Citation
School-Wide Positive Behavioral Interventions and Supports

Material Information

Title:
School-Wide Positive Behavioral Interventions and Supports A Formative Evaluation of First Year Implementation
Creator:
White, Joshua Earl
Place of Publication:
[Gainesville, Fla.]
Florida
Publisher:
University of Florida
Publication Date:
Language:
english
Physical Description:
1 online resource (243 p.)

Thesis/Dissertation Information

Degree:
Doctorate ( Ed.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Educational Leadership
Human Development and Organizational Studies in Education
Committee Chair:
OLIVER,BERNARD
Committee Co-Chair:
ELDRIDGE,LINDA BURNEY
Committee Members:
OLIVER,EILEEN
CROCKETT,JEAN B

Subjects

Subjects / Keywords:
achievement -- behavior -- education -- pbis -- referrals -- reinforcement -- swpbis
Human Development and Organizational Studies in Education -- Dissertations, Academic -- UF
Genre:
bibliography ( marcgt )
theses ( marcgt )
government publication (state, provincial, terriorial, dependent) ( marcgt )
born-digital ( sobekcm )
Electronic Thesis or Dissertation
Educational Leadership thesis, Ed.D.

Notes

Abstract:
School Wide Positive Behavior Interventions and Support (SWPBIS) is a broad approach providing school personnel with tools and skills to enhance student academic achievement and improve socially relevant behavior (OSEP-TAC on PBIS, 2016). SWPBIS is a "technology" with four core features: (1) behavioral strategies; (2) integrated interventions; (3) commitment to long-term outcomes; and (4) system organizational to ensure sustained impacts (Dunlap, Sailor, Horner & Sugai, 2009). This research sought to explore the impact of SWPBIS interventions (particularly tangible reinforcements) from two baseline years to the first implementation year, as measured by the number of negative student behaviors (i.e., minor/pre-referrals and major office disciplinary referrals (ODRs)). The study adds to existing research, with an exploration of relationships between tangible reinforcements and referrals. The research was designed as a formative evaluation, focused on the design of an elementary school SWPBIS model. Archival data on 838 students were collected over three years. Data were collapsed for each year, with limitations to monthly regression analyses discussed. Statistical methods included Chi Squares, critical ratio tests, ANOVAs, and MANOVAs. The school experienced a 20.17% increase in enrollment, 23.75% decrease in total referrals, 20.05% decrease in minor/pre-referrals, and 45.15% decrease in proportion of major ODRs. There was a significant decrease in the proportion of referrals from the second baseline year to the intervention year. The school realized significantly lower proportions of students receiving minor referrals and ODRs in the intervention year. African American / Black students had a significantly higher average number of major ODRs than Caucasian American / White students in both baseline years, but such differences did not persist into the intervention year. The number of tangible reinforcements did not predict and was not significantly related to the number of negative student behaviors measured by minor referrals and ODRs. The method and results of this research should help inform practice and guide future research in the early implementation phases of SWPBIS models. Overall, this study supported the importance of empowering teachers and staff to improve the quality of the educational learning environment through a system of proactive, comprehensive, School-Wide Positive Behavioral Interventions and Supports. ( en )
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis:
Thesis (Ed.D.)--University of Florida, 2017.
Local:
Adviser: OLIVER,BERNARD.
Local:
Co-adviser: ELDRIDGE,LINDA BURNEY.
Statement of Responsibility:
by Joshua Earl White.

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Classification:
LD1780 2017 ( lcc )

Downloads

This item has the following downloads:


Full Text

PAGE 1

SCHOOL WIDE POSITIVE BEHAVIORAL INTERVENTIONS AND SUPPORTS: A FORMATIVE EVALUATI ON OF FIRST YEAR IMPLEMENTATION By JOSHUA E. P. WHITE A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF EDUCATION UNIVERSITY OF FLORIDA 2017

PAGE 2

2017 Joshua E.P. White

PAGE 3

I am forever indebted to my first teacher my mother, Sandra Marie White Pitts. I am forever grateful for your unconditional love and support.

PAGE 4

4 ACKNOWLEDGMENTS I would first like to thank the members of my doctoral committee Dr. Bernard Oliver, Dr. Linda Eldridge, Dr. Eileen Oliver, and Dr. Jean Crockett who were so gracious and willing to serve on my committee and give of themselves through their time, guidance, constructive criticism, and expertise. I will never forget their willingness to support and guide me through this stressful and arduous process to attain my academic goals. I would also like to give a very special thank you to Dr. Charles E. Byrd for his unconditional mentorship, guidance, expertise, and strong encouragement during the dissertation writing process. For without him, I would not have made it to the end of this journey. Finally, I would like to express my deepest appreciation to my dear moth er, Sandra Marie White Pitts. Her unwavering support throughout my life and academic endeavors, personal sacrifice to be the best mother she could be to me and my siblings, and instilling a value of education and the pursuit of knowledge has forever given me the keys to success.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 8 DEFINITION OF TERMS ................................ ................................ .............................. 10 ABSTRACT ................................ ................................ ................................ ................... 12 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 14 Statement of the Problem ................................ ................................ ....................... 14 Purpose and Significance ................................ ................................ ....................... 21 Evidence Justifying the Research Problem ................................ ............................. 21 Deficiencies in Evidence ................................ ................................ ......................... 24 Rese arch Questions ................................ ................................ ............................... 25 Research Question 1 ................................ ................................ ........................ 26 Research Question 2 ................................ ................................ ........................ 26 Research Question 3 ................................ ................................ ........................ 26 Research Question 4 ................................ ................................ ........................ 26 2 REVIEW OF LITERATURE ................................ ................................ .................... 27 Overview ................................ ................................ ................................ ................. 27 Behavioral Psychology: The Foundations of Positive Behavioral Supports ............ 29 Classical Conditioning ................................ ................................ ...................... 29 Operant Conditioning ................................ ................................ ....................... 31 Social Learning Theory ................................ ................................ ..................... 35 Modeling ................................ ................................ ................................ ........... 36 Vicarious Reinforcement ................................ ................................ .................. 40 From Behaviorism to Positive Behavioral Interventions and Support ..................... 43 Token Economies (Systematic Rein forcement) ................................ ................ 44 Immediacy and Satiation ................................ ................................ .................. 46 Continuous versus Intermittent Reinforcement ................................ ................. 48 Token Economies Within PBIS ................................ ................................ ......... 49 Aversive Stimuli and Zero Tolerance ................................ ............................... 50 Applied Behavioral Analysis (ABA) ................................ ................................ ... 54 Sociopolitical Roots of PBIS as a School Wide Intervention Model ........................ 61 Roots in Teaching Students with Disabilities ................................ .................... 61 Laws, Rules and Regulations ................................ ................................ ........... 64 Student Inclusion ................................ ................................ .............................. 66

PAGE 6

6 Positive Behavior Interventions and Supports (PBIS): From Theory to Practice ..... 67 Positive Behavior Interventions and Supports (PBIS) ................................ ...... 68 Multi Tiered Systems of Support (MTSS) ................................ ......................... 69 Functional Behavior Assessment ( FBA) ................................ ........................... 70 Design Considerations for SWPBIS ................................ ................................ 73 Design considerations: staff buy in ................................ ............................ 73 Design considerations: data collection and analysis ................................ .. 76 Design considerations: setting goals and expectations .............................. 81 Design considerations: staff training ................................ .......................... 84 Multi Tiered System of Supports within the Context SWPBIS .......................... 87 Tier 1 (primary universal) interventions ................................ .................... 88 Tier 2 (secondary targeted) interventions ................................ ................ 91 Tier 3 (tertiary intensive) interventions ................................ .................... 93 Impacts and Outcomes of SWPBIS ................................ ................................ ........ 93 Learning Environment and Behavior ................................ ................................ 94 Impacting Behavior with SWPBIS Interventions ................................ ............... 98 Impacting Academic Achievement with SWPBIS Interventions ...................... 107 Summary ................................ ................................ ................................ .............. 111 3 METHODOLOGY ................................ ................................ ................................ 115 Purpose of Study ................................ ................................ ................................ .. 115 Research D esign ................................ ................................ ................................ .. 115 Participants ................................ ................................ ................................ ........... 119 Target Population and Setting ................................ ................................ ........ 119 Participants and Descriptive Statistics ................................ ............................ 120 The School Wide Intervention Model ................................ ................................ .... 132 Development of the Intervention ................................ ................................ ..... 132 Intervention Design ................................ ................................ ........................ 135 Model Implementation Support and Training ................................ .................. 14 3 Measures and Metrics ................................ ................................ ........................... 149 Office Discipline Referrals (ODRs; Major Referrals) ................................ ....... 149 Minor Referrals (Pre Referrals) ................................ ................................ ...... 151 Tangible Reinforcements (Lion Loot) ................................ ............................. 152 Other Data of Interest ................................ ................................ ..................... 155 Proc edures ................................ ................................ ................................ ........... 155 4 RESULTS ................................ ................................ ................................ ............. 157 Section One: Descriptive and Preliminary Analysis (Independent Variable) ......... 158 Section Two: Descriptive and Preliminary Analysis (Dependent Variable) ........... 166 Section Three: Selection of Analyses ................................ ................................ ... 176 Section Four: Research Question One and Two ................................ .................. 180 Section Five: Research Questions Three and Four ................................ .............. 193

PAGE 7

7 5 DISCUSSION ................................ ................................ ................................ ....... 198 Synopsis of Findings ................................ ................................ ............................. 200 Implic ations ................................ ................................ ................................ ........... 205 Referrals and Negative Student Behaviors ................................ ..................... 206 Ethnic Disproportionality of Referrals ................................ ............................. 208 Impact of Tangible Reinforcements (Lion Loot) ................................ .............. 211 Classroom Learning and Vicarious Learning ................................ .................. 213 Research Methodology ................................ ................................ ................... 214 Limitations and Future Research ................................ ................................ .......... 215 APPENDIX: ADDITIONAL RESULTS TABLES ................................ ......................... 220 LIST OF REFERENCES ................................ ................................ ............................. 225 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 243

PAGE 8

8 LIST OF TABLES Table page 3 1 Distribution of ethnicity by year (students with disciplinary actions) .................. 122 3 2 Distribution of ethnicity by year (faculty / staff) ................................ ................. 123 3 3 Distribution of gender by year (students with disciplinary actions) .................... 124 3 4 Distribution of ge nder by year (faculty / staff) ................................ ................... 125 3 5 Distribution of grade level by year (students w/ discipline actions) ................... 127 3 6 Free and reduced price lunch (all students) ................................ ..................... 128 3 7 Free and reduced price lunch (students w/ discipline actions) .......................... 129 3 8 Distribution of special needs by year (students w/ discipline actions) ............... 130 3 9 Years of experience in education (faculty / staff) ................................ .............. 131 4 1 Students receiving tangible reinforcement by grade (lion loot) ......................... 160 4 2 Number of tangible reinforcements by grade level (lion loot) ............................ 162 4 3 Descriptives for independent variable by ethnicity and gender ......................... 163 4 4 Number of tangible reinforcements by month (lion loot) ................................ ... 166 4 5 Descriptive statistics f or dependent variables (discipline) ................................ 167 4 6 Minor and major referrals by school grade ................................ ....................... 171 4 7 Minor and major referrals by student gender ................................ .................... 172 4 8 Minor and major referrals by student ethnicity / race ................................ ........ 174 4 9 Distribution of minor and major referrals by month ................................ ........... 175 4 10 ANOVA table (major office disciplinary referrals) ................................ .............. 183 4 11 MANOVA table (minor, major, and total referrals) ................................ ............ 185 4 12 Proportionality of referrals given by student enrollment ................................ .... 191 4 13 Multiple regression ANOVA table (research question 3) ................................ .. 194 4 14 Correlation matrix: Lion Loot and negative student behaviors .......................... 195

PAGE 9

9 4 15 ANOVA table (Lion Loot reinforcements) ................................ ......................... 196 A 1 Administration of independent variable (Lion Loot) by staff member ................ 221 A 2 Proportionality of reinforcements by ethnicity and grade ................................ .. 223

PAGE 10

10 DEFINITION OF TERMS School Wide Positive Behavior Interventions and Support s Is a systems approach (three tiers of interventions) for establishing the social culture and individualized behavior supports needed for a school to be a safe and effective learning environment for all students (Sailor, Dunlap, Sugai, & Horner, 2009). It is a proactive approach that requires teachers and oth er sch ool staff are trained in using School Wide Positive Behavior Intervention and Supports (SWPBIS), while school rules and expectations are universal, s chool wide, and visible throughout the campus I t is data driven and adjusts based on the behavioral needs that are identified by the SWPBIS Leadership Team, which meets regularly. Tangible Reinforcements school proposed to be studied in this research that are given to recogni school for the positive behavior tokens provided to students. These tangible reinforcements are given by all school staff members for students exhibiting positive behaviors, generally around other students. These are provided for immediate reinforcement of the individual student, as well as immediate vicarious reinforcement of the other students in the class. Such vicarious reinforcement is an i mportant concept, but is not easily measured. As such, the first two research questions do not consider the number of tangible reinforcements provided, as all students may have benefited from vicarious reinforcement. Regardless, as a crude measurement of a dherence to the reinforcement paradigm, the number of Lion Loot given to students is the best measurement available for this archival study. The assumption is that teachers who provide Lion Loot also provide the intangible reinforcements and the other comp onents of the school wide positive behavior interventions and supports model. Intangible Reinforcements In addition to tangible tokens, teachers and staff were trained in immediate method of reinforcing positive behaviors in front of other students, while not relaying on the tangible (and less powerful) reinforcements. Intangible reinforcements included: verbal praise, non verbal praise, and positive gestures (i.e., cheers, thumbs up, high fi ves, pats on the back, smiling at students, etc.). Major Referrals office negative behavior al school record.

PAGE 11

11 Minor Referrals ODR (major) referral but do resul t in some form of formal docum entations, such as a note home or intervention visit from or Response to Intervention for Behavior Database (RtI:B) A supplemental database available to schools in the State of Fl orida participating in SWPB I S. It allows school administrators in charge of behavior to input both major and minor referrals data and run various reports that allow for data driven decision making with SWPBIS Leadership Teams at the school.

PAGE 12

12 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Education SCHOOL WIDE POSITIVE BEHAVIORAL INTERVENTIONS AND SUPPORTS: A FORMATIVE EVALUAT ION OF FIRST YEAR IMPLEMENTATION By Joshua E. P. White December 2017 Chair: Bernard Oliver Co C hair: Linda Eldridge Major: Educational Leadership School Wide Positive Behavior Interventions and Support (SWPBIS) is a broad approach providing school personnel with tools and skills to enhance student academic achievement and improve socially relevant behavior ( OSEP TAC on PBIS, 2016) SWPBIS is with four core features: (1) be havioral strategies; (2) integrated interventions ; (3) commitm ent to long term outcomes; and (4) system organizational to ensure sustained impacts (Dunlap, Sailor, Horner & Sugai, 2009) This research sought to explore the impact of SWPB I S intervention s (particularly tangible reinforcements) from two baseline years to the first implementation year, as measured by the number of negative student behaviors (i.e., minor/pre referrals and major office disciplinary referrals (ODRs)). The study adds to existing research, with an exploration of relation ships between tangib le reinfor cements and referrals. The research was designed as a formative evaluation, focused on the design of an elementary school SWPBIS model. Archival data on 838 students were collected over three years. Da ta were collapsed for each year, with limit ations to monthly regression

PAGE 13

13 analyses discussed. Statistical methods included Chi Squares, critical ratio tests, ANOVAs, and MANOVAs. The school experienced a 20.17% increase in enrollment, 23.75 % decrease in total referrals, 20.05% decreas e in minor/pre referrals, and 45.15% decrease in proportion of major ODRs There was a significant decrease in the proportion of referrals from the second baseline year to the intervention year Th e school realized significantly lower proportion s of students receiving minor referrals and ODRs in the intervention year. African American / Black students had a significantly higher average number of major ODRs than Caucasian American / White students in both baseline years, but such differences did not persist into the int ervention year The number of tangible reinforcements did not predict and was not significantly related to the number of negative student behaviors measured by minor referrals and ODRs. T he method and results of this research should help inform practice and guide future research in the early implementation phases of SWPBIS models. Overall, this study supported the importance of empowering teachers and staff to improve the quality of the educational learning environment through a system of proactive, comp rehensive, School Wide Positive Behavioral Interventions and Supports

PAGE 14

14 CHAPTER 1 INTRODUCTION Statement of the Problem Over the past several decades, American schools have been plagued with a myriad of ever increasing forms of negative and challenging student behaviors (e.g., Carr, Taylor, & Robinson, 1991; Farmer, Quinn, Hussey, & Holohan, 2001; Greenwood, Delquadri, & Hall, 1984; Gunter, et al., 1994; Kazdin, 1987; Wehby, Symons, & Shores, 1995; Walker, Col vin, & Ramsey, 1995) Disrespect toward school personnel, gross insubordination, classroom disruptions, and fighting top the list, with more alarming and serious forms of violence, such as bullying and school shootings, being of major concern to parents, educators, and the general public (e.g., Glew, Fan, Katon, Rivara, & Kernic, 2005; Leary, Kowalski, Smith, & Phillips, 2003) Images in the media of school shootings such as those at Pearl High Sch ool in Mississippi in 1997 Col umbine High School in Co lorado in 1999 and Sandy Hook Element ary School in Connecticut in 2012 demonstrate the pain, suffering, and negative impact that violent student behavior can have on students, families, and igation to provide a safe place for student learning From 1999 to early 2016, ABC News reported that America has witnessed 141 deaths at schools during mass murders or attempted mass murders, with 17 of the murder er s being 15 years of age or younger (Pea rle, 2016) Increasing f ear and co ncern are even felt by students outside of these communities, with results from the National Crime Victimization Survey revealing that 3% of students reported fear of being attacked or harmed at school, while 4.7% of stud ents reported actively avoiding

PAGE 15

15 school activities or classes because they were fearful someone might attack them or harm them (Zhang, Musu Gillette, & Oudekerk, 2016) In 2012, Scholastic and the Bill & Melinda Gates Foundation released a comprehensive a nd detailed report entitled the Teaching Profession This report was based on surveys received from 10,000 educators representing elementary, middle, and high schools from all 50 states The report indicated that 62% of teachers who had been teaching in the same school for at least five years believed that problematic student behavior has worsened in recent years and has resulted in deleterious effects on both teaching and learning Even more concerning was that behavior concerns of teachers were not limited to any particular group of students, and spanned both high and low socioeconomic statuses (Bill & Melinda Gates Foundation, 2012). Moreover, according to a national poll of 1,350 veteran elementary school prin cipals conducted by the National Association of Elementary School Principals in 1997, 80% of principals indicated that ptive NAESP, 1997 p. 19 ) and 81% of teachers polled stated that th eir worst behaved students are a barrier to effective education in their classrooms for all students (Public Agenda, 2004) Indeed, t here is a significant amount of lost instructional time during each instance a classroom is disrupted by inappropriate stu dent behavior both of the students behaving inappropriately and their peers in the classroom (e.g., e.g., Godwin, Almeda, Petroccia, Baker, & Fisher, 2013; Karweit & Slavin, 1981; Lee, Kelly, & Nyre, 1999; Lemov, 2010) The impact for the offending stud ent is even greater if they are sent out of class (e.g., time out in the front office),

PAGE 16

16 excluded from the classroom for an extended period of time (e.g., in school suspension), or excluded from the school for a period of days (e.g., out of school suspensio n, expulsion) Ultimately, t he combined effect of inappropriate student behavior in the classroom can result in student underachievement and overall loss of Indeed, nearly every research study exploring cl assroom behavior management and school culture begins with the same basic tenet poor student behavior distracts from and negatively impacts the learning environment (e.g., Guardino & Fullerton, 2010; McKevitt, Dempsey, Ternus, & Shriver, 2012; Scott & Ba rrett, 2004; Sugai & Horner, 2006) Certainly, it is the current belief that all students deserve an effective learning environment (e.g., USED, 2015), but this was not always the case Prior to the 1980s, students identified with disabilities were ofte n placed in hospitals or alternative schools to receive minimal education, as per the philosophy at the time (Torrey, 1997) However, parents of these students felt their children were not getting the same education as their n on disabled peers in the publ ic school system, and felt they also deserved an equal chance at effective learning environments (e.g., Blankenship, Boon, & Fore, 2007) This brought upon several lawsuits, such as Board of Education of the Hendrick Hudson Central School District v Rowl ey (1982) ; Mills v Board of Education of District of Columbia (1972); and Honig v Doe (1988) Ultimately, courts often ruled in favor of the students with disabilities and upheld the idea that they deserved a public education and protection from being unnecessarily punished or removed from school Along with these court cases, the United States Congress supported the same notion and passed such laws as the Rehabilitation Act of 1973 (amended in 1986 and

PAGE 17

17 1992), the Individuals with Disabilities in Education Act of 1990 (with significant amendments in 1997), and the American with Disabilities Act of 1990 These laws ultimately required all school systems to provide children identified with disabilities a Free and Appropriate Public Education (FAPE) in the least restrictive environment (USED, 2010; 34 C.F.R §300.130) possible (i.e., inclusion) (Blankenship, et al. 2007) The practice of inclusion has proven beneficial for many students with disabilities (e.g., Peetsma, Vergeer, Roeleveld, & Karsten, 2001; Powell, 2015), but brought about significant challenges with students not used to having peers with disabilit ies and teachers not trained in how to best manage their manifest behaviors (Dunlap, Sailor, Horner, & Sugai 2009) Ultimately, this quickly led to the need for effective and efficient behavior management programming for students with disabilities within the regular classroom environment, primarily to maintain the effective learning environment and maximize student achievement (Dunlap, et al. 2009; Sugai & Horner, 2006) Given the need for immediate action, one of the first efforts was to incorporate t he same theory of behavior management and modification utilized with inpatient children (and adults) with severe disabilities Applied Behavior Analysis (ABA) ABA had been shown effective at suppressing the most severe behaviors within these controlled settings, so it was applied to students with disabilities in the regular classrooms (Dunlap, et al. 2009) For the most part, studies suggest that the use of ABA had some impact on these students by reducing the instances of their manifest behaviors (e.g ., Dunlap, et al. 2009) Unfortunately, many of the seemingly effective

PAGE 18

18 methods brought into the educational settings were based on aversive stimuli and punishment driven actions, such as use of electrical shock, sprays of water, and other physical and e motional punishments (e.g., Butterfield, 1985; Linscheid, Iwata, Ricketts, Williams, & Griffin, 1990) The public outcry and moral objections of many educators resulted in the need for a more positive alternative to these barbaric methods of behavior mana gement (Singer & Wang, 2009) In addition to the desire to move away from aversive stimuli and punishment driven methods, there was also a drastic need to be more culturally sensitive and culturally aware Since the landmark United States Supreme Court ruling in Brown vs Board of Education of Topeka (347 U.S 483) in 1954, which ended legal segregation in public schools, minority students have found a system replete with ongoing bias, as well as overt and covert racism (NAACP LDF, 2011) These cultur ally insensitive practices were only exacerbated by the zero tolerance policies of the 1990s (Skiba, 2000) Not only have zero tolerance policies been shown to exacerbate poor behavior (Poe Yamagata & Jones, 2000; Wald & Losen, 2003), but the American Psy chological zero tolerance is an entirely ineffective system (Skiba et al., 2006). In part stemming from such zero tolerance policies and continued insensitivity, Wald and Losen (2003) found that Black students made up 17% of the national student population, but accounted for 34% of those students who had been suspended from school and were 2.6 times more likely to be suspended as White students Moreover, according to the Office of Civil Rights (U.S Department of Education, 2014), African American students without disabilities are more than three times as likely as their white

PAGE 19

19 peers without disabilities to be expelled or suspended from school Furthermore, over 50% of students who were involved in sc hool related arrests or referred to law enforcement are Hispanic or African American (U.S Department of Education, 2014) It is data like these that supports the unfortunate and concerning cycle known as the to kiba, 2000) Such racial disparities in application of behavior management processes and discipline were recognized by the U.S Department of Education (USED) In January of 2014, the USED Office for Civil Rights and the U.S il Rights Division issue d a joint letter to serve as a significant guidance document. The guidance ( USED, 2 014, p. 2), but to also help students succeed through programs that support and reinforce positive student behavior and character development This guidance served to underscore the decades of research and success that went into creating a more positive m ethod to teach students more positive social behaviors that support the overall learning environment and their academic achievement The three aforementioned primary issues led to the development of School Wide Positive Behavior Interventions and Supports (SWPBIS): (1) need to manage behavior of all students in an inclusive environment to maximize the learning environment; (2) moral objections to aversive stimuli and punishment driven methods of control; and (3) need for more culturally sensitive a nd understanding methods to support positive behavior and character Based on a complex web of prior theories and research (discussed in Chapter 2), SWPBIS is a broad approach designed to provide school

PAGE 20

20 personnel with tools and skills necessary to enhance student academic achievement SWPBIS is designed to allow individual schools flexibility to provide a wide variety of evidence based and research based interventions that b est meet the needs of their students Some of the most prominent researchers in PBIS indicate that SWPBIS has four core features: (1) an application of research based behavioral strategies; (2) use of multiple and integrated intervention elements aligned to the needs of the environment; (3) a commitment to sustained and long term outcomes; and (4) support within the organizational systems that ensure sustained impacts of the interventions (Dunlap, et al. 2009) In essence, SWPBIS makes use primarily of p ositive reinforcement of appropriate behaviors, focuses on teaching productive social behaviors (not just suppressing negative behaviors), and is designed with a clear focus on being culturally sensitive ( Sugai & Horner, 2009a) SW PBIS makes use of a thr ee tiered system of supports: (1) Primary, Tier 1 whole school interventions impacting 80 90% of the student population; (2) Secondary, Tier 2 small group interventions impacting 15% of the student population who do not respond to primary tier interven tions; and (3) Tertiary, Tier 3 individual interventions impacting the remaining 5% of the student population who do not respond to primary and secondary tier interventions (McIntosh & Goodman, 2016) Tier 1 interventions are relatively broad and focuse d on several pre established school wide goals, with both tangible and non tangible incentives provided to students to recognize appropriate behavior This research focuses on Tier 1 interventions, with all three tiers discussed in the following literatur e review This study seeks to examine how the implementation of

PAGE 21

21 a Sc hool W ide Positive Behavior Interventions and Support (SWPB I S) approach can help to reduce incidents of negative and problemat ic student behavior which should theoretically in turn incr ease learning time for all students. Purpose and Significance The present research is essentially a formative evaluation study to investigate the initial, first year impacts of implementing a comprehensive School Wide Positive Behavioral Interventions and Support (SWPBIS) model at an elementary school serving students in kindergarten through 5th grade Impacts and outcomes measured within the current study included both minor /pre referrals and major office disciplinary referrals (ODRs), as these metrics ar e most commonly associated with displays of inappropriate and disruptive student behaviors resulting in a deleterious impact on the learning environment The study adds to an already growing body of research on the utilization of SWPB I S models and techniq ues, with a somewhat unique exploration of how the provision of tangible positive reinforcements under a token economy relates to the instances of behavioral referrals In addition, the study has the enhanced purpose to explore potential considerations fo r school personnel regarding demographic characteristics, cultural differences, and racial biases when reinforcing desired and modeled student behavior using tangible incentives under the SWPBIS approach Evidence Justifying the Research Problem With ina ppropriate and distracting student behavior significantly reducing the effectiveness of the learning environment (e.g., Guardino & Fullerton, 2010; McKevitt, et al. 2012; Scott & Barrett, 2004; Sugai & Horner, 2006), the importance of finding positive and proactive methods to help students learn and apply more appropriate behaviors is of great importance throughout the field of education Such negative

PAGE 22

22 impacts are often exponentially worse for stude nts from traditionally defined minority groups, as they a re often given more disciplinary referrals and punishments that remove them from the learning environment for extended periods of time (e.g., Blanchett, 2014; Gonsoulin, Zablocki, & Leone, 2012; Skiba, Arrendondo, & Rausch, 2014; USED, 2014) Fortunately, a great deal of research has shown the positive impacts of School Wide Positive Behavioral Interventions and Supports (SWPBIS) on the behavior of students attending schools (e.g., Fairbanks, Sugai, Guardino, & Lathrop, 2007; George, White, & Schlaffer, 20 03; Nocera, Whitbread, & Nocera 2014 ; Scott & Barrett, 2004; Sugai & Horner, 2006; Waadsorp, Bradshaw, & Leaf, 2012) Preventative and proactive behavior management techniques used to teach positive behaviors and stop the emergence of inappropriate behav ior patterns have long been found effective, though sometimes difficult to implement (e.g., Walker & Shinn, 2002) SWPBIS was specifically developed to be proactive, positive focused, and relatively easy to implement in a consistent and effective manner Ultimately, SWPBIS has been repeatedly and consistently found effective in reducing the instances of negative student behaviors across all student sub groups, communities, and grade levels (e.g., Fairbanks, et al. 2007; George, et al., 2003 ; Nocera, et al. 2014; Scott & Barrett, 2004; Waadsorp, et al., 2012 ) Scott and Barrett (2004) found significant decreases in ODRs and suspensions from baseline to the first year and second year of SWPBIS implementation, resulting in over 12,000 minutes of gained in structional time with students George, White, and Schlaffer (2003) found a simila r decrease in ODRs from baseline to the second year of imple menting a SWPBIS model, with an even more pronounced decrease among students with IEPs Spencer (2015) found a s ignificant decrease in the number of ODRs within one year of

PAGE 23

23 implementing SWPBIS, with decreases in each of eight primary offense categories indicated on ODRs (e.g., disrespect, refusal to obey, physical contact, profanity, etc.) Menendez, Payne, and May ton (2008) also conducted a one year study of a new SWPBIS project and found a reduced number of ODRs, fewer rules based violations committed of students, and fewer punitive consequences used by teachers (e.g., time outs, written reprimands) Luiselli, Pu tnam, Handler, and Feinberg (2005) found that SWPBIS resulted in fewer ODRs and school suspensions over the course of two years Vincent, Swain Bradway, Tobin and May (2011) conducted a study with 153 elementary schools and found schools implementing high fidelity SWPBIS had significantly fewer rates of ODRs across all three years of the research, with the disproportionality of referrals between minority and non minority students significantly lower in schools implementing SWPBIS. As noted, ODRs were selec ted as the primary metric for the present study and ar e most often the primary metric for any study exploring the behavioral impacts of SWPBIS on students Indeed, only a handful of behavioral metrics are traditionally used to assess impact on the learnin g environment: in school suspensions, out of school suspensions, expulsion, detentions, office disciplinary referrals (ODRs), school attendance, and tardiness However, not all of these are appropriate for elementary schools, as some are rare in elementar y school (e.g., expulsions and detention) while others are generally out of and elementary tardiness) As such, ODRs are the most common method for assessing conti nuous behavioral change within elementary s chools im plementing school wide preventative interventions such as that provided by SWPBIS.

PAGE 24

24 ODRs have also been shown to be related to future behavior problems, such as drug use and disorderly conduct in classrooms (e.g., Nelson & Roberts, 2000; Sprague et al., 20 01) Pas, Bradshaw and Mitchell (2011) utilized data from 21 elementary schools regarding ODR data from a centralized database, ODR reports from a teacher survey, and scores from the Teacher Observation of Classroom Adaptation Checklist (TOCA C) The stu dy found ODRs to be positively correlated with both the disruptive behavior and attention problems subscales, as well as negatively correlated with the prosocial behavior subscale of the TOCA C Overall, the researchers determined that ODRs have moderate convergent and divergent validity with behavior ratings on the TOCA C ( Pas, et al., 2011 ) Other researchers have also found validity for using ODRs as a measure of general behavior (e.g., Irvin, Tobin, Sprague, Sugai, & Vincent, 2004; Scott & Barrett, 20 04), with ODRs being related to general misbehavior at school, school attendance, student and teacher perceptions of safety and victimization, classroom orderliness, juvenile delinquency, and behavior disorders Deficiencies in Evidence While the develop ment and application of SWPBIS has been largely shown to significantly reduce the incidents of major ODRs, suspensions, bullying, and other disruptive student behaviors, there are some gaps within existing research that are addr essed within the current stu dy First, nearly every research study exploring ODRs focused on the use of centralized databases to collect archival data about student referrals, which can introduce error in terms of data entry and unwritten policies (Nelson, Benner, Reid, Eptsein, & C urrin, 2002). Instead, the present research uses internally collected data on all behavior instances at the sc hool studied recorded immediately upon the receipt of the minor/pre referral or ODR by the school

PAGE 25

25 administrator responsible for behavior Second referrals (pre referrals) that are not recorded in such centralized databases of student (ODRs) referrals Third, no studies were found tha t explored the potential impact of the token economy and tangible reinforcement component of the overall SWPBIS model, with nearly every study in current literature exploring the impact of SWPBIS without focusing on any specific elements of the process H owever, the present research specifically compares the level of tangible reinforcements provided and the behavioral impacts mentioned above Finally, few studies have included differentiated analysis by race and ethnicity due to complications with the rec ording of such data in centralized databases, while the present research collected all data connected to the individual demographics of each student Overall, the method of addressing these primary gaps and the results of the present research should help inform practice and guide future research in the early implementation of SWPBIS models Research Questions The overarching goal of this study is to e xplore whether a comprehensive SWPBIS model will impact the instances of negative student behaviors among elementary school students within the first year of implementation The secondary goal of this research is to explore potential differences among student sub groups to drive consideration of demographic characteristics, cultural differences, and racial bi ases when reinforcing and teaching student behavior under the SWPBIS approach The tertiary goal is to determine whether the level of tangible reinforcements, as a component of the overall SWPBIS model implemented for this study, was related to any change s in the number of negative student behaviors during the intervention year.

PAGE 26

26 Although a more detailed description of the study design is provided in the third chapter, the research questions that are assessed in the present research are set forth below : Re search Question 1 Regardless of race/ethnicity and student grade level, will there be a significant impact of the SWPB I S intervention from baseline years (pre intervention) and the intervention year (post intervention), as measured by the number of instanc es of negative student behavior by month and by year (i.e., minor referrals and major referrals)? Research Question 2 Is there a difference by race/ethnicity and/or student grade level in the impact of the SWPB I S intervention from baseline to implementatio n on the number of negative student behaviors? Research Question 3 Regardless of race/ethnicity and student grade level, is there a significant relationship between the level of implementation of a SWPB I S initiative (i.e., number of tangible reinforcements provided) and the instances of negative student behavior (i.e., minor refer rals and major referrals [ODR])? Research Question 4 Is there a difference in the impact a SWPB I S intervention (i.e., number of reinforcements) and instances of negative school behavior (i.e., number of minor and major referrals by race/ethnicity and/or grade level?

PAGE 27

27 CHAPTER 2 REVIEW OF LITERATURE Overview Philosophers and teachers have long attempted to understand how best to raging hands on learning within The For thousands of years, the field of education has continued to develop new theorie s and techniques to educate children, while trying to keep with increasing societal demands for enhanced education outcomes Unfortunately, educators have constantly faced the seemingly insurmountable challenge of managing student and classroom behavior t o create the most conducive environment for successful education interventions There have been several attempts with questionable benefits, such as the often criticized school to prison pipeline created by arguably over reaching zero tolerance policies (e.g., Heitzig, 2009; NAACP LDF, 2011) The critiques of past efforts have led to new models for improving student behaviors, such as the creation of Single School Cultures, character e ducation techniques, and School Wide Positive Behavioral Interventions and Supports the topic of this dissertation In attempts to support the overall educational system and address some of the adversities created by other reform efforts, the United States Congress updated the Individuals with Disabilities Education Act in 1997 (IDEA, 1997) to include language specific to the adoption and implementation of Positive Behavioral Interventions and Supports (PBIS) and classrooms, the use of PBIS has sinc e spread to regular education classrooms and

PAGE 28

28 school wide implementation (FLDOE, 2008) A wide array of acronyms is now used to describe this expanding initiative to improve student behavior PBIS, PBS, SWPBIS, SWPBS, and EBS Many of these acronyms are familiar to many educators and practitioners working with children, and each version encompasses the same general approach (FLDOE, 2005) Before focusing on the specific theory and practice of PBIS, it is important to review the theoretical underpinnings that form the foundation for many of the concepts contained within PBIS It is important for any conceptual model to understand and respect past theories, as they often contain specific concepts and nuances that serve to strengthen the understanding and a pplication of the theory of focus As such, this section starts with an overview of the most basic elements of the foundational theories, as well as aspects that are salient to the later discussion of PBIS It is nearly impossible to begin any discussio n of student behavior and education without first discussing the field of psychology The field of psychology as the study of human behavior and thoughts is widely considered to be as old as the field of education, with many psychological concepts datin g back to the ancient Greeks (e.g., Hergenhahn, 1992) However, the development of modern psychology is far more recent, with the first formal laboratory dedicated to psychological research starting in 1879 under Wilhelm Wundt Shortly after, psychology rapidly grew into a wide variety of sub fields, including G Gunn, & Duncan Johnson, 2006), John choanalysis (Fancher, 1996) Among these sub development of behavior modification and behavior management theories Behaviorism

PAGE 29

29 was developed on the shoulders of two primary theories: classical cond itioning and operant conditioning. Behavioral Psychology: The Foundations of Positive Behavioral Supports Classical Conditioning Among the earliest theories within the field of behaviorism was classical conditioning, which is concerned primarily with how new behaviors are learned (Kazdin, 2012a) The concept of classical conditioning is most often credited to the physiologist Ivan Pavlov, who happened upon the concept when conducting research on the salivary response of dogs Pavlov discovered that dogs would begin to salivate when the person who usually fed them would enter the room, without being presented any food result of food in the mouth, pupil constriction in b right light, and muscle flexion in response to pain) to studying how connections were made between various environmental stimuli (Kazdin, 2012a) This led to the well known sequence of studies where the basic facets of classical conditioning were develope d (Kazdin, 2012a; McLeod, 2013). In essence, there are four primary components to classical conditioning (Rescorla, 2014): (1) unconditioned stimulus (e.g., food); (2) unconditioned response (e.g., saliva); (3) conditioned stimulus (e.g., bell ringing); an d (4) conditioned response (e.g., saliva) It is noted that Pavlov argued the saliva produced in the conditioned response differed from that produced in the unconditioned response, so while the conditioned and unconditioned responses are often similar, th ey are not always the same (Kazdin, 2012a) Researchers went on to introduce a number of more complex conditions that strengthen or weaken the pairing of the conditioned stimulus and

PAGE 30

30 response, such as forward conditioning, simultaneous conditioning, highe r order conditioning, backward conditioning, zero contingency conditioning, extinction, stimulus generalization, stimulus discrimination, and many others (Kazdin, 2012a; Rescorla, 2014) if ication was his scientific and learning based explanation of behavior (Kazdin, 2012a), in addition to bringing forward the concept that antecedent actions and the environment can impact the outcome of behavior. In relation to theories and models associated with Positive Behavior Interventions and Supports (PBIS), several concepts from classical conditioning are important to understand how PBIS might work within the school setting (Kappel, Dufresne & Mayer, 2012) For instance, the concept of forward condit ioning (Kazdin, 2012a) might suggest that verbal praise should be presented first when a student behaves well, followed immediately with any tangible rewards (if assuming the tangible rewards produce a desired level of happiness) With this forward condit ioning, the goal would be for the verbal praise alone to eventually be sufficient for the student to feel the same level of happiness as with the tangible reward Second, the concept of classical extinction (Kazdin, 2012a) helps explain why positive behav ioral supports must be consistent across the entire school and maintained across all years while the student is enrolled More specifically, extinction is when the conditioned stimulus (e.g., verbal praise) is not provided for a period of time, such that it no longer produces the desired response process must begin anew (barring any spontaneous recovery, which is beyond the scope of this review) Finally, the concept of s timulus generalization (Kazdin, 2012a)

PAGE 31

31 explains when a teacher successfully pairs a conditioned stimulus (e.g., verbal praise from the teacher) with a desired response, and then finds another conditioned stimulus also elicits the response (e.g., verbal pra ise from another student) Many other concepts exist within the overall classical conditioning framework, but are outside the scope of this research, such as fear conditioning (Fanselow & Sterlace, 2014) Operant Conditioning While classical conditionin g provided some of the earliest experimental findings about how humans might learn new behaviors, some theorists felt it did not adequately explain acquisition of new behaviors particularly those that are not reflexive (Olson & Hergenhahn, 2015) As cla ssical conditioning continued to be researched and as the field of psychology became more robust, E.L Thorndike and B.F Skinner began researching and introducing new ideas about human behavior operant conditioning Thorndike was the first to research the concepts associated with operant conditioning, In essence, Thorndike (also focusi where they needed to do a simple task to escape (e.g., pull a lever) and found they would perform many failed behaviors before doing the necessary task to escape However, each time the cat was placed i nto the puzzle, the cat would have fewer failed behaviors and was quicker to perform the escape task Eventually, the cat would escape the box no sooner than it had been placed inside (Kazdin, 2012a; Keller & Schoenfeld, 2014) In essence, Thorndike explained that behaviors followed by pleasant outcomes are strengthened (more likely

PAGE 32

32 to be repeated), while those followe d by unpleasant outcomes are weakened (less likely to be repeated; Olson & Hergenhahn, 2015) As applied to humans, this became the basis for the theory of operant conditioning While Thorndike helped codify some of these concepts, the basics of operant conditioning are nothing new, and it is widely accepted that parents and caregivers have been using these techniques for thousands of years to teach children acceptable behaviors rewarding those deemed acceptable and punishing those deemed unacceptable (Miltenberger & Crosland, 2014) However, such simple explanations are inadequate to fully understand why some children learn acceptable behavior and some learn unacceptable behavior It became the lifework of B.F operant conditi the complex process of operant conditioning (Olson & Hergenhahn, 2015) While theories were easier to apply to human learning for a range of behaviors, and they were immediately applicable to a variety of fields (e.g., education, psychology, etc.) As such, operant conditioning finds a much larger place in the foundation of PBIS t heories and models (Kappel, Dufresne & Mayer, 2012) In general, operant conditioning explains that human behavior is learned through a system of rewards and punishments, both positive (additive) and negative (subtractive; Miltenberger, 2012) It differ s from classical conditioning, in part, because the behaviors being learned are not reflexive (e.g., salivation, muscle tension, and pupil constriction are all reflexive and hard for an individual to control) Indeed, operant conditioning is more focused on developing and modifying operant behaviors, many of

PAGE 33

33 which are volitional in nature (Miltenberger, 2012) There are two core mechanisms within operant conditioning: reinforcement and punishment In essence, reinforcement is used when the goal is to inc rease the occurrence of a targeted behavior, while punishment is used to decrease the occurrence of the targeted behavior (Miltenberger, 2012) is good or bad; Catania, 1979; Miltenberger, 2012) This creates four primary categories of consequences for the targeted behavior: (1) Positive Reinforcement is when the targeted behavior is rewarded with the addition of a desirable stimulus (e.g., a jelly bean, a pat on the back, etc.); (2) Negative Reinforcement is when the targeted behavior is followed by the subtraction or removal of an undesirable stimulus (e.g., ruler until they sit up straight stopping the tapping is negative reinforcement); (3) Positive Punishment is when the targeted behavior is followed by the addition of an aversive stimulus (e.g., spanking, slapping, yelling, referrals, detentions, suspens ions, grounding, etc.); and (4) Negative Punishment is when the targeted behavior is followed by the removal of a desirable stimulus (e.g., grounding from television, taking away video games, sitting in time out (removal of social stimulation), etc.) (McCo nnell, 2001; Miltenberger, 2012) With these four categories, researchers have been able to explore which is most effective and which work best together For instance, researchers and theorists in behavioral psychology and education have generally found entirely ineffective at bringing about behavioral change, serving only to suppress undesired behaviors in the presence of the punisher, but not in other situations (e.g.,

PAGE 34

34 Smith, 2012) In addition, an often overlooked and mi sunderstood aspect of behavior modification under operant conditioning is that positive reinforcement is matched with negative punishment, while negative reinforcement is matched with positive punishment (McConnell, 2001) This matching essentially explai ns how a single stimulus can be used as an onset or offset to the behavior (e.g., candy can be given (onset) as a positive reinforcement or removed (offset) as a negative punishment) With a focus on positive reinforcement, it is not surprising that the m ost efficient and effective method for behavior change occurs when positive reinforcement is paired with negative punishment or when negative reinforcement is paired with positive punishment as this reduces confusion in the child by associating a more li mited number of stimuli (Miltenberger, 2012) Although Positive Behavioral Interventions and Supports (PBIS) will be further discussed later in this literature review, it is important to mention the linkage between PBIS and operant conditioning beyond th at of Applied Behavioral Analysis (also discussed later) In general, PBIS uses the most widely accepted method of modifying and controlling the behavior of children through non aversive means a combination of positive reinforcement and negative punishme nt (Kappel, Dufresne & Mayer, 2012) Students are given tangible reinforcements and verbal praise for appropriate behaviors, while negative behaviors are largely ignored (thus providing no reinforcement, as attention itself can be reinforcing) When nec essary, desirable situations and stimuli are removed when behaviors are inappropriate, such as giving age appropriate time out or both examples of negative punishment When applied consistently across the entire school, such rewards and

PAGE 35

35 punishments have the greatest theoretical potential to illicit desirable changes in student behaviors (Putnam & Kincaid, 2015) Again, the linkage between operant conditioning and PBIS are further detailed in the following sections of this literature review. Social Learning Theory In addition to the aforementioned theories of conditioning, social learning theory (e.g., Bandura, 1971a) provides additional concepts necessary to understand some of the foundational interventio ns and components associated with models of positive behavioral support (e.g., Chomsky, 1959), in that neither classical nor operant conditioning could adequately explain how humans acq uire new behaviors that they have never performed before (novel behaviors) Chomsky (1959) was among the first to criticize the conditioning theories by focusing on language acquisition (which cannot be explained through pure conditioning theories), thoug h it was Albert Bandura that introduced some of the more instrumental insights into how humans learn behaviors that they would not otherwise perform More specifically, Bandura (1971a) argued that the traditional theories of behavioral learning focused on internal mechanisms (e.g., needs, wants, impulses) and ignored much of the influence from external factors and other people That said, in his early writings, Bandura (1971a) made it clear that social learning theory does not replace the impact of intern al factors on learning, only that human behavior is more complex and theories must also account for the influence of environmental and social factors For instance, specific to student behaviors in an educational setting, a child is unlikely to ever natu rally puff out their cheeks and cross their arms (bubbles and seatbelts) when walking in a line, so theories of conditioning would not easily explain

PAGE 36

36 different setting There must be external factors that provide the initial impetus for the student to perform the behavior, as well as some type of reinforcement to have the behavior continue Explanations of other classroom behaviors from a purely classical learning the ory framework could even be damaging to students, particularly in terms of student misbehavior For instance, a student who hits another student would be considered to have hostile internal impulses according to classical learning theories, while social l earning theory would allow more understanding of the student by considering external influences of the displayed behavior (Bandura, 1971a) While classical and operant conditioning might not adequately explain acquisition of novel behaviors, social learn ing theory provides some insight into possible mechanisms both in the learning and reinforcement of these novel behaviors in social settings Two primary topics from social learning theory helped develop the models of Positive Behavioral Interventions a nd Supports (PBIS): modeling and vicarious reinforcement In order to provide some understanding of these primary elements and their connection to PBIS, they are cursorily described within this literature review Modeling Modeling, or observational learning, was first introduced by Bandura following Bandura & Huston, 1961) to the eventual development of the mor e comprehensive Social Learning Theory The basic idea of modeling is that children are not always internally motivated by wants and needs to perform new behaviors, but are constantly surrounded by models that externally influence their behavior Devoid of any immediate reinforcements or

PAGE 37

37 actions In fact, Bandura (1977), found that children do not just observe and behave immediately, but they encode the modeled behaviors and can access these responses at a later time, thus providing longer term learning of behaviors Bandura (1977) found that children will encode modeled behaviors from a wide variety of sources, including other children, family members, teachers, other adults, film actors, and even cartoon characters on television (Bandura, 1963; Bandura, 1965) Some encoded behaviors are important for the development of the child (regardless of the source), such as learning pro social behaviors instead of anti social b ehaviors Other encoded behaviors are socially motivated, such as masculine or feminine s like blue, girls like pink, boys play with trucks, girls play with dolls, etc.) Whether or not these gender differentiations are desired by the parents, they are impossible to avoid, as they are regularly modeled often on television shows, in advertise ments and toy marketing, by adults, and by other children However, the influence of modeling can be highly beneficial in educational settings across a variety of characteristics from classroom behavior to motivation to academic achievement (e.g., Fiel d 1981; Wentzel & Muenks, 2016) For instance, specific to classroom behavior, when a new student comes into a classroom, it would be nearly impossible for the teacher to start anew teaching the student all the nuance rules for behavior Fortunately, wit hin a very short period of time, the new student will encode many of the necessary behaviors by observing those around her, even without any

PAGE 38

38 teacher input (e.g., bubbles and seat belts, pushing in chair, putting backpack on back of chair, etc.) This beco mes important when working with children in the classroom under a positive behavioral support model, as the teacher must remember that children can learn both adaptive and maladaptive classroom behaviors simply from observing other children (and even the t eacher) There are methods to help mediate the encoding of behaviors through observed positive and negative consequences, with these methods discussed in the next section of this literature review (i.e., vicarious learning) There were many important fi First, his research was conducted using kindergarten children, making his findings even more salient for working in elementary schools (the focus of the present research study) Second, Bandura found that children not only learned the original behaviors which they observed, but would often creatively expand upon that behavior (Bandura, 1977) For instance, Bandura found that modeling hitting a blow up clown with a hammer was copied by children, but then they adapt ed their behaviors to use other objects to hit the the doll and shot him in the head with toy gun none of the models used the gun with the bobo doll) Finally, r esponding to criticism that the bobo doll was just a toy designed to be hit, Bandura repeated the experiment using a live clown model (cited in Yount, 2010) As with the original research, the children watched an adult female model (on video) aggressively attacking the live clown and then were placed in a room with toys and the same live clown Supporting the concept of modeling, the children mimicked punching him, kicking him, and hitting him with a toy hammer

PAGE 39

39 inform many behavioral interventions and paradigms, including Positive Behavioral Interventions and Supports (PBIS) Indeed, many interventions traditionally includ e strategies to either limit, control, or maximize the influence of others on the child of focus However, the findings have also led to large scale debates about how society, For instance, of grea t concern to the field of education and child development, Bandura (1963) found that children would model aggressive behavior regardless of whether it was performed by another child, an adult, an actor on film, cartoon characters, and even written or verba l descriptions of behaviors (Bandura, 1963; Bandura, 1971b; Bandura, 1977) While gender has some impact on modeling, there is little impact on whether the model is attractive, prestigious, or likable (Bandura, 1977) Some might argue that the concept of teaching children not to bully by showing them videos involving other children getting behavior (Jeong & Lee, 2013) The effectiveness of modeling has also been used by some opponents arguing against including sex education in schools, with the belief that discussing sex or having even animated depictions of sex could increase the potential for pre marital relations (Weed & Lickona, 2014) Whether good or bad, the conce pt of modeling has also limited the availability of certain videos, movies, books, and video games to students and teachers in the classroom To some extent, it has also helped shape the content of c some might argue Ba ndura is the reason nobody was ever hurt in older G.I Joe cartoons and why Stormtroopers never actually

PAGE 40

40 hit anyone in Star Wars movies both of which were modeling research Vicarious Reinforcement Closely connected to the concept of modeling is vicarious reinforcement or vicarious learning children can very quickly learn aggressive and inappropriate behaviors by watching their peers at school, their tea chers, and their family members at home Even children that are not inherently aggressive or would not resort to aggressive actions were more likely to respond to frustration through aggression if they had observed a model act aggressively (Bandura, 1977) However, early research studies and explanations of modeling do not provide a complete understanding of how the process works in reality, as they tended to ignore and/or actively remove the impact of behavioral consequences by controlling the environmen t where behaviors were modeled and performed (discussed below) The previous discussion of operant conditioning has already introduced the concept of reinforcement, which is essentially a behavioral consequence applied to increase the frequency of the beh avior (whether it is applying a positive reward or removing a negative stimulus) When discussing modeling alone, focus is usually placed on early modeling For instance without any positive or negative consequences (e.g., they were not punished or praised, the doll did not pop, nobody was injured) The children encoded and reproduced the aggressive behaviors when placed in a room with the bobo doll without any anticipated consequences to their behavior (which incidentally allowed for them to scale up the

PAGE 41

41 aggressive behaviors) (Bandura, 1961; Bandura, 1977) However, realizing that most behaviors do not occur in a vacuum, Bandura expanded the original study by Within this study, the aggressive behavior displayed by the model was met with either rewards (i.e., given (i.e., scolded), or no consequences (i.e., control) As expected, Bandura (1965) found of the models t Most importantly, the study found that the consequences did not influence the learning or encoding of the modeled behaviors, only whether they were performed by the children (Bandura, 1965) In other words, children are constantly learning behaviors displayed by models (e.g., other children in the classroom), but are only likely to perform those behaviors if they are perceived to be rewarded or have no consequences This research formed the basis of the co ncept of vicarious reinforcement and vicarious punishment, though the later concept is not given much attention in popular writings and theories In essence, Bandura (1977) theorized that four elements are necessary for modeling to create new behaviors: ( 1) the child must be paying attention to the model; (2) the child must be able to retain the information that is being modeled; (3) the child must be able to reproduce the behavior being modeled (or an approximation of the behavior); and (4) the child must have some type of motivation to imitate the behavior modeled (Bandura, 1977) This last element is critical, as there is no need for a child to imitate or perform a behavior if there is no internal or external reason to do so It is important to note th at Bandura does not suggest motivation is

PAGE 42

42 required to learn a behavior, only that it will not be displayed without some level of motivation to do so The theory includes three primary motivators: (1) past reinforcement; (2) promised reinforcements; and (3) vicarious reinforcements (Bandura, 1986) Bandura also discusses punishments along identical lines, which are essentially de motivators and discourage the imitation of behavior (Bandura, 1977) Positive Behavioral Interventions and Supports (PBIS) m akes substantial use of Theory, with a special emphasis on vicarious reinforcement (George, Kincaid, & Pollard Sage, 2009) It is noted that PBIS does not explicitly inclu de punishment within the behaviors being ignored could be a de motivator for some children, seeing other students not receiving a reward could also reduce some behaviors) Regardless, the concept of vicarious reinforcement is integral to successful PBIS models, as it would be impossible for any teacher to reward and support every instance of every behavior for every student Rather, the teacher reinforces a positive beha vior in one student, makes a clear announcement that the reinforcement is being given, states the behavior that is being reinforced, and then waits for other students to demonstrate the same behavior At that point, the teacher can provide additional tang ible or verbal reinforcements, but it is not absolutely necessary, as the reinforcement has already occurred vicariously for all students who were paying attention or had their attention grabbed by the announcement of the reinforcement ( George, et al., 200 9 ) It is noted that PBIS can include some level of promised reinforcements, but they are not a typical component of such models (there are some delayed reinforcements,

PAGE 43

43 but these are generally accompanied by immediate reinforcement through a token econom y system) It is also important to note that providing reinforcement for every instance of a behavior in a classroom does not align with behavior modification practices determined to be the best for long lasting change (e.g., Kazdin, 2012a) wherein inte rmittent reinforcement produces results that are the least likely to be extinguished if reinforcement is missed For instance, if a child always receives a reinforcement even if s he is the last to perform a behavior and after being reminded how to behave, then there is little incentive or need to independently perform the behavior Ultimately, the act of bringing attention to the first student to demonstrate the desired behavior models the behavior to other students, provides reinforcement to the student with the proper behavior, and provides several layers of vicarious reinforcement to other students (e.g., verbal praise, tangible reward, increased attention, etc.) In fact, early research by Kazdin (1973) explored the use of vicarious reinforcement in t he inattentive behaviors The researchers found that the observing peer was vicariously reinforced to perform attentive behavior, whether or not the first student had a ttentive or inattentive behavior reinforced (Kazdin, 1973) From Behaviorism to Positive Behavioral Interventions and Support The basic components of operant conditioning (e.g., positive and negative reinforcement and punishment) and social learning theo ry (e.g., vicarious reinforcement) have earned a prominent place within behavioral management models in classroom settings, and the many educational interventions broadly implemented in recent history have been based on many of these principles (Kazdin, 20 12a) Some interventions were adapted to education from non educational fields (e.g., token

PAGE 44

44 economies), others have had significant challenges overcoming critics and negative impacts (e.g., zero tolerance), and others have found challenges in the correct implementation of complex school wide models (e.g., applied behavioral analysis and positive behavioral supports) Because of their impacts on PBIS, as studied within this research, it is important to explore the connection between these interventions and foundational theories of behavior modification Token Economies (Systematic Reinforcement) Amid the emergence of Social Learning Theory and other new theories of behavior and learning in the 1960s, researchers and clinicians were working to apply the co ncepts of classical conditioning to a wide array of social and behavioral problems One such method was the application of primarily operant co nditioning concepts through the use of token economy, which was eventually updated and strengthened with the afo rementioned concepts from social learning theory, particularly when applied in classroom settings (Kazdin, 2012b) Although it has been asserted that token Montesinos, & Precia Bernard, 2004; Liberman, 2000), a general consensus exists that token economies were actually created and implemented within some of the first human civilizations, such as the Greeks, Romans, and Aztecs (Kazdin, 1977; Shorr, 2013) Regardless of when they were first created or codified, it is obvious that early token economies were less concerned with the concepts proposed by social learning theory (which was still being formulated in the 1960 However, more recent token economy models include a better understanding of how modeling, vicarious reinforcement, and social learning can

PAGE 45

45 influence the effectiveness of token economies in educational settings (Kazdin, 2012b; Shorr, 2013) It is al so important to note that not all token economies are created for the same purpose or setting Token economies have traditionally been used for controlling behavior in mental health facilities and correctional institutions, particularly to control aggress ive and unpredictable individuals (Kazdin, 2012b) Within these closed settings, token economies can become extraordinarily complex, with the economies including punishments, token fines, and complex individualized behavior plans However, given the succ ess of token economies in closed settings, they became more common in psychotherapy with children and special education classrooms to control more problematic behaviors (e.g., Dickerson, Tenhula, & Green Paden, 2005; Elliott, Gresham, & Witt, 2013) Event ually, token economies were found effective in teaching appropriate behavior and social skills across a wide array of children, and began to be applied within the general classroom setting (e.g., Filcheck, McNeil, Greco, & Bernard, 2004; Soares, Harrison, Vannest, & McClelland, 2016) Token economies applied in elementary school settings are generally not as complex as those within institutional settings or designed for the purposes of clinical therapy, with school wide token economies generally focused on desired behaviors for all students and including only reinforcement without punishment (Kazdin, 2012b) For the purposes of this research, it is most prudent to focus on the token economies developed and researched for school settings, particularly eleme ntary schools. Regardless of where token economies were developed or where they are applied, there are several generally shared characteristics of this behavior modification

PAGE 46

46 technique (Kazdin, 1977; Kazdin, 2012b) More specifically, all token economies i nclude the token (secondary reinforcer), that for which tokens are exchanged (back up reinforcers), and the targeted behavior which is desired to be increased (Kazdin, 2012b) A token economy is essentially a systematic approach to reinforcing and increas ing desired behaviors by establishing a relationship between a secondary reinforcer and a back up reinforcer The token is specifically designed as a secondary reinforcer under operant conditioning theories, where the reinforcing nature of the token is en tirely conditioned (Kazdin, 2012b) This is an important aspect of the token economy, as the token should have no intrinsic value to the person receiving the reinforcement (food and water are examples of primary reinforcers, whereas a ticket has no intrin sic value absent the back up reinforcer) The back up reinforcer must have some kind of value to the recipient, whether it be candy, a toy, or a special event In can ex change tokens for any number of back up reinforcers Having a wide array of back up reinforcers helps avoid satiation (discussed below) and ensures a school wide token economy will contain at least one back up reinforcers found valuable by each student If any single student does not find at least one of the back up reinforcers valuable, the token economy will fail for that student as the token is only reinforcing if it can be exchanged for a valuable reinforcer Immediacy and Satiation Two concepts a re important to consider in discussing token economies under Positive Behavioral Interventions and Supports (PBIS) models: immediacy and satiation First, it is important to underscore that delayed reinforcement is not as successful as immediate reinforce ment (Kazdin, 2012a), but not all reinforcers can be

PAGE 47

47 applied immediately upon the completion of a desired behavior This is particularly true for reinforcers that are subject to satiation (such as primary reinforcers of food and water) or are expensive A token economy allows for the immediate application of a secondary reinforcer (the token), while also allowing for a delay in the more valuable reinforcer for the recipient (the back up reinforcer) Second, it is important to consider the potential of satiation which is the point at which a reinforcer is no longer valuable because the recipient has too much or no longer cares about the reinforce r (Kazdin, 2012b) In educational settings, a teacher must be careful of satiation of a reinforcer, which o ccurs when too much of the reinforcer is provided so that it loses the impact on Fortunately, satiation eventually diminishes within a short period of time if the child is deprived of the reinforcer However, secondary reinforcers a re generally less likely to suffer the effects of satiation, as their reinforcing value has been conditioned (Kazdin, 2012b) This is particularly true of money and tokens within a token economy, as these items have been conditioned to be associated with a wide variety of other reinforcers (the back up reinforcers) It is noted that the back up reinforcers (that for which the tokens are exchanged) can suffer from satiation, such that 2012b) For instance, a child who has free access to candy at home might not find candy a sufficient reward due to satiation (not to mention children who have been forbidden to eat candy at home and might experience anxiety if rewarded with candy) The concepts of immediacy and satiation work together to help illustrate the benefits of a token economy system For instance, assume a child desires colorful pencils and the teacher wants her to increase the instances of her sitting in her seat

PAGE 48

48 Under a pos itive reinforcement model, the teacher can provide the student with a colored pencil immediately each time she sits in her seat Eventually, the child will have many colored pencils and will no longer find them valuable due to satiation Alternatively, t he teacher may inform the student that she will receive a colored pencil if she sits in her seat all day (or at some point during the day) While this system would not result in satiation, it would also not provide immediate feedback to the child for her behavior, thus reducing the effectiveness of the reinforcement A better method would use a token economy, where the child receives a ticket or a check mark each time she performs the behavior At the end of the day or week, she can then trade her ticket s or checkmarks for a new pencil this provides both immediate feedback and helps prevent satiation Continuous versus Intermittent Reinforcement Another important element to consider within any token economy, particularly with elementary school student s, is the schedule of reinforcement In general, behavioral theories discuss the difference between continuous and intermittent reinforcement schedules (Kazdin, 2012b) Continuous reinforcement is when all (or nearly all) instances of a desired behavior are rewarded, while intermittent rein forcement schedules only reward some instances of the same behavior Both are important within the token economies utilized under Positive Behavioral Interventions and Supports (PBIS) models: (1) continuous reinforceme nt provides for fast learning of new behaviors, but is susceptible to quick extinction when the reinforcement is stopped or paused; (2) intermittent reinforcement provides much slower learning, but is rather robust to extinction, particularly if the there is no clear pattern as to the number of times the behavior must be displayed to be rewarded ( George, et al., 2009 ; Kazdin, 2012b)

PAGE 49

49 As such, within PBIS, it may be most beneficial to first provide continuous reinforcement of the targeted behaviors, where e very instance of the desired behaviors is rewarded with tokens and/or praise The teacher can then slowly transition to an intermittent children performing the desired behaviors Token Economies Within PBIS This literature review provides a general overview of token economies, with specific concepts related to the foundational principles of the P ositive Behavioral Interventions and Supports (PBIS) models and theories There are many concepts related to token economies that are either not relevant to the present review and/or are beyond the scope of this research For instance, Kazdin (2012b) focuses on therapy based token economies, and discusses several additional to pics, such as shaping, chaining, promptin g, fading, and stimulus control Because these additional concepts are generally outside the scope of this research and do not provide additional understanding of token economies used under the PBIS theories, they are not further elucidated within this literature review Regardless, because token economies are primarily concerned with positive reinforcements (and occasionally negative punishments removing positive stimuli to decrease behavior), the concept was ta pped early as an integral component to include in PBIS models being integrated into early concept models and research studies (Kappel, Dufresne, & Mayer, 2012; Ullmann & Krasner, 1975) Under the PBIS framework, token economies combine the standard conc epts from operant conditioning with important elements of social learning theory, particularly modeling and vicarious reinforcement The specific components and

PAGE 50

50 processes in using token economies in PBIS will be discussed later in this literature review w hen introducing the theory and specific components of PBIS models Aversive Stimuli and Zero Tolerance As mentioned earlier, some behavioral interventions have faced significant challenges in overcoming critics and unintended negative impacts of the inte rvention Many of the most challenged and controversial methods utilize aversive stimuli to essentially punish undesired behaviors or exclude the misbehaving students from educational activities (and other students) Perhaps the most well known of such p unishment driven methods is that of zero tolerance The concept of zero tolerance did not start in the field of education, but became The idea of ffenses continued to grow into a variety of areas (e.g., pollution, sexual harassment, etc.), but found the greatest foothold when criminal late 1980s and early 1990s (Ke lling & Coles, 1997) The theory essentially suggests that small offenses (such as a few broken windows) can attract vagrants that break more windows, which then leads to squatters in the buildings, which leads to fires inside the buildings and more serio us destruction of the property (Kelling & Coles, 1997) The concept propelled the idea of immediately addressing minor offenses to prevent the emergence of more serious offenses which underpi nned the implementation of zero tolerance in community policing and within the field of education By the early 1990s, schools and administrators became heavily focused on trying to develop ways to control behavior, increase time on task, and decrease disruptions from other students As such, t hey began to integrate the zero tolerance movement into

PAGE 51

51 schools and classrooms, with the idea of stopping more serious behavioral issues by applying heavy punishments to any instances of specific behavioral issues (Skiba, 2000) The application of punishments, as desc ribed in operant conditioning, is application of an aversive stimulus to decrease the instance of undesirable behavior (e.g., spanking, suspension, forced community servi ce, etc.) While this is not an element of PBIS models, it is included in this review as it helped push education into a need driven mode to develop more effective and less divisive behavioral control methods (Sugai, Horner, & Gresham, 2002) Certainly, zero tolerance still exists, but it has largely lost much of its acceptance in most educational settings In fact, in one of the most wide reaching ad missions of the failure of zero tolerance, the president of the American Federation of Teachers (Randi We ingarten) recently wrote an editorial tolerance policies intended to maintain safety and order not only In essence, as noted above, zero tolerance app lies an aversive stimulus to reduce the occurrence of undesirable behaviors For instance, if a student gets into a fight, they are arrested and sent to court with the idea that swift, inflexible, and heavy punishments will deter behavior of the offendi ng student and other students in the school Other examples are almost comical in th e lack of forethought into zero tolerance policies For instance, suspending a student for truancy infractions (almost rewarding them with endorsing their desire not to a ttend school) There are also sev eral examples where zero tolerance actually exacerbates poor behavior (Poe Yamagata & Jones, 2000; Wald & Losen, 2003), as it is not generally paired with any other behavior

PAGE 52

52 modification techniques (e.g., negative reinforc ement or positive reinforcement) As an example of how negative reinforcement could be applied, the school could have a policy that no students can eat in the courtyard unless there are no fights for the prior week if there are no fights, then they get to eat in the courtyard, thus removing the aversive situation (hence, negative reinforcement) There are certainly inherent problems with zero tolerance policies, and this has led some educational theorists and practitioners to sarcastically rename zero t (Teske, 2012). In addition to challenges with zero tolerance, a number of researchers have demonstrated that other educational interventions often include controversial aversive stimuli and that t hese interventions are relatively ineffective For instance, the most common responses to students with conduct disorders are punishment and exclusion (e.g., Lane & Murakami, 1987; Nieto, 2005; Sprick, Borgmeier, & Nolet, 2002), although other research sh ows that these aversive techniques are not effective in producing long term reduction in problematic behaviors (Costenbader & Markson, 1988) Moreover, a variety of researchers have demonstrated that punishing problematic behaviors without providing a pos itive and proactive support system is counterproductive ( Sugai, et al., 2002 ) Indeed, researchers have found that such one sided punishment can lead to increased instances of undesired behaviors such as aggression, vandalism, truancy, and dropout (e.g., Mayer, 1995; Mayer & Sulzer Azaroff, 1991; Skiba & Peterson, 1999) Additionally, in a review of over 600 research studies, Horner, Sugai, and Dickey (2016) found that one of the least effective methods for reducing school discipline problems and school v iolence was punishment Among these studies, the same

PAGE 53

53 researchers that reported the ineffectiveness of aversive punishment (e.g., Gottfredson, Gottfredson, & Hybl, 1993; Lipsey, 1992; Tolan & Guerra, 1994; Elliott, Hamburg, Williams, 1998) also indicated t he most effective methods for addressing problematic behaviors and school violence includes social skills training and behavioral interventions particularly those found in Applied Behavioral Analysis (ABA) and associated theories (such as Positive Behavior al Interventions and Supports) As t he field of education began to push bac k against the concepts of zero tolerance and the use of aversive stimuli to control student behavior, researchers discovered unacceptable racial disparities in the application of these punishment driven methods For example, Wald and Losen (2003) found that Black students made up 17% of the national student population, but accounted for 34% of those students who had been suspended from school and were 2.6 times more likely to be s uspended as White students not only concluded that zero tolerance is an ineffective system, but (more importantly), there is no evidence that students from minority groups are prone to di sruptive or violent behavior (Skiba et al., 2006) Rather, research suggests that the disparity in application of student punishments (with an overrepresentation of student from traditional minority groups) is most often related to a bias among teachers a nd administrators to more quickly refer and more severely punish these students for minor behavioral issues (Skiba, 2000; Skiba & Knesting, 2001; Weingarten, 2015) Such disproportionate rates of referrals and punishments transfers to the juvenile justic e system, as schools are taking more students in traditional minority groups out of the positive learning environment and punishing often immature behavior (rather than teaching more appropriate behaviors; Skiba, 2000) This is essentially the basis behin d

PAGE 54

54 to punishment driven methods of school and classroom control Essentially, intervention theorists took these critiques to try and develop interventions that would be less co ntroversial and focus on positive behaviors and experiences for students Moreover, in January of 2014, the U.S U.S (USED, 2014) issue d a joint letter to serve as a significant guidance document and strongly urged schools throughout the (p. 1) but to help students succeed throug h programs that support and reinforce positive student behavior and character development Indeed, if applied correctly, Positive Behavioral Interventions and Supports (PBIS) models can help reduce racial disparities and end the school to prison pipeline by focusing on positive reinforcement and teaching students more mature and appropriate behavior Applied Behavioral Analysis (ABA) While token economies and realizations regarding the ineffectiveness of aversive interventions helped encourage the widesp read integration of PBIS across the United States, some of the deepest roots of PBIS lie within Applied Behavioral Analysis (ABA; Dunlap, et al. 2009) In turn, the roots of ABA are found in all three of the aforementioned theories of learning: classical conditioning, operant conditioning, and social learning theory In essence, ABA is the systematic application of interventions based on these learning theories to impact specific behaviors, while simultaneously exploring whether any changes are attributa ble to the intervention process (Baer, Wolf & Risley, 1968) Baer, et al. (1968), who were among the founding theorists of ABA, are often considered to have provided the most comprehensive and standardized

PAGE 55

55 definition of ABA within the first issue of the J ournal of Applied Behavior Analysis Within their article, the researchers indicated that an applied behavior analysis must manipulations which analyze with clarity (p. 97) procedures contributing to that change, the effectiveness of those procedures in making sufficient change for value, and th 97) With such complexity, theoretical underpinnings, and the systematic nature of application, ABA Early in the rapid growth and expansion of ABA, the technology found overwhelming success in modifying and controlling the behavior of institutionalized patients, particularly in terms of manifestations of disabilities, such as aggression and self injurious behaviors (Dunlap, et al. 2009) By the 1980s, ABA had grown from an emerging theory of applied research to a vetted method of organizing and providing services to school age children, particularly those with special needs, disabilities, and severe behavior disorders (Dunlap, et al. 2009) not entirely consistent with the concepts of ABA, and many teachers had some level of misunderstanding about ABA and how it should be applied (Cooper, 1982) In modern concepts of ABA have taken a strong foothold within many classrooms, with an emphasis on assessing the functional relationship between the environment and the behavior of interest (Mace, 1994)

PAGE 56

56 Basic p remises of ABA In general, Applied Behavior Analysis (ABA) is a technology used by behavior analysts to explore and target change in a specific feelings, and behaviors within a standard s equence of antecedents, behaviors, and consequences The analysis is essentially intended to discover the underlying causes and purposes behind the undesirable targeted behavior, with the ultimate hope of applying methods to decrease or stop the undesired behavior (Pierce & Cheney, 2013) For example, a young girl with autism and limited verbal skills is given a literacy activity requiring her to match a grapheme to the initial sound of a word Once the teacher places the worksheet in front of her, she b egins to rock in her chair, stands up, and throws chair In response, the teacher removes her from the other students and places her into time out Under an ABA framework, the analyst would identify the targeted behavior of concern (i.e., throwing the ch air), the antecedents to that behavior (e.g., assigning a challenging literacy activity, rocking in the chair, standing up, heightened anxiety), and the consequences to that behavior (e.g., being placed in time out, having the literacy activity removed, su ccessfully avoiding the assignment, reduction in anxiety) It is possible that an analyst would consider the chair rocking and standing up as additional focus behaviors, depending on how early the teacher should intervene to prevent the behavior escalatio n Regardless, by analyzing the sequence of antecedents, behaviors, and consequences, the analyst could reasonably conclude that the student demonstrated the aforementioned behaviors as a means of avoiding the undesirable literacy task The analyst would then create specific interventions to help prevent the undesirable behavior, such as buckling the child to her seat while

PAGE 57

57 completing literacy work or providing a mild shock when failing to complete (or begin) the assignment within an allotted timeframe Such aversive stimuli and punishments (e.g., shock, pain, discomfort, restraint, etc.) were not uncommon in the earlier years of ABA, a fact that helped lead to the development and wide spread adoption of PBIS (Dunlap, et al. 2009). From aversive to p osit ive ABA was initially used by practitioners in institutional settings to stop self destructive or dangerous behaviors that were unresponsive to medications or other treatments, with most of these individuals having severe disabilities and behavioral disorders (Pierce & Cheney, 2013) As ABA began being used more often within educational settings, practitioners and behavior analysists borrowed a great deal from institutional models and began using more aversive and punishment based interventions (Pier ce & Cheney, 2013) Within the confined environment of an institution, aversive stimuli and punishment were seen as the most efficient and expeditious method of stopping self damaging or dangerous behaviors (e.g., head banging, scratching, etc.) Applied in educational settings, these aversive stimuli and punishments also appeared effective at stopping undesirable behaviors and began to be widely adopted in working with students with disabilities and severe behavioral problems There are many examples of these aversive punishments, such immediately punish self head) using an automatically administered electric s hock (Linscheid, Iwata, Ricketts, Williams, & Griffin, 1990) In addition, an often cited article in the New York Times provided a disturbing look at other aversive stimuli and punishments used by schools in

PAGE 58

58 Massachusetts and New Jersey working with stude nts with severe disabilities, such as the use of restraints, starvation, pinching, spanking, cold showers, an automatic vapor As the use of aversive stimuli and punishments grew across the cou ntry, moral objections to the use of some techniques (particularly the shocking, water spraying, white noise, and physical punishments) and documented permanent damage caused to rch on positive behavior interventions and the development of PBIS as an independent field from ABA (Singer & Wang, 2009) Indeed, as mentioned earlier, a number of researchers have demonstrated that interventions based on aversive stimuli are relatively ineffective at producing long term reduction in problematic behaviors (Costenbader & Markson, 1988; Sugai, et al., 2002 ) Researchers have also found that punishment driven interventions can lead to increased instances of undesired behaviors (e.g., Mayer, 1995; Mayer & Sulzer Azaroff, 1991; Skiba & Peterson, 1999) and are among the least effective methods for reducing school discipline problems and school violence (e.g., Gottfredson, et al., 1993 ; Horner, Sugai, and Dickey, 2016; Lipsey, 1992; Tolan & Guer ra, 1994; Elliott, et al., 1998 ) Other researchers found that punishment is entirely ineffective at bringing about behavioral change, serving only to suppress undesired behaviors in the presence of the punisher, but not in other situations (e.g., Smith, 2012) Fortunately, researchers have found that the most effective methods for addressing problematic behaviors and school violence are social skills training and Positive Behavioral Interventions and S upports (e.g., Gottfredson, et al., 1993 ; Lipsey, 199 2; Tolan & Guerra, 1994; Elliott, et al., 1998 )

PAGE 59

59 Thus, it was primarily the principle of morality that created the offshoot of ABA called PBIS, while ABA researchers created their own journal and professional organization to help defend the use of aversi ve stimuli and punishments in behavior management systems (Singer & Wang, 2009) Unlike many researching and developing ABA theories, researchers and practitioners of PBIS believed that behaviors could be more effectively and humanely changed using positi ve alternatives to aversion In addition, early entrants into PBIS were also committed to improving the overall quality of life for PBIS intervention recipients by teaching a range of productive behaviors beyond just trying to reduce discrete target behav iors (Sailor & Wang, 2009) Ultimately, moving from aversive interventions to positive interventions is the key difference between PBIS and ABA, and is one of the most influential reasons why PBIS has become the national model for intervention across the country Individual to group i ntervention ABA was relatively successful in working with autistic and severely disturbed children in institutional settings, giving it an inroad into the field of education, specifically to support the education of studen ts with disabilities and severe behavioral issues Upon introduction to educational settings, ABA was focused on working with individuals with disabilities who demonstrated inappropriate and challenging behaviors (i.e., maladaptive, violent, and/or social ly inappropriate) with the goal of stopping such behaviors (Dunlap, et al. 2009) As PBIS emerged as a field independent from ABA, initial research and practice was also focused on individuals with disabilities in ESE classrooms, with a focus on individu ally identified behaviors to help the student succeed in the school environment (Sailor & Wang, 2009) The application of PBIS became more prominent as educational settings moved towards

PAGE 60

60 more inclusion of students with disabilities, but remained focused o n individual based interventions based on individualized functional behavior analysis However, the overall success and impact of PBIS with individuals was eventually applied across entire schools, now focusing on supporting and teaching behaviors that wo uld support a safe and productive learning environment for all students in the school (rather than for individual students or classrooms; Sailor & Wang, 2009) This represents the second primary difference between traditional ABA and PBIS a movement fro m individualized interventions for students with disabilities to school wide interventions encompassing every student in the school regardless of disability status The specifics of this school wide PBIS approach will be described in more detail later in this chapter Other differences b etween ABA and PBIS Ancillary to the expanded focus of PBIS to the entire school, the method of data collection to inform which behaviors to target and which interventions to implement is also necessarily different Mo re specifically, as discussed later in this chapter, PBIS focuses on identifying behaviors for intervention, but does not place as much focus on antecedents or consequences as these are much harder to aggregate into meaningful data across an entire schoo l of hundreds of individual students That being said, the interventions within PBIS are most often focused on providing new antecedents to elicit more productive behaviors and, thus, theoretically eliminating problematic behavior by making it unnecessary for the student (Ylvisaker, Feeney, & Hibbard, 2006) PBIS also differs from ABA in focusing on environmental changes, teaching and communication skills, and real world contexts for assessment and intervention (Ylvisaker, Feeney, & Hibbard, 2006) Ultimately, while there are certainly core similarities between ABA and PBIS (e.g., systematic

PAGE 61

61 observation of behavior, understanding of behavior antecedents and consequences, and systematic application of interventions), there are several key differences that have helped shape PBIS into one of the most effective methods of school wide behavior change and improvement (Sailor & Wang, 2009; Ylvisaker, Feeney, & Hibbard, 2006) Sociopolitical Roots of PBIS as a School Wide Intervention Model While the previo us sections of this chapter have focused upon the theoretical underpinnings of Positive Behavioral Interventions and Supports (PBIS), such as operant conditioning, classical conditioning, social learning theory, and applied behavior analysis, it is also im portant to understand how PBIS actually emerged as a wide spread movement in the field of education Some of the reasons for the emergence have already been addressed earlier in this chapter, such as helping to address some of the challenges faced by othe r classroom and behavior management models applied broadly in schools (e.g., focus on punishments, use of aversive stimuli, or restriction to addressing behavior of individual students) These provide some field based explanations as to how and why PBIS w as expanded to school wide models of behavioral support However, it is also important to understand the social and political movements that helped drive the emergence of PBIS as one of the primary methods for school wide interventions to drive positive b ehaviors. Roots in Teaching Students with Disabilities The need to work with children identified with disabilities was one of the most influential sociopolitical movements that led to the eventual emergence of PBIS as the most utilized theory of change in schools The movement and need to provide better services to students with disabilities began in the 1950s with deinstitutionalization of those with mental, developmental, and intellectual disorders (Torrey, 1997) The

PAGE 62

62 movement to deinstitutionalize thes rights movement of the 1960s and several publications revealing the terrible and et al. 2009) Ultimately, the push to get children and adults out of hos pitals and asylums meant they would then be entering communities (and schools) that did not have the infrastructure to support their needs and provide standards of care enjoyed by those without disability, such as education (Torrey, 1997) Over the next 2 0 years (by the 1980s), the population of hospitals and asylums had decreased by nearly half a million people, with fewer than 72,000 remaining in institutions by the mid 1990s (Torrey, 1997) It was no longer acceptable to place children with severe disa bilities into institutions, and they began finding their way into the regular public schools Having students with severe disabilities would certainly create a challenge, as teachers had few tools to work with these types of children Fortunately, about the same time that these children began enrolling in public schools, the field of behavior modification (eventually called Applied Behavior Analysis; ABA) was expanding rapidly with studies reporting amazing success in working with children with severe di sabilities However, as noted by Dunlap, et al. (2009), there was a complicated paradox occurring in the 1980s, as schools across America were banning corporal punishment and much of the ABA practices of the time were focused on aversive stimuli and punis hments Some of the punishments and aversives were adopted from behavior control techniques of the same institutio ns that led to the movement, wit h some of these punishment driven behavior techniques discussed in the prior section of this chapter Howeve r, at the time, these aversive techniques seemed to be effective and efficient at quickly changing

PAGE 63

63 student behavior (recall the earlier discussion of suppressing behaviors versus teaching new prosocial behaviors) Fortunately for students with disabilitie s, many educational professionals found the use of aversives and punishments to be unacceptable in working with any children, particularly those with disabilities (Dunlap, et al. 2009) As such, there was a push among professionals and researchers to fin d alternatives to the punishment driven paradigm of ABA at the time Researchers first began focusing on alternative explanations for behaviors moving from just focusing on reducing the occurrence of behaviors to focusing on why and when these behavior s happen This refocusing of the field led directly to the development of Functional Behavior Analysis (FBA; discussed later in this chapter; Dunlap, et al. 2009) Research continued to show punishment based paradigms were often more damaging than helpf ul, and FBA helped increase awareness that there were logical and understandable reasons for the undesirable behaviors of these students with disabilities Upon a more widespread and public understanding of these behaviors and the students, there became a need to find alternative and non aversive methods for reducing the emergence of undesirable behaviors (Dunlap, et al. 2009) By the late 1980s, researchers and theorists had broken from the field of ABA to create the field of Positive Behavior Intervent ions and Supports (PBIS) The initial research efforts of the 1980s led to the inclusion of PBIS in the Individuals with Disabilities in Education Act Amendments of 1997 (IDEA) as a potential mechanism to work with students with disabilities With this r eauthorization (discussed in more detail below), researchers at the University of Oregon received a grant to create the Center on Positive Behavioral Interventions and Supports to provide technical assistance to all schools and districts wishing to impleme nt (Sugai & Simonsen, 2012)

PAGE 64

64 Laws, Rules and Regulations In addition to the emerging need to find effective methods of working with students with disabilities, the 1980s and 1990s were replete with new rules and regulations requiring schools to provide s pecific services to these students Among the first laws to impact schools in regards to students with disabilities was the Rehabilitation Act of 1973, which was subsequently amended in 1986 and 1992 This law was technically designed to ensure equal opp ortunities for employment among people with disabilities within agencies receiving federal funding, but was eventually extended to a wide range of environments where federal funding is provided (USED, 2010) Of importance to this research, Section 504 of the Rehabilitation Act of 1973, as amended, prohibits discrimination on the basis of disability within educational programs or activities that receive federal funding under the U.S Department of Education (USED, 2015) This section has become part of th e standard terminology in education, supports but not covered under the Individuals with Disabilities in Education Act (IDEA) The second important piece of legislation that drove the incorporatio n of students with disabilities in public schools was the Education for all Handicapped Children Act of 1975, which was an amendment to a previous law from 1966 This law required schools to create an educational plan for handicap ped children that would e nsure them an education as close as possible to that provided to non handicap ped children, while ensuring they were placed in the least restrictive environment that allowed the greatest opportunity to interact with non disabled peers (USED, 2010) This la w was eventually revised and renamed as the Individuals with Disabilities in Education Act of 1990 (IDEA) In essence, this law requires schools and districts to provide all students with

PAGE 65

65 disabilities a Free Appropriate Public Education (FAPE) that provid es specially designed instruction for their specific needs The revised law retains the earlier requirement that the educational services must be provided in the least restrictive environment C.F.R §300.130) Although similar to Section 504 of the Rehabilitation Act of 1973, the requirements for services and accommodations under IDEA are more complex and detailed requiring an Individualized Education Plan (IEP; USED, 2010; 34 C.F.R §300. 340) The breadth of the requirements under IDEA have been solidified by federal courts, where the United States First Court of Appeals ruled that schools were required to provide services to students with disabilities regardless of the severity of the di sability (Timothy W v Rochester, New Hampshire, School District, 1989) While this is a simplified explanation of IDEA, it provides an understanding of the timeline leading to the establishment of PBIS as a primary method for addressing behavior withi n entire schools. The third major legislation to impact the provision of services to students with disabilities was the American with Disabilities Act of 1990 (ADA), primarily Title II The ive national 1990) This act was significantly amended in 2008 and was renamed the American with Disabilities Act Amendments Act of 2008 (ADA Amendments Act) The amended act intended to undo the actions of courts to limit the scope of protection intended by the original law However, neither ADA of 1990 nor the ADA Amendments Act of 2008 provide any greater protection for students with disabilities than the aforem entioned

PAGE 66

66 laws (which are more specific to education; USED, 2015) As such, while ADA is important to protect the rights of individuals with disabilities, it finds less use within the field of education Other legislation was passed to help protect the ri ghts of individuals with disabilities, such as the Telecommunications Act of 1996 and the Fair Housing Act of 1988, though most of these additional laws are not related directly with the education of students with disabilities Student Inclusion The two aforementioned sociopolitical movements first deinstitutionalized and incorporated students with disabilities into schools, then established requirements for schools to provide the same education in the least restrictive environment With momentum in favo r of students with disabilities and laws seemingly requiring all accommodations to ensure these students can receive an equitable education with non disabled students, a significant road block was met with the US Supreme Court ruling in Board of Education of the Hendrick Hudson Central School District v Rowley (1982) In this ruling, the Court determined the law only required school districts to provide students with disabilities with supports necessary for them to demonstrate any level of educational be nefit The Court stated that a school system was not responsible to (p. 21, 458 U S 176) This has helped create a point of contention between parents and schools as to what is restrictive environment (Blankenship, et al. 2007) Schools have been accused of simply seeking to show any educational benefit and selecting placements that represent the least cost in terms of money and personnel (Blankenship, et al. 2007) However, with the support of several court cases and continued efforts by schools to obtain the best results on

PAGE 67

67 sociopolitical change leading to the widespread utilization of PBIS in schools disabilities, with the intent of placing these students appropriately in regular classrooms for the maximum amount of time possible each day Using the most recent data available, t he US Department of Education found that 95% of students served under IDEA (i.e., with IEPs) had been educated in regular classrooms for at least some part of the school day (USED, 2015) In addition, this same report found that 62.1% of students served under IDEA were in regular classrooms for at least 80% of the school Only 5% of the students served under I DEA were educated outside the regular classroom (e.g., self contained classrooms) T his introduces new complications for teachers within inclusive classrooms, as there may now be students with difficult behaviors that are manifestations of their disabilit y Learning methods to work with these students without punishing them or distracting from other students required new techniques and systems, and became a focus of PBIS Indeed, PBIS theorists and researchers developed a paradigm wherein all students in the school are provided positive supports to learn and exhibit positive behaviors, which can help all students (and the teachers) Positive Behavior Interventions and Supports (PBIS): From Theory to Practice As discussed earlier, the inclusion of PBIS i n school settings, particularly in school problematic and challenging student behavior that disrupts the learning environment, environment not just for offending student, but also for other students, instructional

PAGE 68

68 personnel, support staff, and other stakeholders In this section PBIS, more specifically PBIS as a school wide approach, is desc ribed in order to highlight its defining characteristics and features Positive Behavior Interventions and Supports (PBIS) As evidenced by the complex foundational theories and numerous challenges to comprehensive behavior interventions, Positive Behavio r Interventions and Supports (PBIS) has a broad and complicated theoretical foundation and resultant definition Indeed, entire books have been written to explain and describe PBIS, and organizations (e.g., The Technical Assistance Center on Positive Behavioral Interventions and Supports established by the U.S Department of Education's Office of Special Education Programs) and large conferences have been designed around supporting PBIS implementation As such, it would be impossible for this literat ure review to fully and explicitly define PBIS, particularly as definitions can vary depending on the focus of the setting of implementation That being said, a brief explanation of the theory and its basic elements will help to illustrate why PBIS was ch osen as the intervention of focus for this study In essence, PBIS is considered a broad approach designed to provide school personnel with tools and skills necessary to enhance student academic achievement and improve socially relevant behavior (Florida Assistance Center on PBIS, 2016) Rather than providing a single strategy or intervention, PBIS is designed to allow individual schools and organizations the flexibility to provide a wide variety of evidence based and research based interventions that best meet the needs of their populations, specifically students in this case

PAGE 69

69 PBS Project, 2002b, p.1) constantly adapting to the needs of the students, classrooms, and schools through an integrated process of assessment, feedback, and data analysis Some of the most an application of research based behavioral strategies; (2) use of multiple and integrated intervention elements aligned to the needs of the enviro nment; (3) a commitment to sustained and long term outcomes; and (4) support within the organizational systems that ensure sustained impacts of the interventions (Dunlap, et al. 2009) While developing a single sentence definition of PBIS is nearly impos sible, the above overview is enhanced below with a more thorough description of some of the most prominent elements of the model (e.g., Response to Intervention, staff buy in, training, etc.) Multi Tiered Systems of Support (MTSS) In addition to pulling from Applied Behavior Analysis, PBIS has pulled from other For example, Multi Tiered System of Supports (MTSS), previously known as Response to Intervention (RtI), is a critical underlying component of the PBIS approach ( McIntosh & Goodman, 2016; Sugai & Horner, 2009 b ) MTSS essentially utilizes a continuum of three tiers (i.e., primary, secondary, tertiary) that outline differentiated intervention approaches designed to progressively move from those effective with a majo rity of students (primary, Tier 1 impacting 80% of the student population receiving the interventions), to those effective with small groups (secondary, Tier 2 impacting 15% of the student population who do not respond solely to primary tier interventi ons), and to those effective with individuals (tertiary, Tier 3

PAGE 70

70 impacting the remaining 5% of the student population who do not respond to primary and secondary tier interventions; McIntosh & Goodman, 2016) MTSS is used extensively within classrooms wh en working with students to ascertain if additional educational supports or programs are needed to best meet student needs, and has been described as providing outstanding guiding principles for student assessment and intervention decisions under the PBIS approach ( Sugai & Horner, 2009a) The multi tier continuum was originally developed primarily to work on the individual student level, with interventions progressing through the continuum when a lower tier was ineffective for the student (McIntosh & Good man, 2016; Sugai & Horner, 2009a) For example, MTSS can help determine whether students struggling in regular education classrooms are demonstrating deficiencies in their learning to warrant testing for learning disabilities and, if so, whether or not a special education program would be most appropriate for their learning needs (McIntosh & Goodman, 2016) PBIS took the multi tiered approach of MTSS and applied the continuum to entire classrooms and school settings, essentially helping move the concepts to broader application models ( Sugai & Horner, 2009a) Unlike the traditional MTSS approach, PBIS focuses on the entire student body as a whole rather than focusing on individual students In addition, PBIS tends to focus primarily on Tier 1 intervention s, though Tiers 2 and 3 interventions are available for students that need more intensive behavior supports In essence, PBIS borrows the multi tier continuum from MTSS to help enhance the overall effectiveness PBIS approach at a school wide level (McInto sh & Goodman, 2016) Functional Behavior Assessment (FBA) PBIS also relies heavily on the behavioral analysis techniques of Functional Behavioral Assessment (FBA), another product of Applied Behavior Analysis (ABA)

PAGE 71

71 (Singer & Wang, 2009) At its most bas ic level, FBA is a process of collecting and circumstances to determine the function of the behavior (i.e., the purpose of the behavior for the student; Singer & Wang, 2009) Fu nctional Behavioral Analysis (not cognitive or communication disabilities behave in specific maladaptive ways (e.g., children with autism; Dixon, Vogel, & Tarbox, 2012) By focusing on an analysis of behaviors and circumstances, there was less need for the individual to explain why they were exhibiting certain behaviors a characteristic that was particularly important for these individuals that could not communicate we ll enough to explain the reason for their behaviors With the reauthorization of the Individuals with Disabilities in Education Act a required process for every student i dentified with a disability if they have serious misconduct that are determined to be manifestations of their disability and would result in a change of placement exceeding 10 days (e.g., placement in an alternative school; 20 U.S.C §1415(k)(1)(F)(i)) essentially the same, though IDEA used the new term because of the broader base of students for whom it would be required In general, FBA not only focuses on the antecedents and consequences of mala inappropriately, but also on understanding cultural contexts and social meaning being associated with the maladaptive behaviors (Singer & Wang, 2009) In addition, FBA also focuses on us ing positive interventions to develop alternative behaviors that reduce

PAGE 72

72 and/or replace the maladaptive behaviors (Singer & Wang, 2009) For this reason, FBA was integrated into the overall method of designing and implementing school wide PBIS models How ever, as with MTSS, functional behavioral assessment was designed to be used at an individual level and is now required by law to be used at an individual level for all students identified with disabilities (Hanley, Iwata, & McCord, 2003) The most curren t theories and models of school wide PBIS (the focus of this research) took the individualized techniques used in FBA and applied them to a school wide method of assessment (Singer & Wang, 2009) With an individualized method, FBA could be used to determ ine why a student consistently refuses to read during small back, bothers others, touches materials, etc.) until the teacher removes them from the group (e.g., time out) In this situation, FBA might discover that the student is engaged in task avoidance because they are embarrassed about their reading ability, which could lead to positive interventions to reduce the maladaptive behavior (e.g., remediation to bolster confidence, encourage asking for help, etc.) School Wide PBIS (SWPBIS) transposes this technique to focus less on the specific function of behaviors of individual students, and more on behavioral data across the entire school of students Within the SWPBIS model (similar to FBA), data are collect ed to discover which are the most salient and frequent maladaptive behaviors across the entire student body, particularly those that are challenging, socially inappropriate, and negatively impacting on student learning and safety (Dunlap, et al. 2009) W ithin such a school wide transposition, PBIS also develops positive interventions that can reduce or replace the maladaptive behaviors in those students displaying such behaviors, while also adding

PAGE 73

73 the intent to prevent the emergence of the maladaptive beh aviors in students that have not previously behaved in such a way (e.g., Singer & Wang, 2009) Ultimately, this process ensures that all students receive maximum educational benefits Design Considerations for SWPBIS While the theoretical underpinnings of Applied Behavior Analysis, Multi Tier Support Systems, and Functional Behavioral Assessment all provide a general guide to develop and implement School Wide Positive Behavioral Interventions and Supports (SWPBIS), it is important to also consider some o f the most important aspects in the preliminary design of a school wide PBIS intervention model While there are many design considerations for SWPBIS models, the following are four elements that have the most impact on design and ultimate implementation of an effective intervention model This subsection is focused on elements that must be considered before implementation of the interventions, while the next subsections focus on implementation aspects and considerations of the SWPBIS model Design cons iderations: s taff buy in It is common knowledge that having staff and employees change initiative is important with any organization, whether they be new human resources initiatives or new ways of working with students In terms of the systematic and structured changes implemented within SWPBIS, adopting a systems perspective means it is absolutely cri procedure, and initiative ( Sugai & Horner, 2009a) This is essential if t he new techniques are to be consistently implemented with fidelity, particularly in terms of consistently providing the correct positive supports for the proper social behaviors across the entire school Lack of buy in can create confusion among the child ren and

PAGE 74

74 drastically decrease the effectiveness and efficiency of the SWPBIS initiatives both during early implementation and during maintenance phases in future years ( Sugai & Horner, 2009a) The process of staff buy in for SW PBIS must begin with a core group of school personnel who see a need for improvement in overall student behavior, are willing to implement systematic change, and are willing to help pr omote the utilization of SW PBIS ( Sugai & Horner, 2009a) The initial staff buy in does not need to compose the entire school, as there are opportunities for increasing buy in as the design process moves forward These initial staff might be front line staff (such as teachers) or administrators, with the latter group needing to incorporate front line st aff as quickly as possible in the process Regardless of the composition of the initial group, the next step is to obtain Sugai & Horn er, 2009a) The SWPBIS Team should include members from school administration and all ranks of instructional staff, ideally representing each grade level This anchoring group must receive formal on or representing agency, ideally prior to designing any initiatives (FLDOE, 2008) After the SWPBIS Team has been established and trained, they can begin the process of designing the school wide interventions under the model This becomes one of the m ost important steps for promoting more wide spread staff buy in, with the SWPBIS Team discussing and eliciting feedback from other teachers, instructional staff, and non instructional staff (e.g., paraprofessionals, cafeteria workers, and custodians) ( Suga i & Horner, 2009a) Knowing their feedback is being considered will help promote

PAGE 75

75 buy in, while also providing valuable data for determining the school wide goals and interventions to be implemented Overall, it is suggested that at least 80% of school pe intervention philosophy) and agree to implement the PBIS model (once developed) before the school can begin to implement a school wide PBIS initiative (e.g., Newcomer & Barrett, 2009; Putnam et al., 2009) Indeed, achieving staff buy in has been found to be directly related to successful implementation of SWPBIS, with this same research indicating that low staff buy in can result from misperceptions or misunderstandings of the ba sic foundational theories of SWPBIS, poor understanding of the principles and interventions used within PBIS, and lack of support for overall implementation (Kincaid, Childs, Blase, & Wallace, 2007) After this buy in threshold has been achieved, the SWP BIS Team works to draft initial goals and interventions for review by all stakeholders from all areas of the school (e.g., grade level classroom teachers, special area teachers, support staff, school administration, etc.) These stakeholders must then mee t to share their concerns, ideas, analyze frequently occurring behavior concerns, and work to develop a collaborative plan of action ( Sugai & Horner, 2009a) This task helps with giving everyone involved a voice and further helps promote and reinforce sta ff buy in and understanding of the SWPBIS intervention model Ultimately, the importance of staff buy in from as many stakeholders as possible cannot be understated The overall success of a school wide PBIS initiative hinges on having as many staff as p ossible implementing the same interventions and supporting the same behaviors, thus promoting successful systematic changes to student behaviors (FLDOE, 2008; Newcomer & Barrett, 2009).

PAGE 76

76 Design considerations: d ata collection and analysis As discussed above school wide PBIS is closely linked with Functional Behavior Assessment and Applied Behavior Analysis (Singer & Wang, 2009) As such, preliminary and on going data collection and analysis are integral components of the SWPBIS approach ( Sugai & Horner, 20 09a) In essence, school wide behavior data are collected and analyzed to determine the types of challenging and inappropriate social behaviors that students within a school are demonstrating (note that SWPBIS is focused on social behaviors across the sch ool; Sugai & Horner, 2009a) Based on the initial review of data, student behavior priorities are established for intervention under the SWPBIS model Four types of data sources and collection methods to establish these priorities and associated interven tions include: (1) informal observations, (2) formal observations and surveys, (3) school behavioral data, and (4) ongoing data collection methods Each of these four are briefly discussed below. Informal o bservations Data collection most often begins wi th an informal analysis of overall student behaviors that are concerning to school personnel, such as students not respecting others, noisy transitions, etc (FLDOE, 2005) The informal analysis is often completed by the initial group of personnel bringi ng up the need for a school wide model of systematic behavioral change Upon approval of the school administrators to develop a school SWPBIS team, additional informal discussions are had with all stakehol ders (e.g., grade level classroom teachers, special area teachers, support staff, school administration, etc.) in an effort to identify any additional behavioral concerns across the school population ( Sugai & Horner, 2009a) These informal observations ar e critical for the

PAGE 77

77 design of the PBIS model within the school, and they allow for a deeper understanding and discussion of the behaviors to be focused upon by the interventions Formal observations and s urveys Beyond informal observations provided by st akeholders, many schools and districts conduct more formal school climate surveys (Lewis, Barrett, Sugai, Horner, Mitchell, & Starkey, 2016) These surveys are often completed by school personnel and parents, and generally include a section on school discipline and safety Data from these climate surveys can then be utilized by the school staff and parents perceive to be enviro nment Certainly, not every behavior or issue identified in school climate surveys will be addressed by a SWPBIS intervention model, though behaviors that are reported consistently should be considered when designing the model In addition to formal sur veys, nearly every school completes formal observations of teachers during annual evaluation processes These formal evaluations can be Those student behaviors that caused the great est disruption during formal evaluations can become areas of focus for the SWPBIS intervention model In this sense, it is important for the SWPBIS Team and the school administrators to only focus on behaviors that are relatively common disruptions to the learning environment It is also important to select target behaviors that would benefit the most stakeholders, which is important to help enhance staff buy in (discussed above) Ultimately, when formal classroom observations are analyzed, reoccurring b ehavior concerns and themes can be

PAGE 78

78 School behavioral d ata : In addition to data from teachers and staff, the school also has access to a plethora of data on student behaviors (particularly undesirable behaviors; e.g., Sugai & Horner, 2009a) For instance, the school may have access to that can provide an understanding of those behaviors that were problematic and disruptive, but did not rise to the classification of a formal office discipline referral (e.g., talking with peers after being told to stop repeatedly, throwing a crayon, argu ing with a teacher, etc.) In addition to pre referral data, the school often has access to formal office discipline referral (ODR) data, which is often separated into different levels of offenses depending on their severity (as determined by the school, district, and/or state) Taken together, the pre referral and referral data can be analyzed to determine the most common behaviors that teachers find to be problematic and disruptive in the years prior to the implementation of SWPBIS (e.g., Clonan, McDoug al, Clark, & Davison, 2006; Irvin et al., 2006; Irvin, Tobin, Sprague, Sugai & Vincent, 2004) Certainly, other data are available (e.g., suspensions, detentions, school absences), but these are most commonly used for middle and high school students, and the focus of this research study is elementary school Indeed, it is most common for research on school (e.g., Bradshaw, Reinke, Brown, Bevans, & Leaf, 2008; Mclntosh, Campbell, Ca rter, & Zumbo, 2009; Pas, et al., 2011 ) On going data collection and continuous i mprovement School personnel certainly have access to a wide variety of informal and formal student data from daily attendance to behavioral data to academic outcomes All of these data points allow for

PAGE 79

79 explore any potential differences between student groups based on demographics (e.g., ESE status, gender, race/ethnicity, etc.) This al lows for SWPBIS Team to make data based decisions as to which behaviors to target, while ensuring all student sub groups are equally impacted by the positive behavior supports Most definitely, with research showing African American students as being o ver twice as likely than their w hite peers to have more serious consequences for the same behaviors (Skive, Horner, Chung, Rausch, May, Tobin, 2011), it is imperative that sub groups of students are fully considered in all aspects of the SWPBIS design, implem entation, and continuous improvement processes With appropriate SWPBIS positive strategies equally applied to all sub groups and with appropriate implementation training, this may help to reduce the discrepancies seen in disciplinary data across groups ( e.g., Jones, Caravaca, Cizek, Horner, & Vincent, 2006) In addition to using existing data systems in the design of interventions, the data collection and analysis process fo r implementation with the PBIS model ( Sugai & Horner, 2009a) In some instances, schools may already have data systems to monitor student behavior, such as School Wide Information System (SWIS) This type of system can be used specifically for behavior d ata collection and analysis across all three MTSS tiers that make up SWPBIS In other settings, a state might have a centralized system for collecting behavioral data For instance, in Florida, all schools utilizing SWPBIS approaches have access to the S collection and analysis system called Response to Intervention for Behavior (RtI:B; FLDOE, 2009)

PAGE 80

80 information system to collect and analyze studen t behavior data using a variety of criteria such discipline type (i.e., pre referral or ODR), infraction type, location, time, grade level, and other demographic indicators, as well as across all three tiers of interventions (FLDOE, 2009) In additio n to behavioral data, the PBIS T eam should set up necessary procedures to collect information on utilization of those positive behavior supports that can be reliably measured, which can be useful in exploring how some of the more measurable Exploring the utilization of supports by sub group (e.g., teacher, grade level, race/ethnicity, etc.) could help show deficiencies in the overall implementation of positive supports a nd guide the provision of additional training and encouragement to stakeholders implementing the interventions Again, the focus should be on consistently applying the SWPBIS intervention to all students and in all settings ( Sugai & Horner, 2009a) It c annot be over stated that an important hallmark of SWPBIS is the ongoing process of data collection, analysis, and collaboration of all school staff to adjust priorities depending on the behavioral needs of the school (e.g., Sugai & Horner, 2009a) For exa mple, if data demonstrate students are receiving major disciplinary referrals for behavior incidents occurring on the playground, then interventions, such as re teaching of appropriate playground based social behavior and etiquette would be provided Howe ver, if after three months, new data indicate that major disciplinary referrals were now on the rise on the bus, then appropriate interventions to address social behavior concerns on the bus would be added to the intervention model It is the

PAGE 81

81 school based data driven decision making and the frequent presentation of data to all stakeholders that helps set SWPBIS apart from other approaches ( Sugai & Horner, 2009a) Design considerations: s etting goals and expectations After gaining sufficient staff buy in and collecting initial data and determining which behaviors will be at the initial focus of the SWPBIS interventions, the next important step is for the SWPBIS T eam to work with all stakeholders in the development of school goals and expectations ( Sugai & Horner, 2009a) Based off a close analysis referrals, etc.), a proposed set of school wide rules and expectations of behavior are initially developed by the SWPBIS Team, which is composed of representatives from across all grade levels and instructional areas (e.g., Special Areas, ESE, and Regular Education Teachers) In developing the school wide expectations, the SWPBIS T eam must remember the types of behaviors that are most effectively impacted by SWPBIS In essence, at the most basic foundation of successful SWPBIS i ntervention models, the SWPBIS T eam must give priority to ( Sugai & Horner, 2009a) Problematic behavior that d oes not fit under the model of SWPBIS can still be addressed on an individual basis through individualized PBIS, functional behavior assessment techniques, or other appropriate positive based approach ( Sugai & Horner, 2009a) The goals and expectations est ablished for the school should follow a specific set of rules proposed by the developers and leading researchers of SWPBIS ( Sugai & Horner, 2009a) First, there should be relatively few rules (three to five) to help both students and teachers easily and q uickly recognize these standard expectations

PAGE 82

82 Second, they should be brief and phrased with positive language, thus keeping with the Thi rd, the expectations should be universal for the school environment, including all staff and students, and should have the ability to apply succinct and universal expectation) Fourth, the expectations must recognize the foundational premise that SWPBIS is designed to support academic and behavior outcomes ( Sugai & Horner, 2009a) and these desired outcomes must be the focus of Finally, Sugai and Horner (2009) state that all expectation s should be culturally and contextually appropriate for the students and school. Upon establishing the initial goals and expectations, it is important for all stakeholders to have input into whether the chosen expectations are acceptable across all grade l evels, student sub groups, and settings at the school Once a consensus is reached and the expectations approved by school administrators, they become one of the foundational components of Tier I interventions under the SWPBIS approach (discussed below) In essence, these expectations are taught akin to any academic skill, with all school personnel expected to proactively teach, review, model, and re teach these school wide rules and expectations of behavior This process is especially important during t imes when student behavior is expected or anticipated to escalate (e.g., before the holidays, Spring Break, end of the school year, etc.) As mentioned,

PAGE 83

83 the establishment of school wide rules, expectations, and the teaching and re teaching to students mak e up one of the primary Tier I strategies of SWPBIS, which is expected to positively impact the social and academic success of at least 80% of students within the school (FLDOE, 2016). Once school wide rules and expectations are determined and agreed upon, school personnel review current processes and establish new procedures and protocols for the types of behaviors that warrant a formal, office discipline referral or pre referral It is important to note that referrals are not forbidden under the SWPBIS m odel, and inappropriate behaviors are not ignored when they pose a threat to the student or the learning environment However, it is important that the entire school (all staff and settings) form a unified front (through buy in) and respond consistently t o the same behaviors so if cussing receives a referral in one classroom, then it would also receive a referral in the library or playground (U.S Office of Special Education Programs, 2016) The referral procedures and protocols should be a clearly writ ten, agreed upon by all staff as part of a progressive discipline plan, and be followed consistently throughout the school, regardless of area (i.e., classroom, cafeteria, playground, bus loading area, etc.) If warranted, school forms should be reviewed and updated (or created) such that pertinent behavioral data can be collected and aggregated for meaningful use within the SWPBIS initiative Since SWPBIS is data driven, it is imperative that data include the type of behavior, time of day, and location, thus providing information necessary to make any necessary changes and implement meaningful strategies to better serve students Behavior forms should also include demographics to assist school personnel in any future decision making processes

PAGE 84

84 Design c onsiderations: s taff training As noted earlier in this section, staff buy in is one of the most critical elements of successful implementation of SWPBIS, as it helps to ensure consistent application of interventions and a consistent message to students acr oss all stakeholders and all school settings Difficulties with staff buy in most often result from misperceptions or misunderstandings of the basic foundational theories of SWPBIS, as well as poor understanding of the principles and interventions used wi thin PBIS (Kincaid, Childs, Blase, & Wallace, 2007) Given that successful SWPBIS implementation requires at least 80% staff buy in, it is critical to address these potential challenges to getting staff to approve of the core values and overall PBIS intervention philosophy (Newcomer & Barrett, 2009; Putnam et al., 2009) Staff training is one of the most useful and important methods for promoting staff buy in and ensuring an accurate understanding as to how SWPBIS can help the school and students (e. g., FLDOE, 2016; Sugai & Horner, 2009a) The following three types of training are the most commonly utilized: (1) initial training on SWPBIS; (2) intervention training; and (3) periodic retraining Initial SWPBIS t raining Initial training for SWPBIS is designed to provide a general understanding of the model and associated interventions for school administra tors and members of the SWPBIS T eam at the school or schools planning on designing and implementing an SWPBIS intervention model While intereste d school personnel may consider beginning a school process within their school, the Technical Assistance Center on Positive Behavioral Interventions and Supports established by the U.S Department of Education's Office of Special Education Programs recommends that interested schools and districts contact

PAGE 85

85 the tenants, resources, and training available to successfully implement the approach (Lewis, Barrett, Sugai, Horner, Mitchell, & Starkey, 2016) PBS Project recommends that interested schools first establish buy in at the district level and gain administrator support at the school level With tha t accomplished, the FLDOE (2002c) recommends the district complete the District Readiness Checklist and which includes a first tier training on SWPBIS to include the primary supports that must be developed and implemented before the school can develop and implement supplemental or intensive supports (described in the next subsection of this proposal) Ultimately, the initial training is designed to provide a general understanding of SWPBIS to those who will be leading and/or overseeing the SWPBIS interventions, whether they be district or school personnel It is vital for the SWPBIS Team to have an accurate understanding of SWPBIS prior to developing the school expec tations, data collection processes, selected interventions, and other elements and procedures of the SWPBIS model ( Sugai & Horner, 2009a) In Florida, when training is complete, the district and SWPBIS Team members receive access to online resources, incl uding the Florida RtI:B system for data collection and analysis of behavioral data within their school(s) (FLDOE, 2009) Intervention training of school p ersonnel After the SWPBIS Team has received the initial PBIS training, provided basic information to school personnel and stakeholders as to the basic principles and theory of SWPBIS, developed school wide rules and expectations with input from all stakeholders, and selected evidence and researched S Team or representative trains all school personnel on specifics of the chosen model This

PAGE 86

86 intervention training includes a more in depth overview of SWPBIS theory and potential outcomes, thorough review of behavior expectations for students, and methods for consistently implementing school wide strategies for reducing undesired behaviors This training also provides the SWPBIS Team to review any discipline and behavior protocols (new and old) for addressing and documenting specific behaviors and/or rece training can further improve buy in and ensure consistent application of the model i nterventions Re training through continuous improvement and m onitoring As the SWPBIS initiative and associated interventions are being implemented across all school settings, it is imperative that the implementation and outcomes are tracked and evalua ted through a continuous improvement and monitoring process ( Sugai & Horner, 2009a) There are two important facets of this process: fidelity and feedback First, it is important that the SWPBIS Team explore incoming data ensure fidelity in implementing the selected interventions consistently (and correctly) and to ensure all student sub groups are receiving equal treatment under the model Second, it is highly beneficial for teachers and stakeholders to receive regular feedback about their implementatio n of the interventions, successes and changes in their classrooms, and successes and changes across the entire school Within both the fidelity and feedback processes, the SWPBIS Team is collecting data on how well the SWPBIS model is being implemented an d where specific stakeholders can improve the implementation With such data, school As the SWPBIS approach is malleable and has the ability to be differentiated to the needs of the school, the

PAGE 87

87 appr oach can be viewed as an initiative in ever changing metamorphosis and driven by student behavior data and needs As such, during these refresher trainings, the original SWPBIS approach and interventions can be discussed, as well as any new interventions or behavior expectations that have emerged Staff may be re trained individually, in small groups, or as an entire school, depending on school needs. Multi Tiered System of Supports wit hin the Context SWPBIS Once the school has addressed most the aforementioned design considerations, the SWPBIS Team must begin to design the actual interventions used for each of the three tiers of interventions used within SWPBIS The use of three tiers was adopted from Multi Tiered System of Supports (MTSS), previously known as Response to Intervention (RtI), and is a critical underlying component of the SWPBIS approach (McIntosh & Goodman, 2016; Sugai & Horner, 2009a) More specifically, prior to 2007, SWBPBIS utilized essentially the same framework as MTSS: (1) Primary, Tier 1 whole school interventions impacting 80% of the student population; (2) Secondary, Tier 2 small group interventions impacting 15% of the student population who do not respond to primary tier interventions; and (3) Tertiary, Tier 3 individual interventions impacting the remaining 5% of the student population who do not respond to primary and secondary tier interventions (McIntosh & Goodman, 2016) In 2007, however, the SWPBI 5) of preventative interventions, with less focus on static tiers and labels Still similar to the original tiered system, the SWPBIS intervention continuum moves from universal interventions to targeted to intensive (Lewis et al., 2016) with the level of interventions increasing along with the severity of the behavior(s) at focus (Lewis et al, 2016; McIntosh & Goodman, 2016; Sugai & Horner, 2009a) The purpose of using the th ree

PAGE 88

88 tier system in SWPBIS is grounded in helping to organize and coordinate the implementation of specific behavioral interventions As with MTSS, the three continua remain focused on providing interventions and strategies to impact the entire student bod y, with 80% of the entire student body effectively managed by the universal interventions (Tier 1), 15% by the targeted interventions (Tier 2), and 5% in intensive interventions (Tier 3) (FLDOE, 2016) The primary difference between MTSS and PBIS is the t ypes of interventions used within each tier and the school wide focus of PBIS In essence, PBIS borrows the multi tier continuum from MTSS to help enhance the overall effectiveness PBIS approach at a school wide level (McIntosh & Goodman, 2016). Tier 1 ( p rimary u niversal) i nterventions As noted above, Tier 1 of PBIS is designed to provide universal interventions to reach as many members of the student body as possible, with the theoretical expectation of reaching at least 80% of the students (FLDOE, 2016 ) As such, Tier 1 interventions are relatively broad and focused on several pre established school wide and process of setting goals) In fact, setting these school wid e goals and behavior expectations are one of the first interventions under Tier 1, in addition to teaching and re teaching the rules and expectations to school personnel and students In Tier 1 strategies, all staff are trained on how to address student b ehavior by focusing on positives and alternative ways of addressing inappropriate behavior For example, instead of classroom teachers or school staff pointing out those students who are not sitting and waiting quietly, they would find a student model exa mple and focus on the appropriate behavior of that student Poor behaviors are not ignored, but they are addressed in more individualized and approp riate ways, with the emphasis on teaching

PAGE 89

89 more appropriate replacement behaviors and not resorting to offic e disciplinary referrals as a means to remove the student from the (theoretically) desirable atmosphere of the classroom environment and activities It is important to note that Tier 1 interventions under SWPBIS bring forth several concepts from the foun dational theories discussed earlier in this chapter For instance, the concepts of operant conditioning explain that human behavior is learned through a system of rewards and punishments, both positive and negative (Miltenberger, 2012), with the most effe punishment (McConnell, 2001; Miltenberger, 2012) Equally important is social learning theory (e.g., Bandura, 1971a), the concept of modeling where children encode modeled behaviors from a wide v ariety of sources (Bandura, 1963; Bandura, 1965), and the concept of vicarious reinforcement ( George, et al., 2009 ) In addition, SWPBIS applies operant conditioning and social learning theory concepts through the token economy (e.g., Kazdin, 2012b), all of which include the token (secondary reinforcer), that for which tokens are exchanged (back up reinforcers), and the targeted behavior which is desired to be increased (Kazdin, 2012b) As detailed earlier, the token economy is essentially a systematic ap proach to reinforcing and increasing desired behaviors by establishing a relationship between a secondary and back up reinforcer. Specific to SWPBIS, as part of the Tier 1 intervention system, both tangible and non tangible incentives are provided to stude nts to recognize appropriate behavior Tangible incentives in SWPBIS models usually operate under a token economy and interest items such as bubbles, stickers, pencils, balls, art sets, and bracelets

PAGE 90

90 Borrowing from the concepts of token economies, these tokens have no intrinsic value to the students beyond their exchangeability for the items they desire from the school store Having a wide array of back up reinforcers in the PB I S Store helps avoid satiation and ensures a school wide token economy will contain at least one back up reinforcers found valuable by each student If any single student does not find at least one of the back up reinforcers valuable, the token economy will fail for that student as the token is only reinforcing if it can be exchanged for a valuable reinforcer These tickets or tokens can be given by any school personnel to students when that are ctations or when they are behaving in a specifically desirable way (e.g., sitting quietly with hands in their lap) The tokens should be given with enough frequency to sustain the appropriate behavior being opriate behavior It is important to note that providing reinforcement for every instance of a behavior in a classroom does not align with behavior modification practices determined to be the best for long lasting change, as continuous reinforcement is le ss robust than intermittent reinforcement against extinction when a reinforcement is missed (Kazdin, 2012a) Staff should be In addition to tangible rewards, Tier I SWPBIS relie s heavily on non tangible reinforcements Non tangible reinforcements have been found equally reinforcing within both direct reinforcement and vicarious reinforcement paradigms (Kazdin, 2012a) As with tangible reinforcements, non tangible reinforcements are also given to students that are caught demonstrating appropriate behavior in keeping with the school wide behavior expectations and goals Non tangible rewards can consist of verbal praise,

PAGE 91

91 non verbal praise, affirmative gestures, and physical contac t (as appropriate, such as a pat on the back) In some cases, schools may decide to have monthly or quarterly SWPBIS celebration activities (i.e., recognition ceremonies, parties, special lunches, caution is warranted, as such celebrations might not be reinforcing for all students Using non tangible reinforcement, the teacher only needs to reinforce a positive behavior in one student, making a clear announcement that the reinforcement was given, s tating the behavior that was reinforced, and then waiting for other students to demonstrate the same behavior At that point, the teacher can provide additional tangible or non tangible reinforcements, but it is not absolutely necessary, as the reinforcem ent already occurred vicariously for all students who were paying attention or had their attention grabbed by the announcement of the reinforcement ( George, et al., 2009 ). Tier 2 (secondary t argeted) i nterventions The secondary level of the SWPBIS, Tier 2, consists of working with small groups of students using more targeted interventions Depending on the behavior concerns, some Tier 2 interventions can be applied to individual students Tier 1 interventions are still provided to those students needing Tier 2 interventions, as they are still part of the overall school culture ( Sugai & Horner, 2009a) However, students receiving Tier 2 interventions will have additional interventions associated with their individualized needs For instance, the school may develop a point sheet for those students having difficulty remaining in their seats, which can be customized to any specific behaviors that need increased attentio n Other Tier 2 interventions may also consist of academic supports (e.g., small group instruction, tutoring, academic

PAGE 92

92 mentoring) to assist students whose behavior leads to academic deficiencies or social skills groups where students with similar behavior concerns can work on specific skills (e.g., communication) As a result of these interventions, students may have better and more appropriate interactions with adults and/or peers and reduce verbal altercations and confrontations The school may provid e check in/check out supports for students who need more one on one adult support to keep them on track and give them feedback on their behavior by reinforcing desired behavior and helping them to recognize and decrease undesirable behavior ( Fairbanks, et al. 2007) Tier 2 interventions can also include more comprehensive and complex versions of the token economy provided under Tier 1, though more individualized to address specific behaviors ( Sugai & Horner, 2009a) Tier 2 interventions most often take place with a much larger team of individuals with more specialized expertise, such as the school psychologist and occupational therapists, in addition to the student, classroom teacher, and administrator responsible for behavior ( Sugai & Horner, 2009a) S tudents with similar behavior concerns may work in small groups with the school counselor and/or administrator on group character development and lessons on appropriate behavior Such targeted interventions often involve the parent to support the more ind ividualized behavior intervention, and researchers recommend a regular method of communication between all stakeholders (e.g., students, parents, teachers, and administrators) It is expected that an intervention at this level would only be utilized for a pproximately 5 10% (Sugai, 2016), up to 15% of the overall student body at the school (McIntosh & Goodman, 2016)

PAGE 93

93 Tier 3 (tertiary intensive) i nterventions The tertiary level of SWPBIS, Tier 3, consists of highly intensive one on one interventions provided by one or several school personnel, including the classroom teacher, school counselor, administrator for behavior, outside counseling resources (if needed), parents, and possibly additional district level personnel specializing in behavior Tier 1 and Tier 2 level interventions may also still be used with students who are receiving Tier 3 interventions, but only if they are useful and appropriate ( Sugai & Horner, 2009a) At this level, which is only intended for between 1% (Sugai, 2016) and 5% (M cIntosh & Goodman, 2016) of the student population, monitoring of behavior would include frequent check ins by staff There is a higher potential for the student to be moved to a more appropriate educational setting, such as an alternative education setti ng with additional behavioral staff resources and outside behavior counseling services, when appropriate Students in need of Tier 3 supports are those students that demonstrate behavior that is chronic, highly disruptive, results in educational and socia l Sugai & Horner, 2009a) Tier 3 interventions are not further detailed within this chapter, as they are far outside the scope of this research study Impacts and Outcomes of SWPBIS Since the 1980s, a great deal of research has shown the positive impacts of SWPBIS on student behavior, the learning environment, and academic achievement (e.g., Sugai & Horner, 2006) Following the inclusion of Positive Behavioral Interventions and Supp orts (PBIS) in the IDEA Amendments Act of 1997 (20 USC §1400 et seq.), the number of research studies supporting the effectiveness or the processes and interventions associated with PBIS have blossomed (Dunlap, et al.

PAGE 94

94 2009) As PBIS expanded to School Wi de PBIS, the majority of research continued to show positive impacts of implementing the SWPBIS model under the aforementioned continuum of supports (e.g., Fairbanks, et al. 2007; George, et al., 2003 ; Nocera, et al., 2014 ; Scott & Barrett, 2004; Waadsorp et al., 2012 ) Although some research and findings have been discussed in prior sections, it is important to specifically address some of the most recent research regarding the primary elements of SWPBIS under investigation within the present research While not all research discussed within this section is specific to SWPBIS, the studies are generally focused on behavior modification, positive reinforcement, or other specific elements of the SWPBIS model Learning Environment and Behavior As mentione d earlier in this chapter, staff buy in is among the most important first steps to designing and implementing a successful SWPBIS intervention model (Newcomer & Barrett, 2009; Putnam et al., 2009; Sugai & Horner, 2009a) Such buy in requires sound argumen ts and explanations as to how the interventions will support both the students and the teachers to help accomplish the annual goals associated with academic achievement and social learning As such, one of the most useful and impactful arguments for using SWPBIS is that it can positively influence the classroom and school learning environment (e.g., Adelman & Taylor, 2002 Dorman & Adams, 2004; Fraser, 1998; Freiberg, 1999) The learning e nvironment There are certainly many theories and concepts & Cunningham, 2011; Moos, 1979; Patrick, Ryan, & Kaplan, 2007), but it is often considered one that is safe, positive, in tellectually stimulating, interesting, and culturally sensitive (among many other concepts proposed by hundreds of theorists and

PAGE 95

95 researchers) Miller and Cunningham (2011) attempt to break down the environment into three interrelated components: (1) the p hysical environment (e.g., classroom management, collaboration, cooperative groups, etc.); (2) the psychological environment (e.g., interaction between teacher and students, student participation, communication of goals, etc.); and (3) teacher characterist ics and behaviors (e.g., professional development, school culture, etc.) Each of these components could be addressed within a SWPBIS model either directly or indirectly (e.g., improving student behavior could improve each of these) Moos (1979) presents a slightly different three dimension model of strong classroom learning environments, which has been referenced heavily in literature: (1) Relationship Dimension (e.g., nature and intensity of personal relationships); (2) Personal Development Dimension (e .g., opportunities personal growth and self development); and (3) System Maintenance and Change Dimension (e.g., orderly, clear expectations, control, and responsive) Moos (1979) dimensions have even been used in the development of research instruments t o measure perceptions of classroom environments (e.g., Fraser, 2002) Other researchers have found a positive relationship between student motivation and engagement and their perceptions of the degree of social support they felt in the classroom environme nt (Patrick, Ryan, & Kaplan, 2007) Regardless of how the learning environment is defined and measured, the general consensus is that finding ways to maximize and support the learning environment will support the social development and ultimate academic achievement of students Indeed, over the last 40 years, a great deal of focus has been placed on improving the classroom environment, and research has found a positive relationship to desired student outcomes, such as motivation, engagement, and learning (e.g.,

PAGE 96

96 Adelman & Taylor, 2002 Dorman & Adams, 2004) Other researchers have found significant relationships between classroom climate and student behavior, self efficacy, social emotional development, achievement, burnout, and overall perceived school quality (e.g., Fraser, 1998; Freiberg, 1999) However, while it is important to understand how a good learning environment supports students, teachers, and schools, it is also important to understand what influences the learning environment either posit ively or negatively Impacts of student behavior and d iscipline In an attempt to understand what influences the learning environment, many studies have focused on the deleterious influence of negative student behavior on efficacy and efficiency of the school and classroom learning environments Indeed, nearly every study exploring classroom behavior management and school culture begins with the same basic tenet poor student behavior distracts from and negatively impacts the learning environment (e.g. Guardino & Fullerton, 2010; McKevitt, et al. 2012; Scott & Barrett, 2004; Sugai & Horner, 2006) The deleterious effects can be insurmountable for some teachers, including difficulty managing and controlling the classroom, difficulty teaching and prese nting content to the students, loss of focus of students or the entire classroom, and a reduction in the amount of time on task When the teacher stops teaching to deal st udent and other students who attend to the distraction Even more time can be lost for students who are sent to the office on a referral or suspended from school In fact, Scott and Barrett (2004) performed a thorough review of data from an elementary s chool in Baltimore and found that office disciplinary referrals resulted in removal of a student from the classroom for an average of at least 20 minutes (nearly

PAGE 97

97 6% of the learning time provided during the school day), while a suspension resulted in a loss of at least 6 hours of instructional time for each day of suspension Not surprisingly then, according to the National Association of Elementary School Principals, dan students are a barrier to effective education in their classrooms (Public Agenda, 2004) Moreover, the US Department of Education (USED) revealed that secondary schools suspend or expel over 2 million students each year, which USED Secretary Duncan George, 2014) Disproportional i mpacts of discipline on mino rity s tudents The negative impacts of disciplinary procedures for problematic behavior is even more pronounced for students from traditionally defined minority groups and those identified with a special need or disability This fact was underscored by U S Department of Education Secretary Duncan and US Attorney General Holder in 2014, when they revealed that minority students and those with disabilities were disproportionately impacted by school discipline with African American more than three times mor e likely to be suspended or expelled than their w hite peers (USED, 2014) The Discipline Disparities Collaborative (Skiba, Arrendondo, & Rausch, 2014) reported consistent findings that African American students and st udents with disabilities have dispropo rtionately high er rate s of office disciplinary referrals (ODRs), suspensions, and expulsions Importantly, the researchers found that the differences were not fully explained by poverty or rates of misbehavior

PAGE 98

98 This disproportionality has been supported by a number of studies completed in recent years, and none of the studies found any other student demographic, behavior, or school factors to explain the disproportionate disciplinary outcomes (e.g., Eitle & Eitle, 2004; Gonsoulin, Zablocki, & Leone, 2012 ; Rocque, 2010) Other researchers have found African American children to be disproportionately placed in special education settings (e.g., Blanchett, 2014) In general, most researchers looking at disparities and disproportionalities for students from traditional minority groups and those identified with disabilities have also focused on factors that might have caused these social issues Aside from the very real issues of white privilege and racism discussed by some theorists (which are beyond the sco pe of this research), among the most prominent reasons behind the disproportional discipline outcomes are a lack of teacher preparation and training, a focus on punishment and discipline as the first response, and misperceptions of the cultural aspects of students and their communities (e.g., Blanchett, 2014; Gonsoulin, Zablocki, & Leone, 2012; Skiba, Arrendondo, & Rausch, 2014; USED, 2014) Impacting Behavior with SWPBIS Interventions The overwhelming focus on using punishments and severe disciplinary pr ocedures in attempts to control student behaviors and enforce an effective learning environment has met with a great deal of criticism, from the aforementioned to 2011) In fact, the Washington Post quoted USED Secretary Duncan as stating that adding that zero unwelcome in t

PAGE 99

99 have a significant and lasting negative effects on the long term well being of our young George, 2014) Fortunately, other well documented options exist for helping manage s tudent behavior without reliance on reactive and punitive discipline procedures In fact, Walker and Shinn (2002) make a concerted argument as to the importance and overwhelming success of prevention interventions to stop inappropriate behavior patterns b efore they have an impact on the environment, but concede that the greatest challenge is finding the willpower to actually implement preventative strategies One such proactive, preventative, and positive focused intervention model is School Wide Positive Behavioral Interventions and Supports (SWPBIS), which was discussed at length earlier and is the focus of the present research Using the most global statement possible, SWPBIS has been repeatedly and consistently found effective in reducing the instanc es of negative student behaviors across all student sub groups, communities (e.g., urban, rural, etc.), and all student grade levels (e.g., Fairbanks, et al. 2007; George, et al., 2003 ; Nocera, et al., 2014 ; Scott & Barrett, 2004; Waadsorp, et al., 2012 ) This section provides an overview of some such studies, but a complete review of the hundreds of studies that support SWPBIS as a model for reducing problematic student behavior would be outside the scope of this literature review One such study relate d directly to the concepts of student behavior and lost learning time was conducted by Scott and Barrett (2004) at an elementary school in Baltimore, Maryland Scott and Barrett (2004) first provided training and guided the development of a comprehensive SWPBIS model and measured a number of outcome variables, including office disciplinary referrals (ODRs) and school suspensions The study found

PAGE 100

100 that ODRs decreased from 608 at baseline (the year prior to implementation) to 108 in the first year and 46 in the second year of SWPBIS implementation Based on their finding that each ODR resulted in an average of 20 minutes of lost instructional time, this resulted in a decrease from 12,160 minutes of lost instruction at baseline to 2,160 in the first year and 920 in the second year an average gain of 10,620 minutes of instructional time per year In addition to ODRs, the study found that student disciplinary suspensions decreased from 77 at baseline to 32 in the first year and 22 in the second year of imple mentation Assuming each suspension results in a loss of 6 hours of instructional time, this resulted in a decrease of lost instructional time from 462 hours in the baseline year to 192 hours in the first year and 132 hours in the second year of implement ation an average gain of approximately 50 days of instruction each year across the school (Scott & Barrett, 2004) While Scott and Barrett (2004) calculated the cost to the learning environment of suspensions and ODRs, most studies have not gone to that extent and assume any reduction in these disciplinary processes are beneficial to both the student and the learning environment Similar to Scott and Barrett (2004), George, White, and Schlaffer (2003) explored the impact of applying a SWPBIS model to a n elementary school over the course of two years and found a similarly impressive decrease in ODRs More specifically, ODRs decreased from 1,717 during the baseline year to 702 during the first year of SWPBIS implementation and 619 during the second year These researchers also explored afterschool detentions, which decreased from 845 in the baseline year to only 85 in the first year of SWPBIS implementation The researchers further discovered that the impact was even more pronounced for students with IE Ps, decreasing from 298 of the

PAGE 101

101 total ODRs in the baseline year (17.3% of the total) to only 34 in the first year (4.8% of the total ODRs) In another study, Spencer (2015) worked with a small intermediate school to implement SWPBIS after a large number of students were increasingly being sent to the office on ODRs due to disruptive behavior After implementation of the SWPBIS model, the research found a significant decrease in the number of ODRs within each of the eight primary offense categories indicate d on ODRs (e.g., disrespect, refusal to obey, physical contact, profanity, etc.) decreasing from 593 in the baseline year to 268 after the first year of SWPBIS Menendez, Payne, and Mayton (2008) conducted a one year study of a new SWPBIS project at an elementary school and found a reduced number of ODRs, fewer rules based violations of students, and fewer punitive consequences used by teachers (e.g., time outs, written reprimands) Luiselli, Putnam, Handler, and Feinberg (2005) found that SWPBIS result ed in fewer ODRs and school suspensions over the course of two years, with an average of 1.3 ODRs per day at baseline decreasing to 0.7 per day in the first year of implementation and 0.5 per day by the second year of implementation Suspensions also sign ificantly decreased from 0.3 per day in the baseline to 0.25 in the first year and 0.2 in the second year of SWPBIS implementation As noted above, it is impossible to review all the research that shows a positive impact on student behavior after impleme nting a SWPBIS intervention model through the reduction in ODRs (e.g., Bohanon et al., 2006; McCurdy, Mannella, & Eldridge, 2003; Sprague, Walker, Golly, White, Myers, & Shannon, 2001; Taylor Greene, et al., 1997) However, Vincent, Swain Bradway, Tobin a nd May (2011) conducted a study with one of the largest number of elementary schools and is important to mention due to

PAGE 102

102 the unique focus on racial disproportionality The study was an archival analysis of 153 elementary schools, with 72 demonstrating a hi gh fidelity implementation of SWPBIS (as demonstrated by self evaluations or external evaluations) and 81 not implementing SWPBIS (or a low fidelity implementation) Data were collected on major ODRs and student demographics over a three year period, with a requirement being the schools collect and report major ODRs by student race/ethnicity Consistent with other research, the researchers found that the schools implementing high fidelity SWPBIS had significantly fewer rates of ODRs across all three years as compared to the schools not implementing SWPBIS However, this study was focused on determining whether discipline disproportionality was impacted by SWPBIS interventions While the findings showed that a disproportional number of minority students c ompared to white students received ODRs in all schools, regardless of SWPBIS, an important finding is that the level of disproportionality was lower in schools with SWPBIS The researchers suggest that the findings demonstrate the potential of SWPBIS to h elp decrease the discipline gap that currently exists in education (Vincent, Swain Bradway, Tobin, & May, 2011). While much research has been completed at the elementary school level, there are also numerous studies supporting the use of SWPBIS in middle s chools and high schools For instance, Nocera, Whitbread, and Nocera (2014) found that a comprehensive SWPBIS model could be combined with a school wide model of data driven decision making and data teams to have positive impacts on student behavior with middle school students The researchers found that the number of teacher discipline referrals and student suspensions were reduced, including significant improvement among students identified with disabilities In another study, Caldarella

PAGE 103

103 and colleagues (2011) collected data from two middle schools, with one implementing SWPBIS for a period of four years and the other serving as the quasi control Ultimately, the researchers found that teachers provided higher ratings of the school climate at the interv ention school, while teachers in the control school reporting the same or worse feelings regarding the school climate More specific to student behavior, the students in the intervention school had significantly fewer tardiness, absences, and ODRs for dis cipline issues Lassen, Steele and Sailor (2006) explored the implementation of SWPBIS in an urban middle school over the course of three years, finding that the average number of referrals actually increased in the first year of implementation (from 5.22 per student at baseline to 6.84 per student in year 1) However, by the third year of implementation, the average number of referrals per student had decreased to 3.7, which represented a significant decrease from the first year of implementation As su ch, not all studies have shown an immediate improvement in student behavior High schools have also found success implementing SWPBIS interventions, as demonstrated by research such as that completed by Flannery, Fenning, Kato, and McIntosh (2014) Thes e researchers conducted a three year effectiveness trial using 12 high schools, with eight implementing SWPBIS and four serving as a control comparison Overall, the researchers found significant decreases in the number of student ODRs within the SWPBIS s chools and increases in the schools not implementing SWPBIS (Flannery, Fenning, Kato, & McIntosh, 2014) Still other researchers have explored entire school districts and found significant impacts of SWPBIS across all grade levels, such as decreases in problematic behaviors,

PAGE 104

104 decreased numbers of referrals, and increased perceptions of school safety (e.g., McI ntosh, Bennett, & Price, 2011) Certainly, not all studies have explored the impact on student behavior using standardized school data, which has provided for a better understanding of some of the more nuanced changes that result from the implementation of SWPBIS For instance, Waadsorp, Bradshaw, and Leaf (2012) performed a randomized control effectiveness trial to explore the impact of SWPBIS using 37 elementary schools and over 12,000 children The researchers found that children in schools with SWPB IS had lower teacher reported rates of bullying (both the number of children who bullied and were bullied) and were reported by teachers to be less rejected by peers than in schools without SWPBIS It is noted the researchers used a single observation che cklist to assess the impact of SWPBIS, rather than school discipline records However, from the same research program and longitudinal study, schools implementing SWPBIS showed a significant reduction in the number of school suspensions and ODRs over the course the five year study (Bradshaw, Mitchell, & Leaf, 2010) In addition, not all studies use the full SWPBIS approach though there are many studies that have explored whether similar aspects of behavior management are effective in impacting student b ehavior For instance, Fairbanks, Sugai, Guardino, and Lathrop (2007) conducted a small study of two 2nd grade classrooms to explore whether RtI could be effective in reducing inappropriate student behaviors The study in / ch eck behavior reports) designed to provide additional structure and feedback to students Overall, the researchers found that the procedure was effective in reducing negative

PAGE 105

105 student behavior such as inappropr iate physical contact, talk outs, inappropriate placement, noncompliance, and non disruptive off task behavior ( Fairbanks, et al. 2007) Office disciplinary referrals as primary metric of b ehavior In reviewing the hundreds of research studies explorin g SWPBIS and student behavior, it becomes apparent that only a handful of outcome metrics are used to assess impact on the learning environment Indeed, behavioral data are the most commonly used metrics to assess the impact of SWPBIS under continuous imp rovement models, as they are data readily available to the school administrators and teachers These metrics traditionally include in school suspensions, out of school suspensions, expulsion, detentions, office disciplinary referrals, school attendance, a nd tardiness However, not all of these are appropriate for elementary schools, as some are very rare with ES students (e.g., expulsions and detention) while others are generally out of the control of the younger students (e.g., attendance and tardiness) However, what can be readily measured on a continuous basis (daily, if needed) are office disciplinary referrals (ODRs), and these have become the most common method for assessing behavioral change within schools implementing school wide preventative int erventions With ODRs being one of the primary measures of student behavior impacts in research on SWPBIS interventions, it is important to understand whether such a metric is a valid representation of student behavior Certainly, ODRs have been shown t o be related to future behavior problems, such as drug use and disorderly conduct in classrooms (e.g., Nelson & Roberts, 2000; Sprague et al., 2001) However, not all research has been so clear Nelson, Benner, Reid, Eptsein, and Currin (2002) utilized

PAGE 106

106 t he Teacher Report Form (TRF) to assess whether students receiving ODRs met the clinical and borderline cut scores on one or more of the eleven scales and subscales (e.g., withdrawn, social problems, attention problems, delinquent behavior, aggressive behav ior) Unfortunately, students who scored at the clinical and borderline levels on the TRF scales did not necessarily receive ODRs, while students who received ODRs did not necessarily score at these levels on the TRF For instance, of those students with ODRs, only 3.1% were found to be clinically aggressive and 2.5% were found to have clinically impaired attention The results from the Nelson et al (2002) study calls into question the convergent validity of using ODRs as a true measure of student beha vior, though the authors admit to the substantial difficulties in researching validity of a variable with so many uncontrollable contextual variables To help address some of the gaps in research on the validity of using ODRs, Pas, Bradshaw and Mitchell (2011) collected data from 21 elementary schools implementing high fidelity SWPBIS intervention models The researchers collected ODR data from a centralized database, ODR data from a teacher survey, and scores from the Teacher Observation of Classroom Ad aptation Checklist (TOCA C) Overall, the findings revealed that the centralized database of ODRs and the teacher report of whether students were referred to the office were moderately correlated This raises some question about using a centralized datab ase, where the school administrators might not record every instance of a referral to the office (e.g., many administrators do In fact, over 40% of the students where teachers reported having se nt to the office at least two times had one or no ODRs recorded in the centralized system In terms of convergent and divergent validity, the study found ODRs to be positively correlated with both the disruptive

PAGE 107

107 behavior and attention problems subscales, as well as negatively correlated with the prosocial behavior subscale of the TOCA C Overall, the researchers determined that ODRs have moderate convergent and divergent validity with behavior ratings on the TOCA C ( Pas, et al., 2011 ) Other researchers have also found validity for using ODRs as a measure of general behavior (e.g., Irvin, et al., 2004 ; Scott & Barrett, 2004), with ODRs being related to general misbehavior at school, school attendance, student and teacher perceptions of safety and victimiz ation, classroom orderliness, juvenile delinquency, and behavior disorders Ultimately, Irvin et al (2004) found that higher rates of ODRs was related to higher levels of a variety of problematic behaviors in school Impacting Academic Achievement with SWPBIS Interventions The impact of disruptive student behavior on the learning environment and impact of SWPBIS on reducing such behaviors cannot ignore the importance of one of the highest stake metrics used in the United States academic achievemen t using standardized achievement tests (Byrd & Vinson, 2008) Most certainly, academic achievement is one of the most controversial and most important metrics in schools under current law and administration In general, if disruptive behavior negatively impacts the learning environment and reduces time on task, then it would also theoretically impact academic achievement, as implied by US Department of Education Secretary Duncan (USED, 2014) There are a number of research studies that have explored the impact of SWPBIS or associated components on academic achievement in reading, mathematics, and science Some research supporting the use of SWPBIS methods did not specifically explore the implementation of SWPBIS, though did explore behavior management

PAGE 108

108 m ethods that directly relate to those provided under SWPBIS One such study conducted by Abidsereshki, Abkenar, Ashoori, and Mirzamani (2015) compared the effectiveness of tangible and social reinforcements with 8th grade female students identified with in tellectual disabilities in science Although the research was completed in Iran, the results are meaningful and the researchers ultimately found that reinforcements were effective in improving academic achievement Interestingly, the researchers found ta ngible reinforcements to be significantly more effective at increasing student achievement scores in science than social reinforcements, while both were significantly better than the control (no reinforcements) While this research did not specifically re search SWPBIS, the use of tangible and social reinforcements are an integral component of many SWPBIS intervention models. Cotton (2001) completed a meta analysis of 37 reports exploring several interventions common to SWPBIS, such as praise, symbolic rewa rds (e.g., gold stars), token rewards (e.g., points, tickets), tangible rewards (e.g., edibles), and activity rewards (e.g., free time, field trip) Overall, the analysis found that contingent reinforcement was positively related to achievement, while non contingent reinforcement was not In addition, rewards for correct responses was related to achievement, but there was no added benefit for combining rewards with response cost, suggesting response cost is an unnecessary addition to a behavior management program (and rarely part of a SWPBIS intervention) On a similar topic, Broughton and Lahey (1978) is an older study, but found that both positive reinforcement and response cost resulted in improved mathematics performance for elementary school students in a remedial math course, though a combined approach was less impactful as either technique alone

PAGE 109

109 While a number of studies have explored techniques similar to those used in SWPBIS, there are many studies that have specifically explored the impact of SWPBIS on the academic achievement of students For instance, Menendez, Payne, and Mayton (2008) found that within one year of implementing a SWPBIS intervention in an elementary school, the students demonstrated significantly higher scores on the state m andated academic achievement test for 3rd grade students (the school was only K 3) Using a within subjects design, Luiselli, Putnam, Handler, and Feinberg (2005) found that students attending a school with a newly implemented SWPBIS intervention model de monstrated significant improvement in academic achievement, with these students demonstrated an average of 18 percentage point improvement in reading comprehension and 25 percentage point improvement in mathematics know ledge. Nocera, Whitbread, and Nocera (2014) found that middle school students in a school with SWPBIS improved their scores on statewide achievement tests by an average of 25% in reading and 11% in mathematics, suggesting the primary change factor was the implementation of a new SWPBIS interv ention model Lassen, Steele, and Sailor (2006) found that academic achievement of middle school students significantly increased from the first year of implementing a new SWPBIS intervention to the second year of implementation, with an additional increa se in the third year of implementation Furthermore, other studies have explored entire school districts and found increased academic achievement across all grade levels following the implementation of SWPBIS (McIntosh, Bennett, & Price, 2011) Impact o f fidelity on SWPBIS o utcomes Fidelity within SWPBIS implementation is the extent to which school personnel and administrators are faithful to

PAGE 110

110 the established approach and specific interventions designed to impact the school and classroom learning enviro nments Several studies exploring the impact of SWPBIS on behavioral and academic outcomes have also included a measure of the fidelity of the implementation For instance, Flanner, Fenning, Kato and McIntosh (2014) studied eight high schools implementin g SWPBIS and found that an increase in the fidelity of implementation was associated with a significant decrease in the number of ODRs for inappropriate student behavior Other studies have found that schools implementing SWPBIS with high fidelity showed a significant reduction in the number of school suspensions and ODRs over a five year period than did those schools with lower fidelity implementation (Bradshaw, et al. 2010) In a somewhat unique study showing an initial worsening of student behavior a fter first implementing SWPBIS, Lassen, Steele and Sailor (2006) found that the average number of referrals had significantly decreased by the third year implementation Importantly, the researchers measured fidelity of implementing components of SWPBIS u sing the School wide Evaluation Tool (SET; Horner et al, 2004; Sugai, Lewis Palmer, Todd, & Horner, 2001) and found that fidelity was correlated to the disciplinary outcome data with the school only implementing 24.97% of the critical SWPBIS components i n the first year, but 69.64% in the third year Or more connection to the present study and specific to Florida, 54 of 67 (81%) school assistance on providing SWPBIS In the most recent annual report, the project found that schools with higher fidelity had 35% fewer ODRs and 36% fewer out of school

PAGE 111

111 Impact of training on SWPBIS i mpleme ntation The impact of training on the high fidelity implementation of SWPBIS intervention models cannot be understated, with one of the critical components of the design and implementation process being the training of all school personnel and administra tors on SWPBIS (FLDOE, 2016; Sugai & Horner, 2009a) While a number of studies speak to the importance of training or include training as a core component of the studied intervention model (e.g., Scott & Barrett, 2004), fewer studies actually explore the impact of the training itself on the fidelity of SWPBIS implementation For instance, Bradshaw and colleagues (2008) conducted a randomized trial with 37 elementary schools to explore whether training would have an impact on successful implementation of a comprehensive SWPBIS intervention model The researchers utilized the School wide Evaluation Tool (SET; Horner et al, 2004; Sugai, et al. 2001) to measure how well the schools were implementing the various components of SWPBIS with fidelity Overall, t he researchers found that schools receiving training in SWPBIS implemented the model with significantly higher fidelity than did those schools that were not formally trained, both during the first and second years of implementation (Bradshaw, Reinke, Brown Bevans, & Leaf, 2008) Summary For thousands of years, the human race has been concerned with how to educate youth to be productive citizens and useful members of society (e.g., Aristotle, 350 B.C./2000) While society has changed and the extent of ed ucation content has The field has responded with a plethora of educational theories and reform initiatives designed to

PAGE 112

112 address challenges determined the most deleterious to the education of youth For at least the past 100 years, changes in educational environments and class room composition have resulted in an increased focus on effective management and control of behavior to maximize the effectiveness of the learning environment The focus on student behavior and the learning environment became increasingly important as Ame rica progressed through the deinstitutionalization movement of the 1950s, the civil rights movement of the 1960s, the increasingly negative views of corporal punishment s tudent classrooms and associated laws of the 1980s and 1990s Fortunately, as these movements and changes were exerting increasing influence on the education environment, researchers and theorists were developing or adopting a number of behavioral manage ment techniques, such as Applied Behavior Analysis (ABA), Functional Behavioral Assessment, and Positive Behavioral Interventions and Supports Using theoretical foundations from classical conditioning (e.g., Kazdin, 2012a), operant conditioning (e.g., Mi ltenberger, 2012), and social learning theory (e.g., Bandura, 1971a), these techniques drew heavily upon such interventions and behavioral concepts as modeling (e.g., Bandura, 1961; Bandura & Huston, 1961), vicarious reinforcement (e.g., Bandura, 1977), an d token economies (Kazdin, 2012b) Unfortunately, the use of aversive stimuli and punishments began to increase across the country under the use of ABA, as did the moral objections to the use of some techniques (e.g., shocking, water spray, and physical p unishment). These moral objections and documented permanent damage caused to some

PAGE 113

113 behavior interventions and the development of PBIS as an independent field (Singer & Wa ng, 2009) In essence, PBIS is considered a broad approach designed to provide school personnel with tools and skills necessary to enhance student academic OSEP Technical As sistance Center on PBIS, 2016) Some of the most prominent application of research based behavioral strategies; (2) use of multiple and integrated intervention elements alig ned to the needs of the environment; (3) a commitment to sustained and long term outcomes; and (4) support within the organizational systems that ensure sustained impacts of the interventions (Dunlap, et al. 2009) School Wide PBIS (SWPBIS) applies the P BIS technology to focus less on individual students, and more on behaviors across the entire school of students to enhance the learning environment The development and application of SWPBIS has been largely supported with a number of research studies ac ross the world These studies have shown that SWPBIS can significantly reduce the incidents of negative and disruptive student behavior as measured by such metrics as office disciplinary referrals (ODRs; e.g., George, et al., 2003 ; Nocera, et al., 2014 ; P utnam, Handler, & Feinberg, 2005; Scott & Barrett, 2004; Spencer, 2015), school suspensions (e.g., Nocera, et al., 2014 ; Putnam, Handler, & Feinberg, 2005; Scott & Barrett, 2004), and incidents of student bullying ( Waadsorp, et al., 2012 ) In addition, SW PBIS has been shown effective in improving the academic achievement of students (e.g., Abidsereshki, Abkenar, Ashoori, & Mirzamani, 2015; Broughton & Lahey, 1978; Cotton, 2001; Luiselli, Putnam, Handler, & Feinberg, 2005;

PAGE 114

114 Menendez, Payne, & Mayton, 2008) As noted in the prior chapter and outlined in the next chapter, the present research addresses several gaps within past research: (1) using internally collected data on all behavior instances at the school studies ; (2) using (ODRs) ; (3) comparing t he level of tangible reinforcements provided and the behavioral impacts mentioned above ; and (4) analyzing tangible reinforcements, major referrals ( ODRs ), and minor referrals by race and ethnicity, with all data conne cted to the individual demographics of each student Overall, the method of addressing these primary gaps and the results of the present research should help inform practice and guide future research in the early implementation of SWPBIS.

PAGE 115

115 CHAPTER 3 METHOD OLOGY Purpose of Study This study was designed to investigate the initial impacts of implementing a School Wide Positive Behavioral Interventions and Supports (SWPBIS) model on negative student behaviors among elementary school students in Florida The study was conducted during the first year of implementing the SWPBIS model, which was specifically designed to decrease instances of two primary types of disciplinary referrals requiring behavioral intervention by school administrators (i. e., minor pre referrals and major office discipline referrals) The study adds to an already growing body of research on the utilization of SWPB I S models and techniques, with a somewhat unique exploration of the relation between tangible positive reinforcements and discipline referrals (ODRs and minor/ pre referrals ) As noted in the previous chapters, this study is founded with an enhanced purpose to explore demographic differences with both reinforcements and referrals provided to students, as it is imperative tha t school personnel and administrators consider cultural differences and internal biases when applying SWPB I S reinforcements and disciplinary consequences. Research Design The present research study was designed as a formative evaluation study, with a focus on forming and informing the design of a SWPBIS model for elementary school students at the selected school As a formative evaluation study, the intervention and associated processes for implementing the SWPBIS model were investigated during the develop ment of the model, rather than after the model had been fully implemented In fact, some research on SWPBIS approaches suggests that full implementation of the

PAGE 116

116 school wide model (and associated changes in the philosophy as to how schools handle inappropri ate student behavior) can take from between three to five years (Florida Department of Education [FLDOE], 2008; Stormont, Lewis, Beckner, & Johnson, 2008) However, other research and theorists suggest that schools can see positive impacts on student beha vior even within the first year of implementation, though it may still take several years for the philosophy of the school to change (FLDOE, 2008; FLDOE, 2005; George, et al., 2003 ; Nocera, et al., 2014 ; Putnam, Handler, & Feinberg, 2005; Scott & Barrett, 2004; Spencer, 2015) It is noted that any SWPBIS model is often considered more of a process, rather than a defined program that can be immediately implemented at full force and with the exact components needed to impart the desired level of changes in the school Indeed, the process of implementing SWPBIS necessarily includes ongoing adaptations to meet the changing needs of the school and students As such, the time it takes to see results can vary drastically by school, though the general consensus in the field is that the critical components of the model can be implemented within the first three months to one year (FLDOE, 2005) The present research study focused exploration on the immediate impacts only during the first year of implementation and did not consider changes to school philosophy (which can take several years to realize) As such, it is most prudent to consider this a formative evaluation study rather than the assessment of a fully developed SWPBIS intervention model The foundation of this research utilized a non randomized, quasi experimental design for applying and assessing the SWPBIS intervention model The use of a quasi experimental design was necessary, as randomly assigning students to intervention and

PAGE 117

117 non intervention group s would have been entirely impractical and potentially unethical More specifically, the SWPBIS interventions were applied as part of the overall School Improvement Plan, which required the intervention be applied for all students, with all staff and facu lty, and across the entire school Even if the researcher had been able to randomly assign students to intervention and non intervention groups, it would have and so of reinforcements in the presence of other students The potential damage such a method could cause was famously demonstrated where groups of students demonstrated prejudice and discrimination after being assigned to a group based only on their eye color (Peters, 1970) It would also have been relatively impossible to differentiate groups of students, as the intervention was ap plied by all staff members across the school, including staff and faculty that interact with multiple groups of students at once (e.g., cafeteria staff, media center staff, etc.) a necessary element of successful SWPBIS models (Sugai & Horner, 2009a) M oreover, even if the students had been randomly assigned to intervention and control groups, it would have been practically impossible to prevent vicarious learning and contamination of the control group due to their observation of intervention group reinf orcements and/or discussions of such reinforcements among students As such, because all students in the school were included in the initial baseline year and the subsequent intervention year without randomization, the research is considered a pretest po sttest, one group, quasi experimental design (Reichardt, 2009)

PAGE 118

118 The quasi experimental design explored student behavior data (i.e., incidences of major referrals ( ODRs ) and minor (pre) referrals ) for two baseline school years prior to the implementation o f the SWPBIS model (2012 13 and 2013 14) and one intervention school year during the first year of SWPB I S model implementation (2014 15) Realizing that one group, pretest posttest quasi experimental designs are particularly sensitive to threats to intern al validity (Reichardt, 2009), it is important to note that the principal, BRT, and the majority of staff and faculty remained consistent throughout the three years of data collection Several other design characteristics were intended to reduce threats t o internal validity (Reichardt, 2009), such as: (1) using all students in the school, rather than only the students that attended all three years, helps prevent threats ther collecting all data by month (see data collection procedures later in this chapter), the The elementary school selected for this study was located in North Central Florida, with all kindergarten to fifth grade students attending the school included in the total student sample As noted previously, the SWPBIS model implemented during the 2014 2015 academic year at the selected elementary school was part of the School Improvement Plan Indeed, the SWPBIS model would have been applied and most of the data would have been collected regardless of this study However, the school and district had no intention of analyzing the data associated with the SWPBIS model and associated interventions due to the complexity of disaggregating and recording the data metrics (e.g., number of reinforcements per student per month, number of minor pre referrals

PAGE 119

119 per student per mo nth, etc.) As such, this study obtained approval to access, collect, and analyze three years of archival data associated with the two baseline years and the SWPBIS model implementation year Approval was obtained from both (1) the University of Florida, Institutional Review Board (IRB 02) and (2) the Alachua County Public Schools Department of Research, Assessment, and Student Information Primary data metrics collected as part of this study are discussed later in this chapter, with additional archival data approved for collection including such information as school population, gender, race/ethnicity of students, number of 504/IEP students, Free Reduced Lunch percentage of the overall school population, and feedback surveys from the school faculty and s taff members during the course of SWPBIS implementation Participants Target Population and Setting All data were collected and interventions were applied at an elementary school in North Central Florida Although technically a convenience sample, the e ntire population of students attending the school participated in the SWPBIS model and associated interventions No students from kindergarten through fifth grade were excluded from the study, though not all students in the school received tangible reinfo rcements (all students received some level of the SWPBIS invention model, whether through direct or vicarious reinforcements) The school selected is relatively representative of the overall composition of the county in which it resides, with the school e nrolling students far outside the standard enrollment zone For instance, the school has a large number of enrollment zone, in part due to special programing at the scho ol for students with special needs and disabilities, as well as highly desired combined grade level classes

PAGE 120

120 It is important to note the school was selected in part because the researcher was the behavioral resource teacher (BRT), principal intern, and PB I S Coach for the school, and was responsible for the direct oversight of all SWPBIS interventions, supporting and maintaining disciplinary data Participants and Descr iptive Statistics All participants attended the school where the SWPBIS model was implemented The principal agreed to assist with the data collection process, and all data were collected by the BRT and principal investigator on this research study As n oted, all school age students attending the school during the course of the three years of data collection were included in the study sample (the school has a pre kindergarten program which was not included in the study due to non traditional operations) No school age students were excluded from the sample However, due to limitations of the approved archival data collection, the research project was only permitted to collect limited data on students with disciplinary actions during the two baseline year s (i.e., student identification number, limited demographics, and number of minor pre referrals and major office discipline referrals by month) Overall, data on a total of 838 students were collected over the three years explored by this research (i.e., Baseline Year 1, Baseline Year 2, and Intervention Year 3), with 32 students having data in the first year, 150 with data from the second year, and 755 with data from the intervention year Of these students, 89 had multiple years of data, as they were e nrolled during the intervention year and had prior disciplinary actions during at least one baseline year In addition, 78 students had data only from the baseline years, as they had prior disciplinary actions during the baseline years but

PAGE 121

121 were not enroll ed during the intervention year Finally, 671 students were enrolled during the intervention year without any prior disciplinary actions and, as such, did not have any data from baseline years This is not to suggest students without multiple years of da ta were not enrolled during the baseline years, only that they did not have any formal disciplinary actions during those years Regardless, the method of data collection prevented the need to exclude students that only attended the school for a short peri od of time (e.g., those attending only a few days at the beginning of a school year before being transferred on a waiver), as those students would have neither reinforcements nor referrals and would not impact the overall data analysis As such, for the p urposes of descriptive statistics, all students enrolled in the school during the three years are included Race and e thnicity Of the 838 students included in the total sample, the final participant sample consisted of 288 (34.4%) students identified a s African American or Black; 357 (42.6%) identified as Caucasian American or White; 98 (11.7%) identified as Hispanic or Latina(o); 34 (4.1%) identified as Asian American or Pacific Islander, and 59 (7.0%) identified as multiple ethnicities The School Di strict only reports multiple ethnicities within archival data as a single general category, without providing a specific breakdown of which multiple ethnicities are included As such, students considered irely separate group for the purposes of data analysis and descriptive statistics Because of the restrictions of archival data collection, the overall participant demographics can be misleading, such that it is important to consider the overall school p opulation during the intervention year as a more accurate depiction of the school

PAGE 122

1 22 demographics across all three years Ultimately, during the intervention year (when all student demographics could be collected), the 755 students in the sample consisted of 243 (32.2%) students identified as African American or Black; 334 (44.2%) identified as Caucasian American or White; 91 (12.1%) identified as Hispanic or Latina(o); 32 (4.2%) identified as Asian American or Pacific Islander, and 53 (7.0%) identified as mu ltiple ethnicities However, as shown in Table 3.1, when only considering those students with formal disciplinary actions during each year (i.e., at least one minor pre referral and/or major office discipline referral (ODR) ) the distribution of ethniciti es across the three years of data collection are relatively disproportional to the ove rall school demographics For instance, during the intervention year, African American students represented a higher proportion of students receiving referrals (45.7% in the intervention year) than the proportion of total students in the school (32.2% in the intervention year) thus further demonstrating the need for this study. Table 3 1. D istribution of eth nicity by year (students with disciplinary actions) n = 250 Baseline Year 1 Baseline Year 2 Intervention Year 3 Unduplicated All Years n % n % n % n % African American / Black 18 56.3% 79 52.7% 58 45.7% 124 49.6% Caucasian American / White 7 21.9% 40 26.7% 43 33.9% 76 30.4% Hispanic / Latina(o) 3 9.4% 16 10.7% 11 8.7% 24 9.6% Native American 0 0.0% 0 0.0% 1 0.8% 1 0.4% Pacific Islander / Asian Am. 0 0.0% 2 1.3% 3 2.4% 5 2.0% Multiple Ethnicities 4 12.5% 13 8.7% 11 8.7% 20 8.0% Total 32 -150 -127 -250 -Note: Race/Ethnicity categories are those established by t he Local School District Race/Ethnicity categories are presented in alphabetical order 59 Students had disciplinary actions in more than one year, with 250 total students represented. In addition to the race and ethnicity of students, it is important to also look at the demographics of the faculty and staff who provided the interventions associated with the

PAGE 123

123 applied PBIS model Due to limitations of the archival data collec tion process complete demographic data were not available on faculty and staff from the first year of baseline, such that these data are not reported herein However, when exploring the distribution of faculty and staff from the second baseline year and intervention year, the distribution appears quite disproportional compared to the student population More specifically, as shown in Table 3 2, the vast majority of faculty and staff on whom demographics were collected were reported to be Caucasian Americ an / White in both the second baseline year (N=49, 90.7%) and the intervention year (N=56 91.8 %) Only a small percentage of faculty and staff were from the traditionally defined minority groups, with the intervention year including 4.9 % of faculty/staff identified as African American / Black (N=3), 1.6 % identified as Hispanic / Latina(o) (N=1), and 1.6 % identified as Asian American (N=1) Table 3 2. D istribution of ethnicity by year (faculty / staff) Baseline Year 1 Baseline Year 2 Intervention Year 3 n % n % n % African American / Black --2 3.7% 3 4.9 % Caucasian American / White --49 90.7% 5 6 91.8 % Hispanic / Latina(o) --1 1.9% 1 1.6 % Pacific Islander / Asian Am. --2 3.7% 1 1.6 % Total --54 -61 -Note: Race/Ethnicity categories are those established by the Local School District Race/Ethnicity categories are presented in alphabetical order Gender : In addition to race / ethnicity, the archival data process collected data on the gender of students dur ing the two baseline years and the intervention year As with other demographic data, limitations of archival data only allowed gender data to be collected from those students with at least one office disciplinary referral during the respective baseline y ear In looking at all 838 unduplicated students across all three

PAGE 124

124 years of the archival data collection (83 unduplicated students from the two baseline years and 755 from the intervention year), the overall student participant sample consisted of 55.1% (N =462) males and 44.9% (N=376) females As with other demographics, the distribution of students across genders was relatively consistent when looking at only the 755 students that attended during the 2014 2015 intervention year More specifically, 54.2% (N=409) students were reported as male and 45.8% (N=346) students were reported as female during the intervention year G ender is reported by the district according to a dichotomous classification Given the focus of this research on those students receiving office disciplinary referrals, it is important to explore demographics of those students that displayed behavioral issues during each of the three study years More specifically, as shown in Table 3 3, the distributions of stud ent gender in each of the three study years was largely incongruous with the distribution of t he overall student population and the distribution of students enrolled in the 2014 2015 intervention year For instance, during the 2014 2015 intervention year, students with at least one office disciplinary referral were composed of 80.3 % (N= 10 2) males and 19.7 % ( N=25 ) females a substantially higher proportion of males than contained in the overall student enrollment during the same year (i.e., 54.2%; N=409 of 755) Table 3 3. D istribution of gender by year (students with disciplinary actions) n = 250 Baseline Year 1 Baseline Year 2 Intervention Year 3 Unduplicated All Years n % n % n % n % Male Student 26 81.3% 109 72.7% 102 80.3% 182 72.8% Female Student 6 18.8% 41 27.3% 25 19.7% 68 27.2% Total 32 -150 -127 -250 -Note: Gender categories are established and reported by the Local School District 59 Students had disciplinary actions in more than one year, with 250 total students represented.

PAGE 125

125 In addition to gender distribution of students, the archival data collection included the gender of faculty and staff during the second baseline year and the intervention year As with other faculty and staff demographics, data from the first baseline year were not available for collection Regardless, as shown in Table 3 4, the vast majority of faculty and staff during both the second baseline year and the intervention year were female, with 88.9% (N=48) reporting as fema le during the second baseline year and 91.8% (N=56) reporting as female during the intervention year Such a distribution is relatively consistent with that of the overall state For instance, according to the most recent data reported by the Florida De partment of Education across all elementary and secondary schools, 77.4% (N=407,828) of all staff and faculty across the state in 2012 2013 i dentified as females, compared to only 22.6% (N=119,061) identifying as males ( Education Information and Accountabi lity Services ( EIAS ) 2013 ) Exploring only elementary schools during the 2012 2013 academic year (the most recent data provided by the FLDOE) the FLDOE reported elementary teachers across Florida to be composed of 89.9% (N=66,408) females and 10.1% (N=7,479) males (the FLDOE did not break out elementary school gender distribution for all staff members; EIAS, 2013) Table 3 4. D istribution of gender by year (faculty / staff) Baseline Year 1 Baseline Year 2 Intervention Year 3 n % n % n % Male Faculty / Staff --6 11.1% 5 8.2% Female Faculty / Staff --48 88.9% 56 91.8% Total --54 -61 -Note: Gender identification is provided by staff and reported by the Local School District

PAGE 126

126 School grade l evel As with race and ethnicity, the distribution of students across school grade levels can be explored in two methods: (1) all student participants and (2) students with at least one office disciplinary referral First, of the 838 unduplicated students included in the sample across all thr ee study years and with grade level data reported, 36 students (4.3%) were in voluntary pre kindergarten; 185 students (22.1%) were in kindergarten; 118 students (14.1%) were in 1st Grade; 126 students (15.0%) were in 2nd Grade; 129 students (15.4%) were i n 3rd Grade; 112 students (13.4%) were in 4th Grade; and 132 students (15.8%) were in 5th Grade The distribution of student across grade levels is relatively consistent when looking at only those 755 students enrolled during the 2014 2015 academic year ( i.e., intervention year), with 34 students (4.5%) in voluntary pre kindergarten; 170 (22.5%) in kindergarten; 113 (15.0%) in 1st Grade; 115 (15.2%) in 2nd Grade; 116 (15.4%) in 3rd Grade; 108 (14.3%) in 4th Grade; and 99 students (13.1%) in 5th Grade Th e consistency of these distributions is expected, as the only unduplicated students included in the total sample that were not enrolled in 2014 2015 were students with at least one disciplinary referral that left the school prior to the intervention year However it is also important to explore the distribution of only those students with disciplinary referrals across the three years of the research study As with race/ethnicity, Table 3 5 presents the distribution of students with disabilities across bo th baseline years and the intervention year, as well as the distribution of the unduplicated students across all three years Ultimately, the distribution of students with disciplinary referrals is relatively consistent across all three years, with 56.4% (N=141) of unduplicated students with referrals across all three years being in 3 rd grade

PAGE 127

127 or higher This is relatively disproportionate with the overall distribution of the 838 unduplicated students, where only 44.5% (N=373) of students were in 3 rd grade or higher When appropriate, differences by grade level are detailed in the results section of this research Table 3 5. D istribution of grade level by year (students w/ discipline actions) n = 250 Baseline Year 1 Baseline Year 2 Intervention Year 3 Unduplicated All Years n % n % n % n % Voluntary Pre K 0 0.0% 2 1.3% 2 1.6% 4 1.6% Kindergarten 4 12.5% 21 14.0% 24 18.9% 43 17.2% 1 st Grade 7 21.9% 14 9.3% 13 10.2% 25 10.0% 2 nd Grade 7 21.9% 23 15.3% 20 15.7% 37 14.8% 3 rd Grade 6 18.8% 34 22.7% 22 17.3% 45 18.0% 4 th Grade 3 9.4% 27 18.0% 26 20.5% 43 17.2% 5 th Grade 5 15.6% 29 19.3% 20 15.7% 53 21.2% Total 32 -150 -127 -250 -Note: 59 Students had disciplinary actions in more than one year, with 250 total students represented. Free and reduced price l unch As one of the primary indicators of poverty student sample in terms of their free or reduced price lunch status Unlike other demographic d ata, the district collected and stored necessary archival data for all three study years, though the data were not provided by individual student (with the district providing only a final overall distribution for the school for each year) Regardless, as shown in Table 3 6, data from the three years of study suggest relatively consistent proportions of students qualified for free and/or reduced price lunch More specifically, nearly half the students qualified for free or reduced price lunch in each year of study, with 54.5% (N=335) of all the school population qualifying for free or reduced price

PAGE 128

128 lunch in the first baseline year, 60.6% (N=383) qualifying in the second baseline year, and 51.7% (N=390) qualifying during the intervention year Table 3 6. F ree and reduced price lunch (all students) Baseline Year 1 Baseline Year 2 Intervention Year 3 n % n % n % Free Lunch 301 48.9% 346 54.7% 355 47.0% Reduced Price Lunch 34 5.5% 37 5.9% 35 4.6% Paid Lunch 280 45.5% 249 39.4% 365 48.3% Total 615 -632 -755 -Note: Lunch categories are established by the USED and reported by the Local School District As with other demographic variables and given the focus of the re search on students displaying negative school behaviors it is important to explore the free and reduced price lunch distribution of students that displayed behavioral issues during each of the three study years As shown in Table 3 7 for each year of the study, the proportion of students with disciplinary referrals receiving fr ee or reduced price lunch appears higher than the proportion of students receiving free or reduced price lunch across the entire school for each respective year For instance, 72.4% (N=92) of students with discipline actions (N=127) during the interventio n year qualified for free or reduced price lunch, while only 51.7% of the overall student population at this school qualified during the same year This disproportionality is consistent across all three years of the study (i.e., first baseline year, secon d baseline year, and intervention year) Ultimately, exploring only the 250 unduplicated students across all three years, archival data revealed 76.4% of all students with office disciplinary referrals qualified for free or reduced price lunch

PAGE 129

129 Table 3 7. F ree and reduced price lunch (students w/ discipline actions) n = 250 Baseline Year 1 Baseline Year 2 Intervention Year 3 Unduplicated All Years n % n % n % n % Free / Reduced Price 30 93.8% 121 80.7% 92 72.4% 191 76.4% Paid Lunch 2 6.3% 29 19.3% 35 27.6% 59 23.6% Total 32 -150 -127 -250 -Note: Lunch categories are established by the USED and reported by the Local School District 59 Students had disciplinary actions in more than one year, with 250 total students represented. Student special n eeds For the purposes of demographics, archival data were also collected on the special need status of students participating in the research study Two elements of special needs were collected for the purposes of this study: (1) student It is important to note that all students identified as ESE had an Individualized Education Plan (IEP), such that this was not utilized as a separate demographic category Regardless, under the archival data collection process, the district provided aggregate data on these variables for the 2013 2014 academic year (i.e., second baseline year) and 2014 2015 academic year (i.e. intervention year) The district would not provide aggregate data from the 2012 2013 academic year, nor data on all individual students for these years (unless they had received a behavioral referral) Ultimately, data revealed that 127 students (20.1% of 632 students) were identified as ESE students during the second baseline year, while 122 students (16.2% of 755 students) were identified as ESE during the intervention year In addition to ESE, the district reported that 47 students (7.4%) received a ccommodations under Section 504 during the second baseline year, and 44 students (5.8%) received accommodations under Section 504 during the intervention year

PAGE 130

130 In addition to school wide data on student special nee ds, archival data was collected on speci al need status for each individual student that received a behavioral referral during the three study years As shown in Table 3 8, the proportion of students identified as ESE or receiving accommodations under Section 504 were relatively higher than the overall school proportions of all students in these categories For instance, 20.5% (N=26) of students with behavioral referrals during the intervention year were identified as ESE (compared to 16.2% of all 755 students enrolled during the intervention ye ar) and 14.2% (N=18) of students with behavioral referrals receiving accommodations under Section 504 (over twice the proportion within the total student enrollment during the same year: 5.8%) Table 3 8. D istribution of special needs by year (students w/ discipline actions) n = 250 Baseline Year 1 Baseline Year 2 Intervention Year 3 Unduplicated All Years n % n % n % n % ESE / IEP Yes 10 31.3% 28 18.7% 26 20.5% 51 20.4% No 22 68.8% 122 81.3% 101 79.5% 199 79.6% Section 504 Plan Yes 3 9.4% 15 10.0% 18 14.2% 27 10.8% No 29 90.6% 135 90.0% 109 85.8% 223 89.2% Either / Both (ESE / 504) Yes 11 34.4% 43 28.7% 44 34.6% 75 30.0% No 21 65.6% 107 71.3% 83 65.4% 175 70.0% Total 32 -150 -127 -250 -Note: 59 Students had disciplinary actions in more than one year, with 250 total students represented. However, while it is common for students to receive only one categorical designation, it is possible for some students to receive both ESE status and a Section 504 plan As such, Table 3 8 also shows the proportion of students with behavioral referrals that were identified within either or both special need category during each of

PAGE 131

131 the study years, as well as the proportion within the unduplicated student participant sample As shown, across all three years of the study and exploring only unduplicated students, data revealed that 75 students (30.0%) were identified as ESE students and/or received accommodations under Section 504 Years of experience (faculty / staff). Alt hough only a limited amount of demographics were permitted to be collected on faculty and staff, the researcher was able to collect the total years of experience in education for each of the faculty and staff members. As with other demographics on faculty and staff, only the second baseline year and intervention year were accessible for archival data collection. As shown in Table 3 9, experience data were obtained from 54 faculty and staff during the second baseline year and 61 faculty and staff during th e intervention year. The average experience across all faculty and staff, with faculty and staff having an average of 12.8 years of experience during the second baseline year (sd = 10.4), with an average of 14.5 years of experience during the intervention year (sd = 10.7). Table 3 9. Y ears of experience in education (faculty / staff) Baseline Year 1 Baseline Year 2 Intervention Year 3 -n = 54 n = 61 Minimum -1 <1 Maximum -38 39 Mean ( ) -12.8 14.5 Standard Deviation (sd) -10.4 10.7 Note: Years of experience are official records reported by the Local School District Neither s tudent p articipants nor faculty/staff members received compensation for their participation in this research Numerous measures were taken to protect the conf identiality of the participating students and faculty/staff, and no data were connected directly to student names Moreover, all participants and data were treated

PAGE 132

132 in accordance with the Health Insurance Portability and Accountability Act (HIPAA), the Fam ily Education Rights and Privacy Act, and the ethical standards of the American Educational Research Association The School Wide Intervention Model Development of the Intervention As with any comprehensive intervention model, a great deal of planning mu st take place prior to the design and implementation of the interventions It is important to summarize the pre design activities that led to the development of the SWPBIS model applied and investigated under this research study At the beginning of the second baseline year (2013 Behavior Resource Teacher (BRT) approached the new principal about implementing a SWPBIS model, as it had not previously been implemented at the school The principal would not allow the implementation of s uch a model without first (1) establishing a baseline to determine whether a need for SWPBIS existed, and (2) gaining the support of school faculty As such, the initial focus was gaining the support of stakeholders and faculty. The concept of SWPBIS was presented to school grade level team leaders in 2013, who provided positive feedback and guarded interest in implementing a SWPBIS approach The concept of SWPBIS was then presented to all teachers and instructional staff working at the school during the second baseline year, with an overwhelming majority of teachers expressing support for implementing a SWPBIS approach if designed with their input After establishing faculty and staff support, the concept was presented to the Parent Teacher Association ( PTA) with an accompanying request for the provision of student incentives and equipment to create a school store (where interest items) By October 2013, the

PAGE 133

133 PTA had already committed $800 to purchase ne cessary materials and incentives for the implementation of the emerging SWPBIS model After obtaining an overall and initial commitment from school faculty, staff, and parents, the BRT and school administrators began the process of developing the SWPBIS model based on best practices in the field of education A SWPBIS T eam was created to include the BRT, school principal, school counselor, a primary grade (K 2) teacher representative, and an intermediate grade (3 5) teacher representative In December 2013, all SWPBIS T eam members attended a PB I S Coaches Training Florida Within the contex t of this training, the SWPBIS T eam members progressed on developing the log istics of implementing SWPBIS at the intervention school, including necessary roles and responsibilities for implementing the SWPBIS model and methods The SWPB I S T eam also discussed be st methods for ensuring all faculty and staff were properly and consistently implementing any chosen SWPBIS components. In January 2014, the SWPBIS Team began the process of looking at school behavior data and disciplinary referrals to determine areas wher e student behavior could be addressed and improved through common SWPBIS model interventions The SWPBIS T eam solidified a list of seven intervention strategies that could be implemented under a school specific SWPBIS model In addition, the SWPBIS T eam identified the following three school wide student behavioral expectations as part of the overall SWPBIS model: (1) respect yourself and others, (2) respect school property, and (3) keep your hands, feet, and objects to

PAGE 134

134 yourself The SWPBIS T eam also dev eloped the list of materials (e.g., rolling carts for school store, incentives, etc.) needed to start the program, and developed potential SWPBIS group events to further enhance the incentives (e.g., parties, student recognition programs, etc.) The ini tial plan was to begin the SWPBIS model in February 2014 However, the principal and BRT faced substantial faculty push back about adding a new set of responsibilities associated with learning and implementing SWPBIS in the middle of the year, as well as feedback regarding the stress and anxiety associated with upcoming statewide standardized testing As such, the p rincipal and SWPBIS T eam agreed to delay the implementation of the SWPBIS model until the beginning of the 2014 2015 academic year It was be lieved there would be enhanced success and impact, as well as better adherence to the interventions, if the SWPBIS model was rolled out at the very beginning of the next school year Moreover, the delayed implementation was beneficial in allowing the scho ol administration to develop trainings for faculty and staff, as well as develop a plan for providing the necessary training during the required pre planning weeks to ensure sufficient time for absorption and processing of the new information, while also a ddressing concerns and questions. Carefully designed SWPBIS models are only effective if those implementing the interventions do so consistently and completely, such that it is imperative for faculty and school staff to be well trained in the planned inter ventions (FLDOE, 2016; Scott & Barrett, 2004; Sugai & Horner, 2009a) Indeed, a key feature of any effective SWPBIS model is the consistent application of the reinforcements to all students in all settings (Sugai & Horner, 2009a) As noted previously, th e delay in implementing the SWPBIS

PAGE 135

135 model until the beginning of the 2014 2015 academic year allowed for faculty and staff to be trained during the pre planning week of the academic year (before students began attending) During this time, all faculty and staff were trained in SWPBIS, including both in the general concepts and the specific interventions designed by th e SWPBIS T eam for the intervention school During this training, SWPBIS T eam members were introduced as the primary resources for questions o r concerns regarding the SWPBIS model In addition, 2012 2013 and 2013 2014 data on student discipline and referrals was presented to demonstrate the urgent need for the selected SWPBIS interventions to address problematic and challenging student behavior s that resulted in lost instructional time and reduced educational benefits to students in the prior academic years The training was focused on providing an understanding of the SWPBIS model interventions and methods to ensuring consistent and adequate i mplementation of the interventions. Intervention Design As me ntioned previously, the SWPBIS T eam developed seven interventions behavior needs and goals. The seven primary inter vention strategies selected by the SWPBIS T eam were implemented simultaneously at the beginning of the 2014 2015 academic year after the faculty and staff were provided training specific to the consistent implementation of these intervention strategies. The following are the seven primary strategies selected, as well as a brief explanation of each strategy utilized by the school throughout the 2014 15 school year. School wide expectations of b ehavior Within this strategy, teachers and school staff were provided the school wide expectations that were to be prominently

PAGE 136

136 displayed, taught, re taught, and referenced when addressing student behavior throughout the school year. These school wide expectations of behaviors were displayed in every classroom, hallways, and within common use rooms (e.g., cafeteria, media center, auditorium, etc.). This strategy is a standard component of most school wide PB I S models, wherein the recommendation is to create three to five expectations that fully encompass the behavioral concerns i dentified by the school SWPBIS T eam (Sugai & Horner, 2009a). For the purposes of this school intervention, the following behavioral expectations were identified: (1) respect yourself and o thers, (2) respect school property, and (3) keep your hands, feet, and objects to yourself. These expectations were presented to all students in the school at the beginning of the academic year, with posters of these basic expectations posted prominently in all classroom and student use areas (e.g., cafeteria, media center, hallways). Teachers were expected to reinforce these expectations throughout the year using other interventions (e.g., morning meeting would include a review of these expectations). Morning m eeting M orning M M orning M review specific concepts and reintroduce the learning from the prior day. However, for the p designed to enhance the SWPBIS model using concepts from core educational theory (Charney, 2015; Nelson, Lott, & Glenn, 2013), which underscores the importance of having teach ers and students welcome one another, share something with the class, as primary prevention technique to create a more positive school culture, wherein

PAGE 137

137 positive student behav iors can be taught by the teacher, modeled by other students, and consistently reinforced in a small group social interaction setting (Sugai & Horner, was included as th e first 15 minutes of the daily Master Schedule, with the intent for teachers to spend time to get to know their students, build rapport with students, and allow students to build a learning community. Teachers were encouraged to ask students about their weekend, share their favorite foods or music, talk about and share positive experiences they have had with their families, plans for future vacations, or talk about age appropriate headlines from the newspaper. Teachers received reminders rapport with students. Tangible i ncentives in the form tickets (i.e., Lion Loot & school s tore) The concept of tangible incentives is not unique to this research nor positive behaviora l support theory, but is considered a foundational element of school wide PBIS models (Kappel, Dufresne, & Mayer, 2012; Ullmann & Krasner, 1975). The importance of reinforcement paradigms has certainly been underscored by many past researchers and theoris ts, including positive reinforcement under operant conditioning theory (S k inner, 1953) and vicarious learning under social learning theory (Bandura, 1962). Using such concepts, the first models of token economies were developed and applied in mental healt h hospitals, with application to children in educational settings developed more recently. In essence, a token economy has three basic requirements (Kazdin, 1977; Kazdin, 2012b): (1) an object used as the immediate reinforcer with no intrinsic value (i.e.

PAGE 138

138 can be exchanged; and (3) specific and clear criteria for obtaining tokens (e.g., specific behaviors to be modified or controlled). Tokens : For the purposes of this interven tion model, the school developed tickets student behavior. Teachers were also taught how give out tickets in a meaningful way and in a way that the SWPBIS Team could track their distribution. For instance, ittently to avoid predictability, increase the impact of the token system on developing positive student behavior, and increase the longevity of impacts on student behavior (Miltenberger, 2012). Teachers student, so as to increase the vicarious learning of other students and couple the token praise and encouragement (an incentive t hat could not be recorded for this study). To tations printed on the front. They were designed to have no intrinsic value, but could be traded for items from the PBIS school store. Back Up reinforcer (school s tore): The SWPBIS School Store was a venue where students could spend their earned tickets (i.e., Lion Loot) on small, high interest items, such as small toys, trinkets, stickers, bubbles, balls, pencils, erasers, and bracelets. All items for the school store were donated by the PTA or business partners.

PAGE 139

139 The amount of tickets for items varied depending on the size of the item (as students tend to connect cost with size). For example, students could purchase stickers, pencils, and erasers for one ticket, while items such as basket balls and art kits cost upwards of twenty five tickets with m ore savvy students saving their tickets for the entire year to purchase the larger items. All school faculty and staff were responsible for initialing the tickets given to students. The initial of the teacher would be used to validate that it was earned and the student was required to put their name on it as a way for the SWPBIS Team to keep track of the frequency of how teachers and staff were recognizing positive student behavior. The initial plan for the SWPBIS School Store was that it would be open for students to spend their tickets (i.e., Lion Loot) two days per week in the morning; Tuesdays would be for grades K 2nd and Thursdays would for grades 3rd 5th. Student safety patrols in the 5th grade were to be utilized to run the school store, while b eing supervised and assisted by a teacher. Once the SWPBIS School Store began operating in September, the volume of students overwhelmed the store and it was impossible for all students to select items and turn in their tickets before their next class. T he BRT/PB I S Coach and school principal were able to devise a new schedule that allowed for the SWPBIS School Store to be open every day of the week in various locations around the school campus, such that students could spend their tickets and maintain the excitement of earning and spending tickets under the SWPBIS initiative. During subsequent team leader and faculty meetings, some kindergarten and first grade teachers indicated that students who rode the school bus and participated in the reakfast program were still unable to make it to the SWPBIS School Store. The solution devised by the SWPBIS Team was to allow teachers from grade

PAGE 140

140 level teams to pick up the SWPBIS School Store carts and take them to their classrooms, such that primary ag ed students could make purchases. The classroom system by making it easier and more for the bac k up reinforcers (i.e., SWPBIS School S tore items). Positive speak (language and p hrases) In addition to the use of tangible reinforcements under a token economy, teachers and staff members were taught and instructed how and when to use positive language and phrases when addressing challenging or inappropriate behavior. The teachers and staff were taught to try focusing on a student demonstrating appropriate behavior, perhaps giving them a ticket student(s) who are not displaying appropriate behavior for the situation. Teachers were encouraged to avoid using sarcastic or condescending tones if discussing inappropri ate student behavior, instead stating the facts and keeping adult emotions out of the discussion. Teachers were also encouraged to be direct with praise, be genuine with the students, give specific praise about behaviors (rather than general praise about the child as a whole), avoid personal judgements, and believe in the abilities of the students (adapted from Brady, Forton, & Porter, 2015). Verbal and non verbal praise and a ffirmation While positive speak provides a general method for verbally intera cting with students and their behaviors, teachers were also taught how to provide non tangible rewards to the students. Non tangible rewards can consist of verbal praise, non verbal praise, affirmative gestures, and physical contact (as appropriate, such as a pat on the back). The use of verbal praise

PAGE 141

141 verbal praise conjunction with the tang ible reinforcement system (Lion Loot) to further reinforce positive student behavior. Teachers and staff were taught the concept of forward conditioning (Kazdin, 2012a), which suggests that verbal praise be presented first when a student behaves well, fol lowed immediately with any tangible rewards (e.g., Lion Loot). With this forward conditioning, the goal is eventually for the verbal praise alone to be sufficient for the student to feel the same level of happiness as with the tangible reward. Teachers a nd staff were to incorporate verbal praise and non verbal praise/cues as another means to recognize positive student behavior. For example, in recognizing that some children are motivated by the desire to please adults, teachers were taught to use phrases Progressive discipline plan and plan f idelity While the goal of SWPBIS is to provide positive, proactive, and reinforcement based interve ntions to teach students appropriate classroom and school behaviors, it is generally anticipated that some students will have already learned effective disruptive behavior and/or will learn these behaviors and bring them to school. As such, it is importan t that there be a clearly defined and detailed school wide progressive discipline plan to address disruptive and dangerous student behaviors (e.g., Sugai & Horner, 2009b). Teacher s were trained in the discipline plan, which included a progression through the following levels: (1) time out in classroom, (2) time call and/or notification, and (4) if no change, front office intervention through an office

PAGE 142

142 disciplinary referral. Teachers were inform ed of the expectation that the progressive discipline plan be followed with fidelity, using a progressive discipline plan checklist to demonstrate they had attempted to work with the student at least twice prior to requesting the student be removed from th e classroom setting. Teachers were also provided with minor / pre referral forms (i.e., Behavior Notice Forms) that could be used as a formal way of documenting inappropriate behavior and informing parents, but less formal than an office discipline referr al (ODR). Exp ectation of IEP, 504 Plan, and behavior i ntervention plan (BIP) f idelity In addition to a progressive disciplinary plan, teachers were reminded to carefully review student records to ensure that IEPs, 504 Plans, and Behavior Improvement Plans (BIPs) were being followed with fidelity. Teachers were reminded of their responsibil ity to be diligent in reviewing student records to insure they were complying with student IEPs, 504 Plans, and BIPs, if applicable, so as to best ensure the students were being educated and treated in accordance with any additional supports they may need that could impact their behavior in the classroom setting. Teachers were provided coaching and assistance from the BRT and SWPBIS Team in strategies to apply the full SWPBIS model and interventions with all students, regardless of disability or special ac commodations. Teachers were trained in the origins of SWPBIS as an individualized intervention model for students with severe disabilities and behavioral issues, such that SWPBIS should be equally applicable to all students in the school. Other interven tions, incentives, and a ctivities In addition to the aforementioned strategies for the first year of SWPBIS implementation under this formative evaluation study, teachers were also supported with morning announcements

PAGE 143

143 about SWPBIS and the school store to help build interest among the students and reinforce the overall token economy system. In addition, teachers were provided some flexibility to implement alternatives to the SWPBIS School Store, such as eating in the classroom with the teacher, eating out side at the picnic tables, and sitting in special area of the classroom. Alternatives were not used to replace the school store, as some students might not have been motivated by the alternative activities and the tokens e as the back up reinforcers. Model Implementation Support and Training Initial training (pre p lanning) As mentioned above, all teachers and staff were provided detailed training on the SWPBIS model and specific interventions. At the conclusion of the training, teachers and school staff were given time to ask questions regarding the SWPBIS initiative and model, provided with posters of the school wide expectations of student behavior, provided with tickets (i.e., Lion Loot) for recognizing wide behavior expectations set forth by school staff, pre referral forms (minor referrals ), and major ODR forms. Moreover, teachers were asked to discuss the SWPBIS initiative, incl uding the school wide behavior expectations, how tickets could be reviewed with students, and how to begin using tickets starting the first week of school. Furthermore, the BRT made a presentation about the SWPBIS initiative during the school wide morning show broadcast to all classrooms to build excitement about the SWPBIS initiative among students. The SWPBIS School Store and the incentives was also shown on school television, and the BRT explained how incentives could be circuit television show throughout the school year to remind students about the potential

PAGE 144

144 rewards for demonstrating appropriate school and classroom behaviors. Teachers were also encouraged to explain the start of school to inform parents and guardians of the new initiative. Faculty meetings (school b ased, m onthly) Faculty meetings were held at least once per month with the entire school. During these meetings, teachers were presented with overall school wide student behavior data using the Florida Response to Intervention for Behavior Database (RtI:B) website (www.flrtib.org). Both major office discipline referrals (ODRs) and minor pre re ferral data were presented to the faculty by grade level totals, month to month comparisons, comparisons from prior year, type of behavior infraction, location of infraction, time of day, and demographic beak down of disciplined students. Teachers were in formed of data based trends and strategies for helping to curb problematic behaviors and teach more productive behaviors particularly among those students demonstrating repeated occurrences of the same inappropriate behavior. Teachers were informed abou t the number of tickets spent at the PB I S School Store, while also soliciting input regarding student availability to spend tickets (the input was then used to adjust the operations of the school store). In addition, teachers were reminded and retrained a bout how to use tickets (i.e., Lion the completion of homework). Finally, during each meeting, the BRT would review of the SWPBIS approach and the seven strategies detailed above. Particular focus was given to the progressive discipline plan model, distribution of tickets, and use of positive language.

PAGE 145

145 Data chats (grade level, team ba sed, q uarterly) In addition to monthly faculty meetings, all faculty participated in grade level meetings within their respective teams data chats, teachers in each grade level were presented with grade level specific student behavior data using the Florida Response to Intervention for Behavior Database (RtI:B) website (www.flrtib.org). Both major office discipline referrals (ODR) and minor pre referral data were pre sented to teams by individual teacher totals, type of behavior infraction, location of infraction, time of day, and demographic beak down of disciplined students. Teachers were informed of data based trends, while also being provided strategies for helpin g to curb problematic behaviors. Finally, teachers were reminded about how to use tickets (i.e., Lion Loot) during data meetings, and their pattern of awarding Lion Loot was reviewed to help ensure consistency with the SWPBIS model. Focused re minders W hile all faculty were reminded of the selected SWPBIS interventions during meetings, not all faculty attended these meetings regularly (e.g., illness, vacations, etc.) and there were often many other topics discussed during these meetings, such that the SW teachers. As such, the SWPBIS T eam provided several focused reminders to faculty and staff. For instance, during the first week of school, the BRT sent e mails to reiterate and outline the SWPBIS plan to the facul ty and staff, remind faculty and staff about school behavior goals (i.e., reducing pre referrals and major office discipline referrals (ODRs) ), and BRT also sent reminder s to faculty and staff during those times of the year when problematic student behavior typically would spike. The purpose was to remind faculty

PAGE 146

146 and staff of the need to remind and re teach students of school wide behavior expectations and ask that the fa culty and staff be mindful of their own emotions and stress when interacting with excited or panicked students. For example, typical times throughout the school year where problematic student behavior escalated was during Halloween, days prior Thanksgivin days prior to Spring Break, and toward the end of the school year. In addition to such email reminders, the PB I S Team members (e.g., School Counselor and BRT) would periodically e mail and meet with t eachers regarding accommodations and/or plans with fidelity and helping students with known problematic behaviors be successful in class. Within these emails and during these meetings, teachers and staff were reminded as to the importance of implementing the SWPBIS model and interventions consistently across all students, including those with accommodations or plans. Further meetings were held with grade level team leade rs, wherein the BRT would remind team leaders to share behavior trends based on student discipline data with their team and encourage them to remind and reteach school wide behavior expectations to each of their team members. Retraining During the second semester of the school year, the SWPBIS Team provided a retraining of SWPBIS and illustrated improvements in student behavior using comparisons of month to month student data from the second baseline year and the intervention year. At the conclusion of t he second training, teachers were able to ask questions, while some teachers shared their successes with using SWPBIS strategies tangible rewards

PAGE 147

147 worked into the SWPBIS model (e.g.., letting th e entire class eat lunch in the classroom, homework pass, popcorn party, etc.). Direct i ntervention of SWPBIS Team Not all classroom teachers had initial success in implementing the SWPBIS interventions, either due to trepidation, doubt, or lack of spe cific skills. In these situations, the BRT (Behavior Resource Teacher) would provide direct assistance and intervention in the classroom. These supports included in person modeling of SWPBIS interventions and providing in vivo guidance for teachers about the interventions (e.g., some teachers were worried about giving too few (e.g., when and where to give Lion Loot), staff members that were not classroom teachers (incl uding cafeteria staff) required more modeling to show them how to give the actual Lion Loot under the SWPBIS model. Finally, when situations arose that a activity for inap propriate behavior, the administrator would stress the school wide behavior expectations as part of the discipline process. Classroom evaluations (implementation c hecks) During both walk throughs and formal classroom observations, the BRT/PB I S Coach an d school principal ensured school wide behavior expectations were clearly visible in every classroom. More specifically, during the implementation year, the PB I S Coach and Team conducted classroom and campus checks using the PB I S Benchmarks of Quality (Bo Q) tool to assess the level and degree of SWPBIS implementation. These were done in conjunction with the Mid year and End of Year SWPBIS Evaluations completed as part of the process implemented by the FLDOE to support schools with SWPBIS models.

PAGE 148

148 During s uch teacher appraisal and walk throughs from administrators, teachers were monitored to ensure they demonstrated positive relationships and rapport with students. Administrators were mindful to check that SWPBIS strategies (e.g., positive language, verbal praise, Lion Loot, etc.) were being utilized with fidelity for teachers with more minor and/or major referrals for students in their classrooms. Moreover, students were also observed to see that they demonstrated appropriate behavior when interacting wit h peers. If students were off task or demonstrating inappropriate classroom behavior, the administrator would watch to see if that behavior was addressed within the SWPBIS strategies adopted by the school. During follow ups to class walk throughs and for mal classroom evaluations, school administrators would complement teachers regarding behavior management strategies used and, when necessary, discuss areas that needed to be addressed and provide teachers with useful strategies. Teacher individualized di scussions / s upport Faculty and staff were also provided more individualized support and retraining, as needed based on data. For instance, the supervising teacher of the SWPBIS School Store would convey to the BRT/PB I S coach when many tickets were comi ng from the same teacher(s). In other cases, some teachers would report their observations when another faculty or staff were giving out tickets in a non intermittent manner that was inconsistent with the SWPBIS strategy. Members of the SWPBIS Team would tactfully approach teachers where concern had been expressed and ask that they describe their process for distributing tickets. If they were not doing it in a strategic and graduated manner, then they would receive feedback and retraining to better help them utilize the strategy of tangible reinforcements. Some teachers would be approached if they were not giving out any

PAGE 149

149 tickets or few tickets. In those cases, some shared they preferred to use individual and whole class incentives, such as earning the a bility to eat lunch in the classroom with the teacher or at the picnic tables. In other cases, teachers stated they used verbal and non verbal praise because they felt it was be more meaningful to the student in their class instead of receiving tickets. These teachers were reminded as to the school wide model and the importance of all faculty and staff implementing the chosen interventions with fidelity and consistency, as this is a multi year model and their students would be better served by learning th e system before they move to another teacher the next year (or another environment in the school during the intervention year). Finally, to help provide individua lized support, two lead SWPBIS T eam members were assigned to support either primary or inter mediate grades, and they were made available to answer any questions or to support teachers directly. These SWPBIS Team representatives did not initiate conversations, but were the contacts for any questions that emerged. If needed, the teache r s or the t eam leaders could come to the BRT/PB I S Coach for more information. Measures and Metrics Office Discipline Referrals (ODRs ; Major Referrals ) Major ODRs are formal documents created by the school district and used to code of conduct, jeopardizes the safety and/or learning environment of other students, and ultimately results in the student being removed from class by a school administrator. These incidents are formally recorded, require specific actions by the may be writte n by any classroom teacher, school staff member, bus driver, or school

PAGE 150

150 administrator. By policy, once an ODR is written and sent to the school administrators, the administrator is required to investigate the incident, determine its validity, and assign an ODR incidents were recorded into both (1) an internal school district database for use in the SWPBIS planning and implementation process and (2) the Florida Response to Intervention for Behavior (RtI:B) database. Data were collected and compared from bo th resources, which were found to exactly match (as th e Behavior Resource Teacher was responsible for maintaining both databases). These data were used for data driven decision making purposes within the school, as well as for sharing with teachers and ot her stakeholders at quarterly data chats, grade level team meetings, and monthly faculty meetings. Although already discussed in detail within Chapter 2, it warrants repeating the fact that ODRs are among the most prevalent and common outcome measures fo r student behavior within SWPBIS literature and research (e.g., Fairbanks, et al. 2007; George, et al., 2003 ; Nocera, et al., 2014 ; Scott & Barrett, 2004; Sugai & Horner, 2006; Waadsorp, et al., 2012 ). As noted previously, ODRs have been shown to be rela ted to future behavior problems, such as drug use and disorderly conduct in classrooms (e.g., Nelson & Roberts, 2000; Sprague et al., 2001). Pas, Bradshaw and Mitchell (2011) found ODRs to be positively correlated with both the disruptive behavior and att ention problems subscales, as well as negatively correlated with the prosocial behavior subscale of the Teacher Observation of Classroom Adaptation Checklist (TOCA C). Overall, the researchers determined that ODRs have moderate convergent and

PAGE 151

151 divergent va lidity with behavior ratings on the TOCA C ( Pas, et al., 2011 ). Other researchers have also found validity for using ODRs as a measure of general behavior (e.g., Irvin, et al., 2004 ; Scott & Barrett, 2004), with ODRs being related to general misbehavior a t school, school attendance, student and teacher perceptions of safety and victimization, classroom orderliness, juvenile delinquency, and behavior disorders. Minor Referrals ( Pre Referrals) While Major ODRs are recorded within central databases and are the most common focus of outcome research on student behaviors following SWPBIS, this is also a common limitation of past research on SWPBIS. More specifically, past studies most often obtained archival data from centralized databases and generally had li ttle ability to explore data on disciplinary issues that did not rise to the level of a formal office discipline referral for administrator action (i.e., Major ODR). Indeed, with respect the school at focus of the present study, incidents of Major ODRs an d associated actions taken by administrators were available in the Florida Response to Intervention for Behavior (RtI:B) database (as with any school implementing SWPBIS in Florida). D ata on behavioral incidents that were less problematic or did not repres ent a major infraction of the code of conduct are also recorded on the RtI:B database, but are not collected on the district database and are rarely collected within other centralized database systems for student behavior. Minor referrals (pre referrals) are still considered formal documentation, but there are no official forms or the forms are generated at the school level (e.g., Behavior Notice Form, formal note from the teacher, etc.). Minor pre referrals are for those behaviors that do not immediatel y or significantly jeopardize the safety and/or learning environment of other students, but continued display of the behavior and/or escalation

PAGE 152

152 of the behavior could rise to the level requiring a Major ODR. Within the progressive discipline plan for the s cho ol at focus of this study, all m inor pre referrals were required to include a formal note from the classroom teacher, a note from the school staff member addressing the behavior, and/or intervention from the school administration that did not necessaril y require removal from the classroom environment. These incidents were recorded at the school level to assist with data driven decision making and for sharing with teachers and other stakeholders at quarterly data chats, grade level team meetings, and mon thly faculty meetings. The ability of this research study to explore m inor pre referrals separately and together with Major ODRs is a unique extension to prior studies in the SWPBIS literature Tangible Reinforcements (Lion Loot) As noted in the prior c hapter, SWBPBIS makes use of a three tiered system of supports: (1) Primary, Tier 1 whole school interventions impacting 80 90% of the student population; (2) Secondary, Tier 2 small group interventions impacting 15% of the student population who do no t respond to primary tier interventions; and (3) Tertiary, Tier 3 individual interventions impacting the remaining 5% of the student population who do not respond to primary and secondary tier interventions (McIntosh & Goodman, 2016). This research focu ses on Tier 1 interventions, with a primary focus on the tangible reinforcements provided to students under a token economy intervention. Tier 1 interventions are relatively broad and with several pre established school wide goals, with both tangible and non tangible incentives are provided to students to recognize appropriate behavior. Few studies in SWPBIS literature have included a quantitative method for assessing the extent to which a specific intervention was implemented, as it is often difficult to measures the provision of reinforcements. However, while non

PAGE 153

153 tangible reinforcements would be relatively impossible to assess while keeping to the expectations for the behavior be genuinely praised (the method of recording the reinforcement would likely take more time and reduce the incidents of supports), the present study devised a method for quantitatively assessing the administration of All token economies include the tangible token (secondary reinforcer) that for which tokens are exchanged (back up reinforcers), and the targeted behavior which is desired to be increased (Kazdin, 2012b). A token economy is essentially a systematic approach to reinforcing and increasing desired behaviors by establishing a relationship between a secondary reinforcer and a back up reinforcer. For the present research, the ned to the school wide behavior expectations set forth by school staff. The Lion Loot could be exchanged by the student at the SWPBIS School Store, which was opened daily at strategic in terest items for varying amounts of tickets (e.g., pencils, erasers, bubbles, sunglasses, bracelets, necklaces, basket balls, folders, art sets, books, game cards, etc.). These items were the back up reinforcers under the token economy model. The present study collected data on the number of tangible tickets (Lion Loot) given to a specific student during the course of the intervention school year. To support this data collection effort, the classroom teacher or school staff member who gave the Lion Loot to the student(s) was required to write both the inch by 4 inch ticket printed on green paper, with the SWPBIS school goals included on the front).

PAGE 154

154 When the student exch anged the ticket at the SWPBIS School S tore, the supervising teacher and/or student helpers (i.e., 5th grade safety patrols) at the school store insured and not acquired inappro priately (i.e., stolen, found, gifted to another student). At the end of each day, the supervising teacher would put the exchanged tickets from the Behavior Resource Teacher an d SWPBIS Coach. A Microsoft Excel database was created with a separate worksheet for each grade level (K 5th) and each individual classroom by teacher. Worksheets were also created for Special Area teachers (i.e., Art, Music, P.E., and Media) and other school staff who gave out tickets (i.e., cafeteria monitors, front office staff, speech/language pathologists, and school administration). Students could have been listed multiple times across worksheets, thus ensuring the deepest level of data collection and storage. For each day, the number of Lion Loot exchanged by the student was recorded under the teacher who provided the Lion Loot (based on the name inscribed on the back of the ticket). These data were then aggregated by month, such that the number of tickets by student and by teacher were utilized for data reporting and sharing with stakeholders and during SWPBIS meetings and trainings. However, these data were aggregated further for the purposes of the present research, with the final database us ed for statistical analyses showing the number of Lion Loot provided by each teacher and the number of Lion Loot exchanged at the School Store by each student. It is noted that the Lion Loot (tangible reinforcements) did not have the date when the reinfo rcement was provided, such that the database created for the present study could only record the date the Lion Loot was exchanged at the SWPBIS School Store.

PAGE 155

155 This will be further discussed in Chapter 4, as this represented an unexpected limitation for the findings of this study. More specifically, the original intent was to explore provision of tangible reinforcements by month of the academic year, which could then be compared to the number of ODRs provided each month. However, because desirable items, it was realized that the number of Lion Loot provided and the number of Lion Loot exchanged each month would not match. Other Data of Interest In addition to the primary in dicators and demographic data detailed earlier in this chapter, additional data were collected under the approved research study. These additional data were primarily implementation checks for the purposes of ensuring the proper implementation of the SWPB IS model designed by the school SWPBIS Team. The additional data included meeting agenda and minutes from SWPBIS planning and implementation meetings, agenda from faculty meetings and trainings where SWPBIS was discussed and/or focused upon, reports requi red by the FLDOE SWPBIS technical assistance project, and staff surveys on the implementation of SWPBIS. These additional data are not a focus of the present research and are not considered within the analysis of the quantitative and demographic data. Ho wever, some information from these additional data sources may be included in the discussion, insofar as they provide context or qualify the findings of the primary metrics. Procedures As noted previously, this study collected archival data using a conve nience sample of students attending a Florida elementary school that had implemented SWPBIS for one year. As the study explored data from the all kindergarten through 5th

PAGE 156

156 grade students attending this school, there was no random assignment or random sampl ing (as discussed under the participant section of this chapter). The archival data collected spanned a period of three school years (i.e., 2012 2013, 2013 2014, 2014 2015) and consisted of both intervention data and demographic data for all students in t he school. More specifically, intervention data included major office discipline referrals (ODRs), minor referrals (pre referrals) utilization of Lion Loot (tangible reinforcements) by student and teacher, and SWPBIS implementation data (e.g., meeting ag endas and minutes pertaining to SWPB I S planning and implementation, required state SWPB I S formative reports, End of Year Report, and staff surveys). Additionally, school demographic data were collected and included both student demographics (i.e., gender, race/ethnicity, Free & Reduced Lunch Program status, 504/IEP status, and gifted education program status) and staff demographics (i.e., gender, race/ethnicity, and years of experience). Referrals (major and minor) were the ultimate measure for determinin g if the implementation of the SWPBIS model (independent variable) made a significant difference in student behavior (dependent variable). A detailed archival data collection protocol was submitted and approved by both the University of Florida Institutio nal Review Board (IRB) and the Alachua County Public Schools Department of Research, Assessment, and Student Information.

PAGE 157

157 CHAPTER 4 RESULTS This chapter is separated into five sections. The first section presents the independent variable (i.e., tangible reinforcements) in terms of descriptive data and preliminary analyses for this variable. In the second section, descriptive data and preliminary analyses are presented for the primary dependent variables (i.e., number of minor / pre referrals, nu mber of major office disciplinary referrals, and the total number of all referrals). The third section provides an explanation of selected statistical analyses, with the fourth and fifth sections exploring the specific research questions, as originally pr oposed. A discussion of all findings is provided in the discussion chapter rather than within the results section. For all statistical analyses presented and discussed in this chapter, a probability level of p < .05 was used as the criterion for statisti cal significance. S tatistical analyses were run using IBM SPSS Statistics (version 23) or T he R Project for Statistical Computing (R Version 3.3.1, Bug in Your Hair) with the latter being an open source, UNIX based statistical environment developed by a number of promine nt statisticians (Hornik, 2016). Both statistical software packaged are commonly used within social science and educational research. The overarching goal of this study is to e xplore whether a comprehensive SWPBIS model will impact the i nstances of negative student behaviors among elementary school students within the first year of implementation The secondary goal is to explore potential differences among student sub groups to drive consideration of demographic characteristics, cultura l differences, and racial biases when reinforcing and teaching student behavior under the SWPBIS approach The tertiary goal is to determine whether the level of tangible reinforcements, as a component of the overall

PAGE 158

158 SWPBIS model implemented for this stud y, was related to any changes in the number of negative student behaviors during the intervention year. Prior to presenting the detailed findings from this research study, the following are the most salient findings related to the four research questions presented in previous chapters and discussed later in this chapter. First, d uring the study years, the school experienced a 20.17% increase in enrollment, coupled with an 23.75% decrease in total referrals, a 20.05% decrease in minor/pre referrals, and a 45.15% decreas e in proportion of major ODRs. Second, t here was a significant decrease in the proportion of referrals from the second baseline year to the intervention year. Third, t he school realized significantly lower proportions of students receiving minor referrals and ODRs in the intervention year. Fourth, African American / Black students had a significantly higher average number of major ODRs than Caucasian American / White students in both baseline years, but such differences did not persist into the intervention year. However, African American students represented a higher proportion of students receiving referrals (45.7% in the intervention year) than the proportion of total students in the school ( 32.2% in the intervention year). Finally, t he n umber of tangible reinforcements did not predict and was not significantly related to the number of negative student behaviors measure d by minor referrals and ODRs. These primary findings are presented in more detail in the following sections of this diss ertation. Section One: Descriptive and Preliminary Analysis (Independent Variable) The first two results sections focus on descriptive analysis for the independent variable (i.e., number of tangible reinforcements provided to students) and the dependent variables (i.e., number of minor pre referrals and number of major office discipline referrals). Scatter plots were first constructed to ensure there were no

PAGE 159

159 significant outliers that may have corrupted or invalidated th e results of further analyses. While no significant outliers were found, there was a need to exclude pre kindergarten students. First, pre kindergarten students accounted for only four (4) of the 250 unduplicated students across all baseline and intervention years with any recorded instance of disciplinary action (e.g., pre referral and/or major office discipline referral). Second, a single instance of tangible reinforcement (i.e., Lion Loot) was provided to only one pre kindergarten student during the intervention year (0.01% of all Lion Loot provided during the year). Finally, pre kindergarten teachers were not expected to participate in t he SWPBIS model implemented under this research study. As such, all data from pre kindergarten students were excluded from this study and are not further considered within the results and analyses. With the exclusion of pre kindergarten students, the ta bles within this section provide descriptive statistics for the number of (1) tangible reinforcements (Lion Loot) awarded to students during the intervention year (i.e., independent variable). As shown in Table 4 1, students receiving the tangible reinfor cement were generally spread across all grade levels relatively equally, with 86.1% of all students enrolled during the intervention year receiving at least one tangible reinforcement (i.e., Lion Loot). It is possible the remaining students did not receiv e any Lion Loot, or received the tangible (secondary) reinforcement without exchanging it for the backup (primary) reinforcements at the SWPBIS School Store. Because the tangible reinforcement was only recorded when exchanged, the non exchanged Lion Loot are necessarily excluded from consideration within this study. In terms of the independent variable of the present research (i.e., the number of tangible reinforcements), as shown in Table 4 1, a total of 621 students received at

PAGE 160

160 least one tangible reinfor cement during the intervention year (86.01% of the 722 students enrolled during the same year in kindergarten through fifth grade). In comparison to the overall sample of 722 students enrolled during the intervention year (2014 Squa re test of goodness of fit was conducted to determine whether the actual grade level distribution of students receiving at least one tangible reinforcement differed from the theoretical grade level distribution of students that were expected to receive a r einforcement (i.e., theoretically, all students enrolled should receive at least one reinforcement). Ultimately, students receiving tangible reinforcements were equally distributed according to the grade level distribution of the total population of stude nts at the school during the intervention year, 2 (5, N = 722) = 2.706, p = 0.745. Table 4 1. S tudents receiving tangible reinforcement by grade (lion loot) Grade N (All) % (All) N (1+) % (1+) Min Max Mean SD Kindergarten 170 23.58% 152 24.48% 0 89 22.12 17.87 Grade 1 114 15.81% 104 16.75% 0 83 16.11 12.82 Grade 2 116 15.95% 104 16.75% 0 99 22.59 20.42 Grade 3 117 16.09% 107 17.23% 0 128 31.09 27.11 Grade 4 109 14.98% 84 13.53% 0 68 13.05 13.60 Grade 5 100 13.73% 70 11.27% 0 80 13.92 17.02 Total (All Grades) 722 -621 -0 128 20.80 19.70 Note: (1+) statistics only include those students receiving one or more lion loot. Tangible reinforcements were only provided during the intervention year (2014 2015). As such, results reveal no significant concerns that any specific grade level had a higher proportion of students receiving reinforcements than represented in the overall school population, with 152 kindergarten students (24.48 %) of students receiving at l east one reinforcement compared to 170 total kindergarten students (23.58%); 104 first graders (16.75%) with tangible reinforcements compared to 114 (15.81%) of all students; 104 second graders (16.75%) with tangible reinforcements compared to 115

PAGE 161

161 (15.95%) of all students; 107 third graders (17.23%) with tangible reinforcements compared to 116 (16.09%) of all students; 84 fourth graders (13.53%) with tangible reinforcements compared to 108 (14.98%) of all students; and 70 fifth graders (11.27%) with tangibl e reinforcements compared to 99 (13.73%) of all students. In addition to exploring the proportion of students receiving the tangible reinforcements (i.e., Lion Loot), it is also important to explore descriptive statistics related to the distribution of in dividual Lion Loot rewards across student grade levels. As shown in Table 4 2, a total of 14,654 tangible reinforcements (i.e., Lion Loot) were provided to students across all six grade levels included in the research design, with over one quarter given to kindergarten students and an additional quarter given to third grade students. In comparison to the overall sample of 621 students enrolled during the intervention year (2014 2015) and receiving at least one Lion Loot, a Square test of go odness of fit was conducted to determine whether the actual distribution of Lion Loot exchanged differed from the theoretical distribution of tangible reinforcements based on the grade level distribution of students receiving a reinforcement (i.e., theore tically, all students had an equal chance to receive the same level of tangible reinforcement). Ultimately, the number of tangible reinforcements were not equally distributed according to the distribution of the population of students provided with tangib le reinforcements and enrolled at the school during the intervention year, 2 (5, N = 622) = 34.047, p < 0.001. Follow up z score tests revealed no significant difference between the proportion of kindergarten students receiving tangible reinforcements ( 24.48%) and the proportion of total reinforcements given to kindergarten students (25.67%; z = 0.6647, p = 0.509); second graders (respectively, 16.75% and 17.87%, z = 0.7132, p = 0.478), and fifth

PAGE 162

162 graders (respectively, 11.27% and 9.46%, z = 1.5016, p = 0.134). However, first grade students were significantly less likely to receive Lion Loot (12.53%) than they were to be represented in the school population (16.75%; z = 3.092, p = 0.002), as were fourth graders (respectively, 9.46% and 13.53%; z = 3.155 9, p = 0.002) Conversely, third graders were significantly more likely to receive Lion Loot (24.80%) than they were to be represented in the school population (17.23%; z = 4.2957, p < 0.001) These findings must be considered when assessing the researc h questions and discussing the meaningfulness of any findings associated with the administration of tangible reinforcements across the six student grade levels included in this study. Table 4 2. N umber of tangible reinforcements by grade level (lion loot) Grade N (1+) % (1+) # Loot % Loot Max Mean SD Kindergarten 152 24.48% 3,761 25.67% 89 22.12 17.87 Grade 1 104 16.75% 1,836 12.53% 83 16.25 12.79 Grade 2 104 16.75% 2,618 17.87% 99 22.77 20.42 Grade 3 107 17.23% 3,634 24.80% 128 31.33 27.1 0 Grade 4 84 13.53% 1,418 9.68% 68 13.13 13.64 Grade 5 70 11.27% 1,387 9.46% 80 14.01 17.08 Total (All Grades) 621 -14,654 -360 19.86 23.27 Note: (1+) statistics only include those students receiving one or more lion loot. Tangible reinforcements were only provided during the intervention year (2014 2015). In addition to student grade level, administration of Lion Loot to students must be considered in relation to student gender and reported race/ethnicity As shown in Table 4 3, as expected based on the t heoretical model that all students have an equal chance of receiving the tangible reinforcements, an approximately equal proportion of tangible reinforcements (i.e., Lion Loot) was administered to male students (N = 7,580; 51.73%) as to female students (N = 7,074; 43.27%) Square test of goodness of fit showed no significant difference between the distribution of Lion Loot across the

PAGE 163

163 two genders and the proportion of students receiving at least one tangible reinforcement, 2 (1, N = 621) = 0.207, p = 0.649 Table 4 3. D escriptives for independent variable by ethnicity and gender Gender A B H I W M Total Male N (1+) 15 110 37 1 144 20 327 % (1+) 4.6% 33.6% 11.3% 0.3% 44.0% 6.1% -# Lion Loot 467 2462 887 9 3218 537 7580 % Lion Loot 6.2% 32.5% 11.7% 0.1% 42.5% 7.1% -Max 73 125 79 9 97 83 125 Mean 29.19 18.51 21.63 9.00 19.27 19.89 19.64 SD 22.81 19.81 18.29 -19.54 21.91 19.81 Female N (1+) 13 85 40 1 131 24 294 % (1+) 4.4% 28.9% 13.6% 0.3% 44.6% 8.2% -# Lion Loot 448 1998 1106 68 3009 445 7074 % Lion Loot 6.3% 28.2% 15.6% 1.0% 42.5% 6.3% -Max 128 120 77 68 99 64 128 Mean 34.46 19.98 25.72 68.00 19.54 17.80 20.99 SD 33.23 20.87 17.06 -17.51 16.74 19.59 All Genders N (1+) 28 195 77 2 275 44 621 % (1+) 4.5% 31.4% 12.4% 0.3% 44.3% 7.1% -# Lion Loot 915 4460 1993 77 6227 982 14654 % Lion Loot 6.2% 30.4% 13.6% 0.5% 42.5% 6.7% -Max 128 125 79 68 99 83 128 Mean 31.55 19.14 23.73 38.50 19.40 18.88 20.27 SD 27.55 20.24 17.68 41.72 18.56 19.43 19.71 Note: (1+) statistics only include those students receiving one or more lion loot. Tangible reinforcements were only provided during the intervention year (2014 2015). Excludes 361 instances of Lion Loot which were missing student names and information. Categories determined by School Distric t and reported in alphabetical order. When exploring the administration of Lion Loot (i.e., tangible reinforcements) to students across racial and ethnic categories, it initially appears that Lion Loot were n American students (42.5% of all Lion Lion Loot) However, as

PAGE 164

164 shown in Table 4 3, when comparing instances of Lion Loot to the overall distribution of the students receiving at least one tangib le reinforcement, differences in administration of Lion Loot appear consistent with the student population Square test of goodness of fit was conducted to determine whether the actual distribution of students receiving at least one tangible reinforcement by racial and ethnicity categories differed from the theoretical distribution of students that were expected to receive a reinforcement (i.e., theoretically, the proportion of reinforcements for each ethnicity should be equal to the proporti on of students enrolled for each ethnicity) Ultimately, results suggest that students receiving tangible reinforcements were equally distributed across the six racial/ethnic groups according to the distribution of the population of students during the in tervention year that received at least one Lion Loot, 2 (5, N = 621) = 4.778, p = 0.444 As such, administration of Lion Loot by race/ethnicity is not further explored within this section of this chapter Further exploring the administration of tangible reinforcements (i.e., Lion Loot) across students, as shown in Table A 1 (Appendix A), the number of reinforcements provided to students was collected from a total of 60 teachers, staff members, and administrators working with students in kindergarten throu gh fifth grade. Two of the teachers included in the study did not provide any Lion Loot during the school year, and several teachers shared the same job during the course of the year (e.g., the school had more than one music teacher during the course of t he year). It is also important to note that Table A 1 (Appendix A) indicates the number of Lion Loot (reinforcements) exchanged for SWPBIS School Store items each month, while the actual reinforcement could have been administered at any time during or pri or to the month indicated.

PAGE 165

165 Regardless, a total of 15,015 Lion Loot were awarded during the academic year, with 360 being exchanged without connected student information and one (1) exchanged from a pre kindergarten student. These 361 Lion Loot were not i ncluded in descriptive statistics indicated previously in this chapter. As shown in Table A 1, the 58 staff members provided between 6 and 1,050 tangible reinforcements, with t he average staff member providing 250.25 Lion Loot during the course of the yea r (SD = 224.02). Table 4 4 provides the total number of Lion Loot by month, with nearly half (47.2%) of all Lion Loot being recorded in the last three months of the school year. Table 4 4 demonstrates the importance of using caution when interpreting da ta within this research according to distribution by month, with an understanding that data we re only able to be collected when the students exchanged the Lion Loot for backup reinforcements (not when they were awarded) As noted in the prior chapter, Lio n Loot were provided as secondary reinforcements to provide for immediate and consistent reinforcement of desired student behaviors. However, these secondary reinforcements only hold value insofar as they can be exchanged for back up reinforcements (i.e., high interest items provided within the SWPBIS School Store). Because some students saved their Lion Loot until they were able to afford a higher interest or more rewarding item, there were many instances of Lion Loot being exchanged in the months after they were originally awarded. This is further considered in the discussion chapter, but it is important to note the apparent pattern of exchange, with a large number of Lion Loot exchanged in the first month (9.7% of all Lion Loot) and nearly half of all Lion Loot exchanged in the last three months of the academic year. Table 4 4 also provides the

PAGE 166

166 formation and one coming from a pre kindergarten student and excluded from the study. Table 4 4. N umber of tangible reinforcements by month (lion loot) Type Sept Oct. Nov. Dec. Jan. Feb. Mar. Apr. May June Tot. All Lion Loot n 1457 981 888 942 1023 1325 1311 2031 3334 1723 15015 % 9.7% 6.5% 5.9% 6.3% 6.8% 8.8% 8.7% 13.5% 22.2% 11.5% -Identified Lion Loot n 1444 971 884 937 1007 1288 1240 1923 3278 1682 14654 % 9.6% 6.5% 5.9% 6.2% 6.7% 8.6% 8.3% 12.8% 21.8% 11.2% -Note: student identification and one (1) instance where the recipient was in pre kindergarten Section Two: Descriptive and Preliminary Analysis (Dependent Variable) This section focuses on descriptive analysis for the dependent variables (i.e., number of minor pre referrals and number of major office discipline referrals) As with the independent variable, scatter plots were first constructed for each dependent variable to ensure there were no significant outliers that may have corrupted or invalidated the results of further analyses No significant outliers were found within the dependent variables, though pre kindergarten students had already been excluded from this research du e issues already outlined regarding the receipt of Lion Loot (i.e., tangible reinforcements) and the overall lack of participation in the interventions associated with School Wide Positive Behavioral Support (SWPBIS) With the exclusion of pre kindergarte n students, the tables within this section provide descriptive statistics for the number of (1) minor/pre referrals given to students for the second baseline and intervention years, (2) major office disciplinary referrals given to students during all three years of the study, and (3) total referrals given to students for each of the three study years

PAGE 167

167 For all three years of this study (i.e., two baseline years and one intervention year), data were collected on the number of minor or pre referrals and the number of major office disciplinary referrals (ODRs), with the school not collecting data on minor/pre referrals during the first baseline year As shown in Table 4 5, a total of 125 students during the intervention year, 148 during the second baseline ye ar, and 32 during the first baseline year received at least one disciplinary action during the course of the respective academic year All students included in Table 4 5 received at least one referral from at least one of the two categories (i.e., minor o r major) Table 4 5. D escriptive statistics for dependent variables (discipline) Variable / Year N (1+) Min Max Mean (1+) SD (1+) Number of Minor / Pre Referrals 2012 2013 (Baseline 1) -----2013 2014 (Baseline 2) 141 1 24 3.52 4.02 2014 2015 (Intervention) 121 1 44 4.04 5.97 Number of Major Office Disciplinary Referrals 2012 2013 (Baseline 1) 32 1 14 2.41 2.65 2013 2014 (Baseline 2) 48 1 6 1.81 1.38 2014 2015 (Intervention) 36 1 4 1.61 0.87 Number of All Referrals (Total) 2012 2013 (Baseline 1) 32 1 14 2.41 2.65 2013 2014 (Baseline 2) 150 1 29 3.98 4.80 2014 2015 (Intervention) 125 1 45 4.38 6.36 Note: (1+) statistics only include those students with one or more instance under each category. Minor / Pre Referrals were not recorded by the school in 2012 2013. Students receiving at least one referral of any type tended to receive more than one referral, with the average student receiving 2.41 total referrals in the first baseline year, 3.94 total referrals in the second baseline year, and 4.38 total referrals in the intervention year The higher average number of referrals in the intervention year must be considered in conjunction with the lower number of total referrals given during the

PAGE 168

168 intervention year (N=547) than th e number of referrals given during the second baseline year (N=583) This suggests that fewer students required behavioral intervention during the intervention year, but required the intervention more often per student It is also noted that the change i n average total referrals appears driven by changes in the average number of minor or pre referrals, which is expected with the SWPBIS model that encourages the pre referrals before major ODRs are written Ultimately, the data outlined in Table 4 5 reveal no major concerns for proceeding with using both minor/pre referrals and major referrals as the dependent variables for this study While there appears to be a large number of referrals and students receiving referrals, several caveats are important to consider when interpreting results from this research First, during the intervention year when data from the entire school were available, the total number of students receiving at least one referral (N = 125) is only a small percentage (17.3%) of the total student population from Kindergarten to 5th grade (N = 721) such that the vast majority of students did not receive referrals and are not reflected in Table 4 5 Second, a small number of students accounted for a large number of referrals More specifically, intervention year data show that only 17 students rec eived ten or more referrals 13.6% of the 125 students with at least one referral during the intervention year with these students accounting for 50.8% of all referrals written that year (N = 278 of 547 referrals) Similarly, data from baseline year sh ow that only 12 students received 10 or more referrals, representing only 9.6% of the all students receiving referrals, but 33.1% of all referrals given (N = 193 of 583 referrals) These findings are incorporated into the discussion of results, when appro priate, and should be considered when interpreting the results within this chapter.

PAGE 169

169 In addition to general information about instances of disciplinary measures, it is important to explore whether there are problematic or important differences between three primary demographic variables collected for each year of the study (i.e., baseline years and intervention year): (1) student grade level; (2) student race and ethnicity; and (3) student gender For each of these demographic variables, this section explor es the relationship between the school distribution and the distribution of students receiving minor/pre referrals and/or major office disciplinary referrals However, as school wide demographics were only available for the intervention year at the studen t level, only the distributions from the intervention year are compared for the purposes of the preliminary statistics presented in this section Following sections are designed to answer research questions associated with this study, while this section p rovides only preliminary comparisons and explorations for the purposes of improving the interpretation of results Table 4 6 presents the distribution of minor/pre referrals and major office disciplinary referrals. As discussed previously, minor referra ls were not recorded by the school in the first baseline year, while major office disciplinary referrals were recorded for all three years (i.e., first baseline, second baseline, and intervention years). While there appears to be some differences between the number of referrals given each year to students at each grade level, any analysis of the data to explore such differences is reserved for the specific research questions and more sophisticated statistical comparisons detailed below. However, prelimina ry statistics are important to explore how the distribution of referrals across grade levels compares to the distribution of all students attending the school. As noted previously, a total of 721 students attended the school during the intervention year in grades Kindergarten through fifth grade, with 170 in kindergarten

PAGE 170

170 (23.6%), 113 in first grade (15.7%), 115 in second grade (16.0%), 116 in third grade (16.1%), 108 in fourth grade (15.0%), and 99 in fifth grade (13.7%). In comparison to the distribution of all referrals (both minor pre referrals and major office disciplinary referrals), a Square test of goodness of fit was conducted to determine w hether the actual distribution of referrals across student grade levels differed from the theoretical distribution of referrals if all students had equal changes of receiving referrals (i.e., theoretically, the proportion of referrals for each grade level should be equal to the proportion of students enrolled within each grade level ) Ultimately, results suggest that referrals were not proportionally distributed across student grade levels according to the distribution of the population of students during the intervention year, 2 (5, N = 721) = 62.192, p < 0. 001. Post hoc z score tests revealed no significant difference between the proportion of referrals given to second grade students (17.9%) and the proportion of second graders at the school during the intervention year (16.0%; z = 0.9274, p = 0.352), the proportion of referrals to third graders (12.8%) and the proportion of enrolled third graders (16.1%; z = 1.6409, p = 0.101), and the proportion of referrals to fifth graders (11.2%) and the proportio n of enrolled fifth graders (13.7%; z = 1.3699, p = 0.171). However, kindergarten students received a significantly higher proportion of referrals (29.1%) than the proportion of kindergarten students enrolled in the school during the intervention year (23 .6%; z = 2.2085, p = 0.027), as did fourth grade students (24.9% of referrals, 15.0% of enrolled students; z = 4.4218, p < .001). Conversely, first grade students had a significantly lower proportion of referrals (4.2%) than the proportion of first grad e students enrolled during the intervention year (15.7%; z = 6.5360, p < .001). Interestingly, as noted previously, first graders were also less likely to receive the

PAGE 171

171 tangible reinforcements than they were to be represented in the school population, as we re fourth graders. Although proportional differences of referrals are more acceptable due to the individualized nature of referrals, these findings must still be considered when assessing the research questions and discussing the meaningfulness of any fin dings associated with the administration of tangible reinforcements across the six student grade levels included in this study. Table 4 6. M inor and major referrals by school grade Grade Minor / Pre Referrals Major Office Referrals All Referrals (Total ) B1 B2 INT B1 B2 INT B1 B2 INT KG N -71 139 24 13 20 24 84 159 % -14.3% 28.4% 31.2% 14.9% 34.5% 31.2% 14.4% 29.1% Gr. 1 N -43 22 14 9 1 14 52 23 % -8.7% 4.5% 18.2% 10.3% 1.7% 18.2% 8.9% 4.2% Gr. 2 N -81 89 11 14 9 11 95 98 % -16.3% 18.2% 14.3% 16.1% 15.5% 14.3% 16.3% 17.9% Gr. 3 N -99 61 10 16 9 10 115 70 % -20% 12.5% 13% 18.4% 15.5% 13% 19.7% 12.8% Gr. 4 N -79 125 6 21 11 6 100 136 % -15.9% 25.6% 7.8% 24.1% 19% 7.8% 17.2% 24.9% Gr. 5 N -123 53 12 14 8 12 137 61 % -24.8% 10.8% 15.6% 16.1% 13.8% 15.6% 23.5% 11.2% Total N -496 489 77 87 58 77 583 547 Note: B1 and B2 are the two baseline years, INT is the intervention year. Minor / Pre Referrals were not recorded by the school in 2012 2013 (B1) Percentages represent percent of total number of each referral type for each respective year. indicates significant difference from proportion of students in school population. In addition to student grade level, incidents of disciplinary referrals gi ven to students must be considered in relation to student gender and reported race/ethnicity As shown in Table 4 7, the distribution of referrals across student gender categories appears to be disproportional to the overall student population More spec ifically, the population of students attending the school during the intervention year (N = 721) were

PAGE 172

172 composed of 53.4% males (N = 385) and 46.6% females (N = 336) Theoretically, students of all genders have equal chances of receiving referrals, such tha t it would be theoretically expected for the number of referrals to be distributed the same as the overall population of students However, using data shown in Table 4 Chi Square test of goodness of fit using the total of all referrals duri ng the intervention year showed a significant difference in the distribution of the numbers of referrals by gender and the overall distribution of students by gender, 2 (1, N = 721) = 169.3, p < .001 In this light, there was a significantly greater prop ortion of referrals given to male students (87.8% of referrals) during the intervention year than the proportion of students enrolled in the school that same year (53.4% of students) This is largely consistent with past research showing that boys tend to receive more referrals than girls, largely due to the fact that boys engage in a broader range of disruptive behavior (e.g., Skiba, Michael, Nardo, & Peterson, 2000) However, this significant difference between genders must be considered when discussing the results of this research and will be further covered within the discussion section R esearch questions already propose d to analyze differences between gender categories. Table 4 7. M inor and major referrals by student gender Gender Minor / Pre Referrals Major Office Referrals All Referrals (Total) B1 B2 INT B1 B2 INT B1 B2 INT Male N -422 427 68 71 53 68 493 480 % -85.1% 87.3% 88.3% 81.6% 91.4% 88.3% 84.6% 87.8% Female N -74 62 9 16 5 9 90 67 % -14.9% 12.7% 11.7% 18.4% 8.6% 11.7% 15.4% 12.2% Total N -496 489 77 87 58 77 583 547 Note: B1 and B2 are the two baseline years, INT is the intervention year. Minor / Pre Referrals were not recorded by the school in 2012 2013 (B1) Percentages represent percent of total number of each referral type for each respective year. indicates significant difference from proportion of students in school population.

PAGE 173

173 When exploring the distribution of referrals across racial and ethnic categories, the majority of referrals were given to during the first baseline year (57.1% of all referrals), the second baseline year (57.6%), and the intervention year ( 56.7% ) To explore whether the distribution of referrals administered during the interventio n year was consistent with the distribution of enrolled students by reported race and ethnicity, a Square test of goodness of fit was conducted with the theoretical basis that all students had relatively equal chance of receiving a referral d uring the intervention year. Ultimately, results suggest that the distribution of referrals across students by race and ethnicity was significantly different than the distribution of all students across the same racial and ethnic categories, 2 (5, N = 721 ) = 82.378, p < 0. 001. Post hoc z score tests revealed no significant difference between the proportion of referrals given to Asian American students (2.7%) and the proportion of Asian American students at the school during the intervention year (4.0%; z = 1.2334, p = 0.219), the proportion of referrals to Native American and Pacific Island students (0.3%) and the proportion of enrolled Native American and Pacific Island students (0.2%; z = 0.3433, p = 0.728), and the proportion of referrals to students id entified as multiple ethnicities (7.2%) and the proportion of enrolled students identified as multiple ethnicities (6.4%; z = 0.5677, p = 0.569). However, the proportion of referrals given to students identified as Hispanic (4.4%) and White / Caucasian Am erican (29.6%) were both significantly lower than the proportion of enrolled students during the intervention year that were identified as, respectively, Hispanic (11.7%; z = 4.5887, p < .001) and White / Caucasian American (44.5%, z = 5.4132, p < .001). Conversely, the proportion of referrals given to students identified as Black / African American (56.7%) was

PAGE 174

174 significantly higher than the proportion of enrolled students identified as Black / African American (32.3%, z = 8.681, p < .001). Such differenc es are important considerations for the application of discipline at the intervention school, and are relatively consistent with findings of past research ( e.g., Skiba, et al. 2000 ). It is important to note that two of the four research questions include student race and ethnicity within the statistical analysis. Table 4 8. M inor and major referrals by student ethnicity / race Race Minor / Pre Referrals Major Office Referrals All Referrals (Total) B1 B2 INT B1 B2 INT B1 B2 INT A N -2 13 0 0 2 0 2 15 % -0.4% 2.7% 0% 0% 3.4% 0% 0.3% 2.7% B N -283 277 44 53 33 44 336 310 % -57.1% 56.6% 57.1% 60.9% 56.9% 57.1% 57.6% 56.7% H N -34 22 16 4 2 16 38 24 % -6.9% 4.5% 20.8% 4.6% 3.4% 20.8% 6.5% 4.4% I N -0 1 0 0 0 0 0 1 % -0% 0.2% 0% 0% 0% 0% 0% 0.2% W N -115 143 13 15 19 13 130 162 % -23.2% 29.2% 16.9% 17.2% 32.8% 16.9% 22.3% 29.6% M N -62 33 4 15 2 4 77 35 % -12.5% 6.7% 5.2% 17.2% 3.4% 5.2% 13.2% 6.4% Total N -496 489 77 87 58 77 583 547 Note: B1 and B2 are the two baseline years, INT is the intervention year. Minor / Pre Referrals were not recorded by the school in 2012 2013 (B1) Percentages represent percent of total number of each referral type for each respective year. Ethnicity Codes: A order. indicates significant difference from proportion of students in school population. Table 4 9 provides the total number of referrals given to students for each month of the academic school year, with the first and last months of the school year having the lowest incidence of referrals for all three years for which data were collected (i.e., first baseline year, second baseline year, and intervention year). During the first baseline

PAGE 175

175 year, 32.5% of referrals were given during the first four months and 45.5% were given during the last four months, which was relatively consistent in the second baseline year (27.0% and 45.3%, respectively). The intervention year shows a more balanced distribution of referrals, with 35.1% given during the first four m onths, 32.2% given during the middle three months, and 32.8% given during the last four months. Table 4 9. D istribution of minor and major referrals by month Month Minor / Pre Referrals Major Office Referrals All Referrals (Total) B1 B2 INT B1 B2 INT B1 B2 INT Aug. N -2 8 2 5 1 2 7 9 % -0.4% 1.6% 2.6% 5.7% 1.7% 2.6% 1.2% 1.6% Sept. N -37 73 6 13 5 6 50 78 % -7.5% 14.9% 7.8% 14.9% 8.6% 7.8% 8.6% 14.3% Oct. N -43 50 11 7 6 11 50 56 % -8.7% 10.2% 14.3% 8% 10.3% 14.3% 8.6% 10.2% Nov. N -39 47 6 11 2 6 50 49 % -7.9% 9.6% 7.8% 12.6% 3.4% 7.8% 8.6% 9% Dec. N -34 33 6 5 1 6 39 34 % -6.9% 6.7% 7.8% 5.7% 1.7% 7.8% 6.7% 6.2% Jan. N -62 50 2 8 5 2 70 55 % -12.5% 10.2% 2.6% 9.2% 8.6% 2.6% 12% 10.1% Feb. N -46 71 9 7 16 9 53 87 % -9.3% 14.5% 11.7% 8% 27.6% 11.7% 9.1% 15.9% Mar. N -42 43 10 13 6 10 55 49 % -8.5% 8.8% 13% 14.9% 10.3% 13% 9.4% 9% Apr. N -65 60 13 10 6 13 75 66 % -13.1% 12.3% 16.9% 11.5% 10.3% 16.9% 12.9% 12.1% May N -121 52 12 8 10 12 129 62 % -24.4% 10.6% 15.6% 9.2% 17.2% 15.6% 22.1% 11.3% June N -5 2 ----5 2 % -1.0% 0.4% ----0.9% 0.4% Total N -496 489 77 87 58 77 583 547 Note: B1 and B2 are the two baseline years, INT is the intervention year. Minor / Pre Referrals were not recorded by the school in 2012 2013 (B1) Percentages represent percent of total number of each referral type for each respective year. indicates significant difference from Baseline 2 to Intervention in monthly proportion of referrals

PAGE 176

176 Square test of goodness of fit was conducted to compare the distribution of referrals during the second baseline year and the intervention year, with the two distributions shown to be significantly different ( 2 (10, N = 578) = 40.613, p < 0.001 ) Post hoc z score tests revealed no significant differences for most months of the academic year, though did reveal significant differences between the proportion of referrals given in September from the second baseline year (8.6%) to the intervention year (14.3%; z = 3.0125, p = .003); February (9.1% to 15.9%; z = 3.4745, p < .001); and May (22.1% to 11.3%, z = 4.8377, p < .001) Unlike Lion Loot (i.e., the tangible reinforcements), referral data were more accurately collected and recorded in the month where they actually occurred (rather than when the students exchanged the tickets for Lion Loot). This is further considered in the discussion chapter, but it is important to note the apparent pattern of referrals was more balanced a cross the year during the intervention year, whereas baseline years were largely loaded towards the end of the academic years. Section Three: Selection of Analyses There are four research questions associated with this study Each of the research questions are related with research questions one and two directly related, and research questions three and four directly related These two sets of research questions are explored within section four and five of this dissertation, with research questi ons one and two the focus of section four The first two research questions are focused on whether the number of negative student behaviors changed from baseline to intervention, with research question two expanding the exploration by adding difference ac ross race/ethnicity and gender The second two research questions focus on whether the level of Lion Loot (tangible reinforcers) awarded can predict student negative

PAGE 177

177 behaviors Regardless of the research question being explored, all data were first analy zed for normalcy to ensure assumptions of parametric statistics were not violated In addition, data were explored for outliers, with no potential outliers identified that were determined to impact the results, such that all students on whom data were col lected were included in the presented analyses All research questions were explored with a two tailed alpha level of .05. There are some important limitations to the data which must be discussed in the results section and expanded upon in the discussion, as these issues and limitations helped guide the selection of analyses First, as noted previously, for all research questions, negative student behaviors were measured with two specific metrics: (1) number of minor / pre referrals and (2) number of majo r office disciplinary referrals When possible, the number of referrals were measured both as the total number of referrals during the baseline and intervention years, as well as the difference between the baseline and intervention years To explore the difference between baseline and intervention years, the total number of referrals from the intervention year was subtracted from the number of referrals during the baseline year with th ese differences then squared to ensure no negative values It is note d that data and statistical findings involving difference scores were interpreted using the square of the differences It is important to reiterate that the school did not collect data on minor / pre referrals during the first baseline year Moreover, m inor referrals from the first baseline year were not stored in any centralized database and were not stored at the school level, such that there was no possibility to retrieve information on minor / pre referrals from the first baseline year The school b egan collecting information on minor / pre referrals in the second baseline year, specifically to begin the planning and design

PAGE 178

178 process for the School Wide Positive Behavioral Intervention and Support (SWPBIS) model at focus in this research With this li mitation, the originally planned analyzes for all four research questions were impacted, as the first baseline year could not be utilized in analyses exploring minor referrals or total referrals. Within the original analysis plan there was also considerat ion for exploring student behavioral impacts both by year and month, particularly for research questions three and four (e.g., exploring whether the implementation of SWPBIS reinforcements would impact referrals by month) However, it became clear that th ere were some potential complications with a ny type of monthly analysis First, there were some differences by year as to when district diagnostics are provided, when spring break occurs, and when standardized testing occurs all events that can have an impact on referrals and student behaviors In addition to schedule differences, there were also issues with students unexpectedly (Lion Loot) from prior months to afford a more desirab le incentive later in the year (e.g., if there is was dinosaur egg that cost 50 tickets and they only receive 5 tickets a month, they might hold all their tickets until the final month of the year) This was not a foreseen issue, and the study procedures provided no alternative method with which to assess the month during which the reinforcements were earned (only when they were used at the SWPBIS School Store ) There were also possible issues with specific teachers and the number of referrals, where new er teachers may give fewer referrals at the beginning of the year because they do not want to develop a reputation for sending students out of class (or any number of other reasons) or may with none of these individual characteristics a focus of this research As such, based on these primary

PAGE 179

179 concerns and limitations, all four Research Questions were analyzed based on aggregate data for the entire school year, rather than by month In ad dition to time based regression analyses (i.e., exploring data by month), the original plan also called for regression to explore both the number of referrals and the number of tangible reinforcements (Lion Loot) provided to individual students across the course of the baseline and intervention years Although this had the possibility to provide the most interesting results, two primary data characteristics precluded the usefulness of regression to explore these research questions as originally anticipated First, it is noted there were many students with no referrals (major and/or minor) in both the baseline and intervention periods, thus producing a large number of students with scores for referrals and lower levels of Lion Loot created a considerable skew in the distribution (with the overwhelming majority of students having no referrals of any type) which led to the distribution of both dependent and independent variables viola ting the assumption of a Gaussian (or normal) distribution and the mean number of referrals being very near zero As such, exploring data at a st udent level would not only violate the assumption of normality a cornerstone assumption of parametric stati stics but any regression using individual, student level data would likely have a very low R2 and be relatively uninterpretable Fortunately, the original plan of statistical analyses had anticipated this potential scenario and, should regression produc e a result with low R2 because of the having no referrals), non parametric statistics were planned to be used for additional

PAGE 180

180 analyses at the student level or classroom lev el However, non parametric statistics rely upon the median as the measure of central tendency, which was also complicated by the overwhelming number of students with true zero values on the dependent variables (i.e., number of minor pre referrals, number of major office disciplinary referrals, and total number of referrals over the course of the year) In addition to possible non parametric statistics in lieu of regressions, analyses also included Chi Square analyses (presented in prior sections on demo graphic and descriptive statistics), z tests of independent proportions, analyses of variance (ANOVAs), and multi variate analyses of variance (MANOVAs) Although the ANOVA and MANOVA procedures are parametric and hold the same assumption for Gaussian dis tributions as regression analysis, these procedures are more robust than regression when the non normality is secondary to a skewed distribution rather than outliers, such as with the current research study (Tabachnick & Fidell, 1996; French, Macedo, Pouls en, Waterson, & Yu, 2002) Results of any necessary analysis of other assumptions and details of the specific characteristics of each statistical procedure performed are provided in the following sections of this dissertation Results from such procedure s are also detailed in the following sections in this chapter. Section Four: Research Question One and Two The first research question is focused on exploring whether the students attending the targeted school demonstrated fewer referrals during the intervention year than during the baseline years More specifically, the first research question reads: of the SWPBIS intervention from baseline years (pre intervent ion) and the intervention year (post intervention), as measured by the number of instances of negative student

PAGE 181

181 The second research question expands upon the first, with the additi on of race/ethnicity and grade Is there a difference by race/ethnicity and/or student grade level in the impact of the SWPBIS intervention from baseline to implementation on the number of negative student behaviors? As noted in the prior section, the first research question was originally intended to be assessed at the student level through the use of regression analysis by month and year, with inherent data characteristics precluding the analysis of data by month u sing regression analysis As such, analyses of variance (ANOVAs) are the more appropriate parametric method to determine whether there were differences between the three years of the study in terms of the number of negative student behaviors recorded thro ugh punitive behavioral interventions (i.e., minor/pre referrals, major office disciplinary referrals, and total referrals) However, as also discussed in the prior vas t majority of students led to a skewed distribution, with the mean number of referrals being very near zero As such, any results of parametric statistics must be considered with caution due to the violation of the Gaussian Distribution and the reliance o n means in the analysis, though the large sample sizes help mitigate these effects. The first and second research questions were explored with two analyses: an analysis of variance (ANOVA) and a multivariate analysis of variance (MANOVA) Using data from the 848 students with data from at least one of the three data collection years, the initial ANOVA was performed to explore the relationship of academic year (i.e., first baseline year, second baseline year, and intervention year), student grade level, and student race/ethnicity on the instances of negative student behaviors

PAGE 182

182 measured only by major office disciplinary referrals Because the first baseline year did not include minor/pre referral data, it could not be included in a MANOVA or any analyses invo lving minor/pre referral or total referral data As such, a MANOVA was performed using data from only the second baseline year and the intervention year to explore the relationship of academic year, student grade level, and student race/ethnicity on the i nstances of negative student behaviors using all three dependent variables (i.e., minor/pre referrals, major office disciplinary referrals, and total referrals) The first research question focuses only on the main effect of academic year on the instances of negative student behaviors, while the second research question expands upon the first, and is most concerned with the interaction effects between academic year and student categories (i.e., grade level and race/ethnicity) As noted above, a three way analysis of variance (ANOVA) was conducted to explore the number of major office disciplinary referrals (i.e., major negative student behaviors) based on academic year (i.e., first baseline year, second baseline year, and intervention year), student grade leve l (i.e., KG, 1 st 2 nd 3 rd 4 th and 5 th grades), and student race/ethnicity (i.e., African American / Black, Hispanic / Latina(o), Multi Ethnic, and Caucasian American / White) Results of the three way ANOVA are presented in Table 4 10 As shown, specific to the first research question, there was a significant main effect of academic year (F(2,1773)=3.731, p=0.024) based on the omnibus F test performed by the ANOVA However, post hoc analyses using the Tukey HSD test to control for multiple pairwi se comparisons, demonstrated no significant differences between the average number of referrals per student for the first baseline year (M=0.13, SD=0.828), second baseline year (M=0.15, SD=0.636); and intervention year (M=0.08, SD=0.402) Such a result is not unusual, and the omnibus F test performed by the

PAGE 183

183 ANOVA may be the effect of a more complex contrast (Cardinal & Aitken, 2006), and may be due to a lack of statistical power to resolve the more conservative post hoc comparisons performed using the Tuke y HSD method In addition to the main effect of academic year on instances of negative student behaviors (i.e., major office disciplinary referrals), Table 4 10 also provides results of the ANOVA for the second research question The second research que stion focuses on whether there were significant differences across years in instances of negative student behaviors between race/ethnicity and/or student grade level As such, the second research question is primarily concerned with the interactions produ ced by the ANOVA wherein academic year is considered As shown, results of the ANOVA revealed no significant interactions with academic year and student race/ethnicity (MS =0.678, F(6,1773)=1.714, p=0.114); academic year and student grade level (MS =0.481 F(10,1773)=1.215, p=0.276); or with the three way interaction of academic year, student grade level, and student race/ethnicity (MS=0.454, F(30,1773)=1.148, p=0.266) Table 4 10. ANOVA table (major office disciplinary referrals) Source Type III Sum of Sqs df M ean Square F p Year 2.952 2 1.476 3.731 .024 Ethnicity 12.017 3 4.006 10.126 .000 ** Grade 1.149 5 .230 .581 .715 Year Ethnicity 4.067 6 .678 1.714 .114 Year Grade 4.806 10 .481 1.215 .276 Ethnicity Grade 5.696 15 .380 .960 .496 Year Ethnicity Grade 13.620 30 .454 1.148 .266 Error 701.360 1773 .396 Total 767.000 1845 ** Significant finding at the established Alpha level of 0.05

PAGE 184

184 While no interactions were significant, the results of the ANOVA demonstrate a significant main effect of ethnicity (MS=4.006, F(3,1733)=10.126, p<0.001), wherein there were significant differences in the number of major office disciplinary referrals betwe en student race/ethnicity categories when combined across all three academic years of interest (i.e., first baseline year, second baseline year, and intervention year) Post hoc analyses were performed using the Tukey HSD test to control for multiple pair wise comparisons Post hoc analyses demonstrated that African American / Black students were more likely to receive major office disciplinary referrals (M=0.22, SD=0.774) across all three academic years of study combined than were Caucasian American / Whi te (M=0.05, SD=0.354; M diff =0.17, MS e =0.396, p<0.001) There were no significant differences between the number of major office disciplinary referrals given to students within these racial/ethnic groups and those identified as Hispanic / Latina(o) (M=0.09 SD=0.935) or Multi Ethnic (M=0.15, SD=0.656) A three way multivariate analysis of variance (MANOVA) was also conducted to explore the instances of negative student behaviors (i.e., minor/pre referrals, major office disciplinary referrals, and total re ferrals) based on academic year (i.e., second baseline year and intervention year), student grade leve l (i.e., KG, 1 st 2 nd 3 rd 4 th and 5 th grades), and student race/ethnicity (i.e., African American / Black, Hispanic / Latina(o), Multi Ethnic, and Cauc asian American / White) It is noted that the first baseline years is excluded from the MANOVA, as there were no minor/pre referral data calculated (as this is th e summation of the minor/pre referrals and the major office disciplinary referrals) Results of the MANOVA are provided in Table 4 11

PAGE 185

185 Table 4 11 MANOVA table (minor, major, and total r eferrals ) Source Dependent Variable Type III SS df Mean Square F p Year Minor 19.796 1 19.796 2.736 .098 Major 2.740 1 2.740 10.226 .001 ** Total 37.263 1 37.263 4.066 .044 ** Ethnicity Minor 297.729 3 99.243 13.719 .000 ** Major 7.611 3 2.537 9.470 .000 ** Total 399.370 3 133.123 14.527 .000 ** Grade Minor 52.100 5 10.420 1.440 .207 Major 2.383 5 .477 1.779 .114 Total 75.671 5 15.134 1.651 .144 Year Ethnicity Minor 15.147 3 5.049 .698 .553 Major 2.381 3 .794 2.962 .031 ** Total 29.514 3 9.838 1.074 .359 Year Grade Minor 16.365 5 3.273 .452 .812 Major 1.695 5 .339 1.266 .276 Total 21.743 5 4.349 .475 .795 Ethnicity Grade Minor 149.919 15 9.995 1.382 .148 Major 4.792 15 .319 1.193 .270 Total 192.559 15 12.837 1.401 .139 Year Ethnicity Grade Minor 109.597 15 7.306 1.010 .442 Major 2.969 15 .198 .739 .746 Total 120.399 15 8.027 .876 .592 Error Minor 8847.075 1223 7.234 Major 327.630 1223 .268 Total 11207.613 1223 9.164 Total Minor 10260.000 1271 Major 364.000 1271 Total 13046.000 1271 In relation to the first research question, results of the MANOVA demonstrate significant main effects for academic year in terms of major office disciplinary referrals (MS=2.740, F(1,1223)=10.226, p=0.001) and total referrals (summation of minor/pre refer rals and major office disciplinary referrals; MS=37.263, F(1,1223)=4.066, p=0.044). More specifically, students were more likely to receive major office disciplinary referrals

PAGE 186

186 during the second baseline year (M=0.15, SD=0 .636 ) than during the intervention year (M=0.08, SD=0 .402 ), and were more likely to receive any type of referral during the second baseline year (M=1.02, SD= 2.995 ) than during the intervention year (M=0.77, SD= 3.148 ). Results demonstrate no significant difference in receipt of minor/pre referrals from the second baseline year (M=0.87, SD= 2.520 ) to the intervention year (M=0.69, SD= 2.903 ; MS=19.796, F(1,1223)=2.736, p=0.098). As introduced previously, the second research question builds upon the first research question to explore whether differences in negative student behaviors (as measured by minor/pre referrals, major office disciplinary referrals, and total referrals) are further influenced by student grade level and/or student race/ethnicity As shown in Table 4 11, combined data across both the second baseline year and the intervention year demonstrate significant main effects of ethnicity for all three dependent variables: minor/pre referrals (MS=99.243, F(3,1223)=13.719, p<0.001), major office disciplinary referrals (MS=2. 537, F(3,1223)=9.470, p<0.001), and total referrals (MS=133.123, F(3,1223)=14.527, p<0.001) for multiple pairwise comparisons demonstr ate that African American / Black students were significantly more likely to receive minor/pre referrals (M=1.31, SD=3.836) than both Hispanic / Latina(o) students (M=0.34, SD=1.135 ; M diff =0.96, p=.001 ) and Caucasian American / White students (M=0.45, SD=1.821; M diff =0.86, p<.001) across the second baseline year and inte rvention year Similarly, across both years, African American / Black students were more likely to receive major office disciplinary referrals (M= 0.20, SD=0 .676 ) than both Hispanic / Latina(o) students (M=0 .0 4, SD= 0.189; M diff =0.16, p=.004 ) and Caucasian American / White students (M=0 .06 SD= 0.372 ; M diff =0.14, p<.001) Finally, the same pattern was found for the total referrals (the

PAGE 187

187 summation of minor/pre referrals and major office disciplinary referrals), with African American / Black students significan tly more likely to receive any referral (M= 1.51 SD= 4.238 ) than both Hispanic / Latina(o) students ( M=0.38, SD=1.223; M diff =1.13, p<.001 ) and Caucasian American / White students ( M=0.50 SD= 2.109 ; M diff =1.00, p<.001) across the second baseline year and int ervention year. While main effects are important, the second research question is more focused on the interactions between academic year and the two demographic variables (i.e., student race/ethnicity and student grade level) As shown in Table 4 11, the MANOVA produced no significant interactions between student grade level and academic year for minor/pre referrals (MS=3.273, F(5,1223)= 0.452, p=0.812), major referrals (MS=0.339, F(5,1223)=1.266, p=0.276), or the summative total referrals (MS=4.349, F(5,1 223)=0.0.475, p=0.795) There were also no significant three way interactions involving academic year, student grade level, or student race/ethnicity However, the MANOVA revealed a significant interaction for major office disciplinary referrals between the second baseline year and the intervention year for students within different racial/ethnic groups (MS=0.794, F(3,1223)=2.962, p=0.031) Post hoc simple effects analysis using estimated marginal means and the Bonferroni correction for multiple compari sons (used to explore the significant interaction) revealed that African American students received a significantly higher number of major office disciplinary referrals than White in the first baseline year (M AA =0.278, M CA =0.041, M diff =0.237, StdErr=0.061, p=0.001) and in the second baseline year (M AA =0.263, M CA =0.058, M diff =0.205, StdErr=0.059, p=0.003) However, this pattern of significant differences was not found during the intervention year, with African American / Black students receiving a statistic ally similar average number of

PAGE 188

188 major office disciplinary referrals as their fellow Caucasian American / White students (M AA =0.142, M CA =0.059, M diff =0.082, StdErr=0.054, p=.764) Post hoc simple effects analysis on the second baseline year also revealed th at students identified as multi ethnic had a significantly higher number of major office disciplinary referrals on average, than both Hispanic / Latina(o) students (M ME =0.385, M HA =0.051, M diff =0.334, StdErr=0.123, p=0.040) and Caucasian American / White s tudents (M ME =0.385, M CA =0.058, M diff =0.326, StdErr= 0.108, p=0.015) However, based on simple main effects analysis using the Bonferroni correction, there were no significant differences in the average number of major office disciplinary referrals between any of the four racial / ethnic groups explored (i.e., African American / Black; Hispanic / Latina(o); Multi Ethnic, and Caucasian American / White) during the intervention year Test of independent p roportions While ANOVA and MANOVA help illustrate sig nificant differences in average instances of negative student behaviors, the aforementioned issues inherent in the data (e.g., predominantly true zeros, skewed distribution, etc.) necessitate some caution when interpreting the results of statistical analyses at the student level As such, to help answer the first and second research questions, additional analyse s were conducted to explore the school wide proportion of instances of negative student behaviors More specifically, a test of independent proportions was conducted to explore the proportion of minor/pre referrals, major office disciplinary referrals, and total referrals based on the total number of students enrolled during each academic year of interest (i.e., first baseline year, second baseline year, and intervention year) Exploration of annual proportions of referrals is not as heavily impacted by the fact that most students have no referrals, as it explores the propor tion of all students with at least one referral and/or the proportion of referrals across all

PAGE 189

189 students enrolled This method also controls for the differences in number of students per year For instance, t here were 50 fewer total referrals from the seco nd baseline year to the intervention year (8.38% of the total baseline referrals), but this can be misleading due to the fact that there were 121 more students in the intervention year attending the school (a 20.17% increase in enrollment) As such, it i s important to do a test of independent proportions to determine whether there is a difference in proportions of referrals across the two years (the first baseline year does not have minor/pre referrals, such that total referrals cannot be calculated) Th e z test of independent proportions is also known as the t test of independent proportions and the critical ratio test Regardless of the name, it is identical to the chi square test, except that the standard normal deviate is estimated (e.g., Snedecore & Cochran, 1991; Woodward, 1999) It is no ted that research suggests a significant reduction in referrals can be expected within the first year of SWPBIS implementation with most research focused on sheer numbers of referrals rather than proportional anal ysis (e.g., George, et al., 2003 ; Luiselli, Putnam, Handler, & Feinber g, 2005 ; Scott & Barrett, 2004; Spencer, 2015) In terms of the present rese arch, the school realized a 20.17 % increase in enrollment from the second baseline year to the intervention y ear, coupled with an 23.75 % decrease in total referrals, a 20.05 % decrease in minor/pre referrals, and a 45.15 % decrease in major office disciplinary referrals C omparing proportions and changes is not necessarily meaningful if not statistically significant Table 4 12 provides the proportionality of minor/pre referrals, major office disciplinary referrals, and total referrals based on the total student enrollment for each year of st udy (i.e., first baseline year, second baseline year, and intervention year)

PAGE 190

190 As shown in Table 4 12, t he school at focus enrolled 594 students in the first baseline year, 600 students in the second baseline year, and 721 students in the intervention year In terms of major office disciplinary referrals (ODRs), the school administered 77 ODRs to 32 students in the first baseline year, 88 ODRs to 49 students in the second baseline year, and 58 ODRs to 36 students in the intervention year The proporti onality of major office disciplinary referrals provided to the total number of students enrolled was 12.96% in the first baseline year, 14.67% in the second baseline year, and 8.04% in the intervention year A test of independent proportions on major offi ce disciplinary referrals indicates that there was no significant difference in the proportion of major referrals to enrolled students from the first baseline year (12.96% ) to the second baseline year ( 14.67%; Z= 0.8529, p=0.395 ) However, a test of indep endent proportions demonstrated a significant difference between the proportion of referrals given to enrolled students in the second baseline year (the year immediately prior to implementing SWPBIS) and the intervention year with 88 major referrals acro ss 600 students in the second baseline year (14.67%), compared to 58 major referrals among 721 students in the intervention year (8.04%; Z=3.8221, p<0.001) As shown in Table 4 12, the number of major office disciplinary referrals decreased by 30 from th e second baseline year to the intervention year (34.09% reduction) while the proportion of major referrals to enrolled students decreased 45.15% from the second baseline year to the intervention year The significant decrease in proportionality is mainta ined from the s econd baseline year to the intervention year when exploring the proportion of students receiving referrals compared to the total enrollment, with the intervention year having a significantly lower proportion of students receiving major off ic e disciplinary referrals (4.99 %) than the second baseline

PAGE 191

191 year (8.17 %; Z=2.3406 p =0.019 ) This equates to a 38.89 % decrease in the proportion of students receiving major office disciplinary referrals in the intervention year compared to the number receiv ing such major referrals in the second baseline year Table 4 12. P roportionality of referrals given by student enrollment Year N (Total) # Stud w/ Referrals % Stud w/ Referrals Difference Prior Year # Referrals % Ref / All Stud. Difference Prior Year Minor / Pre Referrals 1 st Baseline 594 ------2 nd Baseline 600 143 23.83% -509 84.83% -Intervention 721 121 16.78% 29.58% 489 67.82% 20.05% Major Office Disciplinary Referrals 1 st Baseline 594 32 5.39% -77 12.96% -2 nd Baseline 600 49 8.17% + 51.59% 88 14.67% + 13.14% Intervention 721 36 4.99% 38.89% 58 8.04% 45.15% Total Referrals 1 st Baseline 594 ------2 nd Baseline 600 150 25.00% -597 99.50% -Intervention 721 125 17.34% 30.65% 547 75.87% 23.75% Note: in proportions from the two years compared divided by the proportion of the earlier comparison year This provides the percent difference in proportion of referrals from year to year. indicates significant difference from prior year. Table 4 12 also provides information on changes with minor/pre referrals and total referrals (noting total referrals is the sum of minor/pre referrals and major office disciplinary referrals) As noted pr eviously, the school district and school did not collect data on minor/pre referrals during the first baseline year, such that the first baseline year is excluded from analyses involving minor/pre referrals and total referrals Regardless, a test of indep endent proportions between the second baseline year and the intervention year demonstrated a significant decrease in the proportion of minor/pre referrals given to enrolled students in the second baseline year to the intervention year with 509 minor refe rrals across 600 students in the second baseline year (84.83%)

PAGE 192

192 decreasing to 489 minor referrals across 721 students in the intervention year (67.82% ; Z= 7.1623 p<0.001 ) a 20.05% decrease in the number of minor/pre referrals given based on the total enro llment of the school The significant decrease in proportionality is maintained from the second baseline year to the intervention year when exploring the proportion of students receiving minor/pre referrals compared to the total student enrollment, with t he intervention year having a significantly lower proportion of students receiving minor/pre referrals (16.8%) than the second baseline year (23.8%; Z=3.1909, p=0.001) This equates to a 29.58% decrease in the proportion of students receiving minor/pre re ferrals from the second baseline year to the intervention year Table 4 12 finally provides a breakdown of the proportion of all referrals based on total student enrollment each year and the proportion of students receiving any referral based on total student enrollment each year A test of independent proportions between the second baseline year and the intervention year demonstrated a significant decrease in the proportion of all referrals given to all enrolled students in the second baseline year to the intervention year with 597 referrals across 600 students in the second baseline year (99.50%) decreasing to 489 referrals across 721 students in the intervention year (75.87%; Z=12.555, p<0.001) representing a 23.75% decrease in the number of all r eferrals given based on the total enrollment of the school The significant decrease in proportionality is maintained when exploring the proportion of students receiving any referral based on the total student enrollment, with the intervention year having a significantly lower proportion of students receiving any type of referral (17.34%) than the second baseline year (25.00%; Z=3.4155, p<0.001) thus representing a 30.65% decrease in the proportion of students receiving referrals from the second baseline year to the intervention year.

PAGE 193

193 Section Five: Research Questions Three and Four As noted previously, while research questions one and two are directly related, research questions three and four are also directly related Research questions three and four are the focus of this section As with the first two research questions, all data were first analyzed for normalcy to ensure assumptions of parametric statistics are not violated In addition, no potential outliers were identified that were determined to impact the results Regardless of race/ethnicity and student grade level, is there a significant relationship between the level of implementation of a SWPBIS initiative (i.e., number of tangible reinforcements provided) and the instances of negative student behavior (i.e., minor referrals and major referrals [ODR]). Is there a difference in the impact a SWPBIS intervention (i.e., numbe r of reinforcements) and instances of negative school behavior (i.e., number of minor and major referrals ) by race/ethnicity and/or grade level? The third research question was originally planned to be explored using multiple regression for each month of the academic year, exploring the connection between the number of Lion Loot provided to each student (predictor variable) and the number of referrals provided to each student (outcome variables) As noted in Section Three of this chapter, the original in tent to explore data by month was precluded when students ) As such, because Lion Loot was recorded only when students exch these data are not considered reliable when analyzed by month such that no analysis was completed to explore monthly data

PAGE 194

194 Ultimately, u sing data from the 721 students enrolled during the intervention year, a multiple linear regression was completed to determine the extent to which the number of tangible reinforcement s (Lion Loot) provided to students predicted the three outcome variables of associated with negative student behaviors (i.e., number of mi nor/pre referrals, major office disciplinary referrals, and total referrals) As shown in Table 4 13, The multiple regression was not significant with a very low coefficient of determination (R 2 =0.000; F(2,720)=0.096, p=.909) Table 4 13. M ultiple regression ANOVA table (research question 3) Source Sum of Squares df Mean Square F Sig. Regression 74.529 2 37.265 .096 .909 Residual 279445.526 718 389.200 Total 279520.056 720 R 2 = 0.000; Adjusted R 2 = 0.003 Based on these findings, a regression equation cannot be developed to predict incidence of negative student behaviors with incidence of tangible reinforcers under the school wide positive behavioral interventions and supports model (SWPBIS). This finding is not surprising given t outcome variables, with the overwhelming majority of students receiving no minor/pre referrals or major office disciplinary referrals during the intervention year. As noted previously, such a large numbe distribution, with regression not robust against such inherent data issues. As originally proposed in the analysis plan, should the regression produce insignificant results and/or produce a regression with l ow R 2 data would be analyzed with an ANOVA and correlation Within this procedure, the a priori threshold was set at 0.6 for a meaningful correlation As presented in Table 4 14, a correlational analysis

PAGE 195

195 produced no significant correlation between the n umber of tangible reinforcements provided to students (i.e., Lion Loot) and the number of negative student behaviors measured by minor/pre referrals, major office disciplinary referrals, or total referrals Table 4 14. Correlation matrix: Lion Loot and negative student behaviors Intervention Year 2 nd Baseline Year Correlation with # of Lion Loot # Minor Referrals # Major Referrals # Total Referrals # Minor Referrals # Major Referrals # Total Referrals Pearson Correlation 0.001 0.014 0.003 0.005 0.03 0.001 Sig (2 tailed) 0.973 0.706 0.936 0.888 0.424 0.968 The final research question expanded upon the fourth question, with a focus on exploring whether there was a difference in the impact of the SWPBIS intervention based on student race/ethnicity or student grade level The fourth research question also intended to explore whether instances of negative school behavior were di fferent by student race/ethnicity and student grade level a consideration that was already explored under the analyses for the second research question presented previously in this chapter While the final research question was primarily intended to expl ore whether findings from the third research question were different by student race/ ethnicity and grade level, the lack of significant findings from the third research question negate the need for additional analysis, as there appears to be no statistica lly significant relationship between the provision of tangible reinforcements and the incidences of negative student behaviors measured by minor and major referrals However, the fourth research question also intended to explore whether there were differ ences in the application of the tangible reinforcements (i.e., Lion Loot) between student racial/ethnic groups and/or student grade levels To test for these differences, a two way analysis of variance (ANOVA) was conducted to explore the

PAGE 196

196 number of tangible reinforcements (i.e., Lion Loot) based on student grade leve l (i.e., KG, 1 st 2 nd 3 rd 4 th and 5 th grades) and student race/ethnicity (i.e., African American / Black, Hispanic / Latina(o), Multi Ethnic, and Caucasian American / White) Only the 690 students identified within the aforementioned racial / ethnic groups were included in the analysis, with the small number of Asian American and Native American students excluded due to the small sample size s Results of the two way ANOVA ar e prese nted in Table 4 15 Table 4 15. ANOVA table (Lion L oot reinforcements) Source Type III Sum of Sqs df Mean Square F p Model 28066.131 23 1220.267 3.629 .000 Ethnicity 1280.064 3 426.688 1.269 .284 Grade 15099.895 5 3019.979 8.981 .000 Ethnicity Grade 4178.042 15 278.536 .828 .646 Error 223952.269 666 336.265 Total 522526.000 690 ** Significant finding at the established Alpha level of 0.05 As shown in Table 4 15, specific to the fourth research question, there was not a significant main effect of student race/ethnicity on provision of tangible reinforcements based on the omnibus F test performed by the ANOVA (F(3,666)=1.269, p=0.284), nor wa s there a significant interaction between student race/ethnicity and grade level on the number of tangible reinforcement provided to students (F(15,666)=0.828, p=0.646). For instance, 32.3% of all Lion Loot was awarded to African American / Black students who accounted for 30.4% of students enrolled during the intervention year. Similarly, 44.5% of the Lion Loot was awarded to Caucasian American / White students, who accounted for 42.5% of all students enrolled during the intervention year.

PAGE 197

197 Although st udents were awarded tangible reinforcements (i.e., Lion Loot) at a similar rate across ethnicities, a significant main effect was found for student grade level (F(5,666)=8.981, p<.001), with post hoc analyses using the Tukey HSD test to control for multipl e pairwise comparisons demonstrating a complex pattern of differences between student grade levels Ba sed on post hoc analyses, third grade students were significantly more likely to receive the tangible reinforcement (M 3 =33.162) than were any of the othe r grade levels, including kindergarten (M K =20.745; M diff = 12.714, p<0.001), first grade (M 1 =17.573, M diff =15.590, p<0.001), second grade (M 2 =24.299, M diff =8.863, p=0.012), fourth grade (M 4 =12.484, M diff =20.678, p<0.001), and fifth grade (M 5 =14.036, M diff =19 .126, p<0.001) In addition, second grade students were more likely to receive tangible reinforcements (M 2 =24.299) than f irst grade students ( M 1 =17.573, M diff =6.726, p=0.031), fourth grade students (M 4 =12.484, M diff =11.815, p<0.001) and fifth grade stude nts (M 5 =14.036, M diff =10.263, p=0.002) Finally, kindergarten students were more likely to receive tangle reinforcements (M K =20.745 than both fourth grade students (M 4 =12.484, M diff =8.261, p=0.005) and fifth grade students (M 5 =14.036, M diff =6.709, p=0.030 )

PAGE 198

198 CHAPTER 5 DISCUSSION Previous chapters have established that e ncouraging and establishing an educational atmosphere conducive to learning has long been a struggle within school systems and schools Theories and philosophies on learning and behavior date back to at least the time of Plato and Aristotle sought to codify and formalize methods to shape and control student behavior to improve learning outcomes For instance, William James, the Fa ther of American Psychology (Pajares, 2003), discussed the important relationship between education and the development of prosocial student behavior as early as 1892 (James, 1925) With significant sociopolical changes in America during the 1900s and a d ramatic upswing in undesirable student behaviors the late 1900s saw educational practices focus on controlling student behavior through reactive and aversive management strategies (e.g., Singer & Wang, 2009; Skiba et al., 2006 ; Wald & Losen, 2003 ) The o verall ineffectiveness of these strategies, racial disparities in the application of the aversive interventions, and public outcry over permanent damage caused by some aversive interventions drove the need for a more positive method to teach and encourage Department of Education, 2014) In essence, Positive Behavioral Interventions and Supports (PBIS) was developed as a broad approach designed to provide school personnel with positive tools and skills necessary to enhance student academic achievement and improve Center on PBIS, 2016) Some of the most prominent researchers in PBIS indicate that

PAGE 199

199 based behavioral strategies; (2) use of multiple and integrated intervention elements aligned to the needs of the environment; (3) a commitment to sustained and long term outcomes; and (4) sup port within the organizational systems that ensure sustained impacts of the interventions (Dunlap, et al. 2009) School Wide PBIS (SWPBIS) applies the PBIS technology to focus less on individual students, and more on behaviors across the entire school of students to enhance the learning environment The development and application of SWPBIS has been largely supported with a number of research studies These studies have shown SWPBIS can reduce the incidents of negative and disruptive student behavior as measured by such metrics as office disciplinary referrals (ODRs; e.g., George, et al., 2003 ; Nocera, et al., 2014 ; Putnam, et al. 2005; Scott & Barrett 2004; Spencer, 2015), school suspensions (e.g., Nocera, et al., 2014 ; Putnam, et al. 2005; Scott & Barrett, 2004), and incidents of student bullying ( Waadsorp, et al., 2012 ) However, a s noted in prior chapters, the present research addresses several g aps within past research: (1) using internally collected data on all behavior instances at the school ; (2) using (ODRs) ; (3) comparing t he level of tangible reinforcements provided and the behavioral impacts men tioned above ; and (4) analyzing tangible reinforcements, major referrals (ODRs), and minor referrals by race and ethnicity, with all data connected to the individual demographics of each student Overall, the method of addressing these primary gaps and th e results of the present research should help inform practice and guide future research in the early implementation of SWPBIS.

PAGE 200

200 Synopsis of Findings In essence, it was hypothesiz ed that the implementation of a comprehensive School Wide Positive Behavioral Interventions and Supports (SWPBIS) model with tangible reinforcements would be effective in reducing the incidence of disruptive student behaviors as measured by minor/pre referrals and major office disciplinary referrals Ultimately, all four research questions detailed in the prior chapters were explored and resulted in some important findings relevant to the field of education and the main stream focus on positive behavior interventions and supports It is noted that some research questions were met with a lack of significant findings, which d oes not suggest a lack of meaningful findings Any lack of significance found with the current analyses can also be interpreted when meaningful, and may suggest a potential flaw in the prior research (where the current study does not match past research) a flaw in the present research, or the presence of extraneous variables that were not or could not be measured d ur ing the course of the present study The following provides the most salient and interpretable f indings from this research. The primary goal of this research was to determine whether there was a change in the number of referrals from the baseline years to the intervention year, with the primary change in the school culture being the new implementatio n of the School Wide Positive Behavior interventions and Supports (SWPBIS) model Overall, data from two baseline years and the intervention year demonstrated that students receiving at least one referral of any type tended to receive more than one referr al, with the average student receiving 2.41 total referrals in the first baseline year, 3.94 total referrals in the second baseline year, and 4.38 total referrals in the intervention year The higher average number of referrals in the intervention year mu st be considered in conjunction

PAGE 201

201 with the lower number of total referrals given during the intervention year (N=547) than the number of referrals given during the second baseline year (N=583) This suggests that fewer students required behavioral intervent ion during the intervention year, but required the intervention more often per student It is also noted that the change in average total referrals appears driven by changes in the average number of minor or pre referrals, which is expected with the SWPBI S model that encourages the pre referrals before major office disciplinary referrals are written Over the course of the study, the school realized a 20.17% increase in enrollment from the second baseline year to the intervention year, coupled with an 23 .75% decrease in total referrals, a 20.05% decrease in minor/pre referrals, and a 45.15% decrease in proportion of major office disciplinary referrals to total student enrollment The school administered 77 major office disciplinary referrals to 32 studen ts in the first baseline year, 88 to 49 students in the second baseline year, and 58 to 36 students in the intervention year While there was no difference in the proportion of major referrals to enrolled students from the first to the second baseline yea rs, there was a significant decrease in the proportion of referrals from the second baseline year (the year immediately prior to implementing SWPBIS) to the intervention year Moreover, the intervention year realized a significantly lower proportion of students receiving major office disciplinary referrals than the second baseline year (a 38.89% decrease in the proportion of students receiving major office disciplinary referrals) In addition to major office disciplinary referrals, the school realized a significant decrease in the proportion of minor/pre referrals and total referrals given to enrolled students in the second baseline year to the intervention year The intervention year also had a

PAGE 202

202 significantly lower proportion of students receiving minor /pre referrals than the second baseline year (a 29.58% decrease in the proportion of students receiving minor/pre referrals, and a 30.65% decrease in the proportion of students receiving any type of referral from baseline to intervention). The second goal of this research was to explore whether the impact of the SWPBIS initiative had differential impact on students based on their race/ethnicity and/or grade level in school In terms of student grade level, results show some differences in student grade lev el, but the differences did not reveal a pattern Most notably, when compared with the proportion of total students at each grade level, kindergarten and fourth grade students received a significantly higher proportion of referrals during the intervention year, while first grade students had a significantly lower proportion of referrals Otherwise, when combined and explored across all three study years, students in each grade level were no more likely to receive any type of referral (i.e., minor/pre refe rral, major office disciplinary referrals, or total referrals) than students in any other grade levels In terms of student race and ethnicity, more important patterns of results emerge For instance, during the intervention year, African American studen ts represented a higher proportion of students receiving referrals (45.7% in the intervention year) than the proportion of total students in the school (32.2% in the intervention year) thus further demonstrating the need for this study. The proportion of referrals given to students identified as Hispanic and White / Caucasian American were both significantly lower than the proportion of enrolled students during the intervention year from these groups, while the proportion of referrals given to students id entified as African American / Black was significantly higher than the

PAGE 203

203 proportion of enrolled students identified as such Such differences are important considerations for the application of discipline at the intervention school, and are relatively consi stent with findings of past research (e.g., Skiba, et al. 2000) In addition, African American / Black students had a higher average rate of major office disciplinary referrals across all three academic years of study combined than did Caucasian American / White students African American / Black students also had significantly higher rates of minor/pre referrals and major office disciplinary referrals than both Hispanic / Latina(o) and Caucasian American / White students across the second baseline year and intervention year (combined) Importantly, while African American / Black students had a significantly higher average number of major office disciplinary referrals than Caucasian American / White students in both baseline years, such differences did n ot persist into the intervention year In fact, there were no significant differences in the average number of major office disciplinary referrals between any of the four racial / ethnic groups explored (i.e., African American / Black; Hispanic / Latina(o ); Multi Ethnic, and Caucasian American / White) during the intervention year In addition to student grade level and race/ethnic, students with at least one office disciplinary referral during the intervention year were composed of 87.8% males and 12.2% females a substantially higher proportion of males than contained in the overall student enrollment during the same year In addition, 72.4% of students with at least one referral during the intervention year qualified for free or reduced price lunch, compared to 51.7% of the overall student population Finally, 20.5% of students with referrals during the intervention year were identified as ESE, compared to 16.2% of all students enrolled, and 14.2% received accommodations under Section 504, compared

PAGE 204

204 t o 5.8% of the school population during the same intervention year These are important variables for consideration when designing future research studies, but were outside the scope of the present research. The third goal of this research was to explore t he extent to which the use of tangible reinforcements impacted the change in negative student behavior (e.g., minor/pre referrals, major office disciplinary referrals, and total referrals) As noted in Chapter Four, it is important to use caution when int erpreting data within this research according to distribution by month, as data were only able to be collected when the students exchanged the Lion Loot for backup reinforcements (not when they were awarded) As noted in Chapter Three, Lion Loot were prov ided as secondary reinforcements to provide for immediate and consistent reinforcement of desired student behaviors However, these secondary reinforcements only hold value insofar as they can be exchanged for back up reinforcements (i.e., high interest i tems provided within the SWPBIS School Store) Because some students saved their Lion Loot until they were able to afford a higher interest or more rewarding item, there were many instances of Lion Loot being exchanged in the months after they were origin ally awarded It is important to note the apparent pattern of exchange, with a large number of Lion Loot exchanged in the first month (9.7% of all Lion Loot) and nearly half of all Lion Loot exchanged in the last three months of the academic year Moreov er, it is noted that 86.1% of all students enrolled during the intervention year received at least one tangible reinforcement (i.e., Lion Loot) It is possible the remaining students did not receive any Lion Loot, or received the tangible (secondary) rein forcement without exchanging it for the backup (primary) reinforcements at the SWPBIS School Store Because the

PAGE 205

205 tangible reinforcement was only recorded when exchanged, the non exchanged Lion Loot were necessarily excluded from consideration within this s tudy Ultimately, with the challenges noted above, the number of tangible reinforcements (Lion Loot) did not predict and was not significantly related to the number of negative student behaviors measured by minor/pre referrals, major office disciplinary r eferrals, or total referrals The research also looked to explore whether there were differences in the application of the tangible reinforcements (i.e., Lion Loot) between student racial/ethnic groups and/or student grade levels Results demonstrated n o significant differences between student race/ethnicity groups on the provision of tangible reinforcements, with the proportion of Lion Loot given to student racial/ethnic groups matching the overall proportion of enrollment of s tudents within the same gr oups However, w hile students receiving tangible reinforcements were equally distributed according to the grade level distribution of the total population of students at the school during the intervention year, the number of tangible reinforcements were n ot equally distributed according to the distribution of the population of students provided with tangible reinforcements and enrolled at the school during the intervention year More specifically, third grade students were significantly more likely to rec eive Lion Loot than any other grade level of students, and fourth and fifth grade students less likely to receive Lion Loot than students in kindergarten or second grade. Implications With the rising focus on instructional time for students, the rising stakes of academic achievement tests, and the high need to support teacher in better managing classroom behavior, this research presents clear implications for education Certainly, much research has explored the impact of School Wide Positive Behavior In terventions

PAGE 206

206 and Supports (SWPBIS) and this research both confirms past research and provides new insights and implications for educators and schools implementing such a complex and comprehensive intervention model Based on results from this research and coupled with results from past research, even within the first year of implementation, it appears SWPBIS is a viable method for reducing incidence of negative student behaviors and increasing learning time for those students most prone to receive referral s The following provides the most salient implications from the findings of the present study. Referrals and Negative Student B ehaviors One of the most important implications of this research study is the overall chan ge in the number of minor/pre referra ls and major office disciplinary referrals given to students for demonstration of negative student behaviors Certainly, there was some expectation for the number of referrals to be reduced in the first year, with the primary expectation ( based on past re search ) for the major office disciplinary referrals to have the greatest reduction Results supported findings of several past studies, including Scott and Barrett (2004), who found that office disciplinary referrals (ODRs) decreased from 608 at baseline (the year prior to implementation) to 108 in the first year and 46 in the second year of SWPBIS implementation George, et al. (2003) found ODRs decreased from 1,717 during the baseline year to 702 during the first year of SWPBIS implementation and 619 du ring the second year Spencer (2015) found the number of ODRs decreased from 593 in the baseline year to 268 after the first year of SWPBIS Finally, Luiselli, et al. (2005) found that SWPBIS resulted in fewer ODRs, with an average of 1.3 ODRs per day at baseline decreasing to 0.7 per day in the first year of implementation and 0.5 per day by the second year of implementation Overall, the

PAGE 207

20 7 present research supports the findings of these past studies (and many others) while also demonstrating that the im pact of SWPBIS goes beyond major office disciplinary referrals (ODRs) As discussed in Chapter Two, ODRs are the most common behavioral indicator used in research on SWPBIS (e.g., Pas, et al. 2011), but few (if any) studies have explored internal school data on minor/pre referrals (referrals that do not meet the seriousness of ODRs and are rarely recorded outside of internal student records) More specifically, following the implementation of SWPBIS under the same principal and school administrators as t he baseline years, the school demonstrated significantly fewer minor/pre referrals and major ODRs thus supporting and expanding past research As noted previously the school realized a 20.17% increase in enrollment from the second baseline to the inter vention year, coupled with an 23.75% decrease in total referrals, a 20.05% decrease in minor/pre referrals, and a 45.15% decrease in to proportion of major office disciplinary referrals based on total student enrollment. The importance of decreasing minor/ pre referrals is similar to that of reducing major ODRs, as both create a disruption in the school day for the student and the entire class Reducing these disruptions not only reduces time away from the learning lessons, but also decreases disruptions fo r all students and can positively impact the Indeed, nearly every study exploring classroom behavior management and school culture begins with the same basic tenet poor student behavior distracts from and n egatively impacts the learning environment (e.g., Guardino & Fullerton, 2010; McKevitt, et al. 2012; Scott & Barrett, 2004; Sugai & Horner, 2006) The deleterious effects can be insurmountable for some teachers,

PAGE 208

208 including difficulty managing and controll ing the classroom, difficulty teaching and presenting content to the students, loss of focus of students or the entire classroom, and a reduction in the amount of time on task When the teacher stops teaching to deal ease the time on task for both the individual student and other students who attend to the distraction Even more time can be lost for students who are sent to the office on a referral or suspended from school In fact, Scott and Barrett (2004) performed a thorough review of data from an elementary school in Baltimore and found that office disciplinary referrals resulted in removal of a student from the classroom for an average of at least 20 minutes (nearly 6% of the learning time provided during the sch ool day), while a suspension resulted in a loss of at least 6 hours of instructional time for each day of suspension Not surprisingly then, according to the National Association of Elementary School Principals, 80% of ime is spent on dealing with disruptive and dangerous barrier to effective education in their classrooms (Public Agenda, 2004) This study demonstrates that implementing a comprehensive SWPBIS model can help decrease both major behavior issues and minor behavioral distractions (which still have a negative impact on the learning environment ) Ethnic D isproportionality of R eferrals Three primary issues led to the developmen t of School Wide Positive Behavior Interventions and Supports (SWPBIS): (1) need to manage behavior of all students in an inclusive environment to maximize the learning environment; (2) moral objections to aversive stimuli and punishment driven methods of control; and (3) need for more culturally sensitive and understanding methods to support positive behavior and

PAGE 209

209 character As noted in Chapter Two, since the landmark United States Supreme Court ruling in Brown vs Board of Education of Topeka (347 U.S 483) in 1954, which ended legal segregation in public schools, minority students have endured an educational system replete with ongoing bias, as well as overt and covert racism (NAACP LDF, 2011) as (Skiba, 2000), and several prominent national agencies have discredited these policies (e.g., Skiba et al., 2006) However, the impact of these policies and sociocultural aspects of the education field have resulted in wide spread institutionalized rac ism (Jones, 2000), where African American and Black students are facing barriers to obtaining the highest quality of educational opportunities, in part due to disproportionate punishments that take them out of class and away from lessons Past research ha s found that Black students were more than three times more likely to be suspended as White students (e.g., Wald & Losen, 2003; U.S Department of Education, 2014) It is data like these that supports hool to 2009; Skiba, 2000) The U.S Department of Education (USED, 2014) called on 2), and help students succeed throu gh programs that support and reinforce positive student behavior and character development Findings from the present research related to racial and ethnic disproportionality can be separated into two primary groups: punishments and rewards. For the purpo minor/pre referrals and the number of major office disciplinary referrals administered to enrolled students during the course of the academic years of focus It is noted that

PAGE 210

210 minor/pre referrals could be relatively minor, but still required the attention of the school administrators and often removed the student from the classroom Major office disciplinary referrals were more serious and necessitated a more serious consequenc e and longer time away from the classroom and lessons (either through an extended visit to the office, in school suspension, out of school suspension, or sending the child home) Such time lost from the learning day can be a significant negative impact on student learning, with Scott and Barrett (2004) estimating at least 20 minutes for an office referral and at least 6 hours for suspension s As such, when students from one racial/ethnic group receive more referrals and punishments than other groups, they also receive less quality education and time on task Thus, SWPBIS was partially developed to help reduce overall referrals and, perhaps more importantly, reduce the disproportionality between students from traditional minority groups and those from trad itional majority groups This study demonstrated that, as expected from past research, the African American / Black students were significantly more likely to receive referrals than their Caucasian American / White fellow students, with the average numbe r of referrals for African American / Black students higher than their counterparts and the proportion of African American / Black students receiving referrals significantly higher than their proportion of the overall school It is important to note that, in terms of major office disciplinary referrals, the racial/ethnic differences in average number of referrals appear to dissipate in the intervention year where African American / Black students were similar in the average number of referrals received The number of students with any type of referral (minor or major) decreased from 150 to 127 from the second baseline

PAGE 211

211 year to the intervention year, and the number of African American / Black students with at least one referral decreased from 79 to 58 Wh ile not significant, this still demonstrates some meaningful progress and research suggests that sustained SWPBIS can continue to decrease the number of students with referrals That being said, the fact that the proportion of students receiving referrals still demonstrated ethnic disparities, there continues to be a need for additional training and support to reduce the use of punitive classroom removals (e.g., referrals) and increase the instances of positive behavior supports. Impact of Tangible R einfor cements (Lion Loot) The third and fourth research questions were focused on the provision of Lion Loot and whether it was related to the number of referrals given during the intervention year While the analysis by month was precluded by students collecti ng and holding Lion Loot to save up for better items from the school store (see limitations and recommendations for future research), the overall analysis of Lion Loot suggested that it was not related to the number of referrals when explored across all st udents This was somewhat unexpected from when the research was first designed, but it became an expected result as the intervention model was applied and teachers were trained Most importantly, the entire purpose of the tangible reinforcement is that a ll students should be rewarded for positive behaviors and the reward should be given with a level of fanfare (thus increasing vicarious reinforcement) Moreover, teachers were trained to provide fewer tangible reinforcements (more verbal reinforcements) t o those students behaving well so as to increase the effectiveness of the reinforcement process (see Chapter 2 for a discussion of intermittent reinforcement) and prevent satiation (see Chapter 2 for a discussion of satiation with reinforcements) As such students who

PAGE 212

212 were alway s behaving well would often receive the same number of Lion Loot as students who were often struggling with behavior (as teachers and staff members would more readily see the positive behaviors of more troubled students and would n ot risk satiation with applying the tangible reinforcement ) Ultimately, this means that students on both sides of the spectrum could receive potentially equal number of Lion Loot, such that the lack of significance in relation to the number of referrals is less surprising based on how SWPBIS is implemented While this lack of significance is less exciting, it remains a meaningful finding and helps demonstrate the overall implementation of this component of the SWPBIS model Future research should consid er other methods to measure the implementation of Tier I interventions if desiring to explore how differentiation in implementation impacts outcomes. Based on a complex web of prior theories and research (discussed in Chapter 2), SWPBIS is a broad approa ch designed to provide school personnel with tools and skills necessary to enhance student academic achievement and improve socially relevant Some of the most prominent researchers in PBIS indicate that SWPBIS has four core features (noted at the beginning of this chapter) : (1) an application of research based behavioral strategies; (2) use of multiple and integrated intervention elements aligned to the needs of the environment; (3) a commitment to susta ined and long term outcomes; and (4) support within the organizational systems that ensure sustained impacts of the interventions (Dunlap, et al. 2009) In essence, SWPBIS makes use primarily of positive reinforcement of appropriate behaviors, focuses on teaching productive social behaviors (not just suppressing negative behaviors), and is designed with a clear focus on being

PAGE 213

213 culturally sensitive (Sugai & Horner, 2009a) SWPBIS makes use of a three tiered system of supports: (1) Primary, Tier 1 whole school interventions impacting 80 90% of the student population; (2) Secondary, Tier 2 small group interventions impacting 15% of the student population who do not respond to primary tier interventions; and (3) Tertiary, Tier 3 individual interventions impacting the remaining 5% of the student population who do not respond to primary and secondary tier interventions (McIntosh & Goodman, 2016) As such, it might be that the implementation of tangible reinforcements (Lion Loot) and the School Store were only a small component of the overall SWPBIS model, such that the measurement of Lion Loot as the primary indicator of implementation overestimated the importance of the tangible reinforcements and/or underestimated the importance of the many other compone nts (e.g., data chats, training, etc.). Future research might consider how to better measure implementation using, as noted above, a measure that is not expected to be administered to all students equally. Classroom L e arning and Vicarious L earning While n ot a focus of this research study, i t is also important to note that there was some challenge with explo ring data at the student level the purpose of this research Indeed, it could be argued that students are grouped within a classroom and that every s tudent in the classroom would see the rewards (Lion Loot) and punishments (referrals) provided to all other students in the classr oom (see Chapter Two for research about vicarious learning and modeling) A s such, every time a student receives a tangible r einforcement, there is vicarious learning with all students in the classroom a situation that cannot be accurately measured for research across an entire school It might be important for future research studies to look at the classroom level, as this w ould

PAGE 214

214 potentially help explain the impact of the intervention without having some of the same limitations as looking at the individual student level It is also important to consider whether referrals themselves (given that they were given out significantl y less often in the intervention year) also became stronger deterrents to negative behaviors In other words, referrals were given less often and might appear to be more serious punishments to the students such that they were more focused on not receivi ng referrals and/or the vicarious punishment was more effective at preventing other students from engaging the same behaviors In this sense, it might have been less about the receipt of Lion Loot and more about the severity of referrals that led to the r eduction in overall referrals in the intervention year (or a combination of these interventions) As this is far beyond the scope of the present research, it is left to future researchers to consider ways to measure such subjective thoughts with young chi ldren Research Me thodology While the development and application of SWPBIS has been largely shown to significantly reduce the incidents of major office disciplinary referrals suspensions, bullying, and other disruptive student behaviors, there are some gaps within past research that were addr essed within the current study First, nearly every research study exploring office disciplinary referrals focused on the use of centralized databases to collect archival data about student referrals, which can introduce error in terms of data entry and unwritten policies ( Nelson, et al., 2002 ), while the present research used internally collected data on all behavior instances at the school studied Second, few referrals) that are not be recorded in such centralized databases of student behavior, while the present research used referral s (ODRs) and Finally, few studies have included differentiated

PAGE 215

215 analysis by race and ethnicity due to complications with the recording of such data in centralized databases, while the present research collected data connected to the individual demogr aphics of each student Overall, the method of addressing these primary gaps and the results of the present research should help inform practice and guide future research in the early implementation phases of SWPBIS models Limitations and Future Research Although there were important findings and implications of the present study, there also existed some limitations which could be addressed by future research One such limitation was the method utilized to administer and collect data on the tangible rein forcements (i.e., Lion Loot) The method employed was effective in showing the total number of tangible reinforcements provided to each student by each teacher (as the student name and teacher name were both recorded on the actual ticket) However, three of the original assumptions regarding the exchange of the Lion Loot were not realized: (1) students would not lose their tickets; (2) students would be able to exchange their tickets daily; and (3) students would regularly exchange the tickets at the SWPB IS School Store Of these, the greatest limitation was that students did not regularly exchange their tickets, either because they did not have enough opportunities or because they saved their tickets to purchase a more desirable (and more expensive) item from the SWPBIS School Store Because the methodology was based on counting the tickets during the month they were exchanged, tickets could have been counted in months other than when they were awarded This specifically limited the ability within this research to compare the number of tangible reinforcements directly to the number of referrals by week or month Several directions could be made for future research to eliminate this limitation: (1) have teachers record the tangible reinforcements by stud ent

PAGE 216

216 each day; (2) have teachers write the date awarded on the tangible reinforcement (along with the student and teacher name); and/or (3) color code the tangible reinforcements for each of the 10 academic months (thus providing visual indicators as to the months the tickets were provided) Each of these methods creates increased effort for the teacher and school staff, but would deepen the data and allow for monthly (or weekly) comparison of the intervention with school behavioral data A second limitat ion of this research revolved around the provision of training and development to the teachers and school staff While the researcher provided several trainings and feedback reports to teachers, a survey of staff conducted by the school (the full results are outside the scope of this research) revealed that staff requested additional training and some staff did not entirely understand the overall SWPBIS model and interventions The pattern of tangible reinforcements (see Appendix A) also revealed some dis crepancies across teachers, with some providing only a handful of tickets and some providing nearly 1,000 tickets (one provided more than 1,000) As such, m ore training and theoretical understanding might have helped the teachers in implementing the model such as making a when giving the tangible reinforcements to increase vicarious reinforcement The trainings provided were focused on application, with some theoret ical understanding being assumed and might not have been fully present For instan ce, teachers were assumed to know that students might not be taught prosocial behaviors for classroom while at home, such that the teacher (and school) are responsible for teaching the positive behaviors necessary for a productive learning environment Fu ture research may consider increasing the level of training, both in terms of content and frequency, while implementing a method for assessing the training and including as a variable within the research study

PAGE 217

217 A third limitation and potential direction for future research stems from the need for regular feedback to teachers and staff members, particularly as SWPBIS requires teachers to be proactive in teaching and reinforcing positive student behaviors While reactive disciplinary practices rarely requi re proactive planning, SWPBIS is more data driven in selecting which behaviors to reinforce and how best to reinforce behaviors for each student For instance, if a student is having particular difficulty walking correctly in a line, then the teacher woul d want to plan when and how to apply reinforcements (whether tangible or intangible) to elicit the desired change in this student However, d eveloping individualized behavior modification plans is time consuming and requires substantial data to develop, i nitiate, and monitor the plan Of course, i f the student has not yet behaved in an unproductive manner, then it is not possible to predict that they might mis behave Similarly, developing a behavior modification plan that targets only individual students ignores the potential benefits that SWPBIS can have on those students who are already behaving well, but are not being rewarded (e.g., vicarious reinforcement, modeling, etc.) As such, future research may incorporate feedback methods, such as weekly or monthly, to help teachers become more aware of how they are interacting with their students comparing student groups, grade levels, or individual teachers A data dashboard showing the number of tangible reinforcements compared to behaviora l and academic data could also help provide needed feedback Depending on the depth of the data, the teacher can quickly explore potential paths to take when a student demonstrates unproductive behavior not previously demonstrated As noted in Chapter T wo, not all research has shown ODRs to be a valid and reliable measure of student negative behavior While hundreds of research studies use

PAGE 218

218 these data as the primary measure of student behavior, not all research has shown them to be the most valid measure For instance, Nelson, et al. (2002) utilized the Teacher Report Form (TRF) and questioned the convergent validity of using ODRs as a true measure of student behavior Other studies have combined the use of student behavior data measured by referrals to include additional survey data and rubric scores (e.g., Pas, et al. 2011) As such, future research may consider using a validated and reliable behavioral survey in combination with school level behavioral data in assessing the impact of SWPBIS implemen tation It is noted that research has demonstrated some challenges with teacher assessments, such as only a moderate correlation between teacher reports of whether students were sent to the office and school records showing such referrals ( Pas, et al., 20 11 ), such that it would be important for future research to retain the use of school records (ideally both minor/pre referrals and major office disciplinary referrals) This being said, more research than not has found validity for using ODRs as a measure of general behavior (e.g., Irvin, et al., 2004 ; Scott & Barrett, 2004), with ODRs being related to general misbehavior at school, school attendance, student and teacher perceptions of safety and victimization, classroom orderliness, juvenile delinquency, and behavior disorders. Several past research studies have also utilized fidelity assessments as part of their studies into SW PBIS (e.g., Bradshaw, et al., 2008; Bradshaw, et al. 2010), with the most common assessment being the School Wide Evaluation Tool (SET) developed by some of the foundational theorists of PBIS (Sugai, et al. 2001) The present research did not utilize formalized fidelity checks due to limitations with both human resources and hesitation of school administrators to implement any add itional teacher or classroom evaluation process within the school day Indeed, the SET requires trained

PAGE 219

219 assessors to visit the school and classrooms, providing observational scores on seven separate key features Such a process can be rather time consumi ng and intrusive on the school day, and the present resource did not have the capacity to engage in such an effort However, future research is encouraged to consider using a method to ensure fidelity of the planned interventions and behavioral supports Certainly, as discussed throughout this dissertation, the importance of fidelity, continuity, and consistency are critical to maximize the impact of PBIS models Future studies stemming from the present research can certainly include a plethora of additional variables and enhancements For instance, the inclusion of pre kindergarten students could provide interesting information as to the application of the model to a younger population, and could be explored in longitudinal research to determine w hether such early intervention improves kindergarten readiness and/or behavior in early grades Future research could also include personality measures of teachers and staff administering the tangible reinforcements to determine whether there are personal ity traits that impact the implementation and/or effectiveness of the chosen interventions It would likely be most beneficial to focus on traits that can either be modified or easily measured to enhance the overall SWPBIS model Overall, this study has supported the importance of empowering teachers and staff members to improve the quality of the educational learning environment through a system of proactive, comprehensive, School Wide Positive Behavioral Interventions and Supports

PAGE 220

220 A PPENDIX ADDITIONAL RESULTS TABLES

PAGE 221

221 Table A 1. Administration of independent v ariable (Lion Loot) by staff m ember Teacher Type N Sept Oct. Nov. Dec. Jan. Feb. Mar. Apr. May June Tot. T001 Teacher 46 21 6 7 11 0 3 3 8 46 56 161 T002 Teacher 21 22 19 7 11 1 17 8 14 20 0 119 T003 Teacher 24 101 52 15 0 66 49 0 228 0 94 605 T004 Teacher 57 16 7 20 7 4 20 36 22 36 53 221 T005 Para 169 8 39 46 41 28 30 59 76 52 20 399 T006 Admin 19 0 2 3 0 10 3 0 0 2 0 20 T007 Teacher 27 61 9 32 11 6 42 15 4 11 26 217 T008 Admin 95 9 3 4 6 4 6 14 10 15 7 78 T009 Para 29 0 0 4 4 1 24 6 13 21 20 93 T010 Teacher 17 0 0 0 0 0 0 0 0 0 0 0 T011 Teacher 11 0 0 0 0 0 0 0 0 6 0 6 T012 Teacher 46 18 7 13 11 8 7 10 12 32 0 118 T013 Teacher 37 106 82 44 71 63 62 47 71 53 35 634 T014 Teacher 101 47 10 4 9 12 11 3 8 15 5 124 T015 Teacher 29 0 0 0 0 0 10 28 79 58 142 317 T016 Teacher 106 15 24 16 12 13 16 10 5 48 27 186 T017 Teacher 26 12 17 23 21 27 45 21 38 66 53 323 T018 Teacher 24 31 24 6 51 28 34 31 54 59 1 319 T019 Teacher 22 58 49 17 41 2 38 21 52 66 30 374 T020 Teacher 87 99 62 56 30 73 147 100 95 299 89 1050 T021 Teacher 29 31 55 48 137 135 36 98 118 60 27 745 T022 Teacher 31 58 17 30 49 17 79 21 172 264 7 714 T023 Teacher 20 13 25 29 3 9 15 12 5 50 11 172 T024 Teacher 135 16 2 20 8 12 16 10 29 51 33 197 T025 Teacher 8 0 44 1 0 16 0 27 82 0 18 188 T026 Teacher 155 36 14 9 3 2 25 23 16 50 34 212 T027 Teacher 27 16 32 10 11 29 47 29 20 113 65 372 T028 Teacher 54 31 9 26 5 6 9 8 16 117 24 251 T029 Para 273 6 11 25 31 40 54 55 88 279 75 664 T030 Teacher 28 54 41 25 25 22 48 27 35 36 36 349 T031 Teacher 22 18 14 20 17 10 8 1 23 13 0 124 T032 Teacher 20 28 5 6 17 7 1 24 13 44 0 145 T033 Teacher 20 14 8 15 15 7 47 13 11 52 10 192 T034 Teacher 53 31 10 24 9 4 2 2 9 36 20 147

PAGE 222

222 Table A 1. Continued Teacher Type N Sept Oct. Nov. Dec. Jan. Feb. Mar. Apr. May June Tot. T034 Teacher 53 31 10 24 9 4 2 2 9 36 20 147 T035 Teacher 40 9 2 6 4 2 5 3 3 9 8 51 T036 Teacher 1 0 0 0 1 0 0 9 5 0 2 17 T037 Teacher 22 43 13 35 13 20 26 1 0 20 7 178 T038 Teacher 83 0 0 0 0 0 5 10 26 45 14 100 T039 Teacher 43 36 14 8 10 47 75 132 106 173 103 704 T040 Para 61 3 2 3 15 13 23 13 10 10 1 93 T041 Teacher 13 0 0 0 24 0 1 0 1 28 0 54 T042 Teacher 46 23 2 8 5 2 0 4 8 14 0 66 T043 Teacher 46 18 6 34 14 18 13 6 20 88 29 246 T044 Teacher 7 0 19 0 0 12 0 10 57 0 18 116 T045 Admin 58 14 37 17 14 30 12 45 71 60 105 405 T046 Teacher 16 0 0 0 0 0 0 0 0 0 0 0 T047 Teacher 25 10 25 10 3 12 28 9 6 9 18 130 T048 Teacher 45 2 2 2 6 6 8 7 10 3 8 54 T049 Teacher 24 9 6 3 13 0 7 38 34 51 10 171 T050 Teacher 24 99 14 26 11 30 16 26 7 8 35 272 T051 Teacher 73 13 11 13 9 3 23 7 29 99 71 278 T052 Teacher 33 1 2 2 1 1 5 1 11 1 8 33 T053 Teacher 63 1 0 3 5 7 33 33 28 102 86 298 T054 Teacher 20 7 2 4 2 5 9 3 4 21 0 57 T055 Teacher 60 42 43 37 34 84 35 110 65 156 97 703 T056 Admin 8 0 0 5 0 0 0 0 0 2 1 8 T057 Teacher 22 24 18 25 4 37 20 11 49 49 75 312 T058 Teacher 27 56 12 0 1 1 1 8 0 144 0 223 T059 Teacher 38 7 11 23 32 10 29 14 14 19 5 164 T060 Teacher 23 64 41 19 54 21 0 49 41 153 4 446 TOTAL -1457 981 888 942 1023 1325 1311 2031 3334 1723 15015 Note: Tangible reinforcements were only provided during the intervention year (2014 2015). 361 instances of Lion Loot were missing student information, but could be attributed to a teacher

PAGE 223

223 Table A 2. P roportionality of reinforcements by ethnicity and grade KG 1 2 3 4 5 Total Black # Students 53 36 41 40 28 35 233 # Referrals 62 13 76 22 95 42 310 # Lion Loot 1155 589 700 1171 275 570 4460 Proportion 18.63 45.31 9.21 53.23 2.89 13.57 14.39 White # Students 75 50 45 53 56 42 321 # Referrals 69 3 15 36 33 6 162 # Lion Loot 1504 752 1127 1540 794 510 6227 Proportion 21.80 250.67 75.13 42.78 24.06 85.00 38.44 Asian # Students 9 3 7 4 2 4 29 # Referrals 13 1 1 0 0 0 15 # Lion Loot 350 25 159 262 66 53 915 Proportion 26.92 25.00 159.00 65.50 33.00 13.25 61.00 Hispani c # Students 20 15 13 15 10 11 84 # Referrals 2 2 3 12 3 2 24 # Lion Loot 476 300 443 496 113 165 1993 Proportion 238.00 150.00 147.67 41.33 37.67 82.50 83.04 Native Americ an # Students 1 . 1 2 # Referrals 0 . 1 1 # Lion Loot 68 . 9 77 Proportion 68.00 . 9.00 77.00 Multiple Ethniciti es # Students 12.00 9.00 9.00 4.00 11.00 7.00 52.00 # Referrals 13 4 3 0 4 11 35 # Lion Loot 208 170 189 165 161 89 982 Proportion 16.00 42.50 63.00 41.25 40.25 8.09 28.06 All Ethniciti es # Students 170 113 115 116 108 99 721 # Referrals 159 23 98 70 136 61 547 # Lion Loot 3761 1836 2618 3634 1418 1387 14654 Proportion 23.65 79.83 26.71 51.91 10.43 22.74 26.79 Note: (1+) statistics only include those students receiving one or more lion loot. Tangible reinforcements were only provided during the intervention year (2014 2015). Excludes 361 instances of Lion Loot which were missing student names and information. Race/Ethnicity categories were defined b y the School District, as reported. Proportionality. This is calculated by taking the average lion loot per student enrolled in the specific category (even if they did not receive loot) divided by the average number of referrals per student enrolled in the race/grade category (regardless whether they received a referral) If there were no referrals, then the proportion becomes simply the average lion loot per student Lower proportionalities indicate a lower proportion of lion loot to referrals, indica ting the less desirable situation The goal within SWPBIS would

PAGE 224

224 be for student groups with higher numbers of referrals to have a corresponding higher number of tangible reinforcements A lower proportion suggests that the students within the specific gro up were more likely to receive referrals than tangible reinforcements, which is generally counter to the foundational principles of SWPBIS

PAGE 225

225 LIST OF REFERENCES Abidsereshki, N., Abkenar, S. J., Ashoori, M., & Mirzamani, M. (2015). The effectiveness of using reinforcements in the classroom on the academic achievement of students with intellectual disabilities. Journal of Intellectual Disabilities, 19 (1), 83 93. A DA Amendments Act of 2008, Pub. L. 110 325, 154 Stat. 110 325, codified as amended at § 3406. Adelman, HS., & Taylor, L. (2002). Classroom climate. In W. Lee, A. Lowe, & E. Robinson (Eds.), Encyclopedia of school psychology (pp. 304 312). Thousand Oaks, C A.: Sage. Americans with Disabilities Act of 1990, 42 U.S.C. § 12101. Aristotle (2000). The Nicomachean Ethics (W. D. Ross & D. C. Stevenson, Trans.). Boston, MA: Internet Classics Archive. (Original work published ca. 350 B.C.) Retrieved April 5, 2016, f rom: http://classics.mit.edu//Aristotle/nicomachaen.html Baer, D. M., Wolf, M. M. & Risley, T. R. (1968). Some current dimensions of applied behavior analysis. Journal of Applied Behavior al Analysis, 1 ( 1), 91 97. Bandura, A. (1961). Psychotherapy as a learning process. Psychological Bulletin 58 (2), 143 159. http://dx.doi.org/10.1037/h0040672 Bandura, A. (1962). Social learning through i mitation. In M. R. Jones (Ed.), Nebraska symposium of motivation (pp. 211 269). Lincoln, NE: University of Nebraska Press. Bandura, A. (1963). The role of imitation in personality development. Journal of Nursery Education 18, 205 215. Bandura, A. (1965a). Influence of models' reinforcement contingencies on the acquisition of imitative responses. Journal of Personality and Social Psychology, 1(6) 589. Bandura, A. (1965b). Vicarious processes: A case of no trial learning. Advances in exp erimental social psychology 2 1 55. http://dx.doi.org/10.1016/S0065 2601(08)60102 1 Bandura, A. (1971a). Social learning theory General Learning Press: New York, NY. Bandura, A. (1971b). The nature of reinforcement New York, NY: Academic Press. Bandura, A. (1977). Social learning theory Englewood Cliffs, NJ: Prentice Hall.

PAGE 226

226 Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory. New York, NY: Prentice Hal l, Inc. Bandura, A., & Huston, A. C. (1961). Identification as a process of incidental learning. Journal of Abnormal and Social Psychology 63 (2), 311 318. Bandura, A. Ross, D., & Ross, S. A. (1961). Transmission of aggression through the imitation of ag gressive models. Journal of Abnormal and Social Psychology 63, 575 582. Bandura, A., Ross, D., Ross, S. A. (1963). Imitation of film mediated aggressive models. The Journal of Abnormal and Social Psychology 66 (1), 3 11. Bill & Melinda Gates Foundation. (2012). the teaching profession New York, NY: Scholastic. Blanchett, W. J. (2014). Disproportionate representation of African American students in special education: Acknowledging the role of white privilege an d racism. Educational Researcher, 35 (6), 24 28. Blankenship, T., Boon, R. T., & Fore, III., C. (2007). Inclusion and placement decisions for students with special needs: A historical analysis of relevant statutory and case law. Electronic Journal for Inclusive Education, 2 (1). Board of Education of the Hendrick Hudson Central School District v. Rowley, 458 U.S. 176 (US Supreme Court, 1982). Bohanon, H., Fenning, P., Carney, K. L., Minnis Kim, M. J., Anderson Harris, S., Moroz, K. B., Hicks, K. J., Kas per, B. B., Culos, C., Sailor, W., & Pigott, T. D. (2006). Schoolwide application of positive behavior support in an urban high school: A case study. Journal of Positive Behavior Interventions, 8 (3) 131 145. Bradshaw, C. P., Mitchell, M. M., & Leaf, P. J. (2010). Examining the effects of schoolwide positive behavioral supports on student outcomes: Results from a randomized controlled effectiveness trial in elementary schools. Journal of Positive Behavior Inter ventions, 12 (3), 133 148. Bradshaw, C. P., Reinke, W. M., Brown, L. D., Bevans, K. B., & Leaf, P. J (2008). Implementation of school wide positive behavioral interventions and supports (PBIS) in elementary schools: Observations from a randomized trial. Education and Treatment of Children, 31( 1), 1 26. Retrieved from http://www.jstor.org/stable/42899960 Brady, K., Forton, M. B., Porter, D. (2015). Rules in school: Teaching discipline in the responsive c lassroom (2nd ed.). Turners Fall, MA: Center for Responsive Schools.

PAGE 227

227 Brooks Gunn, J., & Duncan Johnson, A. (2006). G. Stanley Hall's contribution to science, practice and policy: The child study, parent education, and child welfare movements. History of P sychology 9(3), 247. Broughton, S. F. & Lahey, B. B. (1978). Direct and collateral effects of positive reinforcement, response cost and mixed contingencies for academic performance. Journal of School Psychology, 16 (2), 126 136. Brown v. Board of Educati o n of Topeka, 347 U.S. 483 ( 195 4) challenged. New York Times p. A20. Byrd, C. E. & Vinson, T. S. (2008). Achievement testing in the 21st Century: Current status and fu ture directions. In F. T. L. Leong (Ed.) Encyclopedia of Counseling Psychology Thousand Oaks, CA: Sage. Cardinal, R. N. & Aitken, M. R. F. (2006). ANOVA for the behavioural sciences researcher Newark, NJ: Lawrence Erlbaum Associates. Retrieved from: https://egret.psychol.cam.ac.uk/statistics/weblinks_Cardinal_and_Aitken_ANOVA .html Carr, E. G., Taylor, J. C., & Robinson, S. (1991). The effects of severe behavior problems in children on the teaching behavior of adults. Journal of Applied Behavior Analysis, 24 (3) 523 535. Catania, A. C. (1979). Learning Englewood Cliffs, NJ: Prenti ceHall. Charney, R. S. (2015). Teaching children to care: Classroom management for ethical and academic growth, K 8. Turners Falls, MA: Center for Responsive Schools. Chomsky, N. (1959). A review of BF Skinner's Verbal Behavior. Language 35 (1), 26 58. Cl onan, S. M., McDougal, J. L., Clark, K., Davison, S. (2007). Use of office disciplinary referrals in school wide decision making: A practical example. Psychology in the Schools, 44 (1), 19 27. Cooper, J. O. (1982). Applied behavior analysis in education. Theory Into Practice, 21 (2), 114 118. Costenbader, V., & Markson, S. (1998). School suspension: A study with secondary school students. Journal of School Psychology 36 (1), 59 82. Cotton, K. (2001). Instructional reinforcement Portland, OR: Northwest R egional Educational Laboratory. Retrieved from: http://flpbs.fmhi.usf.edu/

PAGE 228

228 Dewey, J. (1986). Experience and education The Educational Forum, 50 (3), 241 252. http://dx.doi.org/10.1080/00131728609335764 Dickerson, F. B., Tenhula, W. N., & Green Paden, L. D. (2005). The token economy for schizophrenia: review of the literature and recommendations for future research. Schizophrenia Research 75 (2), 405 41 6. Dixon, D. R., Vogel, T., & Tarbox, J. (2012). A brief history of functional analysis and applied behavior analysis. In J. J. Matson (ed.), Functional assessment for challenging behaviors (pp. 3 24). New York, NY: Springer Science and Business Media. D orman J. & Adams, J. (2004). Associations between students' perceptions of classroom environment and academic efficacy in Australian and British secondary schools. Westminster Studies in Education, 27 (1), 69 85. Dunlap, G, Sailor, W., Horner, R., & Sugai, G. (2009). Overview and history of positive behavior support. In W. Sailor, G. Dunlap, G. Sugai, & R. Horner (Eds.), Handbook of positive behavior supports New York, NY: Springer Science & Business Media. Education Information and Accountability Services (2013). Gender representation 20D). Tallahassee, FL: Florida Department of Education. Retrieved from: http://www.fldoe.org/eias/ Eitle, T. M. & Eitle, D. J. (2004). Inequality, segregation, and the overrepresentation of African Americans in school suspensions. Sociological Perspectives, 47 (3), 269 287. Elliott, D. S., Hamburg, B. A., & Williams, K. R. (1998). Violence in American schools: A new perspective Cambridge, MA: Cambridge University Press. Elliott, S. N., Gresham, F., & Witt, J. C. (2013). Handbook of behavior therapy in education New York, NY : Springer Science & Business Media. Fairbanks, S., Sugai, G., Guardino, D., & Lathrop, M. (2007). Response to intervention: Examining classroom behavior support in second grade. Exceptional Children, 73 (3) 288 310. Fancher, R. E. (1996). Pioneers of psy chology WW Norton & Co. Fanselow, M. S. & Sterlace, S. R. (2014) Pavlovian Fear Conditioning. In F. K. McSweeney and E. S. Murphy (Eds.), The Wiley Blackwell Handbook of Operant and Classical Conditioning Oxford, UK: John Wiley & Sons. doi: 10.1002/97811 18468135.ch6

PAGE 229

229 Farmer, T. W., Quinn, M. M., Hussey, W., & Holohan, T. (2001). The development of disruptive behavior disorders and correlated constraints: Implications for intervention. Behavioral Disorders, 26 (2) 117 130. Field, T. (1981). Early Peer Rela tions. In Strain, P. S. (1981). P. S. Strain, The utilization of classroom peers as behavior change agents New York, NY: Plenum Press. Filcheck, H. A., McNeil, C. B., Greco, L. A. and Bernard, R. S. (2004), Using a whole class token economy and coaching o f teacher skills in a preschool classroom to manage disruptive behavior. Psychology in the Schools, 41 351 361. doi:10.1002/pits.10168 Flannery, K. B., Fenning, P., Kato, M. M., & McIntosh, K. (2014). Effects of school wide positive behavioral interventio ns and supports and fidelity of implementation on problem behavior in high schools. School Psychology Quarterly, 29 (2), 111 124. Florida Department of Education (Producer). (2008). Discovering school wide PBS: Moving towards a positive future [DVD]. Avail able from http://flpbs.fmhi.usf.edu Florida Department of Education. (2002c). Interested districts Retrieved from: http://flpbs.fmhi.usf.edu/interested.cfm Florida Department of Education. (2005, November). Frequently asked PBS Questions: University of South Florida Retrieved from http://flpbs.fmhi.usf.edu/faq.cfm Florida Department of Education. (2016, April). interventions and support project: A multi tiered system of supports. Tallahassee, http://flpbs.fmhi.usf.edu/pdfs/FLPBIS Facts 4.5.16.pdf upport Project. (2016). Annual report (2014 2015). Tampa, FL: University of South Florida. Retrieved from: http://flpbs.fmhi.usf.edu/pdfs/FLPBIS%20Annual%20Report%20201 4 2015.pdf Florida's Positive Behavior Support Project. (2002a). PBS fact sheet Tallahassee, FL: State of Florida Department of Education. Retrieved from http://flpbs.fmhi.usf.edu/ Florida's Positive Behavior Support Project. (2002b). A multi tiered system of supports Tallahassee, FL: State of Florida Department of Education. Retrieved from http://flpbs.fmhi.usf.edu/ Fraser, B. J. (1998). Classroom environment instr uments: Development, validity, and applications. Learning Environments Research, 1 7 33.

PAGE 230

230 Fraser, B. J. (2002). Learning environments research: Yesterday, today, and tomorrow. In S. C. Goh & M. S. Khine (Eds.), Studies in educational learning environments: An international perspective (pp. 1 26). Singapore: World Scientific. Freiberg, H. J. (Ed) (1999). School climate: Measuring, improving, and sustaining healthy learning environments. London: Falmer Press. French, A., Macedo, M., Poulsen, J., Waterson, T., & Yu, A. (2002). Multivariate analysis of variance (MANOVA). Retrieved January, 18, 2017, from: http://userwww.sfsu.edu/efc/classes/biol710/manova/MANOVAnewest.pdf George H. P., Kincaid, D. & Pollard Sage, P. (2009). Primary tier interventions and supports. In W. Sailor, G. Dunlap, G. Sugai, & R. Horner (Eds.), Handbook of positive behavior supports New York, NY: Springer Science & Business Media. George, M. P., White, G. P., & Schlaffer, J. J. (2007). Implementing school wide behavior change: Lessons from the field. Psychology in the Schools, 44 (1), 41 51. Glew G. M., Fan M., Katon W., Rivara F. P., Kernic M. A. (2005). Bullying, psychosocial adjustment, and academic p erformance in elementary school. Archives of Pediatric Adolescent Medicine, 159 (100), 1026 1031. doi:10.1001/archpedi.159.11.1026 Godwin, K. E., Almeda, M. V., Petroccia, M., Baker, R. S., & Fisher, A. V. (2013). Classroom activities and off task behavior in elementary school children. In M. Knauff, M. Pauen, N. Sebanz, and I. Wachsmuth (Eds.), Proceedings of the 35th Annual Meeting of the Cognitive Science Society 2428 2433. Gonsoulin, S., Zablocki, M., & Leone, P. E. (2012). Safe schools, staff developm ent, and the school to prison pipeline. Teacher Education and Special Education, 35 (4), 309 319. Gottfredson, D. C., Gottfredson, G. D., & Hybl, L. G. (1993). Managing adolescent behavior a multiyear, multischool study. American Educational Research Journal, 30 (1), 179 215. Greenwood, C. R., Delquadri, J., & Hall, R. V. (1984). Opportunity to respond and student academic performance. In W. L. Heward, T. E. Heron, J. Trap Porter, & D. S. Hill (Eds.), Focus on behavior analysis in education (pp. 58 88) Columbus, OH: Charles Merrill. Guardino, C. A. & Fullerton, E. (2010). Changing behaviors by changing the classroom environment. TEACHING Exceptional Children, 42 (6), 8 13.

PAGE 231

231 Gunter, P. L., Denny, R. K., Shores, R. E., Reed, T. M., Jack, S. L., & Nelson, C. M. (1994). Teacher escape, avoidance, and counterc ontrol behaviors: Potential res ponses to disruptive and aggressive behaviors of students with severe behavior disorders Jo urnal of Child and Family Studies, 3 (2) 211 223. Hanley, G. P., Iwata, B. A., & McCord, B. E. (2003). Functional analysis of problem behavior: A review. Journal of Applied Behavior Analysis, 36 (2) 147 185. Heitzig, N. A. (2009). Education or incarcerat ion: Zero tolerance policies and the school to prison pipeline. Forum on Public Policy Retrieved August 8, 2016, from: http://files.eric.ed.gov/fulltext/EJ870076.pdf Hergenhahn, B.R. (1992). An Introduction to the History of Psychology, 2nd ed. Belmont, CA: Wadsworth. Honig v. Doe, 484 U.S. 305 ( 1988) Horner, R. H. Sugai, G., Dickey, C. R. (2016, July). School wide positive behavior support Presentati on retrieved from www.pbis.org Horner, R. H., Sugai, G., Todd, A. W., & Lewis Palmer, T. (2005). School wide positive behavior support: An alternative approach to discipline in schools. In L. Bambara & L. Kern (Eds.), Individualized supports for students wit h problem behaviors: Designing positive behavior plans (pp. 359 390). New York, NY: Guilford Press. Horner, R. H., Todd, A. W., Lewis Palmer, T., Irvin, L. K., Sugai, G., & Boland, J. B. (2004). The school wide evaluation tool (SET): A research instrument for assessing school wide positive behavior support. Journal of Positive Behavior Interventions, 6 (1), 3 12. Hornik, K. (2016). The R FAQ. Retrieved from: https://CRAN.R project.org/doc/FAQ/ Individuals Individuals with Disabilities Education Act Amendments of 1997 [IDEA]. (1997). Retrieved from https://www.congress. gov/105/plaws/publ17/PLAW 105publ17.pdf Irvin, L.K., Horner, R.H., Ingram, K., Todd, A.W., Sugai, G., Sampson, N., & Boland, J. (2006). Using office discipline referral data for decision making about student behavior in elementary and middle schools: An e mpirical investigation of validity. Journal of Positive Behavior Interventions 8(1), 10 23. Irvin, L.K., To bin, T., Sprague, J., Sugai, G., & Vincent, C. (2004). Validity of office discipline referral measures as indices of school wide behavioral status and effects of school wide behavioral interventions. Journal of Positive Behavioral Interventions 6 (3), 131 147.

PAGE 232

232 James, W. (1925). Talks to ideals. New York, NY: Henry Holt and Company. Retrieved from: http://www.gutenberg.org/files/16287/16287 h/ Jeong, S., & Lee, B. H. (2 013). A multilevel examination of peer victimization and bullying preventions in schools. Journal of Criminology, Volume 2013 10 Pages. http://dx.doi.org/10.1155/2013/735397 Jones, C. P. (2000). Level American Journal of Public Health, 90 (8), 1212 1215. Jones, C., Caravaca, L., Cizek, S., Horner, R. H., & Vincent, C. (2006). Culturally responsive schoolwide positive behavior support: A case stud y in one school with a high proportion of Native American students. Multiple Voices for Ethnically Diverse Exceptional Learners, 9 (1), 108 119. Kappel, B., Dufresne, D., & Mayer, M. (2012). From behavior management to positive behavioral support Report, Department of Health and Human services, Administration on Developmental Disabilities. Karweit, N. & Slavin, R. E. (1981). Measurement and modeling choices in studies of time and learning. American Educational Research Journal, 18 (2), 157 171. Kazdin, A. E. (1973). The effect of vicarious reinforcement on attentive behavior in the classroom. Journal of Applied Behavior Analysis 6 (1), 71 78. Kazdin, A. E. (1977). Token economy: A review and evaluation New York, NY: Plenum Press Kazdin, A. E. (1987). T reatment of antisocial behavior in children: Current status and future directions. Psychological Bulletin, 102 (2) 187 203. Kazdin, A. E. (2012a). Behavior modification in applied settings: Seventh edition Long Grove, IL: Waveland Press. Kazdin, A. E. (2 012b). The token economy: A review and evaluation New York, NY: Springer Science & Business Media. Keller, F. S., & Schoenfeld, W. N. (2014). Principles of psychology: A systematic text in the science of behavior (Vol. 2). Cambridge, MA: B.F. Skinner Foundation. Kelling, G. L., & Coles, C. M. (1997). Fixing broken windows: Restoring order and reducing crime in our communities. New York, NY: Touchstone. Kincaid, D., Childs, K., Blase, K. A., Wallace, F. (2007). Identifying barriers and facilitators in implementing schoolwide positive behavior support. Journal of Positive Behavior Interventions 9 (3), 174 184.

PAGE 233

233 Lane, T. W., & Murakami, J. (1987). School pr ograms for delinquency prevention and intervention. In Behavioral approaches to crime and delinquency (pp. 305 327). New York, NY: Springer Science & Business Media. Lassen, S. R., Steele, M. M., & Sailor, W. (2006). The relationship of school wide positi ve behavior support to academic achievement in an urban middle school. Psychology in the Schools, 43 (6), 701 712. Leary, M. R., Kowalski, R. M., Smith, L., & Phillips, S. (2003). Teasing, rejection, and violence: Case studies of the school shootings. Agg ressive B ehavior, 29 (3), 202 214. doi:10.1002/ab.10061 Lee, S. W., Kelly, K. E., & Nyre, J. E. (1999). Preliminary report on the relation of task behavior with completion of school work. Psychological Reports, 84 (1) 267 272. Lemov, D. (2 010). Teach like a champion: 49 techniques that put students on a path to college San Francisco, CA: Jossey Bass Lewis, T. J., Barrett, S., Sugai, G., Horner, R. H., Mitchell, B. S., & Starkey, D. (2016). Training and professional development blueprint f or positive behavioral interventions and supports Washington, OR: National Technical Assistance Center on Positive Behavioral Interventions and Supports Linscheid, T.R., Iwata, B.A., Ricketts, R.W., Williams, D.E., & Griffin, J.C. (1990). Clinical evaluat ion of the self injurious behavior inhibiting system (SIBIS). Journal of Applied Behavior Analysis, 23 (1), 53 78. doi:10.1901/jaba.1990.23 53 Lipsey, M. W. (1992). Juvenile delinquency treatment: A meta analytic inquiry into the variability of effects. I n T. D. Cook, H. Cooper, D. S. Cordray, H. Hartmann, L. V. Hedges, R. J. Light, T. A. Louis, & F. Mosteller (Eds.), Meta Analysis for Explanation (pp. 83 127). New York, NY: Russell Sage Foundation. Luiselli, J. K., Putnam, R. F., Handler, M. W., & Feinber g, A. B. (2005). Whole school positive behaviour support: Effects on student discipline programs and academic performance. Educational Psychology, 25 (2 3), 183 198. Mace, F. C. (1994). The significance and future of functional analysis methodologies. Jou rnal of Applied Behavior Analysis, 27 (2), 385 392. Mayer, G. R. (1995). Preventing antisocial behavior in the schools. Journal of Applied Behavior Analysis 28 (4), 467 478. Mayer, G. R., & Sulzer Azaroff, B. (1991). Interventions for vandalism. In G. Sto ner, M. K. Shinn, & H. M. Walker (Eds.), Interventions for achievement and behavior problems (pp. 559 580). Washington, DC: National Association of School Psychologists Monograph.

PAGE 234

234 McConnell, J. V. (2001). Negative Reinforcement and Positive Punishment. In Griggs, R. A. (Ed.), Handbook for teaching introductory psychology: With an emphasis on assessment (2 nd ed.) (214 216). Mahwah, NJ: Lawrence Erlbaum Associates. McCurdy, B. L., Mannella, M. C., & Eldridge, N. (2003). Positive behavior support in urban schools: Can we prevent the escalation of antisocial behavior? Journal of Positive Behavior Interventions, 5 (3) 158 170. McIntosh, K. & Goodman, S. (2016). Integra ted multi tiered systems of support: Blending RTI and PBIS New York, NY: The Guilford Press. McIntosh, K., Bennett, J. L., & Price, K. (2011). Evaluation of social and academic effects of school wide positive behavior support in a Canadian school district Exceptionality Education International, 21 (1), 46 60. McKevitt, B. C., Dempsey, J. N., Ternus, J., & Shriver, M. D. (2012). Dealing with behavior problems: The use of positive behavior support strategies in summer programs. Afterschool Matters, 15, 16 25. Retrieved from: http://files.eric.ed.gov/fulltext/EJ980189.pdf McLeod, S. A. (2013). Pavlov's dogs Retrieved from www.simplypsy chology.org/pavlov.html Mclntosh, K., Campbell, A., Carter, D., & Zumbo, B. (2009). Concurrent validity of office discipline referrals and cut points used in schoolwide positive behavior support. Behavioral Disorders, 34 (2), 100 113. Retrieved from http://www.jstor.org/stable/43153806 Menendez, A. L., Payne, L. D., & Mayton, M. R. (2008). The implementation of positive behavioral support in an elementary school: Processes, procedures, and outcomes. The Alberta Journal of Educational Research, 54 (4), 448 462. Miller, A. & Cunningham, K. (2011). Classroom environment. Education.com. Retrieved from: http://www.education.com /reference/article/classroom environment/ Mills v. Board of Educati on of the District of Columbia, 348 F.Supp. 866 (D.D.C. 1972) Miltenberger, R. G. (2012). Behavior modification: Principles and procedures (5 th ed.). Belmont, CA: Wadsworth, Cengage Learning. Miltenberger, R. G., & Crosland, K. A. (2014). Parenting. The wiley blackwell handbook of operant and classical conditioning (pp. 509 531). Wiley Blackwell. doi:10.1002/9781118468135.ch20 Mirzamani, S. M., Ashoori, M., & Sereshki, N. A. (2011). The effect of social and token economy reinforcements on academic achievement of students with intellectual disabilities. Iranian Journal of Psychiatry 6 (1), 25 30. Retrieved from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3395934/

PAGE 235

235 Moos, R. (1979) Evaluating educational environments: procedures, measures, findings and policy implications San Francisco: Jossey Bass. NAACP Legal Defense and Educational Fund. (2011). Dismantling the school to prison pipe line New York, NY. Retrieved August 8, 2016, from http://www.indiana.edu/~atlantic/wp content/uploads/2011/11/NAACP Dismantling_the_School_to_Prison_Pipeline.pdf National Association of Elementary School Principals (1997, April 11). NAESP: Tough Discipline Policies a Must for Safe Schools, Say Principals in Nationwide Poll Retrieved from: http://www.prnewswire.com/news releases/naesp -tough discipline policies a must for safe schools say principals in nationwide poll 75101062.html National School Safety Center. (2001). NSSC review of school safety research. Westlake Village, CA: Author. Nelson, J. R., & Roberts, M. L. (2000). Ongoing r eciprocal teacher student interactions involving disruptive behaviors in general education classrooms. Journal of Emotional & Behavioral Disorders, 8 (1) 27 37. Nelson, J. R., Benner, G. J., Reid, R. C., Epstein, M. H., & Currin, D. (2002). The convergent validity of office discipline referrals with the CBCL TRF. Journal of Emotional & Behavioral Disorders, 10 (3) 181 188. Nelson, J., Lott, L., & Glenn, H. S. (2013). Positive discipline in the classroom: Developing mutual respect, cooperation, and respons ibility in your classroom (4 th ed.). New York, NY: Three Rivers Press. Newcomer, L. & Barrett, S., (2009). Administration roles and functions in PBIS high schools. In B. Flannery & G. Sugai (Eds.), SWPBS implementation in high schools: Current practice and future directions (pp 23 42). University of Oregon. Nieto, S. (2005). Public education in the twentieth century and beyond: High hopes, broken promises, and an uncertain future. Harvard Educational Review 75 (1), 43 64. Nocera, E. J., Whitbread, K. M., & Nocera, G. P. (2014). Impact of school wide positive behavior supports on student behavior in the middle grades. Research in Middle Level Education Online, 37 (8), 14 pages. Retrieved from: http://files.eric.ed.gov/ Office of Special Education Programs (OSEP) Technical Assistance Center on Positive Behavioral Interventions and Supports (PBIS). (2016). Positive behavioral interventions and supports. Washington, DC: United States Department of Education, Office of Special Education Programs. Retrieved from http://www.pbis.org/

PAGE 236

236 Olivares Rodriguez, J., Montesinos, L., & Preciado, J. (2005). A 19th Century predecessor of the token economy. Journal of Applied B ehavior Analysis, 38 (3), 427. http://doi.org/10.1901/jaba.2005.181 03 Olson, M. H. & Hergenhahn, B. R. (2015). An introduction to theories of learning (9 th ed.). New York, NY: Psychology Press. Paja res, F. (2003). William James: Our father who begat us. In B. J. Zimmerman & D. H. Schunk (Eds.). Educational psychology: A century of contributions (pp. 41 64). Mahwah, NJ: Lawrence Erlbaum Associates. Retrieved from: https://www.uky.edu/~eushe2/Pajares/PajaresJames.PDF Pas, E. T., Bradshaw, C. P., & Mitchell, M. M. (2011). Examining the validity of office discipline referrals as an indicator of student behavior problems. Psychology in the Schools, 48 (6), 541 555. doi:10.1002/pits.20577 Patrick, H., Ryan, A., & Kaplan, A. (2007). Early adolescents' perceptions of the classroom social environment, motivational beliefs, and engagement. Journal of Educational Psychology, 99, 83 98. Pearle L. (2016, February 12). School shootings since Columbine: By the numbers. ABC News. Retrieved from: http://abcnews.go.com/US/school shootings columbine numbers/ story?id=36833245 Peetsma, T., Vergeer, M., Roeleveld, J., & Karsten, S. (2001). Inclusion in education: Comparing pupils' development in special and regular education. Educational Review, 53 (2), 125 135. Peters, W. (1970). The eye of the storm [Documentary]. USA: American Broadcasting Company (ABC). Pierce, W. D. & Cheney, C. D. (2013). Behavior analysis and learning (3 rd ed.). New York, NY: Psychology Press. Poe Yamagata, E., & Jones, M. J. (2000). And justice for some: Differential treatment of minority youth in the justice system. Washington, DC: Building Blocks for Youth. Powell, J. J. (2015). Barriers to inclusion: Special education in the United States and Germany. New York, NY: Routledge. Public Agenda (2004). Teaching interrupted: Do di schools foster the common good ? New York, NY: Author. Retrieved from: www.publicagenda.org Putnam, R. F., & Kincaid, D. (2015). School wide PBIS: Extending the impact of applie d behavior analysis. Why is this important to behavior analysts?. Behavior Analysis in Practice 8 (1), 88 91.

PAGE 237

237 Putnam, R. F., Horner, R. H., & Algozzine, R. (2006). Evaluation brief. Educational and Community Supports, University of Oregon, Eugene, Oregon. Retrieved from: https://www.pbis.org/newsletters/volume 3 Putnam, R. F. S., Stone, L., (2009). Establishing and maintaini ng staff participation in PBIS high schools. In B. Flannery & G. Sugai (Eds.), SWPBS implementation in high schools: Current practice and future directions (pp. 43 56). University of Oregon. Rehabilitation Act of 1973, 29 U.S.C. § 701. Reichart, C. S. (2 009). Quasi experimental design. In R. E. Millsap & A. Maydeu Olivares (Eds.), Quantitative methods in psychology Los Angeles, CA: Sage. Rescorla, R. A. (2014). Pavlovian Second Order Conditioning (Psychology Revivals): Studies in Associative Learning New York, NY: Psychology Press. Richardson, V. (2002, May 3). Zero tolerance=zero intelligence The Washington Times. Retrieved from http://www.washingtontimes.com/New s/2002/May/13/200205 02465 5038r/print Rocque, M. (2010). Office discipline and student behavior: Does race matter? Journal of Education, 116 (4), 557 581. Sailor, W., Dunlap, G., Sugai G., & Horner, R. (Eds.). (2009 ). Handbook of positive behavior support New York, NY: Springer Science & Business Media. Scott, T. M. & Barrett, S. B. (2004). Using staff time and student time engaged in disciplinary procedures to evaluate the impact of school wide PBS. Journal of Positive Behavior Interventions, 6 (1 ), 21 27. Retrieved from: http://journals.sagepub.com/doi/pdf/10.1177/10983007040060010401 Shorr, A. (2013). Behavior modification. In H. J. Eysenck & I. Martin (Eds.). (2013) Theoretical foundations of behavior therapy (37 56) New York, NY: Springer Science & Business Media. Singer, G. H. S. & Wang, M. (2009). The intellectual roots of positive behavior support and their implications for its development. In W. Sailor, G. Dun lap, G. Sugai, & R. Horner (Eds.), Handbook of positive behavior supports New York: Springer Science & Business Media. Skiba, R. J. (2000). Zero tolerance, zero evidence. An analysis of school disciplinary practice (Policy Research Report SRS2). Bloomingt on, IN: Indiana Education Policy Center. Skiba, R. J., & Knesting, K. (2001). Zero tolerance, zero evidence: An analysis of school disciplinary practice. New directions for youth development 92, 17 43.

PAGE 238

238 Skiba, R. J. & Peterson, R. (1999). The dark side of zero tolerance: Can punishment lead to safe schools?. The Phi Delta Kappan, 80 (5), 372 382. Skiba, R. J., Arrendondo, M. I., & Rausch, M. K. (2014). New and developing research on disparities in discipline Bloomington, IN: The Equity Project at Indiana University, Center for Evaluation and Education Policy. Skiba, R. J., Horner, R. H., Chung, C. G., Rausch, M. K., May, S. L., & Tobin, T. (2011). Race is not neutral: A national investigation of African American and Latino disproportionality in school dis cipline. School Psychology Review, 40 (1), 85 107. Skiba, R. J., Michael, R. S., Nardo, A. C. & Peterson, R. (2000). The color of discipline: Sources of racial and gender disproportionality in school p unishment (Policy Research Report #SRS1). Indianapolis, IN: Indiana Education Policy Center. Skiba, R. J. Reynolds, C. R., Graham, S., Sheras, P., Conoley, J. C., & Garcia Vazquez, E. (2006). Are zero tolerance policies effective in the schools? An evidentiary review and recommendations. A Report by the Ameri can Psychological Associate Zero Tolerance Task Force Washington, DC: American Psychological Association. Skinner, B. F. (1953). Science and human behavior New York: Simon & Schuster. Smith, B. L. (2012). The case against spanking: Physical discipline i s slowly declining as some studies reveal lasting harms for children. Monitor on Psychology, 43 (4). Washington, DC: American Psychological Association. Sned ecor, G. W. & Cochran, W. G. (1989). Statistical methods: 8th edition. Ames, Iowa: Iowa State Press Soares, D. A., Harrison, J. R., Vannest, K. J., & McClelland, S. S. (2016). Effect size for token economy use in contemporary classroom settings: A meta analysis of single case research. School Psychology Review, 45 (4), 379 399. Spencer, J. P. (2015). Effect of positive behavioral interventions and supports on school wide discipline in a Title I intermediate school. The Online Journal of New Horizons in Education, 5 (4). Retrieved from www.toj ned.net Sprague, J., Walker, H. M., Stieber, S., Simonsen, B., Nishioka, V., & Wagner, L. (2001). Exploring the relationship between school discipline referrals and delinquency. Psychology in the Schools, 38 (2) 197 206. Sprague, J., Walker, H., Golly, A ., White, K., Myers, D. R., & Shannon, T. (2001). Translating research into effective practice: The effects of a universal staff and student intervention on indicators of discipline and school safety. Education and Treatment of Children, 24, 495 511.

PAGE 239

239 Spri ck, R., Borgmeier, C., & Nolet, V. (2002). Prevention and management of secondary level behavior problems. In Shinn, M. R., Walker, H. M., Stoner, G. (Eds.). Interventions for academic and behavior problems II: Preventive and remedial approaches (pp. 315 3 50). Washington, DC: National Association of School Psychologists. St. George, D. (2014, January 8). Holder, Duncan announce national guidelines on school discipline Washington Post. Retrieved from: https:// www.washingtonpost.com/ Stormont, M., Lewis, T. J., Beckner, R., & Johnson, N. W. (2008). Implementing positive behavior support systems in early childhood and elementary settings Thousand Oaks, CA: Corwin Press. Sugai, G. (2016). School wide positive beh avior support and response to intervention New York, NY: National Center for Learning Disabilities. Retrieved from: http://www.rtinetwork.org/learn/behavior supports/scho olwidebehavior Sugai, G. & Horner, R. H (2006). A promising approach for expanding and sustaining school wide positive behavior support. School Psychology Review, 35 ( 2), 245 259. Sugai, G. & Horner, R. H (2009 a ). Defining and describing schoolwide positive behavior support In W. Sailor, G. Dunlap, G. Sugai, & R. Horner (Eds.), Handbook of positive behavior supports New York: Springer Science & Business Media. Sugai, G. & Horner, R. H. (2009 b ). Responsiveness to intervention and school wide positiv e behavior supports: Integration of multi tiered system approaches. Exceptionality: A Special Education Journal, 14 (4), 223 237. Sugai, G Horner, R. H., Gresham, F. M. (2002). Behaviorally Effective School Environments. In Shinn, M. R., Walker, H. M., S toner, G. (Eds.). Interventions for academic and behavior problems II: Preventive and remedial approaches (pp. 315 350). Washington, DC: National Association of School Psychologists. Sugai, G., Lewis Palmer, T., Todd, A., & Horner, R. H. (2001). School wid e evaluation tool Eugene: University of Oregon. Sugai, G. & Simonsen, B. (2012). Positive behavioral interventions and supports: History, defining features, and misconceptions Hartford, CT: Center for PBIS & Center for Positive Behavioral Interventions a nd Supports. Tabachnick, B.G., & L.S. Fidell. (1996). Using multivariate statistics Harper Collins College Publishers: New York.

PAGE 240

240 Taylor Greene, S., Brown, D., Nelson, L., Longton, J., Gassman, T., Cohen, J., Swartz, J., Horner, R. H., Sugai, G., & Hall, S. (1997). School wide behavioral support: Starting the year off right. Journal of Behavioral Education, 7 (1), 99 112. Teske, S. C. (2011). A study of zero tolerance policies in schools: A multi integrated systems approach to improve outcomes for adolesce nts. Journal of Child and Adolescent Psychiatric Nursing 24 (2), 88 97. Teske, S. C. (2012). Testimony before the Senate Subcommittee on The Constitution, ngton, DC: United States Government Printing Office. Retrieved from: https://www.gpo.gov/fdsys/pkg/CHRG 112shrg86166/pdf/CHRG 112shrg86166.pdf Timothy W. v. Rocheste r, New Hampshire, School District, 875 F.2d 954 (1st Cir., 1989) Tolan, P., & Guerra, N. (1994). What works in reducing adolescent violence Boulder, CO: The Center for the Study and Prevention of Violence. Torrey, E. F. (1997). Out of the shadows: Confron ting America's mental illness crisis New York, NY: John Wiley & Sons. U.S. Department of Education (2015). 37th annual report to Congress on the implementation of the Individuals with Disabilities Education Act. Washington, DC: Office of Special Education and Rehabilitative Services. Retrieved from: https://www2.ed.gov/about/reports/annual/osep/2015/parts b c/37th a rc for idea.pdf U.S. Department of Education, Office for Civil Rights & U.S. Department of Justice, Civil Rights Division. (2014). Dear Colleague Letter and Guidance: Nondiscriminatory Administration of School Discipline Washington, DC: Authors. U.S. Offi ce of Special Education Programs. (2016, December). Tier 1 FAQs Retrieved from: http://www.pbis.org/school/tier1supports/tier1faqs Ullmann, L. P. & Krasner, L. (1975). A psychological app roach to abnormal behavior (2nd ed.). Englewood Cliffs, NJ: Prentice Hall. United States Department of Education. (2010). Free appropriate public education for students with disabilities: Requirements under section 504 of the Rehabilitation Act of 1973. Wa shington, DC: Author. Retrieved from: https://www2.ed.gov/about/offices/list/ocr/docs/edlite FAPE504.html United States Department of Education. (2015). Protecting studen ts with disabilities Washington, DC: Author. Retrieved from: http://www2.ed.gov/about/offices/list/ocr/504faq.html

PAGE 241

241 Vincent, C. G., Swain Bradway, J., Tobin, T. J., & May, S. (2011). Dis ciplinary referrals for culturally and linguistically diverse students with and without disabilities: Patterns resulting from school wide positive behavior support. Exceptionality: A Special Education Journal, 19 (3), 175 190. Waasdorp, T. E., Bradshaw, C. P., & Leaf, P. J. (2012). The impact of schoolwide positive behavioral interventions and supports on bullying and peer rejection: A randomized controlled effectiveness trial. Archives of Pediatric Adolescent Medicine, 166 (2), 149 156. Wald, J., & Losen, D. (2003). Defining and re directing a school to prison pipeline. In J. Wald & D. Losen (Eds.), New directions for youth development: Deconstructing the school to prison pipeline (pp. 9 16). Hoboken, NJ: Wiley & Sons. Walker H. M., & Shinn, M. R. (2002). Structuring school based interventions to achieve integrated primary, secondary, and tertiary prevention goals for safe and effective schools. In M. R. Shinn, G. Stoner, & H. M. Walker (Eds.), Interventions for academic and behavior problems: Pre ventive and remedial approaches (pp. 1 26). Silver Spring, MD: National Association of School Psychologists. Walker, H., Colvin, G., & Ramsey, E. (1995). Antisocial behavior in school. Strategies and best practices Pacific Grove, CA: Brooks/Cole. Weed, S. E., & Lickona, T. (2014). Abstinence education in context: History, evidence, premises, and comparison to comprehensive sexuality education. In M. Kenny (Ed.), Sex education Hauppauge, NY: Nova Science Publishers. Wehby, J. H., Symons, F. J., & Shores, R E. (1995). A descriptive analysis of aggressive behavior in classrooms for children with emotional and behavioral disorders. Behavioral Disorders, 20 (2) 87 105. Weingarten, R. (2015). Moving past punishment towards support. American Educator, 39 (4), 1 achievement, and social behavior. In Wentzel, K. R. & Ramani, G. B. (Eds.), Handbook of social influences in school contexts: Social emotional, motivation, and cognitive outcomes New York, NY: Routledge. William Moore, J.; Manning, S. A.; Smith, W. I. (1978). Conditioning and Instrumental Learning New York, NY: McGraw Hill Book Company. pp. 52 61. Woodward, M. (2004). Epidemiology: Study design and analysis (2 nd edition). Boca Raton, FL: CRC Press / Taylor & Francis Group. Ylvisaker, M., Feeney, T., & Hibbard, M. (2006). What are positive behavior supports? Albany, NY: The Brain Injury Association of New York. Retrieved from http://www.projectlearnet.org/tutorials/positive_behavior_supports.html

PAGE 242

242 Yount, W. R. (2010). Created to learn Nashville, TN: B&H Publishing Group. Zhang, A., Musu Gillette, L., & Oudekerk, B. A. (2016). Indicators of school crime and safety: 2015 Washington, DC: Institute of Education Sciences: National Center for Education Statistics.

PAGE 243

243 BIOGRAPHICAL SKETCH Joshua Earl P itts White was born in Gainesville, Florida at Shands Teaching Hospital at the University of Florida. He grew up just outside of the small rural town of Bronson, Florida in Levy County. There, he lived with his two parents and six siblings (four brothers and two sisters). He attended Bronson High School and graduated with honors in 1999. Upon graduating high school, Mr. White attended Central Florida Community College in Ocala, Florida, as well as Santa Fe Community College in Gainesville, Florida. In May of 2002, he graduated with high honors wi th an Associate in Arts degree from Central Florida Community College. Upon transferring to the University of Florida, he earned a Bachelor of Music in music e ducation degree in 2005 graduating with highest honors, Summa Cum Laude, a Master of Music degre e with a major in m usic e ducation in 2008 a Specialist in Education in educational l eadership degree in 2013, and fina lly his Doctor of Education in educational l eadership degree in 2017. Since uously worked full time while pursuing his graduate studies serving in various roles as a music teacher, classroom teacher, behavioral resource teacher, and principal intern for Alachua County Public Schools for the last eleven years. Mr. White also served two years as a program evaluator for curriculum, instruction, and afterschool activities for the State of Florida 21st Century Community Learning Centers Project.