Full Terms & Conditions of access and use can be found athttp://www.tandfonline.com/action/journalInformation?journalCode=rpos20

The Journal of Positive PsychologyDedicated to furthering research and promoting good practice

ISSN: 1743-9760 (Print) 1743-9779 (Online) Journal homepage: http://www.tandfonline.com/loi/rpos20

The left side of courage: Three exploratory studieson the antecedents of social courage

Matt C. Howard & Joshua E. Cogswell

To cite this article: Matt C. Howard & Joshua E. Cogswell (2018): The left side of courage: Threeexploratory studies on the antecedents of social courage, The Journal of Positive Psychology

To link to this article: https://doi.org/10.1080/17439760.2018.1426780

View supplementary material

Published online: 17 Jan 2018.

Submit your article to this journal

View related articles

View Crossmark data

The Journal of PosiTive Psychology, 2018https://doi.org/10.1080/17439760.2018.1426780

The left side of courage: Three exploratory studies on the antecedents of social courage

Matt C. Howard and Joshua E. Cogswell

Marketing and Quantitative Methods, The university of south alabama, Mitchell college of Business, Mobile, al, usa

ABSTRACTCourage has been linked to important workplace outcomes, but little is known about the antecedents of courage. To identify possible methods to develop a more courageous workforce, the current article provides initial evidence regarding the antecedents of behavioral social courage from five broad categories: personality, job characteristics, leadership, culture, and demographic characteristics. Three exploratory studies show that certain antecedents from these categories have a significant relationship with behavioral social courage, including grit, proactive personality (personality), social support (job characteristics), empowering leadership (leadership), power distance (culture), and age (demographics). Perceived courage benefits and risks did not mediate the effect of most antecedents on behavioral social courage. These results suggest that certain antecedents may influence behavioral social courage, but the mediators of these relationships remain unknown. Further implications and directions for future research are discussed.

The importance of courage has been recognized through-out history, dating back to the works of Plato and Aristotle (Hobbs, 2000; Snyder, Lopez, & Pedrotti, 2010). These ear-lier works often stressed the importance of courage for extreme occasions, such as leading during war, but recent works have argued that courage may be important for everyday interactions – particularly within the workplace (Koerner, 2014; Pury & Lopez, 2010; Schilpzand, Hekman, & Mitchell, 2014). An employee may need courage to report the abusive supervision of their supervisor, or a supervisor may even need courage to provide an honest performance review of a subordinate. Further, earlier works often pro-vided vague definitions for the construct, such as ‘persis-tence despite fear’ (Howard & Alipour, 2014; Norton & Weiss, 2009), but more recent works have applied sophisticated empirical methods to identify theoretically- and empirical-ly-supported definitions for courage (Pury, Brawley, Lopez, & Burnett, 2016; Pury, Kowalski, & Spearman, 2007; Rate, 2010; Rate, Clarke, Lindsay, & Sternberg, 2007). Perhaps the most supported definition was created by Rate and colleagues (2007; Rate, 2010). This definition specifies that courage is a ‘(a) willful, intentional act, (b) executed after mindful deliberation, (c) involving objective substantial risk to the actor, (d) primarily motivated to bring about a noble good or worthy end’ (Rate et al., 2007, p. 95). A per-son that repeatedly performs such behaviors is believed

to possess the trait of courage (Pury & Starkey, 2010; Rate, 2010; Rate et al., 2007).

Recent studies have begun to empirically study cour-age’s influence on work outcomes (Howard, Farr, Grandey, & Gutworth, 2016; Koerner, 2014; Schilpzand et al., 2014). Koerner (2014) provided a theoretical model suggesting that courageous behaviors influence employees’ personal identities, which subsequently influence the behaviors that they choose to perform. Qualitative evidence supported this proposal, showing that courage behaviors indeed influence subsequent behavioral outcomes (Koerner, 2014). Howard and colleagues (2016) quantitatively showed that social courage, one of the many possible dimensions of courage, significantly relates to organiza-tional citizenship behaviors and prosocial rule breaking even when controlling for conscientiousness. From these findings and others (Baumert, Halmburger, & Schmitt, 2013; Hannah, Avolio, & Walumbwa, 2011), authors have suggested and empirically supported that courage relates to many important work outcomes.

Despite these important findings, much less research has considered the antecedents of courage (Koerner, 2014; Schilpzand et al., 2014); few authors have empiri-cally studied possible antecedents (Howard et al., 2016); and Detert and Bruno (2017) directly call for research on the antecedents of courage in their recent review. In other

© 2018 informa uK limited, trading as Taylor & francis group

KEYWORDScourage; positive organizational behavior; traits; prosocial behaviors; positive psychology; personality

ARTICLE HISTORYreceived 14 June 2017 accepted 14 December 2017

CONTACT Matt c. howard Mhoward@southalabama.edu supplemental data for this article can be accessed https://doi.org/10.1080/17439760.2018.1426780.

2 M. C. HOWARD AND J. E. COGSWELL

specifies four primary aspects of a courageous behavior: (a) intention, (b) deliberation, (c) risk, (d) and noble inten-tions. While the term ″noble″ has been treated in several different manners, researchers often consider it to be syn-onymous with prosocial (Howard & Alipour, 2014; Howard et al., 2016; Rate, 2010). The same perspective is taken in the current article. Further, a behavior that includes these four elements is considered a courageous act, and a person that repeatedly performs these behaviors is then consid-ered to possess trait courage (Rate, 2010; Rate et al., 2007). Thus, Rate and colleagues’ (2007; Rate, 2010) definition is able to identify courage as a trait as well as a behavior.

In research, a greater attention has been paid to trait courage (Howard et al., 2016; Norton & Weiss, 2009; Pury & Starkey, 2010; Rate, 2010), perhaps due to the historic focus on courage as an accolade. In the current article, however, the antecedents of behavioral courage are explored. In the traditional method to gauge trait courage, participants respond independently of their context (i.e. work environ-ment; Howard et al., 2016). As further detailed below, we gauge courage by asking participants to respond specifi-cally to their behaviors in their current work environment, which we consider to be a measure of behavioral courage (Pury & Starkey, 2010).

We analyze behavioral courage for three primary rea-sons. First, prior studies of courage as a trait does not pre-clude the study of courage as a behavior. As mentioned, conceptualizations of trait courage are predicated on the occurrence of courage behaviors, and thereby courage as a behavior is implicit in these prior studies of trait courage. Second, behavioral manifestations of courage, as opposed to the latent trait of courage, have a direct influence on the workplace, and these behavioral manifestations are more proximal to workplace outcomes that are important for both research and practice. Third, behavioral courage is believed to be more malleable than trait courage, which would be more likely to be influenced by antecedents. Thus, behavioral social courage may be more relevant to the study of antecedents than trait social courage.

Additionally, courage is a multidimensional construct (Lopez et al., 2010; Sekerka, Bagozzi, & Charnigo, 2009). Some authors have differed on their methods to identify these dimensions, but most consider the dimensions to be differentiated by the risks (i.e. physical courage, social courage) or goals (i.e. moral courage) involved in a behav-ior rather than the context (i.e. work courage, war courage) or person performing the behavior (i.e. managerial cour-age, subordinate courage; May, 1994; Schilpzand, 2008; Woodard & Pury, 2007). From categorizing risks, authors have consistently identified three types of courageous behaviors that may be important to workplace interac-tions: physical, moral, and social courage (Howard et al., 2016; Sekerka & Bagozzi, 2007).

words, very little is known about ‘the left side’ of courage. To prompt the wider study of courage’s development, the goal of the current article is to provide an explora-tory investigation into (a) the possible antecedents of courage and (b) the manner in which these antecedents may influence behavioral courage. In doing so, we study antecedents from five broad categories (personality, job characteristics, leadership, culture, and demographic characteristics), and we test whether perceived courage benefits and risks serve as mediators of these relation-ships. Providing initial inferences regarding the anteced-ents of courage may allow future research to conduct more directed studies to identify how courage can be developed. This would subsequently allow businesses to foster conducive environments, which could benefit the success of the organization as well as the employees themselves.

To achieve these goals, the current article first provides a review of courage. Second, the current article identifies a dimension of courage with a sufficient theoretical basis and existing measurement, social courage, which indicates that it is prime for empirical research. Third, three empir-ical studies are performed to identify the relationship of possible antecedents with behavioral social courage and possible mediators. Fourth, these findings are integrated with prior research and suggestions are provided for the continued study of the left side of courage.

Background

Courage and social courage

Certain aspects of courage should be considered before identifying possible antecedents. As mentioned, courage is often defined as a type of act or behavior, but it is also con-sidered a trait. Pury and Starkey (2010) consider this the dif-ference between courage as a process and courage as an accolade. Most early definitions of courage treated it only as a trait (e.g. as an accolade), and the label was reserved for those that were different from the rest (Peterson & Seligman, 2004; Pury & Starkey, 2010; Rate, 2010). These labels, however, often applied vague descriptions that could not be clearly applied, and the definitions did not accurately identify those commonly believed to possess courage (Peterson & Seligman, 2004; Pury & Starkey, 2010; Rate, 2010). For this reason, definitions were developed that treated courage as a process.

Definitions of courage as a process can be seen across several disciplines (MacIntyre, 1981; McCain & Salter, 2004; Rate, 2010), and they are commonly used in descriptions of awards for courage (i.e. The Carnegie Medal, The Courage in Journalism Awards; Pury & Starkey, 2010). Of these definitions, Rate and colleagues′ (2007; Rate, 2010) defi-nition is perhaps the most popular and supported, which

THE JOURNAL OF POSITIVE PSYCHOLOGY 3

Physical courage is an intentional, deliberate, and prosocial behavior in which the risks involved are to the actor’s physical well-being (Clancy, 2003; Olsthoorn, 2007; Woodard & Pury, 2007). While physical courage was reg-ularly discussed in early works, as these works often dis-cussed courage on the battlefield (Hobbs, 2000; Snyder et al., 2010), recent authors have argued that it may not be necessary for modern society – especially the workplace (Clancy, 2003; Kidder, 2003; Lachman, 2007). These authors argue that modern employees rarely need to sacrifice their physical well-being for the betterment of their company or coworkers (Clancy, 2003; Kidder, 2003; Lachman, 2007). While this point is arguable, it emphasizes that physical courage may be less important to organizations than other courage dimensions.

Next, moral courage is often discussed in research, but a widely-supported definition has yet to be provided for the construct. Some definitions do not clearly distinguish moral courage from general courage or other ethics-re-lated individual differences (Brooks & Edwards, 2009; Niesta, Greitemeyer, Fischer, & Frey, 2010). Examples of these definitions are ‘willingness to take a controversial stand’ (Corley, 2002, p. 647) or ‘acting according to val-ues’ (Callister & Sudia-Robinson, 2011, p. 3). Other defini-tions define moral courage as overcoming risks to others (Simola, 2015), addressing norm violations (Baumert et al., 2013; Halmburger, Baumert, & Schmitt, 2015), expressing authenticity (Pianalto, 2012; Worline, 2012), or even cour-age in extreme situations (Comer & Vega, 2015; Oliner, 1991). Without a widely-supported construct definition, it is difficult to empirically study moral courage, and other dimensions of courage may be more appropriate to study. It should be noted, however, that moral courage is likely an important dimension to study, especially once a definition has been more widely agreed-upon.

Lastly, social courage is an intentional, deliberate, and prosocial behavior in which the risks involved are to the actor’s esteem in the eyes of others (Howard et al., 2016). Social courage is often included in authors’ proposed dimensions of courage (May, 1994; Schilpzand, 2008; Woodard & Pury, 2007), which provides support for the validity of the construct. Also, social courage has been sug-gested to be the most important dimension of courage to the modern workplace (Geller & Veazie, 2009; Howard et al., 2016; Koerner, 2014; Worline, 2012). Employees are regularly tasked with risking their friendships with coworkers and/or social image for the betterment of their organization, coworkers, or even self. For instance, many employees are afraid to give presentations at work (Geller & Veazie, 2009; Koerner, 2014; White et al., 2015), but the results of a presentation may significantly impact the direction of a work unit or even organization. Further,

research has already empirically shown that social courage is significantly related to important workplace outcomes, and a psychometrically sound and valid measure exists to measure this dimension of courage in the context of the workplace (Howard et al., 2016). Thus, social courage is important to organizations, and it can be empirically studied using existing measures.

Due to these considerations, the current article studies the antecedents of behavioral social courage. It is believed that the antecedents may also relate to other dimensions of courage, but any findings would still be important due to their relation to social courage alone.

Categories of antecedents

Many motivational theories (i.e. expectancy theory; Vroom & Jago, 1988, 2007; Vroom & Yetton, 1973), leadership the-ories (i.e. path-goal theory; House, 1971; House & Mitchell, 1975), need-based theories (i.e. Maslow’s hierarchy; Maslow, 1943, 2013), and a host of other theoretical per-spectives all propose that people weigh the costs and ben-efits of their goal-directed actions. Accordingly, cognitive mediational theory suggests that this cognitive appraisal – which may happen automatically and subconsciously – is a response to a stimulus in the environment that may require action (Lazarus, 1991). This appraisal then elicits a physiological arousal and an emotional response, which may also influence subsequent emotional responses if the stimulus returns. Current empirical research has repeatedly supported these notions and shown that perceptions of benefits and risks are a primary determinant of goal-di-rected risky behaviors (Foster, Shenesey, & Goff, 2009; Moore & Gullone, 1996; Zhang, Zhang, & Shang, 2016).

Similar dynamics may be true for courage. Benefits and risks are central to most all definitions and theoreti-cal models of courage (Hannah, Sweeney, & Lester, 2007; Koerner, 2014; Quinn & Worline, 2008; Rate et al., 2007; Schilpzand et al., 2014; Worline, 2004, 2012). Sekerka and Bagozzi (2007), for example, suggest that an initial step to performing a courageous action is having an affective reaction and cognitively processing information. Affective responses are greatly influences by prior beliefs regarding associated risks and benefits, which results in cognitive processes that more directly weigh these two aspects. From these perspectives, people may habitually perform courageous behaviors because they particularly value the benefits and/or devalue the risks involved, which has been initially supported by Brandstätter and colleagues (2016). These authors viewed courage as a goal-directed process, in which people must perceive the severity of a norm vio-lation that can be rectified (e.g. potential benefits) and the extent of risk involved with intervening. The authors then

4 M. C. HOWARD AND J. E. COGSWELL

2010). Job characteristics theory (Hackman & Oldham, 1976; Oldham & Hackman, 2010) and various work design models (Fried & Ferris, 1987; Humphrey, Nahrgang, & Morgeson, 2007; Morgeson & Humphrey, 2006) suggest that certain aspects of employees’ work environment directly influence their affect, attitudes, intentions, and behaviors (Humphrey et al., 2007; Morgeson & Humphrey, 2006; Oldham & Hackman, 2010). With regard to behav-ioral social courage, social support may be a particularly relevant job characteristic.

Employees with high levels of social support may view their relationships as a resource to buffer against such con-sequences. An employee with few supportive relationships may be unwilling to damage their few social relationships, whereas an employee with many supportive relationships may believe that their social support can withstand short-term conflicts. Also, employees with ample social support may believe that their strong social networks can alleviate any damage to their ‘esteem in the eyes of others.’ Thus, we propose:

Hypothesis 2: Social support is positively related to behav-ioral social courage.

The third category is leadership. Certain leadership styles may influence followers’ expression of traits (Afolabi, Adesina, & Aigbedion, 2009; Colbert & Witt, 2009; Perry, Witt, Penney, & Atwater, 2010). For instance, Colbert and Witt (2009) empirically demonstrated that goal-focused leadership helps enable the expression of subordinates’ conscientiousness. Schilpzand and colleagues (2014) specifically propose that this process occurs with coura-geous behaviors via self-perceptions. That is, certain exter-nal influences, including leadership, may elicit positive self-perceptions, which may cause employees to perform courageous behaviors – possibly because these employees feel more resilient to the risks involved with these behav-iors. More directly, a host of prior authors have suggested that leadership plays a primary role in the performance of courageous behaviors (Hannah et al., 2013; Hannah et al., 2007; Sekerka & Bagozzi, 2007; Sekerka et al., 2009). Leaders have the power to reduce the risks involved with courageous action, such as preventing retaliatory conse-quences, but leaders can also amplify the risks involved with courageous actions, such as belittling or directly punishing (Hannah et al., 2013; Snyder, 2010; Terry, 1993).

Positive aspects of leadership, such as ethical and empowering styles, have both been suggested to influ-ence follower proactive and prosocial behaviors, of which courageous behaviors are included (Hannah et al., 2011; Waldman, Siegel, & Javidan, 2006; Zhu, May, & Avolio, 2004). On the other hand, abusive supervision has been suggested to reduce proactive and prosocial behaviors (Conger, 1999; Hannah et al., 2013; Waldman et al., 2006; Zhu et al., 2004). Thus, we propose:

showed that certain values and self-regulatory affective traits influence these perceptions, but no indicators of courage were measured.

Consistent  with these perspectives and findings, we argue that cognitive appraisals of the potential risks and benefits of social courage behaviors are a critical step in the process of enacting courage in the social context. To illus-trate this point, we consider the following example from the Workplace Social Courage Scale (WSCS; Howard et al., 2016): ‘Although it may damage our friendship, I would tell my superior when a coworker is doing something incor-rectly.’ The respondent has witnessed something that may result in negative consequences for the organization, and the courageous action is to tell the supervisor of this incor-rect behavior. In deciding whether to act, the person must weigh the perceived risk (i.e. damaging a friendship with a coworker) with the potential benefit (i.e. remedy of duress) of performing the action.

Based on this proposal, we identify antecedents that may influence how people appraise risks and benefits associated with social courage behaviors. To study social courage in a systematic manner, the current article iden-tifies five categories of antecedents. These categories are based on prior discussions of courage’s antecedents (Schilpzand et al., 2014) as well as prior investigations into the antecedents of similar constructs, such as psychologi-cal capital (Avey, 2014; John & Srivastava, 1999; Mowen & Carlson, 2003; Thatcher & Perrewe, 2002).

The first category of antecedent is personality. Other individual differences may impact the decision to perform a courageous behavior aside from trait courage – par-ticularly those that describe a tendency to work through difficulties despite personal risks (i.e. risk of failure). Grit describes a predisposition toward perseverance and pas-sion for long-term goals (Duckworth, Peterson, Matthews, & Kelly, 2007). Those high in grit may be more likely to persist through perceived risks (i.e. damage to friendship) in the interest of accomplishing longer-term goals, such as the alleviation of duress in the organization. Proactive individuals are similarly known to persist through difficul-ties (Bateman & Crant, 1993). As part of the appraisal pro-cess (Lazarus, 1991), individuals assess whether they have sufficient resources to meet the threat. High-grit and/or proactive individuals may be more likely to judge their psy-chological resources sufficient to withstand any adverse consequences of courageous action. Thus, we propose the following hypothesis:

Hypothesis 1: (a) Grit and (b) proactive personality are positively related to behavioral social courage.

The second category is job design characteristics. Job characteristics are aspects of employees’ work environ-ment, and include autonomy, skill variety, task identity, and others (Hackman & Oldham, 1976; Oldham & Hackman,

THE JOURNAL OF POSITIVE PSYCHOLOGY 5

legally protected class status, then its inclusion in selec-tion batteries could result in disparate impact – a concern that would need to be addressed. For these reasons, we propose that:

Hypothesis 5: (a) Gender, (b) age, and (c) tenure relate to behavioral social courage

Together, these five categories of antecedents (per-sonality, job characteristics, leadership, culture, and demographic characteristics) are investigated to prompt a more systematic study of the left side of social courage. We recognize, however, that each of these categories contain other possible antecedents that were not directly hypothesized. Therefore, we also investigate other possible antecedents from these categories without firm a priori hypotheses.

Research Question 1: Which other antecedents relate to behavioral social courage?

Mediators

The current article proposed that each antecedent influ-ences behavioral social courage via perceived risks and benefits. For this reason, we test whether perceived risks and benefits are dual mediators between various anteced-ents and behavioral social courage. By doing so, we can provide a more in-depth understanding regarding the role of perceived risks and benefits in social courage actions, which may also provide insights on courage as a goal-di-rected process.

Hypothesis 6: Perceived risks and benefits mediate the relationship between antecedents and behavioral social courage.

Overview of studies

The current article investigates five categories of courage antecedents, each with several predictors. Participant fatigue could be a concern if each of these antecedents were analyzed simultaneously, which would result in random responses, dropout, and biased results (Blasius & Thiessen, 2015; Egleston, Miller, & Meropol, 2011). For this reason, the current article provides three studies. Personality, job characteristics, and demographics are analyzed together in Study 1. Leadership, culture, and demographics are analyzed together in Study 2. In Study 3, the significant antecedents of Studies 1 and 2 are tested together along with perceived courage benefits and risks. Through this three-study process, each predictor can be studied together with reduced participant fatigue, result-ing in a more accurate assessment of courage and its antecedents.

Hypothesis 3: (a) Ethical leadership and (b) empowering leadership are positively related to behavioral social courage. (c) Abusive supervision is negatively related to behavioral social courage.

The fourth category is cultural influences. Culture represents a belief system that is shared by members of a group (Hofstede, 2006; Hofstede & Hofstede, 2001; House, Hanges, Javidan, Dorfman, & Gupta, 2004). While cultural influences are often analyzed across groups, such as comparing participants from Eastern and Western cul-tures, cultural beliefs may vary within a single culture (Avey, 2014; Kashima et al., 2004; Schug, Yuki, & Maddux, 2010). Further, culture has a strong influence on group members’ typical patterns of behaviors, which may even include courageous behaviors as suggested by Sekerka and Bagozzi (2007). While several different aspects of cul-ture may influence courage, the most proximal are likely those related to interpersonal interaction (Hobfoll, 2001; Hofstede, 2006; Hofstede & Hofstede, 2001; Hofstede & McCrae, 2004; House et al., 2004). One such cultural influence is humane orientation. Cultures with a humane orientation reward each other for fair and altruistic treat-ment. In these cultures, members may feel that retaliation for social courage behaviors is unfair, as these behaviors have a prosocial intent – thereby promoting the perfor-mance of such behaviors. Likewise, members of these cultures particularly value the prosocial treatment of oth-ers, and social courage may be actively promoted due to the helping nature of this type of behavior. For this reason, we propose:

Hypothesis 4: Humane orientation positively relates to behavioral social courage.

Lastly, the fifth category is demographic characteris-tics. Throughout history, courage has often been treated as a gendered concept (Hobbs, 2000; Pury & Lopez, 2010; Snyder et al., 2010). That is, men are expected to express courageous behaviors, such as rushing into bat-tle, whereas women are not expected to perform such behaviors. While this notion is largely based on con-jecture and speculation (Pury & Lopez, 2010; Snyder et al., 2010), gender may still influence the expression of behavioral courage through social conditioning. Also, employees may grow bolder and become more cou-rageous as they spend more time in an organization, causing age and/or tenure to also relate to behavioral courage. While organizations may be largely unable to influence demographic characteristics to prompt courage, these relationships could still have important implications. Particularly, organizations could include a measure of courage in selection batteries. If courage is related to certain demographic characteristics, such as a

6 M. C. HOWARD AND J. E. COGSWELL

whatever I begin’ and ‘My interests change from year to year.’ Its Cronbach’s alpha was 0.89.

Proactive personality. Proactive personality was measured via Bateman and Crant’s (1993) 17-item measure. Example items are, ‘If I see something I don’t like it, I fix it’ and ‘I excel at identifying opportunities.’ Its Cronbach’s alpha was 0.94.

Job characteristicsAutonomy, task variety, task significance, task identity, job complexity, skill variety, and social support. All job characteristics were measured via Morgeson and Humphrey’s (2006) measures. Example items are, ‘The job allows me to make a lot of decisions on my own’ (autonomy), ‘The job involves a great deal of task variety’ (task variety), ‘The job has a large impact on people outside the organization’ (task significance), ‘The job allows me to complete work I start’ (task identity), ‘The job comprises relatively uncomplicated tasks’ (job complexity; reverse scored), and ‘The job requires a variety of skills’ (skill variety), and ‘People I work with take a personal interest in me’ (social support). The Cronbach’s alphas of the measures were as follows: autonomy α = 0.95 (9 items), task variety α = 0.96 (4 items), task significance α = 0.93 (4 items), task identity α = 0.93 (4 items), job complexity α = 0.89 (4 items), skill variety α = 0.93 (4 items), and social support α = 0.88 (6 items).

DemographicsGender. Gender was coded as ‘0’ for male and ‘1’ for female. Too few participants responded ‘other’ and provided their gender to be included in analyses.

Age. Age was recorded by asking participants: ‘What is your current age in years?’

Tenure. Tenure was recorded by asking participants: ‘How long have you been working for your current organization in years?’

Results

Before conducting our primary analyses, we investigated the measurement properties of our scales – particularly the item factor loadings and scale discriminant validity. Using structural equation modeling (SEM), we first tested a meas-urement model using a total disaggregation method to create factors, which allowed us to inspect individual item loadings. No items had any factor loadings that were con-cerning (< 0.35). Because our sample size did not meet prior suggestions for SEM, and thereby prevented us to draw overall inferences regarding model fit and comparisons,

Study 1

Method

ParticipantsParticipants (N  =  222, Mage  =  35.43, SDage  =  10.33, 40% female, 95% American) were recruited from MTurk and provided $1.00 of monetary compensation, and 61% participated in the Time 2 survey. MTurk is a website that connects individuals willing to perform tasks on a com-puter, such as taking a survey, with those needing the tasks completed. Several prior studies have shown that results obtained from MTurk samples are reliable and valid, even when studying special populations (Buhrmester, Kwang, & Gosling, 2011; Paolacci & Chandler, 2014; Shapiro, Chandler, & Mueller, 2013). All participants were employed (89% full-time, Mtenure = 5.70, SDtenure = 4.98) in a wide range of industries (24% Other, 19% Business & Information, 13% Education, 10% Finance & Insurance), and those that failed any attention check were removed (31 participants). All statistics, including the sample size reported above, reflect the sample after removing these participants.

ProcedureParticipants signed-up to participate in Study 1 by self-se-lecting based on a short description presented on MTurk. They provided their digital informed consent and com-pleted the first survey online. The first survey included all measures except behavioral social courage. Two days later, they were sent a second survey, which was intended to reduce method biases that could be involved with purely cross-sectional studies. The second survey only included behavioral social courage. Then, participants were debriefed about the study.

MeasuresBehavioral social courage. Behavioral social courage was measured via Howard and colleagues’ (2016) 11-item Workplace Social Courage Scale (WSCS) with modified instructions. Example items are, ‘I would not tolerate when a coworker is rude to someone, even if I make him/her upset’ and ‘Although it makes me look incompetent, I would tell my coworkers when I’ve made a mistake.’ Instead of the traditional instructions, which asks participants to respond independently of their current workplace, participants were told to respond in regards to their behaviors within their current workplace. This method is believed to assess behavioral social courage rather than trait social courage. The scale’s Cronbach’s alpha was 0.87.

PersonalityGrit. Grit was measured via Duckworth and colleagues (2007) 12-item measure. Example items are, ‘I finish

THE JOURNAL OF POSITIVE PSYCHOLOGY 7

support Hypotheses 2 (social support), 5a (gender), 5b (age), and 5c (tenure).

Discussion

These results suggest that other personality traits (grit and proactive personality) may indeed have a significant influence on social courage, whereas job characteristics may have a much smaller influence – if any at all. Before discussing this finding in depth, however, Study 2 investi-gates the relationship of leadership and cultural influences with social courage.

Study 2

Method

ParticipantsParticipants (N  =  215, Mage  =  33.19, SDage  =  10.51, 40% female, 88% American) were recruited from MTurk and provided $0.95 of monetary compensation, and 53% participated in the Time 2 survey. All participants were employed (86% full-time, Mtenure  =  5.24, SDtenure  =  4.81) in a wide range of industries (20% Other, 16% Business & Information, 11% Finance & Insurance, 10% Education, 10% Food & Hospitality), and those that failed any atten-tion check were removed (21 participants). All statistics, including sample size reported above, reflect the sample after removing these participants.

ProcedureParticipants signed-up to participate in Study 2 by self-se-lecting based on a short description presented on MTurk. They provided their digital informed consent and com-pleted the first survey online. The first survey included all

we then tested a measurement model using a parceling method to create factors. This has been suggested to ana-lyze overall fit and comparisons when sample sizes do not meet traditional cutoffs (Bagozzi & Edwards, 1998; Kishton & Widaman, 1994; Little, Cunningham, Shahar, & Widaman, 2002). Each factor was represented by two or three par-cels, which were created by alternating items selected to each parcel. When testing this measurement model, the overall model fit was appropriate (CFI = 0.96, TLI = 0.95, SRMR = 0.04, RMSEA = 0.06, χ2/df < 2), and model fit could not be improved by replacing two factors with a single factor. This supports the discriminant validity of the scales, and suggests that the underlying constructs are unique.

Correlations of all study variables are included in Table 1. Behavioral social courage had a significant relationship with grit (r = 0.47, p < 0.01), proactive personality (r = 0.57, p < 0.01), and all job characteristics (autonomy, r = 0.32, p < 0.01; task variety, r = 0.32, p < 0.01; task significance, r = 0.27, p < 0.01; task identity r = 0.22, p < 0.05; task com-plexity, r = −0.21, p < 0.05; skill variety, r = 0.30, V < 0.01; social support, r = 0.45, p < 0.01). Behavioral social cour-age did not have a significant relationship with any demo-graphic variables. A regression analysis was performed to analyze the joint effect of all predictors in predicting behavioral social courage (Table 2). No variance inflation factor (VIF) was greater than three, which indicates that multicollinearity does not pose concerns with this regres-sion. When analyzed together, only proactive personality (β = 0.350, t = 3.980, p < 0.01) and grit (β = 0.198, t = 2.333, p < 0.05) were significant in predicting behavioral social courage. Of the job characteristics, social support was the closest to approach statistical significance (β = 0.132, t  =  1.458, p  >  0.05). These results support Hypothesis 1a (grit) and 1b (proactive personality), but they fail to

Table 1. correlations and cronbach’s alphas of study 1 measures.

*p < 0.05; **p < 0.01.

  1 2 3 4 5 6 7 8 9 10 11 12 13Outcomes                          (1) Behavioral social

courage0.87                        

Personality                          (2) grit 0.47** 0.89                      (3) Proactive personality 0.57** 0.48** 0.94                    Job Characteristics                          (4) autonomy 0.32** 0.23** 0.32** 0.95                  (5) Task variety 0.32** 0.23** 0.37** 0.37** 0.96                (6) Task significance 0.27** 0.14* 0.36** 0.18** 0.38** 0.93              (7) Task identity 0.22* 0.19** 0.29** 0.11 0.12 0.19** 0.93            (8) Task complexity 0.21* 0.24** 0.10 0.29** 0.46** 0.15* −0.17* 0.89          (9) skill variety 0.30** 0.25** 0.32** 0.40** 0.65** 0.35** 0.09 0.60** 0.93        (10) social support 0.45** 0.34** 0.47** 0.35** 0.41** 0.41** 0.16* 0.27** 0.48** 0.88      Demographics                          (11) age 0.06 0.16* 0.05 0.01 0.02 0.08 0.06 0.10 0.10 0.07 n/a    (12) Tenure 0.09 0.08 0.02 0.18** 0.04 0.10 −0.01 0.18** 0.23** 0.16* 0.49** n/a  (13) gender −0.12 −0.01 −0.02 −0.07 −0.00 −0.07 −0.06 −0.04 −0.04 −0.05 0.19** 0.06 n/a

8 M. C. HOWARD AND J. E. COGSWELL

from the GLOBE studies (House et al., 2004). Example items are, ‘In this society, rank and position in the hierarchy have special privileges’ (power distance), ‘This society has rules or laws to cover most situations’ (uncertainty avoidance), ‘In this society, leaders encourage group loyalty even if individual goals suffer’ (collectivism), ‘In this society, people are generally very concerned about others’ (humane orientation), and ‘In this society, people are generally assertive’ (assertiveness). The Cronbach’s alphas of the measures were as follows: power distance α  =  0.81 (5 items), uncertainty avoidance α  =  0.71 (4 items), collectivism α = 0.57 (4 items), humane orientation α = 0.88 (5 items), and assertiveness α = 0.84.

DemographicsGender, age, and tenure. Demographics were measured with the items from Study 1.

Results

Again, we first tested a measurement model using a total disaggregation method to create factors. One item had a concerning factor loading (< 0.35), which was in the col-lectivism scale. This item was removed for all subsequent analyses. We then tested a measurement model using a parceling method to create factors. The overall model fit was appropriate (CFI = 0.97, TLI = 0.96, SRMR = 0.04, RMSEA  =  0.06, χ2/df  <  2), and model fit could not be improved by replacing two factors with a single factor. This supports the discriminant validity of the scales.

Correlations of all study variables are included in Table 3. Behavioral social courage had a significant relationship with all three leadership styles: ethical leadership (r = 0.35, p < 0.01), empowering leadership (r = 0.43, p < 0.01), and abusive supervision (r = −0.30, p < 0.01). It also had a signif-icant relationship with most cultural influences, including power distance (r = 0.23, p < 0.05), uncertainty avoidance (r = 0.26, p < 0.01), and collectivism (r = 0.24, p < 0.01). It did not have a significant relationship with tenure (r = 0.15, p > 0.05) or gender (r = 0.07, p > 0.05), but it did have a sig-nificant relationship with age (r = 0.22, p < 0.05). A regres-sion analysis was performed to analyze the joint effect of all predictors in predicting behavioral social courage (Table 4). No VIF was greater than four, which indicates that multicollinearity does not pose concerns. When analyzed together, only empowering leadership (β = 0.357, t = 2.388, p < 0.05), power distance, (β = 0.283, t = 3.044, p < 0.01), and age (β = 0.190, t = 2.012, p < 0.05) were significant in predicting behavioral social courage. These results support Hypotheses 3b (empowering leadership) and 5b (age), but they fail to support Hypotheses 3a (ethical leadership), 3c (abusive supervision), 4 (humane orientation), 5a (gender), and 5c (tenure).

measures except behavioral social courage. Two days later, they were sent a second survey, which was intended to reduce method biases that could be involved with purely cross-sectional studies. The second survey only included behavioral social courage. Then, they were debriefed about the study.

MeasuresSocial courage. Behavioral social courage was measured in the same manner as Study 1.

LeadershipEthical leadership. Ethical leadership was measured via Brown and colleagues’ (2005) 10-item measure. Participants respond to items in regards to their current supervisor or manager, and example items are, ‘Can be trusted’ and ‘Makes fair and balanced decisions.’ Its Cronbach’s alpha was 0.94.

Empowering leadership. Empowering leadership was measured via Zhang and Bartol’s (2010) 12-item measure. An example item is, ‘My manager makes many decisions together with me’ and ‘My manager allows me to do my job my way.’ Its Cronbach’s alpha was 0.94.

Abusive supervision. Abusive supervision was measured via Tepper’s (2000) 15-item measure. Participants respond to items in regards to their current supervisor or manager, and an example item is, ‘Ridicules me’ and ‘Lies to me.’ Its Cronbach’s alpha was 0.96.

Cultural influencesPower distance, uncertainty avoidance, collectivism, humane orientation, and assertiveness. All cultural influences were measured via slightly modified measures

Table 2. study 1 regression results.

*p < 0.05; **p < 0.01.

  Behavioral social courage

  B S.E. β t

constant 1.573 0.617   2.551Personality        (1) grit 0.167 0.072 0.198 2.333*(2) Proactive personality 0.322 0.081 0.350 2.980**Job Characteristics        (3) autonomy 0.052 0.052 0.079 1.005(4) Task variety 0.032 0.057 0.055 0.566(5) Task significance 0.003 0.042 0.005 0.063(6) Task identity 0.041 0.044 0.072 0.945(7) Task complexity 0.010 0.051 0.02 0.201(8) skill variety −0.015 0.075 −0.022 −0.199(9) social support 0.102 0.070 0.132 1.458Demographics        (10) age 0.004 0.007 0.050 0.594(11) Tenure 0.003 0.015 0.015 0.175(12) gender −0.096 0.124 −0.055 −0.774R2       0.42

THE JOURNAL OF POSITIVE PSYCHOLOGY 9

survey, 53% participating in the Time 3 survey, and 46% participating in the Time 4 survey. All participants were employed (84% full-time, Mtenure = 5.41, SDtenure = 4.87) in a wide range of industries (26% Other, 17% Business & Information, 11% Education, 10% Finance & Insurance, 10% Food & Hospitality), and those that failed any atten-tion check were removed (16 participants). All statistics, including sample size reported above, reflect the sample after removing these participants.

ProcedureParticipants signed-up to participate in Study 3, which was a four-day study, by self-selecting based on a short descrip-tion presented on MTurk. They provided their informed con-sent and completed the first survey online, which included demographic information alone. On the second day, partic-ipants were sent a second survey that included all anteced-ents. On the third day, participants were sent a third survey that included perceived courage benefits and perceived courage risks. On the fourth day, participants were sent a fourth survey that included behavioral social courage. Then, they were debriefed about the study. Like Studies 1 and 2, the time-separated nature of Study 3 was intended to reduce the method biases involved with purely cross-sec-tional studies, and allow for more accurate inferences regarding the relationships of interest. Similarly, temporal precedence is encouraged for tests of mediation, which can be inferred through this time-separated research design.

MeasuresFor each of the following variables, the same measures from Studies 1 or 2 were applied: grit, proactive person-ality, social support, empowering leadership, power dis-tance, age, tenure, gender, and behavioral social courage. Although the effect of social support was not significant in Study 1, it was included in Study 3 because it was the

Discussion

These results suggest that certain leadership styles (empowering leadership) and aspects of culture (power distance) may indeed have a significant influence on social courage. These results also question the findings of Study 1 in regards to the influence of demographic characteristics, and age may indeed have an effect on behavioral social courage. Before discussing these results, Study 3 reanaly-zes and extends the findings of Studies 1 and 2.

Study 3

Method

ParticipantsParticipants (N  =  395, Mage  =  32.97, SDage  =  9.49, 41% female, 91% American) were recruited from MTurk and provided $1.21 in monetary compensation. Study 3 con-sisted of four waves, with 63% participating in the Time 2

Table 3. correlations and cronbach’s alphas of study 2 measures.

*p < 0.05; **p < 0.01.

  1 2 3 4 5 6 7 8 9 10 11 12Outcomes                        (1) Behavioral social courage 0.87                      Leadership                        (2) ethical leadership 0.35** 0.94                    (3) empowering leadership 0.43** 0.83** 0.94                  (4) abusive supervisor −0.30** −0.59** −0.50** 0.96                Cultural influences                        (5) Power distance 0.23* 0.04 0.05 −0.01 0.81              (6) uncertainty avoidance 0.26** 0.23** 0.26** −0.07 0.28** 0.71            (7) collectivism 0.23* 0.12 0.09 0.02 0.45** 0.44** 0.68          (8) humane orientation 0.08 0.14* 0.20** 0.04 −0.21** 0.14* −0.00 0.88        (9) assertiveness 0.08 0.12 0.09 0.07 0.27** 0.27** 0.10 −0.05 0.84      Demographics                        (10) age 0.22* 0.08 0.11 −0.13 0.03 0.06 0.07 0.03 −0.01 n/a    (11) Tenure 0.15 0.03 0.10 −0.05 −0.06 0.04 0.08 0.11 −0.02 0.50** n/a  (12) gender 0.07 −0.04 −0.07 0.07 0.07 −0.03 0.03 −0.09 0.14* 0.09 0.00 n/a

Table 4. study 2 regression results.

*p < 0.05; **p < 0.01.

 

Behavioral Social Courage

B S.E. β t

constant 1.334 0.791   1.687Leadership        (1) ethical leadership −0.032 0.117 −0.043 −0.272(2) empowering leadership 0.272 0.114 0.357 2.388*(3) abusive supervision −0.123 0.100 −0.134 −1.231Cultural influences        (4) Power distance 0.251 0.082 0.283 3.044**(5) uncertainty avoidance 0.091 0.079 0.105 1.151(6) collectivism 0.135 0.081 0.150 1.675(7) humane orientation 0.050 0.063 0.075 0.798(8) assertiveness −0.055 0.066 −0.075 −0.826Demographics        (9) age 0.015 0.008 0.190 2.012*(10) Tenure 0.002 0.018 0.012 0.126(11) gender 0.028 0.149 0.016 0.186R2       0.35

10 M. C. HOWARD AND J. E. COGSWELL

overall model fit was appropriate (CFI = 0.97, TLI = 0.96, SRMR = 0.03, RMSEA = 0.06, χ2/df < 2), and model fit could not be improved by replacing two factors with a single factor. This supports the discriminant validity of the scales.

Correlations of all study variables are included in Table 5. Behavioral social courage had significant relationships with grit (r = 0.55, p < 0.01), proactive personality (r = 0.62, p < 0.01), social support (r = 0.37, p < 0.01), empowering leadership (r = 0.47, p < 0.01), age (r = 0.27, p < 0.01), per-ceived courage benefits (r = 0.43, p < 0.01), and perceived courage risks (r = −0.18, p < 0.05). It did not have a significant relationship with power distance (r = 0.10, p > 0.05), human orientation (r = 0.10, p > 0.05), tenure (r = 0.12, p > 0.05), or gender (r = 0.08, p > 0.05). A regression analysis was per-formed to analyze the joint effect of all predictors in predict-ing behavioral social courage (Table 6). No VIF was greater than three, which indicates that multicollinearity does not pose concerns. When analyzed tougher, grit (β  =  0.211, t  =  3.076, p  <  0.01), proactive personality (β  =  0.279, t = 3.730, p < 0.01), age (β = 0.230, t = 3.841, p < 0.01), per-ceived courage benefits (β = 0.242, t = 4.267, p < 0.01), and perceived courage risks (β = −0.158, t = −2.954, p < 0.01) had a significant relationship with behavioral social courage. It should be noted, however, that empowering leadership (β = 0.139, t = 1.929, p > 0.05) and power distance (β = 0.104, t = 1.967, p > 0.05) closely approached significance with p-values of below 0.06. These results support Hypotheses 1a (grit), 1b (proactive personality), 5b (age). They partially support Hypothesis 3b (empowering leadership). These results do not support Hypotheses 2 (social support), 3a (ethical leadership), 3c (abusive supervision), 4 (humane orientation), 5a (gender), 5c (tenure).

Bootstrapped estimates of indirect effects and con-fidence intervals were calculated to test for mediating effects  (Hayes, 2013; Preacher & Hayes, 2004), and only

job characteristic that most closely approached statistical significance.

Perceived courage benefits. A method was adapted from prior research to measure perceived courage benefits (Foster et al., 2009; Moore & Gullone, 1996; Zhang et al., 2016). Participants were presented the items of the WSCS; however, instead of reporting their level of agreement to each item, they were instructed to indicate their perceived benefits involved with each item. While not directly specified to respond in regards to one’s current workplace, the instructions directed participants to respond in regards to benefits for ‘yourself, your coworkers, your organization, and other relevant entities.’ The provided rating scale ranged from 1 (No benefit at all) to 7 (As beneficial as possible). The scale’s Cronbach’s alpha was 0.90.

Perceived courage risks. The same method was adapted from prior research to measure perceived courage risks (Foster et al., 2009; Moore & Gullone, 1996; Zhang et al., 2016). Participants were instructed to report their perceived risks involved with each item on the WSCS, and they were directed to respond in regards to ‘how risky the behavior is for yourself.’ The provided rating scale ranged from 1 (No risk at all) to 7 (As risky as possible). The scale’s Cronbach’s alpha was 0.91.

Results

We first tested a measurement model using a total dis-aggregation method to create factors. One item had a concerning factor loading (< 0.35), which was in the proactive personality scale. This item was removed for all subsequent analyses. We then tested a measurement model using a parceling method to create factors. The

Table 5. correlations and cronbach’s alphas of study 3 measures.

*p < 0.05; **p < 0.01.

  1 2 3 4 5 6 7 8 9 10 11Outcomes                      (1) Behavioral social courage 0.86                    Mediators                      (2) Perceived courage benefits 0.43** 0.90                  (3) Perceived courage risks −0.18* 0.08 0.91                Personality                      (4) grit 0.55** 0.17* −0.14* 0.90              (5) Proactive personality 0.62** 0.39** −0.12* 0.564** 0.94            Job characteristics                      (6) social support 0.37** 0.21** −0.11 0.32** 0.43** 0.89          Leadership                      (7) empowering leadership 0.47** 0.32** −0.07 0.36** 0.47** 0.64** 0.94        Cultural influences                      (8) Power distance 0.10 0.06 0.13 −0.07 0.02 −0.08 −0.02 0.77      Demographics                      (9) age 0.27** 0.03 0.00 0.24** 0.07 0.07 0.09 −0.02 n/a    (10) Tenure 0.12 0.01 0.04 0.25** 0.07 0.08 0.14* −0.08 0.53** n/a  (11) gender 0.08 0.09 0.05 0.04 0.03 −0.02 −0.01 0.14* 0.08 −0.04 n/a

THE JOURNAL OF POSITIVE PSYCHOLOGY 11

Discussion

These results suggest that personality (grit and proac-tive personality) and demographics (age) have a signifi-cant influence behavioral social courage, and the effect of leadership (empowering leadership) and cultural influences (power distance) should not be entirely dis-counted. The effect of job characteristics on behavioral social courage, however, was not supported. Also, per-ceived social courage benefits and risks did not have a significant mediating effect for most predictors, although both had significant direct effects on behavioral social courage. A visual representation of all Study 3 results is presented in Figure 1.

mediating effects were tested with antecedents that had a significant relationship with the mediators (perceived ben-efits and/or risks). From a regression analysis including all antecedents, only proactive personality had a significant relationship with perceived courage benefits and/or risks (Supplemental Material A). When controlling for all other antecedents, the indirect effect of proactive personality on behavioral social courage through perceived benefits was significant (z = 0.067, 95% C.I. [0.019, 0.149]), and the direct effect was still significant (t = 4.080, 95% C.I. [0.145, 0.416]). This result suggests that proactive personality has a signifi-cant, partially mediated effect on behavioral social courage through perceived courage benefits. Nevertheless, these results did not support Hypothesis 6 overall.

Table 6. study 3 regression results for behavioral social courage.

*p < 0.05; **p < 0.01.

 

Step 1 Step 2

B S.E. β t B S.E. β t

constant 0.721 0.428   1.684 0.725 0.432   1.679Personality                (1) grit 0.180 0.059 0.220 3.039** 0.181 0.056 0.220 3.247**(2) Proactive personality 0.325 0.067 0.368 4.816** 0.238 0.066 0.270 3.630**Job characteristics                (3) social support 0.032 0.056 0.041 0.573 0.028 0.052 0.036 0.529Leadership                (4) empowering leadership 0.132 0.061 0.166 2.179* 0.110 0.058 0.138 1.905Cultural influences                (5) Power distance 0.086 0.052 0.093 1.654 0.064 0.049 0.102 1.922Demographics                (6) age 0.020 0.006 0.218 3.402** 0.021 0.005 0.231 3.852**(7) Tenure −0.012 0.012 −0.065 −1.006 −0.010 0.011 −0.057 −0.940(8) gender 0.019 0.105 0.010 0.177 0.014 0.099 0.007 0.139Mediators                (9) Perceived courage benefits         0.198 0.046 0.245 4.309**(10) Perceived courage risks         −0.144 0.047 −0.165 −3.086**ΔR2, R2       0.50       0.07, 0.57

All solid arrows represent statistically significant relationships (p < .05), whereas dashed arrows represent relationships that approached significance (.05 < p < .06). All possible indirect effects through perceived courage benefits were statistically significant with a 95% C.I. that excludes 0.

Power Distance

Tenure

Gender

Behavioral Social Courage

Social Support

Grit

Proactive Personality

Empowering Leadership

Age

Perceived Courage Risks

Perceived Courage Benefits

Figure 1. visual representation of study 3 results.

12 M. C. HOWARD AND J. E. COGSWELL

courageous actions and supports prior research on courage as a goal-directed process (Brandstätter, Jonas, Koletzko, & Fischer, 2016). At the same time, however, perceived social courage benefits and risks did not medi-ate the relationship of most antecedents with behavioral social courage, and power distance was related to social courage while the link could not be clearly explained via perceived risks and benefits. This suggests that these antecedents may not influence social courage behaviors in the same manner that antecedents commonly influence goal-directed behavior, and the current article can provide few inferences regarding why these antecedents influence social courage. This finding is not overly surprising given the elusive nature of courage and social courage. Courage has long been considered a mysterious construct, with only recent authors identifying a definition (Rate et al., 2007; Rate 2010) and measures for particular dimensions (Howard & Alipour, 2014; Howard et al., 2016; Woodard & Pury, 2007). Part of the uncertainty regarding courage is also its relationship with other variables. For instance, Howard and colleagues (2016) predicted that social cour-age would be related to performance, but did not find sup-port for this hypothesis. Moving forward, authors should incorporate these null findings into theories surrounding courage and social courage, as these null results may be as informative as significant results.

Also, the current study demonstrated the importance of studying multiple antecedents of a common category together. When these antecedents were analyzed alone, many had significant relationships with social courage; however, when analyzed together, many of the anteced-ents were no longer significantly related. This indicates that many of the antecedents share common variance that should not be attributed to each (Podsakoff, MacKenzie, & Podsakoff, 2012), and future research on courage and social courage should likewise study multiple antecedents together.

At the same time, however, certain antecedents were still significant predictors even when analyzed together, indicating that their relationships with social courage were unique. This suggests that social courage does not have a single primary antecedent. Rather it is influenced by many antecedents from many domains, and social courage may emerge from several aspects of the self and the environment (Koerner, 2014; Pury & Lopez, 2010; Schilpzand et al., 2014).

Future research

Several avenues for future research should be discussed. It is almost certain that other antecedents significantly relate to behavioral social courage beyond those studied in the current article. Future research should strive to identify

General discussion

The current studies provide several important findings regarding the antecedents of social courage. First, behavio-ral social courage is significantly related to certain aspects of personality, leadership, culture, and demographic char-acteristics – thereby suggesting that many constructs may indeed predict employees’ social courage. Second, certain nuances were observed within each of these antecedent categories. That is, social courage was significantly related to only a few antecedents in these categories, rather than being significantly (or not significantly) related to all ante-cedents in a category. Third, when analyzed together via regression, each antecedent category only had one or two significant predictors. These significant predictors included grit, proactive personality, empowering leadership, power distance, and age; however, only grit, proactive person-ality, and age were significant when all categories were analyzed together, whereas empowering leadership and power distance approached significance. Together, these results have several implications for the continued study of courage, social courage, and beyond.

Implications

Although the current studies cannot provide decisive evi-dence for the antecedents of social courage, as discussed further below, they do provide support that certain per-sonal and organizational factors may indeed influence the social courage of employees. Some of these factors are believed to be relatively malleable, such as leadership styles. Linking these antecedents to social courage pro-vides a more complete understanding of their workplace importance, and organizations may be able to alter these antecedents to foster social courage in their employees. For example, organizations could coach their leaders to be more empowering in order to develop the social cour-age of their employees. On the other hand, other factors are believed to be more stable, such as personality. These findings likewise emphasize the importance of these ante-cedents, and organizations may select for these employee characteristics to obtain a more socially courageous work-force. Perhaps more importantly, Study 2 found a signifi-cant relationship of social courage with age and cultural influences. If organizations were to select for social cour-age, then unintentional adverse impact could occur based on age or ethnicity. Thus, organizations should strongly consider whether the benefits of social courage merit the possible concerns when included within a selection battery.

Also, perceived benefits and risks were significantly related to behavioral social courage. This suggests that people weigh these aspects before performing socially

THE JOURNAL OF POSITIVE PSYCHOLOGY 13

Sekerka & Bagozzi, 2007). Only through such focused investigations can these models be supported (or not supported), which would provide valuable steps towards identifying potential methods to develop a courageous workforce. Also, while several of these models are over a decade old, much remains unknown about these models. While the cause of these uncertainties is unclear, recent improvements regarding the conceptualization and oper-ationalization of courage may allow these models to be tested with a newfound ease. Thus, researchers should not be wary of this domain of research.

Limitations

As with any article, certain limitations should be noted. Causality cannot be firmly supported with the design of the current studies, and, while these studies may give support for possible antecedents, they cannot deci-sively identify the antecedents of behavioral social cour-age. Future research should consider the length of time required to substantively alter behavioral social courage, which may take months or even years, and use longitudi-nal research designs spanning this timespan to observe effects. It should nevertheless be emphasized that the time-separated nature of the current studies can investi-gate mediated effects more reliably than cross-sectional designs, albeit not as reliably as intensive longitudinal studies, which likewise improves the accuracy of infer-ences derived from Study 3.

Also, most of the applied measures were directly taken from prior research. The original authors of these scales provided support for their psychometric properties and validity, and the repeated applications of these scales, along with cumulative evidence, provide further sup-port for their validity (Brown, Treviño, & Harrison, 2005; Morgeson & Humphrey, 2006). On the other hand, some of these measures were only adapted from prior research, and the psychometric properties and validity of these measures cannot be guaranteed. Two of these scales meas-ured perceived social courage risks and benefits, which applied the WSCS but asked participants to report the extent of perceived risks and benefits associated with each behavior. It should be noted, however, that this method may provide benefits beyond developing a new measure altogether (Moore & Gullone, 1996; Zhang et al., 2016). This method can help ensure the content validity of the measures. Because these scales directly reference previ-ously supported items, researchers can ensure that their perceived risks and benefits measure targets appropriate behaviors. Relatedly, this method can also help ensure that the measures are relatively free from construct contamina-tion. Foster and colleagues (2009), for instance, noted that some prior risks and benefits measures may inadvertently

these other possible antecedents, which could give a more complete view of the ‘left side of courage.’

The current article investigated the direct relationship of possible antecedents with social courage. It is possible that each of these relationships, both significant and non-sig-nificant, have certain boundary conditions (Busse, Kach, & Wagner, 2016; Dawson, 2014). That is, the strength and/or direction of these relationships may change depending on a third variable. For instance, ample prior research has shown that the strength of the relationship between lead-ership characteristics and follower outcomes depends on the relationship of the leader and follower (Graen & Uhl-Bien, 1995; Schriesheim, Castro, & Cogliser , 1999). Leaders with relatively infrequent contact with followers will have a weaker influence than those with more frequent contact. For this reason, it is likely that the relationship between empowering leadership and social courage is moderated by leader-member exchange or other similar variables (Graen & Uhl-Bien, 1995; Schriesheim et al., 1999), and similar relationships likely exist between other anteced-ents and social courage.

Future research should also investigate methods to elicit social courage from employees. Among the most efficient methods may be through training programs targeting leader behaviors that enhance follower social courage, such as empowering leadership, which would result in more effective leaders as well as more socially courageous employees (Beer, Finnström, & Schrader, 2016; Santos, Caetano, & Tavares, 2015). Perhaps more importantly, how-ever, these investigations would provide more robust sup-port that these factors are antecedents of social courage, as cause-and-effect relationships could be clearly observed.

The mechanisms in which antecedents influence social courage should also be considered. Most authors have proposed that courage and social courage are closely related to decision-making and/or regulatory processes, and employees weigh the risks and benefits of a coura-geous action similar to most other goal-directed behav-iors (Brandstätter et al., 2016; Koerner, 2014; Pury & Lopez, 2010; Schilpzand et al., 2014). This notion was not sup-ported, however. Future research should apply additional theories, or create novel theories, to understand the pro-cesses that are involved with courageous and socially cou-rageous actions.

Relatedly, the current study used a general framework (perceived risks and benefits) to identify possible anteced-ents of social courage behaviors, as individual theoretical models may not be able to sufficiently explain the relation-ship of multiple antecedents from a wide array of domains. Future research, however, should perform focused inves-tigations into prior theoretical models that identify par-ticular courage antecedents and respective mechanisms (Hannah et al., 2007; Koerner, 2014; Schilpzand et al., 2014;

14 M. C. HOWARD AND J. E. COGSWELL

References

Afolabi, O., Adesina, A., & Aigbedion, A. (2009). Influence of team leadership and commitment on teamwork and conscientiousness. Journal of Social Sciences, 21(3), 211–216.

Avey, J. B. (2014). The left side of psychological capital: New evidence on the antecedents of PsyCap. Journal of Leadership & Organizational Studies, 21(2), 141–149.

Bagozzi, R. P., & Edwards, J. R. (1998). A general approach for representing constructs in organizational research. Organizational Research Methods, 1(1), 45–87.

Bateman, T. S., & Crant, J. M. (1993). The proactive component of organizational behavior: A measure and correlates. Journal of Organizational Behavior, 14(2), 103–118.

Baumert, A., Halmburger, A., & Schmitt, M. (2013). Interventions against norm violations dispositional determinants of self-reported and real moral courage. Personality and Social Psychology Bulletin, 39(8), 1053–1068.

Beer, M., Finnström, M., & Schrader, D. (2016). Why leadership training fails – and what to do about it. Harvard Business Review, 94(10), 50–57.

Blasius, J., & Thiessen, V. (2015). Should we trust survey data? Assessing response simplification and data fabrication. Social Science Research, 52, 479–493.

Brandstätter, V., Jonas, K. J., Koletzko, S. H., & Fischer, P. (2016). Self-regulatory processes in the appraisal of moral courage situations. Social Psychology, 47, 201–213

Brooks, A. K., & Edwards, K. (2009). Allies in the workplace: Including LGBT in HRD. Advances in Developing Human Resources, 11(1), 136–149.

Brown, M. E., Treviño, L. K., & Harrison, D. A. (2005). Ethical leadership: A social learning perspective for construct development and testing. Organizational Behavior and Human Decision Processes, 97(2), 117–134.

Buhrmester, M., Kwang, T., & Gosling, S. D. (2011). Amazon’s Mechanical Turk: A new source of inexpensive, yet high-quality, data? Perspectives on Psychological Science, 6(1), 3–5.

Busse, C., Kach, A. P., & Wagner, S. M. (2016). Boundary conditions what they are, how to explore them, why we need them, and when to consider them. Organizational Research Methods, 1094428116641191.

Callister, L. C., & Sudia-Robinson, T. (2011). An overview of ethics in maternal-child nursing. MCN: The American Journal of Maternal/Child Nursing, 36(3), 154–159.

Clancy, T. R. (2003). Courage and today’s nurse leader. Nursing Administration Quarterly, 27(2), 128–132.

Colbert, A. E., & Witt, L. A. (2009). The role of goal-focused leadership in enabling the expression of conscientiousness. Journal of Applied Psychology, 94 (3), 790.

Comer, D. R., & Vega, G. (2015). Moral courage in organizations: Doing the right thing at work. Armonk, NY: Routledge.

Conger, J. A. (1999). Charismatic and transformational leadership in organizations: An insider’s perspective on these developing streams of research. The Leadership Quarterly, 10(2), 145–179.

Conway, J., & Lance, C. (2010). What reviewers should expect from authors regarding common method bias in organizational research. Journal of Business and Psychology, 25(3), 325.

Corley, M. C. (2002). Nurse moral distress: A proposed theory and research agenda. Nursing Ethics, 9(6), 636–650.

Dawson, J. F. (2014). Moderation in management research: What, why, when, and how. Journal of Business and Psychology, 29(1), 1–19.

include valence or motivational justifications within their items, which results in biased responses. Asking for per-ceptions regarding previously supported items helps avoid the inclusion of such biases. Thus, while it would be pre-ferred to exclusively apply established measures, efforts were taken to help ensure the validity of all adapted scales.

Relatedly, the perceived risks and benefits measures did not precisely instruct participants to respond in regards to their current workplace, although they did instruct partici-pants to assess the risks and benefits in regards to ‘yourself’ or ‘yourself, your coworkers, your organization, and other rel-evant entities.’ It is possible that participants responded in regards to the risks and benefits involved with these behav-iors in general, rather than to their specific workplace. For this reason, it should be recognized that a more direct assessment of risks and benefits associated with social courage behaviors in one’s current workplace may serve as a significant media-tor of the relationships tested in the current article.

Lastly, all variables were measured via self-report, as done in most prior studies involving these constructs (Brown et al., 2005; Howard et al., 2016; Morgeson & Humphrey, 2006). This measurement design poses some concern regarding common method bias, but the time-separated nature of the three studies helps alleviate some concerns regarding common method bias (Conway & Lance, 2010; Podsakoff, MacKenzie, Lee, & Podsakoff, 2003; Podsakoff et al., 2012). Future research should nevertheless analyze the anteced-ents of courage and social courage using multi-source and/or observational studies to obtain more accurate inferences.

Conclusion

The goal of the current article was to identify possible ante-cedents of social courage from five different categories: personality, job characteristics, leadership, culture, and demographic characteristics. The results showed that cer-tain variables from each antecedent category had a signifi-cant relationship with social courage, suggesting that both fixed and malleable antecedents may influence employee social courage. Thus, businesses should consider several aspects of the employee as well as work environment to foster social courage, and researchers should consider sev-eral avenues for future research to better understand the antecedents of social courage.

Acknowledgement

We would like to thank the associate editor, Dr Cynthia Pury, and two anonymous reviewers for their detailed and valuable comments throughout the review process for this manuscript.

Disclosure statement

No potential conflict of interest was reported by the authors.

THE JOURNAL OF POSITIVE PSYCHOLOGY 15

House, R. (1971). A path goal theory of leader effectiveness. Administrative science quarterly,.321–339

House, R. J., Hanges, P. J., Javidan, M., Dorfman, P. W., & Gupta, V. (Eds.). (2004). Culture, leadership, and organizations: The GLOBE study of 62 societies. Thousand Oaks, CA: Sage.

House, R. J., & Mitchell, T. R. (1975). Path-goal theory of leadership (No. TR-75-67). Seattle: Department of Psychology, Washington University.

Howard, M. C., & Alipour, K. K. (2014). Does the courage measure really measure courage? A theoretical and empirical evaluation. The Journal of Positive Psychology, 9(5), 449–459.

Howard, M. C., Farr, J. L., Grandey, A. A., & Gutworth, M. B. (2016). The Creation of the Workplace Social Courage Scale (WSCS): An investigation of internal consistency, psychometric properties, validity, and utility. Journal of Business and Psychology, 1–18.

Humphrey, S. E., Nahrgang, J. D., & Morgeson, F. P. (2007). Integrating motivational, social, and contextual work design features: A meta-analytic summary and theoretical extension of the work design literature. Journal of Applied Psychology, 92(5), 1332–1356.

John, O. P., & Srivastava, S. (1999). The Big Five trait taxonomy: History, measurement, and theoretical perspectives. Handbook of personality: Theory and research, 2, 102–138.

Kashima, Y., Kokubo, T., Kashima, E. S., Boxall, D., Yamaguchi, S., & Macrae, K. (2004). Culture and self: Are there within-culture differences in self between metropolitan areas and regional cities? Personality and Social Psychology Bulletin, 30(7), 816–823.

Kidder, R. M. (2003). Moral courage in Newsletter of the President. Institute for Global Ethics.

Kishton, J. M., & Widaman, K. F. (1994). Unidimensional versus domain representative parceling of questionnaire items: An empirical example. Educational and Psychological Measurement, 54(3), 757–765.

Koerner, M. M. (2014). Courage as identity work: Accounts of workplace courage. Academy of Management Journal, 57(1), 63–93.

Lachman, V. (2007). Moral courage: A virtue in need of development? Medsurg Nursing, 16(2), 131.

Lazarus, R. S. (1991). Emotion and adaptation. Oxford: Oxford University Press on Demand.

Little, T., Cunningham, W., Shahar, G., & Widaman, K. (2002). To parcel or not to parcel: Exploring the question, weighing the merits. Structural Equation Modeling, 9(2), 151.

Lopez, S. J., Rasmussen, H. N., Skorupski, W. P., Koetting, K., Petersen, S. E., & Yang, Y. T. (2010). Folk conceptualizations of courage. In C. L. S. Pury & S. J. Lopez (Eds.), The Psychology of Courage: Modern Reesarch on an Ancient Virtue. Washington, DC: APA Press.

MacIntyre, A. (1981). The nature of the virtues. Hastings Center Report, 11(2), 27–34.

Maslow, A. H. (1943). A theory of human motivation. Psychological Review, 50(4), 370.

Maslow, A. H. (2013). Toward a psychology of being. New York, NY: Simon and Schuster.

May, R. (1994). The courage to create. WW Norton & Company.McCain, J., & Salter, M. (2004). Why courage matters: The way to a

braver life. New York, NY: Random House.Moore, S., & Gullone, E. (1996). Predicting adolescent risk

behavior using a personalized cost-benefit analysis. Journal of Youth and Adolescence, 25(3), 343–359.

Detert, J., & Bruno, E. (2017). Workplace courage: Review, synthesis, and future agenda for a complex construct. Academy of Management Annals, 11(2), 593–639, annals-2015.

Duckworth, A. L., Peterson, C., Matthews, M. D., & Kelly, D. R. (2007). Grit: Perseverance and passion for long-term goals. Journal of Personality and Social Psychology, 92(6), 1087.

Egleston, B. L., Miller, S. M., & Meropol, N. J. (2011). The impact of misclassification due to survey response fatigue on estimation and identifiability of treatment effects. Statistics in Medicine, 30(30), 3560–3572.

Foster, J. D., Shenesey, J. W., & Goff, J. S. (2009). Why do narcissists take more risks? Testing the roles of perceived risks and benefits of risky behaviors. Personality and Individual Differences, 47(8), 885–889.

Fried, Y., & Ferris, G. R. (1987). The validity of the job characteristics model: A review and meta-analysis. Personnel Psychology, 40(2), 287–322.

Geller, E. S., & Veazie, B. (2009). The courage factor: Leading people-based culture change. Coastal Training Technologies Corporation.

Graen, G. B., & Uhl-Bien, M. (1995). Relationship-based approach to leadership: Development of leader-member exchange (LMX) theory of leadership over 25 years: Applying a multi-level multi-domain perspective. The Leadership Quarterly, 6(2), 219–247.

Hackman, J. R., & Oldham, G. R. (1976). Motivation through the design of work: Test of a theory. Organizational Behavior and Human Performance, 16(2), 250–279.

Halmburger, A., Baumert, A., & Schmitt, M. (2015). Anger as driving factor of moral courage in comparison with guilt and global mood: A multimethod approach. European Journal of Social Psychology, 45(1), 39–51.

Hannah, S. T., Avolio, B. J., & Walumbwa, F. O. (2011). Relationships between authentic leadership, moral courage, and ethical and pro-social behaviors. Business Ethics Quarterly, 21(04), 555–578.

Hannah, S. T., Schaubroeck, J. M., Peng, A. C., Lord, R. G., Trevino, L. K., Kozlowski, S. W., … Doty, J. (2013). Joint influences of individual and work unit abusive supervision on ethical intentions and behaviors: A moderated mediation model. Journal of Applied Psychology, 98(4), 579.

Hannah, S. T., Sweeney, P., & Lester, P. (2007). Toward a courageous mindset: The subjective act and experience of courage. The Journal of Positive Psychology, 2(2), 129–135.

Hayes, A. F. (2013). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach. New York, NY: Guilford Press.

Hobbs, A. (2000). Plato and the hero: Courage, manliness and the impersonal good. Cambridge, UK: Cambridge University Press.

Hobfoll, S. (2001). The Influence of culture, community, and the nested-self in the stress process: Advancing conservation of resources theory. Applied Psychology, 50(3), 337.

Hofstede, G. (2006). What did GLOBE really measure? Researchers’ minds versus respondents’ minds. Journal of International Business Studies, 37(6), 882–896.

Hofstede, G. H., & Hofstede, G. (2001). Culture’s consequences: Comparing values, behaviors, institutions and organizations across nations. Thousand Oaks, CA: Sage.

Hofstede, G., & McCrae, R. R. (2004). Personality and culture revisited: Linking traits and dimensions of culture. Cross-Cultural Research, 38(1), 52–88.

16 M. C. HOWARD AND J. E. COGSWELL

of courage: Modern research on an anicent virtue (pp. 47–66). Washington, DC: American Psychological Association.

Rate, C. R., Clarke, J. A., Lindsay, D. R., & Sternberg, R. J. (2007). Implicit theories of courage. The Journal of Positive Psychology, 2(2), 80–98.

Santos, J. P., Caetano, A., & Tavares, S. M. (2015). Is training leaders in functional leadership a useful tool for improving the performance of leadership functions and team effectiveness? The Leadership Quarterly, 26(3), 470–484.

Schilpzand, P. (2008). Personal courage: A measure creation study. University of Florida.

Schilpzand, P., Hekman, D. R., & Mitchell, T. R. (2014). An inductively generated typology and process model of workplace courage. Organization Science, 26(1), 52–77.

Schriesheim, C. A., Castro, S. L., & Cogliser, C. C. (1999). Leader-member exchange (LMX) research: A comprehensive review of theory, measurement, and data-analytic practices. The Leadership Quarterly, 10(1), 63–113.

Schug, J., Yuki, M., & Maddux, W. (2010). Relational mobility explains between-and within-culture differences in self-disclosure to close friends. Psychological Science, 21(10), 1471–1478.

Sekerka, L. E., & Bagozzi, R. P. (2007). Moral courage in the workplace: Moving to and from the desire and decision to act. Business Ethics: A European Review, 16(2), 132–149.

Sekerka, L. E., Bagozzi, R. P., & Charnigo, R. (2009). Facing ethical challenges in the workplace: Conceptualizing and measuring professional moral courage. Journal of Business Ethics, 89(4), 565–579.

Shapiro, D. N., Chandler, J., & Mueller, P. A. (2013). Using Mechanical Turk to study clinical populations. Clinical Psychological Science, 1(2), 213–220.

Simola, S. (2015). Understanding moral courage through a feminist and developmental ethic of care. Journal of Business Ethics, 130(1), 29–44.

Snyder, N. (2010). Vision, values, and courage: Leadership for quality management. New York, NY: Simon & Schuster.

Snyder, C. R., Lopez, S. J., & Pedrotti, J. T. (2010). Positive psychology: The scientific and practical explorations of human strengths. Thousand Oaks, CA: Sage.

Tepper, B. J. (2000). Consequences of abusive supervision. Academy of Management Journal, 43(2), 178–190.

Terry, R. W. (1993). Authentic leadership: Courage in action. San Francisco: Jossey-Bass.

Thatcher, J. B., & Perrewe, P. L. (2002). An empirical examination of individual traits as antecedents to computer anxiety and computer self-efficacy. MIS Quarterly, 381–396.

Vroom, V. H., & Jago, A. G. (1988). The new leadership: Managing participation in organizations. Upper Saddle River, NJ: Prentice-Hall.

Vroom, V. H., & Jago, A. G. (2007). The role of the situation in leadership. American Psychologist, 62(1), 17.

Vroom, V., & Yetton, P. (1973). Leadership and decision-making. Pittsburgh, PA: University of Pittsburgh Press.

Waldman, D., Siegel, D., & Javidan, M. (2006). Components of CEO transformational leadership and corporate social responsibility. Journal of Management Studies, 43(8), 1703–1725.

White, K. H., Howard, M. C., Zhong, B., Soto, J. A., Perez, C. R., Lee, E. A., … Minnick, M. R. (2015). The communication anxiety regulation scale: Development and initial validation. Communication Quarterly, 63(1), 23–43.

Morgeson, F. P., & Humphrey, S. E. (2006). The Work Design Questionnaire (WDQ): Developing and validating a comprehensive measure for assessing job design and the nature of work. Journal of Applied Psychology, 91(6), 1321.

Mowen, J. C., & Carlson, B. (2003). Exploring the antecedents and consumer behavior consequences of the trait of superstition. Psychology & Marketing, 20(12), 1045.

Niesta, K. D., Greitemeyer, T., Fischer, P., & Frey, D. (2010). Why mood affects help giving, but not moral courage: Comparing two types of prosocial behaviour. European Journal of Social Psychology, 40(7), 1136–1157.

Norton, P., & Weiss, B. (2009). The role of courage on behavioral approach in a fear-eliciting situation: A proof-of-concept pilot study. Journal of Anxiety Disorders, 23(2), 212–217.

Oldham, G. R., & Hackman, J. R. (2010). Not what it was and not what it will be: The future of job design research. Journal of Organizational Behavior, 31(2–3), 463–479.

Oliner, S. P. (1991). Altruism: Antidote to Human Conflict. Humboldt Journal of Social Relations, 1–37.

Olsthoorn, P. (2007). Courage in the military: Physical and moral. Journal of Military Ethics, 6(4), 270–279.

Paolacci, G., & Chandler, J. (2014). Inside the Turk: Understanding Mechanical Turk as a participant pool. Current Directions in Psychological Science, 23(3), 184–188.

Perry, S. J., Witt, L. A., Penney, L. M., & Atwater, L. (2010). The downside of goal-focused leadership: The role of personality in subordinate exhaustion. Journal of Applied Psychology, 95(6), 1145.

Peterson, C., & Seligman, M. E. (2004). Character strengths and virtues: A handbook and classification. Oxford University Press.

Pianalto, M. (2012). Moral courage and facing others. International Journal of Philosophical Studies, 20(2), 165–184.

Podsakoff, P. M., MacKenzie, S. B., Lee, J. Y., & Podsakoff, N. P. (2003). Common method biases in behavioral research: A critical review of the literature and recommended remedies. Journal of Applied Psychology, 88(5), 879.

Podsakoff, P. M., MacKenzie, S. B., & Podsakoff, N. P. (2012). Sources of method bias in social science research and recommendations on how to control it. Annual Review of Psychology, 63, 539–569.

Preacher, K. J., & Hayes, A. F. (2004). SPSS and SAS procedures for estimating indirect effects in simple mediation models. Behavior Research Methods, 36(4), 717–731.

Pury, C. L. S., Brawley, A. M., Lopez, S. J., & Burnett, E. A. (2016). Courage.  In C. R. Snyder,  S. J. Lopez,  L. M. Edwards, &  S. C. Marques (Eds.),  Oxford Handbook of Positive Psychology (3rd ed.). New York: Oxford.

Pury, C. L., Kowalski, R. M., & Spearman, J. (2007). Distinctions between general and personal courage. The Journal of Positive Psychology, 2(2), 99–114.

Pury, C. L., & Lopez, S. J. (2010). The psychology of courage: Modern research on an ancient virtue. Washington, DC: American Psychological Association.

Pury, C. L., & Starkey, C. B. (2010). Is courage an accolade or a process? A fundamental question for courage research. In C. L. Pury & S. J. Lopez (Eds.), The psychology of courage: Modern research on an ancient virtue (pp. 67–87), Washington, DC.

Quinn, R. W., & Worline, M. C. (2008). Enabling courageous collective action: Conversations from United Airlines flight 93. Organization Science, 19(4), 497–516.

Rate, C. R. (2010). Defining the features of courage: A search for meaning. In C. L. S. Pury & S. J. Lopez (Eds.), The psychology

THE JOURNAL OF POSITIVE PSYCHOLOGY 17

empowerment, intrinsic motivation, and creative process engagement. Academy of Management Journal, 53(1), 107–128.

Zhang, L., Zhang, C., & Shang, L. (2016). Sensation-seeking and domain-specific risk-taking behavior among adolescents: Risk perceptions and expected benefits as mediators. Personality and Individual Differences, 101, 299–305.

Zhu, W., May, D. R., & Avolio, B. J. (2004). The impact of ethical leadership behavior on employee outcomes: The roles of psychological empowerment and authenticity. Journal of Leadership & Organizational Studies, 11(1), 16–26.

Woodard, C. R., & Pury, C. L. (2007). The construct of courage: Categorization and measurement. Consulting Psychology Journal: Practice and Research, 59(2), 135.

Worline, M. C. (2004). Dancing the cliff edge: The place of courage in social life. Unpublished doctoral dissertation: University of Michigan.

Worline, M. C. (2012). Courage in organizations: An integrative review of the ‘difficult virtue’. The Oxford handbook of positive organizational scholarship, 304–315.

Zhang, X., & Bartol, K. M. (2010). Linking empowering leadership and employee creativity: the influence of psychological