ArticlePDF Available

Prekindergarteners Left Behind: Expulsion Rates in State Prekindergarten Systems

Authors:

Abstract

Expulsion is the most severe disciplinary sanction that an educational program can impose. Results are reported from a national study of 3,898 prekindergarten classrooms (81.0% response rate), representing all of the nation's 52 state-funded prekindergarten systems currently operating across 40 states. Weighted results indicated that 10.4% of prekindergarten teachers reported expelling at least one preschooler in the past 12 months, of which 19.9% expelled more than one. Nationally, 6.67 preschoolers were expelled per 1,000 enrolled. Although this rate for state-subsidized prekindergarten is lower than what has been previously reported for child care programs, the prekindergarten expulsion rate is 3.2 times the rate for K-12 students. Rates are reported for each of the states and state prekindergarten systems represented. Significant cross state variability in expulsion rates was found, possibly due in part to differences in how state prekindergarten systems are structured. Rates were highest for older preschoolers and African-Americans, and boys were over 4½ times more likely to be expelled than were girls. Expulsion rates were lowest in classrooms in public schools and Head Start and highest in faith-affiliated centers and for-profit child care. The likelihood of expulsion decreases significantly with access to classroom-based mental health consultation. Although there has been considerable media attention to the issue of young students – kindergarteners and preschoolers – being suspended or expelled from their educational programs, almost no research exists on the topic. As a result, it has been impossible to estimate the number of preschoolers (children ages 3-years to 4-years old) expelled from school or to determine which preschoolers are most at risk for this disciplinary action. As the complete and permanent removal of a student from an entire educational system, expulsion is the most severe disciplinary response that any educational system can impose on a student. Transferring students with behavior problems to other educational settings (e.g., self-contained special education programs, alternative schools, etc.) is not considered expulsion. Rather, expulsion represents a complete cessation of educational services without the benefit of alternative services provided by or through the educational program that has expelled the child. In the case of kindergarten through 12 th grade students in public schools, expulsion is typically the last of a series of disciplinary actions that ultimately culminates in the student being barred from attending any educational programming in that school system. Very little research exis ts on expulsion at any grade level, but the dearth is even more pronounced for children younger than kindergarten. One of the first efforts to collect data on expulsion prior to kindergarten (Grannan, Carlier, & Cole, 1999) was conducted as part of Michigan's Child Care Expulsion Prevention Program, which dispatches mental health consultants to classrooms where children are at immediate risk of expulsion. Surveys were mailed to 127 directors of "preschools and daycare facilities" in the Detroit area. A rate of 27.5 expulsions per 1,000 children enrolled in child care centers was obtained, but the survey suffered from a low response rate of 28%. The first statewide study of expulsion in child care and early education sites was conducted in Massachusetts during the 2001 school year (Gilliam & Shahar, in press). Paper surveys were mailed to a geographically stratified random sample of 185 preschool teachers, of which 64% responded (n = 119). Results indicated that 39% of teachers reported expelling at least one child from her or his class over the previous 12 months. Most of the teachers who reported expelling a child expelled only one child during that time, whereas 25% expelled two or more. When the number of preschoolers expelled was compared to the number of preschoolers enrolled in those classes, an expulsion rate of 27.4 per 1,000 children enrolled was found – a rate nearly identical to that which was previously reported for child care programs in the Detroit, Michigan area. In order to provide a context for this rate, the rate of preschool expulsion was compared to the rate of expulsion in public schools in Massachusetts and across the nation. The rate of expulsion for Massachusetts preschoolers was more than 34 times the rate of expulsions for children K-12 in Massachusetts (0.80 expulsions per 1,000 students) and more than 13 times the national K-12 rate (2.09 expulsions per 1,000 students).
Acknowledgements. The writing of this paper was supported by a grant from the A. L. Mailman Family Foundation. The National
Prekindergarten Study was supported by the Foundation for Child Development and by the National Institute for Early Education Research,
which is funded by the Pew Charitable Trusts. The author wishes to acknowledge the generous support of these foundations and the
thoughtful suggestions and guidance of their staff. The accuracy of the data and conclusions presented in this paper, however, are the sole
responsibility of the author.
Correspondence. Walter S. Gilliam, PhD, Assistant Professor of Child Psychiatry and Psychology, Yale University Child Study Center, 230
South Frontage Road, PO Box 207900, New Haven, CT 06520-7900; Phone: 203-785-3384; E-mail: walter.gilliam@yale.edu.
May 04, 2005
Prekindergarteners Left Behind:
Expulsion Rates in State Prekindergarten Systems
Walter S. Gilliam, PhD
Yale University Child Study Center
Expulsion is the most severe disciplinary sanction that an educational program can impose. Results are
reported from a national study of 3,898 prekindergarten classrooms (81.0% response rate), representing all of
the nation’s 52 state-funded prekindergarten systems currently operating across 40 states. Weighted results
indicated that 10.4% of prekindergarten teachers reported expelling at least one preschooler in the past 12
months, of which 19.9% expelled more than one. Nationally, 6.67 preschoolers were expelled per 1,000
enrolled. Although this rate for state-subsidized prekindergarten is lower than what has been previously
reported for child care programs, the prekindergarten expulsion rate is 3.2 times the rate for K-12 students.
Rates are reported for each of the states and state prekindergarten systems represented. Significant cross state
variability in expulsion rates was found, possibly due in part to differences in how state prekindergarten
systems are structured. Rates were highest for older preschoolers and African-Americans, and boys were over
4½ times more likely to be expelled than were girls. Expulsion rates were lowest in classrooms in public
schools and Head Start and highest in faith-affiliated centers and for-profit child care. The likelihood of
expulsion decreases significantly with access to classroom-based mental health consultation.
Although there has been considerable media attention to the issue of young students kindergarteners and
preschoolers being suspended or expelled from their educational programs, almost no research exists on the topic.
As a result, it has been impossible to estimate the number of preschoolers (children ages 3-years to 4-years old)
expelled from school or to determine which preschoolers are most at risk for this disciplinary action. As the
complete and permanent removal of a student from an entire educational system, expulsion is the most severe
disciplinary response that any educational system can impose on a student. Transferring students with behavior
problems to other educational settings (e.g., self-contained special education programs, alternative schools, etc.) is
not considered expulsion. Rather, expulsion represents a complete cessation of educational services without the
benefit of alternative services provided by or through the educational program that has expelled the child. In the case
of kindergarten through 12th grade students in public schools, expulsion is typically the last of a series of
disciplinary actions that ultimately culminates in the student being barred from attending any educational
programming in that school system.
Very little research exists on expulsion at any grade level, but the dearth is even more pronounced for
children younger than kindergarten. One of the first efforts to collect data on expulsion prior to kindergarten
(Grannan, Carlier, & Cole, 1999) was conducted as part of Michigan’s Child Care Expulsion Prevention Program,
which dispatches mental health consultants to classrooms where children are at immediate risk of expulsion.
Surveys were mailed to 127 directors of “preschools and daycare facilities” in the Detroit area. A rate of 27.5
expulsions per 1,000 children enrolled in child care centers was obtained, but the survey suffered from a low
response rate of 28%.
The first statewide study of expulsion in child care and early education sites was conducted in
Massachusetts during the 2001 school year (Gilliam & Shahar, in press). Paper surveys were mailed to a
geographically stratified random sample of 185 preschool teachers, of which 64% responded (n = 119). Results
indicated that 39% of teachers reported expelling at least one child from her or his class over the previous 12
months. Most of the teachers who reported expelling a child expelled only one child during that time, whereas 25%
expelled two or more. When the number of preschoolers expelled was compared to the number of preschoolers
enrolled in those classes, an expulsion rate of 27.4 per 1,000 children enrolled was found a rate nearly identical to
that which was previously reported for child care programs in the Detroit, Michigan area. In order to provide a
context for this rate, the rate of preschool expulsion was compared to the rate of expulsion in public schools in
Massachusetts and across the nation. The rate of expulsion for Massachusetts preschoolers was more than 34 times
the rate of expulsions for children K-12 in Massachusetts (0.80 expulsions per 1,000 students) and more than 13
times the national K-12 rate (2.09 expulsions per 1,000 students).
Embargoed Until May 17
Prekindergarten Expulsion 2
Further analyses of the Massachusetts data indicated that the likelihood of expelling a preschooler was
significantly associated with several characteristics of the program, classroom, and teacher. Preschool teachers in
either a public school or Head Start were less likely to report expelling a child, relative to teachers in for-profit child
care programs or other community-based non-profit agencies (11% versus 50% and 40%, respectively). Also, the
likelihood of a teacher expelling at least one preschooler was significantly higher when the class size or the
proportion of three-year olds mixed with four-year olds was higher. The teacher’s level of self-reported job stress
also was related significantly to the likelihood of expelling, and contributed to the prediction of expulsion even
when class setting, size, and student age were controlled. When both class size and teacher job stress were low,
however, the likelihood of the teacher expelling a child was also low (12% versus 46% to 50% when either class
size, job stress, or both were higher than the sample median).
The purpose of this paper is to report the rate at which preschoolers are expelled from state-funded
prekindergarten systems across the nation. State-funded prekindergarten systems, operating in 40 states across the
nation, serve nearly one million preschoolers each year. These systems have evidenced dramatic growth since 1980,
when only 10 states funded such programs. Currently, two states Oklahoma and Georgia operate programs that
are universally accessible and serve over 50% of their state’s four-year olds (Barnett, Hustedt, Robin, & Schulman,
2004), and several other states are developing plans for universal expansion. Given the large number of preschoolers
served in these state-funded prekindergarten systems and their mission of enhancing school readiness and
facilitating a smooth transition to kindergarten, these systems play an important role in America’s patchwork of
early education and child care programs.
In order to better understand expulsion rates in state-funded prekindergarten systems, they are compared to
expulsion rates for public school students in grades kindergarten through 12. Also, national prekindergarten rates are
reported on the basis of setting type (public school, Head Start, for-profit child care, etc.); child age, gender, and
ethnicity; and access to classroom-based mental health or behavioral consultation.
Methods
Prekindergarten Data
The data used for these analyses were collected as part of the National Prekindergarten Study (NPS),
consisting of classroom-level data from all 52 state-funded prekindergarten systems operating in the United States
during the 2003 and 2004 academic years.1 These 52 state-funded systems were administered by 40 different states.
The basic methods of the NPS as they pertain to this paper are described below. For a more detailed description of
the sample selection process, measures, and procedures, please see Gilliam and Marchesseault (2004).
Sample. A total of 40,211 state-funded prekindergarten classrooms operating during the study period were
identified and located through the NPS. Target sample sizes within each of the 52 state systems were determined
using a formula developed for the NPS that calculates the sample size needed across systems of varying numbers of
classrooms in order to constrain sampling error to ±5% or less in each state system. Classrooms were randomly
selected at the state level, totaling 4,812 classrooms across the nation. The overall response rate was 81.0%, ranging
from 73.0% in one state system to 100% in four, yielding a final sample size of 3,898 respondents. (See Table 1.)
Response rates did not differ significantly by region of the nation (χ2(3) = 2.89, ns). A statistically
significant difference in response rates was found, however, on the basis of program setting (χ2(2) = 6.19, p < .05),
with teachers in Head Start or public schools being slightly more likely to respond, relative to teachers in other
settings. The effect size of the lower response rate, however, was very small (asymmetric Sommer’s d compared to
Head Start = -.05, compared to public schools = -.03). See Table 2 for a description of the programs, classes,
teachers, and children included in this sample. These descriptions are also presented in the table weighted by the size
of the program to obtain national estimates of the population of classes that were sampled.
Measures. Data were collected over the telephone as part of a comprehensive survey, administered by trained
interviewers using a fully scripted computer-assisted telephone interview (CATI) protocol. The classroom informant
was the lead teacher most responsible for the day-to-day operation of the sampled classroom. In classrooms where
1 For the NPS, state-funded prekindergarten systems were defined as those that (a) are administered and funded (at least in part) by a state agency
or department, (b) serve children in the three- to four-year old age range, but not necessarily this entire age range or exclusively this age range, (c)
have a classroom-based component that meets on a regular basis, (d) have a programmatic goal pertaining to facilitating children’s de velopment,
providing early education, or promoting school readiness, and (e) serve either targeted or non-targeted populations, but do not consist solely of
children with special education disabilities. These selection criteria are similar to those used to identify state-funded prekindergarten systems in
other national studies (Barnett et al., 2004; Cauthen et al., 2000; Gilliam & Ripple, 2004; Gilliam & Zigler, 2000, 2004; Mitchell et al., 1998;
Ripple et al., 1999; Schulman et al., 1999).
Embargoed Until May 17
Prekindergarten Expulsion 3
Table 1. Total Number of Classrooms, Sample Size, and Response Rates by State and System
State N n
Response State Prekindergarten System Name N n
Response
Alabama
69
45
77.6%
Office of School Readiness Prekindergarten
Alaska
103
57
85.1%
Alaska Head Start Program
Arizona
243
84
93.3%
Early Childhood State Block Grant (PreK Component)
Arkansas
177
71
81.6%
Arkansas Better Chance (ABC)
California
5,831
201
77.3%
State Preschool Program
3,128
101
78.3%
California Full Day Preschool Program
2,703
100
76.3%
Colorado
758
91
76.5%
Colorado Preschool Program (CPP)
Connecticut
676
159
85.5%
Connecticut School Readiness and Ch
ild Care Initiative
575
98
84.5%
State Funded Head Start
101
61
87.1%
Delaware
68
40
78.4%
Early Childhood Assistance Program (ECAP)
Florida
1,602
101
86.3%
Florida Partnership for School Readiness
Georgia
3,112
100
73.0%
Georgia Prekinderg
arten Program for Four
-
Year
-
Olds
Hawaii
518
98
82.4%
Preschool Open Doors
506
86
80.4%
State Funded Head Start
12
12
100.0%
Illinois
1,935
99
79.8%
Early Childhood Block Grant
Iowa
128
59
90.8%
Comprehensive Child Development Program
Kansas
211
69
77.5%
At
-
Risk Four
-
Year
-
Old Children Preschool Program
Kentucky
1,024
104
88.1%
Kentucky Preschool Program
Louisiana
268
72
75.0%
Preschool Block Grant
Maine
237
104
81.9%
Two
-
Year Kindergarten
53
36
83.7%
State Funded
Head Start
184
68
81.0%
Maryland
329
78
75.7%
Extended Elementary Education Program (EEEP)
Massachusetts
2,420
153
78.5%
Community Partnerships for Children
2,333
103
77.4%
State Funded Head Start
87
50
80.6%
Michigan
1,110
93
74.4%
Michigan
School Readiness Program
Minnesota
1,157
185
80.8%
School Readiness
707
100
85.5%
State Funded Head Start
450
84
75.7%
Missouri
142
61
83.6%
Missouri Preschool Project
Nebraska
16
16
100.0%
Early Childhood Projects
Nevada
30
29
100.0%
Early Childhood Education Comprehensive Plan
New Jersey
2,787
183
82.1%
Early Childhood Program Aid (ECPA) Abbott
2,471
104
83.2%
Early Childhood Program Aid (ECPA) Non
-
Abbott
316
79
80.6%
New Mexico
40
39
97.5%
Child Development Pr
ogram
25
24
96.0%
State Funded Head Start
15
15
100.0%
New York
4,066
192
78.0%
New York State Universal Prekindergarten Program
3,494
102
76.7%
New York State Experimental Prekindergarten Program
572
90
79.6%
North Carolina
137
66
88.0
%
More at Four
Ohio
1,271
188
82.1%
Public School Preschool
373
92
82.1%
State Funded Head Start
898
96
82.1%
Oklahoma
1,343
188
81.0%
Early Childhood Four
-
Year
-
Old Program
919
94
74.6%
State Funding for Head Start
424
94
88.7%
Oreg
on
460
93
79.5%
Oregon Head Start Prekindergarten
Pennsylvania
88
51
73.9%
Education Aid for Kindergarten for Four
-
Year
-
Olds
South Carolina
608
91
79.8%
Early Childhood Program (Half
-
Day Programs)
Tennessee
177
67
84.8%
Early Childhood Educatio
n Pilot Program
Texas
5,665
101
73.7%
Public School Prekindergarten
Vermont
82
54
84.4%
Early Education Initiative
Virginia
419
90
80.4%
Virginia Preschool Initiative
Washington
304
87
87.9%
Early Childhood Education and Assistance Program
West Virginia
228
90
85.7%
Public School Early Childhood Education
Wisconsin
597
150
77.3%
Four
-
Year
-
Old Kindergarten
463
91
76.5%
State Funded Head Start
134
59
78.7%
NATION 40,211 3,898 81.0%
Note. For each state or state-funded prekindergarten system, N = the total estimated number of classrooms; n
= the number of survey respondents;
Response = the number of survey respondents (n) divided by the number of classrooms sampled.
Embargoed Until May 17
Prekindergarten Expulsion 4
Table 2. Characteristics of Programs, Teachers, and Children
Percentage
Of Sample Weighted for Nation
Program Setting
Head Start 27.48 17.05
School-Based 56.50 63.67
For-Profit Child Care Center 3.24 4.11
Faith-Affiliated Program 1.49 1.70
Other Community-Based 11.30 13.47
Program Length of Year
Partial-Year (1-7 Months) 5.48 3.43
School-Year (8-10 Months) 71.65 72.25
Extended-Year (11-12 Months) 22.87 24.32
Program Length of Day
Part-Day (1.50-4.99 Hours) 35.97 32.68
School-Day (5.00-7.99 Hours) 36.39 38.09
Extended-Day (8.00+ Hours) 27.63 29.23
Teacher Gender
Female 97.83 97.64
Teacher Race/Ethnicity
Asian 2.05 2.69
Latino 8.14 15.21
Native American 2.96 1.66
Black (Non-Latino) 12.02 14.14
White (Non-Latino) 72.87 63.79
Other/Multi-racial 1.95 2.52
Teacher Highest Degree/Credential
High School/GED 3.69 3.75
CDA 11.39 9.08
AA 13.84 14.11
BA 46.97 49.46
MA 24.10 23.61
Child age (on October 1)
Younger than 3 years 2.46 2.17
3 years old 21.95 19.92
4 years old 66.21 68.64
5 years old and older 9.36 9.25
Child racial/ethnic composition
Asian 3.51 3.81
Latino 19.25 29.65
Native American 4.09 2.08
Black (non-Latino) 21.62 22.01
White (non-Latino) 48.36 39.44
Other/Multi-racial 2.94 2.99
Note. Data are presented based on the sample results, as well as results when the sample is weighted based on the number of classrooms in each
state system.
the responsibility was shared among more than one “lead teacher,” the informant was the teacher who had the
highest educational degree in an early education field, the teacher with the highest overall degree, the teacher with
the most years teaching preschool-age children, or the teacher with the most years teaching in that classroom (in that
order of preference). In cases where the identified teacher was not currently working in the classroom for an
extended or indefinite period of time (e.g., due to turnover, retirement, maternity leave, long-term disability, or
death), the informant became whoever assumed full-time, long-term responsibility for the day-to-day operations of
the classroom.
The complete survey averaged 45 to 55 minutes in length, and was either administered in one session or
broken up over several shorter sessions. Skip patterns, fills, and internal validity checks were added to the CATI
program in order to enhance the performance of the survey protocol, reduce the overall length of the survey, and
provide more reliable data collection by validating the responses against known valid ranges and against responses
Embargoed Until May 17
Prekindergarten Expulsion 5
to previous questions. Prior to data collection, the entire protocol was extensively field tested and revised, on the
basis of data collected from a statewide pilot administration and qualitative feedback from preschool teachers, local
and state administrators, and national experts in early childhood education (Gilliam & Marchesseault, 2004). The
survey was translated into Spanish using blind back-translation techniques (Massoubre, Lang, Jaeger, Jullien, &
Pellet, 2002). Trained translators were used for other language preferences.
As part of the NPS survey, teachers were asked to report the number of children in their classroom that
were expelled from attending their prekindergarten setting due to behavioral concerns during the past 12 months.
Expulsion was defined as permanent termination of the child’s participation in the setting. Children who were
transitioned directly from the classroom to a different setting deemed to be more appropriate for the child (e.g.,
special education, transitional classroom, or therapeutic preschool program) were not included. Data from teachers
who responded that they had not taught for at least 12 months were excluded from analyses. For every child who
was expelled, teachers were asked to report the child’s age in years, gender, and race or ethnicity (African-
American, Asian, Latino, White (non-Latino), or Other).
Procedures. Respondents were contacted to schedule the interview at a time that was convenient for the
teacher and did not interfere with her or his classroom duties. When requested, a letter and supporting materials
explaining the project were faxed to the administrator and/or teacher. Teachers were given $10 and a certificate of
participation for completing the interview. All CATI interviewers completed a half-day training on the measures. A
random set of phone interviews were monitored live and rated for quality on a daily basis at the beginning of the
study, and later on a weekly basis. A random subset of study participants also completed a 10-item stamped postcard
response survey at the end of the study to rate the overall experience. Results of live monitoring and postcard
feedback were used to improve ongoing data collection efforts.
K-12 Data
In this paper, prekindergarten expulsion rates are compared to K-12 expulsion rates in the same states. In
order to calculate the rate of K-12 students expelled in each state and for the nation, data were obtained from the
Elementary and Secondary School Survey: 2000 (ESSS; National Center for Educational Statistics, 2001). The
ESSS is a survey of district-level administrators for all of the nation’s over 92,000 public K-12 schools. Data were
obtained for more than 97% of the nation’s schools. Data indicating the total enrollment numbers and numbers of
expulsions for the 1999-2000 school year for all of the nearly 15,000 school districts responding to the ESSS were
downloaded from the U.S. Department of Education. The ESSS defined expulsion as: “The exclusion of a student
from school for disciplinary reasons that result in the student’s removal from school attendance rolls or that meets
the criteria for expulsion as defined by the appropriate state or local school authority. Not suspension.” ESSS data
are expressed for the total number of children grades kindergarten through 12, but are not disaggregated by grade
level.
Data Analysis: Prekindergarten and K-12
Expulsion rates for both prekindergarten students and K-12 students were computed. Prekindergarten
expulsion rates were computed for each state-funded prekindergarten system by dividing the total number of
expulsions in sampled classrooms within that system by the total number of preschoolers enrolled in those sampled
classrooms. Prekindergarten expulsion rates were also computed for each state, by calculating the rate across all
classrooms within each state, weighted by a factor equal to N/n, where N = the estimated number of classrooms in
the state-funded prekindergarten system and n = the number of classrooms sampled from that system. This
weighting yielded results that reflect the relative size differences between state-funded prekindergarten systems.
Expulsion rates for K-12 students were calculated nationally and in individual states by dividing the sum of the
number of children expelled in each school district in the nation or the respective state by the sum of the total
applicable student enrollment.
For ease of reporting, all expulsion rates were multiplied by 1,000 and indicate the number of expulsions
per 1,000 students enrolled. National data were weighted as described above to yield national results that reflect the
relative size differences between the 52 state-funded prekindergarten systems. (For more information regarding the
weighting procedures and computation of weighted sampling errors employed in the National Prekindergarten
Study, see Gilliam & Marchesseault, 2004.)
Embargoed Until May 17
Prekindergarten Expulsion 6
Results
National and State Expulsion Rates
In this sample of state-funded prekindergarten teachers, 9.50% reported having expelled at least one child
in the past 12 months, while 90.50% reported not expelling. When data were weighted to reflect differences in the
relative sizes of the prekindergarten systems each class represented, the percentage of teachers reporting to have
expelled at least one preschooler in the last 12 months was 10.39%. Teachers who responded that they had not
taught yet in their classroom for at least 12 months (19.68%), indicated that they did not know the answer (0.08%),
or skipped this item (1.95%) were excluded from the analyses. Of those teachers reporting to have expelled at least
one preschooler in the past 12 months, 78.28% expelled one, 15.17% expelled two, 5.52% expelled three, and 1.03%
expelled four. When weighted, the proportions were 80.07%, 16.05%, 3.49%, and 0.39%, respectively. No
prekindergarten teacher reported expelling more than four students during the 12-month period.
The weighted national rate of prekindergarten expulsion was 6.67 (± 0.53) per 1,000 preschoolers enrolled,
3.20 times higher than the national rate of expulsion for K-12 students (2.09 per 1,000 enrolled).2 Prekindergarten
expulsion rates and the proportion of prekindergarten teachers expelling at least one preschooler in the past 12
months are reported for both state-funded prekindergarten systems and states. (See Tables 3 and 4, respectively.) In
states with more than one state-funded prekindergarten system, the data are weighted aggregates of all systems in
those states. Across states, prekindergarten expulsion rates varied significantly. The ten states with the highest rates
all exceeded 10 expulsions per 1,000 students, while the ten lowest states were all below 4 per 1,000.
Prekindergarten to K-12 expulsion ratios were computed for all 40 states. Of the 40 states funding prekindergarten,
prekindergarten expulsion rates exceeded K-12 expulsion rates in all but three (Kentucky, South Carolina, and
Louisiana).
As presented in Table 5, an estimated 5,117 prekindergarten students across the nation were expelled from
a total estimated enrollment of 766,907. The number of prekindergarten students that were expelled during the 12-
month period was estimated based on the prekindergarten expulsion rates presented in Table 4. Total state and
national enrollments were estimated based on the average October 1 class enrollment in sampled prekindergarten
classrooms multiplied by the estimated number of classrooms in each state and the nation obtained during the
sampling process described earlier. Student estimates are rounded to the nearest whole number. It is important to
note that the enrollment estimates reflect the estimated number of preschoolers in these state-funded prekindergarten
classes and not the estimated number of preschoolers receiving state subsidies for participation.
Prekindergarten Expulsion Rates By Child Demographic Data
Table 6 presents the weighted national prekindergarten expulsion rates by child age, gender, and
race/ethnicity. Older preschoolers were expelled at a higher rate relative to younger preschoolers. Four-year olds
were expelled at a rate about 50% greater than either 2-year olds or 3-year olds. Children who were either 5 or 6
years old were about twice as likely to be expelled, relative to 4-year olds. These 5 and 6 year old children may have
been either children who were born early in the year or had been retained or held an extra year in prekindergarten
prior to kindergarten entry. African-American preschoolers were about twice as likely to be expelled as European-
American (both Latino and non-Latino) preschoolers and over five times as likely as Asian-American preschoolers.
Boys were expelled at a rate over 4½ times that of girls. The increased likelihood of boys to be expelled over girls
was similar across all ethnicities, except for African-Americans (?2 = 25.93, p < .01), where boys accounted for
91.4% of the expulsions.
Prekindergarten Expulsion Rates by Setting Type and Access to Mental Health Consultation
Program settings were coded as either school-based, Head Start, for-profit child care, faith-affiliated, or
other community-based program. The proportion of classrooms in each of these settings was presented in Table 1.
As shown in Table 7, the percentage of prekindergarten teachers reporting to have expelled at least one student in
the past 12 months varied significantly on the basis of setting (?2(4) = 129.89, p < .001). Teachers in faith-affiliated
(18.06%), for-profit child care (14.55%), and other community-based (13.24%) settings were significantly more
likely to report expelling a preschooler, relative to teachers in either schools (9.19%) or Head Start centers (11.71%).
School-based prekindergarten teachers were less likely than Head Start teachers to expel. Although school-based
prekindergarten teachers reported the lowest rate of expulsion (6.16 per 1,000), the rate of expulsion of preschoolers
in schools is still nearly three times higher than the rate of expulsion of K-12 students in schools (2.09 per 1,000).
2 The 3.20 ratio was obtained with extended precision calculations and is somewhat different than the 3.19 ratio that would be obtained by simply
dividing the two rates with the level of precision presented.
Embargoed Until May 17
Prekindergarten Expulsion 7
Table 3. Prekindergarten Expulsion Rates by State Prekindergarten System
Expulsion Rate PreK Teachers Expelling
State Prekindergarten System Per 1,000
SEM % SEP
1. New Mexico Child Development Program 24.31 2.57 16.67 1.55
2. Maine State Funded Head Start 18.39 3.66 20.69 3.94
3. New Mexico State Funded Head Start 15.75 0.00 28.57 0.00
4. Alabama Office of School Readiness Prekindergarten 14.12 3.45 12.50 2.93
5. Delaware Early Childhood Assistance Program 13.04 3.34 11.54 3.58
6. North Carolina More at Four Pre-kindergarten Program 13.04 3.57 14.71 3.11
7. New York State Experimental Prekindergarten Program 12.67 2.89 16.44 3.59
8. Tennessee Early Childhood Education Pilot Program 12.54 2.76 20.69 3.91
9. Connecticut School Readiness and Child Care Initiative 12.48 2.67 16.25 3.40
10. Missouri Preschool Project 12.02 2.97 20.00 3.85
11. Massachusetts Community Partnerships for Children 11.38 2.86 15.85 3.52
12. Connecticut State Funded Head Start 11.33 2.35 12.00 2.63
13. Virginia Preschool Initiative 10.25 2.74 11.59 3.01
14. Nevada Early Childhood Education Comprehensive Plan 9.48 2.11 9.09 1.03
15. Wisconsin Four-Year-Old Kindergarten 9.10 2.10 14.47 3.31
16. Washington Early Childhood Education and Assistance Program 8.73 2.13 9.72 2.70
17. Georgia Prekindergarten Program 8.58 2.43 13.89 3.40
18. New York State Universal Prekindergarten Program 8.53 2.42 10.98 3.15
19. Vermont Early Education Initiative 8.32 2.01 4.65 1.68
20. Oklahoma Early Childhood Four-Year-Old Program 8.24 2.25 14.67 3.46
21. Hawaii Preschool Open Doors 7.56 2.08 14.86 3.50
22. California State Preschool Half Day Program 7.50 2.12 9.86 2.92
23. California State Preschool Full Day Program 7.48 1.89 14.29 3.43
24. Florida Partnership for School Readiness 6.64 2.03 12.99 3.24
25. New Jersey Early Childhood Program Aid (Abbott Districts) 6.21 2.29 7.79 2.57
26. Oregon Head Start Prekindergarten 6.17 2.08 10.94 2.89
27. Texas Public School Prekindergarten 5.99 1.87 10.71 3.05
28. Maryland Extended Elementary Education Program 5.97 1.97 8.20 2.72
29. Alaska Head Start Program 5.85 1.78 6.38 2.17
30. New Jersey Early Childhood Program Aid (non-Abbott Districts) 5.80 2.05 8.33 2.68
31. Colorado Preschool Program 5.17 1.71 8.54 2.75
32. Massachusetts State Funded Head Start 4.95 1.86 8.11 2.53
33. Ohio State Funded Head Start 4.87 1.88 8.45 2.68
34. Arizona Early Childhood State Block Grant (PreK Component) 4.79 1.46 6.94 2.23
35. Ohio Public Preschool 4.48 1.47 8.22 2.49
36. Arkansas Better Chance 4.40 1.54 7.02 2.35
37. Minnesota State Funded Head Start 4.16 1.67 7.14 2.52
38. Nebraska Early Childhood Block Grant Program 4.15 0.00 7.69 0.00
39. Louisiana 8(g) Preschool Block Grant 3.58 1.53 6.56 2.52
40. West Virginia Public School Early Childhood Education 3.57 1.39 3.17 1.45
41. Pennsylvania Education Aid for Kindergarten for Four-Year-Olds 3.54 1.39 4.88 1.97
42. Minnesota School Readiness 2.92 1.21 3.49 1.70
43. Illinois Early Childhood Block Grant Prekindergarten Program 2.70 1.32 6.58 2.20
44. Michigan School Readiness Program 2.36 1.30 4.11 1.97
45. Iowa Comprehensive Child Development Program (Shared Visions)
2.32 1.20 1.92 1.32
46. Kansas At-Risk Four-Year-Old Children Preschool Program 2.13 1.24 3.51 1.82
47. South Carolina Early Childhood Program 2.04 1.08 2.78 1.60
48. Maine Two-Year Kindergarten 2.00 1.13 2.94 1.61
49. Oklahoma State Funding for Head Start 1.51 0.94 1.41 1.07
50. Wisconsin State Funded Head Start 1.25 0.94 2.17 1.43
52. Hawaii State Funded Head Start 0.00 0.00 0.00 0.00
52. Kentucky Preschool Program 0.00 0.00 0.00 0.00
NATION 6.67 0.53 10.39 0.76
Note. Expulsion rates are tabled in descending order and indicate the number of preschoolers expelled per 1,000 enrolled. National data are
weighted by a factor of N/n, where N = the estimated number of classrooms in the state system and n = the number of classroom teachers
responding.
Embargoed Until May 17
Prekindergarten Expulsion 8
Table 4. Expulsion Rates for Prekindergarten and K-12 by State
PreK
K-12 PreK / K-12 PreK Teachers Expelling
State Expulsions/1,000
SEM
Expulsions/1,000 Ratio % SEP
1. New Mexico 21.10 1.61 1.48 14.22 20.00 1.03
2. Maine 14.73 3.10
0.68 21.50 17.79 2.82
3. Alabama 14.12 3.45
1.03 13.70 12.50 2.93
4. Delaware 13.04 3.34
1.47 8.90 11.54 3.58
5. North Carolina 13.04 3.57
1.90 6.86 14.71 3.11
6. Tennessee 12.54 2.76
3.79 3.30 20.69 3.91
7. Connecticut 12.31 2.62
1.18 10.40 15.52 2.51
8. Missouri 12.02 2.97
0.70 17.22 20.00 3.85
9. Massachusetts 11.15 2.82
0.80 13.93 15.56 2.84
10. Virginia 10.25 2.74
1.49 6.88 11.59 3.01
11. Nevada 9.48 2.11
2.28 4.16 9.09 1.03
12. New York 9.11 2.48
0.47 19.18 11.72 2.27
13. Washington 8.73 2.13
3.71 2.36 9.72 2.70
14. Georgia 8.58 2.43
1.76 4.88 13.89 3.40
15. Vermont 8.32 2.01
1.00 8.31 4.65 1.68
16. California 7.49 2.01
2.52 2.97 12.11 2.26
17. Hawaii 7.39 2.04
0.00 -- 14.63 3.24
18. Wisconsin 7.34 1.84
1.43 5.12 12.08 2.30
19. Florida 6.64 2.03
0.37 18.04 12.99 3.24
20. Oregon 6.17 2.08
3.51 1.76 10.94 2.89
21. New Jersey 6.16 2.26
0.38 16.15 7.85 1.92
22. Oklahoma 6.12 1.84
2.15 2.84 10.41 2.07
23. Texas 5.99 1.87
2.93 2.04 10.71 3.05
24. Maryland 5.97 1.97
0.97 6.18 8.20 2.72
25. Alaska 5.86 1.78
2.12 2.77 6.38 2.17
26. Colorado 5.17 1.71
2.62 1.97 8.54 2.75
27. Arizona 4.79 1.46
1.61 2.98 6.94 2.23
28. Ohio 4.76 1.76
4.17 1.14 8.38 1.87
29. Arkansas 4.40 1.54
1.20 3.66 7.02 2.35
30. Nebraska 4.15 0.00
1.90 2.18 7.69 0.00
31. Louisiana 3.58 1.53
7.78 0.46 6.56 2.52
32. West Virginia 3.58 1.39
1.43 2.51 3.17 1.45
33. Pennsylvania 3.54 1.39
1.06 3.33 4.88 1.97
34. Minnesota 3.41 1.39
0.76 4.51 4.83 1.45
35. Illinois 2.70 1.32
0.96 2.81 6.58 2.20
36. Michigan 2.36 1.30
1.67 1.41 4.11 1.97
37. Iowa 2.32 1.20
0.43 5.45 1.92 1.32
38. Kansas 2.13 1.24
1.80 1.19 3.51 1.82
39. South Carolina 2.04 1.08
7.16 0.28 2.78 1.60
40. Kentucky 0.00 0.00
1.21 0.00 0.00 0.00
Idaho No PreK
1.18 NA No PreK
Indiana No PreK
7.93 NA No PreK
Mississippi No PreK
3.20 NA No PreK
Montana No PreK
1.10 NA No PreK
New Hampshire No PreK
0.72 NA No PreK
North Dakota No PreK
0.56 NA No PreK
Rhode Island No PreK
2.14 NA No PreK
South Dakota No PreK
1.09 NA No PreK
Utah No PreK
3.43 NA No PreK
Wyoming No PreK
1.46 NA No PreK
NATION 6.67 0.53
2.09 3.20 10.39 0.76
Note. Expulsion rates indicate the number of students expelled per 1,000 students enrolled. States are tabled in descending order, based on
prekindergarten expulsion rates. “PreK / K-12 Ratio” indicates the degree to which the PreK expulsion rate surpasses the K-12 expulsion rate
(e.g., in New Mexico, the preK expulsion rate is 14.22 greater than the K-12 expulsion rate). PreK national data are weighted by a factor of N/n.
PreK/K-12 Ratios were determined based on extended precision calculations, and may differ somewhat from division with numbers rounded to
nearest hundredth, as they are presented in this table.
Embargoed Until May 17
Prekindergarten Expulsion 9
Table 5. Estimated Number of Prekindergarten Children Expelled by State
Estimated PreK
Estimated Number Expelled
State Enrollment
Estimate Sampling Error
Alabama 1,217
17 4
Alaska 1,897
11 3
Arizona 4,901
23 7
Arkansas 3,462
15 5
California 134,612
1,008 271
Colorado 14,094
73 24
Connecticut 11,942
147 31
Delaware 1,166
15 4
Florida 30,549
203 62
Georgia 61,234
526 152
Hawaii 10,255
76 21
Illinois 38,181
103 50
Iowa 2,098
5 3
Kansas 3,483
7 4
Kentucky 21,735
0 0
Louisiana 5,014
18 8
Maine 3,579
53 11
Maryland 6,390
38 13
Massachusetts 38,310
427 108
Michigan 19,184
45 25
Minnesota 21,534
73 30
Missouri 2,512
30 7
Nebraska 275
1 0
Nevada 649
6 1
New Jersey 41,409
255 94
New Mexico 668
14 1
New York 69,395
632 172
North Carolina 2,166
28 8
Ohio 23,467
112 41
Oklahoma 25,454
156 47
Oregon 8,220
51 17
Pennsylvania 1,850
7 3
South Carolina 12,561
26 14
Tennessee 3,175
40 9
Texas 110,637
662 207
Vermont 1,337
11 3
Virginia 6,605
68 18
Washington 5,849
51 12
West Virginia 4,312
15 6
Wisconsin 12,393
91 23
NATION 766,907
5,117 406
Note. Estimates are rounded to the nearest whole number. Estimated preK enrollment is the estimated number of classrooms for each state times
that state’s mean October 1 class enrollment. Estimated number expelled is based on the rates and SEM times the estimated preK enrollment.
Embargoed Until May 17
Prekindergarten Expulsion 10
Table 6. Prekindergarten Expulsion by Child Age, Gender, and Race
Expulsions per 1,000
Child Age
2 Years 3.75
3 Years 3.96
4 Years 5.85
5-6 Years 11.57
Child Gender
Female 2.26
Male 10.46
Child Race/Ethnicity
Asian-American 1.82
Latino 4.42
White (non-Latino) 5.77
Other 6.81
African-American 10.04
Note. Rates are weighted based on the relative size of the prekindergarten system each class represents.
Table 7. Prekindergarten Expulsion by Setting and Access to Mental Health Consultation
Proportion of PreK Teachers Expelling Expulsions
% ?2 per 1,000
Classroom Setting 1 129.89***
School (S) 9.19 6.16
Head Start (HS) 11.71 6.59
Other (O) 13.24 7.63
For-Profit Child Care (CC) 14.55 11.93
Faith-Affiliated (FA) 18.06 12.48
Mental Health Consultation Access 2
Psychologist/Psychiatrist 134.83***
On-Site or Regular Visits 7.97 5.68
On-Call 10.29 6.17
No Access 14.34 10.76
Social Worker 106.05***
On-Site or Regular Visits 8.38 6.29
On-Call 11.12 6.55
No Access 13.49 8.56
Note. Percentages are weighted based on the relative size of the prekindergarten system each class represents. (1) FA, CC, & O > HS & S; HS >
S. (2) For both psychologist/psychiatrist and social worker, No Access > On-Call > On-Site or Regular Visits.
*** p < .001.
Prekindergarten expulsion rates were also related significantly to teacher access to classroom-based mental
health consultation. Teachers were asked to rate their access to classroom-based mental health consultation
separately for a psychologist/psychiatrist and for a social worker, as “on-site or regular visits” (22.90% and 38.42%,
respectively), “on-call” (60.33% and 48.13%, respectively), or “no access” (16.77% and 13.46%, respectively).
“On-site or regular visits” was defined as “the professional either has an office in your building or has a predictable
schedule of visits to your site, at least monthly.” “On-call” was defined as “the professional is available only by
request.” “No access” was coded when the teacher indicated that she or he is unable to access the professional or is
not sure whether such a professional is available.
As shown in Table 7, the percentage of prekindergarten teachers reported to have expelled at least one child
in the past 12 months is significantly lower at each level of increased access to classroom-based mental health
consultation provided by either a psychologist/psychiatrist (?2(2) = 134.83, p < .001) or a social worker (?2(2) =
106.05, p < .001). The number of expulsions per 1,000 preschoolers also decreases commensurately with access,
though the greatest decrease in rate appears to be between the access categories of “no access” and “on-call” access.
Discussion
Expulsion is the most severe disciplinary sanction that an educational program can impose on a student.
Prekindergarteners are expelled at a rate that is more than three times that of their older peers in grades kindergarten
Embargoed Until May 17
Prekindergarten Expulsion 11
through 12 (6.67 per 1,000 preschoolers, as compared to 2.09 per 1,000 K-12 students). Prekindergarten expulsion
rates were higher than those found for K-12 students in all but 3 of the 40 states that fund prekindergarten
(Kentucky, South Carolina, and Louisiana). At the classroom level, prekindergarten expulsion is not uncommon,
with 10.4% of prekindergarten teachers expelling at least one preschooler in a given year. The rate of expulsion
found in state-funded prekindergarten classrooms, however, is less than one-fourth the rates that have been
previously reported in samples consisting largely of private child care centers that may not have been participating
in state-funded prekindergarten system (27.4 (Gilliam & Shahar, in press) and 27.5 (Grannan et al., 1999) per
1,000). The difference in expulsion rates between this sample of state-funded prekindergarten classes and previous
studies of child care programs is likely due to two major differences in the samples. First, the current sample
consisted largely of teachers in public school or Head Start centers, whereas the two previous studies consisted
largely of teachers in for-profit child care centers or other centers not in the schools or Head Start. In both the
current study and Gilliam and Shahar (in press), the lowest expulsion rates were found in classrooms located in the
public schools and Head Start, possibly explaining much of the difference in expulsion rates in these samples.
Second, many state-funded prekindergarten systems require private child care providers to meet state
prekindergarten guidelines regarding class size, teacher-child ratios, and teacher credentials in order to accept state
prekindergarten subsidies (Gilliam & Ripple, 2004). These requirements may result in the private child care centers
that participate in state prekindergarten systems being of a higher structural quality than is typically found among
child care programs in general. In the present study, the rate of expulsion in for-profit child care centers (11.9 per
1,000) was less than half the rate reported for child care programs that were not necessarily participating in state
prekindergarten systems, as reported by Gilliam and Shahar (in press) and Grannan et al. (1999).
Overall, the differences in expulsion rates across setting types are rather striking and mirror what was
previously reported for early childhood programs in Massachusetts (Gilliam & Shahar, in press). Whereas teachers
in schools and Head Start centers expelled at the lowest rates, the highest rate of expulsion was reported by teachers
in for-profit child care and faith-affiliated centers, with rates about twice as high.
Variation in Expulsion Rates Between States
Significant variability in expulsion rates was found across states and their prekindergarten systems.
Prekindergarten expulsion rates ranged from a high of 21.1 per 1,000 preschoolers in New Mexico to a low of 0 in
Kentucky. The median expulsion rate among the 10 states with the highest rates was about five times as great as the
median rate among the 10 states with the lowest expulsion rates.
At least part of this variability may be due to differences in the way in which these prekindergarten systems
are structured. Gilliam and Marchesseault (2004) identified seven states as having prekindergarten systems that are
comprised of an exceptionally wide variety of provider types and including a high proportion of classrooms in child
care centers not affiliated with either the public schools or Head Start. All seven of these states have prekindergarten
expulsion rates that exceed the national average. There may be advantages to utilizing a variety of providers when
building state-funded prekindergarten systems (e.g., faster scale-up, less duplication of services and greater
coordination of funding, etc.). The challenge, however, may be in coordinating these various provider types under a
coherent set of policies ranging from structural variables of quality (e.g., teacher credentials, student-teacher ratios,
etc.), to student learning expectations, to responses to severe behavioral challenges. Developing consistent policies
regarding expulsion and other disciplinary actions should be an important step in creating any statewide system of
early education.
Demographic Characteristics of Preschoolers Most At Risk
Clear differences in expulsion rates were found on the basis of child age, gender, and race/ethnicity. The
highest rates of expulsion were reported for preschoolers who were on the older end of the prekindergarten age
spectrum. Prekindergarten programs typically serve children within the 3- to 4-year old age range, though some
systems allow children to enroll just prior to their third birthday and some children may turn 5 or 6 years old while
still enrolled in prekindergarten. Expulsion rates were about 50% greater for children who were 4, as opposed to
those who were 3. The highest rate, however, was found for children who were 5- to 6-years old. These children
were expelled at a rate that is nearly twice as high as 4-year olds and nearly three times that of 3-year olds.
Boys were over four times as likely to be expelled as girls, and African-American preschoolers were about
twice as likely to be expelled as preschoolers of European descent. Although a pattern of particular risk for
expulsion with African-American students has been demonstrated during kindergarten through grade 12 (Holzman,
2004), the pattern of disparity appears to begin much earlier.
Embargoed Until May 17
Prekindergarten Expulsion 12
Access to Mental Health Consultation
When teachers reported having access to a mental health consultant that was able to provide classroom-
based strategies for dealing with challenging student behaviors, the likelihood of expulsion was lower. Having
access to a mental health consultant that was able to come to the classroom in response to a request initiated by the
teacher was better than no access at all, but the lowest rates of expulsion were reported by teachers that had an
ongoing, regular relationship with a mental health consultant either because the teacher and consultant shared a
building or because the consultant paid regular visits to the classroom at least monthly. It is not possible to know
from these data that access to a mental health consultant actually caused the decreased likelihood of expulsion. The
relationship may be due to other factors, such as a greater overall level of resources in programs where consultants
are made available. However, given the rather pronounced differences in expulsion rates when mental health
consultants are available to teachers, the effectiveness of consultancy-based systems of support to preschool teachers
deserves further consideration.
The practice of mental health consultation has been described in detail (Donahue, Falk, & Provet, 2000),
but its effectiveness has not been rigorously studied yet, and there exists little guidance as to how a statewide system
of preschool mental health consultation would be developed (Brennan, Bradley, Allen, Perry, & Tsega, 2005).
Michigan has a relatively long history of statewide response to child care expulsion (Child Care Expulsion
Prevention Program; CCEPP), and a similar statewide initiative recently has been launched in Connecticut (Early
Childhood Consultation Partnership; ECCP).3 These statewide efforts could serve an important role in learning
more about the effectiveness of preschool mental health consultation and the relative merit of consultancy-based
approaches for preventing the expulsion of preschoolers.
Implications for Policy
The implications of this work for developing early education and early intervention policy are myriad. State
prekindergarten systems need to have support services in place that are able to meet the needs of children with
severe behavior problems. This means active collaboration with behavioral consultants that can provide consultation
to teachers regarding individual children’s behaviors, as well as consultation that focuses on more general classroom
behavior management techniques. These support systems should be viewed as an essential component of any
prekindergarten system and funding should be designed so that the service can expand to meet changing needs as
greater numbers of children are provided prekindergarten access.
There may be instances, however, where behavior problems are so sever that they cannot be managed
safely in a typical prekindergarten classroom. For these cases, prekindergarten systems should explore the
effectiveness of alternative settings where children’s behavioral and academic needs can be addressed effectively.
Strong collaboration with preschool special education providers would be a key component. Alternative settings
might include transitional classrooms with highly trained teaching staff, ample support services through school
psychologists and other professionals, and low child-teacher ratios. With adequate access to appropriate alternative
placements, children may be provided early education through alternative settings rather than expelling children
where they may receive no assistance at all only to arrive at kindergarten even farther behind. Also, state
prekindergarten systems should be integrally coordinated with state early intervention services that serve the most
vulnerable infants and toddlers. Early intervention, before children arrive at prekindergarten, may be an effective
way of helping to ensure that all children are ready to take full advantage of prekindergarten.
The data presented in this paper were obtained from classroom teachers participating in state-funded
prekindergarten programs that are considered by their respective states as being important parts of their overall
system of public education. The reality of state-sponsored prekindergarten, however, is that prekindergarten is
offered across a wide variety of setting types public schools, Head Start, for-profit child care, and other
community-based providers. These data show that expulsion rates are much higher in settings outside of the public
schools and Head Start. Within a coherent statewide system of early education, such disparities across setting types
should not exist. States should develop clear procedural guidelines regarding the discipline of prekindergarteners,
and the level of support services that are provided should be similar regardless of where the classroom happens to be
located. An important way to support teachers would be through enhanced preservice and in-service training in
empirically proven methods of classroom behavior management. More needs to be known, however, about the
degree to which teachers are provided training in this area through teacher preparation programs and employer-
sponsored in-service training, as well as the effectiveness of these teacher education programs.
3 Information about Michigan’s CCEPP is available at http://www.earlyonmichigan.org/articles/10-03/CCEP10-03.htm, and information about
Connecticut’s ECCP is available at http://www.abhct.com/casestudies_earlyint.htm.
Embargoed Until May 17
Prekindergarten Expulsion 13
Although the benefits of high-quality early childhood education have been well documented, the effects of
preschool are likely quite small for those children who are unable to participate because of their own challenging
behaviors. The goal of early education is to promote school readiness. Many children may be “unready” for
kindergarten because of difficulties regulating their emotions and behavior, forming friendships, and following adult
directives. For these children, a high-quality school readiness experience is essential to their starting kindergarten
with the skills they need to succeed in school. More work needs to be done to better understand the causes of
preschool expulsion, the impact of preschool expulsion, and how these children who are left behind so early in their
educational experience can be provided a more productive start to school.
References
Barnett, W. S., Hustedt, J. T., Robin, K. B., & Schulman, K. L. (2004). The state of preschool: 2004 state preschool yearbook. New Brunswick,
NJ: National Inst itute for Early Education Research.
Brennan, E. M., Bradley, J. R., Allen, M. D., Perry, D. F., & Tsega, A. (2005, March). The evidence base for mental health consultation in early
childhood settings: Research synthesis and review. Paper presented at the Conference on Establishing the Evidence Base for Early
Childhood Mental Health Consultation, Tampa, FL.
Cauthen, N. K., Knitzer, J., & Ripple, C. H. (2000). Map and track: State initiatives for young children and families 2000 Edition. New York:
National Center for Children in Poverty, Columbia University School of Public Health.
Donahue, P. J., Falk, B., & Provet, A. G. (2000). Mental health consultation in early childhood. Baltimore, MD: Paul H. Brookes Publishing.
Gilliam, W. S., & Marchesseault, C. M. (2004). From capitols to classrooms, policy to practice: State-funded prekindergarten at the classroom
level National Prekindergarten Study technical report #1. New Haven, CT: Yale University.
Gilliam, W. S., & Ripple, C. H. (2004). What can be learned from state-funded prekindergarten initiatives? A data-based approach to the Head
Start devolution debate. In E. Zigler & S. Styfco (Eds.), The Head Start debates (pp. 477 -497). Baltimore, MD: Paul H. Brookes.
Gilliam, W. S., & Shahar, G. (in press). Prekindergarten expulsion and suspension: Rates and predictors in one state. Infants and Young Children.
Gilliam, W. S., & Zigler, E. F. (2000). A critical meta-analysis of all evaluations of state-funded preschool from 1977 to 1998: Implications for
policy, service delivery and program evaluation. Early Childhood Research Quarterly, 15, 441-473.
Gilliam, W. S., & Zigler, E. F. (2004). State efforts to evaluate the effects of prekindergarten: 1977-2003. Available:
nieer.org/resources/research/StateEfforts.pdf.
Grannan, M., Carlier, C., & Cole, C. E. (1999, March). Early childhood care and education expulsion prevention project. Southgate, MI:
Downriver Guidance Clinic, Department of Early Childhood Programs.
Holzman, M. (2004). Public education and black male students: A state report card. Cambridge, MA: The Schott Foundation for Public
Education. Available: www.schottfoundation.org.
Massoubre, C., Lang, F., Jaeger, B., Jullien, M., & Pellet, J. (2002). Translating questionnaires and tests: Techniques and problems. Canadian
Journal of Psychiatry, 47, 63-69.
Mitchell, A., Ripple, C., & Chanana, N. (1998). Prekindergarten programs funded by the states. New York: Families and Work Institute.
National Center for Educational Statistics. (2001). Overview of public elementary and secondary schools and districts: School year 1999-2000
(NCES 2001-339R). Washington, DC: U. S. Department of Education, Office of Educational Research and Improvement. Available:
nces.ed.gov/pubs2001/2001339r.pdf. Accessed June 2002, www.ed.gov/office/OCR/data.html.
Ripple, C. H., Gilliam, W. S., Chanana, N., & Zigler, E. (1999). Will fifty cooks spoil the broth? The debate over entrusting Head Start to the
states. American Psychologist, 54, 327 -343.
Schulman, K., Blank, H., & Ewen, D. (1999). Seeds of success: State prekindergarten initiatives 1998-1999. Washington, DC: Children’s
Defense Fund.
... In exploring correlates of exclusionary discipline in early education settings, prior research has predominantly explored center-level, classroom-level, and teacher-level factors associated with use of exclusionary discipline strategies (Gilliam, 2005(Gilliam, , 2016Gilliam & Shahar, 2006). Existing studies that have examined child-level demographic correlates find Black children (Gilliam, 2005(Gilliam, , 2016, Latino children (Gilliam, 2005(Gilliam, , 2016Graves & Howes, 2011), males (Gilliam, 2005(Gilliam, , 2016, and older children (Gilliam, 2005;Perry et al., 2008) experience disproportionately high rates of exclusionary discipline in early education settings. ...
... In exploring correlates of exclusionary discipline in early education settings, prior research has predominantly explored center-level, classroom-level, and teacher-level factors associated with use of exclusionary discipline strategies (Gilliam, 2005(Gilliam, , 2016Gilliam & Shahar, 2006). Existing studies that have examined child-level demographic correlates find Black children (Gilliam, 2005(Gilliam, , 2016, Latino children (Gilliam, 2005(Gilliam, , 2016Graves & Howes, 2011), males (Gilliam, 2005(Gilliam, , 2016, and older children (Gilliam, 2005;Perry et al., 2008) experience disproportionately high rates of exclusionary discipline in early education settings. As compared to White children, Black and Latino children are more likely to be suspended or expelled from early learning settings (Gilliam, 2005), and male children are more likely to be removed than female children (Gilliam, 2016). ...
... In exploring correlates of exclusionary discipline in early education settings, prior research has predominantly explored center-level, classroom-level, and teacher-level factors associated with use of exclusionary discipline strategies (Gilliam, 2005(Gilliam, , 2016Gilliam & Shahar, 2006). Existing studies that have examined child-level demographic correlates find Black children (Gilliam, 2005(Gilliam, , 2016, Latino children (Gilliam, 2005(Gilliam, , 2016Graves & Howes, 2011), males (Gilliam, 2005(Gilliam, , 2016, and older children (Gilliam, 2005;Perry et al., 2008) experience disproportionately high rates of exclusionary discipline in early education settings. As compared to White children, Black and Latino children are more likely to be suspended or expelled from early learning settings (Gilliam, 2005), and male children are more likely to be removed than female children (Gilliam, 2016). ...
Article
Full-text available
Recent policy briefs have drawn attention to the use of exclusionary discipline in early learning settings; however, little is known about child-level correlates associated with risk of exclusion. This omission is important, as early childhood education may reduce the likelihood of later delinquent and criminal behavior. Additionally, exclusion from early learning may label a child as deviant, contributing to an accumulation of disadvantage that may place the child at greater risk for delinquency and crime over the life-course. The current study applies Moffitt’s (1993) life-course theory to better understand child-level correlates associated with exclusionary discipline in early childhood. Using data from the 2016 National Survey of Children’s Health (N = 5876), results indicate neuropsychological deficits in early childhood are associated with a 1.38 unit increase in odds of removal from early learning environments. Furthermore, in support of Moffitt’s (1993) interactional hypothesis, exposure to adverse experiences (ACEs) was found to moderate the association between neuropsychological deficits and odds of exclusion, such that children with more indicators of neuropsychological deficits and a greater number of ACEs were more likely to experience exclusion than those with fewer ACEs. Implications for policy and directions for future research are discussed.
... Exclusionary discipline from early care and education (ECE) settings throughout the United States has received public interest in recent years. Gilliam first highlighted the issue in 2005 when he reported that early childhood expulsion occurred at a rate three times higher than with school-age children (Gilliam, 2005). Research indicates that exclusionary discipline is a predictor of negative future educational and social-emotional outcomes (Noltemeyer et al., 2015), including decreased math and reading achievement (Lacoe & Steinberg, 2018), disengagement from school, diminished educational opportunity (Skiba et al., 2014), and incarceration (Barnes & Motz, 2018). ...
... Preschool expulsion first received mainstream attention in the United States in 2005 following Walter Gilliam's landmark study (Gilliam, 2005) and remained a critical issue in ECE settings (U.S., 2014). Expulsion is defined as the permanent removal of a child from an educational setting due to a violation of school policies (School Discipline Support Initiative, 2020). ...
... ECE expulsions affect young children disproportionately. For example, the pattern of expulsion among African American students that is evident from kindergarten to 12th grade appears to begin earlier in ECE settings (Gilliam, 2005;Giordano, Interra, et al., 2021). Exclusionary discipline of a preschool child is a strong predictor of future negative educational and social-emotional outcomes. ...
Article
Full-text available
Expulsion has been a well-documented practice in early learning centers throughout the United States. The present study attempted to describe expulsion practices in one state’s community childcare centers during the COVID-19 pandemic. Surveys from 161 childcare program administrators were analyzed and, overall, expulsion rates appeared to be lower than they were pre-pandemic. No association was found between whether a program closed and reopened or remained open; the presence of a waiting list; if a program readmitted all or some children; factors that influenced which teachers were rehired; training provided to teaching staff; perceived frequency and intensity of challenging behavior; and availability of support for children with challenging behaviors and expulsion decisions. Results of the current study are analyzed and discussed in this article along with the results and how they fit into the literature.
... 2). Preschool children are suspended and expelled at a higher rate than school-aged youth (Gilliam, 2005). According to a recent U.S. Department of Education, Office for Civil Rights report (2016), of the 1.4 million children enrolled in public pre-K, 6,743 children received one or more out-of-school suspensions in the 2013-14 school year, and these data do not account for the experiences of children attending private preschools. ...
... Within young children's contexts, available and coordinated support systems (e.g., mental health and behavioral consultation) (Albritton et al., 2019;Gilliam, 2005;Gilliam, 2016;Silver & Zinsser, 2020) and teacher capacity, in terms of teacher levels of stress and well-being (Zinsser et al., 2019;Silver & Zinsser, 2020), are key factors related to exclusionary discipline. Silver and Zinsser (2020) found that when early childhood centers across a racially and ethnically diverse sample had access to and utilized infant/early childhood mental health consultation services, the association between teacher depression and their request for expulsion was attenuated. ...
... Such findings point towards the mediating effect of physical activity-induced alterations in self-regulation on academic outcomes, such as numeracy, in young children. In terms of policy and practice, the work highlights recommendations that efforts targeting the increase of young children's physical activity as one way to reduce inattention and off-task behavior in the classroom, challenging behaviors that have been implicated in minimizing children's access to learning experiences and the expulsion of young children from preschool and early care settings (Gilliam, 2005). ...
... Children with lower self-regulation may miss out on learning opportunities available to them in early educational settings . Even more alarming, low self-regulation, inattention, and lack of inhibitory control contribute to the high levels of expulsion exhibited in preschool programs (Gilliam, 2005;Gilliam & Reyes, 2018). Our finding that children with more days of physical activity during the week have better self-regulation has important implications for understanding this developing skill and for practice in early learning environments. ...
Article
Research Findings: The present study examined the associations among meeting 24-hour movement behavior recommendations set by the World Health Organization (2019) and young children’s self-regulation and quantity estimation skills in a sample of 123 children (n = 65 female; 4.9 ± 0.7 years) in mid-Michigan. Meeting screen time recommendations alone, meeting physical activity recommendations in combination with either sleep or screen time recommendations, meeting more recommendations overall, and being active more days weekly were associated with superior quantity estimation. Meeting more guidelines and accruing more moderate-to-vigorous physical activity daily related to better self-regulation. Improvements in self-regulation partially mediated the relation between physical activity and quantity estimation. Practice or Policy: Our findings identify benefits for adopting specific physical activity guidelines for children (e.g., physical activity, screen time, and sleep duration) and integrating these into early learning standards so both families and schools can support children’s capacity to meet 24-hr movement guidelines and thus support cognitive health. An active lifestyle in early childhood may support young children’s self-regulation and early educational outcomes, with physical activity promotion efforts during early childhood serving as a viable means to address growing expulsion rates in preschool-aged children.
... Despite the seminal work by Gilliam (2005) and Gilliam and Shahar (2006) showing that preschool-aged children (i.e., ages three to five) were expelled at 3 times the rate of K-12 students, preschool expulsion continues to be a national concern. Estimates from the 2016 National Survey of Children's Health indicated that upward of 250 children were suspended or expelled from preschool on a daily basis (Malik, 2017), with Black children being excluded from school at persistently higher rates (Fabes et al., 2021;US. ...
... According to Gilliam (2005), expulsion is defined as "the complete and permanent removal of a student from an entire educational system" (p. 1) and is often the last recourse for sanctioning problem behaviors in children. Although suspension and expulsion decisions are intended to deter future problems, children suspended earlier in school are more likely to be suspended again (Raffaele Mendez, 2003), indicating the ineffectiveness of such disciplinary practices. ...
Article
Full-text available
Despite the known relationship between trauma and academic outcomes, including expulsion risk, for preschoolers, little is known about the role that teachers may play in addressing the effects of childhood trauma within preschool settings. The current study examined the relationship between a teacher’s overall stress, trauma-informed attitudes, and indicators of children’s expulsion decision risk using a sample of preschool lead and assistant teachers (n = 129) recruited from Head Start classrooms in the Mountain West. Multivariate multiple regression was used to determine whether teachers stress and trauma-informed attitudes (trauma-informed knowledge, self-efficacy, and reactions) were related to three indicators of expulsion decision risk using subscales of the Preschool Expulsion Risk Measure (classroom disruption, fear of accountability, and child-related stress) for the most disruptive child in the teacher’s classroom. Higher overall stress significantly predicted higher fear of accountability (β = 0.26, 95% CI = 0.07, .45, p = 0.007). Higher trauma-informed knowledge was significantly related to lower child-related stress (β = −0.40, 95% CI = −0.63, −.17, p = 0.001). Higher trauma-informed self-efficacy was significantly related to lower classroom disruption (β = −0.45, 95% CI = −0.66, −.25, p < 0.001). Multigroup models revealed significantly different pathways for children of color (Black, Latinx, and American Indian children) compared to White children; teacher stress predicted higher expulsion decision risk for children of color and trauma-informed attitudes predicted lower expulsion decision risk for White children. Implications for development and evaluation of trauma-informed approaches for early childhood settings are discussed.
... We do, however, also see some clear differences between different subgroups. If we look at expelled students, for example, we see that Asian Americans are being expelled less often than European Americans, who are being expelled less often than African American preschool students (Gilliam, 2005) . ...
Article
Full-text available
In this paper we analyze societal polarizing trends and link these to mental wellbeing. We assume that the increased (perceived) polarization in American society has an impact on people's mental health, the mental health of more vulnerable communities in particular. We dive deeper in the concepts of polarization of the American society and we explore the concept of marginalized communities. Furthermore, we connect both concepts of polarization with marginalization and address several examples in which these trends are communicating vessels. In the final section of this literature review we discuss several mental health treatment plans that can help both survivors as people who be considered as contributors to these polarizing trends. Hereby, we focus on a model based on occupational therapy, a psychoanalytic model, and a model based on child-centered play therapy. We conclude that polarization is not a new phenomenon, but recent developments seemed to have significantly further increased the gap between different (sub) populations. This development seems to have harsher effects amongst the more vulnerable marginalized individuals. Fortunately, there are approaches that can offer the desired support. Both from a structural societal perspective in training people and by being more inclusive as a society, as via various therapeutic approaches specifically geared towards oppression, polarization, and marginalization.
... Exposure to high levels of adversity, which is often confounded with living in under-resourced and marginalized communities, places children at greater risk for these problems (Blair & Raver, 2015). Children who are perceived to display elevated levels of problem behavior in early care and education (ECE) settings are also at greater risk of being suspended or expelled, and these practices are disproportionately applied to children of color (Gilliam, 2005;Gillaim & Shahar, 2006;US DOE OCR, 2016). Even if ECE programs have policies that prohibit expulsion, social-emotional competence and the capacity for self-regulation help children benefit from their educational experiences and predict future academic and life success (Graziano, et al., 2007;McClelland et al., 2006). ...
Article
Full-text available
Early childhood mental health consultation (ECMHC) is a capacity-building intervention that aims to enhance the quality of young children’s affective environments in order to promote children’s social, emotional, and behavioral health. In this study, the effects of ECMHC on children’s social–emotional and early academic outcomes over the course of one academic year were evaluated in 20 classrooms (15 intervention and 5 comparison) with 38 teachers (29 intervention and 8 comparison), and 390 children (282 intervention and 108 comparison; Mage = 46.71 months old) across 3 schools. Observations, teacher ratings, and direct assessment were used to evaluate children’s social–emotional skills and early academic outcomes. A three-level model that accounted for the nesting of children within classrooms within schools found that children in intervention schools had more positive classroom behavior, fewer observed social–emotional challenges, and higher academic achievement in math, literacy, and writing at the end of the school year. Our findings suggest that this model of ECMHC is an effective way to spread out the expertise of mental health professionals and improve the social, emotional, and academic outcomes for children in the school setting. This is particularly important for marginalized and under-resourced communities who often face higher levels of adversity and mental health needs with fewer available resources as a result of structural factors including racism and underinvestment of public funds.
Article
The persisting issue of racial injustice within disciplinary action referred to as the school-to-prison pipeline has been frequently examined and studied across multiple disciplines spanning education, public policy, criminal justice, and others. The racial school discipline crisis is the disproportionate and differential use of exclusionary action against Black children in school. While disproportionate exclusion occurs throughout the educational continuum, early childhood expulsions and suspensions are a growing concern and are linked to further problems in kindergarten and beyond. With national attention from civil rights organizations drawing eyes to the injustices, educational systems are looking to solve the over-use of suspension and expulsion to address student behavior. Behavior analysts are often tasked with addressing and reducing the behavioral concerns of students; however, there is a gap in the behavior analytic literature on racism in schools. Bringing awareness to anti-Black racism in American schooling is an initial step for behavior analysts to take toward dismantling oppressive systems within education.
Chapter
Full-text available
...“Our Foundational Covenant: Strengthening the Family” [is] the newest addition to The Covenant With Black America. This supplemental chapter is one of several projects that Jamestown is releasing in collaboration with Tavis Smiley to operationalize The Covenant With Black America. ....As the great civil rights pioneer and matriarch, Dr. Dorothy I. Height, reminds us, “we have survived because of family.” Placing this survival in the context of a long-standing historic struggle highlights the unique characteristic of the Black family - resilience.
Article
Data from the US Department of Education clearly documents the chronic and persistent disproportionality of negative educational outcomes for students of color. To move closer to an antiracist system that provides all youth with the resources, protections, and opportunities to which they are entitled through public education, we recommend that mental health clinicians understand the social determinants of education; become familiar with the historical legacy of inequity in schools; identify current trends of racial disparities in education; engage in opportunities for antiracist school transformation; and reflect on their personal practices in providing access, diagnosis, and treatment to underresourced and minoritized youth.
Article
Full-text available
Many researchers and policymakers recognize the importance of preschool education, particularly for low-income children, and the role of state governments has become increasingly prominent in providing services. At the same time, some policy analysts have urged that Head Start should be shifted from federal to state control. Others, however, believe this is inadvisable. We sought to inform the debate over whether entrusting Head Start to the states would be a sound policy decision by examining extant state preschool programs. Our survey of 31 programs that matched our criteria suggested that state-funded initiatives vary widely in design, implementation, and quality. There are some common areas of strength and weakness: States tend to be strong in domains such as classroom quality, but are weak on providing comprehensive services. Overall, we find reason for serious concern regarding Head Start devolution from federal to state control at this time. (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Article
Full-text available
Acknowledgements: The authors acknowledge Steven Abramovitz for his assistance in the collection and organization of the data for updating this survey, as well as the many state officials who provided the information. Thanks, too, to Edward Zigler for his guidance throughout this project, and to Sally Styfco for her fine editorial assistance.
Article
This third biennial "Map and Track" examines state-level efforts to promote positive outcomes for young children, including efforts to enhance children's economic security, healthy growth and development, and school readiness. The report provides information about the following types of state-funded child development and family support efforts in each of the states and the District of Columbia: (1) state supported programs with a central focus on child development or family support; (2) states' efforts to prepare children for school and state kindergarten requirements; and (3) early childhood systems development. Analysis was based on information from the Current Population Survey. Chapter 1 of the report provides an overview of the framework and data collection methods. Chapter 2 provides the findings on state child development and family support efforts. Chapter 3 summarizes secondary source data on state efforts to promote family economic security. Chapter 4 reviews the major findings and provides suggestions for future state early childhood efforts. Following these chapters are individual state profiles for the 50 states and the District of Columbia, accompanied by a reader's guide that provides detailed information about the data, data sources, and dates of data collection. Three appendices include demographic summary tables, technical information on demographic statistics, and tables delineating staff efforts to promote family economic security. (KB)
Article
Rates and predictors of preschool expulsion and suspension were examined in a randomly selected sample of Massachusetts preschool teachers (N = 119). During a 12-month period, 39% of teachers reported expelling at least one child, and 15% reported suspending. The preschool expulsion rate was 27.42 per 1000 enrollees, more than 34 times the Massachusetts K-12 rate and more than 13 times the national K-12 rate. Suspension rates for preschoolers were less than that for K-12. Larger classes, higher proportion of 3-year-olds in the class, and elevated teacher job stress predicted increased likelihood of expulsion. Location in a school or Head Start and teachers' positive feelings of job satisfaction predicted decreased likelihood of expulsion. Expulsion was relatively rare in classes where both class size and teacher job stress were low. A higher proportion of Latino children in the class and lower teacher job satisfaction predicted an increased likelihood of suspension. Implications are discussed regarding policy, prevention, and future research.
Article
The number of state-funded preschool programs for low-income children has increased dramatically over the past few decades, and recent research has indicated that these programs vary considerably along a variety of dimensions. By 1998 only 13 of the current 33 state preschool programs (which serve children 3 to 5, provide some form of classroom-based educational service, and are primarily funded and administered at the state level) had completed a formal evaluation of the program’s impact on child outcomes. This paper presents a critical meta-analytic review of these evaluations, providing measures of standardized effects for all significant impacts to facilitate comparisons across differing domains of outcome and evaluative methods. Although several methodological flaws in these studies are identified, the pattern of overall findings may offer modest support for positive impacts in improving children’s developmental competence in a variety of domains, improving later school attendance and performance, and reducing subsequent grade retention. Significant impacts were mostly limited to kindergarten and first grade; however, some impacts were sustained several years beyond preschool. The results of these studies were similar to evaluations of other large-scale preschool programs for low-income children, such as Head Start. Modest outcome goals are warranted for preschool programs serving low-income children, for example, the promotion of school readiness. Suggestions are presented for improved preschool and early intervention program evaluation.
Article
To summarize the difficulties involved in translating tests, to describe the translation methods and the test validation procedures, and to apply those to a personality test. The revised Freiburg Personality Inventory (FPI-R) was translated, then subjected to the following test validation methods: backtranslation, pretest, and review by a carefully selected expert committee. We used a literature review to clarify FPI-R translation problems. These include in particular the different types of equivalence between the source language and the target language (for example, semantics and idioms, as well as experiential and conceptual equivalence). Statistical validation procedures are employed in principle only. The current method combining translation with backtranslation is not sufficient and must be used with, at least, a pretest and step-by-step review by an expert committee. The presence of unilingual experts to explain the smallest details of the target language, which bilingual experts could miss, seems to be mandatory.
The evidence base for mental health consultation in early childhood settings: Research synthesis and review
  • E M Brennan
  • J R Bradley
  • M D Allen
  • D F Perry
  • A Tsega
Brennan, E. M., Bradley, J. R., Allen, M. D., Perry, D. F., & Tsega, A. (2005, March). The evidence base for mental health consultation in early childhood settings: Research synthesis and review. Paper presented at the Conference on Establishing the Evidence Base for Early Childhood Mental Health Consultation, Tampa, FL.
State efforts to evaluate the effects of prekindergarten: 1977-2003. Available: nieer.org/resources
  • W S Gilliam
  • E F Zigler
Gilliam, W. S., & Zigler, E. F. (2004). State efforts to evaluate the effects of prekindergarten: 1977-2003. Available: nieer.org/resources/research/StateEfforts.pdf.