The Achievement Gap and the Discipline Gap: Two Sides of the Same Coin?

Author(s): Anne Gregory, Russell J. Skiba and Pedro A. Noguera

Source: Educational Researcher, Vol. 39, No. 1, New Perspectives on School Safety and Violence
Prevention (JANUARY/FEBRUARY 2010), pp. 59-68
Published by: American Educational Research Association
Stable URL: http://www.jstor.org/stable/27764554
Accessed: 11-03-2015 23:04 UTC

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://www.jstor.org/page/info/about/policies/terms.jsp
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content
in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship.
For more information about JSTOR, please contact support@jstor.org.

American Educational Research Association is collaborating with JSTOR to digitize, preserve and extend access to Educational
Researcher.

http://www.jstor.org

This content downloaded from 152.15.236.17 on Wed, 11 Mar 2015 23:04:00 UTC
All use subject to JSTOR Terms and Conditions

and the Discipline

Coin?

The Achievement
Gap
Two Sides of the Same

Gap:

Anne Gregory, Russell J.Skiba,and Pedro A. Noguera

The

across

in achievement

gap

focus of education

research

has received

racial

and

ethnic

less attention.This

article

in school

patterns

has been

groups

but the disproportionate

Indian stu

of Black, Latino, and American

and expulsion

suspension
dents

racial and ethnic

for decades,

research

synthesizes

sanctions

and

considers

on
how

achievement
disproportionatedisciplinemight contributeto lagging
students

among

dent, school,

of color.

and

in school

patterns

gap-reducing

Keywords:

It further examines

discipline

it offers promising

and

sanctions,

for stu

to the racial and ethnic

contributors

community

the evidence

directions

for

and practices.

policies

gap;

management;

school

at-risk

classroom

students;

psychology;

student

behavior/

violence

lthoughour national discourse on racialdisparity tends

to focus

on academic

ment gap?in
States,

Black,

Latino,

outcomes?the

so-called

achieve

school districts throughout theUnited

and American

Indian

students

are also

sub

ject to a differentialand disproportionate rateof school disciplin

ary sanctions,rangingfromofficedisciplinary referralsto corporal
punishment, suspension, and expulsion (Krezmien, Leone, &
Achilles, 2006;Wallace, Goodkind, Wallace, & Bachman, 2008).
Ostensibly, the intentof school disciplinary interventions is to

preserveorder and safetyby removing studentswho break school

rulesand disrupt the school learningenvironmentand, by setting
an example of those punished students, to deter other students
from

committing

future

rule

infractions.

However,

schools

tend

to relyheavily on exclusion from the classroom as the primary

discipline strategy(Arcia,2006), and thispractice oftenhas a dis
proportionate impact on Black, Latino, and American Indian
students.The use of school exclusion as a discipline practicemay

contribute to the well-documented racial gaps in academic

achievement.This suggeststhat there is a pressingneed for schol
arlyattention to the racial discipline gap ifeffortsaddressing the

achievement

gap

are to have

income

status,

greater

likelihood

of success.

In this article,we synthesize the research on racial and eth

nic patterns in school discipline, and we suggesthow the racial
Educational Researcher,Vol.

39, No.

low

and

achievement,

rates

of misconduct

contribute towhy Black, Latino, and American Indian students

are overselected and oversanctioned in the
We
discipline system.
argue

that

such

student

characteristics

are

not

adequate

to

explain the largedisparities, and we describe school and teacher

contributors that need to be investigated in future research.
Finally,we identifymethodological challenges to the study of
disproportionality and discuss promising strategies for gap
reducing

achievement

attitude;

discipline gap influencesracialpatterns in achievement.We then

review the evidence on the factors that contribute to the disci
pline gap. Specifically,we examine the degree to which low

interventions.

Safety Efforts and Racial Disproportionality

A largebody of evidence shows thatBlack students are subject to
a disproportionate amount of discipline in school
settings,and a
smaller and less consistent literature suggests disproportionate
sanctioning of Latino and American Indian students in some
schools.1This conclusion has been drawn across awide arrayof
sanctions (e.g., suspensions, officediscipline referrals)and meth
odology (see discussion below). The Children's Defense Fund
(1975) first brought the issue of racial disproportionality to
national

attention,

showing

that Black

students were

two to three

timesoverrepresented in school suspensions compared with their

enrollment

rates in localities

across

the nation.

National

and

state

data show consistent patterns of Black disproportionality in

school discipline over thepast 30 years, specificallyin suspension
(McCarthy & H?ge, 1987; RaffaeleM?ndez, Knoff, & Ferron,

2002), expulsion (KewelRamani, Gilbertson, Fox, & Provasnik,

2007), and officediscipline referrals(Skiba,Michael, Nardo, &
Peterson, 2002). According to a nationally representativestudy
utilizing parent reports,in2003 Black studentswere significantly
more likely to be suspended thanWhite orAsian students <
(p
.001). Specifically, almost 1 in 5 Black students (19.6%) were
suspended, compared with fewer than 1 in 10WTiite students
(8.8%) and Asian and Pacific Islanders (6.4%; KewelRamani et
al., 2007). A nationally representative survey of 74,000 10th

graders found that about 50% of Black students reported that

theyhad ever been suspended or expelled compared with about
20% ofWhite students (Wallace et al., 2008). The study further
showed that,unlike thepatternforother racial and ethnic groups,
suspensions and expulsions of Black students increased from
1991 to 2005 (Wallace et al., 2008).

I,pp. 59-68

DOI: 10.3102/0013189X09357621
?2010

AERA, http://er.aera.net

This content downloaded from 152.15.236.17 on Wed, 11 Mar 2015 23:04:00 UTC
All use subject to JSTOR Terms and Conditions

JANUARY/FEBRUARY
\sT
2010~|

Although disproportionality in school discipline has been

documented forLatino and American Indian students,findings
related to such disparities have been inconsistent.National data
(U.S. Department of Education, National Center forEducation
Statistics,2003) show that,based on parent surveysadministered
in 1999, 20% of Latino students inGrades 7 through 12 had ever

been suspended or expelled,which is a statisticallysignificantly

lowerrate (p < .001) than forBlack students (35%) and a statisti
<
White students
cally significantlyhigher rate (p .001) than for
in
racial
(15%). Analyzing
disparities discipline, Gordon, Delia
Keleher
and
Piana,
(2000) found that, in 3 of the 10 cities stud
rates
the
of
ied,
suspended and expelled Latino studentswere
10% ormore than 10% higher than the percentage of enrolled
Latino students. Inconsistency infindingswas furtherconfirmed
in a studymeasuring disproportionalityusing odds ratios.Based
on state records fromMaryland, Krezmien et al. (2006) found
thatLatino students had similar or lower odds thanWhite stu

dents of being suspended for9 successiveyears (1995-2003).

National and statedata have also shown disproportionality in
discipline forAmerican Indian students, although again there
appears to be some inconsistency(Wallace et al., 2008). Krezmien

et al. (2006) showed thatAmerican Indian andWhite students

had a similar chance of being suspended from 1995 to 1998 in
Maryland. However, from 1998 to 2003, they found that
American Indians had significantlyhigher odds thanWhites of
being suspended (odds ratios ranged from 1.5 to 1.8). The dis
in American

proportionality

Indian

suspension

was

again

docu

mented in nationally representativesamples using school records

(DeVoe &C Darling-Churchill, 2008) and student reports
(Wallace et al., 2008). It isunclearwhether the inconsistentfind
ingson American Indian suspension is a statisticalartifactgiven
their relatively small numbers of suspended students (e.g.,
Krezmien et al., 2006) or if it reflectsactual variability in dispro
portionate

suspension

rates across

time and

school

districts.

of all racial and ethnic groups are more likely than

females to receive disciplinary sanctions. In 2004, only 1% of
Asian Pacific Islander femaleswere suspended, compared with
11% ofAsian Pacific Islandermales (KewelRamani et al., 2007).
White females
Expulsion data from that same year showed that
were half as likely to be expelled asWhite males (p < .001), and
similarly,Black femaleswere half as likelyto be expelled as Black
males (p < .05). Black males are especially at risk for receiving
discipline sanctions,with one study showing that Black males
were 16 times as likelyasWhite females to be suspended (J.E
Males

Gregory, 1997).
Racial Disproportionality
Achievement

and Patterns

receiveat least one suspension,which typicallyresults inmissed

for some,

could

exacerbate

aca
cycle of

demic failure, disengagement, and escalating rule breaking

(Arcia, 2006). In fact, a suspended studentmay miss anywhere
?

in greatest

need

of improvement.

Research shows that frequent suspensions appear to signifi

cantly increase the riskof academic underperformance (Davis &
Jordan, 1994). Arcia (2006) followed two demographically simi
larcohorts (matched on gender, race, grade level,familypoverty,
and limitedEnglish proficiency), contrasting a cohort thathad
received at least one suspension with another thathad received
no

suspensions.

In Year

1, suspended

students

were

three grade

levelsbehind theirnonsuspended peers in theirreading skills,but

were almost 5 years behind 2 years later.Although other unmea
sured risk factorsmay have contributed to cohort differences,
a process ofwith
suspensionmay have initiatedormaintained
drawal from learning in the classroom. In the long term,school
suspension has been found to be a moderate to strongpredictor
of dropout and not graduating on time (Ekstrom, Goertz,
Pollack, & Rock, 1986; Raffaele M?ndez, 2003; Wehlage &
Rutter, 1986).
Discipline sanctions resulting in exclusion from schoolmay
as well. Suspended
damage the learning process in other ways
studentsmay become less bonded to school, less invested in
school rulesand coursework, and subsequently, lessmotivated to
achieve academic success. Studentswho are lessbonded to school
may be more likelyto turn to lawbreaking activities and become
less likely to experience academic success. Consistent findings

highlight the importanceof school bonding for reducing the risk

of delinquency (Hawkins, Smith,& Catalano, 2004). Conversely,
Hemphill, Toumbourou, Herrenkohl, McMorris, and Catalano
(2006) found that taking into account previous violent and
aggressive behavior and a multitude of other risk factors (e.g.,
negative peer group, low grades), school suspension actually
increasedthe riskof antisocial behavior a year later. In sum, dis
some racial and
proportionate school discipline experienced by
ethnic groups has important implications for academic out

comes. There is a need for research to identify

why racialdispro
occurs
in
and
what
types of disciplinary
discipline
portionality
outcomes.
to
academic
exacerbate
be
less
likely
practicesmight

Certain

and American Indian students, albeit less consistent,have rarely

been considered in lightof thewell-documented racial and ethnic
disparities in school achievement (KewelRamani et al., 2007). In
many schools, large proportions of a group (e.g., Black males)
time and,

Horton, & Utley, 2002). The school disciplinary practices used

most widely throughout theUnited Statesmay be contributing
to lowered academic performance among the group of students

Explanations

in

The consistent pattern of disproportionate discipline sanctions

issued to Black students and the trends in sanctions forLatino

instructional

from one class period to 10 ormore school days, depending on

the violation and school policies. One of themost consistent
findings ofmodern education research is the strongpositive rela
tionship between time engaged in academic learningand student
achievement (Brophy, 1988; Fisher et al., 1981; Greenwood,

for the Racial Discipline

demographic

characteristics

that

Gap
are more

common

among some racial and ethnicgroups have been used as a primary

explanation for the racial discipline gap (see, e.g., National
Association of Secondary School Principals, 2000). Low-income
studentswith histories of low achievement,who reside in high
at greater risk for
crime/high-povertyneighborhoods, may be
in
engaging in behavior resulting officedisciplinary referralsand
school suspension. A review of the literaturesuggests that such
characteristics likelyaccount for some proportion of the gap in
sanctions

across

groups.

Yet

there is no evidence

to

suggest demo

are in anyway sufficient to "explain away" the

graphic factors
gap. Teacher and school factorsneed to be considered as possible

RESEARCHER
IEDUCATIONAL

This content downloaded from 152.15.236.17 on Wed, 11 Mar 2015 23:04:00 UTC
All use subject to JSTOR Terms and Conditions

contributors

to

the overselection

and American

Latino,

Indian

status

socioeconomic

oversanction

disproportionality (Wallace et al., 2008). Yet the highly consis

tentfinding that race/ethnicityremains a significantpredictor of
measures of fam
discipline even after statisticallycontrolling for
to explain
not
SES
is
sufficient
ily income suggests that student

of Black,

students.

Characteristics

Poverty and Neighborhood

Race,

and

(SES),

and

characteristics

of

neigh

borhoods associatedwith riskof negative outcomes are frequently

connected in the United States (Duncan, Brooks-Gunn, &
Klebanov, 1994;McLoyd, 1998). The confluence of thesefactors
makes itchallenging to separateout the contributions of each to
the racial discipline gap. Many low-income students living in
urban neighborhoods may experience adversity,such as exposure
to violence and substance abuse, which may increase the likeli
hood of their receiving school sanctions (Brantlinger, 1991;
Bureau of JusticeStatistics,2005). Although there isno evidence
that

exposure

tional

studies

to violence
show

causes

links between

behavior

difficulties,

exposure

to violence

correla
and

stu

dent mental health and behavior in the classroom (e.g.,Kuther

& Fisher, 1998). Many violence-exposed children sufferfrom
stress,and hypervigilence (Gorman-Smith&
anxiety,irritability,
Tolan, 1998). These conditionsmay have a negative effectupon

behavior in classrooms and result in increaseddiscipline referrals.

Exposure toviolence may also influencehow students cope in
school.One copingmechanism toward off the threatof violence
includes

presenting

"tough

front"

or even

arming

to

oneself

ward off futurevictimization (Anderson, 1999; Stewart,Schreck,

& Simons, 2006). The need to negotiate what Anderson has
called the "code of the street"may contribute to behavior prob

lems in school as studentsfromhigh-crime neighborhoods adjust

to a different set of norms in their interactionswith peers and
teachers in school settings (Dance, 2002). Additional research is

needed

to tease apart

community

effects

(e.g.,

concentrated

pov

erty,neighborhood crime, and the stressof low SES) and their

impact on studentbehavior in school.
It is important to distinguish, however, between the role of

poverty in predicting disruptive behavior and theways itmay

contribute to racial and ethnic disparities in discipline. Existing
school discipline research suggests that student SES is limited in
its explanatory power of the racial discipline gap (McCarthy &
H?ge, 1987;Wallace et al., 2008). Whether statisticallycontrol
ling for a measure of SES at the school level (percentage of par
ents unemployed or percentage of students enrolled in free or
reduced-cost meals; Raffaele M?ndez et al., 2002; Wu, Pink,
Crain, & Moles, 1982) or at the student level (parental education
or qualification for free or reduced-cost meals; McCarthy &c
H?ge, 1987; Skiba et ah, 2002), multivariate analyses have
repeatedlydemonstrated that racialdifferences indiscipline rates

remain significant.The most recent of these analyses (Wallace

et al., 2008) used a seriesof logisticregressionsto testracial/ethnic
disparities in officedisciplinary referrals,suspension, and expul
sion. Race/ethnicity remained a significantpredictor of all three
disciplinary
parental

outcomes
education,

even after
accounting
structure
family
(e.g.,

for student-reported
house
single-parent

hold), and urbanicityof neighborhood. In sum, being enrolled in

a schoolwith high ratesof low-income students (Raffaele
M?ndez
et al., 2002; Wu et al., 1982) or being from a low-income family
(McCarthy& H?ge, 1987; Skiba et al., 2002) does increase the
likelihood that a studentwill be subject to punitive forms of
discipline and even appears to make a mild contribution to

the racial discipline gap.

In fact, some research has found an inverse relationship
between studentdemographics and ratesof disproportionality in
school discipline. Rausch and Skiba (2004), examining suspen
sion and

expulsion
students
that Black

across

one Midwestern

are at
greater

risk of suspension

state, reported
com
when

students,not in urban schools but, rather, in

pared withWhite
more

records

suburban

resource-rich

schools.

research

Other

suggests

that the context of school or district racial climatemay have an

influence on rates of disproportionality. Thornton and Trent

(1988) reported that racial disproportionality in school suspen

sionwas greatest in schools thathad been recentlydesegregated,
especially if those schools had a higher SES student population.
Conversely, Eitle and Eitle (2004) found decreased rates of dis
proportionality in school suspension in schools that became
resegregated. Such data suggest that, at the school and district
levels,

financial

resources,

staff perceptions,

and

racial

climate

may be as importantas studentdemographics inpredicting racial

disparity.
Low Achievement
Low achievement is another variable thatmay contribute to the
racialdiscipline gap.A wide body of researchdocuments a persis
tent pattern thatAsian and WTiite students score higher on
achievement

tests

compared

with

Black,

Latino,

and American

students (A. Gregory & Weinstein,

2004; U.S.
Center
for
Education
National
of
Education,
Department
Statistics,2003). Faced with repeated academic struggles,under
performing studentsmay become frustratedand disaffected and
have lower self-confidence, all of which may contribute to a
Indian

higher rate of school disruption (Miles & Stipek, 2006). Low

literacyachievement in the elementary grades is linked to later
aggression in third and fifth grades (Miles & Stipek, 2006).
Similar patterns have been found in latergrades?low achieve
ment inmiddle and high school is linkedwith more serious
formsof aggression a year later (Choi, 2007). Although it is clear
that low achievement ishighly correlatedwith aggressivebehav
ior and disciplinary infractions, such patterns in and of them
selves do not explain disproportionality in discipline. Studies of

the relationship between achievement and student discipline

have shown thatwhen taking into account grade point average,
race remainsa predictor of suspension (Wehlage& Rutter, 1986).
Moreover, it is also possible that any relationship between the
achievement gap and the discipline gap is in fact the product of
othervariables, such as educational disadvantage. Ladson-Billings
(2006) argues thatwhat iswidely viewed as an achievement gap
between White and Black students could more properly be
termed an "education debt" in that educational opportunities in
theUnites States have historicallynever been equalized fordiffer
ent groups.McLloyd (1998) notes that poverty's effectson stu

dents are mediated not simply by family or community risk

factors but also by poor school conditions in disadvantaged
aremore likely to attend
neighborhoods. Poor students of color
schoolswith lowerquality resources and facilities (Kozol, 2005),
201Q~|f?T?
JANUARY/FEBRUARY

This content downloaded from 152.15.236.17 on Wed, 11 Mar 2015 23:04:00 UTC
All use subject to JSTOR Terms and Conditions

higher teacher turnover,and a lowerpercentage of highly quali

fied teachers (Darling-Hammond, 2004). Discrepancies in the
quality of resources available to rich and poor districts arewell
documented, but there is a need for sound policy research that
can specifyhow to address resourcedisparities in order to posi
tivelyaffectboth the achievement gap and thediscipline gap.

Differential Behavior
Another explanation for the racial discipline gap is that students
fromcertain racial and ethnic groupsmisbehave or contribute to
a lack of safety in schoolsmore than students from other racial
and ethnic groups. Studies using both measures of student self
reportand extant school disciplinary recordshave examined this
premise and have generallyfailed tofind evidence of racialdiffer
ences in student behavior (e.g., Skiba et al., 2002; Wehlage &
Rutter, 1986). In one of the earliest longitudinal studies of stu
dent race and school sanctions formisbehavior,Wehlage and
Rutter examined predictors of school sanctions for7th, 9th, and
11th graders over a 3-year period and reported that Black stu

dents did not consistently reportmore misbehavior thanWhite

students.This failure to find consistently large racial and ethnic
differences in student self-reportedbehavior has been corrobo
rated in the literature (McCarthy & H?ge, 1987; Wu et al.,
1982). A recent study using a nationally representative sample

showed few and generally smalldifferencesin self-reportedunsafe

behavior across racial groups compared with the racial discrep
ancy in discipline sanctions (Dinkes, Cataldi, & Lin-Kelly,
2007). There were, for example, no differences in self-reported

weapon carryingamong Black,White, andAmerican Indian stu

dents. Some of themost recent data on school safety (Bauer,
Guerino, Nolle, & Tang, 2008) show thatvictimization by vio
lence or theftisnot statisticallydifferentiatedby race,with simi
lar percentages ofWhite (4.7%), Black (3.8%), and Latino
(3.9%) students reporting that theyhad been victimized in the
past 6 months in school.
The

use of
self-report

data,

however,

can raise

questions

about

theaccuracy of the student reportersand hence thevalidity of the

results.Hindelang, Hirschi, andWeis (1979) hypothesized that
the failure to find differences between Black andWhite self
reportof serious delinquent behavior could be due to underre

porting by Black youth. Studies examining this hypothesis,

however,have failed to find support for it.McCarthy and H?ge
(1987) examined whether Black students,more thanWhite stu
dents, underreported their rule-breakingbehavior. Comparing
student self-report
with a sample of teacher reportsof rule break
a
ingfrom sample of 1,125 7th and 11th graders, the researchers

found no clear pattern that teacher reportswere more highly cor

relatedwith eitherWhite or Black self-reportsof misconduct,
and they concluded thatneither group tended to systematically

under-

or

over-report

their misconduct.

The findings of self-reportdata have also been corroborated

which have
by studies using extant school data on office referrals,
also failed to find substantial differences in rates of disruptive
school behavior by race.McFadden, Marsh, Price, and Hwang
(1992), studyingdiscipline records in a singleFlorida school dis
trict,found no general differencesinbehavior betweenWhite and

Black students and indeed found that

White students engaged in

a higher level of those behaviors (e.g., defiance, fighting, and

or corporal
bothering others) that tended to result in suspension
and
Braden
Shaw
(1990) reported that
punishment. Similarly,
more
White children in a single school districtwere significantly
to
action
for
for
than
Black
children
be
referred
likely
disciplinary
severe

rule violations,

despite

the overrepresentation

of Black

stu

dents in thatdistrict in corporal punishment. Finally,Skiba et al.

(2002) setout specificallyto test thedifferentialbehavior hypoth
esis, using disciplinary referralsfrom all 19middle schools in a

single largeurban district.They found no evidence that either

Black orV/hite studentswere referredto the officeformore seri
ous behaviors.The analyses did show, however, that reasons for
WTiite students tended to be for causes thatwere more
referring
objectivelyobservable (smoking,vandalism, leavingwithout per
mission, obscene language),whereas office referralsforBlack stu
dentswere more likelyto occur in response tobehaviors (loitering,
disrespect,

threat, excessive

noise)

that appear

to be more

subjec

tive in nature. In short, thereappears to be a notable paucity of

evidence thatcould support a hypothesis that the racialdiscipline
gap can be explained throughdifferentialratesofmisbehavior.

Differential Selection
In juvenile justice research, there has been a similar focus on
exploring disproportionateminority contact in the justice system
(Piquero, 2008). Some of this research has sought to identify
whether the high incarceration rates of ethnicminority youth,
compared with the ratesofWTiite youth, are due to theirhigher
rates of illegal behavior or due to institutionalpractices such as
patterns inpolice surveillance, racial profiling,or biased sentenc
ing (Piquero, 2008). This researchprovides a useful framework
for understanding discrimination as a contributor to the racial
discipline gap in schools. Specifically, the "differentialselection"
hypothesis asserts that ethnic minorities aremore likely to be
arrestedbecause theyaremore likely to be picked out forwrong

doing despite similar levels of infractions (Piquero, 2008). This

hypothesis is usefulwhen applied to the school setting; that is,
or
despite relativelysimilar rates of disruption, Black, Latino,
American Indian studentsmay be more likelyto be differentially
selected(or discipline consequences.
There is a fairlysubstantial researchbase suggesting thatdif
ferentialselection at the classroom level contributes in someway
to racial/ethnic disproportionality in school disciplinary out
comes. Consistent findings of disproportionality in office refer
rals (Skiba et al., 2002; Skiba et al., 2008; Wallace et al., 2008)
suggest that racial/ethnicdisparities in discipline begin at the
classroom level. In an ethnographic observational studyof urban

classrooms,Vavrus and Cole (2002) found thatmany office refer

rals leading to school suspension were due towhat the authors
described as a students "violation of implicit interactionalcodes,"
most often a student calling into question established classroom
practices or the teachers authority.Those students singled out in
thisway were disproportionately students of color. Skiba et al.
(2002) reportedon findings of referralsbased on objective versus
stu
subjective reasonsby race.Together with findings thatBlack
to
to
dents aremore likely thanWTiite students be referred the
office fordefiance (A.Gregory &C
Weinstein, 2008) or noncom
pliance (Skiba et al., 2008), these results strongly suggest that

RESEARCHER
~62]| EDUCATIONAL

This content downloaded from 152.15.236.17 on Wed, 11 Mar 2015 23:04:00 UTC
All use subject to JSTOR Terms and Conditions

some process of differentialselection at the classroom levelmay

contribute to disparities in discipline.
Explanations for the overselection of certain students fordis
or negative
ciplinemay include culturalmismatch, implicitbias,
in classrooms

expectations

and

schools.

The

cultural

mismatch

hypothesis suggests that the classroom culture or the teacher's

culture is at odds with the culture of ethnicminority students
(Irvine,2002; Townsend, 2000). For instance,Boykin and col
Western European-based individualism and
leagues argued that
are
thedominant underlying ideologies guiding
competitiveness
orienta
classroom activities (Boykin,Tyler,& Miller, 2005)?an
tion thatmay clashwith a strongeremphasis on communal val
ues inBlack, Latino, and American Indian culture (Gay, 2006).

Gay

further

that

suggested

communicative

tensions

can

arise

through cultural difference. Specifically,differences inways of

communicating

between

Blacks

(e.g.,

animated,

interpersonal)

(e.g., dispassionate, impersonal)may lead to conflict

(Kochman, 1981). In a study of 62White elementary teachers
who taught in two predominantly Black schools,Tyler, Boykin,
andWalton (2006) found that teacherswere more likely to rate
andWhites

vignettesof studentswho exhibited competitive and individual

istic behavior

as motivated

and

achievement

oriented

than

stu

dents who exhibited more communal and vervistic (e.g.,

collaborative and multitasking) behaviors. Such findings, ifvali
dated in actual classroom settings,would indicate a differential
perception on the part of teachers that could well advantage
WTiite students exhibiting competitive behaviors and disadvan
tage Black students exhibiting a more active and community
oriented learning style.
Other scholarshave focused onways inwhich negative teacher
beliefs and expectations can contribute to racially relatedauthor
ityconflicts (R. S.Weinstein, 2002; R. S.Weinstein, Gregory,&
Strambler,2004). In her ethnography of school discipline in an
elementary school, Ferguson (2000) observed patterns in nega
tive teacher?student

interactions

and

argued

that

these

events

were fueledbyWhite teachers'overreactingand relyingon stereo

types to interpretBlack students' language and physical expres
sion. Given stereotypesand media portrayals of Black youth as

dangerous and aggressive (Devine & Elliot, 2000; Noguera &

Akom, 2000), teacher expectations for behavior may also influ
encewhether these students are selected fordiscipline sanctions.
A related area of researchexamines how implicitbeliefsmay neg
atively affectBlack and Latino students. Implicit racial bias,
out of conscious
according to social psychologists, operates
awareness yet influencesdecision making (e.g.,Dovidio, Glick,
& Rudman, 2005). Although no studieshave been conducted on
the implicit bias of teachers and how racemay activate stereo
types,Graham and Lowery (2004) conducted an analogous
experimental studywith police and probation officers.They
found that, compared with officerswho were subliminally
primed

with

neutral,

non-race-related

words,

officers who

had

been subliminally primed with words related to the category

BUck were more likely to recommend harsher punishments for
adolescentswho had committed crimes,as presented in standard
ized, written
Taken

vignettes.
research
together,

on

classroom

processes

suggests

that

Black students are differentiallyselected for discipline referral

(e.g., Skiba et al., 2002), although there is insufficientdata to

establishwhy thismay occur. Several reasonsmay include societal

stereotypes,implicitbias, or culturalmismatch between teachers
and Black

To

students.

advance

research

in this area,

systematic

line ofmixed-methods research is needed, using observational

studies

interactions

of classroom

and

interviews

of teachers

and

students concerning the process of school discipline. Coding of

whether teach
teacher-student interactionscould help identify
ers aremore or less tolerant of racially specific deviations from
implicitbehavioral standards in the classroom.
Differential Processing
The differential
processinghypothesis asserts that discrimination
occurs

in the courts

and

arrest

disproportionate

leads to a
systems, which
rate of minorities

correctional

incarceration

and

(Piquero, 2008). Subjective judgments in sanctioningmay be det

rimental toBlack, Latino, and American Indian youth.Morrison
(Morrison et al., 2001; Morrison & Skiba, 2001) noted that the
as
application of school consequences such suspension and expul
less a discrete

sion represents

outcome
teacher

event

than a complex

process

whose

is influenced simultaneously by student behavior,

classroom

management,

administrator

perspectives,

and

school policy.There is tremendous local flexibility in the typesof

infractions thatmove forward from the classroom to the office
and in the typesof consequences issued by administrators.The
Gun-Free Schools Act of 1994 mandates a 1-yearexpulsion for
thepossession of firearmsat school, but such consequences can be
modified based on the discretion of the district administration.
Thus, in general, there is considerable flexibility in the type and
an infraction.For the same
lengthof sanction students receivefor
offense,

one

administrator

may

decide

to mandate

a conference

with parents or guardians; a differentadministratormay mandate

a 5-day suspension (Noguera & YonemuraWing, 2006).

The most well-documented gap in sanctions isbetween Black

andWhite students.Wehlage and Rutter (1986) found that
Black studentswere more likely thanWhite students to report
were more likely than
s
being sent to the principal office and
White students to reportbeing suspended even though theydid
not report higher incidents of misbehavior, across 2 years of
study.These findings suggest a discrepancy between sanctions
and student-reportedbehavior. Indeed, itmay be thatBlack stu
dents are suspended and punished forbehavior that is less serious
than the behavior of other students.McFadden, Marsh, Price,
and Hwang (1992) reported thatBlack pupils in a Florida school
districtwere more likely thanWhite students to receive severe
punishments (e.g., corporal punishment, school suspension) and
less likely to receive milder

consequences

(e.g.,

in-school

suspen

sion). These results are consistentwith findings that Black stu

dents were referredfor corporal punishment for less serious
behavior than were other students (Shaw & Braden, 1990).
These findings, as a whole, suggest harsh sanctions issued to
Black

students

may

contribute

cipline data.
Methodological

to their
overrepresentation

in dis

Issues and Recommendations

seems
Although the concept of disproportionate representation
straightforward,

its measurement

can

be

complex,

as demon

strated in special education research (Skiba et al., 2008). The com

position index (Donovan &CCross, 2002) compares theproportion
~
JANUARY/FEBRUARY
20?cTj[ 3~

This content downloaded from 152.15.236.17 on Wed, 11 Mar 2015 23:04:00 UTC
All use subject to JSTOR Terms and Conditions

of those served in special education representedby a given ethnic

group with the proportion thatgroup represents in the popula
tion or in school

enrollment.

For

example,

Black

students

account

for33% of students identifiedasmentally retardedat thenational

level, clearly discrepant from their 17% representation in the
school-aged population (Donovan & Cross, 2002). Although an
intuitivemeasure, problems with interpretationand scaling of
the composition indexmeasure have led the field toward use of

the riskindexand riskratio (Coutinho & Oswald, 2000; Skiba et

al., 2008; Westat, 2005). The risk index is the proportion of a
given group in a given category; at the national level,2.64% of
all Black students enrolled in thepublic schools are identifiedas
mentally retarded (Donovan & Cross, 2002). To interpretthe
risk index, a ratio of the riskof the targetgroup to one ormore
groupsmay be constructed, termeda riskratio (Hosp & Reschly,

2003; Parrish,2002). Comparison of Black student riskfor iden

tificationasmentally retarded (2.64%) with the
White risk index
of 1.18% for thatcategoryyields a riskratio of2.24 (2.64/1.18),
suggesting thatBlack studentsare over two timesmore likely to

be served in thecategorymental retardationthan

White students.
The same data can also be used to compute an odds ratio (Finn,
1982), oftendrawn from logisticregression(Wallace et al., 2008).
In contrast
rence

to the risk ratio,

and nonoccurrence

the odds

ratios

assesses

both

occur

data.

Methodological issues in themeasurement of disproportion

ality remain outstanding, including criteria for determining a
significantlevelof disproportionality (Bollmer,Bethel, Garrison
Mogren, & Brauen, 2007; Skiba et al., 2008), the appropriate

comparison group when calculating risk ratios (Westat, 2004),

and the comparability of riskand odds ratios (Davies, Crombie,
& Tavakoli, 1998). In the face of national special education law
mandating the identificationof significantdisproportionality at
the local level,however, criteriaformaking that determination
are necessary.Thus, theU.S. Department of Education Office of
Special Education Programs issued policy guidance to state and

local education agencies regardingthe calculation and interpreta

tion of risk indicesand riskratios (Westat,2004,2005), which has
implications for how disproportionality in discipline sanctions
could be identified.The Office of Special Education Programs
recommends that a risk ratio can be used to understand the rela

tive riskof students receivingspecial education servicesfordiffer

ent racial and ethnic groups (Westat,2005). The office cautions,
when based on
however, that risk ratios are difficult to interpret
small numbers of students in a racial and ethnic group. It further
describes the benefits of a weighted risk ratio,which takes into
account differences in the size of racial and ethnic groups. This
allows for comparison of risk ratios across districtswith varying
racial and ethnic composition.
Improved measurement of the racial discipline gap should

advance

substantive

areas

of inquiry. One

important

area

relates

to the unique contributions of student, teacher,school, and fam

ilyand neighborhood to the racial discipline gap. As of yet, there
have been no comprehensive studies or systematic lines of
research thathave disentangled the unique effectsof these con
tributors.Education researchers
might follow the lead of a recent
study by Sampson, Morenoff, and Raudenbush (2005) on the
gap in community violence betweenWhite, Black, and Latino
young adults, which offers a guide for ecologically sensitive

researchon race and discipline. Using data from almost 3,000

young adults in 180 Chicago neighborhoods, Sampson and col
leagues identifiedthe unique contributions of individual,home,
and neighborhood variables to the relativeodds of self-reported
violence for each racial and ethnic group. The apparentmulti
level causation of disciplinary disproportionality strongly sug
that multivariate

gests

in

procedures,

hierarchical

particular

approaches (Raudenbush & Bryk, 2002), will bemost appropri

ate

in future

next

The

research.

of research

generation

could

simultaneouslyconsider the effectsof studentattitudeand behav

ior, teacher

tolerance

classroom

and

management

skills, adminis

trative leadership, school climate, and school and community

demographics on the racial discipline gap.
Following the lead from researchon the juvenile justice system
on the chain of
(Piquero, 2008), systematic lines of research
events

in

that culminate

and

suspension

are needed.

expulsion

Unfair selection and sanction at various points in thediscipline

process could additively contribute to thediscipline gap.Another
area

crucial
theory

of research

needs

the conscious

regarding

to test mechanisms
unconscious

and

and

develop
that

processes

result in differentialtreatmentof some racial and ethnic groups.

Previous

has

research

teachers

and

students

shown

that

cultural

can contribute

to

mismatch

between
fear,

misunderstandings,

and conflictwith respect to pedagogy (Irvine, 2002; Ladson

on the
Billings 1995; Pollack, 2008); furtherresearch is needed
extent towhich such processes also contribute to inequitable dis
ciplinary

Social

practices.

class,

immigrant

status,

racial

and

eth

nic identity,neighborhood and familial diversity,and educator

trainingand perspectivesmay all affectstudentbehavior, teacher
or their interaction.

responses,
could

truly

sort out

Clearly,

the numerous

and

conducting
interacting

research
sources

that

of vari

ance contributing to disciplinary disproportionality is challeng

ing. Subtle and implicitprocesses related to racial bias, negative
expectations,

or

are not

stereotypes

easily detected

outside

of con

trolled laboratory conditions, and it is not a simplematter to

observe

the complex

and

interactive

social

processes

that can con

tribute to an escalating sequence of actions and reactionsduring

actual

discipline

encounters.

Identifyingthe characteristics of resilient schools is another

importantnext step in researchon racial and ethnic disparities in
school discipline. In thefield of public health, researchhas estab
lished a strong linkbetween community violence andmanifesta
tions of school violence (Ozer, 2005). Not surprisingly,schools
in areaswith a high incidence of crime and violence also tend to

experiencehigher ratesof violence and disorder (Noguera, 2003).

Yet the presence of schools thatdemonstrate positive outcomes
despite their location in high-risk neighborhoods (e.g.,Welsh,
Greene, & Jenkins, 1999) strongly suggests thatneighborhood
and familydisadvantage be approached in researchand practice
as conditions that increase educational challenge, rather than as

a
limitingconditions. In particular, there is need for additional
research on the types of strategies schools can implement to
reduce the effectsof violence inneighboring communities.

Disciplinary Practices,
School Reform

Prevention

Programming,

and

Existing researchon the racialdiscipline gap suggeststhat,similarto

effortsthataddress the achievement gap or thedisproportionate

RESEARCHER
~6?\\ EDUCATIONAL

This content downloaded from 152.15.236.17 on Wed, 11 Mar 2015 23:04:00 UTC
All use subject to JSTOR Terms and Conditions

number ofBlack studentsplaced in special education (Skiba et al.,

2008), no single causal factor can fullyexplain raciallydisparate
discipline, and no single action will thereforebe sufficient to

ameliorate it.Multifaceted strategies

may offerpromise, but there
is as yet no empirical research testing specific interventionsfor

reducing the discipline gap.

Given the lack of systematicresearchaddressing the effective
ness of gap-reducing interventions,promising directionsmust be
extrapolated

from

implemented

with

intervention

other

research.

Freiberg

and

Lapointe (2006) reviewed 40 school-based programs targeting

the reduction of behavior problems in schools.Of those, 29 were
Latino,

Black,

low-income

and

urban,

stu

dents and offered some evidence for their success in increasing

studentproblem solvingand/or reducingdifficultiesin classroom
as a whole.

for participants

management

and

Freiberg

Lapointe

identified commonalities among those effectiveprograms. The

programsmove beyond discipline, emphasizing student learning
and self-regulation,not simply procedures for addressing rule
infractions.
and

Overall,

"school

encourage

They

connectedness"

between

trusting
relationships"
to increase
the programs
try

of

and

schooling
tions to misbehavior.

to move

teachers
students'

from a reliance

away

and
and

"caring
students.

experience
positive
on
reac
punitive

The programmatic commonalities described by Freiberg and

Lapointe (2006) offer a promising direction for lowering the
of Black,

oversanctioning
Yet universal

and American

Latino,

to educational

approaches

Indian

practice

have

students.
frequently

been critiqued fornot specificallyaddressing the racial dynamics,

economic

or other

Stressors,

influences

on

the racial

discipline

gap (Goldstein & Noguera, 2006). In a national sample of

schools at the elementary and middle school level that imple
mented positive behavior supports for at least a year, Skiba et al.
(2008) reportedgenerallypositive findings before disaggregation
by race but significantdisciplinary disproportionality forBlack
and Latino students in both office disciplinary referralsand
administrative
Explicit

when

consequences

the data were

to issues of race and

attention

culture may

disaggregated.
be necessary

for sustained change in racial and ethnic disciplinary disparities.

Studies of successful teachers of Black students support the
idea that teachersdifferfromone another in theirability to elicit
cooperation and diffuse conflict.A. Gregory and Weinstein
(2008) found that teacherswho elicited trustand cooperation
with theirBlack students tended to use an authoritative styleof
teaching?one

inwhich teachers showed both caring and high

These

expectations.

"warm

demanders"

(Irvine,

2002)

may

pro

vide cultural synchronizationbetween authority in thehome and

in the school.
expressions

use

Teachers'

are other

of humor,

avenues

occur

through

emotions,
which

and

cultural

colloquial
synchrony

(Monroe & Obidah, 2004; C. S. Weinstein,

may
Tomlinson-Clarke, & Curran, 2004). Additional research on
preservice teacher training and professional development is
needed

to ascertain

if an

increase

in teacher

cultural

responsive

ness or synchronywith students is linked to lower discipline

referrals

for Black,

Latino,

and American

Indian

students.

Overall, little isknown about the typesof interventionsthat

reduce the racialdiscipline gap.Given the researchon possible con
tributorsto thegap, a varietyof strategies
may be needed, includ
awareness
the
of
teachers
and administratorsof
ing (a) increasing

the potential for bias when issuing referralsfor discipline, (b)

a

utilizing

to behavior

in response

of consequences

range

prob

lems, (c) treatingexclusion as a last resortrather than the firstor

only option, (d) making a concerted effort to understand the
roots of behavior problems, and (e)
findingways to reconnect
students to theeducational mission of schools during disciplinary
events (Noguera, 2007).

Summary
The racial and ethnic disparity in discipline sanctions has not
it deserves.

the attention

received

Few

studies

have

examined

where andwhy disproportionality between Black andWhite stu

dents is on the increase, especially for Black females (Wallace
et al., 2008). Discipline trends for Latinos have been inconsis
tentlydocumented. Given the diversityof Latinos in theUnited
States

status,

immigrant

(e.g.,

country

of

origin),

in-depth

exam

inations of differentLatino groups is needed (e.g., first-genera

tion Mexican
Moreover,

American,
of

comparisons

Cuban

third-generation
schools with

racial

American).

diversity

versus

racial homogeneity would be informative.Such researchwould

then lend itselfto inquiry about why such trendsexist in school
discipline.
Unfortunately, the discourse on racial and ethnic dispropor
tionality seems to be constrained by simplisticdichotomies that
artificiallypit individual student characteristics (e.g., student
aggression,disengagement from school) against systemicfactors
(e.g., school administrators' implicit bias, community violence)
as the reason
why

some

groups

are

in

overrepresented

suspension

or expulsion (Skiba et al., 2008). The

multiple and interacting
variables thatappear to contribute to racial and ethnic disparities
in discipline demand a more comprehensive and nuanced

approach. More sophisticated statisticalmethodologies such as

hierarchical linearmodeling or sequential analysis (Gottman&
Roy, 1990) may prove to be better suited formodeling the com
plexityof inequitable outcomes in school discipline.
At this time, however, little is known about the efficacyor
effectiveness

of possible

of interventions

interventions. What

"gap-reducing"
successfully

might

increase

teacher

and

types

adminis

tratorawareness of thepotential forbiaswhen

issuingreferralsfor
at
as a last
Do
interventions
aimed
exclusion
discipline?
using
resortratherthan thefirstor only option reduce thegap in refer
rals across racial and ethnic groups?Will interventionsaimed at
the achievement

reducing

gap,

such as access

to

curricu

rigorous

lum and caring teacher-student relationships,be accompanied by

a narrowed discipline gap? Can
gap-reducing interventionsdraw
on universal approaches, or do theyneed
targeted,culturallyspe
cific approaches that respond to the students' cultural and socio
economic contexts?Effectivelyaddressing thesequestions poses a
serious

challenge

to researchers,

as it

necessarily

involves

attention

to thecomplex, politically charged, and oftenpersonally threaten

ing topic of race.Yet creativityand perseverancewill be necessary
to craft such

research

ifwe

are to understand

and develop

inter

ventions thatcan effectivelyreduce the racialdiscipline gap.

NOTE
barely
alternative

does

research differentiate

educational

result, this review must

rely

expulsion
resulting in
from such services. As a

between

services or exclusion

on a broad usage of the term

expulsion.

2010 |[ 5~
JANUARY/FEBRUARY

This content downloaded from 152.15.236.17 on Wed, 11 Mar 2015 23:04:00 UTC
All use subject to JSTOR Terms and Conditions

REFERENCES
street:
E. (1999). Code
Decency, violence, and the moral
of the
life of the inner city.New York: W. W. Norton.
and enrollment status of suspended stu
Arcia, E. (2006). Achievement
in a
dents: Outcomes
multicultural
school district. Education
large,
and Urban Society, 38, 359-369.

Anderson,

Bauer, L., Guerino,

ization in U.S.

K. L.,
S. (2008). Student victim
&Tang,
school crime supplement to
schools: Results from the2005
P., Nolle,

theNational Crime Victimization Survey (NCES 2009-306).

DC:

Washington,
of Education

National

Center

Sciences, U.S.

Institute

Statistics.

for Education

of Education.

Department

&

M.

Brauen,
R,
(2007).
J., Bethel, J., Garrison-Mogren,
to assess racial/ethnic
in special
Using the risk ratio
disproportionality
education at the school-district level. Journal of Special Education, 41,
186-198.

Bollmer,

In search of cultural
(2005).
Boykin, A. W, Tyler, K. M., & Miller, O.
themes and their expressions in the dynamics of classroom life.Urban
Education,
40, 521-549.
E.

in adolescents'
(1991). Social class distinctions
reports of
in school. Behavioral Disorders, 17, 36-46.
and punishment
as
Brophy, J. (1988). Classroom management
socializing students into
roles.
1?4.
articulated
Classroom
Interaction,
33(1),
Journal
clearly
of
Bureau of Justice Statistics. (2005). Crime victimization, 2005. Retrieved

Brantlinger,

over
J. (2004).
Inequality, segregation, and the
in
Americans
school
African
of
suspensions. Sociological
representation
Perspectives, 47, 269-287.

Eitle, T. M., &

problems

Ekstrom,

Defense

Fund.

suspensions: Are they helping

Research Project.
and problem behaviors among

School

(1975).

Asian

Pacific

Islander American

Adolescence,

36, 403-415.

M.

J.,& Oswald,

Coutinho,

L.

Tough fronts: The

(2002).

London:

Journal

representa
Disproportionate
Journal
synthesis and recommendations.

ofChild andFamilyStudies,9, 135-156.

Dance,

of Youth and

P. (2000).

D.

tion in special education:

adolescents.

impact of

street culture on

schooling.

L. (2004). Inequality and the right to learn: Access

Darling-Hammond,
to
in California's
teachers
qualified
public schools. Teachers College
Davies,

H.,

J. E., &

M.

L, &Tavakoli,

Crombie,

British Medical

mislead?
Davis,

1936-1966.

106,

(1998). When

Jordan,W.

ratios

effects of school context, struc

P. G., & Elliot, A. J. (2000). Are

trilogy revisited. In C.

inmiddle

and high

racial stereotypes

really fading?
Stangor (Eds.), Stereotypes and
New York: Psychology Press.
prejudice: Essential readings (pp. 86-99).
.E.
DeVoe,
(2008). Status and trends in the
J. F., & Darling-Churchill,
education ofAmerican Indians and Alaska Natives: 2008 (NCES2008
The

Princeton

084). Washington,
Institute of Education
Dinkes,

DC:

for Education

Sciences, U.S.

E. F., &

R., Cataldi,

Center

National

Statistics,

of Education.

Department
(2007).
Lin-Kelly, W.

Indicators

of school

crimeand safety:
2007 (NCES2008-021/NCJ219553). Washington,

DC:

National

Center

for Education

Statistics,

Institute of Education

Sciences, U.S.

of Education,
and Bureau of Justice.
Department
M.
students in
Donovan,
S., & Cross, C. T. (Eds.).
(2002). Minority
DC: National
Academies
special and gifted education. Washington,
Press.
Dovidio,

J. E, Glick,

P. G., &

Rudman,

L.

(2005).

On

the nature of

Maiden, MA: Blackwell.

Allport.
prejudice:Fiftyyearsafter
Duncan,

G.

J., Brooks-Gunn,

deprivation
65, 296-318.

and

J.,&

Klebanov,

early childhood

(1986).

Public schooland themakingofBhck

A. A. (2000). Bad boys:
Ferguson,
Press.

masculinity. Ann Arbor: University ofMichigan

Finn, J. D.
(1982). Patterns in special education
survey. In K. A. Heller, W.

by the OCR

as revealed

placement
&
H. Holtzman,

S. Messick

(Eds.), Placing childrenin special education:A strategy

for equity
National Academy Press.
N.
N., Marliave, R., Cahen,
Filby,
DC:

(pp. 322-381). Washington,

Berliner, D. C,
Fisher, C. W,
&

M.

Dishaw,

time, and

M.

student
J.,&

Teaching
An
achievement:
J.M.

Lapointe,

and

preventing
C. S.Weinstein

academic

L. S.,
learning

of Classroom

overview. Journal

Connections

(2006).

(2006).

solving discipline
(Eds.), Handbook
NJ: Lawrence

786). Mahwah,
Gay, G.

behaviors,

(1981).

17( 1 ), 2-15.

Interaction,
Freiberg, H.

Research-based

problems.

programs
Evertson

In C. M.

ofchssroom management
Erlbaum.

for
&

(pp. 735

and cul
classroom management
Evertson & C. S. Weinstein

between

In C. M.

turally responsive teaching.

classroom management
(Eds.), Handbook
of
NJ: Lawrence Erlbaum.

(pp. 343-372).

Mahwah,

for diversity:
P. (2006, Spring). Designing
& Noguera,
can use cultural competence
in developing
substance
abuse prevention programs for urban youth. New Directions for Youth
pp. 29-40.
Development,

Gordon,

M.,

educators

R., Delia

Pia?a,

consequences: An
schools. Oakland,
Gorman-Smith,
violence

D.,

L., &

examination

Keleher,

T.

Facing the
in U.S.
public

(2000, March).

of racial discrimination
Research Center.

CA: Applied
& Tolan,
P. H.

The

(1998).

and

among
problems
developmental
and Psychopathology, 10, 101-116.
(1990).
J. M., & Roy, A. K.
Sequential

P. K.

development.

Gottman,

for behavioral
Press.
Graham,

S., &

types about

to

role of exposure
inner-city

youth.

(1994).

Economic

Child Development,

researchers. Cambridge,

UK:

analysis: A guide

Cambridge

University

racial stereo
Lowery, B. S. (2004). Priming unconscious
offenders. Law and Human
adolescent
Behavior, 28,

483-504.
C. R., Horton,

Greenwood,

ture, and experiences on African American males

schools. Journal ofNegro Education,
63, 570-587.
Devine,

can odds

989-991.

Journal, 316,
J. (1994). The

Rock, D. A.

&

J.M.,

Development

Routledge.

Record,

E., Pollack,

study. Teachers College Record, 87, 356-373.

How

children? Cambridge, MA: Washington

achievement
Choi, Y. (2007). Academic

R. B., Goertz, M.

Who drops out of high school andwhy? Findingsfroma national

April 20, 2009, fromhttp://www.ojp.usdoj.gov/bjs/pub/pdf/cv05Goldstein,

.pdf
Children's

Eitle, D.

B. T., &

Current

Utley, C. A. (2002). Academic

on research and
practice. School

engagement:
perspectives
Psychology Review, 31, 328-349.
& Weinstein,
R. S. (2004).
Gregory, A.,
home

and

Connection

and regulation at
for adoles

in achievement

in school:

Predicting growth
cents. Journal
ofAdolescent Research, 19, 405-427.
& Weinstein,
R. S. (2008). The discipline gap and African
Gregory, A.,
or
in the
Americans:
Defiance
cooperation
high school classroom.

46, 455-475.
JournalofSchoolPsychology,

strikes and they're out: African American

J. F. (1997). Three
International
American
schools'
and
responses to misbehavior.
boys
Journal ofAdolescence and Youth, 7, 25-34.

Gregory,

Gun-Free

Schools Act of 1994, 20 U.S.C.

free requirements
Hawkins,
ment

J.D.,

Chapter

70, Sec. 8921 Gun

(1994).

Smith, B. H,

& Catalano,

and social and emotional

R. F. (2004). Social develop

learning. Inj. E. Zins, R. P.Weissberg,
(Eds.), Building academic success on

C. Wang, & H. J.Walberg

social and emotional learning: What does the research say? (pp. 135-150).
Press.
New York: Teachers College

M.

S. A., Toumbourou,
J.W, Herrenkohl, T. I., McMorris,
Hemphill,
R. F. (2006). The effect of school suspensions
J., & Catalano,
arrests on
behavior inAustralia
subsequent adolescent antisocial
the United States. Journal ofAdolescent Health, 39, 736-744.

?? IEDUCATIONAL
RESEARCHER

This content downloaded from 152.15.236.17 on Wed, 11 Mar 2015 23:04:00 UTC
All use subject to JSTOR Terms and Conditions

B.
and
and

M.

Hindelang,

& Weis, J.G. (1979). Correlates of delin

J.,Hirschi,T.,
illusion of discrepancy
between self-report and official

The

quency:
measures.

American

Sociological Review, ^, 995?1014.

&
L.,
J.
Hosp,
Reschly, D. J. (2003). Referral rates for intervention and
assessment: A
of racial differences. Journal
meta-analysis
of Special
Education,

37, 67-81.

In search
teachers and
of wholeness: African American
their culturally competent classroom practices. New York:
Palgrave.
S. (2007).
A., Gilbertson,
KewelRamani,
L., Fox, M., & Provasnik,
Irvine, J. J. (2002).

Status and trends in the education

2007-039).
Statistics,

minorities (NCES
of racial and ethnic
DC:
Center
National
for Educational
Washington,
Institute of Education
U.S.
of
Sciences,
Department
Retrieved

Education.

February

from

3, 2009,

University

Press.

Chicago

Kozol, J. (2005). The shameof thenation:The restoration

ofapartheid
inAmerica. New York: Crown.
schooling
Krezmien, M. P., Leone, P. E., & Achilles,
race, and disability:
Journal ofEmotional
Kuther, T. L., &

G. M.

of statewide

Analysis
and Behavioral

(2006).

Suspension,
and reporting.

practices

Disorders,

14,217-226.

Fisher, C. B.

(1998). Victimization
by community vio
lence in young adolescents
from a suburban city. Journal of Early
Adolescence, 18, 53-76.

G. (1995). But that's just good teaching! The case for

Ladson-Billings,
culturally relevant pedagogy. Theory Into Practice, 34, 159-165.
G. (2006). From the achievement gap to the education
Ladson-Billings,
debt:

in U.S.

achievement
Understanding
3-12.

schools.

Educational

Researcher, 35(7),

J.D., & H?ge, D. R. (1987). Social construction of school

Social Forces, 65, 1101-1120.
punishment.
A.
Y. (1992). A
C,
Marsh, G. E., Price, B. J., & Hwang,
McFadden,
race and
in
the
of
bias
of
study
gender
punishment
handicapped

McCarthy,

school children. Urban

Review, 24,

239-251.

V. C. (1998). Socioeconomic
McLoyd,
disadvantage
ment. AmeHcan
Psychologist, 53, 185-204.
S. B., & Stipek, D.
Miles,
(2006). Contemporaneous
associations

between

sample of low-income
103-117.

elementary

and child develop

and

longitudinal
in a
literacy achievement
school children. Child Development,

social behavior

and

77,

influence of cultural syn

J. E. (2004). The
a teacher's
of disruption.
Journal
perceptions
of
Teacher Education, 55, 256-268.

Monroe,

C. R., & Obidah,

on

chronization

G. M.,

Anthony, S., Storino, M., Cheng, J., Furlong, M. E,

as a process and an
R. L. (2001).
School
expulsion
event: Before and after effects on children at-risk for school disci

Morrison,

& Morrison,

pline.

New

Directions

Youth Development:

for

Theory,

Practice,

Research, 92, 45-72.

&

G. M.,

Morrison,

Skiba,

R.

Promises

school misbehavior:

violence

from

J. (2001).
Predicting
and perils. Psychology in the Schooh, 38,

173-184.
National

Association

of Secondary

School

(2000, February).
Principals.
on civil
zero tolerance
Statement
programs.
rights implications
of
to the United
on Civil
States Commission
Testimony
presented
DC.
Rights, Washington,

P. A. (2003). City schooh and

Noguera,
Teachers College Press.
Noguera,

P. A.

(2007).

How

theAmerican
to students

listening
improve. Theory Into Practice, 46, 205-211.
P. A., & Akom, A.
(2000). The
Noguera,
Quarterly, 24,

86.

dream. New
can

help

York:

schools

Closing

Jossey-Bass.
E.
(2005).

to

on urban
The
adolescents:
impact of violence
effects of perceived
and family sup
school connection
ofAdolescent Research, 20, 167-192.

Ozer,

Longitudinal

port, journal
Parrish, T. (2002).

in the identification,
Racial disparities
funding, and
In
&
of
D.
Losen
G.
education.
Orfield
(Eds.),
J.
provision
special
Racial
MA:
inequity in special education
(pp. 15-37). Cambridge,
Press.
Harvard Education
A.

Piquero,

R.

Raffaele

(2008).

M?ndez,

gap. Wilson

Disproportionate

contact.

minority

Future

of

Everyday
L. M.

anti-racism.

London:

New

Press.

of

and
suspension
In J.Wald
investigation.

Predictors

(2003).

outcomes: A
negative school
longitudinal
& D. J. Losen
(Eds.), New directions for youth development: No. 99.
the school-to-prison pipeline (pp. 17?34). San Francisco:
Deconstructing
Jossey-Bass.
Raffaele M?ndez,

L. M.,

&

Knoff, H. M.,

(2002). School
rates: A
demographic
suspension
quanti
tative and qualitative
analysis of large ethnically diverse school dis
in the Schools, 39, 259-276.
trict.
Psychology
variables

& Bryk, A. S. (2002). Hierarchical

linear models:
and data analysis methods (2nd ed.). Thousand
Oaks, CA:

Applications
Sage.

Ferron, J.M.

and out-of-school

S.W,

Raudenbush,

outcomes:
.,& Skiba, R. J. (2004).
Suspensions
Unphnned
IN: Center
for Evaluation
expulsions in Indiana.
Bloomington,
and Education
Policy. Retrieved July 21, 2004, from http://ceep.indi
ana.edu/ChildrenLeftBehind

Rausch,
and

S. (2005). Social anat

J.D., & Raudenbush,
Sampson, R. J.,Morenoff,
in violence. American Journal
omy of racial and ethnic disparities
of
Public Health, 95, 224-232.
Shaw, S. R, &

Braden, J. B. (1990). Race and gender bias in the adminis

tration of corporal punishment. School Psychology Review, 19, 378-384.
R. S., Nardo, A. C, & Peterson, R. L. (2002). The
Skiba, R. J.,Michael,
color of discipline: Sources of racial and
in
gender disproportionality
school punishment. Urban Review, 34, 317?342.
. .,&
Skiba, R J., Simmons, A. B., Ritter, S., Gibb, A. C, Rausch,
in
Cuadrado,
J. (2008). Achieving
equity
special education: History,
status, and current challenges. Exceptional
Children, 74, 264-288.
Stewart, E. A.,

"I ain't gonna let

Schreck, C. J.,& Simons, R. L. (2006).
the code of the street reduce or increase
disrespect me": Does
violent victimization
adolescents? Journal of
among African American
Research in Crime and
43, 427-458.
Delinquency,
no one

C. H., & Trent, W

Thornton,
sion

in East

American

(1988).

School

Parish: A

Baton

Negro Education,
B. L.
Townsend,

Rouge
57, 482-501.

(2000).

The

learners: Suspensions

and suspen

desegregation

report. Journal

preliminary

disproportionate
and expulsions.

discipline
Exceptional

of

of African
Children,

66,381-391.
T. R. (2006). Cultural
consid
Tyler, K. M., Boykin, A. W, & Walton,
erations in teachers' perceptions
of student classroom behavior and
and Teacher Education: An International Journal
Teaching
Research
and
22, 998-1005.
Studies,
of
U.S. Department
of Education, National Center forEducation
Statistics.
achievement.

(2003). Statusand trendsin theeducationofHispanics(NCES 2003?

008). Washington,
E, & Cole,

Vavrus,

construction
opportunity

(2008).

18, 59-79.

Children,
Pollack, M.

http://nces.ed.gov/

pubs2007/2007039.pdf
Kochman, T. (1981). Black and White stylesof conflict.Chicago:
of

.A., & Yonemura

business:
J. (2006).
Wing,
Unfinished
the racial achievement gap in our nations schools. San Francisco:

Noguera,

Wallace,
(2008).

J.M.,

DC:

.
M.

of school

Author.
(2002).

Jr.,Goodkind,

Racial,

ethnic,

"I didn't do nothin":

suspension.

Urban

S., Wallace,

and gender

Review, 34,
C. M.,

differences

&

The

discursive

87-111.
Bachman,

in school

J. G.

discipline

2010 |[ 7~
JANUARY/FEBRUARY

This content downloaded from 152.15.236.17 on Wed, 11 Mar 2015 23:04:00 UTC
All use subject to JSTOR Terms and Conditions

among U.S.

high
Review, 59, 47-62.

school

students:

1991-2005.

Negro

Educational

out: How much do

G. G., & Rutter, R. A. (1986). Dropping
Wehlage,
schools contribute to the problem? Teachers College Record, 87,374-393.
C.

Weinstein,

S., &

S., Tomlinson-Clarke,

conception of culturally responsive

ofTeacher Education, 55, 25-38.

Curran, M.

(2004).

classroom management.

Toward
Journal

R. S. (2002). Reaching higher: The power of expectations in

Press.
schooling. Cambridge, MA: Harvard University
R.
&:
M.
Intractable self
Weinstein,
S., Gregory, A.,
Strambler,
(2004).
v.
Brown
Board
after
Fifty years
of Education.
fulfilling prophecies:
Weinstein,

American Psychologist, 59, 511?519.

Welsh, W. N., Greene, J. R., & Jenkins, P. H.
The
influence of individual,
institutional,

(1999).

School

and community

disorder:
factors.

Criminology, 37, 601-643.

Inc. (2004). Summary of taskforce meeting on racial!ethnic dispro
DC: Author.
portionality in special education. Washington,

Westat,

Inc.

(2005). Methods

Westat,
for assessing racial!ethnic disproportionality
DC:
in special education: A technical assistance guide. Washington,
U.S. Department
of Education Office of Special Education
Programs.
Retrieved

October

5, 2006,

from https://www.ideadata.org/docs/

Disproportionality%20Technical%20Assistance%20Guide.pdf
S., Pink, W, Crain, R. L., & Moles, O. (1982). Student

Wu,

A critical reappraisal. Urban Review,

suspension:

AUTHORS
ANNE

School of
is an assistant professor in the Graduate
at
152
Rutgers University,
Psychology

GREGORY
and

Applied

Frelinghuysen
Her
research

Professional

Road,
Piscataway, NJ 08854;
annegreg@rci.rutgers.edu.
in school discipline
interests include disproportionality

sanctions and the role of teacher-student

eration

relationships

in fostering coop

in the
high school classroom.

RUSSELL J.SKIBA isa professorin theDepartmentofCounseling and

and director of the Equity Project at Indiana

Psychology
IN 47406; skiba@indiana
University, 1900 East 10th Street, Bloomington,
.edu. His research interests include school discipline and school violence,
Educational

and equity
PEDRO

in school discipline

A. NOGUERA

the Steinhardt

School

is the Peter L. Agnew

Education

Education,
His

Professor of Education

at

New
and Development,
Center for
director of theMetropolitan

of Culture,

and executive

York University,
Urban

and special education.

5th Floor, New York, NY

10003;
Broadway,
research focuses on the ways schools are influenced

726

pan6@nyu.edu.
by social and economic

conditions.

Manuscript received June24,2009

Revision

14, 245-272.

RESEARCHER
~6?] IEDUCATIONAL

This content downloaded from 152.15.236.17 on Wed, 11 Mar 2015 23:04:00 UTC
All use subject to JSTOR Terms and Conditions

received October

20,2009

Accepted November 2,2009