<%BANNER%>

Gender Role Conflict and Resilience in Adolescent and Emerging Adult Males

Permanent Link: http://ufdc.ufl.edu/UFE0021163/00001

Material Information

Title: Gender Role Conflict and Resilience in Adolescent and Emerging Adult Males
Physical Description: 1 online resource (112 p.)
Language: english
Creator: Bates, Stephanie Lynn
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007

Subjects

Subjects / Keywords: adolescent, adult, conflict, emerging, gender, male, resilience, role
Family, Youth and Community Sciences -- Dissertations, Academic -- UF
Genre: Family, Youth and Community Sciences thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Gender identity development can be an arduous process for many adolescents and emerging adults. Both the paths of gender development and the resulting identities of young people can have a great effect on the development and later lives of these individuals. One difficulty youth can face at this stage is gender role conflict, an issue arising from conflict within male youth associated with gender socialization process. The purpose of this study was to examine gender role conflict in adolescent and emerging adult males and how it affects their resilience, or their ability to bounce back from adverse circumstances. Specifically, the study investigated the effects of gender role conflict on male adolescent/emerging adult resilience. Results of the study revealed a complex relationship between male gender role conflict and resilience in this sample. Multiple regression modeling showed that as conflict surrounding the Success, Power and Competition (SPC) pattern increased, these male youth reported more resilience. Conversely, as conflict surrounding the Restrictive Emotionality (RE) pattern increased, male youth sampled reported less resilience. Further study is needed to explore these relationships and their effects in adolescent and emerging adult males more closely. More specifically, research should focus on the gender socialization process, the development of gender role conflict, and its effects on male youth throughout the gender development process.
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 Stephanie Lynn Bates.
Thesis: Thesis (M.S.)--University of Florida, 2007.
Local: Adviser: Barnett, Rosemary V.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2017-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0021163:00001

Permanent Link: http://ufdc.ufl.edu/UFE0021163/00001

Material Information

Title: Gender Role Conflict and Resilience in Adolescent and Emerging Adult Males
Physical Description: 1 online resource (112 p.)
Language: english
Creator: Bates, Stephanie Lynn
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007

Subjects

Subjects / Keywords: adolescent, adult, conflict, emerging, gender, male, resilience, role
Family, Youth and Community Sciences -- Dissertations, Academic -- UF
Genre: Family, Youth and Community Sciences thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Gender identity development can be an arduous process for many adolescents and emerging adults. Both the paths of gender development and the resulting identities of young people can have a great effect on the development and later lives of these individuals. One difficulty youth can face at this stage is gender role conflict, an issue arising from conflict within male youth associated with gender socialization process. The purpose of this study was to examine gender role conflict in adolescent and emerging adult males and how it affects their resilience, or their ability to bounce back from adverse circumstances. Specifically, the study investigated the effects of gender role conflict on male adolescent/emerging adult resilience. Results of the study revealed a complex relationship between male gender role conflict and resilience in this sample. Multiple regression modeling showed that as conflict surrounding the Success, Power and Competition (SPC) pattern increased, these male youth reported more resilience. Conversely, as conflict surrounding the Restrictive Emotionality (RE) pattern increased, male youth sampled reported less resilience. Further study is needed to explore these relationships and their effects in adolescent and emerging adult males more closely. More specifically, research should focus on the gender socialization process, the development of gender role conflict, and its effects on male youth throughout the gender development process.
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 Stephanie Lynn Bates.
Thesis: Thesis (M.S.)--University of Florida, 2007.
Local: Adviser: Barnett, Rosemary V.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2017-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0021163:00001


This item has the following downloads:


Full Text

PAGE 1

1 GENDER ROLE CONFLICT AND RESILI ENCE IN ADOLESCENT AND EMERGING ADULT MALES By STEPHANIE LYNN BATES A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2007

PAGE 2

2 Stephanie Lynn Bates

PAGE 3

3 To those whose lives have been negatively a ffected by a rigid interpretation of gender.

PAGE 4

4 ACKNOWLEDGMENTS First, I would like to thank my supervisory committee, with out whom I could never have completed this undertaking. Special thanks go to Dr. Rosemary V. Barnett, my supervisory committee chair, confidant and colleague who has taught, mentored, and supported me throughout this process. I also must thank Dr. Mark Brennan, who endured my incessant questions both about analysis and about Ireland. Finally, Dr. Gl enn Israel has proven to be a most valuable resource as a scholar and teacher, providing insight and guidance into my research and education. I must also thank my entire supervisory committee for assisting me along the way, even when it was far from easy. On a more personal note, my family, friends and loved ones have heard more about this undertaking than they bargained for and deserv e commendation for their assistance, time and tolerance. They provided motivation, support, an d much-needed breaks from this thesis. I could not have done this without them.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........8 LIST OF FIGURES................................................................................................................ .......11 LIST OF TERMS.................................................................................................................. .........12 ABSTRACT....................................................................................................................... ............14 CHAPTER 1 INTRODUCTION..................................................................................................................15 Background..................................................................................................................... ........15 Rationale for Study............................................................................................................ .....16 Purpose of Study............................................................................................................... ......17 Research Questions.........................................................................................................17 Hypotheses..................................................................................................................... .18 Significance of Study.......................................................................................................... ....19 2 LITERATURE REVIEW.......................................................................................................21 Introduction to Gender Role Conflict.....................................................................................21 Gender Role Conflict: History.........................................................................................22 Effects of Gender Role Conflict: Initial Findings...........................................................25 Gender Role Conflict: Contemporary Research..............................................................26 Gender Identity Disorder....................................................................................................... .28 Notes on Sexual Minority Populations...................................................................................29 Theoretical Approach........................................................................................................... ..30 The 40 Developmental Assets.........................................................................................31 Resilience Theory............................................................................................................33 Conclusion..................................................................................................................... .........35 3 MATERIALS AND METHODS...........................................................................................36 Purpose of the Study........................................................................................................... ....36 Research Design................................................................................................................ .....36 Population and Sampling Procedures.....................................................................................36 Data Collection................................................................................................................ .......40 Instrumentation................................................................................................................ .......41 Developmental Assets Profile (DAP)..............................................................................41 Validating the Developmental Assets Profile (DAP)......................................................43 Gender Role Conflict Scale (GRCS)...............................................................................43

PAGE 6

6 Sexuality Rating..............................................................................................................44 Demographic Questions..................................................................................................45 Data Analysis.................................................................................................................. ........46 Limitations and Delimitations................................................................................................47 4 RESULTS........................................................................................................................ .......49 Sample Statistics.............................................................................................................. .......49 Summary Statistics: Demographics.................................................................................49 Summary Statistics: GRCS (Independent Variable).......................................................51 Summary Statistics: DAP (Dependent Variable)............................................................52 Bivariate Analyses............................................................................................................. .....53 Demographics by Resilience...........................................................................................53 Demographics by Independent Variables........................................................................56 Independent Variables by Resilience..............................................................................59 Multivariate Analysis.......................................................................................................... ....61 Summary........................................................................................................................ .........65 5 DISCUSSION..................................................................................................................... ....66 Revisiting Research Questions...............................................................................................67 RQ1: Is there a relationship between Gender Role Conflict (GRC) and Overall Resilience in Adolescent/Emerging Adult Males?......................................................67 RQ2: Is there a relationship between the Success, Power and Competition (SPC) pattern of Gender Role Conflict and Overall Resilience in Adolescent/Emerging Adult Males?................................................................................................................68 RQ3: Is there a relationship between the Restrictive Emotionality (RE) pattern of Gender Role Conflict and Overall Resilience in Adolescent/Emerging Adult Males?......................................................................................................................... .69 RQ4: Is there a relationship between the Restrictive Affectionate Behavior Between Men (RABBM) pattern of Gender Role Conflict and Overall Resilience in Adolescent/Emerging Adult Males?............................................................................70 RQ5: Is there a relationship between the Conflict Between Work and Family Relations (CBWFR) pattern of Gender Role Conflict and Overall Resilience in Adolescent/Emerging Adult Males?............................................................................71 RQ6: Is there a relationship between Demographic Variables and Overall Resilience in Adolescent/Emerging Adult Males?......................................................73 Other Findings.................................................................................................................77 Implications of Research....................................................................................................... .79 Gender Role Conflict Theory..........................................................................................80 Resilience Theory............................................................................................................83 Implications for Future Resear ch, Programs and Public Policy.............................................85 Future Research...............................................................................................................85 Practitioner/Programmatic Suggestions..........................................................................86 Policy Recommendations................................................................................................88 Summary........................................................................................................................ .........90

PAGE 7

7 APPENDIX A INSTRUMENTATION..........................................................................................................92 Explanations................................................................................................................... ........92 Informed Consent Letter..................................................................................................92 Developmental Asset Profile (DAP)...............................................................................92 Gender Role Conflict Scale (GRCS)...............................................................................92 Demographic Questions..................................................................................................93 B RESULTS TABLES...............................................................................................................98 LIST OF REFERENCES.............................................................................................................108 BIOGRAPHICAL SKETCH.......................................................................................................112

PAGE 8

8 LIST OF TABLES Table page 3-1 Developmental Assets Profile Extern al Asset Categories and Item Numbers..................42 3-2 Developmental Assets Profile Intern al Asset Categories and Item Numbers....................42 3-3 Internal Consistency Values for the Developmental Assets Profile..................................43 3-4 Gender Role Conflict Scale Factors and Item Numbers....................................................44 3-5 Gender Role Conflic t Scale Statistics................................................................................44 3-6 Internal Consistency Reliability Values for the GRCS......................................................44 4-1 Summary statistics for race variables................................................................................50 4-2 Age and Sexual Orientation summary statistics................................................................50 4-3 GRCS summary statistics..................................................................................................51 4-4 DAP summary statistics.....................................................................................................53 4-5 Pearsons Correlation r values relating Demographics and External Resilience Patterns and overall External Asset score..........................................................................54 4-6 Pearsons Correlation r values relating Demographi cs and Internal Resilience Patterns and overall Internal Asset Score..........................................................................54 4-7 Effect of academic status on mean DAP score..................................................................55 4-8 Effect of Race on mean DAP score...................................................................................56 4-9 Pearsons r values correlating Demo graphics and Gender Role Conflict..........................57 4-10 Effect of Academic Status on mean SPC pattern score of gender role conflict................58 4-11 Effect of Race on mean pattern a nd total gender role conflict scores...............................58 4-12 Effect of sexual orientation on mean pa ttern and total gender role conflict scores...........58 4-13 Pearsons r values correlating Gender Role Conf lict and External resilience factors and External Asset score....................................................................................................59 4-14 Pearsons r values correlating Gender Role Conflic t and Internal resilience factors, Internal Asset score and overall DAP score......................................................................59 4-15 Significant relationships between RE quartile and resi lience pattern/overall means........60

PAGE 9

9 4-16 Significant relationships between RABBM quartile and resilience pattern means...........60 4-17 Significant relationships between CBWFR quartile and resilience pattern means............61 4-18 Significant relationships be tween GRCS score quartile a nd resilience pattern means......61 4-19 Standardized regression co efficients for demographic and GRCS pattern effect on DAP score...................................................................................................................... ....63 4-20 Standardized regression co efficients for demographic and GRCS pattern effect on External Asset score...........................................................................................................64 4-21 Standardized regression co efficients for demographic and GRCS pattern affect on Internal Asset score........................................................................................................... .64 B-1 DAP summary statistics including context areas...............................................................98 B-2 Pearsons r values for DAP factors, Internal/Ext ernal asset scores, and total score.........98 B-3 Pearsons r values for GRCS factors and total score.........................................................98 B-4 Pearsons r values relating demographics and external resilience.....................................99 B-5 Pearsons r values relating demographics and internal resilience.....................................99 B-6 Pearsons r values correlating demographi cs and Gender Role Conflict..........................99 B-7 Pearsons r values correlating Gender Role Co nflict and external resilience...................99 B-8 Pearsons r values correlating Gender Role Conflic t with internal asset factors and overall resilience............................................................................................................. .100 B-9 Significant ANOVA relationships between academic status and mean resilience..........100 B-10 Effect of race on mean resi lience scores shown by ANOVA..........................................100 B-11 Significant ANOVA relations hips between GRC pattern mean scores, total mean scores and age................................................................................................................. .100 B-12 Significant ANOVA relations hips between Academic Status and SPC score of gender role conflict..........................................................................................................100 B-13 Significant ANOVA relationships between race and pattern/to tal gender role conflict..100 B-14 Effect of sexual orientation on mean pa ttern and total gender role conflict scores shown by ANOVA...........................................................................................................101 B-15 Significant ANOVA relationshi ps between RE quartile a nd resilience pattern/overall means.......................................................................................................................... .....101

PAGE 10

10 B-16 Significant ANOVA relations hips between RABBM quartil e and resilience pattern means.......................................................................................................................... .....101 B-17 Significant ANOVA relations hips between CBWFR quartil e and resilience pattern means.......................................................................................................................... .....101 B-18 Significant ANOVA relations hips between GRCS scor e quartile and resilience pattern means.................................................................................................................. .101 B-19 Standardized regression co efficients for demographic and GRCS pattern effect on DAP score with age.........................................................................................................102 B-20 Standardized regression co efficients for demographic and GRCS pattern effect on DAP score without age....................................................................................................102 B-21 Standardized regression co efficients for demographic and GRCS pattern effect on External Asset score with age..........................................................................................103 B-22 Standardized regression co efficients for demographic and GRCS pattern effect on External Asset score without age.....................................................................................103 B-23 Standardized regression co efficients for demographic and GRCS pattern effect on Internal Asset score with age...........................................................................................104 B-24 Standardized regression co efficients for demographic and GRCS pattern effect on Internal Asset score without age......................................................................................104 B-25 p-values for demographic and GRCS pattern effect on DAP score with age regression..................................................................................................................... ....105 B-26 p-values for demographic and GRCS pattern effect on DAP score without age regression..................................................................................................................... ....105 B-27 p-values for demographic and GRCS pattern effect on External Asset score with age regression..................................................................................................................... ....106 B-28 p-values for demographic and GRCS patte rn effect on External Asset score without age regression................................................................................................................. ..106 B-29 p-values for demographic and GRCS pattern effect on Internal Asset score with age regression..................................................................................................................... ....107 B-30 p-values for demographic and GRCS patte rn effect on Internal Asset score without age regression................................................................................................................. ..107

PAGE 11

11 LIST OF FIGURES Figure page 2-1 Gender Role Conflict Model..............................................................................................25 2-2 Resiliency Model........................................................................................................... ....34 3-1 Sample Size Formula........................................................................................................ .39 3-2 Kinsey Sexuality Scale..................................................................................................... .45 3-3 Data Analysis Model........................................................................................................ ..46 A-1 Informed Consent Letter....................................................................................................93 A-2 Developmental Asset Profile, Page 1.................................................................................94 A-3 Developmental Asset Profile, Page 2.................................................................................95 A-4 Gender Role Conflict Scale...............................................................................................96 A-5 Demographics Questions...................................................................................................97

PAGE 12

12 LIST OF TERMS Adolescence for the purposes of this study, defi ned as the period of development occurring in youth from ages 10 to 18. Developmental Assets according to the S earch Institute concrete, common sense, positive experiences and qualities essential to raising successful young people. These assets have the power during critical adol escent years to influence choices young people make and help them become caring, re sponsible adults, (http://www.searchinstitute.org/assets/, n.d.). Emerging adults for the purposes of this study, youth between ages 19 and 24, who are, according to Arnett, exploring various possibilities in love and work [and developing] a more definite identity an understanding of who they are, what their capabilities and limitations are, what their beliefs and values are, and how they fit into the society around them (2007). While Arnett extends this age range to 25, this study limits this to the traditional undergraduate populations. As nontra ditional student status begins at age 25, 24 was used as the maximum age in this study. Gender conformist individual who does follow soci ally dictated roles prescribed to their sex (i.e. masculine males and feminine females). Gender nonconformist individual who does not follo w socially dictated roles prescribed to their sex (i.e. female tomboys or male sissies). For purposes of this study, transgender individuals are not included in this category. Gender role conflict (GRC) according to ONeil, the author of the Gender Role Conflict Scale, GRC is a psychological state in which socialized gender roles have negative consequences on the person or others. Gender ro le conflict occurs when rigid, sexist, or restrictive gender roles result in personal restrictions, devalua tion, or violation of others or self (ONeil et al., 1995). Heterosexual adolescent adolesce nt romantically and/or sexua lly attracted to a member of the opposite sex. Intersex according to the Intersex Society of North America, is a general term used for a variety of conditions in which a person is born with a reproduc tive or sexual anatomy that doesnt seem to fit the typical definitions of female or male (http://www.isna.org/faq/what_i s_intersex, n.d.). This occurs when genetic or anatomical ambiguity exists, such as having primary sex characteristics of one sex and secondary of the opposite sex, or a mixture of both. Resilience ability to bounce back from advers e situations. According to Richardson, a force within everyone that driv es them to seek self-actua lization, altruism, wisdom, and harmony with a spiritual source of strength (2002).

PAGE 13

13 Sexual minority adolescent Lesbian, Gay, or Bise xual (LGB) adolescent, attracted to either members of their own sex (L G) or both sexes (B). Suicidality according to Kalafat, refers to all suicide-related behaviors and thoughts, including completing or attempting suic ide, and suicidal ideation (232). Transsexual/Transgender according to the Intersex Society of North America, are usually people who are born with typical male or fe male anatomies but feel as though theyve been born into the wrong body, (http://www.isna.org/fa q/transgender, n.d.).

PAGE 14

14 Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science GENDER ROLE CONFLICT AND RESILI ENCE IN ADOLESCENT AND EMERGING ADULT MALES By Stephanie Bates August 2007 Chair: Rosemary V. Barnett Major: Family, Youth and Community Sciences Gender identity development can be an arduous process for many adolescents and emerging adults. Both the paths of gender de velopment and the result ing identities of young people can have a great effect on the developmen t and later lives of th ese individuals. One difficulty youth can face at this stage is gender role conflict, an issue aris ing from conflict within male youth associated with gender socializati on process. The purpose of this study was to examine gender role conflict in adolescent and emerging adult males and how it affects their resilience, or their ability to bounce back fr om adverse circumstances. Specifically, the study investigated the effects of gende r role conflict on male adolescen t/emerging adult resilience. Results of the study revealed a complex relatio nship between male gender role conflict and resilience in this sample. Multiple regression modeling showed that as conflict surrounding the Success, Power and Competition (SPC) pattern increased, these male youth reported more resilience. Conversely, as c onflict surrounding the Restrictiv e Emotionality (RE) pattern increased, male youth sampled reported less resilien ce. Further study is needed to explore these relationships and their effects in adolescent and emerging adult males more closely. More specifically, research should focus on the gende r socialization process, the development of gender role conflict, and its effects on male youth throughout the gender development process.

PAGE 15

15 CHAPTER 1 INTRODUCTION Background Gender identity formation is one of the many tasks associated with the developmental periods of adolescence and emerging adulthood, and is not without its risks and abnormalities. While a majority of adolescents experience norma tive development in this area, that is, they identify with a gender that matches their sex and the social roles associated with it, there are several atypical paths of gender development. One such pa ttern is gender role conflict, a psychological state in which socialized gender roles have negative cons equences on the person or others [that] occurs when rigid, sexist, or restrictive gender roles result in personal restrictions, devaluation, or violation of others or se lf (ONeil, 1995, p. 165-166). Adolescents and emerging adults face physical emotional, hormonal, psychological, and social changes on a daily basis, which can translate to increased vulnerability to a great many stressors among these youth (Pelkonen & Marttu nen, 2003). Additionally, they may experience problems associated with gender role conflict, such as more anxiety, anger, and alcohol use (Blazina & Watkins, 1996), para noia, interpersonal se nsitivity, and depre ssion (Good, Robertson, & Fitzgerald, 1996) above and beyond these base line stressors. As evident in several developmental models, adolescents and emerging a dults need support on many levels in order to overcome the challenges of this period (Bronf enbrenner, 2000; Richardson, 2002, e.g.). Despite the myriad stressors plaguing this population, with support fro m family, friends, peers, and communities, adolescents and emerging adults can overcome many obstacles, displaying resilience in a variety of problem s and situations (Scales, 1999).

PAGE 16

16 Rationale for Study Some potential negative e ffects of non-normative development in adolescence and emerging adulthood include substance use/abuse, risky sexual behavior, and suicide, a problematic behavior that has b een studied extensively among this age group. In fact, the Youth Risk Behavior Surveillance System indicates that suicide is one of the f our leading causes of death among youth ages 10-24 (2005), w ith 9% of Florida deaths in th is age range due to suicide. It is therefore important to iden tify those individuals or groups mo st at risk for suicide, what causes this risk, and how to best minimize th is vulnerability among a dolescents and emerging adults. Some adolescent groups often pres ent rates of problematic beha vior such as suicide higher than that of adolescents in gene ral. Researchers indicate that adolescent sexual minority suicide rates are several times the rate of their heterosexual counterpa rts (Judge, 2004; Kulkin, Chauvin, & Percle, 2000; Remafedi, 1998). Males seem to be more at risk within this population, with bi/homosexual males being 3.5 times more likely to attempt suicide than bi/homosexual females (Remafedi, 1998). Judge notes that the reason for this heightened risk is not sexual orientation alone (2004), while DAugelli et al. report findi ngs of suicide attempts related to sexual orientation (2005). Exploration of the reasons behind this groups increased propensity for suicide is needed. A more recently developed field may shed additional light on this topic. Gender nonconforming adolescents, or those that do not adhe re to roles ascribed by society for their sex, may also be at higher risk for problematic be haviors (Watts & Border s, 2005). This population can be taunted and ostracized in much the same way as the sexual minority group, even by parents, for their gender-atypical behavior (DAugelli et al., 2005) In fact, patterns of gender role conflict are associated with decreased psychological well-being and increased anxiety

PAGE 17

17 (Blazina & Watkins, 1996) as well as depres sion (Good, Robertson, & Fi tzgerald, 1996). While sexual minority adolescents and gender role conf licted adolescents are independent groups, they do overlap, and thus may share some of the increased risk that comes from societal reactions to their respective status. While sexua l minority adolescents may be more at risk if they experience gender role conflict, they may also be less likely to experience this type of conflict as they may have faced similar issues related to their sexuality exploration and disclosure. These are just two groups at higher risk; however, the study of gr oups like these with increased vulnerability will lead to further understanding of risk factors for suicidality and other problematic behaviors in adolescence and emerging adulthood. Purpose of Study The purpose of this study is to examine gende r role conflict in adolescents and emerging adults and how it affects their resilience, or their ability to bounce back from adverse circumstances. Specifically, the study will investigate the effect of the different patterns of gender role conflict on male adolescent/emerging adu lt resilience. The results of this study will further develop the understanding of both gender role conflict in males and resilience in this population. Additionally, the study may elucidate possible new linkages in suicide (i.e. gender role conflict) research in lieu of formerly hypothesized, often di sproved research relationships (i.e. sexual minority status itself). Research Questions RQ 1 Is there a relationship between gender role conflict (GRC) and overall resilience in adolescent/emerging adult males? RQ 2 Is there a relationship between the Success, Power, and Competition (SPC) pattern of gender role conflict and overall resilien ce in adolescent/emerging adult males? RQ 3 Is there a relationship between the Restrictiv e Emotionality (RE) pattern of gender role conflict and overall resilience in adolescent/emerging adult males?

PAGE 18

18 RQ 4 Is there a relationship between the Restri ctive Affectionate Behavior Between Men (RABBM) pattern of gender ro le conflict and overall resilience in adolescent/emerging adult males? RQ 5 Is there a relationship between the Conflict Between Work and Family Relations (CBWFR) pattern of gender ro le conflict and overall resilience in adolescent/emerging adult males? RQ 6 Is there a relationship between demogra phic variables and overall resilience in adolescent/emerging adult males? Hypotheses Ho 1 Gender role conflict does not a ffect overall resilience in adolescent/emerging adult males. Ha 1 The more gender role conflict that adolescent/e merging adult males experience, the less overall resilience they will report. Ho 2 The Success, Power, and Competition (SPC) pattern of gender role conflict does not affect overall resilience in a dolescent/emerging adult males. Ha 2 The more gender role conflict with Success, Power, and Competition that adolescent/emerging adult males experience, th e less overall resilience they will report. Ho 3 The Restrictive Emotionality (RE) pattern of gender role conflict does not affect overall resilience in adolescent/emerging adult males. Ha 3 The more gender role conflict with Restrict ive Emotionality that adolescent/emerging adult males experience, the less ove rall resilience they will report. Ho 4 The Restrictive Affectionate Behavior Be tween Men (RABBM) pattern of gender role conflict does not affect overall resilien ce in adolescent/emerging adult males. Ha 4 The more gender role conflict with Restrict ive Affectionate Behavior Between Men that adolescent/emerging adult males experience, th e less overall resilience they will report. Ho 5 The Conflict Between Work and Family Re lations (CBWFR) patte rn of gender role conflict does not affect overall resilien ce in adolescent/emerging adult males. Ha 5 The more gender role conflict with Conflict Between Work and Family Relations that adolescent/emerging adult males experience, th e less overall resilience they will report. Ho 6 There is no relationship between demogra phic variables and overall resilience in adolescent/emerging adult males. Ha 6 There is a relationship between the demogr aphic variables and overall resilience in adolescent/emerging adult males.

PAGE 19

19 Significance of Study Suicide research on sexual minor ity adolescents has come to a roadblock: researchers have found that sexual minority adolescents consistently have higher rates of suicide than their heterosexual counterparts (Kulkin, Chauvin & Percle, 2000; Remafedi, 1998); however, studies have also suggested that sexual mi nority status is not itself a risk factor for suicide (Judge, 2004). There must be another factor common to gay adol escents that puts them at higher risk for these problematic behaviors. This study hypothesizes that gender role conflic t could be the missing factor in this model, applyi ng not only to homosexual adolescents/emerging adult males, but heterosexual males in this age group experi encing gender role conflict as well. Research on adolescents/emerging adults and both suicide and other problematic behaviors has been largely problem-based. Through applica tion of an asset-based approach, perhaps this study will add to the body of resilience literature as well as to practit ioners intervention and prevention programs. While myriad risk factors for problematic behaviors have been identified, compiling an exhaustive list of th ese factors is almost assuredly impossible. Identification of protective factors or assets th at significantly decrease propen sity for these behaviors, though, could provide similarly valuable information more effectively. This study aims to use this assetbased approach to examine gender role conflict and its effects in adolescents/emerging adults. Watts and Borders have recently applied the ge nder role conflict paradigm to adolescent boys, finding in their study of eleven males that th ese conflicts with societ ally prescribed gender roles did exist, especially in older boys ( 2005). While only a small sample, this research supports the idea that gender role conflict does affect adolescent boys. Gender role conflict in men generally has been associated with negativ e outcomes such as more anxiety, anger, and alcohol use (Blazina & Watkin s, 1996), paranoia, interperso nal sensitivity, and depression (Good, Robertson, & Fitzgerald, 1996) Researchers at the Search Institute have shown that

PAGE 20

20 problems like these are related to a lack of support structures inte rnally and externally in their developmental asset framework and that the more vulnerable youth are, the more they seem to benefit from the protective impact of developm ental assets (Scales, 1999, p. 118). By relating these concepts, this study hopes to create a more structured paradigm through which gender role conflict in different groups of adoles cent/emerging adult males can be studied. This study, then, aims to link gender role conflict to resilience as measured by the developmental assets. Through this model, the research will focus on which patterns of gender role conflict most affect resi lience in adolescent/emerging adu lt males. This has important implications for researchers and practitioners alike, but as a preliminary study, will require further inquiry to establ ish further connections.

PAGE 21

21 CHAPTER 2 LITERATURE REVIEW Introduction to Gender Role Conflict Gender identity development can be an arduous process for many adolescents and emerging adults. Both the paths of gender deve lopment and the resulting identities of these young people can have a great effect on the develo pment and later lives of these individuals. One of the difficulties youth can face at this stag e is gender role conflict, which occurs when rigid, sexist, or restrictive gende r roles result in personal restrict ion, devaluation, or violation of others or the self (ONeil, Good, & Holmes, 1995, p. 166). This chapter presents an overview of gender role conflict, detai ling its history in research, pa st and current implications, and theoretical basis for exploration. Although much of gender identity development occurs during adolesce nce (Arnett, 2007), this is a lifelong process that simply intens ifies during this time. This period of gender development, then, is an appropriate one in which to study gender role conflict; however, as researchers have noted, there is growing attention to the trans ition of youth to young adulthood (Galambos & Leadbeater, 2000, p. 289). Arnett labe led this transitional period of development emerging adulthood and defined it as being from approximately 18 to 25 years of age (Arnett, 2007). In this study, those individuals 25 and olde r were excluded from this study due to their nontraditional status as colleg e students; thus, this for this study, only those aged 18 to 24 were included. This study, then, hopes to focus on gender role conflict in individuals on the adolescent/emerging adult divide, a compromi se between the initial stirrings of gender development in early adolescence with the typi cally more established gender roles of later emerging adulthood.

PAGE 22

22 Gender Role Conflict: History Garnets and Pleck researched the precursor to gender role conflict, sex role strain, or discrepancies between an individuals perception of her or his personal characteristics and her or his standards for herself or himself derivi ng from sex role norms (1979, p. 274-275). Their model supposed that societal norms of sex ro les were impressed upon individuals, that these individuals could to varying de grees accept/conform or reject/resist these norms, and that sex role strain heightened as individuals perceptions deviated from these norms, eliciting resistance (1979). ONeil found a flaw in this model, namely that it did not specify the precise patterns of sex role conflicts that occur when discrepancie s exist between a person s real and ideal selfconcept (1995, p. 169). In order to empirically research this se x role strain, ONeil sought to operationalize these specific patterns in men. Th e result was gender role conflict, originally called sex role conflict, but changed due to Ungers sex/gender distinction (1979). Gender role conflict was put into the cont ext of The Masculine Mystique and Value System, which was comprised of the negative as pects of the masculine stereotype (O'Neil, 1981a, 1981b, 1982) and "was defined as a complex set of values and beliefs that define optimal masculinity learned during early socialization [and] based on rigid stereotypes and beliefs about men, masculinity, and femininity," (O'N eil, 1995. p. 171). According to ONeil, this contains six patterns of gender role conflict "that might emer ge when men experience gender roles in contradictory or incons istent ways," (1995, p. 171): restrictive emotionality; socialized control, power, and competition; homophobia; re strictive sexual and affectionate behavior; obsession with achievement and succe ss; and health care problems. After testing this model of gender role conflic t, 36% of the variance was explained by just four of the factors (ONe il, 1995). Through a series of tests a nd retests, four of the original six

PAGE 23

23 patterns were identified as most accurately assess ing gender role conflict as follows (taken from ONeil, 1995, p. 175): Success, Power, and Competition (SPC) o Success: Persistent worries about pers onal achievement, competence, failure, status, upward mobility, and wealth, and career success. o Power: Obtaining authority, dominan ce, or ascendancy over others. o Competition: Striving against others to gain something or the comparison of self with others to establish ones s uperiority in a given situation. Those who experience this t ype of gender role conflict internalize this drive for achievement. One example may be a man who is very concerned with career prestige and upward mobility, thus at risk fo r devaluing himself or others in the face of adversity in this process. Restrictive Emotionality (RE): Having di fficulty and fears about expressing ones feelings and difficulty finding wo rds to express basic emotions. An individual experiencing th is type of gender role conf lict may have anxiety about expressing emotion or may associat e disclosure of personal feeli ngs as less masculine, which may result in negative ideas about himself or others or a further lack of disclosure of emotional information. Restrictive Affectionate Be havior Between Men (RABBM): Having limited ways to express ones feelings and thoughts with othe r men and difficulty touching other men. Men who experience this pattern of gender role conflict may not feel they have as many channels to express feelings of camaraderie, support, and the like through physical contact with other men. This restricted behavior may resu lt in less fulfilling relationships or increased conflict regarding this physical affection, which may result in problematic outcomes. Conflict Between Work and Family Rela tions (CBWFR): Experiencing difficulties balancing work-school and family relations, re sulting in health problems, overwork, stress, and a lack of leisure and relaxation.

PAGE 24

24 Those individuals experiencing gender role c onflict with this patte rn may feel extreme stress in an attempt to balance work or school an d family. This form of conflict may exist to a lesser degree among college males, as many do not work; however, school has been associated with work in this model. Moreover, many student s attend school and work at the same time. As this is a preliminary study, it will not assess these specific consider ations. Future research might explore these connections further. The patterns of gender role conflict, then, ar e not necessarily positive or negative. The basic premise of the theory is that there are so cially prescribed gender roles for men (whether good or bad) and that conflict as a result of these ge nder roles has been shown, as described above, to be correlated with negative outcomes. This may result from a rigid interpretation of gender internally or social sanctioning processe s associated with nonconformity to these gender roles; however, the effects in many areas have b een shown, highlighting th e pervasive nature of gender role conflict in varied facets of the male life experience. Figure 2-1 graphically represents the interre latedness of these pa tterns of gender role conflict. These patterns then became the basis for study of gender role conflict in males. Some of these studies showed that college-aged men (1722) had significantly mo re gender role conflict than later ages (Mendels on, 1988); that the restrictive affecti onate behavior between men pattern of gender role conflict was higher for the 17-22 year old age group (Men delson, 1988); and that this group experienced more c onflict with the patte rn of success, power, and competition (Cournoyer, 1994). ONeil also chro nicles the use of this paradigm in assessing gender role conflict in fraternity men, men of different races, classes, a nd ethnicities, a nd nationalities (1995).

PAGE 25

25 Figure 2-1. Gender Role Conflic t Model. (ONeil, Good, & Holm es. (1995). Fifteen years of theory & research on mens gender role conflict: new paradigms for empirical research. In R. Levant & W. Poll ack (Eds.) A New Psychology of Men New York: Basic Books. Reproduced with permission. Effects of Gender Role Conflict: Initial Findings Correlates of gender role conflict have also be en explored across vari ed populations. For college-aged men specifically, gender role conflict, specifically c onflict with three patterns of gender role conflict (restrictive emotionality, restrictive affect ionate behavior between men, conflict between work and family relations), has been associated with low self-esteem (Davis, 1987; Sharpe & Heppner, 1991). Also in this population, gender role conflict (specifically restrictive emotionality, and rest rictive affectionate behavior between men in most but not all

PAGE 26

26 cases) was negatively correlated with intim acy (Sharpe & Heppner, 1991; Cournoyer, 1994; Rounds, 1994). Anxiety and depression were also re lated to patterns of ge nder role conflict in the college student population (Davis, 1987 (anxi ety only); Sharpe & Heppner, 1991; Cournoyer, 1994; Good & Mintz, 1990 (depression only)). Similar negative outcomes were seen when relating gender role conflict to personal strain, stress, and problem-solving and attitudes toward help-seeking (ONeil, 1995). This paradigm ha s also been used to study marital/relationship satisfaction (Cournoyer, 1994; Sharpe & Heppne r, 1991; Sharpe, 1993; Campbell and Snow, 1992), contraceptive use (Berlin, 1988), and ne gative attitudes toward homosexuals (Rounds, 1994) and women (Kaplan, 1992; ONeil & Owen, 1993) Overall, for college-age men, gender role conflict has been established through resear ch to be related to low self-esteem, lower intimacy, anxiety, depression, relationship dissatis faction, sexual aggression and hostility toward women and negative attitudes toward homosexuals (ONeil, 1995). Any single factor or combination of the f actors discussed above could lead to poor developmental outcomes for these men. It is important, then, to study not only these problems but also the ability of colleg e-aged men to be resilient in the face of these problems. Contemporary research has used this gender role c onflict paradigm to better understand many and varied issues related to me n, including gender role conflict in adolescent and college-aged (emerging adult) males. Gender Role Conflict: Contemporary Research Gender role conflict can have a significant impact on the deve lopment and vulnerability of adolescents and emerging adults. Watts and Bo rders explore this issue through research on adolescent males (2005). They found that gend er role conflict affected many of these adolescents, resulting in decreased inter-male a ffection, limited emotionality, and an increase in school/family/friend conflict. They also note th at gender role conflict in Western cultures can

PAGE 27

27 mean rejection of anything effeminate or ga y, and suggest that th is homophobic attitude can negatively affect adolescent males. These issu es as well as those di scussed above such as depression, anxiety, and low self-esteem, can a nd almost certainly do play a part in the tumultuous and often dangerous developmental pe riods of change in adolescence and emerging adulthood. Watts and Borders also suggest an asset-base d approach to furthe r examine gender role conflict (2005). It s eems that this framework may provide fresh insight without exploring various risk factors with little in the way of new results; rath er, the asset-based approach may elucidate overarching pr otective factors that may counteract myriad risk factors in this population. Using the theory of re silience, asset identification in these populations may assist in understanding how gender identity development, specifically gender role conflict, affects adolescents and emerging adults as well as what may protect against any vul nerability associated with these issues. Another contemporary study examined males self-reported Gender Ro le Conflict as well as the Gender Role Conflict th ey perceived their peers had (B eatty et al., 2006). Initially, respondents overestimated peer Gender Role Conflic t. An intervention was held for a portion of the sample in which Gender Role Conflict was di scussed in small groups. Post-tests revealed that respondents from the intervention not only reported more accurate gender role conflict for peers but also self-reported lower levels of Rest rictive Emotionality than those participants who did not receive the intervention. Clearly, thes e results suggest that social norms can and probably do play a role in gender role conflict, and that through informational interventions that offer opportunity for discussion, at least for Re strictive Emotionality, gender role conflict experienced by college-age males can be lessened.

PAGE 28

28 Gender Identity Disorder A more severe classification on the spectrum of gender role conflic t is gender identity disorder. The DSM-IV presents two essential fa ctors in diagnosing this disorder: persistent cross-gender identification and evidence of disc omfort about ones sex (Bower, 2001, p. 2). Furthermore, two other criteria come into play: the absence of a concurrent physical intersex condition, and significant distress or impairment in social, occupational, or other areas of functioning, (Bower, 2001, p. 2). These secondary criteria are not necessary for diagnosis and are not present in many cases of gender identity disorder but can and often are present in those individuals diagnosed with the di sorder, also called transsexuals. Transsexuals can have many problems in vari ous phases of their development. Bower details the social isolation commonly seen in males, particularly during adolescence and the frequency of an anxiety-depression syndrome or a personality disorder, although this is less common in girls, as the tomboy female is mo re readily accepted by peers than her male counterpart (2001, p. 4). Development of this disorder is not straightforward, with both individuals that cross-dress as children and ultimately seek sex reassignment surgery and individuals in which the course of transsexual development is interrupted by a phase in early adulthood characterized by a retu rn of gender identity to th e anatomically correct one, accompanied by exaggeratedly male, almost macho behaviour and subsequent return to male gender roles (Bower, 2001, p. 7). The individuals affected by gender identity di sorder represent extreme cases within the gender role socialization paradigm Although there is no diagnostic test designed to specifically identify gender identity disorder, the prevalen ce of this problem may be ascertained from the 1/3,000 males and 1/100,000 females that seek se x reassignment (Bower, 2001). Overall, the concept of gender identity disorder, a relativ ely new diagnosis, is considered somewhat

PAGE 29

29 ambiguous in its diagnostic criteria (Bower, 2001). Thus, those affected by this disorder may experience gender role conflict di fferently than others and may best be studied separately. Notes on Sexual Minority Populations Gender role conflict can occur simultaneously w ith sexuality development, and thus can be seen as an issue in the lesbia n, gay, bisexual, and transgender community as well as for those individuals that are questioning their sexuality. As DAugelli notes, this period of development is a time of gender intensifi cation, when socialization pressures from families and peers encourage the adoption of traditi onal sex-role related behavior, and divergence from traditional behaviors is less tolerated as youth move through adolescence, (2005). This conflict is not exclusive to these populations, but may be experi enced differently by sexual minority individuals than by heterosexual adolescents/emerging adults. One major problem identified in sexual minor ity research is sampling. As this population relative to the at-large populat ion is small, massive random samples or purposive samples would be needed to obtain sufficient sample sizes for analysis. Time, budgetary, and other common constraints often present problems for the fo rmer, while the latter methodology has come under fire for its lack of external validity. Sampling from couns eling centers, support groups, or community organizations yields important pictur es of segments of this population, but is probably not an accurate portrayal of the larger scope of the sexual mi nority community. For this preliminary study relating gend er role conflict and resilience, it was hoped but not expected that a random sample of college men would yiel d sufficient samples of self-identified gay and heterosexual men; however, this was not the focus of the study, and thus analysis focused on the possible connection between these variables in th e college male at-large population rather than specifically targeting sexual minority groups, which could have skewed the results.

PAGE 30

30 In addition to the transsexualism discussed a bove, intersex individuals may comprise a group with differing patterns of gender role conflict Definitions of intersex vary somewhat, but typically refer to those born with XX chromosome s but displaying male external genitalia, those born with XY chromosomes but displaying female ex ternal genitalia, or th e more rare individual born with both male and female external genita lia (Sax, 2002). Applying the definitions of the all-encompassing type approximately 1.7% of the population would be considered intersex (Fausto-Sterling, 2000); the more narrow definiti on yields a percentage of about 0.018% (Sax, 2002). Despite the small proportion of affected persons, these individual s represent a unique group who may experience gender very differently from non-intersex individuals. Differing levels of hormones, pubertal development, a nd anatomical development make this population interesting to study; however, th ese individuals could vary marked ly from transsexuals in ways related not only to social reaction, but to anatomy and biology as well. In fact, the absence of an anatomical sexual abnormality is a sine qu a non for the diagnosis of gender identity disorder, meaning that the absence of inters ex conditions are require d for a possible gender identity disorder diagnosis (Bow er, 2001, p. 2). These intersex individuals, then, would probably be best studied as an independent group and not in comparison with those individuals without possible anatomical or hormonal expl anations for gender role conflict. Theoretical Approach Research on adolescence has seen a recent sh ift from a problem-based orientation to a positivistic framework, with many researchers a dvocating further shift to an asset-based resilience approach (Galambos & Leadbeater, 2000). These and ot her researchers warn that the problem-based approach has designated the abse nce of negative outcomes as the presence of healthy outcomes (Galambos & L eadbeater, 2000, p. 290). Further, the original two dimensions of youth resiliency have been expa nded through resiliency research to more accurately reflect the

PAGE 31

31 domains in which youth can display positive out comes (Cicchetti & Garmezy, 1993; Luthar, Dorenberger & Zigler, 1993). This study hopes to assess the potential eff ects of gender role conflict through an assessment of resilience in many facets of the adolescent/emerging adult experience. The 40 Developmental Assets The Search Institute, a research organization based in Minnesota, has identified forty developmental assets that play into adolescent de velopment. There are twenty internal assets, including personal hope for futu re and reading for pleasure, which are sub-categorized as follows: commitment to learning, positive values, social, and positive identity. Additionally, there are twenty external assets, such as community values youth and parental communication, which are sub-categori zed thusly: support, empowerment, boundaries/expectations, and construc tive use of time. The two types of assets are very different in their intentions of measurement: the extern al assets refer to the positive developmental experiences of relationship and opportunity that adults offer young people [and]the internal assets are competencies, skills, and self-perce ptions that young people develop gradually over time (p. 127). Peter Benson outlines the history and pertinence of the 40 Developmental Assets list, as both a theory and a practical approach (2002). Initially, this framew ork stemmed from the Search Institutes Profiles of St udent Life: Attitudes and Behaviors and included thirty assets; ten additional assets, based on further research a nd analysis, were added in 1996. The three main informational goals of this approach include d risk behavior prev ention, thriving outcome enhancement, and youth resilience. This chapter goes on to discuss the prevalence of the developmental assets. For optimal development, youth should have over 30 assets. Benson (2002) highlights the low mean of 19.3

PAGE 32

32 assets on average in the adolesce nt population. Clearly this dispar ity translates easily into risks for youth. Additional trends appear as well, such as a decrease in number of assets from 23.1 to 18.3 in grades six through twelve, an average of three fewer assets present in males than females, and the average of a three asset difference between adolescents in highly affluent families and those with lower socioeconomic status. Inte restingly, though, population size in the youths community was not related to a change in asset number, according to Benson (2002). Scales has found this framework helpful in complementing and even triangulating some theories of youth development (1999). He found the most common assets experienced by 6470% of youth included positive view of their pers onal future, family support, time spent in a religious community, school engagement, and the value of integrity; th e least common were caring schools, youth as resources reading for pleasure, the co mmunity values youth, and time spent in creative activit ies, experiences by just 19-25% of youth. The average adolescent according to Scales has only eighteen of the fort y assets, and thus is at risk for problematic behaviors in a wide range. The importance of these assets in adolescen t development is presented by Benson as well (2002). He states that assets are cumulative in nature and that as number of assets increase, propensity for risk behaviors dramatically de creases, while thriving behaviors increase. Additionally, a high number of develo pmental assets (31-40) were correlated with high academic performance (Benson, 2002). In fact, Benson found that adolescents with this high number of assets were .5 times more likely to report getti ng mostly As in school compared to students with eleven to twenty assets and about 8 times more likely than those reporting ten or fewer assets (p. 133). The developmental asset and resilience perspective, then, may provide an

PAGE 33

33 informative new framework through which to view gender role conflict, a concept that has been correlated with so many negative outcomes. Contemporary research utilizing the developmenta l assets identifies interesting trends. For example, Busseri et al. (2006) e xplored developmental asset inve ntory over four categories of sexuality (exclusively heterosexual, mostly hete rosexual, bisexual and same-sex attraction) and found that the exclusively hete rosexual population reported the greatest number of assets; however, the asset comparisons we re not significantly different over several domains, including academic orientation as well as peer victimization, which was low. Additionally, researchers have begun to recogn ize the importance of this asset framework under the domain of resilience. This approach is often compared with problem-based research, which identif[ies] deficits a nd pathology before offering aid [and] contrasts sharply with resiliency research and the Positive Youth Deve lopment (PYD) model, (Edwards et al., 2007, p. 29). They go on to say that the developmental assets perspective is clearly under this PYD model and can be used to help at -risk (or maintain low-risk) youth. Resilience Theory The metatheory of resilience and resiliency pr ovides a context for the Search Institutes 40 Developmental Assets. Richards on defines resilience as a force within everyone that drives them to seek self-actualization, altruism, wisd om, and harmony with a spiritual source of strength (2002, p. 313). He breaks the development of this theory down into three waves: first, the identification of resilient qualities, the segment in which the 40 Developmental Assets framework best fits; second, an understanding of the process of attaining these assets, which explains how assets break down and re-form; last innate resilience, identifying the motivation for resiliency through personal characteristics and the drive for self-actua lization. As the waves progress, aspects of the theory are less tangible, tying in aspect s of physics, Eastern medicine,

PAGE 34

34 and religion, among other fields (Richardson, 2002) The Search Institutes framework operates within the first wave of resiliency theory, the identification of resilient qualities, probably the most concrete of the three waves. The model of resilience, discussed in dept h by Richardson in his section on the second wave of resilience, incorporates the potential effects of gender role conflict as well as sexuality issues and disclosure (2002). In this model, adolescents/emerging adults are presented as having a biopsychospiritual homeostasis, a safe home base of sorts in which protective factors and stressors/adversity balance out, re sulting in minimal risk on the indi vidual. Next in the model, a disruption occurs, throwing off the balance of the former homeostasis. Fa ctors that could cause disruption may include self-awareness of gender ro le conflict, treatment or reaction from peers, taunting due to gender nonconforming behavior or myriad other similar events. Figure 2-2. Resiliency Model (Richardson, G. (2002). The metatheory of resilience and resiliency. Journal of Clinical Psychology 58 (3), 307-321.)

PAGE 35

35 The adolescent/emerging adults task is reinte gration, and according to the model, there are four ways to do this. The first is resilient reintegration, in whic h the rebound level is higher than the original homeostasis, usually due to lear ning, growth, or further development. Another outcome option is reintegration back to homeostas is, in which the adolescent simply deals with the disruption and leaves it behind. Yet another possibili ty is reintegration with loss, in which the individual loses motivation as a result of deal ing with this disruption. Lastly is dysfunctional reintegration, in which people resort to substanc es, destructive behaviors, or other means to deal with life prompts (Richardson, 2002, p.312). Since homeostasis is maintained by a stress level less than or equal to protective factors, as these protectors increase in number or intens ity, the youth becomes less likely to experience disruptions. Conversely, if yout h have very few protective fact ors, even small stressors can cause disruptions, resulting in an almost consta nt reintegration process that could result in dysfunctional reintegration. This dysfunctional integration model can lead to problematic behaviors, which would be more likely in adoles cents with few protective factors, according to this model. Conclusion Gender role conflict, then, can have great and lasting eff ects on adolescent and emerging adult development, both in terms of gender ident ity and resilience. This positivistic framework can assist the explora tion of exactly how adverse events can play out in adolescence and emerging adulthood. These results will allude to potential prevention/intervention methodology for practitioners while also giving researchers direction for further study on the topic.

PAGE 36

36 CHAPTER 3 MATERIALS AND METHODS Purpose of the Study The purpose of this study was to examine gende r role conflict in adolescence and emerging adulthood, as well as its effect on development in males as discussed in the previous chapter. Specifically, it investigated the e ffect of the four patterns of ge nder role conflict on resilience in adolescent and emerging adult males. This chap ter contains an overview of the methodological processes used in the study in terms of research design and data collection methods including population, sampling, instrumentation, and procedure and method of analysis as related to the research questions identified in Chapter 1. Research Design A cross-sectional design was used for this study. Bryman defines a cross-sectional research design as one in which several or many cas es are reported on at one single point in time in order to collect data on two or more va riables (2004). This paradigm relies on existing differences between groups, as it does not track differences between or among individuals over time (de Vaus, 2001). These studies examine relati onships between variable s, but the direction of these relationships is not fully explained th rough cross-sectional res earch and thus requires further study (Bryman, 2004). The cross-sectio nal design relies on comparing groups based on identified variables. In this study, adolescents and emerging a dults varying degrees of gender role conflict will be compared and related to as set inventory, thus assessing the outcome variable of resilience through examina tion of these stratified groups. Population and Sampling Procedures This study examined the influence of gender ro le conflict on resilience in adolescence and emerging adulthood. The theoretical population for this study, therefore, was adolescents and

PAGE 37

37 emerging adults. Since males tend to disclo se their sexual minority status within adolescence/emerging adulthood, while females tend to disclose after this developmental period, the sample was restricted to males exclusivel y (Kulkin, 2002). Further, due to the sensitive nature of the topics at hand, this study opted to utilize only those late adolescent males over the age of 18 that could agree to part icipate without parental consent. This age is on the cusp of the adolescent/emerging adult divide. To examine the relationship between ge nder role conflict and resilience in adolescent/emerging adult males, a self-complet ion questionnaire (as appears in Appendix A) was distributed to potential part icipants during the spring 2007 se mester. The original sample for this study was comprised of males in their firs t year of college enrolled in courses at a large coeducational university in the southeast. This was due to both budget and time constraints as well as to the representative nature of this age group with respect to the variab les. To obtain this sample, a list of all general e ducation classes offered during th e data collection semester was acquired, as every student enrolled must meet th ese educational requirements and thus these classes would most likely be most representative of the first-year collegiate population at large. Recruitment through general education courses also ensured that respondents would be from a range of colleges/degree progr ams within the university. To obtain a stratified random samp le of these courses, every 5th course was selected and individual instructors of sele cted courses were contacted vi a e-mail requesting permission to involve their students in the study. As professors all have valid e-mail addresses while not all professors (i.e. adjunct instructors) had cam pus offices with phones, e-mail was the most universal recruitment method. Additionally, e-mail contact info rmation is provided through the university database automatically unless users alte r it themselves. This supported the use of e-

PAGE 38

38 mail rather than other recruitment methods and also allowed for a shortened recruitment time frame. E-mail addresses for these instructors were obtained through the university database. Courses with instructors whose e-mail addresses we re not available were excluded. If, after two follow-up e-mails the professors failed to respon d, those classes were removed from the sample. Courses not offered during the term of data collection were elimin ated. Subsequent waves of the same procedure were conducted until a su fficient sample size was obtained. Of the approximately 450 general education classes offered on campus during the data collection semester, 22 participated in this study and generated 394 comp leted instrumentation packets. While the general education categories span many areas of study, the instructors that allowed their students to part icipated taught predominantly Social/Behavioral Science or Humanities courses (as opposed to math, biological or physical science). Probably due to the nature of these classes, the majority of every participating class was female. Regardless of the proportion of males to females, all male students in attendance were invited to participate. Only a minimal number of males (n=8) chose not to complete the inst rumentation, and often offered a timerather than content-related excuse relate d to their next class or an upcoming quiz/test; moreover, the refusals usually occurred immediately rather than after a perusal of the instrumentation. After specific cases were exclud ed due predominantly to age restrictions but also to intersex or tran sgender status as discussed in the pr evious chapter, 363 cases remained. It is important to note that no participants indicat ed transgender status without also indicating intersex status; thus, all transg ender respondents were excluded ba sed on their intersex status. One respondent filled out the firs t but not the second page of the instrumentation packet, leaving the valid number of cases for the Developmen tal Assets Profile (DAP) at 363 but the Gender Role Conflict Scale (GRCS) at 362 valid cases.

PAGE 39

39 This was determined to be sufficient th rough power sampling, a technique used based upon preliminary variance, set alpha level, and precision level. The formula for this sampling procedure is as follows: Figure 3-1. Sample Size Formula (Israel, 2002) For purposes of this study, the confidence in terval was set at 95% (z=1.96), and the precision was set at 1/6th of the mean score. This precisi on level reflects the nature of the instrument, specifically the most specific scalar response, which ha s six valid response choices. The maximum precision that could reasonably be reported from this instrument, then, is one response category off, hence the 1/6th multiplier. The variance was calculated at multiple levels within the study, and determined that the numbe r of respondents required was approximately 300. Thus, after consulting a committee of social science researchers familia r with this process, the optimal sample size was set at 400, with an ideal range of between 350 and 400 participants. Through preliminary analysis of the data dur ing the collection phase, the sampling frame what changed somewhat. The adolescent/emergi ng adult age range would encompass, at most, ages10-25; however the Internal Review Board onl y approved the study of those individuals over the age of majority, limiting the theoretical population to the 18-25 age range. As this study sought to examine gender role conflict in th e traditional college population, 25-year-old and older undergraduates were excluded, as these studen ts are typically considered nontraditional students. The reasoning behind the original se lection of the 18-19 year old population was based on the Developmental Asset Profile instrumenta tion which had been normed on an 11-18 year old age range. Since the instruct ions indicated to respondents to answer items based on the prior three months, 19 year olds were also include d in the theoretical sa mple, though subject to

PAGE 40

40 statistical validation. Analysis wa s run to determine th e internal consistenc y of the DAP over the 19 year old age range; moreover, as many responde nts were in the 19-2 4 age range, internal consistencies were run on this la rger group as well. This analys is showed the 19-24 year old age group to be about as internally consistent as th e 18 year old group; thus the entire theoretical population, males aged 18-24 years, was included in the sample. Alpha levels for these age groups can be found in Table 2-3. Data Collection After identifying classes for participation in the study, the researcher corresponded with the professors to set a date and time in which to collect data. All male students in attendance in each class were offered the opportunity to voluntar ily participate. The researcher then provided the participants with an overview of the study and instructions for completion, beginning with the required consent form. No content questions were answered during the study. Since this study sought to examine gender ro le conflict, it could not include those individuals that actually have biological traits of both genders (intersex). This population was excluded from the study in order to avoid sk ewing the sample. Similarly, transgender individuals, or those persons that have one sex but identify with or express the other gender, may be viewed as a separate category of individuals with respect to ge nder role conflict, as they have a diagnosable DSM-IV condition; however, this population, relative to the population at large, would most likely be too small to warrant analysis in this study. Future research in this field is necessary for a thorough examination of gender role conflict and/or resili ence issues specific to this population. Due to the sensitive nature of these demographic questions, all students were asked to participate regardless of their status since identifying these groups at the time of administration would have inadvertently identif ied potentially stigmatized groups. Respondents were informed of their rights and responsibilities as rese arch participants, incl uding their right to

PAGE 41

41 withdraw from the study at any time, as well as confidentiality in accord ance with the regulations of the Internal Review Board. Complete conf identiality was ensured by coding instrumentation packets with identification numbers for analysis rather than collecting identifying information such as name or birthdate. Res pondents were also instructed not to participate multiple times in the study so as not to skew or bias results. Instrumentation A self-completion questionnaire was used for th is study. This questionn aire contained four main portions: the Developmental Assets Profile (DAP), the Gender Role Conflict Scale (GRCS), a modified Kinsey sexuality rating, and demographic questions. The questionnaire was pilot tested on three students for ease of readi ng and timing purposes. Feedback from pilot test participants was used for minor modifications to the instrument such as including scalar response options on subsequent pages as well as for clarification of termi nology or concepts referenced in the instrumentation packet. More major change s and wording changes were not made, as these individual instruments have b een established through social sc ience research and have been optimized through these processes. Developmental Assets Profile (DAP) The Developmental Assets Profile (DAP) was cr eated by the Search Institute to inventory the number, types, and combinations of developm ental assets that youth between the ages of 11 and 18 report. These assets were divided into internal (Commitment to Learning, Positive Values, Social Competencies, Positive Iden tity) and external (Support, Empowerment, Boundaries and Expectations, Constructive Use of Time) asset categories. Additionally, the five social contexts (Personal, Social, Family, School, and Community Assets) were assessed. Respondents indicated the freque ncy at which they had these assets with four possible nonnumeric responses: Not At All or Rarely (=0), Somewhat or Sometimes (=1), Very or

PAGE 42

42 Often (=2) and Extremely or Almost Always (=3). Items are simple statements such as [If]eel safe at school or [Is] tand up for what I believe in. The Developmental Assets Profile (DAP) has been tested on youth in the Midwest (n = 1300) and the North (n = 1190), and the instrument has subsequently been used in many research studies (Developmental Assets Pr ofile Abstract, 2004). Chronbachs alpha for the DAP averaged .81 for the asset categories and .88 for the five context scales (Developm ental Assets Profile User Manual, 2005). Validity was assessed using the Search Institutes Attitudes and Behaviors survey, and yielded a correlation of .82 for total asset scores (Dev elopmental Assets Profile User Manual, 2005). The Developmental Asset Prof ile has a reported completion time of 10-15 minutes and is structured as follows: Table 3-1. Developmental Assets Profile Ex ternal Asset Categories and Item Numbers (Developmental Assets Prof ile User Manual, 2005) Support (7 items) 13, 47, 48, 49, 51, 54, 56 Empowerment (6 items) 17, 21, 25, 29, 36, 46 Boundaries and Expectations (9 items) 43, 44, 45, 50, 52, 53, 55, 57, 58 Constructive Use of Time (4 items) 31, 34, 40, 42 Table 3-2. Developmental Assets Profile In ternal Asset Categories and Item Numbers (Developmental Assets Prof ile User Manual, 2005) Commitment to Learning (7 items) 5, 7, 8, 10, 26, 28, 38 Positive Values (11 items) 1, 9, 16, 22, 23, 30, 32, 33, 35, 37, 41 Social Competencies (8 items) 4, 6, 11, 18, 19, 20, 24, 39 Positive Identity (6 items) 2, 3, 12, 14, 15, 27 The DAP was used to determine resilience in adolescents/emerging adu lts participating in this study. Overall resilience was assessed as related to gender ro le conflict patterns of success, power, and competition (RQ1); restrictive emotionali ty (RQ2); restrictive affectionate behavior between men (RQ3); and c onflict between work and family relations (RQ4).

PAGE 43

43 Validating the Developmental Assets Profile (DAP) The Search Institutes Developmental Asset Profile was normed on respondents ages 1118. As this instrument was used on youth ages 11-18, it was important to examine its internal consistency over ages 18-24 years, the range of ages used in the present study. Chronbachs alphas for this study are separately reported for 18 year olds a nd 19-24 year olds in order to compare internal consistency of the two groups. Values for the overall sample are also given. For all three groups, alphas were assessed for each factor of the DAP instrumentation as well as the overall DAP Score (Table 3-3). Table 3-3. Internal Consistency Values for the Developmental Assets Profile. 18 year olds 19-24 year olds Entire sample Support .715 .767 .762 Empowerment .776 .677 .694 Boundaries .834 .785 .792 Constructive use of time .327 .457 .442 Commitment to Learning .699 .750 .743 Positive Values .731 .814 .805 Soc. Competencies .631 .723 .712 Positive ID .859 .807 .816 DAP Total .937 .937 .937 Gender Role Conflict Scale (GRCS) The Gender Role Conflict Scale (GRCS) wa s designed to assess overall gender role conflict in males as well as its component patt erns. Respondents are in structed to indicate through scalar response the degree to which th ey agree or disagree with given statements, ranging from Strongly Disagree (=1) to Stro ngly Agree (=6). These patterns are success, power, and competition (SPC); restrictive emotiona lity (RE); restrictive affectionate behavior between men (RABBM); and conflict between work and family relations (C BWFR). This thirtyseven-item instrument is scored both as an ove rall gender role conflict score and as a patternspecific score for each of the four patterns of gend er role conflict. The distribution of items per pattern factor and statistics based on past use of the instrument are as follows:

PAGE 44

44 Table 3-4. Gender Role Conflict Scale Fa ctors and Item Numbers (ONeil, 1986) 1: Success, Power, Competition ( 13 items) 1, 5, 8, 12, 14, 18, 21, 23, 24, 28, 32, 34, 37 2: Restrictive Emotionality (10 items) 2, 6, 9, 13, 15, 19, 22, 25, 29, 30 3: Restrictive Affectionate Behavior Between Men Homophobia (8 items) 3, 7, 10, 16, 20, 26, 33, 35 4: Conflicts Between Work and Leisure Family Relations (6 items) 4, 11, 17, 27, 31, 36 Table 3-5. Gender Role Conflict Scale Statistics (ONeil, 1986) Factor 1: SPC Factor 2: RE Factor 3: RABBM Factor 4: CBWFR Mean Factor Loading .54 .55 .60 .57 Internal Consistency Reliability .85 .82 .83 .75 Test Retest Reliability .84 .76 .86 .72 Variance Explained 17.2% 7.6% 6.1% 4.6% Internal consistency for this study was measur ed using Chronbachs alpha values for each pattern of gender role conflict and the overall Gender Ro le Conflict Scale (GRCS) score. Values are presented in Table 3-6. Table 3-6. Internal Consistency Reliability Values for the GRCS Factor 1: SPC Factor 2: RE Factor 3: RABBM Factor 4: CBWFR GRCS Total Score Chronbachs Alpha Value .883 .885 .873 .802 .926 The GRCS was used to ascertain how the indivi dual patterns of gender role conflict affect overall adolescent resilience (RQ1-4). Sexuality Rating The sexuality rating for this study was a singl e item designed to ascertain self-reported respondent feelings of sexuality. This item wa s a self-report item based on the Kinsey rating, and had 0 to 6 scalar response ranges from ex clusively heterosexual with no homosexual to exclusively homosexual, (Kinsey, 1948). This rating would have been be used, sample size permitting, to compare sexual minority males with their heterosexual counterparts in terms of gender role conflict and its pot ential effect on resilience.

PAGE 45

45 Figure 3-2. Kinsey Sexuality S cale (Kinsey at al., 1948, p. 638). 0 = Exclusively heterosexual; 1 = Predominantly heterosexual, only in cidentally homosexual; 2 = Predominantly heterosexual, but more than incidentally homosexual; 3 = Equally heterosexual and homosexual; 4 =Predominantly homosexual, bu t more than incidentally heterosexual; 5 = Predominantly homosexual, only inci dentally heterosexual; 6 = Exclusively homosexual. This spectrum of sexuality has been utilized in research in various forms to assess sexuality. Kinsey was challenged on his method of assigning respondents to a particular group; however, the present study only us ed this sexuality as a cont inuum paradigm to create an opportunity for respondents to self-report more specific sexualities on a spectrum. Demographic Questions These questions were used for three reasons First, they ensured that the population studied here was maximally representative of th e adolescent/emerging adult population at this university; that is, that they met the require ments of the study and closely resembled the population of students at the univers ity that could have participated based on these requirements. These included: male sex, aged over 18 (but not a bove 24) years, enrolled in undergraduate study at the university and no t transexual or intersex individuals. The second purpose was to identify those individual respondents who, due to their answers on these questions, must be excluded, namely intersex individuals, which, in this st udy, also excluded the tota lity of self-reported transsexuals, and those under ag e 18 or over age 24. Finally, de mographic variables may result

PAGE 46

46 in differences in resiliency inde pendent of gender role conflict. Demographics were assessed as related to resilience to separate these effect s from the effects of gender role conflict. Data Analysis Figure 3-3. Data Analysis Model. Data was analyzed as per the above model. Independent variables, including demographics appear on the left and the dependent variable, resilience, appears on the right In order to explore initial potential relationships, Pearson s correlations were conducted between the variables in the model in three group s: demographics related to resilience, gender role conflict related to resilience, and demographics related to gende r role conflict. If significant relationships were found through these correlations, ANOVAs were used to independently explore the more specific associat ions between variables. Thes e tests independently compared demographic variables and the patterns of gender role conflict as well as overall gender role

PAGE 47

47 conflict score to resiliency score. Finally, multivariate analyses were conducted to ascertain which of the variables most heavily influenced or predicted resiliency in this population. In these analyses, overall DAP score served as the dependent variable initially. This was then further explored through multiple regression models using both Internal Asset score and External Asset score as dependent variables to de termine more specific relationships. Limitations and Delimitations This study has several potential limitations. First, a cross-secti onal study cannot assess change over time. Thus, this study was unable to track Gender Role Conf lict and its potential effects on the individual at varyi ng developmental periods or its effect on development itself. Additionally, the sensitive nature of the material covered by the instrumentation may result in dishonest answers from respondents, especially if they perceived their peers to be aware of or interested in their responses. This university setting is a la rge campus in the southeast with much diversity in student body, academic programs, and student life. Thus, th e results may not be readily generalizable to other seemingly similar populations. As a la rge university with highly rigorous admission requirements, a different kind of student may be enrolled at this institu tion rather than other universities. When consideri ng colleges, many prospective students make decisions based on size, number of majors offered, rigor of degree programs, reputation, and other factors. The status of this university with respect to these considerations may determine the type of student body comprising the sample for this study. Thus schools of other types (small, fewer degree programs, religiously affiliated, private institutio ns and the like) may have different populations with respect to the vari ables in this study. Also, the length of the instrumentation ma y cause some degree of respondent fatigue, resulting in potentially less valid responses toward the end of the questionnaire. To alleviate as

PAGE 48

48 much of this as possible, demographic que stions requiring less t hought by respondents were asked at the end of th e instrumentation. Further, sampling from a list of general e ducation courses necessarily excludes those students that may have completed these re quirements by some other method (Advanced Placement, International Baccalaureate, community college, etc.) and thus potentially excludes many highly advanced students as well as commun ity college transfer st udents. Additionally, students enrolled in fewer courses were less likely to be includ ed. These individuals may be different from the typical unde rgraduate population of males. Finally, within the classes sampled, participan ts were asked to volunt arily participate. Those that self-selected not to participate may be different from those that did participate in terms of the variables in question, especially if they perceived th at their peers could potentially become aware of their responses to instrument items. This was not thought to affect this study to a large degree, as the few refusals (n=8) often occurred almost immediately, presumably before they could have reviewed the instrumentati on thoroughly, and were often accompanied to timerelated excuses (I have a test next period). Methodology as presented in this chapter was designed to examine the effects of gender role conflict on resilience among adolescents and emerging adults. Complete analysis of the data is presented in Chapter 4.

PAGE 49

49 CHAPTER 4 RESULTS Sample Statistics Descriptive statistics give a general picture of the data. The data here are divided into three categories corresponding to the sections below, which cont ain summary statistics of the variables included in this study. These variables include: demogr aphics, gender role conflict, assessed by the Gender Role Conflict Scale (G RCS), and resiliency, measured using the Developmental Assets Profile. Summary Statistics: Demographics Demographic summary statistics are re ported below based on the sample of 363 respondents and include race, sexual orientation, age, and academic status. Since this study included only male participants, sex was collect ed but was used only for screening purposes. Current academic college was collected for sample validation. Race was reported by participants by indicating on e of the categories listed in Table 4-1. If respondents indicated more than one race or negl ected to respond to this item, their responses were coded as missing data so as not to skew results. There were 35 cases of missing response for this item. Descriptive statistics for race cat egory are presented in ta ble 4-1. Some of the courses that responded positively to the request fo r research assistance we re courses on cultural or ethnic topics, including history, collegiate experience, and religion. This may have contributed to the slight devian ce from the racial distribution, th at is, the larger proportion of racial minorities, of the student body. While at first this may be considered a weakness, it is probable that some of the student s in these courses, although not ra cially representative of this collegiate population, may make the sample more generalizable to the population of 18-24 year old males at large.

PAGE 50

50 Table 4-1. Summary statistics for race variables Respondent Self-Report Race Frequency Sample Percent Population Percent American Indian or Alaska Native 0 0 0.3 Asian 15 4.6 7.3* Black or African American 53 16.2 9.6 Hispanic or Latino/Latina 55 16.8 13.1 Native Hawaiian/Other Pacific Islander 2 0.6 7.3* White 194 59.1 66.1 Other (please specify) 9 2.7 3.5** Total 363 100.0 100.0 These two categories were combined in the available population statistical information ** This category for the population include s unknown race and non-resident aliens Sexual orientation was indicated by responde nts on a scale ranging from exclusively heterosexual with no homosexual (=0) to exclu sively homosexual (=6). Summary statistics appear in table 4-2. Respondents reported age in years as part of completion of the instrument. Some respondents (n=41) neglected to provide an age. As only a minimal percentage of undergraduate respondents fell outside of the 1824 year old age range, instrument ation packets missing data for age were included for analysis bu t resulted in a depleted number of cases for the age variable. Summary statistics appear in Table 4-2. Table 4-2. Age and Sexual Orie ntation summary statistics Mean Range Standard Deviation n Sexual Orientation 0.32 0 61.161350 Age 20.05 18 241.602322 Academic status was reported as fresh man (n=107, or 29.5%), sophomore (n=93, or 25.6%), junior (n=76, or 20.9%), or senior (n=81 or 22.6%). Six responden ts did not indicate their academic status. Few respondents reported that they were some what (n=8) or strongly (n=3) intersex. All respondents who indicated they felt that the defin ition of transgender applied to them somewhat (n=5) or strongly (n=4) also self-reported intersex status. All intersex individuals were excluded

PAGE 51

51 from analysis based on this status. Thus, no tr ansgender individuals were included. Cases in which respondents did not respond to items regardi ng intersex (n=7) and transgender (n=5) status were included in analysis, since the percentage of the population excluded was so small as to probably not affect these few cases. Summary Statistics: GRCS (Independent Variable) The Gender Role Conflict Scale (GRCS) aske d respondents the degree to which they agreed or disagreed (1=Strongly Disagree, 6=St rongly Agree) with given statements. The individual items combined to fo rm four factors or patterns of gender role conflict, including: Success, Power & Competition (SPC); Restrictiv e Emotionality (RE); Restrictive Affectionate Behavior Between Men (RABBM); and Conf lict Between Work and Family Relations (CBWFR). Pattern scores as we ll as overall GRC score were atta ined by averaging the responses of items within that pattern (detailed in Chapter 2). The summary statistics for the patterns of gender role conflict and total GRC score are reported below. Respondents overall reported a moderate level of gender role conf lict, with the notable exception of SPC score, which is slightly elevated. Relationships between these scores and other variab les will be explored through bivariate and multivariate analyses to provide insight into possible reasons for these scores. Table 4-3. GRCS summary statistics Valid Cases Mean Score Range Standard Deviation Alpha Level SPC 362 4.1431 60.9210.883 RE 362 3.1591 61.0260.885 RABBM 362 3.3271 61.1440.873 CBWFR 362 3.5991 61.0820.802 GRC Score 362 3.6171 5.860.7740.926 Pearsons correlations were r un on the individual pattern scor es and the overall GRC score for further instrument validati on. The correlation table with r values can be found in Appendix B. The range of GRC scores falls short of the maximum score of six because no respondent

PAGE 52

52 indicated that he strongly agreed with all items. As this score is an average of all responses on the GRCS, this resulted in a less than maximum score. Summary Statistics: DAP (Dependent Variable) The Developmental Assets Protocol (DAP) in structed respondents to indicate how often they had positive things in their lives, with responses ranging from Not At All or Rarely (=0), Somewhat or Sometimes (=1), Very or Often (=2), to Extremely or Almost Always (=3). These individual items were grouped to form factors as well as internal and external asset scores, classified as shown and abbrevia ted in Table 4-4. They also fo rm context area factors including personal, social, and family, whic h were similarly derived but not included in further analysis. Summary statistics for these factor s appear in Appendix B. Fina lly, all items result in a total DAP score. Table 4-4 reports summary statisti cs for all factors, and the total score. Although most Chronbachs alpha levels are high enough to use these factors independently, this study focuses primarily on the overall resilience of youth, or total overall DAP score as well as External/Internal Asset scores, for which the alpha levels are sufficient for analysis. Analysis using these factors of resilien ce is included at the bivariate level to elucidate possible relationships for further exploration, but these individual comp onent factors are not included in later multivariate analysis. Pearsons correlations were r un on the individual pattern sc ores, context areas and the overall DAP score for further instrument validation. The correlation table with r values can be found in Appendix B. The overall DAP scores reported here indicate higher levels of resilience than the reported average number of assets in high school youth as di scussed in chapter three. This is surprising, although may result from the high-achieving nature of the students admitted to this university.

PAGE 53

53 The relationships between demographics, gender ro le conflict, and resili ence will be explored through bivariate and multivariate analysis. Table 4-4. DAP su mmary statistics FrequencyMean Range Standard Deviation Alpha Level Support (SUP) 363 21.464 5.71 30.00 6.050 0.762 Empowerment (EMP) 363 22.525 5.00 30.00 4.512 0.694 Boundaries & Expectations (B&E) 363 20.596 6.67 30.00 5.011 0.792 Constructive Use of Time (CUT) 363 14.963 0.00 30.00 6.486 0.442 External Asset Score (EA) 363 20.898 8.61 30.00 3.808 0.891* Commitment to Learning (CTL) 363 21.578 5.71 30.00 4.705 0.743 Positive Values (PV) 363 19.813 9.09 30.00 4.500 0.805 Social Competencies (SOC) 363 20.778 8.75 30.00 4.277 0.712 Positive Identity (PID) 363 21.424 1.67 30.00 5.075 0.816 Internal Asset Score (IA) 363 19.887 5.96 37.31 4.435 0.908* Total Score (DAP) 363 40.786 18.85 58.73 7.419 0.926* *Alpha levels are for compone nt items for these factors, not for calculated scores. Bivariate Analyses Bivariate analyses examined the relationships between gender role conflict and resilience and determined the effect, if any, that demographi c variables had on other variables in this study. Demographics by Resilience Demographic variables may affect resilience independent of ge nder role conflict. It is important to separate these influences to correctl y identify clear directional relationships between these variables. Pearsons correlations were used to dete rmine if there was a relationship between demographic variables and resilience. Signifi cant correlations were found with demographic variables and various patterns of resiliency, both external, show n in Table 4-5 and internal, presented in Table 4-6. Overall resilience measured by DAP value was also correlated with sexual orientation ( r = 0.016), academic status ( r = -0.112*), age ( r = -0.098), and race (black ( r = 0.072), Hispanic ( r = -0.032), white ( r = 0.041), and other ( r = -0.093)).

PAGE 54

54 Table 4-5. Pearsons Correlation r values relating Demographi cs and External Resilience Patterns and overall External Asset score SUP EMP B&E CUT EA Sexual Orientation -0.048 -0.010 -0.005 -0.100 0.096 Academic Status -0.186** -0.111* -0.172** -0.076 -0.028 Age -0.157** -0.068 -0.132* -0.118* -0.020 Black -0.023 -0.056 -0.067 0.088 0.150** Hispanic -0.079 -0.024 0.028 -0.110* 0.011 White 0.171** 0.127* 0.107* 0.015 -0.060 Other Race -0.131* -0.094 -0.106* 0.003 -0.072 Significant at the 0.05 level ** Significant at the 0.01 level Table 4-6. Pearsons Correlation r values relating Demographics a nd Internal Resilience Patterns and overall Internal Asset Score CTL PV SOC PID IA Sexual Orientation 0.126* 0.104 0.133* -0.032 -0.056 Academic Status -0.024 0.011 -0.029 -0.047 -0.164** Age -0.038 0.007 0.004 -0.034 -0.149** Black 0.113* 0.185** 0.135* 0.068 -0.008 Hispanic 0.106* -0.037 -0.003 -0.030 -0.064 White -0.064 -0.098 -0.064 0.020 -0.123* Other Race -0.122 -0.008 -0.040 -0.061 -0.095 Significant at the 0.05 level ** Significant at the 0.01 level This analysis shows that demographic variable s were associated with differing levels of resilience independent of gende r role conflict. Specificall y, as respondents indicated an increasing degree of homosexuality on the scale, they also tended to report more assets within two resiliency areas. Higher academic status was a ssociated with fewer asse ts in three areas and with a lower Internal Asset (IA) score and overa ll DAP Score. As age increased, respondents tended to report fewer assets within three asset areas. Finally, Black or African American and White respondents reported comparatively more resilience than their Hispanic or Other racial

PAGE 55

55 counterparts. It is important to further explore these demogra phic relationships to clarify the nature of these relationships. To this end, ANOVAs were run between each independent demographic variable and resilience both by pattern and score to determine any impact these variables may have on resilience. One of the underlyi ng assumptions for the ANOVA test is that the variances between groups compared is similar. While this assumption was met in many ANOVA tests, some individual tests did not meet th is assumption, indicated by a signifi cant Levenes test statistic. The tests that did not meet this assumption were : Academic status by Constructive Use of Time (CUT); Age by Support (SUP), Empowerment (EMP ), and Boundaries & Expectations (B&E); and Race Group by SUP, EMP, B&E, and Social Co mpetencies (SOC). Further relationships among all variables were explored further through multivariate analysis. Significant results are reported below, and tables for all significan t results are presented in Appendix B. ANOVA examined the effect of academic status on resilience. The significant relationships found are summ arized in Table 4-7. Table 4-7. Effect of academic status on mean DAP score Freshman Sophomore Junior Senior F P-value Support 22.388 21.797 21.150 19.583 4.505 .004 Boundaries & Expectations 21.149 21.531 20.490 18.841 4.960 .002 Internal Asset Score 20.399 20.457 19.734 18.456 4.023 .008 These results indicate a general decline in resilience over the academic progression in college within the Support and B oundaries & Expectati ons asset areas, which are external asset areas. Additionally, Internal Asset score followed the same trend, with less resilience in seniors than in freshmen. Post hoc tests in all cases revealed that the significant difference existed between the Freshman/Sophomore and Senior groups.

PAGE 56

56 Race also showed an effect on resilience through ANOVA tests. Si gnificant results are reported in Table 4-8. Table 4-8. Effect of Race on mean DAP score Asian Black Hispanic White Other F P-value Commitment to Learning 20.000 22.857 22.753 21.296 20.404 3.228 .013 Positive Values 18.727 21.825 19.422 19.400 20.059 3.508 .008 EA Score 19.506 22.277 20.996 20.685 20.549 2.539 .040 IA Score 16.974 19.769 19.202 20.356 19.584 2.694 .031 Post hoc tests only helped to eluc idate the relationships identified in two of these tests. For the Positive Values (PV) asset category, th e largest difference was seen between Black respondents and both Hispanic and White responden ts, with Black individu als typically reporting more of these assets. For Internal Asset score, post hoc tests revealed the largest difference existed between Asian and White respondents, in which Asians typically had lower Internal Asset Scores. Demographics by Independent Variables Exploring the relationships between demogra phic variables and patte rns of gender role conflict are important for this st udy. If demographics affect gende r role conflict, which in turn affects resilience, it is imperative to separate these relationships to determine the independent effects of each group of variables. To examine possible effects here, Pearson s Correlations were run among demographic variables, patterns of gender role conflict, and overall GRCS score. These tests indicated that as sexual orientation tended toward homosexuality, gender role conflict generally decreased. Increased academic status was also associated with a decrease in gender role conflict. Also, Black/African American and Other race status corresponded with higher levels of gender role

PAGE 57

57 conflict patterns, while Hispanic and White race status corres ponded with decreased levels of gender role conflict. Table 4-9. Pearsons r values correlati ng Demographics and Gender Role Conflict SPC RE RABBM CBWFR Total Sexual Orientation -0.048 -0.116* -0.251** -0.047 -0.154** Academic Status -0.110* -0.009 -0.012 -0.007 -0.058 Age -0.052 -0.042 0.018 0.001 -0.036 Black 0.094 0.122* 0.134* 0.108* 0.150** Hispanic -0.041 -0.030 -0.127* -0.095 -0.095 White -0.114* -0.105* -0.059 -0.091 -0.120* Other 0.105* 0.055 0.074 0.111* 0.110* Significant at the 0.05 level (2-tailed) ** Significant at the 0.01 level (2-tailed) Next, ANOVAs were run between these demogra phic variables and gender role conflict. Again, some individual tests did not meet the requirements for this test, as they had differing variances between groups, shown by a significant Levenes test st atistic. These tests included both Conflict Between Work and Family Relations and total Gender Role Conflict Scale score by academic status. Significant findings are presented below. ANOVA tested age against gender role conflict. Age affected the SPC pattern and total gender role conflict scores. Po st hoc tests indicated that the 20-year-olds experienced a significantly higher degree of gender role conflict with the SPC pa ttern and overall than their 22year-old counterparts. A complete table for this test can be found in Appendix B. Academic status also affected gender role conflict, specifically th e SPC pattern. Table 410 shows these results. Post hoc tests revealed the largest difference existed between seniors and their freshman and junior counterparts. Race group also affected gender role conflict, both overall score and pattern scores. Results are in Table 4-11. Post hoc tests reve aled the largest differences for both the RABBM

PAGE 58

58 pattern and overall GRC score ex isted between Hispanic and bl ack respondents, with black individuals generally reporting a higher degree of gender role conflict in both categories. Table 4-10. Effect of Academic Status on mean SPC pattern score of gender role conflict Freshman Sophomore Junior Senior F p-value SPC pattern 4.233 4.158 4.325 3.872 3.769 .011 Table 4-11. Effect of Race on mean pattern and total gender role conflict scores Asian Black Hispanic White Other F P-value SPC pattern 3.739 3.701 2.985 3.265 3.444 3.464 .009 RABBM pattern 4.118 3.885 3.355 3.508 3.783 3.222 .013 GRCS total 3.924 3.900 3.444 3.531 3.769 4.207 .002 Sexual orientation was run against gender role conflict by level. Scores from the self report item were grouped as follows: score of 0 (exclusively heterosexua l), score of 1 (only incidentally homosexual), and sc ores of 2-6 (more than incide ntally homosexual). This was grouped due to a low number of sexual minority males for purposes of analysis. Significant effects of sexuality on gender role conflict were tested by ANOVAs. Results are in Table 4-12. Table 4-12. Effect of sexual orientation on mean pattern and total gender role conflict scores Score of 0 Score of 1 Scores 2-6 F p-value RE pattern 3.193 3.370 2.744 3.352 .036 RABBM pattern 3.431 2.757 2.662 9.583 .000 GRC total 3.666 3.390 3.284 4.593 .011 Post hoc test revealed the largest differen ces for the RE pattern and overall gender role conflict scores were between the exclusively heterosexual and the more than incidentally homosexual group, with higher mean gender role conflict scores seen in the former group in both cases. For the RABBM pattern of gender role conflict, post hoc tests showed significant differences between the exclusively heterose xual group and both other groups, with higher gender role conflic t seen in the heterosexual group.

PAGE 59

59 Independent Variables by Resilience Pearsons correlations were used to determin e whether a relationshi p existed between both the patterns of/overall gender ro le conflict and resilience. Re sults for external factors and External Asset score, internal factors, Internal Asset score, and overall DAP score are given in Tables 4-13 and 4-14. Table 4-13. Pearsons r values correlating Gender Role Conf lict and External resilience factors and External Asset score SUP EMP B&E CUT EA SPC 0.051 0.055 0.063 0.077 0.040 RE -0.185** -0.284** -0.190** -0.076 -0.247** RABBM -0.088 -0.136** -0.082 0.045 -0.077 CBWFR -0.117* -0.142** -0.152** 0.026 -0.032 Total -0.096 -0.151** -0.097 0.029 -0.099 Significant at the 0.05 level (2-tailed) ** Significant at the 0.01 level (2-tailed) Table 4-14. Pearsons r values correlating Gender Role Conflic t and Internal resilience factors, Internal Asset score and overall DAP score CTL PV SOC PID IA Total SPC 0.036 -0.009 0.018 0.080 0.077 0.066 RE -0.165** -0.155** -0.240** -0.248** -0.214** -0.254** RABBM -0.106* -0.026 -0.081 -0.040 -0.070 -0.081 CBWFR -0.036 0.020 -0.020 -0.065 -0.107* -0.080 Total -0.081 -0.059 -0.105* -0.080 -0.086 -0.102 Significant at the 0.05 level (2-tailed) ** Significant at the 0.01 level (2-tailed) These correlations show that generally as gende r role conflict increases, resilience tends to decrease. Specific to pattern of gender role confli ct, SPC was not correlated with resilience. RE was negatively correlated with seven of eight resi lience factors as well as Internal, External and overall DAP score. RABBM was associated with a decrease in resilience as well, but was only significant for two resilience factor s. CBWFR was associated with a decrease three resilience factors as well Internal Asset Sc ore. Finally, overall GRC Score was related to a decrease in two resilience factors.

PAGE 60

60 Next, ANOVAs were run between gender role conflict, both component patterns and overall score, and resiliency factors and total scor e. Quartile levels were used for gender role conflict scores, the independent vari ables in this analysis section. Again, some tests did not meet the assumption of homogeneity of variances and we re explored later in multivariate analyses. These tests included: SPC by B&E, CTL, IA scor e, EA score, and overall DAP score; RE by SUP, B&E, and IA score; RABBM by EMP, PI D, and EA score; and CBWFR by PV and EA score. Tables of significant re sults are reported in Appendix B. Significant results are reported for the Restrictive Emotionality pattern level, the RABBM pattern level, the CBWFR pattern level, and overall gender role conflict score level below in Tables 4-15, 4-16, 4-17, and 4-18, respectively. Table 4-15. Significant relations hips between RE quartile and resilience pattern /overall means Level 0 Level 1 Level 2 Level 3 F p-value EMP 24.326 22.800 21.944 21.145 8.736 .000 CTL 22.632 21.849 20.908 21.040 2.666 .048 PV 21.032 19.826 19.358 19.139 3.247 .022 SOC 22.261 21.038 20.067 19.868 6.187 .000 PID 23.296 21.129 21.035 20.308 6.030 .001 EA 22.305 20.960 20.342 20.089 6.932 .000 DAP 43.235 41.004 40.202 38.728 6.121 .000 Post hoc tests revealed that the largest differe nces here were between the lowest level of RE gender role conflict and the highest or highest two levels. Respondents whose RE scores fell into the lowest quartile tended to report more assets. Conversely, respondents reporting higher levels of this pattern of gender role conflict reported lower resilience within these patterns in terms of number of assets. Table 4-16. Significant relations hips between RABBM quartile and resilience pattern means Level 0 Level 1 Level 2 Level 3 F p-value CTL 23.095 21.162 20.867 21.341 4.001 .008 SOC 21.801 20.059 20.911 20.470 2.687 .046

PAGE 61

61 Post hoc tests revealed that the largest diffe rence for the Commitment to Learning pattern existed between respondents with RABBM pattern sc ores in the lowest qu artile and those with RABBM pattern scores within th e two middle quartiles. For Social Competencies, the difference was between the first and second qu artiles. In both cases, as gender role conf lict with this pattern increased, number of assets re ported in these two areas decreased. Table 4-17. Significant relations hips between CBWFR quartile a nd resilience pattern means Level 0 Level 1 Level 2 Level 3 F p-value EMP 23.854 22.210 22.218 22.042 3.033 .029 SOC 21.409 20.697 19.526 21.439 3.983 .008 Post hoc tests revealed that the largest di fference for empowerment existed between the lowest-scoring and highest-scor ing respondents on the CBWFR patte rn. As gender role conflict of this type increased, resilience within th e empowerment area decreased. For social competencies, the largest differences existed be tween the third quartile and both the first and fourth quartiles. A high and low de gree of this type of gender role conflict was associated with increased resilience within the social competencies area. Table 4-18. Significant relationshi ps between GRCS score quartile and resilience pattern means Level 0 Level 1 Level 2 Level 3 F p-value EMP 24.091 21.778 22.509 21.889 5.105 .002 Post hoc tests revealed that the largest diffe rence existed between the lowest quartile and both the second and fourth quartil es. All relationships explor ed thus far will be further investigated through multivariate analysis. Multivariate Analysis Multiple regression was used to determine whic h specific facets of gender role conflict and/or demographic variables most significantly affect resilience in adolescent/emerging adult males in a real-world context. These regression an alyses tested five main concept areas, namely

PAGE 62

62 demographic variables and patte rns of gender role conflict (SPC, RE, RABBM, and CBWFR), against overall resilience (as measured by the DAP) in six separate models. Before running this model, it was important to determine how best to match the model with available data, research questions, and th eory. Age was one major consideration. Many respondents (n=41) did not complete this item, presumably due to its location on the instrumentation packet. A regression model iden tical to the one following in this chapter but with age included revealed sim ilar relationships (see Appendix B) but resulted in a depleted number of cases. Thus, age was excluded from the models. Beyond age, the regression models were tailored to answer the research questions in this study. Each gender role conflict pattern score was independently related to resilience as were demographic variables. For demographic variables (Model 1) in all models, race was coded into white, black, Hispanic, and Other, with the wh ite race used for the reference category. These items neared but did not meet a level of signi ficance. Models 2, 3, 4, and 5 entered gender role conflict pattern scores (SPC, RE, RABBM, and CBWFR, respectiv ely). Model 6 explored the effects of all gender role conflic t patterns and demographics on resilience at the same time. Finally, a reduced model was obtained which highli ghted the most important relationships in the models. Regression models were run to asse ss the effects of gender role conflict and demographics on overall resilience (total DAP sc ore). Table 4-19 shows the results of this analysis. This model shows that the strongest rela tionship between gender role conflict and reslilence is the component patter n of Restrictive Emotionality (RE). This pattern accounts for over 6% of the variance in DAP score. As gende r role conflict with this pattern increases, overall resilience tends to decline. Another rela tionship elucidated by this model is the positive

PAGE 63

63 relationship between the pattern of Success, Power and Competition and overall resilience. This relationship accounts for less than one percent of the variance in resilience independently, although as gender role conflict with this pattern increases, resili ency also tends to increase. This reveals a complex relationship between overall gender role c onflict and resilience. Table 4-19. Standardized regressi on coefficients for demographi c and GRCS pattern effect on DAP score Model 1 Model 2 Model 3 Mode l 4 Model 5 Model 6 Reduced Demographics Black .044 .069 Hispanic -.049 -.042 Other -.093 -.087 Academic Status -.110 -.090 Sexual Orientation .028 .019 SPC Pattern .066 .192** .181*** RE Pattern -.254*** -.347*** -.338*** RABBM Pattern -.081 .050 CBWFR Pattern -.080 -.040 Adjusted R2 .012 .002 .062 .004 .004 .102 .115 Cases 357 361 361 361 361 356 356 ** Significant at the .010 level *** Significant at the .001 level To further discern the relationships in this model, regression models were run using both the Internal Asset Score and External Asset Score, the components comprising overall DAP score, as dependent variables. As gender role conflict and/or demographi c variables may affect these types of assets differentl y, this analysis could expose more specific relati onships between variables. These regression models ar e presented in Tables 4-20 and 4-21. The first of these two regression models ex plored gender role conflict and demographic variables as related to External Asset Score. As in the a bove model, the SPC pattern was positively related to resilience and the RE pattern was negatively related to resilience. The results for both patterns emulated those found in the overall DAP score regression model. An additional significant influence on External Asset score was race. As mentioned before, white

PAGE 64

64 race was included as the reference category. As shown in this model, black respondents tended to have higher External Asset scores than the reference categor y. This was significant both in the demographic model (Model 1) as well as the reduced model. Table 4-20. Standardized regres sion coefficients for demographi c and GRCS pattern effect on External Asset score Model 1 Model 2 Model 3 Mode l 4 Model 5 Model 6 Reduced Demographics Black .146** .168** .176** Hispanic .019 .031 Other -.050 -.046 Academic Status -.022 -.007 Sexual Orientation .090 .080 SPC Pattern .040 .144* .154** RE Pattern -.247*** -.362*** -.329***RABBM Pattern -.077 .072 CBWFR Pattern -.032 .019 Adjusted R2 .020 -.001 .058 .003 -.002 .102 .106 Cases 357 361 361 361 361 356 356 Significant at the .050 level ** Significant at the .010 level *** Significant at the .001 level Table 4-21. Standardized regres sion coefficients for demographi c and GRCS pattern affect on Internal Asset score Model 1 Model 2 Model 3 Mode l 4 Model 5 Model 6 Reduced Demographics Black -.053 -.030 Hispanic -.099 -.098 Other -.114* -.108* Academic Status -.166** -.146** -.147** Sexual Orientation -.032 -.038 SPC Pattern .077 .200*** .181*** RE Pattern -.214*** -.272*** -.284*** RABBM Pattern -.070 .021 CBWFR Pattern -.107* -.084 Adjusted R2 .033 .003 .043 .002 .009 .101 .103 Cases 357 361 361 361 361 356 356 Significant at the .050 level ** Significant at the .010 level *** Significant at the .001 level

PAGE 65

65 Effect of these models on Internal Asset scor e was then assessed as well (Table 4-21). Again, this model shows the continued relations hip between the SPC and RE patterns of gender role conflict and resilience. Demographic info rmation also influenced Internal Asset score, especially Other race status, which was associated with a decrease in Inte rnal Asset score, and academic status, which is inversely related to In ternal Asset score. This relationship between academic status and Internal Asset score is also included in the reduced model, which suggests that as academic status increases, resilience declines. Summary This chapter presented the results of this study from large to small scale. Following descriptive statistics, bivariate correlations iden tified possible relationships between the myriad variables in this study. Then ANO VAs were used to discern the sp ecific directional relationships involved. Next, to understand which variables most strongly affected resi lience in a real-world context, multivariate least squares regression wa s used. The regression model incorporated all content areas including demographics and examin ed their effect on resilience while accounting for their impact on each other. The final redu ced model filtered out all non-significant variables and presented a clear picture of t hose variables that most strongly influenced resilience in this sample. These analyses have helped to elucidate re lationships between gende r role conflict, its component patterns, and resiliency in adolescent/eme rging adult males. All of these results have important implications for the research questions presented in the first chapter as well as for future research, practitioners working with male youth, and pub lic policy. The importance and ramifications of these findings will be presented in the concluding chapter.

PAGE 66

66 CHAPTER 5 DISCUSSION The purpose of this study was to examine the ef fect of gender role conflict on resilience in adolescent/emerging adult males. The transition to college marks the final stages of adolescence and the beginnings of emerging adulthood for most youth (Arnett, 2007). This transition can be a tumultuous period of change for youth, and perhap s one of the final periods of developmental vulnerability associated with a dolescence. Gender identity de velopment continues through this period, and sexuality exploration and disclosure (for males) ha ve been shown to occur around this period as well (Kulkin, Chauvin & Percle, 20 00). It is therefore important to consider potential areas of vulnerability or difficulty for male youth related to these issues. Results of this study indicate that gender role conflict is one of those speci fic stressors that may decrease resilience, thus increasing propensity for problematic behaviors among adolescent/emerging adult males. This relationship is an important but complex one that re quires further research, especially over different developmental age rang es, as it suggests programmatic considerations as well as potential policy changes in the areas of anti-discrimination, hate crimes, and legislation. This concluding chapter will begin by answering research questions identified in the first chapter, followed by a brief discussion of other findings not specifically related to the research questions. An interpretation of the research findings of this study will follow, including theoretical implications related to the literature review. Thirdly, suggested directions for future research stemming from this study as well as programmatic and practitioner recommendations will be discussed. Finally, policy implications w ill conclude the thesis in order to evoke more macro-level change that may, over time, ha ve significant effects on youth resilience.

PAGE 67

67 Revisiting Research Questions RQ1: Is there a relationship between Gender Role Conflict (GRC) and Overall Resilience in Adolescent/Emerging Adult Males? This research question was the main thrust of the study, and varying levels of analysis provided insight into the answer. The alternative hypothesis for th is research question was that the more gender role conflict that adolescent/eme rging adult males experience, the less overall resilience they will report. Correlations reveal ed a generally inverse relationship, although only the relationships with Empowerment (EMP) and So cial Competencies (SOC) were significant. Further, ANOVA points to a more direct rela tionship between the overall gender role conflict score and resilience. These statisti cal tests showed that total GRC score most significantly negatively impacted the Empowerment (EMP) area of resilience. The general pattern of bivariate results remained that as gend er role conflict increased, resilience decreased. Finally, the multivariate models assessed th e four component patte rns of gender role conflict and their effect on resilien ce in terms of total DAP score as well as External and Internal asset scores. These analyses revealed an i nverse relationship (with the exception of the SPC pattern) between gender role confli ct and resilience in adolescent/ emerging adult males; thus, the null hypothesis that there was no relationship betw een gender role conflict and resilience was rejected. This supported the hypothesis that gender role co nflict negatively affect ed resilience. As gender role conflict has pr eviously been shown to be associ ated with many negative outcomes, this is not surprising but is nonetheless important Rather than relating gender role conflict to one or a few outcomes, this study used gende r role conflic t to predict resilience in adolescent/emerging adult males. This allows re searchers to relate gende r role conflict to the numerous problematic behaviors al ready associated with overall re silience. Further, by relating

PAGE 68

68 gender role conflict to specific patterns within res ilience, as in the bivariate section of analysis, closer ties to specific problema tic behaviors can be seen thro ugh the relationships that exist between these problematic behaviors and resiliency patterns already esta blished in resiliency literature. RQ2: Is there a relationship between the Succe ss, Power and Competition (SPC) pattern of Gender Role Conflict and Overall Resilien ce in Adolescent/Emerging Adult Males? The alternative hypothesis for this research que stion was that the more gender role conflict with Success, Power, and Competition that adol escent/emerging adult males experience, the less overall resilience they will report. The Success, Power and Competition (SPC) pattern of gender role conflict showed no significan t relationships to overall resili ence in both the correlational and the ANOVA tests. The multivariate analyses elucidated more complex relationships. The SPC pattern, when entered individually into regressi on models to ascertain the eff ect of this pattern on overall resilience as well as Internal and External asset scores, showed no significant effect. Interestingly, though, this pattern was positively re lated to resilience in the reduced models for overall DAP score (standardized beta = .181) as well as for Exte rnal (standardized beta = .154) and Internal (standardized beta = .181) asset scores. Thus, the null hypothesis, namely that SPC has no effect on resilience, was rejected; how ever, results did not support the alternative hypothesis, namely that the effect of SPC on re silience would be negati ve. The three reduced models indicate the opposite, that the SPC pattern of gender role conflict is associated with increased resilience and may in fact serv e as a protective factor for male youth. This finding is important in that it is unexp ected. The gender role conflict literature has shown that generally, gender role conflict can be a danger to male youth. While this study in part confirms those findings, results indicate that the SPC pattern of gender role conflict may not

PAGE 69

69 have negative effects on young me n; conversely, it suggests that high SPC pattern scores are associated with more resilience, and thus a d ecreased propensity for problematic behaviors. Further research is needed to examine the specific effects of this pattern of gender role conflict. RQ3: Is there a relationship between the Restri ctive Emotionality (RE) pattern of Gender Role Conflict and Overall Resilience in Adolescent/Emerging Adult Males? The alternative hypothesis for this research que stion was that the more gender role conflict with Restrictive Emotionality (RE) that adol escent/emerging adult males experience, the less overall resilience they will report. Across each le vel of analysis, this relationship continued to appear as more significant than most. Speci fically, RE was significan tly negatively correlated with seven of the eight factors of resilience as well as Internal Asset score, External Asset score, and overall score. More specific relationships were illumi nated through ANOVA. High levels of RE consistently translated to lower levels of resilien ce for five out of the eight factors, as well as external and overall asset scores. Finally, the least squares regres sion model showed this patter n as consistently important, explaining over 6% of the variance in overall DAP score independent of ot her variables (adjusted R2= 0.062). In the reduced model, this singular score was the most si gnificant in explaining resilience level (sta ndardized beta = -0.338). Further analys is showed that RE pattern score, when entered individually into regression mode ls, negatively affected both External (adjusted R2= 0.058) and Internal (adjusted R2= 0.043) asset scores; moreover, this relationship was the most significant in the reduced models for both External and Internal asset scores as well (standardized betas = -.329 and -.284, respectively). A strong inverse relationship suggested by the bivariate analyses was confirme d through this multivariate analysis.

PAGE 70

70 Throughout this study, then, higher Restrictiv e Emotionality (RE) scores meant less resilience overall for adol escent/emerging adult males. As su ch, the null hypothesis was rejected clearly this pattern of gende r role conflict means a decrease in overall resilience, which supported the alternative hypothesis for this research question. As the theory suggests, this finding confirms that fear of emotional disclosure generally has negative effects on males. As males are more unwilling to disc uss feelings, or as they feel increasing pressure to internalize those emotions, they experience a sharp de crease in resilience. It is thus important to explore ways in which males may feel less negati ve about expressing their emotions or situations in which ma les may be most likely to do so. RQ4: Is there a relationship between the Rest rictive Affectionate Behavior Between Men (RABBM) pattern of Gender Role Conflict and Overall Resilience in Adolescent/Emerging Adult Males? The alternative hypothesis for this research que stion was that the more gender role conflict with Restrictive Affectionate Behavior Between Men (RABBM) that adolescent/emerging adult males experience, the less overall resilience they will report. Analysis for this question began with correlations, in which this pattern of gender role conflict was signifi cantly inversely related to two factors within resilience: Boundaries & Expectations (B &E) and Commitment to Learning (CTL). Thus, as RABBM conflict increase, resilience declined. To delve into more specific relationships, ANOVA was again used. For these tests, high levels of Restrictive Affectionate Behavior Between Men (RABBM) were associated with less resilience in two of the eight resilience factor s of resilience, namely Commitment to Learning (CTL) and Social Competencies (SOC). The multivariate model was then used to bett er ascertain the effect of RABBM score on resilience. The RABBM model alone accounted fo r less than one percent of the variance in overall DAP score (adjusted R2 = 0.004) as well as External and Internal asset scores (adjusted

PAGE 71

71 R2 = 0.003 and 0.002, respectively). Across all thr ee regression models, RABBM score was not significantly related to resilience in the reduced models. RABBM is a complex pattern with items th at males can interpret in many ways. Comments from participants about this particul ar factors items generally represented two viewpoints: that touching other men is not necessa rily related to sexuality, or that perhaps this instrument may be better suited to homosexual individuals based on the items included. Based on the interpretation of the ite ms, respondents may have provided very different answers. Overall, though, one would expect higher degrees of conflict to cause a decrease in resilience in males based on the theory. In this study, the regr ession models, the most rigorous tests, did not provide sufficient evidence to reject the null hypothesis, namely that RABBM score does not affect resilience. Further resear ch is needed to more closely study ties between these variables, especially while considering re spondent interpretation of the meanings of this patterns component items. RQ5: Is there a relationship between the Co nflict Between Work and Family Relations (CBWFR) pattern of Gender Role Conflict and Overall Resilience in Adolescent/Emerging Adult Males? The alternative hypothesis for this research que stion was that the more gender role conflict with Conflict Between Work and Family Rela tions (CBWFR) that adolescent/emerging adult males experience, the less overall resilience they will report. When independent variables were correlated with resilience, CBWFR was ne gatively correlated with Support (SUP), Empowerment (EMP), and Boundaries & Expectations (B&E). This meant that as gender role conflict of this type increased, resilience in th ese specific external asset categories decreased. Additionally, there was also a negative correlation found between CBWFR and Internal Asset score; thus, this patter n of gender role conflict affected not only the aforementioned external patterns but also had an impact on internal assets.

PAGE 72

72 After establishing that a rela tionship did exist between CB WFR and resilience generally, ANOVA was used to explore specif ic relationships within these two conceptual areas. The CBWFR pattern of gender role conflict most significantly affected the resilience areas of Empowerment (EMP) and Social Competencies (SOC ). Again, as gender role conflict of this type increased, resilience decreased. When entered independently into regression models testing CBWFR pattern score effect on overall DAP score and External Asset score, th is pattern score accounted for less than 1% of the variance in resilience (adjusted R2 = 0.004 and -0.002, respectively). The reduced models in both of these tests failed to reveal a significant impact of CBWFR on resilience. Interestingly, though, CBWFR pattern score did significantly affect Internal Asset score independently (standardized beta = -.107) accounting for almost 1% of the variance (adjusted R2 = 0.009); however, this relationship did not remain signifi cant in the reduced model. Overall, then, CBWFR score did not significantly contribu te to a change in resilience. The null hypothesis for this research question, namely th at CBWFR had no effect on resilience, could not be rejecte d, as the regression model did not provide sufficient evidence of change in resilience due to this pattern of gender role conflict. Conflict Between Work and Family Relations (CBWFR) did have a negative effect on overall resilience; however this effect when comp ared with the other factors within gender role conflict, was the least significant in explaining ch ange in resilience. This may be due to the population sampled. Many undergraduates do not work; however it is possible that those male respondents that were employed were different from those that were not. This was not addressed in this study. Further, most undergraduates ar e separated geographically from their families, which may result in less conflict wi th this pattern of ge nder role conflict. Finally, this study did

PAGE 73

73 not consider whether these respondents might be married or have children, which might also affect this pattern; however, this probably affect s few respondents in this age range. While this factor was not particularly important in this gr oup, it may be more predic tive of resilience in males of different ages or different circumstan ces, such as college males that live with their families and commute for educational purposes. RQ6: Is there a relationship between Demogr aphic Variables and Overall Resilience in Adolescent/Emerging Adult Males? The alternative hypothesis for this research question was that ther e is a relationship between demographic variables and overall resilience in adol escent/emerging adult males. Correlations showed that the spec ific demographic variables were related to resilience. General patterns found in correlational an alysis with demographics and resilience appeared as follows: An increase in sexual orientation level (less ex clusively heterosexual) was associated with an increase in resilience, specifica lly within 3 of the 8 individual resilience factors, two of which were significant; this increase in sexual orientat ion level was also linked with an increase in overall resilience, but this relationship was not significant. Generally, then, less exclusively heterosexual adolescent/emerging adult males showed more resilience. While this portion of the study lacks some explanatory power due to the grouping methods and small sample sizes of sexual minority groups important relationships are still revealed. Sexual minority individuals may be more resilient due to an existing nonconformity to gender roles. That is, if these males, through sexuality experimentation and/or disclosure, have already encountered conflicts with tradit ional gender roles, they may be less affected by these societal prescriptions subsequent to this. Additionally, it may be that sexual minority males internalize these gender roles less, as they presumably woul d not want to restrict affectionate behavior between men, for example, from an early age. It is therefore important to further explore this

PAGE 74

74 minority group to determine why this sexual minority status is associated with increased resilience. Increases in academic status was associated with a decrease in resilience within 7 of the 8 individual resilience factors, three of which were signifi cant; academic status was also significantly inversely related with overall resilience. The resilience literature specific to the DAP points out a decrease in overall resilience ov er the high school years; similarly, the college progression seems to take its toll on resilience as well. Also, as age increased in this sample, resilience in 6 of the 8 individual resilience f actors decreased, and th ree of these inverse relationships were significant; age was also negatively correlated with overall resilience, but not significantly. As academic status and age have a high degree of overlap, many of the same causal relationships probably explain these relationships. Possible reasons for the inverse relationshi p between both age and academic status and resilience include external factors, such as chan ging peer groups, graduating friends, or increased stress levels at the conclusion of degree programs. Internal factors, though, probably play a larger part in this process. As adolescents transition to emerging adulthood, they become more autonomous, depending less on parents and/or caregi vers and shouldering increased levels of responsibility. This may be associ ated with higher levels of stress, weaker ties to family and/or friends, and possible hardship as inexperience w ith life management may lead to less-thandesirable outcomes. Any of these factors may lead to the decrease in resilience associated with academic status and age found in this study. Different race groups showed different re lationships between race and resilience. black/African American race status and white race status were both positively correlated with 5 of the 8 individual resili ence factors. Hispanic race was nega tively correlated with 6 of the 8

PAGE 75

75 individual resilience factors; however one of the two positive correlations (CTL) was significant as well. Those reporting Other race status showed negative correlations with 7 of the 8 individual resilience factors. Race and overall resilience were positively correlated for Black/African American and White race groups and negatively correlated for Hispanic and Other race groups; however, none of these re lationships were significant. There are many possible explanations for the differences seen in resiliency by racial category. One major consideration is the setting of this study. For example, in this environment, even the list of student organiza tions shows a definite unified black/African American culture and a comparatively divided Hispanic culture base d on country of origin. Also, the Other race category had many component groups By diversifying the sample or increasing sample size, more concrete relationships between specifi c minority race groups and resilience may be explored. Additionally, the backgrounds, SES information, and upbringing probably varies by race and would have a tremendous impact on resili ence. Finally, the differences seen here may be indicative of how well-received these race groups are by the campus and geographic communities. Research at different types of inst itutions, such as Historically Black Colleges and Universities, or in different locations, such as major urban areas, may show very different relationships. ANOVA was used, where possible, to further sp ecify any effects that demographics had on resilience. No sexual orientation effect on re silience was found. Increases in academic status translated to decreased resilience, and significantl y so in two of the eight factors (SUP and B&E) as well as Internal Asset score. For all three f acets of resilience, post hoc tests showed that the major difference existed between freshmen/sophomores, who were more resilient, and seniors, who tended to be less resilient. This showed the tendency of assets to decrease over the

PAGE 76

76 academic progression in college, which mimics th e literature on high school adolescents. ANOVA revealed significant relationships between race and resilience, specifically the CTL and PV patterns as well as both internal and external a sset score. Post hoc test s revealed that blacks generally displayed more resilien ce in the Positive Values factor than did whites or Hispanics. Additionally, for internal asset sc ore, post hoc tests showed Asians generally fared less well than did those respondents reporting white race. In the multivariate model, as discussed in th e previous chapter, age was excluded due to high rate of nonresponse by participants, presumab ly due to less than optimal instrumentation design. Multivariate models were run both with and without age for each test; the models were nearly identical, with the only noteworthy difference being that age sometimes appeared as significant in the independent and reduced mode ls when it was included in analysis (See Appendix B). Both models for each test appear in Appendix B, but this study used the models that excluded age, as this vari ables inclusion resulted in a si gnificantly reduced number of cases but a very similar model. The multivariate model testing for demographic effect on overall DAP score did not show any significant effect of demographic variable s on resilience. Independently, demographics accounted for about 1% of the variance (adjusted R2 = 0.012). When combined with gender role conflict and reduced, no signifi cant effect was found between demographics and overall resilience. For the External A sset score regression model, black race was significant in both the independent and reduced models (standardiz ed betas = .146 and .176, respectively), while no other demographics were significant. In the In ternal Asset score regression model, both Other race and academic status were si gnificant in the demographic model (standardized betas = -.114

PAGE 77

77 and -.166, respectively); however, only academic st atus remained significant in the reduced model (standardized beta = -.147). The null hypothesis, namely that demographics have no effect on resilience, was rejected due to the regression models invo lving both External and Internal Asset scores analysis; however the effect that demographics di d have is not as great as othe r variables in this study. While demographics did not play a large role in the present study, further exploration of gender role conflict and resilience th rough samples with entirely different demographic characteristics is needed to determine whether thes e results are sample-specific. Age, for instance, was often omitted here and thus was excluded, but may be si gnificant. As discussed before, race status may mean different things in different educatio nal and/or geographic communities. While these are only two of myriad examples, it is clearly important that further studies examine these relationships more closely so as to determine the effects of demographic variables on overall resilience. Other Findings Beyond the specific research questio ns discussed thus far, further analysis determined the effect demographic variables had on gender role conflict. If there were relationships between demographics and the independent variable, changes in resilience c ould be mistakenly attributed to an incorrect original source. Significant results from correlati on analysis are discussed below. As self-reported sexuality scor e increased (further from excl usive heterosexuality), gender role conflict tended to decrease, particularly within the Restrictive Emotionality (RE) and Restrictive Affectionate Behavi or Between Men (RABBM) pattern s, as well as overall gender role conflict score. This is interesting but not surprising, since, as discussed above, sexual minority individuals may be less likely to interna lize societally prescrib ed gender roles or may have encountered conflicts akin to gender role conflict due to thei r sexuality. While this may at

PAGE 78

78 least partially explain this rela tionship, further in-depth explora tion is needed to determine why this inverse relationship exists. As academic status increased, gender role c onflict decreased, especially within the Success, Power and Competition (SPC) pattern, wh ile age showed no significant relationships. As students progress through the co llege years, they may develop a more accurate or realistic picture of the job market, career success, the va lue of competition, and ot her aspects related to this pattern. This could be the result of expos ure, education, maturation, or a combination of these. This transition in level of gender role conflict over the college years and its associated decrease in resiliency in this population warrant further study to identify specific processes at work. Black/African American race was associated with a significant in crease in overall gender role conflict and three of th e four individual patterns of ge nder role conflict; respondents reporting Other race tended to have increased over all gender role conflict as well as increased conflict with two of the indivi dual patterns. White race was a ssociated with decreased gender role conflict overall and within two of the four specific patterns of gender role conflict; Hispanic race was associated with decreased gender role conflict with onl y the RABBM pattern. As discussed in terms of resilience, relations hips between race stat us and other variables may be difficult to explain. For gender role conflict, though, it is wort h mentioning, in addition to those factors previously addr essed, that cultural values may heavily influence the effect of race on gender role conflict. Expectations of males within specific cultures, support or condemnation of certain behaviors, and other cultural attributes may affect the degree to which males internalize these ge nder roles. This, in turn, would ha ve a strong influence on gender role conflict.

PAGE 79

79 To further explore the associations betw een demographic variables and gender role conflict, ANOVA was again used. Significant results were found among several of the demographic variables when related to gender role conflict, including: Sexual Orientation, with highe r sexuality score groups (le ss heterosexual) reporting less gender role conflict within the RABBM and overall gender role c onflict. Also, the RE pattern showed significant differences between groups. This was consistent w ith correlational findings discussed above, although this test fu rther specified these relationships. Academic Status was again related to the SPC pattern of gender role conflict. Specifically, the sophomores and juniors experienced significan tly higher levels of ge nder role conflict with this pattern than did the seniors. Age was also related to gender role c onflict, with 20-year-olds experiencing much more gender role conflict ov erall and within the SPC patterns than their 22year-old counterparts. This test supported the correlational findings of decreasing gender role conflict over time in the college years. Race was related to both the RABBM and CBWF R patterns as well as overall gender role conflict Post hoc tests revealed th at black respondents tended to have higher RABBM scores than Hispanics and higher overall GRCS scores than both Hispanic and white males. Overall, these tests show that demographics did have an effect on gender role conflict. This was not included in the res earch questions, and this relati onship between demographics and gender role conflict was not specifically entered into the regression model. These tests were simply to assist in understanding the complex relati onships that exist between all variables in this study. Implications of Research Beyond these specific research questions, th is study has practi cal and theoretical implications, both for developmental science resear ch at large and for the specific areas of study

PAGE 80

80 related to gender role conflict and resilience. To provide a context for these implications, the theoretical perspectives offered by Gender Role Conflict Theory and Resilience Theory must be re-examined. Gender Role Conflict Theory The purpose of this study was to examine gender role conflict and its effect on resilience in adolescent/emerging adult males. Results of th is study show that gender role conflict does have an effect on resilience, and that this effect is generally, although not exclusively, a negative one. A more in-depth look at gender ro le conflict reveals mixed results, with some patterns related to an increase in resilience, whereas others resulted in decreased resilience. These results have important implications for the theoretical paradigm of gender role conflict discussed in chapter three. Gender role conflict has been established throug h research to be related to low self-esteem, lower intimacy, anxiety, depression, relationship dissatisfaction, sexual a ggression and hostility toward women and negative attitudes toward hom osexuals (ONeil, 1995). Since adolescent and general developmental science re search has emphasized a more positivistic approach (Galambos & Leadbeater, 2000), this study attempted an examination of gende r role conflict in adolescent/emerging adult males through a resili ence perspective. Rather than looking at specific problematic behaviors, this study examin ed youth resilience in te rms of developmental assets, which has been linked throug h the literature to propensity to engage in or steer clear of problematic behaviors (Benson, 2002). This wide r scope paints a broad picture of positive youth development rather than focusing on one specific problematic behavi or or a small set of problem issues. This study examined the effect of gende r role conflict on resilie nce, which, along with resilience literature, will assist researchers in de termining which effects of gender role conflict to

PAGE 81

81 be most concerned with as well as which proble matic behaviors these de pleted resilience areas are most associated with. The results of this study seem to be in accordance with past research on gender role conflict. While this design did not examine speci fic problem behaviors, it did show that as gender role conflict increased, adolescent/emerging adult males did have an increased risk for a multitude of problematic behaviors in terms of a lack of resilience. By examining specific patterns of gender role conf lict, analysis revealed several interesting relationships with resilience. The pattern of gender role conf lict that most negatively affected resilience was Restrictive Emotionality (RE). This pattern was defined as having difficulty a nd fears about expressing ones feelings and difficulty finding words to ex press basic emotions, (ONeil, 1995, p. 175). The more intense these difficulties and/or fears were the less resilience overall the males in this sample reported. This relationship has importa nt implications, especially considering the findings of the Beatty et al. study (2006), which showed that intervention providing information and discussion on gender role confli ct lessens the self-report RE scores in college-age males. Neither Conflict Between Work and Family Relations, or experiencing difficulties balancing work-school and family relations, resulting in health problems, overwork, stress, and a lack of leisure and relaxation nor Restrictive Affectionate Behavior Between Men, or having limited ways to express ones f eelings and thoughts with other men and difficulty touching other men significantly affected overall res iliency score (ONeil 1995, p. 175). The pattern of Success, Power and Competition (SPC), though, was more complex. This pattern is defined in component parts as: Success: Persistent wo rries about personal achievement, competence, failure, status, upward mobility, and wealth, and career success; Power: Obtaining authority, dominance, or ascendancy over others; Competition: Striving

PAGE 82

82 against others to gain somethi ng or the comparison of self w ith others to establish ones superiority in a given situation. Higher conflict scores with this pattern were actually associated with an increase in resilience, which has many ramifications for bot h gender role conflict theory and positive youth development, which promotes youth resilience. Overall, this study found that high levels of ge nder role conflict resu lt in a decrease in resilience. Due to the exploratory nature of this study, it was impossible to devote substantial analysis to every factor. Due to this studys significant findings, however, it is important for future researchers to examine more carefully the relationships related to resilience by pattern of gender role conflict. Contributions to Gender Role Conflict Theory This study contributes to the l iterature surrounding ge nder role conflict in several specific ways. First, it expands the study of gender role conflict, using resilience as a representation of adolescent/emerging adult male vulnerability or propensity for problematic behavior. The use of resiliency rather than problematic behaviors in research related to gender role conflict can more completely identify many risks to or strengths of the population of males. Additionally, due to the large body of literature on resilience, this pe rspective also will show insight into variables associated with problematic behaviors. For instan ce, the positive identity factor of resilience has been shown to relate to propensity for suicide th rough the literature; thus, a low score in this area associated with gender role conf lict suggests a potential relationship there. This is one of many examples in which resilience literature and re search might expand the study of gender role conflict. Additionally, this study begins to examine gende r role conflict in sexual minority males. A major criticism of studies of gay males as a group is that these in dividuals are recruited

PAGE 83

83 through gay-supportive organizatio ns, resulting in a non-represen tative sample of the sexual minority. This study suffered from a small number of sexual minority males in the sample, but it supports an idea shift in recruitm ent procedures for the study of se xual minority individuals. The recruitment process used in this study may result in a more re presentative literature base in regards to this population. S uggestions for further researc h, programmatic and practitioner recommendations, and policy comments regarding gender role conflict will follow below. Resilience Theory As discussed in chapter three, the Developm ental Assets Profile (DAP) has been used by many researchers to assess youth resilience in gene ral. Results from ge neral populations of high school youth have shown low means of 19.3 assets on average in the a dolescent population and a decrease in number of assets from 23.1 to 18.3 in grades six through twelve (Benson, 2002). In this study, the DAP, usually used on 11-18 year ol ds, assessed resilience in 18-24 year old males after ensuring internal consistency. Interes tingly, in this college-age male population, the average DAP score was 40.8, a much higher resi liency level than the average high school adolescent population reported above. The patter n showing a decreasing number of assets over academic progression held true in this study, however as increases in acade mic status in college predicted a decrease in three factor scor es (Support, Empowerment, and Boundaries & Expectations) as well as the ove rall Internal Asset score. In the adolescent populati on typically responding to the DAP, a high number of developmental assets (31-40) were correlated with high academic performance (Benson, 2002). The individuals in the present study have performed well academi cally, or at least well enough to enroll at a competitive university, which may explai n the higher mean resiliency score. Trends in resiliency fluctuation as a result of gender role conflict would probabl y not be significantly affected by this sample composition, although this is certainly a possible area for future research.

PAGE 84

84 Much research has been conducted on yout h resilience, in particular using the Developmental Assets Profile (DAP). Gender role conflict, as predicted, did have an effect on overall resilience in adolescent/emerging a dult males. As detailed above, though, these relationships are complex. The ge neral findings suggest that as gender role conf lict increases, resilience in adolescent/emerging adult males decreases, although this is not true for all items within the gender role conflict scale. This al ludes to a more complex relationship worthy of further research. Contributions to Resilience Theory This study contributes to the ex isting resilience literature in se veral ways. First, this study expanded the studied population beyond the range of 11-18 years. By first examining the responses of 18-year-old particip ants and comparing them with t hose of their college classmates through age 24, this research created a general profile of resilien ce in adolescent/emerging adult males pursuing their undergraduate education at a research unive rsity. Recommendations made by developmental scientists regarding research in the field support both the use of a resilience perspective and an expansion of the adolescence research arena in to the relatively new area of study encompassed by emerging adulthood (Galambos & Leadbeater, 2000). Additionally, the three main informational goals of the initial resiliency approach included risk behavior prevention, thriving outcome enha ncement, and youth resilience (Benson, 2002). This study and its findings can be used to achie ve all three of these objectives. Prevention programs can be tailored to address gender iden tity development, gender norms, and gender role conflict in many settings from school and youth extracurricular act ivities to parental seminars and information distribution. Now that gender role conflict has been identified partially as a risk factor for male youth, both prevention and interventi on efforts with regard to gender role conflict

PAGE 85

85 can help enhance positive youth outcomes. Fina lly, through an understanding of what specific assets or asset areas are most depleted by conf lict with gender roles, youth workers can help build assets and thus overall resilience in th eir intended populations. Specific suggestions for future research, programs, and public policy stem ming from this research will conclude this chapter. Implications for Future Research, Programs and Public Policy This study has its origin in youth resiliency. An up-and-coming trend in developmental research, the study of youth resilience, can help researchers, practitioners, program developers, and policymakers keep their finger on the pulse of youth development today. This will help ensure that these professionals deliver relevant products based on current research findings in their respective fields. Suggestions based on this specific research and its findings are chronicled below by specific professional area. Future Research Gender role conflict and resilience in adolesce nt/emerging adult males are clearly related, as evidenced by this study. While this research was beneficial in establishing these relationships, a more thorough examination of these relationshi ps would be benefici al in furthering the literature as well as the ramificati ons for youth workers. To this end, a more qualitative research design in which respondents are ab le to give feedback on their sp ecific experiences and attitudes could help explain some of the complex relatio nships found in this study. Perhaps open-ended interview questions could more fully explore these ties. Additionally, respondents in this study answered questions in a classroom setting amidst their peers. While this was a strength in term s of recruiting potentially a full range of sexual minority males rather than the biased sample s generated by support grou p recruitment, some degree of social desirability effect probably played into individual participant response. As the

PAGE 86

86 nature of some of the gender ro le conflict scale as well as the sexuality item were somewhat personal, sometimes even resulti ng in nonresponse, an online survey or private interview may be a more appropriate way to allow participants to hones tly respond to these ite ms without fear of association or lack of confidentiality. Finally, follow-up research on gender role confli ct and its effect on re silience in collegeage sexual minority and transgende r populations would be interesti ng and appropriate to add to the body of literature. A snapshot picture as provided through cross-sect ional research would allow further understanding of what relationshi ps to explore specific to these populations. Practitioner/Programmatic Suggestions The effect of overall gender role conflict on adolescent/emerging adult male resilience is a negative one. As gender identity develops throughout the lifespa n, it is important to understand the implications of societal expectations on youth and the outcomes that these youth may face. While adolescent/emerging adult males were stud ied here, the implications for youth workers extend from childhood through emerging adulthood. Although this study did not focus on developmen tal stages before adolescence, this research suggests that the ri gid gender roles and stereotype s that are formed and adopted throughout childhood can greatly impact resilience in later stages of development. A general message to take away from this study is that th ese societally prescribed roles may actually hurt youth; thus, it is important to encourage exposure to myriad activities, toys, stories, etc. in which these gender roles are not so stringent. Further, as youth spend a majority of their wa king hours in schools, it is imperative that this message be heard in the e ducational setting as well. This can take many forms, including strict anti-hate speech policies, intolerance of blatant ridicule, and increased assistance from guidance administration for those identified as highly conflicted or nonconforming with gender

PAGE 87

87 roles. Additionally, encouraging all youth to enga ge in activities not us ually associated with their gender may help youth explore other in terests or understand ot hers perspectives. Specifically, programs for youth might focu s on male readiness for college beyond academics. Since college-age males showed resilience as Success, Power and Competition (SPC) score increased, practitione rs should empower youth with tools that will assist them in their academic and personal pursuits in college as well as with motivation and support to sustain their drive for achievement. Restrictive Emoti onality (RE) score was shown to be inversely related to resilience. Practitioners might encourage male youth to discuss emotional feelings independently and in a support gr oup setting. This could assist individuals lessen anxiety regarding emotional disclosure individually an d among peers, and could reduce this type of gender role conflict. Completing such a program would support the efforts of parents, teachers, and communities to maximize resilience in young men. While many suggestions from this study on adolescent/emerging adult males have to do with gender development, which starts much earlier there are things practitioners can do to assist college-age males such as those in this sample with issues related to gender role conflict. First, realizing that this period is one in which sexual minority individua ls, particularly males, may be exploring and/or disclosing their sexuality, it is important for youth practitioners including school officials to host open fora in which to discu ss these issues. Interventi ons of this type have been shown to affect at least Restrictive Emotiona lity (Beatty et al., 2006). Further, they would assist both the sexual minority indi viduals involved in these transi tions as well as ensure a more tolerant environment by educating the public at-large about these issues. Further, the importance of individual exposure to activities, events, and discussions not typically aligned with gender norms does not end in pre-college education. The college years

PAGE 88

88 often provide an environment of openness, expe rimentation and exposure. Thus, practitioners within this field should encourage emerging adults to explore the myriad opportunities afforded them both in class and outside of it. This can include but is not limite d to programming within residential communities on college campuses, edu cational information and/or discussion through clubs or organizations, and demons trations in campus areas of hi gh student traffic, such as a student union. Perhaps the faculty may even become involved, working activities, discussion, and video pertaining to this topic into their curriculum. This increased exposure will at a minimum educate youth on these important issues a nd may help to create a climate of tolerance, helping some individuals who would have othe rwise suffered due to gender role conflict. Finally, at all educational levels it is imperative to inform instructors across the age ranges of these relationships. Training in hate speech identification, how these situations should be handled, and appropriate information to dispense during these interactions would benefit teachers and students alike in creating a school climate in which rigid gender roles do not necessarily translate to decreased resili ence. These programs may include important individuals beyond teachers, including coaches, peer mediators a nd mentors, guidance counselors, and other administrators. By creating a more accepting sc hool environment, we may improve problems with bullying, absenteeism, sc hool safety, and victimization. Policy Recommendations Although gender role conflict it self does not bring to mind specific public policies, the societal prescription of traditi onal masculine gender roles trans cends the interrelationships of people and small-scale structures. Public po licy may unknowingly defend the very institutional stigmas that result in this lack of resilience in males. Thus, it is important to incorporate modern findings from studies such as th is one into our public policies to best protect the public they intend to serve.

PAGE 89

89 First, an up-and-coming issue in public policy that serves as an illustrative case in which these findings can be incorporated is that of anti-discrimination clauses. While demographic characteristics such as race and sex are almost always included, it is onl y recently that sexual orientation, gender identity, and ge nder expression have been included in these statements. In fact, Lucent Technologies was the first company to add gender identity and expression to its antidiscrimination clause in 1997 (Joy ce, 2004), and corporations con tinue to add this important factor over time, although slowly. Encouragingly, companies seem to be competing to embrace these nontraditional gender roles, as indicated by the increasing scores on the Corporate Equality Index, a scale used by the Human Rights Campaign. In 2006 alone, almost 100 companies added gender identity/expression to their anti-discr imination clause (Corporate Equality Index, 2006). This organization also encourages divers ity training on these issues, support groups, and providing an accepting environment for all employees By encouraging businesses, educational institutions, and other public operations to in corporate these characte ristics into their antidiscrimination clauses, a strong public policy statement will be made in support of a more fluid interpretation of gender independent of sex. This may help those w ith more strict interpretations of gender and sexuality consider a more openminded approach to these issues and, more importantly, may extend needed recognition to those that do not fit the societally prescribed mold in terms of gender and/or sexuality. Additionally, through recognition of the stigmas potentially associated with nonconformity to gender roles and/or sexual orient ation status, it is important th at public policy weighs in on the anti-hate speech/hate crime traini ng suggested to practitioners above particularly as related to adolescents and emerging adults. The government has begun to realize th e importance of these issues. On May 3, 2007, in a 237 to 180 vote, th e U.S. House of Representatives passed the

PAGE 90

90 Local Law Enforcement Hate Crimes Prevention Act, which will strengthen the ability of federal, state and local governments to investig ate and prosecute hate crimes based on race, religion, color, national origin, ge nder, sexual orientation, disabil ity and gender identity, (U.S. House of Representatives, 2007). The challenge that will arise is how to deal with these crimes committed on or by youth. Instituting a plan by which these issues and offenders would be handled (most probably an educ ational rather than punitive c onsequence) ensures that these issues are not to be dealt with lightly and that la rger organizations are taking this issue seriously, valuing both the substan ce of these matters and th e people affected by them. Finally, policymakers must examine public policy overall to ensure consistency of message. Specifically, if they are to encour age incorporation of se xual orientation, gender identity and/or gender expression in their anti-discrimination clau ses as discussed above, they must be sure that their policies do not discriminate based on these f actors as well. Issues covered can range from the seemingly inconsequential, such as the modification of standardized governmental forms, to the much more controve rsial and politicized issue such as marriage and/or domestic partner rights. While thes e suggestions are quite long-range, they are nonetheless important if we are to establish a society in which one does not have negative developmental outcomes as a result of nonconform ity to societally pres cribed gender roles. Developmental science is, as is public policy, a work in progress, but one in which every carefully contemplated effort can have treme ndous effects in terms of supporting positive youth outcomes. Summary Issues surrounding gender role s including gender role confli ct, gender identity, gender expression, and sexuality are cutting-edge issues in society today. In news, politics, public

PAGE 91

91 policy, higher education, as well as in th e corporate world, these topics are becoming increasingly common, and they have important ramifications. Recognizing that issues relate d to gender have origins within gender identity formation, it is necessary for youth researchers to examine how these variables relate to children, adolescents and emerging adults. This study examined gender role conflict and resilience in adolescent/emerging adult males. The results of this study have many possible implications for youth researchers, practitioners, policymakers and advocates. This research aimed to elucidate relationships between these variab les on a large scale, as no rela tionship between these variables had previously been established. To maximize po sitive youth development, this study aims in its finality to spark further research and program matic development to best serve male youth, especially those who do not conf orm to traditional masculine ge nder roles and may have poorer outcomes because of this.

PAGE 92

92 APPENDIX A INSTRUMENTATION A brief explanation surrounding each of the component parts of the instrumentation packets as well as figures of the actual instrument follow. Explanations Informed Consent Letter This was page one of the instrumentation pa cket and was approved by the Internal Review Board (IRB) for this research. Potential responde nts were instructed to read the consent form and, if they voluntarily chose to pa rticipate, to tear this page off and keep it for their records. Developmental Asset Profile (DAP) The DAP was purchased from the Search In stitute and is a manila, two-sided form complete with instructions a nd prompts for respondents. Th e DAP comprised pages two and three of five in the in strumentation packet. In this study, th e Name, Grade, and Birth Date items were blacked out so as to preserve respondent confidentiality. Gender Role Conflict Scale (GRCS) Instructions read as follows: In the grid below please mark the box corresponding to the number which most closely repres ents the degree that you Agree or Disagree with the statement. There is no right or wrong answer to each stat ement; your own reaction is what is asked for, (ONeil, 1982). This was page four of five of the instru mentation. The bottom of the sheet asked respondents to turn over and complete the back of the form. The origin al layout of the GRCS was changed to a grid and the scalar responses reversed for ease of completion and seamless transition from the DAP to this instrument.

PAGE 93

93 Demographic Questions This was page five of five of the instru mentation packet. The packet ended with demographic information to minimize respondent fatigue, but obtained information used in sorting, excluding, or categorizing responses to the DAP and GRCS. Figure A-1. Informed Consent Letter

PAGE 94

94 Figure A-2. Developmental Asset Profile, Page 1

PAGE 95

95 Figure A-3. Developmental Asset Profile, Page 2

PAGE 96

96 Figure A-4. Gender Role Conflict Scale

PAGE 97

97 Figure A-5. Demographics Questions

PAGE 98

98 APPENDIX B RESULTS TABLES Table B-1. DAP summ ary statistics incl uding context areas FrequencyMean Range Std. Dev. Alpha Support (SUP) 363 21.464 5.71 72.86 6.050 0.762 Empowerment (EMP) 363 22.525 5.00 30.00 4.512 0.694 Boundaries & Expectations (B&E) 363 20.596 6.67 30.00 5.011 0.792 Constructive Use of Time (CUT) 363 14.963 0.00 30.00 6.486 0.442 External Asset Score (EA) 363 20.898 8.61 30.00 3.808 0.891* Commitment to Learning (CTL) 363 21.578 5.71 30.00 4.705 0.743 Positive Values (PV) 363 19.813 9.09 30.00 4.500 0.805 Social Competencies (SOC) 363 20.778 8.75 30.00 4.277 0.712 Positive Identity (PID) 363 21.424 1.67 30.00 5.075 0.816 Internal Asset Score (IA) 363 19.887 5.96 37.31 4.435 0.908* Personal 363 20.195 8.46 30.00 4.177 0.797 Social 363 21.421 10.00 30.00 4.086 0.823 Family 363 22.537 5.00 30.00 5.256 0.857 School 363 21.888 8.00 30.00 4.494 0.802 Community 363 17.363 8.00 30.00 4.831 0.772 Total Score (DAP) 363 40.786 18.85 58.73 7.419 0.926* Alpha values reported for component items, not scores Table B-2. Pearsons r values for DAP factors, Internal/E xternal asset scores, and total score SUP EMP B&E CUT CTL PV SOC PID IA EA DAP SUP 1 0.647 0.792 0.434 0.288 0.409 0.422 0.401 0.869 0.462 0.751 EMP 1 0.641 0.430 0.412 0.548 0.529 0.576 0.805 0.630 0.801 B&E 1 0.411 0.332 0.462 0.489 0.404 0.854 0.511 0.768 CUT 1 0.244 0.561 0.395 0.368 0.737 0.475 0.680 CTL 1 0.504 0.563 0.569 0.383 0.805 0.642 PV 1 0.679 0.511 0.612 0.812 0.780 SOC 1 0.573 0.557 0.846 0.766 PID 1 0.527 0.821 0.735 EA 1 0.631 0.916 IA 1 0.889 ** All values significant at the 0.01 level (2-tailed) Table B-3. Pearsons r values for GRCS factors and total score SPC RE RABBM CBWFR Total SPC 1 0.396** 0.406** 0.395** 0.783** RE 1 0.583** 0.380** 0.795** RABBM 1 0.278** 0.760** CBWFR 1 0.619** Total Score 1 ** Significant at the 0.01 level (2-tailed)

PAGE 99

99 Table B-4. Pearsons r values relating demographics and external resilience SUP EMP B&E CUT EA Sexual Orientation -0.048 -0.010 -0.005 -0.100 0.096 Academic Status -0.186** -0.111* -0.172** -0.076 -0.028 Age -0.157** -0.068 -0.132* -0.118* -0.020 Black -0.023 -0.056 -0.067 0.088 0.150** Hispanic -0.079 -0.024 0.028 -0.110* 0.011 White 0.171** 0.127* 0.107* 0.015 -0.060 Other Race -0.131* -0.094 -0.106* 0.003 -0.072 Significant at the 0.05 level (2-tailed) ** Significant at the 0.01 level (2-tailed) Table B-5. Pearsons r values relating demographics and internal resilience CTL PV SOC PID IA Sexual Orientation 0.126* 0.104 0.133* -0.032 -0.056 Academic Status -0.024 0.011 -0.029 -0.047 -0.164** Age -0.038 0.007 0.004 -0.034 -0.149** Black 0.113* 0.185** 0.135* 0.068 -0.008 Hispanic 0.106* -0.037 -0.003 -0.030 -0.064 White -0.064 -0.098 -0.064 0.020 -0.123* Other Race -0.122 -0.008 -0.040 -0.061 -0.095 Significant at the 0.05 level (2-tailed) ** Significant at the 0.01 level (2-tailed) Table B-6. Pearsons r values correlating demographi cs and Gender Role Conflict SPC RE RABBM CBWFR Total Sexual Orientation -0.048 -0.116* -0.251** -0.047 -0.154** Academic Status -0.110* -0.009 -0.012 -0.007 -0.058 Age -0.052 -0.042 0.018 0.001 -0.036 Black 0.094 0.122* 0.134* 0.108* 0.150** Hispanic -0.041 -0.030 -0.127* -0.095 -0.095 White -0.114* -0.105* -0.059 -0.091 -0.120* Other 0.105* 0.055 0.074 0.111* 0.110* Significant at the 0.05 level (2-tailed) ** Significant at the 0.01 level (2-tailed) Table B-7. Pearsons r values correlating Gender Role Conflict and external resilience SUP EMP B&E CUT EA SPC 0.051 0.055 0.063 0.077 0.040 RE -0.185** -0.284** -0.190** -0.076 -0.247** RABBM -0.088 -0.136** -0.082 0.045 -0.077 CBWFR -0.117* -0.142** -0.152** 0.026 -0.032 Total -0.096 -0.151** -0.097 0.029 -0.099 Significant at the 0.05 level (2-tailed) ** Significant at the 0.01 level (2-tailed)

PAGE 100

100 Table B-8. Pearsons r values correlating Gender Role Conflic t with internal asset factors and overall resilience CTL PV SOC PID IA Total SPC 0.036 -0.009 0.018 0.080 0.077 0.066 RE -0.165** -0.155** -0.240** -0.248** -0.214** -0.254** RABBM -0.106* -0.026 -0.081 -0.040 -0.070 -0.081 CBWFR -0.036 0.020 -0.020 -0.065 -0.107* -0.080 Total -0.081 -0.059 -0.105* -0.080 -0.086 -0.102 Significant at the 0.05 level (2-tailed) ** Significant at the 0.01 level (2-tailed) Table B-9. Significant ANOVA re lationships between academic status and mean resilience Freshman Sophomore Junior Senior F P-value Support 22.388 21.797 21.150 19.583 4.505 .004 Boundaries & Expectations 21.149 21.531 20.490 18.841 4.960 .002 Internal Asset Score 20.399 20.457 19.734 18.456 4.023 .008 Table B-10. Effect of race on mean resilience scores shown by ANOVA Asian Black Hispanic White Other F P-value Commitment to Learning 20.000 22.857 22.753 21.296 20.404 3.228 .013 Positive Values 18.727 21.825 19.422 19.400 20.059 3.508 .008 EA Score 19.506 22.277 20.996 20.685 20.549 2.539 .040 IA Score 16.974 19.769 19.202 20.356 19.584 2.694 .031 Table B-11. Significant ANOVA rela tionships between GRC pattern mean scores, total mean scores and age 18 19 20 21 22 23 24 F p SPC 4.191 4.113 4.349 4.128 3.607 4.253 4.294 2.459 .024 GRC 3.677 3.492 3.772 3.716 3.236 3.548 3.649 2.391 .028 Table B-12. Significant ANOVA re lationships between Academic Status and SPC score of gender role conflict Freshman Sophomore Junior Senior F p-value SPC pattern 4.233 4.158 4.325 3.872 3.769 .011 Table B-13. Significant ANOVA re lationships between race and pattern/total gender role conflict Asian Black Hispanic White Other F P-value RABBM pattern 3.739 3.701 2.985 3.265 3.444 3.464 .009 CBWFR pattern 4.118 3.885 3.355 3.508 3.783 3.222 .013 GRCS total 3.924 3.900 3.444 3.531 3.769 4.207 .002

PAGE 101

101 Table B-14. Effect of sexual orientation on mean pattern and total gender role conflict scores shown by ANOVA Score of 0 Score of 1 Scores 2-6 F p-value RE pattern 3.193 3.370 2.744 3.352 .036 RABBM pattern 3.431 2.757 2.662 9.583 .000 GRC total 3.666 3.390 3.284 4.593 .011 Table B-15. Significant ANOVA relati onships between RE quartile and resilience pattern/overall means Level 0 Level 1 Level 2 Level 3 F p-value EMP 24.326 22.800 21.944 21.145 8.736 .000 CTL 22.632 21.849 20.908 21.040 2.666 .048 PV 21.032 19.826 19.358 19.139 3.247 .022 SOC 22.261 21.038 20.067 19.868 6.187 .000 PID 23.296 21.129 21.035 20.308 6.030 .001 EA 22.305 20.960 20.342 20.089 6.932 .000 DAP 43.235 41.004 40.202 38.728 6.121 .000 Table B-16. Significant ANOVA rela tionships between RABBM quar tile and resilience pattern means Level 0 Level 1 Level 2 Level 3 F p-value CTL 23.095 21.162 20.867 21.341 4.001 .008 SOC 21.801 20.059 20.911 20.470 2.687 .046 Table B-17. Significant ANOVA rela tionships between CBWFR quar tile and resilience pattern means Level 0 Level 1 Level 2 Level 3 F p-value EMP 23.854 22.210 22.218 22.042 3.033 .029 SOC 21.409 20.697 19.526 21.439 3.983 .008 Table B-18. Significant ANOVA re lationships between GRCS sc ore quartile and resilience pattern means Level 0 Level 1 Level 2 Level 3 F p-value EMP 24.091 21.778 22.509 21.889 5.105 .002

PAGE 102

102 Table B-19. Standardized regressi on coefficients for demographi c and GRCS pattern effect on DAP score with age Model 1 Model 2 Model 3 Mode l 4 Model 5 Model 6 Reduced Demographics Black .044 .068 Hispanic -.049 -.040 Other -.093 -.088 Age -.006 -.083 Academic Status -.105 -.020 Sexual Orientation .028 .020 SPC Pattern .066 .197** .189*** RE Pattern -.254*** -.355*** -.344*** RABBM Pattern -.081 .055 CBWFR Pattern -.080 -.040 Adjusted Rsquared .007 .002 .062 .004 .004 .098 .105 Cases 317 361 361 361 361 317 317 ** Significant at the 0.01 level *** Significant at the 0.001 level Table B-20. Standardized regressi on coefficients for demographi c and GRCS pattern effect on DAP score without age Model 1 Model 2 Model 3 Mode l 4 Model 5 Model 6 Reduced Demographics Black .044 .069 Hispanic -.049 -.042 Other -.093 -.087 Academic Status -.110 -.090 Sexual Orientation .028 .019 SPC Pattern .066 .192** .181*** RE Pattern -.254*** -.347*** -.338*** RABBM Pattern -.081 .050 CBWFR Pattern -.080 -.040 Adjusted Rsquared .012 .002 .062 .004 .004 .102 .115 Cases 357 361 361 361 361 356 356 ** Significant at the 0.01 level *** Significant at the 0.001 level

PAGE 103

103 Table B-21. Standardized regressi on coefficients for demographi c and GRCS pattern effect on External Asset score with age Model 1 Model 2 Model 3 Mode l 4 Model 5 Model 6 Reduced Demographics Black .146* .167** .176** Hispanic .019 .033 Other -.050 -.046 Age .016 -.059 Academic Status -.035 .043 Sexual Orientation .090 .081 SPC Pattern .040 .147* .154** RE Pattern -.247*** -.367*** -.329***RABBM Pattern -.077 .077 CBWFR Pattern -.032 .019 Adjusted Rsquared .015 -.001 .058 .003 -.002 .097 .105 Cases 317 361 361 361 361 317 317 Significant at the 0.050 level ** Significant at the 0.01 level *** Significant at the 0.001 level Table B-22. Standardized regressi on coefficients for demographi c and GRCS pattern effect on External Asset score without age Model 1 Model 2 Model 3 Mode l 4 Model 5 Model 6 Reduced Demographic Variables Black .146** .168** .176** Hispanic .019 .031 Other -.050 -.046 Academic Status -.022 -.007 Sexual Orientation .090 .080 SPC Pattern .040 .144* .154** RE Pattern -.247*** -.362*** -.329***RABBM Pattern -.077 .072 CBWFR Pattern -.032 .019 Adjusted Rsquared .020 -.001 .058 .003 -.002 .102 .106 Cases 357 361 361 361 361 356 356 Significant at the 0.050 level ** Significant at the 0.01 level *** Significant at the 0.001 level

PAGE 104

104 Table B-23. Standardized regressi on coefficients for demographi c and GRCS pattern effect on Internal Asset score with age Model 1 Model 2 Model 3 Mode l 4 Model 5 Model 6 Reduced Demographic Variables Black -.053 -.031 Hispanic -.099 -.096 Other -.114 -.108 Age -.028 -.092 -.148*** Academic Status -.143 -.067 Sexual Orientation -.031 -.036 SPC Pattern .077 .205*** .195*** RE Pattern -.214*** -.281*** -.292*** RABBM Pattern -.070 .027 CBWFR Pattern -.107* -.084 Adjusted Rsquared .028 .003 .043 .002 .009 .098 .097 Cases 317 361 361 361 361 317 317 Significant at the 0.050 level *** Significant at the 0.001 level Table B-24. Standardized regressi on coefficients for demographi c and GRCS pattern effect on Internal Asset score without age Model 1 Model 2 Model 3 Mode l 4 Model 5 Model 6 Reduced Demographic Variables Black -.053 -.030 Hispanic -.099 -.098 Other -.114* -.108* Academic Status -.166** -.146** -.147** Sexual Orientation -.032 -.038 SPC Pattern .077 .200*** .181*** RE Pattern -.214*** -.272*** -.284*** RABBM Pattern -.070 .021 CBWFR Pattern -.107* -.084 Adjusted Rsquared .033 .003 .043 .002 .009 .101 .103 Cases 357 361 361 361 361 356 356 Significant at the 0.050 level ** Significant at the 0.01 level *** Significant at the 0.001 level

PAGE 105

105 Table B-25. p-values for demographic and G RCS pattern effect on DAP score with age regression Model 1 Model 2 Model 3 Mode l 4 Model 5 Model 6 Reduced Demographic Variables Black .450 .229 Hispanic .407 .479 Other .115 .126 Age .958 .420 Academic Status .323 .848 Sexual Orientation .622 .715 SPC Pattern .208 .002 .001 RE Pattern .000 .000 .000 RABBM Pattern .125 .429 CBWFR Pattern .128 .515 Adjusted Rsquared .007 .002 .062 .004 .004 .098 .105 Cases 317 361 361 361 361 317 317 Table B-26. p-values for demographic and GRCS pattern effect on DAP score without age regression Model 1 Model 2 Model 3 Mode l 4 Model 5 Model 6 Reduced Demographic Variables Black .421 .196 Hispanic .376 .431 Other .094 .105 Academic Status .038 .076 Sexual Orientation .600 .719 SPC Pattern .208 .001 .001 RE Pattern .000 .000 .000 RABBM Pattern .125 .448 CBWFR Pattern .128 .484 Adjusted Rsquared .012 .002 .062 .004 .004 .102 .115 Cases 357 361 361 361 361 356 356

PAGE 106

106 Table B-27. p-values for demogra phic and GRCS pattern effect on External Asset score with age regression Model 1 Model 2 Model 3 Mode l 4 Model 5 Model 6 Reduced Demographic Variables Black .012 .003 .001 Hispanic .749 .564 Other .398 .424 Age .883 .561 Academic Status .737 .673 Sexual Orientation .111 .145 SPC Pattern .444 .020 .008 RE Pattern .000 .000 .000 RABBM Pattern .145 .273 CBWFR Pattern .539 .756 Adjusted Rsquared .015 -.001 .058 .003 -.002 .097 .105 Cases 317 361 361 361 361 317 317 Table B-28. p-values for demogra phic and GRCS pattern effect on External Asset score without age regression Model 1 Model 2 Model 3 Mode l 4 Model 5 Model 6 Reduced Demographic Variables Black .008 .002 .001 Hispanic .730 .556 Other .369 .397 Academic Status .672 .884 Sexual Orientation .090 .126 SPC Pattern .444 .016 .005 RE Pattern .000 .000 .000 RABBM Pattern .145 .268 CBWFR Pattern .539 .746 Adjusted Rsquared .020 -.001 .058 .003 -.002 .102 .106 Cases 357 361 361 361 361 356 356

PAGE 107

107 Table B-29. p-values for demogra phic and GRCS pattern effect on Internal Asset score with age regression Model 1 Model 2 Model 3 Mode l 4 Model 5 Model 6 Reduced Demographic Variables Black .358 .581 Hispanic .090 .090 Other .051 .058 Age .793 .367 .006 Academic Status .174 .513 Sexual Orientation .576 .512 SPC Pattern .144 .001 .001 RE Pattern .000 .000 .000 RABBM Pattern .187 .699 CBWFR Pattern .041 .170 Adjusted Rsquared .028 .003 .043 .002 .009 .098 .097 Cases 317 361 361 361 361 317 317 Table B-30. p-values for demogra phic and GRCS pattern effect on Internal Asset score without age regression Model 1 Model 2 Model 3 Mode l 4 Model 5 Model 6 Reduced Demographic Variables Black .330 .569 Hispanic .070 .065 Other .038 .045 Academic Status .002 .004 .004 Sexual Orientation .549 .466 SPC Pattern .144 .001 .001 RE Pattern .000 .000 .000 RABBM Pattern .187 .752 CBWFR Pattern .041 .142 Adjusted Rsquared .033 .003 .043 .002 .009 .101 .103 Cases 357 361 361 361 361 356 356 Significant at the 0.050 level ** Significant at the 0.01 level *** Significant at the 0.001 level

PAGE 108

108 LIST OF REFERENCES Arnett, J. (2007). Adolescence and emerging adulthood:A cultural approach 3rd edition. New Jersey: Pearson Education, Inc. Beatty, A., Syzdek, M., & Bakkum, A. (2006). The Saint Johns experience project: challenging mens perceptions of normative gender role conflict. Journal of Mens Studies 14(3), 322326. Benson, P. (2002). Adolescent development in so cial and community context: A program of research. New Directions for Youth Development 2002(95), 123-147. Berlin, C. (1988). Gender role conflict, future time perspective, and contraceptive behavior among unmarried, black which and Hispanic ad olescent fathers and their nonfather peers (Doctoral dissertation, Ne w York University). Dissertation Abstracts International 49/10, 2940. Blazina, C., & Watkins, C. E. (1996). Masculine gender role conflict: Ef fects on college men's psychological well-being, chemical substance usage, and attitudes toward help-seeking. Journal of Counseling Psychology, 43, 461-465. Bower, H. (2001). The gender identity disorder in the DSM-IV classi fication: a critical evaluation. Australian and New Zealand Journal of Psychiatry 35, 1-8. Bronfenbrenner, U., & Evans, G. (2000) Developmental science in the 21st century: emerging questions, theoretical models, resear ch designs and empirical findings. Social Development 9(1), 115-125. Bryman, A. (2004). Social Research Methods (2nd ed.) New York: Oxford University Press Inc. Busseri, M., Willoughby, T., Chalmers, H. & Bog aert, A. (2006). Same-sex attraction and successful adolescent development. Journal of Youth & Adolescence 35(4), 561-573. Campbell, J. & Snow, B. (1992). Gender role conf lict and family environm ent as predictors of mens marital satisfaction. Journal of Family Psychology 6, 84-87. Centers for Disease Contro l and Prevention. (2005). 2005 Youth Risk Behavior Survey from www.cdc.gov/yrbss. Cicchetti, D. & Garmezy, N. (1993). Milest ones in the developm ent of resilience. Development and Psychopathology 5, 497-774. Cournoyer, R. (1994). A devel opmental study of gender role c onflict in men and its changing relationship to psychological well-being (Doctoral dissertati on, Boston College). Dissertation Abstracts International 54/12, 6476.

PAGE 109

109 DAugelli, A., Grossman, A., Salter, N., Vasey, J., Starks, M., & Sinclair, K. (2005). Predicting the suicide attempts of le sbian, gay, and bisexual youth. Suicide and Life-Threatening Behavior 35(6), 646-660. de Vaus, D. (2001). Research Design in Social Research London: Sage Publications. Developmental Assets Profile Abstract. (2004). Minneapolis : Search Institute. Developmental Assets Profile User Manual. (2005). Minneapolis: Search Institute. Davis, F. (1987). Antecedents and consequences of gender role conflict: an empirical test of sexrole strain analysis (Doctoral di ssertation, Ohio State University). Dissertation Abstracts International 48/11, 3443. Developmental Assets Profile. Search Institute. Retrieved 18 Feb 2007 from < http://www.search-institu te.org/surveys/dap.html>. Edwards, O., Mumford, V., & Serra-Roldan, R. (2007). A positive youth development model for students considered at-risk. School Psychology International 28(1), 29-45. Fausto-Sterling, A. (2000). Sexing the body: gender politics and the construction of sexuality. New York: Basic Books. Galambos, N. & Leadbeater, B. (2000). Trends in adolescent research for the new millennium. International Journal of Behavioral Development 24(3), 289-294. Garnets, L., & Pleck, J. (1979). Sex role identi ty, androgyny and sex role transcendence: a sex role strain analysis. Psychology of Women Quarterly 3, 270-283. Good, G. & Mintz, L. (1990). Gende r role conflict and depression in college men: evidence for compounded risk. Journal of Counseling and Development 69, 17-20. Good, G., Robertson, J., & Fitzgerald, L. (1996). The relation between ma sculine role conflict and psychological distress in male uni versity counseling center clients. Journal of Counseling and Development 75 (5), 44-49. Human Rights Council. (2006). Corporate Equality Index from http://www.hrc.org/Template.cfm?Sec tion=Get_Informed2&CONTENTID=31668&TEM PLATE=/ContentManagement/ContentDisplay.cfm>. Israel, G. (1992). Determining sample size (PEOD6) Gainesville: Department of Agricultural Education and Communication, Florida Coopera tive Extension Service, Institute of Food and Agricultural Sciences, University of Florida. Retrieved March 2, 2007 from http://edis.ifas.ufl.edu/PD006. Judge, B. & Billick, S. (2004). Suicidality in Adolescence: review and legal considerations. Behavioral Sciences and the Law 22, 681-695.

PAGE 110

110 Joyce, A. (2004, November 3). Companie s add gender to antibias policies. The Washington Post p. E01. Kalafat, J. (1994). An evaluation of a school-based suicide aw areness intervention. Suicide & Life-Threatening Behavior 24(3), 224-234. Kaplan, R. (1992). Normative masculinity and sexual aggression among college males (Doctoral dissertation, University of Connecticut). Dissertation Abstracts International 53/08, 3005. Kinsey, Alfred C. et al. (1948/1998). Sexual Behavior in the Human Male .. Philadelphia: W.B. Saunders; Bloomington: Indiana U. Press. Kulkin, H., Chauvin, E., & Percle, G. (2000). Suicide among gay and lesbian adolescents and youth adults: a review of the literature. Journal of Homosexuality 40(1), 1-29. Luthar, S., Dorenberger, C. & Zigler, E. (1993) Resilience is not a unidimensional construct: Insights from a prospective st udy of inner-city adolescents. Development and Psychopathology 5, 703-717. Mendelson, E. (1988). An exploratory investiga tion of male gender-role development during early adulthood (Doctoral dissertatio n, University of North Carolina). Dissertation Abstracts International 48, 2119-2120B. ONeil, J. (1986). Gender Role Conflict Scale I. Storrs: School of Family Studies, University of Connecticut. ONeil, J., Good, G., & Holmes, S. (1995). In R. Levant & W Pollack (Eds.). The New Psychology Of Men New York: Basic Books. ONeil, J. & Owen, S. (1994). The manual for the Gender Role Conflict Scale Storrs: School of Family Studies, University of Connecticut. Pelkonen, M. & Marttunen, M. (2003). Child and adolescent suicide: epidemiology, risk factors, and approaches to prevention. Pediatric Drugs 5(4), 243-265. Remafedi, G., French, S., Story, M., Resnick, M. & Blum, R. (1998). The relationship between suicide risk and sexual orientation: results of a populations-based study. American Journal of Public Health 88(1), 57-60. Richardson, G. (2002). The metatheo ry of resilience and resiliency. Journal of Clinical Psychology 58 (3), 307-321. Rounds, D. (1994). Predictors of homosexual intoler ance on a college campus: identity, intimacy, attitudes toward homosexuals and gender role conflict Unpublished masters thesis, Department of Psychol ogy, University of Connecticut. Sax, L. (2002). How common is intersex? A response to Anne Fausto-Sterling. The Journal of Sex Research 39(3), 174-178.

PAGE 111

111 Scales, P. (1999). Reducing risks and building developmental assets: essential actions for promoting adolescent health. Journal of School Health 69 (3), 113-119. Sharpe, M. (1993). Gender role and psyc hological well-being in adult men (Doctoral dissertation, University of Missouri-Columbia). Dissertation Abstracts International, 54/10, 5373. Sharpe, M. & Heppner, P. (1991). Gender role, gender role conflict, and psychological wellbeing in men. Journal of Counseling Psychology 38, 323-330. Unger, R. (1979). Toward a redefinition of sex and gender. American Psychologist 34, 10851094. U.S. House of Representatives pa sses historic hate crimes bill (2007, May 3). Press release. Retrieved on 8 May 2007 from < http://www.hrc.org/Template.cfm?Sec tion=Get_Informed2&CONTENTID=36630&TEM PLATE=/ContentManagement/ContentDisplay.cfm> Watts, R., & Borders, L. (2005). Boys percepti ons of the male role: Un derstanding gender role conflict in adolescent males. The Journal of Mens Studies 13 (2), 267-280. What is intersex? Intersex Society of North America. Retrieved 18 Feb 2007 from . Whats the difference between being transge nder or transsexual and having an intersex condition? Intersex Society of North America. Retrieved 18 Feb 2007 from .

PAGE 112

112 BIOGRAPHICAL SKETCH Stephanie Bates was born in Clearwater, Flor ida. After growing up in Coral Springs, Florida and graduating from J.P. Taravella High School, she attend ed the University of Florida for her undergraduate education. She received dual degrees (a B.S. in family, youth and community sciences and a B.A. in classical studi es) and minors in art history, interdisciplinary leadership, and organizational l eadership for nonprofits in December 2005. She entered the "4 + 1" program and began work toward her M.S. in family, youth and community sciences as an undergraduate. After this degree, she will pursue a J.D. at the University of Florida Levin College of Law with the hope of combining soci al science research wi th public policy and legislation.