Citation
Positive psychology in education : results from the Denver Public Schools whole child student survey

Material Information

Title:
Positive psychology in education : results from the Denver Public Schools whole child student survey
Creator:
Duvall, Carl Garp
Place of Publication:
Denver, CO
Publisher:
University of Colorado Denver
Publication Date:
Language:
English

Thesis/Dissertation Information

Degree:
Doctorate ( Doctor of psychology)
Degree Grantor:
University of Colorado Denver
Degree Divisions:
School of Education and Human Development, CU Denver
Degree Disciplines:
School psychology
Committee Chair:
Crapeau-Hobson, Franci
Committee Members:
Harris, Bryn
Geishler, Lisa

Notes

Abstract:
Previous attempts to promote student mental well-being in schools using positive psychological education models, such as whole child education, have demonstrated academic and behavioral benefits in private school settings (Norrish, Williams, O’Connor, & Robinson, 2013; Roffey, 2015). However, whether these benefits extend to large public school districts has remained unclear to this point. The current study sought to explore the relationship between indicators of whole child education, including student perceptions of feeling challenged, engaged, healthy, safe, supported, and socially-emotionally intelligent, and detrimental behaviors, such as bullying, chronic absences, and out-of-school suspensions in a large, urban public school district. Specifically, it was hypothesized that the indicators of whole child education would negatively associate with behavior outcomes. It was also hypothesized that this trend would continue when specifically examining students attending Title I schools. Data was collected from the Denver Public Schools’ ‘Whole Child Student Survey’ conducted near the end of the 2015-2016 school year. Results from the correlational analysis provided preliminary support for both hypotheses, indicating that when student perceptions of the whole child metrics are higher, they are less likely to experience negative behavioral outcomes. However, substantially fewer and weaker associations we nre shown for students attending Title I schools than those in non-Title I schools. Upon further investigation it was found that Title I school students were significantly less likely to endorse whole child indicators and significantly more likely to experience detrimental behaviors. Findings suggest that monitoring tools, such as the Whole Child Student Survey, are an effective first step in scaling up positive psychological initiatives to large school districts. Future studies may benefit from experimental or qualitative research designs to investigate additional risk and protective factors of student mental wellbeing.

Record Information

Source Institution:
University of Colorado Denver
Holding Location:
Auraria Library
Rights Management:
Copyright Carl Garp Duvall. Permission granted to University of Colorado Denver to digitize and display this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.

Downloads

This item is only available as the following downloads:


Full Text

PAGE 1

POSITIVE P S YCHOLOGY IN EDUCATION: RESULTS FROM THE DENVER PUBLIC SCHOOLS WHOLE CHILD STUDENT SURVEY by CARL GARP DUVALL B.S., Whitman College, 2010 M.S., Lund University, 2013 A thesis submitted to the Faculty of the Graduate School of the Univers ity of Colorado in partial fulfillment of the requirements for the degree of Doctor of Psychology School Psychology Program 2018

PAGE 2

ii This thesis for the Doctor of Psychology degree by Carl Garp Duvall Has been approved for the School Psychology Program by Franci Crepeau Hobson, Chair Bryn Harris Lisa Geisler Date: May 12 th , 2018

PAGE 3

iii Garp Duvall, Carl (PsyD, School Psychology Program) Positive Psychology in Education: Results from the Denver Public Schools Whole Child Student Survey T hesis directed by Associate Professor Franci Crepeau Hobson ABSTRACT Previous attempts to promote student mental well being in schools using positive psychological education models, such as whole child education, have demonstrated academic and behavioral b 2013; Roffey, 2015). However, whether these benefits extend to large public school districts has remained unclear to this point. The current study sought to explore the relationsh ip between indicators of whole child education, including student perceptions of feeling challenged, engaged, healthy, safe, supported, and socially emotion ally intelligent, and detrimental behaviors, such as bullying, chronic absences, and out of school s uspensions in a large, urban public school district. Specifically, it was hypothesized that the indicators of whole child education would negatively associate with behavior outcomes. It was also hypothesized that this trend would continue when specifically examining students attending Title I schools. Data was of the 2015 2016 school year. Results from the correlational analysis provided preliminary support for bot h hypotheses, indicating that when student perceptions of the whole child metrics are higher, they are less likely to experience negative behavioral outcomes. However, substantially fewer and weaker associations we n re shown for students attending Title I sc hools than those in non Title I schools. Upon further investigation it was found that Title I school students were significantly less likely to endorse whole child indicators and significantly more

PAGE 4

iv likely to experience detrimental behaviors. Findings sugge st that monitoring tools, such as the Whole Child Student Survey, are an effective first step in scaling up positive psychological initiatives to large school districts. Future studies may benefit from experimental or qualitative research designs to in vest igate additional risk and protective factors of student mental well being. The form and content of this abstract are approved. I recommend its publication. Approved: Franci Crepeau Hobson

PAGE 5

v TABLE OF CONTENTS I . INTRODUCTION ................................ ................................ ................................ ................... 1 Study Aims and Significance ................................ ................................ ................................ .. 2 II . LITERATURE REVIEW ................................ ................................ ................................ ....... 4 Background ................................ ................................ ................................ .............................. 4 Current Directions ................................ ................................ ................................ ................... 5 Critical Considerations ................................ ................................ ................................ ............ 7 Present Challenges ................................ ................................ ................................ ................... 8 III . METHOD ................................ ................................ ................................ ............................ 10 Design ................................ ................................ ................................ ................................ .... 10 Participants ................................ ................................ ................................ ............................ 10 Measures ................................ ................................ ................................ ................................ 11 Procedure ................................ ................................ ................................ ............................... 11 Analysis ................................ ................................ ................................ ................................ . 12 IV . RESULTS ................................ ................................ ................................ ........................... 14 Normality of Data ................................ ................................ ................................ .................. 14 District Level Data Analysis ................................ ................................ ................................ . 14 Generalizable Trends across Grade Levels ................................ ................................ ........... 15 Comparison of Title I and Non Title I schools ................................ ................................ ..... 16 V . DISCUSSION ................................ ................................ ................................ ...................... 28 Implications ................................ ................................ ................................ ........................... 31 Li mitations and Future Directions ................................ ................................ ......................... 32 REFERENCES ................................ ................................ ................................ .......................... 35

PAGE 6

1 CHAPTER I INTRODUCTION The positive relationship between the mental well being of students and beneficial outc omes, such as improved academic achievement and diminished behavioral challenges , is well documented (e.g., Noble, McCandliss, & Farah, 2007; Weare & Nind, 2011 ). However, mental well being is more than simply the absence of mental illness, but an active p rocess sustained through skill acquisition, support ive environments , and social experiences . It follows then that schools , as supportive and social institutions of learning, are well positioned to nurture student mental well being . This is especially true for students who come from less than ideal home life circumstances and may have few other resources available to them. Beyond the boundaries of a school, there is extreme variability across home environments of students. Unfortunately, a growing proportio n of children and adolescents across the United States are exposed to poverty, toxic stress, and unsafe living conditions ( Federal Interagency Forum on Child and Family Statistics , 2016). The negative impact of these adverse conditions, including diminishe d academic achieve ment, increased school drop out, and poor mental health have been studied at length (e.g., Shonkoff et al., 2012 ; Cataldi, Laird, & Kewal , 2009 ). Moreover, research exploring the neurodevelopmental consequences of stressful environments i ndicates that children and adolescents are especially hard hit (e.g., Doige, 2007 ; Noble et al., 2015 ). Left unchecked, the cascading and cyclical nature of these conditions c ould precipitate devastating outcomes for students, schools , and communities alik e. In an effort to stem the significant mental health threats facing vulnerable student populations, many school districts are partnering with influential research organizations to study

PAGE 7

2 potential interventions for supporting student well being . Reports fr om the Institute of Medicine (2009) , the (2000) , and the World Health Organization (2014) have all indicated that promoting positive mental health attributes in schools , such as competence, social em otional learning, resilience, and supportive relationships aids in the prevent ion and treat ment of mental, emotional, and behavioral disorders . These claims match the growing desire of students, families, and teachers t o measure success as the ability to s olve personal problems and get along well with others , in addition to more traditional indicators such as academic performance and post secondary readiness (Axelrod & Markow, 2000). Taken together, it behooves school districts to allocate more resources to wards programs and init iatives that provide a more comprehensive analysis of student well being . Study Aims and Significance It is clear that there is great demand and potential to examine the capacity for schools to rigorously integrate the promoti on of s tudent well being . But how can schools identify whether they are meeting the mental health needs of their students? One promising framework for doing so, known as whole child education, has emerged from the field of positive psychology (Norrish, Williams, . Specifically, w hole child education emphasizes the promotion of positive mental attributes, such as engagement and social emotional learning, as a preventative strategy and monitoring tool. Therefore , the current study seeks to explore whether whole child education can be incorporated into a school district as a means of stimulating student mental well being and decreasing behavioral challenges. With one of the largest student bod ies in the state of Colorado, Denver Public Scho ols (DPS) reflects the promise and challenges of a large, diverse, and urban school district. Prominently displayed throughout its strategic plan , DPS has committ ed to

PAGE 8

3 surve ying student perceptions of whole child education components, su ch as feeling safe, challenge d , and health y . Th ese data are then compared to rates of chronic truancy, suspension s , and bullying. In turn, Student align planning and future goals with the n eeds of students ( Denver Public Schools , 2017) . The collection of whole child metrics across an entire school district presents an ideal opportunity to study the impact of promotin g student well being on behavioral challenges. In other words, it is now p ossible to analyze whether students who feel better (i.e., safer, challenged, socially emotionally intelligent, engaged, supported, and healthy) are subsequently less likely to experience bullying, miss school, or receive disciplinary suspensions. If there is direct evidence of a negative relationship among these variables , the n there will be abundant incentive for educators and policy makers a cross the nation to prioritize student well being as a fixed component of education in the 21 st century. T he curre nt study seeks to examine the whole child student survey data corresponding to the 2015 2016 school year in order to identify whether significant associations exist between indicators of whole child education and detrimental behavior s . The study also aims to test whether similar associations hold true for students from low income households . W ill an increased emphasis on whole child education allow schools to better respond to the adverse conditions of this population?

PAGE 9

4 CHAPTER II LITERATURE REVIEW Bac kground In their seminal paper on the topic of positive psychology, psychologists Seligman and Csikszentmihalyi (2000) a rgue d for a new or , perhaps , revived framework for investigating the conditions and experiences ne cessary for humans to flourish. This ambition is not unprecedented, however. P rominent h umanistic psychologists , including Abraham Maslow and Carl Rogers , previously advocated for a similar approach to psychological counseling during the mid 20 th century (Waterman, 2013) . Yet , Seligman and Cs ikszentmihalyi contend that the field of psych ology remains dominate d by research in pathology while failing to provide a greater understanding of the positive, protective, and fulfilling qualities of the human experience . In contrast, proponents of positi ve psychology believe that mental health research should incorporate both mental illness and positive attributes (Antaramian, Huebner, Hills, & Valois, 2010). On its surface, a shifting paradigm towards positive psychology may appear to be a n entirely phi losophical matter . H owever, Seligman and Csiksentmihalyi are primarily concerned with the develop ment of p ractical applications within psychology, education , and health . O f particular interest to school based mental health professionals is positive psychol inherent focus on prevention and development . From this perspective Seligman and Csikszentmihalyi mindedness, optimism, interpersonal skill, work ethic, hop the task of prevention in this new century will be to create a science of human strength whose

PAGE 10

5 ( 2000, p. 7). As a result of the direct implications for students, education, health, and development, it may not be surprising that positive psychology has appeal ed to many stakeholders in these fields . Seligman and Csikszentmihalyi (2000) are not alon e in advocating for expanding mental heal th capacities . Over the past few decades, a surge of research has explored such ideas as or uebner, Hills, & Valois, 2010; Vella Brodrick, Rickard, Hattie, Cross, & Chin, 2015; Marx & Wooley , 1998 ). Overlapping each of t hese various derivatives of positive psychology are the key tenants of (2012 being: positive e motions, engagement, relationships, meaning and purpose, and accomplishments . For example, there are clear parallels between the PERMA model and the primary objectives of whole child education as outlined by Centers for Disease Control and Prevention (CDC) and the Association for Supervision and Curriculum Development (ASCD; i.e., student health , engage ment , support, safe ty , and challenge ; Lewallen, Hunt, Potts Datema, Zaza & Giles, 2015 ). Thus, w hile the corresponding approach or audience for the se deriva tives may vary, each primarily emphasizes the promotion of well being and strengthening positive psychological traits, often in the c ontext of schools. Current Directions As a result of the exponential growth in positive psychology research , it has become possible to study generalizable trends within the associated body of literature. In a meta analysis of randomized controlled studies, researchers Bolier et al. (2013) suggest there is evidence that positive psychology interventions may enhance subjective w ell being, psychological well being, and reduce symptoms of depression. Unfortunately, Boiler et al. were not able to include school -

PAGE 11

6 based studies as part of their analys e s and the exact benefits of positive psychology in education continue to be investiga ted. However, a growing body of data indicates that positive education and academic achievement (Roffey, 2015). The Geelong Grammar School (GGS) of Victoria, Au stralia is often cited as a promising school based implementation ( & Robinson, 2013). Specifically , researchers Vella Brodrick, Rickard, Hattie, Cross, and Chin, (2015) have developed a longitu dinal study of whole school approach that embeds positive education training and curriculum. The aim of this project was to determine whether student well being improves as a result of this intervention . In a report evaluating the first two years of a cohort of 9 th grade students, findings demonstrate several improvements across measures of well being in the GGS students when compared with a matched control sample . However, academic engagement and achievement did not show similar progress at the time of reporting (Vella Brodrick et al., 2015) . I ndications that positive psychology can pr omote well being in students have inspired the formation of national committees and education reforms endorsing positive education programming. The CDC and the ASCD joi n tly devised a model that aims to incorporate student well being, academic achievement, and health within education. Known as the Whole School, Whole Community, Whole Child (WSCC) model , proponents of this framework encourage collaboration between institut ions of education and health to improve cognitive, physical, social, and emotional development ( Lewallen et al. , 2015 ). Since its development, supporters of the WSCC model point to the success of states and school districts identified as early a dopters of the framework. For example, the State of Colorado and Denver Public Schools

PAGE 12

7 (DPS) in particular, are touted as communities that exemplify strong collaboration s between education, community, and health stakeholders in order to foster healthy scho ol climates (Chiang, Meagher, & Slade, 2015) . In addition, DPS its 5 year strategic plan ( 2017, p. 782) . Although positive psychology is not explicitly mentioned in the DPS plan, emphas is on social and emotional needs, pursuit of passion and interests, and making responsible decisions reveal clear influences. Critical Considerations With the adoption of its tenants among leaders across disciplines and institutions, it may appear that p os itive psychology has been universally well received . However, both philosophical and empirical scrutiny of positive psychology has revealed criticisms and lingering questions regarding its validity and generalizability. In a critical review , Kristjá nsson ( 2012) asserts positive psychology and positive education lack empirical support, originality, cohesion, and conceptual clarity. For example, Kristjá nsson details the ambiguous definitions of happiness presented by positive psychologists, complicated furthe r by the intricate task of measuring happiness. Regarding the latter concern, the author remarks on the shortcomings associated with report measures . Kristjá nsson also comments that much of the strongest evide nce for the efficacy of positive psychology interventions are linked more closely to individual programs, such as resiliency training (e.g., Penn Resiliency Program) . Still, Kristjá nsson admits these allegations are due in part to the lack of research demo nstrating clear advantages of whole school approaches (e.g., Geelong Grammar School) to enhance well being (2012) . Apart from the forthrightly dissenting views of Kristjá nsson himself, others question the ability of schools to implement positive psychology interventions amid increasing academic

PAGE 13

8 demands. Sanderse, Walker, and Jones (2015) conducted qualitative analy s e s of interviews with and academic pressure . Their findings confirm that m any teachers, despite their best intentions, believe curricul a are too inflexible to accommodate additional, explicit support for character building and social emotional recommendations include concerted efforts to foster constructive relationship s between teachers and students, as well as between students through extra curricular activities. In addition to the se practical limitations , cr itics have express ed concern that positive psychology igno res the diverse backgrounds of traditionally disenfranchised student demographics. Specifically, authors Rao and Donal d son (2015) note in a review of the positive psychology literature a lack of attention to the challenges and succe sses of people of color. The authors also highlight the tendency of positive psychology to define well being according to Western values and historical perspectives . Thus, there remains a degree of uncertainty regarding the cultural responsiveness of posit ive psychology and whole child education. Rao and Donaldson conclude by presenting a call for positive psychology definitions to become more inclusive and utilizing its strengths based approach to empower all students (2015) . Present Challenges T aken toget her, it is clear that school based approaches , including whole child education , present an alluring case for educators and mental health professional s . However, it is important to consider how a shift away from the prevailing deficit based model can be utilized to benefit students of all backgrounds and classes . In a paper outlining the benefits of employing a pro active, universal, well being framework, Roffey (2016) argues that whole child education is well positioned to support child ren through adversity . Specifically, it is

PAGE 14

9 noted that many of the protective factors that help to cultivate resilience during hardship (supportive relationships, high expectations, opportunities to contribute, teaching social and emotional skills, collabor ating with families) are also emphasized in the whole child approach. Roffey also discusses the crucial function of school psychologists in this endeavor (2016) . Yet, d s, it remains to be seen whether these claims will be corroborated by research. Th e challenge to reconcile cultural responsiveness with mental health is certainly faced by other areas of psychology, but relatively r ecent emergence may confer an advantage in this regard. Through out the current literature review it is evident that the definitions and boundaries of positive psychology are still malleable. This presents a unique and significant opportunity to further incorporate issu es pertaining to social justi ce, as well as alloca te greater focus towards vulnerable student populations. To expedite this process , it is critical that further evidence be gathered demonstrating the positive benefits of whole child education towards students of color and those living in poverty. Without this validation it will be difficult to justify the continued implementation of positive psychology interventions in the school s .

PAGE 15

10 CHAPTER III METHOD Design The current study designates correlational analysis as its primary design classification. S pecifically , publicly available survey data w as measured descriptively, directionally, and for strength of statistically significant relationships among the whole child student survey variables. As a r esult, it is possible to gain an enhanced understanding of whether these variables have positive or negative associations with one another . For the purposes of this study, it is hypothesized that schools with higher percentages of students endorsing positive well being indicators will have lower percentages of behavioral challenges. Participants The total survey sample consisted of 3 rd through 12 th grade students attending Denver Public Schools (DPS) during the 2015 2016 school year. DPS is a larg e , urban public school district located in Denver, Colorado. S ur vey data w ere collected from 109 elementary, 66 middle, and 55 high schools. Of these schools, eight (three elementary, three middle, and two high schools) were excluded from the analys e s due to missing or incomplete data. Although all schools represented are publicly funded, both traditional and charter schools are included in the sample. According to the 2016 DPS report of student membership by ethnicity and gender, d istrict wide demographic information of the student population indicates tha t students pr imarily identified as Latino (55%), White (23%), and Black or African American (13%). Genders were approximately equal in distribution (51% male ) .

PAGE 16

11 Information regarding DPS schools receiving federal Title I funding was collected as a measure of high pove rty rates to be used in comparison with non Title I schools. In total, 59 elementary, 7 kindergarten 8 th grade , 15 middle, 19 high, 15 alternative, and 40 charter schools were found to receive Title I support. These schools w ere matched with their respecti ve whole child survey school names in order to be included in analysis. Measures The Whole Child Student Survey (Denver Public Schools, 2017) was created by the DPS Department of Accountability, Research & Evaluation . Two separate versions of the instrum ent are available in order to differentiate between primary (grades 3 5) and secondary (grades 6 12) students. Specifically, the survey is a self report measure used to capture the percentage of students that reported feeling challenged, engaged, healthy, safe, socially emotionally intelligent, and supported by members of the school community. These variables were selected and included in the survey on the basis of their membership in the Association for Supervision and Curriculum Development ( ASCD ) whole c hild support model, with the exception of social emotional intelligence, which was included by DPS as an additional indicator of whole child support (Lewallen et al. , 2015 ) . In addition, the survey also collects information regarding rates of bullying ( sep arated into two categories at the secondary level, physical and sexual orientation, depending on the nature of the bullying incident) , chronic absences, and out of school suspensions, as well as those receiving daily breakfast. Procedure T he survey is ad ministered electronically towards the end of each school year as part of survey to all eligible students in grades 3 12 and held accountable for doing so, in order to p romote a high

PAGE 17

12 participation rate. Data received from each school are then reported as a percentage to indicate the number of students endorsing each survey item. These results a re then compiled and reported as a table on the DPS website. Thus, a ll data use d for the purposes of the current study were retrieved from the DPS website , view able only in aggregate under the title of the corresponding school ( Denver Public Schools , 2017) . As such, n o individual identifying information was utilized and all student d ata was anonymous. Whole Child Student Survey data for the 2015 16 academic year was acquired from the DPS website and transferred to Microsoft Excel spreadsheet s. This data was then further analyzed using IBM Statistical Package for Social Studies (SPSS). Analysis It wa s hypothesized that the whole child variables will correlate negative ly with the behavioral variables. In other words, the current study aimed to identify whether there are statistically significant associations between higher rates of who le child support and lower behavioral outcomes. Additionally, it was hypothesized that similar negative correlations would occur regardless of whether students attend Title I schools or non Title I schools. Prior to the main analyses, the normality of the data within each survey variable was examined. P reliminary, descriptive analys e s were conducted on the survey data . This analysis comprised creating a multiple correlation table, whereby each of the six whole child variables (Challenged, Engaged, Healthy, Safe, Supported, and Socially Emotionally Intelligent), a mean for each of the se variables, reported breakfast intake, and outcome variables (Bullying, Chronic Absence, and Suspensions) were represented . This table then revealed the observed correlation, as well as the strength and direction of that correlation, between each variable.

PAGE 18

13 Subsequent to the multiple correlation analysis, correlations between whole child variables and outcome behavioral variables were tested for statistical significance through multiple linear regression t test analysis . This analysis is conducted automatically as a byproduct of correlation analysis within SPSS . Correlations found to be significant, along with their corresponding threshold for significance, were indicated within the multiple correlation table. Both the correlation and regression analysis procedures were repeated in i solation for schools identified and not identified as receiving Title I funding . Follow up t test s between these latter school groups w ere conducted to determine statistical differences between the mean scores reported by each group.

PAGE 19

14 CHAPTER IV RES ULTS Normality of Data Prior to conducting the correlational analys e s, each variable within each grade level was checked for normal distribu tion utilizing the Kolmogorov Smirnov test of normality. These tests are recommended by Mayer (2013) for sample sizes greater than 50. For each grade level , there were several variables determined to have distributions significantly different from normal ( see Table 1). As a result it was necessary to employ a when analy zing thus can be used to ensure confidence in the statistical sign ificance of correlations when one or both variables do not meet the underlying (Mayer, 2013) . The latter correlation was utilized for the remaining, normally distributed variables. District Level Data Analysis Correlat ional analysis was conducted for each of the three available data sets corresponding to grades 3 5, 6 8, and 9 12. Across grade levels , there were s ignificantly negative correlations observed between behavior and whole child variables, including the mean o f whole child variables and regular breakfast consumption . However, due to the variation in survey items between the separate versions of the survey, it is most accurate to report results as they pertain to each grade level. Results of the grades 3 5 cor relational analysis (see Table 2) indicated significant negative correlations between . In other words, these results reveal that students who

PAGE 20

15 endorsed fe eling challenged, engaged, healthy, safe and socially emotionally intelligent reported significantly fewer instances of bullying. Significant negative correlations were also observed , , and all whole child variables aside from the items, the latter of which showed a significant positive c Grades 6 8 demonstrated significant negative correlations (see Table 3) between , ation , , and all whole child variables. The Finally, upon reviewing the data pertaining to grades 9 12 (s ee Table 4) , showed significant negative correlations with all whole child variables , while also negatively correlated with all whole child variables and days. show ed signif icant negative correlations for Emotionally Generaliz able Trends across Grade Levels Throughout the correlational analysis, it was evident that whole child variables positively correlated amongst one another. The same trend was observed between behavior variables. Regardless of grade level, all behavior var iables negatively correlated with at least one whole child variable, with the for grades 9 12, which only did so for I tems related to bullying tended to show significant negative correlations with nearly all whole child variables. In addition , fewer negative correlations between whole

PAGE 21

16 child and behavior variables were observed for survey respondents in grades 9 12 t han their younger counterparts. Comparison of Title I and Non Title I schools Results of t he correlational analysis comparing Title I and non Title I schools within each grade level indicated significant negative correlations for both populations between some whole child variables and behavior variables (see Tables 5, 6, and 7 for detailed resu lts within each grade level) . However, in nearly all cases, the strength of these correlations w as diminished for Title I schools when compared to non Title I schools . Moreover, there were substantially fewer cases of significant negative correlations betw een whole child variables and behavior variables for grades 3 5 and 6 8. Similar instances of significant relationships were observed across both populations for g rades 9 12. The tendency for Title I schools to have fewer significant and generally weaker correlations was further explored using an independent samples t test for each variable comparing students in Title I schools with those in non Title I schools. During the process of the t test analysis, Test for Equality of Variances was used to reduce the potential influence of unequal regression variance, also known as heteroscedasticity, between groups. Results of the t test analysis uncovered significant discrepancies between students in these two populations (see Tables 8, 9, & 10 for results by grade level) . Specifically, aside from feeling more challenged than their non Title I peers, students attending Title I schools in grade s three through five reported significantly less whole child support. Title I elementary school students were also s ignificantly more likely to experience bullying, chronic absence, and out of school suspension. Across all grade levels, students attending Title I schools reported feeling significantly lower social emotional intelligence and regular breakfast intake . Tit le I students in

PAGE 22

17 grades six through eight also faced significantly higher chronic absent eeism and suspensions, while those in grades nine through twelve encountered more instances of bullying.

PAGE 23

18 Table 1 Kolmogorov Smirnov Tests of Normality Grad e Level 3 5 6 8 9 12 Variable Statistic df Sig. Statistic df Sig. Statistic df Sig. Challenged .109 106 0.003* .123 62 0.021* .113 53 0.090 Engaged .096 106 0.018* .073 62 0.200 .113 53 0.088 Healthy .086 106 0.052 .081 62 0.200 .111 53 0 .151 Safe .052 106 0.200 .083 62 0.200 .100 53 0.200 SEI .076 106 0.161 .072 62 0.200 .065 53 0.200 Supported .098 106 0.013* .091 62 0.200 .079 53 0.200 Whole Child Mean .079 106 0.098 .080 62 0.200 .137 53 0.014* Breakfast (5+ days) .066 106 0.200 .112 62 0.053 .084 53 0.200 Bullying , Reported .069 106 0.200 ------Bullying , Physical ---.118 62 0.031* .104 53 0.200 Bullying , Orientation ---.121 62 0.024* .102 53 0.200 Absences .056 106 0.200 .109 62 0.063 .124 53 0.041* Suspensions .233 106 0.000* .299 62 0.000* .275 53 0.000* *. p <0.05; degrees of freedom (df) ; Socially Emotionally Intelligent (SEI) ; variable not measured at grade level ( -)

PAGE 24

19 Table 2 Correlation analysis across grades 3 5 Descriptive Statistics (N = 106) Variables 1 2 3 4 5 6 7 8 9 10 11 1. Challenged ( p s ) 1 2. Engaged ( p s ) .322 ** 1 3. Healthy .069 .498 ** 1 4. Safe .330 ** .831 ** .594 ** 1 5. SEI .108 .700 ** .663 ** .749 * * 1 6. Supported ( p s ) .500 ** .647 ** .317 ** .595 ** .471 ** 1 7. Mean .435 ** .903 ** .683 ** .924 ** .822 ** .735 ** 1 8. Breakfast 5+ days .176 .479 ** .539 ** .517 ** .643 ** .214 * .530 ** 1 9. Bullying, reported .009 .579 ** .660 ** .816 ** .757 ** .283 ** .727 ** .642 ** 1 10. Chronic Absences .336 ** .338 ** .545 ** .484 ** .532 ** .032 .389 ** .523 ** .668 ** 1 11. Suspensions ( p s ) .060 .380 ** .318 ** .375 ** .507 ** .105 .376 ** .340 ** .408 ** .442 ** 1 * p< 0.0 5 (2 tailed) ** p< 0.01 (2 tailed) Socially Emotionally Intelligent (SEI) ( p s

PAGE 25

20 Table 3 Correlation analysis across grades 6 8 Descriptive Statistics (N = 6 2 ) Variables 1 2 3 4 5 6 7 8 9 10 11 12 1. Chall enged ( p s ) 1 2. Engaged .776 ** 1 3. Healthy .344 ** .704 ** 1 4. Safe .826 ** .755 ** .471 ** 1 5. SEI .375 ** .659 ** .843 ** .503 ** 1 6. Supported .915 ** .805 ** .345 ** .858 ** .373 ** 1 7. Mean .820 ** .937 ** .763 ** .875 ** .766 ** .840 ** 1 8. Breakfast 5+ days .335 ** .478 ** .559 ** .408 ** .685 ** .369 ** .582 ** 1 9. Bullying, physical ( p s ) .622 ** .693 ** .691 ** .670 ** .584 ** .597 ** .777 ** .445 ** 1 10. Bullying, orientation ( p s ) .619 ** .668 ** .397 ** .690 ** .406 ** .648 ** .695 ** .261 * .687 ** 1 11. Chronic Absences .086 .089 .266 * .126 .410 ** .146 .234 .580 ** .244 .165 1 12. Suspensions ( p s ) .308 * .509 ** .590 ** .415 ** .721 ** .341 ** .604 ** .641 ** .437 ** .305 * .635 ** 1 * p < 0.05 (2 tailed) ** p <0.01 (2 tailed) Socially Emotionally Intelligent (SEI) ( p s

PAGE 26

21 Table 4 Correlation analysis across grades 9 12 Descriptive Statistics (N = 53) Varia bles 1 2 3 4 5 6 7 8 9 10 11 12 1. Challenged 1 2. Engaged .650 ** 1 3. Healthy .253 .473 ** 1 4. Safe .572 ** .610 ** .279 * 1 5. SEI .356 ** .613 ** .842 ** .397 ** 1 6. Supported .852 ** .709 ** .260 .7 37 ** .406 ** 1 7. Mean ( p s ) .669 ** .875 ** .547 ** .719 ** .734 ** .779 ** 1 8. Breakfast 5+ days .097 .072 .036 .123 .086 .144 .113 1 9. Bullying, physical .257 .684 ** .438 ** .370 ** .586 ** .406 ** .647 ** .191 1 10. Bullying, orientat ion .303 * .640 ** .211 .558 ** .348 * .445 ** .489 ** .243 .584 ** 1 11. Chronic Absences ( p s ) .148 .272 * .045 .063 .029 .158 .132 .605 ** .199 .415 ** 1 12. Suspensions ( p s ) .073 .158 .166 .205 .320 * .163 .272 * .264 .229 .099 .406 ** 1 * p < 0.05 (2 tailed) ** p <0.01 (2 tailed) Socially Emotionally Intelligent (SEI) ( p s

PAGE 27

22 Table 5 Correl ation analysis across grades 3 5, comparing non Title I and Title I schools Descripti ve Statistics ( Non Title 1 N = 34; Title 1 N = 72 ) Variables 1 2 3 4 5 6 7 8 9 10 11 1. Challenged .506 ** .013 .489 ** .368 ** .647 ** .669 ** .049 .178 .182 .050 2. Engaged .597 ** .227 .749 ** .594 ** .594 ** .866 ** .237 * .352 ** .117 .296 * 3. Hea lthy .120 .593 ** .344 ** .367 ** .175 .465 ** .085 .317 ** .207 .117 4. Safe .506 ** .881 ** .695 ** .611 ** .509 ** .879 ** .224 .724 ** .377 ** .328 ** 5. SEI .294 .775 ** .779 ** .858 ** .389 ** .739 ** .310 ** .543 ** .222 .479 ** 6. Supported .605 ** .686 ** .416 * .655 ** .555 ** .721 ** .056 .122 .205 .025 7. Mean .578 ** .927 ** .767 ** .956 ** .884 ** .755 ** .222 .536 ** .155 .292 * 8. Breakfast 5+ days .133 .644 ** .653 ** .692 ** .721 ** .373 * .703 ** .275 * .085 .073 9. Bullying, reported .273 .721 ** .7 15 ** .885 ** .805 ** .418 * .821 ** .648 ** .516 ** .270 * 10. Chronic Absences .184 .154 .551 ** .314 .480 ** .064 .310 .480 ** .483 ** .245 * 11. Suspensions .062 .325 .402 * .370 * .479 ** .064 .359 * .356 * .461 ** .431 * * p <0.05 (2 tailed) ** p< 0.01 (2 tailed) Socially Emotionally Intelligent (SEI) Title I schools represented in shaded sections

PAGE 28

23 Table 6 Correlation analysis across grades 6 8 , comparing non Title I and Title I schools Descriptive Statistics ( Non Title I N = 22; Title I N = 40 ) Variables 1 2 3 4 5 6 7 8 9 10 11 12 1. Challenged .729 ** .207 .803 ** .126 .919 ** .780 ** .259 .440 ** .523 ** .108 .033 2. Engaged .777 ** .636 ** .664 ** .544 ** .774 ** .928 ** .258 .612 ** .640 ** .036 .177 3. Healthy .401 .759 ** .334 * .821 ** .239 .713 ** .361 * .732 ** .504 ** .147 .305 4. Safe .861 ** .823 ** .573 ** .276 .818 ** .816 ** .042 .587 ** .656 ** .036 .228 5. SEI .538 ** .803 ** .897 ** .692 ** .182 .643 ** .324 * .556 ** .430 ** .180 .439 ** 6. Supported .950 ** .8 32 ** .432 * .896 ** .557 ** .817 ** .119 .427 ** .566 ** .073 .041 7. Mean .850 ** .949 ** .784 ** .923 ** .863 ** .878 ** .280 .721 ** .706 ** .092 .169 8. Breakfast 5+ days .608 ** .685 ** .678 ** .730 ** .753 ** .657 ** .785 ** .207 .061 .406 ** .411 ** 9. B ullying, physical .619 ** .809 ** .806 ** .764 ** .764 ** .690 ** .854 ** .645 ** .778 ** .043 .001 10. Bullying orientation .552 ** .700 ** .581 ** .770 ** .757 ** .638 ** .766 ** .572 ** .580 ** .048 .067 11. Chronic Absences .081 .065 .166 .27 7 .276 .212 .203 .522 * .207 .410 .596 ** 12. Suspensions .551 ** .740 ** .720 ** .716 ** .812 ** .586 ** .789 ** .589 ** .822 ** .665 ** .251 * p <0.05 (2 tailed) ** p<0.01 (2 tailed) Socially Emotionally Intelligent (SEI) Title I schools represented in shaded sections

PAGE 29

24 Table 7 Correlation analysis across grades 9 12 , comparing non Title I and Title I schools Descriptive Statistics (Non Title I N = 12; Title 1 N = 41) Variables 1 2 3 4 5 6 7 8 9 10 11 12 1. Challenged .662** .343* .610** .454** .869** .775** .213 .245 .205 .009 .183 2. Engaged .689* .615** .621** .684** .709** .925** .029 .726** .608** .388* .232 3. Healthy .198 .110 .242 .903** .339* .760** .043 .504** .262 .042 .414** 4. Safe .564 .600* .428 .290 .754** .682** .180 .286 .445** .176 .079 5. SEI .103 .267 .835** .641* .416** .802** .012 .567** .359* .011 .454** 6. Supported .861** .678* .069 .718** .297 .808** .184 .340* .366* .161 .120 7. Mean .629* .736** .494 .900** .772** .747** .075 .603** .491** .181 .280 8. Breakfast 5+ .616* .819** .114 .368 .069 .420 .518 .286 .188 .525** .218 9. Bullying, phys .394 .544 .280 .536 .616* .644* .714** .086 .568** .470** .004 10. Bullying, ori .695* .882** .074 .840** .422 .801** .851** .578* .656* .511** .129 11. Absences .263 .449 .054 .044 .112 .017 .225 .700* .170 .166 .308 12. Suspensions .113 .291 .649* .777** .796** .419 .729** .113 .513 .528 .105 * p <0.05 (2 tai led) ** p <0.01 (2 tailed) Socially Emotionally Intelligent (SEI) Title I schools represented in shaded section

PAGE 30

25 Table 8 Grades 3 5 Means , Standard Deviations and Independent T test Analysis Grade Level 3 5 Mean SD p Whole Child Variables Non Title I (n = 34) Title I (n = 72) Non Title I Title I Challenged 0.86 0.88 0.03 0.04 .007** Engaged 0.84 0.80 0.05 0.05 .000** Healthy 0.80 0.74 0.06 0.04 .000** Safe 0.77 0.71 0.07 0.06 .000** Socially Emotionally Intelligent 0.93 0.89 0.03 0.03 .000** Supported 0.88 0.87 0.04 0.03 .165 Mean of Whole Child Variables 0.85 0.81 0.04 0.03 .000** Breakfast (5+ Days) 0.82 0.68 0.07 0.07 .000** Bullying, Reported 0.30 0.50 0.16 0.11 .000** Chronic Absenteeism 0.10 0.19 0.06 0.06 .000** Suspensions 0.02 0.03 0.02 0.03 .004** ** p <.01 * p <.05 Standard deviation (SD)

PAGE 31

26 Table 9 Grades 6 8 Survey Means, Standard Deviation s , and Independent T test Analysis Grade Level 6 8 Mean SD p Whole Child Variables Non Title I (n = 22) Title I (n = 40) Non Title I Title I Challenged 0.86 0.86 0.07 0.05 .920 Engaged 0.69 0.66 0.09 0.06 .142 Healthy 0.80 0.75 0.07 0.06 .021* Safe 0.82 0.79 0.09 0.07 .224 Socially Emotionally Intelligent 0.76 0.66 0.09 0.06 .000** Supported 0.86 0.84 0.06 0.05 .346 Mean of Whole Child Variables 0.80 0.76 0.07 0.05 .050 Breakfast (5+ Days) 0.67 0.52 0.10 0.08 .000** Bullying, Physical 0.45 0 .47 0.13 0.10 .507 Bullying, Orientation 0.50 0.50 0.15 0.10 .977 Chronic Absenteeism 0.14 0.31 0.08 0.16 .000** Suspensions 0.03 0.13 0.04 0.20 .027* ** p <.01 * p <.05 Standard deviation (SD)

PAGE 32

27 Table 10 Grades 9 12 Sur vey Means, Standard Deviation s , and Independent T test Analysis Grade Level 9 12 Mean SD p Whole Child Variables Non Title I (n = 12) Title I (n = 41) Non Title I Title I Challenged 0.87 0.88 0.03 0.04 .605 Engaged 0.67 0.64 0.08 0.1 0 .363 Healthy 0.75 0.76 0.05 0.06 .764 Safe 0.91 0.89 0.05 0.04 .343 Socially Emotionally Intelligent 0.75 0.70 0.07 0.07 .047* Supported 0.89 0.88 0.04 0.05 .427 Mean of Whole Child Variables 0.81 0.79 0.04 0.05 .329 Breakfast (5+ D ays) 0.51 0.42 0.07 0.13 .000** Bullying, Physical 0.19 0.22 0.09 0.07 .007** Bullying, Orientation 0.30 0.29 0.14 0.11 .002** Chronic Absenteeism 0.38 0.56 0.26 0.26 .232 Suspensions 0.04 0.10 0.04 0.15 .619 ** p <.01 * p <.05 Standar d deviation (SD)

PAGE 33

28 CHAPTER V DISCUSSION This study aimed to examine the relationship between indicators of whole child education, including student perceptions of feeling challenged, engaged, healthy, safe, supported, and socially emotionally in telligent, and detrimental behaviors, such as bullying, chronic absences, and out of school suspensions in a large, urban public school district. Associations among these variables were examined for students in both Title 1 and non Title 1 schools. The cor relational analys e s conducted as part of this study reveal that students typically display less This finding is consistent with previous studies examining the ro le of protective factors, such as perceived support and social emotional intelligence , and improved a cademic and behavioral outcomes for students living both above and below the poverty line ( e.g., Yeung & Leadbeater, 2010 ; Hopson & Lee, 2011; Durlak et al ., 2011). Furthermore, the current study begins to bridge the gap in evide nce needed to support the adoption of whole child education models within large, public school district s . Throughout the positive psychology literature base , it is frequently mention ed that the successes of positive education initiatives within more privileged school settings have not yet been replicated in public school districts (e.g., Rao & Donaldson, 2015; Roffey, 2016). Yet, across the Denver Public School ( DPS ) district , there i s a strong pattern of reduced behavior al challenges when students highly rate indicators of whole child support. Thus, t he clear , negative relationship between these variables suggests that student well being and undesir able conduct appear to be incompatib le with one another irrespective of population .

PAGE 34

29 Beyond addressing the scalability of the whole child framework, its sensitivity to the nuanced perspective of studen ts attending Title I schools has also been questioned (Rao & Donaldson, 2015). Due to the overwhelming evidence corroborating the negative impacts of poverty on developmental and educational outcomes , it is reasonable to suspect that the responses of students from low income households may differ from their peers (e.g., Yoshikawa, A ber, & Beardslee, 2012 ; Akee et al., 2010 ) . However, advocates of positive education models maintain that schools can play a vital role in nurturing resiliency and relationships in disadvantaged youth, and that focusing on the whole child provides a strong foundation to do so (Roffey, 2016). By studying Title I schools in isolation , it is possible to evaluate if whole child education model is a valid framework for students of all socioeconomic backgrounds . Because Title I schools serve a high proportion of students from low income households , they are more likely to reflect the influence s of poverty . When viewing the results of Title I schools, the relationship between student well being and behavior was established . However, the associations were neither as pronounced, nor as frequent as their non Title I counterparts. Thus, when comparing the typical responses from Tit le I schools against non Title I schools , it is not surprising that students in Title I schools report feeling less engaged, healthy, safe, s upported, and socially emotionally intelligent, while experiencing more bullying, chronic absences, and suspensions. Taken together, this finding further highlights the harmful effects of poverty emphasized by the research community and supports the need t o expand measures such as the Whole Child Student Survey in order to monitor and account for these metrics. The observed differences between the responses of students in Title I and non Title I schools is best understood by viewing each of the three grade level categories (i.e., grades 3 5, 6 8, and 9 12) separately. In grades three through five , the discrepancies are most glaring. Only

PAGE 35

30 ed the two population s , while In contrast, elementary students in non Title I schools reported significantly less detrimental behaviors and significantly higher ratings of nearly all whole child variables , including co nsistent breakfast consumption. Although it is not yet possible to conclude why elementary students attending Title I schools report feel ing more challenged than their non Title I counterparts , the overall disproportionality across these student pop ulation s is difficult to overlook. In c ompari ng the middle sc hool and high school aged segment s, there are noticeably fewer significant differences between Title I and non Title I schools. Similar to the elementary population, both levels of secondary students at tending Title I schools report feeling less socially emotionally intelligent and less likely to eat breakfast five or more days per week. Students in Title I middle schools also report feeling less healthy, while experiencing an increased rate of chronic a bsenteeism and out of school suspensions. At the hig h school level, students attending Title I schools are more likely to experience bullying. Taken together, as grade levels increase, there is a clear narrowing of the gap between ratings of Title I and no n Title I students. Across all grade levels, Title I students are less likely to have a regular breakfast intake . This finding matches the previous conclusions of authors Hall, Chai, and Albrecht demonstrating the disparity in eating habits and nutrition knowledge between Title I and non Title I student populations ( 2016 ). In addition, Title I students are also generally less likely to feel socially emotionally intelligent, and more likely to experience detrimental behavior. However, whereas nearly all who le child variables favor non Title I students at the elementary level, there i s only

PAGE 36

31 one significant difference at the high school level. This observed convergence of ratings as students become older can possibly be attributed to the substantial dropout ra te of students living in poverty (e.g., Shonkoff et al., 2012), thereby resulting in an increasingly larger proportion of students with the personal traits, resources , and support that is crucial to continuing their education. The potential impact of attri tion upon the ensuing student body composition and family dynamics warrants further exploration. Overall, while the current study confirms the negative association between whole child variables and detrimental behavior in both Title I and non Title I schoo the primary grade levels, demonstrates that the effects of poverty are wide ranging . Implications Public education in th e United States has historically focused on the aca demic achievement of its students. However, the relatively recent surge of positive psychology, social emotional learning, and whole child education literature has reinvigorated humanistic goals in school districts across the United States, including DPS. These goals in turn align with calls for schools to become more active in fostering social emotional intelligence, problem solving, and critical thinking, which advocates posit are necessary in order to prepare children and adolescents for careers in the 2 1 st century ( Kivunja, 2015). And just as schools are rewarded for fostering academic learning, experts agree that efforts to monitor and promote and well being should also be encouraged (Bonell et al., 2014) . However, regardless of their e xternal recognition, the findings of the present study confirm that schools with bolstered support for the whole child are reaping the benefits of improved behavior and mental well being. The verified trend for schools with higher rates of whole child ind icators to exh ibit lower detrimental behavior is, to a lesser extent, also seen in DPS Title I schools. Significant

PAGE 37

32 differences between reports of students in Title I schools and non Title I schools indicate that the former are likely to need greater suppo rt. With the current allotment of resources across the district, students in Title I schools experience more detrimental behavior and less mental well being. Therefore, a greater emphasis must be placed on supporting the needs of these students, schools, a nd communities. The current study also carries significance within the positive psychology literature. The interrelated link between student well being and behavior highlights a key principle of positive psychology: nurturing positive characteristics in y oung people offers added , practical value to schools . Furthermore, the success of previous implementations of positive psychology in private schools, such as the Geelong Grammar School, can and should be generalized to public education (Vella Brodrick, Ric kard, Hattie, Cross, & Chin, 2015). But in practice, however, Sanderse, Walker, and Jones (2015) brought attention to reservations of educators related to the additional time required to address educating the whole child. These are valid concerns among the increasingly standardized and test driven societies, and it is argued here that improving the mental well being of students may require a shift in pri orities and/or strategic planning from many school districts. Limitations and Future Directions One pot ential limitation of the current study relates to its research design. Due to the reliance on correlation analysis, questions regarding causation are not able to be addressed here. For example, it cannot be implied from the current findings that improving whole child education ratings moderates detrimental behaviors. As a result, it is not possible to rule out the possibility that reducing the presence of bullying, chronic absences, and suspensions allows positive traits to thrive. It is recommended that fu ture studies employ experimental conditions,

PAGE 38

33 such as implementing specific school wide interventions to improve whole child metrics and comparing against similar , control schools without interventions. In a meta analysis of randomized controlled trials, au thors Weiss, Westerhof, and Bohlmeijer provide examples of effective pos itive psychological i nterventions to promote well being (2016) . It is also recommended that follow up qualitative inquiries be conducted to assess the conditions and assets of schools reporting high versus low rates of whole child education indicators. The current research design was also limited to self report, survey style measures that are often criticized for their openness to bias and diminished validity (Coolican, 2017) . However, the measures used here were designed to be sensitive to age differences and demonstrated success in detecting cross school differences. Due to the scope and purpose of this study, it was not feasible to include additional school level factors that may pot entially act as covariates or risk factors . For example, the current study did not account for schools that are near, but not meeting the threshold required to receive Title I status. Schools falling in this category may behave more similarly to those with Title I status than those without. Future studies are also encouraged to examine whether the racial and/or ethnic composition of schools contributes to student ratings. Reports, such as that of English, Lambert, and Ialongo (2016), reveal that experiencin g racial discrimination significantly increases the likelihood of developing depressive symptoms and diminished academic variables. In addition, a s pointed ou t by Rao and Donaldson (2015), cultural backgrounds can impact the definition and importance of the variables studied. These next steps will provide the critical validation necessary to evaluate whether whole child education and positive psychology interve ntions are appropriately meet ing the needs of every student.

PAGE 39

34 As evidenced by the current study and its contribution to the body of literature that emphasizes negative influence on student wel l being, there is a clear need to continue exploring h ow to best support the whole child. For students attending Title I schools , there is often an inherent focus on deficits financial or otherwise . However , the promising results of positive education initiatives represent a n intentional shift to rectify this traditional approach. In addition , with the initiation of monitoring tools such as the Whole Child Student Survey, it is increasingly plausible for schools to identify and promote areas of strength, while also addressing areas of need . School based m ental health professionals can, in turn, utilize these measures to bolster their assessment and prevention efforts. Furthermore, school teams can collaborate with one another to share success es and evaluate resource allocations. In a time of great need , t here i s now a great opportunity to leverage the resiliency of youth against the upbringing .

PAGE 40

35 REFERENCES Akee, R. K., Copeland, W. E., Keeler, G., Angold, A., & Costello, E. J. (2010). Parents' incomes and children's outcomes: a quasi experiment using transfer payments from casino profits. American Economic Journal: Applied Economics , 2 (1), 86 115. Axelrod, A., & Markow, D. (2000). The Metropolitan Life Survey of the American Teacher, 2000: Are We Preparing Students for the 21st Century? Metropolitan Life Insurance Company . Bolier, L., Haverman, M., Westerhof, G. J., Riper, H., Smit, F., & Bohlmeijer, E. (2013). Positive psychology interventions: a meta analysis of randomized controlled studies. BMC P ublic H ealth, 13 (1), 1. Bo nell, C., Humphrey, N., Fletcher, A., Moore, L., Anderson, R., & Campbell, R. (2014). Why Bmj , 348 (7958), g3078. Cataldi, E. F., Laird, J., & Kewal Ramani, A. (2009). High school dropout and completi on rates in the United States: 2007 . Washington, DC: Institute of Education Sciences' National C enter for Education Statistics. Retrieved from http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid=2009064 Chiang, R. J., Meagher, W., & Slade, S. (2015). How the Whole School, Whole Community, Whole Child model works: creating greater alignment, integration, and collaboration between health and education. Journal of S chool H ealth , 85 (11), 775 784. Coolican, H. (2017). Research methods and statistics in psycholog y . New York, NY: Psychology Press. Denver Public Schools Financial Services, Student Submission Team (2016). Report of Student Membership Based on the 2016 Pupil Count Submission . Denver Public Schools. Retrieved from http://thecommons.dpsk12.org/cms/li b/CO01900837/Centricity/domain/31/student%20su bmissions/forms/OC_Membership_2016.pdf Denver Public Schools . (2017, February 21). Denver Public Schools: Whole Child . Retrieved from http://wholechild.dpsk12.org/support for the whole child/whole child stude nt survey/#1476747492382 cbe073ca 3bbb Doidge, N. (2007). The brain that changes itself: Stories of personal triumph from the frontiers of brain science . New York, NY: Penguin. Durlak, J. A., Weissberg, R. P., Dymnicki, A. B., Taylor, R. D., & Schellin ger, K. B. (2011). The based universal interventions. Child development , 82 (1), 405 432.

PAGE 41

36 Federal Interagency Forum on Child and Family Statistics (2016) . ldren in Brief: Key National Indicators of Well Being. Washington, DC: U.S. Government Printing Office. Hall, E., Chai, W., & Albrecht, J. A. (2016). Relationships between nutrition related knowledge, self efficacy, and behavior for fifth grade students attending Title I and non Title I schools. Appetite , 96 , 245 253. Hopson, L. M., & Lee, E. (2011). Mitigating the effect of family poverty on academic and behavioral outcomes: The role of school climate in middle and high school. Children and Youth Serv ices Review , 33 (11), 2221 2229. improve pedagogies of the new learning paradigm. Creative Education , 6 (02), 224. Kline, R. B. (2011). Principles and practice of structural equation modeling (3rd ed.). New York , NY : The Guilford Press. Kristjánsson, K. (2012). Positive psychology and positive education: Old wine in new bottles? Educ ational P sychologist , 47 (2), 86 105. Langford, R., Bonell, C. P., Jones, H. E., Pouliou, T., Murphy, S. M., Waters, E., ... & Campbell, R. (2014). The WHO Health Promoting School framework for improving the health and well being of students and their aca demic achievement. The Cochrane Library . L ewallen, T. C., Hunt, H., Potts Datema, W., Zaza, S., & Giles, W. (2015). The Whole School, Whole Community, Whole Child model: a new approach for improving educational attainment and healthy development for stud ents. Journal of School Health , 85 (11), 729 739. Marx, E. E., & Wooley, S. F. E. (1998). Health is academic: A guide to coordinated school health programs . New York, NY: Teachers College Press. Mayers, A. (2013). Introduction to Statistics and SPSS in P sychology . London: Pearson. National Research Council. (2009). Preventing mental, emotional, and behavioral disorders among young people: Progress and possibilities . Washington, DC: National Academies Press. Noble, K. G., Houston, S. M., Brito, N. H., Bartsch, H., Kan, E., Kuperman, J. M., ... & Schork, N. J. (2015). Family income, parental education and brain structure in children and adolescents. Nature neuroscience , 18 (5), 773.

PAGE 42

37 Noble, K. G., McCandliss, B. D., & Farah, M. J. (2007). Socioeconomic g radients predict individual differences in neurocognitive abilities. Developmental science , 10 (4), 464 480. positive education. International Journal of Well bei ng , 3 (2), 147 161. doi:10.5502/ijw.v3i2.2 Rao, M. A., & Donaldson, S. I. (2015). Expanding opportunities for diversity in positive psychology: An examination of gender, race, and ethnicity. Canadian Psychology/Psychologie C anadienne , 56 (3), 271. Roffey, S. (2015). Becoming an agent of change for school and student well being. Educational & Child Psychology , 32 (1), 21 30. Roffey, S. (2016). Building a case for whole child, whole school well being in challenging contexts. Educational & Child Psychology , 33 (2), 31. Sanderse, W., Walker, D. I., & Jones, C. (2015). Developing the whole child in an age of academic measurement: Can this be done according to UK teachers? Teaching and Teacher Education , 47 , 195 203. Seligman, M. E. (2012). Flourish: A vision ary new understanding of happiness and well being . Simon and Schuster. Seligman, M. E., & Csikszentmihalyi, M. (2000). Positive psychology. An introduction. American P sychologist , 55 (1), 5 14. Shonkoff, J. P., Garner, A. S., Siegel, B. S., Dobbins, M . I., Earls, M. F., McGuinn, L., ... & Committee on Early Childhood, Adoption, and Dependent Care. (2012). The lifelong effects of early childhood adversity and toxic stress. Pediatrics , 129 (1), e232 e246. U.S. Public Health Service. (2000). Report of th Mental Health . Washington, DC: Department of Health and Human Services. Vella Brodrick, D. A., Rickard, N. S., Hattie, J., Cross, D., & Chin, T C. (2015). An Evaluation of Year 10 Positive Education at Geelon g Grammar School: Findings From 2014 . The University of Melbourne, VIC, Australia. Waterman, A. S. (2013). The humanistic psychology positive psychology divide: Contrasts in philosophical foundations. American Psychologist , 68 , 124 133. Weare, K., & Ni nd, M. (2011). Mental health promotion and problem prevention in schools: what does the evidence say?. Health P romotion I nternational , 26 (suppl 1), i29 i69.

PAGE 43

38 Weiss, L. A., Westerhof, G. J., & Bohlmeijer, E. T. (2016). Can we increase psychological well b eing? The effects of interventions on psychological well being: A meta analysis of randomized controlled trials. PloS one , 11 (6), e0158092. Yeung, R., & Leadbeater, B. (2010). Adults make a difference: the protective effects of parent adolescents. Journal of Community Psychology , 38 (1), 80 98. Yoshikawa, H., Aber, J. L., & Beardslee, W. R. (2012). The effects of poverty on the mental, emotional, and behavioral health of children and youth: implications for prevention. American Psychologist , 67 (4), 272.