UNDERSTANDING RESILIENCY—THE RELATIONSHIP BETWEEN
USAFA CADET GRIT-S SCORES AND CADET DEVELOPMENT.
by
JUSTIN R. STODDARD
B.A., University of Colorado—Boulder, 2001
M.A., University of Texas—El Paso, 2011
A dissertation submitted to the Graduate Faculty of the
University of Colorado Colorado Springs
in partial fulfillment of the
requirements for the degree of
Doctor of Philosophy
Department of Education Leadership, Research, and Foundations
2019
Ó 2019
JUSTIN R. STODDARD
ALL RIGHTS RESERVED
This dissertation for the Doctor of Philosophy degree by
Justin R. Stoddard
has been approved for the
Department of Education Leadership, Research, and Foundations
by
Andrea Bingham, Chair
Joseph Taylor
Patricia Witkowsky
Phillip Morris
Christopher Luedtke
____July 29, 2019___ Date
ii
Stoddard, Justin R. (Ph.D., Education, Leadership, and Policy)
Understanding Resiliency—The Relationship Between USAFA Cadet Grit-S Scores and Cadet Development.
Dissertation directed by Assistant Professor Andrea Bingham.
ABSTRACT
Life is filled with adversity and no one is excluded from experiencing setbacks, failures, and even catastrophes. But within all people is the potential to learn and bounce back from these experiences. Resiliency theory describes the promotive and protective factors individuals can use to process learn from setbacks and failures and to grow and bounce back from the disruptive adversities that may otherwise break us down. Using the Grit-S survey, which consists of a passion subscore and a resilience subscore, this study focuses on understanding the nature of resilience by analyzing survey responses from over 5,400 cadets from the United States Air Force Academy spanning over nine years. The quantitative analysis of the associations between Grit-S score, passion subscore, and resilience subscore and over 30 different cadet variables including background, current performance, and participation in USAFA teams and programs provided greater understanding into the cadet characteristics that were associated with their grit variable scores. Significantly associated variables unique to the military learning environment at USAFA were discovered and included performance in leadership development, athletics, and participation in a variety of activities and programs which build real-world life skills and present cadets with realistic challenges. The results confirm the statistically and practically significant associations between the grit variables and both academic and non-academic performance in several areas and also suggest there is considerable value in experiencing challenging scenarios outside the classroom.
iii
DEDICATION
This work is dedicated to my father, who is one of the best examples of resilience
I have ever known. Despite setbacks, failures, and shortcomings, he simply never quits.
He has pushed forward his entire life and has experienced many successes, including his children.
iv
ACKNOWLEDGEMENTS
Several individuals had a significant impact on this research study and guided me to complete this work, and I simply could not have done this without them. I am sincerely grateful to the following individuals for their participation and assistance.
Dr. Andrea Bingham served as my committee chair and asked the hard questions to give this study the correct focus. Dr. Joe Taylor provided essential assistance as my methodologist and helped me navigate SPSS. Dr. Christopher Luedtke from CCLD,
USAFA was my primary connection to USAFA and served as a committee member and
USAFA expert. Dr. Patricia Witkowsky and Dr. Phillip Morris served as committee members and provided important feedback that greatly added to the study. Gene Hagan and Mark Briody both worked to get exactly the data I needed from USAFA. Laura Neal guided me through the USAFA IRB process and made sure all the documentation was correct. LTC Jessica Sullivan and Kathy McHugh provided me with the essential cadet
Grit-S survey data. Dr. Terry McFarlane took the time to discuss cadet resiliency with me and attend a book signing with Dr. Angela Duckworth. Dr. Amanda Lords was a good friend and mentor and helped me structure my research project. Dr. John
Abbatiello and Richard Ramsey provided me with the details of the USAFA Officer
Development System and insight into the various aspects of CCLD programming. Karen
Howard and Paul Torchinsky from the VA Vocational Rehab program who made this degree possible. I especially thank the USAFA cadets for being the excellent individuals they are and for completing the surveys that informed this study. Finally, I am grateful to my wife Miriam who supported me through this study and to my Heavenly Father who gave me the inspiration to build and strengthen others.
v
TABLE OF CONTENTS
CHAPTER
I. INTRODUCTION ...... 1
Background of Resiliency Theory ...... 7
Purpose of the Study ...... 8
Research Questions ...... 9
Research Significance ...... 10
Definitions and Terms ...... 12
Dissertation Structure ...... 15
II. LITERATURE REVIEW ...... 17
Resiliency Theory ...... 17
Origins and Development ...... 18
Resiliency Framework ...... 22
Definitions of Resiliency ...... 28
Resiliency-Building Models ...... 30
Compensatory Model ...... 31
Protective Model ...... 32
Challenge Model ...... 33
Commonalities Between Models ...... 34
Strengths and Weaknesses of Resiliency Theory ...... 35
Review of Relevant Literature ...... 39
Leadership ...... 39
Resiliency in Higher Education ...... 42
vi
The Challenges to Students in Higher Education ...... 42
Efforts to Help Students Overcome Challenges in Higher Education ...... 43
Results of Efforts to Help Students ...... 44
Resiliency in the Military ...... 45
The Challenges to Service Members in the Military ...... 46
Efforts to Help Service Members Overcome Challenges in the Military . 47
Results of Efforts to Help Service Members ...... 49
Resiliency at USAFA ...... 51
The Challenges to Cadets at USAFA ...... 51
Efforts to Help Cadets Overcome Challenges at USAFA ...... 53
Results of Efforts to Help Cadets ...... 58
Emerging Themes and Research Gaps ...... 60
III. METHOD ...... 63
Research Setting and Participants ...... 64
Measures ...... 66
Grit-S Survey Description ...... 66
Reliability and Validity ...... 67
Research Design ...... 69
Procedures ...... 69
Data Collection ...... 69
Variables ...... 73
Missing Values ...... 81
Data Analysis ...... 83
vii
RQ1 – Grit-S score vs Cadet Predictors ...... 83
RQ2 – Performance vs Grit-S Score ...... 86
RQ3 – Attrition vs Grit-S Score ...... 89
RQ4 – Changes in Grit-S Score Over Time vs Cadet Predictors ...... 92
RQ5 – Recommended Future Research, Policies, and Practices ...... 94
Limitations ...... 95
Ethics ...... 97
IV. RESULTS ...... 99
Review of Descriptives and Research Question Results ...... 100
Descriptives ...... 100
RQ1 – Grit-S Score vs Cadet Predictors ...... 103
Grit-S Score ...... 103
Passion Subscore ...... 110
Resilience Subscore ...... 116
RQ2 – Performance vs Grit-S Score ...... 122
Grade Point Average ...... 122
Military Performance Average ...... 127
Physical Education Average ...... 132
Overall Performance Average ...... 137
RQ3 – Attrition vs Grit-S Score ...... 142
Enrollment Status ...... 142
RQ4 – Changes in Grit-S Score Over Time vs Cadet Predictors ...... 148
Average Changes in Grit-S Score Variables ...... 148
viii
Grit-S Gain Score ...... 154
Passion Gain Score ...... 159
Resilience Gain Score ...... 164
Summary of Research Question Results ...... 169
V. DISCUSSION AND CONCLUSION ...... 174
Discussion ...... 175
Discussion of Results ...... 175
RQ1 – Grit-S Score vs Cadet Predictors ...... 175
RQ2 – Performance vs Grit-S Score ...... 184
RQ3 – Attrition vs Grit-S Score ...... 186
RQ4 – Changes in Grit-S Score Over Time vs Cadet Predictors ...... 187
Addressing Earlier Criticisms ...... 194
Implications ...... 196
Implication 1 – Academic and Non-Academic Variables are Associated with Grit ...... 197
Implication 2 – Attrition is not Associated with Grit ...... 201
Implication 3 – Resilience is Critical to Leadership Development ...... 203
Recommendations ...... 205
RQ5 – Recommendations for Future Research, Policy, and Practice ...... 205
Changes to Policies and Practices ...... 206
Future Research ...... 209
Conclusion ...... 211
Final Thoughts ...... 212
REFERENCES ...... 215
ix
APPENDICES
A. USAFA INSTITUTIONAL REVIEW BOARD APPROVAL ...... 234
B. USAFA VOLUNTEER SERVICE AGREEMENT (VSA) ...... 236
C. GRIT-S SURVEY ...... 237
D. USAFA INTERCOLLEGIATE SPORTS ...... 240
E. USAFA INTRAMURAL SPORTS ...... 241
F. USAFA COMPETITIVE PROGRAMS ...... 242
G. USAFA MISSION SUPPORT PROGRAMS ...... 243
H. USAFA PROFESSIONAL PROGRAMS ...... 244
I. USAFA CLUB PROGRAMS ...... 245
J. USAFA RECREATIONAL PROGRAMS ...... 246
K. SPSS LINEAR REGRESSION OUTPUTS ...... 247
x
LIST OF TABLES
TABLE
1. Description of USAFA Grit-S Survey Collected from USAFA-A9 ...... 71
2. Dependent and Independent Variables Used in Regression Analysis ...... 73
3. Statistics of Variables with Imputed Values ...... 82
4. RQ1 Variables Used in Linear Regression Analysis ...... 85
5. RQ2 Variables Used in Linear Regression Analysis ...... 88
6. RQ3 Variables Used in Logistic Regression Analysis ...... 91
7. RQ4 Variables Used in Linear Regression Analysis ...... 93
8. Descriptive Statistics Comparing Grit Variables with Cadet Predictors ...... 101
9. Model 1.1a Regression Results Comparing Grit-S Scores with Cadet Predictors ...... 104
10. Model 1.1b Regression Results Comparing Grit-S Scores with Cadet Predictors ...... 108
11. Model 1.2a Regression Results Comparing Passion Subscores with Cadet Predictors ...... 111
12. Model 1.2b Regression Results Comparing Passion Subscores with Cadet Predictors ...... 114
13. Model 1.3a Regression Results Comparing Resilience Subscores with Cadet Predictors ...... 117
14. Model 1.3b Regression Results Comparing Resilience Subscores with Cadet Predictors ...... 120
15. Model 2.1a Regression Results Comparing GPA with Cadet Predictors ...... 124
16. Model 2.1b Regression Results Comparing GPA with Cadet Predictors ...... 126
17. Model 2.2a Regression Results Comparing MPA with Cadet Predictors ...... 129
18. Model 2.2b Regression Results Comparing MPA with Cadet Predictors ...... 131
xi
19. Model 2.3a Regression Results Comparing PEA with Cadet Predictors ...... 134
20. Model 2.3b Regression Results Comparing PEA with Cadet Predictors ...... 136
21. Model 2.4a Regression Results Comparing OPA with Cadet Predictors ...... 139
22. Model 2.4b Regression Results Comparing OPA with Cadet Predictors ...... 141
23. Model 3.1a Logistic Regression Coefficients Comparing Enrollment Status with Cadet Predictors ...... 144
24. Model 3.1b Logistic Regression Coefficients Comparing Enrollment Status with Cadet Predictors ...... 146
25. Descriptive Statistics Comparing Grit-S, Passion, and Resilience Gain Scores with Cadet Predictors ...... 153
26. Model 4.1a Regression Results Comparing Grit-S Gain Scores with Cadet Predictors ...... 156
27. Model 4.1b Regression Results Comparing Grit-S Gain Scores with Cadet Predictors ...... 158
28. Model 4.2a Regression Results Comparing Passion Gain Scores with Cadet Predictors ...... 161
29. Model 4.2b Regression Results Comparing Passion Gain Scores with Cadet Predictors ...... 163
30. Model 4.3a Regression Results Comparing Resilience Gain Scores with Cadet Predictors ...... 166
31. Model 4.3b Regression Results Comparing Resilience Gain Scores with Cadet Predictors ...... 168
32. Statistically Significant Predictors Associated with Grit Variables ...... 177
33. Statistically Significant Predictors Associated with Grit Gain Score Variables . 188
xii
LIST OF FIGURES
FIGURE
1. Richardson Resiliency Model ...... 26
2. A Comparison of Resiliency Definitions ...... 29
3. Grit-S, Passion, and Resilience Score Trends for Group One ...... 149
4. Grit-S, Passion, and Resilience Score Trends for Group Two ...... 150
5. Grit-S, Passion, and Resilience Score Trends for Group Three ...... 151
xiii
LIST OF ABBREVIATIONS
ABBREVIATIONS
ACES Academy character enrichment seminar AOC Air officer commanding AFSC Air Force Specialty Code AMT Academy military trainer ARSOF Army Special Operations Forces BCT Basic cadet training (at USAFA) CAF Comprehensive airmen fitness CBT Cadet basic training (at USMA) CCLD Center for Character and Leadership Development CSF Comprehensive soldier fitness CSF2 Comprehensive soldier and family fitness DAC Department of the Army civilian DCoE Defense Centers of Excellence of Psychological and Traumatic Brain Injury DRS Dispositional resilience scale DSAT Designated survey and assessment time GAT Global assessment tool GPA Grade point average IRB Institutional review board LIFT Leaders in flight today MPA Military performance average ODS Officer development system OML Order of merit list OPA Overall performance average PEA Physical education average PRP Penn Resiliency Project PTSD Post-traumatic stress disorder R&R Respect and responsibility SERE Survive escape resist evade SES Socioeconomic status SRI Support and resilience inventory STEM Science, Technology, Engineering, Mathematics USAFA United States Air Force Academy USAF United States Air Force USMA United States Military Academy at West Point VECTOR Vital effective character through observation and reflection
xiv
CHAPTER I
INTRODUCTION
Everyone experiences adversity. Adversity is defined as “hard times; a difficult situation or condition; a state or instance of serious or continued difficulty or misfortune”
(Adversity, 2018), and it affects virtually every aspect of life. Adversity often challenges individuals by placing difficult circumstances between them and the goals they want to achieve. As a result, people either learn to confront and grow from adversity or are overwhelmed and overcome by it. Adversity comes from a variety of sources both natural and man-made, and some of the greatest adversities are those presented when countries are at war. The United States has been in a constant state of war since the events of 9/11, and from that time until 2015, 2.77 million service members have answered the call to fight and serve their country on over 5.4 million deployments
(Wenger, O’Connell, & Cottrell, 2018). Contemporary warfighting in environments like
Iraq, Afghanistan, and Syria is difficult and dangerous, and under these arduous conditions, even the most well-prepared, best-trained, and highly experienced individuals are subject to mental fatigue and failure. Being resilient in these environments “is distinct from mere survival, and more than mere endurance. Resilience is often endurance with direction” (Greitens, 2015, p. 25). While it is difficult to imagine the challenges and adversities faced by military service members without actually walking in their boots, it is not difficult to understand the need to support, train, and prepare service members to effectively confront adversity before they head into harm’s way.
Considering the many deployments and austere conditions military men and women serve in, understanding the nature and development of individual resilience is
1
both essential and critical to mission success since it is their resilience that will enable them to overcome their adversities. However, exactly how to develop individual resilience, and how to measure it is a renewed field of study garnering much attention and research. An important aspect of this resiliency research is understanding the relationship between resiliency theory and practices to build resiliency so that military academies tasked with developing young military leaders can have the information necessary to develop the policies and programs designed to develop individual resiliency.
At the United States Air Force Academy (USAFA), the development of resiliency in cadets is a high priority. Exhibiting resiliency in connection with developing grit is a specific proficiency that cadets are expected to improve as they learn to cultivate “. . . a hardiness of spirit and resistance to accept failure despite physical and mental hardships”
(The United States Air Force Academy, 2016c, p. 1). Dr. Duckworth, a prominent researcher on the subject of grit has stated that resiliency is one of the two key components of grit and is also called perseverance of effort (Perkins-Gough, 2013). At
USAFA, transforming theories into operational policies and programs is essential to the goal of developing leaders of character who live honorably and embody the United States
Air Force (USAF) values of “Integrity first, Service Before Self, Excellence in All We
Do” (The United States Air Force Academy, 2016a). Training USAFA cadets to develop their individual resiliency and effectively face and overcome their own adversities contributes to their warrior ethos and is of paramount importance, since warrior leaders must not only face adversities themselves, but also effectively lead their units through those adversities.
2
Both by nature and design, the atmosphere and environment faced by cadets attending USAFA exposes them to a variety of adverse conditions on a regular basis.
USAFA cadets, among other military service cadets, represent future USAF leaders and will provide for the future safety and security of American lives and prosperity. From their first days of entering USAFA to the final days of graduation with the Air Force
Thunderbirds screaming overhead, USAFA cadets experience a unique environment of adversity which challenges them and attempts to develop their leadership skills and abilities. While this adversity-rich environment provides an exciting and challenging developmental adventure for many, it also represents an inhospitable and crushing environment for some who fall beneath acceptable standards, succumb to the academic pressures, and leave the academy either voluntarily or involuntarily.
Despite access to a variety of resources designed to support their successful graduation from USAFA, overall cadet attrition rates have averaged 24.4% over the last 5 years (Zubeck, 2018), compared to an average 35% for other universities in Colorado
(CollegeFactual, 2018a). This is lower than the 28% attrition rate at the United States
Military Academy at West Point (USMA) (CollegeFactual, 2018b) and higher than the
14% attrition rate at the U.S. Naval Academy (Prudente, 2014), but there is more at stake than mere comparisons. A recent report of 1,669 colleges and universities across the nation reported lost revenues close to $16.5 billion with the average school losing almost
$10 million (Raisman, 2013) due to attrition. While military academies are paid for with government budgets and tax dollars instead of receiving revenue from students, the financial loss resulting from attrition is no less a concern. A desire to decrease attrition rates while maintaining standards is of particular interest to USAFA since the Air Force
3
cost per graduate in 2015 was over $534K and financial losses due to attrition prior to graduation are expensive (Zubeck, 2018). However, USAFA has an even larger and more pressing interest in lowering attrition rates to help fill the large shortage of pilots currently faced by the United States Air Force (Losey, 2018a, 2018b). These challenges of minimizing financial losses and producing able and ready officers to fill much-needed pilot slots highlight the importance of reducing attrition rates at USAFA while maintaining high standards. The hope is that as cadet resiliency increases, so will their levels of grit along with their ability to rebuff failure and their ability to continue working to overcome physical and mental challenges. Efforts to find effective methods to help cadets become more successful have included the exploration of cadet resiliency and overall grit to increase cadet performance and decrease attrition despite the adversity inherent to USAFA.
This in part has been the mission and focus for USAFA’s Center for Character and Leadership Development (CCLD), to develop and graduate officers who “Live honorably consistently practicing the virtues embodied in the Air Force Core Values, Lift others to be their best possible selves, and Elevate performance toward a common and noble purpose” (The United States Air Force Academy, 2016a). CCLD is organized into four main divisions consisting of Honor, Cadet Development, Scholarship and
Innovation, and Events, and the joint focus is to improve the overall character and leadership development of USAFA cadets and produce leaders of character. The scholarship and innovation department does this is by conducting scholarly research on the development of leadership skills and strong moral character to inform the
4
development of policies and programs to improve cadet development. This study aids in this effort by researching the nature of cadet resiliency and its connection to performance.
The necessity for this study stemmed from two primary needs: the need to expand and clarify the relationship between resiliency and individual performance factors and the need to understand how USAFA cadet resiliency changes during their educational careers at USAFA and what influences those changes. First, a meta-analysis of 73 studies on grit and its subcomponents of passion and resilience found the relationship between grit and performance was modest at best (Credé, Tynan, & Harms, 2017). The sample sizes of the studies specifically comparing grit with academic performance ranged from 21 to
1,218 and averaged only 351 participants. Conversely, research studies at The United
States Military Academy at West Point (USMA) to understand the connection between grit and performance used one-time grit surveys to take snapshots of two groups of cadets
(N = 1,218 and 1,308) participating in Cadet Basic Training (CBT) (Duckworth,
Peterson, Matthews, & Kelly, 2007) and did find a close relationship between grit and performance. A similar study was also done to understand the hardiness, synonymous with resilience, of 2,383 USMA cadets just beginning their academic careers (Bartone,
Kelly, & Matthews, 2013) and also found close associations between grit and resilience and performance factors. This presents the need to expand and clarify the relationship between grit, passion, and resilience with individual performance factors using a larger sample size and to examine the changes in grit over time instead of a one-time snapshot.
This study addresses these needs by analyzing over 6,900 survey results completed by over 5,400 USAFA cadets over a nine-year period.
5
Second, among the desired outcomes established by USAFA to guide the development of cadets is an objective to instill in cadets a proficiency to “Exhibit grit: a hardiness of spirit and resistance to accept failure despite physical and mental hardships” as one of the ways to help build Warrior Ethos (The United States Air Force Academy,
2016c). The academic experience at USAFA and its programs are designed to present cadets with challenges to navigate through, with the intent of increasing their resiliency and ability to adapt and overcome. This presents the need to evaluate and understand the extent to which these programs are influencing cadet resiliency and how cadet resiliency is changing over time so policies and programs can be modified to continually improve resiliency development. This research study addresses this need by providing a quantitative analysis of cadet characteristics, including demographics, performance measures, attrition data, and participation in various clubs and programs to understand which characteristics and programs were most associated with resiliency development and how they affected cadet resiliency over time.
This study completed a thorough quantitative analysis of the data using several regression models designed to explore the data and identify the primary associations with
Grit-S score and its subscores. This study also includes recommendations for policy and program development, practices to develop resiliency, and future research opportunities to continue the exploration of resiliency theory. The results from this study serve as a baseline of information from which USAFA can conduct future studies to explore the changes and trends in the resiliency development of cadets. It also explores changes in
USAFA cadet resiliency using a larger sample population than previously available to understand the resiliency of cadets more fully at one of the five military academies in the
6
United States. This builds upon prior resiliency research, including studies conducted at
USMA, and expands the understanding of individual resiliency adding to the body of knowledge currently available about resiliency theory and overcoming adversity.
Background of Resiliency Theory
Researchers commonly describe resiliency as the ability of an individual to bounce back and recover after facing adversity (Dyer & McGuinness, 1996). It has been referred to as perseverance of effort (Duckworth et al., 2007; Perkins-Gough, 2013), hardiness (Bartone, 2006; Bartone, Hystad, Eid, & Brevik, 2012), and it is understood to be more accurately described as a process instead of a static characteristic or trait
(Richardson, 2002; Rutter, 1987). In Grit: The Power of Passion and Perseverance, Dr.
Duckworth (2007) defines Grit as a combination of two key subcomponents: consistency of interest, which is also called passion, and perseverance of effort (Duckworth, 2016), which Duckworth later explains as being synonymous with resiliency (Duckworth, 2016;
Perkins-Gough, 2013). Duckworth’s research included the development of a research tool called the Grit Survey, which measured both passion and resiliency to understand the grit levels of individuals including USMA cadets and participants in a national spelling bee (Duckworth, 2016; Duckworth et al., 2007). Later, Dr. Duckworth’s team developed a shorter, more reliable, and more accurate version of the Grit Survey called the Grit-S
Survey (Duckworth & Quinn, 2009). Since 2009, USAFA has used the Grit-S survey to collect data from cadets on their individual resiliency and grit levels.
While research exists exploring the need and value of resiliency in areas such as higher education and doctoral programs (Allan, McKenna, & Dominey, 2014; Cross,
2014; Garza, Bain, & Kupczynski, 2014; Morales, 2014), military leaders experiencing
7
deployments (Meredith et al., 2011; Reivich, Seligman, & McBride, 2011; VanBreda,
2001), and the grit of military cadets experiencing adversity at USMA (Duckworth et al.,
2007), research directly connecting the resiliency subscale of grit and resiliency theory to character and leadership development programs is limited. Military academies provide a unique research environment that combines higher education with military training.
USAFA has gathered survey data over the past several years using a variety of resiliency- related survey tools. However, this data remained unexamined or only initially explored.
This presented a unique, data-rich research environment to understand resiliency theory as it relates directly to the cadets at USAFA.
Purpose of the Study
The purpose of this study was to understand the relationship between USAFA cadet resiliency and a variety of individual characteristics using quantitative regression analyses to determine the extent to which individual characteristics were associated with individual resiliency scores. The intent was to expand the work of previous studies by analyzing a sample population much larger than typically used in resiliency research in order to explore a greater number of individual characteristics and variables possibly associated with individual resiliency. This research adds to the body of knowledge on resiliency theory and provides clarity to prior research which overall has reported mixed results as to the relationship between resiliency and performance. Understanding this enables USAFA leadership and researchers to connect resiliency theory to practice and show which characteristics are the most highly associated with cadet resiliency and deserve additional future research priority. It may also enable decision-makers to develop or modify policies and programs to strengthen or bolster resiliency-weakening
8
factors, capitalize and improve on resiliency-building factors, and bring awareness to the factors affecting cadet resiliency overall. This increased understanding may provide
USAFA leadership the knowledge upon which to improve the overall resiliency of cadets and the units they serve in with the desired outcome of producing a more resilient and grittier fighting force to protect this nation. Regardless of the nature of results, this research serves as a baseline of information and a precursor to future research more closely examining the programs USAFA uses to build cadet resiliency.
Research Questions
The overall research question for this study asked whether there is a relationship between USAFA Cadet Grit-S scores and cadet development. The intent was to answer this question by proposing five specific research questions:
1. To what extent are cadet characteristics, including demographics and
participation in USAFA clubs and programs, associated with Grit-S score?
2. To what extent is Grit-S score associated with cadet overall performance
average (OPA)?
3. To what extent is Grit-S score associated with cadet attrition?
4. What is the change in cadet Grit-S score over time?
5. Based on this research study, what are some recommendations for future
research, policy development, and practice to build cadet resiliency at
USAFA?
The intent of these research questions was to explore the development of USAFA cadet resiliency and to understand the extent to which individual factors affected, were associated with, and predicted individual resiliency. This study analyzed these questions
9
through a quantitative analysis of cadet surveys, and the findings will serve as the basis for identifying program modifications and/or development, as well as determining opportunities for future research. Exploring these questions also provides a deeper understanding of how cadet resiliency is built and developed at USAFA and identifies possible methods of improving programs designed to develop leaders of character.
Research Significance
The significance of this research is at least threefold: it adds to the body of knowledge on the value of non-cognitive factors by clarifying the relationship between resiliency and academic performance, it identifies principles that may increase the effectiveness of military leaders and their development, and it takes advantage of nine years of survey results obtained by USAFA.
First, this study analyzes the relationship between resiliency and individual characteristics and performance factors using a sample population that is both larger than previous studies on similar topics and that examines the changes to resiliency over time instead of a one-time snapshot. This unique research study adds to the global debate discussing resiliency and the development of individual abilities and behaviors that may lead to improved performance and increased success and achievement. It also serves as a firm starting point for future studies delving deeper into specific programs and activities that increase an individual’s ability to persevere through challenging obstacles.
Second, this study seeks to identify and confirm the key principles central to resiliency theory and necessary in the development of cadet resilience at USAFA. The intent is to identify and develop outcomes that military leaders, professionals, educators, and even parents may use to help both individuals and organizations better understand
10
how to overcome adversity regardless of the environment and how to adapt to the ever- changing global and social challenges people face.
Finally, this study takes advantage of the time and resources spent by USAFA over at least nine years conducting cadet surveys at various times throughout the cadets’ academic careers. The time and resources applied by USAFA to understand cadet resiliency are put to work by analyzing the data, uncovering the valuable information it contains, and presenting the results in a way that benefits both USAFA, the military community, and higher education at large. The goal is to identify outcomes that increase the ability of USAFA and other higher education institutions to develop resilient leaders of character regardless of their future career paths.
Direct beneficiaries of this research include the USAFA Superintendent and
Director of CCLD along with their leadership teams as they continually develop and refine the programs and activities designed to develop cadets into leaders of character.
The list of beneficiaries also includes the leadership teams of the Dean of Faculty,
Commandant of Cadets, and Athletic Director, the professors and coaches at USAFA, the cadre of active duty officers and enlisted leaders who provide direct leadership development for USAFA cadets, and finally the cadets themselves. This research also has the potential to provide fundamental substantive principles to guide programs and policies to bring about the development of more adaptive, capable, and resilient cadets ready to face the global military challenges.
11
Definitions and Terms
The following definitions provide clarity for the key terms related to this study as several of the terms have a variety of connotations, interpretations, and implications depending on the audience and specialty field. The intent of these definitions is to clarify their meaning and operationalize each term for use in this study.
Adversity refers to “hard times; a difficult situation or condition; a state or instance of serious or continued difficulty or misfortune” (Adversity, 2018) and includes the various risk factors and obstacles making it difficult to achieve one’s goals.
Grit is defined as the passion and perseverance to achieve long-term goals (Duckworth,
2016).
Grit-S Survey refers to the 8-item survey developed by Dr. Angela Duckworth containing two subscales measuring passion and perseverance.
Resilience is defined as “The ability to withstand, recover and grow in the face of stressors and changing demands” (U.S. Department of the Air Force, 2014, p. 14).
Resilience and resiliency are synonymous with other terms like hardiness and perseverance but are not synonymous with the term “grit.”
Resiliency theory refers to the theoretical process an individual goes through as they face adversity, are influenced by both stabilizing and destabilizing factors, and reintegrate with either a new or unchanged level of resiliency.
Self-efficacy is “the belief in one’s capabilities to organize and execute courses of action required to produce given attainments” (Bandura, 1997, p. 3).
12
Mindset refers to the individual tendency to adopt either the fixed viewpoint, where value is placed on task completion, or the growth viewpoint, where value is placed on individual effort and the learning process (Dweck, 2006).
Promotive factors consisting of internal traits and abilities and external resources or the elements to which the individual can turn for assistance (Fergus & Zimmerman, 2005;
Kiswarday, 2012; Zimmerman, 2013; Zimmerman et al., 2013).
Protective factors include supportive family networks, socioeconomic status, school experiences, supportive communities, and cultural resources (Fleming & Ledogar, 2008) and serve to counteract or ameliorate the effects of adversity and guard the individual from adversity (Braverman, 2001).
Vulnerability factors represent the qualities, experiences, or lack of protective factors that both make individuals more susceptible to adversity and intensify the risk effects
(Braverman, 2001; Fleming & Ledogar, 2008; Luthar, Cicchetti, & Becker, 2000).
Bio-psycho-spiritual-homeostasis is a relatively stable emotional state an individual is in prior to experiencing a difficult situation or an adversity (Richardson, 2002).
Disruption refers to the moment when adversity or adverse conditions impact the homeostasis of an individual and cause the resiliency process to begin.
Reintegration refers to the process by which an individual attempts to regain homeostasis after experiencing disruption.
Compensatory model refers to the outcome-focused resiliency-building model which seeks to strengthen or augment an individual’s promotive factors and ability to employ them in order to better overcome the effects of adversity.
13
Protective model refers to the risk-focused resiliency-building model that focuses on mitigating the risks leading to negative outcomes in order to prevent or avoid the adversity altogether.
Challenge model refers to the adversity-focused resiliency-building model wherein individuals are intentionally exposed to moderate amounts of adversity in controlled environments to build an immunity to the risk (Garmezy, Masten, & Tellegen, 1984).
Attrition refers to the loss of students from an academic institution for any reason.
Warrior Ethos is defined as “the embodiment of the warrior spirit: tough mindedness, tireless motivation, an unceasing vigilance, a willingness to sacrifice one’s life for the country, if necessary, and a commitment to be the world’s premier air, space and cyberspace force” (The United States Air Force Academy, 2016c, p. 1).
Comprehensive soldier and family fitness (CSF2) program is a program initially launched in 2008 and relaunched in 2014 “to increase the resilience and enhance the performance of Soldiers, Families, and DACs,” referring to Department of the Army
Civilians (U.S. Department of the Army, 2014, para 1-5a). The CSF2 program identifies five dimensions of strength to create its primary conceptual pillars of physical, emotional, social, spiritual, and family resiliency to build the overall resiliency of soldiers (U.S.
Department of the Army, 2014).
Bonferroni adjustment is an adjustment made during multiple regression analysis in order to adjust the alpha level for multiple comparisons.
14
Dissertation Structure
I organized the five chapters of this dissertation to provide a thorough review of the purpose, background, and literature related to the concepts inherent to resiliency theory and the connection with leadership development. The methodology provides the analytical underpinnings of this study and describes the framework used to analyze the data. The findings inform the research questions and provide the primary discoveries of this research study, and a discussion follows to highlight the importance of the research and its contribution to the reviewed literature. In Chapter 1, I introduce the focus and intent of the research study, give a brief background of resiliency, and discuss my plan to complete the quantitative analysis to understand the relationship between Grit-S survey scores and cadet characteristics. I also provide a list of definitions and terms to guide the reader, along with an explanation of how I organized this dissertation.
In Chapter 2, I review the literature relevant to this study beginning with the origin, framework, definitions, and debated strengths and weaknesses of resiliency theory followed by a discussion of the key characteristics of leadership. Following this is a discussion of the challenges and research related to individual resiliency in both higher education and the military along with the efforts in each to build resiliency. Then I provide a review of the resiliency-developing elements of the character and leadership development programs used by USAFA, concluding with a discussion of the underlying themes and literature gaps used to develop the research questions for this study.
In Chapter 3, I provide the methodology for the study and a review of the nature of USAFA cadets and the resiliency surveys they have historically completed. I then set the stage for the quantitative analysis of survey scores and their association with
15
individual cadet factors. Following this, I present the data analysis of several groups of
USAFA cadets who took the Grit-S survey from 2009 to 2017 and highlight the several models of regression analyses I utilized to identify how cadet demographic predictors, performance measures, attrition rates, and participation in USAFA clubs and programs are associated with Grit-S survey scores.
In Chapter 4, I present the research findings by discussing each of the research questions and their related findings. I also present and review additional findings discovered during the analysis phase of the study along with a discussion of their relative importance and impact on the overall research study.
In Chapter 5, I present a broader discussion of the research study results in the context of known resiliency and resiliency development literature along with a discussion of the implications and recommendations for policy and/or program modifications. I also present a review of recommended future research strategies before concluding the research study with some personal anecdotal reflections.
16
CHAPTER II
LITERATURE REVIEW
The purpose of the literature review is to provide a detailed description of resiliency theory and the models typically used to try to build resiliency on an individual level. The review begins with a thorough examination of the origin, historical development, essential components, and current definitions of resiliency theory. An examination of the relevant literature follows by first reviewing the key characteristics of leadership and providing a foundation upon which to discuss the application of resiliency theory in higher education and the military. This leads to a discussion of cadet resiliency development at USAFA and the efforts taken to build resilient leadership within the cadet population. A review of the existing literature on these concepts includes discussions of the relevant strengths and weaknesses in each area and justifies the need for this study. I conclude by reviewing and discussing the gaps identified in the literature and presenting them as the foundation for my research questions for this study.
Resiliency Theory
This study uses resiliency theory to analyze and understand the processes individuals go through to operationalize their own resiliency and overcome obstacles when faced with adversity. Resiliency theory has changed over the years since its initial development in 1974 as “stress-resistance” in research conducted by Dr. Norman
Garmezy and Dr. Ann Masten (Garmezy, 1974). Definitions have shifted from focusing on static trait-based theories to more dynamic process-based theories, and researchers continue efforts to measure resiliency, predict it, and explain its effect on other aspects of behavior and performance. The focus of most research studies on resiliency is to identify
17
ways to maximize and predict human performance. This study looks at resiliency challenges in higher education, the military, and more specifically in the leadership development programs at USAFA.
Origins and Development
Research on individual resilience began in the early 1970s with the work of
Norman Garmezy and his investigation into children at risk for severe psychopathology
(Garmezy, 1974). His later work with Ann Masten focused on children raised in severely adverse circumstances and why some of them grew up with serious behavioral challenges while others seemed unaffected or even strengthened by their experiences, demonstrating a kind of stress competence and stress-resistance later termed “resilience” (Garmezy &
Masten, 1986). Early research, including that of Garmezy and Masten (1986) initially defined resilient behavior in terms of traits and characteristics that helped individuals overcome stressors (Fletcher & Sarkar, 2013). This trait-based theory remained the prevailing philosophy until it was later replaced by a concept of dynamic processes of positive adaptation whereby an individual demonstrates resilient behavior and overcomes adversity (Fleming & Ledogar, 2008; Luthar et al., 2000). This new understanding led to describing resiliency as a dynamic process of reacting to and engaging adversity in order to regain a homeostatic state that may be better, worse, or the same as when the adversity began (Richardson, 2002). This new understanding of resiliency theory, and the movement from a static trait-based theory to a process-based theory, led researchers to explore the factors that might lead individuals to develop resiliency, how resiliency influenced performance, how to measure resiliency, and whether it was possible or not to develop resiliency in different populations.
18
Contemporary researchers continue to explore the origins and precursors to resiliency, ways to accurately measure resiliency, and methods to develop resiliency.
Two prominent and contemporary resiliency researchers are Dr. Paul Bartone and Dr.
Angela Duckworth. Dr. Bartone is a retired Army Colonel currently working as a senior research fellow at the National Defense University and has written extensively about concepts surrounding the human ability to overcome adversity, particularly in the military
(Bartone, 2006; Bartone, Eid, Johnsen, Laberg, & Snook, 2009; Bartone et al., 2013;
Bartone, Roland, Picano, & Williams, 2008). Dr. Bartone has also dedicated a great deal of research to the study of resiliency and hardiness by creating, evaluating, and redesigning surveys and studies to measure the levels of hardiness of military personnel with results demonstrating the value of resiliency both in leaders and in those they lead
(Bartone, 2006; Bartone et al., 2013; Bartone et al., 2008).
Dr. Bartone developed a 15-item hardiness scale called the Dispositional
Resilience Scale (DRS), which has been demonstrated a reliable and valid measurement tool to evaluate the hardiness or resilience of a variety of individuals, especially soldiers and military leaders (Bartone, 1995, 2007). The DRS has been used to measure and understand the resilience of military personnel and is one of the surveys offered to cadets at USAFA (Bartone, 1995, 2007). Together, these studies highlight the importance of resiliency, especially in the military, to deal with the stresses of deployments, rigors of military life, and the risk factors associated with combat. It is notable that the terms
“hardiness” and “resilience” are often used interchangeably, showing the close and, in most cases, synonymous use of these terms.
19
Resiliency is one of the two key components of grit as explained by Dr. Angela
Duckworth, a professor at the University of Pennsylvania and author of the popular book
Grit: The power of passion and perseverance. Grit consists of the combination of two key subcomponents: consistency of interest referred to as passion and perseverance of effort simply referred to as perseverance (Duckworth, 2016), which Dr. Duckworth describes as being synonymous with resiliency (Perkins-Gough, 2013). It is important to note that while resiliency is synonymous with perseverance, it is not synonymous with grit. As grit is a combination of both passion and perseverance or resiliency, resiliency is a subcomponent of grit. This important distinction explains why the research study targeted both the passion of an individual to maintain an interest in a subject and their resiliency to persevere through obstacles and setbacks to achieve goals. Studies at the
United States Military Academy at West Point (USMA) show how individual grit—i.e., the passion and perseverance or resiliency of West Point cadets—enabled them to overcome extreme challenges and effectively predicted their successful graduation from
USMA (Duckworth et al., 2007).
Focusing on the resiliency subcomponent of grit, one USMA cadet described the arduous conditions imposed during their initial cadet basic training by stating, “Within two weeks I was tired, lonely, frustrated, and ready to quit—as were all of my classmates” (Duckworth, 2016, p. 25). However, as the author goes on to explain, while some quit, others did not, and the key difference seemed to be, “… a ‘never give up’ attitude” (Duckworth, 2016, p. 26). Resiliency represents the determination to keep trying, keep pushing forward, and never give up despite the many obstacles or adversities that may stand in the way. This quality is particularly valuable in the military where
20
individuals face constant adversities on a routine basis. Despite this, little is known about the accuracy or effectiveness of known resiliency measures, how to develop individual resiliency, or how to accurately predict whether an individual possesses resiliency or not.
Dr. Duckworth created a survey tool called the Grit Survey containing two subscales to measure the elements of passion and resiliency. This survey tool was used to predict student success, including that of USMA cadets attending basic training and students participating in the National Spelling Bee (Duckworth et al., 2007; Duckworth &
Quinn, 2009). Dr. Duckworth developed several versions of the Grit Survey, including the latest eight-item Short Grit survey, known as the Grit-S survey, which demonstrated increased improved psychometric properties, test-retest reliability, and predictive validity
(Duckworth & Quinn, 2009). This test proved accurate in a variety of settings and began to reveal how the combined effects of passion and resilience influenced the outcomes of challenges experienced by the individual. These and other related studies led to the claim that “To be gritty is to keep putting one foot in front of the other . . . to hold fast to an interesting and purposeful goal . . . to invest, day after week after year, in challenging practice . . . to fall down seven times, and rise eight” (Duckworth, 2016, p. 275).
Understanding resiliency is a critical first step to understanding how to change and improve individual resilience, especially in the adversity-filled profession of serving in the military. However, before moving to the application of resiliency theory, it is essential to understand the key elements that make up the framework of this theory and how these elements operate in relation to each other.
21
Resiliency Framework
Resiliency theory is a strengths-based theory that, instead of focusing on deficits, focuses on individual strengths that lead to healthy development and positive outcomes regardless of the level of risk exposure (Fergus & Zimmerman, 2005; Zimmerman,
2013). Resiliency theory involves the concepts of adversity, the promotive factors that include internal assets and external resources, and protective and vulnerability factors which all influence resiliency in various ways. Resiliency theory posits that individuals experience risks or adversities in life and possess promotive factors of varying types and degrees that may support their ability to overcome the adversities (Fergus & Zimmerman,
2005; Zimmerman, 2013). Promotive factors consist of the internal assets and external resources an individual has the knowledge and ability to employ when attempting to overcome adversity. Internal assets include traits inherent to the individual, such as positive identity, competence, hope, self-esteem, self-efficacy, coping skills (Fergus &
Zimmerman, 2005), mindset (Dweck, 2008), and mindfulness (Duckworth, 2016). These internal assets focus on the inherent positive strengths of the individual that “promote” successful resilience when disrupted by risk and adversity. External assets are the resources available to the individual to which they can turn for assistance and include parental support, youth programs, and adult mentors (Fergus & Zimmerman, 2005;
Kiswarday, 2012; Zimmerman, 2013; Zimmerman et al., 2013). They are those elements external to the individual but that the individual can choose to pursue if they decide to.
A keen sense of self-efficacy, fostering a growth mindset, and developing mindfulness are also important internal assets that contribute to resilient behavior. Initial research defined self-efficacy as “the belief in one’s capabilities to organize and execute
22
courses of action required to produce given attainments” (Bandura, 1997, p. 3).
Researchers working with military personnel define self-efficacy as “the belief that you can master your environment and effectively solve problems as they arise” and note its essential role in developing resilience (Reivich & Shatté, 2003, p. 19). Studies have shown that self-efficacy independently predicts resilience (Collishaw et al., 2016), strengthens resilience to challenging experiences (Bandura, 1997), and is closely related to academic performance (Chemers, Hu, & Garcia, 2001; Zajacova, Lynch, &
Espenshade, 2005). Self-efficacy plays a key role in developing resiliency because
“resilience requires some degree of belief in the ability to exert control over the social environment” (Bender & Ingram, 2018, p. 20), and persevering through challenging obstacles requires resilient self-efficacy to manage failure (Bandura, 2012). Overcoming obstacles starts with the internal belief that a person can make decisions, take action, and bring about effects resulting in overcoming obstacles and challenges in a way that brings positive growth. However, without the proper mindset, even self-efficacy can be a weak contributor to a resilient attitude.
Mindset can also be a determining factor in the development of resiliency. The idea behind mindset is that people typically adopt either a fixed viewpoint, wherein intelligence, personality, and character are considered static traits with value placed on the individual achievement of tasks, or a growth viewpoint, wherein intelligence, personality, and character is malleable, gained through both success and failure in completing tasks, with value placed on individual effort and the learning process (Dweck,
2006; Hochanadel & Finamore, 2015; O’Brien & Lomas, 2017). Research demonstrates that individuals’ mindsets can change and develop and that as they do, their resilience
23
increases (Yeager & Dweck, 2012). Researchers studying different groups in an outdoor personal development program found “only the group who engaged with structured processing specific to overcoming challenge with support, application of effort, and highlighting transferable strategies for overcoming setbacks showed a significant change in their resilience levels” (O’Brien & Lomas, 2017, p. 142). Their research supported the conclusion that “interweaving a structured and focused intervention . . . can have a significant impact on participants’ mindset as well as increasing their resilience”
(O’Brien & Lomas, 2017, p. 144). This suggests mindset has a direct influence on an individual’s ability to experience self-efficacy and engage in the types of behaviors that would enable self-improvement and the development of resilient behaviors.
Mindfulness can be an important element in developing resilience as it involves an awareness of the present as a way to understand, evaluate, and deliberately move forward. Researchers describe mindfulness as “a process of regulating attention in order to bring a quality of nonelaborative awareness to current experience and a quality of relating to one’s experience within an orientation of curiosity, experiential openness, and acceptance” (Bishop et al., 2004, p. 234) and an “awareness that emerges through paying attention on purpose, in the present moment, and nonjudgmentally to the unfolding of experience moment by moment” (Kabat‐Zinn, 2003, p. 145). One researcher described the experiences of an accomplished doctor who began developing mindfulness through meditation as a young man and who was later able to pass along what he had learned to patients with serious health problems (Duckworth, 2016, pp. 155-157). This ability to focus attention on each moment with openness and acceptance can empower individuals to experience hardship in a way that enables them to identify, acknowledge, and accept
24
their own abilities to resolve challenges and provide a path to obtaining new skills to handle the challenges ahead. Studies related to Mindfulness-Based Stress Reduction
(MBSR) strategies have found that using MBSR as an intervention helped individuals experiencing stress in a broad range of situations ranging from everyday living to extremely challenging circumstances involving severe disorders (Grossman, Niemann,
Schmidt, & Walach, 2004). Several mindfulness programs have been proposed to build resiliency among military servicemembers (Thomas & Taylor, 2015), and research shows that mindfulness training programs can help individuals learn to mitigate the negative effects of stress (Johnson et al., 2014). Thus, the research shows that self-efficacy, mindset, and mindfulness can be significant internal asset promotive factors which influence both the development and demonstration of resiliency as they affect how an individual values growth and believes that they can, in fact, bounce back from setbacks.
Protective and vulnerability factors are elements that may influence the effects of adversity depending on the individual and their background (Braverman, 2001; Fleming
& Ledogar, 2008; Luthar & Cicchetti, 2000; Luthar et al., 2000). Protective factors are sometimes closely related to promotive factors, but they extend past the individual level to include supportive family networks, socioeconomic status, school experiences, supportive communities, and cultural resources (Fleming & Ledogar, 2008). Protective factors serve to counteract or ameliorate the effects of adversity and thus guard the individual from adversity itself (Braverman, 2001). Vulnerability factors represent the negative qualities, experiences, or lack of protective factors that make individuals more susceptible to adversity and intensify the risk effects. Both factors may be present as a result of the presence or absence of family support and strong community identity and
25
access to or the lack of assistive resources (Braverman, 2001; Fleming & Ledogar, 2008;
Luthar & Cicchetti, 2000; Luthar et al., 2000). Thus, based on their unique life experiences, everyone will have unique combinations of protective factors and vulnerability factors and will likewise handle adversity very differently.
Dr. Glenn E. Richardson, director of graduate studies in the department of health promotion and education at the University of Utah, developed the resiliency model in figure one below; it depicts the process an individual goes through when facing adversity and the possible results.
Figure 1. Richardson Resiliency Model (Richardson, 2002).
26
As Richardson (2002) explains in his resiliency model, all individuals, regardless of demographic differences or prior experiences, begin the resiliency process in a state of bio-psycho-spiritual homeostasis (Richardson, 2002). This comparatively stable state is the starting point for an individual prior to experiencing a difficult situation or an adversity. At this point, the individual has already arrived at the situation with both their internal promotive factors and a measure of both protective factors and vulnerability factors developed from their environment that influence how the individual will deal with adversity. Thus, as adversity disrupts their state of homeostasis, the protective factors serve to strengthen the individual against adversity while vulnerability factors serve to weaken the individual and may intensify the effects of adversity. Of course, every individual has a unique array of experiences and influences, and therefore, each has varying amounts of both protective and vulnerability factors when faced with adversity.
According to Richardson’s model, the individual is then “disrupted” by adversity, which comes from a variety of stressors and life events to which the individual must react. Following this period of disruption begins the reintegrative process whereby the individual processes or reacts to the adversity and attempts to regain a state of stable homeostasis. The end result of this reintegrative process can range from dysfunctional reintegration, where the individual is weaker or more susceptible to adversity than they were before, to resilient reintegration, where the individual is stronger or less susceptible to adversity than when they first encountered it (Richardson, 2002). This process is affected by a variety of factors, including the protective factors, vulnerability factors, and other influences such as the individual’s environment, the promotive factors that build internal problem-solving processes, and the disruptive effects caused by both simple and
27
complex adversities (Zimmerman, 2013). Research has shown that individuals in general are surprisingly good at demonstrating resilient behaviors following loss and trauma
(Bonanno, 2004). However, individual results of this process depend on so many factors, including individual upbringing, that it has historically been difficult to predict individual success or failure given a particular set of adversities.
Definitions of Resiliency
It is good to note there are many definitions of resiliency from a variety of sources. A quick internet search for “resiliency definitions” returns over 4.2 million results, including a vast array of definitions from different fields like business, healthcare, and psychology. Many of these definitions include the concept of bouncing back from adversity (Dyer & McGuinness, 1996; Ledesma, 2014; VanBreda, 2001), returning to a former shape after being bent or pulled (Resilience, 2018), or rising every time you fall
(Duckworth, 2016). There are also several other words often deemed synonymous with resiliency, including “hardiness,” “perseverance,” and even “grit.” However, with so many definitions available, it is necessary for this study to operationalize a definition applicable to military personnel and higher education students.
To quantify, predict, and standardize an understanding of resilience, researchers have developed a variety of definitions and terms that have changed over time as shown in figure two below. A myriad of scales have also been developed and tested to try to capture individual resilience (Windle, Bennett, & Noyes, 2011). Some researchers have argued for definitions of resiliency applicable to specific populations, topical challenges, and research applications instead of global definitions (Ahern, 2006), and this has led to a
28
wide variety of definitions that may appear confusing and disjointed when considering the overall concept of resilience.
Source Population Definition Garmezy (1991) Global The capacity to recover and maintain adaptive behaviors after insult. Greenspan (1982) Infants and children The capacity to successfully undertake the work of each successive developmental stage. Hunter and Chandler (1999) Inner city, vocational high Process of defense using such tactics as insulation, isolation, school adolescents disconnecting, denial, aggression, as a process of survival. Mandleco and Peery (2000) Children and adolescents Capacity to respond, endure, and/or develop and master in spite of experienced life stressors. Markstrom et al. (2000) Rural, low income, Adaptive, stress resistant personal quality that allows the individual Appalachian adolescents to thrive despite unfortunate life experiences. Rew et al. (2001) Homeless adolescents Beliefs in one's personal competence and acceptance of self and life that enhance individual adaptation. Rouse and Ingersoll (1998) Adolescents in high school The ability to succeed, amature, and gain competence in a context of adverse circumstances or obstacles. Wagnild and Young (1993) Adult women The ability to successfully cope with change and misfortune.
Figure 2. Comparison of Resiliency Definitions (Ahern, 2006).
Researchers have cited the strong connection between resiliency and hardiness
(Ahern, 2006; Ledesma, 2014), and many have developed theories surrounding other concepts synonymous with resilience using other terms such as “hardiness” (Bartone,
2006; Bartone et al., 2008). However, reviewing this table and other common definitions from available resilience literature reveals a common thematic cycle wherein an individual possesses and demonstrates the ability to face an adversity, to process and react to that adversity, and to emerge with a new homeostatic state after regaining stability or normalization. While similar, these definitions raise questions of how resilient various populations are, how they became that way and why, and then how to compare the varying severity of adversities these populations each face. The answers vary widely along with the terms used in each case to explain resilience.
As resiliency research has progressed and definitions for resiliency have changed, researchers have moved through several “waves” of theory from identifying resiliency in terms of individual traits and qualities, to protective factors designed to help cope with
29
stressors, and then later combining the two into an internal motivational drive toward self-actualization (Allan et al., 2014; Fletcher & Sarkar, 2013; Masten, Herbers, Cutuli, &
Lafavor, 2008). The latest research has proposed that as resiliency theory moves past the limited understanding of individual character traits or assets (Pashak, Hagen, Allen, &
Selley, 2014; Rutter, 1985, 1987), that resiliency is actually the result of the development of dynamic processes of positive adaptation whereby an individual learns to overcome adversity (Fleming & Ledogar, 2008; Luthar et al., 2000). Thus, resiliency is not merely a static trait or state of being an individual does or does not possess and it is not just the presence or awareness of internal resources, external resources, or protective factors extant in the surrounding environment. So, what exactly is resiliency?
Since this study focuses on the resiliency of USAFA cadets, the operational definition used in this study will be the definition developed by the Defense Centers of
Excellence of Psychological and Traumatic Brain Injury (DCoE) and adopted by the Air
Force to better understand resiliency. They define resiliency as “The ability to withstand, recover and grow in the face of stressors and changing demands” (U.S. Department of the
Air Force, 2014, p. 14). This definition uses the verbs “withstand, recover, and grow” indicating resiliency is a dynamic action instead of a static trait or characteristic. This action-based definition captures the focus embodied by USAFA CCLD and provides a clear reference point from which to study the leadership and resiliency development process of USAFA cadets.
Resiliency-Building Models
In 1984, researchers proposed the compensatory model, the immunity vs. vulnerability or protective model, and the challenge model to aid in evaluating the nature
30
of individual resiliency and to help develop resiliency (Garmezy et al., 1984). These models continued to guide researchers as they applied the principles of resiliency theory to curricular strategies and operational practices to develop and enhance individual resiliency (Fleming & Ledogar, 2008; Ledesma, 2014; Richardson, 2002). All three of these models pursue a systematic approach to evaluating and building resiliency to stress regardless of individual background or experiences and “suggest themselves for the impact of describing stress and personal attributes on quality of adaptation” (Garmezy et al., 1984, p. 102). Resilient adaptation is the goal, and the question is how the promotive factors, protective factors, vulnerability factors, and adversities influence an individual’s ability to adapt.
Compensatory model. The compensatory model is outcome-focused and evaluates an individual’s ability to employ the internal promotive factors to reduce the effects of negative outcomes. This model is based on the idea that with personal attributes remaining the same, either the promotive factors will overpower the effects of the negative outcomes to the benefit of the individual or that the negative outcomes will overcome the promotive factors to the detriment of the individual (Garmezy et al., 1984).
The focus is on the presence, absence, and relative strength of the promotive factors and their ability to directly counteract or “compensate” for the effects of adversities or negative outcomes (Fergus & Zimmerman, 2005; Ledesma, 2014). This is represented in
Richardson’s model by the promotive arrows pushing upwards against the downward risk arrows (Richardson, 2002). The compensatory model proposes ideally that an individual’s promotive factors and ability to employ them can be developed and improved to overcome the effects of adversity more effectively.
31
The applications of this model could include adults mentoring children living in poverty to prevent youth violence where the adult mentorship compensates for the effects of poverty on children who become less inclined to engage in violence (Fergus &
Zimmerman, 2005). In effect, the adult mentorship to the children has a compensatory effect on the negative effects of poverty. The intended result is the children engage less in violence because of the mentorship. The level to which an individual emerges more resilient from the experience depends both upon the presence, strength, and ability to employ their promotive factors and the magnitude of the adversity and its effects
(Zimmerman et al., 2013). Thus, results of using the compensatory model will be unique to every individual.
Protective model. The protective model changes the focus from the outcomes to the risks leading to negative outcomes in an effort to prevent or avoid the adversity altogether. This model focuses on building resiliency by using the same promotive factors to moderate the effects of risks in an attempt to reduce or avoid negative outcomes (Fergus & Zimmerman, 2005; Ledesma, 2014; Zimmerman, 2013). In other words, the individual promotive factors interact with the adversity factors to change the resulting effects and outcomes. Ideally, the protective model capitalizes on the promotive factors in a way that either avoids the risk factors and negative outcomes altogether or changes them in such a way as to avoid their negative effects (Fleming & Ledogar,
2008). Thus, the goal is to change the outcome by employing the promotive factors to minimize the risk factors leading to the negative outcomes.
Application of this model may include drug use (a risk factor) and its connection to sexually risky behavior (negative outcome) that may be counteracted by
32
comprehensive training and education in both the dangers of drug use and sexually risky behavior (Fergus & Zimmerman, 2005). The training and education can modify the individual’s reaction and response to the risk factors and is a type of external resource within the promotive factors toolkit that an individual can call upon to change the negative effects of adversity (Kiswarday, 2012). Thus, success in the model depends on reducing both the effects of risk factors and/or the risk factors themselves.
Challenge model. The challenge model is different from the compensatory or protective models in that instead of seeking to minimize contact with adversity, counteract its effects, or avoid negative outcomes altogether, individuals are intentionally exposed to moderate amounts of adversity in controlled environments to build a kind of immunity to the risk (Garmezy et al., 1984). In this model, individuals can face adversity and wrestle with it under the supervision of others who are ready to help, provide guidance, reinforcement, and aid if necessary, in overcoming the adversity. This type of model serves as an inoculation against adversity that prepares the individual for the next adversity (Ledesma, 2014; Zimmerman, 2013). Researchers note that exposure to levels of adversity too high for the individual to adapt to are counterproductive while exposure to levels of risk too low do not pose a sufficient amount of stress, resulting in little to no positive effect (Fergus & Zimmerman, 2005; Fleming & Ledogar, 2008). After building experience through guided practice and development, individuals may enhance their abilities to process adversity as they develop an understanding of what to do, how to perform, and how to process the adversity (Kiswarday, 2012). This inoculation or steeling process familiarizes the individual with the adversities and prepares them for
33
facing those adversities in real life when restarts are not a possibility and the consequences are real (Fergus & Zimmerman, 2005).
Applications of this model are common especially in the context of athletics or completing physically demanding tasks like football practice, military drills, dress rehearsals, or learning to talk through a fight or argument wherein the stress of similar adversities are present with the expectation of performing a task (Zimmerman, 2013). In this practice environment, individuals can engage in a process whereby they struggle with adversity, experience both success and failure, make mistakes, discuss other strategies, evaluate viable solutions, and are then able to “restart” and face the adversities again.
For coaches, this model is commonly referred to as “doing reps” and is as simple as running athletes repetitively through difficult drills in practice to prepare them for the challenging demands of a real game. For service members in the military, conducting drills and exercises under stressful conditions has the same of effect of preparing them for the adversities they will face while deployed where the demands are extreme, and the outcomes can be lethal.
Commonalities between models. Common to all the models is the concept of repetition and practice. With enough practice, everyone can develop resiliency, as retired
Navy Seal Eric Greitens explains, simply because “We become what we do if we do it often enough” (Greitens, 2015, p. 29). In his discussion of Aristotle’s Ethics, author Will
Durant further discusses the concept of habituation by quoting the ancient philosopher who claims that excellence is a habit:
Excellence is an art won by training and habituation: we do not act rightly because we have virtue or excellence, but we rather have these because we have acted rightly; “these virtues are formed in man by his doing the actions”; we are what we repeatedly do. Excellence, then, is not an act
34
but a habit: “the good of man is a working of the soul in the way of excellence in a complete life . . . for as it is not one swallow or one fine day that makes a spring, so it is not one day or a short time that makes a man blessed and happy.” (Durant, 1961, pp. 61, emphasis added)
Thus, while reviewing the following models, it is important to remember that the key to success in any of them is repetition and re-evaluation until excellence, and resilience becomes a habit. This process of repetitive effort followed by introspective re-evaluation creates a steeling effect on the individual who becomes better able to process adversity.
While these models utilize promotive factors in slightly different ways, have slightly different emphases, and are appropriate in different environments, all have the goal of increasing the individual resiliency necessary to overcome actual adversity, and all try to result in a positive adaptation (Fletcher & Sarkar, 2013). While the models use different strategies, their common goal is to aid individuals in overcoming the adversities they encounter in such a way as to enable them to learn and grow more resilient as they encounter adversities, deal with them, and reintegrate into one of several new states of homeostasis (Fleming & Ledogar, 2008; Richardson, 2002). Thus, whereas the definition of resiliency refers to the ability to withstand, recover, and grow in the face of stressors and changing demands, the resiliency models seek to build and develop this ability.
Strengths and Weaknesses of Resiliency Theory
Researchers have repeatedly reported the merits of resiliency theory and resiliency-related concepts such as grit and hardiness, and several career fields have looked to resiliency theory to improve their various approaches to improve the ability to overcome adversity. Researchers have used resiliency theory to help explain the successful development of adolescents despite being at risk for substance abuse, teen pregnancy, and family dysfunction and the use of successful interventions to build and
35
promote resilient behaviors (Braverman, 2001; Carlos & Enfield, 1998; Fergus &
Zimmerman, 2005; Garmezy, 1993). The medical field has turned to resiliency theory and its elements of vulnerability and protective factors to assist in addressing issues faced by mental health patients and their families (Dyer & McGuinness, 1996). The education field has lauded the effects of teachers applying resiliency theory to assist struggling students (Kiswarday, 2012) and the protective factors including family and institutional support as important factors in the success of low-income, first-generation students
(Mbindyo, 2011). In several studies, Dr. Duckworth demonstrated the effectiveness grit has on predicting the successful graduation of USMA cadets and superior performance of participants in the National Spelling Bee (Duckworth et al., 2007). Grit was also a significant predictor of completing the Army Special Operations Forces (ARSOF) selection course, staying in high school, remaining faithful in marriage, and retaining employment (Eskreis-Winkler, Duckworth, Shulman, & Beal, 2014). Hardiness was determined to effectively predict success of U.S. Army Special Forces candidates
(Bartone et al., 2008) and was shown to increase leader effectiveness and unit cohesion in the military (Bartone, Barry, & Armstrong, 2009).
These examples highlight several aspects of resiliency theory proven effective in understanding human behavior in different settings and in predicting successful behavior and performance of individuals when promotive and protective factors are increased and enhanced to counter the effects of adversity. This concept of using personal strengths to overcome adversity (Perez, Espinoza, Ramos, Coronado, & Cortes, 2009) and working to turn personal weaknesses into strengths resonates with students, teachers, medical professional and their patients, and those serving in the armed forces worldwide.
36
Additionally, the concept of turning weaknesses into strengths to overcome adversity speaks to people of all ages and is a practice anyone can engage in regardless of prior experience or background.
However, while the benefits and strengths of resiliency theory are well documented, some critics of resiliency and grit have challenged the concept and written articles pointing out the weaknesses inherent to the concept of resiliency. One perspective is that resiliency-type theories imply more resilience is always better without recognizing the dangers of being over-resilient. Authors of this argument cite examples of workers attempting to push through unhealthy circumstances they should actually leave, overly resilient leaders who appear unemotional and lack the self-awareness to know when they should change course, and those who push their organizations to complete tasks to the detriment of their employees (Chamorro-Premuzic & Lusk, 2017).
Scholars have also argued that if resiliency is built through experiencing difficulty or hard times, many will assume the more difficulty they experience or the harder the time they put themselves through, the more resilient they will become, leading many to remain in difficult and even dangerous circumstances much longer than they should (Kashdan,
2017). These scholars question whether or not an individual can become too resilient or so resilient that it becomes a detriment to themselves or others.
Another challenge to the concepts closely related to resiliency theory are the emergence of research studies claiming grit and its subscales of passion and perseverance or resiliency have a much smaller impact on performance in a variety of settings than previously claimed. One research study conducted a meta-analysis of over 73 independent research studies on grit using 88 unique samples and 66,807 individuals and
37
found that grit was actually a weak predictor of overall performance, that the grit subscale of passion was not significantly predictive of overall performance, and that the subscale of perseverance or resiliency was not as reliably predictive of academic performance as other study skills and habits, adjustment to college, and class attendance
(Credé et al., 2017). This questions research findings claiming resilience predicts performance.
Another critic asserts building resilience and grit is only available to those who live in privileged communities and have the opportunities and resources available to improve their promotive and protective factors (Schreiner, 2017). The author argues the only familial or societal support that can effectively assist in building grit are those where wealth is a factor and resources are abundant. Where resources are slim, particularly in low-income areas, the family and community support necessary to help build the protective factors and increase resiliency are simply not available. Schreiner also explains how grit theory focuses on a deficit ideology that “identifies the personal shortcomings of people who are struggling, focusing on individual attitudes, behaviors, mindsets, and characteristics that impede their success” (Schreiner, 2017, p. 14). This results in a type of “victim-blaming” where students may be blamed for their own weaknesses. This standpoint supports the position that grit theory and/or resiliency theory unfairly places the responsibility of building grit solely on the shoulders of less gritty individuals without accounting for the nuances and unique circumstances contributing to their deficits in grit. This perspective suggests individuals can do relatively little to change their own grit without addressing other pre-existing and/or
38
current factors in their environment, and that to do so places blame primarily upon the individual for not being or not becoming more resilient.
This study engages these claims with quantitative research questions designed to add clarity to the relationship between resiliency and performance by exploring both positive and negative associations between individual factors and resiliency and the magnitude of those associations. Coupled with an examination of the extent to which changes in resiliency occur over time, this study provides additional insight into the influence of resiliency-building elements built into the character and leadership development programs cadets experience at USAFA. The following review of relevant literature examines leadership and its relationship with resiliency in higher education, the military, and in the leadership development programs at USAFA.
Review of Relevant Literature
Leadership
Leadership is inherent to the military organization, and the process of developing high-quality leaders is a priority, especially to military service academies like USAFA that are tasked with developing future military leaders. The subject of leadership has been discussed for centuries and was of great concern to those leading the earliest civilizations, including the pharaohs in Egypt and the leaders in ancient Greece and Rome
(Avery, 2004). The works of Plato discuss various aspects of leadership; the value of the best leaders and rulers over society; and the attributes, characteristics, and traits that the best leaders should have (Jowett, 1989). After several centuries, leadership theories began to emerge with the Great Man Theory in the 1840s, followed by trait theories, behavioral theories, and theories focusing on the interactions between leaders and those
39
whom they lead (Avery, 2004). Today, the discussion of leadership continues, although the focus has changed slightly with current philosophies focusing on leadership styles and which styles are most effective (Goleman, 2000), how to adopt different leadership styles, and whether leadership is something you learn or are born with (Avery, 2004).
As the discussions over leadership have continued, a multitude of leadership styles have been identified, beginning with the classic “Great Man Theory” leadership style (Avery, 2004; Zaccaro, 2007) and continuing with transactional leadership (Avery,
2004; Northouse, 2016), transformational or visionary leadership (Avery, 2004;
Deichmann & Stam, 2015; Hewitt, Davis, & Lashley, 2014), servant leadership
(Allender, 2006; Beck, 2014; Duncan & Pinegar, 2002; Northouse, 2016), authentic leadership (Avolio, 2007; George, Sims, McLean, & Mayer, 2007; Walumbwa, Avolio,
Gardner, Wernsing, & Peterson, 2007), level-three leadership (Clawson, 2009), principle- centered leadership (Covey, 1992), and the leaderless shared or organic leadership style
(Ancona, Bresman, & Caldwell, 2009; Avery, 2004; Pearce & Conger, 2002). Some leadership styles are regarded as superior to others due to their ability to motivate others, develop close relationships between leaders and subordinates, or result in maximum productivity and mission completion. However, scholars have highlighted research findings demonstrating the benefit and necessity of mastering all leadership styles and knowing how and when to use them in the way a professional golfer would use a variety of golf clubs to put the ball in the cup (Goleman, 2000), suggesting there is no one-size- fits-all style of leadership (Zaccaro, 2007). Leadership, then, is much more than a static or well-defined trait or style; it is an understanding of how to engage a variety of people
40
applying different leadership styles in a process of mutual progression. This is both the art and science of leadership in perfect balance.
Definitions of leadership abound, and most include the idea that some entity influences some other entity to act in a particular way, often resulting in the achievement of a common goal or desired endstate (Avery, 2004). However, current theorists agree that leadership is more than a characteristic, trait, or style. Leadership engages in a process of influencing others to complete a common goal (Northouse, 2016), and effective leadership occurs when leaders work with successful teams to accomplish their missions (Willink & Babin, 2015). Crisis situations add an element of stress, requiring even the greatest leaders to make complex decisions under extreme pressure, chaos, and shifting conditions. These situations pose the greatest test of leadership and test the moral and ethical character under extreme circumstances (C. E. Johnson, 2017). With these chaotic conditions in place, the stage is set for one of the most challenging environments an individual could ever face: that of leading others amidst the hellish chaos of war while influencing them to work together to achieve a common goal. In this environment, the principles of both leadership and resiliency come together in a way to help individuals process the adversity around them and lead their organizations through the adversities they face.
To understand how to apply leadership and resiliency theories in different situations, it is useful to discuss environments where some level of both leadership and resiliency is necessary to overcome the inherent adversities. A discussion of both higher education institutions and the military follows to better understand the variety of adversities and risks present, the policies, training, and programs designed to address the
41
adversities, and the attempts made to measure the success of these programs and evaluate the results. This is followed by a discussion of USAFA which is a uniquely challenging environment in that it combines a higher education institution with leadership development in a military atmosphere. Before looking at USAFA, it is helpful to understand resiliency in the context of both higher education and the military and review efforts to alleviate the challenges posed to university students and military service members alike.
Resiliency in Higher Education
In addition to being a military service academy, USAFA is an institution of higher education that offers 31 academic majors and four minors in a variety of subjects, including physics, computer engineering, aeronautical engineering, mathematics, and military and strategic studies (The United States Air Force Academy, 2018a).
Additionally, over half of the 31 majors offered are in the field of science, technology, engineering, and mathematics (STEM), and in 2016, USAFA was rated by Forbes as the number five STEM university in the nation (Forbes, 2018). This places USAFA in the unique position of developing future military leaders for USAF and also future leaders in
STEM fields and career paths. Being a cadet at a military academy and being a student in a STEM university are each daunting challenges by themselves due to the strenuous nature of the training programs. Combining them into a singular institution poses a uniquely demanding setting for even the best students and provides an excellent research environment to explore the relationship between resiliency and performance.
The challenges to students in higher education. The world of higher education can be challenging for students for a variety of reasons. Students enter the realm of
42
higher education and face a myriad of adversities, termed “risk factors” in resiliency theory, each with the potential of posing significant challenges to their educational progress. Some of these risk factors may include coming from lower socioeconomic status (SES) or low-income families, ethnic minority families with limited access to education, and single-parent homes with parents focused on income instead of academic achievement (Gutman, Sameroff, & Eccles, 2002). In addition to the typical risk factors faced by every individual, students entering higher education experience a completely new world full of adversity ranging from the challenges of admissions, dealing with roommates, professors, financial aid, classes, and ultimately the challenges of graduating
(Allan et al., 2014). Students experiencing significant challenges from this new environment often have low grade point averages and struggle with poor academic performance (Gutman et al., 2002). This can lead to an increase in attrition rates as some students get discouraged and leave the institution prior to graduation because they are unable to adapt to the new environment or maintain academic standards (Garza et al.,
2014). Additionally, STEM programs in particular have experienced high attrition rates with one study reporting 48% of bachelor’s degree students and 69% of associate degree students leaving STEM programs for non-STEM programs between 2003 and 2009
(Chen, 2013). Together, these risk factors make it difficult for many students to succeed, much less thrive in the challenging environment of higher education.
Efforts to help students overcome challenges in higher education. To address these challenges, some academic professionals have turned to the principles of resiliency theory, specifically focusing on developing the promotive and protective factors to build students’ abilities to experience adversity and learn to process and overcome the risk
43
factors (Perez et al., 2009). In higher education, resiliency sometimes refers to the
“interactive and accumulating process of developing different skills, abilities, knowledge and insight that a person needs for successful adaptation or to overcome adversities and meet challenges” (Kiswarday, 2012, p. 94). Being resilient can assist in overcoming these challenges and assist students in experiencing improved performance (Allan et al.,
2014). Researchers have explored ways to increase graduation rates and decrease attrition using resiliency theory to focus on compensatory models using promotive and protective influences to determine what faculty could do in their classrooms to affect and increase the resiliency of their students (Masten et al., 2008). A qualitative research study found that students appreciated teachers who built their individual self-efficacy, helped them effectively self-appraise their own strengths and weaknesses, encouraged help-seeking tendencies, and explained the clear links between academic success and future security (Morales, 2014). Multiple methods, including utilizing a list of developmental assets, have also been identified to understand the factors leading to increased performance in college students (Pashak et al., 2014). These findings explored the adverse risk factors causing students to underperform or drop out, the promotive factors that help them succeed, and the findings suggested actions to staff, faculty, and institutions to build student resilience, therefore reducing attrition and increasing performance.
Results of efforts to help students. However, despite several examples of resiliency theory in action to the benefit of students, the research evaluating resiliency through grit surveys in relation to student performance has mixed results. Researchers have reported findings indicating a close relationship between grit and academic
44
performance and retention (Eskreis-Winkler et al., 2014), including research by Dr.
Duckworth claiming grit is closely related to student performance with grittier students consistently performing better (Duckworth et al., 2007). However, while more recent studies also found a close correlation between performance and the resiliency component of grit, they did not find a close correlation with the consistency of interest or passion aspect of grit (Bowman, Hill, Denson, & Bronkema, 2015; Credé et al., 2017; Wolters &
Hussain, 2015). Other studies have reported no significant correlation between grit and performance, including in STEM programs (Bazelais, Lemay, & Doleck, 2016; Chang,
2014). These conflicting studies highlight the need for further research to understand how grit and its subcomponents of passion and resiliency are related to student performance and to identify and test exploratory models and methods designed to build resiliency. This research study revisits the association between resiliency and student performance and attrition to clarify the mixed results of past research studies and determine the differences and similarities which exist at USAFA.
Resiliency in the Military
Particularly in the military, personal hardiness and resiliency can reduce the effects of stressors commonly found in contemporary military operations, and leaders who improve their own hardiness are better able to influence subordinates in developing their own personal hardiness (Bartone, 2006). Researchers have determined personal hardiness, synonymous with resiliency, to be the largest predictor of military leadership performance for both men and women across different contexts (Bartone, 2006; Bartone,
Eid, et al., 2009; Bartone et al., 2013) and an important indicator of mental health (Eid,
Johnsen, Bartone, & Nissestad, 2008; Ramanaiah, Sharpe, & Byravan, 1999). Leadership
45
in the military is critical for mission success and is the element of combat power that unifies all other elements of combat power (U.S. Department of the Army, 2012). In an attempt to identify styles of leadership most effective in the military, studies have found that some leadership styles foster resilience in subordinates (Eid et al., 2008; Gaddy,
Gonzalez, Lathan, & Graham, 2017). However, despite this understanding, the ever- changing nature of global conflict continues to present new challenges to military service members, and no one was ready for the changes brought about by the events that occurred on the morning of September 11, 2001.
The challenges to service members in the military. Not long after the terrorist attacks on 9/11 and the ensuing war on terror in the Middle East, military leadership increased their attention on the overall psychological resilience of soldiers returning from combat tours. Some soldiers returned from combat with severe physical injuries, others with severe psychological injuries, but all returned permanently changed as a result of the ravages of war. Military leaders took increasing notice and began prioritizing ways to build resilience in soldiers to strengthen them against the adversities of combat. Around the same time in 2003, Dr. Karen Reivich from the Penn Resiliency Project (PRP) at the
University of Pennsylvania and Dr. Andrew Shattee from the Department of Family and
Community Medicine at the University of Arizona published The Resilience Factor: 7
Keys to Finding Your Inner Strengths and Overcoming Life’s Hurdles, which detailed different ways to overcome adversity (Reivich & Shatté, 2003). Military leaders took notice of Dr. Reivich’s research and began working to develop a program to help soldiers and their families.
46
Efforts to help service members overcome challenges in the military.
Understanding the need to engage in the research and development of resilient leaders, both the United States Army and the United States Air Force initiated research into the development of resilient leaders to better understand and develop resiliency within their ranks. The Army stresses the need for every soldier to be resilient and respond positively after facing adversity and argues that leaders should train their units to be resilient now and in preparation for future adversity (U.S. Department of the Army, 2012, 2014). The
Army defines resilience as “The mental, physical, emotional, and behavioral ability to face and cope with adversity, adapt to change, recover, learn, and grow from setbacks”
(U.S. Department of the Army, 2014, para 1-5a). This highlights the elements of facing adversity and being able to withstand, recover, learn, and grow from it.
In 2008, the Army established Army Regulation 350-53: The Comprehensive
Soldier Fitness (CSF) program in order to build soldier resiliency and address the sharply increasing number of soldiers returning from war with post-traumatic stress disorder
(PTSD), alcoholism, drug abuse, and suicidal tendencies (U.S. Department of the Army,
2014). The program was updated in 2014 and was renamed the Comprehensive Soldier and Family Fitness (CSF2) Program, adding the element of family resiliency to the program along with opportunities for families and family members to receive training that would help build family resiliency as they dealt with the difficulties of multiple deployments, soldiers who came home with injuries, soldiers who came home somehow mentally different from when they left, or soldiers who didn't come back at all. As stated in the current regulation, the purpose of the CSF2 program is “to increase the resilience
47
and enhance the performance of Soldiers, Families, and DACs,” referring to Department of the Army Civilians (U.S. Department of the Army, 2014, para 1-5a).
The CSF2 program identifies five dimensions of strength which serve as the primary conceptual pillars of physical, emotional, social, spiritual, and family resiliency to build the overall resiliency of soldiers (U.S. Department of the Army, 2014). The regulation briefly discusses these five dimensions of strength and focuses on the factors of physical, emotional, social, spiritual, and family as the primary factors contributing to individual resiliency.
• Physical resiliency refers to the overall physical health of the individual and
includes aspects of aerobic fitness, strength, body composition, sleep, nutrition,
and training, and their relation to psychological health.
• Emotional resiliency focuses on the ability of individuals to “approach life’s
challenges in a positive, optimistic way and to demonstrate self-control, stamina,
and good character in choices and actions” (U.S. Department of the Army, 2014,
para 2-3).
• Social resiliency refers to the relationship’s individuals have with a variety of
personal and professional contacts and the communication and friendships
developed between them.
• Spiritual resiliency includes “identifying one’s purpose, core values, beliefs,
identity, and life vision” (U.S. Department of the Army, 2014, para 2-5) and
focuses on developing the inner strength to make ethically sound choices and
persevere through adversity. This dimension does not focus on religion so much
48
as it focuses on drawing upon an individual’s core beliefs from a variety of
sources to strengthen his or her basic character.
• Family resiliency refers to building and strengthening the support networks
individuals have at home in order to promote a safe and secure environment and
avoid the tensions and problems associated with domestic difficulties. With this
dimension, resources and training are available to family members as well as
soldiers to aid in the improvement of family resiliency.
When combined and effectively embodied in a soldier, the Army believes these dimensions comprise the primary elements of individual resilience resulting in an
“individual [who] is better able to leverage intellectual and emotional skills and behaviors that promote enhanced performance and optimize their long-term health” (U.S.
Department of the Army, 2014, para 1-5a). This resilience enables leaders at all levels to provide better leadership to their units and strengthen both individual units and the Army as a whole.
Results of efforts to help service members. To measure and assess the resiliency techniques and skills of its members and conduct training to improve and sustain the overall force, the CSF2 program contains three assessment components consisting of an online assessment and self-development programs, specific training for both trainers and individuals in each unit, and a system of metrics and evaluation used to track and report the results of online assessments and training conducted (Reivich &
Shatté, 2003). The Global Assessment Tool (GAT) is the online training and self- assessment tool used to test the individual’s ability in each resiliency dimension and it contains a series of modules individuals can go through to learn more about each
49
dimension and how they can improve their capacity to embody that dimension. As of
2016, the GAT had been taken over 5.2 million times by soldiers, families, and DACs and has proved to be a measuring tool with high reliability (Vie, Scheier, Lester, &
Seligman, 2016). The CSF2 program and GAT remain in use, and research continues to evaluate its overall effectiveness and value to service members and their families in building both individual and family resiliency.
In 2011, the United States Air Force began developing its own program to improve individual resilience in a “Total Force Fitness” approach that included eight pillars of fitness and was infused with the concepts of resilience and how to increase individual resilience (Meadows, Miller, & Robson, 2016). In 2014, the Air Force established the Comprehensive Airman Fitness (CAF) program designed to “enhance the resilience of individuals, families, and communities” (U.S. Department of the Air Force,
2014, p. 1) using four domains consisting of mental, physical, social, and spiritual fitness.
This is a pared-down version of the Army CSF2 program but establishes no other measurement tool to assess the effectiveness of the program besides tools already in place, although it allows for future assessments to be completed on an as-needed basis depending on the needs of the unit (U.S. Department of the Air Force, 2014, para 4.1).
Research focused on the GAT has produced mixed results, with some lauding the program’s effectiveness at decreasing negative behaviors (Lester, Harms, Herian,
Krasikova, & Beal, 2011) while others express concerns about the overall effectiveness of the program to actually build individual resilience (Brown, 2015; Timmons, 2013).
This research study revisited the association between resiliency and military performance
50
to clarify the mixed results of past military studies and determine the differences and similarities which may exist at USAFA.
Resiliency at USAFA
With a review of existing challenges in higher education and military environments and the benefits of individual resiliency in the face of these adversities, the stage is set to explore the challenges and application of resiliency theory at USAFA.
Being a student at one of the five military academies in the United States, cadets at
USAFA experience a unique set of demands that combine the difficulties of higher education with the challenges of preparation for active-duty military service. In such a taxing environment, the ability of cadets to survive and thrive depends on their ability to adapt to their surroundings and overcome a variety of difficult conditions. Put simply, cadets who succeed and prosper are those who learn to demonstrate resiliency and thrive despite the adversities. Thus, efforts to build cadet resiliency have the potential to decrease attrition rates and improve student performance while still maintaining ambitious standards, resulting in more competent and resilient officers prepared for the rigors of active duty.
The challenges to cadets at USAFA. USAFA is a challenging environment for cadets for several reasons. First, it is the intent of the USAFA leadership to create and foster an environment that challenges cadets by deliberate design to build leaders of character. One of the eight key components integral to the essence of USAFA entitled
“Developing Character and Leadership” asserts:
The Academy’s unique opportunities allow cadets to practice leadership theory and learn from their experiences. Daily leadership challenges and opportunities abound to learn, apply, and refine leadership principles. The intentional and integrative nature of this officer development catalyzed by
51
the Center for Character and Leadership Development, but implemented throughout, is pervasive at USAFA and not available anywhere else. The Honor Code guides this leadership development to set cadets on a path of living honorably. (The United States Air Force Academy, 2017a)
This demonstrates the Academy’s commitment to setting high standards and firm expectations on a routine basis to develop the qualities, behaviors, and traits expected from future Air Force officers. This daily “inoculation” of leadership challenges and high standards reminds one of the repetitively habitual actions of excellence previously mentioned by Aristotle as a key to developing good moral character and habits of excellence.
Another challenge inherent to the environment at USAFA is the nature of the cadets themselves. Admissions rates are highly competitive, with an overall admission rate for the 2016 class of 13% with an average high school GPA of 3.85, where 75% of those admitted had a high school GPA of at least a 3.75 (CollegeData, 2016). Average cadet scores in 2016 for the SAT math and critical reading scores were 663 and 633 respectively, with a 1296 composite score, placing USAFA enrollees in the 85th percentile (Staffaroni, 2018), and the average ACT score was 30, which was in the 93rd percentile (Zhang, 2016). Ninety-eight percent of enrollees represented the top half of their class, with 56% in the top 10% of their class, including nine percent valedictorians and eight percent national merit scholars (CollegeData, 2016). The highly selective admissions process results in a group of individuals considered top performers in the country and from around the world. This creates a student body of high-caliber USAFA cadets who have demonstrated academic and athletic accomplishment, provided volunteer service, exhibited strong personal character, and are therefore already highly resilient individuals. While this results in a group of highly capable cadets, it also creates
52
a highly competitive environment where top students compete for top marks to improve their chances at obtaining an Air Force Specialty Code (AFSC) or career field of their choice. Cadets are ranked using order of merit lists (OMLs) according to their performance in academics, military officership, and physical fitness and OML placement has a considerable influence on which AFSC cadets receive thus deciding their careers in the Air Force. This highly competitive environment, in addition to the stresses of a military academy and a STEM institution of higher education, combines a variety of stressors and adverse conditions that cadets often find difficult to manage.
Efforts to help cadets overcome challenges at USAFA. Recognizing the need to continually develop and support cadets through their challenging academic careers,
USAFA established the Officer Development System (ODS) to develop cadets as part of the Air Force’s force development process (The United States Air Force Academy,
2014a). The purpose of the ODS is to provide the framework by which the nine institutional outcomes are accomplished in order to “1) develop each cadet’s appreciation that being an officer is a noble way of life, 2) foster a commitment to character-based officership, and 3) develop competencies essential to this identity as a character-based officer/leader” (The United States Air Force Academy, 2014a, p. 4). The ODS framework includes three main elements, including the foundation (the why), the goal
(the what), and the process (the how) (The United States Air Force Academy, 2017c).
The foundation sets forth the principles and values underlying the educational process at
USAFA; the goals sets forth USAFA’s nine desired institutional outcomes; and the process, describes the means by which cadets will learn to embody the foundational principles and achieve the institutional outcomes.
53
The process to achieve the goals and outcomes include three key components consisting of the Personal, Interpersonal, Team, and Organizational (PITO) model which describes the competencies cadets should develop; the Leadership Growth Model (LGM), which describes the process leaders and followers use to develop the competencies; and the Guiding Principles, which guide the development and implementation of the entire
ODS process (The United States Air Force Academy, 2017c, p. 11). The LGM model itself details four stages that include setting expectations and giving inspiration, providing instruction, providing feedback, and reflection. This growth model enables leaders and followers to work collaboratively using the guiding principles to develop the competencies in the PITO model, resulting in the accomplishment of the institutional outcomes established by USAFA to develop professional leaders of character.
The nine institutional outcomes developed by USAFA, and based on the Air
Force Institutional Competencies, focus on developing cadets into “professional Air
Force officers who think critically, lead with character, and serve the nation” (The United
States Air Force Academy, 2014b, p. 1). USAFA’s nine institutional outcomes are to develop cadets in 1) critical thinking, 2) application of engineering fundamentals, 3) scientific reasoning and the principles of science, 4) the human condition, cultures, and societies, 5) leadership, teamwork, and organizational management, 6) clear communication, 7) ethics and respect for human dignity, 8) national security of the
American republic, and 9) warrior ethos as airmen and citizens (The United States Air
Force Academy, 2014b, p. 1). Each of the outcomes has a list of proficiencies designed to achieve the intent of each outcome, and the proficiencies relate directly to the courses and programs cadets participate in. Outcome number nine, Warrior Ethos as Airmen and
54
Citizens, has eight proficiencies that support the accomplishment of four character traits.
The third character trait, Demonstrate Service before Self as Related to Physical Courage has two proficiencies including “Proficiency 5: Exhibit grit: a hardiness of spirit and resistance to accept failure despite physical and mental hardships” (The United States Air
Force Academy, 2016c, p. 1). This proficiency acknowledges the need for cadets to develop grit and is designed to identify, develop, and evaluate the programs and activities established to achieve that goal.
USAFA has developed and implemented courses, programs, and training opportunities to build cadet character and leadership, challenge decision-making ability, and increase mental and physical toughness. This curriculum is designed to span all four years of the cadet experience with each class of cadets experiencing unique and specific programs. While the full details of specific course and program descriptions are available online (The United States Air Force Academy, 2016a), a brief review of some of the past and current programs is presented below. It is important to note than USAFA constantly strives to capitalize on the latest research and information regarding character and leadership development. It therefore comes as no surprise that USAFA continues to review and update its programs to apply the most current research theories and maintain relevant and effective programs to develop the best officers possible.
USAFA conducts a variety of character and leadership seminars focused on developing the specific needs of each class of cadets from incoming or fourth-year cadets through seniors or first-year cadets, often referred to as “firsties”. Incoming first-year enrollees, called “basic cadets”, experience the physical and mental rigors of basic cadet training (BCT) prior to beginning the academic semester. This six-week introduction to
55
Air Force life consists of garrison and field training designed to challenge individuals physically, mentally, and emotionally. Prior to 2017, the Center for Character and
Leadership Development (CCLD) at USAFA was also conducting several character and leadership seminars focusing on the skills and development of different class years of cadets. The fourth-year cadets participated in a 12-hour, two-phase program called
VECTOR (Vital Effective Character Through Observation and Reflection) where they examined their own values, purpose, vision, and influence, and then committed to a career of honor and integrity. Third-year cadets (sophomores) attended the Respect and
Responsibility (R&R) seminar consisting of an eight-hour seminar/outdoor adventure program designed to expose them to differences and similarities in their own leadership styles, develop communication skills, foster cooperation and trust, and challenge negative views and biases. Second-year cadets (juniors) participated in an eight-hour Leaders in
Flight Today (LIFT) seminar focused on team-building, character, and leadership development and culminating in an exercise with their squadron. The “firsties” attended a culminating eight-hour Academy Character Enrichment Seminar (ACES) program designed to focus on the character and ethical demands placed on Air Force officers, moral and ethical decision-making skills, and their leadership impact on underclassmen.
In 2017, CCLD augmented and consolidated the program into the current ODS program
(The United States Air Force Academy, 2017c). This new system takes elements from all prior seminars and reorganizes, augments, and delivers the material to cadets in a way that better suits the leadership and development needs of each class of cadets. Cadets participating in this study experienced seminars both prior to and after the changes brought about in 2017.
56
Since 1993, CCLD has also hosted the annual National Character and Leadership
Symposium (NCLS) held at USAFA each Spring. NCLS typically has a variety of speakers, including distinguished scholars, military and corporate leaders, athletes, and others who are experienced in the challenges of leadership. While the focus and themes of NCLS changes every year, the goal remains to provide “. . . an opportunity for all
Academy personnel, visiting university students and faculty, and community members to experience dynamic speakers and take part in group discussions to enhance their own understanding of the importance of sound moral character and good leadership” (The
United States Air Force Academy, 2018b). While attendance of NCLS is obligatory to cadets of all classes, the freedom cadets have to select which speakers to listen to makes the event a popular respite from the typically grinding schedule of coursework.
In addition to the courses and programs offered by CCLD, USAFA includes physically challenging programs like boxing, combatives, and a variety of intramural sports. While cadets can choose which sports they participate in, participation in intramural sports is obligatory for all cadets. USAFA also provides longer-term training opportunities in the summers between semesters and these include in part, jump school, glider school, and Survive Escape Resist and Evade (SERE) school. These programs engage the cadets outside the classroom in real-world “live” scenarios where the principles and concepts of leadership and character development learned in the classroom are put into action.
The collective goal of these programs is stated in the USAFA 2015-2016 course catalog under the Center for Character and Leadership Development,
In sum, character development will be a crucial, all-encompassing part of each cadet’s Academy experience. From the time each cadet enters the
57
Academy until graduation, they can expect to see various character and leadership development programs in every aspect of their life with the ultimate goal of inspiring in them an intense inner drive to put integrity first, place service before self, and strive for excellence in all they do. (The United States Air Force Academy, 2016a)
The goals of these programs focus on character development, leadership skill development, and increasing the professionalism and ability of cadets. Thus, resiliency theory, as an essential aspect of character development, is a key element of the development and training of each cadet and demands attention as it appears within the overall framework of character development at USAFA.
Results of efforts to help cadets. USAFA administers a battery of surveys throughout a cadet’s academic career to better understand the effectiveness of its courses and programs and to further understand the nature and development of a variety of cadet traits and characteristics, including resiliency. Several of these surveys, including the
Grit-S survey, the Global Assessment Tool (GAT) used currently by the military, the
Dispositional Resilience Scale (DRS), and the Support and Resilience Inventory (SRI) self-assessment tool sponsored by the North Carolina Department of Health and Human
Services, measure a variety of behaviors and individual characteristics relating to individual resiliency. While these surveys vary in scale and focus, all utilize a method of using self-report survey questions and statements to measure the presence or tendency to demonstrate resiliency-related characteristics.
Initial results from data obtained from USAFA/A9 Department of Testing and
Assessment revealed that cadets from the class of 2016 and 2017 demonstrated average scores of 3.7 out of 5 for the grit survey, 8.2 out of 10 for the SRI, and 1.4 out of 3.0 for the DRS demonstrating average to above average resiliency scores for each of the
58
surveys. However, no comprehensive analysis has been completed to understand why cadets are resilient, how their resilience is changing at USAFA, or which programs demonstrate an impact on survey scores. While a larger meta-analysis of these, and several other surveys would be informative to review their history and development and to conduct a full comparative analysis, initial results from these surveys demonstrate
USAFA cadets display average to higher than average levels of resilience and possess and display resiliency-related qualities.
Though the results of these surveys contain myriad data describing the characteristics, traits, and behavioral development of cadets, they have yet to result in findings upon which to establish clear fundamental principles to guide the development of resiliency-building programs at USAFA. Additionally, while many of the programs build cadet character and leadership, few of them are based upon the principles of resiliency theory or explicitly address individual resiliency development as key aspects of their core frameworks. While these programs are generally considered effective, one recent study analyzing USAFA freshman cadets’ ability to identify with USAFA’s core virtues after completing the VECTOR seminar training produced only mixed results. The qualitative results of interviews with focus groups produced findings that cadets found
VECTOR training positive and impactful, but quantitative survey analysis could find no statistically significant connection (Tate, 2016). To better understand how cadets develop resiliency, this study analyzed cadet survey data to understand how demographics, performance indicators, attrition rates, and USAFA clubs and programs were associated with cadet resiliency.
59
Emerging Themes and Research Gaps
This review began with a discussion of the elements of resiliency theory; the history, development, and current research; and a brief discussion of some strengths and weaknesses of this theory and of how the concepts of self-efficacy and mindfulness further inform resiliency theory. The literature review discussed the need for leadership in the military and highlighted the necessity of developing resiliency to deal with the challenges to service members and their families as a result of warfighting and lengthy deployments. This set the stage for a discussion of resiliency in higher education and the military, listing examples of each and discussing the needs for greater research in each are to clarify and add to the body of existing research exploring individual resiliency.
Following this a review of the challenges faced by service members in the military was discussed noting the significant efforts made by the United States Army and the United
States Air Force to build the resiliency of their forces. The review concluded with a review of the adversities USAFA cadets face and the substantial efforts taken to develop programs and seminars to bolster cadet resiliency and improve student performance.
Several themes emerged requiring the need for further research. First is the need to add clarity to the discussion over the fundamental benefit of resiliency theory and the closely related subject of grit. A large amount of resiliency-based research exists both in higher education and in the military with conflicting findings on both sides of the debate.
This study will clarify these findings and inform current theories by adding credible research to further uncover the relationship between resiliency and performance using
USAFA cadets as its platform.
60
The next apparent gap is the insufficient number of research studies focused specifically on individual resiliency. Of the 73 studies listed researching resiliency and grit (Credé et al., 2017), population sample sizes varied from 21 to 1,554 individuals, with the average number of research study participants being 351 and the vast majority of the studies containing fewer than 1,000 participants. The largest grit studies conducted by Dr. Duckworth were conducted with population sample sizes of 1,218, 1,308
(Duckworth et al., 2007), and 1,554 (Duckworth & Quinn, 2009) and compared participants’ grit scores with other factors at a singular point in time. This study analyzed data from over 5,400 USAFA cadets over a period of nine years, including several groups that completed the Grit-S survey multiple times, providing both a much larger sample size than previously available together with longitudinal data to identify emerging trends and changes in grit and each of its subscales of passion and resiliency.
Another apparent gap in the literature was the lack of research connecting the principles of resiliency and the resiliency models with programs designed to develop individual resiliency. Most of the literature discusses theoretical concepts and provides research detailing the analysis of surveys completed by individuals completing various programs, but little if any mention is made citing specific programs as more or less beneficial to developing individual resiliency. The primary focus in these studies is the resiliency of the individual, not the resiliency-building effectiveness of the programs.
Necessary to the discussion of resiliency development are examples of programs and activities found to influence individual resiliency along with a discussion of effect sizes and outcomes. This research study fills this gap by associating USAFA cadet Grit-S
61
score with cadet participation in the myriad clubs and programs available to cadets with the intent of identifying the types of programs that may increase resiliency.
In addition to filling these gaps, this study fulfills the need to exploit previously unexamined data collected by USAFA, thereby making use of valuable resources used to collect the data over the past several years. Significant time and effort has been spent to develop the surveys to collect the data, provide time to the cadets to complete the surveys, and to ensure the surveys are kept for future analysis. This study finally makes use of this valuable information with the intent of informing USAFA leadership of resiliency-building trends and emerging themes and providing them with the information and recommendations necessary to guide the future leadership development programs that will shape the character and leadership development of future cadets.
This discussion of the thematic elements either absent or requiring further research presents the focus for this research study and provides a justification and necessary connection to the research questions chosen to address these gaps. The quantitative research designed to provide answers to these questions is presented in the next chapter.
62
CHAPTER III
METHOD
The purpose of this chapter is to review the setting and participants who took part in the study, provide a description of the Grit-S survey used to assess cadet resiliency, and set forth the procedures and methods used to analyze the data. I present each research question along with the associated regression model used to examine the relevant data and then provide the data and analysis used to study the several groups of
USAFA cadets who took the Grit-S survey from 2009 to 2017. I present the regression models used to identify how cadet demographic predictors, performance measures, attrition rates, and USAFA clubs and programs were associated with Grit-S survey scores and resilience subscores. A discussion of the analysis follows, answering the proposed research questions and identifying opportunities and strategies for future research studies.
This quantitative study used multiple regression models to explore the level of resiliency and grit in USAFA cadets and explore the relationship between their individual
Grit-S survey scores and other factors, including demographic information, individual performance factors, attrition, and participation in USAFA clubs and summer training programs. It is important to note that while the focus of this study was on cadet resiliency, the overall Grit-S survey that was offered to cadets by USAFA gives information about resilience, passion, and overall grit. Thus, results will be shown for all three factors, although the subject of resiliency will be emphasized. Researchers have performed individual resilience research using surveys demonstrating both reliable and valid results (Eid et al., 2008; Ramanaiah et al., 1999). This study is an exploratory study, and while Grit-S survey results were regressed on cadet performance variables
63
such as grade point average, this study did not seek to replicate findings from other studies. The primary focus of this study was to understand the relationship, if any, between USAFA Cadet Grit-S scores and cadet development while at USAFA. The specific research questions to explore this relationship were:
1. To what extent are cadet characteristics, including demographics and
participation in USAFA clubs and programs, associated with Grit-S score?
2. To what extent is Grit-S score associated with cadet overall performance
average (OPA)?
3. To what extent is Grit-S score associated with cadet attrition?
4. What is the change in cadet Grit-S score over time?
5. Based on this research study, what are some recommendations for future
research, policy development, and practices to build cadet resiliency at
USAFA?
The intent of the study was to examine trends in Grit-S scores to understand how the associations between Grit-S scores and other predictors changed over time, what influenced those changes, and the extent to which individual factors were associated.
Exploring these questions provided a deeper understanding of how cadet resiliency is built and developed at USAFA and identified possible methods of improving programs designed to develop leaders of character.
Research Setting and Participants
The setting for this study was the Unites States Air Force Academy (USAFA) located in Colorado Springs, Colorado. USAFA accepts a pool of unmarried applicants averaging 20 years of age and consisting of a diverse array of students including 34.1%
64
ethnic minority students (CollegeData, 2016). After cadets arrive at USAFA, they become “basic cadets” until they complete the basic cadet training (BCT) held in the summer prior to their first semester. Upon completion of BCT, cadets receive the rank of fourth-year cadets, equal to first-year freshman students at other universities. Third-year cadets are equal to sophomores, second-year cadets are equal to juniors, and first-year cadets, often called “Firsties,” are equal to seniors at other universities. Cadets from all ranks enter squadrons with leadership positions filled by both the cadet ranks and active- duty Air Force leadership personnel. The smallest unit is the “Element” consisting of 10-
12 cadets. Three elements form a “Flight” of 30-35 cadets, and three flights form a
“Squadron” of approximately 100 cadets. USAFA maintains 40 squadrons organized into four “groups” that comprise the “Cadet Wing” representing the entirety of the cadet population ranging in size from 3,900 to 4,100 cadets.
An active-duty leadership team leads each squadron and consists of an Air Officer
Commanding (AOC), who is a commissioned officer, and an Academy Military Trainer
(AMT), who is a non-commissioned officer. This leadership team serves as the United
States Air Force active-duty professional leadership team tasked with developing, supervising, and mentoring every cadet within their squadron in every aspect of their careers while at USAFA (USAFA, 2016). This organizational design represents the social network, community of friends and professional colleagues, and support structure they must work with throughout the duration of their academic careers at USAFA.
From day one, USAFA cadets understand the core principles and critical outcomes of USAFA as they recite the honor code, “We will not lie, steal or cheat, nor tolerate among us anyone who does”; profess the spirit of the code to “Do the right thing
65
and live honorably” (The United States Air Force Academy, 2016b); and strive to fully espouse the United States Air Force core values of “Integrity First, Service before Self, and Excellence in All We Do” (Center for Character & Leadership Development, 2016).
On a daily basis, cadets review the values of the United States Air Force, receive mentorship in leadership, critical-thinking, and problem solving, and are challenged by the physical demands of fitness and extracurricular sporting events. Thus, not only have individuals selected to attend USAFA demonstrated prior ability and strong moral character, they receive constant mentorship in the principles of integrity, service, excellence, and develop a commitment to challenging standards in an environment filled with extraordinary expectations as they prepare to lead in the United States Air Force.
Measures
Grit-S survey description
The survey tool used by USAFA to measure cadet resiliency is a version of the
Grit survey developed by Dr. Duckworth that contains two subscales to test individual levels of both consistency of interest or passion and perseverance of effort or resiliency which combined are referred to as grit (Duckworth et al., 2007). In 2009, USAFA began using Dr. Duckworth’s original seventeen-item Grit survey to understand individual grit as a combination of consistency of interest, perseverance of effort, and ambition. This survey included six questions to measure passion, six questions to measure resilience, and five questions to measure ambition (Duckworth et al., 2007). Beginning in 2016,
USAFA began using the Grit-S survey (Appendix C), a shorter eight-item grit survey developed by Dr. Duckworth and Dr. Quinn that they found to be a more accurate and reliable version of the survey consisting of eight of the original seventeen questions and
66
producing more reliable and valid results (Duckworth & Quinn, 2009). This shorter survey, called the Grit-S survey, includes four questions measuring the passion subscale and four questions measuring the resiliency subscale. A five-point Likert scale is used to answer each question, and responses consist of “very much like me,” “mostly like me,”
“somewhat like me,” “not much like me,” and “not like me at all.” Questions 1, 3, 5, and
6 of the Grit-S survey are reverse scored where the answer “very much like me” receives one point and questions 2, 4, 7, and 8 are normally scored with the answer “very much like me” receiving five points. Overall Grit-S scores are calculated by adding up the points to all the responses and dividing by the total number of questions with a 5.0 being the top possible score. The passion and resilience subscores are calculated by adding up the points from their respective four questions and dividing each by four with a 5.0 also being the top possible score.
Researchers have used the Grit-S survey and its subscales to understand the grit of cadets at West Point, students participating in a national spelling bee, and college undergraduate students. They reported Cronbach’s Alpha values ranging from .73 to .83, indicating a strong level of internal consistency in the survey (Duckworth & Quinn,
2009). While not replicating their study, in a similar fashion I analyzed the results from
Grit-S surveys taken by USAFA cadets from 2009 to 2018 and regressed the Grit-S survey scores and subscale scores on a large number of cadet-level variables.
Reliability and validity
In prior research studies, Dr. Duckworth reported psychometric properties of the
Grit-S survey with an analysis of six separate studies that presented “evidence for the
Grit-S’s internal consistency, test-retest stability, consensual validity with informant-
67
report versions, and predictive validity” (Duckworth & Quinn, 2009, p. 166). Of those six, study one analyzed Grit-S survey results from two samples of cadets from the United
States Military Academy at West Point (USMA). This study found Cronbach’s Alpha values measuring internal consistencies of both the overall Grit-S survey ranging from
.73 to .83 to the sub elements of Persistence of Effort and Consistency of Interest ranging from .60 to .78 and from .73 to .79 respectively (Duckworth & Quinn, 2009, p. 167).
Study two analyzed Grit-S survey results completed by adults over age 25 who reported changing careers and found adequate internal consistencies in Perseverance of Effort
(.70), Consistency of Interest (.77), and Grit-S (.82) (Duckworth & Quinn, 2009, p. 168).
Study three analyzed survey responses from adults over age 25 along with surveys completed by informants found by survey participants. The informants grouped into family members, peers, and self- and internal consistencies were found to be .84, .83, and
.83 respectively. Correlations between the self-reports and family member or peer informants were medium to large: r = .45, p < .001 and r = .47, p = < .001. Study four analyzed survey results from high-achieving middle and high school students who took the survey one year apart with the focus of determining test-retest stability. Internal consistency measured in 2006 and 2007 was strong at .82 and .84 respectively.
Correlation between the two survey periods was good with r = .68, p < .001. Study five analyzed survey results from USMA cadets participating in the summer training program and found the Grit-S survey demonstrated an internal consistency of .77 with a higher predictive value of training completion than the Whole Candidate Score typically used by
USMA (Duckworth & Quinn, 2009).
68
This study of USAFA cadets measured the internal consistency of the Grit-S survey collected over a nine-year span, during ten different survey periods, with a large combined sample population (N = 5,454). Over this period, the internal consistency of the Grit-S survey was measured at 0.733 with the passion and resilience subscores measuring at 0.792 and 0.692 respectively. This demonstrates an acceptable level of internal consistency in all three scales and is very similar to the values presented in Dr.
Duckworth’s research studies.
Research Design
This study includes a quantitative analysis of secondary data previously collected by USAFA during multiple scheduled survey sessions. The historical data was collected from the USAFA-A9 and Cadet Administrative Management Information System
(CAMIS) departments and then cleaned, organized, and prepared for upload into SPSS for regression analysis. I performed several iterations of regressions to analyze the associations between the dependent and independent variables and the results were compared to reveal the best fitting regression models. Details of the analysis for each research question are provided in separate sections later in this chapter. The regression results for each of the research questions is presented in chapter four.
Procedures
Data collection. USAFA administers surveys at various times during the year, including during the biannual designated survey assessment time (DSAT), at basic cadet training (BCT), and by professors conducting research in different classes. Participation in the surveys offered during the DSAT and provided by professors is always considered voluntary, and despite the elevated expectations in both military environments and
69
especially inherent to USAFA, participation can range from low to quite high. Each year group at USAFA begins with slightly more than 1,100 cadets and frequently graduates between 800-900 cadets, thus a quick calculation using the figures in table one shows that participation in this survey ranged between approximately 4.3% and 100%. Participation in the survey prior to BCT is mandatory, hence the 100% participation, and that will be discussed in the limitations section.
Cadets complete the surveys by logging into a secure online system, agreeing to participate in the survey, reviewing the survey items, and clicking on the answers that best match their feelings and opinions. The survey data for this study consisted of survey data collected from cadets from 2009 to 2018 and was not collected by the researcher.
After gaining USAFA IRB approval, I collected the Grit survey data from the USAFA-
A9 survey and analysis department and discovered that 8,331 Grit-S surveys had been competed by USAFA cadets from the Fall semester 2009 to the Fall semester 2018.
Many of these cadets took the surveys multiple times, providing for the analysis of changes in Grit-S scores over time. Table 1 below shows the breakdown of all surveys completed by cadets during their respective survey periods.
70
187 256 582 765 222 469 925 154 574 1684 1482 1031 8331 Surveys Available Available 41 85 Fall Fall 2018 1142 304 (J)* 307 (Sr)* 405 (So)* 0 111 2018 1142 Summer Summer 1031 (B) 2 39 818 2018 Spring Spring urveys 253 (J)* 237 (Sr)* 287 (So)* 0 90 2017 1167 Summer Summer 1077 (B)* 2 12 363 2017 Spring Spring 78 (Sr)* 103 (J)* 168 (So)* 45 77 Fall Fall 435 2016 77 (J)* Grit-S SurveyGrit-S Groups 101 (Sr)* 135 (So)* 0 49 2016 1142 Summer Summer 1093 (B)* 2 23 Fall Fall 450 2015 43 (J)* 62 (F)* 52 (So)* 268 (Sr)* 7 40 2014 1126 Spring Spring 582 (J) 497 (So)* 22 81 Fall Fall 546 2009 187 (J) 256 (F) 2011 2013 2015 2016 2017 2018 2019 2020 2021 2022 Unknown Student Class Student Surveys Total Incomplete data Incomplete Table 1 Table From USAFA-A9 Collected Data Grit-S Survey USAFA of Description s matched groups *denotes containing Sr=Senior; So=Sophomore, F=Freshman, J=Junior, Cadets, B=Basic Note:
71
The USAFA CAMIS department was able to find and provide demographic, performance, attrition, and other cadet data for 5,454 of the cadets, but it found that the remaining cases had blank or erroneous identification information, making it impossible to connect cadet information in CAMIS with the results of the surveys. CAMIS personnel indicated this periodically occurs during survey periods as some cadets, apprehensive about providing their identification information, enter erroneous information to prevent any possible connection to their answers. The data was further inspected for missing and incomplete data, and cases that could not be matched with cadet data, contained erroneous data, or that were missing most, or all of the grit survey values were deleted. This reduced the number of useable surveys by 16.29%, resulting in
6,974 surveys available for analysis. The final result was a large sample population of cadets (N = 5,454) who completed a large number of Grit-S surveys (N = 6,974) from
Fall 2009 to Fall 2018.
After receiving the raw data from USAFA-A9 and the CAMIS department, it was necessary to process the data and prepare it for further regression analysis. Since data regarding cadet Grit-S survey scores, demographics, performance measures, attrition, and participation in USAFA clubs and programs came from more than a single source, the data first had to be organized and arranged so the data for all cases was displayed in a single row. Once the entire dataset was properly organized, a final dataset consisting of all cases (N = 5,454) and surveys (N = 6,974) was compiled in preparation for the analysis of missing data. The multiple-survey cases consisting of those who took the survey twice, three times, and four times (n = 1,335) were used to conduct the trend analysis in research question four.
72
Variables. This exploratory research study used a total of 44 variables throughout the entire analysis, and each research question used a different combination of the dependent and independent variables depending on the requirements of the research questions. Table 2 below shows all the variables used in the study, their regression code, measure, and coding used. A review of each of the variables follows the table and provides more details as to the reason the variables were included in the study.
Table 2
Dependent And Independent Variables Used In Regression Analysis
Variables Code Measure Coding Gender GEN dichotomous 1= female, 0 = male Black BLK dichotomous 1 = yes, 0 = no Caucasian (control) CAUC dichotomous 1 = yes, 0 = no Hispanic HISP dichotomous 1 = yes, 0 = no American Indian AMIN dichotomous 1 = yes, 0 = no Asian ASIA dichotomous 1 = yes, 0 = no Native Hawaiian/Pacific NHPI dichotomous 1 = yes, 0 = no Islander Unknown UNK dichotomous 1 = yes, 0 = no Family education FEDU ordinal 1=High School, 2=Some college, 3=Associates, 4=Bachelors, 5=Graduate, 6=unknown Family income FINC ordinal 1= don't know, 2= <25K, 3= 25K-74,999K, 4= 75K- 124,999K, 5= 125K-174,999K, 6= >175K Family members FMEM ordinal 1=1, 2=2, 3=3, 4=4, 5=5, 6=6, 7=7, 8=8, 9=9, 10=10 or more From single parent FSGP dichotomous 1 = yes, 0 = no First generation college FGEN dichotomous 1 = yes, 0 = no HS GPA HS_GPA% continuous scale HS Athletics HSAH dichotomous 1 = yes, 0 = no Recruited college athlete RCAT dichotomous 1 = yes, 0 = no Honors List HON dichotomous 1 = yes, 0 = no Grade Point Average GPA continuous scale Military Performance Average MPA continuous scale Physical Education Average PEA continuous scale Overall Performance Average OPA continuous scale Grit score GRIT continuous scale Passion subscore PASS continuous scale
73
Dependent And Independent Variables Used In Regression Analysis
Variables Code Measure Coding Resilience subscore RES continuous scale Grit survey gain score GTGN continuous scale Passion gain score PNGN continuous scale Resiliency gain score RYGN continuous scale Not disenrolled (control) NDSN dichotomous 1 = yes, 0 = no Disenrolled voluntarily VOL dichotomous 1 = yes, 0 = no Disenrolled involuntarily INVOL dichotomous 1 = yes, 0 = no Probation PROB dichotomous 1 = yes, 0 = no Probation-Academic PRAC dichotomous 1 = yes, 0 = no Probation-Aptitude PRAP dichotomous 1 = yes, 0 = no Probation-Athletic PRAT dichotomous 1 = yes, 0 = no Probation-Conduct PRCO dichotomous 1 = yes, 0 = no Probation-Honor PRHO dichotomous 1 = yes, 0 = no Intercollegiate sports ICAT dichotomous 1 = yes, 0 = no Intramural sports INTR dichotomous 1 = yes, 0 = no Competitive COMP dichotomous 1 = yes, 0 = no Mission Support MSPT dichotomous 1 = yes, 0 = no Professional PROF dichotomous 1 = yes, 0 = no Club CLUB dichotomous 1 = yes, 0 = no Recreational REC dichotomous 1 = yes, 0 = no Civil Air Patrol, Boy/Girl SRV dichotomous 1 = yes, 0 = no Scouts, Campfire Girls
The gender and race variables are self-explanatory and are designed to place cadets into gender and race groups for comparative purposes. Most questionnaires and surveys completed by USAFA cadets ask for this information, so they were included to be in agreement with common surveying practices.
The family variables consisting of family education, family income, family members, from single parent home, and first-generation college student are exploratory variables designed to identify factors present in the cadet’s lives prior to attending
USAFA. These variables were included to understand how diverse background characteristics such as socioeconomic status (SES) and other factors related to upbringing and family dynamics impact the growth and development of cadets. These variables
74
most closely align with the protective factors within the resiliency theory framework as they point to the family support aspects that influence cadets during their upbringing.
The intent of these variables is to understand to what degree diverse background characteristics influence the associations that may exist between these characteristics and the grit, passion, and resilience levels cadets demonstrate while attending USAFA.
The variables high school GPA percentage of max, high school athletics, and recruited college athlete are exploratory variables designed to identify factors related to the cadet’s individually diverse life experiences prior to attending USAFA. The GPA percentage of max was used instead of straight GPA because USAFA accepts a wide variety of candidates both international and domestic, some of whom complete school with GPAs measured on 3.0, 4.0, 5.0, and even 6.0 scales. To compare all these GPAs evenly, a percentage of max was calculated by dividing each GPA by its respective scale to obtain the GPA percentage each cadet had out of 100%. This made it possible to compare all the GPAs from all the cases evenly and set the groundwork to impute missing variables where applicable. These variables most closely align with the promotive factors within the resiliency theory framework as they point to the individual characteristic’s cadets have obtained and developed throughout their lives. The intent is to identify and compare the cadet’s individual characteristics developed throughout their lives with the grit and resilience levels demonstrated while at USAFA.
The performance variables consisting of honors list, grade point average (GPA), military performance average (MPA), physical education average (PEA), and overall performance average (OPA) are variables designed to understand the relationship between Grit-S score and cadet performance measures. The honors list variable
75
represents cadets who at some point during their career at USAFA were included in one of several different honors lists including the Dean’s list, the Commandant’s list, the
Director of Athletics list, and the Superintendent’s list. The minimum standards require an average GPA of 3.0 for the Dean’s list, an average MPA of 3.0 for the Commandant’s list, an average PEA of 3.0 for the Director of Athletics list, and all three standards must be met for inclusion on the Superintendent’s list. The OPA variable is comprised of the
GPA, MPA, and PEA variables, so to avoid multi-collinearity issues, two different regression models were used to understand the associations between OPA and the independent variables, and then again with each of the GPA, MPA, and PEA variables.
The Grit-S score, passion subscore, and resilience subscore served as the primary dependent variables for research question one and four against which all the independent variables were compared in order to identify any predictive associations. These same variables were included as independent variables in questions two and three. A total of
5,454 cases were included when comparing Grit-S score and its subscores with the other independent variables. For the cadets who completed multiple surveys, only the final survey was included in this analysis. As was mentioned previously, Dr. Duckworth developed and used the Grit-S survey to measure and understand grit levels in a variety of populations, including USMA cadets, elementary school students, and adults
(Duckworth et al., 2007). Because this study had a particular focus on resilience, all the results from the Grit-S surveys completed by cadets were included for regression and comparison. However, in order to avoid multi-collinearity issues, the Grit-S score variable was used as an independent variable in one regression model, and the passion
76
and resilience subscores were used similarly in another variant, and both models were compared for associations.
The Grit-S, passion, and resilience gain scores variables were derived from analyzing the cases where more than one survey was completed. Cadets who took the survey twice (n = 1,157) produced 2,314 surveys, those who took it three times (n = 171) produced 513 surveys, and those who took it four times (n = 7) produced 28 surveys, resulting in an overall multiple-survey group of 1,335 cases who produced 2,855 surveys.
These cadets completed the surveys at various times throughout their careers, at different time intervals between surveys, and were from different graduating classes, so a standardized gain score based on a common time interval between first and last survey was calculated in order to conduct a reasonably even comparison. The differences were standardized by assessing a survey time interval that captured the greatest number of cases and yet did not cover too long or too short a span of time. Analyzing the responses revealed that 73.86% of the cases had survey time intervals of between 16 and 28 months, a difference of only 12 months. Removing the cases outside the interval window
(n = 198) reduced the total number of cases by 14.8% leaving a final group of 1,137 cases for comparison. The gain score was calculated by subtracting the first Grit-S survey score from the last survey score. The same procedure was completed for the passion gain scores and resilience gain scores, and these scores were regressed on the list of independent variables to understand the relative associations. The gain scores identified any longitudinal trends to understand how cadet behavior measured by the Grit-S survey changed as they moved through their educational careers at USAFA. For gain scores, multiple regression analysis compared the gain scores to the list of independent variables,
77
and an additional graph was included showing the growth trend of cadet Grit-S score, passion subscore, and resilience subscore.
The exploratory attrition outcome variables of not disenrolled, voluntarily disenrolled, and involuntarily disenrolled were included to identify any association between Grit-S scores and its subscores and cadets who were voluntarily or involuntarily disenrolled from USAFA. In research questions one, two, and four, these variables were used as independent variables, but in research question three, each of these three variables was used as the dependent variable. “Not disenrolled” refers to those cadets who were not disenrolled from USAFA or those who remained enrolled through graduation, and SPSS automatically used this variable as a control for the other two attrition variables in research questions one, two, and four. Disenrollment from USAFA can either be categorized as voluntary or involuntary depending on the circumstances surrounding the disenrollment. Voluntary disenrollment occurs when cadets who have not yet received their AFSC and fully committed to the Air Force decide on their own to leave USAFA. Reasons for voluntary disenrollment include a change in career goals, medical or administrative turnback, pregnancy, medical injuries, international students who only attend for a few semesters, cadets who left to serve religious missions, psychological challenges like depression or mental distress, minor drug and alcohol infractions, or death of a family member. Involuntary disenrollment involves the mandatory removal of a cadet from USAFA for reasons such as academic deficiencies, forced medical retirement, misconduct, failing to recover from any of the six types of probation (academic, aptitude, athletic, conduct, honor, honor rehab), driving under the
78
influence (DUI), major drug or alcohol violations, cheating, assaulting other cadets, and violations classified under the title of honor code violation.
The exploratory, probation-related variables in this study were included to identify associations between the six types of probations at USAFA and the grit scores and subscores. The variable “probation—all types” is comprised of all probations that can occur at USAFA, namely academic probation, aptitude probation, athletic probation, conduct probation, honor probation, and honor rehab probation. In the 5,454 sample of cadets who completed the surveys, only four cadets had ever been placed on honor rehab probation and regression models excluded it from analysis. Similar to the performance variables and grit variables, regression models were completed using both the singular
“probation—all types” variable and then again with the five probation type variables in order to avoid issues with multi-collinearity.
The exploratory USAFA program variables compared seven categories of activities cadets may participate in white attending USAFA. The goal of including these categories was to identify any associations between Grit-S scores or subscores and the various categories of programs and activities. The seven categories are intercollegiate sports (which include both varsity and junior varsity level sports), intramural sports, competitive programs, mission support programs, professional programs, clubs, and recreational programs and activities. The list of specific activities in each category is long and almost all cadets participate in more than one category during their careers at
USAFA. A full list of specific activities in each category is provided in Appendices C through I, but a brief explanation of each category is instructive.
79
Intercollegiate sports (Appendix C) includes all sports, male and female, in which cadets compete with other Universities. Participation in intercollegiate sports is time consuming with daily practices, competitions on weekends, and extended periods of travel to compete with institutions all over the United States. These sports include football, basketball, swimming, water polo, fencing, tennis, and several others.
Intramural sports (Appendix D) include sports cadets participate in seasonally at the school, competing against other USAFA cadets, and is by far the most inclusive of any category since the majority of cadets are required to participate. The time requirement for these sports is much less than intercollegiate since there are fewer practices and no travel required. These sports include flag football, ultimate frisbee, boxing, rugby, racquetball, team handball, futsal (a form of indoor soccer), and several others.
The remaining categories include activities and programs cadets participate in on their own time or in connection with the classes they may be taking. Competitive activities (Appendix E) include both athletic and non-athletic activities like alpine skiing, performing arts, forensics, cycling, rodeo, parachute/skydiving, triathlon club, coed volleyball, and several others. Mission support programs (Appendix F) include choir, drum & bugle corp., orchestra, combat pistol, cyber warfare, the honor guard, the
Sandhurst Competition Team (which competes annually), the Arnold Air Society (which is an Air Force service organization), and several others. Professional programs
(Appendix G) include astrophysics, future business leaders, history, language club, chemistry, Chinese, tutoring, robotics, forensics, STEM (science, technology, engineering, mathematics), and several others. Club activities (Appendix H) include the
Hispanic/Latino club, Secular Cadet Alliance, Native American Heritage, car club,
80
Karate, the international club, steel script, and a few others. Recreational activities
(Appendix I) include Aikido, karate, archery, equestrian, model building, paintball, mountaineering, trap & skeet, wargaming, some performing arts, and several others.
Finally, the exploratory variable comprised of individuals who had participated in the Civil Air Patrol, Boy and/or Girl Scouts, or Campfire Girls prior to attending USAFA was included to see if the characteristic of service or being active in service organizations was associated with Grit-S score and its subscores. I made the decision to include this variable because from my own experiences in Boy Scouts for over 15 years both as a youth and an adult, I noticed an increase in gritty and resilient behavior from individuals who participated in these types of organizations. This variable is most closely aligned with the promotive factors within the resiliency theory framework as they point to other characteristics cadets may have obtained and developed throughout their lives. The
CAMIS department at USAFA routinely collects this information from applicants, making it possible to include it in regression analysis.
While the total number of variables included in this research study was particularly large, the large sample size (N = 5,454) made it possible to utilize all 41 independent variables while maintaining adequate statistical power. Taking into account, the general guideline of including at a minimum 10 cases per variable, this study far surpassed the necessary 410 cadets, enabling a unique opportunity to explore a wide variety of variables and identify any factors that previously had gone unresearched.
Missing values. While compiling and organizing the data, it became apparent that there were missing values in a few of the variables. An analysis of missing values revealed that 10 of the 41 (24.39%) of the variables were missing at least one value,
81
5,050 of the 5,454 (92.59%) of the cases were missing at least one value, and that 8,214
(3.673%) of all values were missing. Missing data occurred in cadet information because
CAMIS did not record all the information for all the cadets, but a review of the missing value patterns found that no patterns were evident. Conducting Little’s MCAR test of the variables in SPSS was statistically significant at p < .000, and multiple imputation was deemed appropriate. Due to the large sample size, the decision was made to impute values for all 10 variables missing data using 10 imputations for each variable. The imputation data for each of the variables is shown in Table 3.
Table 3
Statistics of Variables with Imputed Values
Missing Percent Total Imputed Variables Data Points Missing Values First generation college study (FGEN) 3,671 67.30% 36,710 Family member education level (FEDU) 2,225 40.80% 22,250 From a single parent home (FSGP) 1,195 21.90% 11,950 Overall performance average (OPA) 295 5.40% 2,950 Family income (FINC) 204 3.70% 2,040 Number of family members (FMEM) 203 3.70% 2,030 Physical education average (PEA) 174 3.20% 1,740 Military performance average (MPA) 110 2.00% 1,100 Grade point average (GPA) 105 1.90% 1,050 High school grade point average percentage (HS_GPA%) 32 0.60% 320 Note: N = 5,454 after imputing missing data
After reviewing the number and percentages for the missing data values, regressions were completed and the pooled regression results were compared using first the dataset without any variable imputations, a second dataset with the seven imputed variables missing less than 5.40% of the data, and a third dataset imputing values for all ten variables. An analysis of the regressions models for each of the research questions
82
using each of these datasets found the best fitting model using the pooled results from the dataset imputing missing data to all ten variables, and this was the model that was selected for the primary analysis in this study. While the first-generation college student variable had a high percentage of missing values (67.3%), the variable was included because doing so resulted in a better fitting model. However, it is important to take this into account when interpreting both the statistical significance and impact of this variable. While the results from the other datasets were reviewed and analyzed for comparison purposes with the other datasets, only the results from the dataset with ten imputed variables will be presented in this study.
Data analysis
Analysis of the combined Grit-S survey data and cadet information data was conducted using the Statistical Package for the Social Sciences (SPSS) version 25, release 25.0.0.1, 64-bit edition (IBM Corp., 2017). Analysis of the five research questions required using different regression models tailored specifically to each question. Each of the regression models is presented separately in relation to the research questions each answered.
Research question 1 (RQ1) – Grit-S score vs Cadet Predictors. The first research question asked which cadet predictors are associated with Grit-S score. To answer this question, I created a linear regression model to regress the continuous dependent variable of USAFA cadet Grit-S survey score, where Y is the grit score of the i-th cadet, on a set of 36 independent variables consisting of the variables shown on Table
4 at the end of this section. The regression model was:
Yi(grit) = β0 + Sβ1-36(X1-36)…+ ei
83
Two similar regression models also compared the relationship between the passion and resiliency subscales and the same list of independent variables. Additionally, a second variant of each of these three regression models was created using the OPA variable in place of the GPA, MPA, and PEA variables, since the three are simply added together to create OPA. Likewise, the “probation—all types” variable was used in place of the five different probation variation variables since combining these together creates the PROB variable. In both cases, this was done to avoid multi-collinearity between variables. Finally, each of these six variations were run without any imputed variables, with seven imputed variables, and with 10 imputed variables, resulting in 18 separate regressions computed and compared to identify model fit and significance. As stated previously, the models using imputed data for all 10 variables demonstrated the best fitting models and were used throughout this study.
Categorical dummy codes were created for non-continuous independent variables, and all independent variables were grand-mean centered. Dummy codes were created for specified demographic variables when more than two categories existed. For binary variables, participation or membership to a group was coded as 1, and non-participation or non-membership was coded as 0. Overall regression results for all RQ1 models are presented in chapter four.
84
Table 4
RQ1 Variables Used in Linear Regression Analysis
Dependent variables Independent variables Abbrev. Codes Grit-S survey score Gender GEN 1 = female, 0 = male (GRIT) Passion subscore Black BLK 1 = yes, 0 = no (PASS) Resilience subscore Caucasian CAUC 1 = yes, 0 = no (RES)
Hispanic HISP 1 = yes, 0 = no American Indian AMIN 1 = yes, 0 = no Asian ASN 1 = yes, 0 = no Native Hawaiian / Pacific NHPI 1 = yes, 0 = no Islander
Unknown UNK 1 = yes, 0 = no Family education FED 1=High School, 2=Some college, 3=Associates, 4=Bachelors, 5=Graduate, 6=unknown Family income FIN 1= don't know, 2= <25K, 3= 25K-74,999K, 4= 75K- 124,999K, 5= 125K-174,999K, 6= >175K Family members FMEM 1=1, 2=2, 3=3, 4=4, 5=5, 6=6, 7=7, 8=8, 9=9, 10=10 or more
From single parent FSP 1 = yes, 0 = no First generation college FGC 1 = yes, 0 = no High school GPA HS_GPA% scale percentage of max High school athletics HAS 1 = yes, 0 = no Recruited college athlete RCA 1 = yes, 0 = no Honors list HON 1 = yes, 0 = no Grade point average GPA scale Military performance MPA scale average Physical education PEA scale average Not disenrolled NDSN 1 = yes, 0 = no Disenrolled voluntarily VOL 1 = yes, 0 = no Disenrolled involuntarily INVOL 1 = yes, 0 = no Academic probation PRAC 1 = yes, 0 = no Aptitude probation PRAP 1 = yes, 0 = no Athletic probation PRAT 1 = yes, 0 = no Conduct probation PRCO 1 = yes, 0 = no Honor probation PRHO 1 = yes, 0 = no Intercollegiate sports ICSP 1 = yes, 0 = no
85
RQ1 Variables Used in Linear Regression Analysis
Dependent variables Independent variables Abbrev. Codes Intramural sports IMSP 1 = yes, 0 = no Competitive programs COM 1 = yes, 0 = no Mission support programs MSPT 1 = yes, 0 = no Professional programs FESS 1 = yes, 0 = no Club programs CLU 1 = yes, 0 = no Recreational programs REC 1 = yes, 0 = no Civil Air Patrol, Boy/Girl SRV 1 = yes, 0 = no Scouts, Campfire Girls
This regression model explored the relationship between Grit-S scores, passion, and resilience subscores, and the list of independent variables, revealing both positive and negative associations with performance scores and the magnitude and statistical significance of each association.
Research question 2 (RQ2) – Performance vs Grit-S Score. The second research question asked to what extent Grit-S score is associated with the cadet overall performance average (OPA). To answer this question, I created a linear regression model to regress the continuous dependent variables of OPA and the subfactors of grade point average (GPA), military performance average (MPA), and physical education average
(PEA), where Y is the score of the i-th cadet, on a set of 35 independent variables consisting of the variables shown on Table 5 at the end of this section. It is important to note that since GPA + MPA + PEA = OPA and since USAFA assigns a different weight to GPA (50%), MPA (40%), and PEA (10%) (The United States Air Force Academy,
2017b), I created unique regression models for each of these factors as the dependent variable. After performing the regressions, I used the Bonferroni correction (Shi, Pavey,
& Carter, 2012) to include alpha levels .05, .025, and .0125 to adjust for multiple comparisons. The resulting regression model for the dependent variables was as follows:
86
Yi(opa) = β0 + Sβ1-35(X1-35)…+ ei
Three similar regression models also compared the relationship between the GPA,
MPA, and PEA subscores with the same list of independent variables. Additionally, a second variant of each of these three regression models was created using the GRIT variable in place of the passion and resilience subscales variables, since both together comprise the overall Grit-S score. Similarly to the previous research question, the
“probation—all types” variable was used in place of the five different probation variation variables since combining these together creates the PROB variable. In both cases, this was done to avoid multi-collinearity between variables. Finally, each of these eight variations were run without any imputed variables, with seven imputed variables, and with 10 imputed variables, resulting in 24 separate regressions computed and compared to identify model fit and significance. As stated previously, the models using imputed data for all 10 variables demonstrated the best fitting models and were used throughout this study.
Categorical dummy codes were created for non-continuous independent variables, and all independent variables were grand-mean centered. Dummy codes were created for specified demographic variables if more than two categories existed. For binary variables, participation or membership to a group was coded as 1, non-participation or non-membership was coded as 0, and the overall regression results for all RQ2 models are presented in chapter four.
87
Table 5
RQ2 Variables Used in Linear Regression Analysis
Dependent variables Independent variables Abbrev. Codes Overall performance Gender GEN 1 = female, 0 = male average (OPA) Grade point average Black BLK 1 = yes, 0 = no (GPA) Military performance Caucasian CAUC 1 = yes, 0 = no average (MPA) Physical education Hispanic HISP 1 = yes, 0 = no average (PEA)
American Indian AMIN 1 = yes, 0 = no Asian ASN 1 = yes, 0 = no Native Hawaiian / NHPI 1 = yes, 0 = no Pacific Islander
Unknown UNK 1 = yes, 0 = no Family education FED 1=High School, 2=Some college, 3=Associates, 4=Bachelors, 5=Graduate, 6=unknown Family income FIN 1= don't know, 2= <25K, 3= 25K-74,999K, 4= 75K- 124,999K, 5= 125K-174,999K, 6= >175K
Family members FMEM 1=1, 2=2, 3=3, 4=4, 5=5, 6=6, 7=7, 8=8, 9=9, 10=10 or more
From single parent FSP 1 = yes, 0 = no First generation college FGC 1 = yes, 0 = no High school GPA HS_GP% scale percentage of max High school athletics HAS 1 = yes, 0 = no Recruited college RCA 1 = yes, 0 = no athlete Honors list HON 1 = yes, 0 = no Passion subscore PASS scale Resilience subscore RES scale Not disenrolled NDSN 1 = yes, 0 = no Disenrolled voluntarily VOL 1 = yes, 0 = no Disenrolled INVOL 1 = yes, 0 = no involuntarily Academic probation PRAC 1 = yes, 0 = no Aptitude probation PRAP 1 = yes, 0 = no Athletic probation PRAT 1 = yes, 0 = no Conduct probation PRCO 1 = yes, 0 = no Honor probation PRHO 1 = yes, 0 = no Intercollegiate sports ICSP 1 = yes, 0 = no
88
RQ2 Variables Used in Linear Regression Analysis
Dependent variables Independent variables Abbrev. Codes Intramural sports IMSP 1 = yes, 0 = no Competitive programs COM 1 = yes, 0 = no Mission support MSPT 1 = yes, 0 = no programs Professional programs FESS 1 = yes, 0 = no Club programs CLU 1 = yes, 0 = no Recreational programs REC 1 = yes, 0 = no Civil Air Patrol, SRV 1 = yes, 0 = no Boy/Girl Scouts, Campfire Girls
This regression model explored the relationship between the performance measures OPA, GPA, MPA, and PEA and the list of independent variables, revealing both positive and negative associations with performance scores and the magnitude and statistical significance of each association.
Research question 3 (RQ3) – Attrition vs Grit-S Score. The third research question asked to what extent Grit-S score is associated with cadet attrition. This question was analyzed using one regression model with two different variants to understand the relationship between Grit-S score for those who were not disenrolled and for those who were disenrolled from USAFA either voluntarily or involuntarily. The first regression model regressed the binary dependent variable of attrition for those cadets who did not leave USAFA prior to graduation, where disenrollees are coded as 0 and non- disenrollees are coded as 1, and where i is the log odds of non-attrition of 퐿푛 the i-th cadet, on a set of 30 independent variables shown in Table 6 at the end of this section. This regression model for non-disenrollees versus independent variables is shown below: