The Leadership Quarterly 24 (2013) 608–622 Contents lists available at SciVerse ScienceDirect The Leadership Quarterly journal homepage: www.elsevier.com/locate/leaqua Questioning universal applicability of transformational leadership: Examining employees with autism spectrum disorder Alissa D. Parr a,⁎, Samuel T. Hunter a, Gina Scott Ligon b a b Pennsylvania State University, 111 Moore Building, University Park, PA 16802, United States Villanova University, 800 E. Lancaster Avenue, Villanova, PA 19085, United States a r t i c l e i n f o Article history: Received 4 September 2012 Received in revised form 7 April 2013 Accepted 10 April 2013 Available online 2 May 2013 Keywords: Transformational leadership Autism Anxiety Commitment a b s t r a c t Challenging an implicit assumption of universal applicability, we propose that a subset of transformational leader behaviors may hamper organizational outcomes for a unique yet growing segment of the workforce: employees on the autism spectrum. Specifically, we tested the hypothesis that emotion-laden communication and social exchanges characterizing the inspirational motivation dimension of the theory are associated with increased feelings of anxiety which, in turn, negatively relate to work outcomes. In contrast, we proposed that other dimensions of transformational leadership, such as individualized consideration, would be associated with reduced levels of anxiety and, ultimately, improved workplace outcomes. Results generally supported the hypothesized indirect relationships for ratings of organizational commitment, but not self-reported job performance which was most strongly predicted by individualized consideration, directly. Implications for managing employees with autism spectrum disorder are discussed as well as the overall applicability of transformational leadership to this growing segment of the workforce. © 2013 Elsevier Inc. All rights reserved. 1. Introduction As a theory, transformational leadership has been shown to be related to a range of outcomes across hundreds of studies, and its impact as a conceptual framework is well summarized in multiple meta-analyses (de Groot, Kiker, & Cross, 2000; Judge & Piccolo, 2004; Lowe, Kroeck, & Sivasubramaniam, 1996) and qualitative reviews (Bennis & Nanus, 1985; Hunt, 1999; Tichy & Devanna, 1986). A number of scholars have suggested that the theory was responsible for taking the field of leadership from the brink of extinction and transforming it into the thriving area of research it is today (Bass & Bass, 2008; Bryman, 1992; Hunt, 1999). In short, there is little debating the importance, impact, or historical significance of vision-based models to our current understanding of, and approach to, leadership. Despite far-reaching empirical support for its impact on several criteria, the theory is not without its detractors. Criticisms include an overemphasis on vision and affective components to the exclusion of other leadership behaviors and cognitions such as planning and strategy development (Hunter, Bedell-Avers, & Mumford, 2007; Yukl, 1999). Rather than a denigration of omission, other scholars have offered more direct criticisms of the theory, suggesting that a focus on affective and emotional arousal via transformational leadership can have detrimental effects on subordinates, leading to outcomes such as burnout, higher levels of stress, dissatisfaction, and reduced performance (Harrison, 1987; Keller, 1992; Seltzer & Bass, 1990). Although empirical support regarding increases in stress and anxiety specifically has been lacking (see Seltzer, Numerof, & Bass, 1989), we ⁎ Corresponding author at: Industrial/Organizational Psychology, Pennsylvania State University, 111 Moore Building, University Park, PA 16802, United States. Tel.: +1 410 591 7252; fax: +1 814 863 7002. E-mail address: aparr@psu.edu (A.D. Parr). 1048-9843/$ – see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.leaqua.2013.04.003 A.D. Parr et al. / The Leadership Quarterly 24 (2013) 608–622 609 suggest that this “non-finding” may be attributable to differences across subordinates not yet investigated. That is, we suggest that an understudied subset of the workforce may be more susceptible to the potential for negative influence from transformational leadership. Moreover, given the range of behaviors comprising transformational leadership it seems reasonable to suggest that some actions on the part of the leader may be ineffective or even harmful, while others are likely to have associations with more positive effects. Central to this article, we propose that some employees, specifically, employees with autism spectrum disorder, will be differentially impacted by the varying dimensions of transformational leadership. In the following sections, a brief overview of the characteristics of employees with autism is presented. Next, given these characteristics, we discuss how transformational leadership behaviors might be received by these employees. Finally, we put forth and test a mediation model whereby transformational leadership behaviors indirectly impact workplace outcomes (organizational commitment and job performance) by operating through anxiety. We begin our discussion by introducing an emerging segment of the workforce: employees with autism. 1.1. Employees with autism spectrum disorder Over the course of the past decade, there has been a significant increase in diagnosed cases of autism spectrum disorder, or ASD. A recent report from the Center for Disease Control and Prevention indicates that approximately 1 in every 50 (2.0%) children meets the diagnostic criteria for the disorder, with that likelihood being much greater (1 in 31) for males (Blumberg et al., 2013). To put these values in perspective, it is useful to consider the prevalence rates of more widely known disabilities, such as vision impairment (1 in 714), hearing impairment (1 in 714), Down syndrome (1 in 800), childhood cancer (1 in 6700), juvenile diabetes (1 in 400) and cerebral palsy (1 in 280; CDC, 2009). Extrapolation from current prevalence rates places the total number of individuals with autism in the United States around 1.5 million, with the Autism Society of America (2006) suggesting that this is a lower bound estimate. Although these values provide some indication of the number of individuals with autism, it is important to note that these values are not static — the number of individuals diagnosed with the disorder is on the rise. The CDC noted that from the years 2002 to 2006, there was a 57% increase in cases of ASD. Additionally, there was a 23% increase in cases from 2006 to 2008. Finally, Blumberg and colleagues (2013) noted a 42% increase in diagnoses between 2007 and 2012. Data are clear in suggesting that diagnoses are likely to stay on an upward trajectory for some time, even with the acknowledgement of currently debated stricter diagnostic criteria that may curb the sharp upward trend. Most central to the present study, a sizable influx of teens and adolescents are entering adulthood and beginning to seek employment (Gerhardt & Lainer, 2011). Improved diagnostic tools as well as early behavioral interventions have led to greater optimism for education and employment resulting in increased numbers of young adults having aspirations of a fulfilling work life (Dawson et al., 2010). For reasons that range from legal (Colella, Paetzold, & Belliveau, 2004; VanBergeijk, Klin, & Volkmar, 2008) to ethical, as this growing population becomes of age to enter the workforce, organizations must increase their preparedness to deal with the unique opportunities and challenges this influx presents. In particular, the social nature of ASD and its high comorbidity with anxiety disorder (Gillott & Standen, 2007; Volker & Lopata, 2008) present noteworthy challenges to those that manage and operate organizations. In short, identifying mechanisms that will facilitate the successful integration and engagement of employees with ASD at work appears to be of paramount and increasing importance. 1.2. Autism and the workplace Autism spectrum disorders (ASDs) are a group of pervasive developmental disorders which range from severe to milder forms. According to the American Psychiatric Association's Diagnostic and Statistical Manual of Mental Disorders, ASD is broadly characterized by impairments in social interaction, deficits in verbal and nonverbal communication, as well as the presence of restricted, repetitive, and stereotyped patterns of behavior and interests (American Psychiatric Association, 2000). ASD is thought to be a permanent developmental disorder creating lifelong challenges for the individual (Gerhardt & Lainer, 2011; Hendricks, 2010). Despite these social impairments, growing research on adults with ASD suggests that this population may possess qualities and attributes that can enable them to excel in the workplace. For example, individuals with ASD often display a keen attention to detail, a willingness to engage in repetitive activities, trustworthiness, reliability, and timeliness (Hillier et al., 2007). Furthermore, having a focused interest can allow them to gain significant expertise on a particular topic, providing a valuable knowledge resource to an organization. Finally, several studies have found that most, if not all, supervisors rated the job performance of their employee with ASD as average or above average (e.g., Hagner & Cooney, 2005; Hillier et al., 2007; Unger, 2003). Taken together, it appears that with sufficient accommodations and supports, individuals with ASD possess the capabilities to be an asset to an organization. Although individuals with autism have the competencies to do well in the workplace, they face significant obstacles in organizations. Overall, individuals with ASD are more likely to be unemployed compared to any other group with disabilities (Dew & Allan, 2007). Specifically, it is estimated that 50%–75% of adults with ASD are unemployed (Howlin, Goode, Hutton, & Rutter, 2004; Hurlbutt & Chalmers, 2004; Mawhood, Howlin, & Rutter, 2000; National Organization on Disability, 2004) and the majority who are employed are not working full-time (Chappel & Somers, 2010). To compound the problem, unemployment difficulties are still common even after taking into account those who are higher functioning and who have postsecondary educational experiences (Gerhardt & Lainer, 2011; Howlin, 2000). Specifically, the National Autistic Society (2011) estimated that 610 A.D. Parr et al. / The Leadership Quarterly 24 (2013) 608–622 approximately 332,600 individuals of working age have ASD in the United Kingdom. Of those, only 6% hold a full-time job and only 12% of those with high functioning autism hold full-time jobs. While acknowledging that core human resource practices, such as training and selection, are critical parts of the broader solution to managing the influx of ASD employees (Gerhardt & Lainer, 2011; Schall, Wehman, & McDonough, 2012), we contend that there are more proximal factors which may directly impact these relationships. In order to better understand the nature of this issue, it is important to determine whether there are person–situation interactions which may be impacting these unemployment rates. In particular, we propose that employee anxiety, which is quite common in individuals with ASD, and leadership, as a contextual influence, may be contributing factors. 1.3. Anxiety in individuals with ASD As mentioned previously, there is a high occurrence of anxiety in individuals with ASD. A recent study by Gillott and Standen (2007) compared levels of anxiety in a group of 34 adults with autism and intellectual disabilities to a matched group of 20 adults with intellectual disabilities without autism. The researchers found that the group with autism had significantly higher overall levels of anxiety, with 73.1% scoring outside the range of the control group. Moreover, researchers have also suggested that these individuals may not have the coping skills to manage the stress or cognitively appraise the situations they find anxiety provoking (Groden et al., 2001). Given that the modern work environment is increasingly more dynamic, uncertain, and team-based, it is likely that the workplace can be a significant stressor for individuals on the autism spectrum. In a series of interviews, employees with ASD cited that one of the factors affecting their employability was increased levels of stress and anxiety in trying to work “in a neurotypical world” (Hurlbutt & Chalmers, 2004, p. 219). The participants stated that they quickly became overwhelmed by not knowing which topics were appropriate to discuss in the workplace and not knowing when to ask for help. As a means to combat these underlying challenges, we contend that leadership will play a central role in the successful engagement of employees on spectrum. Leaders serve as role models, select task assignments, make accommodations to work roles, and provide task and emotional support which all can potentially increase the likelihood of worker success (Bass & Bass, 2008; DeRue, Nahrgang, Wellman, & Humphrey, 2011; Walumbwa, Avolio, & Zhu, 2008; Yukl, 2010). Thus, leaders can serve as an essential support mechanism to help individuals with ASD better adapt to their environment and successfully engage in their job. Additionally, given that leaders are central role models for how to appropriately behave in the organization (Brown, Treviño, & Harrison, 2005; Klimoski & Hayes, 1980) and that employees with ASD report significant challenges in interpreting and understanding the social environment, these social cues from leaders are critical (Müller, Schuler, Burton, & Yates, 2003). Although the impact that a leader can have on employees with ASD is substantial, there currently exist very few empirical studies directly examining the role of leadership in this unique and growing population (Holwerda, van der Klink, Groothoff, & Brouwer, 2012). Additionally, the studies that have been conducted on employees with ASD have been severely limited in their sample size, averaging about three to four participants per study (e.g., Bennett, Brady, Scott, Dukes, & Frain, 2010; Burt, Fuller, & Lewis, 1991; Lattimore, Parsons, & Reid, 2002). Perhaps most critically, although the limited work that has been done has helped lay an essential foundation for understanding the role of leadership in supporting employees with ASD (e.g., Hagner & Cooney, 2005; Hurlbutt & Chalmers, 2004), early efforts have been largely atheoretical with regard to investigations of leader influence. For substantive gains in this area of research to be made, it will be essential to develop research questions grounded in the theoretical framing of leadership literature explicitly (Bacharach, 1989; Cook & Campbell, 1979; Yukl, 2010). Therefore, as an effort to advance current research strategies and to better understand the issue, we turn to how the role of leadership, and in particular transformational leadership, may facilitate healthy well-being and successful engagement at work. 1.4. Transformational leadership Although varying conceptualizations to leadership abound, we chose to focus on transformational leadership (Bass, 1985a) as our primary theoretical framework for three primary reasons: the theory's widespread popularity among modern-day research and practice, its demonstrated capacity to predict key organizational outcomes, and the substantial dimension overlap it displays with elements of ASD (Bass & Bass, 2008; Gardner, Lowe, Moss, Mahoney, & Cogliser, 2010). First, although the theory has its critics, transformational leadership currently represents the most popular leadership framework and is widely researched as well as utilized in leadership development programs. As such, transformational leadership may be widely implemented in work environments given it is appropriate for the masses. Additionally, since there are several risks associated with disclosing their condition that may prevent several employees on the spectrum to notify their employer of their condition (Genchanok & Kunce, 2011), leaders may not be modifying their approach to leadership and instead may be utilizing this more widespread approach. Second, transformational leadership has been empirically demonstrated numerous times to be related to several outcomes important for success in the workplace. Third, when comparing some of the defining features of transformational leadership with typical symptoms associated with ASD, we noticed substantial overlap. In particular, transformational leadership has a strong focus on affect, emotion, and individualized consideration. All of these may be especially relevant to people on the spectrum because they often struggle to interpret social and emotional cues and therefore may require individualized consideration as a result. As such, transformational leadership seemed to be a natural starting point for investigating leader influence on employees with autism. A.D. Parr et al. / The Leadership Quarterly 24 (2013) 608–622 611 As originally proposed by Bass (1985a), the four behavioral dimensions of transformational leadership are: 1) inspirational motivation, 2) idealized influence, 3) intellectual stimulation, and 4) individualized consideration. Since its introduction as a theory, however, there has been sizable debate in the literature regarding the precise factor structure of what behaviors comprise transformational leadership (see Tejada, Scandura, & Pillai, 2001 for discussion). Some researchers, for example, have suggested that a two-factor model be used (Yammarino, Spangler, & Dubinsky, 1998), others have proposed three-factor solution (Heinitz, Liepmann, & Felfe, 2005), and even a nine-factor approach has been offered (Alimo-Metcalfe & Alban-Metcalfe, 2001). It is important to bear in mind, however, that with the exception of a few notable scholars (e.g., Yukl, 1999) these criticisms have largely been levied against the multifactor leadership questionnaire (MLQ) as the primary measure of transformational leadership, rather than the dimensionality of the theory, per se. Other scales, such as the Transformational Leadership Inventory or TLI (Podsakoff, MacKenzie, Moorman, & Fetter, 1990) have produced factor structures congruent with proposed and extended theory. Moreover, there has been some support for a four factor solution, even with respect to the MLQ itself (Bass, 1985b). Thus, although the exact factor structure associated with the theory is open to debate, there is general agreement among leadership scholars that the range of behaviors comprising it are distinct from one another in some form, and that transformational leadership as a theory is indeed multidimensional (Burns, 1978; Schriesheim, Castro, Zhou, & DeChurch, 2006). Central to the present study, these distinct transformational leader dimensions may also have variable influence on subordinate outcomes, and particularly with employees with autism spectrum disorder. We turned to the theoretical framework proposed originally by Bass (1985a) and Bass and Bass (2008) as well as updated recommendations offered by Rafferty and Griffin (2004) for guidance on dimensionality, rather than the conflicting milieu of findings observed within the realm of scaling and measurement. We briefly discuss each of the four proposed transformational leadership dimensions along with their hypothesized relationships with anxiety below. 1.5. Transformational leadership, anxiety, and work outcomes As described earlier, anxiety is common in adults with ASD (Gillott & Standen, 2007) and some of the symptoms associated with ASD may prevent them from executing appropriate coping skills to manage stress and appraise anxiety-arousing situations (Groden et al., 2001). For example, in their study examining the prevalence of anxiety and stressors leading to anxiety in adults with ASD, Gillott and Standen (2007) found that change, anticipation, positive events, and sensory/personal contact were all rated significantly higher as sources of stress in the group of adults with autism than in the matched sample. Given that the modern work environment is increasingly more dynamic and team-based, it is likely that the workplace will be a significant stressor to employees on the autism spectrum. As such, unless mechanisms are in place to help them cope with anxiety producing events in the workplace, it is likely that these stressors will negatively impact their approach to work and performance. Consistent with the wealth of research on neurotypical subordinates (Mumford, 2006; Strauss, 1944), we propose that leaders can also serve as mechanisms to help employees with ASD better cope and appraise the anxiety-provoking situations. For example, knowing that individuals with ASD become anxious when routines change, leaders can prepare their employees for future changes and make subtle, rather than abrupt, changes. Providing careful, individualized support throughout the change event would appear critical and has been shown, via case-study, to help several employees on the autism spectrum (Hurlbutt & Chalmers, 2004). If unaddressed, moreover, such anxiety inducing events could have significant implications on work attitudes and performance. Central to the article, however, is the emerging proposition that leaders themselves may operate as a source of anxiety. That is, the emotionally and affectively laden aspects of transformational leader behaviors may be anxiety inducing antecedents for employees with autism. More precisely, we suggest that certain dimensions of transformational leadership may differentially influence levels of anxiety, with some helping to reduce apprehension and others resulting in anxiety increases. Our proposed conceptual model and our hypotheses are presented in Fig. 1. We discuss each hypothesis in greater detail, below. 1.5.1. Inspirational motivation and anxiety Through inspirational motivation, transformational leaders use affect-laden communication (Bass, 1985a,b). Inspirationoriented behaviors are those that are emotionally driven (Bass, 1999) — behaviors that are often embraced by subordinates that have a high need for messages with an emotional appeal (Howell & Shamir, 2005). Essentially, leaders using inspirational motivation enhance enthusiasm in followers through the use of nonintellectual, emotional qualities (Bass, 1985a). In a study examining how leader positive affect, leader positive expressions, and leader-aroused behaviors, relate to follower affect, Erez, Misangy, Johnson, LePine, and Halverson (2008) found that these positive affective displays were contagious to their followers. Specifically, participants who were exposed to leaders expressing positive charismatic displays indicated experiencing higher levels of positive affect than those not exposed to leaders expressing positive charismatic displays. Given the evidence which suggests individuals with ASD experience sensory overload quite easily (Hurlbutt & Chalmers, 2004), being in contact with someone who expresses these extreme positive displays can be overwhelming to them and may be a source of anxiety. Additionally, transformational leaders communicate their vision using symbols and abstract ideas (Bass & Bass, 2008). However, as noted in a study of employees with ASD, this population has difficulty understanding abstract concepts related to their job (Hillier et al., 2007). In particular, these employees had problems grasping the overall picture of how their job role fit within the broader organization. Moreover, individuals with ASD interpret ideas very literally and may struggle with value-based visions. Therefore, these strategies for motivating employees may not resonate well with employees with ASD, leading us to our first study proposition: Hypothesis 1. Inspirational motivation will be associated with increased levels of anxiety in employees with ASD. 612 A.D. Parr et al. / The Leadership Quarterly 24 (2013) 608–622 Inspirational motivation H1: + Idealized influence Anxiety Work Outcomes H2: H5: H6a & b: Indirect effect H3: + Organizational commitment Job performance Intellectual stimulation H4: - Individualized consideration Hypothesized relationship Tested but not hypothesized Fig. 1. Conceptual framework for how transformational leadership influences work outcomes vis-à-vis anxiety. 1.5.2. Idealized influence and anxiety Despite the potentially deleterious effects of how leaders communicate their vision to employees with ASD, we contend that what they communicate—via the leader's vision—should not have harmful effects on anxiety and, in fact, may provide greater levels of social ease and comfort. Although some research has suggested the combination of inspirational motivation and idealized influence (e.g., Colby & Zak, 1988), scholars such as Barbuto (1997) and Rafferty and Griffin (2004) suggest that the distinguishing feature of behaviors related to idealized influence is to provide a vision in order to enhance goal clarity, task focus, and value congruence (House, 1971). Thus, through idealized influence, leaders clearly articulate a vision providing more structure and guidance to their employees to better understand performance expectations. These behaviors are unlike inspirational motivation because there is a greater emphasis on providing clarity through the content of the message rather than an emphasis on how the message is delivered (i.e., through emotion-laden language; Barbuto, 1997; Rafferty & Griffin, 2004). When asked to give recommendations for enhancing success at work, employees with ASD commented that it is important to clearly describe job duties, responsibilities, expectations, and roles well in advance (Hurlbutt & Chalmers, 2004). Additionally, some effective strategies that supervisors of employees with ASD have used include maintaining a consistent schedule and consistent job tasks, providing organizers to structure work or written instructions for assignments, and being very clear when giving directions (Hagner & Cooney, 2005). These behaviors are consistent with establishing clear objectives and facilitating an environment to increase task focus. In addition, transformational leaders also serve to enhance value congruence and to act as strong role models by displaying high standards for moral behavior. Because individuals with ASD are often quite literal and see ideas in a black-and-white nature (Hurlbutt & Chalmers, 2004), it may be important for their leaders to display honesty, adhere to societal principles, and take into consideration the perspectives of everyone. Having a leader attend to values which individuals with ASD hold closely, may put them more at ease. Therefore, we predict that idealized influence should lessen the anxiety levels of employees with ASD. Hypothesis 2. Idealized influence will be associated with lower levels of anxiety in employees with ASD. 1.5.3. Intellectual stimulation and anxiety Transformational leaders encourage followers to challenge traditions and to inspire intellectual growth and creativity through intellectual stimulation. Leaders accomplish intellectual stimulation in a number of ways, such as through reframing problems through the use of metaphors and similes (Wicker, 1985) or refuting existing points of view (Infante & Gorden, 1985). Quinn and Hall (1983) suggested that intellectual stimulation may come about through rational, existential, empirical, or ideological behaviors. Additionally, Spreitzer and de Janasz (1994) proposed that intellectual stimulation is largely driven by empowering subordinates to formulate their own unique framing to problems. Finally, Seltzer, Numerof, and Bass (1989) suggested that intellectual stimulation may inherently arouse stress because these behaviors push subordinates to use reasoning rather than unsupported beliefs. In support, Seltzer and Bass (1990) found that intellectual stimulation had a negative effect on subordinate satisfaction. Keller (1992) recognized these differences and suggested that contextual factors may influence whether intellectual stimulation has positive or negative effects. Thus, it appears that behaviors associated with this dimension are often abstract, lack high levels of structure, and are stress-inducing, particularly for individuals already susceptible to stress and anxiety. Moreover, A.D. Parr et al. / The Leadership Quarterly 24 (2013) 608–622 613 the emphasis on challenging the status quo is likely to be anxiety inducing for employees who struggle to understand basic social norms, much less how to effectively challenge them. While individuals with ASD are often very intelligent and creative, this push and the use of abstract language may be overly excessive. As such, we predicted that leaders who engage in intellectually stimulating behaviors would increase levels of anxiety in employees on the autism spectrum. Hypothesis 3. Intellectual stimulation will be associated with increased levels of anxiety in employees with ASD. 1.5.4. Individualized consideration and anxiety Finally, through individualized consideration, transformational leaders support subordinates by providing unique and tailored coaching. This attentiveness may be needed, especially with regards to potential frustration and affective breakdowns that can afflict employees with ASD (Hagner & Cooney, 2005). Additionally, because autism is largely a social and communication-based disorder, supervisors may need to provide support and guidance when employees with ASD are interacting with other coworkers or customers. For example, individuals on the autism spectrum often have difficulty interpreting nonverbal cues therefore it may be necessary for the supervisor to assist in their understanding of these cues. Leaders who engage in tailored and individualized behaviors for their employees are likely to recognize the unique needs of individuals on the autism spectrum and, accordingly, assign tasks that suit their skill-set and match their social capacity. In one example, Hagner and Cooney (2005) discussed an employee working as a cashier at a grocery store who became anxious when lines began to form and customers were in a rush. The supervisor recognized the employee's anxiety and scheduled the employee to work on off-peak hours as a means to improve performance and reduce subordinate stress. On the whole, we predicted that leaders who pay special and unique attention to the needs of employees on the autism spectrum would reduce anxiety in these employees, leading to our fourth hypothesis: Hypothesis 4. Individualized consideration will be associated with lower levels of anxiety in employees with ASD. 1.5.5. Anxiety and workplace outcomes In addition to proposing the direct effects of transformational leadership on anxiety, we also hypothesize that levels of anxiety should impact work outcomes of employees with ASD. We chose to focus on one attitudinal variable, organizational commitment, as well as self-reported indicators of organizational performance. As previous research has integrated how transformational leadership affects strain through organizational commitment (Franziska & Felfe, 2011), we focused on the attitudinal variable of organizational commitment due to its relevance. Additionally, organizational commitment is appropriate given the relative underemployment of individuals with ASD (e.g., Howlin et al., 2004; Hurlbutt & Chalmers, 2004; Mawhood et al., 2000; National Organization on Disability, 2004) and may shed light onto potential factors associated with underemployment. We believe the combination of attitudes and performance indicators provides a broader indication of how transformational leadership and anxiety impact work outcomes. Organizational commitment is defined as a “the relative strength of an individual's identification with and involvement in a particular organization” (Mowday, Porter, & Steers, 1982, p. 226). Several studies have examined various antecedents of organizational commitment including individual-level characteristics, role states, job characteristics, leader relations, and organizational characteristics. In a recent study examining the role organizational commitment in stress models, Glazer and Kruse (2008) sought to determine whether organizational commitment moderated the relationship between job-related anxiety and intention to leave. While they were not testing the direct relationship between job-related anxiety and organizational commitment, they did report a significant and negative bivariate correlation between job-related anxiety and organizational commitment (r = − .17). It is important to note that the sample was composed of employees not reporting ASD, with a fairly low mean level of anxiety (M = 2.95, SD = 1.26, on a scale of 1–7; Glazer & Kruse, 2008). Given that individuals with ASD tend to have higher levels of anxiety, it is likely that this correlation would be magnified in a sample expressly focused on such employees. Additionally, there is some indirect evidence which suggests that levels of anxiety may influence organizational commitment. In a meta-analysis examining the antecedents, correlates, and consequences of organizational commitment, Mathieu and Zajac (1990) reported that role ambiguity had a negative and moderate correlation with organizational commitment (mean weighted r = − .22). Role ambiguity, defined as a condition in which an employee lacks clear knowledge of what behaviors are expected in his or her job, has very similar properties as anxiety. In employees with ASD, anxiety may arise when they are put in social situations in which they do not know how to interact. However, anxiety is not just limited to ambiguity but also extends to states or conditions resulting from highly arousing situations. Therefore, while role ambiguity overlaps some properties of anxiety, it encompasses additional elements which may also result in decreased levels of organizational commitment. Additionally, we propose that anxiety may have an influence on performance. At the most general level, role ambiguity has also been shown to have a strong negative correlation with performance in a recent meta-analysis (Gilboa, Shirom, Fried, & Cooper, 2008) indicating that elements of anxiety may decrease not only organizational commitment but also performance. More specifically, anxiety leads to heightened arousal and stress which may impact the cognitive appraisal of situations (Groden et al., 2001). In a recent study by Luke, Clare, Ring, Redley, and Watson (2012), individuals with ASD were compared to a matched control group on their decision-making abilities. Overall, participants with ASD reported higher levels of anxiety than the matched control group. Additionally, analyses revealed that as anxiety increased in participants with ASD, reports of experiencing 614 A.D. Parr et al. / The Leadership Quarterly 24 (2013) 608–622 more problems in decision-making increased as well. Thus, it is likely that anxiety may impede some of the processes critical for effective performance at work. Hypothesis 5. Anxiety will be negatively related to (a) organizational commitment and (b) performance. Finally, we propose that transformational leadership will have an indirect effect on work outcomes, operating through anxiety. That is, we propose that leaders have the ability to ease or increase anxiety levels in employees with ASD, which may ultimately be related to their work attitudes and performance. Drawing on the path goal theory, a partial precursor to transformational leadership (Bass, 1965), we suggest that leaders should seek to remove obstacles in order to enhance the performance of subordinates (House, 1971; Mitchell, 1979). In particular, we suggest that anxiety may be the obstacle and the behaviors of leaders may either reduce that barrier to successful outcomes or potentially augment the barrier ultimately impacting their attitudes and performance. Thus, the critical role of leadership for employees with ASD may be to help manage anxiety in order to ultimately enhance organizational commitment and performance. As noted by House (1971), leaders make the paths to desired work outcomes “easier to travel, by clarifying [them], reducing roadblocks and pitfalls…” (p. 324). Our sixth, and final hypothesis is as follows: Hypothesis 6. Transformational leadership dimensions will be indirectly related to (a) organizational commitment and (b) performance through anxiety. 2. Method 2.1. Participants Participants included 52 employees with ASD (48.1% males and 51.9% females), all of who had a professional diagnosis of ASD. The majority of the workers were employed as human service workers, research support staff, or cleaning and support staff. Furthermore, participants indicated having a more mild form of ASD, on average (M = 1.69, SD = 0.76 using a scale from 1, very mild, to 5, very severe). Finally, 59% of the sample had self-disclosed their condition to their employer. A number of steps were taken to reach as large and broad of a sample as possible. Aside from what the participant experienced regarding recruitment (detailed below), numerous methods and attempts were made at identifying contacts, agencies, and other avenues to gain the appropriate sample. The majority of recruitment efforts were directed at contacting local and national autism agencies as well as broader agencies serving individuals with all developmental disabilities. Participating agencies posted an advertisement on their website or sent announcements to their listservs which targeted individuals with ASD directly. Those unable to do so provided alternative resources to aid in recruitment such as potential points of contact. Also, local support groups for adults with ASD were identified and attended. Finally, more than 20 researchers and authors advocating for adults with ASD were contacted directly to explore additional avenues for recruitment. Nine months of a multi-source recruitment effort resulted in the largest collection of employees with ASD assessed in a leadership study to date. 2.2. Procedure Prior to beginning data collection, we consulted with three licensed psychologists with expertise in working with individuals on the autism spectrum, as well as individuals with ASD working at the university where the research was conducted. Feedback from these individuals was clear in suggesting that traditional means of data collection, in particular use of survey-based measures, would likely be problematic for employees with autism who might need clarification and explanation regarding specific items — a point that was reiterated a number of times with regard to the challenges of assessing the occurrence of transformational leader behaviors, specifically. Given the consistent feedback across health care professionals, advocacy organizations, and individuals with autism, we chose to engage in interviews with the research population rather than utilize either web-based or paper-and-pencil surveys. Data collection began by contacting autism advocacy groups as a means to gain access to autism listservs and internal research posting opportunities. With such access granted, participants were recruited through e-mail and webpage announcements sponsored by various autism advocacy organizations. Interested participants contacted the researchers to schedule a phone interview. As suggested by several members of the health service community that were consulted prior to data collection, all employees with ASD were e-mailed a copy of the interview protocol to ease potential uncertainty of what might be asked of them during the interview. To ensure that all the participants were fully capable of answering all of the questions and differentiating between the items, a thorough intake procedure was followed. The researcher, who was designated to conduct all of the interviews, explained the procedure of the interview and asked follow-up questions to ensure that the participants would be able to competently answer all of the items. Additionally, throughout the interview, the researcher listened for signs of hesitation or uneasiness to ensure that participants fully understood what was being asked. During the interview, participants responded to a series of questions about themselves and their place of work as well as specific behaviors used by their supervisors at work. Participants were also asked to freely respond to a series of questions regarding supervisor behaviors they believe facilitate the success of employees with ASD. After completing the interview protocol, participants were thanked for their involvement. The interview took an average of 30 min to complete. A.D. Parr et al. / The Leadership Quarterly 24 (2013) 608–622 615 2.3. Measures 2.3.1. Transformational leadership To measure transformational leadership, participants rated the frequency that their supervisors engaged in behaviors characteristic of each dimension of transformational leadership. The list of behaviors was created by adapting those described by Yukl (2010) and Rafferty and Griffin (2004). Responses were made on a 5-point Likert scale ranging from 1 (never) to 5 (always). A global assessment of each transformational leadership dimension was utilized over individual items based on feedback and direction from subject matter experts during the piloting phase. Specifically, we found that individuals with ASD were likely to become distracted and anxious by responding to each individual item. To mitigate this challenge, behavioral exemplars were provided (e.g., “Communicates a vision which is exciting and motivating” and “Provides a supportive environment for employees”), followed by requesting participants to provide dimensional ratings. To obtain additional support for the validity of our measure, we administered this scale as well as the transformational leadership scale developed by Podsakoff and colleagues (1990) to a sample of non-ASD undergraduates (n = 171). Results demonstrated that both scales had more than adequate reliability (0.81 and 0.94, respectively) and had a high, positive correlation of r = .83. Furthermore, we assessed reliability by contacting employees approximately one year later and asking them to provide a second set of assessments on transformational leadership. Specifically, participants were asked to provide ratings about the same leader they rated originally. A sample of 19 employees replied to our request, but only 11 reported high levels of confidence that they could accurately recall the leader they provided assessments of during the original interview. Using this sample, test–retest values for all four dimensions were .60 or higher, with an average of .67. Additional studies assessing test–retest reliability for other measures of transformational leadership reported values ranging from .44 to .74 for self-ratings (Mandell & Pherwani, 2003) and .69 to .80 (Popper, Mayseless, & Castelnovo, 2000) using the Multifactor Leadership Questionnaire (Bass & Avolio, 1996) as well as .69 (Conger & Kanungo, 1994) using the Conger Kanungo Scale. Given the time between assessments, small sample size, and relative similarity of values with other measures, we believe that these values provide some support for the reliability of transformational leadership assessments. 2.3.2. Anxiety To assess anxiety, five items from the Generalized Anxiety Disorder-7 Scale (GAD-7) were used (Spitzer, Kroenke, Williams, & Lowe, 2006). Participants were asked to rate how often they experience those five statements as a group on a scale from 1 (never) to 5 (always). Although items were assessed globally in the same way as the transformational leadership items which precluded traditional assessments of internal consistency this scale has been shown to have very good reliability (α = .92) in prior research (Spitzer et al., 2006). It is of note that, using the method described above for assessments of transformational leadership, test– retest reliabilities of anxiety (asked with regard to levels of anxiety at the original time of assessment) were .74. 2.3.3. Work outcomes Organizational commitment and performance were measured. All responses to the scales were made using a 7-point Likert scale ranging from 1 (strongly disagree) to 7 (strongly agree). Organizational commitment was assessed using eight items from the Allen and Meyer (1990) Affective Commitment Scale. The internal consistency for the scale was .84. Finally, self-perceptions of performance were measured by participants responding to the item, “My supervisor would probably rate my overall performance as above average.” Although caution should be taken when interpreting this self-rating of performance, a separate study which matched six of these employees to their supervisors found performance ratings from the employee and the supervisors to have a high correlation (r = .77, p b .05). It is imperative, however, to stress that this measure captures self-perceptions of performance. 2.3.4. Covariates Leader liking has been shown to influence assessments of leaders (e.g., Brown & Keeping, 2005) and as such, leader liking was measured using the Wayne and Ferris (1990) Member Liking of the Leader scale (α = .89). The inclusion of leader liking as a control variable permits greater confidence that observed relationships were not a result of common-method bias (Conway & Lance, 2010; Spector, 2006). 3. Results Descriptive results including means, standard deviations, and intercorrelations for the study variables are presented in Table 1. It is interesting to note that the correlations between transformational leadership dimensions are lower than typically reported for neurotypicals. Since they are still relatively strong and in the positive direction, we feel comfortable in conducting our analyses. Along these lines, much is unknown about scale development for individuals with ASD and there are not any standardized psychometric measures designed or validated specifically for adults with ASD (Ghaziuddin, 2005). However, it can be speculated that since individuals with ASD often are more sensitive to the nuances of behaviors (Hillier et al., 2007), they may be more likely to make distinctions between the behaviors. This pattern is consistent with our projections that individuals on the spectrum interpret some transformational behaviors as positive and others as negative. To test our hypotheses, data were analyzed using bootstrapped tests of direct and indirect effects developed by Preacher and Hayes (2008). This test permits multiple predictors and covariates when assessing predicted relationships. More critically, 616 A.D. Parr et al. / The Leadership Quarterly 24 (2013) 608–622 Table 1 Means, standard deviations, and intercorrelations for study variables (n = 52). 1. 2. 3. 4. 5. 6. 7. 8. Idealized influence Inspirational motivation Intellectual stimulation Individualized consideration Anxiety Commitment Self-reported performance Leader liking M SD 1 3.15 2.94 2.40 3.37 3.52 4.32 5.08 4.92 1.14 1.21 1.13 1.25 0.83 1.25 1.68 1.68 1.00 .49⁎⁎ .45⁎⁎ .70⁎⁎ −.44⁎⁎ .52⁎⁎ 2 .30⁎ .59⁎⁎ 1.00 .51⁎⁎ .44⁎⁎ .01 .51⁎⁎ .37⁎⁎ .49⁎⁎ 3 4 5 6 7 8 1.00 .41⁎⁎ −.19 .41⁎⁎ .22 .33⁎ 1.00 −.45⁎⁎ .34⁎ .50⁎⁎ .69⁎⁎ 1.00 −.31⁎ −.05 −.20 1.00 .22 .44⁎⁎ 1.00 .35⁎⁎ 1.00 Note. All variables, except Anxiety, were rated on a 1–7 point Likert scale. Anxiety was rated on a 1–5 point Likert scale. ⁎ p b .05. ⁎⁎ p b .01 however, this test provides bootstrapped confidence intervals for the tests of indirect effects. Confidence intervals are preferred over methods such as the Sobel test, which makes an unrealistic assumption about the sampling distribution of the indirect test (Hayes, 2009). Supplementing the specific tests of hypotheses were examinations of overall model fit. Specifically, we tested the goodness-of-fit for the model depicted in Fig. 1 using structural equation models applied to both organizational commitment and self-reported job performance. Results indicated that for organizational commitment (χ 2/df = 1.76, CFI = .96, GFI = .93, SRMR = .08) as well as job performance (χ 2/df = 1.76, CFI = .95, GFI = .93, SRMR = .08) fit was acceptable and fell within common thresholds (Hu & Bentler, 1999). It is of note that common fit indices such as RMSEA were not reported as these tests are strongly affected by sample size, where those reported (e.g., CFI) tend to be more stable and independent of number of cases investigated (Bentler, 1990; Hu & Bentler, 1999). On the whole, these results lend some, albeit tenuous, support to the model suggesting that transformational behaviors are related to organizational outcomes both directly as well as through anxiety as a mediating variable. Acceptable fit allowed us to feel more comfortable testing specific hypotheses, particularly given the smaller sample size of the study. We turn now to the results of specific hypotheses. 3.1. Influence of transformational leadership on anxiety To test the relationship between transformational leadership and anxiety, stepwise ordinary least squares regression analysis applying bootstrapping methods was used. As depicted in Table 2, leader liking was added in the first step, followed by a block containing the dimensions of transformational leadership. Adding in liking in the first block allows us to limit the effects of common method bias and rule out the possibility that observed relationships were the result of a third variable (Conway & Lance, 2010; Spector, 2006). Overall, the final model explained a sizable 34% of the variance in anxiety. More specifically, the direct effects of inspirational motivation (β = .23, 95% CI = .05 to .40), idealized influence (β = − .27, 95% CI = − .46 to − .06) and individualized consideration (β = − .28, 95% CI = − .48 to −.09) were all significant and in the hypothesized directions. It is worth noting, however, that the intellectual stimulation dimension did not produce a statistically significant relationship with anxiety. Thus, Hypotheses 1, 2, and 4 were supported. 3.2. Influence of anxiety on work outcomes Turning towards the relationship between anxiety and organizational commitment as well as performance, a similar stepwise regression with bootstrapping was conducted. As may be seen in Table 3, leader liking was added initially, followed by a block containing each of the transformational leadership dimensions. In the final step, anxiety was entered into the model. Results indicate that, when examining organizational commitment as the dependent variable, anxiety produced a significant and negative relationship with organizational commitment (β = − .54, 95% CI = − .99 to −.08) even when controlling for the effects Table 2 Regression results for the direct effects of transformational leadership on anxiety. Variable Step 1 Liking Step 2 Idealized influence Inspirational motivation Intellectual stimulation Individualized consideration SE p .09 .09 .34 −.27⁎ .23⁎ −.05 −.28⁎ .13 .10 .10 .12 .03 .03 .59 .02 β R2 ΔR2 .04 .34 Note. Bootstrap sample size = 1000; SE = standard error. ⁎ p ≤ .05. .30⁎ 617 A.D. Parr et al. / The Leadership Quarterly 24 (2013) 608–622 Table 3 Regression results for the direct effects of transformational leadership and anxiety on outcomes. Variable Commitment Step 1 Liking Step 2 Idealized influence Inspirational motivation Intellectual stimulation Individualized consideration Step 3 Anxiety Performance Step 1 Liking Step 2 Idealized influence Inspirational motivation Intellectual stimulation Individualized consideration Step 3 Anxiety SE p .19 .13 .15 .29 .37⁎ .13 −.35 .17 .17 .18 .21 .09 .04 .48 .11 −.43⁎ .20 .04 −.06 .23 .78 −.15 .28 −.05 .89⁎ .28 .26 .21 .32 .60 .28 .81 .02 .34 .31 .29 β R2 ΔR2 .19 .39 .20⁎ .44 .05⁎ .13 .29 .16⁎ .30 .02 Note. Bootstrap sample size = 1000; SE = standard error. ⁎ p ≤ .05. of transformational leadership, explaining an additional 7% of the variance in organizational commitment over and above leader liking and the four transformational leadership dimensions. In contrast, however, anxiety produced a non-significant relationship with self-reported performance. Therefore, Hypothesis 5a was supported but Hypothesis 5b was not. It is of note that, when examining direct effects of transformational leadership, the inspirational motivation dimension produced a significant positive relationship with organizational commitment. Although the proposed research questions focused on indirect relationships, the observation of a significant direct effect is interesting. It seems that when accounting for the negative influence of anxiety on organizational commitment, inspirational behaviors can also produce positive effects. More directly, it seems that not all inspirational behaviors are harmful to employee commitment — some may result in greater affinity towards the organization. 3.3. Indirect effects of transformational leadership on work outcomes through anxiety Finally, to test for the indirect effects of transformational leadership on organizational commitment and performance through anxiety while controlling for leader liking, an ordinary least squares regression procedure utilizing bootstrapping was used. Specifically, the bootstrapping approach to indirect testing is an extension of the product-of-coefficients strategy, where bootstrapping allows for relaxed assumptions of normality and therefore less biased estimates of indirect effects (Preacher & Hayes, 2008; Shrout & Bolger, 2002). Table 4 presents the results for organizational commitment and performance. Examining the dependent variable organizational commitment, the omnibus test was significant and negative suggesting that overall, transformational leadership had an indirect effect on organizational commitment through anxiety, even while controlling for Table 4 Regression results for the indirect effects of transformational leadership through anxiety on outcomes, controlling for liking. Variable Point estimate SE Lower CI Upper CI Commitment Idealized influence Inspirational motivation Intellectual stimulation Individualized consideration Omnibus .11⁎ −.10⁎ .02 .12⁎ −.11⁎ .07 .06 .05 .08 .08 .01 −.20 −.04 .01 −.27 .24 −.01 .11 .28 −.02 Performance Idealized influence Inspirational motivation Intellectual stimulation Individualized consideration Omnibus −.09 .08 −.02 −.10 .08 .10 .08 .05 .10 .10 −.29 −.03 −.09 −.29 −.04 .03 .23 .06 .04 .28 Note. Bootstrap sample size = 50,000; CI = 95% one-tailed confidence interval; SE = standard error. ⁎ p ≤ .05. 618 A.D. Parr et al. / The Leadership Quarterly 24 (2013) 608–622 leader liking. More specifically, idealized influence, inspirational motivation, and individualized consideration all produced effects in the expected directions. Thus, the indirect and positive effects of idealized influence (point estimate = .11, 95% CI = .01 to .24) and individualized consideration (point estimate = .12, 95% CI = .01 to .28) on organizational commitment through anxiety as well as the indirect and negative effect of inspirational motivation (point estimate = − .10, 95% CI = − .20 to −.01) on organizational commitment through anxiety were observed. Intellectual stimulation did not have a significant indirect effect on organizational commitment. Noting the exception of intellectual stimulation, Hypothesis 6a was generally supported. Alternatively, the omnibus test of the model with performance was not significant, suggesting that none of the proposed indirect effects of transformational leadership on performance through anxiety were observed. Examination of individual tests confirmed findings from the omnibus test. Thus, Hypothesis 6b was not supported. Interestingly, we did find that individualized consideration produced a rather strong direct effect with regard to predicting self-reported organizational performance (β = .89, SE = .20, 95% CI = .34 to 1.36). It is of note that, of the relationships examined in this study, this effect was the largest observed. When predicting organizational performance of employees with autism, it seems that levels of anxiety are less critical than a supervisor who pays special attention to the unique needs of their employees. 4. Discussion Despite a wealth of data suggesting that transformational leadership is a universally beneficial approach to leadership, results from this study suggest that such implicit universality may not hold for all employees. In particular, we observed that one dimension of transformational leadership, inspirational motivation, increased levels of anxiety in employees with ASD, which was related to lower levels of organizational commitment. It is of note, however, that other dimensions such as individualized consideration demonstrated relationships with both reduced levels of anxiety as well a notable direct relationship with organizational performance. Similar findings were also observed for idealized influence with our interpretation being that leaders who offer a clear vision serve to provide goal clarity, thereby limiting perceptions of ambiguity and uncertainty. On the whole, then, results suggest that some transformational behaviors are helpful for employees with ASD, while others may hamper employee well-being in the workplace. At the onset of the article, we suggested that the growing number of employees with ASD dictates that organizations increase their preparedness to manage and support their success in the workplace. Results from this study support the proposition that leaders will play a key role in managing this influx and that a number of behaviors will be particularly crucial in the transition. In particular, results suggest that anxiety may attenuate feelings of organizational commitment. Perhaps most critical to leaders was the finding that transformational leadership behaviors, in positive and negative fashion, were significant predictors of anxiety levels. Although time will tell which leadership behaviors are most critical to employees on the autism spectrum, results from this study suggest that leaders would be well-served to provide clear goals, pay special attention to the needs of this unique workforce, and limit the emotionally laden social interactions often associated with inspirational communication. 4.1. Theoretical implications The results from this study have several theoretical and practical implications for leadership research and workplace issues for individuals with ASD. With respect to theory development and refinement, although transformational leadership has been demonstrated to have positive effects on outcomes for neurotypical employees (e.g., de Groot et al., 2000; Judge & Piccolo, 2004; Lowe et al., 1996), this study highlighted some potential boundary conditions for the positive effects of transformational leadership in other populations. In particular, we found that inspirationally oriented behavior was related to higher levels of anxiety for employees with autism. This is unlike studies on transformational leadership with neurotypicals, which demonstrate inspirational motivation has negative relationships with emotional exhaustion and chronic stress (Densten, 2005; Roswold & Schlotz, 2009) and positive relationships with personal accomplishment and positive affect (Brown & Keeping, 2005; Densten, 2005). The implication of our finding is that when leaders communicate using symbols and emotion-laden messages, they may be negatively affecting employee well-being. Directly, this finding suggests that transformational leadership behaviors may not be universally effective for all employees. Rather, there may be some behaviors which negatively impact various outcomes in individuals, particularly those sensitive to abstract and affectively-laden forms of communication and social interaction. Along these lines, this study demonstrated that the four dimensions of transformational leadership differentially influence anxiety in employees with ASD. This finding underscores the importance of unpacking the dimensionality of transformational leadership as a theory to obtain a more nuanced understanding of relationships with other variables. As the debates continue with respect to the precise dimensionality of the theory, our results suggest that more fine-grained approaches may have merit, particularly with regard to investigating unique workplace populations. From a process-oriented perspective, this study provides insight into the mechanisms driving leadership influence of employees with autism. Specifically, we found that transformational leadership produced an indirect relationship on work attitudes, operating through anxiety, a common symptom of individuals with ASD (Groden et al., 2001). As theory is refined with regard to the processes that shape transformational leader influence, it would appear useful to consider the role of anxiety in such processes — particularly for those employees who trend toward higher levels of anxiousness. Notably, although indirect effects were observed for organizational commitment, the effects were not found for self-ratings of performance. In fact, most of the effects were in the opposite direction hypothesized. Inferred from these trends is the possibility A.D. Parr et al. / The Leadership Quarterly 24 (2013) 608–622 619 that transformational leadership serves to operate indirectly on cognitive and affective variables (i.e., attitudes). If placed in a job position that employees with ASD feel comfortable in and understand the tasks well, it may be the case that leadership and anxiety do not negatively interfere with their task performance. Alternatively, returning to the routinized (and well-known) tasks of the job may ease their anxiety and maintain their level of performance. 4.2. Practical implications Turning now to practical implications, by applying leadership behaviors associated with idealized influence and individualized consideration, results suggest that employees with ASD will experience less anxiety and be more committed to their work. In turn, if they do feel successfully engaged in their work endeavors and feel as though they are contributing to the organization, it is likely that their quality of life will be improved because a sense of independence, self-efficacy, and self-determination will be achieved. Second, understanding these behaviors may have implications not only for employees with ASD but also for the organization. These effects of greater work engagement and productivity are likely to be seen at the organizational level. By having employees who are more successfully engaged and committed to their work, it is likely that organizational effectiveness will also be improved, particularly over time. Noted earlier, individuals with ASD can uniquely contribute to the organization because of their distinct skill sets and having a supported, committed workforce may prove to be a valuable asset to the organization most willing to appropriately manage these employees. Finally, although we have focused most of our attention on employees with ASD, there are some direct benefits to supervisors as well. In particular, if supervisors understand behaviors that work best for engaging individuals with ASD, they can more effectively tailor their actions to those behaviors which have been found to decrease anxiety and increase work attitudes. Thus, rather than experiment using various behaviors, they can more definitively use a set of behaviors thereby increasing the efficiency of their actions, which may produce feelings of efficacy of their leadership skills and may allow them to devote their energy to other matters. Moreover, leaders may be unaware of the appropriate means of engaging employees with autism which may, in turn, cause anxiety in the leaders themselves. Results from this study provide at least initial indication with respect to which leader behaviors may be most useful to begin with when working with employees with ASD. 4.3. Study limitations The results of this study must be considered in light of some limitations. First, the size of the sample is small relative to other studies in organizational behavior. Despite the low numbers, this sample is larger than any other study of its kind. After an analysis of peer-reviewed research studies involving employees with ASD, the largest sample sizes were composed of six individuals with ASD (e.g., Burke, Andersen, Bowen, Howard, & Allen, 2010; Hurlbutt & Chalmers, 2004), with the majority of the studies only including three or four individuals with ASD (e.g., Bennett et al., 2010; Burt et al., 1991; Lattimore et al., 2002). Perhaps most directly, small sample size is a problem only if effect sizes are too small to be detected (Cohen, 1992). In our case, multiple hypotheses were supported, suggesting that the sample was large enough to test the hypothesized effects and that there was acceptable power in the analyses conducted. Moreover, fit indices from two SEM models were supportive of testing the proposed data structure. Nonetheless, a larger sample would lend greater confidence to similar findings and as awareness of autism grows, the procurement of such samples may improve. Second, it is important to note that the measures used in the study utilized self-report ratings. Use of self-report has been criticized for increasing the possibility of common method bias (Organ & Ryan, 1995; Podsakoff & Todor, 1985). However, we did attempt to control for potential third variable issues by covarying the effects of leader liking (Williams & Anderson, 1994), a known biasing variable with respect to impacting assessments of transformational leadership (Brown & Keeping, 2005). Moreover, given the nature of the variables tested, with the exception of performance which demonstrated high correlations with supervisor ratings, self-report indicators appeared most appropriate for the proposed research questions (Conway & Lance, 2010). In regards to performance, this self-report measure is composed of a single item and therefore direct reliability statistics could not be drawn. However, a subsample demonstrated high correlations with supervisor ratings providing some confidence for using this measure. Additionally, past research demonstrated that single item measures of performance do meet acceptable levels of reliability (Wanous & Hudy, 2001). We encourage future research to use objective performance ratings to replicate our findings. Along these lines, because our study involved collecting self-report data at one point in time, we cannot eliminate the possibility of reverse causality. It may be possible that employees with autism who have higher levels of anxiety are less likely to be receptive to a leader's display of inspirational motivation. However, provided that there is evidence to support that inspirational motivation tends to have positive effects on the general population, who may too have high levels of anxiety at times, we feel that the direction of results implied is suitable. Third, this study only focused on one leadership theory. However, there may be other leadership theories that may be applicable to employees with ASD. For example, transactional leadership may be particularly relevant to those on the spectrum because it focuses on elements related to providing structure, clarity, and direction to tasks as well as providing support, coaching, and guidance. Future research is encouraged to explore how other leadership theories may influence employees on the spectrum. Fourth, this study does not include a comparison sample of coworkers to assess the effects of transformational leadership. Due to the sensitive nature of disclosing their condition to employers and coworkers, we were limited in acquiring a comparison sample in order to respect the privacy of our participants. Finally, the majority of the sample consisted of employees who had a 620 A.D. Parr et al. / The Leadership Quarterly 24 (2013) 608–622 mild form of ASD. Future research is encouraged to actively solicit opinions of individuals who may be on the more severe end of the spectrum to see whether their perspectives differ and whether observed relationships may vary by disorder level. 4.4. Directions for future research Both the results and limitations from this study highlight the need for future research in this domain. First, other outcome variables should be investigated in relation to leadership and employees with ASD. While organizational commitment is certainly important, other more objective indices of performance such as absenteeism, errors, and tenure are also likely related to leadership behaviors with this population. In addition, other models of leadership should be examined for generalizability to this population to determine if one style may outweigh the transformational leadership benefits identified. Second, while the majority of ASD research has focused on young children, there is a critical need to conduct more studies using an adult population to facilitate their successful integration into adulthood and independent living. As employment is one of the defining features of independent living and has implications for quality of life (Schall et al., 2012), it is important to direct more research projects towards adults with ASD in the work environment (Gerhardt & Lainer, 2011). In particular, future research should replicate these results with a broader sample, especially with those lower on the spectrum, to see if there are any differences in perspectives. It may be possible that individuals with more severe cases have exaggerated reactions to transformational behaviors. Additionally, if these individuals have more significant social and communication problems, individualized consideration may be even more important to help employees learn workplace norms and navigate this unique territory in a sincere manner. Third, it is interesting to note that several participants had not self-disclosed their condition to their employer. Future research should seek to understand why and when individuals decide to self-disclose and how their employers react. It is possible that individuals may be resistant to self-disclose because they are fearful that they will be discriminated against or put at a disadvantage. Not only does this present issues for the employee by nature of stigmatization and lack of accommodations provided, but it also presents an issue for research on these understudied populations—ASD and all other cognitive and emotional conditions. As researchers, we will need to focus more closely on recruiting efforts and privacy issues to reach out to this sector. Furthermore, as practitioners, it shows a need that we need to focus on creating work climates that are more open and receptive to all disorders. Fourth, there is a need to examine more closely the role of anxiety in work outcomes of employees with ASD. In particular, it is important to better understand the relationship between anxiety and performance. While an overall measure of performance was utilized for this study, it may be the case that there are differential results for task performance and contextual performance. It is possible that moderate levels of anxiety could enhance task performance on the one hand but could harm contextual performance because of the more social nature of contextual performance. Additionally, thinking more broadly, it will be important to examine how leadership styles may influence other individuals with disabilities, for example attention deficit disorder. The affective nature of transformational leadership may also have a negative impact on this population, as compared to neurotypicals. Finally, although the legal landscape is in continual flux, it is currently open to question what types of actions on the part of an organization might constitute reasonable accommodation for employees with social disabilities such as autism. As the number of employees with ASD grows and their legal awareness increases (VanBergeijk et al., 2008), organizations may soon be left searching for the appropriate action to place them in compliance with disability laws. The social nature of disabilities such as autism may require organizations to offer training to supervisors that guides appropriate forms of support and management. Future research is needed to determine what types of training programs are most effective and whether these programs operate within the reasonable accommodation requirements. In summary, this is the first study to empirically examine the relationship between transformational leadership on work attitudes and performance in employees with ASD and to determine how this relationship is impacted by anxiety. Overall, this study showed that transformational leadership may not always be beneficial to employees. We hope that these results will lead to more investigations in how leadership can influence the successful integration and engagement of employees with ASD, a population which will be entering the workforce in the near future, as well as to take a more critical approach to assessing the applicability of commonly accepted constructs in unique populations. References Alimo-Metcalfe, B., & Alban-Metcalfe, R. J. (2001). The development of a new transformational leadership questionnaire. Journal of Occupational and Organizational Psychology, 74, 1–27. Allen, N. J., & Meyer, J. P. (1990). The measurement and antecedents of affective, continuance and normative commitment to the organization. Journal of Occupational Psychology, 63, 1–18. American Psychiatric Association (2000). Diagnostic and statistical manual of mental disorders (Revised 4th ed.). Washington, DC: Author. Autism Society of America (2006). Facts and Statistics. Retrieved September 2, 2012 from Autism Society of America Web site: http://www.autism-society.org/ about-autism/facts-and-statistics.html Bacharach, S. B. (1989). Organizational theories: Some criteria for evaluation. Academy of Management Review, 14, 496–515. Barbuto, J. E. (1997). Taking the charisma out of transformational leadership. Journal of Social Behaviour and Personality, 12, 689–697. Bass, B. M. (1965). Organizational psychology. Boston: Allyn & Bacon. Bass, B. M. (1985a). Leadership and performance beyond expectations. New York: Free Press. Bass, B. M. (1985b). The multifactor leadership questionnaire: Form 5. Binghamton: State University of New York. Bass, B. M. (1999). Two decades of research and development in transformational leadership. European Journal of Work and Organizational Psychology, 8, 9–32. A.D. Parr et al. / The Leadership Quarterly 24 (2013) 608–622 621 Bass, B. M., & Avolio, B. J. (1996). Transformational leadership development: Manual for the Multifactor Leadership Questionnaire. Palo Alto, CA: Consulting Psychologists Press. Bass, B. M., & Bass, R. (2008). The Bass handbook of leadership: Theory, research, and managerial applications (4th ed.). New York: Free Press. Bennett, K., Brady, M. P., Scott, J., Dukes, C., & Frain, M. (2010). The effects of covert audio coaching on the job performance of supported employees. Focus on Autism and Other Developmental Disabilities, 25, 173–185. Bennis, W., & Nanus, B. (1985). Leaders: The strategies for taking charge. New York: Harper and Row. Bentler, P. M. (1990). Comparative fit indexes in structural models. Psychological Bulletin, 107, 238–246. Blumberg, S. J., Bramlett, M. D., Kogan, M. D., Schieve, L. A., Jones, J. R., & Lu, M. C. (2013). Changes in prevalence of parent-reported autism spectrum disorder in school-aged U.S. children: 2007 to 2011–2012. National Health Statistics Reports, 65, 1–11. Brown, D., & Keeping, L. M. (2005). Elaborating the construct of transformational leadership: The role of affect. The Leadership Quarterly, 16, 245–272. 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, 92, 117–134. Bryman, A. (1992). Charisma and Leadership in Organizations. London: Sage. Burke, R. V., Andersen, M. N., Bowen, S. L., Howard, M. R., & Allen, K. D. (2010). Evaluation of two instruction methods to increase employment options for young adults with autism spectrum disorder. Research in Developmental Disabilities, 31, 1223–1233. Burns, J. M. (1978). Leadership. New York: Harper & Row. Burt, D. B., Fuller, P., & Lewis, K. R. (1991). Brief report: Competitive employment of adults with autism. Journal of Autism and Developmental Disorders, 21, 237–242. Centers for Disease Control, Prevention (2009). Retrieved June 2, 2010 from www.cdc.gov Chappel, S. L., & Somers, B. C. (2010). Employing persons with autism spectrum disorders: A collaborative effort. Journal of Vocational Rehabilitation, 32, 117–124. Cohen, J. (1992). A power primer. Psychological Bulletin, 112, 155–159. Colby, A. H., & Zak, R. E. (1988). Transformational leadership: A comparison of army and air force perceptions (Report 88-0565). Maxwell AFB, AL: Air Command and Staff College, Air University. Colella, A., Paetzold, R., & Belliveau, M. (2004). Factors affecting coworkers' procedural justice inferences of the workplace accommodation of employees with disabilities. Personnel Psychology, 57, 1–23. Conger, J. A., & Kanungo, R. N. (1994). Charismatic leadership in organizations: Perceived behavioral attributes and their measurement. Journal of Organizational Behavior, 15, 439–452. Conway, J. M., & Lance, E. C. (2010). What reviewers should expect from authors regarding common method bias in organizational research. Journal of Business and Psychology, 25, 325–334. Cook, T. D., & Campbell, D. T. (1979). Quasi-experimentation: Design & analysis issues for field settings. Boston: Houghton Mifflin Company. Dawson, G., Rogers, S., Munson, J., Smith, M., Winter, J., Greenson, J., et al. (2010). Randomized, controlled trial of an intervention for toddlers with autism: The early start Denver model. Pediatrics, 125, 17–23. de Groot, T., Kiker, S. D., & Cross, T. C. (2000). A meta-analysis to review organizational outcomes related to charismatic leadership. Canadian Journal of Administrative Sciences, 17, 356–371. Densten, I. L. (2005). The relationship between visioning behaviours of leaders and follower burnout. British Journal of Management, 16, 105–118. DeRue, D. S., Nahrgang, J. D., Wellman, N., & Humphrey, S. E. (2011). Trait and behavioral theories of leadership: An integration and meta-analytic test of their relative validity. Personnel Psychology, 64, 7–52. Dew, D., & Allan, G. (2007). Rehabilitation of individuals with autism spectrum disorder. Institute on Rehabilitation Issues Monograph No. 32. Washington D.C.: The George Washington University. Erez, A., Misangy, V. F., Johnson, D. E., LePine, M. A., & Halverson, K. C. (2008). Stirring the hearts of followers: Charismatic leadership as the transfer of affect. Journal of Applied Psychology, 93, 602–615. Franziska, F., & Felfe, J. (2011). How does transformational leadership impact employees' psychological strain?: Examining differentiated effects and the moderating role of affective organizational commitment. Leadership, 7, 295–316. Gardner, W. L., Lowe, K. B., Moss, T. W., Mahoney, K. T., & Cogliser, C. C. (2010). Scholarly leadership of the study of leadership: A review of The Leadership Quarterly's second decade, 2000–2009. The Leadership Quarterly, 21, 922–958. Genchanok, A., & Kunce, L. (2011). Explanatory autism disclosure in the workplace: Perspectives of key stakeholder groups. Honors Projects. Paper 144 (http://digitalcommons. iwu.edu/psych_honproj/144) Gerhardt, P. F., & Lainer, I. (2011). Addressing the needs of adolescents and adults with autism: A crisis on the horizon. Journal of Contemporary Psychotherapy, 4, 37–45. Ghaziuddin, M. (2005). Mental health aspects of autism and Asperger syndrome. London: JKP. Gilboa, S., Shirom, A., Fried, Y., & Cooper, C. (2008). A meta-analysis of work demand stressors and job performance: Examining main and moderating effects. Personnel Psychology, 61, 227–271. Gillott, A., & Standen, P. J. (2007). Levels of anxiety and sources of stress in adults with autism. Journal of Intellectual Disabilities, 11, 359–370. Glazer, S., & Kruse, B. (2008). The role of organizational commitment in occupational stress models. International Journal of Stress Management, 15, 329–344. Groden, J., Diller, A., Bausman, M., Velicer, W., Norman, G., & Cautela, J. (2001). The development of a stress survey schedule for persons with autism and other developmental disabilities. Journal of Autism and Developmental Disabilities, 31, 207–217. Hagner, D., & Cooney, B. F. (2005). “I do that for everybody”: Supervising employees with autism. Focus on Autism and Other Developmental Disabilities, 20, 91–97. Harrison, R. (1987). Harnessing personal energy: How companies can inspire employees. Organizational Dynamics, 16, 5–20. Hayes, A. F. (2009). Beyond Baron and Kenny: Statistical mediation analysis in the new millennium. Communication Monographs, 76, 408–420. Heinitz, K., Liepmann, D., & Felfe, J. (2005). Examining the factor structure of the MLQ: Recommendation for a reduced set of factors. European Journal of Psychological Assessment, 21, 182–190. Hendricks, D. (2010). Employment and adults with autism spectrum disorders: Challenges and strategies for success. Journal of Vocational Rehabilitation, 32, 125–134. Hillier, A., Campbell, H., Mastriani, K., Izzo, M., Kool-Tucker, A. K., Cherry, L., et al. (2007). Two-year evaluation of a vocational support program for adults on the autism spectrum. Career Development for Exceptional Individuals, 30, 35–47. Holwerda, A., van der Klink, J. J. L., Groothoff, J. W., & Brouwer, S. (2012). Predictors for work participation in individuals with an autism spectrum disorder: A systematic review. Journal of Occupational Rehabilitation, 22, 333–352. House, R. J. (1971). A path-goal theory of leader effectiveness. Administrative Science Quarterly, 16, 321–339. Howell, J. M., & Shamir, B. (2005). The role of followers in the charismatic leadership process: Relationships and their consequences. Academy of Management Review, 30, 96–112. Howlin, P. (2000). Outcome in adult life for more able individuals with autism or Asperger syndrome. Autism, 4, 63–83. Howlin, O., Goode, S., Hutton, J., & Rutter, M. (2004). Adult outcome for children with autism. Journal of Child Psychology and Psychiatry, 45, 212–229. Hu, L. -T., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6, 1–55. Hunt, J. G. (1999). Transformational/charismatic leadership's transformation of the field: An historical essay. The Leadership Quarterly, 10(2), 129–144. Hunter, S. T., Bedell-Avers, K. E., & Mumford, M. D. (2007). The typical leadership study: Assumptions, implications, and potential remedies. The Leadership Quarterly, 18, 435–446. Hurlbutt, K., & Chalmers, L. (2004). Employment and adults with Asperger syndrome. Focus on Autism and Other Developmental Disabilities, 19, 215–222. Infante, D. A., & Gorden, W. I. (1985). Superiors' argumentativeness and verbal aggressiveness as predictors of subordinates' satisfaction. Human Communication Research, 12, 117–125. 622 A.D. Parr et al. / The Leadership Quarterly 24 (2013) 608–622 Judge, T. A., & Piccolo, R. F. (2004). Transformational and transactional leadership: A meta-analytic test of their relative validity. Journal of Applied Psychology, 89, 755–768. Keller, R. T. (1992). Transformational leadership and the performance of research and development project groups. Journal of Management, 18, 489–501. Klimoski, R. J., & Hayes, N. J. (1980). Leader-behavior and subordinate motivation. Personnel Psychology, 33, 543–555. Lattimore, L. P., Parsons, M. B., & Reid, D. H. (2002). A prework assessment of task preferences among adults with autism beginning a supported job. Journal of Applied Behavioral Analysis, 35, 85–88. Lowe, K. B., Kroeck, G. K., & Sivasubramaniam, N. (1996). Effectiveness correlates of transformational and transactional leadership: A meta-analytic review of the MLQ literature. The Leadership Quarterly, 7, 385–425. Luke, L., Clare, I. C., Ring, H., Redley, M., & Watson, P. (2012). Decision-making difficulties experienced by adults with autism spectrum conditions. Autism, 16, 612–621. Mandell, B., & Pherwani, S. (2003). Relationship between emotional intelligence and transformational leadership style: A gender comparison. Journal of Business and Psychology, 17, 387–404. Mathieu, J. E., & Zajac, D. M. (1990). A review and meta-analysis of the antecedents, correlates, and consequences of organizational commitment. Psychological Bulletin, 108, 171–194. Mawhood, L., Howlin, P., & Rutter, M. (2000). Autism and developmental receptive language disorder: A comparative follow-up in early adult life, I: Cognitive and language outcomes. Journal of Child Psychology and Psychiatry, 41, 547–559. Mitchell, T. R. (1979). Organizational behavior. Annual Review of Psychology, 30, 243–281. Mowday, R. T., Porter, L. W., & Steers, R. M. (1982). Employee-organization linkages: The psychology of commitment, absenteeism, and turnover. New York: Academic Press. Müller, E., Schuler, A., Burton, B., & Yates, G. (2003). Meeting the vocational support needs of individuals with Asperger syndrome and other autism spectrum disabilities. Journal of Vocational Rehabilitation, 18, 163–175. Mumford, M. D. (2006). Pathways to outstanding leadership: A comparative analysis of charismatic, ideological, and pragmatic leaders. Mahweh, New Jersey: Erlbaum. National Autistic Society (2011). Living with autism: Looking for work. Retrieved December 2, 2012 from http://www.autism.org.uk/living-with-autism/ employment/looking-for-work.aspx National Organization on Disability (2004). Harris survey of Americans with disabilities: Landmark survey finds pervasive disadvantages. Retrieved January 21, 2012 from http://www.nod.org/content.cfm%3Fid=1537 Organ, D. W., & Ryan, K. (1995). A meta-analytic review of attitudinal and dispositional predictors of organizational citizenship behavior. Personnel Psychology, 48, 775–803. Podsakoff, P. M., MacKenzie, S. B., Moorman, R. H., & Fetter, R. (1990). Transformational leader behaviors and their effects on followers' trust in leader, satisfaction, organizational citizenship behaviors. The Leadership Quarterly, 1, 107–142. Podsakoff, P. M., & Todor, W. D. (1985). Relationships between leader reward and punishment behavior and group processes and productivity. Journal of Management, 11, 55–73. Popper, M., Mayseless, O., & Castelnovo, O. (2000). Transformational leadership and attachment. Leadership Quarterly, 11, 267–289. Preacher, K. J., & Hayes, A. F. (2008). Contemporary approaches to assessing mediation in communication research. In A. F. Hayes, M. D. Slater, & L. B. Snyder (Eds.), The Sage sourcebook of advanced data analysis methods for communication research (pp. 13–54). Thousand Oaks, CA: Sage Publications. Quinn, R. E., & Hall, R. H. (1983). Environments, organizations, and policy makers: Toward an integrative framework. In R. H. Hall, & R. E. Quinn (Eds.), Organization theory and public policy: Contributions and limitations. Beverly Hills, CA: Sage Publications. Rafferty, A. E., & Griffin, M. A. (2004). Dimensions of transformational leadership: Conceptual and empirical extensions. The Leadership Quarterly, 15, 329–354. Roswold, J., & Schlotz, W. (2009). Transformational and transactional leadership and followers' chronic stress. Kravis Leadership Institute, Leadership Review, 9. (pp. 35–48), 35–48. Schall, C., Wehman, P., & McDonough, J. L. (2012). Transition from school to work for students with autism spectrum disorders: Understanding the process and achieving better outcomes. Pediatrics Clinic of North America, 59, 189–202. Schriesheim, C. A., Castro, S. L., Zhou, X. T., & DeChurch, L. A. (2006). An investigation of path-goal and transformational leadership theory predictions at the individual level of analysis. The Leadership Quarterly, 17, 21–38. Seltzer, J., & Bass, B. M. (1990). Transformational leadership: Beyond initiation and consideration. Journal of Management, 16, 693–703. Seltzer, J., Numerof, R. E., & Bass, B. M. (1989). Transformational leadership: Is it a source of more burnout and stress? Journal of Health and Human Resource Administration, 12, 174–185. Shrout, P. E., & Bolger, N. (2002). Mediation in experimental and nonexperimental studies: New procedure and recommendations. Psychological Methods, 7, 422–445. Spector, P. E. (2006). Method variance in organizational research: Truth or urban legend? Organizational Research Methods, 9, 221–232. Spitzer, R. L., Kroenke, K., Williams, J. B., & Lowe, B. (2006). A brief measure for assessing generalized anxiety disorder. Archives of Internal Medicine, 166, 1092–1097. Spreitzer, G. M., & de Janasz, S. C. (1994). The transformational capacity of empowered managers. Los Angeles, CA: University of Southern California, Department of Management Organization. Strauss, A. L. (1944). The literature on panic. Journal of Abnormal and Social Psychology, 39, 317–328. Tejada, M. J., Scandura, T. A., & Pillai, R. (2001). The MLQ revisited—psychometric properties and recommendations. The Leadership Quarterly, 12, 31–52. Tichy, N. M., & Devanna, M. A. (1986). The transformational leader. New York: John Wiley & Sons. Unger, D. D. (2003). Employers' attitudes toward persons with disabilities in the workforce: Myths or realities? Focus on Autism and other Developmental Disabilities, 17, 2–10. VanBergeijk, E., Klin, A., & Volkmar, F. (2008). Supporting more able students on the autism spectrum: College and beyond. Journal of Autism and Developmental Disorders, 38, 1359–1370. Volker, M. A., & Lopata, C. (2008). Autism: A review of biological bases, assessment, & intervention. School Psychology Quarterly, 23, 258–270. Walumbwa, F. O., Avolio, B. J., & Zhu, W. (2008). How transformational leadership weaves its influence on individual job performance: The role of identification and efficacy beliefs. Personnel Psychology, 61, 793–825. Wanous, J. P., & Hudy, M. J. (2001). Single-item reliability: A replication and extension. Organizational Research Methods, 4, 361–375. Wayne, S. J., & Ferris, G. R. (1990). Influence tactics, affect, and exchange quality in supervisor–subordinate interactions: A laboratory experiment and field study. Journal of Applied Psychology, 5, 487–499. Wicker, A. W. (1985). Getting out of our conceptual ruts: Strategies for expanding conceptual frameworks. American Psychologist, 40, 1094–1103. Williams, L. J., & Anderson, S. E. (1994). An alternative approach to method effects using latent-variable models: Applications in organizational behavior research. Journal of Applied Psychology, 79, 323–331. Yammarino, F. J., Spangler, W. D., & Dubinsky, A. J. (1998). Transformational and contingent reward leadership: Individual, dyad, and group levels of analysis. The Leadership Quarterly, 9, 27–54. Yukl, G. (1999). An evaluation of conceptual weaknesses in transformational and charismatic leadership theories. The Leadership Quarterly, 10, 285–305. Yukl, G. (2010). Leadership in organizations (7th ed.). Upper Saddle River, NJ: Pearson.