283-310). Oxford, England: Blackwell. Lord, R. G., De Vader, C. L., & Alliger, G. M. (1986). A meta-analysis of the relation between personality traits and leadership perceptions: An application of validity generalization procedures. Journal of Applied Psychology, 71(3), 402-410. Lord, R. G., Foti, R. J., & De Vader, C. L. (1984). A test of leadership categorization theory: Internal structure, information processing, and leadership perceptions. Organizational Behavior and Human Performance, 34(3), 343-378. Lord, R. G., & Maher, K. J. (1993). Leadership and information processing: Linking perceptions and performance. New York: Routledge. Luthans, F. (2002). The need for and meaning of positive organizational behavior. Journal of Organizational Behavior, 23, 695-706. Luthans, F., & Avolio, B. J. (2003). Authentic leadership: A positive developmental approach. In K. S. Cameron, J. E. Dutton, & R. E. Quinn (Eds.). Positive organizational scholarship. San Francisco: Barrett-Koehler. Maltby, J., Wood, A. M., Day, L., & Pinto, D. (2012). The position of authenticity within extant models of personality. Personality and Individual Differences, 52(3), 269- 273. Marsden, P. V. (1990). Network data and measurement. Annual Review of Sociology, 16, 435-463. 49

Martens, W. H. (2007). A multicomponential model of authenticity. Journal of Theoretical and Philosophical Psychology, 27(1), 73-88. McAllister, D. J. (1995). Affect-and cognition-based trust as foundations for interpersonal cooperation in organizations. Academy of Management Journal, 38(1), 24-59. Mirisola, A., Sibley, C. G., Boca, S., & Duckitt, J. (2007). On the ideological consistency between right-wing authoritarianism and social dominance orientation. Personality and Individual Differences, 43(7), 1851-1862. Moorman, R. H., Darnold, T. C., & Priesemuth, M. (2013). Perceived leader integrity: Supporting the construct validity and utility of a multi-dimensional measure in two samples. The Leadership Quarterly, 24(3), 427-444. Moorman, R. H., & Grover, S. (2009). Why does leader integrity matter to followers? An uncertainty management-based explanation. International Journal of Leadership Studies, 5(2), 102-114. Nestler, S., Grimm, K. J., & Schönbrodt, F. D. (2015). The social consequences and mechanisms of personality: How to analyse longitudinal data from individual, dyadic, roundrobin and network designs. European Journal of Personality, 29(2), 272-295. Offermann, L. R., Kennedy Jr, J. K., & Wirtz, P. W. (1994). Implicit leadership theories: Content, structure, and generalizability. The Leadership Quarterly, 5(1), 43-58. Ostroff, C., Kinicki, A. J., & Clark, M. A. (2002). Substantive and operational issues of response bias across levels of analysis: An example of climate-satisfaction relationships. Journal of Applied Psychology, 87(2), 355-368. Palanski, M. E., Cullen, K. L., Gentry, W. A., & Nichols, C. M. (2015). Virtuous leadership: Exploring the effects of leader courage and behavioral integrity on leader performance and image. Journal of Business Ethics, 132(2), 297-310. Palanski, M. E., & Yammarino, F. J. (2011). Impact of behavioral integrity on follower job performance: A three-study examination. The Leadership Quarterly, 22(4), 765-786. Pescosolido, A. T. (2002). Emergent leaders as managers of group emotion. The Leadership Quarterly, 13(5), 583-599. Pillow, D. R., Hale Jr, W. J., Crabtree, M. A., & Hinojosa, T. L. (2017). Exploring the relations between self-monitoring, authenticity, and well-being. Personality and Individual Differences, 116, 393-398. Platow, M. J., & Van Knippenberg, D. (2001). A social identity analysis of leadership endorsement: The effects of leader ingroup prototypicality and distributive intergroup fairness. Personality and Social Psychology Bulletin, 27(11), 1508- 1519. Platow, M. J., Van Knippenberg, D., Haslam, S. A., Van Knippenberg, B., & Spears, R. (2006). A special gift we bestow on you for being representative of us: Considering leader charisma from a selfcategorization perspective. British Journal of Social Psychology, 45(2), 303-320. 50

Podsakoff, P. M., MacKenzie, S. B., Lee, J. Y., & Podsakoff, N. P. (2003). Common method biases in behavioral research: A critical review of the literature and recommended remedies. Journal of Applied Psychology, 88(5), 879-903. Reis, H. T., & Judd, C. M. (Eds.). (2000). Handbook of research methods in social and personality psychology. Cambridge: Cambridge University Press. Riggio, R. E., Riggio, H. R., Salinas, C., & Cole, E. J. (2003). The role of social and emotional communication skills in leader emergence and effectiveness. Group Dynamics: Theory, Research, and Practice, 7(2), 83-103. Robins, R. W., Tracy, J. L., Trzesniewski, K., Potter, J., & Gosling, S. D. (2001). Personality correlates of self-esteem. Journal of Research in Personality, 35(4), 463-482. Rosenbloom, Stephanie. (2011, September 9). Authentic? Get real. Nytimes.com. Retrieved June 30 2019, https://www.nytimes.com/2011/09/11/fashion/for-only- the-authentic-cultural-studies.html Ryan, W. S., & Ryan, R. M. (2019). Toward a social psychology of authenticity: Exploring within-person variation in autonomy, congruence, and genuineness using self-determination theory. Review of General Psychology, 23(1), 99-112. Satici, S. A., Kayis, A. R., & Akin, A. (2013). Investigating the predictive role of social self-efficacy on Authenticity in Turkish University students. Europe’s Journal of Psychology, 9(3), 572-580. Scherer, K. R. (1978). Personality inference from voice quality: The loud voice of extroversion. European Journal of Social Psychology, 8(4), 467-487. Sheldon, K. M., Ryan, R. M., Rawsthorne, L. J., & Ilardi, B. (1997). Trait self and true self: Cross-role variation in the Big-Five personality traits and its relations with psychological authenticity and subjective well-being. Journal of Personality and Social Psychology, 73(6), 1380-1393. Simons, T. (2002). Behavioral integrity: The perceived alignment between managers' words and deeds as a research focus. Organization Science, 13(1), 18-35. Simons, T., Leroy, H., Collewaert, V., & Masschelein, S. (2015). How leader alignment of words and deeds affects followers: A meta-analysis of behavioral integrity research. Journal of Business Ethics, 132(4), 831-844. Schmader, T., & Sedikides, C. (2018). State authenticity as fit to environment: The implications of social identity for fit, authenticity, and self- segregation. Personality and Social Psychology Review, 22(3), 228-259. Steffens, N. K., Mols, F., Haslam, S. A., & Okimoto, T. G. (2016). True to what we stand for: Championing collective interests as a path to authentic leadership. The Leadership Quarterly, 27(5), 726-744. Tavares, G. M., Sobral, F., Goldszmidt, R., & Araújo, F. (2018). Opening the Implicit Leadership Theories’ Black Box: An Experimental Approach with Conjoint Analysis. Frontiers in Psychology, 9, 1-11. Tajfel, H., & Turner, J. C. (1979). An integrative theory of intergroup conflict. In: W. G. Austin & S. Worchel (Eds), The Social Psychology of Intergroup Relations (pp. 33–47). Monterey, CA: Brooks/Cole. 51

Treviño, L. K., & Brown, M. E. (2007). Ethical Leadership: A Developing Construct. In D. L. Nelson & C. L. Cooper (Eds.), Positive organizational behavior (pp. 101– 116). Thousand Oaks, CA: Sage. Tomlinson, E. C., Lewicki, R. J., & Ash, S. R. (2014). Disentangling the moral integrity construct: Values congruence as a moderator of the behavioral integrity– citizenship relationship. Group & Organization Management, 39(6), 720-743. Uhl-Bien, M., Riggio, R. E., Lowe, K. B., & Carsten, M. K. (2014). Followership theory: A review and research agenda. The Leadership Quarterly, 25(1), 83-104. Van Knippenberg, D. (2011). Embodying who we are: Leader group prototypicality and leadership effectiveness. The Leadership Quarterly, 22(6), 1078-1091. Van Knippenberg, D., & Hogg, M. A. (2003). A social identity model of leadership effectiveness in organizations. Research in Organizational Behavior, 25, 243- 295. Zaccaro, S. J. (2007). Trait-based perspectives of leadership. American Psychologist, 62(1), 6-16. Zander, L. (2013). Viewpoint: an authentic jerk. Authentic leadership can be bad leadership. In D. Ladkin & C. Spiller (Eds.), Authentic leadership: Clashes, convergences and coalescences (pp. 279–281). Cheltenham: Edward Elgar. 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. Wanous, J. P., Reichers, A. E., & Hudy, M. J. (1997). Overall job satisfaction: how good are single-item measures? Journal of Applied Psychology, 82(2), 247-252. Watson, D., Suls, J., & Haig, J. (2002). Global self-esteem in relation to structural models of personality and affectivity. Journal of Personality and Social Psychology, 83(1), 185-197. Weischer, A. E., Weibler, J., & Petersen, M. (2013). “To thine own self be true”: The effects of enactment and life storytelling on perceived leader authenticity. The Leadership Quarterly, 24(4), 477-495. Woehr, D. J., Loignon, A. C., Schmidt, P. B., Loughry, M. L., & Ohland, M. W. (2015). Justifying aggregation with consensus-based constructs: A review and examination of cutoff values for common aggregation indices. Organizational Research Methods, 18(4), 704-737. Wood, A. M., Linley, P. A., Maltby, J., Baliousis, M., & Joseph, S. (2008). The authentic personality: A theoretical and empirical conceptualization and the development of the Authenticity Scale. Journal of Counseling Psychology, 55(3), 385-399.

52

ARTIGO 2

COMMITMENT TO PUBLIC VALUES AND CHARISMATIC LEADERSHIP

IN STREET-LEVEL BUREAUCRACIES

“If you don’t believe in the messenger, you won’t believe the message” —Kouzes-Posner first law of leadership.

Introduction

Public values (PV)—broadly defined as “the normative principles on which governments and policies should be based” (Andersen et al., 2013, 294)—is receiving increased attention in public administration research, apparently as a response to the emphasis on “running government like a business” of the 1980s and 1990s (Bryson,

Crosby, & Bloomberg, 2014; Van der Wal, Nabatchi, & De Graaf, 2015). As such, studies have been emphasizing the need for leaders of public organizations to commit to, protect, and promote PV (e.g., Denhardt & Campbell, 2006; Getha-Taylor et al., 2011; Terry,

1990; Tummers & Knies, 2016). However, some aspects of the role of PV to public leadership remains to be investigated, particularly from a behavioral perspective.

In this study, we draw on charismatic leadership (Conger & Kanungo, 1998;

Howell & Shamir, 2005; Shamir, House, & Arthur, 1993) and socio-cognitive leadership theories (Hogg, 2001; Van Knippenberg, 2011) to propose that commitment to PV is not only a public leader’s obligation (from a normative perspective), but it is the basis for the existence of more impactful and motivational forms of leadership in public organizations, particularly in street-level bureaucracies (public schools, social work agencies, public hospitals, etc.). Central to our theorization is the notion that charismatic leadership is not simply the expression of a leader’s personal charisma—a “special gift” (see Platow et al., 53

2006)—but a two-way relational phenomenon that depends on both leaders and followers

(Jermier, 1993). In this sense, we argue that the emergence and dynamics of charismatic leadership are greatly influenced by the existing values, identities, and expectations of followers, as they set the basis for their judgements about and attitudes towards leaders

(Burns, 1978; Klein & House, 1995; Platow et al., 2006).

However, despite the relevance of charismatic leadership to the public sector, wherein transactional options are less available for leaders (Moynihan, Pandey, & Wright,

2014), we argue it should not be equally effective in all public service settings. Indeed, research has provided evidence that charismatic leadership can be especially beneficial in more stressful and challenging situations that demand strong personnel effort and self- sacrifice (Bass, 1985; House, 1977; Klein & House, 1995; Shamir, House, & Arthur,

1993). As such, we also investigate whether in more difficult and demanding public service environments characterized by high levels of poverty and violence, the effect of charismatic leadership on public organizational outcomes becomes more prominent.

We test our hypotheses with data collected from a sample of 87 street-level bureaucracies (specifically, public schools) that were selected using a probabilistic stratification method (Groves et al., 2011). We surveyed public employees working in those organizations using a split-sample design to mitigate response biases (Ostroff,

Kinicki, & Clark, 2002) and used objective secondary data provided by the Board of

Education as our control and outcome variables.

We believe this study advances the public leadership and PV literatures in some ways. First, by concentrating on the role of PV, we answer the call for more publicness in public leadership studies (Crosby & Bryson, 2018; Getha-Taylor et al., 2011; Vogel &

Masal, 2015). Second, differently from previous public leadership studies on PV, we shift 54

the focus to the internal environment of public organizations and shed light on the emergence of charismatic forms of public leadership, especially in street-level bureaucracies. Third, by considering the moderating effect of contextual factors, we can get a more nuanced understanding of when leaders’ commitment to PV—and the resulting charismatic leadership processes—are likely to influence public organizations’ outcomes.

Finally, we believe this study takes another step towards connecting the PV and the public service motivation (PSM; Perry & Wise, 1990) literatures, since we propose that a public leader’s commitment to PV should include behaviors related both to more classical PV and PSM-related values (see Andersen et al., 2013; Maesschalck, Van der Wal, &

Huberts, 2008).

Public Values in Public Administration Research

According to Schwartz and Bilsky (1987), five features are common to most definitions of values: values are “(a) concepts or beliefs (b) about desirable end states or behaviors (c) that transcend specific situations, (d) guide selection or evaluation of behavior or events, and (e) are ordered by relative importance” (551). In the context of public administration research and practice, some values are held in high regard, as they represent the idealized principles upon which the public service should be based

(Andersen et al., 2013). They are usually referred to as Public Values (PV).

However, to date, there is no complete consensus about the definition of PV or its content (Rutgers, 2015). Nonetheless, some PV are recurrent in the literature. Examples are integrity, accountability, impartiality, openness, legality, humanness, altruism, and professionalism (e.g. Jørgensen & Bozeman, 2007; Van Wart, 2013; Van der Wal et al.,

2006; Witesman & Walters, 2014). These values form the basis of what is often labelled 55

as the public (service) ethos (e.g., Jørgensen, 2006), and can be found in professional codes of ethics, such as that of the American Society for Public Administration (ASPA; see Svara, 2014).

Furthermore, the concept of PV overlaps with other well-known concepts, such as public service motivation (PSM; Perry & Wise, 1990). Indeed, the PSM scale explicitly includes “commitment to public values” as one of its dimensions (Kim et al., 2012).

Witesman and Walters (2014) say that their public service values framework

“encompasses and complements the dimensions of the PSM scale” (377), and Andersen and colleagues (2013) assert that the concepts of PV and PSM are profoundly interwoven, although they do not recommend their complete integration. Maesschalck, Van der Wal, and Huberts (2008) concludes that the PSM dimensions mirror many of the values found in the more general PV literature, although “a number of values that concentrate on the integrity of governance are less apparent in public service motivation, such as impartiality, incorruptibility, and transparency” (171). Maesschalck, Van der Wal, and

Huberts (2008) also imply that PSM-related values (e.g., compassion), when not combined with other normative public service principles (e.g., lawfulness) can backfire

(e.g., lead to lack of neutrality and preferential treatment). Thus, in the present paper, we consider both more general, normative PV (e.g., integrity and accountability), and PSM- related values (e.g., self-sacrifice and compassion). Other concepts closely related to PV are public service ethos (e.g., Rayner et al., 2010) and public service ethic (e.g., Brewer

& Selden, 1998).

Based on the notion that the internalization of PV by public servants can affect their judgement and behavior in ways that are beneficial to society, studies have proposed strategies to promote these values in public organizations (e.g., Paarlberg, Perry, & 56

Hondeghem, 2008). In this matter, leadership has been referred to as a critical factor to be considered (Getha-Taylor et al., 2011; Paarlberg, Perry, & Hondeghem, 2008). Indeed, evidence suggests that leadership can both directly affect public employees’ internalization of PV (e.g., Pandey et al., 2016; Park & Rainey, 2008; Wright, Moynihan,

& Pandey, 2012) and capitalize on employees’ existing values to increase their performance (e.g. Bellé, 2013a).

Some studies have focused more specifically on public leaders’ commitment to

PV (e.g., Denhardt & Campbell, 2006; Getha-Taylor et al., 2011; Newell, 2007; Terry,

1990). These studies have usually adopted a normative—rather than a descriptive and empirical—approach, and they are typically not grounded in mainstream leadership theories (see Crosby & Bryson, 2018). Moreover, their focus is often on abstract societal- level outcomes—which are difficult to measure —such as protection of democracy, the public interest (Getha-Taylor et al., 2011), and institutional integrity (Terry, 1990).

However, based on the notion that followers’ perceptions of leaders are critical determinants of (measurable) group-level leadership processes and behavioral outcomes in organizations (Lord & Maher, 1993; Van Knippenberg & Hogg, 2003), we argue that the intra-organizational behavioral effects of perceived leaders’ commitment to PV deserve more theoretical and empirical attention.

Charismatic Leadership in Public Organizations

Charismatic leadership5 is a relational phenomenon (Klein & House, 1995). On the leaders’ part, it is characterized by the articulation of visions that appeal to followers’

5 Two different forms of charismatic leadership have been explored by research: socialized and personalized (e.g., House and Howell 1992; Howell 1988; Watts et al. 2018). According to House and Howell’s (1992, 84), socialized charismatic leadership “(a) is based on egalitarian behavior, (b) serves collective interests 57

values and identities, the exemplification of idealized behaviors, and the promotion of a sense of collective identity. On the followers’ part, it is characterized by a strong identification and engagement with the leader and the mission he/she articulates, an emotional attachment to the leader, and motivational arousal (Antonakis et al., 2016;

Conger & Kanungo, 1998; Gardner & Avolio, 1998; Shamir, House, & Arthur, 1993).

Diverging from simpler conceptions, researchers have been highlighting the idea that charismatic leadership—or charisma—is not something that can be owned by an individual (Jermier, 1993). In other words, charisma does not result simply from the leader’s personality, but from the interplay between leaders’ characteristics and behaviors and followers’ expectations, perceptions and attitudes towards leaders, all of which embedded in a charisma-conducive environment (e.g., Klein & House, 1995). The crucial test of charisma, says Jermier (1993), is the followers’ voluntary engagement with and support to the leader. Importantly, research suggests that it can only happen if followers recognize in their leaders the idealized values shared by the group (Gardner & Avolio,

1998; Shamir, House, & Arthur, 1993; Platow et al., 2006).

The concept of charismatic leadership is inextricably related to the concept of transformational leadership (Burns, 1978), to the point that many researchers treat them as equivalent (see Yukl, 1999). Indeed, “charisma” (i.e., the combination of inspirational motivation and idealized influence) is a central component of transformational leadership

(Bass, 1985). Measures of both charismatic and transformational leadership include very similar leadership behaviors (Knippenberg & Sitkin, 2013). In the present paper,

and is not driven by the self-interest of the leader and (c) develops and empowers others.” On the other hand, personalized charismatic leadership “(a) is based on personal dominance and authoritarian behavior, (b) serves the self-interest of the leader and is self-aggrandizing and (c) is exploitive of others.” Our focus herein is on socialized, positive, forms of charismatic leadership. 58

however, we decided to use the term charismatic (rather than transformational) leadership, because our primary focus is not on the transformation of followers’ values, but the charismatic relationships that are triggered by followers’ perceptions of their leaders’ commitment to their existing PV.

Meta-analyses showed that charismatic leadership can affect followers’ work motivation, performance, extra effort, and job satisfaction (DeGroot, Kiker, & Cross,

2000; Dumdum, Lowe, & Avolio, 2013). In the context of public administration, it is also related to important public organizational outcomes (Orazi, Turrini, & Valotti, 2013; Van

Wart, 2013; Vogel & Masal, 2015). Effects include increased public employee PSM

(Caillier, 2014; Jensen & Bro, 2018; Krogsgaard, Thomsen, & Andersen, 2014; Park &

Rainey, 2008; Wright et al., 2012) and performance (Bellé, 2013a; Trottier, Van Wart, &

Wang, 2008).

Some organizational characteristics can make charismatic leadership even more relevant. For instance, when transactional options (e.g., use of financial rewards) are less available, it can be “left as one of the few tools at leaders’ disposal” (Moynihan, Pandey,

& Wright, 2014, 88). Accordingly, Shamir, House, & Arthur (1993) argue that charismatic leadership is particularly important when performance cannot be easily measured and linked to rewards. The existence of beneficiary contact can also strengthen the effects of charismatic leadership on employees’ work outcomes (Bellé, 2013a; Grant,

2012). It happens because the leader’s vision and inspirational appeals become less abstract to followers: “Beneficiary contact enables employees to see that their contributions to the vision [articulated by the leader] have meaningful consequences for other people” (Grant, 2012, 458). 59

The aforementioned conditions can be found in many public organizations. While the lack of transactional options, or the difficulty to link performance to rewards— especially financial rewards—characterize most types of public organizations, beneficiary contact is usually present in most street-level bureaucracies, since in these organizations public servants deal directly with the public. Therefore, understanding the factors that influence the emergence of charismatic leadership in public organizations— especially in street-level bureaucracies—should be highly relevant. In the next sections, we present our model and the proposed relationships between PV, charismatic leadership, and public organizational outcomes.

Hypotheses Development

Commitment to PV and Charismatic Leadership in Public Organizations

Early behavioral charismatic leadership studies (e.g., House, 1977) have long acknowledged the important role of leaders’ values and the ideological content of their messages to the emergence of charismatic leadership. This premise continued to guide subsequent studies until the present days (see Antonakis et al., 2016). As Conger and

Kanungo (1998, 63) assert, “a leader whose vision fails to incorporate important values and lacks relevance for the organizational context is unlikely to be perceived as charismatic.” Accordingly, Shamir, House, and Arthur (1993) points out that communicating a message that appeals to the existing values and identities of followers is a necessary condition for a leader’s message to have charismatic effects. As such, leaders need to respond to followers (to their values, identities, and expectations) no less than followers respond to them (Shamir, House, & Arthur, 1993). This rationale was further developed by social identity models of leadership (e.g., Van Knippenberg & 60

Hogg, 2003), and we believe it provides a solid explanation for why perceived leader commitment to PV is a critical predictor of charismatic leadership in the public organizations.

Due to attraction-selection-attrition and socialization processes, public employees are expected to internalize and share PV (Paarlberg & Lavigna, 2010; Perry & Wise,

1990; Wright & Christensen, 2010). We believe that the internalization of these public service values is even more likely to occur in street-level bureaucracies, wherein public employees literally serve citizens (i.e., there is beneficiary contact) and observe the prosocial impact of their work (see Bellé, 2013b). Based on the precepts of social identity theory (Ashforth & Mael, 1989; Tajfel & Turner, 1986), to the extent that PV are internalized and shared by public servants, these values are likely to become characteristic of public servants’ sense of collective identity—a sense of “who we are”. In other words, especially in service-oriented street-level bureaucracies, PV are expected to define the group prototype: a fuzzy set of idealized characteristics that define a group (Hogg, 2001;

Van Knippenberg, 2011). According to Van Knippenberg (2011, 1079), group prototypes

“include what the group values, believes, and considers important, and what are seen as appropriate and desirable behaviors and courses of action.” This way, group prototypes represent what is group-normative (Hogg, 2001; Van Knippenberg, 2011).

The importance of group-prototypes for leadership processes is that they serve as a cognitive benchmark for followers to make sense of their leaders (Hogg, 2001; Platow et al., 2006; Van Knippenberg, 2011). A central premise of the social identity theory of leadership is that, to be endorsed by followers—and be granted the credentials to emerge as a charismatic leader— a leader must be perceived to live up to the idealized image of

“who we are” (Platow et al., 2006; Platow et al., 2003; Van Knippenberg & Hogg, 2003). 61

This “we-ness” is necessary because leadership is not developed in a vacuum; rather, it usually happens within a social group with a distinctive social identity characterized by specific shared values, beliefs, and interests (Haslam, Reicher, & Platow, 2010). As such, perceived leader group-prototypicality is normally translated into status within the group, referent influence, social attraction to the leader, and trust in his/her group-orientedness

(Van Knippenberg & Hogg, 2003). This way, the more followers see the leader as

“consistently expressing in their pronouncements and their actions those norms and values that make our group distinctive from other groups […] the more we will pay heed, the more we will follow what such people say, and the more effort we will put into supporting their proposals” (Haslam & Reicher, 2016, 29). In the context of the public service, these “norms and values that make the group” arguably encompass PV.

Thus, diverging from some recent public leadership studies (e.g., Jensen,

Andersen, & Jacobsen, 2018; Krogsgaard, Thomsen, & Andersen, 2014), we propose that a public leader’s enacted values function as a predictor of charismatic leadership, rather than a moderator of its effects (see Conger & Kanungo, 1998; Platow et al., 2006; Shamir,

House, & Arthur, 1993). Prior experimental evidence confirms the premise that leader group-prototypicality predicts perceptions of charismatic leadership (Platow et al., 2006; see Van Knippenberg & Hogg, 2003). This means that leaders who are perceived to exemplify the normative, identity-defining values of a group are also perceived to inspire loyalty, have a sense of mission, and have a vision of the future (Platow et al., 2006).

Therefore, by combining the findings and insights from both charismatic leadership and social identity theory of leadership literatures, we propose that public servants’ perceptions of their leaders’ commitment to PV is a critical predictor of the emergence of charismatic leadership in public organizations. 62

Hypothesis 1: Perceived leader commitment to public values is positively

associated with perceptions of charismatic leadership in public organizations.

Charismatic Leadership and Employee Voluntary Turnover

Charismatic leadership engage and motivate followers by connecting their self- concepts to the organization’s goals (Howell & Shamir, 2005). It occurs because the leader articulates an inspiring vision that appeal to followers’ existing values and identities, thus creating a more meaningful, motivating, and satisfying work experience

(Shamir, House, & Arthur, 1993). Moreover, as charismatic leaders are catalysts in the sense-making process and serve as a source of organizationally relevant information to employees, they can increase employees’ person-organization value congruence

(Hoffman et al., 2011; Meglino, Ravlin, & Adkins, 1989). This is particularly relevant in public organizations, since research shows that public servants can only satisfy their prosocial motivations to the extent to which they perceive their values to be congruent with their organizations’ values (Liu, Tang, & Yang, 2015; Wright & Pandey, 2008). In fact, meta-analyses show that one of the most consistent effects of charismatic leadership is the increase of followers’ job satisfaction (e.g., DeGroot, Kiker, & Cross, 2000;

Dumdum, Lowe, & Avolio, 2013).

For similar reasons, charismatic leadership should also be translated into lower levels of employee turnover. By enhancing public servants’ person-organization value congruence, charismatic leadership is expected to reduce turnover, since individuals prefer to work in organizations that enables them to be involved in meaningful and personally satisfying tasks. Not surprisingly, studies show that low job satisfaction is associated with increased turnover intentions and actual voluntary turnover rates 63

(Griffith, 2004; Grissom, Viano, & Selin, 2016; Judge, 1993; Nyberg, 2010; Pitts, Marvel,

& Fernandez, 2011). Moreover, charismatic forms of leadership lead to higher-quality relationships between leaders and followers, which also translates into lower turnover behavior (Herman, Huang, & Lam, 2013).

Indeed, turnover research in the context of public administration indicates that leadership (or effective management) is one of the most important factors to predict employee turnover (Brown & Wynn, 2009; Grissom, 2011; Grissom, Viano, & Selin,

2016). As such, we hypothesize that, because charismatic leaders are more able to align their followers’ values and identities with their organization’s values and goals, and because they establish higher-quality relationships with followers, they will likely reduce voluntary turnover levels in public organizations.

Hypothesis 2: Charismatic leadership is associated with lower levels of

employee voluntary turnover in public organizations.

The Moderating Role of Contextual Stressors

The view that charismatic forms of leadership can be particularly effective in stressful and challenging situations is present in many seminal works (e.g., Bass, 1985;

Conger & Kanungo, 1988; House, 1977; Klein & House, 1995; see also Van Knippenberg

& Hogg, 2003). Klein and House (1995, 185-186) asserts that “In crises, individuals are uncertain and stressed and, thus, open to the influence of persuasive leaders who offer a hopeful, inspiring vision of the crisis resolved.” Relatedly, Shamir, House, and Arthur

(1993) say that charismatic leadership can be particularly important when “exceptional effort, behavior and sacrifices are required of both leaders and followers” (590). 64

Many contextual factors can create a more stressful and challenging work environment that demands increased effort from public employees. For instance, in the case of street-level bureaucracies (such as public schools or public hospitals), common contextual stressors known to increase employee voluntary turnover are the level of poverty and violence in the neighborhood (Kraft et al., 2012; Simon & Johnson, 2015;

Smith & Smith, 2006). In these precarious contexts, public servants leave mainly because they see their work environment as a threat to their safety (Smith & Smith, 2006) and perceive that their poor working conditions impede their chance to perform well (Simon

& Johnson, 2015).

Interestingly, research shows that working in these precarious contexts can be more meaningful to public servants. Many employees are attracted to serving in poorer contexts due to a “humanistic commitment” (Achinstein et al., 2010, 71; Cochran-Smith et al., 2012), or PSM. However, in these difficult contexts, wherein resources of all sorts are less available, the chances of not satisfying employees’ PSM increase, because they are often not given the necessary support to successfully accomplish their tasks. In a research conducted in a public educational context, Kraft and colleagues (2012) perfectly illustrate this idea. Based on interviews, they report that most teachers working in poorer districts did not express the wish to work with wealthier, whiter, or higher-performing students. Instead, they reaffirmed that helping students to succeed in those challenging contexts was even more satisfying to them. However, when these teachers perceived that their poor working conditions would prevent them from achieving success, they expressed great frustration and willingness to leave that school (Kraft et al., 2012).

Therefore, we expect that in more stressful work contexts characterized by higher levels of poverty and violence, wherein public servants are more likely to have their (often 65

high) prosocial expectations frustrated, charismatic leaders should become even more critical to prevent high levels of voluntary turnover, as they can create a more meaningful and fulfilling work environment that help employees satisfy their prosocial motivations and cope with contextual adversities. We, then, advance the following hypothesis:

Hypothesis 3: The effect of charismatic leadership on employee voluntary turnover will be moderated by the work context, such that in more stressful public service environments charismatic leadership will be more strongly related to lower employee turnover.

Based on the hypotheses set forth above, we provide the theoretical model of our research in Figure 1.

Figure 1. General Study Model

66

Methods

Research Setting and Sampling

To test our model, we established a partnership with the Board of Education of the city of Rio de Janeiro, in Brazil. The public educational system in Rio de Janeiro includes more than 1,000 public elementary schools, making it one of the largest public- school systems in the world6. We chose public schools not only because they are street- level bureaucracies crucially important to society, but also because they are the most common type of public organization (Meier & Bohte, 2003). Furthermore, public schools within the same educational system have similar organizational structures, despite operating in heterogeneous environments (e.g., wealthy and safe vs. poor and violent neighborhoods). As such, they represent the perfect setting to test our hypotheses.

To select the schools that would be included in our sample, a stratified probabilistic sampling was carried out to ensure that relevant subpopulations would not be underrepresented (Groves et al., 2011). Using this procedure, we selected a random sample of 154 public elementary schools from a population of 1009 schools. The number of schools sampled from each of the 11 existing districts was proportional to the districts’ actual number of schools (Table 1). Of the 154 schools selected, 148 provided us with the contacts of their principals and teachers, representing 3,770 potential participants. In total,

1,017 teachers completed the questionnaires sent via text message (27% response rate).

They were mostly female (82%), had an average age of 44.10 (SD = 4.31), and 12.56 (SD

= 4.38) years of experience as teachers in public schools. As our level of analysis is the school, we only kept in our sample schools with at least 4 respondents to have more

6 Information retrieved from the Rio de Janeiro Municipal Board of Education’s website: http://www.rio.rj.gov.br/web/sme/exibeconteudo?id=9095194. 67

reliable aggregate measures. Thus, our final sample included a total of 87 schools (a total of 874 respondents). The average number of participants per school was 10.05 (SD =

5.50).

Table 1 shows the characteristics of our final sample, compared to the population and the initially selected sample. It reveals that the most important difference is the number of teachers per school. This happened because we initially excluded schools with less than 10 teachers7 and, after the data collection, we excluded schools with less than 4 respondents (as explained above). Therefore, our final sample naturally included schools with slightly higher number of teachers (compared to the population and selected sample).

Nevertheless, all other school characteristics (number of students per class, school performance, and Human Development Index of the neighborhood) did not differ in meaningful ways. Also, the proportion of schools per district remained mostly the same as shown in Table 1. As such, we believe that the sampling procedure produced an unbiased and representative sample of the population.

Procedures

Data collection. Two weeks before data collection, we sent banners, letters and e- mails to all the randomly selected schools informing about our research. We then sent the questionnaires’ links to all potential participants via a texting app (Whatsapp)8 based on the telephone numbers provided by the schools. Participants were presented the term of

7 We did not include in the sample schools with less than 10 teachers in order to avoid having schools with low number of participants (our aim was to have at least four respondents per school). During the randomized sampling, when a school with less than 10 teachers was selected, we substituted it by another randomly selected school from the same district. 8 Whatsapp is the primary way people text in Brazil (information available at https://datareportal.com/reports/digital-2019-brazil). 68

informed consent which explained the general purpose of the study, that their participation was voluntary and confidential, and that the research had been approved by the Rio de Janeiro Board of Education and by the Institutional Review Board of the institution at which the leading author was located at the time of data collection. One month after sending the links, we sent new messages, thanking those who participated and inviting those who did not. We also personally called the school principals whose schools had lower participation rates, asking them to incentivize their teachers to participate.

Table 1. Characteristics and Representativeness of the Sample

Selected Final Population Average Characteristics Sample Sample (N = 1009) (N = 154) (N = 87)

Number of Teachers per School 25.55 27.74 29.54 Number of Students per Class 30.50 30.20 30.55 School Performancea 5.14 5.24 5.08 HDI of School Neighborhood 0.80 0.79 0.79

Proportion of Schools per District: District 1 6% 6% 9% District 2 10% 10% 7% District 3 9% 9% 6% District 4 11% 10% 12% District 5 8% 9% 9% District 6 7% 7% 7% District 7 12% 11% 11% District 8 12% 13% 16% District 9 10% 10% 10% District 10 12% 12% 11% District 11 3% 3% 2% Notes. aStandardized test score in, 2015 (range: 0 to 10).

Split-sample design. To limit common source biases and mitigate problems of spurious correlations (see Podsakoff et al., 2003), teachers were randomly assigned to one of two possible questionnaires (i.e., a split-sample design; e.g., Ostroff, Kinicki, & Clark,

2002). One of the questionnaires included questions about their principals’ commitment 69

to PV, while the other measured their perceptions of principals’ engagement in charismatic leadership. Although split-sample designs cannot address all sources of method biases and endogeneity (Favero & Bullock, 2014), it was shown to be effective in mitigating individual-level biases that inflate correlations between study variables

(Ostroff, Kinicki, & Clark, 2002).

Measures

Perceived leader commitment to PV. According to Van der Wal and Huberts

(2008, 266) “broad agreement exists that values cannot be seen or heard and can only be observed in how they manifest themselves through attitudes, preferences, decision making and action.” In other words, followers cannot directly assess their leaders’ values.

However, they can observe some of their leaders’ day-to-day decisions and actions, which, over time, signal to them the extent to which their leaders commit to specific values. We chose to use followers’ perceptions of their leaders’ commitment to PV, rather than collecting this information directly from their leaders (i.e., self-report), for two main reasons. First, self-report measures are more likely to be influenced by social desirability

(Fisher & Katz, 2000). Moreover, as each school has only one principal, they knew that they could be identified, which would also increase their chances of providing only socially desirable answers. Second, for the emergence of charismatic leadership, followers’ perceptions of their leaders are more relevant than the self-report of one’s values (Gardner & Avolio, 1998). In other words, when it comes to charismatic leadership processes, leaders’ values matter to the extent to which employees perceive these values are exhibited in the leaders’ words and actions. 70

Regarding the selection of the specific PV that were used to measure leaders’ commitment to PV, it is important to highlight that “there are no self-evident core [public] values, nor categorizations or hierarchies of values that can simply be applied universally in theory or practice.” As such “the validity and usefulness of any typology relies on the justification of the approach taken, that is, the very context it is stated in” (Rutgers, 2008,

109). As such, we selected six values that are very recurrent in the PV literature that we believe are relevant to our study’s context. We included both classical, normative public service values (integrity, accountability, and openness to participation) and PSM-related values (self-sacrifice, compassion, and service mentality), since we understand they complement each other (see Maesschalck, Van der Wal, & Huberts, 2008). Of course, we do not suggest that these six PV are either the most important ones or that they are universally applicable. We believe, however, that in balancing comprehensiveness and parsimony, they capture important aspects of a public leader’s commitment to PV. Table

2 presents the values and items used to measure leader commitment to PV.

As presented in Table 2, we created two scale items for each of the six values, totaling 12 items. Participants rated their school principals in a five-points scale, ranging from “strongly disagree” (1) to “strongly agree” (5). They evaluated their school principals’ commitment to PV considering their behavior during the last mandate that had just expired at the moment of the data collection9. After performing confirmatory factor analyses (CFA) and goodness-of-fit tests, we estimated the factor scores for the latent variable commitment to PV. The factor scores were then aggregated (averaged) at the school level.

9 In Rio de Janeiro, school principals are internally elected for a 3-years mandate. The mandate under study finished in December 2017. 71

Table 2. Leader Commitment to PV – Scale Items

Public Values Scale Items Frequently ignores legal and ethical standards and prefer to do things Administrative integrity* "his/her own way.” Is highly concerned about the ethicality and legality of all school Administrative integrity employees’ actions. Openness to participation Is open to the participation of employees and local community. Considers the opinion and interests of many people and groups before Openness to participation making decisions. Accountability Is accountable to citizens and superiors and manages transparently. Gets annoyed when he/she is asked to provide information about the Accountability* school management practices. Self-sacrifice Is willing to sacrifice him/herself in order to help the school. Is often absent when the work in the school requires some personal Self-sacrifice* sacrifice. Exemplifies the "service spirit" that public servants are expected to Service mentality live up to. Service mentality Reminds employees about the important social impact of their work. Compassion Demonstrates compassion for the people experiencing difficulties. Compassion* Is not sensitive to the needs of the community.

Notes. *Reverse-coded items.

Charismatic Leadership. Charismatic leadership was measured with the five items used by Wright and Pandey (2009) and Krogsgaard, Thomsen, and Andersen

(2014). This scale derives from House (1998) socialized charismatic leadership scale. The five items are: “Says things that make employees proud to be part of the organization,”

“Leads by setting a good example,” “Clearly articulates his/her vision of the future,” “Has a clear sense of where our organization should be in 5 years,” and “Challenges me to think about old problems in new ways.” Teachers rated their school principals in a five- points scale, ranging from “strongly disagree” (1) to “strongly agree” (5). We used the

CFA’s predicted factor scores as our measure of charismatic leadership. These factor scores were then aggregated at the school level. 72

Voluntary Turnover. The Rio de Janeiro Board of Education provided us with the information of teachers’ actual voluntary turnover for each of the schools being studied.

High levels of employee turnover tend to be problematic to most types of organizations, as it increases costs with recruitment and training, and lead to loss of relevant organizational knowledge and expertise (Brown & Wynn, 2009; Grissom, 2011).

Specifically in the case of public schools, research shows that higher levels of teacher turnover ultimately affect students’ learning outcomes (O’Toole & Meier, 2003; Simon

& Johnson, 2015). In our study, this measure represents the total number of teachers who voluntarily left their schools in, 2017. It does not include teachers who left due to dismissal, retirement, health problems, or death. This measure was obtained after the data collection period.

Contextual Stress. We used objective indicators of poverty and violence of school neighborhoods to create a measure of contextual stress. Both high levels of poverty and violence were shown to be important predictors of teachers’ decision to leave the school

(Simon & Johnson, 2015; Smith & Smith, 2006). To measure poverty and violence in the school area, we used students’ socio-economic status (SES)10 of 2015 (the latest available) and the average number of homicides around the school (within a radius of

500m) from 2014 to 2016, all of which provided by the Rio de Janeiro Board of

Education. The two indicators (SES and number of homicides) where combined with factor analysis. We predicted the factor scores to create a standardized measure of

“contextual stress”.

10 The students’ SES is an index calculated by the Brazilian Government that considers the students’ family income, possession of goods, and the educational level of parents, among other socio-economic indicators. 73

Control Variables. We controlled for (a) school past performance (school standardized test score in 2015); (b) school size, as proxied by the total number of teachers in the school, (c) the average number of students per class, and (d) the gender of school principal. By controlling by past performance, we aimed at mitigating the effect of followers’ attribution of “good/bad leadership” based on organizational performance

(Meindl, Ehrlich, & Dukerich, 1985). We controlled for school size and the number of students per class because these variables can be related both to the complexity of managing the school and to organizational climate variables, and thus they might also confound the results. School principal gender was included as a control because leader gender can also bias followers’ leadership perceptions (Eagly & Karau, 2002; Schein et al., 1996).

Results

Measurement. We tested different configurations of the commitment to PV measure with confirmatory factor analysis (CFA). As shown in Table 3, the CFA revealed that a six-factor model showed significant better fit than a single-factor model, which indicates discriminant validity between the six PV dimensions. Furthermore, a model in which all the six PV dimensions load in a higher-order factor (Figure 2) showed an acceptable goodness-of-fit, indicating that all the six PV dimensions reflect a common latent variable (“commitment to PV”). The Cronbach’s alpha of commitment to PV was

.93.

74

Table 3. Commitment to PV Measure: Goodness-of-fit Tests

Model χ2 χ2( df) SRMR CFI TLI RMSEA 1. Single-factor model χ2(54) = 383.95*** - .04 .90 .88 .11 2. Six-factor model χ2(39) = 211.30*** χ2(15) = 172.64*** .03 .95 .91 .09 3. Second-order model χ2(48) = 254.60*** χ2(9) = 43.30*** .04 .94 .92 .09 Notes. *** p < .001. Model 3 has a significantly better fit than Model 1: χ2(6) = 129.35***.

We then assessed the extent to which teachers of the same school agreed among

themselves in terms of their perceptions of their school principals’ commitment to PV.

The ICC(1) and ICC(2) (Bliese, 1998) of commitment to PV were .44 and .78,

respectively, indicating strong agreement among respondents (Woehr et al., 2015).

Regarding the socialized charismatic leadership measure, all five items showed to

be highly convergent (Cronbach’s Alpha = .95), and CFA results indicated a good fit:

SRMR = .01, CFI = 1.00, TLI = .99, RMSEA = .02. Teachers of the same school also

seemed to agree in terms of their perceptions of their principals’ engagement in

charismatic leadership: ICC(1) = .35 and ICC(2) = .68. The ICCs of both commitment to

PV and charismatic leadership provides support for aggregation and analysis at the school

level (Woehr et al., 2015).

Hypotheses testing. Means, standard deviations, and bivariate correlations among

studied variables are reported in Table 4. The correlations reveal a strong positive

association between commitment to PV and charismatic leadership (r = .67, p < .001).

Both commitment to PV and charismatic leadership were negatively related to turnover

(r = -.29, p < .01 and r = -.31, p < .01, respectively). Also, contextual stress is positively

related to teacher turnover (r = .24, p < .05). These results provided preliminary support

for the hypothesized relationships. 75

Figure 2. Leader Commitment to PV (Second-order Factor)

Notes. Model goodness-of-fit: χ2(48) = 254.60***, SRMR = .04, CFI = .94, TLI = .92, RMSEA = .09.

Table 4. Descriptive Statistics and Intercorrelations

Mean SD 1 2 3 4 5 6 7 Study Variables: a 1 Commitment to PV 4.43 0.60 - a 2 Charismatic Leadership 4.11 0.79 .67*** - b 3 Contextual Stress 0 0.50 -.18 -.10 - 4 Turnover 2.88 2.29 -.28** -.30** .28** - Control Variables: 5 School Past Performance 5.08 0.83 .21* .13 -.28** -.18 - 6 Number Teachers 29.54 10.81 -.39*** -.29** 0.16 .21* -.54*** - 7 Students per Class 30.55 9.09 -.15 -.22* 0.19 -.04 -.51*** .44*** - 8 Principal Genderc 1.91 0.29 .08 .04 0.07 .01 .22* -.24* 0.10 Notes. N = 87. Level of analysis: schools. *p < .05 ** p < .01 ***p < .001. aRaw mean (average of the scale items) are presented. bStandardized factor scores. cMale = 1, female = 2.

76

Hypotheses were then tested with linear (H1) and negative binomial regression

(H2 and H3) using Stata 14 Software. Turnover is a right-skewed count variable (Stanley et al., 2013) and negative binomial regression is the most appropriate method to model this type of variable (Gardner, Mulvey, & Shaw, 1995; Stanley et al., 2013). Additional analyses were conducted with generalized structural equation modeling (GSEM) both as a robustness check of our results and to conduct the conditional path analysis (Hayes,

2013); that is, to test the indirect effect of perceived commitment to PV on turnover through charismatic leadership (mediation).

Table 5 shows the results of our regression analyses. Supporting H1, we found a strong positive association between perceived school principals’ commitment to PV and the existence of charismatic leadership processes in public schools (b = 0.65, SE = 0.08, p < .001). Further, the results revealed that the relationship between charismatic leadership and teacher voluntary turnover was negative and significant (b = -0.22, SE =

0.10, p < .05), meaning that schools managed by more charismatic principals have, on average, lower turnover rates than schools managed by less charismatic leaders, thus providing support to H2.

Noteworthy, commitment to PV also had a negative effect on turnover (b = -0.12,

SE = 0.05, p < .05; see Model 4, Table 5). This effect, however, became null after the inclusion of charismatic leadership in the model (Model 5, Table 5), which suggests that the effect of commitment to PV on turnover may be mediated by charismatic leadership

(Baron & Kenny, 1986). To further analyze this result, we conducted more strict mediation tests with GSEM.

77

Table 5. Regression Results

Dependent Variables Charismatic Leadership Employee Turnover (OLS Linear Regression) (Negative Binomial Regression) Independent Variables Model 1 Model 2 Model 3 Model 4 Model 5 Model 6

Study Variables: 0.65*** -.12* 0.02 0.01 Commitment to PV (0.08) (0.05) (0.10) (0.10) -0.22* -0.16 Charismatic Leadership (0.10) (0.09) 0.06 0.14 0.12 0.09 Contextual Stress (0.10) (0.09) (0.08) (.08) -0.11* Charism. Lead. X Contex. Stress (0.05)

Control Variables: -0.11 -0.19 -0.22 -0.15 -0.21 -0.21 School Past Performance (0.14) (0.13) (0.13) (0.14) (0.13) (0.13) -0.02* -0.00 0.02* 0.01* 0.01* 0.01* Number of Teachers (0.01) (0.01) (0.01) (0.00) (0.01) (0.01) - - -0.02 -0.02 -0.02* -0.03** Students per Class 0.03*** 0.03*** (0.02) (0.01) (0.01) (0.01) (0.00) (0.00) 0.05 0.12 0.41 0.31 0.34 0.29 Principal Gender a (0.29) (0.28) (0.25) (0.27) (0.26) (0.26) Constant 1.72 1.41 2.17 1.66 2.00 2.03 R2 .10 .48 .03 .05 .06 .07 Notes. N = 87. Level of analysis: schools. *p < .05 **p < .01 ***p < .001. Non-standardized coefficients. Robust standard errors in parenthesis. aMale = 1, female = 2.

Regarding H3, we predicted that more stressful and demanding environments

would strengthen the effects of charismatic leadership on turnover. Supporting this

prediction, the negative interaction between contextual stress and charismatic leadership

(b = -.11, SE = 0.05, p < .05) shows that in schools located in more stressful environments,

charismatic leadership can more effectively reduce teacher turnover (see Figure 3).

Confirming that, Figure 4 shows that at higher levels of contextual stress, the effect of

charismatic leadership on turnover is significant and negative (the 95% confidence

interval excludes zero), whereas in less stressful environments, the effect is not

significant. 78

Figure 3. Predicted Teacher Turnover (Interaction Effect)

Figure 4. Marginal Effects of Charismatic Leadership on Turnover

Notes. Shaded area: 95% confidence interval of charismatic leadership’s effect on turnover.

Additional Analysis. As robustness check of our results, we estimated the entire model with GSEM. Also, as commitment to PV was shown to affect charismatic leadership, which, in turn, affects turnover—especially in more stressful contexts—we 79

also tested whether charismatic leadership functions as a mediator of commitment to PV’s effects. To do so, we conducted a conditional path analysis (Hayes, 2013), testing the indirect effects of commitment to PV on turnover (mediated by charismatic leadership), at different levels of contextual stress. We followed the procedures outlined by Hayes

(2013)11. All the paths to charismatic leadership were modeled with linear regression. The paths to employee turnover were modeled with negative binomial regression (Figure 5).

Results confirm that all hypothesized relationships were supported.

Figure 5. GSEM Results – Full Model

Notes. *p < .05 ** p < .01 ***p < .001. Non-standardized coefficients. c': Direct effect of commitment to PV on Turnover.

Further, the results of the conditional path analysis suggest that commitment to

PV significantly decrease turnover to the extent to which it affects the emergence of charismatic leadership, especially in more stressful work environments. Table 6 shows that the indirect effect of commitment to PV on turnover (through charismatic leadership) reaches statistical significance when contextual stress is high (b = -0.18, SE = 0.06, p <

.01). At average and low levels of contextual stress, the indirect effect of commitment to

11 The script for conditional path analysis (moderated mediation) for Stata Statistical Software can be found at https://stats.idre.ucla.edu/stata/faq/how-can-i-do-moderated-mediation-in-stata/. 80

PV on turnover is not significant. Finally, the non-significance of the direct effect of commitment to PV on turnover (c’, Figure 5) suggests full mediation.

Table 6. Conditional Path Analysis (Moderated Mediation) Conditional Indirect Effects of

Leader Commitment to PV on Turnover Contextual Stress Level: b SE 95% CI Low -0.03 0.08 -0.19 0.13 Average -0.10 0.06 -0.23 0.02 High -0.18** 0.06 -0.31 -0.05 Notes. **p < .01. ”Low” = 1 SD below the mean; “Average” = mean; “High” = 1 SD above the mean.

Discussion

Drawing from the literature on charismatic leadership (e.g.; Conger & Kanungo,

1988; Shamir, House, & Arthur, 1993) and social identity theory of leadership (e.g.,

Hogg, 2001; Van Knippenberg, 2011; Platow et al., 2006), we proposed—and our results confirmed—that followers’ perceptions of their leaders’ commitment of PV are a critical ingredient for the emergence of charismatic leadership in the public sector, especially in street-level bureaucracies. This happens because “[t]he character, goals and aspirations of a leader are not developed in a vacuum” (Gini, 1997, 74). Rather, they frequently develop within a dense atmosphere of shared values, beliefs, and expectations that define what is group prototypical and affect the way followers make sense of and respond to their leaders (Van Knippenberg, 2011). In other words, when leaders are not perceived by followers to embody the group’s identity-defining characteristics, their ability to develop charismatic forms of leadership becomes rather limited (Conger & Kanungo,

1988; Klein & House, 1995; Platow et al., 2006; Van Knippenberg & Hogg, 2003). Our 81

results indicated that the values, beliefs, and expectations that make up group prototypes in the context of street-level bureaucracies encompass the idealized principles of public administration—the PV—and by doing so, they foster the emergence of charismatic leadership processes within public organizations.

Moreover, our results confirmed the idea that charismatic forms of leadership can be especially effective in the public sector (Moynihan, Pandey, & Wright, 2014; Paarlberg

& Lavigna, 2010; Wright & Pandey, 2009), particularly in the prevention of high levels of employee voluntary turnover. In our understanding, this happens because PV-based charismatic leadership can create a more meaningful and satisfying work experience for front-line public servants, mainly by connecting their values and identities to the organizations’ goals (Shamir, House, & Arthur, 2013). Our results show, however, that more attention needs to be given to context. Specifically, we found that the effect of charismatic leadership was stronger in more challenging and stressful work environments characterized by higher levels of local violence and poverty. Previous studies have shown that some public servants, motivated by a humanistic commitment (i.e., high PSM), voluntarily choose to work in more difficult environments (e.g., poorer and more violent neighborhoods) with the purpose of “making a difference” (Achinstein et al., 2010;

Cochran-Smith et al., 2012). However, if they do not find the necessary administrative and psychological support to conduct their activities—which is more likely to occur in these precarious environments—they get frustrated and are more likely to leave the organization (Kraft et al., 2013). Our findings show that in such contexts PV-based charismatic leadership can be remarkably effective in preventing high turnover levels.

Furthermore, the results of the path analysis (moderated-mediation model) suggested that public leaders’ commitment to PV decrease employee turnover to the 82

extent to which it affects the emergence of charismatic leadership, especially in more stressful and demanding work environments. In other words, charismatic leadership seems to function as a mediator of perceived commitment to PV’s effect on turnover. In more stable work environments, both the direct effect of charismatic leadership and the indirect effect of perceived commitment to PV on turnover were non-significant.

Although our study was not intended to explore that, maybe in more stable work environments, transactional and task-oriented forms of leadership can be equally predictive of employee voluntary turnover.

Theoretical Contributions

Our study advances the current knowledge of public leadership and the role of PV in public organizations in four major ways. First, differently from previous public leadership research, which adopted a more conceptual-normative approach and focused on the potential effects of leaders’ commitment to PV on abstract societal-level outcomes, we shifted the focus to the internal environment of public organizations and adopted a more empirical approach based on behavioral theories. More specifically, we tackled the role of public servants’ perceptions of their leaders’ commitment to PV. Although socio- cognitive processes have been found to be critical determinants of effective leadership

(Lord & Maher, 1993, Van Knippenberg, 2011), these effects have been largely overlooked by previous public leadership research.

Second, this study helps explain the factors that underlie the emergence of charismatic forms of leadership in public organizations, especially in street-level bureaucracies. Given that such forms of leadership were consistently shown to positively 83

affect public organizations’ outcomes (Orazi, Turrini, & Valotti, 2013; Van Wart, 2013;

Vogel & Masal, 2015), comprehending its antecedents becomes rather relevant.

Third, we show that contextual factors must be taken into account in the study of the effectiveness of charismatic leadership in the public sector. Most of previous studies have considered charismatic leadership to be a universally positive form of public leadership. Our results, however, suggest otherwise.

Finally, we believe our study takes another step towards connecting the PSM and the PV literature. Following Maesschalck, Van der Wal, and Huberts’ (2008) view that

PSM-related values and “classical” PV complement each other (e.g., being self-sacrificial and compassionate may not be beneficial to society if it is not coupled with lawfulness and accountability), we combined both sets of values to form our behavioral measure of leader commitment to PV. Our results provided support for such integration.

Limitations and Future Research

We acknowledge that this work has important limitations, which opens several avenues for future research. First, although the use of a split-sample design and objective outcome variables to mitigate response biases (Ostroff, Kinicki, & Clark, 2002), we could not address all sources of endogeneity that are common to cross-sectional studies.

Moreover, the split-sample design prevented us from using hierarchical models that could have accounted for some individual-level characteristics of respondents. Future work can use experimental or longitudinal multi-level designs to further test the causal relationship between the proposed variables.

Moreover, while we show that our final sample is quite representative of the population being studied, our reduced sample size (87 public schools) and the specificity 84

of the context (public educational arena) points to the need of future replication of our results in order to ascertain their generalizability. Also, we focused on street-level bureaucracies and argued that in these types of public organizations PV are more likely to be shared by public servants. Future research may also investigate if the effect of perceived leader commitment to PV on charismatic leadership is weaker in public organizations that are more distant from the “street level”, since PV may be less salient in such contexts.

Furthermore, we focused only on the effects of perceived leaders’ commitment to

PV on the emergence of charismatic leadership. We argue, however, that such perceptions can have other important effects on followers. For instance, based on social learning theory (Bandura, 1986), ethical leadership theory (Brown, Treviño, & Harrison, 2005) proposes that leaders can influence the ethical conduct of followers via role modeling.

We believe the same rationale can be applied to PV-based leadership. As such, future research may examine the extent to which perceived leader commitment to PV could be modeled and emulated by public servants.

Finally, although we focused on the leadership of those occupying formal positions of authority in public organizations, we believe that our study does not conflict with the contemporary public leadership literature that calls for more distributed, collaborative, and relational conceptualizations of leadership (e.g., Crosby & Bryson,

2010; Ospina, 2017). The existence of participative and democratic “horizontal” leadership processes is definitely not at odds with the existence of a more vertical, formal leadership (Pearce, 2004). On the contrary, assigned leaders, especially when guided by

PV, may act as facilitators and catalysts of collaboration processes, stimulating internal and external stakeholders to actively participate in the creation of public value (Campbell, 85

2018; Luke, 1998). In this regard, future research may explore the role of formal (vertical)

PV-based leadership on the emergence of influential informal leadership networks at all levels of public organizations.

Conclusion

We believe that this work contributes to our knowledge about how and when PV can affect public leadership processes and outcomes. By showing that commitment to PV is not only a public leader’s obligation (from a normative perspective), but also an important predictor of more engaging, inspirational forms of public leadership, we bring more publicness to behavioral public leadership research (see Vogel & Masal, 2015).

Moreover, we address Crosby and Bryson’s (2018) recent recommendation that we should not only pay more attention to PV, but also to existing leadership theories, such as

Burn’s (1978) transformational (or charismatic) leadership and the social identity theory of leadership (Hogg, 2001; Van Knippenberg, 2011), with the objective of integrating these behavioral theories in the public leadership research agenda.

86

References

Achinstein, Betty, Rodney T. Ogawa, Dena Sexton, and Casia Freitas. 2010. "Retaining teachers of color: A pressing problem and a potential strategy for “hard-to-staff” schools." Review of Educational Research 80: 71-107. Andersen, Lotte Bøgh, Torben Beck Jørgensen, Anne Mette Kjeldsen, Lene Holm Pedersen, and Karsten Vrangbæk. 2013. "Public values and public service motivation: Conceptual and empirical relationships." The American Review of Public Administration 43: 292-311. Antonakis, John, Nicolas Bastardoz, Philippe Jacquart, and Boas Shamir. 2016. "Charisma: An ill-defined and ill-measured gift." Annual Review of Organizational Psychology and Organizational Behavior 3: 293-319. Ashforth, Blake E., and Fred Mael.1989. "Social identity theory and the organization." Academy of Management Review 14: 20-39. Bandura, Albert. 1986. Social foundations of thought and action. Englewood Cliffs: Prentice-Hall. Baron, Reuben M., and David A. Kenny. 1986. "The moderator–mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations." Journal of Personality and Social Psychology 51: 1173-82. Bass, Bernard M. Leadership and performance beyond expectations. New York: Free Press, 1985. Bellé, Nicola. 2013a. "Leading to make a difference: A field experiment on the performance effects of transformational leadership, perceived social impact, and public service motivation." Journal of Public Administration Research and Theory 24: 109-36. Bellé, Nicola. 2013b. "Experimental evidence on the relationship between public service motivation and job performance." Public Administration Review 73: 143-53. Bliese, Paul D. 1998. "Group size, ICC values, and group-level correlations: A simulation." Organizational Research Methods 1: 355-73. 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 Theory 8: 413-40. Brown, Kathleen M., and Susan R. Wynn. 2009. "Finding, supporting, and keeping: The role of the principal in teacher retention issues." Leadership and Policy in Schools 8: 37-63. 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: 117-34. Bryson, John M., Barbara C. Crosby, and Laura Bloomberg. 2014. "Public value governance: Moving beyond traditional public administration and the new public management." Public Administration Review 74: 445-56. Burns, James M. Leadership. 1978. New York: Harper and Row. 87

Campbell, Jesse W. 2018. "Efficiency, incentives, and transformational leadership: Understanding collaboration preferences in the public sector." Public Performance & Management Review 41: 277-99. Cochran-Smith, Marilyn, Patrick McQuillan, Kara Mitchell, Dianna Gahlsdorf Terrell, Joan Barnatt, Lisa D’Souza, Cindy Jong, Karen Shakman, Karen Lam, and Ann Marie Gleeson. 2012. "A longitudinal study of teaching practice and early career decisions: A cautionary tale." American Educational Research Journal 49: 844- 80. Conger, Jay A., and Rabindra N. Kanungo. 1988. Charismatic leadership: The elusive factor in organizational effectiveness. San Francisco: Jossey-Bass. Conger, Jay A., and Rabindra N. Kanungo. 1998. Charismatic leadership in organizations. London: Sage Publications. Crosby, Barbara C., and John M. Bryson. 2010. "Integrative leadership and the creation and maintenance of cross-sector collaborations." The Leadership Quarterly 21: 211-30. Crosby, Barbara C., and John M. Bryson.. 2018 "Why leadership of public leadership research matters: and what to do about it." Public Management Review 20: 1265- 86. DeGroot, Timothy, D. Scott Kiker, and Thomas C. Cross. 2000. "A metaanalysis to review organizational outcomes related to charismatic leadership." Canadian Journal of Administrative Sciences/Revue Canadienne des Sciences de l'Administration 17: 356-72. Denhardt, Janet V., and Kelly B. Campbell. 2006. "The role of democratic values in transformational leadership." Administration & Society 38: 556-72. Dumdum, Uldarico Rex, Kevin B. Lowe, and Bruce J. Avolio. 2013. "A meta-analysis of transformational and transactional leadership correlates of effectiveness and satisfaction: An update and extension." In Transformational and Charismatic Leadership: The Road Ahead 10th Anniversary Edition, ed. Bruce J. Avolio, and Francis J. Yammarino, 39-70. Bingley: Emerald. Eagly, Alice H., and Steven J. Karau. 2002. "Role congruity theory of prejudice toward female leaders." Psychological review 109: 573-98. Favero, Nathan, and Justin B. Bullock. 2014. "How (not) to solve the problem: An evaluation of scholarly responses to common source bias." Journal of Public Administration Research and Theory 25: 285-308. Fisher, Robert J., and James E. Katz. 2000. "Socialdesirability bias and the validity of selfreported values." Psychology & Marketing 17: 105-20. Gardner, William, Edward P. Mulvey, and Esther C. Shaw. 1995. "Regression analyses of counts and rates: Poisson, overdispersed Poisson, and negative binomial models." Psychological Bulletin 118: 392-404. Gardner, William L., and Bruce J. Avolio. 1998. "The charismatic relationship: A dramaturgical perspective." Academy of Management Review 23: 32-58. Getha-Taylor, Heather, Maja Husar Holmes, Willow S. Jacobson, Ricardo S. Morse, and Jessica E. Sowa. 2011. "Focusing the public leadership lens: Research 88

propositions and questions in the Minnowbrook tradition." Journal of Public Administration Research and Theory 21: 83-97. Gini, Al. 1997. "Moral leadership and business ethics." Journal of Leadership Studies 4: 64-81. Grant, Adam M. 2012. "Leading with meaning: Beneficiary contact, prosocial impact, and the performance effects of transformational leadership." Academy of Management Journal 55: 458-76. Griffith, James. 2004. "Relation of principal transformational leadership to school staff job satisfaction, staff turnover, and school performance." Journal of Educational Administration 42: 333-56. Grissom, Jason A. 2011. "Can good principals keep teachers in disadvantaged schools? Linking principal effectiveness to teacher satisfaction and turnover in hard-to-staff environments." Teachers College Record 113: 2552-85. Grissom, Jason A., Samantha L. Viano, and Jennifer L. Selin. 2016. "Understanding employee turnover in the public sector: Insights from research on teacher mobility." Public Administration Review 76: 241-51. Groves, Robert M., Floyd J. Fowler Jr, Mick P. Couper, James M. Lepkowski, Eleanor Singer, and Roger Tourangeau. 2011. Survey methodology. Hoboken: John Wiley & Sons. Haslam, S. Alexander, and Stephen D. Reicher. 2016. "Rethinking the psychology of leadership: From personal identity to social identity." Daedalus 145: 21-34. Haslam, S. Alexander, Stephen D. Reicher, and Michael J. Platow. 2010. The new psychology of leadership: Identity, influence and power. London: Psychology Press. Hayes, Andrew F. 2013. Introduction to mediation, moderation, and conditional process analysis: A regression-based approach. New York: Guilford Press. Herman, H. M., Xu Huang, and Wing Lam. 2013. "Why does transformational leadership matter for employee turnover? A multi-foci social exchange perspective." The Leadership Quarterly 24: 763-76. Hoffman, Brian J., Bethany H. Bynum, Ronald F. Piccolo, and Ashley W. Sutton. 2011. "Person-organization value congruence: How transformational leaders influence work group effectiveness." Academy of Management Journal 54: 779-96. Hogg, Michael A. 2001. "A social identity theory of leadership." Personality and Social Psychology Review 5: 184-200. House, Robert J. 1998. "Measures and assessments for the charismatic leadership approach: Scales, latent constructs, loadings, Cronbach alphas, and interclass correlations." In Monographs in Organizational Behavior and Industrial Relations, ed. Fred Dansereau, and Francis J. Yammarino, 23-30. London: JAI Press. House, Robert J. 1977. “A 1976 Theory of Charismatic Leadership.” In Leadership: The Cutting Edge, ed. James G. Hunt, and Lars L. Larson, 189-207. Carbondale: Southern Illinois University Press. 89

House, Robert J., and Jane M. Howell. 1992. "Personality and charismatic leadership." The Leadership Quarterly 3: 81-108. Howell, Jane M. 1988. “Two faces of charisma: Socialized and personalized leadership in organizations.” In Charismatic leadership: The elusive factor in organizational effectiveness, ed. Jay A. Conger, and Rabindra N. Kanungo, 213-36. San Francisco: Jossey-Bass. Howell, Jane M., and Boas Shamir. 2005. "The role of followers in the charismatic leadership process: Relationships and their consequences." Academy of Management Review 30: 96-112. Jensen, Ulrich T., and Louise Ladegaard Bro. 2018. "How transformational leadership supports intrinsic motivation and public service motivation: The mediating role of basic need satisfaction." The American Review of Public Administration 48: 535- 49. Jensen, Ulrich T., Lotte Bøgh Andersen, and Christian Bøtcher Jacobsen. 2019. "Only When We Agree! How Value Congruence Moderates the Impact of Goal Oriented Leadership on Public Service Motivation." Public Administration Review 79: 12-24. Jermier, John M. 1993. "Introduction: Charismatic leadership: Neo-Weberian perspectives." The Leadership Quarterly 4: 217-33. Jørgensen, Torben B. 2006. "Public values, their nature, stability and change. The case of Denmark." Public Administration Quarterly 30: 365-98. Jørgensen, Torben B., and Barry Bozeman. 2007. "Public values: An inventory." Administration & Society 39: 354-81. Judge, Timothy A. 1993. "Does affective disposition moderate the relationship between job satisfaction and voluntary turnover?" Journal of Applied Psychology 78: 395- 401. Kim, Sangmook, Wouter Vandenabeele, Bradley E. Wright, Lotte Bøgh Andersen, Francesco Paolo Cerase, Robert K. Christensen, Céline Desmarais et al. 2012. "Investigating the structure and meaning of public service motivation across populations: Developing an international instrument and addressing issues of measurement invariance." Journal of Public Administration Research and Theory 23: 79-102. Klein, Katherine J., and Robert J. House. 1995. "On fire: Charismatic leadership and levels of analysis." The Leadership Quarterly 6: 183-98. Kouzes, James M., and Barry Z. Posner. 2007. The Leadership Challenge. San Francisco: Jossey-Bass. Kraft, Matthew A., John P. Papay, Megin Charner-Laird, Susan Moore Johnson, Monica Ng, and Stefanie K. Reinhorn. 2012. "Committed to their students but in need of support: How school context influences teacher turnover in high-poverty, urban schools." Unpublished manuscript. Krogsgaard, Julie Alsøe, Pernille Thomsen, and Lotte Bøgh Andersen. 2014. "Only if we agree? How value conflicts moderate the relationship between transformational leadership and public service motivation." International Journal of Public Administration 37: 895-907. 90

Liu, Bangcheng, Thomas Li-Ping Tang, and Kaifeng Yang. 2015. "When does public service motivation fuel the job satisfaction fire? The joint moderation of person– organization fit and needs–supplies fit." Public Management Review 17: 876- 900. Lord, Robert G., and Karen J. Maher. 1993. Leadership and information processing: Linking perception and performance. New York: Routledge. Luke, Jeffrey S. 1998. Catalytic leadership. San Francisco: Jossey-Bass. Maesschalck, Jeroen, Zeger Van der Wal, and L. W. J. C. Huberts. 2008. "Public service motivation and ethical conduct." In Motivation in public management: The call of public service, ed. James L. Perry, and Annie Hondeghem, 157-76. New York: Oxford University Press. Meglino, Bruce M., Elizabeth C. Ravlin, and Cheryl L. Adkins. 1989. "A work values approach to corporate culture: A field test of the value congruence process and its relationship to individual outcomes." Journal of Applied Psychology 74: 424-43. Meier, Kenneth J., and John Bohte. 2003. "Span of control and public organizations: Implementing Luther Gulick’s research design." Public Administration Review 63: 61-70. Meindl, James R., Sanford B. Ehrlich, and Janet M. Dukerich. 1985. "The romance of leadership." Administrative Science Quarterly 30: 78-102. Moynihan, Donald P., Sanjay K. Pandey, and Bradley E. Wright. 2014. "Transformational leadership in the public sector." In Public administration reformation: Market demand from public organizations, ed. Yogesh K. Dwivedi, Mahmud A. Shareef, Sanjay K. Pandey, and Vinod Kumar, 87-104. New York: Routledge. Newell, Terry. 2007. "Values-based leadership for a democratic society." The Trusted Leader: Building the Relationships That Make Government Work, ed. Terry Newell, Grant Reeher, and Peter Ronayne, 19-48. Washington: CQ Press. Nyberg, Anthony. 2010. "Retaining your high performers: Moderators of the performance–job satisfaction–voluntary turnover relationship." Journal of Applied Psychology 95: 440-53. O'Toole, Laurence J., and Kenneth J. Meier. 2003. "Plus ça change: Public management, personnel stability and organizational performance." Journal of Public Administration Research and Theory 13: 43-64. Orazi, Davide Christian, Alex Turrini, and Giovanni Valotti. 2013. "Public sector leadership: new perspectives for research and practice." International Review of Administrative Sciences 79: 486-504. Ospina, Sonia M. 2017. "Collective leadership and context in public administration: Bridging public leadership research and leadership studies." Public Administration Review 77: 275-87. Ostroff, Cheri, Angelo J. Kinicki, and Mark A. Clark. 2002. "Substantive and operational issues of response bias across levels of analysis: an example of climate- satisfaction relationships." Journal of Applied Psychology 87: 355-68. 91

Paarlberg, Laurie E., and Bob Lavigna. 2010. "Transformational leadership and public service motivation: Driving individual and organizational performance." Public Administration Review 70: 710-18. Paarlberg, Laurie E., James L. Perry, and Annie Hondeghem. 2008. "From theory to practice: Strategies for applying public service motivation." In Motivation in public management: The call of public service, ed. James L. Perry, and Annie Hondeghem, 268-93. New York: Oxford University Press. Pandey, Sanjay K., Randall S. Davis, Sheela Pandey, and Shuyang Peng. 2016. "Transformational leadership and the use of normative public values: Can employees be inspired to serve larger public purposes?" Public Administration 94: 204-22. Park, Sung Min, and Hal G. Rainey. 2008. "Leadership and public service motivation in US federal agencies." International Public Management Journal 11: 109-142. Pearce, Craig L. 2004. "The future of leadership: Combining vertical and shared leadership to transform knowledge work." Academy of Management Perspectives 18: 47-57. Perry, James L., and Lois Recascino Wise. 1990. "The Motivational Bases of Public Service." Public Administration Review 50: 367-73. Pitts, David, John Marvel, and Sergio Fernandez. 2011. "So hard to say goodbye? Turnover intention among US federal employees." Public Administration Review 71: 751-60. Platow, Michael J., S. Alexander Haslam, Margaret Foddy, and Diana Grace. 2003. "Leadership as the outcome of self-categorization processes." In Leadership and Power: Identity Processes in Groups and Organizations, ed. Daan Van Knippenberg, and Michael A. Hogg, 34-47. London: Sage Publications. Platow, Michael J., Daan Van Knippenberg, S. Alexander Haslam, Barbara Van Knippenberg, and Russell Spears. 2006. "A special gift we bestow on you for being representative of us: Considering leader charisma from a self categorization perspective." British Journal of Social Psychology 45: 303-20. Podsakoff, N. P. 2003. "Common method biases in behavioral research: A critical review of the literature and recommended remedies." Journal of Applied Psychology 88: 879-903. Rayner, Julie, Helen M. Williams, Alan Lawton, and Christopher W. Allinson. 2010. "Public service ethos: Developing a generic measure." Journal of Public Administration Research and Theory 21: 27-51. Rutgers, Mark R. 2015. "As good as it gets? On the meaning of public value in the study of policy and management." The American Review of Public Administration 45: 29-45. Schein, Virginia E., Ruediger Mueller, Terri Lituchy, and Jiang Liu. 1996. "Think manager—think male: A global phenomenon?." Journal of organizational behavior 17: 33-41. Schwartz, Shalom H., and Wolfgang Bilsky. 1987. "Toward a universal psychological structure of human values." Journal of Personality and Social Psychology 53: 550-62. 92

Shamir, Boas, Robert J. House, and Michael B. Arthur. 1993. "The motivational effects of charismatic leadership: A self-concept based theory." Organization Science 4: 577-94. Shamir, Boas, and Jane M. Howell. 1999. "Organizational and contextual influences on the emergence and effectiveness of charismatic leadership." The Leadership Quarterly 10: 257-83. Simon, Nicole S., and Susan Moore Johnson. 2015. "Teacher turnover in high-poverty schools: What we know and can do." Teachers College Record 117: 1-36. Smith, Deborah L., and Brian J. Smith. 2006. "Perceptions of violence: The views of teachers who left urban schools." The High School Journal 89: 34-42. Stanley, Laura, Christian Vandenberghe, Robert Vandenberg, and Kathleen Bentein. 2013. "Commitment profiles and employee turnover." Journal of Vocational Behavior 82: 176-87. Svara, James H. 2014. "Who are the keepers of the code? Articulating and upholding ethical standards in the field of public administration." Public Administration Review 74: 561-569. Tajfel, H., and John C. Turner. 1986. “The social identity theory of intergroup behavior.” In Psychology of intergroup relations, ed. William G Austin, and Stephen Worchel, 7-24. Chicago: Nelson-Hall. Terry, Larry D. 1990. "Leadership in the administrative state: The concept of administrative conservatorship." Administration & Society 21: 395-412. Trottier, Tracey, Montgomery Van Wart, and Xiao Hu Wang. 2008. "Examining the nature and significance of leadership in government organizations." Public Administration Review 68: 319-33. Tummers, Lars, and Eva Knies. 2016. "Measuring public leadership: Developing scales for four key public leadership roles." Public Administration 94: 433-51. Van der Wal, Zeger, Tina Nabatchi, and Gjalt De Graaf. 2015. "From galaxies to universe: A cross-disciplinary review and analysis of public values publications from 1969 to 2012." The American Review of Public Administration 45: 13-28. Van der Wal, Zeger, and Leo Huberts. 2008. "Value solidity in government and business: Results of an empirical study on public and private sector organizational values." The American Review of Public Administration 38: 264-85. Van der Wal, Zeger, Leo Huberts, Hans Van den Heuvel, and Emile Kolthoff. 2006. "Central values of government and business: Differences, similarities and conflicts." Public Administration Quarterly 30: 314-64. Van Knippenberg, Daan. 2011. "Embodying who we are: Leader group prototypicality and leadership effectiveness." The Leadership Quarterly 22: 1078-91. Van Knippenberg, Daan, and Michael A. Hogg. 2003. "A social identity model of leadership effectiveness in organizations." Research in Organizational Behavior 25: 243-95. Van Knippenberg, Daan, and Sim B. Sitkin. 2013. "A critical assessment of charismatic— transformational leadership research: Back to the drawing board?" The Academy of Management Annals 7: 1-60. 93

Van Wart, Montgomery. 2013. Changing public sector values. New York: Routledge. Vogel, Rick, and Doris Masal. 2015. "Public leadership: A review of the literature and framework for future research." Public Management Review 17: 1165-89. Witesman, Eva, and Lawrence Walters. 2014. "Public service values: A new approach to the study of motivation in the public sphere." Public Administration 92: 375-405. Woehr, David J., Andrew C. Loignon, Paul B. Schmidt, Misty L. Loughry, and Matthew W. Ohland. 2015. "Justifying aggregation with consensus-based constructs: A review and examination of cutoff values for common aggregation indices." Organizational Research Methods 18: 704-37. Wright, Bradley E., and Robert K. Christensen. 2010. "Public service motivation: A test of the job attraction–selection–attrition model." International Public Management Journal 13: 155-176. Wright, Bradley E., Donald P. Moynihan, and Sanjay K. Pandey. 2012. "Pulling the levers: Transformational leadership, public service motivation, and mission valence." Public Administration Review 72: 206-15. Wright, Bradley E., and Sanjay K. Pandey. 2008. "Public service motivation and the assumption of person—Organization fit: Testing the mediating effect of value congruence." Administration & Society 40: 502-21. Wright, Bradley E., and Sanjay K. Pandey. "Transformational leadership in the public sector: Does structure matter?" Journal of Public Administration Research and Theory 20, no. 1 (2009): 75-89. Yukl, Gary. "An evaluation of conceptual weaknesses in transformational and charismatic leadership theories." The Leadership Quarterly 10, no. 2 (1999): 285- 305.

94

ARTIGO 3

DISSATISFIED WHEN MY ORGANIZATION IGNORES POOR

PERFORMERS: THE MODERATING ROLE OF PUBLIC SERVICE

MOTIVATION

Introduction

Dealing with poor performance is a fundamental managerial function (Daley,

2008; Yukl, 2013). However, the Federal Employee Viewpoint Survey’s (FEVS’) annual reports12 reveal a worrying pattern. In the last years, the survey item “In my work unit, steps are taken to deal with a poor performer who cannot or will not improve” has repeatedly received the second lowest level13 of agreement of all survey items (e.g., in

2018, only 32% of participants selected either “agree” or “strongly agree” in this item).

It suggests that most federal employees believe that their superiors may be failing to address poor-performing coworkers. While there can be many reasons for managers to be reluctant to deal with a poor performer—especially in the public sector (see Daley, 2008;

Deng & Leung, 2014; Franklin & Pagan, 2006; York, 2018)—we believe it is important for them to know that the costs of ignoring poor-performing employees are not restricted to these employees’ lost productivity.

By ignoring poor performers, managers can leave the social order unbalanced and negatively affect the job attitudes and behaviors of higher-performing employees

(Atwater, Waldman, Carey, & Cartier, 2001; Niehoff, Paul, & Bunch, 1998; O'Reilly &

12 The FEVS’ yearly reports are available on https://www.opm.gov/fevs/reports/governmentwide-reports/. 13 The FEVS’ item with the lowest level of agreement (28% in 2018) is “Pay raises depend on how well employees perform their jobs.” 95

Puffer, 1989; Skarlicki, O’Reilly, & Kulik, 2015; Treviño, 1992). One social effect of ignoring poor performance concerns the inequity distress experienced by employees who perform at higher levels as they may feel that their greater contributions are not being properly rewarded by the organization (Treviño, 1992; Schnake & Durnler, 1988;

Skarlicki et al., 2015). Feelings of inequity could also be addressed by rewarding high performers (Adams, 1963). However, a common characteristic of public-sector jobs is the standardization of pay and the constraints to the use of performance-based incentives

(Moynihan, Pandey, & Wright, 2014). Then, addressing poor performers may remain as an important managerial strategy to foster a climate of justice in public organizations.

Besides these equity issues, we also expect that when poor performance is systematically ignored in public organizations, employees will likely conclude that their organizations have a low concern for results and do not embrace the values of accountability, efficiency, and effectiveness in the provision of public services. In this case, we expect that employees—especially those with greater PSM, as they are more concerned about making meaningful contributions to society (Perry & Wise, 1990)—will hold a more negative image of their organizations (Dutton, Dukerich, & Harquail, 1994;

Rho, Yun, & Lee, 2015). That negative image should lead to lower employee- organization value congruence (Ashforth, Harrison, & Corley, 2008). Both feelings of inequity and low value congruence are expected to be translated into lower satisfaction with the job (Cohen-Charash & Spector, 2001, Cropanzano & Ambrose, 2015; Edwards

& Cable, 2009).

In the present study, we conducted a survey with a sample of 217 government employees from Brazil to test the hypotheses that perceived organization’s failure to address poor performance negatively affects job satisfaction, particularly when employee 96

PSM is higher (a moderation effect). We also conducted some additional analysis to explore the non-linear relationship between perceiving the organization ignore/address poor performance and job satisfaction.

We believe that this study contributes to public administration theory and practice in some ways. While most studies have investigated the role of PSM as an antecedent of job satisfaction (see Ritz, Brewer, & Newman, 2016), we explore how PSM can affect the way employees make sense and respond (in terms of job satisfaction) to their organizations’ (deficient) practices. We draw attention to the fact that public employees with greater PSM may not only be more willing to perform at higher levels (e.g., Bellé,

2013; Perry & Wise, 1990); they also seem to have high expectations of their organizations setting a high standard of performance for all employees. In practical terms, our results indicate that public managers should be highly aware of the potential social effects of ignoring poor performers as they decide to take (or not to take) steps to deal with them. We also discuss the contributions of this study to the literature on workplace discipline.

Managing Poor Performance and its Social Effects

Dealing with poor performers is an essential, albeit distasteful managerial function

(Daley, 2008; Trahan & Steiner, 1994). A consensus exists that managers’ inaction in the face of an employee’s recurrent poor performance is a dysfunctional supervisory behavior that can have negative consequences for the organization. However, dealing with

“problem employees” is never an easy task (Greer & Labig, 1987; Yukl, 2013). It may lead to confrontation—which is not enjoyable for most people—increases the risk of involving managers or the organization in litigation (Goodhew, Cammock, & Hamilton, 97

2008), and can be an emotionally exhausting process (Daley, 2008). Moreover, there is a conventional wisdom that disciplinary actions are ineffective and should be avoided (see

Butterfield, Treviño, Wade, & Ball, 2005). As a result, managers are frequently reluctant to deal with poor performers, which may create a culture of “chronic niceness” in organizations (Goodhew et al., 2008).

Dealing with poor performance does not necessarily mean resorting to punishment, although the use of disciplinary actions is one of the tools at the managers’ disposal. A manager can use strategies such as meetings with employees to clarify and reaffirm his/her performance expectations, individualized feedback sessions, training, counselling, and verbal reprimand (Bugdol, 2018). More severe actions include written reprimand, holding back pay raises, suspending the employee without pay, and dismissal

(Trahan & Steiner, 1994; Treviño, 1992).

Within the public administration literature, studies have explored various aspects related to the use of disciplinary actions and the challenge of dealing with poor performers

(Daley, 2008; Gaertner & Gaertner, 1984; Guffey & Helms, 2001; Kearney & Hays,

1995; King & Wilcox, 2003; Selden, 2006). One theme that is commonly tackled by researchers concerns the administrative burden required from public managers to uphold formal disciplinary actions (Ban, 1995; Daley, 2008; Franklin & Pagan, 2006; York,

2018). Complex and lengthy grievance and appeal systems common to public organizations can make dealing formally with poor performers a daunting and frustrating task for managers (Daley, 2008; Franklin & Pagan, 2006; York, 2018). As a result, in many cases, public managers prefer to use alternative (off-the-records) disciplinary actions or simply ignore poor performers at all (Franklin & Pagan, 2006). 98

One common objective of using progressive discipline is to rehabilitate a poor- performing employee (Daley, 2008). Some studies, however, question the effectiveness of using disciplinary actions to promote changes in behavior, particularly when compared to the effectiveness of using rewards or other motivational strategies (see Treviño &

Weaver, 2010). While some authors are less pessimistic about it and suggest that when some conditions are met (e.g., giving voice to employees, being interpersonally just, using a sanction that is proportional to the infraction, etc.), disciplinary actions can be successful in changing employee behaviors (Ball, Treviño, & Sims, 1992; Podsakoff,

Podsakoff, & Kuskova, 2010), it should be noted that the importance of dealing with poor performers cannot be restricted to a concern with these employees’ lost productivity. In specific, we highlight the fact that when organizations regularly fail to address poor performers, third-party observers’—especially high performers’—judgements about the organization, job attitudes, and work behaviors can also be negatively affected (Atwater et al., 2001; Liden et al., 1999; Niehoff et al., 1998; O'Reilly & Puffer, 1989; Treviño,

1992).

Effects on Third-party Observers’ Job Satisfaction

Within the stream of literature devoting attention to the factors that influence team success or failure (Cohen & Bailey, 1997; Ilgen, Hollenbeck, Johnson, & Jundt, 2005;

Kozlowski & Bell, 2003), studies have focused on how “bad apples”—i.e., negative team members—can “spoil the barrel” (e.g., Felps, Mitchell, & Byington, 2006; Furnham &

Taylor, 2011; Hung, Chi, & Lu, 2009; Robinson, Wang, & Kiewitz, 2014). It has been pointed out that “sometimes the source of the dysfunction [in a workgroup] is one individual’’ (Keyton, 1999, p. 493). This assertion finds resonance with the idea that “bad 99

is stronger than good” (Baumeister, Bratslavsky, Finkenauer, & Vohs, 2001); that is, negative team members can exert an asymmetric effect on team outcomes (e.g., Camacho

& Paulus, 1995). These findings underscore the importance of managers not to ignore uninvolved employees in their units, even if they are few in number.

According to Felps et al. (2006), three main types of negative team members exist: the interpersonal deviant, the affectively negative individual, and the withholder of effort.

This latter category is more central to the present study. Withholders of effort are poor performing employees who “intentionally dodge their responsibilities to the group and free ride off the efforts of others” (Felps et al., 2006, p. 182). A perception that coworkers perform poorly and free riding off others’ efforts—and are ignored by superiors—can trigger negative cognitive and emotional responses in higher-performing employees (e.g.

Felps et al., 2006; Hung et al., 2009; O'Reilly & Puffer, 1989; Treviño, 1992). In this study, to explain why public organizations’ systematic failure to address poor performers should be negatively related to employees’ job satisfaction, we concentrate on two psychological responses: perceived inequity (Adams, 1963; Skarlicki et al., 2015) and employee-organization value congruence (Edwards & Cable, 2009; Boxx, Odom, &

Dunn, 1991; Hoffman & Woehr, 2006).

Employees’ perceptions of equity (or distributive justice) are normally based on social comparison (Adams, 1963; Cropanzano & Ambrose, 2015; van den Bos, Lind,

Vermunt, & Wilke, 1997) and employees seem to be highly sensitive to peers who get similar rewards for less effort (Adams, 1963; Hung et al., 2009, Treviño, 1992). Then, the presence of poor performers and free riders in workgroups makes more productive group members feel under-rewarded, which creates a sense of inequity. Moreover, when some employees dodge their responsibilities, they may also increase the workload of 100

more committed coworkers, which should further intensify feelings of inequity. Common psychological reactions to inequity include anger, contempt, and indignation (Skarlicki et al., 2015). Felps et al. (2006) highlight that intentional poor performance can be “a major source of distress, as it violates effort norms and takes advantage of other members’ good- faith contributions” (p. 191). Then, more committed coworkers usually interpret such peer behavior as a serious misconduct that should be addressed by the organization (O'Reilly

& Puffer, 1989; Podsakoff et al., 2006; Treviño, 1992). In other words, it creates in the group an expectation for retributive justice (Hogan & Elmer, 1981, Treviño, 1992; Darley

& Pittman, 2003).

However, employees normally do not have the necessary authority to address a negative coworker; then, they naturally expect their superiors to do that (Treviño, 1992).

And the more the deviant employee’s actions—or inaction—are perceived to be harmful to the group, the more coworkers endorse the application of more severe sanctions by their superiors (Treviño & Ball, 1992; Wang & Murnighan, 2017). Then, managers’ failure to deal with poor performers when it is thought to be deserved “leaves the social order unbalanced and is viewed by observers as unjust” (Treviño, 1992, p. 654). When it happens, feelings of inequity and injustice persist and higher-performing employees are likely to experience lower job satisfaction (Cohen-Charash & Spector, 2001; Podsakoff et al., 2006; Skarlicki et al., 2015).

Indeed, the organizational leadership literature shows that managers’ contingent use of discipline is consistently associated to employee perceptions of organizational justice, job satisfaction, satisfaction with the manager, and satisfaction with coworkers

(Podsakoff et al., 2006; Podsakoff et al., 2010), especially if managers treat subordinates with interactional justice (i.e., with respect and humanness; Deng & Leung, 2014). On 101

the other hand, managers’ passive/avoidant behaviors and punishment omission are negatively associated to the above-mentioned outcomes (Hinkin & Schriesheim, 2008;

Holtz & Hu, 2017; Judge & Piccolo, 2004).

Another potential detrimental effect of ignoring poor performers—that has not been explored theoretically so far—concerns public employees’ feelings of low value congruence with their organizations (Boxx et al., 1991; Hoffman & Woehr, 2006; Wright

& Pandey, 2008). Judgements of value congruence (or person-organization fit; Kristof-

Brown, Zimmerman, & Johnson, 2005) can be defined as the perceived level of agreement between one’s own values and those of the organization (Edwards & Cable,

2009). Value congruence is inextricably related to the concept of organizational identification (Ashforth et al., 2008).

Employees’ judgements of value congruence depend on how they view their organizations; that is, on the organizational image held by employees. That image reflects an employee’s beliefs about the distinctive, central, and enduring characteristics of the organization (Dutton et al., 1994). Employees are constantly making sense of their organizations’ goals, practices, and values and forming an organizational image that can be more or less congruent with their self-image (Dukerich, Golden, & Shortell, 2002; Rho et al., 2015). Importantly, due to self-enhancement motives, the more an employee perceive the organization to be committed to virtuous practices and values (i.e., the higher distinctiveness and attractiveness of the organizational image) the more likely he/she is to experience greater value congruence (Ashforth et al., 2008; Dutton et al., 1994).

Research shows that managers’ words and actions can be the most important source of organizationally relevant information for employees as they form impressions about their organizations (Meglino, Ravlin, & Adkins, 1989). Then, we believe that when 102

public managers repeatedly fail to hold employees accountable for their poor performance, higher-performing employees may conclude that their organizations are not seriously committed to promoting a culture of efficiency, hard work, fairness, accountability, and effectiveness in the provision of public services. Consequently, these employees may develop a negative (unattractive) image of their organizations, which is expected to harm their perceptions of value congruence (Ashforth et al., 2008; Dutton et al., 1994; Dukerich et al., 2002; Rho et al., 2015). That is, there will be low consistency between the values employees use to define themselves and the values they use to define their organizations.

The positive impact of employee-organization value congruence on work outcomes has been extensively demonstrated by previous research (Edwards & Cable,

2009; Boxx et al., 1991; Hoffman & Woehr, 2006; Kristof-Brown et al., 2005).

Specifically in the context of public administration, value congruence was shown to be an important factor to help satisfy public employees’ prosocial motivations and lead to positive job attitudes, especially job satisfaction (Christensen & Wright, 2011; Wright &

Pandey, 2008; Kim, 2012; Liu, Tang, & Yang, 2015).

Therefore, by combining the arguments involving both perceptions of inequity and low value congruence, we predict that public organization’s failure to address poor performers should be related to lower employee job satisfaction. Following Herzberg’s two-factor theory (Herzberg, 1968), we believe that organizations’ failure to address poor performers functions as a source of dissatisfaction for employees, while addressing poor performance should not be per se a source of satisfaction (i.e., a hygiene factor, not a motivator). Then, we advance the following hypothesis:

103

Hypothesis 1: Public employees who perceive that their organizations systematically fail to address poor performance will be less satisfied with their jobs, compared to public employees who perceive that their organizations regularly take steps to address poor performance.

The Moderating Effect of Observers’ PSM

PSM has been defined as “an individual's orientation to delivering service to people with the purpose of doing good for others and society” (Hondeghem & Perry,

2009, p. 6) or ‘‘the motivational force that induces individuals to perform meaningful public service’’ (Brewer and Selden, 1998, 417). We believe that employee PSM should moderate the relationship between perceived organization’s failure to address poor performers and job satisfaction for two main reasons: (i) high-PSM employees normally display greater work engagement and job performance (Perry & Vandenabeele, 2015;

Ritz et al., 2016), and (ii) we argue they will see their peers’ deliberate poor performance as more outrageous, and their organizations’ failure to address poor performance as a more serious omission.

Studies have confirmed the relationship between employee PSM and organizational commitment, work motivation, work effort, and job performance (for a review, see Ritz et al., 2016). PSM’s effect on job performance was also confirmed experimentally (e.g., Bellé, 2013). Based on the precepts of the Equity Theory (Adams,

1963), we argue that high-PSM employees, as they are more engaged and perform at higher levels, should be more sensitive to coworkers who perform poorly and get similar rewards. That is, when they perceive that their organizations ignore poor performers, their perceptions of inequity should be accentuated, in comparison to low-PSM employees. 104

That increased inequity distress experienced by high-PSM employees will likely be translated into lower levels of job satisfaction (Cohen-Charash & Spector, 2001).

Moreover, high-PSM employees see their work as more important to society (e.g.,

Stritch & Christensen, 2014) and they are more committed to public values and the public interest (Perry & Wise, 1990; Kim et al., 2012). As such, these employees are more likely interpret peers’ intentional poor performance as a serious misconduct; that is, as a consequential violation of public service values. Perceived consequence severity was shown to increase individuals’ retributive justice expectations following a misconduct

(Treviño, 1992). Relatedly, Skitka (2003) argues that how people interpret others’ misconducts is contingent on what frame they use to understand the situation. When a moral frame is being used, a wrongdoing is more likely to be categorized as moral outrage and people more vigorously want wrongdoers to be addressed (Skitka, 2003). Studies indicate that there is a link between PSM and moral development (Kohlberg, 1984), in the sense that individuals with greater PSM are expected to be in more advanced stages of moral development (e.g., Choi, 2004; Koehler & Rainey, 2008). Then, we believe that public servants with greater PSM are more likely to use a moral frame to judge poor performers. Consequently, if organizations repeatedly fail to address poor performers, high-PSM employees will be more likely to interpret that omission as inacceptable and morally wrong, which can more strongly harm the image they hold of their organizations and, thus, their value congruence.

Altogether, these arguments lead us to conclude that employees with higher PSM

(compared to those with lower PSM) will have their job satisfaction more negatively affected by the perception that their organizations systematically ignore poor performers.

Then, we advance the following hypotheses: 105

Hypothesis 2: The negative effect of perceived organization’s failure to address poor performance on employees’ job satisfaction will be stronger for employees with higher (as opposed to lower) PSM.

Methods Sample and Procedures

Employees from diverse public organizations in Brazil were contacted through professional online forums (restricted to public employees) and via their institutional e-mails14. In total, 217 public employees completed the survey. They were

58% male, with an average age of 42.60 (SD = 9.90), and 78% of them had at least a college degree. On average, they have worked for their current organizations for 12.98 years (SD = 9.16) and had a total of 17.52 years (SD = 9.52) of experience as public servants. They worked for the federal (76%), municipal (16%), and state (8%), governments. Their organizations were majorly subordinated to the Ministry of Defense

(32%), Ministry/Department of Education (30%), and Ministry of Economy (22%).

Participants were presented a term of informed consent saying that their participation was voluntary and anonymous and that their organizations would not be identified either. After providing their informed consent, they were directed to an online questionnaire.

14 Because the first author of this paper is a public servant in Brazil, he had access to online forums and groups (on Facebook and Whatsapp) restricted to public servants, wherein he shared the survey link. In addition to that, we sent the survey link to the institutional e-mails of employees of the Brazilian Institute of Geography and Statistics (IBGE), which is a federal agency. 17% of the responses were provided by IBGE’s employees. The remaining 83% was collected through the online forums and groups. 106

Measures

Perceived organization’s failure to address poor performance. In order to capture participants’ perceptions about how much their organizations ignore (or address) poor performance, we asked them the following question: “According to your perception, how often does each of the following employee behaviors seem to be ignored by your organization when they occur? That is, inform how frequently your organization does not take any steps to address the following counterproductive employee behaviors.” We provided participants with a list of seven counterproductive behaviors (low productivity, poor quality of work, lack of initiative, lateness, missing deadlines, misuse of resources, and absenteeism) which they rated with a 5-points scale ranging from 1 (My organization never ignores) to 5 (My organization always ignore). These behaviors fall in the

“production deviance” category of Robinson & Bennett’s (1995) typology of deviant workplace behaviors. After conducting factor analysis to certify that all items significantly load in a single factor (Figure 1), we used the average of the seven items as our measure of perceived organization’s failure to address poor performance15.

Job satisfaction. We used the three items from Wright & Pandey (2008):

“In general, I like working here,” “In general, I don't like my job,” and “All in all, I am satisfied with my job.” These items were rated using a 5-points scale ranging from 1

(Strongly disagree) to 5 (Strongly agree). These three items were averaged to form our

PSM measure.

15 We randomized the presentation order of the question about organizations’ failure to address poor performance. Half of participants responded to this question at the beginning of the survey and the other half at the end. With this procedure we tried to manipulate (via priming) participants’ perceptions of “my organization ignores poor performance.” However, that priming did not have any effect on participants’ job satisfaction. Independently of that, to be transparent about our procedures, we included the presentation order as a control variable and tested its interaction with the other predictor variables, so that we could control for any confounding effect that might have been created by the presentation order of this question. 107

Public service motivation. We measured PSM with the 5-items scale used by

Wright & Pandey (2008). The items are “Meaningful public service is very important to me,” “I am often reminded by daily events about how dependent we are on one another,”

“Making a difference in society means more to me than personal achievements,” “I am prepared to make enormous sacrifices for the good of society,” and “I am not afraid to go to bat for the rights of others, even if it means that I will be ridiculed.” These items were rated using a 5-points scale ranging from 1 (Strongly disagree) to 5 (Strongly agree). The five items were averaged to form our PSM measure.

Control variables. We controlled for participants’ demographic characteristics like age, gender, tenure, and educational level. As participants worked in different areas of the public sector, to avoid the confounding effects of job type, we also controlled by the Ministry (or Department) they were subordinated to (Defense, Education, and

Economy). We created a dummy variable for each of these Ministries/Departments.

Results

Confirmatory factor analysis (CFA) revealed that the measurement model (Figure

1) has an adequate goodness-of-fit: χ2(87) = 206.42, p < 0.001, RMSEA = .08, SRMR

= .06, CFI = .90. A model in which all items load in a single factor showed a significantly worse goodness-of-fit ( χ2(90) = 529.67, p < 0.001, RMSEA = .15, SRMR = .12, CFI =

.64), which supports the discriminant validity of the variables.

108

Figure 1. Measurement Model

Table 1. Descriptive Statistics and Intercorrelations

Mean SD 1 2 3 4 5 6 7 8 9 10 1 Job Satisfaction 4.05 0.99 2 Fail. to Address 2.44 0.92 -.28*** 3 PSM 3.91 0.67 .41*** -.26*** 4 Tenure 12.98 9.16 .14* -.07 0.16* 5 Age 42.60 9.90 .23*** -.21** 0.11 .51*** 6 Education Level 3.01 0.90 .00 .17* -.01 .06 0.15* 7 Gender 1.42 - .16* -0.14* .12 -.15* .12 .12 8 Min. of Defense 0.32 - -.08 .01 .17** .03 -.36** -.23*** -.31*** 9 Min. of Economy 0.22 - -.09 .08 -.21** .08 .11 .22*** -.06 -.36*** 10 Min. Education 0.30 - .06 -.12 .02 -.17* .18** .03 .37*** -.44*** -.34*** 11 Present. Order 0.48 - -.06 .11 -.03 .07 -.06 .03 -.13 .08 .11 -.07 Notes. N = 217. *p < .05 **p < .01 ***p < .001. Educational Level: 1=elementary school, 2=high school, 2=college, 3=post- graduation, 4=master's degree, 5=PhD. Gender: 1=male, 2=female. Presentation order (of question about organizations ignoring poor performance): before=1, after=0.

Table 1 presents the descriptive statistics and correlations for the variables in the

study. We tested our hypotheses with OLS regression (Table 2). Supporting hypothesis

1, we found that perceived organization’s failure to address poor performance has an 109

average negative effect on job satisfaction (b = -0.17, SE = 0.07, p < .05; Model 2).

Hypothesis 2 predicted that employees with greater PSM would have their job satisfaction more negatively affected by the perception that their organizations systematically ignore poor performance. Confirming this hypothesis, we found a significant PSM by “failure to address” interaction (b = -0.26, SE = 0.10, p < .05; Model 3). Regarding this moderation effect, Figure 2 shows that a high level of job satisfaction is only achieved when both

PSM is high and employees perceive that the organization does not ignore poor performance.

The simple main effect of ignoring poor performance when PSM was 1 SD below average was not significant (b = -0.01, SE = 0.10, p = .93), showing that low-PSM individuals are not negatively affected by the perception that their organizations ignore poor performance. On the other hand, when PSM was 1 SD above the mean, the simple main effect of ignoring poor performance was negative and significant (b = -0.37, SE =

0.10, p < .01).

We conducted some robustness checks to verify whether the presentation order of the survey question about organization’s failure to address poor performance could have affected the results (see footnote 3). According to Model 4 and 5 (Table 2), presentation order did not have a direct effect on job satisfaction, nor did it interact with our study variables. It shows that that our results were not affected by the order participants were presented that question.

110

Table 2. Regression Results

Robustness Checks/ Main Analyses Additional Analyses Model Model Model Model Model Model Independent Variables 1 2 3 3 4 5

Study Variables: -0.17* -0.19** -0.17* -0.18* -0.11 Failure to address (0.07) (0.07) (0.07) (0.07) (0.07) 0.51*** 0.50*** 0.50*** 0.48*** 0.52*** PSM (0.09) (0.09) (0.09) (0.09) (0.09) -0.26* -0.26* -.35** PSM X Failure to address (0.10) (0.10) (0.11) Failure to address X -0.17* Failure to address (0.07)

Control Variables:

0.01 0.00 0.01 0.00 0.00 0.00 Tenure (0.01) (0.01) (0.01) (0.01) (0.01) (0.01)

0.02* 0.02 0.01 0.01 0.01 0.01 Age (0.01) (0.01) (0.01) (0.01) (0.01) (0.01)

-0.03 -0.01 0.00 0.00 0.00 0.00 Educational level (0.07) (0.07) (0.07) (0.07) (0.07) (0.07)

0.33* 0.15 0.17 0.17 0.17 0.18 Gender (0.14) (0.14) (0.14) (0.14) (0.14) (0.14) -0.13 -0.28 -0.30 -0.28 -0.28 -0.30 Ministry of Defense (0.20) (0.20) (0.20) (0.20) (0.20) (0.20) -0.35 -0.20 -0.22 -0.20 -0.20 -0.22 Ministry of Economy (0.21) (0.20) (0.20) (0.20) (0.20) (0.20) -0.21 -0.19 -0.24 -0.22 -0.22 -0.18 Ministry of Education (0.20) (0.20) (0.20) (0.20) (0.20) (0.20) -0.04 0.00 -0.03 0.00 0.00 -0.05 Presentation order (0.13) (0.13) (0.12) (0.06) (0.07) (0.12)

Present. order X Failure -0.05 -0.07

to address. (0.07) (0.07)

0.02 Presentation order X PSM (0.10)

Present. order X Fail. to 0.11 address X PSM (0.11)

Constant -1.07 -0.55 -0.58 -0.57 -0.63 -0.50 R2 0.06 0.20 0.22 0.20 0.22 0.24 Notes. N = 217. *p < .05 **p < .01 ***p < .001. Standard errors in parenthesis. Non-standardized coefficients are presented. Study variables are mean-centered. Gender: Male=1, female=2. Presentation order (of questions about organizations ignoring poor performance): before =1, after=0.

111

Figure 2. Interaction between PSM and Organization’s Failure to Address Poor Performance

Notes. Predicted Job Satisfaction based on Model 3 (Table 2).

Additional Analysis. Although our results show an average negative effect of perceived organization’s failure to address poor performance on job satisfaction, our intuition is that when organizations are overly harsh in the way they deal with counterproductive behaviors, it could also lead lower job satisfaction. By exerting too strict control over employees’ behaviors and by using severe forms of discipline frequently, organizations can both undermine employees’ sense of autonomy (Deci &

Ryan, 2008) and psychological safety (Edmondson & Lei, 2014), which can also negatively affect their job satisfaction. Then, we anticipated that the relationship between ignoring poor performance and job satisfaction might not be linear but follow an inverted

U-shape. To test that, we included a quadratic term for “Failure to address” in the regression (see Model 6, Table 2).

112

Figure 3. Non-linear Effects of Organization’s Failure to Address Poor Performance by PSM Level

Notes. Predicted Job Satisfaction based on Model 6 (Table 2).

Model 6 (Table 2) shows that this quadratic term’s effect is negative and significant (b = -0.17*, SE = 0.07, p < .05), which confirms the inverted U-shape. Figure

3 presents the non-linear relationship between organization’s failure to address poor performance and job satisfaction at different levels of PSM. It shows that for participants with lower PSM (dashed line), after achieving an optimal level of job satisfaction near the median point of the “Failure to address poor performance” scale, the more the organization addresses poor performance—i.e., the closer it gets to the value of 1 (“My organization never ignores”)—the more they get dissatisfied. For participants with high

PSM, however, the perception that the organization addresses poor performance more frequently seems to be more consistently related to greater jobs satisfaction, although after a certain point (around the value of 2: “My organization rarely ignores”) the effect becomes null (see Figure 3). 113

Discussion

In this study, we explored how public employees’ job satisfaction can be affected by the perception that their organizations systematically ignore counterproductive employee behaviors, like poor quality of work, low productivity, lateness, missing deadlines, etc. People holding a more “theory X” view of public employees (i.e., a view that they are lazy and not concerned with their organization’s goals; see Frank & Lewis, 2004; McGregor, 1960) would expect them to be generally satisfied for working in an organization that is not much concerned about performance.

However, our results show that this is not the case. Our analyses revealed that, on average, public employees respond negatively (in terms of job satisfaction) to their organizations’ systematic inattention to poor performance. And we found that this negative effect was significantly stronger for employees with higher levels of PSM. Further analyses of the data revealed that only 4% of participants had such a low level of PSM that would make them react positively to their organizations’ failure to address poor performance.

Based on these findings, we believe that the mentioned FEVS’ results (those indicating that most public employees perceive that their organizations are not taking effective steps to deal with poor performers) should be viewed with concern. By ignoring poor performance, public organizations may be losing not only the contributions of uncommitted employees, but also of potentially high-performing employees with greater

PSM—those who could make a bigger difference for the organization and society. And although our study has concentrated only on employees’ attitudinal responses to organizations’ failure to address poor performance, it should be noted that there are other possible negative social effects of ignoring poor performers. From a social learning standpoint (Bandura, 1977), Treviño (1992) argues that by observing deviant behaviors 114

go unpunished, employees may become more inclined to replicate these behaviors.

Indeed, the vicarious effects of coworker deviance have been confirmed by empirical studies (see Robinson et al., 2014).

Theoretically, this work contributes to the public administration literature by further clarifying the relationship between employee PSM and job satisfaction. Previous studies have explored mainly the role of PSM as an antecedent of job satisfaction (PSM as a predictor). Our study, however, shows that PSM can also affect how employees make sense and respond attitudinally to their organizations’ practices (PSM as a moderator).

Like Quratulain and Khan (2015), we found that PSM exacerbates employees’ negative responses to deficient organizational practices, although we would not consider that a

“dark side” of PSM. When it comes to the relationship between PSM and work performance, it seems that public employees with greater PSM not only have higher performance standards for themselves (Ritz et al., 2016); our results indicate that they also have high expectations of their organizations setting a high standard of performance for all employees. And when these expectations are not met, they may experience frustration and dissatisfaction.

Our results also seem to help reconcile some conflicting arguments regarding the effectiveness of workplace discipline. On the one hand, studies stress the importance of addressing deviant members disciplinarily in order to maintain a climate of justice, clarify roles and goals, and protect the group’s values (Podsakoff et al., 2010; Treviño, 1990).

On the other hand, some studies highlight the potential side effects of using punishment, especially regarding the negative reactions of recipients (anger, sabotage, etc.; Atwater et al., 2001, Butterfield, Treviño, & Ball, 1996). As illustrated by Atwater et al., (2001), “the disciplinee may stop coming to work late, but may have negative attitudes toward the job 115

and/or the supervisor as a result of the disciplinary event. Perhaps as a need to maintain self-esteem or ‘save face,’ disciplinees will harbor resentment, and spread such feelings to other employees.” Moreover, organizations can harm employees’ sense of autonomy

(Deci & Ryan, 2008) and psychological safety (Edmondson & Lei, 2014) by excessively controlling their behaviors and by using more strict forms of discipline too frequently.

Although authors have already recognized the tension between these two perspectives (Atwater et al., 2001) and suggested that future research could investigate the existence of an “optimal point” regarding the use of discipline (Treviño & Weaver,

2010), to our knowledge, no study has explored this issue. We believe that our results showing that the relationship between addressing counterproductive behaviors and job satisfaction follows an inverted U-shape (Figure 3) helps clarify that both perspectives may be correct and that an optimal point may exist. And the finding that the level of employee PSM changes the pattern of this U-shaped relationship (i.e., the higher the employee PSM, the more that optimal point approaches the situation in which the organization never ignores counterproductive behaviors) also helps shed light on these paradoxical effects. Given that organizations will probably want to prioritize the job satisfaction of their most committed members (i.e., those with greater PSM), it seems that the optimal point—at least in terms of job satisfaction—will be somewhere between

"rarely ignores poor performance" and "never ignores poor performance" (see Figure 3).

Based on the results discussed above, we believe this study has an important practical implication. Public managers should be advised about the negative social effects of ignoring employees who intendedly dodge their responsibilities and perform poorly.

However, while we stress the importance of considering these effects, we recognize that addressing poor performers is a complex and challenging supervisory task (Daley, 2008). 116

In this regard, the literature offers strategies and guidelines that may help managers promote the intended changes in behavior of deviant employees (and minimize their negative reactions) while satisfying third-party observers (Brett, Atwater, & Waldman,

2005; Bugdol, 2018; Podsakoff et al., 2010). Also, offering training to public managers about how to deal effectively with difficult employees seems to be an indispensable strategy to prepare them to perform such a difficult task successfully (Daley, 2008;

Franklin & Pagan, 2006).

Limitations and Future Research

We first tackle the methodological limitations of this study. As informed in the

Measures section, we randomized the presentation order of the questions about organization’s failure to address poor performance to try to manipulate participants’ perceptions of “my organization ignores poor performance” (i.e., a priming technique).

A recent study used a similar procedure to manipulate PSM (Meyer-Sahling, Mikkelsen,

& Schuster, 2018). However, the presentation order did not affect any of our study variables. Then, we used the survey questions about organization’s failure to address poor performance as our independent variable. Although the CFA and regression analyses showed that this measure worked well—and although we have controlled for some potentially confounding variables, including PSM—we cannot fully rule out alternative explanations for our results. We believe that future research can use experimental designs based on hypothetical scenarios and vignettes in order to further test our hypotheses.

Another limitation of this work concerns the empirical investigation of the psychological mechanisms underlying our hypotheses. Although we theorized about the mediating role of both value congruence and perceived inequity, we did not test these 117

effects. Future studies can empirically explore these (and other) psychological mechanisms, as well as other outcome variables. Lastly, in this study we did not ask participants to specify what disciplinary actions their organizations used to address counterproductive employee behaviors (we asked only the frequency these behaviors were ignored/addressed). We believe that future research may add to our study by conducting a more detailed analysis of how specific disciplinary actions may affect employees’ perceptions and attitudes differently. It would be helpful investigating which managerial strategies for dealing with poor performance can be most successful in producing the highest level of positive attitudes and behaviors for all organizational members.

118

References

Adams, J. S. (1963). Toward an understanding of inequity. Journal of Abnormal and Social Psychology, 67, 422-436. Ashforth, B. E., Harrison, S. H., & Corley, K. G. (2008). Identification in organizations: An examination of four fundamental questions. Journal of Management, 34(3), 325-374. Atwater, L. E., Waldman, D. A., Carey, J. A., & Cartier, P. (2001). Recipient and observer reactions to discipline: Are managers experiencing wishful thinking? Journal of Organizational Behavior, 22(3), 249-270. Ball, G. A., Treviño, L. K., & Sims Jr, H. P. (1992). Understanding subordinate reactions to punishment incidents: Perspectives from justice and social affect. The Leadership Quarterly, 3(4), 307-333. Ban, C. (1995). How do public managers manage: Bureaucratic constraints, organizational culture, and potential for reform. San-Francisco: Jossey-Bass. Bandura, A. (1977). Social learning theory. Englewood Cliffs, NJ: Prentice Hall. Baumeister, R. F., Bratslavsky, E., Finkenauer, C., & Vohs, K. D. (2001). Bad is stronger than good. Review of General Psychology, 5, 323-370. Bellé, N. (2013). Experimental evidence on the relationship between public service motivation and job performance. Public Administration Review, 73(1), 143-153. Brett, J. F., Atwater, L. E., & Waldman, D. A. (2005). Effective delivery of workplace discipline: do women have to be more participatory than men? Group & Organization Management, 30(5), 487-513. Bugdol, M. (2018). A Different Approach to Work Discipline. Cham: Springer. Butterfield, K. D., Treviño, L. K., & Ball, G. A. (1996). Punishment from the manager's perspective: A grounded investigation and inductive model. Academy of Management Journal, 39(6), 1479-1512. Butterfield, K. D., Treviño, L. K., Wade, K. J., & Ball, G. A. (2005). Organizational punishment from the manager's perspective: An exploratory study. Journal of Managerial Issues, 17(3) 363-382. Boxx, W. R., Odom, R. Y., & Dunn, M. G. (1991). Organizational values and value congruency and their impact on satisfaction, commitment, and cohesion: An empirical examination within the public sector. Public Personnel Management, 20(2), 195-205. Brewer, G. A., & Selden, S. C. (1998). Whistle blowers in the federal civil service: New evidence of the public service ethic. Journal of Public Administration Research and Theory, 8(3), 413-440. Camacho, M. L., & Paulus, P. B. (1995). The role of social anxiousness in group brainstorming. Journal of Personality and Social Psychology, 68(6), 1071-1080. Choi, D. L. (2004). Public service motivation and ethical conduct. International Review of Public Administration, 8(2), 99-106. 119

Cohen, S. G., & Bailey, D. E. (1997). What makes teams work: Group effectiveness research from the shop floor to the executive suite. Journal of Management, 23(3), 239-290. Cohen-Charash, Y., & Spector, P. E. (2001). The role of justice in organizations: A meta- analysis. Organizational Behavior and Human Decision Processes, 86(2), 278- 321. Christensen, R. K., & Wright, B. E. (2018). Public service motivation and ethical behavior: Evidence from three experiments. Journal of Behavioral Public Administration, 1(1), 1-8. Cropanzano, R., & Ambrose, M. L. (Eds.). (2015). The Oxford handbook of justice in the workplace. Oxford: Oxford University Press. Christensen, R. K., & Wright, B. E. (2011). The effects of public service motivation on job choice decisions: Disentangling the contributions of person-organization fit and person-job fit. Journal of Public Administration Research and Theory, 21(4), 723-743. Daley, D. M. (2008). The burden of dealing with poor performers: Wear and tear on supervisory organizational engagement. Review of Public Personnel Administration, 28(1), 44-59. Darley, J. M., & Pittman, T. S. (2003). The psychology of compensatory and retributive justice. Personality and Social Psychology Review, 7(4), 324-336. Deci, E.L. and Ryan, R.M., 2008. Self-determination theory: A macrotheory of human motivation, development, and health. Canadian Psychology/Psychologie Canadienne, 49(3), 182-185. Deng, H., & Leung, K. (2014). Contingent punishment as a doubleedged sword: A dualpathway model from a sensemaking perspective. Personnel Psychology, 67(4), 951-980. Dukerich, J. M., Golden, B. R., & Shortell, S. M. (2002). Beauty is in the eye of the beholder: The impact of organizational identification, identity, and image on the cooperative behaviors of physicians. Administrative Science Quarterly, 47(3), 507-533. Dutton, J. E., Dukerich, J. M., & Harquail, C. V. (1994). Organizational images and member identification. Administrative Science Quarterly, 39(2), 239-263. Edmondson, A. C., & Lei, Z. (2014). Psychological safety: The history, renaissance, and future of an interpersonal construct. Annu. Rev. Organ. Psychol. Organ. Behav., 1(1), 23-43. Edwards, J. R., & Cable, D. M. (2009). The value of value congruence. Journal of Applied Psychology, 94(3), 654-677. Felps, W., Mitchell, T. R., & Byington, E. (2006). How, when, and why bad apples spoil the barrel: Negative group members and dysfunctional groups. Research in Organizational Behavior, 27, 175-222. Frank, S. A., & Lewis, G. B. (2004). Government employees: Working hard or hardly working? The American Review of Public Administration, 34(1), 36-51. 120

Franklin, A. L., & Pagan, J. F. (2006). Organization culture as an explanation for employee discipline practices. Review of Public Personnel Administration, 26(1), 52-73. Gaertner, G. H., & Gaertner, K. N. (1984). Formal disciplinary action in two federal agencies. Review of Public Personnel Administration, 5(1), 12-24. Goodhew, G. W., Cammock, P. A., & Hamilton, R. T. (2008). The management of poor performance by front-line managers. Journal of Management Development, 27(9), 951-962. Greer, C. R., & Labig Jr, C. E. (1987). Employee reactions to disciplinary action. Human Relations, 40(8), 507-524. Guffey, C. J., & Helms, M. M. (2001). Effective employee discipline: A case of the Internal Revenue Service. Public Personnel Management, 30(1), 111-127. Hackman, J. R. (1987). The design of work teams. In J. Lorsch (Ed.), Handbook of organizational behavior (pp. 315–342). Englewood Cliffs, NJ: Prentice-Hall. Herzberg, F. (1968). One More Time: How Do You Motivate Employees? Harvard Business Review, 46(1), 53-62. Hinkin, T. R., & Schriesheim, C. A. (2008). An examination of "nonleadership:" From laissez-faire leadership to leader reward omission and punishment omission. Journal of Applied Psychology, 93(6), 1234-1248. Holtz, B. C., & Hu, B. (2017). Passive leadership: relationships with trust and justice perceptions. Journal of Managerial Psychology, 32(1), 119-130. Hondeghem, A., & Perry, J. L. (2009). EGPA symposium on public service motivation and performance. International Review of Administrative Sciences, 75(1), 5-97. Hoffman, B. J., & Woehr, D. J. (2006). A quantitative review of the relationship between person–organization fit and behavioral outcomes. Journal of Vocational Behavior, 68(3), 389-399. Hogan, R., & Emler, N. P. (1981). Retributive Justice. In M. J. Lerner & S. C. Lerner (Eds.), The justice motive in social behavior (pp. 125-143). New York: Plenum. Hung, T. K., Chi, N. W., & Lu, W. L. (2009). Exploring the relationships between perceived coworker loafing and counterproductive work behaviors: The mediating role of a revenge motive. Journal of Business and Psychology, 24(3), 257-270. Ilgen, D. R., Hollenbeck, J. R., Johnson, M., & Jundt, D. (2005). Teams in organizations: From input-process-output models to IMOI models. Annual Review of Psychology, 56, 517–543. Judge, T. A., & Piccolo, R. F. (2004). Transformational and transactional leadership: a meta-analytic test of their relative validity. Journal of Applied Psychology, 89(5), 755-768. Kearney, R. C., & Hays, S. W. (Eds.). (1983). Public personnel administration: Problems and prospects. Englewood Cliffs: Prentice-Hall. Keyton, J. (1999). Analyzing interaction patterns in dysfunctional teams. Small Group Research, 30(4), 491–518. 121

Kim, S. (2012). Does personorganization fit matter in the publicsector? Testing the mediating effect of personorganization fit in the relationship between public service motivation and work attitudes. Public Administration Review, 72(6), 830- 840. King, K. N., & Wilcox, D. E. (2003). Employee-proposed discipline: How well is it working? Public Personnel Management, 32(2), 197-209. Koehler, M. & Rainey, H. G. (2008). Interdisciplinary Foundations of Public Service Motivation. In J. Perry and A. Hondeghem (Eds.), Motivation in Public Management: The Call of Public Service (pp. 33-55). Oxford: Oxford University Press. Kohlberg, L. (1984). Essays on moral development: Vol. 2. The psychology of moral development. San Francisco: Harper & Row. Kozlowski, S. W. L., & Bell, B. S. (2003). Work groups and teams in organizations. In W. C. Borman, D. R. Ilgen & R. J. Klimoski (Eds), Handbook of psychology: Industrial and organizational psychology, (Vol. 12, pp. 333–375). London: Wiley. Kristof-Brown, A. L., Zimmerman, R. D., & Johnson, E. D. (2005). Consequences of individuals’ fit at work: A metaanalysis of person–job, person–organization, person–group, and person–supervisor fit. Personnel Psychology, 58, 281-342. Liden, R. C., Wayne, S. J., Judge, T. A., Sparrowe, R. T., Kraimer, M. L., & Franz, T. M. (1999). Management of poor performance: A comparison of manager, group member, and group disciplinary decisions. Journal of Applied Psychology, 84(6), 835. Liu, B., Tang, T. L. P., & Yang, K. (2015). When does public service motivation fuel the job satisfaction fire? The joint moderation of person–organization fit and needs– supplies fit. Public Management Review, 17(6), 876-900. McGregor, D. (1960), The human side of enterprise. New York: McGraw-Hill. Meglino, B. M., Ravlin, E. C., & Adkins, C. L. (1989). A work values approach to corporate culture: A field test of the value congruence process and its relationship to individual outcomes. Journal of Applied Psychology, 74(3), 424-443. Meyer-Sahling, J. H., Mikkelsen, K. S., & Schuster, C. (2018). The Causal Effect of Public Service Motivation on Ethical Behavior in the Public Sector: Evidence from a Large-Scale Survey Experiment. Journal of Public Administration Research and Theory, 29(3), 445-459. Moynihan, D.P.; Pandey, S.K.; & Wright, B.E. (2014). Transformational leadership in the public sector: Empirical evidence of its effects. In Y.K. Dwivedi, M.A. Shareef, S.K. Pandey, & V. Kumar (Eds.), Public administration reformation: Market demand from public organizations (pp. 87–104). New York: Routledge/Taylor & Francis Niehoff, B. P., Paul, R. J., & Bunch, J. F. (1998). The social effects of punishment events: The influence of violator past performance record and severity of the punishment on observers' justice perceptions and attitudes. Journal of Organizational Behavior, 19(6), 589-602. 122

O'Reilly, C. A., & Puffer, S. M. (1989). The impact of rewards and punishments in a social context: A laboratory and field experiment. Journal of Occupational Psychology, 62(1), 41-53. Perry, J. L., & Vandenabeele, W. (2015). Public service motivation research: Achievements, challenges, and future directions. Public Administration Review, 75(5), 692-699. Perry, J. L., & Wise, L. R. (1990). The motivational bases of public service. Public Administration Review, 367-373. Podsakoff, P. M., Bommer, W. H., Podsakoff, N. P., & MacKenzie, S. B. (2006). Relationships between leader reward and punishment behavior and subordinate attitudes, perceptions, and behaviors: A meta-analytic review of existing and new research. Organizational Behavior and Human Decision Processes, 99(2), 113- 142. Podsakoff, N. P., Podsakoff, P. M., & Kuskova, V. V. (2010). Dispelling misconceptions and providing guidelines for leader reward and punishment behavior. Business Horizons, 53(3), 291-303. Podsakoff, P. M., & Todor, W. D. (1985). Relationships between leader reward and punishment behavior and group processes and productivity. Journal of Management, 11(1), 55-73. Podsakoff, N. P., Whiting, S. W., Podsakoff, P. M., & Blume, B. D. (2009). Individual- and organizational-level consequences of organizational citizenship behaviors: A meta-analysis. Journal of Applied Psychology, 94(1), 122-141. Rho, E., Yun, T., & Lee, K. (2015). Does organizational image matter? Image, identification, and employee behaviors in public and nonprofit organizations. Public Administration Review, 75(3), 421-431. Ritz, A., Brewer, G. A., & Neumann, O. (2016). Public service motivation: A systematic literature review and outlook. Public Administration Review, 76(3), 414-426. Quratulain, S., & Khan, A. K. (2015). Red tape, resigned satisfaction, public service motivation, and negative employee attitudes and behaviors: Testing a model of moderated mediation. Review of Public Personnel Administration, 35(4), 307- 332. Robinson, S. L., Wang, W., & Kiewitz, C. (2014). Coworkers behaving badly: The impact of coworker deviant behavior upon individual employees. Annu. Rev. Organ. Psychol. Organ. Behav., 1(1), 123-143. Robinson, S. L., & Bennett, R. J. (1995). A typology of deviant workplace behaviors: A multidimensional scaling study. Academy of Management Journal, 38(2), 555- 572. Schnake, M. E., & Dumler, M. P. (1989). Some unconventional thoughts on the use of punishment in organizations: reward as punishment and punishment as reward. Journal of Social Behavior and Personality, 4(1), 97-107. Selden, S. C. (2006). The impact of discipline on the use and rapidity of dismissal in state governments. Review of Public Personnel Administration, 26(4), 335-355. 123

Skarlicki, D. P., O’Reilly, J., & Kulik, C. T. (2015). The Third-Party Perspective of (In)justice. In R. Cropanzano & M. L. Ambrose (Eds.), The Oxford handbook of justice in the workplace (pp. 235-255). Oxford: Oxford University Press. Skitka, L. J. (2003). Of different minds: An accessible identity model of justice reasoning. Personality and Social Psychology Review, 7(4), 286-297. Stewart, G. L., Manz, C. C., & Sims, H. P., Jr. (1999). Team work and group dynamics. New York: Wiley. Stritch, J. M., & Christensen, R. K. (2014). Looking at a Job's Social Impact Through PSMTinted Lenses: Probing the Motivation–Perception Relationship. Public Administration, 92(4), 826-842. Trahan, W. A., & Steiner, D. D. (1994). Factors affecting supervisors' use of disciplinary actions following poor performance. Journal of Organizational Behavior, 15(2), 129-139. Treviño, L. K. (1992). The social effects of punishment in organizations: A justice perspective. Academy of Management Review, 17(4), 647-676. Treviño, L. K., & Ball, G. A. (1992). The social implications of punishing unethical behavior: Observers' cognitive and affective reactions. Journal of Management, 18(4), 751-768. Treviño, L. K., & Weaver, G. R. (2010). Advances in research on punishment in organizations: Descriptive and normative perspectives. In M Schminke (Ed.), Managerial ethics: Managing the psychology of morality (pp. 177-204). New York: Routledge. van den Bos, K., Lind, E. A., Vermunt, R., & Wilke, H. A. (1997). How do I judge my outcome when I do not know the outcome of others? The psychology of the fair process effect. Journal of Personality and Social Psychology, 72(5), 1034-1046. Wang, L., & Murnighan, J. K. (2017). The dynamics of punishment and trust. Journal of Applied Psychology, 102(10), 1385-1402. Wright, B. E., Moynihan, D. P., & Pandey, S. K. (2012). Pulling the levers: Transformational leadership, public service motivation, and mission valence. Public Administration Review, 72(2), 206-215. Wright, B. E., & Pandey, S. K. (2008). Public service motivation and the assumption of person—Organization fit: Testing the mediating effect of value congruence. Administration & Society, 40(5), 502-521. York, J. W. (2018). Strengthening the Federal Workforce Through Increased Accountability. Heritage Foundation Backgrounder, 3325, 1-17. Yukl, G. (2013), Leadership in Organizations. Englewood Cliffs: Prentice Hall.

124

GENERAL CONCLUSION

The three articles of this dissertation reveal that followers' perceptions of a leader's behaviors and values have a strong effect on the way they respond to that leader, which can ultimately affect their job attitudes and behaviors. Importantly, common these articles, is the finding that the values shared by followers greatly impact the way they make sense of their leaders’ and organizations’ practices. In the first article, we demonstrated that perceived authenticity is highly predictive of leader emergence in groups, especially when the target individual embodies and exemplifies the group’s shared values. In the second article, we show the importance of public leaders being committed to public values (values that are likely to be shared by subordinates in the public sector), so they can develop more inspiring and engaging forms of leadership. In the third study, we found that when civil servants—especially those more committed to public service values—perceive that their organizations and leaders fail to address poor performing coworkers, their job satisfaction is negatively impacted. By combining the results of these three articles, we believe that leaders need to be highly aware of the values shared by followers and try to connect—authentically—with these values in order to promote higher levels of employee motivation and performance.