Summer 6

Educational Administration Quarterly 2014, Vol. 50(2) 233 –278

© The University Council for Educational Administration 2013

Reprints and permissions: sagepub.com/journalsPermissions.nav

DOI: 10.1177/0013161X13492795 eaq.sagepub.com

Article

The Rise of Neurotics: Social Networks, Leadership, and Efficacy in District Reform

Alan J. Daly1, Yi-Hwa Liou1, Natalie A. Tran2, Frank Cornelissen1, and Vicki Park3

Abstract Purpose: Increasing evidence suggests the importance of relationships between district and site leaders. However, there is limited empirical evidence regarding the social infrastructure between and among leaders especially as related to the exchange of advice related to reform. Moreover, we have limited understanding regarding the mechanisms that are associated with how certain leaders occupy influential social positions. Method: Using social network data from district and site leaders, we conducted social network analysis and regression models to examine the relationship between a leader’s network position measured by incoming, outgoing, and close ties; personality traits; and leader self-efficacy controlling for demographics. Findings: Findings indicate that leaders with more incoming advice relationships from other leaders were associated with more years of experience in the district, being self-identified as “neurotic,” reporting higher efficacy in leading reform, and less efficacy in management. Leaders with more outgoing advice ties also self-identified as “neurotic” and reported lower efficacy in management. Leaders who were readily sought in terms of advice were explained by the number of years in the district and the self-reported

1University of California, San Diego, La Jolla, CA, USA 2California State University, Fullerton, Fullerton, CA, USA 3San Jose State University, San Jose, CA, USA

Corresponding Author: Alan J. Daly, University of California, San Diego, 9500 Gilman Drive, La Jolla, CA 92093-0070, USA. Email: [email protected]

492795EAQ50210.1177/0013161X13492795Educational Administration QuarterlyDaly et al. research-article2013

at TEMPLE UNIV on November 7, 2016eaq.sagepub.comDownloaded from

234 Educational Administration Quarterly 50(2)

personality traits of “neurotic,” “extraverted,” and “conscientious.” Leaders who were able to more efficiently connect to other leaders for advice were explained by being “neurotic,” “extraverted,” and “conscientious” as well as reporting low efficacy for management. Implications: Results suggest the importance of considering both personality traits as well as perceptions of efficacy in terms of understanding how leaders come to occupy influential social position in an advice network related to reform.

Keywords leadership, social networks, personality, efficacy, reform

In the last decade, there has been an increasing demand for educators to imple- ment systemic reforms in improving teaching and learning. School leaders are often viewed as central to reform efforts and are expected to lead initiatives toward increasing student outcomes or face significant sanction (Daly, 2009). Recent calls also suggest the need for increased coordinated and coherent efforts between central office and site leaders in enacting reforms (Honig, Copland, Rainey, Lorton, & Newton, 2010; Plecki et al., 2009). In response to these calls, a growing body of evidence suggests the importance of exchang- ing knowledge and expertise through “learning partnerships” between central office and school administrators as a way to both support and sustain change efforts (Honig, 2006; Honig, Lorton, & Copland, 2009; Knapp, Copland, Honig, Plecki, & Portin, 2010; Vander Ark, 2002).

In order to develop these vertical (across district and school) and horizontal (within district and school) learning partnerships, relationships must be in place to support the flow of knowledge and expertise necessary to enact reforms. The pattern of these vertical and horizontal ties creates a social network between and among leaders that may support and constrain efforts at change (Daly & Finnigan, 2011). This network structure affords opportunities and constraints not only for the larger network, but also for individual actors depending on their position in the social network infrastructure (Kilduff & Krackhardt, 2008; Krackhardt, 1999). To date, despite calls for improved social ties, relatively little research in the field of educational leadership investigates the patterns of interactions among and between district and school leaders during reform using social network theory and analysis. This study aims to respond to that gap as well as add to the larger field of social network research.

The social network literature provides important theoretical and practical insights on the impact of leadership activities as it suggests an actor’s

at TEMPLE UNIV on November 7, 2016eaq.sagepub.comDownloaded from

Daly et al. 235

position in a network is associated with a variety of outcomes—both for the individual and collective (Daly, 2010, Daly & Finnigan, 2011; Klein, Lim, Saltz, & Mayer, 2004). Individuals, such as leaders in our case, with rela- tively greater number of incoming and outgoing, as well as close, social ties are considered to be “central” in a social structure, while those with fewer ties are more “peripheral” in a network sense (Wasserman & Faust, 1994). Central actors tend to have access to more relational resources (e.g., advice, knowl- edge, and expertise) as well as having disproportionate influence over the movement of those resources given their network position (Kilduff & Krackhardt, 2008; Raider & Krackhardt, 2002). By occupying a more central location, an actor has increased opportunity to effectively access diverse information, which may provide that central actor with greater potential to steer, control, or even broker the flow of resources across a network (Burt, 2005; Kilduff & Krackhardt, 2008). While research suggests the importance of interpersonal interactions and collaborations between leaders in efforts at reform (Agullard & Goughnour, 2006; Daly & Finnigan, 2011), and the influ- ence of those who are “central” in a network (Balkundi & Harrison, 2006; Daly, 2010; Kilduff & Krackhardt, 2008), we have little empirical evidence to suggest the mechanisms through which these educational leaders come to occupy such potentially influential social positions.

In addressing this gap, we examine the extent to which a leader’s position in a social network can be understood through a leader’s personality traits and sense of efficacy. We argue that a leader’s personality traits and sense of effi- cacy may influence the extent to which that individual occupies a central posi- tion in disseminating or receiving relational resources related to efforts at reform, such as advice. In examining these variables, we use an exploratory case study approach that allows us to understand and open up the phenomena for investigation and theory development (Yin, 1984). We first describe the importance of a leader’s position in a social network and its potential influence in supporting and/or constraining resource flow in an advice network. We then generate initial hypotheses that link personality traits and leader self-efficacy with specific network positions as a way to explore associations between these variables. Finally, we present the results of our study of nearly 150 district and school leaders in underperforming systems and report findings from our sur- vey and network analyses. We conclude with a discussion of key findings, limitations, and implications for practice and research directions.

Theoretical Framework Given that an actor’s (in this case, educational leaders) informal network posi- tion may influence the flow of relational resources (advice), it is important to

at TEMPLE UNIV on November 7, 2016eaq.sagepub.comDownloaded from

236 Educational Administration Quarterly 50(2)

understand the factors that may be associated with individuals who occupy a central role in a social network. In the subsequent pages, we first argue the importance of position in a network and then discuss how individual personal- ity traits and leader self-efficacy relate to that leader’s network position.

Social Network Theory Social network theory provides insight into the social processes involved in reform that are stretched across individuals and levels of the educational sys- tem (Daly, 2012). Generally speaking, social network theory is concerned with the pattern of social ties that exists between actors in a social network (Scott, 2000). A social network perspective entails a move from a primary focus on the individual and the attributes of that actor to understanding the more dynamic supports and constraints of the larger social infrastructure (Borgatti & Foster, 2003; Cross, Borgatti, & Parker, 2002; Wellman & Berkowitz, 1998). Social network studies in education (e.g., Coburn & Russell, 2008; Cole & Weinbaum, 2010; Finnigan & Daly, 2012; Frank, Zhao, Penuel, Ellefson, & Porter, 2011; Moolenaar, 2010; Moolenaar, Daly, & Sleegers, 2011), as in other fields, primarily focus on how the constellation of relationships in networks may facilitate and constrain the flow of “rela- tional resources” (e.g., advice, knowledge, expertise, etc.), as well as provid- ing insight into how individuals gain access to, are influenced by, and leverage these resources (Degenne & Forsé, 1999). The network perspective does not supplant the importance of individual attributes in understanding reform but rather offers a complementary perspective and set of methods for better understanding the dynamic influence of social processes.

Several assumptions underlie social network theory and resulting research (Degenne & Forsé, 1999). First, actors in a social network are assumed to be interdependent rather than independent (Degenne & Forsé, 1999; Wasserman & Faust, 1994). Second, relationships are regarded as connections that pro- vide for the exchange of resources (Burt, 1982, 1997; Kilduff & Tsai, 2003; Powell, Koput, & Smith-Doerr, 1996). Third, the structure of a network has influence on the resources that flow to and from an actor (Borgatti & Foster, 2003). Fourth, patterns of relationships, captured by social networks, can act as both opportunities and constraints for individual and collective action (Brass & Burkhardt, 1993; Burt, 1982; Gulati, 1995). It is important to pres- ent these assumptions early on as these tenants are consequential to our approach and methodology.

In this study we use social network theory to conceptualize the social interactions between leaders at the district and site level. In building this argument, we draw on the work of Borgatti (2005, 2006), Burt (1992), Lin

at TEMPLE UNIV on November 7, 2016eaq.sagepub.comDownloaded from

Daly et al. 237

(2009), and other network structuralists1 who suggest that the network posi- tion of an individual to a large degree determines the quantity of social resources that flow to and from that actor (Borgatti & Everett, 1992; Burt, 1976, 1980; Kilduff & Krackhardt, 2008). An individual’s network position is determined by the pattern of incoming (indegree) and outgoing (outde- gree) social ties. Indegree refers to “incoming” relationships, meaning those that end with an actor and reflect the actor being sought by others for a relational resource. Outdegree reflects “outgoing” relationships, reflected in the number of ties that begin with an actor and as such represents the seeking of relational resources. For example, a particular leader may be sought for advice on reform efforts and the number of leaders seeking that specific individual would reflect that actor’s indegree. In contrast, that same focal leader may seek advice about reform efforts from other leaders and the number of relationships that individual seeks would reflect that actor’s outdegree. Individuals therefore can either be sought or seek rela- tional resources to differing degrees. So for example, one leader may be highly sought for advice, but not seek advice; another may do a great deal of seeking, but rarely be sought. Thus, in and out ties, although related, are conceptually distinct as leaders may be highly sought, but not seek, or rarely sought and frequently seek, or have some balance of the two (Clifton, Turkheimer, & Oltmanns, 2012; Wasserman & Faust, 1994; Zhang, Wang, Zhao, & Tang, 2012). It is this pattern of incoming and outgoing ties between and among actors that results in individuals occupying different “positions” in an overall social network. Some individuals in the network will be relatively central in being sought for advice (high indegree), others more central in frequently seeking advice (high outdegree), and still other individuals engaging in relatively few advice relationships moving them more to the periphery of the network.

Individuals who occupy a central position may have greater opportunities to access diverse resources as she/he has a larger number of social ties to other actors. Those who are on the periphery of a social network tend to have relatively fewer ties to others and thus may have less opportunity to receive or access resources held by central actors. It is the pattern of social ties that “determines” an actor’s network position, which in turn supports and con- strains the flow of resources to and from that actor (Burt, 2005). Despite potential outcomes for actors who occupy different network positions, in this study we are not exploring outcomes for individual actors or networks. Instead, we draw on the substantial literature acknowledging the importance of these positions and investigate potential underlying mechanisms that may provide insight into how actors (i.e., leaders) may come to occupy different positions in a network. In understanding network positions, we use individual

at TEMPLE UNIV on November 7, 2016eaq.sagepub.comDownloaded from

238 Educational Administration Quarterly 50(2)

network centrality (as discussed below) to represent the degree of influence (connectedness) a leader has to others in a network of advice about reform.

Informal network position. Network scholars argue that individuals who occupy different network positions also have different opportunities to access resources from other actors and as such may be at comparable advantage or disadvantage. Individuals that occupy a central position in a network are considered “hubs” and/or “connectors” in that they are more likely to have greater advantage to receive and/or disseminate resources (Ibarra & Andrews, 1993; Lin, 2009; Penuel, Riel, Krause, & Frank, 2009). Actors who are hubs in a social network are identified as central and have high-degree centralities (i.e., both incoming and outgoing ties), whereas those with relatively fewer incoming and outgoing ties are regarded as peripheral (Scott, 2000).

Central actors therefore are more likely to be sought by others for resources, as well as have more access to resources via the connections to multiple actors in a network (Alder & Kwon, 2002). By having more connec- tions to (outdegree) and from (indegree) others, these central actors have increased opportunities to seek as well as receive resources such as knowl- edge and information (Balkundi & Kilduff, 2005). In general, actors who occupy central positions tend to have greater influence over the network as the result of increased access to multiple actors and potential connections to new ties (Balkundi & Harrison, 2006; Burt, 1992; Friedkin, 2001; Kowalski, Petersen, & Fusarelli, 2007; Tsai, 2001; Wasserman & Faust, 1994).

In contrast, those actors who are more peripheral (i.e., having fewer incoming and outgoing social ties) in a social system may have less access to relational resources due to their limited social ties. This lack of ties may result in reduced opportunities to benefit from the resources held by those in central network positions (Borgatti & Everett, 1999; Cross & Parker, 2004). For example, if advice for knowledge and expertise about a reform are dis- seminated throughout a social system, those who are on the periphery, and have fewer ties, may have limited access to these resources compared to those who are more central. Therefore, peripheral actors may have less opportunity to receive knowledge from, and disseminate knowledge to, the rest of the social system and as such may have diminished influence. Moreover, peripheral actors are more likely to become “isolates” if the con- necting actors (i.e., central actors), who control resources that flow to the peripheral actors, are removed from the network (Everett & Borgatti, 1999). While peripheral actors have limited access to resources, they also have fewer social constraints on their behaviors in terms of social influence and responsibility to others who are tied by stronger connections. This lack of social constraints may allow these peripheral actors to develop “disruptive

at TEMPLE UNIV on November 7, 2016eaq.sagepub.comDownloaded from

Daly et al. 239

innovations” that enable the actors to access novel and nonredundant knowl- edge and information (Penuel & Riel, 2007).

While network scholars have examined network positions and their influ- ence on the flow and distribution of resources by taking into account the number of interactions among network actors, few examine the influence of “social distance” between these actors (Bourgeois & Friedkin, 2001; Friedkin, 1998). Closeness centrality, a network measure, takes into account the “social distance” between network actors. The social distance between actors can influence how efficiently an actor can directly reach others across the net- work for resources (Camagni, 1995; Helmsing, 2001; Knoke & Yang, 2008; Lawson, 1999; Mok, Wellman, & Basu, 2007). Social distance in this sense refers to the geodesic distance2 between actors. The extent to which one can reach all other network members through shorter distance denotes one’s closeness (Knoke & Yang, 2008), which may improve or dampen the effec- tiveness of resource flow (Uzzi, 1996). The higher an actor’s closeness cen- trality, the “closer” that actor is to others, and the more efficiently the resources dispensed by that actor will reach all other actors in the network (Knoke & Yang, 2008). Being close to others is considered to be beneficial as that actor has more efficient ties to a large number of individuals in the net- work (Maskell & Malmberg, 1999). In other words, actors who are close to each other (i.e., have the shortest paths to one another) in a network sense3 tend to be able to effectively and efficiently exchange resources across a net- work (Borgatti, 1995, 2005; Hanneman & Riddle, 2005; Okamoto, Chen, & Li, 2008; Wasserman & Faust, 1994). Therefore, the closer an actor is to other individuals, the better positioned that actor is to efficiently distribute (out- closeness) and receive (incloseness) advice from others within the network. Similar to outdegree and indegree, these closeness measures, while being related, are conceptually distinct in a directed network (Wasserman & Faust, 1994). In sum it is important to note that in- and outdegree and in- and out- closeness are also conceptually related as they are based on the pattern of incoming and outgoing ties among actors in a network. As such we would expect these measures to be correlated. However, in a directed network, these measures, while related, are also distinct as degree represents total in and out ties for an actor, while closeness is a measure of how efficiently an actor may be accessed by others in the overall network. These subtle nuances are poten- tially important in understanding the flow of advice in a network.

Given that these central and close actors are influential in moving resources across a network, exploring the factors associated with occupying these social positions is important to understand. In this study, network position (as measured by indegree, outdegree, incloseness, and outcloseness) serves as our dependent variable. In the next sections, we outline the independent

at TEMPLE UNIV on November 7, 2016eaq.sagepub.comDownloaded from

240 Educational Administration Quarterly 50(2)

variables of personality traits and leader self-efficacy. We hypothesize that traits and self-efficacy may be associated with a leader occupying a more central and close position in an advice network.

Personality Traits In addition to the importance of social structure, scholars outside of educa- tion have also suggested that personality traits of individuals may influence organizational change (Barrick, Mount, & Judge, 2001; Lau & Woodman, 1995; Salgado, 2003). Personality traits are reported to guide individual actions when performing executive tasks such as information screening, deci- sion-making, and advice seeking (Gino, Brooks, & Schweitzer, 2012). This suggests that these traits influence the way individuals enact change as well as orientations toward change (Arnulf, 2012; Elizur & Guttman, 1976; Oreg, 2006; Vakola, Tsaousis, & Nikolaou, 2003). Although a number of studies have examined personality structure and traits on organizational change, most of that work is outside of education (Caspi, Roberts, & Shiner, 2005; Cervone, 2005; Gino et al., 2012) and as such the role of personality traits has not been widely examined in terms of its role on educational reform efforts.

Personality has been conceptualized in the field of psychology by a vari- ety of theoretical perspectives noted for both its breadth and multiple dimen- sions (John, Hampson, & Goldberg, 1991; McAdams, 1995). Although there is significant debate, personality traits have been argued to be consistent, context-independent (McCrae & Costa, 1990) noncontingent in nature (Asendorpf & Wilpers, 1998), and thus are potentially useful in explaining various outcomes related to leadership across contexts (Hooker & McAdams, 2003). After decades of research, the Five Factor Model of Personality (FFM), sometimes referred to as the “Big Five,” codifies one such consensus (John & Srivastava, 1999), and is widely used and reported to measure per- sonality traits (Benet-Martinez & John, 1998; John, Donahue, & Kentle, 1991; John, Naumann, & Soto, 2008; Rammstedt & John, 2005, 2007).4

The Big Five model consists of five distinctive dimensions measured through a self-reported questionnaire: extraversion, conscientiousness, agree- ableness, openness, and neuroticism. Extraversion refers to an individual who reports being assertive, active, talkative, upbeat, energetic, optimistic (Costa & McCrae, 1992), sociable, and gregarious (Barrick & Mount, 1991). Conscientiousness is identified with being cautious, deliberate, self-disci- plined, well-organized (Costa & McCrae, 1992), dependable, hard-working, achievement-oriented, persevering, careful, and responsible (Barrick & Mount, 1991). Agreeableness refers to one reporting being cooperative, trust- ing, gentle, kind (Graziano & Eisenberg, 1997), courteous, flexible,

at TEMPLE UNIV on November 7, 2016eaq.sagepub.comDownloaded from

Daly et al. 241

good-natured, forgiving, and tolerant (Barrick & Mount, 1991). Openness is related to being creative, introspective, imaginative, resourceful, insightful (John & Srivistava, 1999), curious, and open-minded (Digman, 1990). Neuroticism refers to excessive feelings of fear, anxiety, worry, and insecu- rity (Barrick & Mount, 1991; Costa & McCrae, 1992).

Personality traits may influence the way individuals interact with others and the external environment. Research has examined the impact of personalities on individuals’ actions.5 Scholars have also investigated the association between the characteristics of effective leaders using personality traits and leadership (Judge, Bono, Ilies, & Gerhardt, 2002; Zaccaro, 2007). These stud- ies found that extraversion, conscientiousness, agreeableness, and openness have positive association with leadership (Bass, 1998; Bono & Judge, 2004; Hogan & Ones, 1997; Judge & Bono, 2000), whereas leaders who self-report as neurotic are less likely to exhibit strong leadership characteristics (Bass, 1985; Northouse, 1997). Extraverted leaders tend to be more positive, flexible, and influential and thus are more likely to generate energy and enthusiasm among others (Watson & Clark, 1997), which suggests they may have many outgoing and incoming social ties. Leaders who self-report as conscientious tend to be more dependable, hardworking (Witt, Burke, Barrick, & Mount, 2002), as well as being detailed and goal oriented (Hogan & Ones, 1997). They also tend to initiate changes and engage in management, goal setting, and attainment (Bass, 1998). Thus, it is likely that conscientious leaders may receive more incoming ties from others regarding organizational performance (e.g., reform in school system) as well as seek additional relationships.

Leaders who report being highly agreeable tend to be available to others and display kindness, modesty, trustworthiness, and individualized consid- eration (Costa & McCrae, 1992), and as such may have many incoming social relationships given their prosocial dispositions. Leaders rating high on openness tend to be creative, imaginative, and insightful (Judge & Bono, 2000) and may be well-positioned to innovate as they exhibit interest in a variety of novel ideas. Open leaders also tend to develop and share visions for the future of the organization (McCrae & John, 1992) and as such may possess more outgoing ties. Leaders who self-identify as neurotic tend to be associated with fewer positive leadership characteristics (Barrick, Parks, & Mount, 2005; Kumar, Bakhshi, & Rani, 2009; Strang & Kuhnert, 2009) as they often come across as anxious and tense resulting in a high degree of worry and concern, which often inhibits effective action (McCrae & John, 1992) and the development of relationships (Klein et al., 2004). Although research on personality traits suggests influence on a number of leadership characteristics and suggests certain patterns of ties, we have limited empiri- cal knowledge in educational leadership about how personality traits may be

at TEMPLE UNIV on November 7, 2016eaq.sagepub.comDownloaded from

242 Educational Administration Quarterly 50(2)

related to a leader’s position in a social network regarding the exchange of advice related to reform.

Relationship between personality traits and network positions. Based on the relevant literature described above, we synthesize key personality traits and their potential associations with leader positions in a social network. We pro- pose that leaders who are extraverted and conscientious in particular may have relatively more incoming and outgoing social ties (in- and outdegree) with others within a network as well as being able to efficiently exchange relational resources (in and out closeness) as they tend to initiate changes, generate excitement, exhibit dependability, and support others to reach goals. We hypothesize that leaders who are agreeable tend to have more incoming and close ties (indegree and incloseness) from a wide variety of other lead- ers as they make themselves available and are considered trustworthy. Based on the literature that suggests that open individuals tend to develop future visions and seek novel interactions and experiences, we propose that leaders who identify as open will be characterized by more outgoing and close ties to others (outdegree and outcloseness). Finally as the research indicates, neu- rotic leaders may not exhibit strong leadership characteristics (e.g., Kumar et al., 2009) as their high level of anxiety may be debilitating and inhibit pro- ductive action (see Barrick et al., 2001; McFarlane, 1989). We offer, as others scholars have suggested (Klein et al., 2004), that this high anxiety and tension will constrain overall incoming, outgoing, and close ties (in- and outdegree, and in- and outcloseness), which may limit these individuals from interac- tions and make them more peripheral in a network. In light of the literature, the following hypotheses are generated:

Hypothesis 1: The personality traits of extraversion and conscientiousness will be positively and significantly associated with the number and efficiency of incoming and outgoing advice ties (indegree, outdegree, incloseness, and outcloseness).

Hypothesis 2: The personality trait of agreeableness will be positively and significantly associated with the number and efficiency of incoming advice ties (indegree and incloseness).

Hypothesis 3: The personality trait of openness will be positively and sig- nificantly associated with the number and efficiency of outgoing advice ties (outdegree and outcloseness).

Hypothesis 4: The personality trait of neuroticism will be negatively and significantly associated with the number and efficiency of incoming and outgoing advice ties (indegree and outdegree, incloseness and outcloseness).

at TEMPLE UNIV on November 7, 2016eaq.sagepub.comDownloaded from

Daly et al. 243

Leader Self-Efficacy The increasing demands for schools to be accountable for the learning out- comes of the students have positioned school and district leaders to be even more active in reform efforts as well as experience the consequences for lack of performance. As stated earlier, personality traits may influence the way individuals exchange advice regarding reform. In addition, a leader’s percep- tion of her or his own self-efficacy to lead may also be related to both seeking and being sought for advice.

Self-efficacy in general refers to one’s belief that he or she can successfully take actions to accomplish certain tasks or achieve goals (Bandura, 1993, 1997). Bandura (1986) described self-efficacy as a person’s belief in their ability to “organize and execute courses of action required to attain designated types of performance” (p. 391). Efficacy has also been described to include “one’s capa- bilities to organize and execute the courses action required to manage prospec- tive situations” (Bandura, 1995, p. 2). Self-efficacy differs from personality traits, in that it reflects a potentially changing sense of confidence about one’s ability to accomplish particular activity, tasks, or goals within a specific context (e.g., school reform), whereas personality traits are considered to be a more gen- eral, somewhat fixed orientation to the broader world. As a result, a person’s self-efficacy may influence how much effort is devoted to a task and goals and how long one persists in those efforts in a particular setting (Bandura, 1977, 1986), while personality traits are the orientation toward those settings.

Efficacy as described has been studied widely in psychology, management, and other fields. In education, researchers have investigated several aspects of efficacy including students’ self-efficacy (Pajares, 1996, 1997), teachers’ self- efficacy (Tschannen-Moran, Woolfolk Hoy, & Hoy, 1998), and collective effi- cacy (Goddard, Hoy, & Woolfolk Hoy, 2000). However, very few have studied leaders’ self-efficacy (Daly, Der Martirosian, Ong-Dean, Park, & Wishard, 2011; Gareis & Tschannen-Moran, 2005; Smith, Guarino, Strom, & Reed, 2003), which has been identified as important in facilitating reform (Leithwood & Jantzi, 2008; Smith et al., 2003).

The few efficacy studies that have examined educational leaders (Daly et al., 2011; Imants & De Brabander, 1996; Leithwood & Jantzi, 2008; Louis, Leithwood, Wahlstrom, & Anderson, 2010; McCullers & Bozeman, 2010; Smith et al., 2003; Tschannen-Moran & Gareis, 2004, 2005) have identified at least two types of leader efficacy: instructional leadership and manage- ment. Efficacy for instructional leadership refers to one’s confidence and ability to lead and engage others in developing vision, promoting positive interactions, and generating enthusiasm for change. In contrast, management efficacy is related to handling paperwork, controlling schedules, and shaping operational policies. Both types of leadership efficacy are important in a

at TEMPLE UNIV on November 7, 2016eaq.sagepub.comDownloaded from

244 Educational Administration Quarterly 50(2)

reform context as educational leaders need to not only believe they can craft the vision, set direction, and develop the team, but also have the confidence to manage the change itself (McCormick, 2001; McCullers, 2009). However, studies have suggested that most leaders devote more time engaging in tech- nical management practices (e.g., paperwork, discipline, and parent issues) than they do in practicing instructional leadership, particularly in lower socioeconomic contexts (Amodeo & Taylor, 2004; Barnett, 2004; Smith et al., 2003). Since instructional leadership and management represent two distinctive types of efficacies that are associated with different skill sets and abilities, we suspect that they uniquely influence the leaders’ position within the network. As such, depending on the leader’s disposition, practice, and efficacy, he or she will interact with actors within the network accordingly, thus shaping his or her position within the network.

In a reform context, educational leaders are regarded as key agents of change within organization, and perceptions about their own ability to carry out change efforts with others are important (Carr, 2008; Cohen, 2006; Fullan, 2002, 2007; Lunenburg & Ornstein, 2008; Paglis & Green, 2002). A highly efficacious leader motivates others to develop, implement, and sustain change efforts (Pitre, 2003) and as such may be regarded as a source of exper- tise particularly in providing advice related to the effort as well as managing the technical elements of reform (Bandura, 2000; Tschannen-Moran & Gareis, 2004, 2005). As educational leaders play key roles in initiating change, creating buy-in, and implementing reforms, their confidence and belief in their ability to achieve these goals through and with others suggest the social aspect of efficacy (Tschannen-Moran & Gareis, 2004).

In support of the relationship between efficacy and the social nature of change, Penuel et al. (2009) found that leaders with higher self-efficacy tended to be central in a teacher social network as these educators often sought leaders for useful expertise. These efficacious leaders regarded them- selves as knowledgeable and skilled mentors who were able to motivate teachers and provide them with useful resources to improve their practices (Penuel et al., 2009). Expanding on the empirical research in this area, our work investigates the impact of two types of leader self-efficacy (instruc- tional leadership and management) on leaders’ incoming and outgoing advice ties between and among district and site leaders. Based on the literature, we offer the following hypothesis:

Hypothesis 5: Leader self-efficacy in instructional leadership and manage- ment will be positively and significantly associated with the number and efficiency of incoming and outgoing advice ties (indegree, outde- gree, incloseness, and outcloseness).

at TEMPLE UNIV on November 7, 2016eaq.sagepub.comDownloaded from

Daly et al. 245

Method We administered a Leadership Network Survey questionnaire to all princi- pals and district leaders in three school districts located in Southern California who serve similar student populations, are under sanction for continued underperformance under No Child Left Behind, and are implementing simi- lar literacy reform efforts to address student underperformance. We asked respondents to report on their demographic information and perceptions about the various social aspects related to their work, including “To whom do you turn to for reliable advice concerning the district’s literacy reform effort?” The questionnaire also asked respondents to self-assess personality traits and report perceptions of their own instructional leadership and management- related self-efficacy. We approach this work as an exploratory case study focused on understanding the relationship/interaction between social net- works, personality traits, and self-efficacy in underperforming systems (Yin, 1984). Our goal is to add to the limited empirical base both in education and in the social network literature regarding the interplay between the study variables and to better inform the development of theoretical propositions about the relationships between and among these concepts in terms of educa- tional leadership.

Sample In this exploratory study, we focused on school administrators in an effort to understand the overall leadership networks, thus expanding the research on the role of leadership in reform efforts in underperforming systems (DeVita, Colvin, Darling-Hammond, & Haycock, 2007; Honig et al., 2010). Seventy- two district leaders (i.e., superintendent, assistant superintendent, directors, and supervisors) and 76 school principals across three districts participated in the study, totaling 148. Survey response rates averaged 96% over the three districts with a range from 93% to 99%. Seventy-two of the participants came from District 1, 45 from District 2, and 31 from District 3. Females repre- sented 63% of the respondents in the overall sample. About half of the par- ticipants were White, approximately 32% of the participants were Latino, 9% Asian American, 7% Black, and 1% Other. Participants ranged in their years of working experience with 7% reporting serving as administrators for more than 25 years, 10% 20 to 25 years, 8% 15 to 19 years, 23% 10 to 14 years, 35% 5 to 9 years, and 15% less than 4 years of administrative experience. In general, participants had an average of 3 years of working experience in their district and less than 2 years working at the current site. Descriptive informa- tion for participants in the study is provided in Table 1.

at TEMPLE UNIV on November 7, 2016eaq.sagepub.comDownloaded from

246 Educational Administration Quarterly 50(2)

Procedure Participating school and district leaders were asked to complete an online survey in a 2-week period during the spring of 2011. Through the survey, we collected information about individual characteristics (gender, ethnicity, job title, work place, degree, and years of working experience) and measures of leader self-efficacy related to instructional leadership and management. In addition, leaders were asked to rate their own personality (extraversion, con- scientiousness, agreeableness, openness, and neuroticism) using a 5-point Likert-type scale (strongly disagree to strongly agree). We also collected a variety of social network questions about the frequency of advice seeking around reform efforts.

For our network data, we used a bounded/saturated approach (Lin, 1999; Scott, 2000), meaning that we included all the members of each of the leader- ship team (district and school administrators) as this strategy, coupled with high response rates, provides a more complete picture and more valid results (Scott, 2000). We provided each respondent with a list of all the leadership team members for each district. Following this procedure, we gathered com- plete network data in each of the districts.

Table 1. Characteristics of Participants in the Study.

N % M SD

Female 93 63 — — Male 55 37 Ethnicity White 74 50 — — Latino 48 32 — — Asian American 13 9 — — Black 11 7 — — Other 2 1 — — Highest degree attained — — Bachelor’s degree 13 9 — — Master’s degree 32 22 — — Master’s degree + 30 units 74 50 — — Doctoral degree 12 8 — — Working experience Years as administrator 148 — 2.93 1.55 Years in the district 148 — 3.03 1.46 Years at current site 148 — 1.76 1.08

Note. n = 148 leaders (72 district administrators and 76 school principals). Of all the partici- pants, 72 came from District 1, 45 from District 2, and 31 from District 3.

at TEMPLE UNIV on November 7, 2016eaq.sagepub.comDownloaded from

Daly et al. 247

Measures Dependent variables. In developing and validating our social network ques- tions, we drew upon previous research on network that used “advice” (Cross et al., 2002; Cross & Parker, 2004; Daly & Finnigan, 2010, 2011; Hite, Wil- liams, & Baugh, 2005; Klein et al., 2004). Advice relationships are important in disseminating knowledge and information by recognizing the advisory role of the experts, who often serve as a key channel for leaders to convey important messages regarding reforms. We piloted our relational questions with 49 practicing administrators from another district that were representa- tive of our sample, but not included in this study. For the final network sur- vey, respondents were asked to quantitatively assess their advice relationships with other administrators (district office and school) within their own district on a 4-point interaction frequency scale ranging from 1 (within the past 2 months) to 4 (1-2 times a week), based on the following prompt: “Please check the interaction frequency of those administrators to whom you turn for reliable advice concerning the district’s literacy reform effort.”

Indegree, outdegree, incloseness, and outcloseness. We calculated four net- work measures related to individual centrality in the advice network: inde- gree, outdegree, incloseness, and outcloseness (Borgatti, Jones, & Everett, 1998; Burt, 1983) using UCINET 6.0 (Borgatti, Everett, & Freeman, 2002). The four types of centrality offer conceptually distinct, although related, per- spectives on leaders’ network position. Advice indegree of an actor refers to the number of incoming ties/connections around advice an actor receives from other actors. By contrast, advice outdegree is the number of outgoing ties/connections regarding advice an actor possesses. For each of the degree measures, we normalized the scores to enable comparisons across districts. The normalized degree centrality ranges between 0 and 100 as percentage.

We also measured advice closeness centrality, which measures how close an actor is to the other actors in the network and how efficiently an actor can exchange advice with other actors (Wasserman & Faust, 1994). Closeness is the standardized form of farness, which measures the sum of the geodesic distances an actor has from and/or to the other actors in the network. The closeness index ranges between 0 and 1. The greater the closeness, the “closer” an individual is to all other actors, and as such the more efficiently and “quickly” that actor may be able to exchange resources. In this study, we are interested in the directional relations among school and district leaders in terms of advice-seeking and advice-receiving regarding school reforms. Thus, the closeness measure in a directional network can be divided into advice incloseness and advice outcloseness. Incloseness of actor n1 refers to

at TEMPLE UNIV on November 7, 2016eaq.sagepub.comDownloaded from

248 Educational Administration Quarterly 50(2)

the proportion of normalized number of shortest distances/paths required for other individuals in a network to access actor n1 in a directed network. Outcloseness of actor n1 refers to the proportion of normalized number of shortest distances/paths required for actor n1 to access other actors in a directed network. In this manner, we may regard the incloseness of an actor as an index of how “efficiently” an actor receives advice ties from others and the outcloseness as of how “efficiently” an actor seeks advice from others. The dependent variables therefore represent the degree of connectedness (in/ out-degree) and the minimum shortest efficient paths between these connec- tions (in/out-closeness) among leaders within each district regarding advice about reform efforts.

Independent variables. Based on our hypotheses, we selected variables that would allow us to explain the variance of the network position among the leaders within the school districts. Independent variables were selected to reflect two self-reported components: personality traits and perceptions of leader self-efficacy controlling for demographics variance.

Personality traits. The variables supporting this dimension were derived from the Big Five personality survey items regarding leaders’ self-assess- ment of their personality traits (extraversion, conscientiousness, agreeable- ness, openness, and neuroticism). The traits were derived from survey items, which asked the leaders to rate the extent to which they agree or disagree (1 = strongly disagree; 5 = strongly agree) with such statements as “I see myself as someone who is outgoing, sociable” and “I see myself as someone who is reserved” (reverse coded).

We administered the short version of the Big Five scale (BFI) since an increasing number of researchers have used this version of personality instru- ment, which has been demonstrated to be valid and reliable in measuring personality traits (Burisch, 1997; Gosling, Rentfrow, & Swann, 2003; John, Donahue et al., 1991). The structural validity of BFI scale has been shown to be replicable and the pattern of loadings is similar to those observed using the full version Big Five scale (FFM) (Rammstedt & John, 2007). We conducted factor analysis of the personality measures using principal component analy- sis of the survey items. Five factors were extracted and rotated using Varimax rotation, explaining 70% of the variance. Each component explains 8% to 25% of the variance with the Cronbach alphas ranging from .60 to .73.

Leader self-efficacy. We drew upon the Leadership Efficacy Scale used in Tschannen-Moran and Gareis’s (2004, 2005) studies as well as in our previ- ous studies (Daly et al., 2011) that examined the perceptions of both district

at TEMPLE UNIV on November 7, 2016eaq.sagepub.comDownloaded from

Daly et al. 249

office and principal self-efficacy, which allows us to capture leaders’ self- efficacy. Collectively, the efficacy scale comprised 18 items. All items began with the sentence stem, “In your current role as a leader, to what extent can you. …” Responses for each item were based on a 9-point Likert-type scale ranging from none at all to a great deal.

The efficacy measures were generated using principal component analysis for 18 survey items. The factors were rotated using Promax rotation, resulting in two factors with Eigenvalues greater than 1. The 18 original items were reduced to a scale of 15 items that explained 61% of the variance. Items that were removed had a communality of less than 0.50, or loaded on more than one factor, which resulted in two factor subscales. The first factor included nine items that were related to efficacy in instructional leadership practices with Eigenvalue of 6.65 and factor loadings ranging from 0.58 to 0.84, repre- senting 44% of the total variance in the efficacy items. Instructional leader- ship practices examined one’s confidence and ability to lead and perform leadership activities (e.g., motivation, culture building, leading change/ reform, etc.). An example item was “Promote district initiatives?” The sec- ond factor included six items that described efficacy in management practices with Eigenvalue of 2.49 and factor loadings ranging from 0.53 to 0.84, repre- senting 17% of the total variance of the efficacy items. Management practices examined the extent to which one believes in his/her ability to manage tech- nical job activities (e.g., paperwork, scheduling, prioritizing demands, etc.). An example item was “Maintain control of your daily schedule?” Table 2 provides factor coefficients of each item for efficacy variable.

Control variables. We controlled for demographic information that included position, degree, and years of working experience (i.e., the number of years working as an administrator, the number of years working in the district, and number of years at the current site).6 We chose to include all of these vari- ables in an effort to better understand advice exchanges as these social inter- actions may be due in part to the length of time one has been in a system or in a formal position as a proxy for knowledge and experience.

Data Analysis First, we conducted correlation analyses to examine the relationships between leaders’ network position, demographics, self-efficacy, and personality traits. Second, we conducted multiple regression analyses to test the relationships between leaders’ network positions and self-efficacy and personality traits after controlling for demographic characteristics. The model is specified as follows:

at TEMPLE UNIV on November 7, 2016eaq.sagepub.comDownloaded from

250 Educational Administration Quarterly 50(2)

Outcome = β0 + β1(Position) + β2(Degree) + β3(Years Administration) + β4(Years District) + β5(Years School) + β6(Extraversion) + β7(Conscientiousness) + β8(Agreeableness) + β9(Openness) + β10(Neuroticism) + β11(Instructional Leadership Practices) +

β12(Management Practices) + ε

Finally, we provided a network sociogram from one of the study school dis- tricts as a representative illustration of our findings regarding the relationship between a leaders’ network position and personality traits. We used NetDraw (Borgatti, 2002) to generate the sociogram that contains nodes (individual leaders), ties (advice-seeking and -receiving connections between leaders), and outcome information of our variables in the social network, which is discussed in detail in the Results section.

Before conducting analyses, we addressed two important methodological concerns: (1) examining the nested nature of the data and (2) assumption of independence of observations while handling social network data. First, since we selected three school districts in our sample, we performed mean

Table 2. Exploratory Factor Analysis for Leader Efficacy Scales.

Survey Items 1 (α = .90) 2 (α = .87)

Efficacy: Instructional leadership practices Motivate district or school staff? .842 .160 Generate enthusiasm for a shared vision? .813 .106 Promote positive interactions among school or

district members? .800 .190

Create a positive environment? .771 .272 Promote district initiatives? .699 .314 Facilitate learning? .686 .014 Improve student achievement? .678 .025 Promote ethical behavior? .650 .299 Manage change? .582 .476 Efficacy: Management practices Cope with the stress of the job? .074 .837 Prioritize among competing demands of the job? .170 .827 Handle the paperwork required of the job? .087 .823 Maintain control of your daily schedule? .186 .784 Handle the time demands of the job? .164 .778 Shape the necessary operational policies and procedures to manage your site or your department?

.350 .532

Note. n = 148 leaders. Bold type corresponds with the factors identified with each scale.

at TEMPLE UNIV on November 7, 2016eaq.sagepub.comDownloaded from

Daly et al. 251

comparison tests as well as running similar models for each district to examine the consistency of results. The results indicate that the independent variables were not significantly different across the three sites (F = 0.62 to 2.96, p > .05) and that the models offered similar findings. Given these results, we pooled the data from both district and school leaders for subsequent analyses.

Regarding the second concern in terms of statistical tests and violations of assumptions of independence given the intercorrelated nature of network data (Borgatti et al., 1998), we considered various statistical approaches7 to the data. We considered p* models, which would enable us to predict dyadic rela- tions among pairs of actors, but that was not the intent of this study. Instead we were interested in determining the associations between the number of ties an actor has in a directed network and the independent variables. In order to understand the association between different outcome measures (in/out-degree and in/out-closeness) and the study variables (personality traits and self-effi- cacy), we use conventional regression that allows us to study the number of relationships, as opposed to predicting dyadic ties that school leaders have with their colleagues. We recognize that network measures are interrelated, but we argue based on the literature they are conceptually distinct as noted in several network studies (Clifton et al., 2012; Zhang et al., 2012). Therefore, using regression models is appropriate since we are mostly interested in understand- ing the number and type of social interactions within a network. Given this strategy (Borgatti et al., 1998), we may also assume less concern about inflat- ing standard errors based on multicollinearity between network measures.

Results

Descriptive Analyses We summarized descriptive statistics for demographics, leader self-efficacy, and personality traits (see Table 3). Findings indicate that leaders reported themselves as extraverted, conscientious, agreeable, and open, and tended to regard themselves with less neuroticism. Leaders also indicated that they are efficacious in dealing with instructional leadership (M = 7.33, SD = 1.18) and management (M = 6.99, SD = 1.33), meaning that they feel competent to perform practices in both areas. In terms of network position in degree centrality, leaders receive and seek advice from about 3% of all possible leaders (mean normalized indegree = 3.26; mean normalized outdegree = 3.41, respectively) in the network. The degrees of how efficiently (i.e., the shortest distance) one actor can efficiently receive (incloseness) and seek (outcloseness) advice are similar. Leaders were able to connect to 9% of other leaders at the minimum shortest paths to receive advice (mean

at TEMPLE UNIV on November 7, 2016eaq.sagepub.comDownloaded from

252 Educational Administration Quarterly 50(2)

normalized incloseness = 0.09, SD = 0.12). Similarly, leaders were able to connect to 9% of all potential connections at the minimum shortest paths when seeking advice (mean normalized outcloseness = 0.09, SD = 0.11) related to reform practices. Results suggest that leaders possess similar net- work position in terms of receiving and seeking advice related to reforms as well as perceive higher levels of efficacy and all personality traits with the exception of neuroticism.

Relationship Between Network Position, Personality Traits, and Leader Self-Efficacy Correlation analysis as shown in Table 4 provided evidence in explaining our hypotheses. The results of correlation analysis indicated that personality traits such as extraversion, conscientiousness, agreeableness, openness, and neu- roticism were positively and significantly correlated with the distinct advice network measures (indegree, outdegree, incloseness, and outcloseness). Among the relationships between personality traits and network position, neu- roticism, somewhat surprisingly, is positively and highly correlated with incloseness and outcloseness, suggesting that leaders who perceived them- selves as neurotics tended to efficiently receive and seek advice from others at a short distance. In addition, we found negative significant associations

Table 3. Descriptive Statistics for Variables Used in the Analysis.

Variables Minimum Maximum M SD

Personality traits Extraversion 1.0 5.0 3.58 1.12 Conscientiousness 1.0 5.0 4.43 1.09 Agreeableness 1.0 5.0 4.37 0.97 Openness 1.0 5.0 3.85 1.14 Neuroticism 1.0 5.0 2.74 1.41 Leader efficacy Instructional leadership practices 2.3 9.0 7.33 1.18 Management practices 3.5 9.0 6.99 1.33 Leader network position Advice indegree 0.00 44.44 3.26 6.54 Advice outdegree 0.00 30.36 3.41 5.51 Advice incloseness 0.01 0.55 0.09 0.12 Advice outcloseness 0.01 0.44 0.09 0.11

Note. n = 148 leaders.

at TEMPLE UNIV on November 7, 2016eaq.sagepub.comDownloaded from

Daly et al. 253

between leader self-efficacy in management practices and network measures (i.e., advice outdegree and outcloseness). This suggests that those with a high level of self-efficacy in management practices are less likely to seek advice from others and less likely to have the opportunity to seek advice from others at a short distance. Self-efficacy in instructional leadership practices was not significantly associated with any of the outcome variables. These correlations indicate that an educational leader’s network position in this sample is associ- ated with individual personality traits and his/her beliefs in the ability to address managerial tasks. We also found no significant relationship between leader self-efficacy and personality traits as has been suggested in studies out- side of education (e.g., Thoms, Moore, & Scott, 1996).

The Role of Demographic Variables in Leaders’ Social Network Position To understand whether demographic characteristics of district and school leaders played a role, we examined the extent to which demographic vari- ables were associated with leaders’ network position (see Table 5). We found that leaders who had more years of experience in the district tended to occupy a central and close position within the advice network. That is, those leaders with longer tenure in the district tended to have greater percentage of incom- ing ties (β = .17, p < .05) as well as close advice receiving relationships with others (β = .23, p < .01). However, tenure is not a significant predictor of advice-seeking behaviors (outdegree and outcloseness). All other demo- graphic variables (years in administration and the school) were found to have no effect on seeking and receiving advice.

Table 4. Correlation Matrix for Variables Used in the Analysis.

1 2 3 4 5 6 7 8 9 10 11

1. Extraversion — 2. Conscientiousness .40** — 3. Agreeableness .43** .66** 4. Openness .38** .50** .62** — 5. Neuroticism .09 .17* .21* .43** — 6. Efficacy: leadership .14 .06 .03 –.03 .12 — 7. Efficacy: management .15 .26** .12 –.14 .11 .46** — 8. Advice indegree .26** .20* .23** .39** .27** .10 –.11 — 9. Advice outdegree .22** .23** .24** .28** .18* –.02 –.18* .36** — 10. Advice incloseness .37** .32** .37** .65** .38** –.04 –.13 .62** .61** — 11. Advice outcloseness .31** .35** .38** .62** .39** –.08 –.18* .50** .71** .88** —

Note. n = 148 leaders. *p < .05. **p < .01.

at TEMPLE UNIV on November 7, 2016eaq.sagepub.comDownloaded from

254

T ab

le 5

. R

es ul

ts F

ro m

t he

M ul

ti pl

e R

eg re

ss io

n A

na ly

se s

by D

ep en

de nt

V ar

ia bl

e.

A dv

ic e

In de

gr ee

A dv

ic e

O ut

de gr

ee A

dv ic

e In

cl o se

ne ss

A dv

ic e

O ut

cl o se

ne ss

V ar

ia bl

e B

SE B

β B

SE B

β B

SE B

β B

SE B

β

Po si

ti o n

–0 .7

0 0.

73 –0

.1 0

0. 06

0. 64

0. 01

–0 .0

1 0.

15 –0

.0 0

–0 .0

0 0.

15 –0

.0 0

D eg

re e

0. 88

0. 56

0. 17

0. 64

0. 49

0. 15

0. 07

0. 12

0. 05

0. 12

0. 11

0. 09

Y ea

rs in

a dm

in is

tr at

io n

0. 03

0. 39

0. 01

–0 .3

4 0.

34 –0

.0 9

–0 .0

9 0.

08 –0

.0 8

–0 .0

4 0.

08 –0

.0 4

Y ea

rs in

d is

tr ic

t 0.

80 0.

40 0.

17 *

0. 28

0. 35

0. 07

0. 23

0. 08

0. 18

** 0.

10 0.

08 0.

08 Y

ea rs

in s

ch o o l

–1 .0

9 0.

63 –0

.1 5

–0 .6

6 0.

55 –0

.1 1

–0 .2

6 0.

13 –0

.1 3

–0 .1

6 0.

13 –0

.0 9

Ex tr

av er

si o n

0. 80

0. 51

0. 14

0. 58

0. 45

0. 12

0. 43

0. 11

0. 26

** 0.

24 0.

10 0.

16 *

C o ns

ci en

ti o us

ne ss

1. 05

0. 78

0. 17

1. 15

0. 69

0. 23

0. 39

0. 16

0. 23

* 0.

39 0.

16 0.

25 *

A gr

ee ab

le ne

ss 0.

10 0.

84 0.

01 0.

49 0.

73 0.

08 –0

.0 7

0. 18

–0 .0

3 0.

12 0.

17 0.

07 O

pe nn

es s

–0 .0

1 0.

63 0.

00 –0

.3 6

0. 55

–0 .0

7 –0

.1 6

0. 13

–0 .0

9 –0

.0 5

0. 13

–0 .0

3 N

eu ro

ti ci

sm 1.

65 0.

42 0.

35 **

0. 86

0. 37

0. 22

* 0.

87 0.

09 0.

64 **

0. 68

0. 09

0. 55

** Ef

fic ac

y: le

ad er

sh ip

1. 16

0. 50

0. 21

* 0.

46 0.

44 0.

10 0.

05 0.

11 0.

03 0.

03 0.

10 0.

02 Ef

fic ac

y: m

an ag

em en

t –0

.9 2

0. 45

–0 .1

8* –1

.0 5

0. 40

–0 .2

5* *

–0 .1

1 0.

10 –0

.0 7

–0 .2

3 0.

09 –0

.1 8*

R2 0.

31 0.

24 0.

62 0.

56 F

4. 60

** 3.

25 **

17 .3

1* *

13 .3

3* *

N ot

e. S

ta nd

ar di

ze d

be ta

. n =

1 48

le ad

er s.

V ar

ia nc

e in

fla ti o n

fa ct

o rs

( V

IF s)

fo r

ea ch

m o de

l r an

ge fr

o m

1 .3

t o 2

.9 .

*p <

.0 5.

* *p

< .0

1.

at TEMPLE UNIV on November 7, 2016eaq.sagepub.comDownloaded from

Daly et al. 255

The Relationships Between Advice Indegree, Personality, and Leader Self-Efficacy The model was significant in explaining 31% of the variance in advice inde- gree.8,9 Each of the independent variables contributes independently to the pre- diction of advice indegree. The results indicate that after controlling for position, advanced degree, years of experiences, self-efficacy in instructional leadership and management, and other personality traits, advice indegree is positively associated with neuroticism (β = .35, p < .01). Similarly, after con- trolling for demographic characteristics, personality traits, and self-efficacy in management, self-efficacy in instructional leadership practices is positively associated with advice indegree (β = .21, p < .05). In contrast, after controlling for demographic characteristics, personality traits, and self-efficacy in instruc- tional leadership practices, self-efficacy in management practices is negatively associated with advice indegree (β = -.18, p < .05). In sum, leaders who self- identified as neurotic along with those who perceived themselves having the efficacy to lead reforms and those leaders who have less managerial abilities were more likely to have others come to them for advice. Contrary to Hypotheses 2 and 3, agreeableness and openness did not contribute to the explanation of a leader’s indegree. Similarly but surprisingly, Hypothesis 4 is rejected because our data indicate that a leader’s advice indegree can be positively explained by personality traits such as neuroticism. Partially supporting Hypothesis 5, which indicates that individuals exchange advice with more efficacious leaders in instruction and management, only those reporting greater leader self-efficacy for instructional leadership tended to be sought for advice, and those who have greater self-efficacy for management were less likely to be sought for advice.

The Relationships Between Advice Outdegree, Personality, and Leader Self-Efficacy The model was significant in explaining 24% of the variance in advice outde- gree.10 Each of the independent variables contributes independently to advice outdegree. Controlling for position, degree, and years of experience, self- efficacy in instructional leadership and management practices, and other per- sonality traits, the results showed that neuroticism is positively associated with advice outdegree (β = .22, p < .05). In contrast, after controlling for position, degree, and years of experience, self-efficacy in instructional lead- ership, and other personality traits, self-efficacy in management practices is negatively associated with advice outdegree (β = -.25, p < .01). Leaders who reported neuroticism were more likely to seek advice on reform, and those who perceived to have greater self-efficacy in managerial abilities were less

at TEMPLE UNIV on November 7, 2016eaq.sagepub.comDownloaded from

256 Educational Administration Quarterly 50(2)

likely to seek advice from others. Hypotheses 2 and 3 were rejected, as agree- ableness and openness were not associated with a leader’s outdegree network position. This finding is also contrary to Hypothesis 4 and suggests that lead- ers who self-identified as neurotic are positively associated with advice out- degree. Hypothesis 5 was not supported as the result indicates that a leader with higher self-efficacy in managerial practices is less likely to seek advice from others. In other words, those who are confident in their ability to man- age technical and procedural aspects of their work tended not to seek advice from many of their colleagues regarding reform.

The Relationships Between Advice Incloseness, Personality, and Leader Self-Efficacy The model was significant in explaining 62% of the variance in advice incloseness.11 Each of the independent variables contributes independently to advice incloseness. The results showed that controlling for position, degree, and years of experiences, self-efficacy in instructional leadership and man- agement practices, and other personality traits, extraversion was positively associated with advice incloseness (β = .26, p < .01). Similarly, after control- ling for position, degree, and years of experiences, self-efficacy in instruc- tional leadership and management practices, and other personality traits, conscientiousness was positively associated with advice incloseness (β = .23, p < .05). Lastly, after controlling for position, degree, and years of experi- ences, self-efficacy in instructional leadership and management practices, and other personality traits, neuroticism was positively associated with advice incloseness (β = .64, p < .01). This can be interpreted as leaders who have identified themselves as being extraverted, conscientious, and neurotic are more likely to have a greater number of close advice ties with others. In other words, such leaders tended to be efficiently connected to others through the minimum shortest paths and act as a type of information “drain” for advice on reform. Moreover, neuroticism explained a leader’s advice incloseness almost three times more than did the extraversion and conscientiousness, indicating the significant role neurotics play in understanding a leader’s advice incloseness. Consistent with Hypothesis 1, the personality traits of extraversion and conscientiousness are positive and significant predictors of a leader’s advice incloseness. Hypotheses 2 and 3 were rejected as agreeable- ness and openness did not significantly explain a leader’s incloseness net- work position. Contrary to Hypothesis 4, neuroticism was positively related to a leader’s incloseness network position. However, these results did not offer support for Hypothesis 5 as a leader’s advice incloseness is not explained by either aspect of leader self-efficacy.

at TEMPLE UNIV on November 7, 2016eaq.sagepub.comDownloaded from

Daly et al. 257

The Relationships Between Advice Outcloseness, Personality, and Leader Self-Efficacy The model was significant in explaining 56% of the variance in advice out- closeness.12 Each of the independent variables contributes independently to the prediction of advice outcloseness. The results showed that controlling for position, degree, and years of experience, self-efficacy in instructional lead- ership and management practices, and other personality traits, extraversion is positively associated with advice outcloseness (β = .16, p < .05). Similarly, controlling for position, degree, and years of experience, self-efficacy in instructional leadership and management practices, and other personality traits, conscientiousness is positively related to advice outcloseness (β = .25, p < .05). Controlling for the same characteristics, neuroticism is positively associated with advice outcloseness (β = .55, p < .01). Interestingly, after controlling for position, degree, and years of experience, self-efficacy in instructional leadership practices, and other personality traits, self-efficacy in management practices is negatively associated with advice outcloseness (β = -.18, p < .05). These results suggest that leaders who reported being extra- verted, conscientious, neurotic, and those who perceived themselves to have low self-efficacy in managerial practices are more likely to efficiently dis- seminate or seek advice to others through the minimum shortest paths. Thus, these leaders have efficient access to multiple actors across the network regarding advice about reform efforts. Again, neuroticism had a stronger association to advice outcloseness compared to extraversion and conscien- tiousness, indicating the significant role that neuroticism plays in understand- ing a leader’s network position. As predicted in Hypothesis 1, a leader’s outcloseness network position is explained by personality traits such as extra- version and conscientiousness. Hypotheses 2 and 3 were rejected, as agree- ableness and openness were not associated with a leader’s advice outcloseness. However, once again contrary to Hypothesis 4, neuroticism did not nega- tively explain a leader’s network position, but rather, was positively related to outcloseness. Hypothesis 5 was also not supported in predicting a leader’s self-efficacy in management since the result indicated that a leader with higher self-efficacy in management practices reported having less efficient advice-seeking relationships from other leaders in the network.

Advice Network Sociogram: The Relationship Between Network Position and Personality The results from our network analyses are also displayed in a sociogram. We provide a representative example of advice network from one of our sample

at TEMPLE UNIV on November 7, 2016eaq.sagepub.comDownloaded from

258 Educational Administration Quarterly 50(2)

school districts (see Figure 1) to illustrate the relationship between a leader’s network position (e.g., indegree) and personality traits (e.g., neuroticism). Nodes in Figure 1 represent individual district and school leaders, and the lines between nodes represent advice ties between those leaders. The size of nodes indicates the relative number of advice ties an individual leader receives from others, with larger nodes reflecting more incoming ties. These nodes were also shaped by three levels of neuroticism, with a diamond repre- senting a high level of neuroticism, up triangle reflecting midrange neuroti- cism, and a down triangle describing the least amount of reported neuroticism. The advice network indicates that leaders with middle to high levels of neu- roticism tend to have a greater number of incoming ties from others regarding reform advice. These leaders tended to be sought for advice regarding reform and have close relationships (closely connected by the minimum shortest paths) with others. With regard to relationships between other types of leader network position and leader self-efficacy as well as other personality traits, we found a similar network pattern of interactions among these leaders.

Table 6 summarizes our results regarding the study’s five hypotheses. Of the five hypotheses, two were partially supported and three were not sup- ported. Partial support in the area of personality traits of extraversion and con- scientiousness denote that the personality traits only had significant effects on two of the four outcome variables (i.e., incloseness and outcloseness) and thus only partially supported our hypothesis. The other partially supported

Figure 1. Advice Network of Indegree Leader Centrality Note. Network size = 37. Nodes were sized by indegree and shaped by level of neuroticism (high = diamond; middle = up triangle; and low = down-triangle).

at TEMPLE UNIV on November 7, 2016eaq.sagepub.comDownloaded from

Daly et al. 259

hypothesis had to do with the fact that the leader self-efficacy variable had two distinct features—instructional leadership and management practices—which may influence the direction of the significant effect on the outcome variables. Hypotheses 2 and 3 were not supported by the data, as we found no significant effects of agreeableness and openness on the network outcome variables. Finally, Hypothesis 4 was not supported, as we found the effect of neuroticism on all four outcome variables was completely counter to our hypothesis.

Discussion A number of studies have referenced the importance of interpersonal social relationships (Borgatti & Foster, 2003; Daly & Finnigan, 2012; Liou & Daly, in press; Penuel et al., 2009; Penuel et al., 2010) in enacting reforms, but to date we have only limited knowledge about the mechanisms that may be

Table 6. Results of Study Hypotheses.

Hypothesis Results Findings

Hypothesis 1: The personality traits of extraversion and conscientiousness will be positively and significantly associated with the number and efficiency of incoming and outgoing advice ties (indegree, outdegree, incloseness, and outcloseness).

Partially supported Extraversion and conscientiousness positively explained the incloseness and outcloseness of a leader.

Hypothesis 2: The personality trait of agreeableness will be positively and significantly associated with the number and efficiency of incoming advice ties (indegree and incloseness).

Not supported Agreeableness did not significantly explain a leader’s incoming advice ties.

Hypothesis 3: The personality trait of openness will be positively and significantly associated with the number and efficiency of outgoing advice ties (outdegree and outcloseness).

Not supported Openness did not significantly explain a leader’s outgoing advice ties.

Hypothesis 4: The personality trait of neuroticism will be negatively and significantly associated with the number and efficiency of incoming and outgoing advice ties (indegree and outdegree, incloseness and outcloseness).

Not supported Neuroticism positively and significantly explained a leader’s incoming and outgoing advice ties.

Hypothesis 5: Leader self-efficacy in instructional leadership and management will be positively and significantly associated with the number and efficiency of incoming and outgoing advice ties (indegree, outdegree, incloseness, and outcloseness).

Partially supported Leader self-efficacy in instructional leadership was positively associated with the indegree of a leader. However, self-efficacy in management practices was negatively related to indegree, outdegree, and outcloseness of a leader.

at TEMPLE UNIV on November 7, 2016eaq.sagepub.comDownloaded from

260 Educational Administration Quarterly 50(2)

associated with leaders who play influential roles in social systems. This study not only extends our understanding of the personality traits found in other research domains (i.e., management, psychology, etc.), but connects the idea of leader self-efficacy as an additional and important variable in understand- ing how leaders come to occupy different network positions. Our overall find- ings suggest that leaders who occupy a central position in an advice network was associated with years working in the district, a higher level of neuroti- cism, and to a lesser degree extraversion and conscientiousness, as well as efficacy in instructional leadership. The other finding points toward a leader’s central position in an advice-seeking network, which was explained by neu- roticism, extraversion, conscientiousness, and self-efficacy in management practices. A number of these findings relate to the existing literature, while some run counter to what may be expected. We will discuss the major findings below and consider implications of this work.

Leader Personality and Social Network Structure Our analysis offers interesting findings that are partially consistent with pre- vious research on personality traits and network centralities (i.e., measures of actor’s social status and network position; see Bendersky & Shah, 2012; Wehrli, 2008). Our overall findings indicate that personality traits, such as extraversion, conscientiousness, and neuroticism, each make significant con- tribution in explaining a leader’s advice seeking and receiving in terms of the number of in and out ties. This result is even stronger on the level of effi- ciently connecting to other leaders (in- and outcloseness). We will discuss each of the major traits and our specific findings below.

Extraverted and conscientious leaders are more likely to be close to other leaders and efficiently exchange (i.e., both seek and receive) advice through minimum shortest paths. This may not be a surprise considering extraverted leaders tend to be more positive, energetic, interpersonally dominant (Caspi et al., 2005; DeYoung, Quilty, & Peterson, 2007) as well as engaging others in building coalitions (Cable & Judge, 2003)—characteristics that may influ- ence the way these individuals interact with colleagues. Thus, these extra- verted leaders are more likely to be sought for and seek advice. However, there may also be a negative effect of having an extraverted personality, in that these leaders may be perceived as being too assertive or bold. These extraverted individuals may not be aware of the negative perceptions others have toward them and thus may develop negative relationships over time (Bendersky & Shah, 2012; Klein et al., 2004).

As for conscientious leaders, previous research suggests that they tend to receive more advice ties from others regarding organization performance as

at TEMPLE UNIV on November 7, 2016eaq.sagepub.comDownloaded from

Daly et al. 261

well as disseminate advice to others regarding change efforts (Bass, 1998; Witt et al., 2002). Our findings support this point in that educational leaders who reported being consciousness both seek and are sought for advice. In addition to having more overall advice ties, these conscientious leaders also tend to efficiently connect with others within the shortest path distance, meaning that these leaders can easily reach all other leaders more efficiently in terms of exchanging advice related to reform.

However, contrary to research that equates neuroticism with lower levels of both seeking and being sought for advice (Barrick et al., 2005; Bono & Judge, 2003; Klein et al., 2004; Kumar et al., 2009; McCrae & John, 1992; Strang & Kuhnert, 2009), our results indicate that neuroticism explained much of the leader centrality around the reform advice network both in terms of seeking and being sought for advice. Our data suggest that leaders who reported being neurotic tended to have more and efficient (close) connections to and from others for advice. This finding runs counter to what may have been expected based on the literature, and as such we offer an alternative explanation, which draws on the importance of context.

The Potential Role of Context on Social Processes and Personality We developed our hypothesis based on the personality trait literature that sug- gests the relationships between traits, leadership, and organizational change (Judge, Thoresen, Pucik, & Welbourne, 1999; Vakola et al., 2003). The major- ity of this work has been conducted in organizations outside of education (Caspi et al., 2005; Cervone, 2005; Gino et al., 2012) and typically suggests that personality traits are generally fixed and are context independent (Asendorpf & Wilpers, 1998; Hooker & McAdams, 2003; McCrae & Costa, 1990), meaning that one’s personality is consistent regardless of the larger context in which that individual resides. We therefore framed this study, as others outside of education have, in terms of understanding the mechanisms that may provide insights into how a leader comes to occupy an influential social position without regard to the larger context in which these individuals operate. Although we limited the scope of our study by focusing on districts in similar organizational contexts in terms of underperformance, as a way to “account” for context, we acknowledge that other types of contexts (e.g., fed- eral policy, state systems, and district) may influence how leadership practices and personality traits interact, as these layered contexts are not necessarily cohesive, coherent, or mutually exclusive. Furthermore, we did not take into account how the larger context may actually play a role in terms of personality and advice exchanges. Therefore, the role of multiple and sanction-heavy

at TEMPLE UNIV on November 7, 2016eaq.sagepub.comDownloaded from

262 Educational Administration Quarterly 50(2)

contexts, as we will suggest below, may well provide a reasonable alternative hypothesis to our findings as well as an important additional perspective that needs to be taken into account in terms of understanding the interplay and effects on advice exchanges, personality, and efficacy.

It seems reasonable to suspect as an alternative explanatory hypothesis that the intense and pressure-filled environments in which these leaders inter- act may affect personality and the role in which “neurotics” end up playing in social interactions between and among educational leaders. Previous research has found that neuroticism can negatively affect social relationships, but this work has typically taken place without attending to the larger context in which these individuals operate (Barrick et al., 2005; Kumar et al., 2009; Strang & Kuhnert, 2009). As federal and state accountability pressure increases, district and school leaders are required to respond to the demands of school improvement or face potential sanction, which may also have per- sonal implications including loss of employment. Perhaps leaders with char- acteristics of neuroticism (e.g., heightened anxiety, tension, nervousness, and worry) about meeting the demands for improving performance in a context of sanction may have become better “positioned” or even more “motivated” due to anxiety to both seek and be sought for advice about reform. In this sense, the larger context may have played a role in elevating these leaders to become “hubs” of advice. It may also be that the larger context served to “enhance” already existing traits in a leader or even “suppress” other traits. In this way, perhaps the larger context influenced how leaders saw and reported about themselves in terms of personality.

In support of these ideas, previous work in underperforming systems sug- gests that in these high-pressure environments school leaders may become more bureaucratic and rule bound and as such inhibit risk-taking and devel- opment of novel linkages with other leaders and educators (Daly, 2009; Daly & Finnigan, 2011, 2012; Finnigan & Gross, 2007). Moreover, as this study took place in California, a system marked by deep budget cuts and rationing of resources, this may also add to the pressure in these systems and in turn on educational leaders. It may be this “perfect storm” of public scrutiny, account- ability, increasing sanction, dwindling resources, and personal threat to pro- fessional futures that may actually influence educational leaders to self-identify as worried, anxious, and insecure, all traits of the neurotic. Alternatively, those who are already neurotic may just move more to the fore- ground in these settings. Either way this may not portend well for these sys- tems as those leaders that tend to be anxious also tend to not seek new opportunities, novel connections, or engage in risk taking, the same charac- teristics that may be necessary for systems to innovate and improve (Moolenaar, Daly, & Sleegers, 2010). Therefore, while other work suggests

at TEMPLE UNIV on November 7, 2016eaq.sagepub.comDownloaded from

Daly et al. 263

that a sanction-heavy environment may have unintended consequences for improvement, this current study may also indicate that the same high-inten- sity environment may undermine productive social processes and potentially suppress or enhance existing personality traits. Perhaps ultimately these lead- ers are merely reflecting the larger “neurotic” environment, meaning the tense, high-pressure, and anxiety-filled, accountability context that often sur- rounds underperforming systems. We offer that this line of inquiry would be an excellent area for additional work, particularly given the upcoming reau- thorization of accountability policies.

This alternative explanation suggests, as other authors have written (Knapp & Feldman, 2012; Louis & Robinson, 2012; Pitts & Spillane, 2009), that greater attention must be paid to contexts in general, policy environment in particular (i.e., the rewards and sanctions associated with performance), and the overall district/school context. We suspect that these contextual fac- tors may suppress or enhance personality traits and consequently the relation- ship between personality traits and social interactions. If this is the case, then it is reasonable to speculate that personality traits may not be as context inde- pendent as is suggested in the literature. Therefore, it is likely that the rela- tionship between these variables may be an interactive process that evolves over time within a specific context. Future research including longitudinal studies of network change, personality, and leadership practices will enable us to better disentangle the important ideas that have been raised in this exploratory study.

Leader Self-Efficacy and Relationship With Network Position Our work is unique in that it builds on some of the earliest research that examines the relationship between self-efficacy and network centrality (Penuel et al., 2009). Our results indicate that leaders who were confident in leading reforms, motivating action, and engaging staff were more likely to be sought for advice (high indegree) concerning reforms. However, these leaders were limited by a low sense of self-efficacy in managing technical and procedural tasks such as time management, scheduling, and paperwork. We suspect that these leaders do not have to be competent with manage- ment-oriented tasks, which can be delegated to others, but they need to have confidence in their ability to successfully lead change through instruc- tional leadership. As for those leaders with more efficient advice seeking (high outdegree and outcloseness), they tended to report low self-efficacy in managing technical and procedural practices. Perhaps this is not surpris- ing given that those who have confidence in their ability to manage

at TEMPLE UNIV on November 7, 2016eaq.sagepub.comDownloaded from

264 Educational Administration Quarterly 50(2)

technical and procedural work may not feel the need to efficiently reach out to others for advice.

Our results also suggest that leader self-efficacy plays an influential role for leaders who hold central network positions (i.e., more incoming advice ties) and act as connectors. Our results support previous studies on leaders’ sense of self-efficacy in that these efficacious leaders tend to facilitate col- lective learning environments (McCullers, 2009; Smith et al., 2003), and a sense of support for reforms (Tschannen-Moran & Gareis, 2005). Building on these previous studies, our results offer a nuanced understanding of dis- tinct types of leader self-efficacy in support of instructional leadership prac- tices and relationship between those beliefs and the social structure of an advice network.

Limitations and Implications Our findings are exploratory, promising, and cautionary. Although two of the four models only explained one third of the variance in leaders’ network position, the other models generated factors that explained a significant amount of the variance in network position. Despite the different ranges of variance explained, we are cautious in drawing definitive conclusions about the relationship between leaders’ network positions, personality traits, and self-efficacy, as these results are the first of their kind in the educational leadership literature and we have limited base of comparison. While this study provided a unique approach to analyzing the important role of leaders using a social network analysis, it also had limitations in design and as such provides useful directions for future studies. We did not collect qualitative data (i.e., interviews, observations, and relevant document review) to trian- gulate quantitative results, which would have been useful in further investi- gating how leaders in central positions perceived their ability and personalities as well as the specific “quality” of advice that was exchanged. Additional work will benefit from such mixed methods in further unpacking this complex phenomenon. Future studies may also collect data from a sam- ple of leaders that represent a variety of different network positions and personality types and then conduct a comparative study between those cen- tral and peripheral leaders to understand individual perceptions as well as the quality of advice exchanges.

In addition, in this study we collected data from leaders at three school districts identified as underperforming. Future studies may increase the sam- ple size for a more elaborate multilevel analysis to better control the vari- ance across contexts as well as take into account contextual factors more directly from a variety of performance levels. Moreover, although we have

at TEMPLE UNIV on November 7, 2016eaq.sagepub.comDownloaded from

Daly et al. 265

examined the point in time effect of demographic characteristics, years of experiences, personalities, and leader self-efficacy, additional variables may be included within a larger sample such as the proximity of leaders’ work site to others, subgroup effect (e.g., specific leadership team), collective efficacy (Goddard et al., 2000; Goddard, Hoy, & Hoy, 2004; Leithwood & Jantzi, 2008; Tschannen-Moran & Barr, 2004), the effect of intermediary organizations and/or external agencies, and other organizational-level vari- ables such as district size, improvement status, level of poverty, urbanicity, and so forth (Louis et al., 2010). Moreover, examining these variables over time may provide insight into the directionality of these variables as well as trajectories of change. Furthermore, examining these same constructs at a school level may also provide additional opportunities to build on previous studies and build theory. Finally, while we examined advice-seeking and -receiving relationships, future studies may investigate other network rela- tionships, such as functional expertise (e.g., Aronson, Reilly, & Lynn, 2006; Bunderson, 2003), innovation, and friendship as a way to test out additional relational hypothesis.

Our study attempts to shed new light on leadership characteristics and aims to connect three important veins of research across disciplines—social networks, personality, and leader self-efficacy—supported by empirical evi- dence. Particular personality traits (e.g., extraversion, conscientiousness, and neuroticism) and leader self-efficacy emerged as important factors in deter- mining key network positions of a leader in their advice-exchanging behav- iors. With increasing numbers of underperforming districts enacting reforms, it is crucial to have a better understanding of the leaders who occupy central advice positions in enacting change as well as a better understanding of the influence on personality and social processes of the broader context in which reform takes place.

In terms of application to the field, it may be useful to align leadership tasks and/or activities with personality and self-efficacy factors, which can improve communication and exchange of knowledge and information to individuals within a network. With knowledge of the leaders’ personality traits and a pro- file of self-efficacy, a school system may be able to develop better targeted coaching and mentorship for leaders who take on lead roles in reform. While we suggest a potential use of the leadership profile that contains leaders’ per- sonality traits and self-efficacy information, we also call for greater attention to the influence of those leaders who play important roles in an informal social network through which social resources may flow. By focusing on these influ- ential network leaders, school systems may better be able to steer the direction of reform, as these key leaders possess greater opportunities to disseminate and receive knowledge and information related to efforts at change. A better

at TEMPLE UNIV on November 7, 2016eaq.sagepub.comDownloaded from

266 Educational Administration Quarterly 50(2)

understanding and leveraging of those key leaders may facilitate the reform process and better spread knowledge and expertise.

Finally, our findings raised an intriguing question around those neurotic leaders who occupied central positions in the advice network. Most research on leadership and management relates to the negative effect of neurotic personal- ity on leaders’ behaviors and consequences of their actions on a system. Additionally, rarely do those studies take place in “high-stakes” contexts as may be argued is the case in our sample. Our study indicates that neurotic lead- ers were often central in the advice network and if we expand on the research in other fields as reported above, then one may be concerned that the influence that these leaders possess may not be ultimately supportive of successful reform effort. This proposition goes beyond our data, but raises important questions about the influence of the larger policy environment and organizational context in shaping relationships. The interplay between social interactions, individual personality traits, leadership beliefs about the ability to effect change, and the larger context in which reform takes place provides a rich and novel line of inquiry we hope other scholars and practitioners will continue to explore.

Declaration of Conflicting Interests The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding The authors received no financial support for the research, authorship, and/or publica- tion of this article.

Notes 1. See Borgatti and Foster (2003), Borgatti and Cross (2003), Borgatti and Everett

(2006), Borgatti, Mehra, Brass, and Labianca (2009), Brass (2009), Freeman (1979), Hanneman and Riddle (2005), Kilduff and Brass (2010), Kilduff and Krackhardt (2008), and Wasserman and Faust (1994).

2. Geodesic distance refers to the length of the shortest path between two actors in a social network.

3. Geodesic closeness centrality of a vertex n1 (actor) in a network graph, from a social network perspective, refers to the inverse of the average shortest path dis- tance from the vertex n1 to all other vertices reachable from n1. It can be regarded as an index of efficiency of each actor in spreading resources and/or information to other actors (Okamoto et al., 2008) as well as an index of the expected time an actor needs to flow something in or out from other actors across the network (Borgatti, 1995, 2005).

at TEMPLE UNIV on November 7, 2016eaq.sagepub.comDownloaded from

Daly et al. 267

4. In addition, a shorter version of personality measures (i.e., Big Five Inventory, BFI) has been demonstrated to be a useful and valid instrument (Burisch, 1997). The BFI is a relatively shorter self-report instrument of FFM. Reliability coef- ficients for the BFI scores typically range from .75 to .90, and over a 3-month test-retest correlations range between .80 and .90 (John & Srivastava, 1999). See Costa and McCrae (1992), Goldberg (1992), John and Srivastava (1999), and a more recent study done by Rammstedt and John (2007) for further information regarding the validation of the BFI measure.

5. For example, extraverted individuals are often more successful in sales and management positions and tend to have many relationships (Barrick, Mount, & Gupta, 2003; Klein et al., 2004); conscientious individuals tend to be success- ful in conventional jobs (Barrick et al., 2003) and are often more reliable at work (Barrick & Mount, 1991); agreeable individuals tend to be sought more for friendships and have fewer adversarial relationships (Klein et al., 2004); open individuals tend to have few friendship ties and more novel connections (Klein et al., 2004); and neurotic individuals tend to have fewer ties related to advice- seeking and friendship-building.

6. More detailed demographics information follows: position (1 = being in the dis- trict; 2 = school site), degree (1 = bachelor, 2 = master’s, and 3 = doctoral), and years of working experience (1 = less than 4 years; 2 = 5 to 9 years; 3 = 10 to 14 years; 4 = 15 to 19 years; 5 = 20 to 25 years; 6 = 25 years or more).

7. Statistical analysis of social networks using p* model handles the dependence between dyads (i.e., pairs of actors) in a directed and binary network and takes into account actor- and network-level attributes in the model (Pattison & Wasserman, 1999; Zijlstra, van Duijn, & Snijders, 2006). Alternatively, conven- tional regression analysis can be used to examine social measures separately in order to avoid issues of multicollinearity. It is important to note that these methodological approaches not only differ in technique but also in the interpreta- tion of the results—the former predicts the probability of observing specific ties between actors, while the latter examines the relationships regarding the number of interactions between actors.

8. Multicollinearity test for this model indicates that there is no variable in the model that is measuring the same relationship as is measured by another variable(s). The variance inflation factors (VIFs) range from 1.07 to 2.90.

9. It should be noted that for all of our models, we tested for an interaction effect between a leader’s personality traits and sense of efficacy as suggested in Gino et al.’s (2012) study. However, our data indicated no such effect as each of the inde- pendent variables contributed independently to the prediction of network positions.

10. Multicollinearity test for this model indicates that there is no variable in the model that is measuring the same relationship as is measured by another variable(s). The variance inflation factors (VIFs) range from 1.09 to 2.89.

11. Multicollinearity test for this model indicates that there is no variable in the model that is measuring the same relationship as is measured by another variable(s). The variance inflation factors (VIFs) range from 1.07 to 2.93.

at TEMPLE UNIV on November 7, 2016eaq.sagepub.comDownloaded from

268 Educational Administration Quarterly 50(2)

12. Multicollinearity test for this model indicates that there is no variable in the model that is measuring the same relationship as is measured by another variable(s). The variance inflation factors (VIFs) range from 1.09 to 2.87.

References Adler, P. S., & Kwon, S. W. (2002). Social capital: Prospects for a new concept.

Academy of Management Review, 27(1), 17-40. Agullard, K., & Goughnour, D. S. (2006). Central office inquiry: Assessing organiza-

tion, roles, and actions to support school improvement. San Francisco: WestEd. Amodeo, A. J., & Taylor, A. (2004). Virtual supervision model tips the scales in favor

of instructional leadership. THE Journal, 31(7), 22. Arnulf, J. K. (2012). Organizational change capacity and composition of management

teams: A visualization of how personality traits may restrain team adaptability. Team Performance Management, 18(7/8), 433-454.

Aronson, Z. H., Reilly, R., & Lynn, G. (2006). The impact of leader personality on new product development teamwork and performance: The moderating role of uncertainty. Journal of Engineering and Technology Management, 23, 221-247.

Asendorpf, J. B., & Wilpers, S. (1998). Personality effects on social relationships. Journal of Personality and Social Psychology, 74, 1531-1544.

Balkundi, P., & Kilduff, M. (2005). The ties that lead: A social network approach to leadership. The Leadership Quarterly, 16, 941-961.

Balkundi, P., & Harrison, D. (2006). Ties, leaders, and time in teams: Strong infer- ences about network structures effects on team viability and performance. Academy of Management Journal, 49(1), 49-68.

Bandura, A. (1977). Self efficacy: Toward a unifying theory of behavioural change. Psychological Review, 84(2), 191-215.

Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory. Englewood Cliffs, NJ: Prentice-Hall.

Bandura, A. (1993). Perceived self efficacy in cognitive development and function- ing. Educational Psychologist, 28(2), 117-148.

Bandura, A. (1995). Exercise of personal and collective efficacy in changing societies. In A. Bandura (Ed.), Self-efficacy in changing societies (pp. 1-45). Cambridge, UK: Cambridge University Press.

Bandura, A. (1997). Self-efficacy: The exercise of control. New York: W. H. Freeman and Company.

Bandura, A. (2000). Cultivate self-efficacy for personal and organizational effective- ness. In E. A. Locke (Ed.), Handbook of principles of organization behavior (pp. 120-136). Oxford, UK: Blackwell.

Barnett, D. (2004). School leadership preparation programs: Are they preparing tomorrow’s leaders? Education, 125(1), 121-129.

Barrick, M. R., & Mount, M. K. (1991). The big five personality dimensions and job performance: A meta-analysis. Personnel Psychology, 44, 1-26.

at TEMPLE UNIV on November 7, 2016eaq.sagepub.comDownloaded from

Daly et al. 269

Barrick, M. R., Mount, M. K., & Gupta, R. (2003). Meta-analysis of the relationship between the Five-Factor Model of personality and Holland’s occupational types. Personnel Psychology, 56, 45-74.

Barrick, M. R., Mount, M. K., & Judge, T. A. (2001). Personality and performance at the beginning of the new millennium: What do we know and where do we go next? International Journal of Selection & Assessment, 9, 9-30.

Barrick, M. R., Parks, L., & Mount, M. K. (2005). Self-monitoring as a modera- tor of the relationship between personality traits and performance. Personnel Psychology, 58(3), 745-767.

Bass, B. M. (1985). Leadership and performance beyond expectations. New York: Free Press.

Bass, B. M. (1998). Transformational leadership: Industrial, military, and educa- tional impact. Mahwah, NJ: Erlbaum.

Bendersky, C., & Shah, N. P. (2012). The downfall of extraverts and rise of neurotics: The dynamic process of status allocation in task groups. Academy of Management Journal, 56(2), 387-406.

Benet-Martinez, V., & John, O. P. (1998). Los Cinco Grandes across cultures and ethnic groups: Multitrait multimethod analyses of the Big Five in Spanish and English. Journal of Personality and Social Psychology, 75, 729-750.

Bono, J. E., & Judge, T. A. (2003). Core self-evaluations: A review of the trait and its role in job satisfaction and job performance. European Journal of Personality, 17, S5-S18.

Bono, J. E., & Judge, T. A. (2004). Personality and transformational and transactional leadership: A meta-analysis. Journal of Applied Psychology, 89, 901-910.

Borgatti, S. P. (1995). Centrality and AIDS. Connections, 18(1), 112-114. Borgatti, S. P. (2002). Netdraw network visualization. Harvard, MA: Analytic Technologies. Borgatti, S. P. (2005). Centrality and network flow. Social Networks, 27, 55-71. Borgatti, S. P. (2006). Identifying sets of key players in a network. Computational,

Mathematical and Organizational Theory, 12(1), 21-34. Borgatti, S., & Cross, R. (2003). A relational view of information seeking and learn-

ing in social networks. Management Science, 49(4), 432-445. Borgatti, S. P., & Everett, M. G. (1992). Notions of position in social network analy-

sis. Sociological Methodology, 22, 1-35. Borgatti, S. P., & Everett, M. (1999). Models of core periphery structures. Social

Networks, 21, 375-395. Borgatti, S. P., & Everett, M. G. (2006). A graph-theoretic framework for classifying

centrality measures. Social Networks, 28(4), 466-484. Borgatti, S. P., Everett, M. G., & Freeman, L. C. (2002). Ucinet for Windows: Software

for social network analysis. Harvard, MA: Analytic Technologies. Borgatti, S., & Foster, P. (2003). The network paradigm in organizational research: A

review and typology. Journal of Management, 29(6), 991-1013. Borgatti, S. P., Jones, C., & Everett, M. G. (1998). Network measures of social capi-

tal. Connections, 21(2), 27-36. Borgatti, S. P., Mehra, A., Brass, D. J., & Labianca, G. (2009). Network analysis in

the social sciences. Science, 323, 892-895.

at TEMPLE UNIV on November 7, 2016eaq.sagepub.comDownloaded from

270 Educational Administration Quarterly 50(2)

Bourgeois, M., & Friedkin, N. E. (2001). The distant core: Social solidarity, social distance, and interpersonal ties. Social Networks, 23, 245-260.

Brass, D. J. (2009, April). Social networks: Clusters and bridges. Paper presented at the University of Kentucky, Gatton College Invest with the Best and Hall of Fame.

Brass, D. J., & Burkhardt, M. E. (1993). Potential power and power use: An investiga- tion of structure and behavior. Academy of Management Journal, 36, 441-470.

Bunderson, J. S. (2003). Recognizing and utilizing expertise in work groups: A status characteristics perspective. Administrative Science Quarterly, 48, 557-591.

Burisch, M. (1997). Test length and validity revisited. European Journal of Personality, 11, 303-315.

Burt, R. S. (1976). Positions in networks. Social Forces, 55(1), 93-122. Burt, R. S. (1980). Innovation as a structural interest: Rethinking the impact of net-

work position on innovation adoption. Social Networks, 2(1980), 327-355. Burt, R. S. (1982). Toward a structural theory of action. New York: Academic Press. Burt, R. S. (1983). Range. In R. S. Burt & M. J. Minor (Eds.), Applied network analy-

sis (pp. 176-194). Beverly Hills, CA: Sage Publications. Burt, R. S. (1992). Structural holes: The structure of competition. Cambridge, MA:

Harvard University Press. Burt, R. S. (1997). A note on social capital and network content. Social Networks,

19(4), 355-373. Burt, R. S. (2005). Brokerage and closure: An introduction to social capital. Oxford,

UK: Oxford University Press. Cable, D. M., & Judge, T. A. (2003). Managers’ upward influence tactic strategies:

The role of manager personality and supervisor leadership style. Journal of Organizational Behavior, 24, 197-214.

Carr, D. (2008, December). Being a change agent: What the research says [PowerPoint presentation]. Charleston, WV: Edvantia. Presentation slides from ARCC Rally.

Camagni, R. (1995). The concept of innovative milieu and its relevance for public policies in European lagging regions. Papers in Regional Science, 74(4), 317-340.

Caspi, A., Roberts, B. W., & Shiner, R. L. (2005). Personality development: Stability and change. Annual Review of Psychology, 56, 453-484.

Cervone, D. (2005). Personality architecture: Within-person structures and processes. Annual Review of Psychology, 56, 423-452.

Clifton, A., Turkheimer, E., & Oltmanns, T. F. (2012). Personality disorder in social networks: Network position as a marker of interpersonal dysfunction. Social Networks, 31(1), 26-32.

Coburn, C., & Russell, J. (2008). Getting the most out of professional learning communities and coaching: Promoting interactions that support instructional improvement. Learning Policy Brief, 1(3), 1-5.

Cohen, S. (2006). Change agents bolster new practices in the workplace. Nursing Management, 37(6), 16-17.

Cole, R. P., & Weinbaum, E. H. (2010).Changes in attitude: Peer influence in high school reform. In A. J. Daly (Ed.), Social network theory and educational change (pp. 77-96). Cambridge, MA: Harvard University Press.

at TEMPLE UNIV on November 7, 2016eaq.sagepub.comDownloaded from

Daly et al. 271

Costa, P. T., Jr., & McCrae, R. R. (1992). Revised NEO personality inventory (NEO- PI-R) and NEO five- factor inventory (NEO-FFI) professional manual. Odessa, FL: Psychological Assessment Resources.

Cross, R., Borgatti, S. P., & Parker, A. (2002). Making invisible work visible: Using social network analysis to support strategic collaboration. California Management Review, 44(2), 25-46.

Cross, R., & Parker, A. (2004). The hidden power of social networks: Understanding how work really gets done in organizations. Cambridge, MA: Harvard Business School Press.

Daly, A. J. (2009). Rigid response in an age of accountability. Educational Administration Quarterly, 45(2), 168-216.

Daly, A. J. (Ed.). (2010). Social network theory and educational change. Cambridge, MA: Harvard Education Press.

Daly, A. J. (2012). Data, dyads, and dynamics: Exploring data use and social net- works in educational improvement. Teachers College Record, 114(11), 1-38.

Daly, A. J., Der Martirosian, C., Ong-Dean, C., Park, V., & Wishard, A. (2011). Leading under sanction: Principals’ perceptions of threat-rigidity, efficacy, and leadership in underperforming schools. Leadership and Policy in Schools, 10(2), 171-206.

Daly, A. J., & Finnigan, K. (2010). A bridge between worlds: Understanding network structure to understand change strategy. Journal of Educational Change, 111, 111-138.

Daly, A. J., & Finnigan, K. (2011). The ebb and flow of social network ties between district leaders under high stakes accountability. American Educational Research Journal, 48, 39-79.

Daly, A. J., & Finnigan, K. (2012). Exploring the space between: Social networks, trust, and urban school district leaders. Journal of School Leadership, 22(3), 493- 530.

Degenne, A., & Forsé, M. (1999). Introducing social networks. London: Sage Publications.

DeYoung, C. G., Quilty, L. C., & Peterson, J. B. (2007). Between facets and domains: 10 Aspects of the Big Five. Journal of Personality and Social Psychology, 93, 880-896.

DeVita, M. C., Colvin, R. L., Darling-Hammond, L., & Haycock, K. (2007, October). A bridge to school reform. Paper presented at The Wallace Foundations National Conference, New York, New York.

Digman, J. M. (1990). Personality structure: Emergence of the five factor model. Annual Review of Psychology, 41, 417-440.

Elizur, D., & Guttman, L. (1976). The structure of attitudes toward work and techno- logical change within an organization. Administrative Science Quarterly, 21(1), 611-623.

Everett, M. G., & Borgatti, S. P. (1999). The centrality of groups and classes. Journal of Mathematical Sociology, 23(3), 181-201.

at TEMPLE UNIV on November 7, 2016eaq.sagepub.comDownloaded from

272 Educational Administration Quarterly 50(2)

Finnigan, K., & Daly, A. J. (2012). Mind the gap: Learning, trust, and relationships in an underperforming urban district. American Journal of Education, 119(1), 41-71.

Finnigan, K., & Gross, B. (2007). Do accountability policy sanctions influence teacher motivation? Lessons from Chicago’s low-performing schools. American Educational Research Journal, 44(3), 594-629.

Frank, K. A., Zhao, Y., Penuel, W. R., Ellefson, N., & Porter, S. (2011). Focus, fiddle, and friends: Experiences that transform knowledge for the implementation of innovations. Sociology of Education, 84(2), 137-156.

Freeman, L. C. (1979). Centrality in social networks: Conceptual clarification. Social Networks, 1, 215-239.

Friedkin, N. E. (1998). A structural theory of social influence. Cambridge, UK: Cambridge University Press.

Friedkin, N. E. (2001). Norm formation in social influence networks. Social Networks, 23, 167-189.

Fullan, M. (2002). The change leader. Educational Leadership, 59(8), 16-20. Fullan, M. (2007). The new meaning of educational change (4th ed.). New York:

Teachers College Press. Gareis, C. R., & Tschannen-Moran, M. (2005, November). Cultivating principals’

sense of efficacy: Supports that matter. Paper presented at the annual meeting of the University Council for Educational Administration, Nashville, TN.

Gino, F., Brooks, A. W., & Schweitzer, M. (2012). Anxiety, advice, and the ability to discern: Feeling anxious motivates individuals to seek and use advice. Journal of Personality and Social Psychology, 102(3), 497-512.

Goddard, R. D., Hoy, W. K., & Hoy, A. W. (2004). Collective efficacy beliefs: Theoretical developments, empirical evidence, and future directions. Educational Researcher, 33(3), 3-13.

Goddard, R. D., Hoy, W. K., & Woolfolk Hoy, A. (2000). Collective teacher efficacy: Its meaning, measure, and impact on student achievement. American Educational Research Journal, 37, 479-508.

Goldberg, L. R. (1992). The development of markers for the Big-Five factor structure. Psychological Assessment, 4, 26-42.

Gosling, S. D., Rentfrow, P. J., & Swann, W. B., Jr. (2003). A very brief measure of the Big Five Personality domains. Journal of Research in Personality, 37, 504-528.

Graziano, W. G., & Eisenberg, N. (1997). Agreeableness: A dimension of personal- ity. In R. Hogan, J. A. Johnson & S. R. Briggs (Eds.), Handbook of personality psychology (pp. 795-824). San Diego, CA: Academic Press.

Gulati, R. (1995). Does familiarity breed trust? The implications of repeated ties for contractual choices in alliances. Academy of Management Journal, 35, 85-112.

Hanneman, R. A., & Riddle, M. (2005). Introduction to social network methods. Riverside, CA: University of California, Riverside (published in digital form at http://faculty.ucr.edu/~hanneman/).

Helmsing, A. H. J. (2001). Externalities, learning and governance: New perspectives on local economic development. Development and Change, 32, 277-308.

at TEMPLE UNIV on November 7, 2016eaq.sagepub.comDownloaded from

Daly et al. 273

Hite, J., Williams, E., & Baugh, S. (2005). Multiple networks of public school admin- istrators: An analysis of network content and structure. International Journal on Leadership in Education, 8(2), 91-122.

Hogan, J., & Ones, D. S. (1997). Conscientiousness and integrity at work. In R. Hogan, J. A. Johnson & S. R. Briggs (Eds.), Handbook of personality psychology (pp. 849-870). San Diego, CA: Academic Press.

Honig, M. (2006). Street-level bureaucracy revisited: Frontline district central- office administrators as boundary spanners in education policy implementation. Educational Evaluation and Policy Analysis, 28(4), 357-383.

Honig, M. I., Copland, M. A., Rainey, L., Lorton, J. A., & Newton, M. (2010, April). School district central office transformation for teaching and learning improve- ment (a report to the Wallace Foundation). Seattle, WA: The Center for the Study of Teaching and Policy.

Honig, M. I., Lorton, J. S., & Copland, M. A. (2009). Urban district central office transformation for teaching and learning improvement: Beyond a zero-sum game. Yearbook of the National Society for the Study of Education, 108(1), 21-40.

Hooker, K. S., & McAdams, D. P. (2003). Personality reconsidered: A new agenda for aging research. Journal of Gerontology: Psychological Sciences, 58B(6), P296-P304.

Ibarra, H., & Andrews, S. B. (1993). Power, social influence, and sense mak- ing: Effects of network centrality and proximity on employee perceptions. Administrative Science Quarterly, 38, 277-303.

Imants, J. G., & De Brabander, C. J. (1996). Teachers’ and principals’ sense of effi- cacy in elementary schools. Teaching and Teacher Education, 12(2), 179-195.

John, O. P., Donahue, E. M., & Kentle, R. L. (1991). The big five inventory: Versions 4a and 54. Berkeley: University of California at Berkeley, Institute of Personality and Social Research.

John, O. P., Hampson, S. E., & Goldberg, L. R. (1991). Is there a basic level of per- sonality description? Journal of Personality and Social Psychology, 60, 348-361.

John, O. P., Naumann, L. P., & Soto, C. J. (2008). Paradigm shift to the integrative big-five trait taxonomy: History, measurement, and conceptual issues. In O. P. John, R. W. Robins & L. A. Pervin (Eds.), Handbook of personality: Theory and research (pp. 114-158). New York: Guilford.

John, O. P., & Srivastava, S. (1999). The big-five factor taxonomy: History, measure- ment, and theoretical perspectives. In L. A. Pervin & O. P. John (Eds.), Handbook of personality: Theory and research (2nd ed., pp. 102-138). New York: Guilford.

Judge, T. A., & Bono, J. E. (2000). Five-factor model of personality and transforma- tional leadership. Journal of Applied Psychology, 85, 751-765.

Judge, T. A., Bono, J. E., Ilies, R., & Gerhardt, M. W. (2002). Personality and lead- ership: A qualitative and quantitative review. Journal of Applied Psychology, 87(4), 765-780.

Judge, T. A., Thoresen, C. J., Pucik, V., & Welbourne, T. M. (1999). Managerial cop- ing with organizational change: A dispositional perspective. Journal of Applied Psychology, 84, 107-122.

at TEMPLE UNIV on November 7, 2016eaq.sagepub.comDownloaded from

274 Educational Administration Quarterly 50(2)

Kilduff, M., & Brass, D. J. (2010). Organizational social network research: Core ideas and key debates. In J. P. Walsh & A. P. Brief (Eds.), Academy of management annuals (pp. 317-357). New York: Routledge.

Kilduff, M., & Krackhardt, D. (2008). Interpersonal networks in organizations: Cognition, personality, dynamics, and culture: Structural analysis in the social sciences. Cambridge, UK: Cambridge University Press.

Kilduff, M., & Tsai, W. (2003). Social networks and organizations. London: Sage. Klein, K. J., Lim, B. C., Saltz, J. L., & Mayer, D. M. (2004). How do they get there?

An examination of the antecedents of network centrality in team networks. Academy of Management Journal, 47, 952-963.

Knapp, M. S., Copland, M. A., Honig, M. I., Plecki, M. L., & Portin, B. S. (2010). Learning-focused leadership and leadership support: Meaning and practice in urban systems (a report to the Wallace Foundation). Seattle, WA: The Center for the Study of Teaching and Policy.

Knapp, M. S., & Feldman, S. B. (2012). Managing the intersection of internal and external accountability: Challenge for urban school leadership in the United States. Journal of Educational Administration, 50(5), 666-694.

Knoke, D., & Yang, S. (2008). Network analysis (2nd ed.). Thousand Oaks, CA: Sage Publications.

Kowalski, T. J., Petersen, G. J., & Fusarelli, L. D. (2007). Effective communication for school administrators: A necessity in an information age. Lanham, MD: Rowman & Littlefield Education.

Krackhardt, D. (1999). The ties that torture: Simmelian tie analysis in organizations. In S. B. Bacharach, S. B. Andrews & D. Knoke (Eds.), Research in the sociology of organizations (pp. 183-210). Stamford, CT: JAI Press.

Kumar, K., Bakhshi, A., & Rani, E. (2009). Linking the “Big Five” personal- ity domains to organizational citizenship behavior. International Journal of Psychological Studies, 1(2), 71-81.

Lau, C. M., & Woodman, R. W. (1995). Understanding organizational change: A schematic perspective. The Academy of Management Journal, 38(2), 537-554.

Lawson, C. (1999). Towards a competence theory of the region. Cambridge Journal of Economics, 23, 151-166.

Leithwood, K., & Jantzi, D. (2008). Linking leadership to student learning: The contributions of leader efficacy. Educational Administration Quarterly, 44(4), 496-528.

Lin, N. (1999). Building a theory of social capital. Connections, 22(1), 28-51. Lin, N. (2009). Social capital: A theory of social structure and action (8th ed.). New

York: Cambridge University Press. Liou, Y., & Daly, A. J. (in press). Closer to learning: Social networks, trust, and pro-

fessional communities. Journal of School Leadership. Louis, K., Leithwood, K., Wahlstrom, K., & Anderson, S. (2010). Learning from

leadership: Investigating the links to improved student learning. New York: The Wallace Foundation.

at TEMPLE UNIV on November 7, 2016eaq.sagepub.comDownloaded from

Daly et al. 275

Louis, K. S., & Robinson, V. M. (2012). External mandates and instructional leader- ship: School leaders as mediating agents. Journal of Educational Administration, 50(5), 629-665.

Lunenburg, F., & Ornstein, A. (2008). Educational administration: Concepts and practices (5th ed.). Belmont, CA: Wadsworth/Thompson Learning.

Maskell, P., & Malmberg, A. (1999). Localized learning and industrial competitive- ness. Cambridge Journal of Economics, 23, 167-185.

McAdams, D. P. (1995). What do we know when we know a person? Journal of Personality, 63, 365-396.

McCormick, M. J. (2001). Self-efficacy and leadership effectiveness: Applying social cognitive theory to leadership. Journal of Leadership Studies, 8(1), 22-33.

McCrae, R. R., & Costa, P. T. (1990). Personality in adulthood. New York: Guilford. McCrae, R. R., & John, O. (1992). An introduction to the five-factor model and its

applications. Journal of Personality, 60(2), 174-214. McCullers, J. F. (2009). Self-efficacy beliefs of Florida school principals regarding

federal and state accountability measures. Unpublished doctoral dissertation, University of Central Florida, Orlando, Florida.

McCullers, J. F., & Bozeman, W. (2010). Principal self-efficacy: The effects of No Child Left Behind and Florida school grades. NASSP Bulletin, 94(1), 53-74.

McFarlane, A. C. (1989). The aetiology of post-traumatic morbidity: Predisposing, pre- cipitating and perpetuating factors. British Journal of Psychiatry, 154, 221-228.

Mok, D., Wellman, B., & Basu, R. (2007). Did distance matter: A pre-internet analy- sis. Social Networks, 29(3), 430-462.

Moolenaar, N. M. (2010). Ties with potential: Nature, antecedents, and consequences of social networks in school teams. Unpublished doctoral dissertation.

Moolenaar, N., Daly, A. J., & Sleegers, P. (2010). Occupying the principal position: Innovation in schools. Educational Administration Quarterly, 46, 623-670.

Moolenaar, N., Daly, A. J., & Sleegers, P. (2011). Ties with potential: Social network structure and innovation in Dutch schools. Teachers College Record, 113(9), 1983-2017.

Northouse, P. G. (1997). Leadership: Theory and practice. Thousand Oaks, CA: Sage. Okamoto, K., Chen, W., & Li, X. Y. (2008). Ranking of closeness centrality for large-

scale social networks. Lecture Notes in Computer Science, 5059, 186-195. Oreg, S. (2006). Personality, context and resistance to organizational change.

European Journal of Work and Organizational Psychology, 15(1), 73-101. Paglis, L. L., & Green, S. G. (2002). Leadership self-efficacy and managers’ motiva-

tion for leading change. Journal of Organizational Behavior, 23, 215-235. Pajares, F. (1997). Current directions in self-efficacy research. In M. L. Maehr &

P. R. Pintrich (Eds.), Advances in motivation and achievement (Vol. 10, pp. 1-49). Greenwich, CT: JAI Press.

Pajares, F. (1996). Self-efficacy beliefs in academic settings. Review of Educational Research, 66, 543-578.

Pattison, P. E., & Wasserman, S. (1999). Logit models and logistic regressions for social networks, II. Multivariate relations. British Journal of Mathematical and Statistical Psychology, 52, 169-194.

at TEMPLE UNIV on November 7, 2016eaq.sagepub.comDownloaded from

276 Educational Administration Quarterly 50(2)

Penuel, W. R., & Riel, M. (2007). The new science of networks and the challenge of school change. Phi Delta Kappan, 88(8), 611-615.

Penuel, W. R., Riel, M., Joshi, A., Pearlman, L., Kim, C. M., & Frank, K. A. (2010). The alignment of the informal and formal organizational supports for reform: Implications for improving teaching in schools. Educational Administration Quarterly, 46(1), 57-95.

Penuel, W. R., Riel, M. R., Krause, A., & Frank, K. A. (2009). Analyzing teachers’ professional interactions in a school as social capital: A social network approach. Teachers College Record, 111(1), 124-163.

Pitts, V., & Spillane, J. P. (2009). Using social network methods to study school leadership. International Journal of Research and Method in Education, 32(2), 185-207.

Pitre, L. (2003). The Mississippi Student Achievement Improvement Act and the No Child Left Behind Act: Leadership and motivation. Retrieved from Dissertations & Theses database.

Plecki, M. L., Knapp, M. S., Castaneda, T., Halverson, T., LaSota, R., & Lochmiller, C. (2009). How leaders invest staffing resources for learning improvement. Seattle: Center for the Study of Teaching & Policy, University of Washington.

Powell, W. W., Koput, K., IV., & Smith-Doerr, L. (1996). Interorganizational col- laboration and the locus of innovation: Networks of learning in biotechnology. Administrative Science Quarterly, 41, 116-145.

Raider, H., & Krackhardt, D. J. (2002). Intraorganizational networks. In J. A. C. Baum (Ed.), The Blackwell companion to organizations (pp. 58-74). Malden, MA: Wiley-Blackwell.

Rammstedt, B., & John, O. P. (2005). Short version of the Big Five Inventory (BFI- K): Development and validation of an economic inventory for assessment of the five factors of personality. Diagnostica, 51, 195-206.

Rammstedt, B., & John, O. P. (2007). Measuring personality in one minute or less: A 10-item short version of the Big Five Inventory in English and German. Journal of Research in Personality, 41, 203-212.

Salgado, J. F. (2003). Predicting job performance with FFM-based and non-FFM based personality inventories. Journal of Occupational and Organizational Psychology, 76, 323-346.

Scott, J. (2000). Social network analysis (2nd ed.). London: Sage Publications. Smith, W., Guarino, A. J., Strom, P., & Reed, C. (2003). Principal self-efficacy

and effective teaching and learning environments. School Leadership and Management, 23(4), 505-508.

Strang, S., & Kuhnert, K. W. (2009). Personality and leadership development levels as predictors of leader performance. The Leadership Quarterly, 20, 421-433.

Thoms, P., Moore, K. S., & Scott, K. S. (1996). The relationship between self-effi- cacy for participating in self-managed work groups and the big five personality dimensions. Journal of Organizational Behavior, 17, 349-362.

Tsai, W. (2001). Knowledge transfer in intraorganizational networks: Effects of net- work position and absorptive capacity on business unit innovation and perfor- mance. Academy of Management Journal, 44(5), 996-1004.

at TEMPLE UNIV on November 7, 2016eaq.sagepub.comDownloaded from

Daly et al. 277

Tschannen-Moran, M., & Barr, M. (2004). Fostering student achievement: The rela- tionship between collective teacher efficacy and student achievement. Leadership and Policy in Schools, 3, 187-207.

Tschannen-Moran, M., & Gareis, C. (2004). Principals’ sense of efficacy: Assessing a promising construct. Journal of Educational Administration, 42, 573-585.

Tschannen-Moran, M., & Gareis, C. R. (2005, November). Cultivating principals’ sense of efficacy: Supports that matter. Paper presented at the annual meeting of the University Council for Educational Administration, Nashville, TN.

Tschannen-Moran, M., Woolfolk Hoy, A., & Hoy, W. K. (1998). Teacher efficacy: Its meaning and measure. Review of Educational Research, 68, 202-248.

Uzzi, B. (1996). The sources and consequences of embeddedness for the economic performance of organizations: The network effect. American Sociological Review, 61(4), 674-698.

Vakola, M., Tsaousis, I., & Nikolaou, I. (2003). The role of emotional intelligence and personality variables on attitudes toward organizational change. Journal of Managerial Psychology, 19(2), 88-110.

Vander Ark, T. (2002). Toward success at scale. Phi Delta Kappan, 84(4), 322. Wasserman, S., & Faust, K. (1994). Social network analysis: Methods and applica-

tions. Cambridge, UK: Cambridge University Press. Watson, D., & Clark, L. A. (1997). Extraversion and its positive emotional core. In R.

Hogan, J. A. Johnson & S. R. Briggs (Eds.), Handbook of personality psychology (pp. 767-793). San Diego, CA: Academic Press.

Wehrli, S. (2008). Personality on social network sites: An application of the five fac- tor model. ETH Zurich Sociology, Working Paper No. 7.

Wellman, B., & Berkowitz, S. D. (1998). Social structures: A network approach. Cambridge, MA: Cambridge University Press.

Witt, L. A., Burke, L. A., Barrick, M. R., & Mount, M. K. (2002). The interactive effects of conscientiousness and agreeableness on job performance. Journal of Applied Psychology, 87, 164-169.

Yin, R. (1984). Case study research: Design and methods (1st ed.). Beverly Hills, CA: Sage Publishing

Zaccaro, S. J. (2007). Trait-based perspectives of leadership. American Psychologist, 62(1), 6-16.

Zhang, W., Wang, Z., Zhao, D., & Tang, X. (2012). Graph degree linkage: Agglomerative clustering on a directed graph. Computer Vision and Pattern Recognition, arXiv:1208.5092v1.

Zijlstra, B., van Duijn, M., & Snijders, T. (2006). The multilevel p2 model: A random effects model for the analysis of multiple social networks. Methodology, 2(1), 42-47.

Author Biographies Alan J. Daly is an associate professor and Vice Chair of the Department of Education Studies at the University of California, San Diego. His research and teaching focus on leadership, educational policy, district reform, and social network theory. Recent

at TEMPLE UNIV on November 7, 2016eaq.sagepub.comDownloaded from

278 Educational Administration Quarterly 50(2)

studies have focused on research evidence and the supports and constraints of social networks at multiple levels of the educational system ranging from the elementary schools to higher education. In addition, he has published an edited volume on social networks titled Social Network Theory and Educational Change with Harvard Press.

Yi-Hwa Liou is a postdoctoral researcher at the University of California, San Diego. Her research focuses on the ways in which practitioners access, receive, exchange, and advance contextually situated practices in enacting reform efforts from the lens of social/human/intellectual capital theory and the use of social network theory and anal- ysis. Liou aims to bring the research methods and practices of social network analysis to the world of leadership development, school reform, and policy reflection to under- stand several complex areas of organization-level research such as leadership for organizational development, teacher leadership, professional learning community, data-driven decision making, and professional development.

Natalie A. Tran is an assistant professor at California State University, Fullerton. Her research focuses on evaluating the effectiveness of curriculum and services related to science, technology, engineering, and mathematics (STEM) education and examining factors that affect students’ learning experiences both in the classroom and out-of- school settings. Her other research interests include integrating the practice of mind- fulness in K-12 classrooms and exploring issues related to diversity in higher educa- tion. These areas align with her methodological research interests that include hierarchical linear modeling, experimental design, quasi-experimental design, and survey studies.

Frank Cornelissen is a postdoctoral research fellow at the University of California, San Diego, and the University of Cambridge. He worked in the field of education as a teacher, consultant, lecturer, and senior policy advisor. After completing his dis- sertation at the Eindhoven University of Technology (the Netherlands) about the developing, sharing, and using of research-based knowledge in school-university research networks he received a Marie Curie Fellowship grant from the European Union to continue his study on this topic in the United States and United Kingdom. His research work and interests include social network analysis, collaborative knowl- edge processes, educational reform, teacher research, and teacher professionalism.

Vicki Park is an assistant professor in the Department of Educational Leadership at San Jose State University’s Connie L. Lurie College of Education. She is also the Co-Director of Research for the Pathways to Postsecondary Education Success: Maximizing Opportunities for Youth in Poverty project. Her research focuses on urban school reform and leadership, policy implementation, and the ways in which class, race, and gender shape educational opportunities for diverse students.

at TEMPLE UNIV on November 7, 2016eaq.sagepub.comDownloaded from