Introduction
With the rise of telecommuting technology and flexible work arrangements, particularly during the COVID-19 pandemic, employees are increasingly relying on digital communication platforms for work instead of traditional face-to-face interactions (Szurawitzki, Reference Szurawitzki2022; Yang, Murad, Mirza, Chaudhary, & Saeed, Reference Yang, Murad, Mirza, Chaudhary and Saeed2022). In China, there are over 540 million online workers, representing a remarkable increase of 56% since 2020 (China Internet Network Information Center, 2023). As a result, leaders and followers increasingly rely on digital communication apps like WeChat, which boasts more than 1.3 billion active monthly users in China, to facilitate timely and efficient work-related communication.
Compared to traditional communication methods, digital platforms offer interactions characterized by increased frequency, variety, vividness, and promptness (Erle, Schmid, Goslar, & Martin, Reference Erle, Schmid, Goslar and Martin2022; Imran, Shahzad, Butt, & Kantola, Reference Imran, Shahzad, Butt and Kantola2020). Notably, a key feature of communication on these digital platforms is the ability to express emotions through the use of emojis (e.g., 
, 
, 
, and 
), compensating for the absence of nonverbal cues (Beattie, Edwards, & Edwards, Reference Beattie, Edwards and Edwards2020; Boutet, LeBlanc, Chamberland, & Collin, Reference Boutet, LeBlanc, Chamberland and Collin2021). Originally coined in Japan, the term ‘emoji’ combines ‘e’ for ‘picture’ and ‘moji’ for ‘letter’ (Sugiyama, Reference Sugiyama2015).Footnote 1 Emojis encompass various pictorial expressions, such as a smiling face 
or a flower 
. Emojis have been recognized as the primary nontextual cues in virtual communication on digital platforms (Arafah & Hasyim, Reference Arafah and Hasyim2019; Fischer & Herbert, Reference Fischer and Herbert2021), with over 6 billion emojis being sent daily through communication apps and 90% of online users frequently using them (Swyft Media, 2018).
Considering the prevalence of emoji usage in virtual communication at work, understanding the potential impact of leader emoji usage in virtual communication is crucial for organizations striving to foster effective leadership in virtual contexts and maintain strong relationships with their employees. However, while research on virtual leadership has shed light on various important topics, such as the effectiveness of traditional leadership styles (e.g., transformational leadership and empowering leadership) in virtual settings (e.g., Banks, Dionne, Mast, & Sayama, Reference Banks, Dionne, Mast and Sayama2022; Hoch & Kozlowski, Reference Hoch and Kozlowski2014) and leader social media usage (e.g., Ernst et al., Reference Ernst, Banks, Loignon, Frear, Williams, Arciniega, Gupta, Kodydek and Subramanian2022; Grant, Hodge, & Sinha, Reference Grant, Hodge and Sinha2018), there has been limited investigation into the impact of leaders’ emoji usage on employees in virtual communication (for recent reviews, see Banks et al., Reference Banks, Dionne, Mast and Sayama2022; Höddinghaus, Nohe, & Hertel, Reference Höddinghaus, Nohe and Hertel2023).
Concurrently, scholars in the fields of communication, linguistics, and psychology have shown increasing interest in studying the effects of emoji usage in virtual communication (e.g., Ferré, Haro, Pérez-Sánchez, Moreno, & Hinojosa, Reference Ferré, Haro, Pérez-Sánchez, Moreno and Hinojosa2022; Logi & Zappavigna, Reference Logi and Zappavigna2023; Stark & Crawford, Reference Stark and Crawford2015). Emojis come in different types and carry distinct meanings (Algharabali & Taqi, Reference Algharabali and Taqi2018; Riordan, Reference Riordan2017). Among them, smile emojis (e.g., 
) are the most commonly used emojis in virtual communication (Edelman & van Knippenberg, Reference Edelman and van Knippenberg2017; Wantong, Reference Wantong2021). Smile emojis refer to emojis that contain smiling faces with smiling eyes in order to express pleasure, kindness, and positive feelings (Bai, Dan, Mu, & Yang, Reference Bai, Dan, Mu and Yang2019; Ebel & Dutra, Reference Ebel, Duta, Hanlon and Tuten2022; Sun, Lasser, & Lee, Reference Sun, Lasser and Lee2023). Smile emojis are typically depicted with upward-curving mouths and vary in style from simple grins to more detailed expressions with additional features like eyes, teeth, and rosy cheeks. A general consensus in the existing literature is that using smile emojis is beneficial for interpersonal communication (Aritajati & Rosson, Reference Aritajati and Rosson2021; Smith & Rose, Reference Smith and Rose2020), including expressing friendliness (Ernst & Huschens, Reference Ernst and Huschens2019), fostering a more harmonious communication atmosphere (Pricilla, Lestari, & Dharma, Reference Pricilla, Lestari and Dharma2018), and building positive relationships (Beattie et al., Reference Beattie, Edwards and Edwards2020). What remains understudied, however, is whether the benefits of using smile emojis in virtual communication hold true in the leader–follower communication context and whether there are potential negative impacts in this context.
To address these limitations in the literature, this research aims to focus on the leader–follower virtual communication context and explore the potential impact of leader smile emoji usage on followers’ perception of leaders. Following previous research (Edelman & van Knippenberg, Reference Edelman and van Knippenberg2017; Wantong, Reference Wantong2021), we define leader smile emoji usage as leaders' use of smile emojis in digital communication with followers. To explore its impact on followers, we draw on expectancy violations theory (EVT; Burgoon, Reference Burgoon, M. E. Roloff, J. P. Dillard and D. Solomon2015; Burgoon & Le Poire, Reference Burgoon and Le Poire1993), which is an interpersonal communication theory. EVT suggests that individuals hold certain expectations about appropriate and inappropriate behavior from the other party during interpersonal communication. When these expectations are violated, it leads to uncertainty and prompts the recipient to evaluate the violation and the communicator (Burgoon & Hale, Reference Burgoon and Hale1988). EVT further suggests that expectation violations are not always negative (i.e., falling short of expectations) and can also be positive (i.e., exceeding expectations), leading to positive evaluations from the recipients (Campo, Cameron, Brossard, & Frazer, Reference Campo, Cameron, Brossard and Frazer2004; White, Reference White, Baxter and Braithwaite2008). Moreover, whether a violation is perceived as positive or negative is dependent on social norms associated with the communication context (Burgoon, Reference Burgoon, M. E. Roloff, J. P. Dillard and D. Solomon2015; Burgoon & Le Poire, Reference Burgoon and Le Poire1993).
Based on EVT, we posit that leaders’ smile emoji usage may violate followers’ expectations in different ways, leading to different impacts on followers’ perception of the leaders. Moreover, the strength of these impacts is dependent on follower power distance orientation, which captures followers’ understanding of the social norms in the leader–follower interaction context (Auh, Menguc, Spyropoulou, & Wang, Reference Auh, Menguc, Spyropoulou and Wang2016; Farh, Hackett, & Liang, Reference Farh, Hackett and Liang2007). Power distance orientation describes the extent to which an individual accepts the unequal distribution of power in their organizations (Auh et al., Reference Auh, Menguc, Spyropoulou and Wang2016; Farh et al., Reference Farh, Hackett and Liang2007; Kirkman, Chen, Farh, Chen, & Lowe, Reference Kirkman, Chen, Farh, Chen and Lowe2009; Lian, Ferris, & Brown, Reference Lian, Ferris and Brown2012). Followers with different levels of power distance orientation hold different expectations for their leaders in leader–follower interactions, which may impact their perception of their leaders’ use of smile emoji.
Specifically, followers hold two different sets of expectations in their interactions with leaders. On one hand, followers usually expect leaders to maintain authority and distance when interacting with them (Eagly, Wood, & Diekman, Reference Eagly, Wood, Diekman, Eckes and Trautner2000; Offermann, Kennedy, & Wirtz, Reference Offermann, Kennedy and Wirtz1994; Subramaniam, Othman, & Sambasivan, Reference Subramaniam, Othman and Sambasivan2010). Leaders’ use of smile emojis may violate these expectations and make their followers feel that they fail to demonstrate the authority and seriousness expected of an authority figure, and express vague information. Such a negative violation of expectations likely undermines followers’ perceptions of the leaders’ power and, in turn, leads to lower perceptions of leader effectiveness. Moreover, these negative impacts are likely stronger for followers with higher power distance orientation, who tend to hold higher expectations of their leaders to maintain authority and distance when interacting with followers (Kirkman et al., Reference Kirkman, Chen, Farh, Chen and Lowe2009; Li, He, Yam, & Long, Reference Li, He, Yam and Long2015) and are thus more likely to view leaders’ use of smile emoji as inappropriate, representing a strong negative violation of their expectations.
On the other hand, followers also usually expect that leaders build good relationships with them (Bass, Reference Bass1990; Offermann et al., Reference Offermann, Kennedy and Wirtz1994). Leaders’ use of smile emojis may exceed, that is, positively violate these expectations of followers regarding the closeness between them and their leaders in communication. Such a positive violation likely has positive impacts on followers’ perception of leader intimacy and, in turn, leads to higher satisfaction with their leaders. Moreover, these positive impacts are likely stronger for employees with lower power distance orientation, who tend to view leaders and followers as equals and are thus less likely to view leaders’ use of smile emoji as inappropriate and more likely to view it as an unexpected positive violation.
In sum, based on EVT, we propose that leader smile emoji usage is positively related to follower satisfaction with the leader through perceived leader intimacy, while it is negatively related to follower perception of leader effectiveness via decreased perception of leader power. Moreover, follower power distance orientation moderates the effects of leader smile emoji usage on follower perceptions of their leaders. Our theoretical model is depicted in Figure 1.
Theoretical model
Our study makes several important contributions to the current literature on smile emoji and leadership. First, we contribute to the literature on leadership by taking an initial step to explore how leaders’ use of emojis affects follower outcomes. Specifically, while the literature on leadership in the virtual context is fruitful and growing rapidly (Banks et al., Reference Banks, Dionne, Mast and Sayama2022; Höddinghaus et al., Reference Höddinghaus, Nohe and Hertel2023), it has yet to explore how leaders’ linguistic characteristics, such as emoji usage, in virtual communication affect follower outcomes. This research is among the first to explore leader smile emoji usage and its effect on follower outcomes, expanding our understanding of the complex dynamics of leadership in the digital age.
Second, we contribute to the literature on smile emojis by challenging the prevailing conclusion that using smile emoji in virtual communication is always beneficial for interpersonal interaction (Edelman & van Knippenberg, Reference Edelman and van Knippenberg2017; Ge & Gretzel, Reference Ge and Gretzel2018; Riordan, Reference Riordan2017). We propose a theoretical model that uncovers when and why leader smile emoji usage leads to positive versus negative judgments from followers. By doing so, we offer a more balanced and comprehensive understanding of the impact of smile emoji usage in virtual communication.
Third, this research also extends EVT. EVT suggests that a behavior can violate the recipient’s expectations in either a positive or negative way (Burgoon, Reference Burgoon, M. E. Roloff, J. P. Dillard and D. Solomon2015; Burgoon & Le Poire, Reference Burgoon and Le Poire1993). However, the current literature on EVT has largely overlooked the possibility that a behavior can simultaneously violate the other party’s expectations in both positive and negative aspects. Our study proposes and reveals that leader smile emoji usage can lead to both positive and negative violations of followers’ expectations of their leaders. By uncovering this nuanced dynamic, our findings expand the theoretical underpinnings of EVT.
Theoretical Background and Hypotheses
Leader Smile Emoji and EVT
Traditional personality traits and cognitions, such as extraversion and self-efficacy, which can prompt specific leadership in face-to-face conversation, may have a much weaker influence in the virtual context (Charlier, Stewart, Greco, & Reeves, Reference Charlier, Stewart, Greco and Reeves2016; Matthews, Matthews, Wang, & Kelemen, Reference Matthews, Matthews, Wang and Kelemen2022; Serban et al., Reference Serban, Yammarino, Dionne, Kahai, Hao, McHugh, Sotak, Mushore, Friedrich and Peterson2015). Hence, leaders need to embrace the digital era and possess new communication characteristics. Previous research on leader virtual communication has mainly been conducted in situations such as telephone, email, and online meetings (e.g., Brodsky, Reference Brodsky2021; Maruping & Agarwal, Reference Maruping and Agarwal2004; Newman, Ford, & Marshall, Reference Newman, Ford and Marshall2020), which suggests that virtual communication lacks nonverbal cues compared to face-to-face communication, leading to information loss and difficulties in communication between leaders and followers (Banks et al., Reference Banks, Dionne, Mast and Sayama2022). For example, Bennett, Campion, Keeler, and Keener's (Reference Bennett, Campion, Keeler and Keener2021) study found that the lack of nonverbal cues in videoconference meetings increases employees’ cognitive processing load and consequently increases their fatigue. However, the emergence of emojis, which contain visual elements (e.g., colorful and vivid faces 
,
,
,
,
), complements the lack of nonverbal cues (Tang & Hew, Reference Tang and Hew2019; Tigwell & Flatla, Reference Tigwell and Flatla2016; Zhou, Hentschel, & Kumar, Reference Zhou, Hentschel and Kumar2017), and offers new avenues for leaders to provide more enriched information in virtual communication.
Emojis provide social cues by displaying intuitive iconography which helps to enrich visual representation and expression of thoughts in virtual communication (Cappallo, Mensink, & Snoek, Reference Cappallo, Mensink and Snoek2015). According to media richness theory (Ernst et al., Reference Ernst, Banks, Loignon, Frear, Williams, Arciniega, Gupta, Kodydek and Subramanian2022; Grant et al., Reference Grant, Hodge and Sinha2018), using smile emojis might be helpful in facilitating communication since it increases the richness of information. Most current research has explored emoji usage, especially smile usage, in the field of mass communication (e.g., Sampietro, Reference Sampietro2020), social psychology (e.g., Marengo, Giannotta, & Settanni, Reference Marengo, Giannotta and Settanni2017), and marketing (e.g., Ge & Gretzel, Reference Ge and Gretzel2018). Past research in general found positive effects of smile emojis and revealed three important functions of using smile emojis (Al Rashdi, Reference Al Rashdi2018; Beattie et al., Reference Beattie, Edwards and Edwards2020; Kaye, Wall, & Malone, Reference Kaye, Wall and Malone2016; Lu, Reference Lu2019): (1) expressing positive emotions, (2) softening message tone, and (3) reducing misunderstanding, such as complaints or hostility. Based on these functions, scholars found that the use of smile emojis has a positive impact on building a good impression (Ernst & Huschens, Reference Ernst and Huschens2019; Pricilla et al., Reference Pricilla, Lestari and Dharma2018), increasing the likelihood of receiving a reply (Aritajati & Rosson, Reference Aritajati and Rosson2021; Stein, Reference Stein2023), and strengthening relationships (Chang, Reference Chang2022; Smith & Rose, Reference Smith and Rose2020).
However, our understanding of the impact of leader smile emoji usage is still limited. As one of the first studies to explore the influence of leader emoji usage, we focus on leader smile emoji use. There are more than ten different types of smile emojis (e.g., 
,
,
) on a typical digital platform, which have nuanced effects and meanings (Miller et al., Reference Miller, Thebault-Spieker, Chang, Johnson, Terveen and Hecht2016). Taking 
as an example, past research found that this smile emoji is ambiguous; some people regard it as ‘positive’ while others think it expresses ‘ironic’ and ‘despising’ meanings (Huang, Reference Huang2017; Shi, Reference Shi2022; Violaris, Reference Violaris2020). Research also found that the interpretation of smile emojis varies among different digital platforms (e.g., Android and IOS) (Tigwell & Flatla, Reference Tigwell and Flatla2016). Thus, as one of the first research to explore the effect of leader smile emoji usage, we focus on positive smile emojis in general instead of a specific type of smile emojis.
We draw upon EVT and propose the dual-path model of the impact of leader smile emoji usage on followers. Developed by Burgoon and Jones (Reference Burgoon and Jones1976), EVT initially focused on proxemics violations, which involve the interpretation of space and distance during interpersonal interactions. Over the past four decades, EVT has expanded to encompass nonverbal behaviors and has been applied in various contexts (Burgoon, Reference Burgoon, M. E. Roloff, J. P. Dillard and D. Solomon2015), including workplace interpersonal communication (e.g., Heyden, Gu, Wechtler, & Ekanayake, Reference Heyden, Gu, Wechtler and Ekanayake2023; Wayne, Sun, Kluemper, Cheung, & Ubaka, Reference Wayne, Sun, Kluemper, Cheung and Ubaka2023). EVT differentiates between positive and negative violations, suggesting that positive violations can yield favorable outcomes while negative violations lead to undesirable outcomes (Burgoon & Jones, Reference Burgoon and Jones1976). Different from previous research context of smile emoji usage, there is a natural power gap and different social role expectations in the leader–follower communication context (Antonakis & Atwater, Reference Antonakis and Atwater2017; Hollander & Offermann, Reference Hollander and Offermann1990). The same leader behavior may negatively violate one aspect of follower expectations while positively violating another aspect. The degree of negative or positive violation is dependent on follower perception of related social norms in the leader–follower communication context, such as power distance. Thus, EVT provides a useful framework for understanding whether, how, and when leader smile emoji usage affects followers’ perception of their leaders in positive or negative ways.
Leader Smile Emoji Usage and Perceived Leader Intimacy
Perceived intimacy refers to a sense of connectedness and sharing toward another individual (Kark, Reference Kark, Cameron and Spreitzer2012). It is a personal experience that emerges from an interpersonal, transactional process characterized by self-disclosure and the other person's responsiveness (Laurenceau, Rivera, Schaffer, & Pietromonaco, Reference Laurenceau, Rivera, Schaffer, Pietromonaco, Mashek and Aron2004). Followers tend to perceive a high level of leader intimacy when their leaders are willing to share their emotions, disclose about themselves, show empathy, and express the intention to build a close relationship with them (Ellis, Carmon, & Pike, Reference Ellis, Carmon and Pike2016; Kark, Reference Kark, Cameron and Spreitzer2012; Laurenceau, Barrett, & Rovine, Reference Laurenceau, Barrett and Rovine2005). Past research has highlighted that a distinguishing feature of emojis is to express intimate intention (Elder, Reference Elder2018; Gesselman, Ta, & Garcia, Reference Gesselman, Ta and Garcia2019; Stein, Reference Stein2023). As noted by Rodrigues, Cavalheiro, and Prada (Reference Rodrigues, Cavalheiro and Prada2022), emojis serve as icebreakers to convey intimate sentiment. Based on EVT, we argue that leader smile emoji usage in virtual communication with follower is positively related to follower perceived leader intimacy.
EVT suggests that, in interpersonal communication, positive expectation violations lead to more desirable reactions from recipients than expectation confirmations (Burgoon, Reference Burgoon, M. E. Roloff, J. P. Dillard and D. Solomon2015; Burgoon & Le Poire, Reference Burgoon and Le Poire1993). In the leader–follower communication context, followers typically expect that leaders should build good relationships with followers (e.g., Bass, Reference Bass1990; Offermann et al., Reference Offermann, Kennedy and Wirtz1994). In contrast to pure texts, the use of smile emojis confirms such expectations. Moreover, leaders’ use of smile emojis, compared with other types of emojis (e.g., anger emojis), further signals affinity and affection toward followers (Fischer & Herbert, Reference Fischer and Herbert2021; Pricilla et al., Reference Pricilla, Lestari and Dharma2018). Thus, leader smile emoji usage not only confirms but also surpasses followers’ expectations regarding the formal relationship and emotional tone of communication between them and their leaders. Such a positive violation of expectations, in turn, is likely to result in followers perception of intimacy with the leaders.
Moreover, smile emoji (e.g., 
) can mitigate illocutionary force in conversations and reduce the perceived rudeness of the interlocutor (Dresner & Herring, Reference Dresner and Herring2010; Skovholt, Reference Skovholt and Darics2015; Wantong, Reference Wantong2021). Hence, leaders’ smile emoji usage helps cultivate an unexpectedly pleasant and comfortable communication atmosphere and signals that they are willing to engage with followers and have a sense of camaraderie, leading to higher follower-perceived leader intimacy. In addition, using smile emojis to express emotions is a common self-disclosure approach in virtual communication (Tang & Hew, Reference Tang and Hew2019). Previous research has shown that self-disclosure can improve mutual understanding (Ickes, Reference Ickes1997) and lead to perceived approachability and intimacy (Janssen, Ijsselsteijn, & Westerink, Reference Janssen, Ijsselsteijn and Westerink2014). In line with our arguments, past research also provides evidence for the relationship between smile emoji usage and intimacy perception (e.g., Coyle & Carmichael, Reference Coyle and Carmichael2019; Griggio, Sato, Mackay, & Yatani, Reference Griggio, Sato, Mackay and Yatani2021; Konrad, Herring, & Choi, Reference Konrad, Herring and Choi2020). Thus, we propose that:
Hypothesis 1 (H1): Leader smile emoji usage is positively related to follower perceived leader intimacy.
Perceived Leader Intimacy and Leader Satisfaction
We further suggest that perceived leader intimacy is positively related to leader satisfaction. Leader satisfaction refers to the extent to which followers are satisfied with the work relationships with their leaders (Gerstner & Day, Reference Gerstner and Day1997; Grover, Reference Grover2014; Medcalf, Reference Medcalf2008). Humans have the innate need for connectedness and belonging (Baumeister & Leary, Reference Baumeister and Leary1995; Deci & Ryan, Reference Deci and Ryan2000). Followers who perceive a high level of leader intimacy feel connected with their leaders (Kark, Reference Kark, Cameron and Spreitzer2012). By fostering a sense of closeness and shared understanding, leader intimacy helps meet followers’ need for connectedness and belonging and thus leads to higher leader satisfaction. In addition, followers who have more intimate relationships with leaders are more likely to experience open and constructive communication with their leaders, which may foster a sense of partnership and lead to greater satisfaction with the leader. Furthermore, intimacy helps develop shared identity between followers and leaders, leading followers to view their leaders as members of the same social category, which will promote satisfaction with the leader (Ashforth & Mael, Reference Ashforth and Mael1989).
Based on these arguments, we propose that perceived leader intimacy is positively related to leader satisfaction. Moreover, as we expect that leader usage of smile emoji is positively related to follower perceived leader intimacy, we propose that:
Hypothesis 2 (H2): Leader smile emoji usage has a positive indirect relationship with leader satisfaction through perceived leader intimacy.
Leader Smile Emoji Usage and Perceived Leader Power
Drawing upon EVT, we further argue that leader smile emoji usage is negatively related to follower perceived leader power. We define perceived leader power based on the psychological perspective of power (Anderson, John, & Keltner, Reference Anderson, John and Keltner2012; Ju et al., Reference Ju, Huang, Liu, Qin, Hu and Chen2019). This perspective suggests that power is not simply ‘the control over resources or composed solely of one's social position’ but also ‘a psychological state – a perception of one's capacity to influence others’ (Anderson et al., Reference Anderson, John and Keltner2012: 314; Galinsky, Gruenfeld, Magee, & Dovidio, Reference Galinksy, Gruenfeld, Magee and Dovidio2003). In line with this perspective, previous research has shown that, although leaders have more formal power than followers because of their leadership position, followers vary in their perception of how powerful their leaders are (Anderson & Berdahl, Reference Anderson and Berdahl2002; Hoogervorst, De Cremer, Van Dijke, & Mayer, Reference Hoogervorst, De Cremer, Van Dijke and Mayer2012; Magee, Gruenfeld, Keltner, & Galinsky, Reference Magee, Gruenfeld, Keltner, Galinsky, Messick and Kramer2005). Thus, following research on the psychological perspective of power (e.g., Anderson et al., Reference Anderson, John and Keltner2012; Ju et al., Reference Ju, Huang, Liu, Qin, Hu and Chen2019), we define perceived leader power as followers’ perception of their leader’s ability to influence them. Followers are likely to perceive high leader power when their leaders act in ways that show authority and credibility as leaders who are authoritative and credible are more likely to be influential in their interactions with followers.
In the workplace, the leader–follower relationship is generally considered as relatively formal (Glikson, Cheshin, & Van Kleef, Reference Glikson, Cheshin and Van Kleef2018; Hurlburt, Reference Hurlburt2018; Schaubroeck & Shao, Reference Schaubroeck and Shao2012). Followers are likely to expect their leaders to adhere to formal communication norms as a way of demonstrating their authority (Eagly et al., Reference Eagly, Wood, Diekman, Eckes and Trautner2000; Offermann et al., Reference Offermann, Kennedy and Wirtz1994; Subramaniam et al., Reference Subramaniam, Othman and Sambasivan2010). Leaders’ use of smile emoji, however, is likely to lead to a negative violation of these expectations. Specifically, using emojis, as compared with pure text, is generally viewed as informal and likely violates followers’ expectations for leaders to demonstrate authority. Such negative violations may lead to doubts about the leaders’ authority, which in turn leads to reduced perception of leader power. In line with these arguments, previous research has shown that using emoji in relatively formal contexts (e.g., work email, professional communication) may be viewed as inappropriate and less competent (Cavalheiro, Prada, Rodrigues, Lopes, & Garrido, Reference Cavalheiro, Prada, Rodrigues, Lopes and Garrido2022; Glikson et al., Reference Glikson, Cheshin and Van Kleef2018; Tigwell & Flatla, Reference Tigwell and Flatla2016). Moreover, using smile emojis can undermine followers’ expectations for clear communication with their leaders, as these symbols often convey vague information. In addition, some research has shown that inappropriate emotional expression in a particular context results in recipients’ doubts about and challenges to the communicators’ competence (Glikson et al., Reference Glikson, Cheshin and Van Kleef2018; Van Kleef, Reference Van Kleef2016), which is closely related to credibility.
Furthermore, smile emojis can distract the focus on the core message in conversations (Konrad et al., Reference Konrad, Herring and Choi2020). When leaders use smile emojis in their conversations with followers, followers are less likely to interpret the leaders’ messages, including those related to work arrangements, as commands (Dresner & Herring, Reference Dresner and Herring2010). In this case, followers’ perception of leaders’ authority over them and influence on them are likely to be reduced, leading to a decrease in perceived leader power. In line with this, Schaubroeck and Shao (Reference Schaubroeck and Shao2012) proposed that leader emotion expression might be regarded as lacking control over the event, sending a signal of a lack of power. Thus, we suggest that:
Hypothesis 3 (H3): Leader smile emoji usage is negatively related to follower perceived leader power.
Perceived Leader Power and Leader Effectiveness
We further propose that perceived leader power is positively related to perceived leader effectiveness. Leader effectiveness refers to followers’ evaluation of how their leaders are capable of completing tasks including creating positive effect and guiding activities associated with their units toward desired goals (Hu, Jiang, Huang, & Liu, Reference Hu, Jiang, Huang and Liu2022; Paustian-Underdahl, Walker, & Woehr, Reference Paustian-Underdahl, Walker and Woehr2014). High power is consistent with the implicit expectations of followers for a prototypical and effective leader (Hoogervorst et al., Reference Hoogervorst, De Cremer, Van Dijke and Mayer2012).
First, perceived leader power reflects the extent to which followers believe that their leaders are able to influence others. Thus, followers who perceive their leaders to be powerful are more likely to believe that their leaders can make impactful decisions and direct team members’ efforts toward team goals and thus view their leaders as legitimate and effective (Antonakis & Day, Reference Antonakis and Day2017; Mooijman, van Dijk, van Dijk, & Ellemers, Reference Mooijman, van Dijk, van Dijk and Ellemers2019). Second, followers who perceive high leader power are also more likely to have confidence in their leaders’ ability to gain resources and overcome obstacles. Such confidence and trust in their leaders may make followers more likely to accept their leaders’ decisions, strategies, and actions, thereby enhancing their perceived leader effectiveness (Breevaart & Zacher, Reference Breevaart and Zacher2019; Norman, Avolio, & Luthans, Reference Norman, Avolio and Luthans2010). Third, implicit leadership theory (Offermann et al., Reference Offermann, Kennedy and Wirtz1994) suggests that being powerful is an important characteristic of the prototype of an effective leader (Abdalla & Al-Homoud, Reference Abdalla and Al-Homoud2001; Offermann & Coats, Reference Offermann and Coats2018). Leaders with high power are more likely to be considered persuasive and charismatic (Yukl & Falbe, Reference Yukl and Falbe1991) which helps them to be more effective in leading programs (Kipnis, Schmidt, & Wilkinson, Reference Kipnis, Schmidt and Wilkinson1980). As Barksdale (Reference Barksdale2008: 22) noted, ‘it may be necessary for a leader to manifest his power bases to establish a positive reputation of overall leadership effectiveness’. Moreover, previous research has also linked perceived leader power and leader effectiveness (Anderson & Brown, Reference Anderson and Brown2010; Dhar & Mishra, Reference Dhar and Mishra2001; Tost, Gino, & Larrick, Reference Tost, Gino and Larrick2013). Based on these arguments and H3, we propose that:
Hypothesis 4 (H4): Leader smile emoji usage has a negative indirect effect on leader effectiveness through perceived leader power.
The Moderating Role of Power Distance Orientation
EVT further suggests that whether a behavior is seen as a positive or negative violation by recipients is dependent on the recipients’ perceived social norms in the communication context (Burgoon, Reference Burgoon, M. E. Roloff, J. P. Dillard and D. Solomon2015; Burgoon & Le Poire, Reference Burgoon and Le Poire1993). A key individual difference that determines follower perception of the social norm in the leader–follow communication context is power distance orientation (Auh et al., Reference Auh, Menguc, Spyropoulou and Wang2016; Farh et al., Reference Farh, Hackett and Liang2007). Power distance orientation refers to the extent to which an individual accepts the unequal distribution of power in their organizations (Hofstede, Reference Hofstede1980; Kirkman et al., Reference Kirkman, Chen, Farh, Chen and Lowe2009; Lian et al., Reference Lian, Ferris and Brown2012). Based on EVT, we argue that follower power distance orientation moderates the effect of leader smile emoji usage on the follower perception of leader intimacy and leader power.
Followers with high power distance orientation tend to show great respect and deference to their leaders (Farh et al., Reference Farh, Hackett and Liang2007; Song, Gu, Wu, & Xu, Reference Song, Gu, Wu and Xu2019). They attach great importance to leader’s authority and tend to expect that leaders should be professional and authoritative (Kirkman et al., Reference Kirkman, Chen, Farh, Chen and Lowe2009; Schaubroeck, Lam, & Cha, Reference Schaubroeck, Lam and Cha2007), and keep social distance from followers (Farh et al., Reference Farh, Hackett and Liang2007; Tyler, Lind, & Huo, Reference Tyler, Lind and Huo2000). Although smile emojis show friendliness, past research suggests that emoji usage is considered informal and unprofessional (Hurlburt, Reference Hurlburt2018; Skovholt, Reference Skovholt and Darics2015). As such, followers with high power distance orientation are likely to view leaders’ use of smile emojis as inappropriate, representing a negative violation of their expectations of leaders (Lian et al., Reference Lian, Ferris and Brown2012). Consequently, leader usage of smile emojis is less appealing and more unacceptable to these followers. In this case, the positive effect of leader smile emoji usage on follower perceived leader intimacy is likely to be weakened while the negative effect of leader smile emoji usage on follower perceived leader power is likely to be strengthened.
In contrast, followers with lower power distance orientation tend to view leaders and followers as equals and expect to develop more equal relationships with their leaders (Daniels & Greguras, Reference Daniels and Greguras2014; Lin, Chen, Tse, Wei, & Ma, Reference Lin, Chen, Tse, Wei and Ma2019; Yu, Reference Yu2017). Although they may not expect leaders to use smile emojis with them in digital communications, they are less likely to find it inappropriate if leaders do so and more likely to feel that it exceeds their expectations. In other words, for these followers, leaders’ use of smile emojis is less likely to be viewed as negative violation and more likely to be viewed as positive violation. In this case, the positive effect of leader smile emoji usage on follower perceived leader intimacy is likely to be strengthened while the negative effect of leader smile emoji on follower perceived leader power is likely to be weakened. Based on these arguments, we propose the following hypotheses:
Hypothesis 5 (H5): Follower power distance orientation moderates the positive relationship between leader smile emoji usage and follower perceived leader intimacy, such that the positive relationship is weaker when follower power distance orientation is high rather than low.
Hypothesis 6 (H6): Follower power distance orientation moderates the negative relationship between leader smile emoji usage and follower perceived leader power, such that the negative relationship is stronger when follower power distance orientation is high rather than low.
Thus far, we have proposed that leader smile emoji usage is positively related to follower satisfaction with leader via perceived leader intimacy (H1 and H2) and negatively related to follower perceived leader effectiveness via perceived leader power (H3 and H4). Furthermore, we have proposed that these relationships are moderated by follower power distance (H5 and H6). Integrating these hypotheses, we propose two moderated mediation hypotheses. Through proposing and testing these moderated mediation hypotheses, we aim to explore how (i.e., the mediating effect of follower perceived leader intimacy and leader power) and when (i.e., the moderating effect of follower power distance orientation) leader usage of smile emoji is associated with followers’ positive or negative perception of the leaders. Thus, we propose the following hypotheses:
Hypothesis 7 (H7): Power distance orientation moderates the indirect effect of leader smile emoji usage on leader satisfaction through perceived leader intimacy, such that the positive effect is weaker when power distance orientation is high rather than low.
Hypothesis 8 (H8): Power distance orientation moderates the indirect effect of leader smile emoji usage and leader effectiveness, such that the negative relationship is stronger when power distance orientation is high rather than low.
Research Overview
We conducted two studies to test our theoretical model. Study 1 was an experiment in which we tested the causal effect of leader smile emoji usage on follower perception of leader intimacy and leader power. Study 2 was a field survey study in which we collected time-lagged data from employees to test our whole model. By adopting this multimethod design, we aim to strengthen the internal and external validity of this research. These two datasets are used solely for the purposes of this research, and the variables and samples within the data have not been used in our previously published research. We have updated the research material, data, code, and supplementary analysis results of these two studies on the open science framework (OSF) website (https://osf.io/zkgjc/?view_only=4c9732d66d8c4aeaa1f2b3d51d2fb196).
Study 1 Methods
Sample and Procedure
We used G*Power to determine the sample size and the results showed that 231 participants were required to achieve 90% power. In order to surpass this requirement, we recruited 300 full-time employees through Credamo, a widely used data collection platform (e.g., Gong et al., Reference Gong, Lu, Schaubroeck, Li, Zhou and Qian2020; Tang, Zhang, & Ning, Reference Tang, Zhang and Ning2023). This research was pre-registered on AsPredicted (https://aspredicted.org/K89_JQC). After excluding two participants who failed the attention-check item (i.e., ‘please choose strongly agree’), the final sample consisted of 298 participants. Among them, 63.76% were females and 76.17% had a college degree. The average age was 31.50 (SD = 7.15) and the average tenure was 5.60 years (SD = 4.87). Each participant was compensated 5 RMB for their time (approximately 0.7 USD).
Participants first rated their power distance orientation. Then, to manipulate leader smile emoji usage, we randomly assigned participants to one of two conditions: the leader smile emoji usage condition (n = 150) or the control condition (n = 148). Participants were first required to read a scenario and assume the role of a team member. Specifically, they were asked to imagine that they were a newcomer named Wang in Hua Xing Electronic Company. Their directed leader was Liu. Wang usually used WeChat to communicate with Liu. Then, they were presented with a phone screenshot showing that Liu sent a message to Wang in WeChat.
In the leader smile emoji usage condition, participants were presented with a phone screenshot (see Figure 2) showing that Liu sent a message with smile emojis to Wang. In the control condition, participants were presented with a similar phone screenshot (see Figure 2) which showed that Liu sent the same message without smile emoji. Following the scenario, the participants rated perceived leader intimacy, perceived leader power, leader satisfaction, and leader effectiveness toward Liu. Finally, they completed the measure of manipulation check and reported their demographics.
Materials in leader smile emoji usage condition and control condition
Note: The message in the screenshot is ‘Liu: Wang, could you please prepare the material for the project presentation next week, thank you’.
Measures
For the manipulation check, participants were required to choose whether the leader used smile emojis in the message (0 = ‘Yes’, 1 = ‘No’). All participants chose the correct answer and passed the manipulation check. All scales were measured using a 5-point Likert scale unless otherwise specified (1 = ‘Strongly disagree’, 5 = ‘Strongly agree’).
Power distance orientation
We measured power distance orientation using a three-item scale developed by Lee, Pillutla, and Law (Reference Lee, Pillutla and Law2000). A sample item is ‘When a performance appraisal made by the supervisor does not align with subordinates’ expectations, the employees should feel free to discuss it with the supervisor’ (reverse-coded item) (α = 0.70).
Perceived leader intimacy
Perceived leader intimacy was measured with four items based on the research of Sinclair and Dowdy (Reference Sinclair and Dowdy2005). Participants were asked about the extent to which Liu (the leader in the scenario), ‘would be willing to share thoughts and feeling with me’, and ‘would be willing to accept me as who I am’ (α = 0.79).
Perceived leader power
We measured perceived leader power with four items used by Ju et al. (Reference Ju, Huang, Liu, Qin, Hu and Chen2019). These items asked participants the extent to which Liu, ‘would get me to listen to what he says’, and ‘if he wants to, he would get to make the decisions’ (α = 0.84).
Leader satisfaction
Leader satisfaction was measured with a three-item scale adapted from the Michigan Organizational Assessment Questionnaire (Cammann, Fichman, Jenkins, & Klesh, Reference Cammann, Fichman, Jenkins, Klesh, Seashore, Lawler, Mirvis and Camman1983). A sample item is ‘All in all, I am satisfied with my leader (Liu)’ (α = 0.83).
Leader effectiveness
We measured leader effectiveness using a three-item scale from Douglas and Ammeter (Reference Douglas and Ammeter2004). A sample item is ‘This leader (Liu) is effective in meeting the job-related needs of work unit members’ (α = 0.71).
Study 1 Results
Table 1 reports the descriptive statistics and correlations among key variables. We conducted OLS to test our hypotheses, and the results are presented in Table 2.
Means, standard deviations, and bivariate correlations in Study 1
Notes: n = 298. M, mean; SD, standard deviation. Interaction term was calculated by multiplying leader smile usage and centered power distance orientation. For leader smile emoji usage: 1 = smile emoji condition, 0 = control condition. *p < 0.05, **p < 0.01.
Results of regression analyses in Study 1
Notes: n = 298. Robust standard errors are shown in parentheses. *p < 0.05; **p < 0.01.
Consistent with H1, leader smile emoji usage was positively related to perceived leader intimacy (b = 0.53, SE = 0.07, p < 0.001) in Model 1 of Table 2. Furthermore, perceived leader intimacy was positively associated with leader satisfaction (b = 1.15, SE = 0.05, p < 0.001) in Model 5 of Table 2. Leader smile emoji usage had a significant indirect effect on leader satisfaction through perceived leader intimacy (indirect effect = 0.61, 95%CI = [0.46, 0.77]). Thus, H2 received support.
As shown in Model 3 and Model 7 in Table 2, leader smile emoji usage was negatively related to perceived leader power (b = −0.21, SE = 0.10, p = 0.03), and perceived leader power was positively associated with leader effectiveness (b = 0.39, SE = 0.04, p < 0.001). The results showed that leader smile emoji usage has a significant indirect effect on leader effectiveness through perceived leader power (indirect effect = −0.08, 95%CI = [−0.16, −0.01]). Thus, H3 and H4 were supported.
Model 2 of Table 2 indicated that the interaction of leader smile emoji usage and power distance orientation related negatively to perceived leader intimacy (b = −0.22, SE = 0.08, p = 0.01). We further plotted this interaction in Figure 3 and the simple slope test revealed that leader smile emoji usage was more positively related to leader intimacy (b = 0.70, t = 7.72, p < 0.001) when power distance orientation was low (−1 SD) than when power distance orientation was high (+1 SD) (b = 0.36, t = 3.99, p < 0.001). Thus, H5 was supported.
The moderating effect of power distance orientation in the relationship between leader smile emoji usage and perceived leader intimacy in Study 1
The interaction of leader smile emoji usage and power distance orientation was negatively related to perceived leader power (b = −0.33, SE = 0.12, p < 0.01) in Model 4 of Table 2. We graphed this interaction effect in Figure 4. A simple slope test showed that the relationship between leader smile emoji usage and perceived leader power was significantly negative when power distance orientation was high (b = −0.46, t = −3.47, p = 0.001), while the relationship was not significant when power distance orientation was low (b = 0.05, t = 0.40, p = 0.69). Thus, H6 was supported.
The moderating effect of power distance orientation in the relationship between leader smile emoji usage and perceived leader power in Study 1
Furthermore, the indirect effect of leader smile emoji usage on leader satisfaction via perceived leader intimacy was also more positive when follower power distance orientation was low (indirect effect = 0.81, 95%CI = [0.60, 1.03]) than the high condition (indirect effect = 0.42, 95%CI = [0.21, 0.63]). The difference between these indirect effects was also significant (Δindirect effect = 0.39, 95%CI = [0.10, 0.69]). The indirect effect of leader smile emoji usage on leader effectiveness via perceived leader power was negative when follower power distance orientation was high (indirect effect = −0.18, 95%CI = [−0.30, −0.08]) but nonsignificant when power distance orientation was low (indirect effect = 0.02, 95%CI = [−0.08, 0.13]). The difference between these indirect effects was significant (Δindirect effect = 0.21, 95%CI = [0.06, 0.37]). The results provided support for H7 and H8.Footnote 2 To enhance the external validity of our findings, we conducted a field study in Study 2 and examined our full model again.
Study 2 Methods
Sample and Procedures
We recruited participants through the alumni network of a university in east China. Participants were asked to report whether they used WeChat to communicate with their direct leaders regularly and all chose yes, meeting the criteria of our study. Research assistants invited them to a WeChat group of participants. In this group, we explained that the aim of the research was to explore employee attitudes at workplace. We also assured anonymity and told all participants that they could voluntarily withdraw at any time. Online questionnaire links were sent to the participants through the WeChat group at two time points. At Time 1, 282 participants reported power distance orientation, perceived leader intimacy, and perceived leader power. After that, they were asked to review their communication with their direct leaders in WeChat during the last week and count the number of smile emojis (their leaders sent to them). One month later, at Time 2, 255 participants rated leader satisfaction and leader effectiveness. We used the WeChat ID to match the data collected at two time points. To encourage engagement, participants were compensated with 10 RMB (about 1.46 USD) after completing each survey.
After matching the data across two waves, we obtained a final sample of 225 employees. Among them, 62.67% were females, and 79.56% had a bachelor's degree. They had an average age of 25.60 (SD = 5.86). They were from various industries, including information technology (20.89%), manufacturing (14.67%), education and training (14.22%), energy (14.22%), service (12.89%), and others (23.11%). The results of response analysis showed that there were no significant differences in demographics between our final sample and those who dropped of the analyses (p > 0.05).
Measures
Following Brislin's (Reference Brislin, Triandis and Berry1980) translation–back translation procedure, we translated all English scales into Mandarin Chinese. One research assistant majoring in English first translated the English scales to Mandarin Chinese, and then another research assistant translated it back into English. Authors of this paper discussed the differences between the translated version and the original version, and improved the translation. Unless otherwise specified, all measures were rated using a 5-point Likert-type scale ranging from 1 (Strongly disagree) to 5 (Strongly agree).
Leader smile emoji usage
As we mentioned above, we asked participants to calculate the number of positive smile emojis their leader sent to them last week via WeChat communication. We provided the definition of positive smile emojis and listed some examples (
).Footnote 3 The number of the leader smile emojis used ranged from 0 to 50, with an average number of 5.75 (SD = 5.95). To strengthen the robustness of our empirical findings, we also measured the frequency of leader smile emoji usage directly. We asked participants to rate how frequently their leaders used smile emojis in their communication with them (1 = never, 5 = very often). The objective measure and the subjective measure had a high correlation (r = 0.42, p < 0.001). More importantly, the pattern of our hypothesis-testing results remained unchanged when we used the subjective measure of leader smile emoji usage to replace the objective measure in the analyses. Full results can be found in Appendix D on the OSF.
Power distance orientation
We measured power distance orientation using the same scale as in Study 1 (α = 0.71).
Perceived leader intimacy
We used the same scale as in Study 1 to measure perceived leader intimacy (α = 0.75).
Perceived leader power
We used the same scale as in Study 1 to measure perceived leader power (α = 0.70).
Leader satisfaction
Leader satisfaction was measured using the same scale as in Study 1 (α = 0.79).
Leader effectiveness
We measured leader effectiveness using the same scale as in Study 1(α = 0.70).
Control variables
We controlled for leader age (in years) and gender (0 = female, 1 = male) since previous research suggested that older leaders are considered to have higher power and prestige than younger ones (Buengeler, Homan, & Voelpel, Reference Buengeler, Homan and Voelpel2016) and male leaders might be regarded as more powerful than female leaders, who might be regarded as more communal (Diekman, Goodfriend, & Goodwin, Reference Diekman, Goodfriend and Goodwin2004; Rosette & Tost, Reference Rosette and Tost2010). In addition, we controlled for leader–follower dyadic tenure (years) since it might influence follower perception of leader competence as well as followers’ relationship with the leader. Our results remained stable after excluding all control variables.
Analytic Strategy
We used ordinary least squares (OLS) and path model in STATA 15.0 to test our hypotheses. The continuous explanatory variables were grand-mean centered to reduce collinearity (Hofmann & Gavin, Reference Hofmann and Gavin1998). The interaction terms were calculated using the mean-centered independent variable and moderator. We further used the Monte Carlo simulation (Selig & Preacher, Reference Selig and Preacher2008) to calculate confidence intervals (CIs) to test the indirect effects (Edwards & Lambert, Reference Edwards and Lambert2007).
Study 2 Results
We first conducted confirmatory factor analysis (CFA) to test the discriminant validity of our measures. The results showed that our hypothesized five-factor model (power distance orientation, perceived leader intimacy, perceived leader power, leader satisfaction, and leader effectiveness) had a better model fit than other alternative models (χ 2[109] = 196.38, p < 0.001; comparative fit index [CFI] = 0.90; standardized root mean-square residual [SRMR] = 0.06, root mean square error of approximation [RMSEA] = 0.06), providing support for the discriminant validity of our measures.
Table 3 presents descriptive statistics and correlations of study variables. The results of the OLS analysis are shown in Table 4. H1 and H2 predict the influence of leader smile emoji usage on follower perceived leader intimacy and leader satisfaction. As shown in Model 1 and Model 5 in Table 4, leader smile emoji usage had a significant positive relationship with perceived leader intimacy (b = 0.03, SE = 0.01, p < 0.001) which, in turn, was significantly and positively related to leader satisfaction (b = 0.34, SE = 0.07, p < 0.001). Furthermore, results of the Monte Carlo method with 20,000 simulations showed that leader smile emoji usage had a significant indirect effect on leader satisfaction through perceived leader intimacy (indirect effect = 0.01, 95% CI = [0.004, 0.015]). Thus, H1 and H2 were supported.
Means, standard deviations, and bivariate correlations in Study 2
Notes: n = 225. M, mean; SD, standard deviation. For gender, male = 0, female = 1. *p < 0.05, **p < 0.01.
Results of regression analyses in Study 2
Notes: n = 225. Robust standard errors are shown in parentheses. *p < 0.05; **p < 0.01.
H3 and H4 proposed that leader smile emoji usage is negatively related to perceived leader power and leader effectiveness. Model 3 and Model 7 in Table 4 indicated that leader smile emoji usage was negatively related to perceived leader power (b = −0.01, SE = 0.01, p = 0.017) and that leader perceived power was positively related to leader effectiveness (b = 0.20, SE = 0.06, p < 0.01). Moreover, Monte Carlo results revealed that leader smile emoji usage had a significantly negative indirect effect on leader effectiveness via perceived leader power (indirect effect = −0.003, 95% CI = [−0.006, −0.001]). Hence, H3 and H4 were supported.
H5–8 proposed the moderating role of power distance orientation. As shown in Model 2 in Table 4, the interaction term of leader smile emoji usage and power distance orientation was negatively related to perceived leader intimacy (b = −0.03, SE = 0.02, p = 0.03). We further plotted this interaction in Figure 5. Simple slope analysis indicated that the relationship between leader smile emoji usage and perceived leader intimacy was significant among followers with low power distance orientation (b = 0.03, t = 4.80, p < 0.001), but not significant among followers with high power distance orientation (b = −0.01, t = −0.68, p = 0.50). Thus, H5 was supported. Moreover, the Monte Carlo results showed that the indirect effect of leader smile emoji usage on leader satisfaction via perceived leader intimacy was positive when follower power distance orientation was low (indirect effect = 0.01, z = 3.59, 95% CI = [0.006, 0.019]) rather than high (indirect effect = −0.01, 95% CI = [−0.021, 0.009]). The difference between these indirect effects was also significant (Δindirect effect = 0.02, 95% CI = [0.002, 0.036]). Therefore, H7 was supported.
The moderating effect of power distance orientation in the relationship between leader smile emoji usage and perceived leader intimacy in Study 2
Furthermore, as shown in Model 4 in Table 4, the interaction term of leader smile emoji usage and power distance orientation was negatively related to perceived leader power (b = −0.04, SE = 0.01, p < 0.01). As presented in Figure 6, the negative relationship between leader smile emoji usage and perceived leader power was significant for followers with high power distance orientation (b = −0.06, t = −2.99, p < 0.01), but not significant for followers with low power distance orientation (b = −0.01, t = 0.86, p = 0.39). Therefore, H6 was supported. Moreover, results of the Monte Carlo method revealed that the indirect effect of leader smile emoji usage on leader effectiveness via perceived leader power was negative when power distance orientation was high (indirect effect = −0.01, 95% CI = [−0.025, −0.003]) but not significant when power distance orientation was low (indirect effect = −0.001, 95% CI = [−0.004, 0.002]). The difference between these indirect effects was also significant (Δindirect effect = 0.01, 95% CI = [0.002, 0.024]). Thus, H8 was supported. To enhance the robustness of our findings, we used path analysis to test the full model and found that the pattern of our results remained unchanged.
The moderating effect of power distance orientation in the relationship between leader smile emoji usage and perceived leader power in Study 2
Supplementary Analysis
Although WeChat communication is prevalent in the workplace, the usage of emojis is not the sole method for leaders to express their emotions. For example, leaders can convey their positive expressions via other expressive methods, such as face-to-face communication and textual expression. Thus, to verify the incremental predictive validity of smile emoji usage, we recruited 180 MBA students who were full-time employees through the network of authors’ university. Participants were asked to rate leader smile emoji usage in WeChat, face-to-face positive emotional display, and text positive emotional display. We found that leader smile emoji usage still predicted follower perceived intimacy and power after controlling for the effect of leader positive affect display in the face-to-face contexts and in text expressions. Full results can be found in Appendix E on the OSF.
Discussion
Based on an experiment and a field study, we found that leader smile emoji usage had differing effects on follower perception and assessment of the leader. On one hand, it had a positive effect on leader satisfaction through an increased perception of leader intimacy. On the other hand, it had a negative effect on leader effectiveness through a reduced perception of leader power. Moreover, the positive effect of leader smile emoji usage was stronger for followers with low power distance orientation. Conversely, the negative effect was stronger for followers with high power distance orientation. Our research provides several important theoretical contributions to the literature and offers practical implications for managers.
Theoretical Contribution
First, our research contributes to the literature on leadership by uncovering the impact of leader smile emoji usage on follower outcomes. Specifically, the current literature on leadership in the virtual context has discussed leader emergence (e.g., Balthazard, Waldman, & Warren, Reference Balthazard, Waldman and Warren2009; Charlier et al., Reference Charlier, Stewart, Greco and Reeves2016; Serban et al., Reference Serban, Yammarino, Dionne, Kahai, Hao, McHugh, Sotak, Mushore, Friedrich and Peterson2015), the effectiveness of traditional leadership styles (e.g., Banks et al., Reference Banks, Dionne, Mast and Sayama2022; Kelley & Kelloway, Reference Kelley and Kelloway2012; Hoch & Kozlowski, Reference Hoch and Kozlowski2014), leadership functions (Cortellazzo, Bruni, & Zampieri, Reference Cortellazzo, Bruni and Zampieri2019; Kohntopp & McCann, Reference Kohntopp and McCann2019), and leader social media usage (e.g., Matthews et al., Reference Matthews, Matthews, Wang and Kelemen2022; Tur, Harstad, & Antonakis, Reference Tur, Harstad and Antonakis2021) in virtual settings. In addition, this line of literature has emphasized the importance of virtual communication skills for effective leadership in virtual contexts (Bell, McAlpine, & Hill, Reference Bell, McAlpine, Hill and Landers2019; Lubich, Rawlings, & Menefee, Reference Lubich, Rawlings and Menefee2022; McCann & Kohntopp, Reference McCann and Kohntopp2019). However, the current literature has mainly treated technology as a mere context, overlooking its socio-material nature. Our research shifts this paradigm by embracing the socio-material nature of technology, recognizing that digital tools actively shape leadership dynamics. Our approach aligns with and extends a few studies that do consider technology as socio-material (Larson & DeChurch, Reference Larson and DeChurch2020). These studies have primarily focused on verbal and textual information in virtual communication, highlighting the significance of language expression, particularly for leader emergence. Research on the use of nonverbal cues, such as emojis, within the leader-follower virtual communication context remains very limited. Thus, our research extends this line literature by taking an initial step to explore how leaders’ usage of emojis in virtual communications affects follower outcomes. Overall, this exploration expands our understanding of the complex dynamics of leadership in the digital age.
Relatedly, we contribute to the leader–follower virtual communication literature by investigating this issue in the context of instant messaging (IM) application. While IM applications have been broadly applied in workplace communication, the existing literature on leader–follower communication has mainly focused on face-to-face communication (Meinecke, Lehmann-Willenbrock, & Kauffeld, Reference Meinecke, Lehmann-Willenbrock and Kauffeld2017) and email communication (Kato, Kato, & Akahori, Reference Kato, Kato and Akahori2007; Riordan & Glikson, Reference Riordan and Glikson2020). Unlike formal communication tools such as email, the communication between leader and follower on IM applications (e.g., WeChat) is prompter and more diverse. In addition, past research suggests that virtual communication via email and video conference lacks nonverbal cues and it could reduce communication motivation and result in depletion and adverse effects (Bennett et al., Reference Bennett, Campion, Keeler and Keener2021; Gajendran, Loewenstein, Choi, & Ozgen, Reference Gajendran, Loewenstein, Choi and Ozgen2022). Our research thus extends the scope of research on leader–follower communication by exploring this issue in a new communication context and revealing the influences of new technology in leader–follower communication.
Second, our research extends our understanding of emoji literature by revealing the potential negative effect of smile emoji usage (Bai et al., Reference Bai, Dan, Mu and Yang2019; Tang & Hew, Reference Tang and Hew2019). The existing literature has highlighted the positive effect of smile emoji usage in virtual communication such as making conversation vivid, promoting communication proficiency, and reducing distance (e.g., Skovholt, Reference Skovholt and Darics2015; Tu, Reference Tu2022; Zhao, Liu, Chao, & Qian, Reference Zhao, Liu, Chao and Qian2020). Our research confirms the positive effects by showing that leader smile emoji usage is related to increased follower perceived leader intimacy. However, we make a further step by uncovering a unique dynamic in leader-follower interactions: leader smile emoji usage could also lead to negative effects on employee perceived leader effectiveness. As such, our research depicts a more comprehensive picture of the influence of leader smile emoji usage on followers’ evaluations of their leaders, deepening the interpretation of smile emoji's function and detailing how it influences followers’ perceptions of their leaders.
Moreover, the existing literature on emoji usage has mainly focused on the context of conversation between close partners, friends, and even strangers (Rodrigues et al., Reference Rodrigues, Cavalheiro and Prada2022; Sugiyama, Reference Sugiyama2015; Wiseman & Gould, Reference Wiseman and Gould2018). Research on emoji usage in communication at workplace, especially in the leader–follower context, has been very limited (Bai et al., Reference Bai, Dan, Mu and Yang2019; Shandilya, Fan, & Tigwell, Reference Shandilya, Fan and Tigwell2022). Therefore, this research enriches the literature on emoji usage by testing the effect of emoji usage in the context of workplace communication (i.e., leader–follower communication).
Third, we also contribute to the leadership literature by exploring leader emotion display in the virtual communication context. Specifically, past research has focused on leader emotional display in face-to-face contexts (Gardner, Fischer, & Hunt, Reference Gardner, Fischer and Hunt2009; Lewis, Reference Lewis2000; Wang & Seibert, Reference Wang and Seibert2015). Leader usage of smile emojis, as an informal yet common way of emotional display, and its effects on follower has largely been overlooked. Our research extends the existing literature not only by shifting the focus to the virtual communication context, but also by investigating leaders’ usage of smile emojis as a new form of leader emotional display. In addition, previous research on leader emotion display has shown that leaders’ positive emotional display leads to positive outcomes such as organizational citizenship behavior (Edelman & van Knippenberg, Reference Edelman and van Knippenberg2017; Johnson, Reference Johnson2008). Our research extends the literature and provide a more comprehensive understanding of leader emotion display by revealing that leaders’ positive emotion display (i.e., using smile emojis) can also lead to negative outcomes (i.e., reduced follower perception of leader power and leader effectiveness).
Fourth, our research not only utilizes EVT to understand the dual impact of leader smile emoji usage on follower perception and assessment of leaders, but also contributes to the theory itself. Specifically, EVT suggests that in interpersonal communication, a behavior can violate the recipient's expectations in either a positive or negative manner (Burgoon, Reference Burgoon, M. E. Roloff, J. P. Dillard and D. Solomon2015; Burgoon & Le Poire, Reference Burgoon and Le Poire1993). Previous research drawing upon EVT has primarily focused on how one party's behavior positively or negatively violates the expectations of the other party within interpersonal interactions (e.g., Chandler, Anglin, Kanwal, & Short, Reference Chandler, Anglin, Kanwal and Short2024; Heyden et al., Reference Heyden, Gu, Wechtler and Ekanayake2023; Lanaj & Hollenbeck, Reference Lanaj and Hollenbeck2015; McClean, Kim, & Martinez, Reference McClean, Kim and Martinez2022; Polin, Doyle, Kim, Lewicki, & Chawla, Reference Polin, Doyle, Kim, Lewicki and Chawla2024; Schaumberg & Flynn, Reference Schaumberg and Flynn2017). However, an important aspect overlooked in previous EVT research is the fact that individuals often hold multiple expectations of each other. Consequently, it is possible for one behavior to simultaneously violate the other party's expectations in both positive and negative ways. Building on EVT, our study proposes and reveals that leader smile emoji usage leads to a dual violation of followers’ expectations toward their leaders. This research represents a pioneering endeavor to explore the paradoxical phenomenon where a behavior can simultaneously violate the expectations of the other party in both positive and negative manners. Thus, our study contributes to the theoretical underpinnings of EVT by shedding light on this nuanced dynamic.
Fifth, our research contributes to the field of interactional sociolinguistics theory. This theory posits that specific linguistic components, such as language, grammatical rules, volume, timbre, and vocal tones, contribute to meaning-making and provide social cues in communication (Gumperz, Reference Gumperz, Tannen, Hamilton and Schiffrin2005; Rampton, Reference Rampton and Tusting2019; Stubbe et al., Reference Stubbe, Lane, Hilder, Vine, Vine, Marra, Holmes and Weatherall2003). However, previous research has largely overlooked the role of emojis in interactional sociolinguistics theory, possibly because it focused on text-based communication and face-to-face interactions. With the advancements of technology, emojis have emerged as a new linguistic component in virtual communication and have functions of emotional expression, conveying requests, and initiating conversations (Al-Jahdali, Reference Al-Jahdali2023; Al Rashdi, Reference Al Rashdi2018). These studies consider emojis as contextualization cues that enhance individuals’ interpretations of communication. Thus, our findings offer a fresh perspective to expand traditional interactional sociolinguistics theory by considering emojis as a novel linguistic component.
Finally, our research offers a more detailed description of the management context in China. As digital technology continues to permeate Chinese enterprises, Chinese enterprises are increasingly relying on messaging applications, such as WeChat and DingTalk, to communicate and work online (Sun, Wang, & Jeyaraj, Reference Sun, Wang and Jeyaraj2020; Zhong, Lu, Zhu, & Chen, Reference Zhong, Lu, Zhu and Chen2022). This presents a unique opportunity to explore how the adoption of these technologies affects the effectiveness of management in Chinese enterprises. By focusing on leader smile emoji usage, our study addresses the need to examine virtual communication within Chinese enterprises (e.g., Xie, Zuo, & Hu, Reference Xie, Zuo and Hu2021) and offers theoretical insights for Chinese enterprise management.
Practical Contributions
Our findings also offer several important implications for practice. First, our findings suggest that leaders should pay attention to emoji usage in face of the tendency where digital communication is becoming more and more popular and young employees are gradually becoming the dominant group in the workplace. Emoji usage is considered common and necessary in computer-mediated communication (CMC) and can be viewed as a potential tool for leaders to manage a better visual impression and foster efficiency in the virtual workplace. For instance, leaders can use emojis to create a warm impression to strengthen the leader–follower relationship, contribute to mutual understanding and build an atmosphere of flexible and relaxing communication at work.
Second, our findings indicate that leader smile emoji usage in communication with followers may help leaders to build more intimate relationships with followers. In addition, doing so may lead to reduced follower perception of leader power and rating of leader effectiveness. Accordingly, we suggest that leaders who intend to build intimate relationships with followers to use smile emojis in their communication. With that said, these leaders should also pay attention to the potential costs of using smile emojis with followers and take other actions to enhance followers’ perception of leader power.
Third, our findings show that follower power distance orientation plays a critical role in followers’ response to leader smile emoji usage. Specifically, follower power distance orientation weakens the positive effects of leader smile emoji usage on satisfaction with leader while strengthens its negative effect on ratings of leader effectiveness. Accordingly, we suggest that leaders should adjust their use of smile emojis based on follower power distance. That is, they should use smile emoji more when communicating with followers who have low power distance orientation. Conversely, when interacting with followers who have high power distance orientation, leaders should be aware that using smile emojis may lead to undesirable employee evaluations.
Fourth, our findings on the negative effect of emoji usage on perceived leader power highlight the importance of avoiding biases when evaluating leaders. This reminds followers that in order to avoid bias, they need to apply a more comprehensive perspective to evaluate the leader instead of just focusing on the specific linguistic components in conversations with leaders on digital platforms. In addition, the insights from our findings suggests that nuanced nonverbal cues in visual communication can lead to distinct and paradoxical interpretations. Thus, organizations should provide training for leaders on the effective use of digital communication cues, including emojis and stickers, and thus promote the accuracy of online emotional display. Also, organizations can create a special emoji series that symbolizes the organization and encourage members to use them at work, which may enhance their sense of belonging and build more vivid organizational branding.
Limitations and Future Research Directions
There are also several limitations and future directions that warrant discussion. First, there are many different types of smile emojis on digital platforms (Sun et al., Reference Sun, Lasser and Lee2023). In this research, as an initial attempt to explore how leaders’ use of smile emojis affects followers, we chose to focus on leaders’ use of smile emojis in general, rather than specific types of smile emojis (e.g., 
and 
). As such, we do not differentiate between specific types of smile emojis such as typical smile emojis (e.g., 
) and laughter emojis (e.g., 
). However, as an anonymous reviewer suggested, the nuanced differences among different smile emojis may convey different levels of emotional arousal. While the use of emojis within the same category may impact recipients in similar ways, subtle differences between individual emojis may lead to differentiated emotional reactions and interpretations. Hence, future research needs to balance between focusing on a category of emojis (e.g., smile emojis) or focusing on a specific emoji (e.g., a particular type of smile emoji). Although focusing on a broad category of emojis may yield insights into general trends and common effects, it may overlook the nuanced differences between specific emojis. Conversely, while focusing on a specific emoji helps to uncover unique impacts, it may limit the generalizability to other emojis within the same category. Hence, researchers must carefully consider these factors to balance the need for generalizability with the precision of detailed analysis when designing their studies. This is both a challenge and an opportunity for new research. Specifically, when past research has provided a systematic understanding of the impact of a certain category of emojis at work, new research may shift the focus to some specific emojis to gain a more nuanced understanding.
Second, while we focus on leader smile emoji usage in this research, we encourage future research to further explore how leader usage of other emojis affects follower responses. Specifically, there are many different types of emojis ranging from faces to objects (Riordan, Reference Riordan2017), such as crying emojis (e.g., 
) and angry emojis (e.g., 
), demonstrating both diversity in kinds and subtlety of meanings. Furthermore, the way emojis are used, such as the content of the message (Bai et al., Reference Bai, Dan, Mu and Yang2019) and the position of emojis within messages (Wu, Chen, & Li, Reference Wu, Chen and Li2021), may also impact recipients’ perception. In addition, a message may contain multiple emojis, creating more complex scenarios (Riordan, Reference Riordan2017). The combinations of emojis within messages may also lead to different impacts on recipients, warranting special attention in emoji research.
Third, we focused on WeChat as the platform in which digital communication take places. WeChat is the most widely used social media application in China (Zhang, Wen, Liang & Lei, Reference Zhang, Wen, Liang and Lei2017). Work communication between leaders and subordinates on WeChat is common in the workplace (Tian, Reference Tian2021; Zhang, Wang, Li, & Chen, Reference Zhang, Wang, Li and Chen2021), as 57.22% of new contacts users add are work-related and over 80% of users show work-related content on their accounts (TechWeb, 2017). Although WeChat is the most frequently used digital platform in our sample to communicate with their leaders, more studies could be conducted to test whether our research findings could be applied to other digital platforms in different countries, such as DingTalk and Feishu. Furthermore, exploring whether leaders express similar emotions through different communication approaches is worth considering. For example, a leader may adopt a serious demeanor in face-to-face interactions while frequently use smile emojis in virtual communication. It would be interesting to investigate such discrepancy and its impact on followers.
Fourth, we employed an objective measure of total leader smile emoji usage by asking participants to count how many smile emojis they received from their leaders. We chose to measure this after participants rated the two mediators to avoid potential interference with their ratings. Also, since the measure for the independent variable is objective, the order of the measures should not affect the results. However, methodological diversity is encouraged to track leader emoji usage and examine its subsequent effects on followers’ attitudes and behaviors. Other influential mechanisms such as warmth, psychological distance, and leader-member exchange (LMX) should also be further investigated to enrich our understanding of the leader emoji usage.
Relatedly, our research focuses on satisfaction with leader and leader effectiveness as the dependent variables, and perceived leader intimacy and perceived leader power as mediators. Nevertheless, it is possible that satisfaction with leader has a reversed effect on perceived leader intimacy, and that leader effectiveness might affect follower evaluation of leader power. We have conducted two additional experiments to test these causal relationships (Please find Appendix F on the OSF). We also encourage future research to conduct longitudinal studies to test the causal relationships proposed in our research model. In addition, perceived leader intimacy may also influence perceived leadership effectiveness. We focus on perceived leader power as it is a more optimal predictor. Specifically, power is integral to leadership as it enables leaders to influence others and allocate resources effectively (Offermann et al., Reference Offermann, Kennedy and Wirtz1994). Moreover, although LMX theory emphasizes the importance of high-quality relationships, these relationships characterized by intimacy are not the sole determinants of leader effectiveness. Effective leaders are expected to demonstrate competence and facilitate team performance (Kenney, Schwartz-Kenney, & Blascovich, Reference Kenney, Schwartz-Kenney and Blascovich1996), and thus, perceived intimacy is more related to relational outcomes like satisfaction and commitment rather than task outcomes. Hence, leaders perceived as powerful are likely seen as more effective in achieving organizational goals. Future studies could further explore the relationship between leader–follower intimacy and leader effectiveness.
Fifth, in this research, we examined the moderating role of follower power distance orientation. We encourage future research to explore other potential boundary conditions to draw a more comprehensive picture of emoji research in the field of organizational behavior. Specifically, the same emoji may be interpreted in significantly different ways in various communication contexts (Bai et al., Reference Bai, Dan, Mu and Yang2019). For instance, the relationship between the leader and follower, the purpose of the communication, and the follower preference for emoji usage may influence the impact of leader emojis usage. The complexity related to various contextual factors presents both a challenge and an opportunity for future research. By exploring the interactions and nuanced impact of contextual factors, future research can achieve a more comprehensive understanding of the diverse roles and impact of emojis in digital communication at work.
Finally, our experiment and field study were conducted in China. While China provided a suitable context to examine the effect of digital communication in the workplace, further investigation is needed to assess the generalizability of our findings to other cultural contexts, as culture plays a significant role in individuals’ interpretation and selection of emojis (Sun et al., Reference Sun, Lasser and Lee2023). The same emoji can convey different meanings and leads to different interpretations depending on the cultural background. For instance, the ‘face with tears of joy’ emoji is commonly used in Western cultures to express laughter or amusement. However, in some countries such as Nigeria, it is usually used to convey a sense of disappointment, sadness, or even hopelessness. While cultural context may limit the generalizability of certain research findings in this field, it also presents new opportunities for in-depth research. By exploring cross-cultural differences in emoji usage at work and its impact, researchers can uncover the unique ways in which emojis are used and interpreted across cultures. This understanding can help global organizations better manage and facilitate communication within multicultural teams, enhancing collaboration efficiency and employee satisfaction worldwide.
Data availability statement
The data that support the findings of this study are available at https://osf.io/zkgjc/?view_only=4c9732d66d8c4aeaa1f2b3d51d2fb196
Funding statement
This research was supported by the National Natural Science Foundation of China (Grant 72372033, 72002038, 72072032, 72172034, 72372012, and 72472005), the Humanities and Social Sciences Foundation of the Ministry of Education of China (No. 24YJA630112), the fundamental research funds for the central universities, Peking University (No. LMEQF), and Beijing Philosophy and Social Science Foundation (No. 24DTR027).
Appendix A
Additional Experiment in Study 1
Following the reviewer's suggestion, we have conducted a new experiment with a 2-by-2 design. In this new experiment, we manipulated both leader smile emoji usage and power distance orientation.
Sample
We collected data via Credamo, which is a widely used data-collection platform in China (e.g., Gong et al., Reference Gong, Lu, Schaubroeck, Li, Zhou and Qian2020; Tang, Zhang, & Ning, Reference Tang, Zhang and Ning2023). We recruited 250 participants. The final sample consisted of 241 participants, as 9 participants who failed attention check were excluded. Their average age was 31.85 (SD = 6.17), average tenure was 5.84 years (SD = 4.11), and 60.54% of them were females.
Process
We manipulated both power distance orientation and leader smile usage, using a 2 (power distance orientation: high vs. low) × 2 (leader smile usage: use vs. not use). Participants were randomly assigned to four conditions. We asked participants to image they were an employee Wang worked in HuaXing company and their direct leader was Liu. Then they read the description of Wang reflecting power distance orientation. Next, they were presented with the leader smile usage manipulation. After that, participants completed the measures of perceived intimacy, perceived power, manipulation checks, and demographic information.
Manipulations
Power distance orientation. We followed the definition of power distance orientation and related research (van den Bos, Brockner, van den Oudenalder, & Kamble, Reference van Den Bos, Brockner, van Den Oudenalder, Kamble and Nasabi2013) to design the manipulation material. In the high power distance orientation condition, participants read the following statement:
Liu is your direct leader, who holds formal authority over you due to his higher position. You believe that the unequal power distribution is important in your interactions with Liu. You think that maintain a formal hierarchical relationship and a certain social distance with Liu. In addition, followers should obey the directives of leaders to complete tasks, leaders should avoid soliciting followers' opinions, and followers should not question the leaders' decisions.
In the low power distance orientation condition, participants were presented with the following statement:
Liu is your direct leader, who holds formal authority over you due to his higher position. You believe that the unequal power distribution is not important in your interactions with Liu. You think that in addition to formal work relations, there should be informal interactions between leader and follower. In addition, followers should work alongside leaders to accomplish tasks, communicate on an equal footing, leaders should not consult their followers’ opinions, and followers could even question the leaders' decisions.
We used the same leader emoji usage manipulation and manipulation check as in our Study 1. Perceived leader intimacy (α = 0.87), leader power (α = 0.77), and power distance orientation manipulation check (α = 0.82) were also used the same scales as in our Study 1.
Results
The results of ANOVA showed that participants in high power distance orientation rated higher on power distance orientation (M = 3.40, SD = 0.96) than those in the low power distance orientation condition (M = 2.36, SD = 0.79), F (239) = 85.66, p < 0.001, d = 1.18. Thus, our manipulation of power distance was successful.
Participants in leader smile usage condition rated higher leader intimacy (M = 3.78, SD = .67) than those in the control condition (M = 3.17, SD = 1.01), F (239) = 30.29, p < 0.01, d = 0.71. The rating of leader power was lower in the leader smile usage condition (M = 3.99, SD = 0.71) than the control condition (M = 4.24, SD = 0.36), F (239) = 12.63, p < 0.01, d = 0.44.
Furthermore, results of two-way ANOVA showed that the interaction between leader smile emoji usage and power distance orientation on leader intimacy was significant (F = 26.73, p < 0.01). We presented the results in Figure A1. Specifically, in the low power distance orientation group, participants rated higher on leader intimacy (M = 3.88, SD = 0.68, n = 62) in leader smile usage condition than the control condition (M = 2.74, SD = 0.87, n = 61), t = 7.89, p < 0.01. While, in the high power distance orientation group, there was no significantly different on leader intimacy between leader smile emoji usage condition (M = 3.66, SD = 0.66, n = 56) and control condition (M = 3.59, SD = 0.96, n = 62), t = 0.49, p = 0.63.
Leader intimacy in four conditions
In addition, the two-way ANOVA also showed that the interaction between leader smile emoji usage and power distance orientation on leader power was also significant (F = 26.73, p < 0.01). The results are presented in Figure A2. In the high power distance orientation group, participants rated lower on leader power (M = 3.72, SD = 0.84, n = 56) in leader smile usage condition than the control condition (M = 4.35, SD = 0.26, n = 62), t = 6.44, p < 0.01. In the low power distance orientation group, there was no significantly different on leader power between leader smile emoji usage condition (M = 4.23, SD = 0.47, n = 62) and control condition (M = 4.14, SD = 0.41, n = 61), t = 0.95, p = 0.34. Thus, the new experiment also supported our hypotheses.
Leader power in four conditions
Appendix B
Different Smile Emoji Use Frequencies
There are many smile emojis in WeChat so for, including 
,
,
,
,
,
,
,
,
,
. Among the ten smile emojis we list above, several of them are quite normal in our daily lives but seem unusual in the context of the workplace. We conducted a survey to figure out the frequency of leader's usage of these smile emojis in their daily work communication.
We recruited 160 employees via Credamo, a widely used data collection website in China (e.g., Gong et al., Reference Gong, Lu, Schaubroeck, Li, Zhou and Qian2020; Tang, Zhang, & Ning, Reference Tang, Zhang and Ning2023). The participants (26.25% male) have average age of 29 years old and work in various industries, including manufacturing (36.88%), information technology (35.63%), service (11.88%), finance (6.25%), and others (9.38%). And they are asked to rate the frequency of leader usage of specific emojis. We used a 5-point Likert-type scale ranging from 1 (rarely) to 5 (always). The results are presented in the following table.
Therefore, we observed a variation in the usage of smile emojis among leaders. Emojis that convey jokes and sarcasm, such as 
,
,
,
, are uncommon in leader–follower communication. On the other hand, emojis that represent pure and general positive emotions are more prevalent in the workplace setting. Consistent with this observation, the examples (
) we provided to survey participants excluded emojis with ambiguous meanings, ensuring clarity in their responses.
Appendix C
The Effects of Different Smile Emojis on Followers
We conducted another survey to explore followers' perceptions of leaders’ different smile emoji usage. Specifically, we try to explore three smile emojis (i.e., 
,
,
) which are our examples in the survey and we hope they have the least difference in affecting followers’ perception.
We recruited 341 participants, who were all workers from Credamo. Among them, 65.40% are female and more than 93.26% has a bachelor's degree or received higher education. Their average age is 32 years old and they come from various industries, including manufacturing (36.95%), information technology (21.41%), service (9.97%), finance (7.92%), and others (23.75%).
We presented the above ten types of smile emojis leader used in their conversation on WeChat to participants and then we asked them to evaluate their perception of leader intimacy, leader power, leader warmth, leader competence, and psychological distance toward the leader using a specific smile emoji. All measures were rated using a 5-point Likert-type scale ranging from 1 (Strongly disagree) to 5 (Strongly agree).
Perceived leader intimacy. We adapted the scale from Sinclair and Dowdy (Reference Sinclair and Dowdy2005) and the single item we used is ‘I have a close relationship with the leader’.
Perceived leader power. We adapted the scale from Ju et al. (Reference Ju, Huang, Liu, Qin, Hu and Chen2019) and the single item we used is ‘I think the leader is powerful’.
Perceived leader warmth. We adapted the scale from Porath, Gerbasi, Schorch, and Chen. (Reference Porath, Gerbasi, Schorch and Chen2015) and the single item we used is ‘I think the leader is very warm’.
Perceived leader competence. We adapted the scale from Porath et al. (2015) and the single item we used is ‘I think the leader is very competent’.
The following Figure C1 shows the average of perceived intimacy and power for ten kinds of smiles. Average score of the intimacy and power of smile emojis.
Average score of the perceived intimacy and perceived power of smile emojis
We conducted one-way repeated measures ANOVA to compare the effect of different smile emojis on the perception of workers and the following Table C1 presents the result. We first test the three kinds of smile emojis that we consider are congruent with our definition: 
,
,
. The difference in perception of leader intimacy (F (2,680) = 2.00, p = 0.136) and perception of leader power (F (2,680) = 0.45, p = 0.639) among the three conditions are not significant. And we also find that perception of competence (F (2,680) = 1.76, p = 0.174), and psychological distance (F(2,680) = 0.47, p = 0.627) among the three conditions are also not significantly different. The results demonstrate that our conceptional construction is solid and the conclusion of our research has certain universe significance.
One-way repeated measures ANOVA for smile emojis (
,
,
)
We also analyzed other kinds of smile emojis by one-way repeated measures ANOVA and also found some interesting phenomena. The full results are presented in the following Table C2. For example, focusing on the perception of intimacy, we paid attention to the predictive margins of groups and found that under emoji conditions 
and 
, participants were less likely to perceive intimacy. The phenomenon also existed when we analyzed the predictive margin of perception of warmth. Though these smile emojis are not congruent with our definition of leader smile emoji usage, we encourage further study to explore their effect on leader–follower communication.
Comparing perceptions of different emoji smiles
Appendix D
The Effects of Different Smile Emojis on Followers
We have conducted analyses using subjective smile usage as IV. The results were stable and all hypotheses were supported. We have listed the regression analysis results in the following table.
Specifically, subjective leader smile usage had a positive indirect effect on leader satisfaction through perceived leader intimacy (indirect effect = 0.06, z = 2.90, p = 0.004) and had a negative indirect effect on leader effectiveness through perceived leader power (indirect effect = −0.02, z = 2.11, p = 0.04).
In addition, when power distance orientation was high (+1 SD), the positive indirect effect of leader smile usage on leader satisfaction through perceived leader intimacy was not significant (indirect effect = 0.02, z = 0.82, p = 0.41), and such indirect effect become positive (indirect effect = 0.09, z = 3.23, p = 0.001) when power distance orientation was low (−1 SD). The difference between high and low power distance orientation conditions was also significant (indirect effect change = 0.08, z = 2.38, p = 0.02). Leader smile emoji usage had a negative indirect effect on leader effectiveness through perceived leader power when power distance orientation was high (indirect effect = −0.03, z = −2.16, p = 0.02), whereas such indirect effect becomes nonsignificant when power distance orientation was low (indirect effect =−0.002, z = 0.25, p = 0.81). The difference between high and low power distance orientation conditions was also significant (indirect effect change = −0.04, z = 2.05, p = 0.04).
Appendix E
Compare Different Leader Positive Emotion Expression Approaches
We want to test whether leader smile emoji usage has unique influence after controlling emotional expression in other approaches. We recruited 180 MBA students who were full-time employees through the network of authors’ university. Their average age was 32.71, average tenure was 5.97, and 65.56% were males. Participants were asked to rate the leader smile emoji usage in WeChat, leader face-to-face positive emotional display, leader text positive emotional display, perceived intimacy, and perceived power.
We adapted the five-item positive affect scale developed by Wang and Seibert (Reference Wang and Seibert2015) to measure leader positive express through different modules. A five-item Likert scale was used to measure the frequency of each item (1 = never, 5 = always). We used five items to measure leader positive emotional display in the face-to-face context and a sample item is ‘My direct leader displays happiness toward me in the face-to-face condition’ (α = 0.79). We used five items to measure leader positive emotional display in textual expression, and a sample item is ‘My direct leader displays optimistic in the text condition (e.g., email)’ (α = 0.83). We used four items to measure leader smile emoji usage in social media and a sample item is ‘My leader often use positive smile emojis in the WeChat communication’ (α = 0.84). Perceived leader intimacy (α = 0.72) and perceived leader power (α = 0.85) were measured using the same scales as in the study of manuscript. The results of OLS are presented in the following table.
We found that leader smile emoji usage was still positively related to perceived leader intimacy (b = 0.22, SE = 0.06, p < 0.001) after controlling the effect of positive emotional display in face-to-face context and in text expression. In addition, when controlling the effect of positive emotional display in face-to-face context and in text expression, leader smile emoji usage was negatively related to perceived leader power (b = −0.38, SE = 0.15, p = 0.01). As such, it provides evidence for the influence of leader smile usage in affecting employees’ perception of their leaders.
Inspired by your suggestion, we have also discussed this issue in the Supplementary Analysis in Study 2 and Limitation and Future Direction section. An interesting topic for future studies is what if leader presents different affect in different approaches. For instance, a leader is serious in the workplace but present positive affect and use smile on virtual communication. We encourage future studies to further explore the unique effect of leader emoji usage on followers.
Appendix F
Additional Studies About Mediators and Dependent Variables
The Effect of Leader Intimacy on Leader Satisfaction
Sample and procedure
We recruited 99 participants, who were all workers from Credamo. Among them, 70.71% are females and 87.88% have a bachelor's degree or received higher education. Their average age is 32.44 years old and their tenure is 6.31 years. They were from various industries, including manufacturing (12.12%), information technology (10.10%), service (20.20%), finance (9.09%), education (10.10%), construction business (7.07%), and others (31.31%).
All participants were randomly divided into two groups: the experiment group and the control group. We adopted the manipulation of intimacy perception based on the definition of intimacy (Kark, Reference Kark, Cameron and Spreitzer2012). In both groups, participants were required to imagine they were an employee in the Huaxing Company with a leader Liu. In the experiment group, participants read that:
During the work, Liu shares a close relationship with you. Liu is open and willing to confide in you, share his emotions, and demonstrate empathy towards you. In times of difficulty, Liu expresses care and support, indicating his willingness to forge a strong bond with you. Both of you possess a deeper understanding of each other, fostering a sense of mutual closeness and comprehension.
While in the control group, participants were shown that:
During the work, you feel a sense of detachment in your relationship with Liu. He is reluctant to open up to you, share his emotions, or display any empathy toward you, and there is no indication of his willingness to establish a close bond with you. Even in times of difficulty, you seldom receive any concern from Liu.
After reading the materials, participants were required to rate leader satisfaction, finish the manipulation check, and report demographic information.
Measures
All variables are measured using a 5-point Likert scale (1 = ‘Strongly disagree’, 5 = ‘Strongly agree’).
Leader satisfaction. We adapted the three-item scale from Michigan Organizational Assessment Questionnaire (Cammann, Fichman, Jenkins, & Klesh, Reference Cammann, Fichman, Jenkins, Klesh, Seashore, Lawler, Mirvis and Camman1983) and the sample item is ‘All in all, I am satisfied with leader Liu’ (α = 0.95).
Manipulation check. We adapted the four-item scale from Sinclair and Dowdy (Reference Sinclair and Dowdy2005) and the sample item is ‘I have a close relationship with leader Liu’ (α = 0.96).
Results
We conducted one-way ANOVA to test the effect of perceived intimacy on leader satisfaction and plotted the results in Figure F1. Results displayed that participants under the experimental condition rated significantly higher levels of leader satisfaction (M = 4.38, SD = 0.40) than those in the control condition (M = 2.33, SD = 1.04), F (1,97) = 169.65, p < 0.001, d = 2.60, which supported our hypothesis that perceived leader intimacy is positively related to leader satisfaction.
The Effect of Perceived Leader Power on Leader Effectiveness
Through Credamo, we recruited 100 participants, who were all working adults. Among them, 78.00% were females, their average tenure is 6.54 years, and their average age is 32.09 years. 90.00% of them had a bachelor's degree and they were from various industries, including manufacturing (34.00%), information technology (20.00%), service (11.00%), finance (6.00%), education (10.00%), construction business (5.00%), and others (14.00%).
All participants were randomly distributed into the experiment and control groups. We manipulated leader power following previous research (Tost et al., Reference Tost, Gino and Larrick2013; Mooijman, Kouchaki, Beall, & Graham, Reference Mooijman, Kouchaki, Beall and Graham2020; Mooijman et al., 2019). Specifically, in both groups, participants were required to imagine they were an employee in the Huaxing Company with a leader whose name is Liu. In the experiment group, participants were shown that:
Liu is a leader within the organization, possessing significant job capabilities, and is recognized as one of the most influential individuals within the organization. Compared to others in the organization, Liu controls the most resources and has the ability to dictate decisions and task assignments for others within the organization. Others within the organization report to Liu for their work, and he evaluates their job performance. Additionally, others within the organization rely on Liu to achieve their goals and obtain the resources they require.
While in the control group, participants were shown that:
Liu is a leader within the organization, possessing certain job capabilities, but is recognized as one of the least influential individuals within the organization. Compared to others in the organization, Liu controls the fewest resources, and his ability to make decisions and undertake tasks is subject to the control of others. Liu reports his job performance to others within the organization and is evaluated by them. Additionally, Liu relies on others within the organization to achieve the goals and obtain the resources he desires.
After reading the materials, participants were required to rate leader effectiveness, finish the manipulation check, and report demographic information.
Measures
All variables are measured using a 5-point Likert scale (1 = ‘Strongly disagree’, 5 = ‘Strongly agree’).
Leader effectiveness. We adapted the three-item scale from Douglas and Ammeter (Reference Douglas and Ammeter2004) and the sample item is ‘Leader Liu is effective in meeting the job-related needs of work unit members’ (α = 0.87).
Manipulation check. We adapted the four-item scale from Ju et al. (Reference Ju, Huang, Liu, Qin, Hu and Chen2019) and the sample item is ‘Leader Liu would get me to listen to what he says’ (α = 0.91).
Results
Then we applied the one-way ANOVA to test the effect of perceived leader power on leader effectiveness and plotted in Figure F2. The results showed that employees rated higher on the leader effectiveness (M = 4.16, SD = 0.52) when perceived leader power is high than perceived leader power is low (M = 2.83, SD = 1.00), F (1,98) = 70.38, p < 0.001, d = 1.67, which was consistent with our hypothesis that perceived leader power is positively associated to leader effectiveness.
Leader satisfaction in two conditions
Leader effectiveness in two conditions
Shengming Liu (shengmingliu@fudan.edu.cn) is an associate professor at the School of Management, Fudan University. He received his PhD in organizational management from the Guanghua School of Management, Peking University. His research interests include leadership, social media, and high-tech venture management. He has published research in the Journal of Applied Psychology, Journal of Management, Journal of Management Studies, and Journal of Business Ethics.
Mingpeng Huang (hmp@uibe.edu.cn) is a professor at the Business School, University of International Business and Economics. He received his PhD in organizational management from the Guanghua School of Management, Peking University. His research interests include leadership and ethics. His research has been published in top academic journals such as Academy of Management Journal, Journal of Applied Psychology, Organizational Behavior and Human Decision Processes, Journal of Business Ethics, and Journal of Organizational Behavior.
Dong Ju (dongju@bnu.edu.cn) is an associate professor at the Business School of Beijing Normal University. She received her PhD in organizational management from the Guanghua School of Management, Peking University. Her current research interests include leadership, ethics, and employee well-being. She has published her research in Academy of Management Journal, Journal of Applied Psychology, Organizational Behavior and Human Decision Processes, and Journal of Business Ethics.
Zhiying Shi (zyshi24@m.fudan.edu.cn) is a doctoral student at the School of Management, Fudan University. Her research interest lies in leadership and identity.
Minya Xu (minyaxu@gsm.pku.edu.cn) is an Associate Professor at the Guanghua School of Management, Peking University. Her research interests lie in employees’ workplace behavior and performance, especially their behavior changes in the new technology era. Her research has been published in top academic journals such as Journal of Management, Journal of Applied Psychology, Personnel Psychology, Journal of Organizational Behavior, The Leadership Quarterly, Journal of Occupational and Organizational Psychology, Human Relations, and Journal of Business Ethics.
means “I love you”. Paper presented at the 2018 CHI Conference on Human Factors in Computing Systems, New York, April 2018.
