Need corrections made

davidifediba4

Disproportionate_Exclusionary_2.pdf

Home >Psychology homework help >Need corrections made

DISPROPORTIONATE EXCLUSIONARY DISCIPLINE PRACTICES AND

THE IMPLEMENTATION AND FIDELITY OF PBIS IN HIGH SCHOOLS

A THESIS

Presented to the Department of Advanced Studies in Education and

Counseling California State University, Long Beach

In Partial Fulfillment

of the Requirements for the Degree

Educational Specialist in School Psychology

Committee Members:

Kerri Knight-Teague, Ph.D. (Chair)

Jacob Olsen, Ph.D.

Troya Ellis, Ed.D.

College Designee:

Shireen Pavri, Ph.D.

By Caitline T. Castillo

B.S., 2018, California State University San Marcos

May 2022

ABSTRACT

For several decades, African American students have experienced disproportionate rates

of punitive discipline that removes them from their learning environment. Students who

experience exclusionary discipline can face a wide range of harmful effects to their education and

future. African American students are more at risk to experience these effects due to the

discipline gap. Alternatives to exclusionary discipline, such as Positive Behavioral Interventions

and Supports (PBIS), have been found to reduce the use of punitive discipline practices. In this

study, 41 public high schools implementing PBIS were selected to compare rates of exclusionary

discipline (i.e., suspension and expulsion) with public high schools that did not implement PBIS.

This study also examines the relationship fidelity levels may have on exclusionary discipline

practices. Findings indicate a continued existence of the discipline gap, differing exclusionary

discipline rates between PBIS and non-PBIS schools, and differing expulsion rates between

different levels of fidelity. This study extends the current research on the implementation of PBIS

and its relationship with exclusionary discipline rates. Findings imply a need for a change in

disciplinary practices to reduce the discipline gap and its harmful effects.

iii

ACKNOWLEDGEMENTS

I would like to acknowledge and give my warmest thanks to my supervisor, Dr. Erika

Kato, who made this work possible. Her guidance and advice carried me through all the stages of

writing my thesis. I would also like to thank Dr. Powers and Dr. Knight-Teague for their patience

and support. Their guidance and help contributed to my continued persistence in completing my

thesis. Additionally, I would like to thank my committee members, Dr. Jacob Olsen and Dr.

Troya Ellis, for contributing to a positive defense experience and for all their comments and

suggestions.

I would also like to give a special thanks to my forever partner, Ryan Stanchfield, for his

support and understanding in taking on this research and writing project. Your continued support

kept me going each day and I would not have been able to sit down and concentrate without your

kind, motivating words. I would also like to acknowledge my family, friends, and cohort family

for allowing me to express my worries and always meeting those worries with support and

motivation.

TABLE OF CONTENTS

ABSTRACT.................................................................................................................................. ii

ACKNOWLEDGEMENTS......................................................................................................... iii

LIST OF TABLES......................................................................................................................... v

1. INTRODUCTION ....................................................................................................... 1

2. REVIEW OF LITERATURE .................................................................................... 10

3. METHODOLOGY .................................................................................................... 25

4. RESULTS .................................................................................................................. 33

5. DISCUSSION............................................................................................................. 42

REFERENCES ............................................................................................................................ 52

LIST OF TABLES

1. Descriptive Statistics: Enrollment, Race, Grade Level, and English Learner Status ....... 34

2. Exclusionary Discipline Rates by Ethnicity ..................................................................... 35

3. Descriptive Statistics for PBIS Implementation on Exclusionary Discipline................... 37

4. Median, Skewness, and Kurtosis of PBIS Implementation on Exclusionary Discipline... 38

5. Independent-Samples Mann-Whitney U Test for PBIS Implementation and Exclusionary Discipline....................................................................................... 38

6. Descriptive Statistics for PBIS Fidelity on Exclusionary Discipline................................. 39

7. The Kruskal-Wallis test of PBIS Fidelity Implementation and Exclusionary Discipline Rates..................................................................................................................... 40

CHAPTER 1

INTRODUCTION

On May 17, 1954, the U.S. Supreme Court Justice, Earl Warren, issued the landmark

decision, in the case of Brown v. Board of Education (1954), to desegregate schools. This

decision was made on the basis that previous educational inequalities deprived students of color

protection of the laws guaranteed by the fourteenth amendment (Brown v. Board of Education,

1954). The significance of this decision meant that the United States public school system would

provide integrated and equal education for all of its students. However, over 65 years have

passed since this landmark decision and studies have shown persistent lower academic

achievement, higher rates of placement in special education, and higher rates of suspension and

expulsion among students of color in comparison to the experiences of White students (Burchinal

et al., 2011; Sullivan & Bal, 2013; Skiba et al., 2014). The persistence of disproportionality

between students of color and White students, both academically and in discipline, demonstrates

that the decision made in the Brown v. Board of Education (1954) has not been fully adopted.

The quality of education within public schools is not equal for all students.

In the 2014-15 school year African American students made up 15.5% of all public

school students but represented 39% of students suspended from school (United States

Government Accountability Office, 2018). More recently, data from the 2017-18 school year

demonstrated that African American students made up 15.1% of total student enrollment in K-12

but represented 33.3% of expulsions without educational services and 38.8% of expulsions with

educational services (Civil Rights Data Collection [CRDC], 2021). The CRDC (2021) also

reported that 31.4% of African American students received one or more in-school suspensions

(ISS) and 38.2% of these students received one or more out-of-school suspensions (OSS) during

the 2017-18 school year. These data indicate there has been little to no change in the

overrepresentation of African American students experiencing suspensions and expulsions when

comparing data from 2014-15 to 2017-18 school years. OSS, ISS, and expulsion are all forms of

exclusionary discipline, or a disciplinary practice that physically removes students from their

learning environment (Gagnon et al., 2016).

Exclusionary discipline and its effects impact students of color at greater,

disproportionate rates compared to White students. In particular, students of color tend to be

disciplined at higher rates, and for offenses that are less severe and more subjective, when

compared to their White peers (Skiba et al., 2011). This disparity in punitive practice is known

as the discipline gap. The presence of a discipline gap has been showcased as early as 1975 in a

report by the Children’s Defense Fund (1975). This report used data submitted to the federal

office for civil rights by over 2,000 school districts in order to investigate the relationship among

race, sex, achievement, and suspension from the 1974-75 academic school year. The findings

indicate that African American students were 3 times more likely at the elementary level, and 2

times more likely at the secondary level, to be suspended compared to White students. In

addition, the Children’s Defense Fund (1975) found that there was also a relationship between

underachievement and discipline, such that those who received exclusionary discipline were also

categorized as students who were underachieving. These results indicate that both race and

underachievement were related to suspension. Over 45 years later, evidence continues to

demonstrate the existing presence of a discipline gap across all academic levels (Gopalan &

Nelson, 2019). This discipline gap has existed across several decades and should be addressed

due to the harmful consequences that arise from it.

The removal of students from a learning environment can be detrimental to their

education and can have long-term, harmful effects. Students who have experienced out-of-school

suspension or expulsion are at an increased risk for school disengagement, poor academic

achievement and behavior outcomes, dropout, or failure to graduate on time, and involvement

with the juvenile justice system (Skiba et al., 2014). Skiba et al. (2014) report that exclusionary

discipline has a negative effect on both educational opportunity and school engagement which are

two predictors of academic achievement. Academic achievement is negatively associated with

exclusionary discipline in the areas of state accountability testing, reading achievement, writing

achievement, and academic grades (Arcia, 2006; Raffaele Mendez et al., 2002; Rocque, 2010).

This means that students who experience disciplinary actions that remove them from the learning

environment are less likely to succeed in school. Furthermore, research demonstrates that being

suspended increases students’ likelihood of dropping out or failing to graduate on time (Christle

et al., 2005; Suh et al., 2007). According to Christle et al. (2005), students who dropout are 8

times more likely to be incarcerated. Exclusionary discipline has been shown to be associated

with many long-term effects for the individual students involved. More importantly, due to the

discipline gap, these negative consequences are affecting students of color at higher rates

compared to their White peer counterparts.

Over the decades, several strategies have been proposed to reduce or eliminate the

discipline gap. These alternative approaches to discipline include relationship-building,

emotional literacy, and structural interventions (Skiba et al., 2014). These three approaches are

components of school climate and school discipline that are predicted to lead to a reduction in

the discipline gap. Relationship building approaches, such as restorative practices, have been

shown to reduce the use of exclusionary discipline and narrow the discipline gap (Gregory et al.,

2016). Gregory et al. (2016) demonstrated that high school teachers with high ratings on

restorative practice implementation tended to have narrower racial discipline gaps, were

perceived as more respectful by their students, and issued fewer exclusionary discipline referrals

to African American and Hispanic/Latino students when compared to teachers with low ratings

on restorative practice implementation. Emotional literacy, such as the implementation of

empirically validated social and emotional learning programs, have been shown to reduce

negative behavioral incidents by almost 50% and have shown a district-wide reduction in the use

of out-of-school discipline by 60% (Osher et al., 2013). However, despite this implementation,

disproportionality in discipline remained present.

System-levels approaches, such as a positive behavior intervention and supports (PBIS)

framework, is another type of approach that has been examined in the effect of reducing disparity

in discipline. School-wide positive behavioral interventions and supports (SWPBIS) is a

framework for delivering whole-school social culture and behavior support needed to improve

educational and social outcomes for all students (Horner & Sugai, 2015). PBIS implementation

utilizes proactive support, rather than relying only on reactive punishment, and has been shown to

reduce the use of exclusionary discipline (Skiba et al., 2014). In a 5-year randomized controlled

study on PBIS implementation in 35 middle schools, it was shown that suspension rates were

reduced for Hispanic/Latino and American Indian/Alaska Native students, but not for African

American students (Sprague et al., 2013). Extending the knowledge of research in this area can

lead to interventions and discipline policy changes that could be made in districts indicating

higher prevalence of unproportionate discipline rates.

Statement of the Problem

Exclusionary discipline (e.g., suspension, expulsion) has been a topic of concern because

of its negative, long-term effects especially among students of color. At first glance, it may seem

like exclusionary discipline is an effective way to improve school culture and increase safety for

students by removing potentially harmful behaviors to the school environment. However, what

often goes unnoticed is that exclusionary discipline occurs at disproportionate rates for students

of color and has long-term harmful effects (e.g., higher dropout rates, school-to-prison-pipeline)

for those students who receive this type of discipline. By rethinking our approach to regulating

managing difficult behaviors students engage in, we can address the increased, disproportional

rates of discipline for students of color. PBIS is a school-wide system that can be used as a

preventative and alternative measure to exclusionary discipline.

Purpose of the Study

This study aims to investigate the current levels of disparity in discipline as well as

examine the effects of PBIS implementation on the use of exclusionary discipline and the

potential gap between ethnicity groups. Firstly, this study will utilize public data on suspension

and expulsion rates across public high schools within Los Angeles county in order to analyze the

current level of disproportionality that exists between different race/ethnicity groups. Secondly,

this study will compare the rates of exclusionary discipline between schools that implement PBIS

and schools that do not implement PBIS. In doing so, this study will present the relationship

between PBIS implementation and the rates of suspension and expulsion use. Lastly, of the

schools that are implementing PBIS, this study will compare the rate of exclusionary discipline

across different levels of fidelity.

Research Questions

1. At what level does exclusionary discipline occur within LA County public high schools

by race?

2. Do rates of exclusionary school discipline within LA County public high schools differ

based on whether PBIS is implemented or not?

3. Do rates of exclusionary discipline differ based on level of PBIS implementation (i.e.,

none, not awarded, bronze, silver, gold, and platinum) within LA County public high

schools?

Significance

The current literature has demonstrated an existence of the discipline gap and the negative

outcomes that result. The literature suggests that these effects demonstrate a need for more

research on practical interventions or educational reform in order to decrease the disproportionate

rates of discipline by race/ethnicity (Bottiani et al., 2017; Mizel et al., 2016; Skiba et al., 2011).

There is limited research on interventions conducted at a school-wide, structural level that focus

on decreasing the discipline gap that is present between non-White and White students. This

study increases the knowledge of literature regarding the current state of the discipline gap

among affecting students of color. Furthermore, this study investigates the potential impact of a

systematic, preventative intervention on the use of punitive, exclusionary discipline. The

intention of this study was to develop knowledge that could support schools who present high

and consistent rates of disparity in discipline. It also aims to inform district and school

administration about the potential benefits of implementing a school-wide intervention targeting

behavior. Given that discipline policy makers, school district officials, and school administration

can influence the type of discipline practices being implemented in the classroom and across

school campuses, this study also aims to provide knowledge for this audience in order for them to

make informed decisions regarding systematic discipline policies and practices.

Theoretical Framework

The overarching theoretical and analytical frameworks that are used to explore the

discipline gap and PBIS include Bronfenbrenner’s ecological systems theory and the Critical

Race Theory. Bronfenbrenner and Morris’s (2007) ecological systems theory were utilized to

examine the possible influences of PBIS Systems on the existing discipline gap. In addition, the

Critical Race Theory was used to explore the influence of race and ethnicity on disparities within

discipline practices and why they may be present.

The ecological systems theory states that development is influenced by various

ecosystems one is exposed to (Bronfenbrenner & Morris, 2007). These ecosystems can include

family or home systems, school systems, and societal and cultural systems. This model states that

these ecosystems interact with each other and influence all aspects of an individual’s life. Five

levels of ecosystems were identified and they include: (1) the microsystem (i.e., immediate

environment), (2) the mesosystem (i.e., connections), (3) the exosystem (i.e., indirect

environment), (4) the macrosystem (i.e., social and cultural values), and (5) the chronosystem

(i.e., changes over time. Apter and Conoley (1984) outlined four fundamental assumptions in the

ecological perspective: (1) individuals are an inseparable part of a system; (2) trouble is not seen

as a disease within a child, instead it reflects trouble within the system; (3) disturbances reflect

incompatibility between the skills and knowledge of a child and their environmental demands;

and (4) interventions should focus on creating the most effective systems. Based on these

assumptions and the ecological systems theory, it can be recognized that student behavior reflects

the interactions between the individual student and their environmental demands. Interventions

that target environmental, or systematic change, might help to influence the presence of student

behavior outcomes in a school system. PBIS is a framework that is meant to target behavior

outcomes through a whole-school and individual approach.

The PBIS system originated as a preventative intervention for students with significant

disabilities who engaged in extreme forms of self-injury or aggression and was as an alternative

to harsher interventions (Carr & Durand, 1985). Although it originated in special education, PBIS

is being implemented at the general education level and is being applied to whole school systems

in order to improve social culture, behavioral outcomes, and school climate. The idea of tackling

behavior and culture at a school system level reflects the ecological perspective and how issues

are system-based as opposed to student-based. The high rates of exclusionary discipline can be

explored through the lens of tackling behavioral issues as an entire school system rather than at

each individual student-level. In addition to the school-level influence on the discipline gap, the

ecological perspective also describes the influence of the macrosystem, or the influence of

societal and cultural values. This idea connects to the Critical Race theory and how disparity in

discipline may reflect the inherent cultural norms that have remained present in U.S. culture for

decades.

The Critical Race Theory (CRT) originated as a movement from scholars and activists

who wanted to further analyze and critique the relationship between law, race, racism, and power

(Delgado & Stefancic, 2017). There are five core components of CRT: (1) racism is ordinary and

a common, everyday experience; (2) the idea of interest convergence; (3) race is a social

construct; (4) the idea of differential racialization; and (5) the dominant group, typically White

individuals, are the true recipients of civil rights legislations. Although the CRT movement

originated in the field of law, it currently extends to many different fields, including education.

The existence of social justice issues within education (e.g., the academic gap, the discipline gap,

or school-to prison pipeline) calls for the need to address the underlying mechanisms the lead to

racial inequality in the first place, and this can be explored through a CRT lens. The CRT sheds

light to the ideas that racial stigmatization, stereotyping, and implicit biases – all influenced by

racial prejudice – may be influencing the objective definitions of appropriate behavior and

therefore lead to how exclusionary discipline and similar punishments are implemented within

the school system (Simson, 2013).

Conclusion

Given the persistent and pervasive presence of systemic racism in our country and the

impact it has on our minority, non-White populations, it is important to have a greater

understanding of its existence in school systems and the possible strategies that can be used to

combat its presence. For almost 50 years, there has been an existence of a discipline gap in our

country. African American students are 2 to 3 times as likely to be removed from their learning

environment as a means of discipline. These discipline strategies are intended to be utilized to

promote school safety and increase positive student climate. Instead, exclusionary discipline

practices are ineffective and harmful towards student achievement, school culture, and

individual, long-term success. This research can increase the understanding of potential

intervention strategies that can help reduce or eliminate the discipline gap. Specifically, PBIS, a

school-wide preventative approach, might be a tool to help change the current outcomes of

discipline practices.

CHAPTER 2

REVIEW OF LITERATURE

In this chapter, the current and relevant literature on the existence and impact of the

discipline gap is discussed. In addition, the implementation of PBIS and its potential impact on

reducing exclusionary discipline is explored. First, this chapter will describe historical and

current information on exclusionary discipline practices and its disproportionate effects on

students of color. Second, evidence on the harmful outcomes of exclusionary discipline will be

described. Last, this chapter will define PBIS and explore the positive impact its implementation

can have on exclusionary discipline practices as well as promising factors that may lead to

greater success in reduction of the discipline practices.

History of Exclusionary Discipline

Exclusionary discipline is described as any discipline practice that removes a student

from their educational setting (American Psychological Association, n.d.). Examples of

exclusionary discipline include office discipline referrals, in- and out-of-school suspensions, and

expulsions. The discipline gap, or discipline disparities, refer to cases in which students from a

specific demographic (e.g., race/ethnicity, sex, disability status) are subjected to discipline at

greater rates than students who belong to other demographic groups.

The increase in the use of exclusionary discipline stems from the rise in popularity of the

“zero tolerance” philosophy. Overtime, the definition of zero tolerance has evolved and does not

have a specific, written definition. However, the concept of zero tolerance initiated in the 1980s

during the war on drugs era in which punishment was implemented for all offenses regardless of

the level of severity (Skiba, 2000). This philosophy was accepted by the education system in the

1990s because educators feared a perceived increase in violence. The zero tolerance policy

became an accepted response to offenses such as gang affiliation and possession or usage of

drugs and weapons. During the Clinton Administration, the Gun Free Schools Act of 1994 was

signed into law and zero tolerance became a national policy. The law mandated that all local

education agencies (LEAs) receiving federal funds were required to implement a 1-year calendar

expulsion on students who were found to have brought a weapon to school (Cerrone, 1999). In

addition, a referral to the criminal or juvenile justice system had to be made. Originally, this law

only covered the possession of firearms, but over time amendments extended the bill to include

any instrument that might be used as a weapon. Furthermore, local school districts included

drugs and alcohol, fighting, threats, or swearing into their zero tolerance policies (Skiba, 2000).

Zero tolerance policies were being implemented in a wide variety of ways such that some

schools might punish for major and minor offenses equally, while others consider the

seriousness of an offense before deciding the severity of the consequence.

In 2006, over a decade after the Gun Free Schools Act of 1994, the American

Psychological Association (APA) began to investigate the effectiveness of zero tolerance. The

APA assigned a Zero Tolerance Task Force to evaluate the effectiveness of the policy. Findings

indicate that, overall, there was no evidence to improve school climate or school safety (APA

Zero Tolerance Task Force, 2008). These policies assumed that removal of students and their

disruptive behavior would deter others from engaging in the same behavior and improve school

climate for students remaining in the schools (Ewing, 2000; Good, 2004). The Zero Tolerance

Task Force (2008) examined data on five key assumptions of zero tolerance policies including (1)

violence in schools is out of control or increasing, (2) zero tolerance has increased the

consistency of school discipline, (3) school climate will be more conducive to learning for

students after the removal of those who violate rules, (4) the swift and certain punishments of

zero tolerance will reduce likelihood of disruption, and (5) parents overwhelmingly support

implementation of zero tolerance policies. There was no evidence to support the first four

assumptions and evidence for the fifth assumptions were inconclusive. Some parents were in

favor of the policies, but others perceived zero tolerance policies to be a threat to their students’

rights to education. Furthermore, the task force found evidence that students of color were

overrepresented in suspension and expulsion and believed the cause was a lack of classroom

management preparation, lack of training in culturally competent practices, or racial stereotypes.

Overall, the APA Zero Tolerance Task Force found that zero tolerance policies, and its

assumptions, were ineffective and led to disproportionate discipline practices for students of

color. Furthermore, the APA Zero Tolerance Task Force (2008) recommended alternatives to the

zero tolerance philosophy, including PBIS which will be discussed later in this chapter.

Although findings from the Zero Tolerance Task Force were published over a decade

ago, the current literature demonstrates a continued existence of the discipline gap across racial

groups (Martin et al., 2018; Mizel et al., 2016; Skiba et al., 2011). The discipline gap is also

found to persist across varying grade levels (Skiba et al., 2011). Skiba and colleagues (2011)

utilized data from a nationally representative sample to examine disparities in discipline. The

sample included 436 schools with a reported total enrollment of 120,148 students in elementary

grades (K-6) and 60,522 students in middle school grades (6-9). Findings indicated African

American students were 2 times as likely, at the elementary level, and 4 times as likely, at the

middle school level, to receive office discipline referrals compared to White students. In the

same study, it was found that Hispanic students were overrepresented in office discipline

referrals at the middle school level. This study demonstrated that the discipline gap affected

students of color from several groups despite the grade level they were in.

In addition to race and ethnicity, other demographic factors might have an effect on the

likelihood of receiving exclusionary discipline. Previous research has also explored the influence

of other demographic factors on rates of exclusionary discipline. A study conducted by Mizel and

colleagues (2016) examined the association of individual and family factors with office referrals,

suspension, and expulsion. A survey on 2,539 10th and 12th grade students demonstrated that

specific family factors (i.e., parental involvement, discussing homework, and engaging in

activities with children) were not related to school punishment (Mizel et al., 2016). Instead, the

role of the family was an important factor on school completion rather than for school discipline.

However, risk factors including marijuana use, exposure to an adult role model using alcohol,

lowered parental monitoring, and lowered beliefs in the importance of grades were significantly

associated with being suspended or expelled from school when students had parents with low

educational backgrounds. Therefore, students with parents of low educational backgrounds

seemed to be at a higher risk of being expelled or suspended (Mizel et al., 2016). In addition to

race and ethnicity, a variety of demographic and risk factors can influence rates of exclusionary

discipline.

In addition to parent education levels, exclusionary discipline practices may be influenced

by the ethnic makeup of a school’s population. A study conducted by Payne and Welch (2010)

explored the effects of racial threat on school discipline practices. The researchers explored how

racial composition in a school influences discipline policies such as punitive and restorative

discipline and zero tolerance. Payne and Welch (2010) utilized data from the National Study of

Delinquency Prevention in Schools (Gottfredson et al., 2000) on 294 public, nonalternative

secondary schools. Their measures included five scales representing different degrees of

disciplinary responses, percent of African American students, level of school crime and disorder,

socioeconomic status (SES), percentage of Hispanic and male students, and other variables

related to school demographics. It was found that schools with a higher percentage of African

American students were more likely to implement punitive responses and less likely to

implement restorative responses. This finding demonstrates that there exists a relationship

between ethnic makeup of schools and the type of discipline policy implemented.

In addition to demographic factors and ethnic school makeup, Gregory and Mosely (2004)

investigated the relationship teachers’ perspectives may have on the reason for discipline

problems that exist. The researchers conducted qualitative interviews with 19 high school

teachers from a large, urban high school with 3,300 students in the year 2000. According to their

discipline data analysis, African American students at this school represented 80% of on-campus

suspensions but only accounted for 37% of the population. Therefore, it was noted that a

discipline gap did exist at the time of the interviews. A large majority of the teachers (80%)

attributed discipline issues to adolescent development and felt that defiance was normal (Gregory

& Mosely, 2004). This theory does not acknowledge the overrepresentation of African American

students receiving suspension and takes on a colorblind approach. The colorblind ideology

believes that by ignoring factors related to race, culture, and ethnicity, and treating everyone as

equally as possible, discrimination will end (Williams, 2011). In addition to a developmental

attribution, the teachers in this study believed discipline were related to low achievement (i.e.,

students acting out due to low achievement), school organization (e.g., class sizes), school culture

(e.g., inconsistent discipline across classrooms), and teacher practices (e.g., poor pedagogy).

Similarly, these reasons do not explain why there is a racial disparity in discipline within the

school (Gregory & Mosely, 2004). Forty-five percent of teachers shared the belief that discipline

issues stemmed from African American students within lower SES homes who brought their

frustrations into school. Gregory and Mosely (2004) observed that teachers were putting the

problem on external factors such as African American students, families, and communities but

continued to ignore the internal culture of the school.

The current literature demonstrates the existence of a discipline gap based on

demographic factors, including race/ethnicity, ethnic makeup of a school, and the colorblind

teacher perspectives (Gregory & Mosely, 2004; Mizel et al., 2016; Payne & Welch, 2010). It is

important to explore the potential outcomes the discipline gap is associated with. Exclusionary

discipline can lead to harmful, long-term effects that negatively impact the lives of students

being removed from the classroom. It is important to note that because of the discipline gap,

students from specific racial/ethnic subgroups are at higher risk of experiencing these effects

compared to others. Some of these negative outcomes include student perceptions in sense of

belonging and equity, the school-to-prison pipeline, and the achievement gap.

Exclusionary discipline may have a negative impact on students who are most likely to be

removed from their learning environment. Bottiani et al. (2017) conducted a study that examined

the discipline gap between African American and White students among 58 high schools. The

participants included 19,726 adolescents from Maryland. Bottiani et al. (2017) conducted a

multilevel, secondary analysis of data from a survey on school climate. First, analysis was

conducted at an individual student level. Factors included perceived equity, school belonging,

adjustment problems and student demographics. Second, data was analyzed at the school level

which included the racial gap in out-of-school suspension, school SES, student diversity, and

condition of school-level intervention. Findings indicated African American students, compared

to White students, were more likely to perceive lower school equity and school belonging, and

perceive higher rates of adjustment issues. It was found that as risk for suspension increases,

African American students’ perception of equity and belonging decrease at significant rates

compared to White students (Bottiani et al., 2017). It can be interpreted that the discipline gap is

associated with a higher likelihood for African American students to have a negative perception

on school climate, school equity and their sense of belonging. Other potentially negative

outcomes from the disproportionate rates of discipline are related to the success of a student’s

future.

Racial discipline disparities can lead to an increased likelihood for students to drop out of

school and get arrested or incarcerated, also known as the school-to-prison pipeline (Wald &

Losen, 2003). Carmichael et al. (2005) conducted a study investigating the school-to-prison

pipeline and found students involved in at least one disciplinary incident were 23.4 times more

likely to be referred to the juvenile justice system. Furthermore, each additional incident of

discipline after that increased the likelihood by 1.5% and each day a student was suspended

increased the likelihood by 0.1% (Carmichael et al., 2005). Findings demonstrate the impact

exclusionary discipline can have on a student’s future success. Fowler (2011) found that African

American students are overrepresented in all categories of discretionary expulsions from public

school. Since African American students are more likely to be punished with exclusionary

discipline, they are also more likely to be subject to future involvement in the juvenile justice

system which can further increase racial disparities within the prison system.

The discipline gap is not only related to disparities in the prison system. Disproportionate

rates of discipline are also related to the achievement gap. Gregory et al. (2010) conducted a

review and found that research demonstrated a correlation between the academic and discipline

gaps. Students who received at least one suspension also missed instructional time and this has

been shown to lead to a pattern of low academic achievement, disengagement with the school

system and increased participation in rule breaking behaviors (Arcia, 2006). Arcia (2006)

followed two demographically similar groups of students who were matched on gender, race,

grade level, SES, and English proficiency. One group of students had received at least one

suspension and the other group did not. In the first year, Arcia (2006) found that students who

had been suspended fell three grade levels behind the non-suspended students, and within three

years, these students were almost five grade levels behind. Findings indicate that exclusionary

discipline practices are related to academic achievement among students. Those who are

suspended may experience lower connections with their school, lower engagement with school

rules or course work, and lower levels of motivation to achieve academically (Gregory et al.,

2010). The relationship between exclusionary discipline and these negative effects can be

reasons why the achievement gap and discipline gap are closely intertwined.

The discipline gap has been shown to be related to negative, individual outcomes.

However, disparities in discipline are also related to negative effects on the social economy.

Marchbanks et al. (2014) investigated the economic costs of increased grade retention and school

dropout as it relates to exclusionary discipline. The researchers followed almost one million

seventh grade students through their 12th-grade year. Grade retentions associated with student

discipline were estimated to cost $76 million per year while student dropout rates associated with

discipline were estimated to cost between $750 million and $1.35 billion over the lifetime of

student cohorts. The research indicates that exclusionary discipline does not only have

detrimental effects to individual students, but also to society and the economy. Therefore, it is

important to understand ways to reduce the use of exclusionary discipline.

The current literature establishes the existence of the racial discipline gap, underlying

causes to its continued existence, and the negative, long-term outcomes related to the discipline

gap. There is a consensus that interventions or educational reform to decrease the

disproportionate rates of discipline by race/ethnicity is needed (Bottiani et al., 2017; Mizel et al.;

2016; Skiba et al., 2011). The following studies describe possible interventions that demonstrate

these outcomes.

Current Literature on Suggested Interventions

The long-standing existence of the discipline gap and its negative, harmful effects call for

a need to reform punitive discipline practices in schools. Bottiani et al (2018) conducted a

literature review and summarized previous studies implementing interventions aimed to combat

disparities in discipline. Two classroom-level interventions were reviewed: the Double Check

and the GREET-STOP-PROMPT. The Double Check intervention was a self-assessment which

asked teachers and school staff to reflect on and address five core components related to

culturally responsive practices which include reflective thinking, authentic relationship,

connection to curriculum, effective communication, and sensitivity to student’s cultural and

situational messages (Hershfeldt et al., 2009). Findings demonstrated that combining classroom

based coaching and professional development in Double Check was related to reductions in

office discipline referrals among African American students. The GREET-STOP-PROMPT

intervention provided proactive classroom behavior management strategies in order to minimize

implicit bias in decision-making related to discipline (Cook et al., 2018). Findings demonstrate

that GREET-STOP-PROMPT is associated with reductions in office disciplinary referrals for

African American, male students.

Furthermore, Bottiani et al. (2018) reviewed school-wide, or system-wide, interventions

that address the discipline gap. A study by Gregory et al. (2018), utilized restorative practices. A

restorative practice approach allows students to face the nature of the problem, or student

behavior from relationally focused perspective that avoids punitive and exclusionary discipline.

Findings indicated the restorative approach benefited all groups of students in decreasing

exclusionary discipline rates. However, the results did not demonstrate better outcomes for

African American students or the discipline gap (Gregory et al., 2018). Another school-wide

approach that has been examined is the use of a threat assessment (Cornell et al., 2018). This is a

systematic process of evaluation and intervention for students who have made threats against

other students, both verbal and physical. Findings indicate that this approach led to lower racial

differences in regard to being suspended or expelled. Cornell et al. (2018) report that the threat

assessment approach can be utilized to reduce the school discipline gap.

The use of classroom- and system-wide level interventions has resulted in positive

outcomes associated with a reduction in office discipline referrals among African American

students or a reduction in suspensions overall (Bottiani et al., 2018). However, in both the

classroom- and system-wide level interventions, Bottiani et al. (2018) revealed challenges in

implementation and progress monitoring of the noted interventions. Firstly, for classroom

interventions, Bottiani et al. (2018) noted concerns related to accurate measurement of teachers’

progression after implicit bias and culturally responsive trainings. These concerns were made

because progress would be measured through teacher self-report. Furthermore, although the

investigated classroom interventions led to reductions in office referrals, these interventions are

not consistent across a school system. Therefore, a system-wide level intervention may have a

greater impact.

Among the system-wide interventions that were explored by Bottiani et al. (2018), the

restorative approach demonstrated an overall reduction in suspensions for all students but did not

address the disproportionate rates of suspension among African American students. The threat

assessment approach did show a specific reduction among suspensions in African American,

male students which is a promising approach to tackling the discipline gap. The threat

assessment approach used a system-wide strategy in which students who demonstrated greater

risk of concerning behaviors were provided an early intervention. Similarly, a PBIS framework

offers a continuum of supports that allow students’ needs to be matched with appropriate

instruction or intervention. In the PBIS approach, students with a higher need will be provided

intensive strategies in order to increase positive behavior.

Positive Behavioral Interventions and Support (PBIS)

PBIS is defined as a framework that incorporates evidence-based practices within a

multi-tiered system of support framework that is aimed at improving student success through the

relationship between academic and social behavior (Sugai & Simonsen, 2012). This framework

emphasizes the relationship between positive school-wide and classroom-wide culture and

individual student success (Sugai & Simonsen, 2012). Sugai and Simonsen (2012) noted that

research in the PBIS framework originated when there was a need to identify and implement

interventions for students with behavior disorders (BD). As a response, the University of Oregon

began conducting studies in the 1980s and, a decade later, the University of Oregon developed

the PBIS Center when provided a grant awarded through the reauthorization of the Individuals

with Disabilities Act of 1997. This grant was created with the intention to assist schools on

evidence-based practices for improving supports for students with BD. Today, PBIS has

developed to support entire school systems rather than a specific student population.

PBIS has many defining characteristics. First, student outcomes (e.g., academic and

social) are used to determine the practice, data collection, and intervention evaluation each

system uses. Second, PBIS emphasizes the use of evidence- or research-based practices to

support students throughout the school-wide system including non-classroom, classroom, and

individual student levels. Third, PBIS follows a continuum of behavior support practices and

systems, including continuous professional development, monitoring based on phase of

implementation, and systems-based competence supports. One of the key principles of PBIS is

their continuum of supports from the least to most intensive. The interventions (e.g., positive

reinforcements) are implemented among all students at a tier 1, or universal, level. For students

who are unresponsive to these supports, a tier 2, or more targeted approach, is in place to prevent

at-risk students from engagement in inappropriate behaviors. Following this, the tier 3, or most

intensive, individualized interventions are implemented for students who need one-on-one

support. Finally, the last defining characteristic of PBIS is that all previous characteristics are

based on data which make the framework completely data driven. For example, data collection

and analysis are considered essential for creating the support framework geared toward academic

and social success.

In addition to classroom- versus system-wide interventions, research has also examined

the implementation of a problem-solving, behavioral approach using school-wide data. McIntosh

et al. (2018), examined a four-step approach (i.e., problem identification, problem analysis, plan

implementation, plan evaluation) in utilizing school discipline data to identify behaviors that are

more susceptible to implicit bias and disproportionate discipline. Through a behavioral lens, the

authors examined school discipline data (i.e., rates of ODRs) and found that using data as a guide

helped reduce the rates of ODRs across all racial groups. Although the data-guided approach did

not reduce the discipline gap, further research can be done to examine the use of this approach by

applying it to specific sub-groups. If data shows specific behaviors are more likely to lead to

discipline among a specific sub-group (e.g., African American students), then a data-guided

approach may be beneficial to implement in order to reduce that behavior and reduce the rates of

discipline. PBIS utilizes a problem-solving approach that is based on data.

Existing research on PBIS indicates some success in reducing overall suspension rates

among students within schools implementing PBIS (Chin et al., 2012; Curtis et al., 2010; Pas et

al., 2019). For example, Chin et al. (2012) conducted a pilot and case study that demonstrated a

correlation between an alternative to suspensions (ATS) model based on a PBIS framework and a

reduction in suspensions at an elementary school level. The ATS model promotes learning and

reduces future incidents of behavioral problems as needed. When a student committed an

offense, the behavioral function was assessed through a debriefing and reflection assignment.

Furthermore, interviews and observations were conducted to evaluate what reinforced students’

behavior. The intervention was then implemented based on the function of the problem behavior.

ATS implemented interventions to nine students from a preschool through 6th grade school. The

participating school served a student population that was 94% Latinx, had 92% of students

eligible for free/reduced-price lunch, and had 79% of students identifying as English language

learners. Data indicated that ATS interventions were related to a decrease in suspensions.

Students’ suspensions were compared to their previous five years of behavioral entries. However,

results were not statistically significant. Of the nine students provided interventions, two

reoffended. The other seven did not. Although the study did not demonstrate statistically

significant results, it showed support for success of an alternative to suspension program in

reducing suspensions and re-offenses.

PBIS can also help reduce behavioral issues which, in turn, can help decrease

instructional days lost through exclusionary discipline. In a 4-year longitudinal study conducted

by Curtis et al. (2010), researchers examined the effects of a PBIS system in a public elementary

school (K-5th grade). The sample consisted of 523 students during the academic school years

between 2003-2004 to 2006-2007. The researchers examined the effect of the PBIS program by

collecting data on behavior referrals to the principal, extended timeouts within the school day,

out-of-school suspensions, and instructional days lost. Curtis et al. (2010) found that there was a

40-67% decrease in behavior referrals and out-of-school suspensions after implementing PBIS.

This research indicates that PBIS might have decreased overall suspensions, extended timeouts,

and instructional days lost (Curtis et al., 2010). Based on this research, an age appropriate PBIS

program at a high school level is promising in decreasing the discipline gap.

Another study examined PBIS at both elementary and secondary school levels. Pas et al.

(2019) examined the effects of Tier 1 (or universal) implementation of PBIS among schools in

the state of Maryland during the academic years of 2006-2007 through 2011-2012. During this

time, the state implemented a “scale-up” program to incorporate SWPBIS in multiple schools

across the state. Pas et al. (2019) examined student outcomes in elementary (i.e., grades K-5, K 6,

K-8) and secondary schools. (i.e., grades 6-8, 9-12) trained in PBIS and compared it to schools

that were not. Findings demonstrate significantly lower suspensions in elementary schools

trained in PBIS as well as significantly lower suspensions in secondary schools. Results indicate

that among elementary schools, the change in discipline did not appear until the 4th and 5th years

of the study. However, in secondary schools, reduced suspensions only occurred during the

second year of the study (Pas et al., 2019). The researchers report that the training in PBIS among

secondary schools may be affected by lower fidelity scores when compared to elementary

schools. These results are promising, but it is important to further investigate implementation at

secondary school levels and the effect fidelity of PBIS implementation may have. The current

study strives to fill in these gaps. This study will explore the implementation of PBIS within

traditional high schools, varying levels of implementation fidelity, and the overall relationship

with exclusionary discipline.

Summary

Students of color facing disproportionate discipline practices have existed for decades and

further escalated during the Zero Tolerance era. Despite the lack of evidence in improving school

climate or safety, zero tolerance policies continue to remain in effect and continue to have long-

lasting, negative effects for students of color, particularly for African American students. The

discipline gap is present across all grade levels from preschool through high school. There are

many underlying causes that influence the continued existence of the discipline gap. These

include students’ demographic factors, ethnic makeup of schools, and “colorblind” teacher

perspectives. The discipline gap can lead to negative, harmful effects on students of color. These

include negative beliefs in sense of belonging, increased risk of the school-to-prison pipeline, and

likelihood of poor academic achievement. Although its existence is well documented, few studies

examine the impact of interventions implemented to combat disproportionate punitive discipline.

Research on classroom-level (e.g., implicit bias and classroom management training) and

school-wide level (e.g., restorative intervention or threat assessment approaches) have been

conducted and results are indicative of promising interventions for tackling the discipline gap.

However, implementation of these interventions were met with challenges such as lack of

objective measures or inconsistency across a school-level system. Therefore, school wide PBIS

is proposed as an alternative to these interventions. PBIS is implemented across a three-tiered

framework, it follows five core characteristics, it requires evidence-based practices, and it is

data-driven. These characteristics along with previous history in improving academic and

behavioral outcomes in schools are reasons why PBIS would be an optimal intervention to

examine. The current study aims to expand on the current literature by investigating the effects

of PBIS at a high school level (grades 9-12) and at different rates of fidelity implementation. In

doing so, this study hopes to understand the influence of PBIS and its fidelity on the reduction of

exclusionary discipline for students of color.

CHAPTER 3

METHODOLOGY

Introduction

The present study aimed to investigate the rates of exclusionary school discipline (i.e.,

suspensions and expulsions) within schools implementing PBIS and compare it to discipline

rates of schools that did not implement PBIS. Additionally, this study investigated the difference

in exclusionary discipline rates in schools implementing PBIS at different levels of fidelity (i.e.,

platinum, gold, silver, bronze, and N/A). The overall rates of suspension and expulsion were

investigated. The following research questions were examined:

1. At what level does exclusionary discipline occur within LA County public high schools

by race?

2. Do rates of exclusionary school discipline within LA County public high schools differ

based on whether PBIS is implemented or not?

3. Do rates of exclusionary discipline differ based on level of PBIS implementation (i.e.,

none, not awarded, bronze, silver, gold, and platinum) within LA County public high

schools?

Definitions/Terminology

Discipline gap: The existence of higher and disproportionate rates of discipline for

students of color compared to White students.

Positive behavioral intervention and supports: A school wide framework of integrated

behavioral support that offers a continuum support across tiers.

Procedures/Design

This study used a non-experimental, nonequivalent group design. A non-experimental

research design is a study that examines the relationship between two variables by comparing

different group outcomes but does not manipulate the independent variable (Gravetter &

Forzano, 2016). In this particular study, the predictor variables were the race/ethnicity of

students, the implementation of PBIS, and the level of fidelity PBIS is implemented. It is not

ethical to assign one school PBIS and fail to provide the same treatment to another school.

Therefore, the predictor variables were being measured as they would naturally occur. A

nonequivalent group design is a study in which the researcher is unable to use random

assignment to place participants in different groups or conditions (Gravetter & Forzano, 2016). In

this study, the high schools that implemented PBIS were not randomly assigned by the

researcher. Additionally, the level of fidelity implemented was not assigned by the researcher.

Therefore, this study used a nonequivalent group design. This study utilized existing, non

identifiable data obtained from public data sources described below. Therefore, this study did not

meet the definition of research with human subjects and there was no need to go through IRB

approval.

School Sample

This study consisted of a sample of 265 public, traditional high schools within Los

Angeles county during the 2018-2019 academic school year. A traditional high school was

defined as a school that served students from 9th through 12th grades. Schools that participated in

this study met the following criteria in the 2018-2019 academic school year: (a) submitted data

to DataQuest on the California Department of Education (CDE) website; (b) reported school’s

race and ethnicity to DataQuest; (c) reported school’s suspension and expulsion to DataQuest,

(d) not a private or charter high school, (e) a public high school with students in 9th through 12th

grade only. After schools were filtered using the above criteria, an additional 24 schools were

removed from the dataset because grade enrollment data was missing or schools had less than

one percent population in one or more of their grade levels (e.g., 0% of students were in 9th and

10th grade). If this data were included, it may have led to differing results and may have skewed

data analysis results. After accounting for these criteria, a total of 241 public high school data

were analyzed for this study. The high schools within Los Angeles county were categorized into

subgroups of high schools that implemented PBIS and high schools that did not implement PBIS

in the 2018-2019 academic school year.

Data/Instruments

Secondary data were utilized to examine exclusionary discipline rates, PBIS

implementation, and fidelity of implementation. The data included school demographic

information that was publicly available from the CDE website. The CDE website releases

publicly available data on state, county, district, school, and SELPA demographics through their

DataQuest site (https://dq.cde.ca.gov/dataquest/dataquest.asp). DataQuest is the CDE’s web

based data reporting system for publicly reporting information about students, teachers, and

schools in California. DataQuest allows public access to a variety of reports, including student

enrollment, English learner, and student misconduct data. Reports generated from the DataQuest

web source provided information regarding annual grade level data, suspension, and expulsion

data based on ethnicity/race. This data helped to inform possible disparities in exclusionary

discipline by race.

School level data was obtained on DataQuest and included annual enrollment,

suspension, and expulsion data. Data was downloaded on a school-by-school basis for high

schools during the 2018-19 academic year. The following datasets were downloaded:

1. Percentage of enrollment by grade

2. Percentage of enrollment by English Language Acquisition Status (ELAS) for

English Learner, Reclassified Fluent English Proficient, and to be determined status

students

3. Suspension Rate – disaggregated by Ethnicity

4. Expulsion Rate – disaggregated by Ethnicity

From these datasets, the suspension and expulsion data included other background

information such as cumulative school enrollment, total enrollment by ethnicity, and percentage

enrollment by ethnicity. After downloading these data, the following information was combined

into one dataset per school: cumulative enrollment, enrollment total by ethnicity, enrollment

percentage by ethnicity, overall suspension rate, suspension totals by ethnicity, suspension

percentage by ethnicity, suspension rate by ethnicity, overall expulsion rate, expulsion total by

ethnicity, expulsion rate by ethnicity, percentage enrollment by English language learner status,

and percentage enrollment by grade level. Each school was assigned with an identification

number from 001 to 275. Next, the public data from California PBIS Coalition’s (CPC’s) PBIS

Tiered Fidelity Inventory (further described below) was added to the dataset. If schools

participated in PBIS, they were indicated as a PBIS school and their level of fidelity was

assigned based on CPC’s award for the 2018-2019 school year.

The California PBIS Coalition (CPC) is an organization that utilizes evidence-based,

culturally relevant practices to support stakeholders in the implementation of PBIS. Under a

multi-tiered systems framework, CPC follows the National PBIS Blueprints for professional

development, implementation, and evaluation. The purpose of CPC is to create a network for

stakeholders and to support those who wish to implement PBIS with a standard of practice. In

collaboration with the CPC, school-level data on the PBIS Tiered Fidelity Inventory (TFI) Scores

were utilized for this study. This dataset provides valid, reliable, and efficient measures of the

extent to which school personnel were applying the core features of School-wide PBIS.

The TFI is divided into three tiers (i.e., Tier 1: Universal SWPBIS Features, Tier 2:

Targeted SWPBIS Features, and Tier 3: Intensive SWPBIS Features; Algozzine et al., 2019). Tier

1 implementation includes universal, SWPBIS such as a school organizational chart explicitly

stating expected school behavior, staff and student handbook for behavioral expectations and

teaching expectations, and clearly defined school procedure or policy for addressing office-

managed vs. staff-managed problems. These features are geared towards implementation for the

entire school population. Tier 2 implementation includes targeted features of SWPBIS such as

decision rules for identification of students who require Tier 2 support, tier 2 behavior support

interventions and progress monitoring of interventions, and staff handbook for how to refer

students to tier 2 intervention. These intervention features are designed to be implemented for the

at-risk students who are not responding to the tier 1 interventions. Lastly, tier 3 implementation

includes intensive SWPBIS features such as student support teams for individual students

receiving tier 3 intervention, adequate staffing assigned to facilitate individualized plans for

students in tier 3, and behavior support plans. These intervention features are implemented with

students who are unresponsive to both tier 1 and tier 2 interventions.

The CPC awarded schools based on their level of fidelity for overall implementation of

PBIS based on the TFI score earned at each tier. A school could earn a maximum of 30 points

across 15 features (e.g., behavioral expectations). Each feature was rated 0, 1 or 2 depending on

the school’s ability to meet scoring criteria where 0 = Not Implemented, 1 = Partially

Implemented, and 2 = Fully Implemented. Each feature also has unique criteria per category

(e.g., 1 = Behavioral expectations identified but may not include a matrix or be posted). A school

can earn a maximum of 26 points across 13 categories for tier 2 and 34 points across 17

categories for tier 3. To find the TFI score for each tier, the total points possible was divided by

the total points awarded and multiplied by 100 to create a percentage. For example, if a school

earned 20 points on tier 1, their tier 1 TFI score would be 66% (i.e., 20/30 x 100 = 66%). The

CPC awarded schools a bronze, silver, gold, or platinum fidelity ranking based on the TFI score

earned at specific tiers. A school was awarded bronze if their tier 1 score was 40% or higher.

Schools were awarded silver if their tier 1 score was 70% or higher. Schools were awarded gold

if they met the criteria for silver and obtained a score of 70% or higher for either tier 2 or tier 3.

Schools were awarded platinum if they obtained a score of 70% or higher at all three tiers. There

were some schools that were Not Awarded (i.e., N/A) for several different reasons including

inability to confirm TFI scores, missing office discipline referral data, missing action plan

documentation, etc.

Data Analysis

Demographic data, exclusionary discipline rates, and PBIS TFI scores were gathered for

statistical analysis. IBM Statistical Package for Social Sciences (SPSS) for Macintosh, Version

28.0 was utilized as the statistical program for the analysis of this study’s data.

To answer research question 1, descriptive analyses were run on suspension and

expulsion rates between race/ethnicities reported to CDE for the 2018-2019 academic school

year. For this particular study, public data indicated the rates of suspension and expulsion by

race/ethnicity for each school and descriptive statistics were used to indicate any existing

patterns. However, assumptions four through six were not met. Outliers were present in the data

and the data did not present a normal distribution or shape. Therefore, non-parametric tests were

used to run the analyses. A non-parametric test is one that makes fewer assumptions by ranking

the data in order to overcome issues with normality and eliminate the effect of outliers

To answer research question 2, the researcher intended to run an independent-samples t-

test. However, the data included outliers and did not represent a normal distribution or shape.

Therefore, the data did not meet all assumptions to run an independent-samples t-test. Due to the

voided assumptions, an independent-samples Mann-Whitney U test was utilized to compare

overall school suspension and expulsion rates, between schools implementing PBIS and schools

not implementing PBIS. The Mann-Whitney U test is the non-parametric equivalent of the

independent samples t-test which ranks data in order to account for the unmet assumptions

(Fields, 2013). In this study, the overall suspension and expulsion rates of schools implementing

PBIS were compared to schools that did not implement PBIS.

To answer research question 3, the researcher intended to run a one-way analysis of

variance (ANOVA) to compare the means of overall exclusionary discipline rates at different

levels of fidelity for PBIS. However, the dataset had outlier and did not present a normal

distribution or shape so it did not meet the criteria to run an ANOVA. The Kruskal-Wallis test

was used to compare overall suspension and expulsion rates between schools implementing

PBIS at four different levels of fidelity (i.e., gold, silver, bronze, N/A). The Kruskal-Wallis test

is the non-parametric equivalent for a One-Way ANOVA and ranks data in order to account

for the unmet assumptions (Field, 2013).

Validity and Reliability

This study utilized the PBIS Tiered Fidelity Inventory (TFI) to measure the fidelity

implementation of PBIS in schools. Algozzine et al. (2014) developed the TFI as a

comprehensive assessment measuring fidelity in multiple areas. It measured fidelity as a single

score across three tiers, as individual scores for each tier, and as scores for 10 subscales across

the three tiers. Massar et al. (2019) tested the internal consistency and content validity of TFI

assessing school wide PBIS implementation with existing fidelity measures. The findings

conducted by Massar et al. (2019) support the use of TFI to measure fidelity of overall

implementation, implementation at each tier level, and implementation for each subscale.

Limitations

This study utilized secondary data collected from the CDE website’s Dataquest. This data

is publicly available, therefore there are some concerns about privacy of students and the CDE

website suppressed the data in regard to ethnicity in order to maintain privacy for individual

students. Several categories (i.e., total enrollment, total suspensions, total expulsions, percentage

of enrollment, percentage of suspensions, percentage of expulsions, suspension, and expulsion

rates by ethnicity) were marked with an asterisk (*) in order to protect student privacy. If the

cumulative enrollment for a selected student population was 10 or less, the data was suppressed.

Additionally, if there were one or more ethnicity groups suppressed, any other ethnicity report

“not reported” was also suppressed (https://dq.cde.ca.gov/dataquest/). For this study, in the cases

where schools reported zero total suspensions and expulsions, the researcher noted zero total

suspensions and expulsions, 0% suspended and expelled, and zero suspension and expulsion rates

for all ethnicities. Among schools that reported at least one or more total suspensions and

expulsions, the researcher left the suppressed data (i.e., the asterisk) as is. Due to data

suppression, the data within this study might not accurately reflect the total, percentage, or rates

of exclusionary discipline in the 2018-2019 academic school year.

This study analyzed the data within traditional, public high schools in Los Angeles

county. Data from non-traditional schools (e.g., charter and special education), non-public

schools, and schools with lower levels (e.g., K-8th grade, K-12th grade) were omitted. The

following analysis may only have generalizability for traditional public high schools and may not

reflect accurate relationships between PBIS implementation and discipline rates presented at

these other types of schools.

Of the high schools participating in this study, only 41 schools were reported to

implement PBIS. The sample size between the two groups is unequal and may affect the results

of the data analyses. However, other studies have run analyses on the effect of PBIS

implementation with similar sample sizes including a study with 15 high schools (Freeman et al.,

2013) and a study with 37 elementary schools (Bradshaw et al., 2012). Both studies compare the

effects of PBIS implementation on total student population among all schools. The current study

follows a similar procedure and examines the effect of overall suspension and expulsion on each

schools’ student population. The average cumulative enrollment per school was 1,162 students.

CHAPTER 4

RESULTS

The purpose of this study was to further understand and analyze the significant rates of

disproportionality in exclusionary discipline practices and its relationship to school-wide

interventions, such as PBIS. This study also investigated differences between the fidelity

implemented in the interventions. Quantitative analyses were conducted to investigate the

implementation of PBIS in LA County High Schools and the relationship implementation and

level of fidelity had on exclusionary discipline rates. The following research questions were

addressed:

1. At what level does exclusionary discipline occur within LA County public high schools

by race?

2. Do rates of exclusionary school discipline within LA County public high schools differ

based on whether PBIS is implemented or not?

3. Do ratees of exclusionary discipline differ based on level of PBIS implementation (i.e.,

none, not awarded, bronze, silver, gold, and platinum) within LA County public high

schools?

Descriptive statistics were used to answer research question 1, an independent sample t test

was used to answer research question 2, and a one-way analysis of variance was used to

answer research question 3.

Sample Descriptive Statistics

Descriptive analyses indicated that, of the 241 high schools included in the present study,

the average cumulative enrollment was 1,162.7 (SD = 980.9) students (see Table 1). The lowest

number of students enrolled at a school was 28 and the greatest number was 4,807. Of the 241

schools, 240 had students identifying as Hispanic or Latinx, 171 had African American students,

149 had White students, 111 had Asian students, 106 had Filipino students, 85 had students

identifying with more than one race, and less than 20 schools had students identifying as Pacific

Islander or American Indian/Alaska Native. The ethnicities with the highest average enrollment

per school were Hispanic/Latinx (M = 735.5, SD = 70.8), White (M = 265.5, SD = 15.4) and

Asian (M = 249.6, SD = 13.3). Descriptive analyses also indicated that, per school, the average

percentage of students enrolled was about the same across each grade level. On average, 21.2%

of students were enrolled in the 9th grade, 24.3% in 10th grade, 26% in 11th grade, and 28.5%

were in 12th grade.

TABLE 1. Descriptive Statistics: Enrollment, Race, Grade Level, and English Learner

Status

Variable N %

Race/ethnicity Mean Enrollment

African American (n = 171) 95.93 10.0

American Indian/Alaska Native (n = 6) 10.67 0.4

Asian (n = 111) 249.59 13.3

Filipino (n = 106) 65.92 4.3

Hispanic/Latinx (n = 240) 735.50 70.8

Pacific Islander (n = 17) 21.63 2.0

White (n = 149) 265.51 15.4

Multiple Race (n = 85) 73.46 4.6

Grade Level

Grade 9 21.2

Grade 10 24.3

Grade 11 26.0

Grade 12 28.5

English Learner Status

English Learner 13.0

Reclassified Fluent English Proficient 37.3

To be determined 0.1

Exclusionary Discipline Rates by Ethnicity

Research question 1 sought to understand the current levels of disparity in exclusionary

discipline rates by ethnicity. Suspension and expulsion rates were obtained from existing school

reports and descriptive statistical analyses were run to determine the mean differences in

exclusionary discipline rates (see Table 2). The data demonstrated that African American

students (M = 6.54, SD = 9.36) were almost twice as likely to be suspended compared to students

identifying as Hispanic/Latinx (M = 3.56, SD = 7.81), White (M = 3.53, SD = 6.50), and with

more than one race (M = 2.53, SD = 6.50). In addition, students who identified as African

American were estimated to be 3 times as likely to be suspended compared to American

Indian/Alaska Native (M = 2.11, SD = 5.48) and Pacific Islander students (M = 1.83, SD = 3.82).

Furthermore, Asian (M = 1.61, SD = 7.96) and Filipino (M = 0.72, SD = 1.42) students were the

least likely to be suspended compared to other ethnicities. Overall, African American students

were more likely to be suspended in LA County public high schools compared to their peers.

TABLE 2. Exclusionary Discipline Rates by Ethnicity

Variable M SD

Suspension Rates

African American (n = 181) 6.54 9.36

American Indian/Alaska Native (n = 28) 2.11 5.48

Asian (n = 131) 1.61 7.96

Filipino (n = 125) 0.72 1.42

Hispanic/Latinx (n = 239) 3.56 7.81

Pacific Islander (n = 40) 1.83 3.82

White (n = 160) 3.53 6.50

Multiple Race (n = 82) 2.53 4.51

Expulsion Rates

African American (n = 196) 0.36 2.85

American Indian/Alaska Native (n = 80) 0.00 0.00

Asian (n = 154) 0.26 2.60

Filipino (n = 155) 0.00 0.00

Hispanic/Latinx (n = 241) 0.09 0.30

Pacific Islander (n = 89) 0.00 0.00

White (n = 185) 0.10 0.53

Multiple Race (n = 128) 0.06 0.37

The data also demonstrated that African American students (M = 0.36, SD = 2.85) were

more than 3 times as likely to be expelled compared to students identifying as American

Indian/Alaska Native (M = 0.00, SD = 0.00), Filipino (M = 0.00, SD = 0.00), Pacific Islander (M

= 0.00, SD = 0.00), Hispanic/Latinx (M = 0.09, SD = 0.30), White (M = 0.09, SD = 0.53), and

more than one race (M = 0.06, SD = 0.37). In addition, students identifying as Asian (M = 0.26,

SD = 2.60) were more than twice as likely to be expelled compared to American Indian/Alaska

Native, Filipino, Pacific Islander, Hispanic/Latinx, White, and multiple race students. Overall,

African American students were most likely to be expelled, followed by Asian students. African

American students also had the highest rates of both suspension and expulsion.

Assumptions

Prior to running statistical analyses, the data were checked to see if assumptions were met

for an independent t-test and a one-way ANOVA. For both analyses, six assumptions were tested

and only the first three assumptions were met. These three assumptions were that the dependent

variables were measured on a continuous scale, the predictor variables included categorical,

independent group, and the participants (i.e., the high schools) could not be in both groups.

However, assumptions four through six were not met. Outliers were present in the data and the

data did not present a normal distribution or shape. Therefore, non-parametric tests were used to

run the analyses. A non-parametric test is one that makes fewer assumptions by ranking the data

in order to overcome issues with normality and eliminate the effect of outliers (Field, 2013).

PBIS Implementation

Research question 2 investigated whether or not the implementation of PBIS was

significantly related to differences in exclusionary discipline rates in LA County Public High

schools. Data on the overall suspension and expulsion rates and data from the PBIS coalition was

used to answer the second research question. Schools were categorized as PBIS or non-PBIS.

Descriptive statistics related to type of school and exclusionary discipline is summarized in Table

3. There was a total of 200 non-PBIS schools and 41 PBIS schools being analyzed. Overall

suspension rate was reported through the CDE DataQuest portal. Suspension rate was the

quotient of unduplicated count of student suspensions and cumulative enrollment multiplied by

100. For example, Alhambra High school had 2,506 students enrolled during the 2018-19 school

year and 15 unduplicated counts of students suspended. The suspension rate ([15/2,506]*100)

was 0.6%. Similarly, the overall expulsion rate was reported through the CDE DataQuest Portal.

Expulsion rate can be determined by finding the quotient of unduplicated count of students

expelled and cumulative enrollment and multiplying by 100.

TABLE 3. Descriptive Statistics for PBIS Implementation on Exclusionary Discipline

PBIS Implementation N Min Max M SD

Non-PBIS School

Expulsion Rate 200 0.00 2.84 0.07 0.27

Suspension Rate 200 0.00 34.80 2.67 4.92

PBIS School

Expulsion Rate 41 0.00 0.98 0.12 0.22

Suspension Rate 41 0.00 60.30 6.82 9.98

Of the schools observed, 200 did not implement PBIS. Table 4 illustrates the skewness

and kurtosis of the data. When addressing research question 2, the difference between the

median suspension and expulsion rates were used. The median was used because the data

presented outlier, was positively skewed, and did not meet normality assumptions. The Mann

Whitney non-parametric test was used in order to account for the unmet assumptions. The

Independent-Samples Mann-Whitney U test is the non-parametric equivalent of the independent

t-test (Fields, 2013). The results are summarized in Table 5. Results indicate that suspension

rates in PBIS schools (Mdn = 5.10) differed significantly from non-PBIS schools (Mdn = 0.80),

U = 6,389.5, z = 5.675, p < .001, r = 3.66. Results also indicate that expulsion rates in PBIS

schools (Mdn = 0.00) differed significantly from non-PBIS schools (Mdn = 0.00), U = 5,142.0, z

= 3.228, p = .001, r = 0.21. Schools that implemented PBIS had higher rates of suspension and

expulsion compared to schools that did not implement PBIS.

TABLE 4. Median, Skewness and Kurtosis of PBIS Implementation on Exclusionary

Discipline

PBIS Implementation N Mdn Range Skewness Kurtosis

Non-PBIS School

Expulsion Rate 200 0.00 2.84 7.30 64.36

Suspension Rate 200 0.80 34.80 3.83 17.76

PBIS School

Expulsion Rate 41 0.00 0.98 0.12 0.22

Suspension Rate 41 0.00 60.30 6.82 9.98

TABLE 5. Independent-Samples Mann-Whitney U Test for PBIS Implementation and Exclusionary Discipline

Test Statistics Suspension Rates Expulsion Rates

Total N 241 241

Mann-Whitney U 6389.50** 5142.00**

Test Statistic 6389.50** 5142.00**

Standard Error 403.46 322.78

Standardized Test Statistic 5.68 3.23

Asymptotic Sig. (2-sided test) <.001 0.001

Note. *p < .05, **p <. 01

PBIS Fidelity

Research question 3 examined whether the level of fidelity of PBIS schools was

significantly related to differences in exclusionary discipline rates in LA County Public High

schools. The same suspension and expulsion rate data used for research question 2 was utilized

for research question 3. Schools that implemented PBIS were awarded one of five awards based

on their level of fidelity. The awards were Not Awarded (N/A), Bronze, Silver, and Gold.

Schools that were not implementing PBIS were categorized as “None.” Table 6 depicts the

descriptive statistics for PBIS fidelity. Of the 41 schools that implemented PBIS, 14 were not

awarded, 7 were awarded bronze, 14 silver, 4 gold and 2 platinum.

TABLE 6. Descriptive Statistics for PBIS Fidelity on Exclusionary Discipline

PBIS Implementation N Min Max M SD

None

Suspension Rate 200 0.00 34.80 2.68 4.92

Expulsion Rate 200 0.00 2.84 0.07 0.27

N/A = Not Awarded

Suspension Rate 14 0.90 6.00 3.50 2.06

Expulsion Rate 14 0.00 0.96 0.07 2.56

Bronze

Suspension Rate 7 1.80 9.90 5.84 2.79

Expulsion Rate 7 0.00 0.42 0.14 0.15

Silver

Suspension Rate 14 0.00 60.30 12.00 15.84

Expulsion Rate 14 0.00 0.98 0.12 0.26

Gold

Suspension Rate 4 2.00 5.50 3.98 1.75

Expulsion Rate 4 0.14 0.25 0.20 0.06

Platinum

Suspension Rate 2 1.10 3.30 2.20 1.56

Expulsion Rate 2 0.10 0.20 0.15 0.07

Similar to research question 2, the suspension and expulsion data presented outliers, was

positively skewed, and did not meet normality assumptions. Therefore, the Kruskal-Wallis non-

parametric test was used in order to account for the unmet assumptions. The Kruskal-Wallis test

is utilized to assess for differences between groups of scores from multiple independent groups

from different populations, and when the assumptions have been violated (Fields, 2013). The

results are summarized in Table 7. Results indicated that suspension rates were not significantly

affected by the level of fidelity in PBIS implementation, H(4) = 8.53, p = .074. Alternatively,

results indicated that expulsion rate was significantly affected by the level of fidelity in PBIS

implementation, H(4) = 12.70, p = .013. To follow-up, pairwise comparisons were run and the

adjusted p-values were used for assessing data. Pairwise comparisons are depicted in Table 8.

The adjusted p-values showed that there were no significant differences between schools not

awarded compared to schools awarded silver (p = 1.00, r = .23), bronze (p = .25, r = .35), and

platinum (p = .57, r = .29). Additionally, there were no significant differences between schools

awarded silver compared to schools awarded bronze (p = 1.00, r = .17), platinum (p = 1.00, r = -

.18), and gold (p = .41, r = -.32). There were also no significant differences between schools

awarded bronze when compared to schools awarded platinum (p = 1.00, r = -.08) and gold (p =

1.00, r = -.16). Furthermore, schools that were awarded platinum were not significantly different

from schools that were awarded gold (p = 1.00, r = .05). However, results indicate that there was

a significant difference between schools that were not awarded and schools that were awarded

gold (p = .027, r = .47).

TABLE 7. The Kruskal-Wallis test of PBIS Fidelity Implementation and Exclusionary Discipline Rates

Test Statistics Suspension Rates Expulsion Rates

Total N 41 41

Test Statistic 8.53 12.70

Degrees of Freedom 4 4

Asymptotic Sig. (2-sided test) .074 .013

Note. *p < .05, **p <. 01

General Discussion

The purpose of this research was to determine the relationship PBIS implementation has

on exclusionary discipline rates in schools. Based on the results of this analysis, it can be

concluded that students identifying as African American are more likely to be suspended and

expelled compared to their peers. Additionally, Filipino and Asian students are less likely to be

suspended compared to their peers, and Filipino, Native American/Alaska Native, and Pacific

Islander students are all less likely to be expelled compared to their peers.

Furthermore, this data demonstrated that schools implementing PBIS were more likely to

have higher suspension and expulsion rates when compared to schools that did not implement

PBIS. In addition, data indicated that the level of fidelity in implementing PBIS was related to a

significant difference in expulsion rates for schools not awarded for fidelity when compared to

schools awarded a gold-level of fidelity. Lastly, suspension rate was not significantly affected by

the level of fidelity implemented at each school. Implications of results will be discussed in the

context of some significant limitations in Chapter 5.

CHAPTER 5

DISCUSSION

The purpose of this study was to extend the literature on PBIS implementation and its

impact on exclusionary discipline among high school students. The use of punitive, exclusionary

discipline is a reactive method of addressing student behavior and has impacted students of color

at greater rates. However, PBIS is a preventative method of addressing student behavior and

research indicates promising results in using PBIS to reduce exclusionary discipline practices.

When implemented with fidelity, PBIS can result in greater effects on improving school climate

and reducing behavior-related discipline. It was hypothesized that the discipline gap continues to

exist and that implementation of PBIS at the high school level could be related to reduction in

the use of exclusionary discipline practices. Additionally, it was hypothesized that when PBIS

was implemented with greater fidelity, there would be lower rates of exclusionary practices

compared to schools which implemented PBIS with lower fidelity. The following research

questions were addressed in the present study:

1. At what level does exclusionary discipline occur within LA County public high schools

by race?

2. Do rates of exclusionary school discipline within LA County public high schools differ

based on whether PBIS is implemented or not?

3. Do ratees of exclusionary discipline differ based on level of PBIS implementation (i.e.,

none, not awarded, bronze, silver, gold, and platinum) within LA County public high

schools?

Chapter 5 will present and interpret the findings of the data analysis conducted to address

each research question. Additionally, this chapter will provide recommendations for policy,

practice, and future research.

Existing data gathered through CDE’s DataQuest and the CPC were used for data

analysis. It was hypothesized that data would indicate continued existence of a discipline gap

within high schools. Additionally, it was hypothesized that high schools implementing PBIS

would also have fewer exclusionary discipline rates compared to high schools that were not.

Lastly, it was hypothesized that schools implementing PBIS with greater rates of fidelity would

have reduced exclusionary discipline rates compared to those with lower rates of fidelity. The

results of the data analyses revealed expected and unexpected findings across the three central

research questions of this study.

Levels of Exclusionary Discipline by Ethnicity

There is an extensive amount of literature on the existence of the discipline gap across

grade levels ranging from preschool through high school (Bottiani et al., 2018; Fowler, 2011;

Gopalan & Nelson, 2016). The first research questions asked: at what level does exclusionary

discipline occur within LA County public high schools by race? The purpose of the first

research question was to re-establish the continued existence of the discipline gap among public

high schools in Los Angeles. When analyzing suspension rates by ethnicity, the current study

found the discipline gap continued to be true. Descriptive statistics (see Table 2) indicated that

African American students were twice as likely to be suspended compared to students

identifying as Hispanic/Latinx, White, or multiracial. Furthermore, these students were 3 times

as likely to be suspended compared to American Indian/Alaska Native and Pacific Islander

students. Similarly, when analyzing expulsion rates by ethnicity, it was found that African

American students were more than 3 times as likely to be expelled compared to students

identifying as American Indian/Alaska Native, Filipino, Pacific Islander, Hispanic/Latinx, and

multiracial. In addition, Asian students were more than twice as likely to be expelled compared

to American Indian/Alaska Native, Filipino, Pacific Islander, Hispanic/Latinx, White and

multiracial students. Overall, among LA County public high schools, African American students

had the highest rates of exclusionary discipline and were more likely to be suspended or

expelled, followed by Asian students being the second most likely to be expelled.

These data indicate a continued existence of the discipline gap. Similar to previously

conducted research (Skiba et al., 2014; Sprague, 2018), these findings re-establish the consistent

overrepresentation of African American students receiving exclusionary discipline. These

findings also indicate some concern for students identifying as Asian. The first research question

asked the levels of exclusionary discipline by ethnicity and findings match the researcher’s

hypothesis in that the discipline gap continues to exist.

Research indicates students of color, specifically African American students, were more

likely to face exclusionary discipline due to behaviors described as defiant or disruptive

(Fenning & Rose, 2007). The lack of objectivity or consensus on how to measure defiant or

disruptive behaviors can lead to the influence of implicit biases when labeling a behavior as

defiant or disruptive. These biases can be indicative of the discipline gap that continues to exist

among high schools in California. Additionally, of the 50 states, 32 continue to allow

suspension or expulsion for defiant or disruptive behavior, including California (Rafa, 2018).

Data analyses in the current study could be explained by the continued use of punitive discipline

and the non-objective reasons for this discipline practice.

Implementation of PBIS and Exclusionary Discipline Rates

After establishing the continued existence of the discipline gap, this study aimed to

provide evidence to support the implementation of PBIS in order to reduce the discipline gap and

the continued use of exclusionary discipline. The second research question asked: do rates of

exclusionary school discipline within LA County public high schools differ based on whether

PBIS is implemented or not? The results of the data analyses indicate that suspension and

expulsion rates were significantly higher in schools that implemented PBIS compared to the

schools that did not implement PBIS. These findings did not match the researcher’s hypotheses.

It should be noted that this analysis was done through a non-parametric statistical testing due to a

low population of schools implementing PBIS. This study compared the exclusionary discipline

rates of a group of 41 schools implementing PBIS to a group of 200 schools not implementing

PBIS. The non-parametric analysis was conducted in order to compensate for the unmet

assumptions in the unequal groups, lack of normality and presence of outliers in the data. The

results of this data analysis may be different if similar comparison groups are acquired.

In addition to an unequal group size, the high schools implementing PBIS may have been

at different stages of the implementation process. For example, some schools may be in their first

year of implementing the model while others may be well-versed in the framework of PBIS. In

the study conducted by Pas et al. (2019), secondary schools trained in school wide PBIS only

demonstrated reduced suspensions in the second year of the study compared to elementary

schools who demonstrated a change in discipline during the fourth and fifth years of the study.

The findings indicate that the implementation of PBIS is an on-going process that takes time

before seeing results in school climate or discipline rates. Additionally, Pas et al. (2019)

hypothesized that lower fidelity scores could be a cause due to the lack of change in discipline

practices. The current findings may be affected by similar factors such as a lack of foundation for

implementation of PBIS in its schools or low fidelity scores for implementation.

Fidelity of PBIS Implementation and Exclusionary Discipline Rates

In order to further investigate the relationship between PBIS and exclusionary discipline,

this study aimed to address the differences in fidelity of PBIS implementation. As described, Pas

et al. (2019) hypothesized a lack of change in discipline at the high school level was due to lower

fidelity scores compared to the elementary level. The third research questions asked: do rates of

exclusionary discipline differ based on level of PBIS implementation (i.e., none, not awarded,

bronze, silver, gold, and platinum) within LA County public high schools? This study considered

differences in fidelity as awarded by the CPC. The results of the analysis indicate that suspension

rates were not significantly affected by differing levels of fidelity in PBIS implementation, but

expulsion rates were. Specifically, schools awarded a gold level of fidelity had higher rates of

expulsion compared to schools who were not awarded. All other groups of fidelity did not

demonstrate significant differences when compared. Again, it is important to note that of the 41

schools implementing PBIS, only 14 were awarded silver, 7 bronze, 4 gold, 2 platinum, and 14

were not provided an award. Therefore, results may not be indicative of the relationship between

exclusionary discipline rates and varying fidelity levels of PBIS implementation. The sample

sizes of schools awarded in each category was smaller than the researcher anticipated.

Additionally, schools identified as “N/A” or not awarded may not have been given an award for

various reasons including failure to report exclusionary discipline rates. Therefore, these schools

may appear to have lower rates of exclusionary discipline compared to the gold level fidelity

because they may not have reported their discipline referrals to the CPC prior to the deadline for

awarding.

Limitations

There were notable limitations to this study. The first is related to sample size. This study

was mostly limited by the sample size of only 41 high schools implementing PBIS. Additionally,

when examining research question 3, this small sample size was broken down into even smaller

groups in order to compare levels of fidelity. The researcher conducted this studying utilizing

existing, public data. There was no control over assigning high schools to implement PBIS which

could have provided the researcher with a greater sample size. Additionally, the researcher chose

to examine public high schools located in Los Angeles County which limits the potential for a

larger sample size.

Second, the findings of this study are limited by the selection of the sample chosen. The

researcher excluded non-traditional high schools (e.g., private and charter schools, K-12th grade

schools, schools that serve 11th through 12th grade only) which could potentially exclude smaller

sized schools that had the freedom to implement PBIS with high fidelity. These schools were

excluded to create a more comparable group and to ensure that school type and school level were

not confounding factors to the analysis. In addition to excluding schools, this study examined

schools fully implementing PBIS in the 2018-19 academic school year and excludes schools who

may have started implementation in the middle of the year or in the academic school years after.

Furthermore, the data that was examined included data from existing, public sources. Therefore,

if exclusionary discipline rates or PBIS implementation and fidelity were not reported or

recorded through these data sources then those schools were automatically excluded from this

study.

Third, there is a lack of previous research studies on the relationship between PBIS

implementation and fidelity of implementation on suspension and expulsion rates in high

schools. Previous research on PBIS is typically conducted at an elementary or middle school

level and measures exclusionary discipline through office referrals (McNiell et al., 2016; Pas et

al., 2019). Therefore, the foundational research for the questions addressed in this study was

limited.

Fourth, the level of fidelity for PBIS examines the overall TFI score for PBIS

implemented at all three tiers. PBIS is a proactive method for intervening student behavior.

Therefore, PBIS is strongly implemented at the general student population level, or tier one, and

for those students at-risk, or in tier two due to the preventative nature of the framework. Students

in tier three are typically receiving intervention through behavior plans or counseling services

and data on the fidelity of PBIS implementation may not be recorded. In the current data

collected from CPC, 34 of the 41 schools had scores of 0 for TFI Tier 3 scores. Therefore, the

overall fidelity score may not be representative of implementation fidelity for schools. It would

be beneficial to analyze fidelity of PBIS implementation based on each tier’s TFI score rather

than a combined TFI score for all three tiers.

Fifth, the suspension and expulsion rate data that was obtained contained significant

outliers that affected the normality of its distribution. Due to outliers, an alternative, non

parametric statistical analyses were conducted. These analyses included the outliers within the

data set which may have significantly impacted the results of the data. However, the non

parametric analyses were utilized in order to reduce the impact of this limitations and reduce the

risk of affect in testing with the unmet assumptions.

Implications and Recommendations

The overall focus of this study was reducing suspension and expulsion rates through

PBIS implementation. The purpose of this study is to further examine the relationship between

implementation of PBIS and the rates of exclusionary discipline present in Los Angeles public

high schools. The study’s research questions were created to gather data on implementation and

implementation fidelity. The results of this study can be used for recommendations in policy,

practice, and future research.

Policy

Findings from this study indicate that the existence of the discipline gap is still present

among multiple groups of students of color. Therefore, districts should work to implement

alternatives to exclusionary discipline in order to reduce potential harm that is disproportionately

affecting students of color. Although this study only examined PBIS implementation, there are

several approaches to discipline that have been proven to reduce the need for exclusionary

discipline. A combination of restorative justice programs, PBIS and SEL has been shown to

reduce exclusionary discipline practices by having a preventative program in place while also

having methods that target behavior change rather than exclusion (Mergler et al., 2014).

Another recommendation is to modify CDE’s DataQuest to make data easily available to

explore topics such as this. DataQuest provides the public with a wide range of information on

schools from state test scores to discipline data. However, the downloadable datasets did not

include individual school-level data for exclusionary discipline rates. Instead, the researcher had

to look up each high school individually and input exclusionary discipline rates manually. This

may hinder other researchers from exploring this research in the future. The researcher was also

unable to gather other demographics related such as school climate rating. It would be beneficial

for DataQuest to make these data readily available to download.

One limitation this study encountered was the lack of consistency in the timing of

awarding fidelity levels of PBIS to schools. The lack of consistency opened the potential for

comparing fidelity among schools at different stages of system implementation. It may be

beneficial to determine the award of fidelity at a specified time (e.g., end of first term).

Additionally, it may be helpful to include the stage of implementation in the award dataset.

Either of these options can help future researchers compare schools in similar stages of

implementation. For example, schools implementing PBIS for the first year can be compared

within a group.

Practice

Results of the study indicate a need for change in discipline approaches for high school

students. High school staff should analyze their school discipline data to determine whether their

discipline approach is contributing to the continued existence of the discipline gap. Data can be

used to determine the likelihood a specific group of students will face suspension or expulsion

compared to other groups of students. Each schools’ student population and discipline trends will

vary and high school administration should work to reduce disproportionality in discipline among

student groups. As mentioned, there are several alternative discipline approaches that have a

strong research foundation in the reduction of exclusionary discipline, including PBIS, restorative

justice programs and peer mediation (McNiell et al., 2016). Peer mediation is an approach that

can be implemented immediately where as PBIS and restorative justice require training and time

to develop and implement. By having multiple alternative approaches, high schools can reduce

the need for suspension and expulsion.

Future Research

As previously indicated, there is a lack of research on the implementation and

implementation fidelity of PBIS and suspension or expulsion data at the high school level. Future

research should replicate this study with a larger sample size in order to obtain greater accuracy

on the relationship between PBIS and high school suspension and expulsion reductions.

Additionally, it would be beneficial to observe the effects of implementation of PBIS over-time.

As mentioned, Pas et al. (2019) noted that it took at least four years of PBIS implementation at

the elementary school level before there were changes present in school discipline policies. By

observing multiple schools over multiple school years, future studies can provide information on

the effect of PBIS implementation and the speed at which it will affect discipline rates.

Findings from this study indicate that students identifying as Asian were second-most

likely to be expelled following Black/African American students. This is an unexpected finding

that differs from the established current literature. One recommendation for future research is to

follow-up on this finding and further investigate potential reasons. One contributing factor may

be that the Asian demographic represents a diverse range of individuals who are typically

underrepresented (e.g., South Asians, Southeast Asians). Future research should consider

exploring exclusionary discipline among smaller, more specific demographic groups to bring

awareness to potential disparities.

Lastly, future research should explore the level of implementation of PBIS at each

individual tier level. This study utilized a combined PBIS score to define implementation of

fidelity. However, the schools included in this data had scores of 0 at their tier 3 level. This

indicates that data was not collected or implementation was not observed for students in the tier

three level. To expand on this study’s findings, future research can analyze the implementation

score awarded at specific tier levels (e.g., tier 1 or tier 2), and compare them to levels of

exclusionary discipline.

In summary, this study aimed to examine the existence of the discipline gap, the

relationship PBIS implementation has with rates of exclusionary discipline, and the relationship

implementation fidelity has with rates of exclusionary discipline in LA County public high

schools. This study revealed that the discipline gap continues to exist among students identifying

as African American. These students were two to 3 times more likely to be suspended compared

to their peers. In addition, African American students were more than 3 times as likely to be

expelled compared to other students and Asian students were twice as likely. Findings also

indicate that suspension and expulsion rates differed in schools implementing PBIS such that

these schools had higher rates of exclusionary discipline compared to schools that did not

implement PBIS. Lastly, results indicate that expulsion rates differed depending on the level of

implementation fidelity of PBIS. While there existed limitations in data availability and data

assumptions, the study was a worthwhile examination of the relationship of the implementation

and fidelity of PBIS and exclusionary discipline practices among LA County public high schools.

The study justifies the need for further implementation of PBIS as well as further research on this

topic.

REFERENCES

American Psychological Association. (n.d.). The pathway from exclusionary discipline to the school to prison pipeline. https://www.apa.org/advocacy/health-disparities/discipline-

facts.pdf

American Psychological Association Zero Tolerance Task Force (2008). Are zero tolerance

policies effective in the schools?: An evidentiary review and recommendations. American

Psychologist, 63(9), 852-862. https://doi.org/ 10.1037/0003-066X.63.9.852

Arcia, E. (2006). Achievement and enrollment status of suspended students: Outcomes in a large,

multicultural school district. Education and Urban Society, 38(3), 359-369.

Bronfenbrenner, U., & Morris, P. A. (2007). The bioecological model of human development.

John Wiley & Sons.

Brown v. Board of Education, 347 U.S. 483 (1954). ttps://www.oyez.org/cases/1940-

1955/347us483

Burchinal, M., McCartney, K., Steinberg, L., Crosnoe, R., Friedman, S. L., McLoyd, V., &

Pianta, R. (2011). Examining the black-white achievement gap among low-income

children using the nichd study of early child care and youth development. Child

Development, 82(5), 1404–1420. https://doi.org/10.1111/j.1467-8624.2011.01620.x

Bottiani, J. H., Bradshaw, C. P., & Gregory, A. (2018). Nudging the gap: Introduction to the

special issue “Closing in on Discipline Disproportionality.” School Psychology Review,

47(2), 109–117. https://doi.org/10.17105/SPR-2018-0023.V47-2

Bottiani, J. H., Bradshaw, C. P., & Mendelson, T. (2017). A multilevel examination of racial

disparities in high school discipline: Black and White adolescents’ perceived equity,

school belonging, and adjustment problems. Journal of Educational Psychology, 109(4),

532–545. https://doi.org/10.1037/edu0000155

Carmichael, D., Whitten, G., & Voloudakis, M. (2005). Study of minority over-

representation in the Texas juvenile justice system. Public Policy Research

Institute, Texas A & M University.

Carr, E. G. & Durand, C. M. (1985). Reducing behavior problems through functional

communication training. Journal of Applied Behavioral Analysis, 18, 111-

126.

Cerrone, K. M. (1999). The Gun-free Schools Act of 1994: Zero tolerance takes aim at

procedural due process. Pace Law Review, 20, 131-188.

Children's Defense Fund. 1975. School suspensions: Are they helping children? Washington

Research Project.

Chin, J. K., Dowdy, E., Jimerson, S. R., & Rime, W. J. (2012). Alternatives to suspensions:

Rationale and recommendations. Journal of School Violence, 11(2), 156-173.

https://doi.org/10.1080/15388220.2012.652912

Christle, C. A., Jolivette, K., & Nelson, C. M. (2005). Breaking the school to prison pipeline:

Identifying school risk and protective factors for youth delinquency. Exceptionality,

13(2), 69-88.

Civil Rights Data Collection. (2021). An overview of exclusionary discipline practices in public

schools for the 2017-18 school year [PowerPoint slides]. U.S. Department of Education,

Office for Civil Rights. https://ocrdata.ed.gov/assets/downloads/crdc-exclusionary

school-discipline.pdf

Cook, C. R., Duong, M. T., McIntosh, K., Fiat, A. E., Larson, M., Pullmann, M. D., & McGinnis,

J. (2018). Addressing discipline disparities for black male students: Linking malleable

root causes to feasible and effective practices. School Psychology Review, 47(2), 135-152.

Cornell, D., Maeng, J. L., Burnette, A. G., Jia, Y., Huang, F., Konold, T., Datta, P., Malone, M.,

& Meyer, P. (2018). Student threat assessment as a standard school safety practice:

Results from a statewide implementation study. School Psychology Quarterly, 33(2), 213.

https://doi.org/10.1037/spq0000220

Curtis, R., Van Horne, J. W., Robertson, P., & Karvonen, M. (2010). Outcomes of a school-wide

positive behavioral support program. Professional School Counseling, 13(3),

2156759X1001300303. https://doi.org/10.1177/2156759X1001300303

Delgado, R., & Stefancic, J. (2017). Critical race theory. New York University Press.

Ewing, C. P. (2000). Sensible zero tolerance protects students. Harvard Education Letter, 16(1).

Fenning, P., & Rose, J. (2007). Overrepresentation of African American students in exclusionary

discipline the role of school policy. Urban Education, 42(6), 536-559.

https://doi.org/10.1177/0042085907305039

Field, A. (2013). Discovering statistics using IBM SPSS statistics. Sage.

Fowler, D. (2011). School discipline feeds the “pipeline to prison.” The Phi Delta Kappan, 93(2),

14–19.

Gagnon, D. J., Jaffee, E. M., & Kennedy, R. (2016). Exclusionary discipline highest in New

Hampshire’s urban schools. (Regional Issue Brief No. 46).

https://scholars.unh.edu/carsey/267/

Good, C. (2004). Public agenda. https://files.eric.ed.gov/fulltext/ED485312.pdf

Gravetter, F. J., & Forzano, L. A. B. (2018). Research methods for the behavioral sciences.

Cengage learning.

Gopalan, M., & Nelson, A. A. (2019). Understanding the racial discipline gap in schools.

AERA Open, 5(2), 2332858419844613. https://doi.org/10.1177/2332858419844613

Gottfredson, G. D., Gottfredson, D. C., Czeh, E. R., Cantor, D., Crosse, S. B., & Hantman, I.

(2000). National study of delinquency prevention in schools. Final Report.

https://www.ojp.gov/pdffiles1/nij/grants/194116.pdf

Gregory, A., Clawson, K., Davis, A., & Gerewitz, J. (2016). The promise of restorative

practices to transform teacher-student relationships and achieve equity in school

discipline. Journal of Educational and Psychological Consultation, 26(4), 325-353.

https://doi.org/10.1080/10474412.2014.929950

Gregory, A., & Mosely, P. M. (2004). The discipline gap: Teachers' views on the over

representation of African American students in the discipline system. Equity &

Excellence in Education, 37(1), 18-30.

Gregory, A., Skiba, R. J., & Noguera, P. A. (2010). The achievement gap and the discipline

gap: two sides of the same coin? Educational Researcher, 39(1), 59–68.

https://doi.org/10.3102/0013189X09357621

Gun Free Schools Act 1994, 20 USC § 7961 (1999). https://uscode.house.gov/view.xhtml?re

q=granuleid:USC-prelim-title20-section7961&num=0&edition=prelim

Hershfeldt, P. A., Sechrest, R., Pell, K. L., Rosenberg, M. S., Bradshaw, C. P., & Leaf, P. J.

(2009). Double-Check: A framework of cultural responsiveness applied to classroom

behavior. Teaching Exceptional Children Plus, 6(2), 2-18.

Horner, R. H., & Sugai, G. (2015). School-wide PBIS: An example of applied behavior

analysis implemented at a scale of social importance. Behavior Analysis in Practice,

8(1), 80–85. https://doi.org/10.1007/s40617-015-0045-4

Marchbanks III, M. P., Blake, J. J., Smith, D., Seibert, A. L., & Carmichael, D. (2014). More

than a drop in the bucket: The social and economic costs of dropouts and grade

retentions associated with exclusionary discipline. Journal of Applied Research on

Children, 5(2), 17.

Martin, J. L., Sharp-Grier, M., & Smith, J. B. (2018). Alternate realities: Racially disparate

discipline in classrooms and schools and its effects on black and brown students.

Leadership and Research in Education: The Journal of the Ohio Counsil of Professors of

Educational Administration, 3(1), 16-33.

Massar, M. M., McIntosh, K., & Mercer, S. H. (2019). Factor validation of a fidelity of

implementation measure for social behavior systems. Remedial and Special

Education, 40(1), 16-24. https://doi.org/10.1177/0741932517736515

McIntosh, K., Ellwood, K., McCall, L., & Girvan, E. J. (2018). Using discipline data to enhance

equity in school discipline. Intervention in School and Clinic, 53(3), 146–152.

https://doi.org/10.1177/1053451217702130

McNeill, K. F., Friedman, B. D., & Chavez, C. (2016). Keep them so you can teach them:

Alternatives to exclusionary discipline. International Public Health Journal, 8(2), 169-

181.

Mergler, M. S., Vargas, K. M., & Caldwell, C. (2014). Alternative discipline can benefit

learning. Phi Delta Kappan, 96(2), 25-30. https://doi.org/10.1177/0031721714553406

Mizel, M. L., Miles, J. N.V., Pedersen, E. R., Tucker, J.S., Ewing, B. A., & D'Amico, E. J.

(2016). To educate or to incarcerate: Factors in disproportionality in school discipline.

Child Youth Services Review, 70, 102-111.

Osher, D. M., Poirier, J. M., Jarjoura, G. R., Brown, R., & Kendziora, K. (2013). Avoid Simple

Solutions and Quick Fixes: Lessons Learned from a Comprehensive Districtwide

Approach to Improving Student Behavior and School Safety: (534832013-001) [Data set].

American Psychological Association. https://doi.org/10.1037/e534832013-001

Pas, E. T., Johnson, S. R., Debnam, K. J., Hulleman, C. S., & Bradshaw, C. P. (2019).

Examining the relative utility of PBIS implementation fidelity scores in relation

to student outcomes. Remedial and Special Education, 40(1), 6-15.

https://doi.org/10.1177/0741932518805192

Payne, A. A. & Welch, K. (2010). Modeling the effects of racial threat on punitive and

restorative school discipline practices. Criminology, 48(4), 1019-1062.

Rafa, A. (2018). Alternative School Discipline Strategies. Policy Snapshot. Education

Commission of the States. https://files.eric.ed.gov/fulltext/ED581578.pdf

Raffaele Mendez, L. M., Koff, H. M., Ferron, J. M. (2002). School demographic variables and

out-of-school suspension rates: A quantitative and qualitative analysis of a large,

ethnically diverse school district. Educational Practices and Problems, 39(3), 259-277.

Rocque, M. (2010). Office discipline and student behavior: Does race matter?. American Journal

of Education, 116, 557-581.

Skiba, R. J. (2000). Zero tolerance, zero evidence: An analysis of school disciplinary practice.

(Policy Research Report SRS2): Indiana Education Policy Center.

Skiba, R. J., Arredondo, M. I., & Williams, N. T. (2014). More than a metaphor: The

contribution of exclusionary discipline to a school-to-prison pipeline. Equity &

Excellence in Education, 47(4), 546–564.

https://doi.org/10.1080/10665684.2014.958965

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 discipline. School Psychology Review, 40(1), 24.

Simson, D. (2013). Exclusion, punishment, racism, and our schools: A critical race theory

perspective on school discipline. UCLA L. Rev, 61, 506.

Sprague, J. R., Scheuermann, B., Wang, E., Nelson, C. M., Jolivette, K., & Vincent, C. (2013).

Adopting and adapting PBIS for secure juvenile justice settings: Lessons learned.

Education and Treatment of Children, 36(3), 121-134.

Sugai, G., & Simonsen, B. (2012). Positive behavioral interventions and supports: History,

defining features, and misconceptions. https://assets-global.website-

files.com/5d3725188825e071f1670246/5d82be96e8178d30ae613263_pbis_revisited

_june19r_2012.pdf

Suh, S., Suh, J., & Houston, I. (2007). Predictors of categorical at-risk high school dropouts.

Journal of Counseling and Development, 85(2), 196-203.

Sullivan, A. L., & Bal, A. (2013). Disproportionality in special education: Effects of individual

and school variables on disability risk. Exceptional Children, 79(4), 475–494.

https://doi.org/10.1177/001440291307900406

United States Government Accountability Office. (2018, March 22). K-12 education: Discipline

disparities for black students, boys, and students with disabilities.

https://www.gao.gov/products/gao-18-258

Wald, J., & Losen, D. J. (2003). Defining and redirecting a school‐to‐prison pipeline. New

Directions for Youth Development, 2003(99), 9-15.

Williams, M. T. (2011). Colorblind ideology is a form of racism: A colorblind approach allows

us to deny uncomfortable cultural differences. Psychology Today.

https://www.psychologytoday.com/blog/culturallyspeaking/201112/colorblind-ideology

is-form-racism

ProQuest Number:

INFORMATION TO ALL USERS The quality and completeness of this reproduction is dependent on the quality

and completeness of the copy made available to ProQuest.

Distributed by ProQuest LLC ( ). Copyright of the Dissertation is held by the Author unless otherwise noted.

This work may be used in accordance with the terms of the Creative Commons license or other rights statement, as indicated in the copyright statement or in the metadata

associated with this work. Unless otherwise specified in the copyright statement or the metadata, all rights are reserved by the copyright holder.

This work is protected against unauthorized copying under Title 17, United States Code and other applicable copyright laws.

Microform Edition where available © ProQuest LLC. No reproduction or digitization of the Microform Edition is authorized without permission of ProQuest LLC.

ProQuest LLC 789 East Eisenhower Parkway

P.O. Box 1346 Ann Arbor, MI 48106 - 1346 USA

29066821

2022