951940

research-article2020
PBIXXX10.1177/1098300720951940Journal of Positive Behavior InterventionsGreen et al.

Article
Journal of Positive Behavior

Examining Commitment to Prevention,

Interventions
2021, Vol. 23(3) 137­–148
© Hammill Institute on Disabilities 2020
Equity, and Meaningful Engagement: Article reuse guidelines:
sagepub.com/journals-permissions

A Review of School District Discipline DOI: 10.1177/1098300720951940

https://doi.org/10.1177/1098300720951940
jpbi.sagepub.com

Policies

Ambra L. Green, PhD1 , Heather Hatton, PhD2,

Sondra M. Stegenga, PhD3, Bert Eliason, EdD4,
and Rhonda N. T. Nese, PhD4

Abstract
Although there is increasing awareness of policy decisions contributing to disproportionality in exclusionary practices, few
studies have empirically examined common elements of discipline policies across the nation. We utilized a methodological
review and the Checklist for Analyzing Discipline Policies and Procedures for Equity (CADPPE) to examine the extent
to which current policies reflect recommendations from research regarding best practices for encouraging appropriate
behaviors and preventing undesired behaviors, as well as correlations between those policies and exclusionary disciplinary
outcomes for all students of color and students of color with disabilities. Data came from 147 district discipline policies
and disciplinary outcomes (i.e., suspension and expulsion) from all 50 U.S. states and the District of Columbia. The analyses
indicated the majority of policies do not include most of the research-based recommendations for preventing the overuse
of exclusionary practices. Furthermore, there was no correlation found between CADPPE ratings and the risk ratios for
exclusionary discipline for students of color and students of color with disabilities. Implications for policy development and
implementation and limitations are provided.

Keywords
public policy, law/legal issues, behavior(s), positive behavior, support(s)

The evidence detailing the disproportionate use of exclusion- district discipline policies and explore correlations between
ary practices (e.g., suspension and expulsion) with students policies and the use of exclusionary discipline.
of color (i.e., students from any non-White racial or ethnic
group), particularly Black students, has been documented
extensively in the literature (Skiba et al., 2013). Due to the
Background and Context
disparities in access to instruction, academic achievement, In 1978, the Safe School Study was conducted by the
and discipline outcomes between White and Black students National Institute for Education in response to a request
(U.S. Department of Education, Office for Civil Rights, from Congress for information regarding the extent of crime
2018), the issue of school disciplinary practices has garnered or violence in schools and how these actions could be pre-
the attention of educators, administrators, researchers, state vented. Findings from the report indicated that among other
and federal education agencies, and educational organiza- practices, administrators who instituted “a firm, fair, and
tions. While focus around this issue has mainly included the
need to use alternative practices, such as positive behavioral 1
The University of Texas at Arlington, USA
interventions and supports (PBIS) or restorative practices, 2
University of Missouri, Columbia, USA
less attention has been paid to addressing these issues through 3
The University of Utah, Salt Lake City, USA
4
policy development. Due to the inequitable receipt of exclu- University of Oregon, Eugene, USA
sionary practices by students of color and lack of proven Corresponding Author:
effectiveness of such punitive practices, district discipline Ambra L. Green, The University of Texas at Arlington, 701 Planetarium
policies or codes of conduct, especially those that include Place, Arlington, TX 76019, USA.
exclusionary practices, warrant evaluation and revision. The Email: [email protected]
purpose of this study was to examine commonalities between Action Editor: Maureen Conroy
138 Journal of Positive Behavior Interventions 23(3)

consistent system of discipline” was the difference between to a discipline gap, in which students of color and students
safe schools and violent schools (National Institute of with disabilities receive harsher and more frequent suspen-
Education, 1978, p. iv). The Safe School Study report sions and expulsions than their peers from dominant cultural
resulted in the largest shift toward the consistent use of and socioeconomic environments (Kennedy-Lewis, 2014).
written policies on discipline and codes of conduct within The focus of this study was on the use of out-of-school sus-
schools and districts (National Institute of Education, 1978). pensions and expulsions, as these practices have been found
Swiftly following the report, the National School Resource to yield the most negative of long-term student outcomes,
Network (1980) published a set of guidelines on developing including academic failure, dropout, substance abuse, and
effective codes of conduct for discipline. These early efforts incarceration (Noltemeyer et al., 2015; Skiba et al., 2013).
made clear that discipline codes of conduct were to be used Furthermore, disparities between the rates of exclusionary
as a positive way of providing clear guidelines for behavior practices received by students of color and their White
that would likely result in educational and rehabilitative peers have been strongly documented. For example,
practices, the consistent and equitable application of rules, researchers have found that school administrators often
and safe schools (Fenning & Bohanon, 2006). apply zero-tolerance expulsion policies disproportionately
After a series of shootings and killings in U.S. schools, for males, students of color from low-income environ-
federal legislators passed the Gun-Free Schools Act (GFSA) ments, and individuals receiving special education ser-
of 1994 that mandated school districts develop and adopt vices (Losen & Skiba, 2010).
zero-tolerance gun policies to remove firearms from schools
(Mongan & Walker, 2012). The passing of the GFSA served
as an expansive approach to increase social control in
Suspension
U.S. schools and included two main objectives: (a) reduce Disparities in rates of suspension can be found among stu-
possession of weapons on school campuses and (b) reduce dents in K–12 schools. Risk ratios are commonly used to
school violence and violence at school-sponsored events measure and summarize disparity data. Risk ratios compare
(Mongan & Walker, 2012). Any violation of the policy the proportion of students in a reference group who receive
resulted in mandatory expulsion for the student for up to 1 a specified exclusionary practice (i.e., in-school suspension
year unless the expulsion would violate the Individuals with [ISS], out-of-school suspension [OSS], and expulsion),
Disabilities Education Act (IDEA) or if the local education compared with the proportion of a comparison group who
agency modified the requirement for extenuating circum- also receive the specified exclusionary practice (Albrecht
stances. The adoption of zero-tolerance gun policies assisted et al., 2012). A risk ratio greater than 1.0 communicates an
other temporal trends (e.g., get tough on crime and mass overrepresentation of the reference group, while a risk ratio
shootings) in the development of discipline policies (also less than 1.0 communicates an underrepresentation of the
termed codes of conduct) that continued to validate the reference group for the specified practice. Findings from
removal of students for a vast array of behaviors. the Civil Rights Data Collection survey indicate Black
After more than a decade of focusing on school shoot- students are 3.8 times more likely to receive one or more
ings, the search continued for effective methods to curb OSS compared with their White peers (U.S. Department of
other undesired behaviors found in schools. U.S. schools Education, Office for Civil Rights, 2016). Other male stu-
began adopting zero-tolerance policies for infractions such dents of color (i.e., Alaska Native, Latinx, Native Hawaiian
as fighting, drug or alcohol use, gang activity, possessing or other Pacific Islander, and multiracial) and Native
over the counter medications, disrespect of authority, sex- American students combined represent 15% of the popula-
ual harassment, verbal threats, vandalism, and other behav- tion but reflect 19% of students who received OSS. These
iors considered to threaten the functioning of the school disparities continue for students of color with disabilities.
environment and the learning of other students (Skiba & Students with disabilities represent about 12% of the
Peterson, 2000). However, zero-tolerance policies are the- K–12 population but are more than 2.0 times as likely to
oretically unsound and exacerbate the overrepresentation receive one or more OSS as students without disabilities
of students of color and students with disabilities receiving (U.S. Department of Education, Office for Civil Rights,
exclusionary practices for minor and subjective behaviors 2016). More than one of five Native American or Alaska
unrelated to weapons or drugs (Losen, 2013; Mongan & Native (23%), Native Hawaiian or other Pacific Islander
Walker, 2012; Skiba & Peterson, 2000). (23%), Black (25%), and multiracial (27%) males with dis-
Although definitions vary, general consensus posits abilities received one or more OSS, compared with one of
exclusionary practices are methods of varying levels of 10 White males with disabilities (U.S. Department of
intensity (e.g., time-out to expulsion) used by schools and Education, Office for Civil Rights, 2016). About 20% of
districts to remove a student from their least restrictive envi- multiracial females with disabilities received OSS, com-
ronment (e.g., classroom or school) after having exhibited pared with 5% of White females with disabilities (U.S.
undesired behavior. Exclusionary practices have contributed Department of Education, Office for Civil Rights, 2016).
Green et al. 139

Expulsion (i.e., laws and policies that require expulsion for a given
offense) were more common in districts serving high per-
Students of color also receive a disproportionate number of centages of students of color (Curran, 2017). Disparities in
expulsions from school. According to the Civil Rights Data practices such as these have led to federal regulation
Collection survey (2016), when compared with their White changes that include practices aimed at alleviating the dis-
peers, Black students receive 1.9 times more expulsions parity gaps in discipline, such as the use of PBIS.
from school and are 2.2 times more likely to receive a refer-
ral to law enforcement or be subject to school-related arrests
(U.S. Department of Education, Office for Civil Rights, Federal Regulations: The Call for PBIS
2016). Due to reports of consistent disparities and negative out-
The application of exclusionary practices further exacer- comes for students of color, the U.S. Department of Education
bates the negative outcomes experienced in schools by stu- (ED) provided several guidance documents to assist local
dents of color and students with disabilities (e.g., increased education agencies (LEAs) in mitigating discipline gaps. In
risk of poor academic achievement due to time away from 2009 with the start of the Obama Administration, the Civil
the learning environment, more likely for school dropout; Rights Data Collection survey expanded its variables to
Losen, 2018; Skiba et al., 2011). Furthermore, data suggest include disaggregated discipline data by race and ethnicity,
that discipline disparities across racial groups may be sex, IDEA disability categories, and other subgroups. As a
reflective of differential use of policies by school personnel result, findings of inequitable discipline practices between
(Skiba et al., 2011). students of color and their White peers emerged, leading to
several documents (e.g., “Dear Colleague Letters”) from ED
addressing these disparities (U.S. Department of Education,
Previous Exclusionary Policy Research Office for Civil Rights, 2016). “Dear Colleague Letters”
While an abundance of research has focused on alternatives provided guidance to LEAs on how to address and reduce
to exclusionary practices to mitigate the discipline gap disciplinary disparities, which included the revision of disci-
between students of color and their White peers (Gregory pline policies. The Dear Colleague Letters strongly urged
et al., 2016; Smolkowski et al., 2016), less research has districts to abandon zero tolerance and exclusionary disci-
explored the relationship between discipline policies and pline policies in codes of conduct (Lhamon & Gupta, 2014)
the likelihood of receiving exclusionary practices. Hoffman and replace them with preventive and instructional alterna-
(2014) conducted one of the first studies examining the tives. In addition, the letters provided direction on how
causal impact of expanding discipline policies to include schools could manage discipline equitably and recommend
zero-tolerance policies on racial disparities in school disci- the use of multitiered behavioral frameworks such as PBIS to
pline practices among secondary students in one mid-size improve school climate, culture, safety, and achievement for
urban district. Results indicated that recommendations for all students.
expulsion were worsened under the expansion of the dis- While previous research has analyzed school disciplin-
trict’s zero-tolerance policy. For example, Black students ary policies and codes of conduct (Curran, 2016, 2017;
comprised approximately 75% of the increase in recom- Hoffman, 2014), this study extends the research base by (a)
mendations for expulsion under the expanded zero-­tolerance reviewing district codes of conduct across the United States
policy despite representing just under 25% of the secondary and (b) directly aligning the analysis of codes of conduct to
students in the district. Additional findings indicated that empirical and evidence-based practices used within the
Black students lost an additional 700 days of instruction PBIS framework. This is important given the aforemen-
after the expansion of zero-tolerance policies. Similar tioned links between systems-level supports and long-term
results were found by Curran (2016). Using state data from sustainability and scale-up (Saldana et al., 2012). Ultimately,
all 50 U.S. states, Curran (2016) examined the effect of if policies are not founded on evidence-based practices,
state-level zero-tolerance laws on rates of suspension. such as the use of PBIS frameworks, there is decreased
Findings from the study indicated state zero-tolerance poli- chance of sustainability and long-term impact.
cies were predictive of a 0.5% increase in district suspen- The purpose of this exploratory study was to examine
sion rates as well as larger increases in suspension rates for commonalities between district discipline policies and
Black students compared with White students. In another explore the relationship between policies and the use of
study, Curran (2017) compared explicit zero-tolerance laws exclusionary discipline. The following research questions
and policies (i.e., laws and policies explicitly termed “zero were addressed:
tolerance”) across federal and state law, district policy, and
media interpretation. Findings demonstrated that explicit Research Question 1 (RQ1): What common elements
zero-tolerance laws and policies appeared in less than 14% exist in district discipline policies across the 50 U.S.
of states or districts. Mandatory expulsion laws and policies states and the District of Columbia?
140 Journal of Positive Behavior Interventions 23(3)

Table 1. Enrollment by Race/Ethnicity for All Students and Students With Disabilities.

African American Native Two or more

Descriptives White Hispanic American Asian Indian Hawaiian races/ethnicities
All students
Total 2,754,361.49 2,523,497.39 2,108,936.61 510,619.76 68,639.24 40,389.85 311,422.47
M 19,261.27 17,646.83 14,747.81 3,570.77 479.99 282.45 2,177.78
Median 13,089.60 5,411.61 4,152.75 1,005.77 226.76 66.32 1,040.88
SD 17,437.97 45,237.84 23,899.32 7,299.81 696.39 742.79 3,549.92
Min. 408.07 37.37 44.74 29.15 0.00 0.00 0.00
Max. 91,434.27 426,613.57 157,485.53 59,825.58 4,088.50 6,001.31 27,201.37
Students with disabilities
Total 333,270.04 308,148.00 296,319.44 26,601.66 24,456.13 8,928.45 26,016.35
M 2,330.56 2,154.88 2,072.16 186.03 171.02 62.44 181.93
Median 1,591.82 652.75 620.89 24.51 0.00 0.00 55.16
SD 2,130.64 5,235.00 3,334.99 471.34 1,486.59 676.31 308.56
Min. 7.53 0.00 0.00 0.00 0.00 0.00 0.00
Max. 10,953.51 48,858.58 24,055.64 3,899.23 17,758.00 8,087.30 2,214.87

Note. All statistics calculated from data retrieved from the 2013–2014 Civil Rights Data Collection survey. New York State school districts are
excluded from the sample. Min. = minimum value; Max. = maximum value.

Research Question 2 (RQ2): What is the relationship Notes related to the usability of the CADPPE were taken to
between themes found in a district’s discipline policy make revisions in the next phase. The third policy was used
and the risk ratio of ISS, OSS, and expulsion for students to qualify (i.e., required to meet at least 80% reliability with
of color with disabilities compared with White students the first author) research team members as coders. In Phase
with disabilities? 5, we made the indicated revisions from Phase 4 to the
Research Question 3 (RQ3): What is the relationship CADPPE and analyzed each policy with the finalized
between themes found in a district’s discipline policy CADPPE measure in electronic form. In phase six, the sec-
and the risk ratio of ISS, OSS, and expulsion for all stu- ond author conducted a search of the Civil Rights Data
dents of color compared with all White students? Collection survey website for the 2013–2014 school year
and downloaded the Discipline Report for each district for
the 2013–2014 school year. Demographic and risk ratio sta-
Method tistics were calculated from the retrieved reports.
This review was conducted in six phases beginning in the
Fall of 2016 and concluded in the Spring of 2017. Phase 1
Sample
consisted of the research team (i.e., five doctoral candidates
and the first author) developing the Checklist for Analyzing The sample of 147 policies included documents from the
Discipline Policies and Procedures for Equity (CADPPE; three largest public-school districts by student enrollment in
see attached) by reviewing and adapting previous checklists all 50 states of the United States and the District of
and recommendations for discipline policies found in the Columbia. Table 1 contains descriptive statistics for enroll-
literature (Fenning & Bohanon, 2006; Green et al., 2015; ment of all students and students with disabilities across all
Longstreth et al., 2013). In Phase 2, the research team gath- districts included in the sample. The districts included in the
ered a list of the three largest school districts in each state sample represented 0.24%–100% of the public-school dis-
and the District of Columbia (D.C.) by enrollment size tricts in each state (M = 7.88%, SD = 23.43) and 0.84% of
through a search of the National Center for Educational the public-school districts in the nation. We selected this
Statistics (NCES) website. In Phase 3, the research team sample to represent a diverse student body and the largest
searched each district’s webpage and collected the most number of students in each state. In addition, larger districts
current, available discipline policy documents (i.e., code of are more inclined to have functioning websites where codes
conduct and/or parent and student handbook). In Phase 4, of conducts are updated and accessible to the public.
the research team participated in training on the use of Hawaii and the District of Columbia each contained only
CADPPE. In training, research team members indepen- one public school district within their geographical bound-
dently coded three of the included policies to ensure consis- ary, and the three largest districts in New York used the
tent interpretation of the checklist. The first and second same policy. As a result, the review included a single disci-
policies were used as practice for coding and clarification. pline policy from the District of Columbia, the state of
Green et al. 141

Hawaii, and New York City Public Schools, respectively. While these risk indices provide a measure of the risk of
Public school policies were used because disciplinary poli- an exclusionary practice involving a student from the target
cies within private, parochial, and charter schools vary group, they do not allow comparison between multiple tar-
greatly as does the extent to which they are required to meet get groups (Boneshefski & Runge, 2014; Girvan et al.,
federal requirements. Publication dates attributed to poli- 2019). Therefore, we also calculated risk ratios. We calcu-
cies reviewed ranged from 1991 to 2016. lated risk ratios using the following equation:

Risk Index for Target Group

Demographic and Discipline Data Risk Ratio = .
Risk Index for Comparison Group
Identifying variables collected from the Civil Rights Data
Collection survey included state name, district name, and As with the demographic data, we calculated risk indices
school level(s) for each policy (e.g., primary, secondary, and and risk ratios for the total student enrollment and repeated
early childhood). After recording identifying variables for the calculations for students with disabilities. For the risk
the included school districts, we calculated the district ratios for all students, White students served as the compari-
enrollment for all students in each racial/ethnic groups as son group. For the risk ratios for students with disabilities,
defined by the U.S. Department of Education, Office for White students with disabilities served as the comparison
Civil Rights (2014), and calculated the number of exclusion- group.
ary discipline incidents (i.e., ISS, OSS, and expulsion) for
each racial/ethnic group. The survey included the total num-
ber of students enrolled and a percentage of students from Policy Coding
each racial/ethnic group. Our enrollment data were calcu- The CADPPE was used to analyze the district policies. The
lated by multiplying the total enrollment by the percentage CADPPE was developed using Green et al.’s (2015)
associated with the racial/ethnic group. Similarly, the survey “7 Key Elements of Effective Policy to Enhance Equity
included a total number of incidents for each exclusionary in School Discipline” and revised items from previously
practice and percentages of incidents associated with each developed policy assessments and checklists (Fenning &
racial/ethnic group. Again, we calculated the number of each Bohanon, 2006; Longstreth et al., 2013) and is aligned with
type of incident for each racial/ethnic group by multiplying Rethinking School Discipline: School District Leader
the total number of incidents by the percentage associated Summit on Improving School Climate and Discipline,
with the group. The survey also included the same demo- Resource Guide for Superintendent Action (U.S. Department
graphic and disciplinary information reported for only of Education, 2015). The CADPPE is composed of 45
students with disabilities. We repeated demographic and checklist items scored dichotomously (i.e., present and not
exclusionary practice calculations for students with disabili- present) and covers the following eight elements: (1) spe-
ties in each school district. The Civil Rights Data Collection cific commitment to equity; (2) family/community partner-
surveys did not distinguish between the districts within New ships in policy and procedure development and practice;
York City Public Schools. As a result, New York City Public (3) focus on implementing positive, proactive behavior
Schools data were not included in the data set. support practices; (4) clear, objective discipline procedures;
Once all demographic and exclusionary practice values (5) absence, removal, or reduction of exclusionary prac-
were calculated, we calculated risk indices and risk ratios tices; (6) graduated discipline systems with instructional
using formulas provided by the Office of Special Education alternatives to exclusion; (7) procedures with accountability
Programs (OSEP) Technical Assistance Center on PBIS for equitable student outcomes; and (8) equity in early
(McIntosh, Barnes, Eliason, & Morris, 2014). Typically, the childhood settings.
risk index is described as the proportion of students from a
target group who have experienced the outcome of interest
(Girvan et al., 2019). The Civil Rights Data Collection sur- Interrater Reliability
vey reports the number of incidents rather than the number To measure reliability, 28.5% (n = 42) of the district poli-
of individual students experiencing exclusionary practices. cies were independently coded twice by two members of
Therefore, for the purposes of this study, we calculated the the research team for exact agreement on each CADPPE
risk index for incidents, as described by Girvan et al. (2017) item. Interrater reliability (IRR) was calculated using per-
using the following equation: cent agreement; dividing the number of agreements by the
total number of items on the checklist and multiplying by
No.of Incidents Associated with
100 (agreements ÷ 45 total checklist items × 100). When
Studentsin theTarget Group there was a disagreement between the two coders, a third
Risk Index = .
Total No.of Studentsin theTarget Group member of the research team independently coded the
142 Journal of Positive Behavior Interventions 23(3)

discrepancy item within the policy a third time to reach more incidents compared with their White peers with dis-
100% agreement. abilities. A similar trend was evident for OSS where risk
ratios ranged from 1.23 to 3.65. However, the risk ratio for
expulsion ranged from 0.30 to 3.39.
Data Analysis
Correlations between the score on the CADPPE and risk
District discipline policies were coded to examine a variety ratios for each form of exclusionary practice for students with
of policy components referred to as Elements on the disabilities are reported in Table 4. None of the correlations
CADPPE. The CADPPE was used to analyze policies that between the CADPPE ratings and the risk ratios for exclu-
may promote equitable school discipline practices. For sionary practice were statistically significant (p = .05).
example, policies including family and community involve- In addition, all the correlations were extremely small, rang-
ment in the creation and implementation of evidence-based ing from −0.19 to 0.23.
discipline practices may promote a more equitable environ-
ment when compared with policies that include the blind
RQ3: What is the Relationship Between Themes
use of exclusionary discipline.
After all policies were coded and data entry completed, Found in a District’s Discipline Policy and the
basic descriptive statistics were calculated for each question Risk Ratio of ISS, OSS, and Expulsion for All
and each scale of CADPPE. In addition, a Pearson product– Students of Color Compared With All White
moment correlation was run to determine the relationship Students?
between the total percentage score on the checklist and the
risk ratios for ISS, OSS, and expulsion. Table 3 contains descriptive statistics regarding the risk
ratios for ISS, OSS, and expulsion for all students. For all
students, the risk ratios for ISS ranged from 1.09 to 4.00
Results indicating students from all racial/ethnic groups were dis-
proportionately involved in more incidents compared with
RQ1: What Common Elements Exist in District their White peers. Similar trends were evident for OSS
Discipline Policies Across the 50 U.S. States and where risk ratios ranged from 0.87 to 5.82, and expulsions
the District of Columbia? where risk ratios ranged from 0.84 to 7.69. In particular, the
The average score for each element of the CADPPE is risk ratio for each type of exclusionary practice was highest
reported in Table 2. Table 2 also includes the number of for students identifying as African American. Correlations
policies reviewed for each item and the number of policies between the score on the CADPPE and risk ratios for each
in which the item was rated present are reported. The mean form of exclusionary practice for all students are reported in
percentage of items rated present, for the entire instrument, Table 4. None of the correlations between the CADPPE rat-
was 19.14% (SD = 12.76, range = 0%–58.54%). The mean ings and the risk ratios for exclusionary practice were statis-
percentage of items rated present for each element ranged tically significant (p = .05).
from 4% to 59.18%. Element 4 (clear, objective discipline
procedures) and Element 5 (absence, removal, or reduction IRR
of exclusionary practices) had the highest average scores
(59.18% and 33.16%, respectively). Element 7 (procedures For the policies coded by two coders, mean agreement was
with accountability for equitable student outcomes) had the 87.93% (range = 66.66%–100.00%). After the third coder
lowest average score at 3.99%. reviewed the discrepancies, there was a discussion between
the third coder and the coder whose coding was in disagree-
ment, and agreement was reached yielding 100% agreement
RQ2: What is the Relationship Between Themes on all policies coded for IRR. Most disagreements were
Found in a District’s Discipline Policy and from checklist Item 1 (i.e., Does the discipline policy/proce-
the Risk Ratio of ISS, OSS, and Expulsion for dure contain a mission/goal statement?); Item 2 (i.e., Does
Students of Color With Disabilities Compared the mission statement have specific language that expresses
a commitment to equity or equitable outcomes [e.g., racial,
With White Students With Disabilities? cultural, ability, decrease suspension for a certain group]?);
Table 3 contains descriptive statistics regarding the risk Item 13 (i.e., Are behavioral expectations provided?); and
ratios for ISS, OSS, and expulsion for students with dis- Item 14 (i.e., Are the behavioral expectations defined posi-
abilities. For students with disabilities, the risk ratios for tively, clearly identifying what successful demonstration of
ISS ranged from 0.86 to 3.89 indicating students from most skills looks like in context, vs. identifying what not to do?)
racial/ethnic groups were disproportionately involved in where criteria or definitions were less clear.
Green et al. 143

Table 2. CADPPE Element Average Score and Item Frequencies.

Item Descriptor # Scored # Present (%)

Element 1: Specific commitment to equity
M = 0.56 (SD = 0.91, range = 1–3)
1 Mission/goal statement 147 49 (33.33)
2 Language expressing commitment to equity 49 15 (30.61)
3 Explicit language related to equity 49 3 (6.12)
4 Clearly articulated long-term objective 49 16 (32.55)
Element 2: Family/community partnerships in policy and procedure development and practice
M = 1.67 (SD = 1.35, range = 0–6)
5 Seeking input from a range of families 147 17 (36.17)
6 Writing plan for providing information to families 147 44 (29.93)
7 Regular evaluation of efficacy of family involvement 147 7 (4.76)
8 Opportunity to develop, implement, review, and revise 147 20 (13.60)
9 Proactive collaborations to promote prosocial behavior 147 26 (17.68)
10 Reactive collaborations to promote prosocial behavior 147 126 (85.71)
11 Obtaining contextually/culturally relevant information 147 3 (2.04)
12 Embedding behavior plan goals/objectives into home routines 147 2 (1.36)
Element 3: Focus on implementing positive, proactive behavior support practices
M = 1.19 (SD = 1.42, range = 0–5)
13 Behavioral expectations provided 147 64 (43.53)
14 Defined positively 64 49 (33.33)
15 Behavior expectations which vary across setting 147 21 (14.29)
16 Prosocial behaviors operationally defined 147 12 (8.16)
17 Prevention model with instructional foci
18 Prosocial behaviors culturally relevant
19 Focus on support structure for implementation 147 16 (10.88)
20 Practices to reinforce prosocial behaviors 147 13 (8.84)
Element 4: Clear, objective discipline procedures
M = 1.77 (SD = 1.05, range = 0–3)
21 Clear delineations between major and minor incidents 147 75 (51.02)
22 Problem behaviors operationally defined 147 110 (74.83)
23 Rights and responsibilities for adults and students defined 147 76 (51.70)
Element 5: Absence, removal, or reduction of exclusionary practices
M = 1.33 (SD = 0.98, range = 0–4)
24 Zero tolerance included as practice/strategy 147 106 (72.11)
25 Use of exclusionary discipline restricted for nonviolent behaviors 147 13 (8.84)
26 Clear communication of limitation of suspension expulsion 147 21 (14.29)
27 Descriptions of guidelines for using alternative to suspension 147 55 (37.41)
Element 6: Graduated discipline systems with instructional alternatives to exclusion
M = 1.11 (SD = 1.75, range = 0–6)
28 Provides possible instructional responses to replace punitive responses 147 41 (27.89)
29 One or more evidenced-based strategies 41 22 (53.65)
30 Universal strategies included 147 32 (21.77)
31 Secondary strategies included 147 23 (15.65)
32 Tertiary strategies included 147 24 (16.33)
33 Process for assessing academic support needs 147 22 (53.65)
Element 7: Procedures with Accountability for Equitable Student Outcomes
M = 0.31 (SD = 1.07, range = 0–7)
34 Communicates a process for preparation/training of staff 147 10 (6.80)
35 Communicates a plan for data analysis 147 12 (8.16)
36 Communicates need for disaggregated data 12 9 (75.00)
37 Communicates need for data analysis through team meetings 12 3 (25.00)
38 Communicates need for data-based decision making 12 9 (75.00)
39 Includes a plan for regular sharing with stakeholders 12 3 (25.00)
40 Describes how success of policy/procedure is measured 12 1 (8.33)
41 Provides for formative and summative evaluations 12 0 (0.00)

Note. Items indented are only scored if the item immediately preceding was scored “Present,” in the same policy. # Scored = number of policies
reviewed for presence of the item, # Present = number of policies scored as containing the item. CADPPE = Checklist for Analyzing Discipline
Policies and Procedures for Equity.
144 Journal of Positive Behavior Interventions 23(3)

Table 3. Risk Ratios for Exclusionary Practices for All Students and Students With Disabilities.

All students Students with disabilities

Race/ethnicity M Median SD M Median SD

In school suspension
Hispanic 2.50 1.51 6.12 2.10 1.36 2.60
African American 4.00 2.89 6.34 3.22 2.39 3.69
Asian 1.09 0.54 3.29 1.27 0.68 2.60
American Indian 2.77 1.91 4.95 3.89 0.00 8.13
Native Hawaiian 1.93 0.91 4.42 0.86 0.00 3.33
Two or More Races/Ethnicities 2.34 1.69 4.24 3.20 2.36 4.48
Out-of-school suspension
Hispanic 3.74 1.56 14.48 2.17 1.35 3.50
African American 5.82 3.32 25.01 3.65 2.60 4.87
Asian 0.87 0.55 3.27 1.23 0.67 3.19
American Indian 4.43 2.24 20.44 2.89 1.36 4.26
Native Hawaiian 3.31 0.82 21.92 1.32 0.00 4.08
Two or more Races/ethnicities 4.39 2.01 22.27 3.46 2.57 3.69
Expulsion
Hispanic 1.74 1.13 3.61 3.39 0.75 14.26
African American 7.69 2.76 24.67 3.35 1.68 7.78
Asian 0.91 0.00 3.46 0.30 0.00 1.09
American Indian 1.62 0.00 4.28 0.58 0.00 2.59
Native Hawaiian 0.84 0.00 4.27 1.87 0.00 12.77

Note. Risk ratios calculated from 2013–2014 Civil Rights Data Collection survey using White students as the comparison group for all students and
White students with disabilities for Students with Disabilities. New York school districts are excluded from the sample.

Discussion science (Bauer et al., 2015). New scales developed by imple-

mentation researchers have identified critical elements in the
A review of 147 district discipline policies was conducted stages of implementation that predict full scale use and sus-
using a 45-item checklist across the three largest school tainment of evidence-based practices (Saldana et al., 2012).
districts, by enrollment size, in all 50 U.S. states and D.C. Clear guidelines and policies are two critical elements pre-
This review adds to the literature base of discipline policy dictive of improved long-term uptake and sustainability of
reviews (Curran, 2016, 2017; Fenning & Rose, 2007; evidence-based practices (Saldana et al., 2012). Given the
Longstreth et al., 2013; Payne & Welch, 2015; Welch & link between implementation-related factors and long-term
Payne, 2010, 2012) as the first to methodically explore poli- sustainability, the OSEP Technical Assistance Center on
cies across the nation using a systematic approach. District PBIS has been especially attuned to the necessary partner-
discipline policies were reviewed and scored for recom- ship between policy and practice as well as the importance
mended elements using the CADPPE. In general, findings of using implementation science. The Center regularly pro-
indicated school disciplinary policies across the sample vides resources related to the use of implementation science
included less than half of the recommended elements. These for improving the integration of PBIS practices (e.g., Lewis
results align with previous research demonstrating that et al., 2010). In fact, the national PBIS website (pbis.org)
district policies have moved away from their positive and provides many resources on integrating policy and special-
proactive foundational intent (Fenning & Bohanon, 2006) ized implementation supports for facilitating improved use
to more punitive and reactive practices that disproportion- and sustainment of its practices (e.g., Green et al., 2015;
ately target and harm students of color and students with Swain-Bradway et al., 2018). Yet, in spite of these well-
disabilities (Curran, 2016). developed resources, there is little information available on
As research continues to evolve related to the understand- whether local education agencies have integrated these sup-
ing of how interventions and supports are actually imple- ports and frameworks into their policies and practices for
mented with success in complex systems, such as education, reducing disparities. Therefore, it becomes critical to exam-
we are beginning to better understand elements that are criti- ine local policies and guidance on discipline processes and
cal to the adoption and scaling up of evidence-based prac- determine whether they are aligned with evidence-based
tices. This area of specialization is known as implementation practices.
Green et al. 145

Table 4. Correlation Between Risk Ratios for Exclusionary Practices and Total Checklist Score for All Students and Students With
Disabilities.

Exclusionary Practices 1 2 3 4 5 6 7
In school suspension
1. Hispanic risk ratio – −.06 −.13 .02 .02 .04 −.11
2. African American risk ratio .14 – .341** −.03 .10 .05 −.01
3. Asian risk ratio .807** −.06 – .01 .04 −.02 .03
4. American Indian risk ratio .244** .757** .03 – .262** −.04 .01
5. Native Hawaiian risk ratio .306** −.03 .504** .05 – −.10 .02
6. Two or more races/ethnicities risk ratio .227** .691** .05 .758** .04 – −.08
7. CADPPE percent score −.02 .02 −.07 .07 −.10 .07 –
Out-of-school suspension
1. Hispanic risk ratio – .01 −.08 −.09 .08 .15 −.14
2. African American risk ratio .771** – −.03 −.02 .02 .11 −.08
3. Asian risk ratio .773** .984** – −.08 .02 .10 .08
4. American Indian risk ratio .767** .983** .982** – .294** .09 −.10
5. Native Hawaiian risk ratio .764** .989** .985** .985** – −.05 .00
6. Two or more races/ethnicities risk ratio .752** .981** .976** .974** .979** – .05
7. CADPPE percent score −.12 −.04 −.06 −.04 −.02 .00 –
Expulsion
1. Hispanic risk ratio – .16 0.15 −0.04 −.02 −.07 .09
2. African American risk ratio 0.02 – 0.04 −0.01 −.06 −.06 .05
3. Asian risk ratio −0.04 .02 – 0.05 −.04 .23 −.17
4. American Indian risk ratio 0.03 −.06 −0.01 – −.02 −.03 −.04
5. Native Hawaiian risk ratio −0.03 −.04 −0.04 .367** – −.07 .03
6. Two or more races/ethnicities risk ratio −0.06 −.01 −0.04 0.00 −.04 – .23
7. CADPPE percent score −0.06 .00 −.19 .10 .09 −.08 –

Note. The results for students with disabilities are shown above the diagonal. The results for all students are shown below the diagonal. Risk ratios
calculated from 2013–2014 Civil Rights Data Collection survey using White students as the comparison group for All Students and White students
with disabilities for Students with Disabilities. New York school districts are excluded from the sample. CADPPE = Checklist for Analyzing Discipline
Policies and Procedures for Equity.
**p < .01 level.

Implications for Policy Development support (universal, secondary, and tertiary); (d) embed lan-
guage in policies to eliminate use of exclusionary discipline
For districts to develop a truly comprehensive and evidence- for nonviolent behaviors to avoid disproportionate impact
based system of supports related to school discipline, dis- to particular groups of students (Lhamon & Gupta, 2014;
trict policy makers must engage in a review of disciplinary McIntosh, Girvan, Horner, & Smolkowski, 2014); (e)
policy. Furthermore, ongoing policy review plays an essen- include clear definitions of expected behaviors and opera-
tial role in verifying that policies continue to function to tionally define the differences between major and minor
increase positive, preventive, and equitable practices for behaviors (Green et al. 2015); and (f) consider the develop-
all students (Curran, 2017). Efforts to review and reform ment of policies including comprehensive processes specific
policy should (a) evaluate current policies using tools to all ages, including early childhood systems as students of
and checklists, such as CADPPE, that function to increase color in preschool are also receiving disproportionate rates
positive, preventive, and equitable practices (Fenning & of exclusionary practices (Gilliam & Shahar, 2006).
Bohanon, 2006; Green et al., 2015; Longstreth et al., 2013);
(b) create or review discipline policies with key stakehold-
ers that include families and individuals representative of Implications for Policy Implementation
the community served to ensure policies promote and sup- Despite a growing research base indicating more effective
port evidence-based, culturally responsive, and equitable and less discriminatory alternatives to the use of exclusion-
practices in school discipline (Fenning & Bohanon, 2006); ary practices (Losen, 2013; McIntosh, Girvan, Horner,
(c) create or review policies to ensure inclusion of methods Smolkowski, & Sugai, 2014), the findings from this review
and procedures for proactively and directly teaching appro- indicate the possibility of a research-to-practice gap in the
priate prosocial, replacement behaviors along all tiers of area of school discipline policy. This gap mirrors the
146 Journal of Positive Behavior Interventions 23(3)

research indicating difficulty with implementation of address both policy and practice. Easily implementable
evidence-­ based practices in applied settings (Cook & and effective means of revising school discipline policies
Odom, 2013; Fixsen et al., 2009). Therefore, efforts to may improve implementation of systems-wide processes,
decrease a potential research-to-practice gap related to dis- and more importantly, potentially improve the lives of and
cipline policies should (a) embed procedures to establish outcomes for students, particularly those from historically
proactive relationships and communication systems with marginalized groups. By establishing policies of support
families into school discipline policies (Green et al., 2019); that span the developmental continuum and range of
(b) establish procedures for regular review of data to ensure needs, we can improve outcomes for all students and cre-
ongoing evaluation of needs and supports related to disci- ate a more equitable and effective society.
pline practices (McIntosh, Girvan, Horner, Smolkowski, &
Sugai, 2014); (c) create a system of ongoing training and
professional development for staff related to school disci-
Limitations
pline (e.g., sworn law enforcement officers [SLEOs], school Limitations must be considered when evaluating the find-
resource officers [SROs], and crisis management team) ings of this policy review. First, although the CADPPE was
including teaching preventive strategies and clear alterna- developed as a checklist based on other checklists and sup-
tives to suspension and expulsion (Losen, 2013); and (d) ported by research, it is not empirically validated; however,
ensure staff and school leadership are trained in and under- researchers and technical assistance providers informally
stand vulnerable decision points (i.e., specific contextual reviewed the checklist for face validity. The preliminary use
events or elements where implicit bias may yield increased of the CADPPE in this study provides opportunities for
rates of disproportionate disciplinary practices) and implicit more rigorous research and the validation of the measure.
bias (i.e., unconscious beliefs about an individual; McIntosh, Furthermore, while the checklist scores were low overall,
Girvan, Horner, Smolkowski, & Sugai, 2014). findings demonstrated decreases in disparities among dis-
A history of research clearly demonstrates the long- tricts with higher scores on the CADPPE. Future research
standing benefits of PBIS compared with traditional disci- should seek to validate the CADPPE. With validation, given
plinary techniques (Mitchell et al., 2018); yet, traditional the low reliability of some of the checklist items, future
practices often remain at the forefront of policy and practice research should seek to provide operational definitions of
in the field of education (Sugai & Horner, 2006). A recent each checklist item with examples and nonexamples.
review indicated lower rates of disproportionality in schools A second limitation would be that the review examined
implementing PBIS compared with those without PBIS district policies. Therefore, generalization of results and
with rates of 3.67 and 4.33, respectively, for the Black/ interpretations of practices cannot be made at the school
White OSS risk ratio (McIntosh et al., 2018). Although this level. This is an important area of study as exclusionary
puts schools implementing PBIS at a lower than average practices and potential disproportionality occur at the school
rate of disproportionality, there is still significant room for level beginning in the classroom with office discipline
improvement. Difficulties related to implementation, such referrals.
as fidelity, may ultimately underlie why supports such as Third, the review did not assess the extent to which poli-
PBIS have not had a larger impact on disproportionate prac- cies included guidelines for SLEO or SRO on campus. With
tices in discipline (McIntosh et al., 2018). After all, if an the overrepresentation of students of color referred to law
intervention works but is not actually being implemented enforcement or subject to officer-related arrests (U.S.
with fidelity, it means it is not fully reaching the individuals Department of Education, Office for Civil Rights, 2016), it
for whom it was intended, limiting its impact on outcomes is critical to evaluate whether policies include training and
(Flannery et al., 2014). guidelines for SLEOs and SROs.
One of the five PBIS implementation drivers is “Policy Fourth, although the study initially set out to include
and Systems Alignment” (OSEP Technical Assistance early childhood policies in the review, those policies were
Center on PBIS, 2015). However, supports and direction difficult to obtain and were often separate from a district’s
for the former have been limited. CADPPE, as it is based K–12 policies. Due to the disparities found among preschool
on effective practices advocated within the PBIS frame- children of color (U.S. Department of Education, Office for
work, is an example of what that support might look like. Civil Rights, 2016), future research should replicate this
Furthermore, research suggests more effective and less study using early childhood policies as the variable of focus.
discriminatory alternatives to the frequent use of exclu- Fifth, policies from charter schools were not included in
sionary practices exist. Based upon changing legislation the sample. Future research should seek to analyze policies
related to zero tolerance, other exclusionary practices, and found in charter schools as these schools make up large por-
evidence-based practices related to school discipline, tions of districts in various settings and discipline can look
areas for district improvement should include action to different in these contexts.
Green et al. 147

Summary Fenning, P. A., & Bohanon, H. (2006). Schoolwide discipline

policies: An analysis of discipline codes of conduct. In C. M.
Data from this study demonstrate the importance of Evertson & C. S. Weinstein (Eds.), Handbook of classroom
exploring the connection between policy implementation management: Research, practice, and contemporary issues
and discipline outcomes at the district level. Exclusionary (pp. 1021–1039). Lawrence Erlbaum.
practices are not effective for changing student behavior Fixsen, D. L., Blase, K. A., Naoom, S. F., & Wallace, F. (2009).
and have harmful student outcomes (Noltemeyer et al., Core implementation components. Research on Social Work
2015). Revising discipline policies to include key elements Practice, 19(5), 531–540. https://doi.org/10.1177/10497315
like those assessed in this study could have a positive impact 09335549
Flannery, K. B., Fenning, P., Kato, M. M., & McIntosh, K.
on large numbers of students through a decrease in exclu-
(2014). Effects of school-wide positive behavioral interven-
sionary practices. tions and supports and fidelity of implementation on prob-
lem behavior in high schools. School Psychology Quarterly,
Declaration of Conflicting Interests 29(2), 111–124. https://doi.org/10.1037/spq0000039
The author(s) declared no potential conflicts of interest with respect Gilliam, W. S., & Shahar, G. (2006). Preschool and child care
to the research, authorship, and/or publication of this article. expulsion and suspension: Rates and predictors in one state.
Infants & Young Children, 19(3), 228–245. https://doi.
Funding org/10.1097/00001163-200607000-00007
Girvan, E. J., Gion, C., McIntosh, K., & Smolkowski, K. (2017).
The author(s) received no financial support for the research, The relative contribution of subjective office referrals to racial
authorship, and/or publication of this article. disproportionality in school discipline. School Psychology
Quarterly, 32(3), 3392–3404.
ORCID iD Girvan, E. J., McIntosh, K., & Smolkowski, K. (2019). Tail, tusk,
Ambra L. Green https://orcid.org/0000-0001-5363-1939 and trunk: What different metrics reveal about racial dispro-
portionality in school discipline. Educational Psychologist,
54(1), 140–159.
References Green, A. L., Cohen, D. R., & Stormont, M. (2019). Addressing and
Albrecht, S. F., Skiba, R. J., Losen, D. J., Chung, C.-G., & preventing disproportionality in exclusionary discipline prac-
Middelberg, L. (2012). Federal policy on disproportional- tices for students of color with disabilities. Intervention in School
ity in special education: Is it moving us forward? Journal and Clinic, 54(4), 1241–1245. https://doi.org/10.1177/1053
of Disability Policy Studies, 23(1), 14–25. https://doi. 451218782437
org/10.1177/1044207311407917 Green, A. L., Nese, R. N., McIntosh, K., Nishioka, V., Eliason,
Bauer, M. S., Damschroder, L., Hagedorn, H., Smith, J., & B., & Canizal Delabra, A. (2015). Key elements of poli-
Kilbourne, A. M. (2015). An introduction to implementation cies to address disproportionality within SWPBIS: A guide
science for the non-specialist. BMC Psychology, 3(1), Article for district and school teams. Office of Special Education
32. https://doi.org/10.1186/s40359-015-0089-9 Programs Technical Assistance Center on Positive Behavioral
Boneshefski, M. J., & Runge, T. J. (2014). Addressing dispro- Interventions and Supports.
portionate discipline practices within a school-wide positive Gregory, A., Clawson, K., Davis, A., & Gerewitz, J. (2016). The
behavioral interventions and supports framework: A practi- promise of restorative practices to transform teacher-student
cal guide for calculating and using disproportionality rates. relationships and achieve equity in school discipline. Journal
Journal of Positive Behavior Interventions, 16(3), 3149–3158. of Educational and Psychological Consultation, 26(4),
Cook, B. G., & Odom, S. L. (2013). Evidence-based practices 325–353. https://doi.org/10.1080/10474412.2014.929950
and implementation science in special education. Exceptional Hoffman, S. (2014). Zero benefit: Estimating the effect of zero
Children, 79(2), 135–144. https://doi.org/10.1177/0014 tolerance discipline polices on racial disparities in school
40291307900201 discipline. Educational Policy, 28(1), 69–95. https://doi.
Curran, F. C. (2016). Estimating the effect of state zero tolerance org/10.1177/0895904812453999
laws on exclusionary discipline, racial discipline gaps, and stu- Kennedy-Lewis, B. L. (2014). Using critical policy analysis to
dent behavior. Educational Evaluation and Policy Analysis, examine competing discourses in zero tolerance legisla-
38(4), 647–668. https://doi.org/10.3102/0162373716652728 tion: Do we really want to leave no child behind? Journal of
Curran, F. C. (2017). The law, policy, and portrayal of zero tol- Education Policy, 29(2), 165–194. https://doi.org/10.1080/02
erance school discipline: Examining prevalence and char- 680939.2013.800911
acteristics across levels of governance and school districts. Lewis, T. J., Barrett, S., Sugai, G., & Horner, R. H. (2010).
Educational Policy, 33(2), 319–349. https://doi.org/10.1177 Blueprint for school-wide positive behavior support training
/0895904817691840 and professional development. National Technical Assistance
Fenning, P., & Rose, J. (2007). Over-representation of African Center on Positive Behavior Interventions and Support.
American students in exclusionary discipline: The role of Lhamon, C., & Gupta, V. (2014). Dear Colleague Letter:
school policy. Urban Education, 42(6), 536–559. https://doi. Civil Rights in Juvenile Justice Residential Facilities. US
org/10.1177/0042085907305039 Department of Justice.
148 Journal of Positive Behavior Interventions 23(3)

Longstreth, S., Brady, S., & Kay, A. (2013). Discipline policies Office of Special Education Programs Technical Assistance
in early childhood care and education programs: Building an Center on Positive Behavioral Interventions and Supports.
infrastructure for social and academic success. Early Education (2015, October). Positive behavioral interventions and sup-
& Development, 24(2), 253–271. https://doi.org/10.1080/1040 ports (PBIS) implementation blueprint: Part 1—Foundations
9289.2011.647608 and supporting information. University of Oregon. Available
Losen, D. J. (2013). Discipline policies, successful schools, racial from: www.pbis.org
justice, and the law. Family Court Review, 51(3), 388–400. Payne, A. A., & Welch, K. (2015). Restorative justice in schools:
https://doi.org/10.1111/fcre.12035 The influence of race on restorative discipline. Youth & Society,
Losen, D. J. (2018). Disabling punishment: The need for rem- 47(4), 539–564. https://doi.org/10.1177/0044118X12473125
edies to the disparate loss of instruction experienced by Saldana, L., Chamberlain, P., Wang, W., & Brown, C. H. (2012).
black students with disabilities. Civil Rights Project/Proyecto Predicting program start-up using the stages of implementa-
Derechos Civiles. https://www.parentcenterhub.org/disabling tion measure. Administration and Policy in Mental Health and
-punishment/ Mental Health Services Research, 39(6), 419–425. https://doi.
Losen, D. J., & Skiba, R. J. (2010). Suspended education: Urban org/10.1007/s10488-011-0363-y
middle schools in crisis. Southern Poverty Law Center. Skiba, R. J., Albrecht, S., & Losen, D. (2013). CCBD’s position
McIntosh, K., Barnes, A., Eliason, B., & Morris, K. (2014). Using summary on federal policy on disproportionality in special
discipline data within SWPBIS to identify and address dis- education. Behavioral Disorders, 382(2), 108–120. https://
proportionality: A guide for school teams. Office of Special doi.org/10.1177/019874291303800202
Education Programs Technical Assistance Center on Positive Skiba, R. J., Horner, R. H., Chung, C., Rausch, M. K., May, S. L.,
Behavioral Interventions and Supports. https://www.pbis. & Tobin, T. (2011). Race is not neutral: A national investi-
org/resource/using-discipline-data-within-swpbis-to-identify- gation of African American and Latino disproportionality in
and-address-disproportionality-a-guide-for-school-teams school discipline. School Psychology Review, 40(1), 85–107.
McIntosh, K., Gion, C., & Bastable, I. (2018). Do schools imple- Skiba, R. J., & Peterson, R. (2000). School discipline at a crossroads:
menting SWPBIS have decreased racial and ethnic dispro- From zero tolerance to early response. Exceptional Children,
portionality in school discipline? Office of Special Education 66(3), 335–346. https://doi.org/10.1177/001440290006600305
Programs Technical Assistance Center on Positive Behavioral Smolkowski, K., Girvan, E. J., McIntosh, K., Nese, R. N., &
Interventions and Supports. https://www.pbis.org/resource/ Horner, R. H. (2016). Vulnerable decision points for dispro-
do-schools-implementing-swpbis-have-decreased-racial-and- portionate office discipline referrals: Comparisons of dis-
ethnic-disproportionality-in-school-discipline cipline for African American and White elementary school
McIntosh, K., Girvan, E. J., Horner, R., & Smolkowski, K. (2014). students. Behavioral Disorders, 41(4), 178–195. https://doi.
Education not incarceration: A conceptual model for reduc- org/10.17988/bedi-41-04-178-195.1
ing racial and ethnic disproportionality in school discipline. Sugai, G., & Horner, R. R. (2006). A promising approach for
Journal of Applied Research on Children Informing Policy expanding and sustaining school wide positive behavior sup-
for Children at Risk, 5(2), 1–22. port. School Psychology Review, 35(2), 245–259.
McIntosh, K., Girvan, E. J., Horner, R. H., Smolkowski, K., & Swain-Bradway, J., Freeman, J., Kittelman, A., Nese, R.,
Sugai, G. (2014). U.S. Department of Education Office for & Technical Assistance Center on Positive Behavioral
Civil Rights, 2014: Recommendations for addressing dis- Interventions Supports. (2018). Fidelity of SW-PBIS in
cipline disproportionality in education. Office of Special high schools: Patterns of implementation strengths and
Education Programs Technical Assistance Center on Positive needs. Technical Assistance Center on Positive Behavioral
Behavioral Interventions and Supports. Interventions and Supports.
Mitchell, B. S., Hatton, H., & Lewis, T. J. (2018). An examina- U.S. Department of Education. (2015). Rethink school discipline:
tion of the evidence-base of school-wide positive behavior School district leader summit on improving school climate
interventions and supports through two quality appraisal and discipline: Resource guide for superintendent action.
processes. Journal of Positive Behavior Interventions, 20(4), U.S. Department of Education, Office for Civil Rights. (2014, March).
239–250. https://doi.org/10.1177/1098300718768217 Civil rights data collection. U.S. Department of Education.
Mongan, P., & Walker, R. (2012). “The road to hell is paved with https://www2.ed.gov/about/offices/list/ocr/data.html
good intentions”: A historical, theoretical, and legal analy- U.S. Department of Education, Office for Civil Rights. (2016). A
sis of zero-tolerance weapons policies in American schools. first look: Key data highlights on equity and opportunity gaps
Preventing School Failure, 56(4), 232–240. https://doi.org/10 in our nation’s public schools. U.S. Department of Education.
.1080/1045988X.2011.654366 U.S. Department of Education, Office for Civil Rights. (2018,
National Institute of Education. (1978). Violent schools-safe March). Civil rights data collection. U.S. Department of
schools: The safe school study report to the congress. Super- Education. Available from: ocrdata.ed.gov/
intendent of Documents. Welch, K., & Payne, A. A. (2010). Racial threat and punitive
National School Resource Network. (1980). Resource handbook school discipline. Social Problems, 57(1). 25–48. https://doi.
on discipline codes. ERIC Clearinghouse. org/10.1525/sp.2010.57.1.25
Noltemeyer, A. L., Ward, R. M., & Mcloughlin, C. (2015). Welch, K., & Payne, A. A. (2012). Exclusionary school punish-
Relationship between school suspension and student outcomes: ment: The effect of racial threat on expulsion and suspen-
A meta-analysis. School Psychology Review, 44(2), 224–240. sion. Youth Violence and Juvenile Justice, 10(2), 2155–2171.
https://doi.org/10.17105/spr-14-0008.1 https://doi.org/10.1177/1541204011423766