<%BANNER%>

Sport as a Promising Factor in Reducing the Risk of Suicide Among Young People

Permanent Link: http://ufdc.ufl.edu/UFE0024737/00001

Material Information

Title: Sport as a Promising Factor in Reducing the Risk of Suicide Among Young People
Physical Description: 1 online resource (168 p.)
Language: english
Creator: Taliaferro, Lindsay
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: adolescent, sport, suicide
Health Education and Behavior -- Dissertations, Academic -- UF
Genre: Health and Human Performance thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Suicide affects everyone. Suicide and the need for suicide prevention represent significant public health concerns. The present work focuses on the relationship between sport participation and reduced risk of suicide among young people. The investigations extend the current literature and address limitations in previous research. Chapters 2 through 4 provide a foundation for the research presented in Chapters 5 and 6. The research described in the latter chapters address three methodological weaknesses in the extant literature. First, the studies test an original theoretical model depicting potential mechanisms that account for the relationship between sport participation and reduced risk of suicidal ideation among young people. Second, the research includes improved assessments of independent and/or dependent variables. Third, the studies compare effects of sport participation to involvement in other extracurricular activities on psychosocial outcomes. Findings show direct effects of sport participation on suicide risk. Tests of a theoretical model suggest that vigorous activity, social support, and school connectedness at least partially account for the relationship between sport participation and reduced risk of suicidal ideation. This research also found that participating in sports provides increased protection against suicidal ideation more effectively than does participating in other extracurricular activities. This line of inquiry provides a foundation for research to examine pathways through which sport participation relates to reduced risk of suicidal behavior. The association between sport participation and reduced risk of suicide suggests opportunities for health professionals, school officials, and public health planners. Eventually, practitioners may incorporate findings from this line of research into intervention efforts to prevent suicide among young people. Recommendations for future research are provided.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Lindsay Taliaferro.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Rienzo, Barbara A.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2011-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024737:00001

Permanent Link: http://ufdc.ufl.edu/UFE0024737/00001

Material Information

Title: Sport as a Promising Factor in Reducing the Risk of Suicide Among Young People
Physical Description: 1 online resource (168 p.)
Language: english
Creator: Taliaferro, Lindsay
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: adolescent, sport, suicide
Health Education and Behavior -- Dissertations, Academic -- UF
Genre: Health and Human Performance thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Suicide affects everyone. Suicide and the need for suicide prevention represent significant public health concerns. The present work focuses on the relationship between sport participation and reduced risk of suicide among young people. The investigations extend the current literature and address limitations in previous research. Chapters 2 through 4 provide a foundation for the research presented in Chapters 5 and 6. The research described in the latter chapters address three methodological weaknesses in the extant literature. First, the studies test an original theoretical model depicting potential mechanisms that account for the relationship between sport participation and reduced risk of suicidal ideation among young people. Second, the research includes improved assessments of independent and/or dependent variables. Third, the studies compare effects of sport participation to involvement in other extracurricular activities on psychosocial outcomes. Findings show direct effects of sport participation on suicide risk. Tests of a theoretical model suggest that vigorous activity, social support, and school connectedness at least partially account for the relationship between sport participation and reduced risk of suicidal ideation. This research also found that participating in sports provides increased protection against suicidal ideation more effectively than does participating in other extracurricular activities. This line of inquiry provides a foundation for research to examine pathways through which sport participation relates to reduced risk of suicidal behavior. The association between sport participation and reduced risk of suicide suggests opportunities for health professionals, school officials, and public health planners. Eventually, practitioners may incorporate findings from this line of research into intervention efforts to prevent suicide among young people. Recommendations for future research are provided.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Lindsay Taliaferro.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Rienzo, Barbara A.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2011-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024737:00001


This item has the following downloads:


Full Text

PAGE 1

1 SPORT AS A PROMISING FACTOR IN REDUCING THE RISK OF SUICIDE AMONG YOUNG PEOPLE By LINDSAY A. TALIAFERRO A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2009

PAGE 2

2 2009 Lindsay A. Taliaferro

PAGE 3

3 To my Mom without her support, this a ccomplishment would not have been possible

PAGE 4

4 ACKNOWLEDGMENTS I thank m y committee members, Drs. Barbara A. Rienzo, R. Morgan Pigg, Jr., Virginia J. Dodd, and M. David Miller, for their patience, encouragement, and unwavering support. I appreciate each person for his/her unique gifts and special talents. I am grateful for the attention shown to me, time devoted to me, and sage advice shared with me.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........9 LIST OF FIGURES.......................................................................................................................11 ABSTRACT...................................................................................................................................12 CHAP TER 1 INTRODUCTION AND LITERATURE REVIEW..............................................................14 General Introduction........................................................................................................... ....14 Prevalence of Suicidal Ideation and Beha vior Among Adolescents and Y oung Adults........ 16 Risk Factors for Suicide..........................................................................................................17 Depression.......................................................................................................................17 Hopelessness....................................................................................................................18 Loneliness/Isolation.........................................................................................................19 Low Self-Esteem.............................................................................................................21 Protective Factors Against Suicide......................................................................................... 22 Social Support.................................................................................................................22 Social Integration or Connectedness............................................................................... 23 Research Investigating Asso ciations Am ong Physical Activ ity, Sport Participation, and Suicide Risk................................................................................................................... .....24 Physical Activity and Sport Examined in Combination.................................................. 25 Physical Activity and Sport Examined Separately.......................................................... 26 Sport Participation Examined Independently.................................................................. 28 Limitations in Research Investigating A ssociations Am ong Physical Activity, Sport Participation, and Suicide Risk...........................................................................................29 Research Design..............................................................................................................29 Precision of Measurement............................................................................................... 31 Exploratory Models......................................................................................................... 32 Comparison of Sport Participation versus Involvem ent in Other Extracurricular Activities.............................................................................................................................33 Rationale and Organization of the Present Studies ................................................................. 34 2 SPORT PARTICIPATION: A VEHICLE FOR PROMOTI NG RESILIENCE IN YOUTH.......................................................................................................................... ........38 Introduction................................................................................................................... ..........38 Methods..................................................................................................................................39 Promoting Resilience Through Sport..................................................................................... 40 Developing Interpersonal and Intr apersonal Skills Am ong Youth................................. 40 Social competence/communication skills................................................................ 40

PAGE 6

6 Problem-solving skills/resourcefulness.................................................................... 40 Self-esteem/self-concept..........................................................................................41 Self-efficacy/positive sense of ones capabilities ..................................................... 42 Internal locus of control........................................................................................... 44 Persistence and ambition.......................................................................................... 44 Cognitive functioning/academic achievement/goals, hopes, and plans................... 45 Promoting Youth Participation in the Environment in Which They Live....................... 45 Emphasizing Intergenerational Relati onships Between Youth and Adults .....................46 Establishing Clear Norms and Expectations for Youth................................................... 47 Implications for Practice...................................................................................................... ...48 Recommendations for Future Research..................................................................................49 3 ASSOCIATIONS BETWEEN PHYSICAL ACTIVITY AND RE DUCED RATES OF HOPELESSNESS, DEPRESSION, AND SUICIDAL BEHAVIOR AMONG COLLEGE STUDENTS......................................................................................................... 50 Introduction................................................................................................................... ..........50 Depression and Hopelessness as Risk Factors for Suicidal Behavior ............................. 51 Physical Activity as a Protective Factor Against Suicidal Behavior ............................... 52 Methods..................................................................................................................................55 Procedures and Participants............................................................................................. 55 Measures..........................................................................................................................56 Hopelessness............................................................................................................56 Depression................................................................................................................56 Suicidal behavior......................................................................................................56 Physical activity....................................................................................................... 56 Data Analysis...................................................................................................................57 Results.....................................................................................................................................57 Discussion...............................................................................................................................60 Limitations.................................................................................................................... ..........62 Conclusions.............................................................................................................................63 4 HIGH SCHOOL YOUTH AND SUICIDE RISK: EXPLORI NG PROTECTION AFFORDED THROUGH PHYSICAL ACTI VITY AND SPORT PARTICIPATION........ 70 Introduction................................................................................................................... ..........70 Methods..................................................................................................................................72 Procedures and Participants............................................................................................. 72 Measures..........................................................................................................................73 Suicidal behavior......................................................................................................73 Physical activity....................................................................................................... 74 Sport participation.................................................................................................... 74 Data Analysis...................................................................................................................74 Results.....................................................................................................................................75 Discussion...............................................................................................................................77 Physical Activity............................................................................................................. 78 Sport Participation...........................................................................................................80

PAGE 7

7 Conclusions.............................................................................................................................82 5 EXPLORING POTENTIAL MEDIATI NG P ATHWAYS THROUGH WHICH PARTICIPATION IN SPORTS RELATES TO REDUCED RISK OF SUICIDAL IDEATION....................................................................................................................... ......90 Introduction................................................................................................................... ..........90 Methods..................................................................................................................................93 Procedures and Participants............................................................................................. 93 Measures..........................................................................................................................94 Involvement in other extr acurricular activities ........................................................ 94 Antecedent variable.................................................................................................. 94 Intervening variables................................................................................................ 95 Criterion variable......................................................................................................97 Data Analysis...................................................................................................................97 Results.....................................................................................................................................98 Sample Characteristics....................................................................................................98 Preliminary Analyses....................................................................................................... 99 Bivariate Analyses......................................................................................................... 100 Confirmation of the Path Analysis................................................................................ 100 Tests of Mediation.........................................................................................................101 Sport Participation versus Involvement in Other Extracurri cular Activities .................103 Discussion.............................................................................................................................103 Findings.........................................................................................................................103 Contribution to the Literature........................................................................................104 Limitations.................................................................................................................... .106 Future Research.............................................................................................................107 Conclusions...........................................................................................................................109 6 EXPLORING PROTECTIVE RELATIONS HIPS BET WEEN PARTICIPATION IN SPORTS AND RISK FACTORS ASSOCIA TED WITH ADOLESCENT SUICIDE........ 116 Introduction................................................................................................................... ........116 Methods................................................................................................................................118 Procedures and Participants........................................................................................... 118 Measures........................................................................................................................119 Involvement in other extr acurricular activities ...................................................... 119 Antecedent variable................................................................................................ 120 Intervening variables.............................................................................................. 121 Criterion variable....................................................................................................123 Data Analysis.................................................................................................................124 Results...................................................................................................................................124 Sample Characteristics..................................................................................................124 Preliminary Analyses .....................................................................................................125 Confirmation of the Path Analysis................................................................................ 126 Tests of Mediation.........................................................................................................126 Sport Participation versus Involvement in Other Extracurri cular Activities .................128

PAGE 8

8 Discussion.............................................................................................................................129 Conclusions...........................................................................................................................131 7 SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS ....................................... 140 Summary...............................................................................................................................140 Conclusions...........................................................................................................................143 Recommendations for Future Research................................................................................145 APPENDIX A DEMOGRAPHIC AND BAC KGROUND INFORMATION ............................................. 147 B PHYSICAL ACTIVITY, SPORT PARTICIPATION, AND INVOLVEMENT IN OTHER EXTRACURRICU LAR ACTIVITIES .................................................................. 148 C SCHOOL CONNECTEDNESS SCALE..............................................................................153 D SCALES THAT CANNOT BE REPRODUCED................................................................154 LIST OF REFERENCES.............................................................................................................155 BIOGRAPHICAL SKETCH.......................................................................................................168

PAGE 9

9 LIST OF TABLES Table page 3-1 Physical activity, hopelessness, depre ssion, and suicidal behavior am ong college students....................................................................................................................... .......65 3-2 Hopelessness, depression, and suicidal behavior am ong college students of various ages....................................................................................................................................66 3-3 Hopelessness, depression, and suicidal behavior am ong college students of various races and ethnicities.......................................................................................................... .67 3-4 Relative risk of hopelessne ss, depression, and suicidal behavior for college students engaging in various levels of aerobic activ ity co mpared to thos e not reporting any aerobic activity............................................................................................................... ....68 3-5 Relative risk of hopelessne ss, depression, and suicidal behavior for college students engaging in various levels of strength or toning activity com pared to those not reporting any toning activity.............................................................................................. 69 4-1 Physical activity, team sports participation, feelings of hopelessness, and suicidality in adolescent m ales and females........................................................................................ 85 4-2 Physical activity, team sports participation, feelings of hopelessness, and suicidality in adolescent m ales and females........................................................................................ 86 4-3 Relative risk of hopelessness and suicidality for adolescents of different races/ethn icities compared to white adolescents................................................................ 87 4-4 Relative risk of hopelessne ss and suicidality for adolescen ts with various levels of physical activity com pared to thos e not reporting any physical activity........................... 88 4-5 Relative risk of hopelessne ss and suicidality for athletes com pared to non-athletes........ 89 5-1 Intercorrelations among the inte rvening and criterion variables .....................................112 5-2 Means and independent samples t-test values comparing students involved in university-run sports and students not involved in university-run sports ........................113 6-1 Percentage of students w ho participated in different school and community activities during high school............................................................................................................ 133 6-2 Percentage of students w ho participated in different sport activities during high school ...............................................................................................................................134 6-3 Standardized regression co efficients for physical activity and psychosocial variables predicted from participation in sports during high school...............................................135

PAGE 10

10 6-4 Standardized regression co efficients for physical activity and psychosocial variables predic ted from participation in different school and community activities during high school...............................................................................................................................136 6-5 Standardized regression co efficients for physical activity and psychosocial variables predic ted from total participation in extr acurricular activities and time spent in school activities during high school.................................................................................137

PAGE 11

11 LIST OF FIGURES Figure page 5-1 Model presenting proposed mechanisms th rough which sport participation influences the risk of suicidal ideation ..............................................................................................111 5-2 Standardized regression co efficients dem onstrating th e direct effects of sport participation on the intervening and criterion variables.................................................. 114 5-3 Path model with standardized regression coefficients..................................................... 115 6-1 Model presenting proposed mechanisms th rough which sport participation influences the risk of suicidal ideation ..............................................................................................138 6-2 Path model with standardized regression coefficients..................................................... 139

PAGE 12

12 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy SPORT AS A PROMISING FACTOR IN REDUCING THE RISK OF SUICIDE AMONG YOUNG PEOPLE By Lindsay A. Taliaferro August 2009 Chair: Barbara A. Rienzo Major: Health and Human Performance Suicide affects everyone. Suicide and th e need for suicide prevention represent significant public health concerns. The present work focuses on the relationship between sport participation and reduced risk of suicide am ong young people. The investigations extend the current literature and address limitations in pr evious research. Chapters 2 through 4 provide a foundation for the research presented in Chapters 5 and 6. The research described in the latter chapters address three methodological weaknesses in the extant litera ture. First, the studies test an original theoretical model de picting potential mechanisms th at account for the relationship between sport participation and reduced risk of suicidal ideation among young people. Second, the research includes improved a ssessments of independent and/or dependent variables. Third, the studies compare effects of sport participation to involveme nt in other extracurricular activities on psychosocial outcomes. Findings show direct effects of sport participation on suicide risk. Tests of a theoretical model suggest that vigorous activity, social supp ort, and school connecte dness at least partially account for the relationship between sport particip ation and reduced risk of suicidal ideation. This research also found that part icipating in sports provides increased protection against suicidal ideation more effectively than does participa ting in other extracurricular activities.

PAGE 13

13 This line of inquiry provides a foundation for research to examine pathways through which sport participation relates to re duced risk of suicidal behavior The association between sport participation and reduced risk of suicide sugge sts opportunities for health professionals, school officials, and public health planners. Eventual ly, practitioners may incorporate findings from this line of research into intervention e fforts to prevent suicide among young people. Recommendations for future research are provided.

PAGE 14

14 CHAPTER 1 INTRODUCTION AND LITERATURE REVIEW For all children aged 1 to 19 years, a signi ficant increase (in cause of death) was registered only for suicide (18.2%). Pediatrics A nnual Summary of V ital Statistics: 2005, February 2007 Public health researchers and suicide prevention practitioners need to learn more about both the risk factors for suicideand effective strategies for suicide prevention. Centers for Disease Control and Prevention, Morb idity and Mortality Weekly Report, September 7, 2007 General Introduction Suicide among young people in the United States represents a national tragedy and grave public health concern. Suicide ranks as the third leading cause of death for American youth [1]. Yet, most professionals view the number of re ported youth suicides as an underestimate of the amount of adolescents who purposefully take th eir own lives each year [2]. Though myriad interrelated socio-psychological f actors affect adolescent suicide, researchers uniformly highlight several correlates including depression, hopele ssness, loneliness, a nd low self-esteem. Conversely, social support and social integrat ion or connectedness may constitute important factors that protect adolescents against suicidal tendencies. Substantial research suggests that physic al activity provides mental health benefits. However, researchers have only recently examined possible pr otective associations between physical activity and sport participation, and suic ide risk. Research examining general physical activity and risk of suicide has yielded equivoc al findings. Yet, research focused on sport participation has produced consistent evidence that indicates young people involved in sports show significantly lower odds of emotional distress [3-9], and de monstrate significantly lower risk of suicidal behavior, compared to non-partic ipants [4-7, 10-13]. Thus, involvement in sports may create a distinct form of protection against ri sk factors associated with adolescent suicide.

PAGE 15

15 The present work seeks to investigate in-dep th the unique psychosocial benefits afforded to young people involved in organized sports. Th e research focuses on th e association between sport participation and suicide risk among youth. Chapter 1 presen ts a review of the relevant literature. This chapter details the prevalence of suicidality (suicida l thoughts or behavior) among young people, describes risk factors and pr otective factors associated with adolescent suicide, and discusses the extant literature focused on physical activity, sport participation, and suicide risk. Chapters 2 through 6 describe investigations that re port findings from a literature review, analyses of secondary data and original research studies. This research was developed into manuscripts and submitted for publication in scholarly journals. The investigation in Chapter 2 sought to document evidence that supports involvement in sports as a vehicle to enhance resilience in youth. The research in Chapter 3 examined relationships between physical activity and rates of hopelessne ss, depression, and suic idal behavior among a national sample of college students. In Chapter 4, associations be tween physical activity an d hopelessness, suicidal thoughts, suicide plans, and suic ide attempts were examined among U.S. high school students. This research in this chapter also examined associations between sport participation and hopelessness and suicidality. Chapters 5 and 6 desc ribe studies that investigated whether or not participating in sports provides young people with greater protection against suicidal ideation than participating in other extracurricular activitie s. Moreover, these studies represented the first effort to test theoretical mode ls that incorporated potential mechanisms through which sport participation may protect young people against risk factors associated with suicide. Chapter 7 presents a synthesized discussion of majo r findings across the studies, and proposes recommendations for future research.

PAGE 16

16 Prevalence of Suicidal Ideation and Behavior Among Adolescents and Young Adults Suicide ranks as the third l eading cause of death for youth aged 15 to 24 [1], and the second leading cause of death for college stude nts [14, 15]. Suicides acco unt for 12.9% of all deaths among persons aged 15 to 24 [1]. During 1990-2003, the combined suicide rate for young people aged 10 to 24 declined 28.5%, from 9.48 to 6.78 per 100,000 persons [16]. However, from 2003 to 2004, the rate increased by 8.0%, fr om 6.78 to 7.32 [16]. This represented the largest single-year increase during 1990-2004 [16]. Furthermore, during this same time period, the suicide rate for persons under age 20 increased 18%, and suicide constituted the only cause of death that increased among adolescents [17] In 2004, youth aged 15 to 24 represented 14.2% of the U.S. population and comprised 13.3% of the deaths by suicide [1]. Data from the Centers for Disease Control and Preventions 2007 Youth Risk Behavior Survey (YRBS) [18] indicated that, during the 12 months preceding the surv ey, 14.5% of high school students seriously considered attempting suicide, 11.3% planned to attempt suicide, and 6.9 % actually attempted suicide. An estimated 1 in 12 college students has made a suicide plan, and approximately 1,000 students die by suicide on college campuses each year [1]. In 2007, 9.8% of college students seriously considered attempting suicide, and 1.5% actually attemp ted suicide at least once during the previous school year [19]. Females attempt suicide more often than males, but males aged 15 to 19 are 3.6 times more likely than females to complete suicide [1]. However, from 2003-2004, females aged 10 to 14 and 15 to 19 demonstrated the largest and second largest, respectively, percentage increase in suicide rates (75.9% and 32.3%, re spectively) among persons 10 to 24 years [16]. Furthermore, for each completed suicide, an estimated 100 to 200 young people attempte d to take their own lives [1]. Thus, Healthy People 2010, a set of health objectives for the nation, specifically targets reducing the rate of completed suicide (O bjective 18-1) and the rate of attempted suicide

PAGE 17

17 among adolescents (Objective 18-2) [20]. The National Institute of Mental Health urges researchers to focus on decreasing the adolescen t suicide rate by studyi ng risk and protective factors [21]. Risk Factors for Suicide Suicidal behavior is considered multicausal in origin because researchers consistently distinguish several correlates as sociated with adolescen t suicide that occur with regularity [22, 23]. The myriad interrelated so cio-psychological factors associat ed with adolescent suicide makes pinpointing distinct factors as causes impo ssible [22]. Rather, research suggests that the development of suicidal behavior results from a convergence of multiple predisposing and concurrent risk factors [24, 25]. Risk factors associated with suicidality among young people include depression, hopelessness, lo w self-esteem, and loneliness/isolation. Frequently, suicide occurs as the end-point of adverse life sequen ces that involve cumulative exposure to multiple risk factors [24, 25]. Therefore, efforts to prev ent suicidal behavior should target points along the pathways that link childhood and adolescent adversity to suicidal behavior [23]. Depression Depression represents the stronge st and m ost consistent predic tor of adolescent suicidality [26]. Psychological autopsies show that 3 out of 4 adolesce nts who died by suicide were depressed [22]. Depressive disorders pose the gr eatest proximal risk for suicidal behavior among young people [26], and they represent the most prevalent disorders among adolescent suicide victims, ranging from 49% to 64% [27]. About 90% of adolescent suicide completers had a diagnosable psychiatric disorder at the time of their deaths [28] Young people with psychiatric disorders possess a 9-fold increa sed risk of suicide [29]. Significant relationships exist between depression and both suicidal thoughts and suicide behavior. According to findings from the 2004 National Survey on Drug Use and Health [30],

PAGE 18

18 during their worst or most recent major depres sive episode, about 9% (2.3 million) of youth thought they would prefer to be dead, 7% ( 1.8 million) thought about killing themselves, 3.6% (900,000) made a plan to kill themselves, and 2.9 % (712,000) actually tried to kill themselves. Prinstein et al. [31] assessed psychiatric inpatients demonstrat ing suicidal tendencies and found both direct and indirect effects of depressive symptoms on suicid al ideation. Kande l et al. [32] found similar results among high sch ool students. In their study, Kandel et al. [32] identified depressive symptoms as the stronge st correlates and predictors of suicidal ideation. Field et al. [33] assessed a relatively small sample of high school seniors, but found a very high rate (37%) of students who scored above th e clinical cutoff for depressed mood. These researchers also found that depressed adolescents experienced more frequent suicidal thoughts than nondepressed youth [33]. Konick a nd Gutierrez [34] tested a path model to predict suicidal ideation in college students. Results from this study show ed that depressive symptoms exerted a stronger effect than other risk factors in predicting suicidal thoughts [34]. Two recent review papers found that depr essed mood consistently emerged as the condition most frequently and strongly associat ed with adolescent suicidal behavior [24, 35]. Using a community sample of adolescents, findings from Fotti et al. [36] support this position. These researchers found that depression represen ted the most significant factor in suicidal behavior [36]. Using a longitudi nal design, Fergusson et al. [25] examined several predictors associated with suicidal ideation and suicid e attempt among young people. These researchers also identified depression as the strongest predictor of suicid al behavior [25]. Hopelessness Relatively few research studies have addr essed the association between hopelessness and adolescent suicidal behavior [37]. Furthermor e, most studies to date yielded mixed findings [37]. Thompson et al. [37] concluded: Wheth er depression or hopelessness or both factors

PAGE 19

19 simultaneously predict suicidal behaviors differs for adolescents compared to adults, for males versus females, and with the outco me variable measured [p.16]. Though studies have established a relati onship between hopelessness and adolescent suicidality, hopelessness does not c onsistently predict suicidal be havior once researchers control for depression [27, 29, 35, 38]. For example, Ko nick and Gutierrez [34] confirmed depressive symptoms and hopelessness as predictors of suicid al ideation in college students. However, in this study, depressive symptoms exerted a str onger influence on suicid al ideation than did hopelessness [34]. Yet, Thompson et al. [37] found direct effects of depression and hopelessness on suicidal behaviors for males, but only a direct effect of hopelessne ss not depression for females. Similarly, Kumar and Steer [39] found that hopelessness explained slightly more than depression explained about suicid al ideation among adolescents. Swedo et al. [40] concluded that only hopelessness discriminated between adolescents with a history of a suicide attempt and adolescents at-risk for suicide, despite the larg e number of variables that differentiated these adolescents from a control group of healthy vo lunteers. Spirito and Esposito-Smythers [35] suggested that hopelessness may place adolescents at risk for suicidal behavior for only a limited period of time during a depressive episode. Th us, the time-limited nature of hopelessness may explain the contradictory results found in many studies [35]. Loneliness/Isolation Adolescents who lack soci al support and experience isol ation may engage in selfinjurious behavior [41]. Adolescents who attemp t suicide often become isolated from important people in their lives, and they experience intense f eelings of isolation shor tly before their attempt [22]. Compared to community controls, adolesce nts hospitalized after a suicide attempt reported more problems such as loneliness and depressive a ffect [42]. Adolescents at risk for suicidal behavior engage in more solita ry activities than their non-suicid al risk peers [43]. High school

PAGE 20

20 youth involved in the Florida Suicide Prevention Study [44] identified social isolation as a primary factor that contri butes to youth suicide. Prinstein et al. [31] found that lower levels of close friendship support and greater levels of perceived peer rejection each directly related to suicidal id eation among adolescents. In addition, perceived peer rejecti on indirectly related to suicidal ideation via symptoms of depression [31]. Examining the structural model ge nerated by Prinstein et al. [31] suggested that family functioning represented a weaker correlate of suicidal ideation th an did peer functioning. In contrast, Field et al. [33] found that parental re lationships represented a more important contributing factor to adolescents psychological h ealth than peer relations. However, findings from this study also showed that peer relati ons, number of friends, and popularity were less optimal for depressed adolescents, compared to their non-depressed peers [33]. Allberg and Chu [22] suggested that suicide risk increases for isolated adolesce nts, who lack support from either family or peers. Young females, especially, may demonstrate increased vulnerability to relational strains that contribute to suicidal behavior. Female s social support systems include parents, other significant adults, and friends [45]. Only family relations represent significant factors associated with suicidality for males [45]. Fotti et al. [36] found that poor parental relationships were positively associated with suicidal ideation and suicide attempts in both males and females. However, findings from Fotti et al. [36] and Watt and Sharp [45] suggested that the effects of family strains on suicide risk remained larger for females than for males. Poor peer relationships also may contri bute to increased suicide risk for females compared to males. Bearman and Moody [46] fo und that females isolated from the adolescent community demonstrated greater risk for suicid al thoughts than females embedded in cohesive

PAGE 21

21 friendship groups. Conversely, adolescent males a ppeared more impervious to social context [46]. Results from Fotti et al. [36] support these findings. For females in Fotti et al.s [36] study, poor peer relationships rema ined significantly associated with suicidal behavior, even after researchers controlled for depression and pa rental relationships. Overall, empirical evidence suggests that dysfunctiona l relationships demonstrate a di fferential impact on male and female adolescents [36, 45]. Low Self-Esteem Research has estab lished a relationship be tween low self-esteem and both current and future suicidal behavior among youth [29]. Using a case-control design, Groholt et al. [42] found suicidal adolescents posse ssed lower self-esteem than non-suicidal adolescents across seven domains (physical appear ance, close friends, social co mpetence, romantic appeal, scholastic competence, athletic competence, and gl obal self-worth). Similarly, Wild et al. [47] found associations between low self-esteem in six different domains (peer, school, family, body image, sports/athletics, and global self-worth) and increased suicide risk. McGee and Williams [48] discovered that self-esteem in preadolescence significantly predicted linear trends in adolescent reports of multiple health-compromis ing behaviors, including suicidal ideation. A study by McGee et al. [49] identified a weak, but direct pathway from low self-esteem in childhood to later suicidal ideation among females. Overholser et al. [50] found that, even after controlling for levels of depres sion and hopelessness, self-esteem demonstrated a direct impact on suicidal ideation. In this study, low self-esteem in high sc hool students contributed to a substantial increase in rates of suicidal ideation and previous suicide attemp ts [50]. Among adolescents with low self-esteem, 33% had previ ously attempted suicide at least once [50].

PAGE 22

22 Protective Factors Against Suicide Prevention programs that attempt to decrease the incidence and prevalence of high-risk behaviors among adolescents require an understanding of factors affecting the occurrence of these behaviors [7]. This strategy focuses on fact ors that both increase and decrease the chance of engaging in high-risk behavior [7]. Most st udies addressing adolescent suicide focus on risk factors [51]. Protectiv e factors constitute a less well-understood area [51]. Yet, protective factors constitute dimensions that moderate the effects of individua l vulnerabilities or environmental hazards so that a given developmen tal trajectory reflects more adaptation in a given domain than would occur if these proce sses were not operating [52, 53]. Resiliency research suggests that the presence of protective factors can reduce the imp act of risk factors on youth violence [54]. Thus, promoti ng protective factors offers an effective approach to prevent adolescent suicide [55]. Significant protective factors associated with reduced suicide risk among young people include social support and social integration (a lso termed connectedness). Social support and integration represent two factors associated with enhanced resi lience in youth. Resilient youth demonstrate reduced suicide risk [56]. Social Support Social support prom otes healthy adolescen t functioning [41]. Adolescents who report strong social support exhibit higher levels of resi lience and lower levels of suicide risk [41]. Adolescents demonstrate less suicide risk if they perceive their family, friends, and peers as more accepting, and if they have more positive friendshi ps [41]. Randell et al [57] found that, among potential high school drop-outs, general satisfact ion with support, higher levels of support for school, and support when feeling depressed or thi nking about suicide predic ted lower levels of suicidal behavior.

PAGE 23

23 Social support may mitigate the effect of two variables highly correlated with youth suicide risk isolation and hope lessness [58]. Kaslow et al. [59] found that social support represented a significant independe nt predictor of suicide attemp ts. Thompson et al. [37] found that family support influenced adolescent suicid al behavior through its direct effects on three mediating factors depression, hopelessness, an d anxiety. High school youth involved in the Florida Suicide Prevention Study [44] identified the presence of a tr ustworthy adult, friends to talk with, and a sense of belonging as protective factors against youth suicide. Social Integration or Connectedness According to Rink and Tricker [60], resilie n cy research has embraced the concept of connectedness as a means to promote resilience in youth. Doan et al. [56] support this view. These researchers explain that the following prot ective factors enhance resilience and help to prevent adolescent suicide: pa rent/family support and connecte dness, good relationships with peers, support from relevant adults, social inte gration/opportunities to pa rticipate, and perceived connectedness to school [59]. A wi dely cited study, Resnick et al [61], found that parent/family connectedness and perceived school connectedne ss protected against multiple health risk behaviors among youth, includi ng adolescent suicidality. Youth strongly integrated into their families show reduced risk of suicidal behavior [62]. Family cohesion, positive parent-child connectio n, spending time together, parental supervision, and high parental academic and be havioral expectations all repr esent protective factors [29]. Borowsky et al. [55] found that parent/family connectedness represented a protective factor for attempting suicide that cross-cut sex and racial/ethnic groups of a dolescents. In a recent review, Gould et al. [27] identified family cohesion as a protective factor agai nst suicidal behavior among middle school, high school, and college stude nts. Furthermore, students who described their family life in terms of mu tual involvement, shared interest s, and emotional support were 3.5

PAGE 24

24 to 5.5 times less likely to demonstrate suicidal be havior than were adoles cents from less cohesive families who reported the same levels of depression or life stress [27]. Youth disconnected from major support syst ems (school, work, and family) show very high risk for suicide [29]. Conversely, positive connection to school represents a protective factor against suicidal behavior [29, 55, 61, 63]. Two longitudina l studies support the protective influence of school connectedness on mental health and risk beha viors. Bond et al. [64] found that students with good school and social c onnectedness were less likely to experience subsequent mental health problems or engage in health risk beha vior. Ozer [65] showed that adolescents who felt more connect ed to school demonstrated reductions from grades 7 to 8 in self-reported depressive symptoms, teacher-rate d anxiety and depressive symptoms, and selfreported perpetration of violence. To counter negative peer influe nces associated with suicidal behavior, high school youth suggested that schools help young people build healthy peer relationships by promoting their involvement in positive youth groups and structured after school activities [44]. Research supports the ability of extracurricular activities, particularly sport participation, to promote increased school conne ction [66-69]. Overall, these findings support a twofold role for schools, to nurture both academic proficiency as well as a sense of connectedness among students [55]. Research Investigating Associations Among Ph ysical Activity, Sport Participation, and Suicide Risk Researchers have only recently examined poten tial positive effects of physical activity and sport participation on suicidal behavior. Th e few studies published to date yielded equivocal findings. The results becom e more complex when examined by sex. The extant literature can be summarized in three categories based on method used to determine the independent variable: 1) physical activity and sport examined in combinat ion as one variable, 2) physical activity and

PAGE 25

25 sport examined separately as two variables, and 3) participation in organized sports examined independently. Physical Activity and Sport Examined in Combination Simon et al. [70] used a case-control de sign to investigate the relationship between involvement in physical activity and suicide risk. Researchers asked four questions to assess level of physical activity. First, participants pr ovided information regard ing whether or not they participated in any physical activ ities or exercises such as runni ng, calisthenics, golf, gardening, or walking for exercise in the pa st month. Participants then reported the type of activity they spent the most time doing, amount of time they spent engaged in the activity, and frequency of this activity. Findings suggested a strong prot ective association between physical activity and nearly-lethal suicide attempts (NLSAs). Even after adjusting for the full set of potential confounders [sociodemographic characteristics, depression, hopelessness, a serious medical condition, alcoholism, BMI, and social support] risk for NLSAs was five times higher among those who had not been physically active in the past month [70, p.312]. Regardless of intensity, frequency, or duration of a primary act ivity, participants who reported engaging in physical exercise demonstrated lower suicide risk. Though Paffenbarger et al. [71] used the term sportsplay, their definition of the variable more closely depicted general physical activity (i.e., golf, walking, gardening, carpentry, swimming, tennis, jogging, or running). These researchers investigated the incidence of depression and suicide among Harvard male al umni aged 35-74 during a 23-27 year follow-up period. Findings indicated that physically ac tive individuals demonstr ated lower rates of depression. However, suicide rates remained unrelated to anteced ent physical activity. Two international studies co mbined physical activity and sports to create a composite independent variable. Tomori and Zalar [51] asked respondents how often they engaged in

PAGE 26

26 sports or physical activity and defined physic al activity as a leis ure time recreational activityas well as club and/or school team sport [p.408]. Comparing female suicide attempters to female non-attempters, researcher s found that females who never attempted suicide more frequently turned to sports as a coping be havior in distress. Among males, non-attempters reported a significantly higher frequency of sport participation and physical activity. Ferron et al. [4] incorporated a similar me thod of assessment that included two physical activity questions. The first question focused on frequency of sport activity, and the second question focused on belonging to a sports club. Athletic adolescents included youth who participated in sports 2-3 times per week and be longed to a sports club, or who participated in sports daily whether or not they belonged to a sp orts club. Non-athletic adolescents included individuals who engaged in sports once a week or never. Findings from this study determined that youth with the highest frequency of sport activity experienced superi or psychological wellbeing. The most athletic adol escents felt sad, depressed, or desperate less often, and a lower proportion reported past suicidal thoughts or suicide attempts. Physical Activity and Sport Examined Separately Brosnahan et al. [72] examined varyi ng levels of physical activity, as well as participation in physical educat ion and team sports. They cr eated a total physical activity variable to differentiate students who engaged in physical activity at least 6 days per week versus less active students. Though spor t participants showed lower risk of sadness and suicidal behavior, results failed to reach statistical significance. Howeve r, youth who participated in a greater number of total physical ac tivity sessions showed signifi cantly less risk of considering suicide. In addition, higher le vels of vigorous physical activit y, strength and toning activity, and total physical activity all were associated w ith a decreased risk of planning suicide.

PAGE 27

27 Brown and Blanton [10] also compared young people demonstrating varying levels of physical activity, as well as athlet es versus non-athletes, with rega rd to suicide risk. Brown and Blanton [10] classified college students into 5 activity levels based on their responses to two physical activity questions. One question assesse d participation in vigorous activity, and one other provided an estimate of participation in moderate-intensity activity. Sport participation was dichotomized (yes/no) by participants respons es to the question: D uring this school year, on how many college sports teams (intramural or extramural) did you participate? Compared to inactive males, males who reported participating in low-intensity activity demonstrated reduced risk of suicidal behavior. However, neither moderately nor vigorously active males showed reduced suicide risk in this study. Furthermore, moderately or vigorously active females demonstrated increased risk of suicidal behavior compared to less active females. Yet, for both males and females, sport participation related to decreased suicidal behavior. Compared to sport participants, male non-participan ts had 2.5 times the odds of reporting suicidal behavior, and female non-participants were 1.67 times more likely to report suicidal behavior. Unger [9] found similarly complex relationships. This study determined the relative risk of suicidality associated with 6 physical activity/team sport combinations compared to a no physical activity group. Findings indicated that physical activity, especially when combined with team sports, related to lower rates of suic idal behavior for males. Yet, high levels of physical activity, especially without team sports, related to higher rates of suicidal behavior for females. In contrast, a recent study by Brown et al [13] found that high school females who participated in regular, vigorous-i ntensity physical activity or on a sports team were less likely to attempt suicide, than were their inactive or sports team non-participant peers. In addition, among

PAGE 28

28 males, frequent, vigorous-intensity physical activity and sport participation were associated with a lower likelihood of suicidal ideation and suicide attempts. Sport Participation Examined Independently Several studies examined the association be tween suicidality and involvement in sports without considering physical activity as an additional independent variable. Focusing on adolescent males, Sabo et al. [ 73] found that athletes were less likely than non-athletes to consider suicide, make a suicide plan, or attemp t suicide. For both males and females, Sabo and colleagues [12] found a signi ficant association between athletic participation and reduced risk of seriously considering suicide. In particular, highly involved athl etes (those participating on 3 or more sport teams) of both sexes showed significa ntly less risk of considering suicide than nonathletes. Furthermore, female athletes were significantly less likel y than their non-athlete counterparts to make a suicide plan. Miller et al. [6] also identified sport partic ipation as a protective factor militating against suicidal tendencies among yout h. These researchers found lowe r rates of suicidal ideation, suicide planning, and suicide attempts among male and female athletes. Similar to Sabo and colleagues [12], highly involved athl etes reported the lowest rates of suicidal behavior. Male athletes were significantly less lik ely than male non-athletes to cons ider, plan, or attempt suicide. Compared to their non-athlete counterparts, female athletes exhibited si gnificantly less risk of considering or planning suicide. Page et al. [11] found that male and female students who participat ed on one or two sport teams were significantly less likely to have attemp ted suicide in the past 12 months. Compared to non-athletes, Baumert et al. [3] found athletes reporte d less frequent feeli ngs of hopelessness, and Oler et al. [7] and Steine r et al. [8] found athletes show ed significantly less risk of depression and suicidal ideation. However, similar to Page et al. [11], a nd contrary to Sabo and

PAGE 29

29 colleagues [12] and Miller et al. [6 ], Oler et al. [7] identified female athletes as less likely than non-athletic females to have attempted suicide. Harrison and Narayan [5] examined relationships among sport participation, involvement in other extracurricular activiti es, psychological factors, and su icidal behavior. Findings indicated that those involved in team sports at school (a lone or combined with other activities)were most likely to report high self-e steem and were least likely to report sadness, anxiety, and suicidal behavior [5, p.118]. These researchers c oncluded that while engaging in any organized activity imparted some benefit, pa rticipation in sports afforded unique positive advantages. Limitations in Research Investigating Asso ciations Amon g Physical Activity, Sport Participation, and Suicide Risk Research Design Cross-sectional survey research dominates the literature examining associations among physical activity, sport participation, and suicide ri sk. Of the reviewed studies, only one, Simon et al. [70], included a case-control design. No st udy to date incorporated a longitudinal design to prospectively examine the effect of sport pa rticipation on subsequent suicidal behavior. However, since sport participation is define d by self-selection [67, 74] researchers may be seeing a selective aggregation of individuals who a priori brin g fewer problems across a wide range of domains of functioning [8]. Conve rsely, involvement in sports may provide psychosocial benefits to participan ts [8]. In a single wave of data collection, researchers cannot determine whether sport participation represents cause or effect, nor can they separate true causal effects from preexisting differences [67, 74]. Cross-sectional research designs limit researchers ability to disentangle effects of selec tion versus the true impact of sport participation [8]. Thus, cross-sectional data collection and analysis preclude establishing a causal

PAGE 30

30 relationship, or temporality of sequence, between sport particip ation and psychosocial variables or health-risk behaviors [4, 5, 8, 67, 69, 74, 75]. The goal of most research involves inferring the consequences of s port participation, so many researchers decry the predominant single -wave designs [67, 74]. Longitudinal designs would serve two purposes: 1) to help establish causal directionali ty in the relationship between sport participation and suicidality, and 2) to identify the mechanisms through which sport participation reduces suicide risk. In general and, in particular, within research investigating the relationship between sport participation and su icide risk, researchers have not tested the processes or variables hypothesi zed to mediate between the eff ects of sport participation and other variables [67, 74]. In summary, according to Marsh [74], a better design for inferring effects of sport participation involves large-scale, longitudinal studies in which re searchers relate changes in the same variables assessed on multiple occasions to involvement in sports. Investigators should conduct multi-wave studies that relate the effects of sport participation to a set of outcomes, controlling for preexisting differences in a pa rallel set of outcomes and background variables [74]. Ideally, researchers would coll ect data from participants in at least three waves: 1) before they become involved in sports, to establish their baseline self-concept an d involvement in health risk behaviors; 2) during their period of participation, to assess the immediate impact of sport participation on intervening and criterion variables; and 3) after th e period of sport participation has ended, to assess the longevity of effects [76]. This design w ould yield longitudinal data that come closer to illuminating causal links betw een sport participation, proposed intervening variables, and suicidal behavior. In addition, this design woul d help unravel the physiological

PAGE 31

31 and psychosocial mechanisms through which sport pa rticipation contributes to a lower risk of suicide [73]. Precision of Measurement Researchers routinely identify self-r eport m easures as a methodological limitation [9, 10, 23, 28, 36, 57, 77]. Self-report data do not allo w for actual observation or measurement of behavior and, thus, may be subject to inherent biases compared to external measures [77]. However, adolescents remain the most reliable source of information on their internal states such as depression and suicidal thoughts as well as perceptions of thei r personal experience and risk behaviors [47, 57]. Furthermore, parents typicall y remain unaware of their childs suicidality [26, 57], which limits the usefulness of their data as an additional source of valid information. A more worrisome problem involves the measurement of adolescent suicidality. Reynolds and Mazza [2] caution that suicidal ideati on, as a component of suicidal behavior, must be viewed as a multifaceted construct that requires much more than a single item regarding whether or not adolescents ha ve thought about killing them selves. According to these investigators, research on suicidal ideation or group assignment predicated on one or two questions regarding having thought s of suicide lacks sufficient specificity for defining this domain of suicidality and obscures the relative im portance of suicidal idea tion as a component of suicide risk in youth [2, p.533]. However, many researchers investigating adolescent suicide have incorporated only a few items, or even a si ngle item, to measure suicidality. For example, Fergusson et al. [25] asked part icipants if they had thought about taking their lives by suicide during the past year. McGee et al. [49] asked: Have you felt so low you thought about committing suicide? Rueter and Kwon [78] asked if, during the past week, adolescents were distressed or bothered by thoughts of ending their lives. Items such as these represent far from

PAGE 32

32 optimal measures of suicidal ideation, and they likely miss many young people who have thought about killing themselves. A large void exists in the lit erature investing physical activ ity, sport participation, and suicide risk regarding precise measurement of adolescent su icidality. To date, only one study assessed suicidality using more than the standard three YRBS [79] items: seriously considered attempting suicide, planned to commit suicide, a nd actually attempted suicide. Oler et al. [7] used five items to measure suicidal ideation and its relationship with sport participation. Multiple indicators of a constr uct produce a more precise and robust measurement, which increases reliability. Future investigations should addre ss this methodological flaw by conducting studies that incorporat e instruments with sufficient sp ecificity to answer definitive questions regarding the association between participation in organized sports and risk of suicide. Exploratory Models Progress in s uicide research requires both analyzing existing data sets and implementing new, theoretically informed studies [28]. An important aspect of future research involves developing and testing multiple pathways to suicid al behavior [35]. Exploratory research using causal models to examine direct effects and buffering effects of different types of daily activities will help guide development of suicide prev ention and intervention programs [35, 43]. Researchers can theoretically identify precursors of suicidal behavior, and use models to develop interventions that attenuat e suicide risk [29]. Most research addressing physical activity, sp ort participation, and adolescent suicidality originated from larger studies that examined a number of adolescent health issues using existing data sets [3-6, 8-13, 72, 73]. Few researchers have designed studi es focused on investigating the relationship between sport particip ation and risk of suicide. S till, research examining sport participation and suicidality re gularly found inverse relationships Researchers have proposed

PAGE 33

33 hypothetical mechanisms to explain the associati on between sport particip ation and reduced odds of suicidal behavior [80]. However, no study specifically tested these mechanisms. Thus, researchers have not conducted a study to test a model depicting potential pathways that explain the relationship between sport participation and reduced su icidal tendencies among young people. Comparison of Sport Participation versus Invol vement in Other Extracurricu lar Activities One of the most important factors for predicting positive youth development involves how adolescents spend their out-of-school time [81]. Adolescents glean many benefits from their involvement in extracurricul ar activities. According to a review in Marsh and Kleitman [67], researchers associate participation in extrac urricular activities, particularly involvement in sports, with reduced rates of high school dropou t; higher attendance records; fewer discipline referrals; fewer mental and general health probl ems; fewer eating and di etary problems; greater subsequent wages, occupational status, and educ ational attainment; high er grades in school; more encouragement by significant others; and greater general self-c oncept, educational aspirations, and subsequent colleg e attendance. Participation in extracurricular activities relates to higher levels of perceived strengths and se lf-esteem, improvement in academic performance, reduction in risk behaviors, and an or ientation toward the future [82-84]. Extracurricular activities facilitate adolescents developmental need for social relatedness by expanding peer networks and facilitating th e cultivation of peer bonds [82, 85]. Youth activities help young people deve lop communication skills, an d provide youth with an opportunity to experience loyalty and intimacy w ith their peers [85]. Involvement in school programs links adolescents to a set of similar peers, provides shared experiences and goals, reinforces friendships between peers, and may strengthen the relationship between the individual and family [82, 84].

PAGE 34

34 In contrast with other activities, sport participation may represent a particularly valuable source of learning outside the cl assroom [86]. Athletes may e xperience greater social support when they become integrated into a networ k that includes teammates, coaches, health professionals, family, and community [12]. Incr eased integration into these protective social networks may buffer adolescents ag ainst the effects of life stress [9, 12]. In addition, Marshs [74] Identification/Commitment Model argues that sport participation may enhance school identification, involvement, and commitment. This connection to school positively affects more narrowly defined academic outcome s, as well as nonacademic outcomes such as self-concept [67, 74]. Abundant evidence support s the positive effects of sport participation on a variety of broadly defined academic, social, and psychological outcomes [67, 74]. Furthermore, little or no evidence suggests that sport participation produces ne gative effects [67, 74]. Despite research that supports the positive role sport participation plays in the lives of youth, only Harrison and Narayan [5] compared effects of sport part icipation to involvement in other extracurricular activities on suicide risk. Ye t, to identify sport as a distinct protective factor against suicide, research must demons trate that sport partic ipation provides unique benefits compared to involvement in other extracurricular activities. Rationale and Organization of the Present Stu dies Involvement in organized sports may provi de distinct protecti on against adolescent suicide through multiple mechanisms. Sport pa rticipants benefit from the psychological advantages of physical activity including improve ments in overall emotional well-being [87-89], depressed mood [88, 90, 91], anxiety and stress [88, 92-94], and self-esteem [90, 95, 96]. Youth athletes also experience greater social integration and support when they become members of an extended social network [12]. In addition, sport participatio n may cultivate an increased commitment to, involvement in, or identification with school and school values [67-69]. Thus,

PAGE 35

35 the synergy of multiple mechanisms may account fo r the distinct protecti on sport participation affords against risk factors a ssociated with suicide. The present work served three purposes: 1) to highlight potential wa ys sport participation may enhance resilience in youth and, thus, reduce the risk of suicide; 2) to further explore associations among physical activity, sport participation, and suicid e risk; and 3) to address key methodological limitations in the extant literature. The investigation in Chapter 2 presents a systematic literature review that detailed benefits of youth particip ation in organized sports. Specifi cally, this study explored potential mechanisms through which sport participa tion may enhance resilience in youth. Chapters 3 and 4 describe studies that sought to substantiate and extend the extant research investigating physical activity, sport partic ipation, and suicide risk. The research presented in Chapter 3 explored associations between physical activity and risk factors associated with suicide among young people. Data for this study derived from the 2005 American College Health Association Nati onal College Health Assessment (ACHA-NCHA) [97]. This investigation examined the relations hip between different types of physical activity and rates of hopelessness, depression, and suic idal behavior among co llege students. The research presented in Chapter 3 provided a foundation for the next investigation. The study described in Chapter 4 built on findings from the fi rst study by including sport participation as an additional independent variable This study examined associ ations among physical activity, sport participation, and suicide risk among adoles cents using data from CDCs 2005 YRBS [79]. The assumption that involvement in organized spor ts affords distinct protection against suicide risk guided the analysis.

PAGE 36

36 Chapters 5 and 6 present studies based on original data that extended the foundational research detailed in Chapters 2 through 4. These studies sought to test the hypothesis that sport participation protects against risk factors asso ciated with suicide through multiple mechanisms. This research addressed three limitations in the ex tant literature. First, both studies improved on the precision of measurement of suicidality by assessing thoughts about suicide us ing a valid and reliable measure that provided sufficient specific ity to answer definitive questions regarding the association between sport participation and suicid al ideation. Second, both investigations tested theoretical models that include d potential mechanisms through which sport participation may protect against risk factors associated with suicide among young people. Third, both studies compared the psychosocial benefits afforded to st udents involved in sports to those afforded to individuals involved in other extracurricular activities. Chapter 5 describes a study conducted with students enrolled in undergraduate classes at the University of Florida, Gainesville, Florida. This study tested an original model that highlighted the mediating effects of physical activ ity and social support on relationships between collegiate sport participation (intramural, club, a nd varsity sport teams), risk factors associated with suicide (depression, hopelessness, loneliness and low self-esteem), and suicidal ideation. Analyses also compared effects of sport part icipation to involvement in other university organizations on psychosocial variables. Based on findings generated from the invest igation presented in Chapter 5, the study described in Chapter 6 tested an enhanced model that included a construct that assessed perceived connection to school. This study also tested the mo del using involvement in high school sports as the antecedent variable. Furthe rmore, more comprehensive assessments of sport participation and involvement in other extracurric ular activities allowed for a detailed analysis

PAGE 37

37 regarding types of activi ties that afford the greatest benefit. Participants in this study included freshman students, aged 18 or 19, attending the Univ ersity of Florida, Gainesville, Florida.

PAGE 38

38 CHAPTER 2 SPORT PARTICIPATION: A VEHICLE FO R PROMOTI NG RESILIENCE IN YOUTHi Introduction Researchers exam ining resilience in youth as k: What protects them? How can we enhance adolescents ability to resist stressfu l life events, and promote positive adaptation and competence? [98]. The study of resilience focu ses on assets within individuals, communities, schools, and families that buffer young people against adopting negative health behaviors, and that foster health-promoting behaviors [60]. Researchers frame the ability to rebound from adversity in the context of prot ective factors that enhance resi lience [52]. Pr otective factors constitute individual or environm ental safeguards that enhance an individuals ability to resist stressful life events while adapting to the situa tion and developing competency in dealing with it [98, p.296]. Protective factors af fect adolescent health behavior in two ways: 1) they buffer adolescents from harm by decreasing the likelihood that youth will engage in high-risk health behavior; and 2) they serve as moderators of ri sk by modifying the relationship between risk and problem behavior [60]. The Association of Mate rnal and Child Health Programs suggested that health educators more closely examine prot ective factors in youth, families, schools, and communities [99]. This approach provides a basis for developing adolescent health policies and programs that emphasize preventive and positive in terventions designed to cultivate assets and competencies in adolescents [60]. Such de velopmental assets help inoculate youth from engaging in high-risk behavior [100]. A recent study found that participation in team sports positively correlated with resilience scores among adolescents [101]. Few other em pirical investigations have examined the i The Health Education Monograph Series, 25, 3, 20-26, 2008. Reprinted with permission from Eta Sigma Gamma. Published by Eta Sigma Gamma, 2000 University Avenue, Muncie, IN, 47306. Copyright 2008.

PAGE 39

39 association between sport participation and resi lience in youth. Yet, sport may represent an intervention activity that increase s protective factors associated w ith enhanced resilience [102]. Rink and Tricker [60] identified six groups of common protective factor themes that promote resilience in youth: 1) Deve loping interpersonal and intrap ersonal skills among youth; 2) Promoting youth participation in the environment in which they live; 3) Emphasizing intergenerational relati onships between youth and adults; 4) Establishing clear norms and expectations for youth; 5) Providing access to information and services for youth in their community; and 6) Involving youth in policy cha nges that support positive development. The purpose of this investigation involved exploring potential mechanisms through which sport participation may enhance resilience in youth. On e research question drove this inquiry: Does sport represent a vehicle for enhancing protec tive factors comprising Ri nk and Trickers [60] first four themes? The last two themes were excluded because they did not pertain to sport participation. Methods A literature search was conducted through Ps ycINFO, PubMed, ERIC, Social Sciences Index, and SPORTdiscus over the time period 1990 to 2007. Initially, the search terms resilience, sport, and extracurricular activit ies were applied. In subsequent searches, sport and specific protective factors comprisi ng each theme such as self-esteem, selfefficacy, or locus of control were applied. The reference li sts of all retrieved publications were scanned for other relevant documents and then these works were obtained. The 31 publications found using these search criteria, including empiri cal, review, and theoretical articles, papers, and reports, comprise the review of the literature summarized in this paper.

PAGE 40

40 Promoting Resilience Through Sport Developing Interpersonal and Intra personal Skills Among Youth Resilient youth demonstrate resourcefulne ss and good problem-solving skills [52, 60]. Resilient adolescents posse ss goals, hopes, and plans for the futu re, and they have the persistence and ambition to bring them to fruition [52]. Furthermore, strong positi ve correlations exist between resilience in youth and social competence, self-esteem, self-efficacy, internal locus of control, and cognitive functioning [52, 60]. Social competence/communication skills According to research presented in Marsh and Kleitman [74], sport participation involves many robust positive effects, and the largest effects relate to social self-c oncept. A report from the Presidents Council on Physi cal Fitness and Sports promoted sport as a means to develop important social skills [103] Through their involvement in sports, young people form new acquaintances, and they learn how to cooperate, negotiate, work t ogether as a team, and respect others [84, 85, 104]. Sport participants lear n about leadership, res ponsibility, tolerance, selflessness, and legitimate ex pression of talent [84, 85, 104]. Youth activities help young people develop communication skills, empathy, and an understanding for others viewpoints [85]. School activities al so provide youth with an opportunity to experience loyalty and intimacy with their peers [85]. Problem-solving skills/resourcefulness Sport can teach young people m any beneficial lessons including problem-solving skills [103]. Sport participa tion may represent a mechanism that demands athletes respond to any given situation or problem, and f acilitates athletes ability to ma rshal their resources to meet a variety of challenges [105]. Pa rticipation in extracurricular ac tivities also provides adolescents with an opportunity to learn em otional self-regulation Dworkin et al. [85]. Youth activities,

PAGE 41

41 especially team sports, involve goal-directed behavior, and require pa rticipants to develop competencies for limiting the disruptive effect of emotions on attention and performance [85]. However, compared to participation in other ac tivities, sport participation relates to increased self-knowledge and improved emotional self-man agement [106]. Youth highly involved in sports describe themselves as more psychologi cally resilient, or ab le to bounce back after experiencing problems [107]. Sport participants gain cr itical knowledge regarding how to achieve a goal including autonomous decision-making, and action and time management [85]. Furthermore, youth athletes learn how to interpre t and negotiate competitive losses [105]. Following a losing game, coaches and players review the overall game strate gy, identify problem areas, and devise a plan of action for future practices designed to impr ove performance in problem areas [105]. This strategy significantly reduces th e likelihood of negative chain reac tions following game losses, and promotes constructive analysis of strengths and weaknesses with an eye toward improvement [105]. Self-esteem/self-concept Participa tion in extracurricular activities relates to higher levels of perceived strengths and self-esteem [84]. However, in contrast with other activ ities, sport participation may represent a particularly valuable source of lear ning outside the classroom [86]. Pedersen and Seidman [108] described the team sport enviro nment as a fertile ground for adolescent selfesteem development, and Kaufmann [86] highlight ed enhanced self-esteem as a benefit gained from sport participation. Self-esteem and self -efficacy derive largely from an individuals successful accomplishment of tasks the individual perceives as impor tant [105]. Sport represents an important activity to many young people who devote considerable time, energy, and other resources to its pursuit [105]. Furthermore, unlike many other activitie s, sport participation

PAGE 42

42 usually results from a young pers ons personal decision rather than the decision of parents and teachers [105]. Donaldson and Roman [104] found that sport positively impacted specific self-concept domains (i.e., athletic, social, a nd physical), as well as overall self-concept. Findings from Marsh and Kleitman [67], Pedersen and Seidman [108], and Tiggemann [109] support the association between participati on in organized sports and in creased self-esteem. Mastery experiences and successes in sport contribute to improved self-esteem. However, even losses within the sport arena may c onvey a sense of accomplishment [105]. A well-played game generates respect between teams, players, and the wider student population [105]. Such positive experiences not only reduce the impact of loss, bu t also contribute to pl ayers self-esteem and self-efficacy [105]. Resilient youth demonstrate areas of competen ce and perceived efficacy valued by self or society [60]. American society values athleticism and sport skills [12]. Therefore, experiences in sport may influence the development of h ealthy self-esteem through feedback youth receive from others regarding their competence in a valu ed domain. In addition, participation in sport generates a sense of physical well-being. Invest igators identified an association between a positive view of ones body and increased self-est eem [69]. Research also supports a positive relationship between sport participation and enhanced body image among adolescents [68, 103, 109]. Richman and Schaffer [76] suggested that sport may promote self-worth by fostering such contributors to self-esteem as physical competence, a favorable body image, and more flexible sex attitudes. Self-efficacy/positive sense of ones capabilities Im proved personal competency may constitute the most significant benefit derived from sport participation. In a recent review, Bailey [110] stated that one of the especially relevant

PAGE 43

43 findings concerning sport relates to the deve lopment of perceived physical competence. Hjemdal et al. [101] also found an association between sport particip ation and the personal competence factor of the Resilience Scale for A dolescents. According to Bell and Suggs [111], four conditions promote self-esteem: 1) a sense of connectedness, 2) a sense of uniqueness, 3) a sense of power, and 4) a sense of models or dynamics used to understand experiences. Involvement in organized sport encourages all of these conditions, and ca n especially encourage a sense of power [111]. Through supportive interactions with coach es, youth athletes learn to call on their reserve of heart to go the extra mile without external motivation [111]. The ability to call upon this reserve provides youth with a sense of power, or a feeling of competence or learned self-helpfulness [111]. Competency-enhancing activit ies are highly organized and adult-supervised, promote increased involvement with prosocial peers, prov ide opportunities for achiev ement, and facilitate the development of autonomous decision-making skills [108]. Compared to individual sports or informal physical activity, team sports typically include all of the characteristics of competencyenhancing activities [108]. Partic ipation in most organized sports encourages the development of initiative, as well as a str ong sense of individual accomplishm ent or mastery [108]. Sport participation builds confidence, helps youth gain feelings of personal control, and creates expectations of success [103]. Tiggemann [109] suggested that playing sports may confer a sense of body competence, whereby bodies can r un fast or jump high, and generally have a function other than merely objects to examine. Findings from Marsh [112] and Stein, Fisher, Berkey, and Colditz [113] support this view. In Marshs [112] study, elite high school athletes demonstrated systematically higher physical sel f-concept than non-athletes Stein et al. [113]

PAGE 44

44 found that increased physi cal activity in youth was positively a ssociated with improved athletic self-perception, and decreased physical activity related to reduced athletic self-perception. Internal locus of control Findings from Snyder and Spreitzer [75] showed a positive relationship between high school athletic commitment and internal control over life events. Yin and Moore [114] also found that participation in inters cholastic sports positively related to scores for locus of control among adolescents. Similarly, Marsh and Kleitman [67] discovered that involvement in high school extramural/team sports benef ited internal locus of control. Persistence and ambition Youth activities, especially sports, represent a context in which young people learn to set achievable goals, and learn about effort and pers everance [85]. Involvem ent in sports helps youth develop personal resources for achieving soci ally approved goals [12]. Athletic training requires consistent investment in the form of practice and conditioning [105]. Athletes learn to push themselves, work hard, demonstrate discipline, and remain focused [85]. Unlike unstructured and unorganized activitie s, sport requires discipline and continuity of effort [115]. Research suggests that organized sports influence behaviors such as perseverance and risk-taking to a greater extent than do informal physical ac tivities [104]. Team spor t participants usually establish goals they strive to obtain throughout the season. In pursuit of their goals, athletes experience physical, emotional, a nd mental challenges [102]. Spor t serves as th e ideal context for individuals to experience adve rsity and use challenges for self-enhancement [116]. Athletes learn to persist in the face of losses, and to analyze constructively and compensate for competitive weaknesses [105]. To th rive within sports, athletes must cultivate perseverance and resilience [102].

PAGE 45

45 Cognitive functioning/academic ac hievement/goals, hopes, and plans Participation in extracurricular activit ies during high school improves academic performance, reduces involvement in risk behavi or, and relates to an orientation toward the future [82, 84]. Participation in sport favorably affects academic self-concept, academic pursuits and achievements, time spent on homework, commi tment to school, and college attendance and completion rates [66-68, 82, 86, 105]. Physical activity and lear ning go hand-in-hand [103]. Sport can elevate academic performance by incr easing blood flow to the brain, enhancing mood, increasing mental alertness, and improving self-e steem [110]. In short, sport participation enhances time, energy, and commitment to academic pursuits [67]. Promoting Youth Participation in the Environment in Wh ich They Live Students who feel engaged at sc hool, and feel connected to an adult in their school, show increased likelihood of graduating [117]. Through affiliation and identification with close friends, adolescents benefit from companions hip, emotional and motivational support, role modeling, and a sense of belongi ng [52]. Engaging youth in real opportunities for involvement and leadership provides young people with experiences that increase their so ciability within their peer group [60]. Participation in extracurricular activities may play an important part in the socialization process by linking adolescents to larger social forces with in the community [84]. The availability of bonding opportunitie s in the school environment enhances students social bonding and reduces risk-taking behavior [11]. Involvement in school programs links adolescents to a set of similar peers, provi des shared experiences and goals, reinforces friendships between peers, and may strengthen the relationship between the individual and family [82, 84]. Extracu rricular activities facilitate adolescents developmental need for social relatedness, and may contribute to young peoples identities as im portant and valued members of

PAGE 46

46 the school community [82]. Participation in school activities expands peer networks, and represents an important context for developing valuable peer bonds [85]. Sport creates important opportuni ties for students to excel and meaningfully contribute to the school community [105]. The team sport e nvironment encourages increased involvement with peers in a prosocial context [108]. The pe rsonal and social behavior associated with sport participation also may increase students' leader ship potential [118]. Athletes become more socially involved than non-athletes, and they de monstrate increased partic ipation in other school and community activities [75, 102, 103]. Involvement in sports relates to be tter attendance records, and to reduced likelihood of dropping out [67, 103, 114, 119]. Sport participation may cultivate an increased commitment to, involvement in, or identification with school and school values [67-69]. School sports may enhance bonding by increasing opportunities for students to feel a sense of belonging, atta chment, and participation in th eir social environment [11]. Emphasizing Intergenerational Relati onships Betw een Youth and Adults Access to caring adults and positive adult role models reduces the likelihood that adolescents will engage in negative health beha viors [60]. Programs that provide adolescents with access to adults enable adults and youth to view each other in a positive light, and allow adults to model prosocial behavior [60]. Participation in most organized sport pr ograms provides youth with regular opportunities to interact with caring adult ro le models [103, 108]. Since adults typically run organized sports, youth athletes become exposed to the care of ad ults beyond their immediate family [84]. The therapeutic support base for athletes extends beyond the head coach, and may include athletic trainers, strength coaches, academic counselors, and other team members [10]. Hawkins and Mulkey [105] suggested that close relationships between coaches and athletes provide youth with a rationale and resource for sustained academic effo rt even in the face of difficulties. McHale et

PAGE 47

47 al. [115] found that free-time spent with parents and non-parent al adults related to positive adjustment among high school youth. Furthermore, time spent in sports positively related to time spent with parents, peers, and, in particular, non-parental a dults [115]. The relationships sport participants form with their coaches and te ammates represent one of the building blocks for developing resilience [102]. Establishing Clear Norms and Expectations for Youth Young people need adults to establish clear norms and expectations to successfully navigate through adolescence [60]. Establishing norm s and expecta tions for youth provides them with an understanding of rules acceptable to the environment in which they live [60]. School activity participation may moderate anti social behavior for some students [120]. Organized activities may expose youth to particular sets of values including acceptance of authority and rule-gover ned behavior [84]. Most researchers found inverse relationships between organized sport participation and social misconduct problems, delinquent behavior illicit drug use and drug abuse, discipline problems, aggression, peer-related problems, a nd involvement with police [68, 104, 105]. Sport participation may affect moral development thro ugh opportunities to engage in prosocial, or ethical, behavior [103]. Organized sports involve longer duration and greater intensity, frequency, and discipline than do informal activiti es [104]. In addition, coaches usually expect athletes behavior to reach a st andard that will allow for team cohesion without many disruptions [104]. Youth athletes also must adhere to rule s of fair competition [ 105]. Thus, in organized sports, youth must behave better than in informal sports where rules may not play an important role [104].

PAGE 48

48 Implications for Practice Findings fro m a literature review revealed empirical evidence suggesting that, for youth interested in sports, participation may prom ote numerous protective factors that enhance resilience. However, pract itioners should know that negative sport experiences increase both physical and psychology health ri sks among youth [121]. Some yout h involved in sport report engaging in more problem behavi ors such as lying and substance use, and high-risk health behaviors such as speeding or not using a bicycle or motorcycle helmet, and report higher rates of negative experiences such as negative peer interactions an d inappropriate adult behavior [122]. Yet, many studies that found negative effects of sport participation did not examine the nature and quality of the experience for youth [ 122]. The downsides of participation are not intrinsic to sports. Factors that likely infl uence a young persons experience in sport include coaches training; support the young person receives from coaches, family members, and peers; and the type and competitive level of a sport [122]. For example, perceived parental pressure decreases enjoyment of sport activity, and increases anxiety and risk of injury [121]. Involvement in sports can represent a productive out-of-school experience if adults make an effort to conduct activities, establish envir onments, and develop sustained peer-peer and youthadult relationships that intentionally focus on building capacity in youth [122]. Health educators may use findings from th e present investigation to advocate for increased opportunities to involve interested yout h in sports. Findings would support efforts to assess the impact that the erosion of funds and social support for school and community sport programs have on young peoples health. Furthermore, the empirical research presented in this paper provides guidance for coach es and health educators in es tablishing youth sport programs. Since few youth possess the talent to play compe titive sports at the highest level, the primary goals of youth sport programs should involve developing general physical competence, and

PAGE 49

49 promoting physical activity, fun, socioemotional development, life skills, sportsmanship, and good health [122]. Coaches should nourish a sens e of independence, self-reliance, courage, positive expectancy, and self-confidence in youth at hletes [111]. Sport programs that enhance these skills and attributes will help cultivate resilience in youth. Recommendations for Future Research A paucity of research has exam ined the a ssociation between sport participation and resilience in youth. To advance the science in this area, res earchers should explore the broad relationship between sport and resilience, as well as variables presumed to intervene in the relationship between sport participation and th e development of resilience. For example, protective factors addressed in this investiga tion may represent mediating variables between sport participation and global measures of resilien ce. In addition, different types of sports may enhance resilience in different ways. For ex ample, Marsh and Kleitman [67] found that extramural and team sports produ ced more positive effects than in tramural and individual sports. Future research should evaluate potential differen tial effects across various forms of sport, and whether effects vary by sex. Opportunities to participate in and contribu te to sport programs focused on positive youth development provide young people with a sense of belonging, mastery, genero sity, and mattering [122]. Involvement in sports may represent an optimal activity to promote mechanisms that enhance resilience in youth. More re search in this area is needed.

PAGE 50

50 CHAPTER 3 ASSOCIATIONS BETWEEN PHYSICAL ACTIVITY AND RE DUCED RATES OF HOPELESSNESS, DEPRESSION, AND SUIC IDAL BEHAVIOR AMONG COLLEGE STUDENTSi Introduction Suicide is the third leading cause of death am ong Americans aged 10 to 24 years [16]. Suicides account for 12.9% of all deaths am ong young people aged 15 to 24 years [1]. An estimated 1 in 12 college students has made a suicide plan, and approximately 1,000 students die by suicide on college campuses each year [1]. According to recent data from the Centers for Disease Control and Prevention, during 1990-2003, the combined suicide rate for persons aged 10 to 24 years declined 28.5% [16]. However, from 2003 to 2004, the rate increased by 8.0%. This represented the larges t single-year increase during 1990-2004 [16]. In 2005, 10.2% of college students seriously consid ered attempting suicide, and 1.5% actually attempted suicide at least once during the previous school year [123]. In this study, we examined the association between physical activity and suicidality in a national sample of 43,499 college students. Physical activity was defined as frequency of vigorous or moderate exercise a nd strength or toning exercises. We performed analyses using data from the 2005 National College Health Assessment (NCHA) conducted by the American College Health Association (ACHA) [97]. We specifica lly explored associations between levels of aerobic and toning or strength activities a nd risk of hopelessness, depression, and suicidal behavior. i Journal of American College Health, 57, 4, 427-436, 2009. Reprinted with permission from the Helen Dwight Reid Educational Foundation. Published by Heldref Publications, 1319 Eighteenth Street NW, Washington, DC 200361802. Copyright 2009.

PAGE 51

51 Depression and Hopelessness as Ri sk Factors for Suicidal Behavior The myriad of interrelated socio-psychol ogical factors associated with suicide among young people makes pinpointing specific factors as causes impossible [124]. Therefore, prevention researchers focus on correlates of su icide, particularly hopelessness and depression [1]. Depressive disorders represent the most consistent disorders among young suicide victims [125]. About two-thirds of individuals w ho died by suicide experienced depressive symptoms at the time of their deaths [1]. Therefore, researchers r ecognize depression as a fundamental risk factor for suicidality among young people [3234, 36, 37, 124, 126-128]. In 2005, approximately 38% of college students reporte d feeling that things were hopeless, and more than a quarter (28%) attributed trouble functioni ng during the academic year to depressive symptoms [123]. The association between hopelessness and suicidal behavior among young people has only recently received attention, with disparate findings [37]. Thompson et al. [37] concluded, Whether depression or hopelessnes s or both factors simultaneously predict suicidal behaviors differs for adolescents compared to adults, fo r males versus females, and with the outcome variable measured [p.16]. In their studies of college students, Stephenson et al. [128] and Konick and Gutierrez [34] confirmed depressive symptoms and hopelessness as predictors of suicidal ideation among males and females. However, in Konick and Gutierrezs [34] study, depressive symptoms exerted a stronger influence on suicidal ideation than did hopelessness. Conversely, Thompson et al. [37] fo und direct effects of depression and hopelessness on suicidal behavior for males, but only a direct effect of hopelessness not depression for females. Kumar and Steer [39], also found hopelessness was more strongly related to suicidal ideation among adolescents than was depression.

PAGE 52

52 Physical Activity as a Protective Factor Against Suicidal B ehavior Research clearly indicates a positive asso ciation between exercise and psychological health. Physical activity promotes positive emotional well-being [87-89] including improvements in depressed mood [88, 90, 91], anxiety and stress [88, 92-94], and self-esteem [90, 95, 96]. Evidence suggests that exercise promotes a positive self-image, especially among young people with low self-esteem [95]. Physi cal activity may protect against suicidality through its effect on ps ychological well-being. Researchers have only recently examined poten tial positive effects of physical activity on suicidal behavior. The few studies published to date yielded equivocal findings. The results become more complex when examined by sex. In addition, the existent literature focuses on the association between physical activity and adolescent suicidal ity. Only 3 studies included young adults of typical college ag e [4, 10, 70], and only one study sp ecifically addressed suicidal behavior among college students [1 0]. Furthermore, only Brosnahan et al. [72] examined the effect of strength or toning ac tivities on suicidal behavior. Simon et al. [70] used a case-control de sign to investigate the relationship between involvement in physical activity and suicide risk among indivi duals aged 13 to 34 years. Researchers asked 4 questions to assess level of physical activity. First, participants provided information regarding past-month participation in any physical ac tivities or exercises such as running, calisthenics, golf, gardening, or walk ing for exercise [p.312]. Participants then reported the type of activity th ey spent the most time doing, amount of time they spent engaged in the activity, and frequency of this activity. Findings suggested a strong protec tive association between physical activity and nearly -lethal suicide attempts (NLSAs). Even after adjusting for the full set of potential confounders [sociodemogra phic characteristics, de pression, hopelessness, a serious medical condition, alcoholism, BMI, and social support], risk for

PAGE 53

53 NLSAs was five times higher among those who ha d not been physically active in the past month [70, p.312]. Regardless of intensity, frequency, or duration of a primary activity, participants who reported engaging in physical exercise demonstrated lower suicide risk. Ferron et al. [4] surveyed young people aged 15 to 20 years. They incorporated 2 questions to assess physical activ ity. The first question focused on frequency of sport activity, and the second question focused on belonging to a sports club. Athletic adolescents were young people who participated in sports 2 to 3 times per w eek and belonged to a sports club, or who participated in sports daily whether or not they belonged to a sports club. Non-athletic adolescents were individuals who engage d in sports once a week or never. Findings from this study determined that young people with the highest frequency of physical activity experienced superior psychological well-being. The most athletic adolescen ts felt sad, depressed, or desperate less often, and a lower proportion reported past suicidal thoughts or su icide attempts. Brosnahan et al. [72] examined the asso ciations among varying levels of different physical activities and suicidal behavior among high school students. To our knowledge, these investigators were the only ones to examine the effect of strength or toning activities on suicidal behavior. Findings indicated th at young people who par ticipated in a greater number of total physical activity sessions showed significantly le ss risk of considering suicide. In addition, higher levels of vigorous physical activity, stre ngth and toning activ ity, and total physical activity all were associated with decreased risk of planning suicide. Brown and Blanton [10] published the only inve stigation we are aware of that addresses the effect of physical activity on suicide risk am ong college students. They divided the college student sample into 5 activity level groups on the bases of responses to 2 physical activity questions. One question assessed participation in vigorous activity, and one other provided an

PAGE 54

54 estimate of participation in moderate-intensity activity. Compared with inactive men, males who reported participating in low-intensity activity dem onstrated reduced risk of suicidal behavior. However, neither moderately nor vigorously activ e males showed reduced suicide risk in this study. Furthermore, moderately or vigorously ac tive females demonstrated increased risk of suicidal behavior compared to less active females. Unger [9] found similarly complex relationships among high school students. Findings from this study indicated that increased physical activity was related to lower rates of suic idal behavior for males. Yet, high levels of physical activity related to higher rates of suicidal behavior for females. Though definitive reasons for these disparate results by sex remain unclear, researchers recognize that, in young females, exercise behavi or affects mental health in complex ways. Thome and Espelage [129] found that incr eased physical activity related to positive psychological health in college females, but only when exer cise was not associated with an eating disorder. Tiggeman and Williamson [130 ] observed significant negative relationships between increased exercise, and body dissatisf action and self-esteem among young females. Strelan et al. [131] found that females who endorsed appearance -related reasons for exercise (i.e., weight control, body tone attractiveness) also experi enced increased levels of body dissatisfaction and reduced self-esteem. These findings support the role of body appearance as an important determinant of physical and globa l self-esteem, especially among females [132]. Thus, factors related to poor body image may cont ribute to low self-esteem, depression, and suicidal feelings that likely influence the re lationship between physical activity and suicidal behavior [9, 10]. Conversely, depressed or suicidal individua ls may choose not to engage in physical activity. Symptoms associated with depressive disorders in cluding diminished pleasure in

PAGE 55

55 daily activities, preference for time alone, fatigue and psychomotor agitati on or retardation [133] may prevent young people from exercising. Tomori and Zalar [51] found that adolescent females demonstrated reduced suicide risk if th ey perceived physical act ivity as an important determinant of health and a beneficial coping behavi or during times of distress. For males in this study, reduced suicidality related to the attitude that exercise repr esents a healthy activity and to higher frequency of involvement [51]. Methods Procedures and Participants The NCHA addresses student health habits, behaviors, and perceptions related to substance use; sexual health; wei ght, nutrition, and exercise; mental health; and personal safety and violence. ACHA conducts the survey twice a year, using national samples of college students from numerous colleges and universities across the United States, with institutional data sets ranging from 8 to 117 school s. We obtained data relate d to the study questions by submitting a research proposal to the ACHA-NC HA Program Office (for instructions visit: http://www.acha-ncha.org/research.html). Participants in the spring 2005 administration of the NCHA included 43,499 college students. The student s ranged in age from 18 to 25 years (mean = 20.4 years, SD = 1.8 years). The sample included more females (n = 28,090, 64.5%) than males (n = 15,409, 35.5%). Participants reported their race/ethnicity as 76.3% white, 9.9% Asian, 5.2% Hispanic, 3.8% African American, 0 .9% American Indian, and 3.9% Other. We collapsed responses from partic ipants in the last 2 groups in to a single category of other.

PAGE 56

56 Measures Hopelessness ACHA researchers assessed hopelessness with 1 item: Within the last school year, how many times have you felt things were hopeless. We categorized participants as feeling hopeless if they answered 1 or more times. Depression Researchers assessed d epression with an item that asked how many times within the last school year students felt so de pressed that it was difficult to function. We categorized participants as feeling depressed if they answ ered 1 or more times. Suicidal behavior Two ACHA-NCHA items assessed suicide t houghts and attempts. We characterized participants as suicidal if they reported seri ously considering attempti ng suicide or had actually attempted suicide 1 or more times in the past sch ool year. We collapsed these data into a single variable because only about 1% of students reported a previous suicide attempt. Physical activity Researchers assessed frequency of aerobic ac tivity with 1 item: On how many of the past 7 days did you participate in vigorous exerci se for at least 20 minutes or moderate exercise for at least 30 minutes? They assessed fr equency of strength or toning activity termed toning activities with 1 other item: O n how many of the past 7 da ys did you do exercises to strengthen or tone your muscles, such as push-ups, sit-ups, or weight lift ing? On the basis of their responses to each item, we grouped partic ipants into 4 categories: 0 times/week, 1 to 2 times/week, 3 to 5 times/week, and 6 to 7 times/w eek. Thus, the categories were not mutually exclusive.

PAGE 57

57 Data Analysis We used descriptive statistics to compare rates of hopelessness, depression, and suicidal behavior for 2 groups: males who engaged in aer obic or toning activities at least once per week (n = 12,797) with males who were inactive (n = 2,609), and females who engaged in aerobic or toning activities at leas t once per week (n = 22,296) with fe males who were in active (n = 5,792). We conducted logistic regressi on analyses by sex to determine the independent effects of aerobic and toning activities on the risk of hopelessne ss, depression, and suic idal behavior. We performed subsequent analyses, controlling for participation in the alternate activity. This analysis helped clarify the association between t ype of activity and mental health. We included age and race as covariates in all models. The analyses determined the relative risk of hopelessness, depression, and suicid al behavior associated with different levels of aerobic and toning activities. Results Of 43,499 c ollege students, 65.4% reported feel ings of hopelessness, and 46.1% stated they felt so depressed it was difficult to functi on, 1 or more times during the past school year. Approximately 11% had seriously considered su icide or actually attempted suicide. Three quarters of these students engaged in aerobic ac tivity, and 64.2% engaged in strength or toning exercises, at least once per week. Table 31 shows the distribution of physical activity, hopelessness, depression, and suicid al behavior by sex. Males were more likely than females to participate in aerobic activity (78.3% vs. 74.7%, respectively, [N = 43,499, 1] = 70.01, p < .001) and toning exercises (67.1% vs. 62.6%, respectively, [N = 43,371, 1] = 87.98, p < .001) at least once per week. Compared with male s, females demonstrated increased rates of hopelessness (57.7% vs. 69.7%, respectively, [N = 43,509, 1] = 636.52, p < .001), depression (40.0% vs. 49.6%, respectively, [N = 43,461, 1] = 364.28, p < .001), and suicidal behavior

PAGE 58

58 (9.4% vs. 11.2%, respectively, [N = 43,593, 1] = 34.56, p < .001). For males and females, the prevalence of hopelessness decreased with age (Table 3-2). However, only younger female students reported significantly lowe r rates of depression and suicidal behavior. Participants in the categories of Asian, Hispanic, and other repo rted higher levels of hopelessness, depression, and suicidal behavior fo r both sexes (Table 3-3). Our findings demonstrated that physically active males experien ced lower rates of hopelessness, depression, and suicid al behavior than did their inact ive counterparts. More than half (57%) of physically active males experienced hopelessness, co mpared with 61% of inactive males ( [N = 15,329, 1] = 13.5, p < .001). About 39% of active males and 45% of inactive males reported feeling so depressed they had difficulty functioning ( [N = 15,306, 1] = 29.9, p < .001). Almost 9% of physically active males and 12% of inactive males had thought about suicide or attempted suicide ( [N = 15,357, 1] = 20.6, p < .001). We found similar patterns among females. More than half (69%) of active females experienced feelings of hope lessness, compared with 72% of inactive females ( [N = 27,985, 1] = 17.3, p < .001). About 49% of physically active fema les reported feeling so depressed they had difficulty functioning, compared with 54% of inactive females ( [N = 27,964, 1] = 46.8, p < .001). Only 11% of females who engaged in physical activity reporte d suicidal behavior, whereas about 14% of inactive females re ported suicidal thoughts or attempts ( [N = 28,042, 1] = 52.9, p < .001). Tables 3-4 and 3-5 present the relative risk of hopelessness, depr ession, and suicidal behavior for males and females who engaged in various levels of both aerobic and toning activities. Males and females who engaged in so me aerobic or toning activity each week showed reduced risk of feeling hopeless or depressed compared with their c ounterparts who did not

PAGE 59

59 engage in each activity. Males who performed aerobic activity also demonstrated reduced suicide risk: 1 to 2 times /week (odds ratio [OR] = .74, p < .001), 3 to 5 times/week (OR = .64, p < .001), 6 to 7 times/week (OR = .65, p < .001). Likewise, females who engaged in aerobic exercise each week showed reduced risk of suicidal behavior: 1 to 2 times/week (OR = .82, p < .001), 3 to 5 times/week (OR = .73, p < .001), 6 to 7 times/week (OR = .70, p < .001). Compared with the effects of aerobic activity, the effects of toning activities exhibited a dose response for both sexes. For males, engaging in moderate levels of toning activities produced optimal results. Males who performed toning acti vities 3 to 5 times/week were significantly less likely to demonstrate suic idal behavior (OR = .76, p < .001) than were males who did not engage in these activities. However, females demonstrated reduced suicide risk if they engaged in toning activities at low (1 to 2 times/week) and moderate (3 to 5 times/week) levels (OR = .84, p < .001 and OR = .79, p < .001, respectively) compared with females who did not perform toning activities. We found no interaction betw een aerobic activities and toning activities in any of our logistic regression models for ei ther sex. However, the two activities were strongly correlated: males (r = .68, p < .001) and females (r = .67, p < .001). Therefore, we examined the effect of each activity on levels of hopelessness, depressio n, and suicidal behavior while controlling for the effect of the alternate activity. For males, toning showed no significant effects on hopelessness, depression, or suicidal behavior after we accounted for participation in aerobic activity. For females, frequent toning activity (6 to 7 times/week) related to increased risk of hopelessness (OR = 1.17, p < .05) after we controlled for aerobic activity. In contrast, females who engaged in low (1 to 2 times/week) or mode rate (3 to 5 times/week) levels of toning activities were less likely to feel depressed (OR = .93, p < .05 and OR = .90, p < .01,

PAGE 60

60 respectively) than were their inactive counterparts, even after we controlled for aerobic activity. We found no significant effects of strength or toning activities on suicidal behavior for females after controlling for aerobic ac tivity. However, for both sexes, aerobic activity remained significantly associated with reduced risk of hope lessness, depression, and suicidal behavior after controlling for participation in toning activities. To more fully capture the complex relations hip between exercise behavior and mental health, we examined the interaction effects betw een physical activity and both exercising to lose weight and dieting to lose weight. Males who performed toning activities 3 or more times/week with the intent to lose weight were more likely to feel hopeless and depressed (OR = 1.17, p < .05 and OR = 1.31, p < .001, respectively). In addition, male s who engaged in aerobic or toning activities 3 or more times/week to lose weight demonstrated increased suicide risk (OR = 1.43, p < .05 and OR = 1.64, p < .001, respectively). Females w ho engaged in aerobic or toning exercises 3 or more times/week to lose weight showed an increased ri sk of hopelessness (OR = 1.25, p < .01 and OR = 1.41, p < .001, respectively) and depression (OR = 1.24, p < .01 and OR = 1.42, p < .001, respectively). Females who perfor med some toning activities each week with the intent to lose weight al so showed increased risk of suicidal behavior (OR = 1.29, p < .01). Furthermore, females who performed aerobic or toning activities 3 or more times/week and dieted to lose weight demonstrated increased risk of suicide (OR = 1.22, p < .05 and OR = 1.26, p < .01, respectively). Discussion In a recent national survey, c ounseling center directors identi fied the increase in reports of student self-injury (54.9% in 2004 vs. 68.7% in 2005) as a big concern [134]. The publication Healthy Campus 2010, which outlines U.S. public health priorities, sets goals to reduce the rate

PAGE 61

61 of completed suicide (Objective 18-1) and the rate of attempte d suicide among college students (Objective 18-2) [135]. To our knowledge, we are the first to examine the association between strength or toning activity and suicide risk among college students, and this is one of only two studies to examine the effects of aerobic activity on suicide risk among college students. Finings provide empirical evidence that establishes the association between physical activity and reduced suicidality among young adults. Our findings also document that ce rtain levels of physical activity relate to reduced risk of hopelessness and de pression, two major risk factors associated with suicidal behavior. This study adds to the extant literatu re by describing the effects of different types of physical activity on suicide risk. Though strength or toning activities dem onstrated a significant positive effect on suicidal behavior for both sexes, the association became non-significant after we controlled for participation in aerobic activ ity. Thus, our findings suggest that aerobic activity compared with strength or toning activities affords di stinct protection against suicide risk among college students. Our findings also support previous research ers who examined the association between physical activity and suicide risk. For example, similar to findings by Brosnahan et al. [72], Ferron et al. [4], and Simon et al. [ 70], we found that student s who participated in frequent aerobic activity showed significantly le ss risk of suicidality. Our results for males also support the findings of Unger [9] and Brown and Blanton [10]. In contrast, our results for females did not support relationships found by Unger [9] or Brow n and Blanton [10]. These researchers found that female exercisers, especi ally those who performe d high-intens ity activity, demonstrated increased suicide risk. Our findings showed that female college students benefited from the psychological advantages of physical exercise. Females who reported engaging in

PAGE 62

62 some physical activity each week de monstrated reduced risk of hopelessness, depression, and suicidal behavior. This study also adds to the literature regarding who receives psychol ogical health benefits from physical activity and whose me ntal health may suffer as a resu lt of frequent exercise [129]. Though our study confirmed the association betw een physical activity a nd enhanced mental health, we also found that motivation underlying ex ercise behavior must be considered. We observed significant associations between engaging in frequent exer cise with the intent to lose weight and increased risk of hopelessness, depre ssion, and suicidal behavior. Furthermore, our findings indicated that females who attempted to lose weight by combining frequent physical activity and deleterious dieting be havior also demonstrated great er suicide risk. Our results support theories that propose a complex relationship between body image and mental health, especially among females [9, 10, 129]. Strelan et al. [131] found that reasons for exercise behavior among young females mediat ed the relationships betwee n self-objectif ication and body dissatisfaction, and self-esteem. In Strelan et al.s [131] study, fe males who exercised for weight control, body tone, or attractiv eness experienced increased body dissatisfaction and reduced selfesteem, compared with females who valued functi onal reasons for exercise (e.g., health, fitness, mood, enjoyment) [131]. Wild et al. [47] found that young fe males with low body image selfesteem, and those who reported low athletic self -esteem, were at increas ed risk of exhibiting suicidal behavior. Research addressing abnor mal eating behaviors among females also supports positive associations between frequent exercise and disordered eating symptomatology [136], as well as between eating disord ers and suicide risk [29]. Limitations These data originated from a cross-sectional survey, so we were unable to determine causal relationships thro ugh the analyses. For in stance, physical activity may mediate the lower

PAGE 63

63 risk of suicidality or, conversely, suicidal young people may choose not to engage in physical exercise. Several symptoms associated with de pressive disorders may prevent individuals from exercising [133]. Longitudinal and case-control studies can he lp identify potential cause-andeffect relationships between physical ac tivity and suicidal behavior. Future investigations should explore reasons why young people part icipate in physical activity to determine if differences in involvement and non-involvement exist based on suicide status. Our sample included young people currently attending college, which, by default, excluded individuals who did not attend college due to emotional or behavioral problems and those who succeeded in their suicide attempts. Pa rticipants who failed in their suicide attempts may possess different characteristic s than do individuals who died by suicide. In addition, these data were obtained through self-report rather than objective measures and, because of procedures followed for the survey, we could not validate the self-reports. Participants completed a confidential survey, but they may have altered th eir responses to avoid embarrassment. Physical activity assessments based on use of movement de vices, as well as medical record information documenting attempted suicides, can produce greater precision of measurement. Conclusions These findings highlight th e prevalence of hopelessness, depression, and suicidal behavior among college students. They also indicate that younge r students, females, and some racial minority groups experience in creased prevalence of suicidalit y. Research suggests sex and race differences in the experience of and respons e to social strains associated with suicidal behavior [45, 137]. Thus, researchers need to account for sex and race differences when investigating suicidal be havior among young people. Our findings showed significantly lower rate s of hopelessness, depr ession, and suicidal behavior among college students wh o engaged in physical activity, compared with their inactive

PAGE 64

64 counterparts. We also found si gnificant relationships betwee n physical activity, especially aerobic activity, and reduced risk of hopelessness, depression, and suicidal behavior among male and female college students. Our findings may s upport the use of physical activity as an adjunct modality in the treatment of depression [138] However, practitioners working with young adults should assess their reason s for exercising to determine whether their motivation relates to appearance or health and enj oyment. Helping young people change their motivations for physical activity from a means to look good to an opportunity to feel good may improve their body satisfaction and self-esteem [131], as well as reduce their suicide risk. Furthermore, individuals in positions to detect disordered eating patterns and unfavorable perceptions of body appearance should become knowledg eable about the links to suicid e risk and make referrals as necessary [103]. Future investigations should identify the mediating f actors between physical activity and decreased suicide risk. Exploring the intricate connec tion between reduced levels of suicidality and participation in physical activity holds promise for improving our understanding of unique protective factors. Understanding the mechanisms th at confer protection against suicidality will benefit suicidal individuals, pare nts, university personnel, practitioners working in community organizations, and mental health clinicians working with young adults. Findings from methodologically sound research may provid e a foundation for interven tion research in the prevention of suicide.

PAGE 65

65 Table 3-1. Physical activity, hopelessness, depression, and suicidal behavior among college students Males Females Variable n % n % Chi-square Aerobic activity No aerobic activity 3,350 21.7 7,114 25.3 99.33*** 1 to 2 times/week 4,826 31.3 8,691 30.9 3 to 5 times/week 5,643 36.6 9,931 35.4 6 to 7 times/week 1,590 10.3 2,354 8.4 Strength or toning activity No strength or toning activity 5,051 32.9 10,49137.4 316.31*** 1 to 2 times/week 4,043 26.3 8,413 30.0 3 to 5 times/week 5,168 33.7 7,810 27.9 6 to 7 times/week 1,083 7.1 1,312 4.7 Hopelessness, depression, and suicidal behavior Felt hopeless 8,891 57.7 19,60369.7 636.52*** Felt depressed 6,153 40.0 13,91749.6 364.28*** Thought about suicide or attempted suicide 1,450 9.4 3,156 11.2 34.56*** *** p < 0.001.

PAGE 66

66 Table 3-2. Hopelessness, depres sion, and suicidal behavior am ong college students of various ages Males Females Variable n % Chi-square n % Chi-square Felt hopeless 18-19 years old 3,105 58.0 21.05*** 7,548 72.0 78.27*** 20-22 years old 4,558 58.8 9,793 69.5 23-25 years old 1,228 53.5 2,262 64.2 Felt depressed 18-19 years old 2,104 39.4 1.70 5,275 50.4 6.78* 20-22 years old 3,137 40.5 6,955 49.3 23-25 years old 912 39.7 1,687 47.9 Thought about suicide or attempted suicide 18-19 years old 531 9.9 5.25 1,285 12.2 24.96*** 20-22 years old 729 9.4 1,542 10.9 23-25 years old 190 8.2 329 9.3 p < 0.05, ** p < 0.01, *** p < 0.001.

PAGE 67

67 Table 3-3. Hopelessness, depres sion, and suicidal behavior am ong college students of various races and ethnicities Males Females Variable n % Chi-square n % Chisquare Felt hopeless White 6,708 56.5 37.38*** 14,422 68.6 73.09*** Asian or Pacific Islander 926 61.9 2,110 75.0 Hispanic 471 63.6 1,089 71.7 African American 278 59.4 781 68.0 Other 438 63.0 1,021 75.0 Felt depressed White 4,595 38.8 49.57*** 10,152 48.3 86.97*** Asian or Pacific Islander 684 45.9 1,535 54.7 Hispanic 338 45.7 799 52.7 African American 177 37.9 539 46.8 Other 317 45.7 774 56.9 Thought about suicide or attempted suicide White 1,047 8.8 30.79*** 2,201 10.4 53.80*** Asian or Pacific Islander 179 12.0 400 14.2 Hispanic 92 12.4 185 12.2 African American 39 8.3 140 12.1 Other 84 12.1 195 14.3 *** p < 0.001.

PAGE 68

68 Table 3-4. Relative risk of hopelessness, depression, and suicidal behavior for college students engaging in various levels of aerobic activ ity compared to thos e not reporting any aerobic activity Felt hopeless Felt depressed Thought about suicide or attempted suicide Level of aerobic activity OR 95% CI OR 95% CI OR 95% CI Malesa 1 to 2 times/week 0.88** (0.81, 0.97) 0.82*** (0.75, 0.89) 0.74*** (0.64, 0.86) 3 to 5 times/week 0.78*** (0.72, 0.86) 0.74*** (0.68, 0.81) 0.64*** (0.55, 0.74) 6 to 7 times/week 0.75*** (0.66, 0.84) 0.72*** (0.64, 0.82) 0.65*** (0.53, 0.80) Femalesa 1 to 2 times/week 0.94 (0.88, 1.01) 0.91** (0.85, 0.97) 0.82*** (0.75, 0.91) 3 to 5 times/week 0.83*** (0.77, 0.89) 0.80*** (0.75, 0.85) 0.73*** (0.66, 0.80) 6 to 7 times/week 0.81*** (0.73, 0.89) 0.76*** (0.69, 0.83) 0.70*** (0.60, 0.82) Note. aThough both are included in the same table, the data for males and females represent separate analyses. bOdds ratios were adjusted for age and race/ethnicity. ** p < 0.01, *** p < 0.001.

PAGE 69

69 Table 3-5. Relative risk of hopelessness, depression, and suicidal behavior for college students engaging in various levels of strength or toning activity compared to those not reporting any toning activity Felt hopeless Felt depressed Thought about suicide or attempted suicide Level of strength or toning activity OR 95% CI OR 95% CI OR 95% CI Malesa 1 to 2 times/week 1.01 (0.93, 1.01) 0.98 (0.90, 1.07) 0.98 (0.85, 1.13) 3 to 5 times/week 0.89** (0.82, 0.96) 0.84*** (0.78, 0.91) 0.76*** (0.66, 0.87) 6 to 7 times/week 0.83** (0.73, 0.95) 0.85* (0.74, 0.98) 0.82 (0.65, 1.03) Femalesa 1 to 2 times/week 0.92** (0.86, 0.98) 0.89*** (0.84, 0.94) 0.84*** (0.77, 0.92) 3 to 5 times/week 0.89*** (0.84, 0.95) 0.81*** (0.76, 0.86) 0.79*** (0.72, 0.87) 6 to 7 times/week 0.96 (0.84, 1.09) 0.85** (0.76, 0.96) 0.88 (0.73, 1.06) Note. aThough both are included in the same table, the data for males and females represent separate analyses. bOdds ratios were adjusted for age and race/ethnicity. p < 0.05, ** p < 0.01, *** p < 0.001.

PAGE 70

70 CHAPTER 4 HIGH SCHOOL YOUTH AND SUICIDE RISK : EXPLORI NG PROTECTION AFFORDED THROUGH PHYSICAL ACTIVITY AND SPORT PARTICIPATIONi Introduction Suicide represents the th ird leading cause of death for youth 15 to 24 years, accounting for 12.9% of all deaths in this age range [1]. According to th e Centers for Disease Control and Prevention (CDC), between 2003 and 2004, the suic ide rate increased 18% for youth under age 20, and suicides constituted the only cause of de ath that increased among adolescents [17]. In 2004, youth 15 to 24 years represented 14.2% of th e U.S. population and comprised 13.3% of the suicides [1]. Though females attempt suicide mo re often than males, young males aged 15 to 19 are 3.6 times more likely than females to complete suicide [1]. Furthermore, for each completed suicide, an estimated 100 to 200 adolescents atte mpted to take their own lives [1]. Thus, Healthy People 2010 specifically targets reducing the rate of completed suic ide (Objective 18-1) and the rate of attempted suicide among adolescents (Obj ective 18-2) [20]. The National Institute of Mental Health urges researchers to focus on decreasing the adolescent suicide rate by studying risk and protective factors [21]. Research indicates that physical activity affords the same psychological benefits to adolescents as to adults. Physical activity promotes positive em otional well-being [89] including improvements in depressed mood [88, 91, 139], a nxiety and stress [88, 94] and self-esteem [89, 91, 140]. Therefore, through its effect on psychological well-being, physical activity may protect against suicidality. Physical activity in the context of team sports may afford additional protection by facilitating soci al support and integration [12] Conversely, youth involved in i Journal of School Health, 78, 10, 545-553, 2008. Reprinted with permission from Blackwell Publishing. Published by Wiley-Blackwell, 9600 Garsington Road, Oxford OX4 2DQ, UK. Copyright 2008.

PAGE 71

71 sports may benefit from psychosocial advantages that increase the like lihood of participation [141]. Researchers have only recently examined th e possible protective association between physical activity, sport participation, and suicid e risk. However, the few studies completed yielded equivocal findings. The results become more complicated when examined by sex. For example, Brosnahan et al. [72] f ound adolescents significan tly less likely to plan suicide if they engaged in frequent, vigorous activity. Yet, they did not find a significa nt association between sport participation and suicidal behavior. Conversely, Brown and Blan ton [10] found that, compared to inactive males, college students who reported low activity levels demonstrated reduced risk of suicidal behavior. However, neither moderately nor vigorously active males showed reduced suicide risk. Furthermore, m oderately and vigorously active females actually showed increased levels of suicidal behavior, co mpared to inactive females. Notably, this study revealed that sport participation protected agains t suicidal behavior in both males and females. Compared to sport participants, male non-participants had 2.5 times the odds of reporting suicidal behavior, and female non-participants were 1.67 times more likely to report suicidal behavior. Unger [9] found similarly complex relationshi ps. Physical activity, especially when combined with team sports, related to lower rates of suicidal behavior for males. Yet, frequent, vigorous aerobic activity, especially without team sports, related to higher rates of suicidal behavior for females. Though definitive reasons for these disparate results by sex remain unclear, researchers recognize that in young females, exercise behavior impacts mental health in complex ways. Thome and Espelage [129] found that exercise related to positive psychological health in college

PAGE 72

72 females, but only when exercise remained unassociated with an eating disorder. Hayes et al. [132] highlighted body appearance as an important determinant of physical and global selfesteem, especially among females. Therefore, factors related to poor body image may contribute to low self-esteem, depression, and suicidal f eelings that could influence the relationship between physical activity and su icidal behavior [9, 10]. Conversely, depressed or suicidal youth may choose not to engage in physical activity. Symptoms associated with depressive disorders may prevent indivi duals from exercising including diminished pleasure in daily activi ties, preference for time alone, fatigue, and psychomotor agitation or retardation [133]. Tomo ri and Zalar [51] found that adolescent females demonstrated reduced suicide risk if they percei ved physical activity as an important determinant of health, and as a beneficial coping behavior during times of di stress. For males in this study, reduced suicidality related to the attitude that ex ercise represents a healthy activity, as well as to higher frequency of involvement. Methods Procedures and Participants This study investigated the relationships between physical activit y, sport participation, and adolescent suicidal beha vior based on data from the 2005 Youth Risk Behavior Survey (YRBS) [79]. This epidemiologic surveillance system, established by the CDC, monitors six categories of priority health-risk behaviors am ong youth, including behaviors that contribute to unintentional injuries and violence, tobacco use, alcohol and other drug use, sexual behaviors, unhealthy dietary behaviors, and physical inactivity. The YRBS constitutes a cross-sectional survey conducted every 2 years among representative samples of 9thto 12th-grade students from schools across the United States.

PAGE 73

73 Based on the preceding findings, this study posed 4 research questions: 1) Do adolescent males who engage in physical activity demonstrate reduced risk of hopelessness and suicidal behavior compared to inactive males?; 2) Do adolescent females who engage in vigorous physical activity demonstrate increased risk of ho pelessness and suicidal behavior compared to inactive females?; 3) Do athletes of both se xes demonstrate reduced risk of hopelessness and suicidal behavior compared to non-athletes?; and 4) Do a subset of highly involved athletes of both sexes demonstrate reduced risk of hopelessn ess and suicidal behavior compared to nonathletes? Participants included 13,857 students in gr ades 9 to 12 who completed the 2005 YRBS. Students ranged in age from 12 to 18 years (mean = 16.2 years, SD = 1.2 years). The sample included slightly more females (51.9%) than males (48.1%). Respondents reported their race/ethnicity as 44.7% white, 24.5% African American, 15.1% Hispanic, 8.4% multiple Hispanic, 2.9% multiple non-Hispanic, 2.7% Asian, 1.1% American Indian, and 0.7% Native Hawaiian. Measures Suicidal behavior To assess suicidal ideation and behavior, part icipants responded to 3 questions regarding their thoughts and actions during th e previous 12 months. Students indicated whether or not (yes or no) they had ever felt sad or hopeless almost everyday for two weeks in a row, seriously considered attempting suicide, or made a suicide plan, and how many times they actually attempted suicide. This assessment was found to have a good internal reliability (Cronbachs alpha = .81). Few youth reported attempting suic ide more than 2 or 3 times (1.5%), so we created 2 dichotomous measures one indicating whether or not a participant had ever attempted suicide, and one indicating whether or not a participant had attempted suicide more than once.

PAGE 74

74 Physical activity Frequency of physical activity was assessed with 1 item: On how many of the past 7 days did you exercise or participate in physical activity for at least 20 minutes that made you sweat and breathe hard, such as basketball, so ccer, running, swimming laps, fast bicycling, fast dancing, or similar aerobic activit ies? Based on their response s, we grouped students into 4 categories: zero times/week, 1 to 2 times/w eek, 3 to 5 times/week, and 6 to 7 times/week. Sport participation Sport participation was assessed by 1 item: During the past 12 months, on how many sports teams did you play? (Incl ude any teams run by your school or community groups.). First, we divided participants into 2 groups: thos e who participated in an y number of sports, and those who did not participate. We subsequent ly grouped participants based on level of sport involvement: non-athletes (participated on no team s), moderately involved athletes (participated on 1 or 2 teams), and highly i nvolved athletes (participate d on 3 or more teams). Data Analysis Descriptive statistics were calculated to compare rates of hopelessness and suicidal behavior for male athletes (n = 3818) with male non-athletes (n = 2444), and for female athletes (n = 3281) with female non-athletes (n = 3609). Logistic regression analyses by sex were conducted to determine if physical activity and sport participation decreased the risk of feeling hopeless, suicidal thoughts, suicide plans, suicid e attempts, or multiple suicide attempts. Age and race were included as covariates in all mode ls. We also controlled for sport participation during the physical activity anal yses and for physical activity dur ing the sport analyses. Our analysis determined the relative risk of hopelessness and suicidal ity associated with different levels of physical activity as well as sport participation.

PAGE 75

75 Results Of 13,857 a dolescents, 16.9% had seriously co nsidered attempting suicide, 13.3% had made a suicide plan, 9.0% had attempted suicide, and 4.2% had attempted suicide multiple times. Most (81.8%) had engaged in physi cal activity at least once per w eek, and 54.0% participated in sports. Table 4-1 shows the dist ribution of physical activity, team sports, and suicidal behavior by sex. Males were more likely than females to participate in physical activity at least once per week (87.1% vs. 77.0%, respectively, = 229.7, p < .001). Males also showed higher rates of sport participation than did fema les (61.0% vs. 47.6%, respectively, = 235.4, p < .001). Females demonstrated increased prevalence of suicidality, with 21.8% reporting thoughts or behavior, compared w ith 11.7% of males ( = 250.6, p < .001). For females, but not males, the risk of hopelessness and suicidalit y decreased with age (Table 4-2). Table 4-3 presents the relative risk of hopelessne ss and suicidality for students reporting different races, compared to their white counterparts. Among both sexes, American Indian, multiple Hispanic, and multiple non-Hispanic students showed greater suicide risk. Among females only, Hispanic and Hawaiian stude nts also demonstrated increased risk. However, American Indian and Hawaiian stude nts were not oversampled, so these groups contain the smallest samples a nd the widest confidence interval s. Among both sexes, African American students demonstrated less suicide risk. Our findings revealed that male athletes experienced lower rates of hopelessness and suicidality. Hopelessness was e xperienced by 18% of athletes, compared to 25% of non-athletes ( = 39.8, p < .001). About 9% of male athletes had seriously considered suicide, compared to 14% of male non-athletes ( = 29.9, p < .001). Only 8% of athletes had a suicide plan, while over 11% of non-athletes had a suicide plan ( = 20.5, p < .001). Relatively few males actually

PAGE 76

76 attempted suicide: 4% of athletes and 7% of non-athletes had attempted suicide ( = 16.9, p < .001). We found similar patterns among females. Hopelessness was experienced by 34% of athletes, compared to 41% of non-athletes ( = 30.8, p < .001). About 20% of female athletes and 23% of female non-athletes ha d seriously considered suicide ( = 6.9, p < .05). Similarly, 15% of athletes and 18% of non-athletes had created a suicide plan ( = 8.4, p < .01). No significant difference was found for attempting suicid e. Approximately 9% of female athletes had attempted suicide, compared to about 10% of female non-athletes. Table 4-4 presents the relative risk of suicidality for male and female adolescents who engaged in various levels of phys ical activity. After controlling for sport participation, we found no relationship between low or m oderate levels of physical activ ity and suicidality among male adolescents. However, males who exercised 6 to 7 times/week showed reduced risk of planning suicide (odds ration [OR] = 0.67, p < .01), attempting suicide (OR = 0.65, p < .05), and attempting suicide multiple times (OR = 0.49, p < .01). Conversely, females who exercised 1 to 2 times/week were more likely to feel hopeless (OR = 1.22, p < .01), compared to inactive females. Though we did not find a significant differe nce in the rates of physical activity between females who reported suicidal behavior and t hose who did not, our fi ndings revealed a link between suicidality and intention motivating ex ercise behavior. Specifically, females who exercised 1 to 2 or 6 to 7 times/week to lose we ight demonstrated increased suicide risk (OR = 1.5, p < .05 and OR = 1.7, p < .05, respectively). Furthermore, females who engaged in some physical activity each week were more likely to consume fewer calories or foods low in fat (OR = 1.6, p < .001) and to take diet supplements (OR = 1.4, p < .05) to lose weight. These 2

PAGE 77

77 behaviors significantly increased suicide risk (OR = 1.71, p < .001 and OR = 3.13, p < .001, respectively). Sport participation was significantly associ ated with reduced the odds of hopelessness and suicidal behavior among both sexes (Table 4-5). Male athl etes were less likely to feel hopeless (OR = 0.71, p < .001), and consider (OR = 0.71, p < .001), plan (OR = 0.76, p < .01), or attempt (OR = 0.70, p < .01) suicide. Highly involved male athletes had reduced odds of feeling hopeless (OR = 0.63, p < .001) and of considering (OR = 0.66, p < .01) or planning (OR = 0.68, p < .01) suicide than non-athletes. Female athletes demonstrat ed reduced risk of hopelessness (OR = 0.76, p < .001), consideri ng suicide (OR = 0.80, p < .001), planning suicide (OR = 0.79, p < .001), and attempting suicide multiple times (OR = 0.76, p < .05). Compared to non-athletes, highly involved female athletes were less likely to feel hopeless (OR = 0.63, p < .001) or consider suicide (OR = 0.79, p < .05). Discussion These data originated from a cross-sec tional survey, so one limitation involves the inability to determine causal relationships. Fo r instance, sport participation may mediate the lower risk of suicidality or, conversely, suicidal youth may choose not to become involved in sports. High school athletes enjoy elevated popula rity [141] and self-estee m [5]. Youth at risk for suicide often feel socially isolated and re jected by their peers [142] Therefore, possessing confidence and feeling socially accepted may represent factors that prompt youth to participate in sports, rather than outcomes of participa tion. Longitudinal and cas e-control studies would help clarify the direction of causality. Also, our sample included youth currently attending high school, which excluded adolescents who do not atte nd school because of emotional or behavioral problems, and youth who succeeded in their suic ide attempts. The youth in our sample who failed in their suicide attempts may possess differe nt characteristics than suicide completers.

PAGE 78

78 These data also reflect self-report rather than objective measures, a nd we did not validate self-reports. Though respondents co mpleted a confidential survey, they may have altered their responses to avoid embarrassment. Physical ac tivity assessments based on movement devices, and medical record information documenting attempted suicides, would produce greater precision of measurement. Psychometric tests indicated our assessment of suicidality was reliable; however, the 3question format could represent a limitation. Reynolds and Mazza [2] caution that research on suicidal ideation or group a ssignment predicated on one or two questions regarding having thoughts of suicide lacks sufficient specificity for defining this domain of suicidality and obscures the relative importance of suicidal ideation as a component of suicide risk in youth [2, p.533]. Despite these limitations, our findings c ontribute to understandi ng physical activity and sport as protective factors against adolescent suicide. Though our analyses did not focus on age and race, noteworthy patterns emerged among these variables. Consistent with previous research, su icide risk decreased w ith age for females, but not for males [12]. Our findings regarding race also support previ ous research [143, 144]. Compared to white students, American Indian, multiple Hispanic, and multiple non-Hispanic students demonstrated increased suicide risk. Hispanic and Hawaiian females also showed increased risk, compared to white females. However, among both sexes, African American students were less likely to c onsider and plan suicide than their white counterparts. Physical Activity Our first research question te sted whether or not engaging in physical activity decreased the risk of suicidality among males. After cont rolling for sport, we found that only frequent exercise significantly reduced suicide risk. Al so tested was whether or not vigorous activity

PAGE 79

79 increased suicide risk among females. We foun d only one statistically significant relationship between low levels of activity and increased risk of hopelessness. Only 3 studies published to date have investig ated the separate eff ects of physical activity and sport [9, 10, 72]. Other investigations either combined phys ical activity and sport into a composite variable [4, 51, 70, 71] or examined sport independently [5-8, 11, 12, 145]. Though Brosnahan et al. [72] found that youth who par ticipated in higher levels of vigorous activity showed less suicide risk, the researchers did not perform sex-specific analyses. Therefore, our results regarding males, but not females, s upport the findings of Br osnahan et al. [72] Our findings also support relationships found by Unger [9] and by Brown and Blanton [10]. Unger [9], who determined the relative risk of suicidality associated with 6 physical activity/team sports combinations found consistently lower rate s of suicidal behavior among males who engaged in various levels of physical ac tivity. However, female exercisers, especially those who performed high-intensity activity, dem onstrated increased suicide risk. Though our study did not find as many statistically significan t relationships between physical activity and suicidal behavior, our results for males and females generally support Ungers [9] findings. Similar to methods by Brown and Blanton [10] we controlled for s port participation in our physical activity analysis. Their study found that male college students who participated in low-intensity activity demonstrated reduced suic ide risk. However, neither moderately nor vigorously active men showed decreased risk in Brown and Blantons [10] study. Similar to findings by Unger [9], Brown and Blanton [ 10] found that physically active females demonstrated increased risk comp ared to less active females. One possible explanation for inconsistenc y between our findings and findings by Brown and Blanton [10] involves methods used to de termine physical activity and suicide behavior.

PAGE 80

80 Brown and Blanton [10] categorized students into 5 activity levels based on their responses to 2 questions: one assessed vigorous activity, and one other estimated moderate-intensity activity. In contrast, we categorized our participants into 4 physical activity groups based on their response to 1 question assessing hi gh-intensity activity. Brown and Blanton [10] also collapsed 3 items that assessed suicide thoughts, plans, and attempts into a single va riable. Our analysis examined these behaviors separately. Another ex planation involves disparate samples. Whereas Brown and Blanton [10] addressed behavior among college stude nts, we examined high school youth behavior. Consistent with previous research [51], we did not find a significant difference in the rates of physical activity behavi or between females who reported suicidal tendencies and those who did not. However, for females, exercising to lose weight, as we ll as restricting calories or taking diet supplements to attain the same goal, significantly increased suic ide risk. Therefore, among females in our sample, findings regarding suicide risk and physical activity appear to support theories that suggest a complex relatio nship between body image and suicidal behavior [9, 10, 129]. Our findings indicate that physical activit y alone may not protect against adolescent suicidality, especially among young females. Po sitive relationships between physical activity and reduced suicidality found in past research th at used a composite independent variable may reflect the effects of sport rather than the benefits provided by physical activity. Sport Participation Findings regarding sport participation support previous investigations that compared the relative risk of suicidal behavior between athletes and non-athletes [6, 9, 11, 12]. After controlling for physical activity, sport participatio n remained a significant factor in reducing suicidality among both sexes. Compared to non-ath letes, male athletes sh owed reduced risk of

PAGE 81

81 feeling hopeless and considering, planning, or at tempting suicide. Young males involved in multiple sports attained even greater protection against hopele ssness and suicidality than nonathletes. Compared to non-athl etes, female athletes demonstrated reduced risk of hopelessness, considering suicide, planning suicide, and attempting suicide multiple times. Highly involved female athletes were less likel y to feel hopeless or consider su icide than non-athletes. These findings support Brown and Blantons [10] investig ation of sports effect on suicidal behavior among college students. Research suggests that physical activity provides significant mental health benefits. Our findings indicate that, in addition to physical activity, sport may protect against suicidality through other mechanisms. We hypothesize th at sport offers unique protection against adolescent suicidality by pr oviding social support and integr ation. Youth who report strong social support exhibit higher levels of resilien ce, less hopelessness, and re duced suicide risk [41, 58]. Adolescents demonstrate less suicide risk if they perceive family, friends, and peers as accepting; possess more positive friendships; and f eel connected to school [146]. Bearman and Moody [46] found that the friends hip environment affected suicidality for both males and females. However, for females, social networ k effects played an especially prominent role. Females who were socially isolated from the adolescent community demonstrated significantly greater suicide risk than females embedde d in cohesive friendship groups [46]. Extracurricular activities provide adolescents an opportunity to establish positive social relationships and networks [ 147]. Harrison and Narayan [5] examined relationships among sport, other extracurricular activ ities, psychological factors, and suicidal behavior. Findings indicated that those involved in team sports at school (a lone or combined with other activities)were most likely to report high self-e steem and were least likely to report sadness,

PAGE 82

82 anxiety, and suicidal behavior [5, p.118]. These authors co ncluded that engaging in any organized activity imparted some benefit, but sport participation afforded unique positive advantages [5]. Athletes may experience greater social in tegration when they become members of a social network that includes teammates, coach es, health professionals, family, and community [12]. The team sport environment represents a fertile ground for adolescent self-esteem development because teams provide opportunities fo r youth to engage with adults and peers to achieve collective goals [108]. Th rough its capacity to foster f eelings of social support and integration, sport participation ma y create a distinct form of protection against risk factors associated with adolescent suicide. Females, especially, may benef it from physical activity performed in the context of team sports. In addition to athletic competition, females also participate in sports to improve their fitness levels, improve thei r athletic skills, and enjoy group interaction [103]. A report from The President s Council on Physical Fitn ess and Sports [103] noted that sport participation can enhance mental health by offering adolescent females positive feelings about body image, improved self-estee m, tangible experiences of competency and success, and increased self-confidence [p.15]. Research supports the positive association between sport participation and improved self-concept among young females. For example, Pedersen and Seidman [108] found that females team sport achievement experiences in early adolescence related to increased self-esteem in middle adolescence. Similarly, Richman and Shaffer [76] found that high school sport participation contribute d to higher self-esteem among female college students. Conclusions This study confirmed findings from prev ious research on the high prevalence of hopelessness and suicidal behavior among adoles cents. Our findings also reinforced past

PAGE 83

83 research indicating that younger students, female s, and some racial minorities experience increased suicide risk. Research suggests se x and race differences in the experience of and response to social strains associat ed with suicidal behavior [45, 137]. Thus, researchers need to account for sex and race differences when investigating adolescent suicide. Our findings showed significantly lower rates of hopelessness and suicidality among male and female athletes, compared to their non-athlete counterparts. We also found a significant relationship between fr equent, vigorous activity and re duced risk of hopelessness and suicidality among male adolescents. However, low levels of ac tivity were actually associated with increased risk of feeli ng hopeless among females. Furthermore, among females, exercising and engaging in deleterious diet ing behavior to lose weight were associated with increased suicide risk. Individuals in a primary position to recognize disordered eating patterns and unfavorable perceptions of body appearance must become knowledgeable about these links and make referrals as necessary. For both sexes, we found that sport particip ation related to reduced risk of hopelessness and suicidal behavior. By controlling for physical activity, while examining the effect of sport, our study makes a unique contributio n to the literature re garding adolescent suicidality and sport participation. Our findings suggest that mechanis ms other than physical ac tivity contribute to the protective association between s port and reduced suicidality. Fu ture investigations should identify the mediating factors between spor t involvement and decr eased suicide risk. Researchers should examine the effect of social connectedness within sport involvement and how this effect specifically rela tes to suicidal behavior. Rese archers also should conduct casecontrol studies to identify potential cause-andeffect relationships between physical activity, sport participation, and adolescent suicidality. In addition, future investigations should explore

PAGE 84

84 reasons why adolescents partic ipate in physical activity to determine if differences in involvement and non-involvement exist based on suicidal status. Findings that indicate athletes of both sexes display less suic ide risk compared to their non-athlete counterparts s uggest opportunities for pr evention and merit seri ous consideration in public health policy and planning [12]. Explorat ion of the intricate c onnection between reduced suicidality and adolescent participation in spor ts holds promise for improving our understanding of unique protective factors. Understanding the mechanisms th at confer protection against adolescent suicidality will benefit suicidal you th, parents, school pe rsonnel, community youth organizations, and mental health clinicians wo rking with youth. Findings from methodologically sound research may provide a foundation for in tervention research in the prevention of adolescent suicide.

PAGE 85

85 Table 4-1. Physical activity, team sports participation, feelings of hopelessness, and suicidality in adolescent males and females Males Females Variable n % n % Chi-square Physical activity No physical activity 810 12.9 1590 23.0 229.7*** 1 to 2 times/week 1024 16.2 1603 23.2 3 to 5 times/week 2149 34.1 2396 34.7 6 to 7 times/week 2319 36.8 1313 19.0 Sport participation No sport 2444 39.0 3609 52.4 235.4*** Sport participant 3818 61.0 3281 47.6 Level of sport involvement No Sport 2444 39.0 3609 52.4 319.7*** Moderately involved (1-2 teams) 2744 43.8 2664 38.7 Highly involved (3 + teams) 1074 17.2 617 9.0 Hopelessness and suicidal behavior Felt sad or hopeless 1417 21.5 2719 38.1 447.7*** Thought about suicide 772 11.7 1558 21.8 250.6*** Planned suicide 641 9.7 1187 16.6 141.9*** Attempted suicide 371 6.3 745 11.5 101.8*** Multiple suicide attempts 174 3.0 349 5.4 44.9*** *** p < 0.001.

PAGE 86

86 Table 4-2. Physical activity, team sports participation, feelings of hopelessness, and suicidality in adolescent males and females OR 95% (Confidence Interval) Age group (years) Felt sad or hopeless Thought about suicide Planned suicide Attempted suicide Multiple suicide attempts Malesa < 12-14 years old 1.00 (referent) 1.00 (referent) 1.00 (referent) 1.00 (referent) 1.00 (referent) 15-16 years old 1.05 (0.82, 1.35) 1.16 (0.84, 1.61) 1.13 (0.80, 1.59) 1.32 (0.82, 2.11) 1.30 (0.66, 2.55) 17 years or older 1.15 (0.89, 1.47) 1.18 (0.85, 1.64) 1.00 (0.70, 1.41) 1.00 (0.62, 1.61) 1.24 (0.63, 2.45) Femalesa < 12-14 years old 1.00 (referent) 1.00 (referent) 1.00 (referent) 1.00 (referent) 1.00 (referent) 15-16 years old 1.03 (0.86, 1.23) 1.01 (0.82, 1.23) 0.95 (0.77, 1.19) 0.97 (0.75, 1.27) 0.91 (0.63, 1.32) 17 years or older 0.94 (0.78, 1.12) 0.76** (0.61, 0.93) 0. 71** (0.57, 0.90) 0.66** (0.50, 0.87) 0.65* (0.44, 0.96) Note. aThough these analyses are presented in a separate table, the variab les were included as covariates in the physical activity and sport analyses. p < 0.05, ** p < 0.01.

PAGE 87

87 Table 4-3. Relative risk of hopelessness and suicidality for adolescents of different races/ethnicities compared to white adolescents OR 95% (Confidence Interval) Race/ ethnic group Felt sad or hopeless Thought about suicide Planned suicide Attempted suicide Multiple suicide attempts Malesa White 1.00 (referent) 1.00 (referent) 1.00 (referent) 1.00 (referent) 1.00 (referent) American Indian 1.74* (1.06, 2.86) 1.90* (1.10, 3.30) 1.83* (0.99, 3.38) 3.99*** (2.08, 7.65) 4.01** (1.66, 9.68) Asian 1.35 (0.94, 1.93) 1.04 (0.66, 1.64) 1.13 (0.69, 1.84) 1.34 (0.71, 2.55) 1.59 (0.67, 3.77) African American 1.04 (0.88, 1.22) 0.53*** (0.42, 0.67) 0.53*** (0.41, 0.69) 1.09 (0.79, 1.49) 1.14 (0.73, 1.80) Hispanic 1.45*** (1.21, 1.74) 0.81 (0.63, 1.04) 0.86 (0.66, 1.13) 1.20 (0.84, 1.71) 1.53 (0.95, 2.47) Hawaiian 2.17* (1.10, 4.28) 1.90 (0.86, 4.19) 2.06 (0.89, 4.75) 1.18 (0.28, 5.02) 1.29 (0.17, 9.64) Multiple Hispanic 1.59*** (1.28, 1.97) 1.26 (0.97, 1.64) 1.42** (1.07, 1.88) 1.87*** (1.29, 2.70) 2.72*** (1.70, 4.35) Multiple NonHispanic 1.47* (1.02, 2.13) 1.77** (1.18, 2.65) 2.13*** (1.40, 3.24) 1.45 (0.74, 2.83) 0.96 (0.29, 3.13) Femalesa White 1.00 (referent) 1.00 (referent) 1.00 (referent) 1.00 (referent) 1.00 (referent) American Indian 0.78 (0.44, 1.40) 1.73* (0.99, 3.04) 0.98 (0.47, 2.02) 2.09* (1.00, 4.38) 2.25 (0.87, 5.79) Asian 1.09 (0.79, 1.51) 0.84 (0.57, 1.25) 1.01 (0.67, 1.55) 0.85 (0.47, 1.52) 0.79 (0.34, 1.84) African American 1.14* (1.00, 1.29) 0.73*** (0.62, 0.85) 0.84* (0.70, 1.00) 0.96 (0.77, 1.20) 0.77 (0.55, 1.08) Hispanic 1.47*** (1.28, 1.70) 1.09 (0.92, 1.29) 1.25** (1.04, 1.50) 1.74*** (1.40, 2.16) 1.65*** (1.22, 2.23) Hawaiian 1.46 (0.80, 2.67) 1.73 (0.91, 3.31) 2.07* (1.05, 4.08) 2.56* (1.16, 5.66) 2.46 (0.86, 7.07) Multiple Hispanic 1.69*** (1.41, 2.02) 1.30** (1.06, 1.59) 1.26* (1.00, 1.59) 1.91*** (1.47, 2.48) 1.77** (1.23, 2.55) Multiple NonHispanic 1.46** (1.09, 1.96) 1.69*** (1.24, 2.32) 1.93*** (1.39, 2.69) 1.75** (1.15, 2.66) 1.93* (1.12, 3.33) Note. aThough these analyses are presented in a separate table, the variables were included as covariates in the physical activity an d sport analyses. p < 0.05, ** p < 0.01, *** p < 0.001.

PAGE 88

88 Table 4-4. Relative risk of hopelessness and suicidality for a dolescents with various levels of physical activity compared to thos e not reporting any physical activity OR 95% (Confidence Interval) Level of physical activity Felt sad or hopeless Thought about suicide Planned suicide Attempted suicide Multiple suicide attempts Malesa 1 to 2 times/week 1.08 (0.86, 1.35) 0.93 (0.70, 1.23) 1.12 (0.83, 1.50) 1.00 (0.68, 1.48) 0.79 (0.47, 1.33) 3 to 5 times/week 0.91 (0.74, 1.11) 0.79 (0.61, 1.02) 0.76 (0.58, 1.00) 0.75 (0.52, 1.07) 0.63 (0.39, 1.01) 6 to 7 times/week 0.84 (0.68, 1.03) 0.77 (0.60, 1.00) 0.67** (0.50, 0.89) 0.65* (0. 45, 0.95) 0.49** (0.29, 0.81) Femalesa 1 to 2 times/week 1.22** (1.06, 1.42) 1.10 (0.92, 1.31) 1.02 (0.85, 1.24) 1.08 (0.85, 1.37) 0.93 (0.67, 1.29) 3 to 5 times/week 1.04 (0.90, 1.19) 1.01 (0.85, 1.19) 0.84 (0.70, 1.00) 0.94 (0.76, 1.19) 0.88 (0.64, 1.20) 6 to 7 times/week 0.99 (0.84, 1.17) 1.06 (0.88, 1.29) 1.09 (0.88, 1.34) 1.02 (0.78, 1.32) 0.89 (0.61, 1.29) Note.aThough both are included in the same table, the data for males and females represent separate analyses. bOdds ratios were adjusted for age, race/et hnicity, and sport participation.* p < 0.05, ** p < 0.01.

PAGE 89

89 Table 4-5. Relative risk of hopelessness and suicidality for athletes compared to non-athletes OR 95% (Confidence Interval) Felt sad or hopeless Thought about suicide Planned suicide Attempted suicide Multiple suicide attempts Malesa Sport participationb 0.71*** (0.62, 0.81) 0.71*** (0.60, 0.84) 0.76** (0.63, 0.92) 0.70** (0.55, 0.89) 0.85 (0.60, 1.19) Level of sport involvementb Moderately involved (1-2 teams) 0.74*** (0.64, 0.85) 0.73*** (0.61, 0.87) 0.79* (0.65, 0.95) 0.69* (0.53, 0.90) 0.83 (0.57, 1.19) Highly involved (3 + teams) 0.63*** (0.51, 0.77) 0.66** (0.51, 0.86) 0.68** (0.51, 0.91) 0.73 (0.50, 1.06) 0.92 (0.55, 1.53) Femalesa Sport participationb 0.76*** (0.69, 0.85) 0.80*** (0.71, 0.91) 0.79*** (0.68, 0.91) 0.84 (0.71, 1.00) 0.76* (0.59, 0.97) Level of sport involvementb Moderately involved (1-2 teams) 0.79** (0.71, 0.88) 0.81** (0.71, 0.92) 0.77*** (0.67, 0.90) 0.84 (0.70, 1.01) 0.73* (0.56, 0.95) Highly involved (3 + teams) 0.63*** (0.52, 0.77) 0.79* (0.63, 0.99) 0.87 (0.68, 1.11) 0.83 (0.61, 1.14) 0.89 (0.58, 1.36) Note. aThough both are included in the same table, the data for males and females represent separate analyses. bThough both are included in the same table, the data for the sport participation variable and the level of sport involvement variable represent two separate analyses. cOdds ratios were adjusted for age, race/ethnicity, and physical activity. p < 0.05, ** p < 0.01, *** p < 0.001.

PAGE 90

90 CHAPTER 5 EXPLORING POTENTIAL MEDIATI NG P ATHWAYS THROUGH WHICH PARTICIPATION IN SPORTS RELATES TO REDUCED RISK OF SUICIDAL IDEATION Introduction Suicide is the thirteenth lead ing cause of death worldwide [ 148], and the eleventh leading cause of death in the United States [18]. A ccording to the World H ealth Organization (WHO), suicide rates am ong young people ha ve increased so dramatically that now youth represent the group at highest risk in a th ird of countries developed and developing countries [149]. Suicide ranks as the third leading cause of death for American youth 15 to 24 years [1], and the second leading cause of death among U.S. college stude nts [14, 15]. Prevalence data suggest that approximately 1,100 suicides occur annually among U.S. college students aged 18 to 24 [15]. Suicidal ideation and attempts remain even more prevalent [14]. An estimated 1 in 12 college students has made a suicide plan [1], and a pproximately 24,000 suicide attempts occur annually among U.S. college students [15]. In 2007, 9.8% of college students seriously considered suicide, and 1.5% actually attempted suicid e during the previous school year [150]. Healthy Campus 2010, based on public health priorities for the U.S., sets goals to reduce the rate of completed suicide (Objective 18 -1) and the rate of attempte d suicide (Objective 18-2) among college students [135]. Involvement in sports may provide distin ct protection against suicide through multiple mechanisms. Sport participants benefit from the psychological advantages of physical activity including improvements in overall emotional well-being, depressed mood, anxiety and stress, and self-esteem [88, 89, 92, 93, 151]. Sport part icipants also experience greater social integration and support when they become me mbers of an extended social network [12]. Brown [80] reviewed the litera ture regarding physical activity, sport participation, and suicidal behavior. Most research found inverse relationships between sport participation and suicidality (i.e.,

PAGE 91

91 suicidal thoughts or behavior) among adolescents and young adults. Compared to nonparticipants, sport participants report less frequent feelings of hopelessness [3], and less risk of sadness, depression, and suicid al behavior [5-13, 73]. However, this field remains in its infa ncy and, thus, suffers from methodological weaknesses that often accompany emerging areas of study [80]. Many studies incorporated limited assessments and definitions of dependent and independent variables [80]. Most research that investigated sport particip ation and suicidal behavior included only one question to assess suicidal ideation: seriously considered attempting suicide in the pa st 12 months. In fact, only [7] used items from a standardized suicide questionn aire and index. Yet, Reynolds and Mazza [2] caution that suicidal id eation, as a component of suicidal behavior, must be viewed as a multifaceted construct that requires much more than a single item regarding whether or not young people have thought about killing themselves. According to these i nvestigators, research on suicidal ideation or group assignment predicat ed on one or two questions regarding having thoughts of suicide lacks sufficient specificity for defining this domain of suicidality and obscures the relative importance of suicidal ideation as a component of suicide risk in youth [2, p.533]. Thus, to advance the science in this ar ea, future research s hould include efforts to achieve greater precision of measurement and, in turn, c onfidence in study outcomes [80]. In addition, most research addressing sport participation a nd suicide risk has remained atheoretical, and originated from secondary data analysis of existing data sets that contained survey information about a number of health issu es [80]. However, as Brown [80] explains, relying on existing data sets to ev aluate this topic relegates researchers to using variables that may lack the precision necessary to identify associati ons of interest [p.496]. Thus, researchers have not conducted original research focused on investigating the relationship between sport

PAGE 92

92 participation and suicidal behavior, systematically evaluated hypothetical mechanisms to explain the association between sport part icipation and reduced suicidality, or placed research questions into a theoretical framework, especially am ong college students. Research evaluating relationships between sport participation and suicidal behavior mu st account for primary factors that mediate between a studys dependent and independent variables [80]. Finally, Harrison and Narayan [5] were the only investigators to examine relationships among sport, other extracurricular activities, and suicidal behavior. Yet, to identify sport as a distinct protective factor against suicidality, research must dem onstrate that sport participation provides unique benefits compared to involvement in other extracurricular activities. The present investigation furt her explored the effect of s port participation on suicidal ideation, and addressed key met hodological limitations in prev ious research. This study addressed the limitation in measurement of suicid ality by incorporating a standardized 25-item measure of suicidal ideation that demonstrated good reliability. The auth ors also filled a void by designing a study with the purpose to test a model that incor porated potential mechanisms through which collegiate sport pa rticipation relates to reduced risk of suicidal ideation among college students (Figure 5-1). Risk factors fo r suicide include depressi on [34, 128], hopelessness [41, 128], low self-esteem [42, 48], and loneliness [46, 58]. The theoretical framework underlying the model suggests sport participation pr oduces mental health be nefits that protect against these risk factors through both physiologi cal and social support mechanisms. Physical activity enhances self-esteem a nd ameliorates depressive sympto ms [89, 151]. Social support positively impacts depression, hopelessness, and loneliness [58, 152]. A study hypothesis posited that the synergistic effect of these mechanis ms protects sport participants against suicidal ideation. Finally, in addition to sport participation, measures assessed involvement in other

PAGE 93

93 university-sponsored activities. Thus, analyses al so tested the hypothesis that participating in sports provides college students increased protec tion against suicidal ideation more effectively than does participating in other activities. Methods Procedures and Participants The protocol was approved by the university Institutional Review Board. Questionnaires were com pleted as a voluntary, in class activit y. The principal inve stigator surveyed a convenience sample of students enrolled in unde rgraduate lecture courses: four from the Department of Health Education and Behavior three from the Department of Psychology, and two from the Department of Sociology. Each in structor determined wh ether or not his/her students received extra credit for participating in this study. Participants received the survey from the principal investigator in a large catal og envelope. Each partic ipant also received a consent form that stated consent for part icipation was implied upon completion of the anonymous survey. Participants were not requir ed to sign the consent form. Students were instructed to use the informed consent form as a cover sheet, to place their completed questionnaires into the envelope prior to submitting the survey, and to take their consent form with them after completing the survey. Partic ipants were assured non-participation would not adversely affect their course grade. To guarantee anonymity, no names were included on questionnaires. Participants were informed that the study measures may elicit distressing feelings, and the informed consent form included information regarding me ntal health services. Participants required approximately 20 minutes to complete the study measures. The sample included 522 students enrolled in undergraduate courses at a large southeastern university in the United States. After eliminating invalid or incomplete

PAGE 94

94 questionnaires, 450 participants remained in th e study. Students whose data were eliminated from the analyses did not differ significantly from participants who returned valid and complete questionnaires regarding sex, age, or race/ethnicity. Most part icipants were female (n = 333, 74%). Participants ranged in age from 18 to 24 years ( M = 19 years, SD = 1.49). Approximately 61% reported their class standi ng as freshman or sophomore. Participants reported their race/ethnicity as 61% white/non-Hispanic, 14% Hispanic, 13% African American, 8% Asian, and 5% American Indian, Native Hawaiian, or Other (categorized as Other). Measures Involvement in other extracurricular activities Involvem ent in other extracurric ular activities was assessed w ith 1 item created for this research: During the past 12 months, how many [university]-sponsored clubs or organizations (not including sport teams) have you been a me mber of? Among the students who provided information regarding the types of non-sport activities in which they participated, most reported participating in academic or career clubs; or religious, social, or service organizations. Participants were divided into 2 groups: those who participated in other university activities, and those who did not participate. Antecedent variable Sport participation: Sport participation during the prev ious 12 m onths was determined by 2 items developed for this research: Dur ing the past 12 months, on how many varsity or club sports teams run by [the university] did you play? ( extramural sports ); and During the past 12 months, on how many intramural sports teams run by [the unive rsity] did you play? First participants were grouped into 3 categories: those who participated in intramural sports, those who participated in extramural sports, a nd those who did not par ticipate in university sports. Participants were subsequently divided into 2 groups: those who participated in sports,

PAGE 95

95 and those who did not participate. The latter dichotomy corresponded to the assessment of sport participation as measured by the Centers for Disease Control and Prev entions (CDC) National College Health Risk Behavior Survey (NCHRBS) (i.e., involvement in intramural or extramural college sports) [153]. The specific rationale for collapsing the data for intramural and extramural sport participants is provided in the results section. Intervening variables Physical activity: Participants were asked two physic al activity questions drawn from CDCs Youth Risk Behavior Survey [79]: 1) On how many of the past 7 days did you exercise or participate in physical activity for at least 20 minutes th at made you sweat and breathe hard, such as jogging, swimming laps, fast bicycling, fast dancing, or similar aerobic activities? ( vigorous exercise ); and 2) On how many of the past 7 days did you participate in physical activity for at least 30 minutes that did not make you sweat or breathe hard, such as fast walking, slow bicycling, skating, pushing a la wn mower, or mopping floors? ( moderate exercise ). Social support: The 25-item Personal Resource Qu estionnaire85 (PRQ85) Part II measures relational provisions in Weisss [152] definition of social support (attachment, social integration, opportunity for nurturing behavior, reassurance of worth, a sense of reli able alliance, and obtaining information and guidance in st ressful situations) [154]. Respondents rate statements on a 4-point Likert scal e (strongly agree to st rongly disagree). Scores range from 25 to 100, with higher scores indica ting higher perceived social s upport. The scale produced an internally consistent measure of perceived social support ( = .93). The data distribution was also normal (skewness = -.72; kurtosis = .09). Self-esteem: The Rosenberg Self-Esteem Scale (RSES) includes 10 general statements that assess global aspects of self-w orth, confidence, and self-appr oval [155]. Respondents rate statements on a 4-point Likert scal e (strongly agree to st rongly disagree). Scores range from 10

PAGE 96

96 to 40, with higher scores indicating higher se lf-esteem. This measure demonstrated an acceptable level of internal consistency ( = .92), and the skewness (-.63) and kurtosis (-.09) values were in the acceptable range. Depression: The 20-item Center for Epidemiological Studies Depression Scale (CES-D) measures depressive symptomatology [156]. Sc ores range from 0 to 60, with higher scores indicating greater frequency and number of depre ssive symptoms. Internal consistency of the measure in the current study was = .91. The skewness value was 1.17, and the kurtosis value was 1.04. Hopelessness: The 20-item Beck Hopelessness Scale (BHS) measures negative expectations about the future [157] Each item elicits a true/false response. Total scores range from 0 to 20, with higher scores indicating a gr eater degree of hopelessness. In this study, internal consistency for this scale was acceptable ( = .79). However, the skewness (2.40) and kurtosis (7.62) values were sli ghtly out of range. Therefore, as suggested by Tabachnick and Fidell [158] and Howell [159], the Logarithmic (Log 10) transformation method was applied to correct for the substantially positively skewed distribution [NEWX = LG10(X + C), where C = 1.5 so the minimum value was greater than zero (min. = .18)]. The transformation procedure produced acceptable skewness (.61) and kurtosis (.13) values. Loneliness: The Revised UCLA Lone liness Scale (UCLA-LS), a measure of subjective experience of loneliness, consists of 20 items on a 4-point summated rating scale [160]. Scores range from 20 to 80, with higher scores indicatin g higher levels of loneliness. Reliability analysis indicated that this measure demonstrated good internal consistency in this study ( = .93). The skewness (1.08) and kurtosis (.90) values also were acceptable.

PAGE 97

97 Criterion variable Suicidal Ideation: The 25-item Adult Suicidal Ideation Questionnaire (ASIQ) measures adults thoughts about suicide within the past month. Respondents rate each item on a 7-point scale ranging from 0 ( I never had this thought) to 6 (I have this thought almost every day ) [161]. Scale content ranges from wishes of being d ead or never being born, to thoughts of how and when to kill oneself. Summing item values pr oduces a maximum score of 150. Research validated the scale for use with college students [34, 162]. Internal consistency for the scale was high in this sample ( = .97). Yet, the distribution indicated substantial positive skewness (skewness = 3.88; kurtosis = 21.06). Therefore, again, the Logarithmic (Log 10) transformation method was applied to correct for the failure of normality [NEWX = LG10(X + C), where C = 1.5 so the minimum value was greater than zero (min. = .18)]. After the transformation, the skewness and kurtosis values we re .41 and -.81, respectively. Data Analysis Intern al consistency reliability values, skewness values, and kurtosis values were calculated for the study measures. Data transfor mations were then applied as appropriate. Cooks Distance values also were generated for regression analyses perf ormed with the total sample. The mean Cooks Distance value for all tests was less than 1, indi cating the data did not contain significant outliers. Descriptive statistics also were calculated for the overall study sample. In preliminary analyses, the effects of dem ographic variables on the study measures were evaluated. The primary data analysis was performe d in four stages. First, independent samples t-tests and simple correlations tested hypothe sized bivariate associ ations among the study variables. Second, standard path analysis evalua ted the proposed model. Third, as stipulated by Baron and Kenny [163], formal test s for mediating effects of th e intervening variables were

PAGE 98

98 performed. Fourth, independent samples t-tests and regression analyses compared effects of sport participation to involvement in other extracurricular activities on the intervening and criterion variables. Results Sample Characteristics About 23% (n = 103) of participants reported involvement in intramural sports, 6% (n = 29) reported involvement in extramural sports, and 2% (n = 9) reported involvement in both intramural and extramural sports. Even after accounting for inequality of variances on some variables, only one significan t difference emerged between intramural and extramural sport participants regard ing vigorous activity F (3, 446) = 25.57, p < .001. Participants only involved in extramural sports reported hi gher levels of vigorous activity ( M = 4.8, SD = 2.09) than students only involved in intramural sports ( M = 3.4, SD = 1.80). Since the sample size for extramural sport participants was small, and only one significant difference emerged, the groups were combined in subsequent analyses. Thus analyses explore differences between students involved in university-run intram ural, club, or varsity sports (n = 141, 31%), and those not involved in university sports. Fifteen percent of participants (n = 68) repor ted involvement in both university-run sports and other organizations. Slightly more students reported only invol vement in university sports (n = 73, 16%). Most students indicat ed they were either not involv ed in university-run sports or other organizations (n = 153, 34%), or they were only involved in other university organizations (n = 156, 35%). On average, pa rticipants reported engaging in vigorous exercise 2.8 days/week (range = 0-7 days), and moderate exercise 3.5 days/week (range = 0-7 days). Scores for the total sample on the risk measures indicated participants exhibited relatively lo w levels of depression, hopelessness, loneliness, and suicidal ideation (CES-D: M = 12.11, SD = 9.29; BHS: M = 2.04

PAGE 99

99 (original)/2.94 (transformed) SD = 2.55; UCLA-LS: M = 33.63, SD = 9.92; ASIQ: M = 7.41 (original)/5.04 (transformed), SD = 11.97). Mean level of suicidal ideation fell below normative college samples ( M = 10.91, SD = 13.89) [161]. Th ese students also demonstrated high levels of social support ( M = 84.84, SD = 10.38) and self-esteem ( M = 33.37, SD = 5.37). Preliminary Analyses Prior to examining relationships among the pr imary study variables, the effects of three demographic variables, sex, age, and race/ethnicity, were evaluated. A few differences emerged between males and females. Males demonstr ated higher levels of vigorous activity t (448) = 3.95, p < .001 (males: M = 3.4, SD = 2.11; females: M = 2.6, SD = 1.97) and sport participation ( 2 = 22.21, p < .001; males: n = 57, 49%; females: n = 84, 25%). Conversely, females reported higher levels of social support t (448) = 2.04, p < .05 (males: M = 83.16, SD = 11.27; females: M = 85.43, SD = 9.99). No significant interactions emerged regarding sex, and sex was not significant in any of the models that predicted suicidal idea tion. All preliminary findings remained consistently in the same direction for females as findings for males and females combined. Older age related to higher levels of hopelessness ( r = .12, p < .01) and suicidal ideation ( r = .15, p < .01), and to lower levels of vigorous activity ( r = -.10, p < .05). One significant difference emerged among participants reporting different races/ethnicities. Findings indicated a significant main effect for vigorous activity F (4, 445) = 4.00, p < .01. Subsequent post hoc analyses revealed African American females as significantly less likely to engage in vigorous activity than females of all other race/ethnicity categories F (4, 328) = 7.06, p < .001: African American ( M = 1.24, SD = 1.2), white ( M = 2.90, SD = 2.0), Hispanic ( M = 2.50, SD = 2.0), Asian ( M = 2.29, SD = 1.7), and Other ( M = 2.92, SD = 2.2).

PAGE 100

100 Based on preliminary analyses, different covariates were included in each model depending on the dependent variable under invest igation. When the relationship between sport participation and vigorous activity was examined, sex, age, and race/ethnicity were controlled. In models that predicted social support or hopelessness, sex or age, respectively, were controlled. When suicidal ideation represented the out come variable models included age. Bivariate Analyses Bivariate correlations among the intervening an d criterion variables were calculated to determ ine the degree to which they were interr elated, and to establis h a foundation for further analyses. Moderate exercise on ly related to vigorous activity ( r = .34, p < .001). Intercorrelations among the remaining continuou s variables appear in Table 5-1. All the intervening variables related to each other and correlated with suicidal ideation. Thus, the proposed intervening variables were reliably associ ated with the criteri on variable. Table 5-2 presents bivariate associations between sport participation and the intervening and criterion variables. Since the model presen ts two sets of intervening vari ables (i.e., 2 protective factors and 4 risk factors) the relationship between spor t participation and each set of variables were examined separately. Findings demonstrated that the antecedent variable in the model, sport participation, was related to the criterion vari able, and to all the variables, except depression, presumed to intervene in the re lationship between sport participation and suicidal ideation. Confirmation of the Path Analysis Regression analyses evaluated the ef fect of sport participation on suicidal ideation, and explored the extent to which the proposed intervening variables influenced this relationship. In separate equations, each of the in tervening variables and the criter ion variable were regressed on sport participation (Figure 5-2). Next, suicidal ideation represented the outcome variable, and the intervening variables were treated as predictors (Figure 5-3). As Figure 5-2 demonstrates,

PAGE 101

101 sport participation predicted both sets of inte rvening variables and the criterion variable. Furthermore, all the intervening variables represented significant predictors of suicidal ideation (Figure 5-3). Tests of Mediation A prim ary hypothesis of this study posited that physical act ivity and social support would mediate relationships between sport participati on and risk factors associated with suicidal ideation. Therefore, thes e variables were formally evaluated as mediators in the path model for the total sample. Baron and Kenny [34, 163] outline a series of regression analyses to test for mediation. First, regress the mediator on the in dependent variable. Next regress the dependent variable on the independent variable. Finally, regress the dependent variable on both the independent variable and the mediator. The foll owing conditions are required for mediation to occur. First, the independent variable must infl uence the mediator in the first analysis. Second, the independent variable must influence the depende nt variable in the sec ond analysis. Third, the mediator must influence the depe ndent variable in the third anal ysis. For partial mediation, the effect of the independent variable on the depende nt variable decreases in the third analysis compared to the second analysis. For full media tion, the independent variab le has no effect after the mediator is controlled. Analyses revealed that all conditions for establishing mediation were satisfied. Adjusted R square [AR2] values for each mediation model, and regression coefficients [ = Beta] for the mediating variables are presented. When vigorous physical activity was tested as a mediator, the variable fully mediated rela tionships between sport partic ipation and self-esteem (AR2 = .04) F (2, 447) = 10.93, p < .001 ( vig activity = .17, p < .001) and depression (AR2 = .03) F (2, 447) = 7.08, p < .001 ( vig activity = -.14, p < .01). Social support also fully mediated relationships between sport participation and depression (AR2 = .33) F (2, 447) = 111.53, p < .001 ( social support

PAGE 102

102 = -.57, p < .001), hopelessness (AR2 = .29) F (3, 445) = 61.52, p < .001 ( social support = -.52, p < .001), and loneliness (AR2 = .64) F (2, 447) = 398.76, p < .001 ( social support = -.79, p < .001). In separate analyses, self-esteem and depression fu lly mediated the relationship between vigorous activity and suicidal ideation (AR2 = .27) F (3, 445) = 57.09, p < .001 ( self-esteem = -.50, p < .001); (AR2 = .28) F (3, 445) = 59.07, p < .001 ( depression = .51, p < .001), respectively. Depression, hopelessness, and loneliness each partially mediat ed the relationship between social support and suicidal ideation (AR2 = .30) F (3, 445) = 64.76, p < .001 ( depression = .41, p < .001); (AR2 = .23) F (3, 445) = 44.40, p < .001 ( hopelessness = .24, p < .001); (AR2 = .22) F (3, 445) = 42.41, p < .001 ( loneliness = .30, p < .001), respectively. The path model, including demographic characteristics, accounted for a significant portion of the variance in stude nts suicidal ideation (AR2 = .34, p < .001). However, the only variables that remained significant in the final model were age ( = .10, p < .05), self-esteem ( = -.23, p < .001), depression ( = .29, p < .001), and, surprisingly, sport participation ( = -.09, p < .05). When the mediating effects of the intervening variables we re examined separately, sport participation actually mediated the relationship between vigorous activity and suicidal ideation (AR2 = .05) F (3, 445) = 8.15, p < .001 ( sport = -.13, p < .01). Conversely, so cial support, selfesteem, depression, hopelessness, and loneline ss each partially mediated the relationship between sport participation and suicidal ideation (AR2 = .20) F (3, 445) = 37.61, p < .001 ( social support = -.40, p < .001); (AR2 = .28) F (3, 445) = 59.06, p < .001 ( self-esteem = -.50, p < .001); (AR2 = .29) F (3, 445) = 61.52, p < .001 ( depression = .50, p < .001); (AR2 = .18) F (3, 445) = 33.37, p < .001 ( hopelessness = .38, p < .001); (AR2 = .22) F (3, 445) = 41.97, p < .001 ( loneliness = .42, p < .001), respectively.

PAGE 103

103 Sport Participation versus Involvement in Other Extracurricular A ctivities To determine if sport provides distinct psyc hosocial benefits, analyses compared effects of sport participation to involve ment in other activities. Independent samples t-tests showed no significant mean differences on a ny study variable between student s who participated in other extracurricular activities and students who did not participat e in university organizations. Regression analyses confirmed th e unique positive effects of spor t participation on psychosocial variables. Even after controlli ng for participation in other unive rsity activities, e ffects of sport participation on the intervening and criterion variables remained significant. Only one significant interaction occurred be tween sport participation and i nvolvement in other activities. Sport participants reported significantly higher levels of vigorous activity ( M = 4.25, SD = 1.95) than students who did not participat e in any university organizations ( M = 2.25, SD = 1.91), students only involved in othe r university organizations ( M = 2.45, SD = 1.87), and students involved in both university-run sports and other organizations ( M = 3.40, SD = 1.91). Discussion Findings Research in vestigating associations between sport participation and suicide risk has not tested variables hypothesized to mediate the rela tionship. However, expl oratory research using models to examine direct effects and buffering effects of different types of daily activities will help guide development of suicide prevention and intervention programs [43]. Therefore, this study tested a model to explain how sport partic ipation may provide dis tinct protection against suicidal ideation. A primary hypothesis posited that physical activity and social support would mediate relationships between sport participation and signif icant risk factors associ ated with suicidality. Consistent with this hypothesis, vigorous physical activity mediat ed relationships between sport

PAGE 104

104 participation and self-esteem and depression. So cial support mediated relationships between sport participation and depressi on, hopelessness, and loneliness. Furthermore, self-esteem and depression each fully mediated th e relationship between vigorous activity and suicidal ideation; and depression, hopelessness, and loneliness each partially mediat ed the relationship between social support and suicidal idea tion. However, an unexpected findi ng revealed that none of the variables fully mediated the relationship between sport participation a nd suicidal ideation. This study also compared effects of spor t participation to involvement in other extracurricular activities. In ge neral, study participants scored low on measures of risk and high on measures of protective factors. Still, participation in non-s port university activities did not provide students with the same psychosocial bene fits as did sport partic ipation. Furthermore, effects of sport participation on suicidal ideati on and related risk and protective factors proved robust. Even after accounting for students involvem ent in other university activities, the effects of sport participation remained significant. These results support findings from Harrison and Narayan [5] who examined similar relationshi ps among adolescents. These investigators examined relationships among sport participation, involvement in other extracurricular activities, psychological factors, and suicidal behavior. Findings indicated that those involved in team sports at school (alone or co mbined with other activities)were most likely to report high selfesteem and were least likely to report sadness, anxiety, and suicid al behavior [5, p.118]. Contribution to the Literature This investigation contributes to the extant literature in three ways. First, the study repres ents the first effort to test a theoretically based model depicting potential mechanisms through which sport participation relates to reduc ed risk of suicidality. Brown et al. [13] discovered that the relationship between sport pa rticipation and suicidal behavior became nonsignificant among high school students after they controlled for feeling sad or hopeless (1 item)

PAGE 105

105 and other non-psychosocial variable s. According to thes e researchers, this finding suggested that distress/depressed mood may medi ate activity-suicide associations, and reinforced the importance of accounting for mood disturban ce in studies on th is topic [13, p.2254]. The present investigation built on these findings by formally testing depression as a mediator between sport participation and suic idal ideation, and by e xploring two protective factors vigorous activity and social support that account for the relationship between sport participation and depressed mood. This study also tested mediating effects of other risk factors on the relationship between sport participation a nd suicidal ideation, and the influence of sport participation and associated protective factors on these measures of risk. Second, this study incorporated thoroughly validated, highly reliable instruments that provided sufficient specificity to answer definitive questions regarding the association between sport participation and suicidal ideation. To date, Oler et al. [7] were th e only other in vestigators to administer a standardized questionnaire to e xplore the relationship between sport participation and suicidal ideation. Furthermore, no othe r study included several multi-item measures assessing psychosocial vari ables to explore mediating effects of risk and protective factors. Third, to our knowledge, this study represente d only the second inve stigation to compare effects of sport participation to involvement in other activities on suicidal ideation. Our findings support conclusions offered by Harrison and Narayan [5] who explored similar relationships among high school students. Thes e investigators concluded that engaging in any organized activity imparted some benefit, but sport participation afforded unique positive advantages. Progress in suicide research requires both analyzing existi ng data sets and implementing new, theoretically informed studies [28]. Furthe rmore, an important aspect of future research involves developing and detecting multiple pathways to suicidal behavior [35]. Consistent with

PAGE 106

106 previous research [6, 7, 10-13], findings from this study confirmed direct effects of sport participation on suicidality for males and females. This research also accounted for several psychosocial factors responsible for the relationship between spor t participation and suicidal ideation. Therefore, though effect sizes rema ined small, this study provides a foundation for research to examine pathways through which s port participation relate s to reduced risk of suicidal behavior. Limitations Five study lim itations should be noted. First, these data originated from a cross-sectional survey, so analyses could not determine causal re lationships. For instan ce, sport participation may confer protection against su icidal ideation or, conversely, suicidal young people may choose not to participate in sports. Longitudinal and case-control studies can help identify potential cause-and-effect relationships between sport part icipation and risk of suicidal behavior. Second, the sample primarily comprised female college students (74%) enrolled in health education, psychology, and sociology courses, whic h represent a limited cross-section of the student population. Students who choose to take these courses, especially the health education classes, may engage in more health-promoting behaviors that positively impact mental health and reduce the risk of suicidal ideation th an do students enrolled in other courses. Third, these data were obtained through self -report rather than objective measures, and no mechanisms existed to validate the self-reports. Though the survey remained anonymous, some participants may have provided socially desirable answers to study questionnaires. Fourth, the assessment of physical activity may have provided an incomplete picture of participants activity levels. For example, items only asked about physical activity during the previous 7 days. Participants activity during the preceding week may have been atypical for many reasons. Furthermore, some participants ma y have intentionally reported engaging in more

PAGE 107

107 physical activity than they actually performed. Thus, longitudinal data on students activity levels and physical activity assessments based on movement devices would produce greater precision of measurement. Fifth, the questions used to assess involvement in sports and other extracurricular activities were developed specifically for this resear ch. Therefore, no validity data exist for these items. This research attempted to obtain more precise measurements of sport participation and involvement in other activities than did two national surveys of U.S. college students: the NCHRBS [153] and the National Su rvey of Student Engagement (NSSE) [164]. This research could not include an item from the NSSE to a ssess involvement in other activities because the NSSE item groups intercollegiate and intramural s ports among all other co-curricular activities. Also, researchers in this area have consis tently applied a global definition of sport participation (e.g., During the past 12 months, on how many sports teams did you play?) [80]. In this investigation, we sought to improve the assessment of sport participation by including separate items to assess involvement in intramural sports and extramural sports. However, lack of participants involved in extramural sports, information regarding type of sport in which participants were involved, or data regarding participants du ration or frequency of sport involvement, limited the evaluation of potential differential effects across various forms of sport or participants inte nsity of involvement. One signif icant difference emerged between intercollegiate athletes and intr amural sport participants, but considerable differences may exist between these groups. Future Research In the future, investigations should include a larger cross-section of the college student population; incorporate enhanced assessments of sport participation, physical activity, and suicidal behavior (e.g., history of suicide attemp ts); test the proposed model using longitudinal

PAGE 108

108 designs; and validate the model w ith adolescents and with a larger sample of males. The relationship and mediating pathwa ys between sport participation and suicidal ideation (or other measures of suicidal behavior) may differ by se x. If differences exis t, this study may have lacked sufficient power to detect such differences. Furthermore, though findings from this research did not show a signifi cant relationship between moderate physical activity and suicidal ideation, other investigators have found associations between low or moderate levels of activity and reduced risk of suicidal behavior [10, 70] Given public health priorities to encourage moderate-intensity activity, future research should conti nue exploring the mental health benefits afforded by different types and levels of physical activity. Investigators should explore additional me diating pathways as well. For example, increased school connectedness may represent an other intervening variable. Positive school connection protects against suicide [56, 61] Especially among younger students, sport participation may cultivate increased commitment to, involvement in, or identification with school and school values [67]. School sport participation enha nces connectedness by increasing opportunities for students to feel a sense of be longing, attachment, and pa rticipation in their social environment [11]. These feelings may operate as protective factors by buffering stress, enhancing social integrati on and, in turn, decreasing risk-taking behavior [11]. Future research also should examine ways through which sport participation enhances resilience in young people, thus help ing to reduce the risk of suic ide. Measures that enhance resilience help to prevent suicid e [56]. Participation in sports may represent an intervention activity that increases protective factors associated with enhanced resilience [102]. Cohu [102] found that intercollegiate athlet es demonstrated significantly hi gher resilience scores, compared

PAGE 109

109 to non-athletes. However, little empirical resear ch specifically examines the association between sport and resilience. Conclusions The growing body of research investigating the relationship between sport participation and suicidal behavior suggests th at involvement does not cause ha rm, and may actually relate to beneficial psychosocial outcomes. However, this study and related research in the area involved individuals with predominantly good mental health. Future research should explore the association between sport participation and su icidality among young people with a broader range of mood disturbances and mental health problems to determine wh ether or not research with a more distressed population yields similar findings Then, if the inverse relationship remains, practitioners may consider incorporating this line of research into intervention efforts to prevent suicide among young people. Multi-compartmental approaches to prevent suicide may include opportunities for young people to engage in physical activity, as we ll as to become involved in sports. For example, Brown [80] describes th e Sports Challenge Inte rvention program used successfully in Australia and Singapore to improve self-concept and self-esteem, reduce destructive behaviors, and promote mental health among at-risk youth. Researchers and practitioners should not oversimplify or overstate the associations between sport participation and suicide risk, bu t findings from this line of research suggest opportunities for prevention and merit considerati on in public health po licy and planning [12]. Health professionals may use findings from the present study to advocate for increased opportunities to involve young people in sports a nd vigorous activity. Fi ndings would support efforts to assess the impact that the erosi on of funds and social support for school and community sport programs have on young people s mental health. Practitioners working specifically with young females also should consider the finding that African American females

PAGE 110

110 demonstrated significantly lower levels of physical activity, compared to their female counterparts. Investigators suggest that racial disparities in fema les exercise behavior originate during early adolescence [165, 166] and conti nue through adulthood [167, 168]. These findings highlight an intervention opport unity for practitioners to targ et exercise behavior among young African American females. Program planners should identify perceived barriers to exercise among African American females, and tailor prog rams to increase levels of physical activity among this subgroup of young people. Exploring the intricate connections between reduced suicidality a nd sport participation holds promise for improving understanding of unique protective factors. Findings from methodologically sound research can provide a foundation for intervention research in preventing suicide. Eventually, physical activity and sport program interventions that possess potential to enhance positive mood and self-estee m, or to promote positive behaviors that compete with negative or self-destructive beha viors, may [become] adjuncts to acceptable behavioral or pharmacological suic ide treatment strategies [80, p.498].

PAGE 111

111 Figure 5-1. Model presenting proposed mech anisms through which sport participation influences the risk of suicidal ideation + + + + Sport Participation Physical Activit y Social Su pp ort Depression Loneliness Suicidal Ideation Hopelessness Self-Esteem + +

PAGE 112

112 Table 5-1. Intercorrelations among the intervening and criterion variables Variable 1 2 3 4 5 6 7 Protective factors 1. Vigorous physical activity .19*** .20*** -.17*** -.10* -.21*** -.14** 2. Social support .69*** -.58*** -.53*** -.80*** -.42*** Risk factors 3. Self-esteem -.68*** -.56*** -.68*** -.52*** 4. Depression .53*** .67*** .52*** 5. Hopelessness .54*** .40*** 6 Loneliness .45*** Criterion variable 7. Suicidal ideation Note. Correlations performed on transformed data for hopelessness and suicidal ideation. p < 0.05, ** p < 0.01, *** p < 0.001.

PAGE 113

113 Table 5-2. Means and independent samples t-te st values comparing students involved in university-run sports and students not involved in university-run sports Sport participants (n = 141) Non-participants (n = 309) Variable Mean (SD) Mean (SD) t df p Protective factors Vigorous physical activity (days/week) 3.84 (1.97) 2.35 (1.89) 7.64 448 .000 Social support 86.71 (9.95) 83.98 (10.47) 2.60 448 .010 Familywise error rate controlled at .025 (.05/2 dependent variables = .025). Risk factors Self-esteem 34.50 (5.22) 32.86 (5.37) 3.03 448 .003 Depression 10.57 (9.12) 12.82 (9.30) -2.39 448 .017 Hopelessness 1.60 (original)/2.65 (transformed) (2.11) 2.24 (original)/3.08 (transformed) (2.71) -2.70 304 .007 Loneliness 31.55 (8.66) 34.58 (10.32) -3.24 319 .001 Familywise error rate controlled at .013 (.05/4 dependent variables = .013). Criterion Variable Suicidal ideation 5.33 (original)/3.95 (transformed) (11.24) 8.36 (original)/5.63 (transformed) (12.19) -3.54 298 .000 Note. Statistical tests were run on transformed data for hopelessness and suicidal ideation. However, the unconverted (original) standard deviations are presented for these variables. T-test statistics for equal variances not assumed are presented for hopeless ness, loneliness, and suicidal ideation because Levenes Test for Equality of Variances was significant.

PAGE 114

114 Note. p < 0.05, ** p < .01, *** p < 0.001 a = sex controlled; b = age controlle d; c = race/ethnicity controlled Figure 5-2. Standardized regres sion coefficients demonstrating the direct effects of sport participation on the intervening and criterion variables Sport Participation Vigorous Activity Social Support Depression Loneliness Suicidal I de ati o n Hopelessness Self-Esteem .29***a b c .15**a .14** -.11* -.12* b -.14** -.16** b

PAGE 115

115 Note. p < 0.05, ** p < 0.01, *** p < 0.001 a = sex controlled; b = age controlle d; c = race/ethnicity controlled Figure 5-3. Path model with stan dardized regression coefficients Sport Participation Vigorous Activity Social Support Depression Loneliness Suicidal Ideation Hopelessness Self-Esteem .29***abc .15**a .20*** -.17*** -.58*** -.53***b -.51***b -.80*** .39***b .44***b .51***b -.41***b -.13**b

PAGE 116

116 CHAPTER 6 EXPLORING PROTECTIVE RELATIONSHIPS BET WEEN PARTICIPATION IN SPORTS AND RISK FACTORS ASSOCIATED WITH ADOLESCENT SUICIDE Introduction Research supports positive associations between participation in cer tain extracurricular activ ities and favorable academic, psychologi cal, and behavioral outcomes [82, 169, 170]. Extracurricular activities facilitate adolescents developmental need for social relatedness by expanding peer networks and facil itating peer bonds [85]. Involvement in school programs links adolescents to a set of similar peers, provi des shared experiences and goals, reinforces friendships between peers, and may strengthen re lationships between indi viduals and family [82, 84]. Compared to other activities, sports may re present a particularly valuable source of learning outside the classroom [86] Athletes may experience greater social support when they become integrated into a network that includes te ammates, coaches, health professionals, family, and community [12]. Integration into protective social networks may buffer adolescents against effects of life stress [12]. Sport participation also may enhance school identification, involvement, and commitment [74]. This connec tion to school positively affects more narrowly defined academic outcomes, as well as nonacademic outcomes such as self-concept [67]. Despite research supporting the positive role of sports for adolescents, investigators only recently explored associations between sport partic ipation and suicide risk. Suicide is the third leading cause of death for American youth aged 15 to 24, and suicides account for 12.9% of all deaths among persons in this age range [1]. Research regarding sport participation and suicidality (suicidal thoughts or behavior) re gularly found inverse relationships [5-13, 73]. However, this field suffers from methodologi cal weaknesses that often accompany emerging areas of study [80].

PAGE 117

117 Most research in this ar ea included only one question to assess suicidal ideation: seriously considered attempting suicide in the past 12 months. Researchers also regularly applied a global definition of s port participation (e.g., During the past 12 months, on how many sports teams did you play?) [80]. In addition, most research addr essing sport participation and suicide risk has remained atheoretical, and originated from secondary data analysis of existing data sets that contained survey information about a number of health issues [80]. Finally, few researchers compared effects of sport participation to involve ment in other extracurricular activities on suicide risk. In one recent study, the authors addressed key methodological limitations in previous research [171]. The study addressed limitations in measurement of suicidality by incorporating a standardized 25-item suicidal ideation measure. The authors also designed a study to explore mechanisms through which collegia te sport participation (varsit y, club, or intramural sports) related to reduced risk of suicidal ideation amon g college students. Finally, analyses compared effects of sport participation to invol vement in other college organizations. Participants included unde rgraduate students aged 18 to 24. Regression analyses confirmed a path model and tested for medi ation effects. Vigorous activity mediated relationships between sport participation and self-esteem and depression; and self-esteem and depression mediated the relationship between vi gorous activity and suicidal ideation. Social support mediated relationships between sport pa rticipation and depression, hopelessness, and loneliness. Each of these risk factors partiall y mediated the relationship between social support and suicidal ideation. However, no variable fully mediated the relationship between sport participation and suicidal ideati on. Notably, participation in non-sport university activities did not provide students with the same psychosoc ial benefits as did sport participation.

PAGE 118

118 The present investigation built on previous research by examining effects of sport participation during high school. Improved assessme nt of sport participation enabled exploration of differential effects across participants involv ement in intramural, extramural, and community sports. This study included enhanced assessm ents of involvement in non-sport school and community extracurricular activities as well. Therefore, analyses te sted the hypothesis that participating in sports provides young people incr eased protection against suicidal ideation more effectively than does participating in other extracurricular activities. This study also tested an enhanced model that incorporated potential mechanisms through which high school sport participatio n relates to reduced risk of su icidal ideation (Figure 6-1). Risk factors for suicide include depression [34, 128], hopelessness [41, 128], low self-esteem [42, 48], and loneliness [46, 58]. The theoretical framework underlying the model suggests sport participation produces mental h ealth benefits that protect against these risk factors through physiological, social support, and school connect edness mechanisms. Physical activity enhances self-esteem and ameliorates depressive symp toms [89, 151]. Social support and school connectedness positively impact depression, hopelessness, and loneliness [58, 61, 152]. A study hypothesis posited that the synergistic effect of these mechanisms protects sport participants against suicidal ideation. Methods Procedures and Participants The university Institutional Re view Board approved the pr otocol. Students completed questionnaires as a voluntary, in class activit y. The principal inve stigator surveyed a convenience sample of students in college cl asses: 7 from Enh ancing the Freshman Experience program, and 10 from the College of Health and Human Performance. Each instructor determined whether hi s/her students received extra credit for participating in this

PAGE 119

119 study. Participants received the survey in a large catalog envelope, a nd they placed their completed questionnaires into the envelope before submitting the survey. Participants received a consent form that implied consent by completing the survey. To guarantee anonymity, no names were included on questionnaires. Participants we re informed that the study measures may elicit distressing feelings, and they were provided in formation regarding mental health services. Participants required approximately 20 minutes to complete the study measures. The sample included 1029 students enrolled in courses at a large southeastern university in the U.S. Data for the present study was glean ed from 282 freshman students included in the total sample. After eliminating invalid freshman questionnaires, 269 participants remained in the study. Students whose data were eliminated did not differ signi ficantly from participants who returned valid questionnaires rega rding sex, age, or race/ethnic ity. Most participants were female (n = 197, 73.2%). Partic ipants were aged 18 or 19, ( M = 18.9 years, SD = 0.40), but most were aged 18 (n = 217, 80.7%). Participants reported their race/ethni city as 54.1% white/nonHispanic; 17.5% Hispanic; 16.4% African Ameri can; 9.0% Asian; and 3.0% American Indian, Native Hawaiian, some other race, or multi-racial (Other). Measures Involvement in other extracurricular activities Involvem ent in other activities during high school was assessed through items from the National Education Longitudinal Study (NELS) [172]. Participants were provided 2 lists that excluded sport activiti es: school-sponsored activities and community activ ities (see Table 6-1). For each activity, participants indicated whether: 1) they participated, 2) they did not participate, or 3) their high school or community did not have the activity. Scores for each activity on the respective lists were summed to indicate the total number of school and community activities (2 separate scores) in which student s participated. Activity groups then were created based on level

PAGE 120

120 of involvement in school and commu nity activities. In addition, a total extracurricular activity participation (TEAP) score was created by summing the total number of sports, school activities, and community activities in which students participated. Again, activity groups were created based on involvement in extracurricular activities. School and community activities were subs equently categorized to focus on actual content or domain of activity. School activities were clustered into 6 categories: performance activities (music group and/or drama); school involvement (student governme nt and/or yearbook or newspaper); academic activities (honor societies and/or academic clubs); prosocial activities (religious groups and/ or service clubs); vocational activities (vocational education or professional clubs); and hobby clubs (photography, chess, etc.). Community activities also were clustered into 6 categories: outdoor activities (scouting and/or 4-H); religious groups ; hobby clubs; neighborhood clubs (neighborhood programs, Boys/Girls Clubs, YMCA); summer programs ; and other One NELS item estimated time participants spent in school-sponsor ed extracurricular activities: During the past 12 months, how mu ch total time did you spen d on all extracurricular activities sponsored by your high sc hool in a typical week? Respons e options: none, < 1 hr/wk, 1-4 hrs/wk, 5-9 hrs/w k, 10-19 hrs/wk, and > 20 hrs/wk. Involvement in other extracurricular activities in college was assessed with 1 item: During the past 12 months, were you a member of any other college organizations (not including sport teams)? Antecedent variable Sport participation: Sport participation during high school was assessed through item s from NELS [172]. Participants were provided a li st of sports (see Table 6-2). For each activity, participants marked all that applie d: 1) their high sc hool did not have the sport, 2) they did not participate, 3) they participated in intramural sp orts, 4) they participated on a junior varsity or

PAGE 121

121 freshman team, 6) they participated on a vars ity team, and/or 6) they participated on a community team. Response option participated as a captain/co-captain included in NELS was changed to participated on a community team. Total scores were generated for sport pa rticipation, intramural sports, extramural (interscholastic) sports, and community sports. Participants then were divided into 2 groups for each variable: those who participated, and those w ho did not participate. Finally, for total sport participation and extramural sport participation, students were grouped based on involvement: not involved, moderately involved (1-2 teams), and highly involved (> 3 teams). An insufficient number of students involved in intramural and community sports precluded examination based on level of involvement. Sport participation during college was assess ed with 1 item: During the past 12 months, were you a member of any college intr amural, club, or varsity sport teams? Intervening variables Physical activity: Participants responded to 2 sl ightly m odified physical activity questions drawn from CDCs YRBS [79]: 1) During your last year in high school, on how many days during a typical week did you participate in physical ac tivity for at least 30 minutes that did not make you sweat or breathe hard, su ch as fast walking, slow bicycling, skating, pushing a lawn mower, or mopping floors? ( moderate activity); and 2) During your last year in high school, on how many days during a typical week did you exercise or participate in physical activity for at least 20 minutes that made you sw eat and breathe hard, such as jogging, swimming laps, fast bicycling, fa st dancing, or similar aerobic activities? ( vigorous activity ). Social support: The 25-item Personal Resource Ques tionnaire85 Part II measures relational provisions in Weisss [152] definition of social support (attachment, social integration, opportunity for nurturing behavior, reassurance of worth, a sense of re liable alliance, and

PAGE 122

122 obtaining information and guidance in stressful situati ons) [154]. Respondents rate statements on a 4-point Likert scale (strongly agree to strongly disagree). Scores range from 25 to 100, with higher scores indicating higher pe rceived social support. The scale produced an internally consistent measure of perceived social support ( = .92). Data distribution was normal (skewness = -.74; kurtosis = .22). School connectedness: School connectedness was construc ted from responses to 6 items from the National Longitudinal Study of Adolescent H ealth [173] phrased in the past tense: 1) I felt close to people at my high school, 2) I felt like I was part of my high school, 3) I was happy to be at my high school, 4) The teachers at my high school treated students fairly, 5) I felt socially accepted at my high school, and 6) I felt safe at my high school. Response options used a 4-point Likert sc ale (strongly agree to strongly di sagree). Responses to the 6 statements were summed; scores range from 6 to 24. The measure showed good internal consistency ( = .88). The skewness (-.87) and kurt osis (.61) values were acceptable. Self-esteem: The Rosenberg Self-Esteem Scale includes 10 general statements that assess global aspects of self-worth, conf idence, and self-approval [155]. Respondents rate statements on a 4-point Likert scale (strongly agree to strongly disagree). Scores range from 10 to 40, with higher scores indicating higher self-esteem. The measure demonstrated good internal consistency ( = .91), and the skewness (-.53) and kurtosi s (-.63) values were in the acceptable range. Depression: The 20-item Center for Epidemiological Studies Depression Scale (CES-D) measures depressive symptomatology [156]. Sc ores range from 0 to 60, with higher scores indicating greater frequency and number of depre ssive symptoms. Internal consistency of the

PAGE 123

123 measure in the current study was = .85. The skewness value was 1.10, and the kurtosis value was .99. Hopelessness: The 20-item Beck Hopelessness Scale (BHS) measures negative expectations about the future [157] Items elicit true/false respons es. Total scores range from 0 to 20, with higher scores indica ting a greater degree of hopelessne ss. In this study, internal consistency for this scale was fair ( = .68), and the skewness (1.64) and kurtosis (2.49) values were within the acceptable range. Loneliness: The Revised UCLA Lone liness Scale (UCLA-LS), a measure of subjective experience of loneliness, consists of 20 items on a 4-point summated rating scale [160]. Scores range from 20 to 80, with higher scores indicatin g higher levels of loneliness. Reliability analysis indicated that this measure demonstrated good internal consistency in this study ( =.90). Skewness (.98) and kurtosi s (.65) values were acceptable. Criterion variable Suicidal ideation: The 25-item Adult Suicidal Ideation Questionnaire (ASIQ) measures adults thoughts about suicide within the past month. Respondents rate each item on a 7-point scale ranging from 0 ( I never had this thought) to 6 (I have this thought almost every day ) [161]. Scale content ranges from wishes of being d ead or never being born, to thoughts of how and when to kill oneself. Summing item values produces a maximum score of 150. Research validated the scale for use with college students [162]. Internal consistency was = .97. Distribution indicated substantial positive skewness (skewness = 4.22; kurtosis = 34.74). Therefore, as suggested by Tabachnick and Fide ll [158] and Howell [159] the Logarithmic (Log 10) transformation method corrected for the subs tantially positively skewed distribution [NEWX = LG10(X + C), where C = 1.5 so the minimum valu e was greater than zero (min. = .18)]. The procedure produced acceptable skewne ss (.99) and kurtosis (-.18) values.

PAGE 124

124 Data Analysis Intern al consistency reliability, skewness, and kurtosis values were calculated, then data transformations were applied as appropriate. Descriptive statistics were calculated, and preliminary analyses determined effects of demographic variables on the study measures. Regression analyses evaluated eff ects of types of sport particip ation on the outcome variables. Standard path analysis evaluated the proposed m odel, and formal tests for mediating effects of the intervening variables were pe rformed. Finally, regres sion analyses compared effects of sport participation to involvement in othe r activities on the outcome variables. Results Sample Characteristics Tables 6-1 and 6-2 describe patterns of stude nts involvement in sports and other activities during high school. Relatively few students participated in intramural sports (n = 50, 18.6%; M = 0.34, SD = 0.84; range: 0-5 sports) or community sports (n = 89, 33.1%; M = 0.42, SD = 0.69; range: 0-5 sports). C onversely, most students participat ed in extramural sports (n = 193; 71.7%; M = 1.33, SD = 1.22; range: 0-5 sports). Overall, 79.9% (n = 215) of students participated on some type of sports team ( M = 2.09, SD = 1.73; range: 0-8 sports). Almost all students participated in n on-sport school activ ities (n = 266, 98.9%; M = 4.07, SD = 1.74; range: 0-9 activities). In fact, many students partic ipated in 5 or more school activities (n = 113, 42.0%). Most students were involved in academic (n = 253, 94.1%) and/or prosocial activities (n = 208, 77.3%). Many students participated in non-sport community activities (n = 212, 78.8%; M = 1.74, SD = 1.41; range: 0-7 activities). Over ha lf participated in 1 to 2 community activities (n = 146, 54%), pa rticularly religious groups (n = 110, 40.9%). All participants were involved in at leas t one extracurricular activity during high school ( M = 7.90, SD = 3.38; range: 1-19 activities). Almo st half (n = 122, 45.4%) reported spending

PAGE 125

125 10 or more hours each week in school-sponsored extracurricular activities (range: 0 to > 20 hrs/wk). On average, during high school, participants reported engagi ng in vigorous activity 4.29 days/wk (SD = 2.08; range: 0-7 days) and moderate activity 3.76 days/wk (SD = 2.24; range: 0-7 days). Just over thirty percent of students (n = 90) participated in college sports, and over half were involved in other un iversity organizations (n = 141, 52.6%). Scores on the risk measures indicated partic ipants exhibited relati vely low levels of depression, hopelessness, loneliness and suicidal ideation (CES-D: M = 10.79, SD = 7.20; BHS: M = 1.55, SD = 1.90; UCLA-LS: M = 31.74, SD = 8.57; ASIQ: M = 4.87 (original)/3.34 (transformed), SD = 10.10). Mean level of su icidal ideation fell below normative college samples ( M = 10.91, SD = 13.89) [161]. These students also demonstrated high levels of social support ( M = 87.17, SD = 9.31), self-esteem ( M = 34.10, SD = 4.95), and school connectedness ( M = 20.13, SD = 3.53). Preliminary Analyses Prior to examining relationships among the primary study variable s, effects of two demographic variables, sex and race/ethnicity, were evaluated. Several differences emerged. For example, males were more likely than females to participate in intram ural sports during high school ( 2 = 7.27, p < .01). Yet, females were more likely to participate in school involvement activities than males ( 2 = 15.94, p < .001). Some interactions emerged regarding sex. Significant race/ethnicity differences appeared, es pecially among females. For instance, Asian females were less likely to partic ipate in extramural sports ( 2 = 24.72, p < .001), and females categorized as other reported higher levels of loneliness F (4, 178) = 3.21, p < .05. Thus, sex and race/ethnicity were controlled in all models. College sport pa rticipation and involvement in other college organizations also we re controlled in models that examined effects of participation in sports or other activ ities during high school.

PAGE 126

126 Confirmation of the Path Analysis Table 6-3 presents results from regression an alyses evaluating effect s of high school sport participation on variables in the path model (Fig ure 6-1). Findings showed few statistically significant relationships between participation in intramural or community sports and psychosocial variables. However, extramural sp ort participation related to higher levels of physical activity, social support, school connectedness, and self-e steem, and to lower levels of depression, hopelessness, loneline ss, and suicidal ideation. Ex tramural sport participation (dummy coded yes/no) was tested as the antecedent variable in the model. Regression analyses confirmed engagement in vigorous activity as the most appropriate physical activity variable to include in the path model moderate activity was not related to any psyc hosocial variables. Thus, in separate equations, each intervening variable (vigorous activity, social support, school connectedness, self-esteem, depression, hopelessness, and l oneliness) and the criterion variable (suicidal ideatio n) were regressed on extramural spor t participation (Table 6-3). Next, suicidal ideation represented the outcome variable, and the interven ing variables were treated as predictors (Figure 6-2). The path model, incl uding demographic character istics and participation in sports or other organizations during college, accounted for a significant portion of the variance in students suicidal ideation (adjusted R square [AR2] = .33, p < .001). As Table 6-3 demonstrates, extramural sport participation pr edicted all the intervening variables and the criterion variable. Furthermore, all the interven ing variables represented significant predictors of suicidal ideation (Figure 6-2). Tests of Mediation A prim ary hypothesis posited that physic al activity, social support, and school connectedness would mediate relationships betw een sport participation and risk factors associated with suicidal ideation. However, vigorous activity was not associated with self-

PAGE 127

127 esteem or depression. Analyses could not eval uate vigorous activity as a mediator between extramural sport participation and these variables. Still, using a series of regression analyses outlined by Baron and Kenny [163], all the protec tive factors were formally evaluated as mediators between extramural spor t participation and suicidal id eation. Yet, only social support and school connectedness were evaluated as medi ators between extramural sport participation and certain risk factors associ ated with suicidal ideation. When vigorous physical activity was tested as a mediator between extramural sport participation and suicidal ideation, the variable fully me diated the relationship (vig activity = -.16, p < .05), but the overall mode l was not significant (AR2 = .03) F (9, 253) = 1.74, p = .08. Social support partially mediated relationships between extramural sport participation, and depression (AR2 =.24) F (9, 231) = 9.37, p < .001 and hopelessness (AR2 = .24) F (10, 234) = 8.83, p < .001. Yet, social support fully mediat ed relationships between extram ural sport participation, and loneliness (AR2 = .63) F (9, 239) = 45.30, p < .001 and suicidal ideation (AR2 = .17) F (9, 240) = 6.75, p < .001. Similarly, school connectedness part ially mediated the relationship between extramural sport particip ation and hopelessness (AR2 = .18) F (10, 239) = 6.33, p < .001, but fully mediated relationships between extramural sport participation and depression (AR2 =.16) F (9, 234) = 6.24, p < .001, loneliness (AR2 = .25) F (9, 233) = 10.18, p < .001, and suicidal ideation (AR2 = .09) F (9, 245) = 3.66, p < .001. In separate analyses, loneliness fully medi ated the relationship between social support and suicidal ideation (AR2 = .27) F (7, 230) = 13.70, p < .001, but depression and hopelessness each partially mediated this relationship (AR2 = .27) F (7, 232) = 13.65, p < .001 and (AR2 = .24) F (7, 236) = 12.19, p < .001, respectively. Similarly, depressi on and hopelessness each partially mediated the relationship between school connectedness and suic idal ideation (AR2 = .26) F (7,

PAGE 128

128 236) = 12.92, p < .001 and (AR2 = .22) F (7, 242) = 10.92, p < .001, respectively. However, loneliness fully mediated this relationship (AR2 = .28) F (7, 235) = 14.09, p < .001. Vigorous activity was not related to self -esteem or depression, so analys es could not evaluate these variables as mediators between vigo rous activity and suicidal ideation. Sport Participation versus Involvement in Other Extracurricular A ctivities To determine if sports provide distinct psyc hosocial benefits, analyses compared effects of sport participation to invol vement in other activities. Ta ble 6-4 presents results from regression analyses that evalua ted effects of involvement in non-sport school and community activities on the outcome variables. Few significant relationships emerged between participation in other extracurricular activiti es and psychosocial measures. Findings remained consistent regardless of students level of involvement in school or community activities (results not shown), or type of activity in which students participated. However, sport participation, particularly extramural sports, related to positive psychosocial outcomes. Beneficial effects of extramural sport participation remained am ong students who reported high involvement (> 3 teams) (Table 6-3). Even after controlling fo r involvement in other activities, effects of extramural sport participati on (yes/no) on the outcome vari ables remained significant. Number of extracurricular activities did not re late to most psychosocial variables. Yet, time spent in all school-sponsored activities during participants last year in high school did relate to the outcome variables (Table 6-5). Specifically, participants who spent 10 or more hours per week during the previous 12 months in extracurricula r activities sponsored by their high schools reported higher levels of physical act ivity, social support, self-esteem, and school connectedness, and lower levels of depression, hopelessness, loneliness, and suicidal ideation. Furthermore, a greater percentage of students wh o participated in extramural sports reported spending 10 or more hours per week in extracurricular activities ( 2 = 54.96, p < .001; n = 111,

PAGE 129

129 57.5%), compared to students who pa rticipated in school activities ( 2 = 12.36, p < .01; n = 122, 45.9%) or in commun ity activities (2 = 8.34, p < .05; n = 103, 48.6%). Discussion Contrary to the authors previous investigation, this study did not show significant relationships between vigorous activity and se lf-esteem or depression. Vigorous activity mediated the relationship between extramural spor t participation and suicidal ideation, yet when vigorous activity was tested as a mediator in the initial model, the variable became nonsignificant. Sport participation actually mediated the relationsh ip between vigorous activity and suicidal ideation. Vigorous activity related to behavior perf ormed in high school. Conversely, in the earlier study, vigorous activity related to be havior performed during the previous 7 days. Some participants may have reported engaging in more vigorous activity during high school than they actually performed. Both studies documented that social s upport either partially or fully mediated relationships between sport part icipation and depression, hopelessn ess, loneliness, and suicidal ideation. School connectedness, measured in this study, mediated (partially or fully) relationships between sport participation and these outcome variab les. Depression, hopelessness, and loneliness each mediated (fully or partially) relationships between social support and school connectedness, and suicidal ideation. Participation in non-sport extracurricular ac tivities did not yield the same psychosocial benefits as sport participation, particularly extramural sports After accounting for students involvement in other activities, effects of extram ural sport participati on remained significant. These results support findings from research th at examined similar relationships among high school students. Harrison and Na rayan [5] found that those invol ved in team sports at school (alone or combined with other activities)were most likely to report high self-esteem and were

PAGE 130

130 least likely to report sadness, a nxiety, and suicidal behavior [p.118] Fredricks and Eccles [170] found that involvement in school clubs did not rela te to either self-esteem or depression, but sport participation predicted positive psychological adjustment. This investigation also addressed key limitati ons in previous resear ch. First, the study tested an enhanced model depicting potenti al mechanisms through which high school sport participation related to reduced ri sk of suicidal ideation. Consiste nt with previous research [6, 7, 10-13], findings confirmed direct effects of spor t participation on suicid ality. Taken together, findings from the authors tests of a theoretical model suggest that vigorous activity, social support, and school connectedness at least partia lly account for the relationship between sport participation and suicidal ideation. This study adds to the foundation of research examining pathways through which sport participation relates to reduced ri sk of suicidality. Second, this study incorporated validated, highly reliable inst ruments with specificity to answer questions regarding the association betw een sport participation and suicidal ideation. This research represented only the third study to include a sta ndardized measure of suicidal ideation to examine effects of sport participation [7, 171]. Meas uring variables more precisely may produce greater confidence in study outcomes [80]. Third, this study assessed differential effect s across various types of sports and other activities on suicidal ideation and related risk and protective factors. No related research included such detailed assessments of sport participation or involvement in other activities. This study determined that sport participation, part icularly extramural sports, affords students enhanced psychosocial benefits, compared to in volvement in other activities. Furthermore, greater time spent in school-spons ored activities during a typical w eek students last year in high school yielded stronger results than total numbe r of activities throughout high school. Fredricks

PAGE 131

131 and Eccles [170] found that breadt h of participation in extracurricular activities positively related to concurrent indicators of psyc hological adjustment. However, findings from this study suggest that time spent in school-sponsor ed activities each week produces stronger effects than number of activity contexts. Four study limitations warrant recognition. First, the cross-sectional survey could not determine causal relationships. Second, data we re obtained through self-report. Participants remained anonymous, but some may have provided so cially desirable answers. Third, some data relied on recall. Participants graduated less than 12 months before completing the survey. However, some participants may have exaggera ted or forgotten involvement in activities. Fourth, the sample included college students wh o may differ from those not attending college regarding emotional health or activity involvement. Investigators should test the proposed mode l using longitudinal designs, and validate the model with high school students and with larger samples of males. The relationship and mediating pathways between sport participation and suicidal ideation may differ by sex, and this study may have lacked sufficient power to detect di fferences. Investigators also should continue exploring relationships between involvement in different activities and psychosocial variables. Limited research examines effects of involvement in extracurricular activ ities on mental health outcomes, compared to effects on health risk behaviors or academic outcomes. In addition, future research should explore the association between sport participation and suicidality among young people with a broader range of mood dist urbances and mental health problems to determine whether or not research with a more distressed population yields similar findings. Conclusions Educators and policymakers must decide whether or not to allocate limited funds to sport programs. This paper provides critical informati on to help inform these decisions. Researchers

PAGE 132

132 and practitioners should not overs implify or overstate associati ons between sport participation and suicide risk, but findings from this line of research suggest opportunities for prevention and merit consideration in public health policy a nd planning [12]. Despite benefits of sport participation, current state and district budget co nstraints likely result in decreasing opportunities for some students to participat e in school-sponsored sports [174176]. Thus, school officials and public health planners can use thes e findings as a tool to evaluate health costs associated with eliminating sport activities, especially as progr ams face cuts during econo mic difficulties [177]. Practitioners also may incorporate these findings into intervention efforts to prevent suicide among young people. Multi-compartmen tal approaches to suicide prevention may include opportunities for young people to engage in physical activity, as well as to become involved in organized spor ts. Eventually, physical activity a nd sport program interventions that possess potential to enhance positiv e mood and self-esteem, or to promote positive behaviors that compete with negative or self-destructive beha viors, may [become] adjuncts to acceptable behavioral or pharmacological suicid e treatment strategies [80, p.498].

PAGE 133

133 Table 6-1. Percentage of student s who participated in different school and community activities during high school Activity Males Females High school-sponsored activities Performance Band, choir, orchestra, or other music group 16.7 23.9 School play or musical, or drama club 15.3 17.3 School involvement Student government 19.4 40.6 School yearbook, newspaper, or other literary magazine 13.9 27.9 Academic National Honor Society or other academic honor society 84.7 87.8 Academic clubs (Art, Computer, Debate, Foreign languages, Science, Psychology, Math, Philosophy, etc.) 59.7 70.1 Procoscial Service clubs (AFS, Key club, etc.) 58.3 72.6 Religious organization or group 29.2 37.0 Vocational Vocational educatio n or professional clubs 25.0 24.9 Hobby Hobby clubs (photography, chess, etc.) 45.8 22.3 Community activities Outdoor Scouting 8.3 2.5 4-H 2.8 2.5 Religious Religious youth groups 37.5 42.1 Hobby Hobby clubs 33.3 22.3 Neighborhood club Neighborhood clubs or programs 20.8 12.7 Boys clubs or Girls clubs 5.6 1.5 YMCA or other youth groups 8.3 14.2 Summer program Summer programs such as workshops or institutes in science, drama, language, etc. 29.2 36.5 Other Other 33.3 37.1

PAGE 134

134 Table 6-2. Percentage of stude nts who participated in differe nt sport activities during high school Males Females Activity Intramural Extramural Activity Intramural Extramural Activity Baseball/softball 9.7 15.3 9.7 0.0 9.6 3.0 Football 12.5 19.4 2.8 3.0 3.6 1.0 Basketball 15.3 11.1 8.3 1.5 11.7 3.6 Soccer 4.2 12.5 8.3 4.1 18.3 10.2 Swim team 1.4 12.5 4.2 0.5 11.7 4.6 Cheerleading 0.0 0.0 1.4 1.5 13.7 3.0 Dance/drill team 0.0 1.4 0.0 2.5 7.6 6.1 Other team sport (hockey, volleyball, etc) 13.9 23.6 5.6 6.1 23.9 3.6 Other individual sport (cross-country, track, golf, gymnastics, wrestling, tennis, etc) 5.6 40.3 6.9 5.1 31.5 5.1 Note. Intramural sports means the sport teams competed within students own schools (not including PE classes). Extramural sports means the sport teams co mpeted with other school teams (freshman, junior varsity, and/or varsity sports). Community sports means the sport teams were run by community organizations.

PAGE 135

135 Table 6-3. Standardized regres sion coefficients for physical ac tivity and psychosocial variables predicted from participation in sports during high school Activity High school moderate activity High school vigorous activity Social support School connectedness Selfesteem Depression Hopelessness Loneliness Suicidal ideation Sport participation (yes/no) .12 .52/.20* 18** .22** .17* -.19** -.19** -.16* -.14* Level of involvement Moderately involved (1-2 teams) .05 .58/.28 ** .21* .21* .20* -. 25** -.22* -.18* -.20* Highly involved (> 3 teams) .27** .72*** .24** .35*** 23* -.23* -.25** -.22* -.13 Extramural sports (yes/no) .14* .47*** .19** .27*** .14* -.22** -.17/-.46* -.14* -.16* Level of involvement Moderately involved (1-2 teams) .15 .47*** 21** .27** .13 -.23** -.20* -.14 -.17* Highly involved (> 3 teams) .16* .51*** .16* .31*** .19* -. 20** -.11/-.29* -.09/-.18* -.15* Intramural sports (yes/no) .10 .10 .02 03 -.01 -.07 -.07 .01 .04 Community sports (yes/no) .17** .19** .09 .10 .07 .04 -.09 -.08 -.02 Note. Though both are included in the same table, the data for the s port participation variables and the level of sport involvement variables represent separate analyses. Also, data for each sport activity represents a separate analysis. Two coefficients indi cate female and male relationships, respectivel y, that resulted from a significant interaction effect. The p-value presented for the se models represents the significance level for the interaction. Single co efficients indicate the sex interaction was not significant in the model. Sex, race/ethnicity, college sport participa tion, and involvement in other college organizations were controlled in all models. p < .05, ** p < .01, *** p < .001.

PAGE 136

136 Table 6-4. Standardized regres sion coefficients for physical ac tivity and psychosocial variables predicted from participation in different school and community activities during high school Activity High school moderate activity High school vigorous activity Social support School connectedness Selfesteem Depression Hopelessness Loneliness Suicidal ideation School activities Performance .13* -11/.06* .01 -.04 -.02 .09 .04/ -.21** .04 .06 School involvement .10 .08 10 .07 .06 .02 -.13* -.02 -.04 Academic -.06 -.02 -.08 .21/-.38* .04 .05 .04 .05 -.03 Prosocial .15* .19** .12 .07 -.03 .01 -.07 -.10 -.05 Vocational .02 -.07 .05 -.06 -.06 .01 .04 .02 .04 Hobby .14* .13* .02 .04 -.05 .06/.21** .13 .04 -.00/.21* Community activities Outdoor .02 .01 .04 .09 04 .01 -.09 -.06 -.01 Religious .01 .10 .05 -.01 02 -.02 -.02 .02 .01 Hobby .11 .08 -.07 -.05 -. 11 .11 .02 .07 .04 Neighborhood club .20** .18** .03 -.04 -.06 .06 -.01 .02 .02 Summer program .13* .01 .13* .02 .05 .05 -.02 -.12 .05 Other .01 .05 .04 -.02 -.14* .15* .16** -.01 .06 Note. Though all are included in the same table, the data for each activity represen ts a separate analysis. Two coefficients indicate female and male relationships, respectivel y, that resulted from a significant interaction effect. The p-value presented for the se models represents the significance level for the interaction. Si ngle coefficients indicate the se x interaction was not signific ant in the model. Sex, race/ethnicity, college sport participation, and invo lvement in other college organizations were controlled in all models. p < .05, ** p < .01, *** p < .001.

PAGE 137

137 Table 6-5. Standardized regres sion coefficients for physical ac tivity and psychosocial variables predicted from total participation in extrac urricular activities and time spent in school activities during high school Activity High school moderate activity High school vigorous activity Social support School connectedness Selfesteem Depression Hopelessness Loneliness Suicidal ideation Total # of extracurricular activities (TEAP) Level of involvement Moderately involved (2-5 activities) .26 -.45 -.86* -.65 -.51 .08 -.35 .42 .32 Highly involved (6-8 activities) .31 -.30 -.83 -.60 -.55 -.02 -.42 .33 .35 Extremely involved (> 9 activities) .55 -.07 -.79 -.53 -. 58 .07 -.48 .36 .34 Time spent in school activities 1 to 4 hrs/wk .00 -.10 17 -.13 .20 -.18 -. 25/-.03** -.18 -.14 5 to 9 hrs/wk .08 .11 .27* -.10 21 -.27* -.33** -.17 -.27* > 10 hrs/wk .30* .45*** .32** .16* .33** -.34** -.36** -.32** -.35** Note. Though both are incl uded in the same table, the data for the extracurricular activity par ticipation variable and time spent in extracurricular activities variable represent separate analyses. Two coefficients indicate female and male relationships, respe ctively, that resulted from a significant interaction effect. The p-value presented for these models represents the significance level f or the interaction. Single coefficients indicate the sex interaction was not significant in the model. Sex, race/ethnicity, college sp ort participation, and involvement in other college organizations were controlled in all models. p < .05, ** p < .01, *** p < .001.

PAGE 138

138 Figure 6-1. Model presenting proposed mech anisms through which sport participation influences the risk of suicidal ideation Sport Participation Physical Activity Social Support Depression Loneliness Suicidal Ideation Hopelessness Self-Esteem + + + + + School Connectedness + +

PAGE 139

139 Note Sex and race/ethnicity were controlled in all models. Also, involvement in sports or other organizations during college were controlled in models that included extramural sport participation. Two coefficients indicate female and male relationships, respectively, that resulted from a signi ficant interaction effect. The p-value presented for these models represents the significance leve l for the interaction. Single coefficients indicate the sex interaction was not significant in the model. p < .05, ** p < .01, *** p < .001. Figure 6-2. Path model with standardized regression coefficients Social Support (SS) .47*** ( SC ) -.48*** School Connectedness (SC) ( SS ) -.79*** ( SC ) -.20/-1.33** ( SS ) -.45*** ( SS ) -.47*** ( SC ) -.37*** .27*** ( SS ) -.42*** ( SC ) -.30*** Depression Loneliness Suicidal Ideation Hopelessness Self-Esteem -.09 .11 -.18** -.50*** .49*** .45*** .53*** ( SS ) -.42*** Extramural Sport Participation High School Vigorous Activity .19**

PAGE 140

140 CHAPTER 7 SUMMARY, CONCLUSIONS, AND RE COMMENDATIONS Summary The present work focused on the relationshi p between sport participation and risk of suicide among young people. Specifically, the work served three purposes: 1) to highlight potential ways sport participation may enhance re silience in youth and, th us, reduce the risk of suicide; 2) to further explore associations among physical ac tivity, sport participation, and suicide risk; and 3) to address key methodological limitations in the extant literature. Five investigations extend the current literature a nd address weaknesses in previous research. Chapter 2 described an inves tigation that synthesized the empirical evidence regarding sport participation and protective factors associat ed with resilience in youth. Findings suggested that involvement in sports may represent an optimal activity to promote mechanisms that enhance resilience in youth. Th e article titled Sport Participation: A Vehicle for Promoting Resilience in Youth was published in The Health Education Monograph Series November 2008. The research presented in Chapter 3 examin ed relationships between different types of physical activity and rates of hopelessness, de pression, and suicidal behavior among college students. Data for this paper derived from the 2005 American College Health Association National College Health Assessment. This study represented the first i nvestigation to examine the association between strength or toning activity and suicide risk among college students, and one of only two studies to examine the effects of aerobic activity on suicide risk among college students. Findings provided empirical evidence th at establishes the asso ciation between physical activity, especially aero bic activity, and reduced risk of su icidal behavior among young adults. However, findings also suggested that for some students, engaging in frequent exercise with the

PAGE 141

141 intent to lose weight related to increased risk of emotional distress. Also, females who combined frequent exercise and deleterious dieting behavior showed increas ed suicide risk. The article based on this research, titled Associations Be tween Physical Activity and Reduced Rates of Hopelessness, Depression, and Suicidal Beha vior Among College Students, was published in the January/February 2009 issue of the Journal of American College Health The study presented in Chapter 4 examined associations among physical activity, sport participation, and suicide risk in adolescents us ing data from the Centers for Disease Control and Preventions 2005 Youth Risk Behavior Survey. In October 2008, the Journal of School Health published findings from this research in the article High School Youth and Suicide Risk: Exploring Protection Afforded Through Physical Activity and Sport Particip ation. To test the assumption that involvement in organized sports affords distinct protection against suicide risk, sport participation was controlled for during the physical activity analyses, and physical activity was controlled for during the sport analyses. By controlling the effect of one variable while examining the relationship between the other va riable and the outcome measures, this study added to the literature and provided a more cl ear understanding of re lationships among the variables. Findings from this study confirme d that mechanisms other than physical activity contribute to the protective association between s port participation and reduced risk of suicide. Furthermore, findings indicated that, for females, ex ercising to lose weight, as well as restricting calories or taking diet supplements to attain the same goal, significantly increased suicide risk. Chapter 5 described research conducted with students enrolled in undergraduate courses at the University of Florida. This study sought to test the hypothesis that sport participation protects against risk factors a ssociated with suicide through multip le mechanisms. Analyses also

PAGE 142

142 tested the hypothesis that partic ipating in sports provides college students increased protection against suicidal ideation more effectively than does participating in other activities. This study represented the first effort to test a model depicting potential mechanisms through which collegiate sport pa rticipation relates to reduced risk of suicidal ideation among college students. Analyses confirmed a path mo del and tested for mediat ion effects. Vigorous activity mediated relationships between sport pa rticipation and self-est eem and depression; and self-esteem and depression mediat ed the relationship between vigorous activity and suicidal ideation. Social support mediat ed relationships between spor t participation and depression, hopelessness, and loneliness; and each of these ri sk factors partially me diated the relationship between social support and suicidal ideation. However, no variable fully mediated the relationship between sport particip ation and suicidal ideation. This study also compared effects of spor t participation to involvement in other extracurricular activities. Findings indicated th at participation in non-s port university activities did not provide students with th e same psychosocial benefits as did sport participation. Findings from this research appear in the manuscript titled Exploring Potent ial Mediating Pathways Through Which Participation in S ports Relates to Reduced Risk of Suicidal Ideation, accepted for publication in Research Quarterly for Exercise and Sport Chapter 6 discussed an investigation that te sted an enhanced model examining the effects of involvement in sports during high school. Improved assessments of sport participation and involvement in other extracurricul ar activities allowed for a detail ed analysis regarding types of activities that afforded the greatest benefit. This study reve aled partial support for the proposed model. Vigorous activity was not related to self-esteem or depression, but vigorous activity mediated the relationship between extramural spor t participation and suicidal ideation. Social

PAGE 143

143 support and school connectedness medi ated relationships between ex tramural sport participation and depression, hopelessness, loneliness, and suicidal ideation. Furthermore, each of these risk factors mediated the relationship between the protective factors and suicidal ideation. Findings also showed that, compared to other activities, extramural sport participation related to higher levels of protective factors and to lower levels of risk factors a ssociated with suicide risk. The manuscript based on this research, titled E xploring Protective Relationships Between Participation in Sports and Risk Factors A ssociated with Adolescent Suicide, has been submitted for publication. Conclusions Findings from this line of research suppor t the contention that sport participation favorably affects a variety of broadly defined outcom es including social and psychological wellbeing [67, 74]. Researchers identified a relationship between sport participation and reduced risk of suicidal behavior [5-13, 73]. Researchers al so proposed hypothetical mechanisms to explain the association between sport participation and reduced odds of suicidal ity [80]. However, no study specifically tested these mechanisms. This research advanced the science in this area by exploring potential mechanisms that account for the relationship between sport participation and reduced risk of suicidal ity among young people. Consistent with previous research [4, 70, 72], findings from the present studies support an association between physical activity and decreased risk of suicidal behavior. However, after controlling for sport participation, relationshi ps sometimes changed. For example, among females in one study, no significant differences em erged between physical activity and suicidal behavior after sport participat ion was controlled [178]. Also, two studies that tested the mediating effect of physical activ ity on the relationship between sport participation and suicidal ideation yielded inconsistent findi ngs. When sport pa rticipation, physical ac tivity, and suicidal

PAGE 144

144 ideation were included in one model, the first st udy showed that sport pa rticipation mediated the relationship between physical act ivity and suicidal ideation. Conversely, in the second study, physical activity mediated the relationship betwee n sport participation and suicidal ideation. The two investigations explored re lationships between suicidal id eation, and physical activity and sport participation using different independent variables. Theref ore, reasons for the dissonant results remain uncertain. Furthermore, among some female students, e ngaging in frequent physic al activity to lose weight, and/or engaging in delete rious dieting behavior, related to increased suicide risk. These findings may explain relationships found in previous research between high levels of physical activity and increased risk of suicidal behavi or among females [9, 10]. Thome and Espelage [129] found that exercise related to positive psychological he alth in college females, but only when exercise was unassociated with an eating disorder. Thus, findings from the present research support the theory that factors related to poor body imag e may contribute to low selfesteem, depression, and suicidal feelings that likely influence the relationship between physical activity and suicidal behavior [9, 10]. The inverse relationship between sport particip ation and suicide risk remained consistent across the three studies in this dissertation that ex amined sports effects on suicidality. Findings support previous research that showed direct eff ects of sport participatio n on suicide risk [6, 7, 10-13]. Furthermore, tests of a theoretical model suggest that vigorous activity, social support, and school connectedness at least partially account for the relationship between sport participation and reduced risk of suicidal ideation. These three factors influence the relationship between sport participation and suicidal ideation by protecting against significant risk factors low self-esteem, depression, hopelessness, and lonelin ess associated with adolescent suicide.

PAGE 145

145 The two studies that compared effects of s port participation to involvement in other extracurricular activities yielded consistent findings. Specifi cally, these studies found that participation in non-sport activities did not provide students with the same psychosocial benefits as did sport participation. These findings s upport previous research that examined similar relationships among high school st udents [5, 170]. Furthermore, an examination of differential effects across types of sports identified extramural sport participation as more beneficial than involvement in intramural or community sports. This finding supports fi ndings from Marsh and Kleitman [67] and Broh [179] who determined th at extramural/interscholastic sports produced more positive effects than intramural sports. In conclusion, this research advances the sc ience regarding sport part icipation and risk of suicide. Findings from this line of inquiry provide a foundation for research to examine pathways through which sport part icipation relates to reduced ri sk of suicidal behavior. Researchers and practitioners shoul d not oversimplify or overstate the associations between sport participation and suicide risk, but findings suggest opportunities for prevention and merit consideration in public health policy and planning [12]. Hea lth professionals may use this research to advocate for increas ed opportunities to involve young people in sports and vigorous activity. School officials and public health planners can use these findings as a tool to evaluate health costs associated with eliminating sport acti vities, especially as programs face cuts during economic difficulties [177]. Eventually, practitioners may incorporate these findings into intervention efforts to preven t and treat suicidal tendencie s among young people [80]. Recommendations for Future Research Test the proposed m odel using longitudinal designs. Validate the model with high school students an d with a larger sample of males.

PAGE 146

146 Incorporate enhanced assessments of physical activity and suicidal behavior (e.g., history of suicide attempts). Empirically study ways through which sport pa rticipation enhances resilience in young people, thus helping to redu ce the risk of suicide. Continue exploring relationships between involvement in differe nt types of extracurricular activities and psychos ocial variables. Explore the association between sport participation and suicidality among young people with a broader range of mood disturbances and mental health problems to determine whether or not research with a more di stressed population yields similar findings.

PAGE 147

147 APPENDIX A DEMOGRAPHIC AND BACKGR OUND INFORMATION 1. What is your sex? O Female O Male 2. How old are you? ______ years 3. What is your class level? (check one circle) (item included only for Chapter 6) O Freshman O Sophomore O Junior O Senior O Graduate 4. In what semester year did you graduate from high school? (check one circle) (item included only for Chapter 6) O Winter (end of fall) 2007 O Summer 2007 or earlier O Spring 2008 O Did not graduate, but passed GED test O Summer 2008 5. From what kind of high school di d you graduate? (check one circle) (item included only for Chapter 6) O Public school (not charter or magnet) O Private religious/parochial school O Public charter school O Private independent college-prep school O Public magnet school O Home school 6. Which of the following best describes yo ur ethnic background? (check one circle) O Hispanic or Latino O Not Hispanic or Latino 7. Which of the following best describes your race? (check one circle) O White or Caucasian O Black or African American O Asian O American Indian or Alaskan Native O Native Hawaiian or Other Pacific Islander O Some Other Race O Multi-Racial (option included only for Chapter 6)

PAGE 148

148 APPENDIX B PHYSICAL ACTIVITY, SPORT PARTICIPATION, AND INVOLVEMENT IN OTHER EXTRACURRICULAR ACTIVITIES The following item s were included for Chapters 5 and 6. 1. On how many of the past 7 days did you participate in physical activity for at least 30 minutes that did not make you sweat or breathe hard, such as fast walking, slow bicycling, skating, pushing a lawn mower, or mopping floors? (check one circle) O 0 O 1 O 2 O 3 O 4 O 5 O 6 O 7 2. On how many of the past 7 days did you exercise or participate in physical activity for at least 20 minutes that made you sweat and breathe hard, such as jogging, swimming laps, fast bicycling, fast dancing, or similar aerobic activities? (check one circle) O 0 O 1 O 2 O 3 O 4 O 5 O 6 O 7 (CDCs YRBS, 2006 [79]) The following items were included only for Chapter 5. 3. During the past 12 months, how many UF sponsored clubs or organizations (not including sport teams) have you been a member of? O 0 O 1 O 2 O 3 O 4 or more For the following questions, do NOT include sport classes you took for credit. 4. During the past 12 months, on how many varsity or club sports teams run by UF did you play? O 0 O 1 O 2 O 3 O 4 or more 5. During the past 12 months, on how many intramural sports teams run by UF did you play? O 0 O 1 O 2 O 3 O 4 or more The following items were included only for Chapter 6. 6. During your last year of high school on how many days during a typical week did you participate in physical activity for at least 30 minutes that did not make you sweat or breathe hard, such as fast walking, slow bicycling, skating, pushi ng a lawn mower, or mopping floors? (check one circle) O 0 O 1 O 2 O 3 O 4 O 5 O 6 O 7

PAGE 149

149 7. During your last year of high school on how many days during a typical week did you exercise or participate in physical activity for at least 20 minutes that made you sweat and breathe hard, such as jogging, swimming laps, fast bicycling, fast dancing, or similar aerobic activities? (check one circle) O 0 O 1 O 2 O 3 O 4 O 5 O 6 O 7 8. During the past 12 months, were you a member of any college intramural, club, or varsity sports teams ? O No O Yes 9. During the past 12 months, were you a member of any other college organizations (not including sports teams)? O No O Yes 10. During the past 12 months, how much total time did you spend on all extracurricular activities sponsored by your HIGH SCHOOL in a typical week? (check one circle) O None O Less than 1 hour per week O 1 to 4 hours per week O 5 to 9 hours per week O 10 to 19 hours per week O 20 hours or more per week

PAGE 150

150 The following 4 scales were included only for Chapter 6. Participation in Organized Sports Please mark all that apply for each inters cholastic, intramural, and community sport activity you participated in during high school INTERSCHOLASTIC means your school team competed against other school teams (i.e., freshman, junior varsity, or varsity team). SCHOOL INTRAMURAL means the teams were within your own school (do not include PE classes). COMMUNITY ACTIVITY means the teams were run by community organizations, not your high school (i.e., YMCA, county league, other community recreation center). MARK ALL THAT APPLY ON EACH LINE: School Did Not Have Did Not Participate Participated in Intramural Sports Participated on a Junior Varsity or Freshman Team Participated on a Varsity Team Participated on a Community Team a. Baseball/ Softball O O O O O O b. Football O O O O O O c. Basketball O O O O O O d. Soccer O O O O O O e. Swim team O O O O O O f. Cheerleading O O O O O O g. Dance/Drill team O O O O O O h. Other team sport (hockey, volleyball, etc.) O O O O O O i. Other individual sport (cross-country, track, golf, gymnastics, wrestling, tennis, etc.) O O O O O O

PAGE 151

151 Participation in Other School-Sponsored Activities Please check one circle for each school-sponsored club or group you participated in during high school MARK ONE ON EACH LINE: School Did Not Have Did Not Participate Participated a. Band, choir, orchestra, or other school music group O O O b. School play or musical, or drama club O O O c. Student government O O O d. National Honor Society or other academic honor society O O O e. School yearbook, newspaper, or other literary magazine O O O f. Service clubs (AFS, Key club, etc.) O O O g. Academic clubs (Art, Computer, Debate, Foreign languages, Science, Psychology, Math, Philosophy, etc.) O O O h. Hobby clubs (photography chess, frisbee, etc.) O O O i. Vocational education or professional clubs O O O j. Religious organization or group O O O

PAGE 152

152 Participation in Other Community Activities Please check one circle for each community group or organization, not including sport teams, you participated in during high school MARK ONE ON EACH LINE: Did Not Have Did Not Participate Participated a. Scouting O O O b. Religious youth groups O O O c. Hobby clubs O O O d. Neighborhood clubs or programs O O O e. Boys clubs or girls clubs O O O f. 4-H O O O g. YMCA or other youth groups O O O h. Summer programs such as workshops or institutes in science, drama, language, etc. O O O i. Other O O O (National Center for Education Statistics, 1988 [172])

PAGE 153

153 APPENDIX C SCHOOL CONNECTEDNESS SCALE Indicate how m uch you agree or disagree with each of the following statements regarding your high school experience. Check one circle for each Strongly Disagree Disagree Agree Strongly Agree 1. I felt close to people at my high school. O O O O 2. I felt like I was part of my high school. O O O O 3. I was happy to be at my high school. O O O O 4. The teachers at my high school treated students fairly. O O O O 5. I felt socially accepted at my high school. O O O O 6. I felt safe at my high school. O O O O (Bearman, Jones, & Udry, 1998 [173])

PAGE 154

154 APPENDIX D SCALES THAT CANNOT BE REPRODUCED Personal Resource Questionnaire85 Part II (W einert, 1987 [154]) Adult Suicidal Ideation Ques tionnaire (Reynolds, 1991 [161]) Beck Hopelessness Scale (Beck, 1974 [157]) Rosenberg Self-Esteem S cale (Rosenberg, 1965 [155]) Center for Epidemiologic Studies Depression Scale (Radloff, 1977 [156]) Revised UCLA Loneliness Scale (Russell, 1980 [160])

PAGE 155

155 LIST OF REFERENCES 1. American Association of Suicidology. Y outh suicide fact sheet. 2007, Available at: http://www.suicidology.org. Accessed: 12 December 2007. 2. Reynolds W, Mazza J. Suicide and suicidal behaviors in children and adolescents. In: Handbook of Depression in Children and Adolescents. Reynolds WM, Johnston HF (eds). New York, NY: Plenum Press. 1994, 525-537. 3. Baumert P, Henderson J, Thompson N. Health risk behaviors of adol escent participants in organized sports. J Adolesc Health 1998; 22: 460-465. 4. Ferron C, Narring F, Cauderay M et al. Sport activity in adolescence: associations with health perceptions and experimental behaviours. Health Educ Res 1999; 14 : 225-233. 5. Harrison P, Narayan G. Differences in behavior, psychological factors, and environmental factors associated with particip ation in school sports and other activities in adolescence. J Sch Health 2003; 73: 113-120. 6. Miller K., Sabo D, Melnick M et al. The Women's Sports Foundation Report: Health Risks and the Teen Athlete. East Me adow, NY: Women's Sports Foundation. 2000. 7. Oler M, Mainous A, Martin C et al. Depression, suicidal id eation, and substance use among adolescents. Are athletes at less risk? Arch Fam Med 1994; 3: 781-785. 8. Steiner H, McQuivey R, Pavelski R et al. Adolescents and sports : risk or benefit? Clin Pediatr 2000; 39: 161-166. 9. Unger J. Physical activity, participation in t eam sports, and risk of suicidal behavior in adolescents. Am J Health Promot 1997; 12: 90-93. 10. Brown D, Blanton C. Physical activity, sports participation, and suic idal behavior among college students. Med Sci Sports Exerc 2002; 34: 1087-1096. 11. Page R, Hammemeister J, Scanlan A et al. Is school sports part icipation a protective factor against adolescent health risk behaviors? J Health Educ 1998; 29: 186-192. 12. Sabo D, Miller K, Melnick M et al. High school athletic part icipation and adolescent suicide: a nationwide US study Int Rev Sociol Sport 2005; 40 : 5-23. 13. Brown D, Galuska D, Zhang J et al. Physical activity, sport pa rticipation, and suicidal behavior: U.S. high school students. Med Sci Sports Exerc 2007; 39: 2248-2257. 14. Applebaum P. "Depressed? Get out!": de aling with suicidal students on college campuses. Psychiatr Serv 2006; 57: 914-916. 15. Lamberg L. Experts work to prevent college suicides. JAMA 2006; 296: 502-504.

PAGE 156

156 16. Centers for Disease Control and Prev ention. Suicide trends among youths and young adults aged 10-24 years United States, 1990-2004. MMWR 2007; 56: 905-908. 17. Hamilton B, Minino A, Martin J et al. Annual summary of vital statistics: 2005. Pediatrics 2007; 119: 345-360. 18. Centers for Disease Control and Prevention. Youth Risk Behavior Surveillance United States, 2007. Surveillance Summaries, June 6. MMWR 2008; 57: No. SS-#4. 19. American College Health Association. Amer ican College Health Association National College Health Assessment Spring 2007 Re ference Group Data Report (abridged). J Am Coll Health 2008; 56: 469-479. 20. US Department of Health and Human Se rvices. Healthy People 2010. Washington, DC: US Government Printing Office. November 2000. 21. National Institute of Mental Health. Interventions for suicidal youth. 2001, Available at: http://grants.nih.gov/gr ants/guide/pa-files/PA-00-077.html. Accessed: 17 February 2006. 22. Allberg W, Chu L. Understanding adolescen t suicide: correlates in a developmental perspective. Sch Couns 1990; 37: 343-350. 23. Beautrais A, Joyce P, Mulder R. Risk f actors for serious suicide attempts among youths aged 13 through 24 years. J Am Acad Child Adolesc Psychiatry 1996; 35: 1174-1182. 24. Beautrais A. Risk factors for suicid e and attempted suicide among young people. Aust N Z J Psychiatry 2000; 34: 420-436. 25. Fergusson D, Woodward L, Horwood L. Risk factors and life processes associated with the onset of suicidal behavior dur ing adolescence and early adulthood. Psychol Med 2000; 30: 23-39. 26. Foley D, Goldston D, Costello E et al. Proximal psychiatric risk factors for suicidality in youth: The Great Sm oky Mountains Study. Arch Gen Psychiatry 2006; 63 : 1017-1024. 27. Gould M, Velting T, Shaffer D. Youth suicide: a review. Prev Res 2006; 13 : 3-7. 28. Conner K, Duberstein P, Conwell Y et al. Psychological vulnerab ility to completed suicide: a review of empirical studies. Suicide Life Threat Behav 2001; 31 : 367-385. 29. Bridge J, Goldstein T, Brent D. A dolescent suicide and suicidal behavior. J Child Psychol Psychiatry 2006; 47: 372-394.

PAGE 157

157 30. Office of Applied Studies. Results fro m the 2004 National Survey on Drug Use and Health: National Findings. Suicidal Thoughts among Youths Aged 12 to 17 with Major Depressive Episode. Rockville, MD: Substa nce Abuse and Mental Health Services Administration. 2005. 31. Prinstein M, Boergers J, Spirito A et al. Peer functioning, family dysfunction, and psychological symptoms in a risk factor mode l for adolescent inpatients' suicidal ideation severity. J Clin Child Psychol 2000; 29: 392-405. 32. Kandel D, Raveis V, Davies M. Suicidal ideation in adolescence: depression, substance use, and other risk factors. J Youth Adolesc 1991; 20: 289-309. 33. Field T, Diego M, Sanders C. Adol escent depression and risk factors. Adolescence 2001; 36: 491-498. 34. Konick L, Gutierrez P. Testing a model of suicide ideation in college students. Suicide Life Threat Behav 2005; 35: 181-192. 35. Spirito A, Esposito-Smythers C. Attempte d and completed suicide in adolescence. Annu Rev Clin Psychol 2006; 2: 237-266. 36. Fotti S, Katz L, Afifi T et al. The associations between peer and parental relationships and suicidal behaviours in early adolescents. Can J Psychiatry 2006; 51: 698-703. 37. Thompson E, Mazza J, Herting J et al. The mediating roles of anxiety, depression, and hopelessness on adolescent suicidal behaviors. Suicide Life Threat Behav 2005; 35: 1434. 38. Goldston D, Sergent-Daniel S, Reboussin B et al. Cognitive risk factors and suicide attempts among formerly hospitalized adoles cents: a prospective naturalistic study. J Am Acad Child Adolesc Psychiatry 2001; 40: 91-99. 39. Kumar G, Steer R. Psychosocial correlates of suicidal ideation in adolescent psychiatric inpatients. Suicide Life Threat Behav 1995; 25: 339-346. 40. Swedo S, Rettew D, Kuppenheimer M et al. Can adolescent suicide attempters be distinguished from at-risk adolescents? Pediatrics 1991; 88: 620-629. 41. Rutter P, Behrendt A. Adolescent suic ide risk: four psyc hological factors. Adolescence 2004; 39: 295-302. 42. Groholt B, Ekeberg O, Wichstrom L et al. Suicidal and nonsuicidal adolescents: different factors contribute to self-esteem. Suicide Life Threat Behav 2005. 35(5): p. 525-535. 43. Mazza J, Eggert L. Activity involvement am ong suicidal and nonsuicidal high-risk and typical adolescents. Suicide Life Threat Behav 2001; 31: 265-281.

PAGE 158

158 44. Lazear K, Nations L, Vaughn D et al. Florida Youth Suicide Prevention Study. Tampa, FL: Department of Child and Family Studies, Louis de la Parte Florida Mental Health Institute, University of South Florida. 1999. 45. Watt T, Sharp S. Gender differences in stra ins associated with su icidal behavior among adolescents. J Youth Adolesc 2001; 30 : 333-348. 46. Bearman P, Moody J. Suicide and friendships among American adolescents. Am J Public Health 2004; 94: 89-95. 47. Wild L, Flisher A, Bhana A et al. Associations among adoles cent risk behaviours and self-esteem in six domains. J Child Psychol Psychiatry 2004; 45: 1454-1467. 48. McGee R, Williams S. Does low self-esteem predict health compromising behaviours among adolescents? J Adolesc 2000; 23: 569-582. 49. McGee R, Williams S, Nada-Raja S. Low self-esteem and hopelessness in childhood and suicidal ideation in early adulthood. J Abnorm Child Psychol 2001; 29: 281-291. 50. Overholser, J, Adams D, Lehnert K et al. Self-esteem deficits and suicidal tendencies among adolescents. J Am Acad Child Adolesc Psychiatry 1995; 34: 919-928. 51. Tomori M, Zalar B. Sport and physical activity as possible protectiv e factors in relation to adolescent suicide attempts. Int J Sport Psychol 2000; 31: 405-413. 52. Everall R, Altrows J, Paulson B. Creating a future: a study of resilienc e in suicidal female adolescents. J Couns Dev 2006; 84: 461-470. 53. Masten A, Best K, Garmezy N. Resilien cy and development: contributions from the study of children who overcome adversity. Dev Psychopathol 1990; 2: 425-444. 54. Resnick M, Rinehart P. In fluencing Behavior: The Powe r of Protective Factors in Reducing Youth Violence. Minneapolis, MN: University of Minnesota, Center for Adolescent Health and Development. 2004. 55. Borowsky I, Ireland M, Resnick M. Adolescen t suicide attempts: ri sks and protectors. Pediatrics 2001; 107: 485-494. 56. Doan J, Roggenbaum S, Lazear K. Youth Suicide Prevention School-Based Guide Issue Brief 3a: Risk Factors: Risk and Protectiv e Factors, and Warning Signs. Tampa, FL: Department of Child and Family Studies, Divi sion of State and Local Support, Louis de la Parte Florida Mental Health Institute, University of South Florida. 2003. 57. Randell B, Wang W, Herting J et al. Family factors predicting categories of suicide risk. J Child Fam Stud 2005; 15: 255-270.

PAGE 159

159 58. Rutter P, Soucar E. Youth suic ide risk and sexual orientation. Adolescence 2002; 37: 289299. 59. Kaslow N, Sherry A, Bethea K et al. Social risk and protec tive factors for suicide attempts in low income African American men and women. Suicide Life Threat Behav 2005; 35: 400-412. 60. Rink E, Tricker R. Promoting healthy be haviors among adolescents : a review of the resiliency literature. Am J Health Stud 2005; 20: 39-46. 61. Resnick M, Bearman P, Blum R et al. Protecting adolescents fr om harm: findings from the National Longitudinal Study on Adolescent Health. JAMA 1997; 278: 823-832. 62. Thorlindsson T, Bjarnason T. Modeling Durkheim on the micro level: a study of youth suicidality. Am Sociol Rev 1998; 63: 94-110. 63. Doan J, Roggenbaum S, Lazear K. Youth Suicide Prevention School-Based Guide Issue Brief 2: School Climate. Tampa, FL: Depart ment of Child and Family Studies, Division of State and Local Support, Louis de la Parte Florida Mental Health Institute, University of South Florida. 2003. 64. Bond L, Butler H, Thomas L et al. Social and school connect edness in early secondary school as predictors of late teenage substan ce use, mental health, and academic outcomes. J Adolesc Health 2007; 40: 357.e9-357.e18. 65. Ozer E. The impact of violence on urban a dolescents: longitudinal effects of perceived school connection and family support. J Adolesc Res 2005; 20 : 167-192. 66. McNeely C, Nonnemaker J, Blum R. Promo ting school connectedness: evidence from the National Longitudinal Study of Adolescent Health. J Sch Health 2002; 72 : 138-146. 67. Marsh H, Kleitman S. School athletic part icipation: mostly gain with little pain. J Sport Exerc Psychol 2003; 25: 205-228. 68. Sabo D, Miller K, Melnick M et al Her Life Depends on It: Spor t, Physical Activity, and the Health and Well-Being of American Girls. East Meadow, NY: Women's Sports Foundation. 2004. 69. Tracy A, Erkut S. Gender and race patterns in the pathways from sports participation to self-esteem. Sociol Perspect 2002; 45 : 445-466. 70. Simon T, Powell K, Swann A. Involvement in physical activity and risk for nearly lethal suicide attempts. Am J Prev Med 2004; 27: 310-315.

PAGE 160

160 71. Paffenbarger R, Lee I, Leung R. Physical act ivity and personal characteristics associated with depression and suicide in American college men. Acta Psychiatr Scand 1994; Suppl. 377 : 16-22. 72. Brosnahan J, Steffen L, Lytle L et al. The relation between phys ical activity and mental health among Hispanic and nonHispanic White adolescents. Arch Pediatr Adolesc Med 2004; 158: 818-823. 73. Sabo D, Melnick M, Miller K et al. Athletic participation and the health risks of adolescent males: a national study. Int J Mens Health 2002; 1: 173-189. 74. Marsh H. The effects of participation in sport during the last tw o years of high school. Sociol Sport J 1993; 10: 18-43. 75. Snyder E, Spreitzer E. Social psychological c oncomitants of adolescents' role identities as scholars and athletes: a longitudinal analysis. Youth Soc 1992; 23: 507-522. 76. Richman E, Schaffer D. "If you let me play sports": how might sport participation influence the self-esteem of adolescent females? Psychol Women Q 2000; 24: 189-199. 77. Anteghini M, Fonseca H, Ireland M et al. Health risk behaviors and associated risk and protective factors among Brazilian adolescents in Santos, Brazil. J Adolesc Health 2001; 28: 295-302. 78. Rueter M, Kwon H. Developmental trends in adolescent suicidal ideation. J Res Adolesc 2005; 15: 205-222. 79. Centers for Disease Control and Prevention. Youth Risk Behavior Surveillance United States, 2005. Surveillance Summaries, June 9. MMWR 2006; 55: No.SS-5. 80. Brown D. Physical activity, sports particip ation, and suicidal be haviors: a literature review. Int J Sport Exerc Psychol 2005; 3: 484-500. 81. Zarrett N, Lerner R. Ways to promote the positive development of children and youth. 2008, Available at: http://www.childtre nds.org. Accessed: 24 July 2008. 82. Eccles J, Barber B. Student council, volunt eering, basketball, or marching band: what kind of extracurricula r involvement matters? J Adolesc Res 1999; 14: 10-43. 83. Marsh H. Extracurricular activities: beneficial extension of the traditional curriculum or subversion of academic goals? J Educ Psychol 1992; 84: 553-562. 84. McGee R, Williams S, Howden-Chapman P et al. Participation in clubs and groups from childhood to adolescence and its eff ects on attachment and self-esteem. J Adolesc 2006; 29: 1-17.

PAGE 161

161 85. Dworkin J, Larson R, Hansen D. Adolescents' accounts of growth experiences in youth activities. J Youth Adolesc 2003; 32: 17-26. 86. Kaufmann A. Interscholastic sports particip ation as a predictor of academic success for high school students. Doctoral dissertation. Milwaukee, WI: University of Wisconsin. ProQuest 2002; AAT3047511. 87. Plante T, Rodin J. Physical fitn ess and enhanced psychological health. Curr Psychol Res Rev 1990; 9: 1-22. 88. Salmon P. Effects of physical exercise on anxi ety, depression, and se nsitivity to stress: a unifying theory. Clin Exerc Rev 2000; 21: 33-61. 89. Briddle S, Fox K, Boutcher S. (eds). Physical Activity and Psychological Well-Being New York, NY: Routledge. 2000. 90. Joiner T, Tickle J. Exercise and depressive and anxious symptoms: what is the nature of their interrelations? J Occup Rehabil 1998; 8: 191-198. 91. Dishman R, Hales D, Pfeiffer K et al. Physical self-concept and self-esteem mediate cross-sectional relati ons of physical activity and spor t participation with depression symptoms among adolescent girls. Health Psychol 2006; 25: 396-407. 92. Byrne A, Bryne D. The effect of exercise on depression, anxiety, and other mood states: a review. J Psychosom Res 1993; 37: 565-574. 93. Gauvin L, Spence J. Psychological research on exercise and fitness: current research trends and future challenges. Sport Psychol 1995; 9: 434-448. 94. Nabetani T, Tokumnaga M. The effect of short-term (10and 15-min) running at selfselected intensity on mood alteration. J Physiol Anthropol Appl Human Sci 2001; 20 : 231-239. 95. Fox K. The effects of exercise on self-percep tions and self-esteem. In: Briddle S, Fox K, Boutcher S. (eds). Physical Activity and Psychological Well-being New York, NY: Routledge. 2000, 88-117. 96. Sonstroem R. Exercise and self-esteem. Exerc Sports Sci Rev 1984; 12: 123-155. 97. American College Health Association. Amer ican College Health Association National College Health Assessment: Reference Group Data Report Spring 2005. Baltimore, MD: American College Health Association. 2005. 98. Blinn-Pike L. Why abstinent adolescents report they have not ha d sex: understanding sexually resilient youth. Fam Relat 1999; 48: 295-301.

PAGE 162

162 99. Fine A, Large R. A Conceptual Framew ork for Adolescent Health. Washington, DC: Association of Maternal and Child Health Programs. 2002. 100. Benson P. Building Developmental Assets : A New Strategy for Preventing High-risk Behavior. Minneapolis, MN: Search Institute, Center for the App lication of Prevention Technologies. 2000. 101. Hjemdal O, Friborg O, Stiles T et al. A new scale for adolescent resilience: grasping the central protective resources behind healthy development. Meas Eval Couns Dev 2006; 39: 84-96. 102. Cohu E. Participation in athle tics and development of certain traits related to resiliency theory. Doctoral dissertation. Jackson, TN: Union University. ProQuest 2005; AAT3197274. 103. President's Council on Physical Fitness and Sports. Physical Activ ity and Sport in the Lives of Girls: Physical a nd Mental Health Dimensions from an Interdisciplinary Approach. Minneapolis, MN: The Center for Research on Girls and Women in Sport, University of Minnesota. Spring 1997. 104. Donaldson S, Ronan K. The effects of sports participation on young adolescents' emotional well-being. Adolescence 2006; 41: 369-389. 105. Hawkins R, Mulkey L. Athletic investment and academic resilience in a national sample of African American females and males in the middle grades. Educ Urban Soc 2005; 38 : 62-88. 106. Hansen D, Larson R, Dworkin J. What adolescents learn in organized youth activities: a survey of self-reported developmental experiences. J Res Adolesc 2003; 13 : 25-55. 107. Bartko W, Eccles J. Adolescent participation in structured and unstructured activities: a person-oriented analysis. J Youth Adolesc 2003; 32: 233-241. 108. Pedersen S, Seidman E. Team sports achievement and self-esteem development among urban adolescent girls. Psychol Women Q 2004; 28: 412-422. 109. Tiggemann M. The impact of adolescent girls' life concerns and le isure activities on body dissatisfaction, disordered eating, and self-esteem. J Genet Psychol 2001; 162: 133-142. 110. Baily R. Physical education and sport in schools: a review of benefits and outcomes. J Sch Health 2006; 76: 397-401. 111. Bell C, Suggs H. Using sports to strengthe n resiliency in children: training heart. Child Adolesc Psychiatr Clin N Am 1998; 7: 859-865.

PAGE 163

163 112. Marsh H. Age and gender effects in physical self-concepts for adoles cent elite athletes and nonathletes: a multicohort-multioccasion design. J Sport Exerc Psychol 1998; 20: 237-259. 113. Stein C, Fisher L, Berkey C et al. Adolescent physical activity and perceived competence: does change in activit y level impact self-perception? J Adolesc Health 2007; 40: 462.e1-462.e8. 114. Yin Z, Moore J. Re-examining the role of in terscholastic sport part icipation in education. Psychol Rep 2004; 94: 1447-1454. 115. McHale S, Crouter A, Tucker C. Free-tim e activities in middle childhood: links with adjustment in early adolescence. Child Dev 2001; 72: 1764-1778. 116. Galli N, Vealey R. Thriving in the face of adversity: applying the resiliency model to high-level athletes. J Sport Exerc Psychol 2007; S162-S163. 117. Freudenberg N, Ruglis J. Reframing sc hool dropout as a public health issue. 2007, Available at: http://www.cdc.gov/pcd/i ssues/2007/oct/07_0063.htm. Accessed: 27 September 2007. 118. Dobosz R, Beaty L. The relationship between athletic participation and high school students' leadership ability. Adolescence 1999; 34 : 215-220. 119. Bailey R. Physical education and sport in schools: a review of benefits and outcomes. J Sch Health 2006; 76: 397-401. 120. Mahoney J. School extracurricular activit y participation as a moderator in the development of antisocial patterns. Child Dev 2000; 71: 502-516. 121. Hamstra K, Cherumbini J, Swanik B. Athl etic injury and parental pressure in youth sports. Athl Ther Today 2002; 7: 36-41. 122. Le Menestrel S, Perkins D. An overview of how sports, out-of -school time, and youth well-being can and do intersect. New Dir Youth Dev 2007; 115: 13-25. 123. American College Health Association. Amer ican College Health Association National College Health Assessment Spring 2005 Re ference Group Data Report (abridged). J Am Coll Health 2006; 55: 5-15. 124. Walter A, Lily C. Understanding adolescent suicide: correlates in a developmental perspective. Sch Couns 1990; 37: 343-350. 125. Gould M, Greenberg T, Velting D et al. Youth suicide risk and pr eventive interventions: a review of the past 10 years. J Am A cad Child Adolesc Psychiatry 2003; 42: 386-405.

PAGE 164

164 126. Birmaher B, Ryan N, Williamson D et al. Childhood and adolescent depression: a review of the past 10 years. J Am Acad Child Adolesc Psychiatry 1996; 35: 1427-1439. 127. Russell S, Joyner K. Adolescent sexual orientation and suicide risk: evidence from a national study. Am J Public Health 2001; 91: 1276-1281. 128. Stephenson H, Pena-Shaff J, Quirk P. Predic tors of college student suicidal ideation: gender differences. Coll Stud J 2006; 40: 109-117. 129. Thome J, Espelage D. Relations among exercise, coping, diso rdered eating, and psychological health among college students. Eat Behav 2004; 5: 337-351. 130. Tiggemann M, Williamson S. The effect of exercise on body satisfaction and self-esteem as a function of gender and age. Sex Roles 2000; 43: 119-127. 131. Strelen P, Mehaffey S, Tiggemann M. Self -objectification and esteem in young women: the mediating role of reasons for exercise. Sex Roles 2003; 48 : 89-95. 132. Hayes S, Croker P, Kowalski K. Gender differences in physical self-perceptions, global self-esteem and physical activ ity: evaluation of the physic al self-perception profile model. J Sport Behav 1999; 22: 1-14. 133. Evans D, Foa E, Gur R et al. (eds). Treating and Preventing Adolescent Mental Health Disorders: What We Know and What We Don't Know New York, NY: Oxford University Press, Inc. 2005. 134. Gallagher R. National Survey of Counse ling Center Director s. Arlington, VA: The International Association of Counseling Services, Inc. 2005. 135. American College Health Association Task Force on National Health Objectives for 2010. Healthy Campus 2010: Making It Happe n. Baltimore, MD: American College Health Association. 2002. 136. Iannos M, Tiggemann M. Personality and the excessive exerciser. Pers Individ Dif 1997; 22: 775-778. 137. Watt T, Sharp S. Race differences in strains associated with suicidal behavior among adolescents. Youth Soc 2002; 34: 232-256. 138. Ernst C, Olson A, Pinel J et al. Antidepressant effects of exercise: evidence for an adultneurogenesis hypothesis? J Psychiatry Neurosci 2006; 31: 84-92. 139. Motle R, Bimbaum A, Kubik M et al. Naturally occurring changes in physical activity are inversely related to depressive symptoms during early adolescence. Psychosom Med 2004; 66: 336-342.

PAGE 165

165 140. Crews D, Lochbaum M, Landers D. Aerobic physical activity e ffects on psychological well-being in low-income Hispanic children. Percept Mot Skills 2004; 98: 319-324. 141. Suitor J, Carter R. Jocks, nerds, babes and thugs: a research note on regional differences in adolescent gender norms. Gender Issues 1999; 3: 87-101. 142. Shagle S, Barber B. A socio-ecological analysis of adolescent suicidal ideation. Am J Orthopsychiatry 1995; 65 : 114-124. 143. National Adolescent Health Information Cent er. Fact Sheet on Suicide: Adolescents & Young Adults. San Francisco, CA: Universi ty of California, San Francisco. 2006. 144. Duarte-Velez Y, Bernal G. Suicide beha vior among Latino and Latina adolescents: conceptual and methodological issues. Death Stud 2007; 31: 435-455. 145. Choquet M, Kovess V. Suicidal thoughts among adolescents: an intercultural approach. Adolescence 1993; 28: 649-660. 146. Harter S, Marold C, Whitesell N et al. A model of the effects of perceived parent and peer support on adolescent false self behavior. Child Dev 1996; 67: 360-374. 147. The Task Force on Youth Development and Community Programs. A Matter of Time: Risk and Opportunity in the Non-School Hour s. New York, NY: Carn egie Corporation of New York. 1992. 148. Krug E, Dahlberg L, Mercy J et al. World Report on Violen ce and Health. Geneva, Switzerland: World Health Organization. 2002, 185-212. 149. World Health Organization. Suicide preventi on and special programs. 2007, Available at: http://www.who.int/mental_health/preventio n/suicide/suicideprevent/en/index.html. Accessed: 1 March 2008. 150. American College Health Association. Amer ican College Health Association National College Health Assessment Spring 2006 Re ference Group Data Report (abridged). J Am Coll Health 2007; 55: 195-206. 151. Blumenthal J, Babyak M, Doraiswamy M et al. Exercise and pharmacotherapy in the treatment of major de pressive disorder. Psychosom Med 2007; 69: 587-596. 152. Weiss R. The provisions of social relationships. In: Rubin E (ed). Doing Unto Others Englewood Cliffs, NJ: Prentice Hall. 1974, 17-26. 153. Douglas K, Collins J, Warren C et al. Results from the 1995 National College Health Risk Behavior Survey. J Am Coll Health 1997; 46 : 55-66. 154. Weinert C. A social support measure: PRQ85. Nurs Res 1987; 36: 273-277.

PAGE 166

166 155. Rosenberg M. Society and the Adol escent Self-Image. Princeton, NJ: Princeton University Press. 1965. 156. Radloff L. The CES-D Scale: A self-report depression scale for research in the general population. Appl Psychol Meas 1977; 1: 385-401. 157. Beck A, Weissman A, Lester D et al. The measurement of pessimism: The Hopelessness Scale. J Consult Clin Psychol 1974; 42 : 861-865. 158. Tabachnick B, Fidell L. Using Multivariate Statistics (5th ed). Boston, MA: Allyn and Bacon. 2007. 159. Howell D. Statistical Methods for Psychology (6th ed) Belmont, CA: Thomson Wadsworth. 2007. 160. Russell D, Peplau L, Cutrona C. The Revi sed UCLA Loneliness Scale: concurrent and discriminant validity evidence. J Pers Soc Psychol 1980; 39: 472-480. 161. Reynolds W. Adult Suicidal Ideation Ques tionnaire Professional Manual. Odessa, FL: Psychological Assessment Resources. 1991. 162. Reynolds W. Psychometric characteristics of the Adult Suicidal Id eation Questionnaire in college students. J Pers Assess 1991; 56: 289-307. 163. Baron R, Kenny D. The moderator-mediator dis tinction in social ps ychological research: conceptual, strategic, and statistical considerations. J Pers Soc Psychol 1986; 51: 11731182. 164. Community College Survey of Student E ngagement. The College Student Report. 2007, Available at: http://www.ccsse.org/aboutsurv ey/aboutsurvey.cfm. Accessed: 26 February 2008. 165. Fahlman M, Hall H, Lock R. Ethnic and soci oeconomic comparisons of fitness, activity levels, and barriers to exercise in high school females. J Sch Health 2006; 76: 12-17. 166. Felton G, Dowda M, Ward D et al. Differences in physical activity between black and white girls living in rural and urban areas. J Sch Health 2002; 72: 250-255. 167. Kelley G, Kelley K. Physical activity habits of African-American college students. Res Q Exerc Sport 1994; 65: 207-212. 168. Masse L, Anderson C. Ethnic differences among correlates of physical activity in women. Am J Health Promot 2002; 17: 357-360. 169. Marsh H, Kleitman S. Extracurricular school activities: the good, the bad, and the nonlinear. Harv Educ Rev 2002; 72: 464-514.

PAGE 167

167 170. Fredricks J, Eccles J. Is ex tracurricular participation associ ated with beneficial outcomes? concurrent and longitudinal relations. Dev Psychol 2006; 42: 698-713. 171. Taliaferro L, Rienzo B, Miller MD et al. Exploring potential me diating pathways through which participation in sports relates to reduced risk of suicidal ideation. Res Q Exerc Sport in press. 172. National Center for Education Statistics. National Education L ongitudinal Study of 1988 (NELS 88). 1988, Available at: http://www .nces.ed.gov/surveys/nels88/. Accessed: 8 January 2008. 173. Bearman P, Jones J, Udry J. The Nationa l Longitudinal Study of Adolescent Health: research design. 1998, Available at: http: //www.cpc.unc/edu/addhealth. Accessed: 3 October 2007. 174. Buhain V. $2 million budget gap threatens sp orts teams: sports activities for middle schools and high schools in peril. 2008, Available at: http://www.theolympian.com/news/st ory/433108.html. Accessed: 28 April 2008. 175. Hurley R. Physical education, extracurricular sports suffer under budget strains. 2004, Available at: http://www.weac.org/ne ws/2003-04/jun04/phyed.htm. Accessed: 13 October 2008. 176. Netter S. Schools open with fewer teacher s, sports, buses: school boards nationwide forced to make frustrating decisions. 2008, Available at: http://abcnews.go.com/US/Story?id=5789558& page=1. Accessed: 15 September 2008. 177. Staples A. School sports programs fight to stay alive in struggling economy. 2008, Available at: http://www.sportsillustrated.cnn.com. Accessed: 21 July 2008. 178. Taliaferro L, Rienzo B, Miller MD et al. High school youth and su icide risk: exploring protection afforded through physical activity and sport participation. J Sch Health 2008; 78: 545-553. 179. Broh B. Linking extracurricular programmi ng to academic achievement: who benefits and why? Sociol Educ 2002; 75: 69-91.

PAGE 168

168 BIOGRAPHICAL SKETCH Lindsay A. Taliaferro was born in 1976 in Ch icago, Illinois. She grew up in northern Virginia and graduated from Oakton High School in 1994. Lindsay attended the Virginia Polytechnic Institute and State University (Virginia Tech). In 1998, Lindsay earned a Bachelor of Science degree from Virginia Tech in Ps ychology. After graduation, Lindsay enrolled at Florida International University. She earned a Master of Science degree in Sports Management. In 2001, Lindsay enrolled at the University of South Florida (USF). Wh ile pursuing a master's degree at USF, Lindsay worked as a Graduate Research Assistant for the Florida Prevention Research Center. Lindsay earned a Master of Public Health in Health Education from USF in 2003. After graduation, Lindsay worked as a Research Study Coordinator for the Health Outcomes and Behavior Program at the H. Lee Moffitt Cancer Center and Research Institute. Lindsay entered the Ph.D. program in the Depa rtment of Health Education and Behavior at the University of Florida (UF) in 2005. Du ring her first year at UF, Lindsay clarified her research focus. She developed a line of res earch investigating the relationship between participation in sports and reduced risk of suicide among young people. Lindsay's doctoral dissertation includes several scien tific papers based on her resear ch submitted for publication to scholarly journals. Lindsay will be granted a Doctor of Philosophy in Health and Human Performance with an emphasis in Health Behavior in August 2009.