AN EMPIRICAL INVESTIGATION OF AN EMERGING YOUTH-DRIVEN MODEL OF LEADERSHIP: THE COLLECTIVE CHANGE YOUTH LEADERSHIP FRAMEWORK  By  Jennifer Mortensen               A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements  for the degree of  Psychology - Doctor of Philosophy  2016      ABSTRACT AN EMPIRICAL INVESTIGATION OF AN EMERGING YOUTH-DRIVEN MODEL OF LEADERSHIP: THE COLLECTIVE CHANGE YOUTH LEADERSHIP FRAMEWORK  By  Jennifer Mortensen  The primary purpose of this study was to empirically examine an emerging, youth-defined framework of leadership, the Collective Change Youth Leadership Framework, with a national sample of youth.  This framework was developed from two initial studies with youth, one that explored ideas about leadership through Photovoice (Mortensen, et al., 2014) and a second that used Concept Mapping to understand how youth define leadership. The proposed anyone can be a leader, creating change, collective action, modeling and mentoring, strong character, and motivation and ambition. By testing this framework on a national sample, this study aimed to deepen current understandings of what leadership means to youth, and add to the literature a youth-defined framework of leadership. The second purpose of this study was to researchers have linked to youth civic engagement including sense of empowerment, critical consciousness, adult support, and background characteristics such as and age were explored. These findings can inform future leadership development efforts so more youth will engage in these opportunities and take on leadership roles within their communities both now, and as adults.   iii ACKNOWLEDGEMENTS   First and foremost, I am eternally grateful to my husband Tim who has supported me and encouraged me throughout this entire journey. Whenever I had doubts about myself or my work, I could count on you to push me to keep going. I cannot imagine life without you. To my family, Dad, Mom, Kim, Geoff, and Sam. Thank you for always believing in me and being the best cheerleaders anyone could ask for. To my advisor, Dr. Pennie Foster-Fishman, I have learned so much from you and am sincerely grateful for the years of guidance and your commitment to my development and success. To my committee, Dr. Hiram Fitzgerald, Dr. Kevin Ford, and Dr. Francisco Villarruel, thank you for your thoughtful feedback and engaging conversations throughout this project. Finally, thank you to the youth who shared their ideas about leadership with our team. Without them, this project would not be possible.     iv TABLE OF CONTENTS   LIST OF TABLES ......................................................................................................................... vi  LIST OF FIGURES ..................................................................................................................... viii  OVERVIEW .......................................................................................................................................9 LITERATURE REVIEW ....................................................................................................................12 DEFINITION OF YOUTH ...............................................................................................................12 CURRENT STATE OF DEFINING YOUTH LEADERSHIP ...................................................................14 FEMINIST STANDPOINT THEORY. ............................................................................................20 WHAT WE KNOW ABOUT YOUTHS DEFINITIONS OF LEADERSHIP ..............................................21 BENEFITS OF YOUTH LEADERSHIP ..............................................................................................26 BENEFITS TO YOUTH IN TAKING ON LEADERSHIP ROLES. .......................................................26 BENEFITS OF PROGRAMS THAT FOSTER YOUTH LEADERSHIP SKILLS. ....................................29 BENEFITS TO SCHOOLS AND THE COMMUNITY. .......................................................................32 SUMMARY. ...............................................................................................................................38 STRENGTHS AND LIMITATIONS OF YOUTH LEADERSHIP RESEARCH. .......................................39 COMMUNITY SHORTCOMINGS IN PROMOTING YOUTH LEADERSHIP ...........................................49 COMMUNITIES NEED MORE YOUTH LEADERS. ..........................................................................49 CRITIQUES OF YOUTH LEADERSHIP PROGRAMS. ......................................................................53 SUPPORTING YOUTH AS LEADERS. ..........................................................................................57 EMPIRICAL INVESTIGATION OF THE COLLECTIVE CHANGE YOUTH LEADERSHIP FRAMEWORK ...59 COLLECTIVE CHANGE YOUTH LEADERSHIP FRAMEWORK. .........................................................60 ADULT THEORIES OF LEADERSHIP ..........................................................................................68 VARIABLES CONTRIBUTING TO YOUTHS DEFINITION OF LEADERSHIP ...................................75 STUDY RATIONALE ..................................................................................................................87 RESEARCH QUESTIONS ............................................................................................................88 RESEARCH DESIGN AND METHODS ...............................................................................................89 OVERVIEW ..................................................................................................................................89 SETTING ......................................................................................................................................90 PARTICIPANTS .............................................................................................................................91 PROCEDURES ..............................................................................................................................96 RECRUITMENT .........................................................................................................................96 DATA COLLECTION .................................................................................................................97 DATA STORAGE .......................................................................................................................97 DATA ENTRY ...........................................................................................................................97 MEASURES  YOUTH SURVEY .....................................................................................................98 MEASURES  ADULT ADVISOR SURVEY ....................................................................................107 DATA ANALYTIC APPROACH ....................................................................................................111 RESULTS .......................................................................................................................................113 RESEARCH QUESTION 1 ............................................................................................................115 RESEARCH QUESTION 2 ............................................................................................................125 SCALE DESCRIPTIVES AND CORRELATIONS. ..........................................................................125 v MULTIPLE REGRESSION. .......................................................................................................127 DISCUSSION .................................................................................................................................140 CONCLUSION. ............................................................................................................................157  APPENDICES ............................................................................................................................159 APPENDIX A ..........................................................................................................................160 APPENDIX B ..........................................................................................................................174 APPENDIX C ..........................................................................................................................185  REFERENCES ............................................................................................................................192    vi LIST OF TABLES   Table 1. Example Definitions and Conceptualizations of Youth Leadership ................................16  Table 2. Selected Adult Theories* and the Collective Change Youth Leadership Framework ....72  Table 3. Youth Demographic Data ................................................................................................92  Table 4. Youth Leadership Experience ..........................................................................................94  Table 5. Collective Change Youth Leadership Framework Dimensions and Survey Items .........99  Table 6. Coalition Characteristics (n=41) ....................................................................................109  Table 7. Advisor-Reported Youth Participation on Coalitions ....................................................111  Table 8.  .........................115  Table 9. Item Parcel Descriptive Statistics ..................................................................................118  Table 10. Conceptually Parceled First Order CFA Factor Loadings ...........................................121  Table 11. Standardized Coefficients for Second Order Confirmatory Factor Analysis ..............124  Table 12. Experience/Attitude and Leadership Scale Descriptive Statistics ...............................126  Table 13. Collective Change Youth Leadership Subscale and Predictor Scale Correlations ......127  Table 14. Unstandardized and Standardized Coefficients for Creating Change Multiple Regression Model ........................................................................................................................129  Table 15. Unstandardized and Standardized Coefficients for Collective Action Multiple Regression Model ........................................................................................................................130  Table 16. Unstandardized and Standardized Coefficients for Modeling and Mentoring Multiple Regression Model to Explore Suppressor Variable .....................................................................134  Table 17. Unstandardized and Standardized Coefficients for Strong Character Multiple Regression Model to Explore Suppressor Variable .....................................................................136  Table 18. Unstandardized and Standardized Coefficients for Motivated and Ambitious Multiple Regression Model ........................................................................................................................137  Table 19. Unstandardized and Standardized Coefficients for Leadership Definition Multiple Regression Model ........................................................................................................................139 vii  Table C1. Anyone can be a Leader Item Correlations .................................................................185  Table C2. Creating Change Item Correlations .............................................................................186  Table C3. Collective Action Item Correlations ...........................................................................187  Table C4. Modeling and Mentoring Item Correlations ...............................................................188  Table C5. Strong Character Item Correlations ............................................................................189  Table C6. Motivated and Ambitious Item Correlations...............................................................190  Table C7. Youth Leadership Subscale Correlations ....................................................................191      viii LIST OF FIGURES   Figure 1. Conceptually Organized First Order CFA Model  .......................................................122 Figure 2. Second Order CFA Model for Collective Change Youth Leadership Framework ......124      9 OVERVIEW  Youth leaders are an essential part of communities. Youth are often more willing to discuss sensitive topics, such as equity issues, than adults (Mitra, 2006), frequently know what is going on in the community as many as three years ahead of adults (Stoneman, 2002), and are more creative and willing to take risks to solve problems and create change (Stoneman, 2002; London, et al., 2003; Libby, et al., 2005). In other words, youth leadership is a critical part of community development, and youth are eager to engage as change agents within their communities (Frank, 2006; Corsi, 2002; Sutton & Kemp, 2002).   Yet, while  unique insights and willingness to create change position them to engage in their communities (Stoneman, 2002; London, Zimmerman, & Erbstein, 2003; Libby Rosen, & Sedonaen, 2005), they often have limited leadership opportunities because the adults in their communities have not tapped into or developed youth leadership potential or included youth voices (Christens & Dolan, 2011; London, et al., 2003). Certainly, youth can take on leadership roles in the clubs, sports, and extracurricular activities they are involved in (Roberts & Treasure, 1992; Wehman, 1996), and are engaged in intentional spaces for leadership such as youth councils (Taft & Gordon, 2013), community organizing efforts (Christens & Dolan, 2011), and community coalitions (CADCA, 2009). Yet, as Stoneman (2000) noted, By ignoring youth voice, communities lose a valuable resource and potentially design solutions that are misaligned with youth needs.   In response to the need for more youth engagement, many programs are emerging in the United States that are designed to promote youth leadership and encourage youth engagement in their communities (Camino & Zeldin, 2002; Flanagan, 2004; Christens & Dolan, 2011). One  10 problem with many of these programs is they are designed from an adult perspective, without Roach, et al., 1999, Siedentop, 1995; McNae, 2010; Dempster & Lizzio, 2007; Kress, 2006; Klau, 2006).  This approach to promoting youth leadership may be a significant misfit for youth because youth and adults often have different viewpoints and opinions (Erikson, 1950; Miller & Glass, 1989), and how they think about leadership may be no exception (Roach, 1999; Dempster & Lizzio, 2007; Mortensen, et al., 2014). Programs that are misaligned with participation, or, may only appeal to those youth whose ideas about leadership align with the models defined by adults. Because there is little research that has explored how youth think about leadership (Dempster & Lizzio, 2007; Kress, 2006; MacNeil, 2006; Matthews, 2004; Conner & Strobel, 2007), the field is not yet equipped to design youth leadership development programs that fully meet youth needs and desires (Dempster & Lizzio, 2007). Feminist standpoint theory highlights how individuals social positions shape how they experience the world (Campbell & Wasco, 2000; Intemann, 2010; Wylie, 2003), which may be perceptions of and experiences with leadership.  Youth are socially located in society to have little power or influence in their communities and do not have the ability to take on traditional leadership roles like adults, such as elected government positions, managerial roles within organizations, or neighborhood association leadership roles so must take on other forms of leadership that align with their current lives and position within their communities. In for becoming adult leaders rather than youth leaders (MacNeil, 2006; Yu, Lewis-Charp, & Gambone, 2007).  Ylived experiences with leadership, such as the opportunities they have had (or not had) to act as leaders or various leaders they have learned from or looked up to, are  11 different  (Roach, et al., 1999). Because of this, feminist standpoint theory (Wylie, 2003) would argue that youth and adults would likely define leadership differently.  Understanding these differences is important because the research on adult leadership, which represents the majority of research on leadership to date (MacNeil, 2006; Dempster & Lizzio, 2007), and is often used to design youth leadership programs, may not represent youth experiences with leadership. Gaining a deeper understanding of how youth define leadership is critical to create programs and opportunities that are responsive to their views of leadership and current social positions.  The primary purpose of this study was to empirically examine an emerging, youth-defined framework of leadership, the Collective Change Youth Leadership Framework, with a national sample of youth.  This framework was developed from two initial studies with youth, one that explored ideas about leadership through Photovoice (Mortensen, et al., 2014) and a second that used Concept Mapping to understand how youth define leadership. The proposed framework includes six dimensions of leadership that reflect youths perceptions of leadership: anyone can be a leader, creating change, collective action, modeling and mentoring, strong character, and motivation and ambition. By testing this framework on a national sample, this study aims to deepen current understandings of what leadership means to youth, and add to the literature a youth-defined framework of leadership. These findings can inform future leadership development efforts so more youth will engage in these opportunities and take on leadership roles within their communities both now, and as adults. The second purpose of this study is to understand variables that relate to youth perceptions of leadership. Several variables previous researchers have linked to youth civic engagement including sense of empowerment (Zimmerman, 1995; Rutter, 1993; Peterson, et al., 2011), critical consciousness (Diemer, et al.,  12 2014; Christens & Dolan, 2011; Foster-Fishman, Lichty, Chen, & Mortensen, 2012), adult support (Larson & Angus, 2011; Grossman, Campbell, & Raley, 2007; Durlak & Weissberg, 2007), and background characteristics such as gender (Cassell, Huffaker, Tversky, & Ferriman, 2006; Jenkins, 2005) and age (Flanagan & Levine, 2010; Flanagan & Sherrod, 1998; Sears & Levy, 2003; Niemi & Hepburn, 1995) were explored.  This is an important area to investigate so we can understand relate to their ideas about leadership, and learn how to support them in ways that align with their leadership goals.  LITERATURE REVIEW   The following literature review will discuss the current state of youth leadership in the scholarly literature, including definitions that exist across leadership programs, what we know about how youth define leadership, the benefits associated with youth leadership, and the shortcomings of communities in engaging youth as leaders. Next, the Collective Change Youth Leadership Framework is presented along with a description of the two studies that informed the development of the framework, as well as how it compares to the adult leadership literature.  After that is a description of the study that was completed to empirically test this framework and begin to understand how it aligns with youth definitions of leadership. Secondary survey data collected from a national sample of youth who participated in a leadership training conference was included in this study.  DEFINITION OF YOUTH  To begin a discussion on youth leadership, it is first important to define who qualifies as The Interagency Working Group on Youth Programs, a collaboration between twelve federal departments and five federal agencies, developed a strategic plan to identify their common goals for youth (Interagency Working Group on Youth Programs, 2013). In this plan,  13 they defined youth as those individuals between the ages of 10 and 24 (though do acknowledge some partner agencies target different age ranges). Although the definitions of youth nearly always use the metric of chronological age, there is great variability across the youth development literature as to what this age range is. Variation in who is and is not included in the definition of youth can be seen in the youth leadership literature where youth samples have included participants as young as 8 and as old as 30. For example, ievaluation of five leadership programs, the programs included youth between the ages of 14 and 30 (Zeldin & Camino, 1999).  The YouthBuild Programs included youth ages 16-24, focusing primarily on youth completing their GEDs or high school diplomas while learning job skills through building affordable housing (YouthBuild USA, 2012). The Generation With Promise program (Richards-Schuster, Checkoway, & Christian, 2012) involved middle school aged (generally 12-14 years old) youth in Michigan as they worked to transform their schools and communities, and the Ohio 4-H Leadership Camp invited youth ages 15-18 to participate (Ohio State University, 2012). The inconsistency across the literature and within society of who qualifies as youth suggests there may be a blurry line youth and young adulthood.  The definitactions. There are different ideas about when youth can be held accountable, with some arguing youth are not mature enough to be taken seriously until they are older, while others believe youth have the agency to make decisions and should be held accountable for their actions at a younger age. This is especially true in the criminal justice system, which often pushes to lower the age at which youth can be charged as adults, such as in some states, where youth under 18 can be charged as adults (Reppucci, 1999; Prout, 2000; Scott, 2000).   14 For the purpose of this study, youth are defined as individuals between the ages of 12 and 20. This age range was selected because though youth are often considered adults at 18 in the United States (e.g. voting, joining the military without parental consent), they do not enjoy the full rights of adults (e.g., holding some public offices, legal age to drink) until they are older. The youth leadership literature also provides support for selecting this age range as many youth development programs include individuals over the age of 18 (Zeldin & Camino, 1999; YouthBuild USA, 2012; Lewis-Charp, Yu, Soukamneuth, & Lacoe, 2003; Roach, et al., 1999) and past Kohlhagen, 2000; Mortensen, et al., 2014). In addition, in many instances, youth can be considered adults at a younger age (e.g., many states consider 17 the age when youth can be criminally charged as adults, youth under 18 can consent to some medical care without parent permission), or not until they are over 18 (e.g., minimum age to legally drink alcohol is 21, minimum age requirements to hold some government positions range from 25 to 35). For these reasons, this study defines years old.  CURRENT STATE OF DEFINING YOUTH LEADERSHIP Across the youth leadership literature, there is not an agreed upon definition of youth leadership (Ricketts & Rudd, 2002; Libby, Sedonaen, & Bliss, 2006; Edelman, et al., 2004; Zeldin & Camino, 1999; Connor & Strobel, 2007).  Various definitions have been proposed and have been used to guide the development of youth leadership programs. For example, some see -Charp, et al., 2003, p. 102). This lack of clarity may cause confusion across the field because the  15 terminology is not used consistently, making it difficult to understand if programs are focusing on promoting leadership in the same way, or if programs emphasize certain areas of leadership over others. The lack of a consistent definition also makes researching the impact of these programs and furthering our understanding of youth leadership a challenge.  A consistent operationalization of youth leadership could make it easier to compare programs that promote leadership and gain a deeper understanding of what youth leadership is and how best to support it.   Table 1 below includes a summary of prominent definitions of youth leadership across the literature. As this table illustrates, leadership has been defined in various ways and each definition has different implications for how leadership development programs train youth. Van emphasis on the ability to influence others and help them act on their own beliefs (p. 7). When using this definition of youth leadership, training might focus on teaching skills and tactics for influencing others. (1999) definition of youth leadership was developed from an evaluation of five leadership programs. It emphasizes the importance of the context in which youth are working, and focuses on a cause selected by youth. Their definition also includes skill building, action, membership, and modeling (Zeldin & Camino, 1999). This guiding definition would give youth authority in selecting a cause they wanted to work toward and programs would support youth in building their skills to take action in their community. On the other hand, more recent literature on youth leadership has focused more on working collaboratively toward a common goal (e.g., Rutgers Cooperative Extension, 2003; Edelman, et al., 2004; Komives, Lucas, McMahon, 2009). Using this conceptualization of leadership as a guide would lead programs to train youth on ways to work within a group and set goals, and  16 make plans for working together to achieve those goals (Leadership Initiative, CADCA, 2009). Similarly, some definitions specifically focus on taking Lewis-Charp, Yu, Soukamneuth, & Lacoe, 2003; Social Policy Associates, 2003). Programs using this approach would emphasize civic engagement in the community and learning the skills necessary to engage in group processes to work with others in their communities.  Table 1. Example Definitions and Conceptualizations of Youth Leadership  Source Definition/Conceptualization Key Leadership Dimensions Van Linden & Fertman (1998)  who think for themselves, communicate their thoughts and feelings to others, and help others understand and act on their own beliefs. They influence others in an ethical and socially responsible way. For many, leadership is best described as a physical sensation: a need to share ideas, energy, and creativity, and not let personal insecurities be an obstacle. Being a leader   Communicate ideas to others  Influence others  Share ideas  Trust instincts Zeldin & Camino, 1999 conceptualizes youth leadership as follows: occurring in the context of a cause, chosen by young people; gearing programming toward an explicit set of outcomes; developing in the context of certain day-to day experiences, specifically skill building, action, -11)   Focused on a social cause and outcomes  Incorporates skill-building, action, membership and modeling Rutgers Cooperative Extension, 2003 a common goal. It is a way of thinking and a way of   Influence others  Work toward a common goal  Way of thinking and acting    17   Source Definition/Conceptualization Key Leadership Dimensions Lewis-Charp, Yu, Soukamneuth, & Lacoe, 2003 of youth in the political and/or social sphere (i.e. civic   Engage in political and/or social sphere Social Policy Research Associates, 2003 about individual achievement as it is about learning how to participate in group processes, build consensus, and subsume personal interests and ideas   Participate in group processes  Focus on collective interests  Edelman, et al., 2004 as the ability to envision a goal or needed change, to take initiative or action to achieve the goal, to take responsibility for outcomes, and to work well with, (p. 4)  Take action to achieve goals  Take responsibility for outcomes  Work with others Wehmeyer, Agran, & Hughes, 1998 action, influence the opinion and behavior of other (Edelman, et al., 2004, p. 4)  Guide/direct others  Influence others Adolescent Employment Readiness Center, Hospital, n.d. weaknesses, set personal and vocational goals, and have the self-esteem to carry them out. It includes the ability to identify community resources and use them, not only to live independently, but also to establish support networks to participate in community life and 2004, p. 4)  Analyze personal strengths/ weaknesses  Set goals  Self-esteem  Identify resources  Establish support networks and participate in community  Effect positive social change        18  Source Definition/Conceptualization Key Leadership Dimensions Libby, Sedonaen, & Bliss, 2006 "A full definition of youth leadership must encompass values, power, and action; without power sharing, a theory of change, and action, youth are not exercising leadership, but taking steps to plan and implement activities prescribed by adults." (p. 23)  Values driven  Power sharing with adults  Action focused Komives, Lucas, McMahon, 2009    Work with others toward positive change   Across youth programs, different definitions of leadership may be used, which may lead to different program goals and expectations, and may cause programs to train youth in a variety of different skills and label them as leadership. This could result in conflicting or unclear program goals or ideas about what leadership is. Overall, a consistent guiding definition of youth leadership across programs would benefit the youth leadership field in several ways. First, it could provide guidance when creating programs that promote leadership, and help programs tailor their activities to the needs and goals of youth participants, emphasizing the aspects of leadership they are most interested in. Next, a consistent definition could improve our ability to research youth leadership and expand our understanding of this important concept. Finally, pcurrent and future leadership activities, and programs could learn from each other how best to promote various components of leadership.  While a consistent framework or definition of youth leadership would provide clarity to the field and allow for more systematic comparisons of youth leadership programs, there is also merit in leadership programs emphasizing various aspects of youth leadership to best meet youth  19 needs and goals. The aim of these programs could then be to develop different types of youth leaders to work on assorted social problems or to lead in various settings. A common youth leadership framework could still guide programs, and focus on various leadership components that are most valuable to the youth they serve. These programs could then be compared to understand how diverse youth benefit from different types of programs, or how emphasizing different components of an overarching model may lead to different types of outcomes for youth and their communities. Another approach could be to create multiple youth leadership definitions or theories, similar to the adult leadership literature, that may be used in different settings or for different groups of youth.  Overall, the above illustrates that across the field of youth leadership, no single definition has been adopted or widely used by researchers, and when leadership is defined, it is defined by adults. Understanding how youth define leadership and how they understand this concept in their lives can provide guidance to youth leadership programs to ensure they are driven by and aligned , which may lead to increased engagement in these types of programs and ultimately more youth leaders within communities (Dempster & Lizzio, 2007). Part of being youth-driven is understanding youth perspectives and responding to their contexts and experiences so programs are representative of youth ideas and promote leadership that is meaningful to youth (Walker, Marczak, Blyth, & Borden, 2005). Feminist standpoint theory provides a useful frame for exploring how youth define leadership, position within society influences the way they view their world and their opportunities (Campbell & Wasco, 2000; Intemann, 2010; Wylie, 2003). Currently, the youth leadership literature is influenced by adult models of leadership, and no youth-driven model has been developed. Youth and adults vary in their opportunities and contexts in which to take on  20 leadership roles, so learning about leadership from a youth perspective would better position organizations to support youth in ways that align with their goals. Using a youth-driven definition, and understanding how different youth may view leadership could help promote buy-in and encourage youth to take on meaningful leadership roles both now and in the future.  FEMINIST STANDPOINT THEORY.  Feminist standpoint theory is informed by post-positivist critical theory that posits no single objective truth exists (Campbell & Wasco, 2000). According to this theory, reality is constructed by factors such as race, gender, class, and sexual orientation, but standpoints are more than simply individual within society (Intemann, 2010). Standpoints are developed through critical reflections of power structures and how those power structures socially locate individuals, affecting how they see and experience the world (Intemann, 2010; Wylie, 2003). Intemann (2010) noted this critical consciousness is not developed by individuals alone, but rather as communities, where people come together to understand patterns of experiences among their lives that reflect power relations within society.  Some authors argue marginalized groups, such as women and racial minorities, have an advantage of being able to see a more accurate view of reality because they are forced to live within the dominant culture as well as their own oppressed culture, something members of the dominant culture do not fully understand (Campbell & Wasco, 2000; Nielsen, 1990; Westkott, 1990).  Because youth are socially located within a power structure that marginalizes them and prevents them from fully participating in civil society (Langhout & Thomas, 2010), their social location may play an important role in conceptualizing leadership in a different way than adults do.  In addition, because the field varies in how communities currently provide leadership opportunities to youth (e.g. in programs or the community) or support youth (e.g. directive  21 versus facilitative support), these experiences may also shape how youth see their role as leaders.  e their definitions of leadership so adults can support them in appropriate ways and promote specific components of leadership that may be more salient for some youth as compared to others based on those standpoints. This study explored how a number of variables, including background characteristics, past support relate to the way youth think about leadership.  By gaining a deeper understanding of variables that may influence their thoughts on leadership, we will be able to offer guidance to youth leadership practitioners for creating programs and opportunities for leadership that are tailored to the needs of various youth.  WHAT WE KNOW ABOUT YOUTHS DEFINITIONS OF LEADERSHIP  In the few studies that have asked youth about their views of leadership, youth typically focus on aspects of leadership that are different from the focus on power and control often emphasized in adult theories of leadership (Mortensen, et al., 2014; Yu, Lewis-Charp, & Gambone, 2007; Culp & Kohlhagen, 2000). Instead, youth tend to emphasize creating change, working collectively toward their goals, and acting as role models and mentors for others.  A study of youth and adults at a 4-H Leadership Camp asked these groups to define leadership through a series of activities, such as brainstorming key traits of leaders, using these ideas to create a definition of leadership, identifying leaders and their leadership characteristics, and describing ways to exert leadership in their families, clubs, and communities (Culp & Kohlhagen, 2000). The authors found youth emphasized listening to others, being a good role model, and helping others when asked to identify leader characteristics (Culp & Kohlhagen, 2000). These ideas are action oriented and focus on working together and creating change.  In  22 contrast, adults in the study talked about the ability to speak confidently in front of people and being responsible as characteristics of leaders (Culp & Kohlhagen, 2000).  These ideas are less action oriented and more focused on individual traits than the youth definitions.  Similarly, as an initial attempt to create a youth generated definition of leadership, researchers from Michigan State University engaged youth in a Photovoice project to understand their ideas about leadership (Mortensen, et al., 2014; Mortensen, Foster-Fishman, Lichty, Harfst, Hockin, Warsinske, 2012). The youth emphasized creating change, working collectively toward goals, modeling behaviors to and mentoring others, having strong character, and the idea that anyone has the potential to be a leader (Mortensen, et al., 2014).  Roach and team studied youth across over 120 youth organizations for ten years to understand what youth leadership looks like. Their team collected data through field notes and audio recordings of daily activities, as well as interviews about leadership with both youth and adults. Youth emphasized working collaboratively and focusing with the group rather than specific traits of a leader which are often emphasized in the adult leadership literature (Roach, et al., 1999, p. 18). Listening and communicating with others (Roach, et al., 1999) and representing the group (Cassell, Huffaker, Tversky, & Ferriman, 2006; Roach, et al., 1999), are also common ideas that have been identified in studies that explored youth perceptions of leadership. While the adult-created definitions and guiding conceptualizations of leadership that exist in the youth leadership literature align in some ways with what we are beginning to learn about youth definitions of leadership, they also vary quite a bit from how youth define leadership.  Some of the existing adult-driven definitions within the youth leadership literature emphasize working collectively toward a common goal (Zeldin & Camino, 1999; Rutgers Cooperative Extension, 2003; Social Policy Research Associates, 2003) and some explicitly mention creating change (Zeldin &  23 Camino, 1999; Komives, et al., 2009; Hospital, n.d.). These were major themes in the studies on youth-defined leadership.  On the other hand, some definitions of youth leadership do not include these core components that youth value, and focus more on  (Van Linden & Fertman, 1998; Fertman & Van Linden, 1999; Wehmeyer, Agran, & Hughes, 1998) as well as sharing power (Libby, et al., 2006).    Without engaging youth in defining what leadership means to them, adult-driven definitions may not produce the most effective programs and opportunities to help youth become leaders in their communities. Roach and colleagues (1999) warn that without understanding how youth view leadership, adults risk overlooking youth who may display leadership outside of a traditional setting or alienate those youth who would benefit from a deeper understanding of leadership (p. 16). Also, some research suggests a decline in youth participation, often due to negative stereotypes adults hold about youth (e.g., Zeldin, et al., 2007) or dissatisfaction with conventional spheres of participation (e.g., Dempster & Lizzio, 2007), suggesting programs may not align with youth goals about leadership. However, others have discussed the various ways youth take on leadership roles, often outside of formal settings (e.g., Roach, et al., 1999), and have noted an increase in participation in new ways and settings (Dempster & Lizzio, 2007; Sander & Putnam, 2010; Taft & Gordon, 2013). Sander and Putnam (2010) discuss the increase -at the time of the 9/11 attacks seem to have strengthened their civic conscious and become more civically engaged. One note of caution from the authors is the engagement divide between upper-middle class white youth and their working or lower class white peers (the authors also acknowledge the Black-White engagement gap that persists; Sander & Putnam, 2010). These  24 findings point to the importance of engaging diverse youth in understanding what leadership means to them and learning how best to support them through their efforts to act as leaders. A youth defined model of leadership could provide guidance to programs and communities in developing opportunities that align with youth from different backgrounds to work toward reducing the engagement gap that currently exists.  Studies that have engaged youth to learn about their definitions of leadership have generally included youth participating in leadership programs or organized out-of-school programs. For example, Culp & Kohlhagen (2000) worked with youth in a 4-H camp, Mortensen and colleagues (2014) engaged youth who were participating in a year-long leadership training, -of-school organizations in underserved and at-risk communities, and Cassell and colleagues (2006) followed youth participating in an online leadership forum. These studies engaged youth who were already leaders or actively working to gain leadership skills. Though these studies may not represent all youth, these participants were primed to think about leadership and can speak to their experience when defining what it means to be a leader. The samples in these studies are a strength, and varied in size, geography, and racial/ethnic make-up, providing a diverse set of perspectives from youth around the world. Culp and Kohlhagen (2000) and Mortensen and colleagues (2014) engaged primarily white, rural students, Roach, et al.over 30,000 youth from 120 youth organization across the United States (demographic information was not provided for the youth engaged in these programs), and Cassel(2006) studied over 3,000 youth from 139 different countries around the world. The youth engaged in these studies come from very different locations and backgrounds, which provides a wide array of perspectives from around the world. The similarities in defining leadership across  25 these studies suggests youth may have comparable ideas about leadership regardless of their demographic characteristics or geographic location. Similarly, the methodology varied across studies, from structured activities to Photovoice to interviews, and actively engaged youth in the process of understanding how they defined leadership. Collecting data through a variety of methods and finding similar results may suggest the information collected was not dependent on or influenced by the methods, and provides more compelling evidence that these studies youth in defining leadership, the variety in samples and methods provides evidence that youth from various backgrounds may experience and define leadership in similar ways, which lends support to the need for creating a youth-defined framework of leadership such as the proposed Collective Change Youth Leadership Framework that was explored in this study.  Despite the strengths of these studies, the literature on how youth define leadership also has limitations. For example, the studies that have looked at youth leadership provide a snapshot in time of how youth think about leadership, but definitions may change over time or how different age/gender/racial groups may think about the youth over this time period, nor do they provide much information about the youth they worked with (e.g., demographic information, types of leadership roles they were engaged in, the online Junior Summit, but asked youth about the qualities of leaders when they participated in the Summit and not how their ideas may have changed over time. Of the studies that have asked youth to define leadership, none have put forth a formal framework of youth leadership. Also, the research that has begun to explore leadership from a youth perspective has not  26 permeated the youth leadership world, as scholars (e.g., Dempster & Lizzio, 2007; Kress, 2006; MacNeil, 2006) are continually critiquing the adult-driven models of youth leadership that are promoted across these programs despite what we began to learn nearly twenty years ago with Roach  Learning more about differences across youth in how they define leadership can provide researchers and practitioners a guide to create programs and opportunities that support youth in meaningful ways by aligning programs with the needs of youth, and promoting increased engagement of youth from diverse backgrounds. Engaging youth as leaders within communities can provide missing viewpoints into identifying and solving community issues that affect youth, therefore making communities more supportive of and responsive to youth (Stoneman, 2002; Soleimanpour, Brindis, Geierstanger, Kandawalla, & Kurlaender, 2008; Libby et al., 2005).   BENEFITS OF YOUTH LEADERSHIP There is evidence that youth participation in leadership roles and programs is beneficial to both the participants and their communities (e.g., Christens & Dolan, 2011; Ozer and Douglas, 2013). This section will discuss the benefits to youth of taking on leadership roles as well as benefits associated with participating in programs that foster youth leadership skills. Following that, how schools and the community benefit from youth leadership efforts will be discussed.    BENEFITS TO YOUTH IN TAKING ON LEADERSHIP ROLES. Youth leadership provides important benefits to youth who engage in leadership roles in various community and school settings. Youth leadership, as exemplified through engagement in community change efforts, and taking on leadership roles within school, clubs, sports, programs, and organizations, has been studied for years and the research offers strong support for the positive benefits to youth associated with participation in these leadership roles (Schwartz & Suyemoto, 2012; Mitra,  27 Serriere, & Kirshner, 2014; Ozer & Douglas, 2013; Chan, Ou, & Reynolds, 2014; Scales & Leffert, 1999; Christens & Dolan, 2011; Mitra, 2006).  Across the various ways youth engage as leaders in their communities, such as in school leadership opportunities, or in community organizing initiatives, the positive outcomes associated with participation in leadership roles are quite similar and are discussed below.  First, research has demonstrated youth gain significant psychological benefits from taking on leadership roles including: higher self-esteem, increased motivation, gains in confidence, improved sense of personal control and self-identity, and increased empowerment (e.g., Schwartz & Suyemoto, 2012; Mitra, et al., 2014; Biddle & Mitra, 2015; Christens & Dolan, 2011; Kirshner, ).  For example, Schwartz and Suyemoto -led, adult-supported community  (p. 3). Youth in the Youth Force program engage in trainings to develop their knowledge and skills around community organizing, and then use their leadership skills to engage in community actions such as rallies and community forums. Select youth move on to the Core Members (CM) program. Here, they receive additional leadership training, and take on leadership roles through training other youth in community organizing and working part time as community organizers. The youth in this study focused on youth employment, and advocated for increased funding for youth jobs through meeting with local representatives and community members, and holding community forums (Schwartz & Suyemoto, 2012). As a result of their efforts, the youth secured $10,000 in funding from a local bank to support youth jobs (Schwartz & Suyemoto, 2012). Through interviews with youth participants and quantitative survey measures, the authors found youth were empowered to work to address issues in their community, developed their sense of self-concept, saw themselves as an agent of change, and  28 felt more confident and able to take action (Schwartz & Suyemoto, 2012).  Similarly, Biddle and Mitra (2015) conducted a qualitative study with middle school youth participating in youth-adult partnerships within their school focused on either community health (Getting to Y program) or changing the school culture around expectations for student learners (Great Expectations program). Youth in the Getting to Y program used school health data to identify community assets and challenges related to healthy behaviors. They held a Community Dialogue night to discuss healthy behaviors in their community with youth, parents, and school staff, and then developed action plans to increase healthy behaviors. In focus group conversations, youth participating in the Getting to Y program felt the action research process gave them a structure to use to create change (Biddle & Mitra, 2015). They also discussed developing their sense of belonging by building relationships with peers and adults, feeling more confident about their ability to lead change, and feeling empowered and developing their self-efficacy (Biddle & Mitra, 2015). These examples from Schwartz and Suyemoto (2012) and Biddle and Mitra (2015) demonstrate important psychological benefits youth can gain when taking on youth leadership roles within their schools and communities.   Next, when youth have opportunities to take on leadership roles and share their voice in planning, problem solving, and implementing various aspects of programs and projects, research indicates they increase their intentions and motivation to engage and increase their participation in addressing community issues (Ozer & Douglas, 2013; Schwartz & Suyemoto, 2012; Chan, Ou, & Reynolds, 2014; Borden & Serido, 2009, Foster-Fishman, et al., 2012).  Ozer and Douglas (2013) conducted an experimental study of the impact of participating in a youth-driven participatory research elective class in five high schools. The authors compared outcomes on surveys between youth in the YPAR course and youth in a comparison classroom that trained  29 youth to be peer educators and mentors (Ozer & Douglas, 2013). Both classes began with a six week training on communication, team-building, and identifying school/community problems. The YPAR class then selected an issue to focus on (e.g., prevention of school dropout, stress, cyber bullying, etc.), engaged in a research phase using surveys, interviews, observations, and multi-media to collect data, and then developed action plans to begin to address the selected issue (Ozer & Douglas, 2013).  Ozer and Douglas (2013) found a statistically significant effect for the YPAR class as compared to the control class for sociopolitical skills, motivation to influence their schools/communities, and participatory behavior. This suggests youth gained skills and motivation, and continued to participate in their schools and communities as a result of taking on leadership roles to identify and address problems in their schools. Another example of increasing intentions to continue to participate in the school/community after taking on youth leadership roles comes from training and community organizing program, described above. The youth in the Youth Force program discussed their increased sense of civic responsibility and commitment to their role as community organizers, indicating their commitment to civic engagement within their communities (Schwartz & Suyemoto, 2012). The results of these studies suggest the importance of providing youth with meaningful leadership opportunities to promote positive developmental outcomes.  Engaging youth in a variety of leadership and civic engagement efforts early in life may play an important role in encouraging later community engagement and leadership as adults.  BENEFITS OF PROGRAMS THAT FOSTER YOUTH LEADERSHIP SKILLS. Many youth programs Research often shows a relationship between youth involved in these types of programs and positive academic achievement outcomes. Though the majority of this research is not causal, so  30 alternative explanations for positive gains cannot be ruled out, the literature suggests youth engaged in programs to develop their leadership skills may earn higher grades and develop an interest for furthering their education (Pachen, 2010; Edelman, et al., 2004; Weiss, Little, & Bouffard, 2005). Durlak and colleagues (2010) completed a meta-analysis of after school programs that sought to teach students personal and soproblem-solving, conflict resolution, self-control, leadership, responsible decision-making, or skills related to the enhancement of self-efficacy or self- esteemp. 296). They found afterschool program participation, especially in programs that followed SAFE (sequenced, active, focused, explicit) practices, was related to significant improvements in (2014) completed an evaluation of the YMCA Youth Institute (YI) in Long Beach and identified a matched comparison group of non-participants to learn how the YI program affected GPA, attendance, and English language arts and math standard test scores. The Youth Institute included an eight week summer component that focused on training in leadership, team building, cultural diversity, decision making, life sciences, and information technology skills followed by an optional school year program and/or college readiness program. The overall goals of YI focused on building skills, improving academic achievement, and promoting relationships with The authors found Youth Institute participants overall had significantly higher test scores and somewhat fewer absences than students included in the matched comparison group. For active participants (those students who participated for at least 30 days in the year-round program over the past two years), their academic GPAs and math test scores were significantly higher than those students in the matched comparison group. These results suggest the benefits of participation generally, as well as the  31 increased benefits associated with higher levels of participation. The meta- demonstrate youth participation in programs that promote leadership skills is related to positive academic achievement. In addition,  and Camino (2005) suggest the social capital youth gain through having opportunities to share their voice opens the door to new opportunities for education, employment, and enrichment, which are crucial for success in adolescence as well as in adulthood.   In addition to academic achievement, youth in programs that develop their leadership skills learn many more general life skills including public speaking, decision-making, problem-solving, job responsibility (Scales & Leffert, 1999; Edelman, et al., 2004), strategic thinking (Larson & Angus, 2011), and communication and collaboration skills (Hadley & Trechter, 2010), which promote achievement both within and outside the school setting. The Chicano-empower, and develop leadership skills and qualities in Chicano-Ganey, Alba, Quintero, and Alvarez Alcantara, 2003, p. S46). As part of the program, youth participated in a two-day Leadership Institute. The authors conducted pre/post surveys of youth participants, and found a significant increase in their social and leadership skills (Bloomberg, et al., 2003). Youth reported gaining skills in leadership, identifying and delivering information for presentations, developing action plans, and identifying problems and solutions in their communities (Bloomberg, et al., 2003). The skills the youth gained through their leadership training are generalizable and can be translated to other areas of their life. Another example of (2010) study of a 4-H leadership program, which surveyed youth from across Wisconsin. Youth who participated in 4-H reported the program helped them to increase their communication  32 skills, work with a group/team, work in partnership with an adult, and improve their ability to treat people who are different from them with respect. As these examples demonstrate, the skills gained by participating in programs that focus on developing leadership skills can support youth in the program and beyond.   BENEFITS TO SCHOOLS AND THE COMMUNITY. Communities and organizations that include often neglected youth perspectives in decision-making and design processes can foster meaningful community change that is responsive to youth needs (Stoneman, 2002; Soleimanpour, Brindis, Geierstanger, Kandawalla, & Kurlaender, 2008; Libby et al., 2005).  Schools have begun involving youth in school reform efforts, and beyond the school setting, some communities have created youth councils to encourage engagement and participation in issues that are meaningful to them (Taft & Gordon, 2013). When communities engage youth voices in decision-making and leadership roles, they have the potential to include fresh perspectives to adult conversations, improve programs and community conditions to reflect the needs and ideas of young people, develop more effective solutions to community problems, and create a just community for everyone (Stoneman, 2002; Libby, et al., 2006).  In addition to creating environments that are more reflective of youth needs and goals, engaged youth can Dolan, 2011, p. 544) to the larger community. Schools and communities have experienced community changes responsive to youth needs as a result of youth leaders, specifically creating new programs/services to meet the needs of youth, policy changes driven by youth, and improved perceptions of youth and communication between youth and adults. Examples of these community benefits will be discussed below.   33 First, research demonstrates that youth engaged in various youth programs and community efforts have worked to identify needs in their schools and communities and develop programs and services to address those needs (e.g., Mitra, 2005; Soleimanpour, et al., 2008; Christens & Dolan, 2011; Ren & Langhout, 2010; Checkoway, 2015; Richards-Schuster & Checkoway, 2009; Chen, Weiss, Nicholson, 2010; Peterson, Baker, Leatherman, Newman, & Miske, 2012; Ramey, 2013; Ozer & Douglas, 2013). Schools have begun to include youth in various ways as they have seen school reform efforts can be more successful with student involvement (Mitra, 2005). One example of developing new programs as a result of youth participation comes from -year longitudinal study of school reform efforts. This study occurred in one school and used multiple qualitative methods (interviews, observations, and document review), to understand how adults in the school could support youth voice and involvement in school reform efforts. One of the groups involved in this study focused on providing a tutoring and translation program for newcomer students because they identified this as the greatest need for Latino newcomers (Mitra, 2005). Similarly, the Youth Dialogues on Race and Ethnicity in Metropolitan Detroit Program promoted dialogue between young people of different racial/ethnic groups, enabled youth to plan projects that challenged discrimination, involved supportive adults, and helped to build organizational capacity to work on issues related to race/ethnicity (Checkoway, 2009; Checkoway, 2015). One youth in this program advocated for a youth dialogues course in her school that was implemented as a required course and spread to other school districts in the area (Checkoway, 2015). In addition, students felt teachers needed further training in facilitating these sensitive dialogues, and as a result of their request to school officials, professional development workshops were delivered to more than 30 school districts (Checkoway, 2015). These examples demonstrate the benefits that can emerge through engaging  34 youth leaders to identify needs in their schools and work to create change through programs to better serve the students in their schools.  In addition to the school setting, youth take on leadership roles in their communities and advocate for new programs and services to meet their needs. Richards-Schuster and Checkoway (2009) describe three communities in Michigan that worked to engage youth in local policy issues through youth councils. These participants gave voice to youth in their communities, and advocated for a number of local changes. Examples of these community changes include the development of a skate park in response to an ordinance that banned skateboarding in one community, and in two of the communities, the creation of a teen center to help address the lack of teen activities available in the area. Likewise, Chen and colleagues (2010) worked with girls involved in five local affiliates of Girls, Inc. on a participatory evaluation project to understand the meaning and impact of the organization, and how it could be improved. Two of the affiliate sites took action on the gir. One site created a teen club to engage older teens in programs focused on peer leadership and peer support systems. At the other affiliate site, girls identified the importance of math and science skills and requested more science programs which the organization implemented through securing additional grant funding (Chen, et al., 2010). These examples demonstrate the opportunity for creating programs and services responsive to youth needs by involving youth in community leadership roles.    Youth have also affected policy changes in their schools and communities (e.g., Soleimanpour, et al., 2008; Ren & Langhout, 2010; Christens & Dolan, 2011; Powers & Allaman, 2012; Jacquez, Vaughn, & Wagner, 2013). For example, Soleimanpour, and team (2008) conducted two case studies of students engaged in research projects to improve their School Based Health Centers (SBHC). The students in these projects surveyed their peers,  35 presented their research to leaders in the school, and affected policy change related to condom distribution in the school. The new policy allowed condoms to be available on-site for students, and allowed health educators to distribute condoms after presentations at school. This policy change was put in place district-wide and reduced the barrier to accessing condoms they faced in the past (Soleimanpour, et al., 2008). Also, Ren and Langhout (2010) worked with elementary aged students to use a participatory action research approach to improving the school playground. Through playground observations conducted by the research team and focus groups perspectives on recess and suggestions for improvement. As a result of this work, funding was reallocated to purchase additional equipment and to hire more recess aides to lead games. In addition, the indoor recess options were expanded from just movies to additional activity options such as playing board games and drawing (Ren & Langhout, 2010). These changes were driven by the children that participated in the study through identifying the problems with recess and suggesting solutions to these problems. Another example comes from The Youth United for Change (YUC) group in Philadelphia. Youth in this program organized a campaign to restructure two large high schools into smaller schools in order to hold them more accountable to the needs of students (Powers & Allaman, 2012). These youth conducted site visits to other schools across the country that successfully accomplished this goal, and ultimately had their proposal accepted by the school district to implement the change (Powers & Allaman, 2012). As demonstrated by these school changes, youth from elementary school through high school have had success in creating policy changes that reflect their needs. Policy change also occurs in the broader community as a result of youth leadership efforts. Christens and Dolan (2011) conducted interviews with youth and adult supporters  36 engaged in a youth organizing initiative and reviewed press coverage and organizational documents to explore the impact of youth organizing on youth and communities. They found policy change occurred as a result of youth efforts, specifically youth influenced the city budget by pressuring local officials to fund schools and parks, and worked with the local police to create community policing policies and practices (Christens & Dolan, 2011). Similarly, youth involved in the Youth Affordabili(T) Coalition (YAC) campaigned to extend the hours in which students could use their school-issued public transportation passes (Powers & Allaman, 2012). They engaged in a Participatory Action Research project that gathered information from their peers through surveys and interviews to inform their policy demands (Powers & Allaman, 2012). Additional examples of policy changes in other communities include those related to mall curfews (Richards-Schuster & Checkoway, 2009), air-quality standards (Minkler, 2014), and integrating youth into organizational and decision-making processes (Ramey, 2013). Engaging youth in policy change efforts may lead to lasting changes within schools and communities that can benefit individuals, schools, and organizations for years to come.  Next, studies of youth organizing have found youth engagement can change erceptions of youth (Zeldin, Petrokubi, Collura, Camino, & Skolaski, 2009).  In  described above, one group of youth focused on improving the communication between students and teachers. She discussed the importance of youth and adults working in partnership and balancing power, as well as the role of adult advisors in helping to manage the relationship between the youth and the broader school. Working collaboratively helped to improve perceptions of youth and communication between youth and teachers in the school.  37 viewing students as able to identify problems and define solutions to improve the school. These examples demonstrate how youth have affected adult perceptions of youth within school settings. Within communities, youth engaged in change efforts have also altered perceptions of adults to be more positive about youth and their contributions to the community, and in some cases changed the behaviors of adults as well. Christens and Dolan (2011) explain that as a result of youth efforts, adults in the community saw the youth organizing initiative as a respected institution in the community, that focused on using careful research, effective tactics, and persistence to achieve change (Christens & Dolan, 2011). The study provides evidence that youth leaders not only play an important role in creating community change, such as the policy changes related to the city budget and community policing described above, but also have the power to shift perceptions and behaviors of adults within the community. Similarly, adults supporting the 4-H Got Farms! Project in which youth worked to learn about and take action on community concerns surrounding food security, quality, and cost reported viewing youth as valuable community assets as a result of participating in this program (Peterson, et al., 2012) These youth held a forum to learn about community concerns, and took action through building raised bed gardens, repairing the school greenhouse, and holding a large food festival (Peterson, et al., 2012). The community gained valuable food resources and increased their knowledge around important issues. decision-roved youth-adult interactions, more opportunities for youth to participate, and a shift in attitudes toward youth in recognizing their capacities and overcoming negative stereotypes as common outcomes reported in the reviewed studies (Ramey, 2013). She goes on to discuss that as a result of youth participation, organizations demonstrated a  38 stronger connection between youth needs and the organization (e.g. in policies and programs that align with youth needs) and more involvement by youth in decision-making through youth councils and the overall organizational and governance structure (Ramey, 2013). These studies provide examples of how youth leaders have positively affected adult perceptions of youth and in some cases have changed the behaviors of adults.  SUMMARY. The literature reviewed above provides numerous examples of how youth and communities benefit from youth leadership. Youth who engage in leadership roles develop important psychological benefits such as increased self-esteem, confidence, and sense of empowerment (Schwartz & Suyemoto, 2012; Mitra, et al., 2014; Biddle & Mitra, 2015; ). Youth leaders also increase their motivation to continue to participate in community change efforts. These are important skills and attitudes to develop that may lead to an increase in youth leaders now and adult leaders in the future. As will be discussed below, a critique of many youth leadership programs is the lack of opportunity to practice youth leadership beyond the program setting (Yu, et al., 2007). These studies show that engaging youth in leadership roles, driven by their interests, can lead to important outcomes for these youth. These findings suggest providing youth with opportunities to take on leader roles is an important part of developing youth skills and attitudes toward leadership.  Youth who engage in programs that foster youth leadership skills develop both academic and life skills (e.g., public speaking, decision making, strategic thinking, communication, etc.). These are important skills for all youth to develop, and provides strength to the argument that youth development programs that teach leadership skills are beneficial for all youth. Research suggests that youth in low-income communities often have fewer opportunities to participate in  39 organized programs such as these, leaving them at a further disadvantage from their higher income peers. This research should be used to advocate for additional opportunities for all youth as the skills they develop can benefit both their academic achievement, and life both now and in the future.  Finally, schools and communities benefit from youth engagement. As youth work to create new programs and policies within their schools and communities, they demonstrate their abilities as leaders, and research suggests adults have more positive views of youth. This may lead to a greater willingness for adults to work with youth and engage them in school and community change efforts (Ramey, 2013). Increasing youth engagement can lead to schools and communities being more responsive to youth needs and may encourage further opportunities for more youth to act as leaders.  While the above studies provide a view into youth leadership, it is important to remember the field of youth leadership is still much smaller than that of adult leadership (MacNeil, 2006), and requires additional research to continue to grow our understanding of this topic. Below, strengths and limitations of the studies reviewed will be discussed.  STRENGTHS AND LIMITATIONS OF YOUTH LEADERSHIP RESEARCH. The studies described above include examples of how youth leadership can benefit both youth and the communities they live in. These studies demonstrate strengths by including diverse samples and a variety methods, but also show a number of limitations. These include small sample sizes, cross sectional rather than longitudinal designs, a lack of experimental studies, and a limited representation of youth due to the lack of availability of programs in some communities. These strengths and limitations will be discussed below.   40 To begin, one of the strengths of the youth leadership research reviewed here is the diverse youth included across various studies. Youth ranged in age from elementary school (e.g., Ren & Langhout, 2010) to youth out of high school (e.g., Christens & Dolan, 2011). Youth also varied in their race/ethnicity across the studies reviewed. Some studies included very diverse samples (e.g., Schwartz & Suyemoto, 2012; Ozer & Douglas, 2013; Larson & Angus, 2011; Soleimanpour, et al., 2008) and others targeted specific groups of youth or worked within communities that were made up of primarily one racial/ethnic group (e.g., Chan, et al., 2014; Hadley & Trechter, 2010; Christens & Dolan, 2011). In addition, geographical location varied across studies, and included youth working on the East Coast (e.g., Biddle & Mitra, 2015; Schwartz & Suyemoto, 2012), the Midwest (e.g., Richards-Schuster & Checkoway, 2009), and West Coast (e.g., Ozer & Douglas, 2013). Finally, youth participants came from both urban (e.g., Schwartz & Suyemoto, 2012; Ozer & Douglas) and rural schools and communities (e.g., Peterson, et al., 2012; Hadley & Trechter, 2010).  Overall, the diverse samples in these studies, including youth from various racial groups, age ranges, and locations demonstrates the positive personal and community outcomes described above that are associated with youth leadership are not limited to certain groups or types of youth. The variety of youth engaged in these studies also lends strength to the argument that communities should continue to create programs and opportunities that engage diverse youth leaders. However, despite the strengths of the samples included in these studies, there are some limitations. For example, not all of the articles reviewed described specific demographic information about participants (e.g., Ren & Langhout, 2010; Richards-Schuster & Checkoway, 2009). Without information about the youth participants, the reader is not able to compare findings to other studies or gain a complete understanding of the youth engaged in these programs. Similarly, one  41 critique of the youth leadership literature is that programs are often less available for youth in low income communities (Flanagan, 2009). While some of these studies intentionally engaged specific ethnic/racial groups and worked in diverse communities, youth in low income communities are likely still underrepresented in the research because they often have fewer opportunities to engage as leaders. Research should continue to engage a range of youth and explore how youth leadership programs and community opportunities to act as leaders align with the needs of youth as well as the outcomes associated with their participation. Another strength is the variety of methods used across, and often within, these studies of youth leadership. Many of the authors focused on qualitative methods and case studies, incorporating interviews and focus groups with youth, and often adults involved in the work. Qualitative approaches such as interviews and focus groups allow the researcher to gain an in-depth understanding the relationship between the individual and the setting they are in (Singleton & Straits, 2005). In these studies, the authors learned directly from youth about the process and outcomes associated with participation in leadership roles and programs in their schools and communities (e.g., Ren & Langhout, 2010; Ozer & Douglas, 2013; Larson & Angus, 2011). One possible drawback of these methods is the extensive time and personnel required to collect and analyze qualitative data, which may limit the number of participants researchers can engage with (Singleton & Straits, 2005).  Document review (e.g., organizational documents, community news stories, school academic records) was also common within these studies (e.g., Christens & Dolan, 2011; . By reviewing documents both internal and external to programs, researchers are able to gain a more objective understanding of program goals and individual and  42 community impacts. These documents can also provide a variety of perspectives, including those of program staff, youth, and the broader community. Document review also allows researchers to understand impacts beyond the individual, and consider the broader social structure and social changes that may occur beyond programs and their participants (Singleton & Straits, 2005). One possible limitation in reviewing existing documents is the fact that the documents were not goals (Singleton & Straits, 2005). Another possible limitation of including existing documents as part of these studies is the quality of the information that is included (Singleton & Straits, 2005). Studies that did review documents also included other methods (e.g., interviews) to triangulate their findings and provide a more thorough description of programs and participants, so may have reduced these possible limitations to reviewing existing documents.  Another common method, observations (e.g., Biddle & Mitra, 2015; Schwartz & Suyemoto, 2012; Ren & Langhout, 2010), allowed researchers to view programs from an outsiders perspective, and similar to the focus groups and interviews, provided an in-depth understanding of the participants and the settings they worked within (Singleton & Straits, 2005). However, observations are often done by one or a few individuals, are dependent on the skills and interpretation of those individuals, may be difficult to replicate, and could be less generalizable than other methods (Singleton & Straits, 2005). Many of these studies included multiple methods, so again, the drawbacks of observations may be limited by the addition of other data collection methods.   Fewer studies included surveys in their work on youth leadership. Some surveys incorporated existing, validated measures to measure youth outcomes such as self-efficacy, empowerment, and self-esteem (e.g., Ozer & Douglas, 2013; Schwartz & Suyemoto, 2012)  43 which added strength to the validity of their outcomes. However, not all surveys included validated scales, and instead were created by the authors and may include more measurement error than validated measures (e.g., Hadley & Trechter, 2010; Chan, et al., 2014). In addition, some studies focused on a review of existing youth leadership research and summarized findings from the literature on the benefits associated with youth participation in leadership roles and programs (e.g., Durlak, et al., 2010; Powers & Allaman, 2012). This approach allowed for a comparison of outcomes discussed across a number of programs and settings and a summary of Finally, a very small portion of more recent studies included comparison groups of youth not engaged in the leadership program under examination (e.g., Ozer & Douglas, . These studies compared student outcomes focused on empowerment and self-esteem (Ozer & Douglas, 2013) and school GPA, test scores, and re students that engaged in specific youth leadership opportunities to those that did not. A major benefit of this approach is that it allowed for a more causal discussion of the outcomes associated with participating in youth leadership roles, which the large majority of youth leadership studies lack. The positive benefits associated with their participation suggests these students did experience both personal and academic gains such as increased sense of empowerment, test scores, and GPAs than non-participants.     Overall, case studies represented the large majority of studies on youth leadership and were an appropriate choice in these instances. Case studies allowed for youth voice to be heard about their experiences through interviews and focus groups, and the information was reviewing program and community documents. The variety of methods applied both within and  44 across studies provides robust evidence for the positive outcomes associated with youth leadership for both individuals and the community. While these studies point to the importance of engaging youth to improve school reform efforts and encourage youth participation in community leadership roles, these studies do have limitations. First, some of the studies included small sample sizes, which provide valuable information about a small number of youth leaders, but a limited understanding of how leadership may impact youth more broadly. One drawback of small sample sizes for quantitative studies is the potential for a lack of statistical power for analyses (Sullivan & Feinn, 2012). The majority of studies reviewed here were qualitative, and provided an in-depth picture of how these programs affected the youth participants. While small samples are common in qualitative studies, there is no set acceptable sample size for these studies (Elo, et al., 2014). A common recommendation is to seek saturation in the data, in which similar ideas are repeated across participants, as a way to identify the optimal sample size (Guthrie, Yongvanich, & Ricceri, 2004; Elo, et al., 2014). However, in most cases, data analysis is completed after the data has been fully collected, or researchers are limited by the number of participants in a program/setting, so it is not possible to adjust the sample size to reach saturation (Elo, et al., 2014). For example, youth at one time, Ren and Langhout (2010) engaged 30 students across eight focus groups to learn about problems that occurred during recess and potential solutions to address those problems, and Biddle and Mitra (2015) included 19 youth in focus groups to learn about their experiences in a youth-adult partnership program. Further, some of the studies did not provide specific information on the sample size or characteristics of participants (e.g., Powers & Allaman, 2012; Richards-Schuster & Checkoway, 2009), so a full understanding of the youth  45 who were engaged in these programs, or the ability to compare these studies to other youth leadership research is not possible.  Another limitation of the youth leadership literature is the short time frame over which these studies occur, and the limited longitudinal studies that exist (Chan, et al., 2014). Cross sectional and short longitudinal studies provide a valuable snapshot of youth at one point or a short timeframe in their lives and can capture their experiences and immediate outcomes associated with their leadership activities, as evidenced by the literature presented above. However, currently we do not have an understanding of how these types of programs and leadership opportunities shape youth later in their life.  Authors have speculated that engagement in leadership roles as youth may influence future actions (Benson & Scales, 2009) but longitudinal studies that have explored the impact on youth over time have generally only examined change over the course of the program, often only a school year or shorter time frame (e.g., Schwartz & Suyemoto, 2012; Ozer & Douglas, 2013). Overall, our current short term understanding of the impact leadership programs and other leadership opportunities can have on youth is valuable as organizations and communities work to develop youth leaders, but conclusions about their impact should be tempered by the lack of long term follow up that we currently have on these youth. Without tracking youth beyond the program or the evaluation of an initiative they are involved in, we cannot know how these initiatives may impact their future actions or engagement as leaders in the community which could help inform youth programming to ensure it supports leadership in their current and future lives.  Another shortcoming of the youth leadership research to date is the lack of experimental and quasi-experimental research studies that include randomly assigned control groups. Implementing these types of more rigorous research methods would allow for causal statements  46 to be made about the impact of leadership programs on participants. As discussed above, qualitative methods and case studies were appropriate methods for these studies given the importance of learning about both the process and outcomes associated with youth leadership. These studies demonstrated positive outcomes for both youth and their communities. As the field develops, moving to more rigorous methods will be beneficial. Two of the more recently published studies reviewed here did include experimental designs in which they compared the outcomes of youth involved in leadership activities to those who were not (e.g., Ozer & Douglas, )academic achievement for youth engaged in youth leadership programs as compared to those not participating. These studies demonstrate the feasibility of incorporating comparison groups to understand the impact of youth leadership on youth participants. The current literature suggests positive outcomes related to youth leadership, but additional, more rigorous research designs are needed.  More experimental and quasi-experimental studies such as the two examples discussed here can allow for more causal statements to be made about programs and rule out alternative explanations for the outcomes achieved such as differential selection into the program or other initiatives that may be occurring in the community.  A final limitation of the youth leadership literature is the possibility that youth programs are not available to all youth, so the research is only reflective of a specific type of youth that engages in leadership opportunities. The is the gap in civic opportunities that exists for youth in high poverty communities as compared to wealthier communities (Chan, et al., 2014; Flanagan, 2009). Youth in these communities often have fewer organized activities, and schools have fewer civic learning activities (Hart & Atkins, 2002). Because of this, the literature may not be representative of all youth, but rather those in communities that support  47 youth leadership. It also may be the case that those youth who do participate in leadership opportunities are higher achieving and more motivated youth than those who do not participate (Kress, 2006). The literature reviewed here demonstrates some bias in this area, engaging select youth over others, but does not fully support these assertions because some of the studies intentionally engaged a range of youth or those youth who are typically not included as leaders. For example, change provides an example of intentionally engaging a variety of youth in leadership opportunities. In the study, one adult advisor recruited a cross section of students with various demographics, academic achievement, and friend groups while the other advisor intentionally recruited students who were disconnected from the school, often cutting classes and not participating in school activities, largely due to family responsibilities (Mitra, 2005). Similarly, Chen and colleagues (2010) worked with Girls, Inc. affiliates who engaged a wide variety of girls, including a majority of girls from minority backgrounds, many of which came from low income neighborhoods, girls with learning disabilities, or those who were in foster care, but also included some girls who were straight A students. As an organization, Girls, Inc. aims to serve all girls, but generally does engage low-income, minority girls in local affiliates across the country. These examples demonstrate an intentional effort to engage youth who were not the typical leaders in their schools and communities, and may suggest the findings are more generalizable to adolescents more broadly.  Though a few studies intentionally engaged youth who may not traditionally participate studies did acknowledge the fact that the youth in their studies may not represent the general youth population. For example, Mitra and Biddle (2015) explain the youth who participated in  48 their study of youth-adult partnerships included students who were engaged in other leadership groups (e.g., student council, student leadership teams, state-sponsored tobacco program) and of students who had demonstrated burgeoning leadership skills or who were motivated and that used an experimental design, in which elective classrooms in a school were randomized into a treatment (YPAR class) or control condition (peer education class), the authors note a similar limitation. They explain rse on peer education and the classrooms were then randomized into the traditional peer education or YPAR alternative (p. 73). As these examples demonstrate, the youth leadership literature may not represent all youth, but include those who are more apt to seek out or have more access to leadership or civic engagement opportunities. leadership programs affect youth in communities where youth leadership opportunities may be less accessible. In conclusion, research indicates that youth involvement as leaders within their schools and communities promotes dynamic, interactive relationships between youth and their context (Theokas, Almerigi, et al., 2005; Lerner, et al., 2005; Lerner, Lerner, & Benson, 2011).   Therefore, promoting positive opportunities for youth, such as leadership roles and programs, should simultaneously promote positive outcomes for youth and the community. The benefits seen across diverse youth and communities suggests the need to continually promote youth leadership as one way to promote community change that is based on youth needs.    49 COMMUNITY SHORTCOMINGS IN PROMOTING YOUTH LEADERSHIP Research on leadership abounds, but the research on youth leadership is sparse (MacNeil, 2006). It is likely this shortage of research on youth leadership is due in part to two reasons.  First, much of the leadership literature comes from the business and management world (Avolio, Walumbwa, & Weber, 2009).  Organizations are often the focus of leadership research, looking at patterns of leaders versus followers, and understanding what makes an within the workplace.  Since youth spend the majority of their time in a school setting and are not part of these workplace settings, they are naturally excluded from research in this arena. Next, currently there is not a consistent definition of youth leadership. A clear, youth-driven definition of leadership will improve efforts to engage youth in meaningful leadership roles and promote consistency across programs that aim to build youth leadership capacity and opportunities.   The following section will discuss shortcomings of our current efforts to promote youth leadership, including the need for more youth leaders, leadership programs, and adult support of youth leaders.  These shortcomings point to the importance of furthering our understanding of youth leadership and engaging youth in meaningful ways to increase the number of youth taking on leadership roles to create change within their communities.  COMMUNITIES NEED MORE YOUTH LEADERS. Opportunities for youth to be engaged in community change efforts are often limited by the separation of youth and community development programs (London, et al., 2003), and because adults are sometimes unsupportive of youth involvement or do not know how to engage youth in meaningful ways (Checkoway & Richards-Schuster, 2003; Frank, 2006).  This lack of involvement can cause three major problems.  First, by not engaging youth in change efforts, communities risk perpetuating power  50 differences between youth and adults, discouraging youth from current and future engagement and leadership (Stoneman, 2002). Next, communities working to address youth-focused issues without youth input may create changes that do not meet the needs of youth or target the issues that are important to youth (London, et al., 2003).  Finally, limiting engagement has caused a shortage of youth leaders in our country (Frank, 2006, London, et al., 2003), which could potentially lead to a future with less adult engagement and leaders. Thankfully, this trend of excluding youth voices is beginning to change. Communities and organizations are increasingly seeing the benefits of encouraging youth involvement to improve outcomes for youth and the broader community (Zeldin, Christens, and Powers, 2013) and have begun to actively reduce the power barriers between youth and adults, and engage youth in leadership roles and opportunities for authentic decision-making (National League of Cities, 2010). Engaging youth in decisions that affect their lives helps to ensure issues that are important to them are being addressed (Taft & Gordon, 2013), and research shows that youth engagement often leads to future adult engagement (Flanagan & Christens, 2011), which may address the shortage of youth leaders in some areas.   Aligned with the need to engage more youth leaders is a desire by youth for more opportunities to act as leaders and change agents. A number of studies of youth leaders point to the importance of creating change to these youth (e.g., Mortensen, et al., 2014; Christens & Dolan, 2011; Taft & Gordon, 2013). This research provides diverse perspectives from around the world, including primarily white, rural youth from one US state (Mortensen, et al., 2014), teenage girl activists from five cities across North and South America (Taft & Gordon, 2013), white, middle-class youth from Portland, OR (Taft & Gordon, 2013), poor students of color in Oakland, CA (Taft & Gordon, 2013), and youth activists in California (Christens & Dolan,  51 2011).  The consistency across groups of youth in demonstrating their desire to create change supports the argument that youth do want to be involved in their communities and can fill the need for more youth engagement that has been identified in the literature (Frank, 2006, London, et al., 2003).  For example, and youth social movements, found activist youth were dissatisfied with youth councils as opportunities to create real community change. These youth sought outlets where their voices could be heard and they had influence over decisions, such as youth social movements that existed outside a formal political system (Taft & Gordon, 2013). Another example comes from . Youth in their study demonstrated a desire to engage in creating community change through their voluntary participation in community initiatives to improve conditions for youth by actively identifying local problems and creating solutions to those issues. Similarly, in defining what it means to be a leader, Mortensen and discussions of leadership. Youth saw the purpose of leadership as creating change and improving their communities. Youth in these studies came from very different contexts, and were working on a variety of issues in their communities, but all of these groups of youth focused on the importance of creating change as a key part of leadership.  While a shortage of youth leaders may be evident in some programs or communities, youth have been leaders in other ways outside of leadership programs or adult-organized leadership opportunities. As Taft and Gordon (2013) found in their study, youth may seek leadership opportunities outside of formal programs or roles offered by adults because they are dissatisfied with the potential to create real change. Also, as social media becomes more and  52 more popular, youth have access to a much broader audience of peers to reach out to engage in grassroots community change efforts. From United We Dream network in recent years, youth activists have engaged in their communities to fight against oppression, often through grassroots efforts organized by youth (Costanza-Chock, 2012). These youth have played important leadership roles in mobilizing their peers and communities, often through social media channels, to create powerful networks and organizations that lead to positive community change (Costanza-Chock, 2012). One example of this comes from undocumented youth activists, such as those engaged in the United We Dream network, who staged sit-ins that eventually led to a national decision to defer deportations for undocumented youth (Costanza-Chock, 2012; Zimmerman, 2011). Another recent example comes from the Black Lives Matter movement, which youth have played a key role in through protests and spreading messages through social media (Chang, 2016).The civic activism of these youth aligns with Paulo Freirepromotion of critical consciousness in analyzing the surrounding social conditions of marginalized people and acting to create change (Hopkins, Diemer, & Voight, 2011). So, while communities may have a shortage of youth leaders in leadership programs or adult-organized settings, for decades, youth have taken on leadership roles through grassroots civic activism to address critical issues in their communities. The model of youth leadership described in this study focuses on areas that align with these types of leadership efforts. For example, the model includes collective action and creating change. Youth engaged in efforts such as the Black Lives Matter movement or the United We Dream network work together with other youth to advocate for change that will affect themselves, their peers, and the broader community. Though the study includes youth involved in a leadership training program, the Collective Change Youth Leadership Framework may be a viable model for youth in other  53 types of leadership roles as well. Empirically testing the model with the study sample can provide initial evidence for its applicability to youth, and future research could engage youth in these grassroots efforts to learn how the model aligns with their ideas about leadership and may suggest additional components in order to align with a broader range of youth leaders.  These ideas of having a voice in decision making and creating community change depart from some of the current youth leadership development efforts underway in the US, which primarily emphasize skill building and preparing youth for leadership roles in the future (MacNeil, 2006; Yu, Lewis-Charp, & Gambone, 2007).  By learning how youth think about leadership, communities will be better equipped to build youth capacity to engage in their communities to create change, and can help promote youth leaders in a way that is consistent with their views of what leadership is, and what leaders should do. This study will begin to fill this gap by testing an emerging leadership framework developed from youth ideas about leadership. The goal is to provide an empirically tested framework of leadership from a youth perspective to help inform the creation of youth leadership programs to ensure they align with youth ideas about leadership and provide supports that match these definitions.  Programs that align with youth viewpoints will be more successful in engaging youth to participate and encouraging them to use the skills they learn in their current and future lives. As research shows, when programs do not align with youth desires and interests, they often do not attend (Grossman, Price, Fellerath, Jucovy, Kotloff, Raley, and Walker, 2002). CRITIQUES OF YOUTH LEADERSHIP PROGRAMS. Youth programs and initiatives vary in their purpose, mission, setting, and activities, yet research has shown these diverse programs can promote positive outcomes for youth, help youth develop supportive relationships, encourage contribution to the community, and promote civic engagement despite their differences  54 (MacNeil, 2006; Independent Sector, 2002).  Though these programs have been beneficial to youth across the country, the research limitations discussed above and the program limitations described below suggest there is room for improvement to ensure programs meet the needs of  (Taft & Gordon, 2013; MacNeil, 2006).    One shortcoming of programs that center on leadership development is a future orientation to youth leadership, in which programs focus on preparing youth to take on leadership roles as adults rather than acting as youth leaders in their communities (MacNeil, 2006; Yu, Lewis-Charp, & Gambone, 2007). Adult leadership development efforts emphasize the importance of learning skills (ability) and practicing those skills (authority) in order to develop as a leader (MacNeil, 2006). In contrast, many youth programs do not provide meaningful or authentic opportunities to engage youth in leadership positions, but rather teach youth about leadership, suggesting they can become leaders when they are adults rather than currently as youth (MacNeil, 2006; Yu, et al., 2007). MacNeil (2006) advocates for the need to provide youth with such opportunities that have true consequences and impact in order to develop their leadership potential and voice, a common practice in adult leadership programs. This future orientation to leadership may stem from commonly held negative perceptions of youth and from the social position youth hold in US society. Research on youth is deeply rooted developing as a result of crisis (Erikson, 1968). Even today, though positive youth development is a dominant narrative in the youth literature, many adults still view youth as problems to be dealt with rather than valuable resources to a community (Checkoway, 2011; Finn, 2001; Roth, Brooks-Gunn, Murray, & Foster, 1998). As a result of these negative perceptions of youth,  55 leadership programs may view their role as developing leadership skills in youth who will use them when they move past the   Another reason leadership development efforts may adopt a future orientation to leadership is because of the social position youth occupy in our society. Traditional adult theories of leadership (e.g. trait, behavioral) emphasize authority and power (Komives & Dugan, 2010). These theories dominated early leadership literature, which often was based in business and managerial settings. Though many of the more contemporary theories of adult leadership are moving toward a collaborative, distributive view of leadership (Komives & Dugan, 2010), adults still wield more power and have opportunities to take on decision-making roles in society such as holding public office or voting. Youth hold little power or authority over their lives or community, so often do not have opportunities to practice their leadership skills in meaningful settings or include their voice in community efforts. As mentioned above(2013) study of youth councils found activist youth saw these councils as less effective at providing opportunities for creating community change because the councils often did not have any power to affect the larger community. expression, spaces for voice but without any real obligation for the adult representatives to 94). According to these youth, the councils provided a space to but the actions and ideas of the councils were of little consequence to creating change within the community (Taft & Gordon, 2013). Activist youth found efforts in the broader community, such as youth social movements, provided more authentic and meaningful change opportunities that were not limited by adults or political power structures (Taft & Gordon, 2013). While the authors included diverse perspectives from youth in a number of countries in the Americas, and engaged in ethnographic studies of these groups of  56 youth, their sample included a small number of activist youth whose experiences and viewpoints may not represent those of other youth leaders. Despite this limitation, according to the youth leadership literature, it is clear youth want to engage in their communities and act as real change agents (Mortensen, et al., 2014; Christens & Dolan, 2011; Taft & Gordon, 2013). As described above, youth can take on leadership roles outside of formal programs, through civic engagement opportunities in their schools and communities (e.g., Schwartz & Suyemoto, 2012; Mitra, et al., 2014; Ozer & Douglas, 2013; Chan, et al., 2014; Christens & Dolan, 2011). These studies demonstrate the potential for youth to act as change agents and have influence over decisions and changes in their schools and communities. Programs that focus on developing leadership skills may benefit from incorporating more deliberate opportunities for youth to actively work on issues that are important to them and use their leadership skills to affect change in their schools and communities.  Another shortcoming of these programs is they often promote engagement and taking on leadership roles within the program setting (Yu, et al., 2007) and do not always help youth translate the skills they learn in the program to the larger community.  While encouraging leadership roles within leadership programs is critical, it is insufficient to ensure youth act as leaders within their communities. Some emerging programs are beginning to help youth act as real change agents in their communities. The Community Anti-Drug Coalitions of America (CADCA) work with youth to build their leadership capacity and engage them in identifying community problems and developing plans to work to alleviate them (CADCA, 2009).  Youth collaborate with local substance abuse coalitions to address youth issues such as underage  57 action plans to create change.  Youth across the country who are engaged in these efforts have been successful in raising awareness about youth issues, engaging more youth in their efforts, , and helping remove prescription drugs frfor safe disposal of unused prescriptions to prevent youth from using pills that were not prescribed to them (CADCA, 2009).  A better understanding of how youth view their role as a leader could provide guidance to existing programs and help create new programs and settings that engage youth in leadership positions that are meaningful to them and create community change.  SUPPORTING YOUTH AS LEADERS. promote youth leadership is how adults support youth in both developing their leadership capacity and helping them engage as leaders within various settings.  A great deal of literature exists on the ways in which adults can support youth (Libby, et al., 2005; Denner, Meyer, & Bean, 2005; Mitra, et al., 2010; Zeldin, Christens, & Powers, 2013; Mitra, 2005; McLaughlin, 2000), but less empirical work exists on how adults can help facilitate and support youth leadership specifically (Mitra, 2005). Understanding how youth think about leadership may help inform how adults support youth in ways that will promote the leadership activities and skills that are important to them and align with their ideas about leadership.    Research on relationships between youth and adults is useful to draw on to understand how various ways of supporting youth can be beneficial.  Larson and Angus (2011) suggest different approaches to supporting youth, directive and facilitative assistance, can lead to different outcomes.  A great deal of research has explored directive assistance, which promotes youth engagement and ownership over the work, within an adult-driven setting (Larson &  58 Angus, 2011; Durlak & Weissberg, 2007; Roberts & Treasure, 1992).  A directive assistance approach is based on the belief that adults have more knowledge and experience than youth and can provide meaningful learning experiences for youth that are driven by adults (Larson & Angus, 2011).  -analysis of after-school programs found support for this directive approach, promoting the development of general personal and social skills in youth, and Larson and colleagues found adult-driven programs facilitated the development of specific talents based on the goals of the program (Larson, Walker, & Pearce, 2005).  Larson and Angus (2011) interviewed 108 youth in various programs (six leadership programs and five arts and media arts programs) to understand how their experiences in these programs affected their development.  They found in programs that used a directive approach, where advisors offered support through reminders about deadlines and prodded youth to keep them focused, youth gained mobilizing skills (Larson & Angus, 2011).  In contrast, in programs that incorporated a facilitative approach to working with youth, in which youth had primary control over the work and adults played a supportive role, youth developed strategic thinking skills (Larson & Angus, 2011). Earlier work on youth-driven programs, which give youth primary control over the work, found these types of programs developed youth as participants in the programs, helped them to learn more (Lansdown, 2001; Larson, 2000), empowered them to create change and use the skills they gained in settings beyond the programs they participated in (Edelman, et al., 2004), promoted youth ownership over the work, and developed leadership and planning skills (Larson, et al., 2004).        As the research has demonstrated, supporting youth in different ways can lead to different outcomes, and there should be a balance between giving youth autonomy and the chance to take on meaningful leadership roles, while still supporting their efforts and providing guidance  59 (Mitra, 2005; McLaughlin, 2000). More research exists on the directive approach to supporting youth because this has been more common in youth programs to date (Larson & Angus, 2011). The facilitative approach to supporting youth is worth further exploration as this approach seems to align with what we are beginning to learn about how youth think about leadership, as a collaborative endeavor that focuses on listening to others and representing the group (Mortensen, et al., 2014), which may be modeled by adults in a facilitative rather than directive approach to support.  The current study explored how different support approaches (directive and facilitative) in past leadership experiences relate to youth definitions of leadership.  Youth that experience these various relationships may shape their ideas about leadership based on how adults support them as they model behaviors to youth.  Programs that promote adult-directed models of leadership may lead youth to see leadership as more directive and hierarchical, and possibly less accessible to them, than youth engaged in programs in which adults give youth autonomy and promote a less hierarchical approach to their work.  EMPIRICAL INVESTIGATION OF THE COLLECTIVE CHANGE YOUTH LEADERSHIP FRAMEWORK The Collective Change Youth Leadership Framework was developed to understand leadership is a) available to anyone in any context; and involves b) creating change, c) collective action, d) modeling and mentoring, e) a strong character, and f) motivation and ambition.  To create the Collective Change Youth Leadership Framework, the author and a team from Michigan State University conducted two initial studies to begin to understand how youth define leadership and to develop a youth-informed framework. The first study engaged 130 youth (78 females, 49 males, average age 14.95, 90% white) from across one rural state who were involved in a year-long leadership program. As part of the program, youth participated in an online Photovoice  60 project that prompted them to take photos and write narratives about their ideas of leadership (See Lichty, Foster-Fishman, Kornbluh, Mortensen, Pollard, & Hockin, 2011, and Mortensen, et al., 2014 for a description of this study).  Youth posted their photos and narratives to an online blog, and engaged in group discussions in that space.  The second study engaged 44 youth (demographic data is not available for youth who generated statements about leadership) who were attending a week-long leadership conference in a concept mapping process in which they generated statements about their definition of leadership.  Both initial studies prompted youth with the same framing questions, What makes someone a leader? What does leadership look like?. In total, 174 youth participated in these studies, contributing their ideas about leadership.  Grounded Theory (Charmaz, 2003; Glaser & Strauss, 1967) was used to explore the dominant ideas that emerged across these studies and a multidimensional framework of youth leadership, the Collective Change Youth Leadership Framework, was created. There was strong converging evidence from these two studies that according to youth, leadership involves six dimensions. The purpose of this study is to empirically test the Collective Change Youth Leadership Framework to understand if the proposed six dimensions represent leadership for a national sample of youth and to explore the extent to which different characteristics and experiences relate to how youth think about leadership.  COLLECTIVE CHANGE YOUTH LEADERSHIP FRAMEWORK  The Collective Change Youth Leadership Framework includes six dimensions that emerged from the Photovoice narratives and concept mapping statements generated in the studies described above (Mortensen, et al., 2014). The ideas that came through in these initial studies were compared to a variety of youth leadership definitions and program components to see how they were similar to what the literature currently says about youth leadership and to add  61 additional ideas that aligned with the youth-defined dimensions. Table 5 below includes a definition of each dimension and the ideas that came from the initial youth studies that fall within each dimension. Then, to illustrate how these ideas connect to current youth leadership literature, examples from the literature are provided to show their alignment with this framework.  Below is a description of the six dimensions of the Collective Change Youth Leadership Framework as they emerged in the initial studies and how the literature on youth leadership supports these dimensions. Following the presentation of this framework, common adult theories of leadership will be discussed, and the Collective Change Youth Leadership Framework will be compared to those adult leadership theories.  Anyone can be a leader at any time. The first dimension of the Collective Change Youth Leadership Framework, anyone can be a leader, includes the notion that all youth have the potential to be leaders and can take on any number of roles to act as a leader.  According to one youth in the initial studiesway. I believe that if you wanchange lives and make differences in your community or state or maybe even the country or who knows the world even, thedo (quote from Photovoice participant).  Research shows youth are often more inclusive than adults (The Pew Research Center, 2007), which may shape their view of leadership as available to anyone who is willing to take on the responsibility and not only to a select few that are born with leadership potential or who demonstrate specific traits.   People can make a difference and be extraordinary leaders from a captain of a team to a  62  (quote from Photovoice participant). This quote exemplifies these ideas, including both the fact that anyone can be a leader, and youth can take on various roles in their communities to make a difference. Overall, this suggests a youth belief that leadership is not limited to a select few, but rather available to all youth who are interested in becoming leaders.   The idea that anyone can be a leader is apparent in the youth leadership literature, as current theories and research on positive youth development suggest all youth can benefit from engaging in their communities in a variety of roles, including taking on positions of leadership (Van Linden & Fertman, 1998; Fertman & Van Linden, 1999; MacNeil, 2006). Reviews of prior youth leadership efforts suggest the belief that anyone can be a leader is an implicit value or norm, with these programs working to promote leadership across diverse youth and building their skills so they can successfully take on leadership roles (Van Linden & Fertman, 1998; Fertman & Van Linden, 1999; Edelman, et al., 2004). Creating Change. The second component of the Collective Change Youth Leadership Framework is creating change. This dimension of the framework is outcome focused and related to accomplishing goals.  According to the youth involved in the initial studies, the role of a leader is to recognize problems and work to improve them. Youth explained, like someone taking a stand and making a change (quotes from Photovoice participant). Caring for and helping others is also a key component of this dimension as both an important action that leaders take and a motivation for creating change.  Recent literature and the media make it clear youth are interested in improving their communities, for example working on issues that improve the environment, and want to take an  63 active role in making their world a better place (Ginwright & James, 2002; Chawla, 2002; Pittman, 2000).  This theme is consistent with definitions of youth leadership adopted by some researchers and practitioners. tion of youth leadership includes a focus on a social issue and taking action to create change. Similarlyyouth leadership programs found programs emphasized developing specific skills to build their capacity to create strategic plans to change their communities and programs and communicate a vision. Finally, ability to analyze their communities and create action plans in response to those identified needs to create community change (CADCA, 2009). The ideas within this theme also align with 2014) conceptualization of critical consciousness in that youth discussed both critical reflection in identifying problems and critical action in taking steps to address those problems. Though not all scholars of youth leadership have defined youth leadership as focusing on creating change, this emphasis on creating change across some leadership programs and definitions is consistent with youth conceptualizations of leadership, suggesting programs and communities must them achieve their leadership goals.  Collective Action. The next dimension of the Collective Change Youth Leadership Framework defines leadership as rooted in collective action.  As one youth describedleadership is people coming together, hand in hand, and addressing the problems in their (quote from Photovoice participant). Not only do youth see their role as collaborating with others, but they see collaboration as necessary to achieving their goals and creating change. Youth standpoints based on their social location may explain this collective  64 approach to leadership because they lack formal authority roles that adults have access to (e.g. elected government positions, management roles in organizations), and youth often work collectively (e.g. through community organizing or coalitions) to create change.  Youth see change as requiring collective action by a group, and do not define leadership as holding (quote from Photovoice participant). Youth do not perceive a lack of formal power as a barrier to acting as leaders, instead, they expand leadership to include people who come together to actively pursue a goal.   Yu and colleagues (2007), talked about leadership with youth in their evaluation of a leadership development initiative, and found similar ideas about collective action.  Though they do not provide an in-depth analysis of the themes that emerged around the What does it me, the data they share does align with this component in the Collective Change Youth Leadership Framework. For example, one youth in their study explained, your own (Yu, et al., 2007).  Additional ideas cooperating with others, building consensus, and resolving conflict (Yu, et al., 2007; Edelman, et al., 2004). These ideas were explored in the current study in addition to those suggested by youth in the Photovoice and concept mapping studies.   Research on youth leadership programs describes collaborative experiences and teamwork as a core component of these programs, and has found some of the outcomes associated with these programs focus on the ability to share power and distribute tasks and work as a team (Edelman, et al., 2004).    65 Modeling and Mentoring.  According to the Collective Change Youth Leadership Framework, leaders provide guidance to others through mentoring and modeling rather than will help point you in the  (quote from Photovoice participant).  Additional ideas within this dimension include leaders leading by example, being someone that others look up to, and not being forceful in their leadership role. Youth do not focus on maximizing control over others, but rather encouraging others to lead as well.  This de-control over them such as at school or at home, so in their role as leaders, they use a more supportive approach of mentoring and modeling to transmit skills to other youth.  Another key idea within this theme is an awareness of youth leaders responsibility to their current situations as well as future generations.  One youth behind.  This process will keep going on the generations to come but each leader will add  (quote from Photovoice participant).  Youth leadership research also supports these ideas of modeling and mentoring as a key dimension of leadership.  Some of the skills that are commonly developed in youth leadership fluence others (Edelman, et al., 2004).  It is possible youth learn to model and mentor, and value that as a critical component of leadership if they are involved in facilitative relationships with adults (Larson & Angus, 2011).  dy found younger participants often looked to older youth to learn how to be a leader. When adults model behaviors but do not force their ideas or opinions on youth or take a directive approach to leadership, youth may internalize those behaviors and  66 incorporate them into their views of leadership and enact similar behaviors in leadership roles they take on. One idea that is less evident in the current youth leadership definitions is the responsibility of leaders to pass on their skills to others which was an idea youth discussed in the Photovoice study. The relationship between past adult support in leadership roles and youth definitions of leadership was explored to further our understanding of the factors that relate to how youth view leadership.   Strong Character. The idea of having a strong character is the fifth dimension of the Collective Change Youth Leadership Framework.  Within this dimension, someone who thinks about others before themselves, and always acts in the best interest of their (quote from Photovoice participant). Some of the specific character attributes included in this dimension are: responsibility, patience, respectfulness, honesty, and loyalty.   Some youth leadership research discusses the importance of developing strong character traits. includes facilitating the development of ethics, values, and ethical reasoning as one of the core principles, which does align with yout Yu and colleagues (2007) suggest the ability to empathize is also critical for leaders.  The idea of empathy was added to the current study because it was not initially discussed by youth in the Photovoice or concept mapping studies. More recently, one of s developing a moral compass/character, and includes respect, responsibility, helpfulness, and working hard as specific measures of this indicator (Benson & Scales, 2009).  The alignment of the literature , suggests the importance of developing this dimension within the youth-driven framework to help youth become effective leaders.    67 Motivation and Ambition. Within the motivation and ambition dimension, leaders are confident, motivational, focused, ambitious, and determined. These characteristics are important to the other dimensions of leadership as well.  For example, youth must be motivated and ambitious in order to work to create change within their communities.  Also, modeling and mentoring requires time and patience, which would suggest youth are motivated to share their skills with others through their mentoring roles.  Youth leadership development programs often motivate others (Edelman, et al., 2004).  Developing this as a skill within youth may help them to encourage others to work collaboratively with them to achieve a goal.  Being able to motivate others is an important component of youth leadership because of the collective nature that youth engage in leadership roles.  Without others to support their efforts, youth may not be as successful in achieving their change goals.    Summary of Collective Change Youth Leadership Framework. Overall, these two initial studies conducted by Michigan State University (MSU) began to develop our understanding of what leadership means to youth, but they are not without their limitations.  The Photovoice study included youth from one state, and 90% of youth were White and from rural communities.  Though they did have various amounts of past leadership experience, the group was quite similar overall in terms of their demographic backgrounds.  The concept mapping study included a small sample size (n=44), and demographic information was not collected at the time of the initial data collection when youth generated statements about what leadership means to them. Other studies have touched on youth definitions of leadership through interviews (Roach, et al., 1999; Cassell, et al., 2006) and organized activities (Culp & Kohlhagen, 2000), often as part of a larger study, but developing a framework of youth leadership was not the ultimate goal of these projects.   68 These studies faced similar sample limitations as the MSU studies. Culp and Kohlhagen (2000) included a small sample of primarily white, rural youth, and Roach, et al. (1999) and Cassell, et al. (2006) did not report specific demographics for the youth they engaged in their work. Without an understanding of who those youth were, or a clearly defined youth leadership framework, it is difficult to draw conclusions about how youth define leadership and generalize their work.  The limitations of these studies suggest the need for a larger, more diverse sample of youth, a youth-defined framework of leadership, as well as a more general population of youth.  This study addressed the first two limitations by including a larger more diverse sample than some of the prior studies (e.g. Mortensen, et al., 2014; Culp & Kohlhagen, 2000) as well as a clearly defined framework of leadership that was informed by youth. This study did not address the third limitation of including a more general population of youth because it sampled participants from a leadership conference. So, though the group was more diverse and larger than some of the previous studies, and did report demographic data for the sample, it still includes a convenience sample of youth and may not be generalizable to the broader youth population. The purpose of the present study is to expand the current understanding of how youth define leadership by empirically testing the emerging framework of youth leadership with a national sample of youth. ADULT THEORIES OF LEADERSHIP  It is essential to examine some of the many adult theories of leadership in a discussion of youth leadership, as these theories have informed some of the youth leadership development efforts to date (MacNeil, 2006). The Collective Change Youth Leadership Framework identifies six important dimensions of leadership defined by youth, some of which align with contemporary adult theories of leadership (e.g. Servant, Social Exchange Model) while others are in stark contrast to more traditional theories (e.g. Trait, Power-Influence). This section will  69 provide an overview of the vast, evolving adult leadership literature as well as a discussion of how the emerging youth framework compares to adult theories (See Table 2 for an overview of the adult leadership theories). Extensive research on leadership exists, and this subject has been studied for decades.  As our sociopolitical context has changed, so too have our theories of leadership. Komives and Dugan (2010) suggest a distinction between traditional and contemporary theories of leadership, which reflects the evolution of this literature. Early research on leadership viewed a leader as a  an individual or a small group holding power and influence over others (e.g. Power-Influence Theory).  There was a clear distinction between leader and followers, which allowed only a few individuals to take on leadership roles.  Other dominant early theories included the Trait and Great Man theories, which saw leadership effectiveness as dependent on a leader possessing specific qualities or characteristics (Stogdill, 1948; Mann, 1959).  As the field evolved, scholars began to explore the role of followers as well, and theories such as Behavioral, Contingency, Situational, Path-Goal, and Leader-Member Exchange (LMX) emerged (Komives & Dugan, 2010).  For example, Behavioral theory posits that leaders perform certain actions (Halpin & Winer, 1957), and LMX theory says leadership is dependent on the relationship between leaders and followers (Grean & Cashman, 1975).  These theories continued to distinguish between leaders and followers, but placed less emphasis on specific traits as required for leadership.  The leadership field began to shift to explore a more collaborative and reciprocal approach to leadership, which introduced a number of contemporary leadership theories. Komives and Dugan (2010) saw this shift in theory as a response to the philosophical shift within our country that began to explore social constructivism, postmodernism, critical theory, and  70 feminist theory. These theories included groups whose voices have been excluded and began to explore power dynamics that perpetuate this marginalization. Theories that fall within this group include: Transformational leadership, which focuses on the relationship between leaders and followers, aiming to elevate followers to become leaders themselves, and includes charisma, inspirational motivation, intellectual stimulation, and individualized consideration (Bass, 1985; Bass, 1999; Bass & Avolio, 1993). Servant leadership, which sees the role of leaders as serving takes a distributive approach to leadership and minimizes power differences to work collaboratively to achieve goals (Komives, Lucas, & McMahon, 2009), and Social Exchange Model, a non-hierarchical, values-based, and collaborative approach that focuses on creating positive change (Astin & Astin, 1996). Similarly, Team leadership within organizations focuses on the process of leadership in which leaders help team members develop individually and as a team using task specific learning experiences to encourage ownership over their own work and leadership in that role (Horner, 1997; Kozlowski, Gully, Salas, & Cannon-Bowers, 1996). In addition, Authentic leadership centers on authenticity and self-reflection, and focuses on developing follower capacity (Avolio & Gardner, 2005), and Complexity theory, which suggests leadership is a complex dynamic that comes from a need for change, and strives to create system-level outcomes (Uhl-Bien & Marion, 2007). Overall, these contemporary theories begin to address power differences and include traditionally excluded individuals and groups, and many focus on creating change. Though the more traditional theories have many similarities to youth-driven ideas about leadership, many are still based within organizational settings rather than the community, and have focused on adult leaders, so likely are not fully aligned with youth views of leadership.    71 How the Collective Change Youth Leadership Framework relates to Adult Theories.  The Collective Change Youth Leadership Framework suggests youth definitions of leadership do not align with traditional adult theories of leadership that focus on hierarchy and having power and using it to influence others.  While youth do discuss ideas about taking a stand and creating change, which may include influencing others, they focus on shared power and influence rather than hierarchical relationships between leaders and followers.  On the other hand, many of the ideas across the contemporary adult theories of leadership align well with what we are beginning to learn about youth leadership. While there is alignment with contemporary theories, it is still important to learn from youth directly what leadership means to them and create a theory of leadership based on those ideas.  In comparing youth ideas to traditional theories of leadership, youth did not emphasize having specific inherent traits, power and authority, or relying on followers to be a leader.  Instead, youth saw leadership as an inclusive opportunity that is available to anyone who is motivated to engage as a leader and create change.  Also, youth saw leadership as a collective effort rather than an individual, power-based endeavor, which is common in the traditional theories of adult leadership.  Finally, traditional theories of leadership do not focus on creating positive change as a key goal of leaders, as the youth in these studies discussed.   Moving on to contemporary theories, the ideas put forth by these theories align well with the various ideas about leadership youth have begun to define. One contemporary theory includes a number of the dimensions described by the youth framework in this study. The Social Exchange Model, emphasizes a non-hierarchical approach to leadership that focuses on working collaboratively to create positive change (Astin & Astin, 1996). The Collective Change Youth Leadership framework aligns in a number of ways, such as the belief that anyone can be a leader,  72 focusing on values, working collaboratively, and aiming to create positive change. Other contemporary theories also show alignment with the proposed youth leadership framework. For example, the Shared/Relational theory aligns n of collective action, suggesting a non-hierarchical approach to leadership. Similarly, an important aspect of the Team leadership theory is developing teams in a way to promote collaborative success. In addition, nd their focus on helping people is consistent with values-driven adult theories such as Servant leadership. Youth also see leadership as available to anyone. Relatedly, a key component of Transformational leadership is the relationship between leaders and followers and building follower capacity to help them become leaders themselves. Finally, youth discussed modeling and mentoring as a key role that leaders play so they can develop future leaders. Authentic leadership theory, Transformational leadership, and Team leadership align with this mission, with an emphasis on growing new leaders and developing follower capacity (Komives & Dugan, 2010).   Table 2. Selected Adult Leadership Theories* and the Collective Change Youth Leadership Framework Theory Name (citations) Komives & Dugan (2010) Categorization Example Characteristics Alignment with Collective Change Youth Leadership Framework Trait  (Stogdill, 1948; Mann, 1959) Traditional  Leadership effectiveness as dependent on specific qualities and characteristics (e.g. demographics, intelligence, extraversion, dominance, assertiveness)  Clear distinction between leaders and followers    Not aligned  Leadership is not dependent on inherent individual traits       73     Theory Name (citations) Komives & Dugan (2010) Categorization Example Characteristics Alignment with Collective Change Youth Leadership Framework Power-Influence  (French & Raven, 1959) Traditional  Leaders have power and authority over others and situations  Power is held by one or a few Not aligned  Leadership is not solely defined by individuals or small groups with power and influence Behavioral  (Halpin & Winer, 1957) Traditional  Leaders perform certain actions  Often categorized into Consideration behaviors (e.g. concern for followers) and Initiation of Structure behaviors (e.g. using standard procedures) Somewhat aligned  Leaders do perform specific behaviors (e.g. creating change, working collaboratively), but behaviors are not the only component of leadership  Leader-Member Exchange  (Graen & Cashman, 1975) Traditional  Leadership is dependent on the relationships between leaders and followers  Achieving outcomes depends on effective mutual relationships  Not aligned  Leadership is not dependent on followers, and is less hierarchical than traditional adult theories  Servant  (Sendjaya & Sarros, 2002) Contemporary  Leaders serves others and of their own Aligned  Leaders put before their own Shared/Relational  (Komives, Lucas, & McMahon, 2009) Contemporary  Distributive approach to leadership, reduces power differences  Focuses on working collaboratively to achieve goals Aligned  Leadership is non-hierarchical and is focused on working together to achieve goals           74     Theory Name (citations) Komives & Dugan (2010) Categorization Example Characteristics Alignment with Collective Change Youth Leadership Framework Social Exchange Model of Leadership Development  (Astin & Astin, 1996) Contemporary  Non-hierarchical, values-based, collaborative approach to leadership  Focuses on creating positive change Aligned  Leadership is a collaborative endeavor that is focused on creating positive change Team Leadership  (Horner, 1997; Kozlowski, et al., 1996) Contemporary  Help team members develop individually and as a team  Encourage ownership over work and leadership in their role Aligned  Leaders mentor others to develop their skills  Anyone can act as a leader in various positions Transformational Leadership    Contemporary  Leaders work with others to help them develop their leadership skills  Leaders motivate and individualize their efforts Aligned  Leaders build capacity to help others become leaders Complexity  (Uhl-Bien & Marion, 2007) Contemporary   Leadership is a complex dynamic that results from a collective need for change which comes from organizational interactions   Strives to create system-level adaptive outcomes (e.g. increased innovation, creativity, learning) Somewhat Aligned  Leaders work to create change, but not specifically system-level change *Table adapted from Mortensen, et al., 2014. Overall, the emerging youth-informed leadership framework is a departure from traditional theories of leadership that emphasize having specific traits, being born a leader, or having power over others (Stogdill, 1948; Mann, 1959; French & Raven, 1959), yet shows a great deal of alignment with contemporary leadership theories. This alignment is not surprising based oginalized position in society. Contemporary theories of leadership have begun to consider marginalized groups, but not youth specifically, as they develop in response to  75 the sociopolitical shifts in our country around power and voice (Komives & Dugan, 2010).  Despite the alignment between contemporary adult theories of leadership and what we are beginning to learn about how youth define leadership, no single contemporary theory captures the six dimensions of leadership put forth by youth and these theories were created for adults in primarily organizational contexts. These theories can provide important insight into youth leadership given the decades of adult leadership research that exists. However, gaining a clearer understanding of different youth perspectives on leadership and creating a youth-defined theory or framework will move the field forward and promote better support in developing youth leaders.  VARIABLES CONTRIBUTING TO YOUTHS DEFINITION OF LEADERSHIP. This study explored the differences between youth experiences and viewpoints in how they define leadership so communities can effectively engage and encourage youth to develop as leaders now and in the future. Five variables were included to understand how they may relate to the way in which youth define leadership.  These variables were: background characteristics, adult support, critical consciousness, empowerment, and civic engagement.  Background characteristics. Feminist standpoint theory suggests a number of demographic variables that influence how individuals construct their realities (Campbell & Wasco, 2000; Intemann, 2010; Wylie, 2003).  According to this theory, it is reasonable to suggest that demographic variables such as race, gender, and class may influence how youth define leadership.  MacNeil (2006) argues against a linear model of leadership development that is age dependent, suggesting the importance of considering other factors that could influence leadership such as experience and context.  Though she does not explicitly address the suggested demographic characteristics as influential in leadership development, her argument can be  76 generalized to explore these individual differences as possible contributing factors to youth conceptualizations of leadership.  Beginning early in life, children experience gender differences in socialization that impact their behaviors and attitudes.  From early theories of sex differences based in biology to theories about social influences of gender, researchers have put forth numerous theories to explain how children develop their gender identities. For example, social cognitive theory of gender development considers both the cognitive construction of ideas about gender and the social influences from various environments, including the family, peer groups, and educational settings (Bussey & Bandura, 1999). Social learning theory suggests children learn sex appropriate behaviors through rewards, punishments, modeling, and vicarious learning (Mischel, 1970), while biologically oriented theories explain gender role development through the different biological roles males and females play in reproduction (Buss, 1995; Trivers, 1972).   Observational learning is an important component of social cognitive theory, and is affected by four processes (Bandura, 1986). First, attentional processes explain what is selectively observed by an individual. Based on cognitive preconceptions, girls and boys pay attention to certain aspects of modeled behavior, generally focusing on same sex models, and observe the social reactions to those behaviors to help inform their mental models of gender appropriate behavior (Bussey & Bandura, 1999). Retention processes explain what observed behaviors are retained, which is influenced by processing the information learned from the models and creating a set of rules that relate to behaviors (Bandura, 1986). Retention is improved elf-efficacy to perform the modeled behaviors (Bandura, 1986). Once a conceptual model is developed, production processes allow individuals to use their models of gender appropriate behavior to  77 guide their actions. These mental models must be a general set of rules so that behaviors can be modified as needed to match the setting in which they are engaged (Bandura, 1986). Finally, motivational processes determine what is performed versus what is learned. People do not perform everything they learn, and behaviors are regulated by direct and vicarious experiences. Behaviors are reinforced if they are met with reward or acceptance, while behaviors that are unrewarding or negatively consequential are generally discontinued (Bandura & Barab, 1971; Hicks, 1968). Applying these processes to youth leadership development, it is possible that girls and boys attend to the leadership behaviors of same sex models and learn what are considered appropriate leadership actions for their gender. Research has shown that women in leadership roles within organizations (e.g. managers) that demonstrate certain leadership behaviors that are typical of men, such as exerting control or dominance over others are often evaluated more negatively than men (Eagly & Karau, 2002; Eagly, Makhijani, & Klonsky, 1992). Because of the negative perceptions and potential for being passed over for hiring and promotions (Burgess & Borgida, 1999; Rudman & Glick, 2001), female leaders tend to demonstrate behaviors that are more consistent with the typical female gender role such as being more collaborative, communal, and less hierarchical as compared to men in similar roles (Eagly, Johannesen-Schmidt, & van Engen, 2003; Hal & Friedman, 1999). Youth may perceive these differences between male and female leadership styles and learn what is typical behavior of a successful adult leader. From these observations, youth can develop that define what leadership looks like for boys and girls. Given opportunities to use these rules to guide their behavior, youth learn how to successfully adapt the leadership behaviors they have observed to various settings. behaviors in various settings may determine what actions youth perform that they have observed in others.   78 Social cognitive theory points out the importance of the social environment in developing an understanding of the differences between boys and girls.  Gender socialization is impacted by parents, siblings, friends, and the media, to name a few (Ruble, Martin, & Berenbaum, 2006; Epstein & Ward, 2011; Blakemore, Berenbaum, & Liben, 2008; Witt, 1997; Leaper, Anderson, & Sanders, 1998; McHale, Crouter, & Whiteman, 2003; Martin, Ruble, & Szkrybalo, 2002; Mischel, 1966). Even brain research shows that gender differences that are attributed to hormones in the brain (e.g. cognitive abilities) are not permanent and are influenced by socialization (Jordan-Young & Rumiatipreschool children explored how -segregation in selecting same-sex had the largest impact on who children decided to play with, bigger than the effect of gender-typed activities (Martin, et al., 2013). The authors go on to discuss the implications of sex segregation early in life; one view is that girls and boys develop within separate cultures, which provide different social experiences, styles of interaction, and opportunities for skill . According to social cognitive theory of gender development (Bussey & Bandura, 1999), peer socialization of gendered characteristics within same-sex peer groups, reinforcement of gender typical behaviors, and sanctioning of deviations Ewing & Troop-Gordon, 2011, p. 90). The cognitive construction of gender begins early in life and persists throughout as social influences reinforce or reject attitudes and behaviors (Bussey & Bandura, 1999).   -efficacy (Bandura, 2001). critical for subsequent aspirations and actions (Bandura, Barbaranelli, Caprara, Pastorelli, 2001; Bandura, 2001; Lips, 2000; Lips,  79 2001). -efficacy through witnessing others in positions they may aspire to (Beaman, Duflo, Pande, & Topalova, 2012; Rudman & Phelan, 2010aspirations and educational attainment found that in those villages that had female leaders, the gender gap related to career aspirations was reduced and the gap related to educational success was eliminated. These findings suggest the female leader may have provided a strong role model to girls and increased their sense of self-efficacy to succeed. As gendered cognitions drive behaviors, boys and girls differentiate their roles in all areas of their lives. Social cognitive theory may lend explanation to potential differences in the way boys and girls think about leadership. As they develop in to adolescents, they have more opportunities to act as leaders and their cultural surroundings suggest appropriate behaviors, modes of interaction, and opportunities to take on various roles in their lives based on their gender. This may happen through direct action, learning firsthand how their actions are perceived, or vicariously through watching adult leaders act in gender specific ways, further strengthening their leadership beliefs and actions.  Research has suggested that the way boys and girls interact is quite different as well, with boys being more rough and active with other boys (Di Pietro, 1981; Fabes, Hanish, & Martin, 2003) and girls acting more cooperative in their play with other girls (Maccoby, 1990). These sex differences in socialization and the preference for same-sex peers have an impact on youth as they continue to develop and differences between the sexes become stronger. Research shows that the preference for same-sex peers persists throughout childhood and adolescence (Bukowski, et al., 1993; Schofield, 1981; Hoffman & Powlishta, 2001; Ruble, et al., 2006; Mehta & Strough, 2009), which impacts how youth develop their preference for various activities and their attitudes and behaviors. Extending these patterns to youth leadership, gender socialization may then also  80 impact how youth think about and enact leadership, with girls enacting a more cooperative and collaborative style to leadership, and boys a more active and engaged approach. Social cognitive theory would suggest these gender differences are reinforced and rewarded by acting in a way that aligns with gender typical behaviors, and yleadership would then align with those behaviors that are rewarded. To date, little is known about how youth define leadership, and fewer studies have explored gender differences in these definitions. Cassell and colleagues (2006) did consider gender differences in online language use during the Junior Summit. They found some differences in language use between girl and boy delegates, such as girls using more apology words and social niceties than boys, and boys synthesizing the ideas of others more than girls (Cassell, et al., 2006). Though their study provides a starting point for understanding how girls and boys may differ in their ideas about leadership, there is still a gap in the literature surrounding this topic, and other studies of youth definitions of leadership have not specifically considered gender differences (e.g., Mortensen, et al., 2014; Culp & Kohlhagen, 2000). The current study will begin to add to our understanding of differences between girls and boys by exploring differences in how they define leadership according to the proposed framework. Specifically, it is hypothesized that girls will value a more collective approach to leadership than boys. Past research suggests women in leadership roles are viewed more positively when they exhibit characteristics aligned with the female gender role (e.g., collaborative, communal, less hierarchical; Eagly, Johannesen-Schmidt, & van Engen, 2003; Hal & Friedman, 1999) so young women may see female leaders demonstrating these collective behaviors and incorporate them into their views on leadership. Similarly, research on young children and adolescents also suggests girls demonstrate more collaborative behaviors in play (Maccoby, 1990) and associate  81 with same sex peers (Bukowski, et al., 1993; Schofield, 1981; Hoffman & Powlishta, 2001; Ruble, et al., 2006; Mehta & Strough, 2009), so these interactions may influence their thoughts on leadership and encourage collaboration. Exploring these differences may point to different ways in which girls and boys view leadership and value various components of the model.  In addition to exploring gender differences, age was also considered to understand how it may relate to leadership definitions. Developmentally, youth are quite different at age 12 versus age 20. Adolescence is a time of rapid growth and change (Eccles & Harold, 1993), and as youth get older, they may have more sophisticated ideas about leadership. Older youth may increase their social and political awareness, and feel compelled to take action on issues they were unaware of earlier in adolescence (Flanagan & Levine, 2010). Older adolescents may also have more opportunities to engage in different ways, for example in the political process by voting once they are 18, which may also affect their views of their communities (Flanagan & Levine, 2010; Flanagan & Sherrod, 1998; Sears & Levy, 2003; Niemi & Hepburn, 1995), and could influence how they think about leadership. In addition, in the few studies that have explored youth definitions of leadership (e.g., Mortensen, et al., 2014; Culp & Kohlhagen, 2000), differences across younger and older youth have not been explored. Understanding how youth of various ages define leadership could help tailor programs to more closely align with youth across the developmental spectrum to ensure they better meet their needs. For these reasons, this study explored how age may relate to how youth define leadership.  Adult Support. Adults can take different approaches to supporting youth in their leadership work, from being more controlling and directive to being more facilitative and behind the scenes (Larson & Angus, 2011; Grossman, Campbell, & Raley, 2007; Durlak & Weissberg, 2007). As discussed previously, supporting youth in different ways can lead to different  82 outcomes. Working with adults and seeing them model behaviors to youth may influence how youth think about leadership. Research suggests that youth who have experienced more facilitative adult support, in which youth and adults work more collaboratively and give youth more autonomy in driving their work, feel empowered to create change and use the skills they developed in settings beyond the programs they were involved in (Edelman, et al., 2004). These youth also felt ownership over their work and developed leadership and planning skills (Larson, et al., 2004). Given past research in this area, it is hypothesized that youth who have had more facilitative support may see Modeling and Mentoring  as an important part of leadership  since the adults they have interacted with likely modeled leadership behaviors to them and acted as mentors as they pursued their leadership goals.  One relationship that was explored is how youth have been supported by adults to see how facilitative relationships and directive relationships relate to Learning from the adults that support them in their leadership efforts may translate into how they think about leadership, and understanding how about leadership could inform future programming to ensure adults provide support in ways that align with how youth define leadership.  Critical Consciousness. Critical consciousness (CC) is a concept developed by Paulo as critical reflection about and taking action on societal inequalities (Freire, 1993; Diemer, et al., 2014). Community Psychology values critical consciousness as a key tenet in the field and is often viewed as a precursor to creating change (Prilleltensky, 2012; Checkoway & Richards-Schuster, 2003; Foster-Fishman, Nowell, Deacon, Nievar, & McCann, 2005). Critical consciousness is an in-outside forces (e.g. politics, power structures, society) affect individuals (Prilleltensky, 2012;  83 Diemer, et al., 2014; Watts, Diemer, & Voight, 2011). CC is often conceptualized as two components: critical reflection, including both perceived inequality and egalitarianism, and critical action (Diemer, et al., 2014; Prilleltensky, 2012; Watts, et al., 2011).  Critical Reflection includes both a critical analysis of perceived social inequalities related to race, ethnicity, gender, etc. and how those may impact opportunities related to education/occupation (perceived inequality), as well as the belief that all people should be equal (egalitarianism; Diemer, et al., 2014). Critical Action includes individual or collective action to create sociopolitical change (Diemer, et al., 2014). Once individuals develop a deeper understanding of their worlds, they may feel empowered to take action to change unjust conditions in their lives and the lives of others.        Critical consciousness has been associated with a number of positive individual and community level outcomes and impacts, many of them similar to those associated with youth leadership.  CC has been linked to positive mental health in urban adolescents (Zimmerman, Ramirez-Valles, & Maton, 1999), academic achievement and engagement for urban African Connor, 1997) and Puerto Rican youth (Ramos-Zayas, 2003), and positive career outcomes (Chronister & McWhirter, 2006; Diemer, 2009).  At the community level, CC is related to promoting change through community organizing, positive youth development, and social action efforts (Diemer, et al., 2014; Christens & Dolan, 2011).  How youth engage in their communities and develop their critical consciousness may relate to the way they define leadership, as their experiences may shape what they view as important to being a leader. In Foster-Fishof a youth leadership program, they found social justice orientations grew throuparticipation in the training, and the more these orientations grew, the more youth became  84 involved in their communities. In addition, early research on youth definitions of leadership suggests youth prioritize helping others and creating change (Culp & Kohlhagen, 2000; Mortensen, et al., 2014) as important aspects of leadership.  When youth develop their critical consciousness through their participation in youth leadership programs (Foster-Fishman, et al., 2012), and develop their understanding of inequalities and the need for change within communities (Diemer, et al., 2014; Zimmerman, 1995; Wang & Burris, 1997; Checkoway & Richards-Schuster, 2003; Foster-Fishman, et al., 2005), CC may influence the way they define leadership.  As youth become more aware of social conditions in their communities and develop their ideas about leadership, they may become more motivated to take action as leaders within their communities. In one study, Diemer and colleagues (2014) found an unexpected negative relationship between egalitarianism and action (though did find a positive relationship between perceived inequality and action). This unexpected negative relationship may suggest the perception of social inequalities (e.g., how SES/race/gender impact opportunities) motivates individuals to work to create change, but individual egalitarian views (e.g., the belief that all people should be treated equally) do not have the same effect (Diemer, et al., 2014). Perhaps experiencing or witnessing social inequalities is more motivating than simply holding egalitarian views, and has a more positive impact on taking action to address social inequalities. Based on this research, it is hypothesized that youth who have a strong sense of critical consciousness, specifically as it relates to perceived inequality, may value creating change and collective action to address inequalities. Research has shown that holding views about perceived inequalities is related to civic action (Diemer, et al., 2014), so may also be related to the importance of leaders working collaboratively to create change. Similarly youth who have egalitarian views may endorse the belief that anyone can be a leader as this aligns with treating people equally and  85 ensuring equal opportunities. This study explored relates to their definition of leadership according to the Collective Change Youth Leadership Framework.   Empowerment. Psychological empowerment closely aligns with critical consciousness, critical reflection) as well as active engagement in the community (similar to critical action; Zimmerman, 1995).  Individual empowerment develops through empowering processes and settings that allow individuals to develop their skills, participate in decision-making, and exert control over issues that affect their lives (Zimmerman, 1995; Maton & Salem, 1995). Often, to create change (Zimmerman, 1995; Holden, Messeri, Evans, Crankshaw, & Ben-Davis, 2004; Peterson, Peterson, Agre, Christens, & Morton, 2011).  Leadership development programs and opportunities may provide these types of settings for youth, and encourage them to develop their critical consciousness and sense of empowerment.   A commonly explored component of psychological empowerment is sociopolitical control (SPC) (Peterson, et al., 2011; Diemer, 2012; Speer, 2000; Zimmerman & Zahniser, 1991; Zimmerman, 1995; Zimmerman, Ramirez-Valles, & Maton, 1999)-efficacy to create change are core components of many youth leadership programs, which may in turn Research shows sociopolitical control often leads to action and engagement in the community (Rutter, 1993; Peterson, et al., 2011). As the research suggests, 1995), so it is hypothesized that individuals with more perceived sociopolitical control may be  86 more likely to take an active role in creating change (Rutter, 1993; Peterson, et al., 2011), a key component of the Collective Change Youth Leadership Framework.  Feeling empowered to take an active role in their community may influence how youth think about leadership, as they see the impact their actions have and develop ideas about how to enact leadership in a meaningful way. Youth who feel empowered to act as leaders may also see leadership as available to anyone as they reflect on how they have been empowered to engage in their communities and take on leadership roles, and see the possibility of all youth developing these skills to act influence different aspects of their lives. The current study explored how perceived sociopolitical control relates to  Civic Engagement.  Civic engagement is an important component of being a change agent in communities, and research shows civic engagement is beneficial to both youth and communities (Larson, 2000; Scales & Leffert, 1999; Christens & Dolan, 2011; Mitra, 2006; Pancer, Pratt, Hunsberger, & Alisat, 2007).  Often, civic engagement efforts occur within groups, as in community organizing or participation in a coalition.  Foster-Fishman and (2012) study of a youth leadership program found increased levels of both civic and political activism among youth participants, and an increase in intended future civic and political activism. These findings suggest the importance of developing youth as leaders and including them in meaningful leadership roles to encourage their civic engagement both now and in the future. This study explored how youth engagement in their communities relates to their definitions of leadership.  Research shows civic engagement often leads to further civic engagement as individuals see the impact their work has on their lives and the lives of others (Borden & Serido, 2009, Jennings, et al., 2006). thoughts about leadership as they reflect on those experiences and how their work may have  87 impacted them and their communities. It is hypothesized that youth who have been more civically engaged may see collective action as an important component of leadership, because youth often work collaboratively with others in their schools and communities when taking on leadership roles (e.g., Soleimanpour, et al., 2008; Powers & Allaman, 2012; Christens & Dolan, 2011). Higher levels of civic engagement may also relate to creating change, which is a common goal of youth leadership efforts as demonstrated by examples described above such as youth working to create programs and change policies in their schools and communities that reflected their needs. Finally, youth who have been more engaged in their communities may also place higher value on motivation and ambition because they were motivated to get involved as a leader and they see how their engagement can have an impact which may motivate them to continue to work toward change.  The relationship between civic engagement and leadership definition was investigats connect with their views of leadership.  STUDY RATIONALE. Communities are increasingly seeing the value of engaging youth in leadership roles that affect their lives, and youth are motivated to take on these roles and participate in leadership opportunities in their schools and communities. However, many of the programs designed to develop youth leaders are created by adults, without input from youth about what they need to be supported as a leader or how they define leadership in their lives. From a feminist standpoint theory, youth have different ideas about their lives than adults based on their social locations and experiences, and from what we are beginning to learn about youth leadership, how youth define leadership is no exception. Creating leadership programs from an adult perspective may lead to a disconnect between youth and the programs and discourage participation by youth who do not share those views.   88 One key shortcoming of the current youth leadership literature is the lack of a youth defined framework for leadership. This study is a critical first step to better understand how youth think about leadership and how their past experiences and attitudes relate to their definitions of leadership. The information gained from this study will prove valuable in and support their leadership development. Researchers and community-based organizations will be able to more effectively develop the next-generation of leaders and engage more youth in their efforts when programs are informed by a youth-driven definition of leadership and supports are reflective of the needs of various youth. This study focused future work can identify necessary supports for youth leaders with a solid grounding in how youth define leadership. RESEARCH QUESTIONS. The current study seeks to answer the following research questions. 1. To what extent does the concept of youth leadership encompass the six dimensions suggested by the Collective Change Youth Leadership Framework?  a. Do the proposed six dimensions within the Collective Change Youth Leadership  constructions of leadership? b. Do the individual dimensions of the Collective Change Youth Leadership Framework map on to a higher order variable, Youth Leadership? 2. How do background characteristics (age and gender), adult support, critical consciousness, empowerment, and civic engagement relate to definitions of leadership? This study will also explore a few specific hypotheses for the multiple regression models based on relationships discussed above between the experience/attitude variables and the leadership subscales.  89 1. Girls will report stronger beliefs in the need for leaders to support Collective Action than boys.  2. Youth who have had experiences of Facilitative Adult Support will demonstrate stronger ideas related to the importance of leaders acting as Models and Mentors.  3. Youth with stronger beliefs related to Critical Reflection-Perceived Inequality will also demonstrate mores support for leaders working toward Creating Change and Collective Action. 4. Youth with views that align with Critical Reflection-Egalitarianism will see leadership as available to Anyone.  5. Youth who feel Empowered to act as leaders will display stronger beliefs that Anyone can be a Leader and leaders should work to Create Change.   6. Youth who have been more Civically Engaged will see Creating Change, Collective Action, and Motivation and Ambition as crucial aspects of youth leadership.  RESEARCH DESIGN AND METHODS  OVERVIEW  To investigate the proposed research questions, this study built on the two initial studies discussed in the Introduction (Photovoice and Concept Mapping), and used secondary data from a survey to empirically test the proposed leadership framework with a national sample of youth. The data in this study came from a survey that was collected -Year Training Institute in 2013 (described below).  Differences between youth in how they define leadership according to the Collective Change Youth Leadership Framework were explored by looking at how background characteristics (gender and age), adult support, critical consciousness,   90 SETTING  CADCA is an international organization that trains community anti-drug coalitions and provides technical assistance, evaluation, and capacity building to coalitions around the world.  -wide change through targeting laws, policies, practices, systems, and attitudes to promote drug-free communities (CADCA, 2009). More than 5,000 coalitions exist within this network, and each coalition brings diverse community leaders together (e.g. educators, law enforcement, faith community, social service providers, government officials, parents, youth) to transform their communities and advocate for change locally and nationally. CADCA emphasizes the importance of engaging youth in community change efforts and created the National Youth Leadership Initiative (NYLI) to train youth. Multi-sector coalitions across the country often work in marginalized communities to address issues related to substance abuse that are prevalent in their neighborhoods. These coalitions engage a range of youth participants, and intentionally recruit disenfranchised youth to build their capacity as leaders. These youth may demonstrate risk factors such as attending alternative high schools, living in high crime or low income communities, or having limited access to resources due to living in rural areas. Using the Strategic Prevention Framework (SPF), CADCA trains youth to create logic models that identify community problems, develop action plans to address those program that trains both youth and adults, emphasizing a supporting role for adults and allowing youth to drive the work and decision-making. Training in NYLI occurs during their annual conferences and through individual training and consultation with communities.  CADCA hosts two large conferences each year, the National Leadership Forum and the Mid-Year Training Institute, which provide training to youth and adults across the country.  For  91 this study, survey data was collected at the Mid-Year Training Institute in Austin, TX in July 2013. At this training conference, 1698 individuals attended, including 221 youth and 71 youth advisors.  In total 705 coalitions were in attendance, and 65 coalitions had youth participants engaged in the NYLI training track.    PARTICIPANTS  Youth connected to their local substance abuse coalitions from across the country were -Year Training Institute. Data collection took place at one of three times: 1) during the youth orientation session the evening before the first day of the conference, prior to the delivery of any training content; 2) on the first day of the conference during a youth social event; or 3) on the second day of the conference if youth did not attend the orientation session or social event (e.g. due to arriving late at the conference or choosing not to Mid-Year Training Institute to avoid having the training influence their responses on the survey.  In total, 177 youth participated in the survey (80% of the 221 Mid-Year Training Institute youth attendees). The remaining youth were unable to complete the survey due to travel restrictions that caused many participants to arrive later in the evening the day before the conference or the following day, or because the CADCA staff that collected additional surveys was not able to connect with the youth on the second day of the conference. Of these surveys, 147 were included in the analyses. Three surveys were excluded from the analyses because the youth filled in response bubbles in a pattern to spell out words or drew a line down a column of response bubbles, and the 27 surveys that were collected by CADCA staff during the second day of training were also excluded because the youth had received at least a full day of leadership training prior to completing the survey. Youth came from 30 states across the country and 52  92 coalitions. Youth who completed the survey were primarily female (65%), White (61%), and lived in rural communities (48%).  The average age of participants was 15.90 years (Range = 12-20, SDand just over half of the youth and/or their siblings did not receive free/reduced price lunch (54%). See Table 3 for additional demographic data. Demographic data was not collected by CADCA at the time of registration for the Mid-Year Institute (only age and ethnicity were collected on parent consent forms which were not available to MSU), and because the survey was anonymous, a comparison of the demographics of youth who did versus those who did not take the survey cannot be done.   Table 3. Youth Demographic Data  Demographics Percent   Gender  Female 65% Male 33%   Race  White 61% Black 12% Hispanic 15% American Indian/Alaskan 2% Asian/Pacific Islander 3% Multiracial 3% Member of a race not listed 3%   Community Type  Rural 48% Urban 25% Suburban  20%        93   Age M = 15.90 SD = 1.57 Range = 12-20   Grades in past year   58%  36%  5%  1%   Free/Reduced Lunch Status  Yes 41% No 54%  5%  -Year Training Institute or Leadership Initiative before (86%). However, a majority of youth had participated in some sort of leadership training program prior to the Mid-Year (78%).  In terms of their leadership experience, 78% of youth saw themselves as current leaders in their communities, and in their most recent leadership positions, 42% took on formal roles (e.g. class president), while 46% took on informal leadership roles (e.g. leading school projects).  Youth were asked to identify places they had acted as leaders, selecting all of the settings that they had been leaders in.  Youth served as leaders most often in school settings, for example, youth selected school (81%), on a sports team (48%), and in a student leadership group (e.g. student council; 47%), as the places they had acted as leaders.  Nearly half of the youth had also been leaders within a community organization (45%). See Table 4 for a summary of leadership experiences of the youth participants.       94 Table 4. Youth Leadership Experience  Leadership Experience Percent   Previously attended Mid-Year Training Institute   Yes 11% No 88%   Previously participated in NYLI  Yes 12% No 86%   Previously participated in a leadership training  Yes 78% No 21%   Current Leaders in the Community  Yes 78% No 16%   Type of Leadership Role  Formal 42% Informal 46%   Settings in which youth were leaders  School 81% Church 36% Sports Team  48% Academic Club 22% Service Club 26% Student Leadership Group 47% Other Extracurricular Activity 19% Community Organization 45% Job 22% Other 6% None 5%  Adult advisors were also surveyed at the same time; during their orientation session, on the first day of the training at the beginning of the youth social event, or on the second day of training by a CADCA staff member. Seventy one registered adult advisors participated in the Mid-Year, and 97% (N=69) of these adults completed a survey. On average, advisors were  95 female (77%), and middle-aged (M=40.55, SD=10.66); most were college educated (65% had Nearly all of the advisors considered themselves to be leaders in their communities and/or local substance abuse coalitions (87%) and had participated in a leadership training before (71%).   In addition, most had prior experience mentoring youth, with 90% of the advisors reporting they had mentored youth, and one quarter reporting they had served as youth advisors for their coalition for three or more years.  Advisors represented a number of roles on their local coalitions: parent/family member (7%), law enforcement (3%), organization that provides services to youth (15%), communitybased organization (16%), elected city/government official (4%), human services organization (16%), school system (13%), volunteer organization (6%), and interested citizen (6%).  One third of advisors had participated in their coalitions for three or more years, 25% had participated for one to three years, and 33% had participated for 0-12 months.  Some coalitions brought more than one advisor to the Mid-Year Training Institute (and some coalitions did not have a registered advisor), so if coalitions had more than one advisor, the advisor who had been a youth advisor for the coalition the longest was selected to have their data included in analyses about coalition characteristics and youth recruitment data.  For coalitions that had multiple advisors that had served for the same amount of time, one advisor was randomly selected.  In total, 41 advisors, representing 41 coalitions, were included for the analyses on coalition and youth recruitment data. Twenty four surveys were excluded because they were from duplicate coalitions, and 4 surveys were excluded because they did not report a coalition.    96 PROCEDURES  RECRUITMENT. Youth survey participants were recruited at the Mid-Year Training Institute at the orientation session for NYLI on the evening before the first day of training (n= 119), a social event the first day of the conference (n= 28), or on the second day of training during free time between training sessions (n=27).  The orientation session was led by CADCA staff and provided an overview of the training to youth attendees.  At the end of this overview, the MSU project director (author) described the survey and invited youth to participate. Similarly at the social event, the purpose of the survey was described and youth were invited to participate at the beginning of the session. On the second day of training, a CADCA staff member connected with youth from coalitions that had not participated in the orientation or social event and invited those youth to participate.  In all three settings, youth were provided with assent forms (See Appendix A) tcontact information.  Because youth were attending the conference without their parents, parental . All youth were required to have parental consent to attend the Mid-Year Training Institute. Youth were free to decide whether or not they would like to participate in the survey.  Adult advisors were also recruited during their orientation session, which was separate explaining the purpose and answering any questions that arose. Those adult advisors who completed their survey at the social event were recruited by the project director who explained the purpose of the study, obtained consent, and administered the survey. Some advisors (n=8) completed the survey on the second day of the training along with the youth who had not participated in the initial survey data collection sessions, and a CADCA staff member administered the consent process and survey.   97 DATA COLLECTION.  For the survey, each youth who agreed to participate was provided an assent form and a survey (See Appendix A), and each adult advisor was provided with a consent form and a survey (See Appendix B).  At the beginning of administration, the project director or advisor walked the participants through the assent/consent form, explaining the purpose of the study, the risks and benefits associated with their participation, as well as their rights as a participant to skip questions and stop participating at any time.  Once participants assented/consented to participate, they completed the paper and pencil surveys. To minimize risk and to encourage honest responses, the survey was confidential.  Participants were not asked to put their name on the survey, rather the surveys were assigned a random ID number after all surveys were collected. Signed asset forms were the only place particcollected. The MSU team and CADCA staff were available to answer questions, and CADCA staff was on-site to assist in collecting and organizing completed assent forms and surveys.  The surveys took approximately 30 minutes to complete.   DATA STORAGE. All physical information gathered from this study (e.g. paper surveys, assent and consent forms) is stored in locked filing cabinets in a locked office at Michigan State University. Electronic files are saved to password-protected computers that only members of the research team have access to. Only IRB certified research team members have access to paper or electronic files.   DATA ENTRY. Survey data was entered manually and quality checked by the project director.  A codebook was created with details about each scale as well as instructions for data entry, quality checking, and cleaning.   Missing Data and Scale Construction. Overall, there was very little missing data. For example, for the 68 leadership items (described below), missing data ranged from 0-6  98 participants for each item, with most items missing only 1-2% of data. Mean substitution was used to deal with missing values prior to completing additional analyses. While mean substitution does have its limitations (e.g., reducing variability; Roth, 1994; Little & Rubin, 1987), some research has found it is an adequate method for addressing missing data when there is a small amount of missing data points, as was the case in this study. For example, Widamen (2006) recommends using single imputation (e.g., mean substitution) when missing data is very identical results across different methods of missing data, and the detrimental effects of using sample mean or individual mean substitution will not occur because of the small number of  MEASURES  YOUTH SURVEY  Collective Change Youth Leadership Framework. The Collective Change Youth Leadership Framework scale included 68 items that map on to the six dimensions of the framework (see Table 5 for a description of each of these dimensions and a list of survey items that correspond with each dimension).  These items were generated from two initial studies about youth leadership (Photovoice and concept mapping, described in the Introduction), along with the existing youth leadership literature that aligns with these dimensions, as well as adult theory items to test for discriminant validity.  Thirty-one items were created from the statements youth brainstormed during the concept mapping process. These items were edited to add clarification based on the collection of ideas put forth by youth in this study. In addition, 16 items were developed using Photovoice narratives by creating survey items that describe a leader based on how youth defined leadership in their blog posts. Themes that emerged across blog posts that were not evident in the concept mapping statements were included in this list. Ten items were  99 added to this list from the youth leadership literature. These items emerged from other definitions of leadership and leadership development program components to create a comprehensive description of each dimension. Finally, 11 items that matched ideas from the adult leadership literature, but were not aligned with the six scales of the Collective Change Youth Leadership Framework, were added to this section of the survey to test for discriminant validity of the scales. It is anticipated that the average scale score for the Adult Leadership items will be lower than the average scale scores for the youth leadership subscales (this relationship was explored through paired samples t-tests). Approximately 10% of the final items were reverse scored (see Table 5 for items). A group of seven youth pilot tested the survey. They provided feedback on items that were confusing and made suggestions for revising those items.  Originally more items were reverse scored, but many of these items were confusing to the youth who piloted the survey, so some were returned to their original wording to avoid confusion. For all items, youth were asked to assess the importance of each component to their definition of leadership on a 5 point likert type scale (1=Not at All Important to 5=Very Important).  Table 5. Collective Change Youth Leadership Framework Dimensions and Survey Items  Dimension and Definition Items Developed from Concept Mapping  Items Developed from Photovoice narratives  Items Developed from the literature    Items marked with an X were endorsed by Photovoice or the Literature Anyone can be a leader  Anyone has the potential to be a leader and leaders can show leadership in a variety of situations. Can be anyone. X   Can act as a leader in many different roles, both large and small. X  100     Dimension and Definition Items Developed from Concept Mapping  Items Developed from Photovoice narratives  Items Developed from the literature (A  Creating Change  The ultimate goal of leadership is to create positive change within communities.   Stands up for what they believe in. X  Is someone who takes charge. X  Is someone who inspires others to create change. X  Helps other people. X  Cares about others. X   Tries to improve problems in their community.  X  Has a strong opinion about change that needs to occur. X  Works to create positive change.  X   Can create and communicate a vision.   Works to create change to address inequalities in the community.   Values social justice.    Collective Action  Leadership is not an individual effort but rather includes working together with others to achieve a common goal.  Is kind to people.   Is a team worker.  X  Works with others to achieve a common goal.  X  Learns from others.     101  Dimension and Definition Items Developed from Concept Mapping  Items Developed from Photovoice narratives  Items Developed from the literature  Collective Action  Leadership is not an individual effort but rather includes working together with others to achieve a common goal.  others.*   Never follows.*    Tries to build consensus in a group.   Is able to resolve conflicts.    Respects and tolerates diversity (e.g. people of different races/cultures).   Knows how to share power and distribute tasks to others.  Modeling and Mentoring  Leaders guide others to do what is right, acting as a mentor and role model, leading by example, and providing support to others.        Can influence others. X X Is open to all ideas/opinions/perspectives.   Is someone that people look up to.   Leads by example.   Is easy to talk to.   Is forceful when interacting with others.* X   Provides guidance to others.   Teaches or demonstrates leadership skills to others.    Acts as a positive role model for others.  X           102 Table 5     Dimension and Definition Items Developed from Concept Mapping  Items Developed from Photovoice narratives  Items Developed from the literature  Modeling and Mentoring Leaders guide others to do what is right, acting as a mentor and role model, leading by example, and providing support to others.    Provides support to others    Leaves skills behind for new leaders.     Seeks out role models that have been leaders. Strong Character  Leaders have a good character and put others before themselves. Is responsible. X  Is positive and optimistic/hopeful.   Is a peacekeeper.   Is happy.   Has patience.   Is respectful of other people.   Is honest.   Is loyal.   Leads for the good of themselves.*   Gives in to peer pressure.*    Is truthful.   Puts themselves before others.* X   Can empathize with others (e.g. understand, relate to someone else). Motivated and Ambitious  Leaders are hard-working, and determined to achieve their change goals.  Has confidence in what they do. X  Is courageous. X  Motivates and encourages others.  X Is ambitious (e.g. has a strong desire to succeed).   Is determined and motivated.        103     Dimension and Definition Items Developed from Concept Mapping  Items Developed from Photovoice narratives  Items Developed from the literature  Motivated and Ambitious  Leaders are hard-working, and determined to achieve their change goals.  Is passionate about their work.   Is good at setting goals.  X  Perseveres to achieve goals.    Values continuous improvement. Adult Leadership Theory  These items come from the adult leadership literature and were added for divergent validity.    Is authentic (true to themselves).   Makes decisions for the group.   Needs followers to achieve their goals.   Is clearly distinguished from (e.g. more powerful than) their followers.   Manages other people.    Must have specific traits (e.g. dominance, extraversion, assertiveness).   Is one person or a small group of people.    Has power and authority over others.   Is born a leader.   Feels the need to have power.   Is charismatic (charming, likeable, energizing, enthusiastic). Note: Items marked with an * were reverse scored.  104 Background Characteristics. A number of demographic variables were included on the survey to learn about the characteristics of the sample, and to understand how these age, gender, race, grade in school, academic performance, type of community and state participant lived in, and socioeconomic status. Youth were asked to report their age in years, and gender as male or female.  Race was measured according to census categories (White, Black or African American, American Indian or Alaska Native, Asian, Native Hawaiian or Other Pacific Islander) with the addition of Hispanic/Latino and Multiracial. Grade in school was reported as the last grade youth completed, and academic performance was measured by typical grades in school the previous year.  Youth were asked to report the type of community they lived in by selecting urban, suburban, or rural, and they were also asked to write in the name of the state they live in and coalition they partici were used as proxies for SES. This section also asked youth about their prior experience in youth leadership training programs, including wtraining programs. For this study, the relationship between these variables (gender and age) and how youth define leadership were explored using multiple regression. See Appendix A for Youth Leadership Survey (Section A includes all Background Characteristics questions).  Prior Leadership Experience. Youth were asked if they had ever been in any leadership roles within their communities (including school, local organizations, etc.).  They were asked where they took on those roles and were provided with a list of responses.  Youth were asked to select all of the settings where they acted as leaders in the past. See Appendix A for Youth Leadership Survey (Section C includes all Prior Leadership Experience questions).  105 Adult Support. This section included questions about adult engagement (if adults were present), asking youth to rate the extent to which the setting they acted as a leader in was youth or adult driven, as well as the type of support they received (facilitative or directive).  This scale ition of these two approaches to supporting youth leaders. Ten items asked about directive support (e.g., Adults set the goals for the work; Adults determined the activities youth would work on) and eight items asked about facilitative support (e.g., Adults supported youth having control over activities; Adults gave youth the freedom to control their projects).  Each item was rated on a 5-point likert scale ranging from 1=Strongly Disagree to 5=Strongly Agree. A mean score was created for each subscale, with higher scores indicating more of that type of support. See Section C of the Youth Leadership Survey in the Appendix A for Leadership Experience questions.  Critical Consciousness. As discussed above, Critical Consciousness includes both critical reflection and critical action (Prilleltensky, 2012; Watts, et al., 2011).  Diemer and colleagues (2014) developed a 22-item Critical Consciousness Scale (CCS) to measure these two components of Critical Consciousness. Using exploratory and confirmatory factor analysis, they found the items fit a three factor structure that included: Critical Reflection: Perceived Inequality, Critical Reflection: Egalitarianism, and Critical Action: Sociopolitical Participation.  pecifically as it relates to Critical Reflection, the current study included Critical Reflection items. The items on the Critical Action: Sociopolitical Participation factor align closely with the items used to measure Civic Engagement (discussed below) so were not included in this study. Critical Reflection: gender and how they impact education and job opportunities.  This subscale included 8 items  106 (e.g., Certain racial or ethnic groups have fewer chances to get a good high school education; Women have fewer chances to get good jobs). Critical Reflection: Egalitarianism included 5 ld be treated equally (e.g., All groups should be given an equal chance in life; We would have fewer problems if we treated people more equally).  These two subscales were measured on a six-point likert scale (1=Strongly Disagree to 6=Strongly Agree). A mean score was created for each of the subscales, with higher scores indicating higher levels of critical reflection. See Section D of the Youth Leadership Survey in the Appendix A for Critical Consciousness questions. Empowerment.  Empowerment was measured using the 17-item Sociopolitical Control Scale for Youth (SPCS-Y; Peterson, et al., 2011).  Sociopolitical Control is a commonly measured element of psychological empowerment, and this scale was created specifically for youth.  Two subscales make up this measure.  Leadership Competence (8 items; e.g., I am often a leader in groups; I would prefer to be a leader rather than a follower) and Policy Control (9 items; e.g., My opinion is important because it could someday make a difference in my community or school; I enjoy participation because I want to have as much say in my community or school as possible).  All items were measured on a 5-point likert scale ranging from 1=Strongly Disagree to 5=Strongly Agree. A mean score for each of the subscales was calculated, with higher scores indicating higher levels of sociopolitical control. These mean subscale scores were then be averaged together to create a single Empowerment scale score. See Section E of the Youth Leadership Survey in the Appendix A for Empowerment questions. Civic Engagement. Civic Engagement was measured using 2 subscales of the Youth Inventory of Involvement scale (Pancer, et al., 2007).  This is a 30-item scale that measures youth involvement in a variety of settings and activities, and includes four subscales: political  107 activities, community activities, passive involvement, and helping activities.  This study included the political activities (7 items; e.g., Worked on a political campaign; Ran for a position in student government) and community activities (8 items; e.g., Participated in a church-connected engagement. Each of the items was rated on a five-point likert scale (0=never did this to 4=did this a lot), asking youth to indicate how often they were involved with each of the activities in the past 12 months.  A sum score for each of the two subscales was calculated to represent the level of civic engagement youth had for political and community activities. These sum subscale scores were then added together to create a single Civic Engagement score (15 items). See Section F of the Youth Leadership Survey in the Appendix A for Civic Engagement questions. MEASURES  ADULT ADVISOR SURVEY Adult advisors were surveyed to learn more about how youth were selected to attend the Mid-Year Training Institute and about the coalitions they came from to provide additional contextual information.  Background Characteristics. A number of background questions were asked to understand the characteristics of the sample of adult advisors, including questions about their educational background (e.g. highest degree earned), economic background (e.g. employment status), various demographic questions (gender, marital status, age, race), and past experience as a leader in the community (e.g. how many years they have been a leader) and a mentor for youth. Advisors were asked to report their birthday, their gender as male or female, and their marital status as single, living with male or female companion, married, separated, divorced, or widowed.  Race was measured according to census categories (White, Black or African American, American Indian or Alaska Native, Asian, Native Hawaiian or Other Pacific Islander)  108 with the addition of Hispanic/Latino and Multiracial. Finally, advisors noted the role they represented on the coalition (e.g. parent, community-based organization), how long they had been involved in the coalition, how long they had been a youth advisor, and how many coalition and workgroup meetings they attended in the past 12 months. See Section D of the Advisor Survey in the Appendix B for Background Characteristics questions.  Coalition Characteristics. Advisors reported on various coalition characteristics, including tprograms, exclusively focusing on environmental/policy change), how the coalition identified itself (e.g. prevention-focused, school-based), and the geographic area that the coalition targeted (See Section B of the Advisor Survey in the Appendix B for Coalition Characteristics questions). Nearly half of the advisors reported their coalitions split their work between prevention programs/services and environmental change (49%). Advisors were also asked how their coalitions identified themselves in terms of the scope of their work (selecting all of the areas that applied to their coalition). The vast majority of advisors reported their coalitions were prevention focused (85%), and many said they were school-based (22%) and served an economically disadvantaged community (17%). In terms of geographic location, coalitions tended to be organized at the county (49%) or city level (10%).             109 Table 6. Coalition Characteristics (n = 41)  Coalition Status Percent Exclusively providing or coordinating the provision of prevention programs or services. 2% Primarily the provision of prevention programs, with some involvement in environmental or policy change 12% About evenly split between prevention programs/services and environmental change 49% Primarily focusing on environmental/policy change, some involvement in prevention programs 15% Exclusively focusing on environmental or policy change 7% Coalition Identity (select all that apply) Percent Prevention focused 85% Faith-based 5% School-based 22% Serving an economically disadvantaged community 17% Other 10% Geographic Area Targeted Percent County 49% City 10% Indian Reservation 2% Neighborhood/Community 2% Multiple Counties 5% School District 7% Multiple School Sites 2% Statewide 2% Other 2%  Youth Recruitment. Advisors were asked to report how youth were recruited to attend the Mid-Year Training Institute (e.g. for their leadership potential, participation in coalition), and were allowed to select all options that were applicable.  They also reported on the number of youth invited to attend the Institute, the number of youth attending, and whether the youth knew each other or not before arriving at the training conference (See Section C of the Advisor Survey in the Appendix B for Youth Recruitment questions). Advisors reported coalitions recruited students from their local substance abuse coalition or local schools. Over half of the advisors noted that coalitions selected youth because they thought the youth had leadership potential  110 (59%), about half of the advisors said coalitions included youth who participated on their local substance abuse coalitions (51%), a third of advisors said coalitions selected youth from school 22% of advisors said their coalitions had other methods for selecting youth (e.g. writing an essay about why they wanted to attend, members of other youth groups). On average, advisors reported 4.10 youth (SD=0.38) were invited to attend the Mid-Year Training Institute from each coalition, and 3.59 attended (SD=0.31) though there was significant variety across coalitions.  The number of youth who attended the training from each coalition ranged from one (2 coalitions) to 11 youth (1 coalition).  Advisors noted some or all of the youth attending the Mid-Year knew each other prior to going to the training (34% and 61%, respectively).   Youth Engagement in the Coalition. To understand how youth were engaged in each coalition, advisors noted the number of youth that were regularly involved in the coalition, and how many youth participated in various activities (e.g. attending meetings, having a voice in decision making). These items were scored on a four-point likert scale, (ranging from 0=None to 3=All); See Section A of the Advisor Survey in the Appendix B for Youth Engagement in the Coalition questions). On average, advisors reported coalitions had 8.37 youth (SD=0.58) youth were engaged in various activities within the coalitions. Patterns of engagement are quite consistent across the various activities, with the exception of serving as an officer.  Forty four percent of advisors said none of the NYLI youth participants served as officers, and 37% said only a few had. This may suggest the presence of a power structure within the coalitions that voice in decision-making, as nearly three quarters of advisors said a few, most, or all youth had a  111 voice in decision-The survey asked about NYLI youth participation rather than youth participation in general, so it is possible the youth invited to attend the Mid-Year training were those that were more actively involved in the coalition. Overall, this data tells us that according to the NYLI advisors, the youth participants in this study came from their local schools and substance abuse coalitions, and of the youth involved in coalitions, a few took an active role in the coalition such as attending and talking at meetings, and helping to organize coalition activities. The sample provides some variability across youth since not all were involved in their local substance abuse coalitions.   Table 7. Advisor-Reported Youth Participation in Coalitions NYLI Youth Participant Coalition Activity In the past 12 months, how many of the NYLI youth you  None A Few Most All Attended coalition meetings. 15% 42% 22% 17% Talked at coalition meetings. 15% 42% 24% 15% Helped organize coalition activities. 15% 34% 27% 20% Served as an officer.  44% 37% 10% 5% Had a voice in coalition decision-making.  24% 24% 22% 24% Had a efforts. 20% 32% 24% 17% Had opportunities to be a leader in the coalition.  22% 34% 17% 22% Became official members of the coalition.  22% 20% 24% 29% DATA ANALYTIC APPROACH Survey Data Analysis. Descriptive analyses, Correlation, Confirmatory Factor Analysis, and Multiple Regression were used to explore the research questions in this study. Specific information on each statistical analysis is described below in the Results section. Power. The literature does not provide a clear set of rules for estimating power and sample size for CFA studies (MacCallum, Widaman, Zhang, & Hong, 1999; Kline, 2010; Westland, 2010), but general guidelines have been suggested.  Due to the exploratory nature of this study, these general guidelines were  112 models, which requires a sample size of 10 for each observed indicator (Chin, 1998; Chin & Newsted, 1999), but other subject-to-variables ratios have been put forth as well (e.g. 20 to one (Hair, Anderson, Tatham, & Black, 1995), five to one (Bryant & Yarnold, 1995; Bentler & Chou, 1987), three to one (Cattell, 1978), and two to one (Kline, 1979)). General sample size recommendations have also been made, suggesting a sample size of at least 100 (Gorsuch, 1983; Kline, 1979), 150 (Hutcheson & Sofroniou, 1999), 200 (Kline, 2010), and more (Cattell, 1978; Comrey & Lee, 1992). Simulation studies of different ratios of indicators per factor (e.g. comparing 2, 3, 4, 6 items loading on a latent factor) have been conducted to understand how the number of indicators per factor and sample size affect CFA models (Boomsma, 1982; Marsh, Hau, Balla, & Grayson, 1998)colleagues (1998) simulation study recommend a minimum of N=100 when the ratio of indicators per factor is 3, which was the case in this study, where three item parcels were loaded on to the proposed latent variables. Studies have found that in general, larger sample sizes are preferred, but simulation studies support the minimum sample size of 100 as sufficient for a CFA such as the one presented here.  A power analysis was run, using a medium effect size (0.15) and a desired power of 0.8, to calculate the sample size needed for the multiple regressions to understand the relationship between the independent variables and the dependent variables (Soper, 2013; Abramowitz & Stegun, 1965; Cohen, 1988; Cohen, Cohen, West, & Aiken, 2003). These models included six independent variables (Civic Engagement, Critical Reflection-Perceived Inequality, Critical Reflection-Egalitarianism, Empowerment, Facilitative Adult Support, Directive Adult Support).  To test the multiple regressions, the required sample size was 97. Overall, using the minimum sample size of 100 recommendation (Gorsuch, 1983; Kline, 1979) and the findings from CFA  113 simulation studies (Marsh, Hau, Balla, & Grayson, 1998; Boomsma, 1982), there is sufficient power for testing the proposed CFA model with the current sample (N=147). The power analyses for the multiple regression and t-test also suggests sufficient power for testing the relationship to the dependent variable, Youth Leadership.  RESULTS This section will describe the analyses and results of each of the two research questions. The section begins with an exploration of the items on the youth leadership subscales, then presents results from the first and second order CFA models, and ends with a description of the series of multiple regression models run to understand the relationship between the experience/attitude scales and the youth leadership subscales and overall leadership scale.   Youth Leadership Subscale Analyses. To begin, item-total correlations were run for the Youth Leadership subscales to identify any items that may not have performed well on the within a subscale related to each other. Overall, the reverse score items displayed unusual patterns of correlation with the other items on their subscale (See Appendix C for correlation tables). For example, reverse score items tended to correlate more highly with each other within a subscale (e.g. On the Strong Character subscale, 50R. Leads for the good of themselves and 57R. Gives in to peer pressure were moderately and significantly correlated with each other; r=.34, p<.001), they had near zero correlations with all other items (e.g. On the Collective Action from r=-.08 to .13), or they were negatively correlated with other items on the scale while all non-reverse scored items displayed small to large, positive correlations (e.g. On the Modeling and Mentoring subscale, 25R. Is forceful when interacting with others and 48. Seeks out role models that have been leaders were negatively and significantly correlated with each other;  114 r=-.21, p<.05). The inconsistent correlations between positively worded and reverse score items suggests these items performed poorly and may have been confusing or difficult for youth to answer. There is some disagreement in the literature about the need for reverse scored items in surveys. Some argue it is important to include negatively worded items to prevent response bias such as acquiescence bias (agreeing with all items) or extreme response bias (selecting the extremes of a rating scale rather than choices near the middle; e.g. Nunnally; 1978; Singleton & Straits, 2005), On the other hand, negatively worded items are often misinterpreted by respondents or they make mistakes in responding (e.g. Accidentally agreeing with a negative statement; Sauro & Lewis, 2011). Based on the correlation patterns, the reverse scored items did not perform well and within scales and did not demonstrate a strong relationship with the positively worded items. Also, when reliability analyses were run, item-total statistics suggested an improvemesubscales. For these reasons, the reverse scored items were excluded from analyses. Finally, the Anyone can be a Leader subscale included only two items, and they were only moderately (though significantly) correlated with each other (r =.30, p<.001). Due to the low correlation between the two items, the Anyone can be a Leader subscale was excluded from the CFA analyses described below. Table 8 below includes the reliability statistics for the items on the youth leadership subscales.            115 Table 8.    Scale Initial Alpha Final Alpha Anyone can be a Leader    .41 Creating Change    .81 Collective Action   .68 .80 Modeling and Mentoring   .77 .84 Strong Character   .71 .86 Motivated and Ambitious   .81  RESEARCH QUESTION 1  To what extent does the concept of youth leadership encompass the six dimensions suggested by the Collective Change Youth Leadership Framework? First- and second-order Confirmatory Factor Analyses (CFAs) were run in Stata 14 to understand if the youth leadership scales were conceptually organized as proposed (first-order CFA), and if those factors loaded on to a higher order factor, Youth Leadership (second-order CFA). These analyses addressed Research Question 1a: Do the proposed six dimensions within the Collective Change Youth  and Research Question 1b: Do the individual dimensions of the Collective Change Framework map on to a higher order variable, Youth Leadership?. As mentioned above, the Anyone can be a Leader subscale was excluded from CFA analyses due to the scale including only two items that were moderately correlated with each other. For all subsequent analyses, the remaining five leadership subscales were included.  116 Model fit was evaluated using Chi-squared test of misfit, comparative fit index (CFI, 0.90 or greater suggests adequate fit, 0.95 and above suggests good fit; Kline, 2010), Tucker-Lewis Index (also known as the nonnormed fit index, NNFI; assessed the same way as the CFI; Tanaka, 1993), the root mean square error of approximation (RMSEA, 0.08 or below suggests reasonable fit, .05 or below suggests good fit; Browne & Cudeck, 1993), and the Coefficient of Determination (CD; a CD of 1.0 suggests a perfect fitting SEM model). The Coefficient of Determination (CD), is a measure of the effect of the exogenous variables on the endogenous variables in SEM (Bollen, 1989; Joreskog & Sorbom, 1996), represents the overall R2 for the (Doll, Zia, & Torkzadeh, 1994, p. 456). Where conceptually appropriate, modification indices were added to the CFA models to improve model fit. The first level of the CFA models included item parcels as manifest variables rather than individual items. The item parcels were then loaded on to each of the latent variables in the model. Item parcels are a recommended approach to CFA models when sample size is relatively small (Little, Cunningham, Shahar, Widaman, 2002; Hall, Snell, & Foust, 1999). Three item parcels were included for the five latent variables (leadership subscales) in the model: Creating Change, Collective Action, Modeling and Mentoring, Strong Character, and Motivated and Ambitious. Little, et al. (2002), describe random assignment of items as one approach to g from a domain sampling rationale, one simple method for constructing parcels is to assign each item, randomly and without replacement, to one of the parcel groupings. Depending on the number of items to be assigned, two, three, or possibly four parcels, or groupings of items, could be created. Random  117 assignment of items to parcels should, on average, lead to parcels that contain roughly equal  Given the exploratory nature of this study, two approaches to item parceling were used: first, items were organized conceptually, and then randomly. For the conceptual item parceling, the items within each of the leadership subscales were randomly assigned to one of the item parcels for that subscale, items were then averaged together to create the parcel, and then those parcels were loaded on to a latent variable to represent the proposed leadership factor. For example, three item parcels were created for Creating Change, using the individual items from that subscale. Parcel 1 included items 3, 4, 16, and 9; Parcel 2 included items 5, 11, 36, and 12; and Parcel 3 included items 7, 14, and 8. A mean score was calculated for each parcel. These parcels were then used as the manifest variables that loaded on the latent variable in the first and second order CFA models (see Figures 1 and 2). Similarly, for the random approach to parceling, the same number of parcels were created, but items from across all leadership scales were randomly assigned to a parcel rather than including only conceptually similar items within a subscale. For example, Factor 4 in the random CFA model included items 48 (modeling and mentoring), 65 (strong character), and 14 (creating change) in Parcel 1; items 16 (creating change), 11 (creating change), and 66 (motivated and ambitious) in Parcel 2; and items 54 (strong character), 64 (motivated and ambitious), 47 (modeling and mentoring), and 46 (modeling and mentoring) in Parcel 3. These items were randomly assigned to each parcel from across all the leadership subscales. Again, items within each parcel were averaged together to create a single manifest variable, which was then loaded on the five random latent variables. Table 9 includes descriptive statistics for the conceptually organized item parcels.       118 Table 9.  Item Parcel Descriptive Statistics Scale Min Max Mean SD Skew Kurtosis      Parcel 1 (items 3, 4, 16, 9) 2.50 5.00 4.41 0.52 -0.95 0.60 Parcel 2 (items 5, 11, 36, 12) 2.41 5.00 4.44 0.54 -1.27 1.78 Parcel 3 (items 7, 14, 8) 2.67 5.00 4.47 0.58 -1.20 1.15      Parcel 1 (items 17, 28, 27) 2.00 5.00 4.40 0.62 -1.35 2.35 Parcel 2 (items 21, 31, 32) 1.67 5.00 4.57 0.54 -2.34 8.08 Parcel 3 (items 26, 20) 2.00 5.00 4.42 0.68 -1.55 2.47      Parcel 1 (items 43, 19, 42) 1.67 5.00 4.41 0.65 -1.71 4.26 Parcel 2 (items 33, 44, 47, 37) 2.25 5.00 4.45 0.56 -1.16 1.31 Parcel 3 (items 40, 48, 39, 46) 1.75 5.00 4.33 0.62 -1.29 2.36      Parcel 1 (items 24, 65, 49) 1.33 5.00 4.43 0.63 -1.74 4.72 Parcel 2 (items 23, 35, 53) 1.67 5.00 4.63 0.57 -2.76 9.51 Parcel 3 (items 30, 51, 52, 54) 2.25 5.00 4.42 0.57 -1.38 2.47      Parcel 1 (items 56, 64, 58) 2.00 5.00 4.36 0.59 -0.88 1.11 Parcel 2 (items 60, 15, 61) 2.67 5.00 4.60 0.46 -1.37 1.97 Parcel 3 (items 62, 66, 67) 2.33 5.00 4.51 0.56 -1.17 1.30  Once conceptual and random item parcels were created, two first order confirmatory factor analyses (conceptual and random) were run to address Research Question 1a. For the first order model, Maximum Likelihood (ml) was used as the estimator, all item parcels loaded on only one of the latent variables, the loading of the first manifest variable for each latent variable was constrained to one, error terms were not allowed to correlate, but all latent variables were correlated with each other (see Figure 1).  The conceptually organized first-order CFA model had close to reasonable fit 2 80, N=147) = 170.92, p<.001; CFI= .93, TLI= .91, RMSEA = .088, CD=.99), and all factor loadings  119 were significant. Ideally, the chi-square statistic would not be significant, but it is sensitive to large sample sizes, and is very often significant in CFA models (Hooper, Coughlan, & Mullen, 2008; Bentler and Bonnet, 1980). For this reason, the relative/normed chi-square (2/df) is often reported along with the other fit indices introduced above to assess model fit. The smaller the ratio of the 2 to df, the better the model fit, though no absolute consensus exists as to what that ratio should be (Hooper, et al., 2008). Tabachnick and Fidell (2007) suggest the ratio be below 2. In this model, the normed chi-square was 2.14. The RMSEA was just above the .08 cutoff for reasonable fit (Browne & Cudeck, 1993). Modification indices were explored to try to improve model fit. Incorporating modification indices should only be done if it makes conceptual or theoretical sense (Schreiber, et al., 2006), and in this case, correlating the error terms for the leadership subscales was a reasonable modification because the items within the subscales were conceptually similar and measured in the same way. In this model, error terms were correlated for Collective Action item parcels 2 and 3, Strong Character parcels 1 and 3 and parcels 2 and 3. Figure 1 includes the correlated error terms that were added based on modification indices. The revised model had improved overall model fit 2 77, N=147) = 135.35, p<.001; normed 2 =1.76; CFI= .96, TLI= .94, RMSEA = .072, CD=.99). Though the chi-square was still significant, the normed chi-square moved below 2, the RMSEA dropped below the .08 cutoff for reasonable fit, the CFI moved above .95, and the TLI remained above .9, so this model had reasonable fit overall. Table 10 provides standardized factor loadings and standard errors for the first order CFA model. Next, another first order CFA was run using the randomly organized item parcels. This model included five latent variables with three item parcels each. Similar to the conceptually organized CFA, Maximum Likelihood (ml) was used as the estimator, all item parcels loaded on  120 only one of the five latent variables, the loading of the first manifest variable for each latent variable was constrained to one, error terms were not allowed to correlate, but all latent variables were correlated with each other. The model did not have acceptable fit, as the RMSEA was well above the .08 cutoff: 2 80, N=147) = 194.55, p<.001; normed 2 =2.43; CFI= .93, TLI= .91, RMSEA = .099, CD=.98). Modification indices were explored, and one was added to the model to allow error terms of two manifest variables that loaded on the same latent variable to correlate, but this did not improve model fit much, and the RMSEA remained above the .08 threshold for reasonable fit: 2 79, N=147) = 179.52, p<.001; normed 2 =2.27; CFI= .94, TLI= .92, RMSEA = .093, CD=.95). Though some of the fit indices did meet minimum requirements for acceptable fit (e.g., CFI was greater than .90), the conceptually organized CFA model demonstrated better fit on more of the reported indices, and CFA models should be driven by theory, which was the case in the conceptually organized model. These results suggest the conceptually organized first-order CFA is a reasonable representation of the Collective Change Youth Leadership Framework, which includes five components: Creating Change, Collective Action, Modeling and Mentoring, Strong Character, and Motivated and Ambitious.        121 Table 10. Conceptually Parceled First Order CFA Factor Loadings  Latent Variable and Indicators  Standardized Estimate SE Factor One:      Parcel 1 (items 3, 4, 16, 9) 0.80** 0.04 Parcel 2 (items 5, 11, 36, 12) 0.77** 0.04 Parcel 3 (items 7, 14, 8) 0.65** 0.06 Factor Two:      Parcel 1 (items 17, 28, 27) 0.85** 0.03 Parcel 2 (items 21, 31, 32) 0.72** 0.05 Parcel 3 (items 26, 20) 0.64** 0.06 Factor Three:  and     Parcel 1 (items 43, 19, 42) 0.72** 0.04 Parcel 2 (items 33, 44, 47, 37) 0.85** 0.03 Parcel 3 (items 40, 48, 39, 46) 0.82** 0.03 Factor Four:      Parcel 1 (items 24, 65, 49) 0.89** 0.03 Parcel 2 (items 23, 35, 53) 0.76** 0.04 Parcel 3 (items 30, 51, 52, 54) 0.81** 0.04 Factor Five:  and     Parcel 1 (items 56, 64, 58) 0.74** 0.05 Parcel 2 (items 60, 15, 61) 0.74** 0.05 Parcel 3 (items 62, 66, 67) 0.74** 0.05 Note: **p<.001 Note: Item Parcels were created by taking the mean of the individual items in parentheses above; item numbers correspond to the youth leadership questions on the survey.       122 Figure 1. Conceptually Organized First Order CFA Model   Note: Circles in above figure represent error terms.  To test the next research question, 1b. Do the individual dimensions of the Collective Change Youth Leadership Framework map on to a higher order variable, Youth Leadership?, a second order confirmatory factor analysis was run to understand if the first order factors load onto an underlying, higher order factor, Youth  Leadership (see Figure 2). Similar to the first order model, the estimator was Maximum Likelihood (ml), and all item parcels loaded on only one latent variable. The loading of the first manifest variable (item parcel) was constrained to 1  123 for each of the first-order latent variables, in addition, the first latent factor loading (Creating Change) was also constrained to 1 in the second-order model. The model did not allow error terms to correlate, with the exception of the correlated error terms in the first order CFA presented above (see Figure 1 above). The model had reasonable fit 2 82, N=147) = 152.65, p<.001; normed 2 =1.86; CFI= .95, TLI= .94, RMSEA = .077), as demonstrated by an RMSEA below the .08 cutoff, a normed chi-square below 2, a CFI at the .95 cutoff, and a TLI above .90. All standardized factor loadings were statistically significant. Table 11 includes the standardized factor loadings and standard errors for the latent variables that loaded on to Youth Leadership (Appendix C includes a correlation table for the youth leadership subscales). Given the results of the second order CFA, it is reasonable to suggest the five dimensions of the proposed framework do comprise one underlying factor, Youth Leadership.    The literature suggests loadings of .7 and higher are recommended to confirm the variables represent a single factor, as this value corresponds to explaining nearly half (49%) of the variance in the indicator by the factor, though smaller loadings have also been seen as acceptable (e.g. .4 or .6) (Raubenheimer, 2004; Hair, Anderson, Tatham, & Black, 1998). This was the case for all latent variables. These results suggested the Collective Change Youth Leadership framework includes five dimensions, and these five dimensions make up a single leadership variable represented by this framework. However, limitations of CFA, such as competing models (Loehlin, 2004; Lee & Hershberger, 1990; MacCallum, Wegener, Uchino, & Fabringar, 1993; Raykov & Marcoulides, 2001), should be considered and these results interpreted with caution. These limitations will be presented in the Discussion section below.      124 Table 11.  Standardized Coefficients for Second Order Confirmatory Factor Analysis   Standardized Estimate SE Creating Change .84** 0.04 Collective Action .93** 0.03 Modeling & Mentoring .99** 0.02 Strong Character .94** 0.03 Motivated & Ambitious  .88** 0.04 Note: **p<.001  Figure 2. Second Order CFA Model for Collective Change Youth Leadership Framework  Note: Circles in above figure represent error terms.      125 RESEARCH QUESTION 2  How do background characteristics (age and gender), adult support, critical consciousness, empowerment, and civic engagement relate to definitions of leadership? The relationships between the experience/attitude scales and each of the youth leadership subscales was explored, as well as the relationship between the experience/attitude scales and the overall youth leadership scale SCALE DESCRIPTIVES AND CORRELATIONS. Experience/attitude scale descriptive data, including minimum and maximum values, scale means and standard deviations, skew, kurtosis, and reliabilities for the scales are presented in Table 12, below. This table also includes the youth leadership subscales and the overall Youth Leadership scale. The subscales were created by averaging together the items retained (no reverse scored items) for each of the leadership subscales, and the overall Youth Leadership scale was calculated by averaging together the scores on the five leadership subscales to create a composite score.  To test for discriminant validity, a handful of items related to adult leadership theories were included on the survey. An Adult Leadership scale was created and compared to the youth leadership scales to see if youth rated the youth-defined components of leadership higher than the adult leadership items. Interestingly, the means for each of the youth leadership theory subscales and overall Youth Leadership scale were significantly greater than the mean of the adult leadership theory scale (M=3.12), as was indicated by a series of paired samples t-tests (see Table 12 for paired samples t-test results). Overall, these higher means indicate most youth, on average, reported the five components of the model as more important in their definitions of leadership than the items on the adult theory scale. All of the means for the youth leadership  126 (5.0) to youth, while the average of the adult scale was 3.12, indicating those items were  to youth.  Table 12.  Experience/Attitude and Leadership Scale Descriptive Statistics  Scale Min Max Mean SD Skew Kurtosis Alpha t Experience/Attitude Scales       Civic Engagement 2.00 56.00 23.35 11.07 0.70 0.42 0.82  Critical Reflection-Perceived Inequality  1.00 5.38 2.65 1.31 0.30 -1.11 0.94  Critical Reflection-Egalitarianism  2.60 6.00 5.43 0.73 -1.56 1.96 0.74  Facilitative Adult Support  1.63 5.00 4.03 0.59 -0.67 1.38 0.83  Directive Adult Support  1.00 5.00 3.33 0.69 -0.20 0.51 0.83  Empowerment  1.00 5.00 4.07 0.53 -1.34 7.00 0.78  Youth Leadership Scales     Creating Change 2.82 5.00 4.44 .45 -1.00 .92 0.81 -22.78** Collective Action 1.88 5.00 4.47 .52 -2.10 6.77 0.80 -20.63** Modeling and Mentoring 2.18 5.00 4.40 .53 -1.55 3.54 0.84 -20.14** Strong Character 2.10 5.00 4.49 .52 -2.16 6.18 0.86 -20.18** Motivated and Ambitious 2.83 5.00 4.49 .45 -0.83 0.47 0.81 -21.76** Youth Leadership 2.50 5.00 4.46 0.43 -1.60 4.31 0.92 -22.22** Note: Data in the t column represent paired t-test t values comparing youth theory scale means and adult theory scale mean. **p<.001  Next, correlations were conducted among the youth leadership subscales and the predictor variables that were included in the subsequent regression analyses. As Table 13 shows, Critical Reflection-Egalitarianism and Empowerment were moderately and significantly  127 correlated with all of the leadership subscales and the overall leadership variable. Also, Directive Adult Support and Facilitative Adult Support demonstrated positive correlations with some of the youth leadership subscales and the overall leadership scale.   Table 13. Collective Change Youth Leadership Subscale and Predictor Scale Correlations  Note: *p<.05, **p<.001  MULTIPLE REGRESSION.  First, to understand the relationship between the experience/attitude scales (independent variables) and the individual youth leadership subscales (dependent variables), and to test hypotheses one through six, a series of multiple regressions were completed. For the multiple regression models, the independent variables included Age and Gender in the first step of the model, and Civic Engagement, Critical Reflection-Perceived Inequality, Critical Reflection-Egalitarianism, Facilitative Adult Support, Directive Adult  Gender Age Civic Engagement Critical Reflection-Perceived Inequality Critical Reflection - Egalitarianism Facilitative Adult Support Directive Adult Support Empowerment Creating Change  .05 .08 .11 -.06 .28** .16* .08 .32** Collective Action  .02 .11 .03 -.04 .30** .23** .07 .30** Modeling and Mentoring  .00 .10 .03 -.04 .24** .16 .21* .35** Strong Character  .18* .04 -.07 -.07 .22** .10 .19* .22** Motivated and Ambitious  .04 .12 .16 .01 .29** .09 .20* .45** Youth Leadership .07 .10 .06 -.04 .31** .17* .17* .38**  128 Support, and Empowerment in the second step. A regression model was conducted for each of the five leadership subscales.  In the first multiple regression, Creating Change was included as the dependent variable. Step 1 of the model was not significant (F(2, 142)=0.61, n.s.; R2=.01), and neither gender nor age were significant predictors of Creating Change.  Step 2 was significant (F(8, 136)=3.77, p<.001; R2=.18), and the addition of the attitude/experience scales led to a statistically significant increase in R2 R2=.17, p<.001), accounting for 17% more variance in step two of the model than in step one of the model. Results suggest Critical Reflection-Egalitarianism (=.24, p<.05), and Empowerment (=.29, p<.05) were significant predictors of how youth define leadership as it relates to Creating Change. This supports hypothesis five which predicted youth who felt Empowered to act as leaders would display stronger beliefs that leaders should work to Creating Change. However, these findings do not support hypothesis three which predicted youth with stronger beliefs related to Critical Reflection-Perceived Inequality would demonstrate more support for leaders working to Create Change. The results from this multiple regression are presented in Table 14 below.           129 Table 14. Unstandardized and Standardized Coefficients for Creating Change Multiple Regression Model   Unstandardized Coefficients Standardized Coefficients    B SE B  R2 R2 Step 1    .01 .01 (Constant) 4.36 .08    Gender .05 .08 .05   Age .07 .08 .08   Step 2    .18 .17** (Constant) 2.28 .44    Gender .02 .08 .02   Age .05 .08 .06   Civic Engagement .00 .00 -.05   Critical Reflection-Perceived Inequality -.02 .03 -.06   Critical Reflection-Egalitarianism .15 .05 .24*   Facilitative Adult Support .05 .06 .07   Directive Adult Support .06 .05 .09   Empowerment .25 .08 .29*     Note: * p<.05, ** p<.001 Next, Collective Action was included as the dependent variable in the multiple regression model. Similar to the model above, step 1 of the model was not significant (F(2, 142)=0.88, n.s.; R2=.01), and gender and age were not significant predictors of the dependent variable. Step 2 of the model was significant (F(8, 136)=4.68, p<.001; R2=.22), and the addition of the attitude/experience scales led to a statistically significant increase in R2 R2=.20, p<.001), accounting for 20% more variance in step two of the model than in step one of the model. Results suggest Critical Reflection-Egalitarianism (=.28, p<.001), and Empowerment (=.29, p<.05) were significant predictors of the Collective Action component of the youth leadership framework. It was hypothesized (hypothesis one) that girls would report stronger beliefs in the need for leaders to support Collective Action, but that relationship was not supported by this model. Hypothesis three anticipated youth with stronger Critical Reflection-Perceived Inequality  130 beliefs would demonstrate more support for youth leaders working collectively, but that relationship did not emerge. Similarly, hypothesis six predicted youth who had been more Civically Engaged would see Collective Action as a critical aspect of youth leadership, but this pattern was not confirmed. The results from this multiple regression are presented in Table 15 below. Table 15. Unstandardized and Standardized Coefficients for Collective Action Multiple Regression Model   Unstandardized Coefficients Standardized Coefficients    B SE B  R2 R2 Step 1    .01 .01 (Constant) 4.39 .09    Gender .02 .09 .01   Age .12 .09 .11   Step 2    .22 .20** (Constant) 1.74 .49    Gender -.04 .09 -.04   Age .10 .09 .10   Civic Engagement -.01 .00 -.16   Critical Reflection-Perceived Inequality -.01 .03 -.03   Critical Reflection-Egalitarianism .20 .06 .28**   Facilitative Adult Support .12 .07 .14   Directive Adult Support .06 .06 .07   Empowerment .28 .09 .29*    Note: *p<.05, ** p<.001 The third multiple regression included Modeling and Mentoring as the dependent variable. Step 1 of the model was not significant (F(2, 142)=0.70, n.s.; R2=.01), and gender and age were not significantly related to Modeling and Mentoring. Step 2 was significant (F(8, 136)=5.40, p<.001; R2=.24), and the addition of the attitude/experience scales led to a statistically significant increase in R2 R2=.23, p<.001), accounting for 23% more variance in step two of the model than in step one of the model. Civic Engagement (=-.18, p<.05), Critical  131 Reflection-Egalitarianism (=.20, p<.05), Adult Directive Support (=.22, p<.05), and Empowerment (=.38, p<.001) were significant predictors of the Modeling and Mentoring subscale of the youth leadership framework. It was hypothesized (hypothesis two) that youth who had experienced Facilitative Adult Support would demonstrate stronger ideas related to the importance of leaders acting as Models and Mentors, but that was not the case and instead, Directive Adult Support was a significant predictor of this scale. The results from this multiple regression are presented in Table 16 below.  It is interesting that Civic Engagement was a negative predictor of the Modeling and Mentoring aspect of the youth leadership framework. Looking more closely at the correlation coefficients, it is possible that suppression is playing a role here. Horst (1966) defined suppressor but which do measure some of the variance in the predictor measures which is not found in the criterion measures. They measure invalid variance in the predictor measures and serve to -criterion relationship, even predictor variables and outcome variables by accounting for invalid variance. Civic Engagement and Modeling and Mentoring had a near zero correlation (r=.03, n.s.), so it would be expected that the regression coefficient would be similar, but in this model, Civic Engagement was a negative, significant predictor of Modeling and Mentoring. To further explore the effect of Civic Engagement on the other variables, the regression was re-run without Civic Engagement and  132 beta weights were compared to see how the addition of Civic Engagement (original regression) may have changed the other predictor variables.  Hierarchical regression is one approach for exploring suppression in a regression model. For example, Reiner, Anderson, Hall, and Hall (2010) used hierarchical regression models to explore gender as a potential suppressor in the relationship between adult attachment anxiety and God attachment anxiety to perceived stress. Similarly, Assari and colleagues (2014) found that SES acted as a suppressor in the relationship between HIV knowledge and frequency of sex with injecting drug users using hierarchical regression models in their study of Iranian female sex workers. Pandey and Elliott (2010) explored suppression in multiple regression using four statistics (R2, sum of squares, regression weight, and comparing zero-order correlations with respective semipartial correlations). They found hierarchical regression, and the significant change in R2 1978 (independent) and population per physician in 1980 (suppressor) were included, the R2 was significantly higher than when the model included only death rate of a country in 1978. In other words, R2 increased from 6.69% to 13.19% by adding a variable in the model that was not related with the dependent variable at the bivariate level. The model would have suffered if the variable for population per physician in 1980 had been eliminated after examining the bivariate ship between an independent variable and the outcome variable improves after the addition of the 2010, p. 38). In this model (Modeling and Mentoring as DV), step one of the model, which included Age and Gender was not significant (F(2,142)=0.70, n.s., R2=.01). Step two of the model was  133 significant and included all of the predictor variables except Civic Engagement (F(7, 137)=5.41, p<.001, R2=.22). Finally, step three of the model, which was originally step two of the first regression model run and included Civic Engagement, was also significant (F(8,136)=5.40, p<.001, R2=.24). In comparing beta weights across step two and three of the model, it appears that Civic Engagement is acting as a suppressor for Empowerment, as demonstrated by an pp<.001) with the addition of Civic Engagement to the model. This pattern suggests the addition of Civic Engagement helps to strengthen the relationships between Empowerment and the dependent variable (Modeling and Mentoring) despite not having a strong impact on the dependent variable on its own, a pattern typical of classical suppression (Lancaster, 1999; Watson, Clark, Chmielewski, & Kotov, 2013).  Civic Engagement and Empowerment were correlated (r=.49, p<.001), which is another indicator that suppression is occurring. Finally, the addition of Civic Engagement to the regression model did lead to a statistically significant increase in R2 R2=.02, p<.05), accounting for 2% more variance in step three of the model than in step two of the model. This significant change is another indicator that Civic Engagement is acting as a suppressor variable in this equation.                 134 Table 16. Unstandardized and Standardized Coefficients for Modeling and Mentoring Multiple Regression Model to Explore Suppressor Variable    Unstandardized Coefficients Standardized Coefficients    B SE B  R2 R2 Step 1    .01 .01 (Constant) 4.34 .09    Gender -.01 .09 -.01   Age .11 .09 .10   Step 2    .22 .21** (Constant) 1.72 .49    Gender -.01 .09 .00   Age .10 .09 .09   Critical Reflection-Perceived Inequality -.03 .03 -.07   Critical Reflection-Egalitarianism .14 .06 .19*   Facilitative Adult Support .05 .07 .06   Directive Adult Support .18 .06 .23*   Empowerment .29 .08 .29**   Step 3    .24 .03* (Constant) 1.59 .49    Gender -.02 .09 -.02   Age .12 .09 .11   Critical Reflection-Perceived Inequality -.02 .03 -.06   Critical Reflection-Egalitarianism .15 .06 .20*   Facilitative Adult Support .04 .07 .04   Directive Adult Support .17 .06 .22*   Empowerment .37 .09 .38**   Civic Engagement -.01 .00 -.18*   Note: *p<.05, ** p<.001 Next, Strong Character was included as the dependent variable. As was the case in the above analyses, step 1 of the model was not significant (F(2, 142)=2.47, n.s.; R2=.03), and gender (=.18, p<.05) was a significant predictor of Strong Character, but age was not. Step 2 was significant (F(8, 136)=4.15, p<.001; R2=.20), and the addition of the attitude/experience scales led to a statistically significant increase in R2 R2=.16, p<.001), accounting for 16% more  135 variance in step two of the model than in step one of the model. Gender (=.17, p<.05), Civic Engagement (=-.21, p<.05), Critical Reflection-Egalitarianism (=.17, p<.05), Directive Adult Support (=.22, p<.05), and Empowerment (=.31, p<.001) were significant predictors of how youth define leadership as it relates to Strong Character. In this model, females had higher ratings on the Strong Character scale than males as indicated by a positive regression coefficient, and through the exploration of groups with a t-test comparing males (M=4.35, SD=.69) and females (M=4.55, SD=.41; t(143)= -2.20, p<.05).  No specific hypotheses were stated for how  views on the attitude/experience scales would relate to Strong Character. Results from this multiple regression are presented in Table 17 below.  Similar to the model in which Modeling and Mentoring was the dependent variable, it appears Civic Engagement may be acting as a suppressor in this model as well. The regression was rerun, moving Civic Engagement to step three of the model to see how the regression coefficients of the other variables changed with the addition of Civic Engagement to the model. Results were similar to those described above. Step one of the model, which included Age and Gender, was not significant (F(2,142)=2.47, n.s., R2=.03). Step two of the model was significant and included all of the predictor variables except Civic Engagement (F(7, 137)=3.87, p=.001, R2=.17). Finally, step three of the model, which was originally step two of the first regression model run and included Civic Engagement, was also significant (F(8,136)=4.15, p<.001, R2=.20). The addition of Civic Engagement to the regression model did lead to a statistically significant increase in R2 R2=.03, p<.05), accounting for 3% more variance in step three of the model than in step two of the model. The beta weight for Empowerment was again larger in step 1, p1, p=.001), suggesting Civic Engagement strengthened the relationship between Empowerment and the  136 dependent variable (Strong Character), but did not have a positive beta weight in the regression model nor was Civic Engagement correlated with the dependent variable (r=-.07, n.s.).  Table 17. Unstandardized and Standardized Coefficients for Strong Character Multiple Regression Model to Explore Suppressor Variable   Unstandardized Coefficients Standardized Coefficients    B SE B  R2 R2 Step 1    .03 .03 (Constant) 4.33 .09    Gender .20 .09 .18*   Age .03 .09 .03   Step 2    .17 .14** (Constant) 2.39 .51    Gender .21 .09 .19*   Age .05 .09 .05   Critical Reflection-Perceived Inequality -.04 .03 -.09   Critical Reflection-Egalitarianism .11 .06 .15   Facilitative Adult Support .00 .07 .00   Directive Adult Support .18 .06 .23*   Empowerment .20 .08 .21*   Step 3    .20 .03* (Constant) 2.25 .50    Gender .19 .09 .17*   Age .07 .09 .07   Critical Reflection-Perceived Inequality -.03 .03 -.07   Critical Reflection-Egalitarianism .12 .06 .17*   Facilitative Adult Support -.01 .07 -.01   Directive Adult Support .17 .06 .22*   Empowerment .30 .09 .31**   Civic Engagement -.01 .00 -.21*    Note: *p<.05, ** p<.001 The last multiple regression for youth leadership subscales included Motivated and Ambitious as the dependent variable. Step 1 of the model was not significant (F(2, 142)=1.03, n.s.; R2=.01), and gender and age were not significant predictors of Motivated and Ambitious.  137 Step 2 was significant (F(8, 136)=7.95, p<.001; R2=.32), and the addition of the attitude/experience scales led to a statistically significant increase in R2 R2=.30, p<.001), accounting for 30% more variance in step two of the model than in step one of the model. In this model, Critical Reflection-Egalitarianism (=.23, p<.05), Adult Directive Support (=.23, p<.05), and Empowerment (=.47, p<.001) were significant predictors of the Motivated and Ambitious subscale of the youth leadership framework. Hypothesis six predicted youth who were more Civically Engaged would see Motivation and Ambition as critical components of youth leadership. However, these results did not support this hypothesis. The results from this multiple regression are presented in Table 18 below.  Table 18. Unstandardized and Standardized Coefficients for Motivated and Ambitious Multiple Regression Model   Unstandardized Coefficients Standardized Coefficients    B SE B  R2 R2 Step 1    .01 .01 (Constant) 4.41 .08    Gender .03 .08 .03   Age .10 .08 .11   Step 2    .32 .30** (Constant) 1.87 .39    Gender .04 .07 .04   Age .09 .07 .09   Civic Engagement .00 .00 -.09   Critical Reflection-Perceived Inequality -.01 .03 -.01   Critical Reflection-Egalitarianism .14 .04 .23*   Facilitative Adult Support -.05 .06 -.06   Directive Adult Support .15 .05 .23*   Empowerment .39 .07 .47**    Note: *p<.05, ** p<.001 Next, in order to explore the relationship between the experience/attitude variables and how youth define leadership according to the Collective Change Youth Leadership Framework, a  138 multiple regression was run using a single outcome variable, Youth Leadership. As described above, the outcome variable, Youth Leadership, was created by averaging together the five leadership scales to create a composite score. Similar to above, age and gender were controlled for by entering them into the first step of the regression, and the remaining attitude/experience scales were added in to step two. For this analysis, step 1 of the model was not significant (F(2, 142)=1.02, n.s.; R2=.01), and gender and age were not significant predictors of the overall Leadership scale. Step 2 was significant (F(8, 136)=6.44, p<.001; R2=.28), and the addition of the attitude/experience scales led to a statistically significant increase in R2 R2=.26, p<.001), accounting for 26% more variance in step two of the model than in step one of the model. Results suggest Critical Reflection-Egalitarianism (=.26, p=.001), Directive Adult Support (=.20, p<.05), and Empowerment (=.40, p<.001) were significant predictors of how youth define leadership. No specific hypotheses were stated for how  attitudes/experiences would relate to their overall views on Youth Leadership. The results from this multiple regression are presented in Table 19 below.                     139 Table 19. Unstandardized and Standardized Coefficients for Leadership Definition Multiple Regression Model    Unstandardized Coefficients Standardized Coefficients    B SE B  R2 R2 Step 1    .01 .01 (Constant) 4.37 .08    Gender .06 .08 .06   Age .09 .07 .10   Step 2    .28 .26** (Constant) 1.95 .39    Gender .04 .07 .04   Age .09 .07 .10   Civic Engagement -.01 .00 -.17   Critical Reflection-Perceived Inequality -.02 .03 -.05   Critical Reflection-Egalitarianism .15 .04 .26**   Facilitative Adult Support .03 .06 .04   Directive Adult Support .12 .05 .20*   Empowerment .32 .07 .40**   Note: *p<.05, **p<.001 Overall, these results suggest youth who have strong egalitarian views and those who feel empowered to participate in their communities and share their opinions are more likely to have definitions of leadership that align with the individual leadership subscales and the overall Collective Change Youth Leadership Framework than youth who have not had these experiences or hold these opinions. In some instances, directive adult support also demonstrated a significant adership. In nearly all the regression models, gender was not a significant predictor of the youth leadership subscale. One exception was the Strong Character model, in which gender was a significant predictor and females scored higher than males on this subscale. In three of the models, those that included Modeling and Mentoring, Strong Character, and Leadership as the dependent variables, Civic Engagement appeared to act as a suppressor for Empowerment. When included in the models, Civic Engagement had a small  140 to moderate negative beta weight, suggesting it did not contribute in a way that aligned with the zero-order correlation, but did strengthen the relationship between Empowerment and the outcome variable. Finally, views about perceived inequality, facilitative adult support, and age, did not contribute to how youth define leadership in terms of the proposed framework.  DISCUSSION The literature on a youth-defined conceptual framework of youth leadership is sparse in comparison to that of adult leadership, and what has been written comes primarily from an adult perspective.  To date, a framework for youth leadership has not been put forth, though some research has begun to identify differences in youth leadership as compared to adult theories of leadership (Roach, et al., 1999; Culp & Kohlhagen, 2000; Mortensen, et al., 2014; Mitra, et al., 2010). The current study begins to fill this gap in the literature by putting forth a youth-driven framework of leadership. Communities would benefit from including more youth in leadership roles, but the first step is understanding how youth define leadership so they can be supported in ways that align with their mental models of leadership. The current study had two primary goals; 1) to empirically test a youth defined framework of leadership, the Collective Change Youth Leadership Framework (RQ1: To what extent does the concept of youth leadership encompass the six dimensions suggested by the Collective Change Youth Leadership Framework?), and 2) to understand what experiences/attitudes con(RQ2: How do background characteristics (age and gender), adult support, critical consciousness, empowerment, and civic engagement relate to definitions of leadership?).   The first goal of this study was to empirically test the Collective Change Youth Leadership Framework, which includes six dimensions of leadership (Anyone can be a Leader, Creating Change, Collective Action, Modeling & Mentoring, Strong Character, and Motivated & Ambitious).  Overall, the findings suggest a five dimension framework may be a fitting model  141 for a youth-defined leadership framework, with these five dimensions representing a common underlying variable, Youth Leadership.  This study supports prior research in this area that has asked youth to define leadership for themselves (Mortensen, et al., 2014; Lichty, et al., 2011). The five included dimensions of the framework had strong, positive loadings on the second order latent variable, supporting prior research on the importance of these scales to youth leadership, and suggesting they encompass a single underlying construct, the Collective Change Youth Leadership Framework. For example, much of the youth leadership research discusses the importance of collective action, and working collaboratively to achieve goals (Yu, et al., 2007; Edelman, et al., 2004; Mortensen, et al., 2014).  to participate in their communities, and because youth often work collaboratively to create change (e.g. through community organizing and coalitions).  Also, the idea of modeling and mentoring is common in the youth leadership literature, as many programs promote positive role modeling as a responsibility of youth leaders (Edelman, et al., 2004). Similarly, youth who take on leadership roles in their communities demonstrate strong character and motivation and ambition as these opportunities are often driven by internal ideas about the need for social change. The alignment of the findings from this study and past research in this area suggests incorporated into youth leadership programs.  While this study provided evidence that this framework may provide a model for how youth conceptualize leadership, it is also possible that there could be better ways to measure these ideas to, perhaps eliminating some of the survey items that may be redundant or restructuring the scales in different ways by testing different factor models. Related, the Anyone  142 can be a Leader scale was excluded because it only included two items that were moderately correlated. This scale could be expanded by adding additional ideas from youth about who they see as current and potential leaders. Another approach would be to survey additional, larger samples of youth which would allow for CFA analyses using the individual items of the youth leadership framework. This would provide a deeper understanding of the individual survey items which the item parcels could not provide. This study began to fill this void by proposing one  complete, and indicates an opportunity, and need, to continue engaging youth to better understand how they define leadership so programs can adapt to their ideas about leadership and meet their needs to support them as leaders now and in the future. This study offers an initial framework to launch this conversation, and provides a preliminary understanding of how youth define leadership according to the Collective Change Youth Leadership Framework. Though the findings support past research in this area, one limitation of the current study is the sample that was surveyed. For both of the initial studies (Photovoice and concept mapping) that were completed to develop the proposed leadership framework and the current study, participants included youth who were attending a leadership training offered by the Community Anti-Drug Coalitions of America (CADCA). So while the data may be representative of youth engaged in organized leadership programs, the findings may not hold in a more general sample or align with youth who take on alternative leadership roles. It is important to recall that the few other studies that have asked youth about their ideas on leadership have also been with youth who were already leaders in their schools or communities or were attending a leadership program. Some of the previous studies that have reported on youth ideas of leadership have not reported specific demographics of the youth involved (e.g. Roach, et al., 1999; Mitra, et al.,  143 2010), so it is unclear who the youth were that engaged in these studies except that they were involved in a leadership program. Mortensen, et al. (2014) and Culp and Kohlhagen (2000) do report on demographics, with both studies including primarily rural, white samples of the same age (M=14.95 years and 16.6 leadership community included youth from across the world, of various socioeconomic backgrounds and levels of computer literacy, who had applied to be a part of the online Junior Summit.  The youth in their study were similar in terms of age range and varied geographic location as other studies of youth leadership, and somewhat similar in terms of racial diversity. Much of the recent literature on youth leadership has explored low-income communities and programs that target racial minorities (e.g. Lewis-Charp, et al., 2003; Christens & Dolan, 2011; Boyd, 2001), so those studies naturally over sample minority youth.  The youth in the program in the current study were not intentionally recruited based on income or racial background, so this sample aligns more closely with other studies that have not focused on specific racial groups (e.g. Dawes & Larson, 2011; Borden & Serido, 2009) but reflect the demographics of youth in the local communitand recruit a variety of youth, often intentionally recruiting disenfranchised youth who may be attending alternative high schools or come from low income or high crime neighborhoods. As will be discussed below, the racial background of youth in this sample is quite similar to the US population of youth under 18, so may be reflective of the larger population from which they were selected.  The current study addresses some of the limitations of past samples, including specific demographics on youth participants, and including a more racially diverse sample than those with primarily white, rural youth or reviews of programs that intentionally engage certain groups  144 of youth.  Though some of these limitations are addressed, the youth involved in this study were still part of a leadership program, most were invited to attend because of their leadership potential or engagement with their local substance abuse coalitions, and they were also high Though the sample is a limitation of this study, it is also true the youth engaged were primed to think about leadership because they were attending a conference about building their leadership skills. It is possible these youth have thought more about leadership than if a less intentional group of youth were selected to participate who may not have had opportunities to think about leadership before so may have had fewer or less developed ideas about leadership. This model does not include opinions from youth engaged in less organized leadership roles, and may not fully align with all youth leaders. Future research could compare how youth in more radical or alternative leadership roles think about and define leadership. Is it similar to this framework? How does it differ? What do they value that youth in formal leadership programs may not emphasize in their definitions? This framework provides one model of youth leadership, but additional youth in varying leadership roles should be engaged in a similar manner to learn how their ideas of leadership may or may not align with the model put forth by the youth in  Another limitation of the included sample was that they were high achievers academically. This may limit the generalizability of these findings to a more general youth population that includes a broader spread of academic grades. As was discussed in the Introduction, the youth who participate in leadership programs may be more high achieving youth in general (Kress, 2006), so additional research should be conducted with youth who are not engaged in a leadership role or program, and youth with more varied academic grades, to  145 understand if their ideas about leadership align with those of the youth in this study. Despite these limitations, the current study indicates the Collective Change Youth Leadership Framework may provide a model for understanding leadership from a youth perspective and guiding program development to meet the needs of youth. Suggestions for youth leadership development programs are discussed in the Implications for Practice section below.  The second goal of this study was to explore how different experiences and attitudes may relate to how youth define leadership. The results show critical consciousness as it relates to in some cases, directive adult support, were significantly related to how youth define leadership according to the Collective Change Youth Leadership Framework. Interestingly, Civic Engagement acted as a suppressor in two of these models, as it showed a close to zero correlation with the dependent variable, yet a significant, negative relationship in the regression models. Suppressor variables account for invalid variance in other independent variables in regression equations, strengthening the regression coefficients of those variables in the equation. This was evident in this study, as Civic Engagement had a close to zero correlation with the youth leadership subscales and the composite youth leadership scale, but when added to the hierarchical regression, it strengthened the relationship between Empowerment and the dependent variables. Each of the relationships between the experience/attitude scales and youth leadership subscales will be discussed below.  Gender. Gender was included in the regression models to understand if it was related to how youth define leadership according to the Collective Change Youth Leadership Framework. In all but one regression model, that which included Strong Character as the dependent variable, gender was not a significant predictor of youth leadership. It was hypothesized that girls and  146 boys would differ in their views of Collective Action due to girls demonstrating more collaborative patterns of play and interaction (Maccoby, 1990), but this did not hold in the current study. This pattern suggests that despite the impact of gender socialization on so many 1999; Ewing & Troop-Gordon, 2011), gender does not appear to relate to how youth think about leadership according to the Collective Change Youth Leadership Framework. This study explored how youth think about leadership and did not find a significant relationship between gender and leadership definitions, but future research could also explore how youth enact leadership in various ways and see if gender has an impact on how youth act as leaders in their communities. Age. Youth in this study ranged from twelve years old to twenty, which represents a large window of rapid development throughout adolescence (Eccles & Harold, 1993). As youth grow and experience more within their social and political surroundings, this may influence how they define leadership. In the current study, youth were categorized as young versus old, and Age was included in the regression models to understand if it was related to how youth define leadership. In none of the models was age a significant predictor, suggesting that in this sample, youth did not vary by age in their definitions of leadership. Future research should continue to explore this variable to ensure leadership programs and opportunities meet the needs of youth of varying ages.   Civic Engagement. Two of the regression models suggested Civic Engagement, as signatures for a political campaign) or community (e.g. participating in church-connected groups, doing things to improve the neighborhood) activities was a significant, though negative predictor  147 of how they define leadership as it relates to Modeling and Mentoring and Strong Character. However, further exploration indicated that Civic Engagement acted as a suppressor variable in the regression equations.  Suppressor variables do not measure variance in the outcome variable, but do account for invalid variance in other predictor variables (Horst, 1941; Horst, 1966; Lancaster, 1999). Horst (1966) was the first to define suppressor variables with a classic example navigating skills. He found that adding verbal ability to the regression equation along with tests of technical abilities increased the proportion of explained variance in the equation, but verbal ability had a negative regression coefficient and a near zero correlation with the outcome variable, navigating skills (Horst, 1966). The explanation for this was that verbal skills were necessary to complete the written assessments, and including it in the regression helped to remove some of the influence of verbal abilwas not predictive of their navigating skills (Horst, 1966).  Exploring Civic Engagement as a suppressor variable in this study through hierarchical regressions indicated that adding Civic Engagement to the models helped to explain more variance and strengthened the relationship of one of the predictor variable, Empowerment, with completing the written assessments to become a pilot, it is possible that Civic Engagement is similarly important in its relationship to Empowerment in this model. Civic Engagement was significantly correlated with Empowerment, so adding it to the regression model helped to account for invalid variance in the Empowerment variable, but by itself, Civic Engagement did not relate to or predict how youth define leadership. It is somewhat surprising that Civic Engagement was not related to youth leadership because this definition of youth leadership  148 emphasized the importance of working to create change and actively engaging with others, which are similar to ideas of political and community engagement explored here. It is possible simply engaging in the community may not be sufficient for youth leaders. Past research shows youth participation in leadership roles is often token (Stoneman, 2002), and youth have voiced dissatisfaction and frustration with not being truly involved in creating change or affecting their communities (Taft & Gordon, 2013; MacNeil, 2006).  This could be the case with this sample of youth; that they have felt discouraged in past leadership roles, and perhaps those experiences alone did not influence the way they thought about leadership, but is related to their sense of empowerment, as evidenced by the significant correlation between these variables.  As was demonstrated by the reports from adult advisors about how youth were engaged in their coalitions, it appeared many coalitions only engaged a few youth in various processes. Though we know that in many coalitions, only a few youth were actively involved, it is unclear whether it was due to youth not wanting to be involved or the coalition not engaging them.  It is also possible many of the youth were too young to have participated in some of the political activities that were included on the survey, so political activities were not related to how youth conceptualize leadership. However, despite the near zero correlation between Civic Engagement and the dependent variables, including it in the regression model helped to account for some invalid variance in the Empowerment variable, and explained more variance in the overall model.  Critical Consciousness. Research shows Critical Consciousness affects how youth view society, and their understanding of what needs to change to address unjust conditions (Diemer, et al., 2014; Zimmerman, 1995; Wang & Burris, 1997; Checkoway & Richards-Schuster, 2003; Foster-Fishman, et al., 2005).  This study supports past research on critical consciousness,  149 and how they define leadership according to this framework.  Specifically, youth who held more egalitarian views, such as the belief that all groups should be given an equal chance in life and that we would have fewer problems in society if people were treated more equally, saw leadership as aligned with the Collective Change Youth Leadership Framework, which includes creating change and improving conditions for all as key aspects of leadership. It was hypothesized that youth with stronger beliefs related to Critical Reflection-Perceived Inequality would demonstrate more support for leaders working toward Creating Change and Collective Action, but Perceived Inequality was not a significant predictor in any of the regression models. On the other hand, Critical Reflection-Egalitarianism was a significant predictor in all the models, suggesting this aspect of Critical Reflection is an important predictor in how youth think about leadership. This finding does not fully align with Diemer and colleagues (2014) study that found Egalitarianism was not related to Civic Action, as the proposed leadership framework is action oriented and focused on working to create change. Additional research should continue to explore the relationship between Critical Consciousness and youth leadership to fully understand  Adult Support. In this study, both directive adult support (e.g. adults keeping youth focused on their work and setting the goals for the work) and facilitative adult support (e.g. adults giving youth the freedom to control projects, allowing youth to learn through trial and error) were explored as predictors of how youth define leadership. Based on what we know about how youth think about leadership (e.g. collective action, modeling and mentoring; Mortensen, et al., 2014), and their generally powerless position in society (Langhout & Thomas, 2010), it was hypothesized that youth who had more facilitative support from adults would demonstrate  150 stronger ideas related to the importance of leaders acting as Models and Mentors as they may have learned about leadership from adult role models in their lives. However, Directive Adult Support was a significant predictor of Modeling and Mentoring, Strong Character, Motivated and Ambitious, and the overall Leadership scale, and Facilitative Adult Support was not a significant predictor in any of the models. More clarity about this concept and how it plays out in youth leadership programs may increase our understanding of the relationship between adult support  t interactions with adults in leadership roles leadership. Past research shows that the family context influences youth civic engagement and political participation (Andolina, Jenkins, Zukin, and Keeter, 2003; Cicognani, Zani, Fournier, Gavray, & Born, 2012; Schulz, Ainley, Fraillon, Kerr, & Losito, 2010; Amna, 2012; Watts & Guessous, 2006). For example, Andolina and colleagues (2003) found that in homes where family members volunteered and in homes where politics were discussed, youth were highly involved in civic life and more attentive to and engaged in politics. Additional background ould provide a stronger understanding of how adults in their lives may influence their thoughts on leadership.  Empowerment.  In all of the regression models, Empowerment was a significant predictor of how youth define leadership subscales and the overall leadership variable that represented the empowerment develops from participating in decision making, having control over issues in their lives, and engaging in empowering settings (Zimmerman, 1995; Maton & Salem, 1995), which are common components of youth leadership programs. In this study, it appears feeling  151 empowered is related to how youth define leadership according to this model. Youth that felt more empowered were more likely to believe leadership was focused on working collaboratively to create change, involved being a role model and mentor to others, and required motivation and reate change and feel empowered to act as leaders to promote continued involvement in leadership roles in the community.  Limitations and Implications for Future Research. Though this study provides evidence for the Collective Change Youth Leadership Framework as a potentially useful model for youth leadership, the findings should be interpreted while considering a number of limitations to the study.   First, one concern is that it is possible there could be another configuration of the model that would fit just as well as or in addition to the proposed model with the current sample or a different sample of youth (Loehlin, 2004; Lee & Hershberger, 1990; MacCallum, Wegener, Uchino, & Fabringar, 1993; Raykov & Marcoulides, 2001). The discussion or exploration of competing alternative models is largely ignored in empirical studies that employ structural equation modeling (MacCallum, et al., 1993) despite the topic being studied by statisticians for decades (Raykov & Marcoulides, 2001). Currently available fit indices do not allow researchers to definitively select the most appropriate model, so the use of non-statistical information (e.g. data collection, design, theory) is often needed to help justify selecting one model over another (Raykov & Marcoulides, 2001). One alternative model was tested by creating random item parcels and running the first order CFA, and while some of the fit indices did meet minimum  152 acceptable thresholds, the conceptual model appeared to be a better fit overall1. Additional research could explore various configurations of the model and see how these competing models compare to the proposed model. Also, testing the model in a more general population of youth, including those who were not engaged in a leadership training program, may suggest a different configuration of the model, or point to the need to learn more about this concept from a more diverse sample.  Next, this study included secondary survey data from a national sample of youth from 30 states to understand how they define leadership and how various experiences and beliefs relate to their ideas on leadership according to the Collective Change Youth Leadership Framework.  The sample included in this study was selected because of previous collaborations with CADCA, and the convenience of surveying a large group of youth at a single event. Youth were recruited to participate in the survey because they attended a week-long leadership training conference.  Though these youth were primed to think about leadership due to their participation in the conference, the fact that the majority of youth were invited to attend the conference because they were thought to have leadership potential or they participated in their local substance abuse coalition is a major limitation of this study. In addition, the youth in this study were past school year. Youth who attend leadership programs or engage in opportunities to act as leaders may be more high-achieving than youth who do not (Kress, 2006). The youth engaged in the two prior studies that generated ideas about leadership were also recruited from similar                                                            1 A four-factor model that combined the Strong Character and Motivated and Ambitious subscales into one factor was also tested. The model demonstrated similar fit with the five factor model 2 77, N=147) = 135.24, p<.001; normed 2 =1.69; CFI= .96, TLI= .95, RMSEA = .069, CD=.99), but the proposed five factor model was retained because it aligned with the dimensions put forth by the youth in previous studies.  153 leadership trainings, so the statements included on the survey about leadership may reflect the views of a small segment of youth leaders engaged in formal training programs. These ideas could vary from youth engaged in other types of leadership roles, or youth who are not currently acting as leaders. The sample included here may limit the generalizability of the study to the larger youth population, and future research should include youth in varying leadership roles as well as youth who do not identify as leaders.  The youth in this study did show variability in their demographic backgrounds, as well as previous experience as leaders and with leadership training. Compared to youth (under age 18) in the United States, the sample that was included in this study was quite representative in terms of race (White: 52% US, 61% study; African American: 14% US, 12% study; Hispanic 25% US, 15% study; American Indian/Alaskan Native: 1% US, 3% study; Asian/Pacific Islander 5% US, 3% study; Multiracial 4% US, 3% study; U.S. Census Bureau, 2015) and SES status, with 41% of youth (or their siblings) eligible for free or reduced lunch in this study compared to 49.6% nationally (National Center for Education Statistics, 2013). This demonstrates some variability across the sample, but future research should include an even larger sample of youth who were not surveyed because they were attending a leadership conference. A larger sample would also allow for comparisons across different types of youth, for example across race, geography, academic performance, previous leadership experience, etc., to learn how other background characteristics may relate to youth definitions of leadership. Understanding how youth from a more general population define leadership, and if the Collective Change Youth Leadership Framework holds in a broader sample would provide valuable information for youth leadership development programs so programs can be tailored to meet the needs of youth.    154 Another limitation to this study was the cross sectional design.  This study provides a clearer understanding of how youth define leadership based on this model at one point in their life, but it does not allow for an understand of how definitions of leadership may change over time or be affected by later life experiences or mindset shifts.  Youth development is an ongoing process, and youth change their views and gain experiences at a rapid rate (Eccles & Harold, 1993), so a longitudinal study of youth leadership would provide a deeper understanding of how age and experience may relate to leadership definitions. This more refined understanding of how definitions of leadership may shift over time could influence youth leadership development programs for youth with various past experiences or youth of different ages to better meet their leadership needs at different points in their lives.  Finally, as with any new tool, constructing a measure and testing a new framework has challenges and limitations. Measurement error is a common source of error in surveys, and occurs because of characteristics of participants, questions, and questionnaires (Dillman, 1991). Satisficing is a common issue with self-administered surveys, in which participants select responses that are satisfactory or acceptable, and may not expend as much psychological energy as needed to answer a question optimally and consider each response option (Stern, Dillman, and Smyth, 2007; Krosnick & Alwin, 1987). Also, research has shown that participants often select responses that are early in a list (primacy effects) on written surveys (Dillman, et al., 1995; Krosnick & Alwin, 1987), but that was not the case in this survey, as the most commonly agreeable (Singleton & Straits, 2005), may have also contributed to measurement error in this study as evidenced by the high averages for the various dimensions of the youth leadership  155 framework. All of the scales averaged above 4.0 on a one to five scale, suggesting youth tended to select responses on the positive end of the likert scale (e.g. Important or Very Important) for most items on the survey. Measurement error associated with the new framework that was tested in the current study may have affected the outcomes of this survey. Future work could alter the visual layout of the survey, perhaps breaking up the questions related to leadership definition, or reducing the number of items if a factor analysis confirmed the items loaded strongly onto the proposed dimensions as strategies for reducing measurement error and participant burden in completing the survey.  Opportunities for Future Research. While this study includes a number of limitations, it also offers a variety of opportunities for future research. The framework presented here was developed based on ideas about leadership from youth engaged in a leadership training program. program that used the same model for training them as leaders. Future research could explore whether or not this framework fits for youth who engage in more emergent leadership positions, such as becoming involved in the school or community as a reaction to negative events in their lives. This model could also be tested in untraditional settings such as gangs or athletic teams, where youth may be seen as leaders in that context, but perhaps not in other settings. Testing this model in additional settings with a variety of youth leaders would help us understand if the proposed subscales are general aspects of leadership that are important to youth, or if the model should be expanded to more accurately reflect the ideas of different youth, or if separate models of leadership should be created for different types of settings. The model presented here provides a springboard for numerous avenues that could be studied by engaging a wide variety of youth in diverse leadership roles and settings.        156 Implications for Practice. Despite the limitations of the current study, the findings have potentially important implications for practice.  The creation of a framework for youth leadership has potential to impact future leadership development efforts in a way that could encourage more youth engagement and produce more youth leaders. Research on youth engagement tells us youth are more likely to attend programs when they align with their desires and interests (Perkins, et al., 2007), and programs that are youth-driven are more effective at empowering youth to use their skills beyond the program setting (Edelman, et al., 2004).  This study, built upon the two prior studies that developed of the Collective Change Youth Leadership Framework, points to five dimensions of leadership that are important to youth.  The current study addressed some of the limitations of the previous studies (e.g. homogenous sample, small sample size), and provided support for this model as a guide for youth leadership programs to be more responsive to youth and supportive of their development as leaders. As discussed in the introduction, there is no consistent definition of leadership guiding programs for youth (Ricketts & Rudd, 2002; Libby, Sedonaen, & Bliss, 2006; Edelman, et al., 2004; Zeldin & Camino, 1999), and no theories have been established for youth leadership like those for adult leadership. Feminist Standpoint Theory (Campbell & Wasco, 2000; Intemann, ideas about leadership are likely to differ as well. The proposed framework begins to fill that gap in the literature, offering a model of youth leadership that could be further explored in larger samples and used to develop programs for youth.   Based on the Collective Change Youth Leadership Framework, future youth leadership programs should focus on the five dimensions of leadership that youth have identified. First, youth should be given meaningful opportunities to create change in their communities.  Youth  157 see positive change as the goal of leadership, so programs should offer opportunities to be could be through existing committees or school or community groups that focus on issues youth value.  Giving them the chance to affect change may empower youth to continue to act as leaders both now and in the future.  In working to create change, youth value collective action, so these efforts should be collaborative, allowing youth to work together and encouraging partnerships with adults and organizations that will support youth in their efforts.   Programs that promote youth leadership should also provide opportunities for youth to be role models and mentors to other youth.  One piece of the Modeling and Mentoring component of the framework is passing along skills to others.  Youth in leadership roles could act as mentors for younger youth looking to develop their skills and become leaders as well.  Youth value modeling and mentoring as an important part of being a leader, so this should be incorporated regularly into leadership roles for youth.   Having strong character and being motivated and ambitious were key pieces to the leadership model according to youth as well.  These components could be incorporated as discussion points in a training on youth leadership, allowing youth to talk about characteristics they could work on strengthening as leaders, and talk about ways to show these characteristics in their leadership positions.   CONCLUSION Overall, this study provides empirical support for a youth-defined model of leadership, leadership.  This study begins to fill some of the gaps within the youth leadership literature and provides guidance for developing programs and opportunities that more closely align with the  158 important components of leadership as they are defined by youth.  Additional research is certainly needed to continue to develop the youth leadership literature to reflect the needs and desires of youth today.     159                APPENDICES        160 APPENDIX A Youth Leadership Survey  Thank you for agreeing to participate in this survey! This survey asks your opinion about and experiences with a number of things in your life, including your ideas about leadership, your past experience with leadership, and your involvement in your community. Your answers to these questions will be anonymous. That means no one will know your answers. To help us keep your answers secret, please do not write your name on this survey form.  This survey is voluntary. That means you do not have to take it. If you choose to take it, you may skip   Instructions  1. This is not a test. There are no right or wrong answers.  ly to you, just leave it blank. If you are not sure what it means, raise your hand and we will do what we can to explain what the questions mean.  3. Mark your answers clearly:  Completely fill in the circles.  unless you are instructed to do so.  4. Some of the questions on this survey have the following format:  EXAMPLE  Wechoose one response.  How much do you agree with the following statements? Strongly Disagree Disagree Neither Disagree nor Agree Agree Strongly Agree Pepperoni pizza is my favorite food O O O  O  EXAMPLE  Which of the following best describes the community you live?   O   Suburban O   Rural PLEASE TURN THE PAGE TO BEGIN THE SURVEY!   161 SECTION A Directions: Please answer the following questions about yourself.   1. How old are you?  Age                         2. What is your gender?  O Male O Female      3. What was the last grade you COMPLETED? (e.g., if you are in 11th grade now, then the last grade you completed was 10th grade)? O   6th  O   7th  O   8th  O   9th  O   10th  O   11th  O   12th  O   1st year in college or technical school  O   2nd year in college or technical school     4. Putting them all together, what were your grades like last year?   O       5. Which of these best describes you?   O   White, not Hispanic/Latino O   Black/African American, not Hispanic O   Hispanic/Latino O   American Indian or Alaskan O   Asian or Pacific Islander O   Multiracial (Having parents of more than one race) O   Member of a race not listed above.    6. What county and state is your coalition from?  ________________________________  7. What is the name of your coalition?   ________________________________  8. What county and state do you live in?  ________________________________ 9. Which of the following best describes the community you live? O  Urban O  Suburban O  Rural  162 10. What is the highest grade in school your mother (or female guardian) completed? O   She did not finish high school O   GED O   High school diploma O   Trade/technical school degree O   She had some college -year college) O   -year college)  O   Doctoral Degree (Ph.D., JD, MD) O   Don't know  11. Do you, or any of your siblings, receive free or reduced price lunch at school?  O  No O  Yes   12. National Youth Leadership Initiative (NYLI) before?  O  No O  Yes      13. Have you ever attended the CADCA Mid-Year Training Institute or another CADCA training before?  O  No O  Yes   14. Have you ever participated in a leadership training program before?  This could include any type of training the taught you about leadership or how to be a leader.  These types of trainings can be offered through schools, community organizations, churches, or other community settings.   O  No O  Yes  If YES, please answer question 14a. If NO, skip to question 15.   14a. What was it called? (if it had a name)   _______________________________________  _______________________________________     15. In your opinion, what makes someone a leader? What does leadership look like?  _____________________________________________________________________________________  _____________________________________________________________________________________  _____________________________________________________________________________________  _____________________________________________________________________________________  _____________________________________________________________________________________  163 SECTION B Directions: Not all items will be equally important to everyone, so please consider the importance of each item to your personal ideas about leadership. It may be helpful to skim all of the items before rating them to identify those that are most and least important to your definition of leadership. How important is each of these statements to your definition of leadership? (Please choose one response)  Not Important at All A Little Important Somewhat Important Important Very Important 1. Can be anyone. O O O O O 2. Can act as a leader in many different roles, both large and small. O O O O O 3. Stands up for what they believe in. O O O O O 4. Is someone who takes charge. O O O O O 5. Is someone who inspires others to create change. O O O O O 6. Is authentic (true to themselves) O O O O O 7. Helps other people. O O O O O 8. Cares about others. O O O O O 9. Tries to improve problems in their community. O O O O O 10. Makes decisions for the group O O O O O 11. Works to create positive change. O O O O O 12. Can create and communicate a vision. O O O O O 13. Puts themselves before others. O O O O O 14. Works to create change to address inequalities in the community. O O O O O 15. Is determined and motivated. O O O O O 16. Values social justice. O O O O O 17. Is kind to people. O O O O O 18. Needs followers to achieve their goals O O O O O 19. Leads by example. O O O O O 20. Is a team worker. O O O O O 21. Works with others to achieve a common goal. O O O O O  164  Not Important at All A Little Important Somewhat Important Important Very Important 22. Is clearly distinguished from (e.g. more powerful than) their followers. O O O O O 23. Is positive and optimistic/hopeful. O O O O O 24. Is honest. O O O O O 25. Is forceful when interacting with others. O O O O O 26. Learns from others. O O O O O 27. Tries to build agreement in a group. O O O O O 28. Is able to resolve conflicts. O O O O O 29. Manages other people O O O O O 30. Is happy. O O O O O 31. Respects and tolerates diversity (e.g. people of different races/cultures). O O O O O 32. Knows how to share power and distribute tasks to others. O O O O O 33. Can influence others. O O O O O 34. Must have specific traits (e.g. dominance, extraversion, assertiveness) O O O O O 35. Is responsible. O O O O O 36. Has a strong opinion about change that needs to occur. O O O O O 37. Is open to all ideas/opinions/perspectives. O O O O O 38.  O O O O O 39. Is someone that people look up to. O O O O O 40. Is easy to talk to. O O O O O 41. Is one person or a small group of people O O O O O 42. Provides guidance to others. O O O O O 43. Teaches or demonstrates leadership skills to others. O O O O O 44. Acts as a positive role model for others. O O O O O 45. Has power and authority over others O O O O O 46. Provides support to others. O O O O O 47. Leaves skills behind for new leaders. O O O O O 48. Seeks out role models that have been leaders. O O O O O 49. Is a peacekeeper. O O O O O  165  Not Important at All A Little Important Somewhat Important Important Very Important 50. Leads for the good of themselves. O O O O O 51. Has patience. O O O O O 52. Is respectful of other people. O O O O O 53. Is truthful. O O O O O 54. Can empathize with others (e.g. understand, relate to someone else). O O O O O 55. Is born a leader O O O O O 56. Has confidence in what they do. O O O O O 57. Gives in to peer pressure. O O O O O 58. Is courageous. O O O O O 59. Feels the need to have power O O O O O 60. Motivates and encourages others. O O O O O 61. Is ambitious (e.g. has a strong desire to succeed). O O O O O 62. Is passionate about their work. O O O O O 63. Never follows. O O O O O 64. Is good at setting goals. O O O O O 65. Is loyal. O O O O O 66. Perseveres to achieve goals. O O O O O 67. Values continuous improvement. O O O O O 68. Is charismatic (charming, likeable, energizing, enthusiastic) O O O O O     166 SECTION C Directions: Please answer the following questions about your previous leadership experiences.   1. Have you taken on any leadership roles in the following settings?  These could be formal (e.g. class president, team captain) or informal (e.g. positive role model, leading class projects) positions in which you acted as a leader. Please select all that apply O  School O  Church O  Sports team O  Academic club O  Service club O  Student leadership group (e.g. student council) O  Other extracurricular activity: _______________________________________ O  Community organization O  Job O  Other:  ___________________________________________ O  NONE of the above (please skip to Section D on page 9)  2. Do you currently see yourself as a leader in your community? O  No      O  Yes     3. Thinking about your most recent experience taking on a leadership role, would you consider that a formal or informal leadership role? O  Formal (e.g. class president, team captain)      O  Informal (e.g. positive role model, leading class projects)  4. Thinking about your most recent experience taking on a leadership role, how much control and decision-making power did youth have? (choose one response) No Control or Decision-Making Power Some Control and Decision-Making Power Quite a Bit of Control and Decision-Making Power Complete Control and Decision-Making Power O O O O   5. Thinking about your most recent experience taking on a leadership role, how much control and decision-making power did adults have? (choose one response) No Control or Decision-Making Power Some Control and Decision-Making Power Quite a Bit of Control and Decision-Making Power Complete Control and Decision-Making Power O O O O experience.  For each statement, choose one response.  167 How were you supported by adults in this leadership position? Strongly Disagree Disagree Neither Disagree nor Agree Agree Strongly Agree 6. Adults told youth what to do and were in charge. O O O O O 7. Adults set the goals for the work. O O O O O 8. Adults determined the activities youth world work on. O O O O O 9. Adults pushed and prodded youth. O O O O O 10. Adults taught youth, and explained things to them. O O O O O 11. Adults gave youth options of things to do. O O O O O 12. Adults defined the roles youth would play. O O O O O 13. Adults kept the youth focused on their work. O O O O O 14. Adults repeatedly reminded youth about deadlines. O O O O O 15. Adults directed the work.  O O O O O 16. Adults gave youth the freedom to control their projects. O O O O O 17. Adults allowed youth to learn through trial and error and discovery. O O O O O 18. Adults supported youth having control over activities.  O O O O O 19. Adults supported youth ideas and encouraged them to make decisions. O O O O O 20. Adults provided initial training to youth.  O O O O O 21. Adults offered suggestions and contributed ideas to the youth-driven work. O O O O O 22. Adults alerted youth to problems they had not anticipated. O O O O O 23. Adults provided back up assistance when youth got stuck, or could not do something on their own. O O O O O    168 SECTION D Directions: Please respond to the following statements by selecting how much you agree or disagree with each statement. For each statement, choose one response.  Strongly Disagree Mostly Disagree Slightly Disagree Slightly Agree Mostly Agree Strongly Agree 1. Certain racial or ethnic groups have fewer chances to get a good high school education O O O O O O 2. Poor children have fewer chances to get a good high school education O O O O O O 3. Certain racial or ethnic groups have fewer chances to get good jobs O O O O O O 4. Women have fewer chances to get good jobs O O O O O O 5. Poor people have fewer chances to get good jobs O O O O O O 6. Certain racial or ethnic groups have fewer chances to get ahead O O O O O O 7. Women have fewer chances to get ahead O O O O O O 8. Poor people have fewer chances to get ahead O O O O O O 9. It is a good thing that certain groups are at the top and other groups are at the bottom O O O O O O 10. It would be good if groups could be equal O O O O O O 11. Group equality should be our goal O O O O O O 12. All groups should be given an equal chance in life O O O O O O  169  Strongly Disagree Mostly Disagree Slightly Disagree Slightly Agree Mostly Agree Strongly Agree 13. We would have fewer problems if we treated people more equally O O O O O O  SECTION E Directions: Please rate the extent to which you agree with each of the following items. For each statement, choose one response.  Strongly Disagree Disagree Neither Disagree nor Agree Agree Strongly Agree 1. I am often a leader in groups O O O O O 2. I would prefer to be a leader rather than a follower O O O O O 3. I would rather have a involved in a group project O O O O O 4. I can usually organize people to get things done O O O O O 5. Other people usually follow my ideas O O O O O 6. I find it very easy to talk in front of a group O O O O O 7. I like to work on solving a problem myself rather than wait and see if someone else will deal with it O O O O O 8. I like trying new things that are challenging to me O O O O O 9. I enjoy participation because I want to have as much say in my community or school as possible O O O O O 10. Youth like me can really with my community or school O O O O O 11. I feel like I have a pretty good understanding of the important issues which confront my community or school O O O O O  170  Strongly Disagree Disagree Neither Disagree nor Agree Agree Strongly Agree 12. Youth like me have the ability to participate effectively in community or school activities and decision making O O O O O 13. My opinion is important because it could someday make a difference in my community or school O O O O O 14. There are plenty of ways for youth like me to have a say in what our community or school does O O O O O 15. It is important to me that I actively participate in local teen issues O O O O O 16. Most community or school leaders would listen to me O O O O O 17. Many local activities are important to participate in O O O O O  SECTION F Directions: The following is a list of school, community, and political activities that people can get involved in.  For each statement, choose one response.   For each of these activities, please use the following scale to indicate how often you were involved in each of these activities in the last year.  How often were you involved in each of these activities in the last 12 months? You never did this You did this once or twice  You did this a few times You did this a fair bit You did this a lot 1. Participated in a church-connected group.  O O O O O 2. Participated in an ethnic club or organization.  O O O O O 3. Participated in a political party, club or organization.  O O O O O  171 How often were you involved in each of these activities in the last 12 months? You never did this You did this once or twice  You did this a few times You did this a fair bit You did this a lot 4. Helped organize neighborhood or community events.(e.g., carnivals, hot dog days, potluck dinners)  O O O O O 5. Helped prepare and make verbal and written presentation to organizations, agencies, conferences, or politicians.  O O O O O 6. Did things to help improve your neighborhood (e.g., helped clean neighborhood).  O O O O O 7. Wrote a letter to a school or community newspaper or publication.  O O O O O 8. Attended a demonstration.  O O O O O 9. Collected signatures for a petition drive.  O O O O O 10. Contacted a public official (phone or mail) to tell him/her how you felt about a particular issue.  O O O O O 11. Joined in a protest march, meeting, or demonstration.  O O O O O 12. Got information about community services from a local community information center.  O O O O O 13. Helped people who were new to your country.  O O O O O 14. Worked on a political campaign.  O O O O O 15. Ran for a position in student government. O O O O O    This is the last page.  Thank you!  172 Mid-Year Training Institute Leadership Survey  YOUTH ASSENT FORM  WHAT IS THIS PROJECT AND WHAT WILL HAPPEN IF I PARTICIPATE?  understanding what leadership means to youth. A better understanding of youth leadership from a youth perspective will help both researchers and community-based organizations effectively develop the next-generation of leaders by framing leadership development efforts in ways that resonate with youth definitions of leadership.  For this project, we will ask you to complete an anonymous 30-minute survey to learn more about what leadership means to youth. The survey asks questions about your ideas of what makes someone a leader, your past experiences as a leader, ways you have been involved in your community, and your thoughts on social issues and creating change. The survey also asks some background questions about you (e.g. age, race, gender, school performance). This survey data will be used to better understand how youth define leadership and explore a youth leadership framework that was created based on our prior work with youth.  ARE THERE RISKS?  Participant safety is our highest priority. To ensure safety and anonymity, all of the work you do with us will be anonymous, meaning your name will NEVER be linked with your responses, and information that you individually share with us will be combined with the responses of the other youth participants. We do not anticipate any risks associated with participating in this study.  ARE THERE BENEFITS?  There are no direct benefits associated with your participation, but the information you share with us could inform youth leadership development efforts in your community.  WHAT ABOUT CONFIDENTIALITY?  Your participation in this evaluation is anonymous. Only members of the Michigan State University research team and Institutional Review Board (IRB) will have access to your materials. To protect you, your survey data will be anonymous. That means your name will NEVER be linked with your answers. Your name or individually identifying information will NEVER be used in any reports of the findings. All project data will be stored in locked file cabinets and/or on password protected computers on the campus of Michigan State University for a minimum of three years after the close of the project. Your confidentiality will be protected to the maximum extent allowable by law.  DO I HAVE TO PARTICIPATE?  Your participation is voluntary, you may choose not to participate at all, or you may refuse to participate in certain procedures or answer certain questions or discontinue your participation at any time without penalty or loss of benefits.  WHO DO I CALL WITH QUESTIONS ABOUT THE EVALUATION OR TO REPORT ANY PROBLEMS?  If you have questions about this project or need to report any problems or injuries, please contact the principal investigator, Pennie Foster-Fishman, toll-free at 1-866-343-5279, or e-mail: fosterfi@msu.edu, or regular mail at: 316 Physics Rd., East Lansing, MI 48824.  If you have any questions or concerns about your role and rights as a research participant, would like to obtain information or offer input, or would like to register a complaint about this project, you may contact, anonymously if you wish, Michigan State University Human Research Protection Program at 517-355-2180, FAX 517-432-4503, or e-mail irb@msu.edu, or regular mail at: 408 W. Circle Dr., East Lansing, MI 48824.  This consent form was approved by a Michigan State University Institutional Review Board. Approved 07/09/13  valid through 01/31/2014. This version supersedes all previous versions. IRB # 13-085.   173 ASSENT  Your signature below indicates your voluntary agreement to participate in this study.  ____________________________________________ ____________  Signature of Participant Date  ____________________________________________  First and Last Name (Please Print)     174 APPENDIX B NYLI Advisor Survey  Thank you for agreeing to participate in this survey. This survey asks your opinion about a number of things, including your relationship with the youth in your coalition, information about your coalition, and information about your community. Your answers to these questions will be anonymous. That means no one will know your answers. To help us keep your answers private, please do not write your name on this survey form.  This survey is voluntary. That means you do not   Instructions  1. This is not a test. There are no right or wrong answers.  to you, just leave it blank. If you are not sure what it means, raise your hand and we will do what we can to explain what the questions mean.  3. Mark your answers clearly:    unless you are instructed to do so.  4. Some of the questions on this survey have the following format:  EXAMPLE  choose one response.  How much do you agree with the following statements? Strongly Disagree Disagree Neither Disagree nor Agree Agree Strongly Agree Pepperoni pizza is my favorite food O O O  O  Is English the language you speak at home most of the time?     No  O   Yes  What is the month, day, and year you were born?  Month Day Year Jan Feb Mar Apr May June July Aug Sept Oct Nov Dec  O O O O O O O O O O 0 1 1 9 9 0                                                                    175 SECTION A  experience as an NYLI advisor.  1. How long have you served as a youth advisor for your coalition?    O Never before O  0-6 months O 7-12 months O  1-2 years O 2-3 years O More than 3 years   2. FOR WHICH COALITION ARE YOU AN NYLI ADVISOR? ____________________________________________   3. What county and state is your coalition from? ________________________________________________________________  There are many ways for youth to be involved in the Substance Abuse Coalition in your community. Please answer the following questions about youth involvement in their coalition and your community.  4. How many youth are regularly involved in your coalition?  O        0      O        1 O        2 O        3 O        4 O        5 O        6 O        7 O        8 O        9 O        10 O        11 or more  5. How old are these youth? (check all that apply)  O  13-14 O  15-16 O  17-18 O  19-22 Please think about  involvement with your coalition.  176  In the past 12 MONTHS, HOW MANY of the NYLI YOUTH  None A few Most All 6. Attended coalition meetings? O O O O 7. Talked at coalition meetings (made comments, expressed ideas, etc.)? O O O O 8. Helped organize coalition activities (other than meetings)?  O O O O 9. Served as an officer (e.g., treasurer, secretary, workgroup/committee chair or lead)?  O O O O 10. Had a voice in coalition decision-making? O O O O 11. Had a role in determining the direction efforts? O O O O 12. Had opportunities to be a leader in your coalition? O O O O 13. Became official members of your coalition? O O O O   SECTION B  The next set of questions ask about your local Substance Abuse Coalition.  If you have never attended a Substance Abuse Coalition meeting or are unfamiliar with your local coalition, please go to Section C.     1. Some coalitions focus on programs or services, others focus on changing aspects of their community environment. Which of the following do you think best describes the current status of your coalition (choose the one answer that best describes your coalition today):  O     Exclusively providing or coordinating the provision of prevention programs or services. O     Primarily providing or coordinating the provision of prevention programs, with some involvement in environmental or policy change. O     About evenly split between prevention programs/services and environmental change. O     Primarily focusing on environmental or policy change, with some involvement in prevention programs and services. O     Exclusively focusing on environmental or policy change.      When was your coalition first started?    ________Month  ________Year  O       177 2. How does your coalition identify itself (i.e., which definitions best reflect your coalition regarding scope) (check all that apply)  O    Faith-based  O     Prevention-focused O     Primarily serving a special population(s) (e.g., Latino youth, refugee community, people with disabilities, etc.).   O     School-based O     Serving an economically disadvantaged community O     Recovery-focused O     Treatment focused O     Other (specify)__________________________ O     Other (specify)__________________________   3. Which category BEST describes the geographic area that your coalition targets (i.e., where you focus your effort; population you want to impact)? (MARK ONLY ONE)  O     City O     Congressional District  O     County  O     Indian Reservation  O     Neighborhood/Community  O     Multiple Counties  O     Multiple Neighborhoods O     School District  O     School Site  O     Multiple School Sites O     Regional  O    Statewide O    Other (please specify)_______________________________________________________    178 Now we would like to know a bit more about you and your involvement in your substance abuse coalition.   3. Which do you represent on your local substance abuse coalition? (MARK ONLY ONE)?  O  Parent or Family member O  Law Enforcement (e.g. Police Department, Courts, Probation) O Organization that provides services and supports to youth (e.g. Afterschool club) O  Business Sector O  Philanthropic and/or United Way O Faith-Based Community O Community-Based Organization (e.g., Neighborhood Center, Grassroots Organization, Ethnic Group/Organization) O Elected City or Government Official (e.g. Local, County, Tribal, State, Federal) O  Human Services Organization (e.g. Hospitals, Public Health, Public Assistance, Housing Authority, Employment Services) O  School System (PreK-12) O  Volunteer Organization  O  Interested citizen  6. How long have you been involved in your coalition (MARK ONLY ONE)? O     Not involved O     0-6 months O     7-12 months O     13-18 months O     19-24 months O     2-3 years O     Over 3 years    7. How many coalition meetings did you attend in the past 12 months?   __________Coalition Meetings  8. How many coalition workgroups or committees have you served on in the past 12 months?  __________Workgroups/Committees   9. How many coalition workgroup or committee meetings did you attend in the past 12 months?   __________Workgroup/Committee Meetings  179 SECTION C Now we would like to learn about the youth attending the Mid-Year Training Institute from your coalition.  10. How were youth from your coalition selected to attend the Mid-Year Training Institute? (check all that apply) O   Recruited because they have leadership potential  O   Youth participate in our local substance abuse coalition  O   Recruited from school O    O   Other: ________________________________________________________________________________________ O   Other: ________________________________________________________________________________________  11. How many youth from your coalition/county were invited to attend the Mid-Year Training Institute? O        0      O        1 O        2 O        3 O        4 O        5  O        6 O        7 O        8 O        9 O        10 O        11 or more 12. How many youth are attending? __________________________ 13. Did the youth know each other before attending the Mid-Year Training Institute? O   No, none of the youth knew each other O   Yes, some of the youth knew each other O   Yes, all of the youth knew each other    180 SECTION D  To help us understand more about our participants, we would like to know some background information about you. Please answer the following questions.Education Background     1. What is the highest degree you have received? (Circle one)  O Did not graduate from high school O GED O High school diploma, trade, or training certificate O Certification of Completion O  O  O  O Ph.D., M.D. or J.D. O Other  _____________________________  2. If you attended school after high school, what type of schooling was it? (circle all that apply)  O Training at my work O Trade or vocational school O Community college/2-year college O 4-year college  3. Are you currently a student?  O YES (Answer Question 3a) ONO (Skip to Question 4)  3a. What type of degree or certificate are you trying to obtain?  O GED O Specialized training  O  O  O Other _________________________  Economic Background   4. Are you currently employed?  O YES   ONO  (go to question 8)  5. How many jobs do you have?      O 1 O 2 O 3 O 4 O 5 O 6 or more  6. How long have you had regular employment?  O 0-6 months O 7-12 months O 13-18 months O 19-24 months O 2-3 years O 3 or more years  7. On average, how many hours do you work each week:     O 0-5 hours O 6-10 hours O 11-15 hours O 16-20 hours O 21-25 hours O 26-30 hours O 31-35 hours O 36-40 hours O More than 40 hours   181 8. Other than your Substance Abuse Coalition, are you a volunteer in other activities in your community?  O YES   O NO     9. Excluding your involvement in your Substance Abuse Coalition, how many hours do you volunteer a month?       O 0-5 hours O 6-10 hours O 11-15 hours O 16-20 hours O 21-25 hours O 26-30 hours O 31-35 hours O 36-40 hours O More than 40 hours   10. What is your annual income?   O Less than $10,000 O $10,000 to $15,000 O $15,000 to $20,000 O $20,000 to $25,000 O $25,000 to $35,000 O $35,000 to $45,000 O $45,000 to $60,000 O $60,000 to $80,000 O $80,000 and above          Your Background  11. What is your gender? O Female O Male  12. What is your marital status?  O Single O Living with male or female companion O Married O Separated O Divorced O Widowed   13. What is the month and year you were born?  Month Year  Jan Feb Mar Apr May June July Aug Sept Oct Nov Dec   O  O  O  O O O O O O O O O 1 9                                                     182 14.  What is your racial background?  (Mark ALL that apply)  O White (Not Hispanic/Not Latino) O Black or African American  O American Indian or Alaska Native   O Native Hawaiian or Pacific Islander O Asian O Hispanic or Latino O Multiracial (Having parents of more than one race) O Member of race not listed above:  __________________  15. Do you consider yourself a leader in your community or local substance abuse coalition?   O YES (If yes, go to Question 15a)    O NO (If no, go to Question 16)    15a. How many years have you been a leader in your community or local substance abuse coalition?   O Less than 1 year  O 1-5 years O 6 -10 years O Over 10 years   16. How good do you think you are as a leader?  O Not good at all O A little O Somewhat  O Quite Good O Great  17. Have you ever participated in a leadership training?  O YES     O NO 17a. If YES, which ones:  _________________________________ _________________________________ _________________________________   18. Have you ever mentored youth before?  O YES  (If yes, go to Question 19)  O NO   (If no, go to Question 21)   19. Overall, would you say mentoring was a positive experience?   O Not at all O A little O Somewhat O Quite a Bit O A great Deal  O Completely  20. How good do you think you are as a mentor?  O Not good at all O A little O Somewhat  O Quite Good O Great     21. What county and state do you live in? ____________________________________  22. How long have you lived there? ____________________________________This is the last page.  Thank you!  183 Mid-Year Training Institute Advisor Survey  ADVISOR CONSENT FORM   WHAT IS THIS PROJECT AND WHAT WILL HAPPEN IF I PARTICIPATE?  understanding what leadership means to youth. A better understanding of youth leadership from a youth perspective will help both researchers and community-based organizations effectively develop the next-generation of leaders by framing leadership development efforts in ways that resonate with youth definitions of leadership.  For this project, we will ask you to complete an anonymous 20-minute survey to learn more about the youth from your coalition that are participating in the Mid-Year Training. The survey asks questions about how youth were selected to participate, how many youth are attending, as well as some background questions about you (e.g. age, race, gender) and your community. This survey data will be used to better understand who is attending the Mid-Year Training Institute.   ARE THERE RISKS?  Participant safety is our highest priority. To ensure safety and anonymity, all of the work you do with us will be confidential, meaning your name will NEVER be linked with your responses, and information that you individually share with us will be combined with the responses of the other participants. We do not anticipate any risks associated with participating in this study.   ARE THERE BENEFITS?  There are no direct benefits associated with your participation, but the information you share with us could inform youth leadership development efforts in your community and other communities.   WHAT ABOUT CONFIDENTIALITY?  Your participation in this evaluation is anonymous. Only members of the Michigan State University research team and Institutional Review Board (IRB) will have access to your materials. To protect you, your survey data will be anonymous. That means your name will NEVER be linked with your answers. Your name or individually identifying information will NEVER be used in any reports of the findings. All project data will be stored in locked file cabinets and/or on password protected computers on the campus of Michigan State University for a minimum of three years after the close of the project. Your confidentiality will be protected to the maximum extent allowable by law.   DO I HAVE TO PARTICIPATE?  Your participation is voluntary, you may choose not to participate at all, or you may refuse to answer certain questions or discontinue your participation at any time without penalty or loss of benefits.   WHO DO I CALL WITH QUESTIONS ABOUT THE EVALUATION OR TO REPORT ANY PROBLEMS?  If you have questions about this project or need to report any problems or injuries, please contact the principal investigator, Pennie Foster-Fishman, toll-free at 1-866-343-5279, or e-mail: fosterfi@msu.edu, or regular mail at: 316 Physics Rd., East Lansing, MI 48824.  If you have any questions or concerns about your role and rights as a research participant, would like to obtain information or offer input, or would like to register a complaint about this project, you may contact, anonymously if you wish, Michigan State University Human Research Protection Program at 517-355-2180, FAX 517-432-4503, or e-mail irb@msu.edu, or regular mail at: 408 W. Circle Dr., East Lansing, MI 48824.   184 This consent form was approved by a Michigan State University Institutional Review Board. Approved 07/09/13  valid through 01/31/2014. This version supersedes all previous versions. IRB # 13-085.     CONSENT Your signature below indicates your voluntary agreement to participate in this study.  ____________________________________________ ____________  Signature of Participant Date  ____________________________________________  First and Last Name (Please Print)   185  APPENDIX C   Youth Leadership Subscale Item Correlation Tables Table C1.  Anyone can be a Leader Item Correlations   1 2 1. Can be anyone 1  2. Can act as a leader in many different roles, both large and small .30** 1  Note: *p>.05, **p<.001     186 Table C2.  Creating Change Item Correlations  Note: *p>.05, **p<.001  3 4 5 7 8 9 11 12 14 16 36 3. Stands up for what they believe in 1           4. Is someone who takes charge .13 1          5. Is someone who inspires others to create change .26* .45** 1         7. Helps other people .31** .11 .24* 1        8. Cares about others .25* .01 .25* .75** 1       9. Tries to improve problems in their community .19* .27** .37** .35** .40** 1      11. Works to create positive change .39** .13 .34** .30** .25* .40** 1     12. Can create and communicate a vision .18* .17* .25* .07 .00 .31** .35** 1    14. Works to create change to address inequalities in the community .15 .18* .18* .24* .28** .43** .38** .35** 1   16. Values social justice .37** .22* .19* .50** .33** .37** .42** .34** .34** 1  36. Has a strong opinion about change that needs to occur .39** .37** .29** .25* .20* .32** .30** .42** .27** .23* 1  187  Table C3. Collective Action Item Correlations  Note: *p>.05, **p<.001   17 20 21 26 27 28 31 32 38R 63R 17. Is kind to people 1          20. Is a team worker  .24* 1         21. Works with others to achieve a common goal .13 .63** 1        26. Learns from others  .24** .30** .34** 1       27. Tries to build agreement in a group  .30** .26** .29** .50** 1      28. Is able to resolve conflicts  .31** .37** .37** .41** .47** 1     31. Respects and tolerates diversity (e.g. people of different races/cultures)  .38** .38** .15 .20* .28** .44** 1    32. Knows how to share power and distribute tasks to others .35** .33** .18** .37** .37** .39** .49** 1   others .13 -.01 .02 -.08 -.06 -.03 .03 -.01 1  63R. Never follows .00 -.08 .01 .09 .03 -.02 -.06 -.11 .30** 1  188 Table C4. Modeling and Mentoring Item Correlations Note: *p>.05, **p<.001  19 25R 33 37 39 40 42 43 44 46 47 48 19. Leads by example 1            25R. Is forceful when interacting with others -.14 1           33. Can influence others .08 -.25* 1          37. Is open to all ideas/opinions/ perspectives  .20* .02 .23* 1         39. Is someone that people look up to .15 -.14 .25* .20* 1        40. Is easy to talk to  .19* -.10 .29** .38** .36** 1       42. Provides guidance to others  .36** -.13 .40** .42** .23* .26* 1      43. Teaches or demonstrates leadership skills to others .41** -.07 .33** .35** .30** .32** .62** 1     44. Acts as a positive role model to others .40** -.02 .13 .51** .31** .46** .37** .50** 1    46. Provides support to others .28** .01 .22** .61** .28** .43** .50** .48** .73** 1   47. Leaves skills behind for new leaders .24* -.17* .28** .29** .21* .22* .36** .30** .36** .46** 1  48. Seeks out role models that have been leaders .39** -.21* .36** .34** .35** .31** .50** .51** .44** .45** .52** 1  189 Table C5.  Strong Character Item Correlations  Note: *p>.05, **p<.001  13R 23 24 30 35 49 50R 51 52 53 54 57R 65 13R. Puts themselves before others 1             23. Is positive and optimistic/hopeful .00 1            24. Is honest .05 .49** 1           30. Is happy  -.06 .46** .25* 1          35. Is responsible  .08 .55** .46** .38** 1         49. Is a peacekeeper  .11 .41** .28** .42** .27** 1        50R. Leads for the good of themselves  .23* -.05 .04 -.23** .04 -.20* 1       51. Has patience .14 .47** .18* .34** .45** .29** -.11 1      52. Is respectful of other people .14 .47** .32** .31** .56** .43** -.04 .51** 1     53. Is truthful .07 .52** .44** .42** .60** .27** -.06 .55** .62** 1    54. Can empathize with others (e.g. understand, relate to someone) .03 .49** .25* .29** .35** .20* .04 .25* .36** .32** 1   57R. Gives in to peer pressure .17* .07 .08 -.08 .04 -.10 .34** .12 .04 .08 -.07 1  65. Is loyal .05 .57** .42** .39** .60** .46** -.08 .54** .60** .56** .37** .05 1  190  Table C6:. Motivated and Ambitious Item Correlations   Note: *p>.05, **p<.00  15 56 58 60 61 62 64 66 67 15. Is determined and motivated 1         56. Has confidence in what they do .38** 1        58. Is courageous  .26* .40** 1       60. Motivates and encourages others  .23* .32** .27** 1      61. Is ambitious (e.g. has a strong desire to succeed)  .10 .41** .24* .30** 1     62. Is passionate about their work  .21* .28** .27** .23* .47** 1    64. Is good at setting goals .17* .48** .40** .13 .34** .35** 1   66.  Perseveres to achieve goals .39** .30** .35** .20* .21* .32** .30** 1  67. Values continuous improvement .37** .47** .39** .30** .33** .39** .43** .55** 1  191  Table C7. Youth Leadership Subscale Correlations           Note: *p>.05, **p<.00  Creating Change Collective Action Modeling & Mentoring Strong Character Motivated & Ambitious Creating Change 1     Collective Action .71** 1    Modeling & Mentoring .67** .73** 1   Strong Character .64** .69** .81** 1  Motivated & Ambitious .62** .62** .73** .64** 1  192           REFERENCES    193 REFERENCES  Abramowitz, M. and Stegun, I.A., eds. (1965). Handbook of Mathematical Functions. New York, NY: Dover.  Forum. Washington, DC.  Amnå, E. (2012). How is civic engagement developed over time? Emerging answers from a multidisciplinary field. Journal of adolescence, 35(3), 611-627. Andolina, M. W., Jenkins, K., Zukin, C., & Keeter, S. (2003). Habits from home, lessons from school: Influences on youth civic engagement. Political Science and Politics, 36(02), 275-280. Assari, S., Rezazade, M., Ahmadi, K., & Sehat, M. (2014). Socio-economic status may suppress the effect of knowledge on sexual risk among female sex workers. International Journal of Health & Allied Sciences, 3(2), 84. Astin, H.S., & Astin, A.W. (1996). A social change model of leadership development: Guidebook (3rd ed). Los Angeles, CA: Higher Education Research Institute, University of California. Avolio, B. J., & Gardner, W. L. (2005). Authentic leadership development: Getting to the root of positive forms of leadership. The Leadership Quarterly, 16(3), 315-338. Avolio, B. J., Walumbwa, F. O., & Weber, T. J. (2009). Leadership: Current theories, research, and future directions. Annual Review of Psychology, 60, 421-449. Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory. Englewood Cliffs, NJ: Prentice-Hall. Bandura, A. (2001). Social cognitive theory: An agentic perspective. Annual review of psychology, 52(1), 1-26. Bandura, A., Barbaranelli, C., Caprara, G. V., & Pastorelli, C. (2001). Selfefficacy beliefs as shapers of children's aspirations and career trajectories. Child development, 72(1), 187-206. Bass, B. M. (1985). Leadership and performance beyond expectations. New York: Free Press. Bass, B. M. (1999). Two decades of research and development in transformational leadership. European Journal of Work and Organizational Psychology, 8(1), 9-32. Bass, B. M., & Avolio, B. J. (1993). Transformational leadership: A response to critiques. In M. M. Chemers & R. Ayman (Eds.), Leadership theory and research: Perspectives and directions (pp. 4980). New York: Free Press.  194 Beaman, L., Duflo, E., Pande, R., & Topalova, P. (2012). Female leadership raises aspirations and educational attainment for girls: A policy experiment in India. Science, 335(6068), 582-586. Benson, P. L., & C. Scales, P. (2009). The definition and preliminary measurement of thriving in adolescence. The Journal of Positive Psychology, 4(1), 85104.  Bentler, P.M. and Bonnet, D.C. (1980), "Significance Tests and Goodness of Fit in the Analysis of Covariance Structures," Psychological Bulletin, 88 (3), 588-606. Bentler, P. M., & Chou, C. P. (1987). Practical issues in structural modeling. Sociological Methods & Research, 16(1), 78-117. Biddle, C., & Mitra, D. (2015). Implementing Middle School Youth-Adult Partnerships: A Study of Two Programs Focused on Social Change. Middle Grades Review, 1(2), 6. Blakemore, J.E.O., Berenbaum, S.A., & Liben, L.S. (2008). Gender development. New York, NY: Psychology Press. Bloomberg, L., Ganey, A., Alba, V., Quintero, G., & Alvarez Alcantara, L. (2003). Chicano-Latino youth leadership institute: An asset-based program for youth. American Journal of Health Behavior, 27(1), S45-S54. Bollen, K. A. (1989). A new incremental fit index for general structural equation models. Sociological Methods & Research, 17(3), 303-316. Boomsma, A. (1982). The robustness of LISREL against small sample sizes in factor analysis models. In K. G. Jöreskog & H. Wold (Eds.), Systems under indirect observation: Causality, structure, prediction (Part 1, pp. 149173). Amsterdam: North-Holland. Borden, L., & Serido, J. (2009). From program participant to engaged citizen: a developmental journey. Journal of Community Psychology, 37(4), 423-438. Boyd, B. L. (2001). Bringing leadership experiences to inner-city youth. Journal of Extension, 39(4), 1-5. Browne, M. W., & Cudeck, R. (1993). Alternative ways of assessing model fit. Sage Focus Editions, 154, 136-136. Bryant, F. B., & Yarnold, P. R. (1995). Principal components analysis and exploratory and confirmatory factor analysis. In L. G. Grimm & R R. Yarnold (Eds.), Reading and understanding multivariate statistics (pp. 99-136). Washington, DC: American Psychological Association. Bukowski, W. M., Gauze, C., Hoza, B., and Newcomb, A. F. (1993). Differences and consistency between same-sex and other-sex peer relationships during early adolescence. Develop. Psychol. 29: 255263.  195 Burgess, D., & Borgida, E. (1999). Who women are, who women should be: Descriptive and prescriptive gender stereotyping in sex discrimination. Psychology, Public Policy, and Law, 5(3), 665-692. Buss, D. M. (1995). Psychological sex differences: Origins through sexual selection. American Psychologist, 50, 164-168. Bussey, K., & Bandura, A. (1999). Social cognitive theory of gender development and differentiation. Psychological review, 106(4), 676. Camino, L., & Zeldin, S. (2002). From Periphery to Center: Pathways for Youth Civic Engagement in the Day-To-Day Life of Communities. Applied Developmental Science, 6(4), 213220.  Campbell, R., & Wasco, S. M. (2000). Feminist approaches to social science: epistemological and methodological tenets. American Journal of Community Psychology, 28(6), 77391.  Cassell, J., Huffaker, D., Tversky, D., & Ferriman, K. (2006). The language of online leadership: Gender and youth engagement on the internet. Developmental Psychology, 42(3), 436-449.  Cattell, R. B. (1978). The Scientific Use of Factor Analysis. New York: Plenum. Chan, W. Y., Ou, S. R., & Reynolds, A. J. (2014). Adolescent civic engagement and adult outcomes: An examination among urban racial minorities. Journal of youth and adolescence, 43(11), 1829-1843. Chang, B. (2016, July 14). Chicago Magazine. Retrieved from http://www.chicagomag.com/city-life/July-2016/Black-Lives-Matter-Chi-Youth-Sit-In-Rally/.  Charmaz, K. (2003). Grounded theory. Strategies of Qualitative Inquiry, 2, 249. and youth into human settlement development. Environment and Urbanization, 14(2), 11-22. Checkoway, B. (2011). What is youth participation?. Children and Youth Services Review, 33(2), 340-345. Checkoway, B. (2015). Research as community-building: Perspectives on the scholarship of engagement. Gateways: International Journal of Community Research and Engagement, 8(1), 139-49. Checkoway, B., & Richards-Schuster, K. (2003). Youth Participation in Evaluation Research. American Journal of Evaluation, 24(1), 2133. Chen, P., Weiss, F. L., & Nicholson, H. J. (2010). Girls study Girls Inc.: Engaging girls in evaluation through participatory action research. American Journal of Community Psychology, 46(1-2), 228-237.  196 Chin, W. W. (1998). Commentary: Issues and opinion on structural equation modeling. MIS quarterly, vii-xvi. Chin, W. W., & Newsted, P. R. (1999). Structural equation modeling analysis with small samples using partial least squares. Statistical strategies for small sample research, 1(1), 307-341. Christens, B. D., & Dolan, T. (2011). Interweaving Youth Development, Community Development, and Social Change Through Youth Organizing. Youth & Society, 43(2), 528548.   Chronister, K. M. & McWhirter, E. H. (2006). An experimental examination of two career interventions for battered women. Journal of Counseling Psychology, 53(2), 151-164. Cicognani, E., Zani, B., Fournier, B., Gavray, C., & Born, M. (2012). Gender differences in social and civic participation. Journal of adolescence, 35(3), 561-576. Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd Edition). Hillsdale, NJ: Lawrence Earlbaum Associates.  Cohen, J., Cohen, P., West, S.G., and Aiken, L.S. (2003). Applied multiple regression/correlation analysis for the behavioral sciences (3rd edition). Mahwah, NJ: Lawrence Earlbaum Associates. Comrey, A. L., & Lee, H. B. (1992). A first Course in Factor Analysis. Hillsdale, NJ: Erlbaum. Community Anti-Drug Coalitions of America (CADCA). (2009). National Youth Leadership Initiative. Retrieved from http://www.cadca.org/trainingevents/training_coalitions/national-youth-leadership-initiative.  Conner, J. O., & Strobel, K. (2007). Leadership development: An examination of individual and programmatic growth. Journal of Adolescent Research, 22(3), 275-297. Corsi, M. 2002. The child friendly cities initiative in Italy. Environment & Urbanization 14(2): 16979. Costanza-Chock, S. (2012). Youth and Social Movements: Key lessons for allies. Berkman Center Research Publication. Culp III, K., & Kohlhagen, B. (2000). Identifying, defining, applying, analyzing & synthesizing leadership opportunities with adolescents. The Journal of Leadership Studies, 7(2), 50-57. Dawes, N.P., and Larson, R. (2011). How youth get engaged: Grounded-theory research on motivational development in organized youth programs. Developmental Psychology, 47(1), 259-269.   197 Denner, J., Meyer, B. and Bean, S. (2005) Young Women's Leadership Alliance: Youth-Adult Partnerships in an All-Female After-School Program. Journal of Community Psychology, 33 (1), 87100. Dempster, N., & Lizzio, A. (2007). Student leadership: Necessary research. Australian Journal of Education, 51(3). Diemer, M. A. (2009). Pathways to Occupational Attainment Among Poor Youth of Color The Role of Sociopolitical Development. The Counseling Psychologist, 37(1), 6-35. Diemer, M.A., Rapa, L.J., Park, C., & Perry, J.C. (2014). Development and validation of the critical consciousness scale. Youth & Society. Dillman, D. A. (1991). The design and administration of mail surveys. Annual Review of Sociology, 225-249. Dillman, D. A., Brown, T. L., Carlson, J. E., Carpenter, E. H., Lorenz, F. O., Mason, R., Saltiel, J. and Songster, R. L. (1995), Effects of Category Order on Answers in Mail and Telephone Surveys. Rural Sociology, 60: 674687. DiPietro, J. A. (1981). Rough and tumble play: A function of gender. Developmental Psychology, 17(1), 50-58. Doll, W. J., Xia, W., & Torkzadeh, G. (1994). A confirmatory factor analysis of the end-user computing satisfaction instrument. MIS Quarterly, 453-461. Durlak, J. A., & Weissberg, R. P. (2007). The impact of after-school programs that promote personal and social skills. Durlak, J. A., Weissberg, R. P., & Pachan, M. (2010). A metaanalysis of afterschool programs that seek to promote personal and social skills in children and adolescents. American journal of community psychology, 45(3-4), 294-309. Eagly, A. H., Johannesen-Schmidt, M. C., & Van Engen, M. L. (2003). Transformational, transactional, and laissez-faire leadership styles: a meta-analysis comparing women and men. Psychological Bulletin, 129(4), 569-591. Eagly, A. H., & Karau, S. J. (2002). Role congruity theory of prejudice toward female leaders. Psychological review, 109(3), 573-598. Eagly, A. H., Makhijani, M. G., & Klonsky, B. G. (1992). Gender and the evaluation of leaders: A meta-analysis. Psychological bulletin, 111(1), 3-22. Eccles, J., & Harold, R. (1993). Parent-school involvement during the early adolescent years. The Teachers College Record, 94(3), 568-587. Edelman, A., Gill, P., Comerford, K., Larson, M., & Hare, R. (2004). Youth Development & Youth Leadership. Elo, S., Kääriäinen, M., Kanste, O., Pölkki, T., Utriainen, K., & Kyngäs, H. (2014). Qualitative content analysis. Sage Open, 4(1), 1-10.  198 Epstein, M., & Ward, L. M. (2011). Exploring parent-adolescent communication about gender: Results from adolescent and emerging adult samples. Sex roles, 65(1-2), 108-118. Erikson, E. H. (1950). Childhood and society. New York: Norton. Erikson, E. H. (1968). Identity: Youth and crisis. New York: Norton. Ewing Lee, E. A., & TroopGordon, W. (2011). Peer socialization of masculinity and femininity: Differential effects of overt and relational forms of peer victimization. British journal of developmental psychology, 29(2), 197-213. Fabes, R. A., Hanish, L. D., & Martin, C. L. (2003). Children at play: The role of peers in understanding the effects of child care. Child development, 74(4), 1039-1043. Fertman, C. I., & van Linden, J. A. (1999). Character education: An essential ingredient for youth leadership development. Nassp Bulletin, 83(609), 9-15. Finn, J. L. (2001). Text and turbulence representing adolescence as pathology in the human services. Childhood, 8(2), 167-191. Flanagan, C. A. (2004). Volunteerism, leadership, political socialization, and civic engagement. Handbook of Adolescent Psychology, 2, 721-745. (Ed.), Handbook of youth and young adulthood: New perspectives and agendas (293-300). New York, NY: Routledge. Flanagan, C. A., & Christens, B. D. (2011). Youth civic development: Historical context and emerging issues. In C. A. Flanagan & B. D. Christens (Eds.), Youth civic development: Work at the cutting edge. New Directions for Child and Adolescent Development, 134, 19. Flanagan, C., & Levine, P. (2010). Civic engagement and the transition to adulthood. The Future of Children, 20(1), 159-179. Flanagan, C. A., & Sherrod, L. R. (1998). Youth political development: An introduction. Journal of social issues, 54(3), 447-456. Foster-Fishman, P., Nowell, B., Deacon, Z., Nievar, M. A., & McCann, P. (2005). Using methods that matter: The impact of reflection, dialogue, and voice. American Journal of Community Psychology, 36(3-4), 275-291. Foster-Fishman, P., Lichty, L., Chen, I., & Mortensen, J. (2012). Shifts in Youth Social Change Attitudes, Skills, and Behaviors following a Year-Long Leadership Training. In Foster-Fishman, P. (Chair), Yang, E. (Discussant), A Participatory and Complimentary Mixed-Method Approach to Evaluating a Youth Leadership Program. Symposium presented at the 26 of the American Evaluation Association, Minneapolis, MN. Frank, K. I. (2006). The potential of youth participation in planning. Journal of Planning Literature, 20(4), 351-371.  199 Freire, P. (1993). Pedagogy of the oppressed (Rev. ed.). New York: Continuum, 1970. French, J., & Raven, B. H. (1959). The bases of social power. In D. Cartwright (Ed.), Studies of social power. Ann Arbor, MI: Institute for Social Research. Ginwright, S., & James, T. (2002). From assets to agents of change: Social justice, organizing, and youth development. New Directions for Youth Development, (96), 2746. Glaser, B. G., & Strauss, A. L. (1967). The discovery of grounded theory: Strategies for qualitative research. London: Wiedenfeld and Nicholson. Gorsuch, R. L. (1983). Factor analysis (2nd ed.). Hillsdale, NJ: Erlbaum. Graen, G.B., & Cashman, J. (1975). A role-making model of leadership in formal organizations: A developmental approach. In J.G. Hunt & L.L. Larson (Eds.), Leadership frontiers (pp. 143-166). Kent, OH: Kent State University Press. Grossman, J. B., Campbell, M., & Raley, R. (2007). Quality after-school time: What instructors can do to enhance learning. Philadelphia, PA: Public/Private Ventures.  Grossman, J. B., Price, M. L., Fellerath, V., Jucovy, L. Z., Kotloff, L. J., Raley, R., & Walker, K. E. (2002). Multiple Choices after School: Findings from the Extended-Service Schools Initiative. Gruber, J., & Trickett, E. J. (1987). Can we empower others? The paradox of empowerment in the governing of an alternative public school. American Journal of Community Psychology, 15(3), 353-371. Guthrie J., Yongvanich K., Ricceri F. (2004). Using content analysis as a research method to inquire into intellectual capital reporting. Journal of Intellectual Capital, 5, 282-293. Hair, J. F. J., Anderson, R. E., Tatham, R. L., & Black, W. C. (1995). Multivariate data analysis (4th ed.). Saddle River, NJ: Prentice Hall. Hair, J.F., Jr.l Anderson, R. E.; Tatham, R. L.; & Black, W. C. (1998). Multivariate data analysis with readings, 5th ed.. Englewood Cliffs, NJ: Prentice-Hall. Hadley, S. & Trechter, D. (2010). Developing leadership within 4-H and UW-Extension youth programs survey report. Hall, J. A., & Friedman, G. B. (1999). Status, gender, and nonverbal behavior: A study of structured interactions between employees of a company. Personality and Social Psychology Bulletin, 25(9), 1082-1091. Hall, G. S. (1904). Adolescence: Its psychology and its relation to physiology, anthropology, sociology, sex, crime, religion, and educa-tion (Vols. 1-2). New York: Appleton. Hall, R. J., Snell, A. F., & Foust, M. S. (1999). Item parceling strategies in SEM: Investigating the subtle effects of unmodeled secondary constructs. Organizational Research Methods, 2, 233256.  200 Halpin, A.W., & Winer, B.J. (1957). A factorial study of the leader behavior descriptions. In R.M. Stogdill and A.E. Coons (eds), Leader behavior: Its description and measurement. Columbus, OH: Bureau of Business Research Ohio State University. Hart, D., & Atkins, R. (2002). Civic competence in urban youth. Applied Developmental Science, 6(4), 227-236. Hicks, D. J. (1968). Effects of co-observer's sanctions and adult presence on imitative aggression. Child Development, 39, 303-309. Hoffman, M. L., & Powlishta, K. K. (2001). Gender segregation in childhood: A test of the interaction style theory. Journal of Genetic Psychology, 162, 298313. Holden D, Messeri P, Evans D, Crankshaw E, & Ben-Davis M. (2004). Conceptualizing youth empowerment within tobacco control. Health Education and Behaviour, 31(5), 548-563. Hooper, D., Coughlan, J., Mullen, M.: Structural Equation Modelling: Guidelines for Determining Model Fit. Electronic Journal of Business Research Methods, 6(1), 53-60. Horner, M. (1997). Leadership theory: past, present and future. Team Performance Management: An International Journal, 3(4), 270-287. Horst, P. (1941). The role of the predictor variables which are independent of the criterion. Social Science Research Council, 48, 431-436. Horst, P. (1966). Psychological measurement and prediction. Belmont, CA: Wadsworth. Hutcheson, G. D., & Sofroniou, N. (1999). The multivariate social scientist: Introductory statistics using generalized linear models. Sage. Independent Sector. (2002). Engaging youth in lifelong service. Washington, DC: Independent Sector.  Intemann, K. (2010). 25 years of feminist empiricism and standpoint theory: Where are we now?. Hypatia, 25(4), 778-796. Interagency Working Group on Youth Programs (US). (2013). Pathways for youth: draft strategic plan for federal collaboration. Jacquez, F., Vaughn, L. M., & Wagner, E. (2013). Youth as Partners, Participants or Passive Recipients: A Review of Children and Adolescents in CommunityBased Participatory Research (CBPR). American Journal of Community Psychology, 51(1-2), 176-189. Jennings, L.B., Parra-Medina, D.M., Hilfinger-Messias, D.K., & McLoughlin, K. (2006). Toward a critical social theory of youth empowerment. Journal of Community Practice, 14(1-2), 31-55.  Jordan-Young, R., & Rumiati, R. I. (2012). Hardwired for sexism? Approaches to sex/gender in neuroscience. Neuroethics, 5(3), 305-315.  201 Jöreskog, K. G., & Sörbom, D. (1996). LISREL 8: User's reference guide. Scientific Software International. Kirshner, B., O'Donoghue, J. L., & McLaughlin, M. W. (Eds.). (2002). Youth participation: Improving institutions and communities (No. 96). Jossey-Bass Inc Pub. Klau, M. (2006).Exploring youth leadership in theory and practice. New Directions for Youth Development, 2006(109), 57-87. Kline, P. (1979). Psychometrics and psychology. London: Academic Press. Kline, R. B. (2010). Principles and practice of structural equation modeling. The Guilford Press. Komives, S. R., & Dugan, J. P. (2010). Contemporary leadership theories. Political and civic leadership: A reference handbook, 1, 111-120. Komives, S. R., Lucas, N., & McMahon, T. R. (2009). Exploring leadership: For college students who want to make a difference (2nd ed). San Francisco: Jossey-Bass. Kozlowski, S. W., Gully, S. M., Salas, E., & Cannon-Bowers, J. A. (1996). Team leadership and development: Theory, principles, and guidelines for training leaders and teams. In M. Beyerlein, D. Johnson, & S. Beyerlein (Eds.), Advances in interdisciplinary studies of work teams: Team leadership (Vol. 3, pp. 251-289). Greenwich, CT: JAI Press. Kress, C. A. (2006). Youth leadership and youth development: Connections and questions. New Directions for Youth Development, 2006(109), 45-56. Krosnick, J. A., & Alwin, D. F. (1987). An evaluation of a cognitive theory of response-order effects in survey measurement. Public Opinion Quarterly, 51(2), 201-219. Lancaster, B. P. (1999). Defining and Interpreting Suppressor Effects: Advantages and Limitations. Langhout, R. D., & Thomas, E. (2010). Imagining participatory action research in collaboration with children: an introduction. American Journal of Community Psychology, 46(1-2), 606.  Lansdown, G. (2001). Promoting children's participation in democratic decisionmaking. Florence, Italy: Innocenti Research Center, UNICEF.   programs: learning to think strategically. Child development, 82(1), 27794.  Larson, W. (2000). Toward a Psychology of Positive Youth Development, 55(I), 170183.  Larson, R., Walker, K., & Pearce, N. (2005). A comparison of youthdriven and adultdriven youth programs: balancing inputs from youth and adults. Journal of Community Psychology, 33(1), 57-74. their children: A meta-analysis. Developmental Psychology, 34, 327.  202 Lee, S., & Hershberger, S. (1990). A simple rule for generating equivalent models in covariance structure modeling. Multivariate Behavioral Research, 25(3), 313-334. Lerner, R. M., Lerner, J. V., Almerigi, J. B., Theokas, C., Phelps, E., Gestsdottir, S., ... & von Eye, A. (2005). Positive Youth Development, Participation in Community Youth Development Programs, and Community Contributions of Fifth-Grade Adolescents Findings From the First Wave Of the 4-H Study of Positive Youth Development. The Journal of Early Adolescence, 25(1), 17-71. Lerner, Richard M, Lerner, J. V, & Benson, J. B. (2011). Positive Youth Development: Research and Applications for Promoting Thriving in Adolescence. Advances in Child Development and Behavior, 41, 117.  Lewis-Charp, H., Yu, H.C., Soukamneuth, S., & Lacoe, J. (2003). Extending the Reach of Youth Development through Civic Activism: Outcomes of the Youth Leadership Development Initiative. Social Policy Research Associates. Libby, M., Rosen, M., & Sedonaen, M. (2005). Building youth-adult partnerships for community change: Lessons from the Youth Leadership Institute. Journal of Community Psychology, 33(1), 111120.  Libby, M., Sedonaen, M., & Bliss, S. (2006). The mystery of youth leadership development: The path to just communities. New Directions for Youth Development, 2006(109), 1325.  Lichty, L., Foster-Fishman, P., Kornbluh, M., Mortensen, J., Pollard, M., & Hockin, S. (2011). Lessons learned from a blog-based photovoice project. In Foster-Fishman, P. (Chair), Taking photovoice online: Lessons learned from an innovative evaluation of a youth leadership training. Multipaper session presented at Evaluation 2011, Anaheim, CA.  Lips, H. M. (2000). College students' visions of power and possibility as moderated by gender. Psychology of Women Quarterly, 24(1), 39-43.  Lips, H. M. (2001). Envisioning positions of leadership: The expectations of university students in Virginia and Puerto Rico. Journal of Social Issues, 57(4), 799-813.  Little, T. D., Cunningham, W. A., Shahar, G., & Widaman, K. F. (2002). To parcel or not to parcel: Exploring the question, weighing the merits. Structural Equation Modeling, 9(2), 151-173. Little, R. J., & Rubin, D. B. (1987). Statistical analysis with missing data (Vol. 539). New York: Wiley. Loehlin, J.C. (2004). Latent Variable Models: An Introduction to Factor, Path and Structural Equation Analysis. Lawrence Erlbaum Assoc Inc. London, J. K., Zimmerman, K., & Erbstein, N. (2003). Youth-led research and evaluation: Tools for youth, organizational, and community development. New Directions for Evaluation, 2003(98), 33-45.  203 MacCallum, R. C., Widaman, K. F., Zhang, S., & Hong, S. (1999). Sample size in factor analysis. Psychological methods, 4, 84-99. MacCallum, R.C., Wegener, D.T., Uchino, B.N., & Fabrigar, L.R. (1993). The problem of equivalent models in applications of covariance structure analysis. Psychological Bulletin, 114(1), 185-199. Maccoby, E. E. (1990). Gender and relationships: A developmental account. American psychologist, 45(4), 513. McLaughlin, M. (2000). Community counts: How organizations matter for youth development. Washington, DC: Public Education Network. leadership development. New Directions for Youth Development, 2006(109),  Mann, R. D. (1959). A review of the relationships between personality and performance in small groups. Psychological Bulletin, 56, 241270. Marsh, H. W., Hau, K. T., Balla, J. R., & Grayson, D. (1998). Is more ever too much? The number of indicators per factor in confirmatory factor analysis. Multivariate Behavioral Research, 33(2), 181-220. Martin, C.L., Ruble, D.N., & Szkrybalo. J. (2002). Cognitive theories of early gender development. Psychological Bulletin, 128, 903933. Martin, C. L., Kornienko, O., Schaefer, D. R., Hanish, L. D., Fabes, R. A., & Goble, P. (2013). The Role of Sex of Peers and GenderTyped Activities in Young Children's Peer Affiliative Networks: A Longitudinal Analysis of Selection and Influence. Child development, 84(3), 921-937. Maton, K. I., & Salem, D. A. (1995). Organizational characteristics of empowering community settings: a multiple case study approach. American Journal of Community Psychology, 23(5), 63156. Matthews, M. S. (2004). Leadership education for gifted and talented youth: A review of the literature. Journal for the Education of the Gifted, 28(1), 77-113. McHale, S.M., Crouter, A.C., & Whiteman SD. (2003). The family contexts of gender development in childhood and adolescence. Social Development, 12, 125148. McNae, R. (2010). Young women and the co-construction of leadership. Journal of Educational Administration, 48(6), 677-688. Mehta, C. M., & Strough, J. (2009). Sex segregation in friendships and normative contexts across the life span. Developmental Review, 29, 201220. Miller, R.B., & Glass, J. (1989). Parent-child attitude similarity across the life course. Journal of Marriage and the Family, 51(4), 991-997.  204 Minkler, M. (2013). Enhancing Data Quality, Relevance and Use through Community-Based Participatory Research. In What Counts: Harnessing Data for America's Communities (1st ed.). United States: Federal Reserve Bank of San Francisco Mischel, W. (1973). Toward a cognitive social learning reconceptualization of personality. Psychological review, 80(4), 252. Mitra, D. L. (2005). Adults Advising Youth: Leading While Getting Out of the Way. Educational Administration Quarterly, 41(3), 520553.  Mitra, D. L. (2006). Student Voice Or Empowerment? Examining The Role Of School-Based Youth-Adult Partnerships As An Avenue Toward Focusing On Social Justice. International Electronic Journal for Leadership in Learning, 10(22). Mitra, D. L., Sanders, F. C., & Perkins, D. F. (2010). Providing spark and stability: The role of intermediary organizations in establishing school-based youth-adult partnerships. Applied Developmental Science, 14(2), 106123.  Mitra, D., Serriere, S., & Kirshner, B. (2014). Youth participation in US contexts: Student voice without a national mandate. Children & Society,28(4), 292-304. Mortensen, J., Foster-Fishman, P., Lichty, L., Harfst, S., Hockin, S., & Warsinske, K. (2012). Youth Perceptions of Leadership: Comparing Youth and Adult Ideas of Leadership and Youth Shifts in Definition following a Leadership Training Program. In Foster-Fishman, P. (Chair), Yang, E. (Discussant), A Participatory and Complimentary Mixed-Method Approach to Evaluating a Youth Leadership Program. Symposium presented at the 26th Annual Conference of the American Evaluation Association, Minneapolis, MN. Mortensen, J.E., Lichty, L., Foster-Fishman, P. Harfst, S., Hockin, S., & Warsinske, K., (2014). Leadership through a youth lens: Understanding youth conceptualizaitons of leadership. Journal of Community Psychology. National Center for Education Statistics, 2013. http://nces.ed.gov/pubs2015/2015011.pdf National League of Cities. (2010). Authentic youth civic engagement: A guide for municipal leaders. Washington, DC: National League of Cities. Nielsen, J. M. (Ed.). (1990). Feminist research methods: Exemplary readings in the social sciences. Boulder, CO: Westview Press. Niemi, R. G., & Hepburn, M. A. (1995). The rebirth of political socialization. Perspectives on Political Science, 24(1), 7-16. Nunnally, J. C. 1978 Psychometric Theory. New York: McGraw Hill. inner city: A case analysis of six African-American high school students. American Educational Research Journal, 34(4), 593-632.  205 for youth. Madison, WI: Community Youth Connection, University of Wisconsin Extension. O'Donnell, J., & Kirkner, S. L. (2014). Effects of an outofschool program on urban high school youth's academic performance. Journal of Community Psychology, 42(2), 176-190. Ozer, E. J., & Douglas, L. (2013). The impact of participatory research on urban teens: An experimental evaluation. American journal of community psychology, 51(1-2), 66-75. Pancer, S. M., Pratt, M., Hunsberger, B., & Alisat, S. (2007). Community and political involvement in adolescence: What distinguishes the activists from the uninvolved? Journal of Community Psychology, 35(6), 741759.  Pandey, S., & Elliott, W. (2010). Suppressor variables in social work research: Ways to identify in multiple regression models. Journal of the Society for Social Work and Research, 1(1), 28-40. Perkins, D. F., Borden, L. M., Villarruel, F. A., Carlton-Hug, A., Stone, M. R., & Keith, J. G. (2007). Participation in structured youth programs: Why ethnic minority urban youth choose to participateor not to participate. Youth & Society, 38(4), 420-442. Peterson, D. J., Baker, B., Leatherman, J., Newman, M. E., & Miske, S. (2012). Engaging youth, serving community: Social change lessons from a 4-H rural youth development program. Berkman Center Research Publication, (2013-7). Peterson, N. A., Peterson, C. H., Agre, L., Christens, B. D., & Morton, C. M. (2011). Measuring youth empowerment: Validation of a sociopolitical control scale for youth in an urban community context. Journal of Community Psychology, 39(5), 592-605. Pittman, K. (2000). Balancing the equation: Communities supporting youth, youth supporting communities. Community Youth Development Journal, 1(1), 33-36. Powers, C. B., & Allaman, E. (2012). How participatory action research can promote social change and help youth development. Berkman Center Research Publication, (2013-10). Prilleltensky, I. (2012). Wellness as fairness. American Journal of Community Psychology, 49(1-2), 121.  Prout, A. (2000). Children's participation: Control and selfrealisation in British late modernity. Children & Society, 14(4), 304-315. Quinn, J. (1999). Where need meets opportunity: Youth development programs for early teens. The future of Children, 96-116. Ramos-Zayas, A. Y. (2003). National performances: The politics of race, class and space in Puerto Rican Chicago. Chicago, IL: University of Chicago Press.  206 Ramey, H. L. (2013). Organizational outcomes of youth involvement in organizational decision making: a synthesis of qualitative research. Journal of Community Psychology, 41(4), 488-504. Raubenheimer, J. E. (2004). An item selection procedure to maximize scale reliability and validity. South African Journal of Industrial Psychology, 30 (4), 59-64. Raykov, T., & Marcoulides, G. A. (2001). Can there be infinitely many models equivalent to a given covariance structure model?. Structural Equation Modeling, 8(1), 142-149. Reiner, S. R., Anderson, T. L., Elizabeth Lewis Hall, M., & Hall, T. W. (2010). Adult attachment, God attachment and gender in relation to perceived stress. Journal of Psychology and Theology, 38(3), 175. Ren, J. Y., & Langhout, R. D. (2010). A recess evaluation with the players: Taking steps toward participatory action research. American Journal of Community Psychology, 46(1-2), 124-138. Reppucci, N. D. (1999). Adolescent development and juvenile justice. American Journal of Community Psychology, 27(3), 307-326. RichardsSchuster, K., & Checkoway, B. (2009). Youth participation in public policy at the local level: New lessons from Michigan municipalities. National Civic Review, 98(4), 26-30. Richards-Schuster, K., Checkoway, B., Christian, B. (2012). Youth leadership in generation with promise [Report to Generation With Promise-Michigan Department of Community Health, Office of Surgeon General]. Ann Arbor: Michigan Youth and Community Program, University of Michigan. Ricketts, J. C., & Rudd, R. D. (2004). Leadership development factors leading to the success of former Florida state FFA officers. Journal of Southern Agricultural Education Research, 54(1), 242-253. Roach, A.A., Wyman, L. T., Brookes, H., Chavez, C., Heath, S. B., & Valdes, G. (1999). Leadership giftedness: Models revisited. Gifted Child Quarterly, 43(1), 13-24. Roberts, G. C., & Treasure, D. C. (1992). Children in sport. Sport Science Review, 1(2), 46-64. Roth, P.L. (1994). Missing data: A conceptual review for applied psychologists. Personal Psychology, 47, 537-560. Roth, J., Brooks-Gunn, J., Murray, L., & Foster, W. (1998). Promoting healthy adolescents: Synthesis of youth development program evaluations. Journal of Research on Adolescence, 8(4), 423-459. Ruble, D., Martin, C., & Berenbaum, S. (2006). Gender development. In N. Eisenberg (Ed.), Social, emotional, and personality development. In W. Damon & R. M. Lerner (Eds.). Handbook of child psychology (Vol. 3, pp. 858932). Hoboken, NJ: John Wiley.  207 Rudman, L. A., & Glick, P. (2001). Prescriptive gender stereotypes and backlash toward agentic women. Journal of Social Issues, 57(4), 743-762. gender beliefs and career aspirations. Journal of Social Psychology, 41, 192202. Rutgers Cooperative Extension. (2003). Teens take the lead in the New Jersey 4-H Teen Leadership project. Retrieved April 30, 2013 from http://njaes.rutgers.edu/pubs/pdfs/4h/e148/appendix-h.pdf Rutter, M. (1993). Resilience: Some conceptual considerations. Journal of Adolescent Health, 14(8), 626-31. Sander, T. H., & Putnam, R. D. (2010). Still bowling alone?: The post-9/11 split. Journal of Democracy, 21(1), 9-16. Sauro, J., & Lewis, J.R. (2011). When designing usability questionnaires, does it hurt to be positive? In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (pp. 2215-2224). ACM.  Scales, P., & Leffert, N. (1999). Developmental assets: A synthesis of the scientific research on adolescent development. Minneapolis, MN: Search Institute. Schofield, J. W. (1981). Complimentary and conflicting identities: Images of interaction in an interracial school. In Asher, S. A., and Gottman, J. M. (eds.), The Development of Children's Friendships. Cambridge University Press, New York, pp. 5390. Schreiber, J. B., Nora, A., Stage, F. K., Barlow, E. A., & King, J. (2006). Reporting structural equation modeling and confirmatory factor analysis results: A review. The Journal of Educational Research, 99(6), 323-338. Schulz, W., Ainley, J., Fraillon, J., Kerr, D., & Losito, B. (2010). Initial Findings from the IEA International Civic and Citizenship Education Study. International Association for the Evaluation of Educational Achievement. Herengracht 487, Amsterdam, 1017 BT, The Netherlands. Schwartz, S., & Suyemoto, K. (2013). Creating change from the inside: Youth development within a youth community organizing program. Journal of community psychology, 41(3), 341-358. Scott, E. S. (2000). The legal construction of adolescence. Hofstra Law Review, 29, 547-598. Sears, D.O., & Levy, S. (2003). Childhood and adult political development. In D. O. Sears, L.  Huddy, & R. Jervis (Ed.), Oxford Handbook of Political Psychology (pp. 60109). Oxford: Oxford Univ. Press Sendjaya, S., & Sarros, J. C. (2002). Servant leadership: Its origin, development, and application in organizations. Journal of Leadership & Organizational Studies, 9(2), 57-64.  208 Siedentop, D. (1995). Improving sport education. Australian Council for Health, Physical Education and Recreation, 42(4), 22-23. Singleton, R.A., & Straits, B.C. (2005). Approaches to Social Research. Fourth Edition. New York: Oxford University Press. Social Policy Research Associates for the Innovation Center for Community and Youth Development. (2003, December). Lessons in leadership: How young people change their communities and themselves. Chevy Chase, MD: The Innovation Center for Community and Youth Development. Soleimanpour, S., Brindis, C., Geierstanger, S., Kandawalla, S., & Kurlaender, T. (2008). Incorporating youth-led community participatory research into school health center programs and policies. Public Health Reports, 123(6), 709-716. Soper, D.S. (2013). A-priori Sample Size Calculator for Multiple Regression [Software]. Available from http://www.danielsoper.com/statcalc  Speer, P. W. (2000). Intrapersonal and interactional empowerment: Implications for theory. Journal of Community Psychology, 28(1), 51-61. StataCorp, L. P. (2013). Stata multilevel Mixed-effects Reference Manual. Stern, M. J., Dillman, D. A., & Smyth, J. D. (2007). Visual design, order effects, and respondent characteristics in a self-administered survey. Survey Research Methods,1(3), 121-138. Stogdill, R. M. (1948). Personal factors associated with leadership: A survey of the literature. Journal of Psychology, 25, 3571. Stoneman, D. (2000). Leadership Development: A Handbook from YouthBuild USA and the Youth Action Program and Homes, Inc. Stoneman, D. (2002). The role of youth programming in the development of civic engagement. Applied Developmental Science, 6(4), 221-226. Sullivan, G. M., & Feinn, R. (2012). Using effect size-or why the P value is not enough. Journal of Graduate Medical Education, 4(3), 279-282. Sutton, S.E, & Kemp, S.P. (2002). Children as partners in neighborhood placemaking: Lessons from intergenerational design charettes. Journal of Environmental Psychology 22:17189. Tabachnick, B.G. and Fidell, L.S. (2007), Using Multivariate Statistics (5th ed.). New York: Allyn and Bacon.  Taft, J. K., & Gordon, H. R. (2013). Youth activists, youth councils, and constrained democracy. Education, Citizenship and Social Justice, 8(1), 87100.  Tanaka, J. S. (1993). Multifaceted conceptions of fit in structural equation models. In K.A. Bolen & J.S. Long (Eds.), Testing Structure Equation Models (10-39). Newbury Park, CA: Sage.  209 The Pew Research Center. (2007). How young people view their lives, futures and politics: A portrait of "Generation Next." Theokas, C., Almerigi, J. B., Lerner, R. M., Dowling, E. M., Benson, P. L., & Scales, P. C. (2005). Conceptualizing and Modeling Individual and Ecological Asset Components of Thriving in Early Adolescence. The Journal of Early Adolescence, 25(1), 113143.  Trivers, R. L. (1972). Parental investment and sexual selection. In B. Campbell (Ed.), Sexual selection and the descent of man 1871-1971 (pp. 136-172). Chicago: Aldine. Uhl-Bien, M., & Marion, R. (2008). Complexity leadership (Vol. 1). Charlotte, NC: Information Age Pub Incorporated. U.S. Census Bureau (2015). http://datacenter.kidscount.org/data/tables/103-child-population-by-race?loc=1&loct=1#detailed/1/any/false/573,869,36,868,867/68,69,67,12,70,66,71,72/423,424.  Van Linden, J.A., & Fertman, C.I. (1998). Youth leadership: A guide to understanding leadership development in adolescents. San Francisco: Jossey-Bass. Walker, J., Marczak, M., Blyth, D., & Borden, L. (2005). Designing youth development programs: Toward a theory of developmental intentionality. Organized activities as contexts of development: Extracurricular activities, after-school and community programs, 399-418. Wang, C. & Burris, M. (1997). Photovoice: Concepts, methodology, and use for participatory needs assessment. Health Education & Behavior, 24(3), 369-387.  Watson, D., Clark, L. A., Chmielewski, M., & Kotov, R. (2013). The value of suppressor effects in explicating the construct validity of symptom measures. Psychological assessment, 25(3), 929. Watts, R. J., Diemer, M. A., & Voight, A. M. (2011). Critical Consciousness: Current Status and Future Directions. New Directions for Child and Adolescent Development, 2011(134), 4357.  Watts, R. J., & Guessous, O. (2006). Sociopolitical development: The missing link in research and policy on adolescents. Beyond resistance, 59-80. Wehman, P. (1996). Life beyond the classroom: Transition strategies for young people with disabilities. Baltimore, MD: Paul H. Brookes Publishing Co. Wehmeyer, M. L., Agran, M., & Hughes, C. (1998). Teaching self-determination to students with disabilities: Basic skills for successful transition. Baltimore, MD: Paul H. Brookes Publishing Co. Weiss, H. B., Little, P. M. D., & Bouffard, S. M. (2005). More than just being there: Balancing the participation equation. New Directions for Youth Development, 2005(105). 15-31.  210 Westkott, M. (1990). Feminist criticism of the social sciences. In J. M. Nielsen (Ed)., Feminist research methods: Exemplary readings in the social sciences (pp. 5867). Boulder, CO: Westview Press. Westland, C.J. (2010). Lower bounds on sample size in structural equation modeling. Electronic Commerce Research and Applications, 9(6), 476-487. Widaman, K. F. (2006). III. Missing data: What to do with or without them. Monographs of the Society for Research in Child Development, 71(3), 42-64. Witt, Adolescence, 32, 253259. Woyach, R. B., & Cox, K. J. (1997). Principles for youth leadership development programs. Leadership Link.(Spring, 1996). Columbus, OH: The Ohio State University Leadership Center. Wylie, A. (2003). Why standpoint matters. Science and other cultures: Issues in philosophies of science and technology, 26-48. Youthbuild, USA. (2012). Retrieved from http://www.youthbuild.org.  Yu, H. C., Lewis-Charp, H. K., & Gambone, M. A. (2007). Evaluating Youth Leadership Development Through Civic Activism. The Handbook of Leadership Development Evaluation (pp. 377402). San Francisco: John Wiley & Sons. Zeldin, S., & Camino, Li. (1999). Youth Leadership: Linking Research and Program Theory to Exemplary Practice. New Designs for Youth Development, 15(1), 1015. Zeldin, S., Camino, L., & Calvert, M. (2007). Toward an understanding of youth in community governance: Policy priorities and research directions. Análise Psicológica, 25(1), 77-95. Zeldin, S., Christens, B. D., & Powers, J. L. (2013). The psychology and practice of youth-adult partnership: Bridging generations for youth development and community change. American Journal of Community Psychology, 51(3-4), 385-397. Zeldin, S., Petrokubi, J., Collura, J., Camino, L., & Skolaski, J. (2009). Strengthening Communities Through Youth Participation: Lessons Learned from the ACT for Youth Initiative. Ithaca, NY: ACT for Youth Center of Excellence. Zimmerman, M. A. (1995). Psychological empowerment: issues and illustrations. American Journal of Community Psychology, 23(5), 58199.  Zimmerman, M. A., & Zahniser, J. H. (1991). Refinements of spherespecific measures of perceived control: Development of a sociopolitical control scale. Journal of Community Psychology, 19(2), 189-204. Zimmerman, M., Ramírez-Valles, J., & Maton, K. I. (1999). Resilience among urban African American male adolescents: a study of the protective effects of sociopolitical control on their mental health. American Journal of Community Psychology, 27(6), 73351.   211 Zimmerman, A. M. (2011). A dream detained: Undocumented Latino youth and the DREAM movement. NACLA Report on the Americas, 44(6), 14-17.