Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2021 Jun 15.
Published in final edited form as: Child Youth Serv Rev. 2018 Jul 18;93:117–125. doi: 10.1016/j.childyouth.2018.07.017

A longitudinal analysis of school discipline events among youth in foster care

Brianne H Kothari a,*, Bethany Godlewski a, Bowen McBeath b, Marjorie McGee b, Jeff Waid c, Shannon Lipscomb a, Lew Bank b
PMCID: PMC8204670  NIHMSID: NIHMS1597058  PMID: 34135541

Abstract

Youth in foster care experience major deficits on standardized measures of academic functioning, are at high risk of academic failure, and are more likely than their non-foster peers to be disciplined at school. School discipline-related problems increase risk of problematic educational and behavioral outcomes including dropping out of school, repeating a grade, and engagement in delinquent and criminal behavior. Identifying which youth are at greatest risk for experiencing school discipline is needed in order to improve the educational experiences of youth in foster care. The current investigation examined the effects of youth and contextual characteristics on school discipline events among 315 youth in foster care. Results revealed that being male, in a higher-grade, and a student of color, living apart from one's sibling, and school mobility significantly predicted discipline events. An additional statistical model divided youth into groups based on race, sex, and disability status taking into account the multiple identities youth have. These results suggest that gender, race, and disability status cumulatively inform school discipline experienced among youth in foster care.

Keywords: School discipline, Foster Care, Child welfare

1. Introduction

Schools are expected to provide the most appropriate and least restrictive learning environment for children (Jacob & Hartshorne, 2007). Providing this learning environment for children and youth involved in the formal child welfare system, however, may be particularly challenging due to the previous maltreatment, placement instability, and disruptions in school stability they experience (Berger et al., 2015; Ferguson & Wolkow, 2012; Trout, Hagaman, Casey, Reid, & Epstein, 2008). Compared to their non-foster peers, youth in foster care show major deficits on standardized measures of academic functioning, are at high risk of school failure (Trout et al., 2008), and are less likely to move on to post-secondary education (Pecora et al., 2006). Foster youth are also more likely than their counterparts to experience grade retention and absenteeism (Stone, 2007). The cumulative risks foster youth face including maltreatment, poverty and parental mental health challenges predict the poorest educational outcomes (Crozier & Richard, 2005).

In addition, research suggests that foster youth are three times more likely than their peers to experience disciplinary events in school settings (Kortenkamp and Erhle, 2002). A longitudinal study that tracked youth over a six year period demonstrated that youth in foster care reported more discipline problems in school compared to children raised in non-foster families (Blome, 1997). Exclusionary school discipline events such as suspensions and expulsions lead to lost class time and less exposure to academic subjects that subsequently help students pass state achievement tests and help prepare them for graduation (Marrus, 2015). Discipline-related problems also place youth in foster care at greater risk of negative educational outcomes including repeating a grade, dropping out of school, (Lee, Cornell, Gregory & Fan, 2011) and engaging in delinquent and criminal behavior (Gregory, Skiba, & Noguera, 2010; Marrus, 2015).

School discipline events and their academic and social consequences may be particularly problematic for specific subgroups of youth. For example, the U.S. Department of Education reported that male and female black students were expelled three times more often than white students, and students with disabilities were twice as likely as their peers to receive out-of-school suspension (U.S. Department of Education Office of Civil Rights, 2014). Suspensions and expulsions may intensify academic deterioration; when students are provided with no immediate educational alternative, student alienation, distrust of teachers, and delinquency may also result (Christie, Jolivette, & Nelson, 2005).

Child welfare legislation requires agencies to address the educational wellbeing of all youth in foster care (Gustavsson and Ann, 2012), and paying closer attention to school discipline events may be an important step towards this goal. Evidence indicates that many youth in foster care experience school discipline events which can have considerable consequences for their educational wellbeing. Youth in foster care are a heterogeneous group, however, and certain subgroups of youth in care might be more likely than others to experience school discipline events than others. Therefore, there is a pressing need to better understand factors associated with school discipline events among youth in foster care. By identifying which youth are at greatest risk for experiencing school discipline trajectories, it may be possible to improve the educational experiences of youth in foster care. The current study concerns the administration of discipline in public school settings for youth in foster care. The purpose of this research is to understand how youth and contextual characteristics are related to school discipline events and subsequently inform research and practice with the systems that serve youth in foster care.

2. Literature review

2.1. Understanding school discipline among youth in Foster Care

Research has established that exclusionary school discipline events hinder academic performance and contribute to racial disparities in achievement (Morris & Perry, 2016). However, there are few empirical peer-reviewed studies on school discipline among youth in foster care. School-related discipline events have significant consequences; they can lead to missed learning days, loss of educational opportunities, and even referrals to the juvenile justice systems (Marrus, 2015). For youth in foster care, the consequences of school discipline events can be exacerbated by the instability they experience in their home and family lives, and could potentially impact youth's permanency plans (American Bar Association, Education Law Center, and Juvenile Law Center, 2014).

Much of the prior research on school discipline utilize administrative educational data (e.g., Blome, 1997; Festinger, 1983; Sawyer & Dubowitz, 1994; Zima et al., 2000) and most do not go beyond descriptive reporting of discipline event prevalence rates. Scherr's (2007) international meta-analysis of disciplinary action only incorporated 10 studies that examined school discipline events. The majority were unpublished reports (i.e., 5 reports, 3 journal articles, 1 book, and 1 dissertation) and all were from the United States and Australia. In these studies discipline events were often examined as one of many ways to look at educational performance or wellbeing outcomes rather than the primary focus of the study. Results of the meta-analysis revealed that 24% of students in foster care had been suspended or expelled from school at least once (Scherr, 2007).

There is significant methodological heterogeneity among studies of school discipline for youth in care, particularly in how school discipline is operationalized, sampling, and timing of data collection (e.g., retrospective research design). Suspensions and expulsions are commonly included in the operationalization of school discipline, but some studies have included all disciplinary events. For example, Smithgall et al. (2005) included violations of CPS District's Uniform Discipline Code reported by schools to the CPS Bureau of Safety and Security., Some studies of school discipline have focused on self-reported data (Blome, 1997), and most of these self-reported studies rely on retrospective reports (Courtney et al., 2004; McMillen, Auslander, Elze, White, & Thompson, 2003). Other studies have relied on different reporting agents such as foster parents (Zima et al., 2000), teachers (Sawyer & Dubowitz, 1994), or administrative data (e.g., Castrechini, 2009; Smithgall et al., 2005).

In summary, the extant research is very limited and reflects methodological challenges relating to: a) variation in how discipline events have been operationalized; b) variation in the reporting agent/source of information used to gather information (i.e., self-report, foster parent report, administrative data, etc.); and c) the lack of prospective, longitudinal, and published studies.

2.2. Factors related to school discipline events and educational outcomes

Given the limited research focused on school discipline events among youth in foster care, this section reviews what is known about the factors related to school discipline and other educational outcomes more generally. Previous studies have more commonly examined educational outcomes such as test scores, grade retention and graduation rates (see O'Higgins et al., 2017), and two domains in both the general and child welfare literature include specific factors that have shown to be associated with school discipline events as well as other educational outcomes. These include (1) youth characteristics; and (2) contextual characteristics. The following subsections review the specific youth and contextual characteristics that have been associated with school discipline and/or educational outcomes among both the general population as well as youth in foster care.

2.2.1. Youth characteristics

National and state data indicate that males, students of color, and students with disabilities tend to be overrepresented among children and youth experiencing discipline events (Krezmien et al., 2006; Marrus, 2015; CRDC, 2014; Vincent et al., 2012). African American or Black male students are particularly overrepresented in regards to suspensions in school (Costenbader & Markson, 1998; Krezmien et al., 2006; Raffaele Mendez & Knoff, 2003). These racial and gender differences for school discipline remain even when controlling for socioeconomic status (Skiba et al., 2002). Students with disabilities also experience more discipline events than their peers, particularly those with learning disabilities (Shifrer et al., 2011). In addition, students of color, particularly black males, are more likely to be labeled emotionally disturbed (Osher et al., 2002; Parrish, 2002); and students of color with disabilities identified as having emotional and behavior difficulties are more likely to be suspended as compared to peers without disabilities (Krezmien et al., 2006).

Among youth in foster care, associations between youth characteristics and school discipline and educational outcomes have also been explored. O'Higgins, Sebba and Gardner (2017) conducted a recent systematic review across a 26-year period focused on factors that have been associated with educational outcomes generally for children in foster and kinship care. They concluded that male gender, ethnic minority status, and youth with special education needs consistently predicted poor educational outcomes (O'Higgins et al., 2017). It should also be noted that youth in foster care are overrepresented in special education, and the number of foster youth qualifying for special education services has steadily increased over the past couple of decades (Scherr, 2007). In addition, a study which focused on school discipline events among youth in foster care found that older age, being male, and receiving a positive screening for a clinical behavior problem significantly increased the odds of ever being suspended or expelled from school (Zima et al., 2000). These findings emphasize that key youth characteristics are related to school discipline events.

2.2.2. Contextual characteristics

Ecological features of youths' microsystem have also been studied in relation to youths' educational outcomes and school discipline events. In the general population, school mobility has shown to have negative effects on youth's educational outcomes (Grigg, 2012; Herbers et al., 2012; Mehana & Reynolds, 2004). Previous research has also shown that family (e.g., parental supervision), school (e.g., student-teacher trust), and neighborhood contextual factors (e.g., collective efficacy) are correlated to school suspension (Kirk, 2009). Contextual disadvantage has also been associated with school discipline events; for example, Sheryl et al. (2014) demonstrated that the socioeconomic status of the school was related to school suspension. Empirical examination of contextual characteristics in relation to school discipline, however, has been limited.

Among child welfare populations, home/placement characteristics such as stability, placement with kin, and placement with sibling(s) are commonly examined contextual characteristics (Hegar & Rosenthal, 2009, 2011; Winokur et al., 2014). Foster care placement instability has been shown to predict lower educational outcomes for children and youth in foster care (Villegas, Rosenthal, O'Brien, & Pecora, 2014). Zima et al. (2000) found that in addition to youth characteristics (i.e., older age, male gender, clinical behavior problem), youth living in foster care for longer periods of time were more likely to have been suspended or expelled at school. Winokur et al. (2014) systematic review demonstrated no significant differences between youth in kinship placements and youth in non-relative placements for educational attainment; however, Rubin et al. (2008) found that children in kinship care, had fewer behavioral problems three years later than did youth in non-relative care.

Researchers have also reported a relationship between sibling co-placement and academic outcomes; Hegar and Rosenthal (2011) found that teacher ratings of academic performance, were higher in groups of co-placed siblings compared to youth who were separated from their sibling(s). Similarly, sibling co-placement was uniquely associated with higher educational competence among a sample of emancipated foster youth (Richardson & Yates, 2014).

Similar to research focused on youth in the general population, school mobility has shown key factor related to educational success among youth in foster care. Youth in foster care have a school transfer rate that is twice as high as for other youth (The National Working Group on Foster Care and Education, 2006). Pears et al. (2015) demonstrated that school mobility was associated with poorer socio-emotional competence among children in foster care. School mobility has shown to be significantly associated with an increase in behavioral problems among youth in foster care (Sullivan, Jones & Mathieson, 2010), and more school moves are also associated with a higher dropout rate among youth in foster care (Schroeter et al., 2015). These findings indicate that contextual factors have shown to be related to educational outcomes, and there is a clear need to better understand the ways in which foster youth's contexts might be related to school discipline events experienced.

2.2.3. Intersectional approach

Intersectionality is a growing multidisciplinary framework that provides a prism for understanding multiple dimensions of individuals' lived experience (Cole, 2009). This approach highlights the role inequality plays by examining multiple categories of group membership (e.g., gender, race, disability status) and how they position individuals and impact their experiences and outcomes (Cole, 2009). A framework informed by intersectionality assumes no individual can be defined by one personal characteristic or social status, and lived experiences vary by the intersection of their identities including but not limited to race, gender, and disability.

In quantitative research, taking an intersectional approach helps to provide a “signal” or marker of who is most impacted by a particular phenomenon. In this study, intersectionality is used as a social-justice oriented analytic tool (Cole, 2009) to help draw attention to foster youth with multiple layers of marginalization (e.g., race, gender, disability status). Researchers have utilized quantitative data to examine intersectionality of various social inequities (Sen, Iyer, & Mukherjee, 2009) and recommend creating indicator variables for each intersecting category, so that each can be compared, and also recommend treating one group as a reference category (e.g., white females without disabilities). In this way, ‘social inequities [can be examined] across the spectrum, not just between the extremes’ (p. 397). This study provides an opportunity to contextualize the extent to which youth in care are disciplined at school, and considers youth's social statuses and identities as well as contextual factors such as living apart from a sibling, placement away from kin, and school mobility.

3. Current investigation

Longitudinal research examining associations between youth and contextual factors and school discipline events is necessary to better understand and improve educational outcomes for youth in foster care. Using an intersectional approach, the current investigation examined key youth and contextual characteristics to identify which youth are more likely to experience discipline events at school, and considers the consequences for youth having multiple marginalized identities or statuses in their discipline experiences. Therefore, two primary questions motivated this investigation:

  1. Which specific youth and contextual characteristics predict school discipline events for youth in foster care over two academic years?

  2. Does an intersectional approach help to explain the impact of youth's characteristics on discipline events among youth in care?

This study utilized data from an existing study of siblings in foster care, merged with state administrative educational data on these same youth—to examine the effects of youth and contextual characteristics on discipline events over time. It was hypothesized that youths' personal characteristics including being in an older grade, male, a student of color, and having a disability (measured as being enrolled in special education) would be associated with more school disciplinary events. It was also hypothesized that contextual characteristics such as placement with non-kin, placement apart from siblings, and school instability would be associated with school discipline events. Finally, youth who intersected multiple marginalized demographic characteristics (i.e., males, students of color, and students with disabilities) were expected to experience more disciplinary events compared to their counterparts.

4. Method

4.1. Supporting Siblings in Foster Care

Data for this investigation were drawn from Supporting Siblings in Foster Care (SIBS-FC; Kothari et al., 2014; Kothari et al., 2017; McBeath et al., 2014) study, which is the first large scale randomized clinical trial to evaluate a sibling relationship development intervention for preadolescent and adolescent foster youth. SIBS-FC was a five year RCT and each participant was followed for an 18-month period of time. Sibling dyads living together and apart participated in the study. The study utilized a universal recruitment strategy that drew from the population of sibling pairs from a three county metropolitan region in Oregon. To be eligible for participation, youth had to be in the legal custody of the local child welfare agency at the time of study enrollment, reside in the three county region, and speak English. The study targeted middle-adolescent aged youth, with older siblings in care between 11 and 15 years and younger sibling within four years of age (Kothari et al., 2014; Kothari et al., 2017; McBeath et al., 2014).

Siblings in the intervention group were taught and practiced communication, cooperation, and problem solving skills. They also learned ways to improve their sibling relationship and advocate for themselves. Feasibility results suggest the intervention was delivered with a high degree of fidelity and well-liked by foster youth (Kothari et al., 2014). Efficacy results indicated that sibling relationship quality improved among those in the intervention group at greater rates than controls (Kothari et al., 2014). SIBS-FC utilized a multiple method, multiple indicator data collection and measurement strategy (Chamberlain & Bank, 1989); data were collected from all youth (i.e., older and younger siblings), foster parents, teachers, caseworkers, and outside observers on the major project domains (i.e., mental health, education, quality of life and sibling relationship quality) ds across the 18-months families were enrolled in the study.

At the end of study participation, administrative data were collected from the Oregon Department of Human Services (DHS) and the Oregon Department of Education (ODE). DHS data were gathered by the DHS study liaison, who played a vital role in the SIBS-FC study and was employed by DHS (Kothari et al., 2014). A memorandum of understanding (MOU) was established with ODE to access educational administrative data on all SIBS-FC youth. ODE data collected in the SIBS-FC included variables on academic performance (i.e., math, reading, and ELPA scores), attendance, and discipline events.

4.2. Current study sample

This investigation includes the 315 youth from the larger SIBS-FC with Oregon Department of Education (ODE) data. As ODE data only accounted for students attending public schools in Oregon, 13 of the 328 youth in the SIBS-FC were not able to be matched to educational administrative data because they were either home schooled or had moved to another state. Table 1 presents details about youth characteristics for the whole sample, and for the subgroups of youth with and without discipline events during the time period of the current investigation.

Table 1.

Sample characteristics associated with ever being disciplined at school during study period (n = 315).

Total sample
n (%)		 With discipline events
n (%)		 Without discipline events
n (%)		 χ2(df), p
All youth 315 (100.0) 109 (33.23) 206 (62.8)
Treatment
 Intervention 161 (51.1) 54 (49.5) 107 (51.9) 0.16 (1), p = 0.685
 Control 154 (48.9) 55 (50.5) 99 (48.1)
Grade at baseline 28.51 (9), p = 0.001
 First 2 (0.6) 1 (0.9) 1 (0.5)
 Second 14 (4.4) 2 (0.18) 12 (5.8)
 Third 20 (6.3) 4 (3.6) 16 (7.8)
 Fourth 41 (13.0) 6 (5.5) 35 (17.0)
 Fifth 54 (17.1) 18 (16.5) 36 (17.5)
 Sixth 66 (21.0) 30 (27.5) 36 (17.5)
 Seventh 47 (14.9) 24 (22.0) 23 (11.2)
 Eighth 36 (11.4) 17 (15.6) 19 (9.2)
 Ninth 21 (6.7) 6 (5.5) 15 (7.3)
 Tenth 14 (4.4) 1 (0.9) 13 (6.3)
Gender 15.53 (1), p = 0.000
 Male 160 (50.8) 72 (66.1) 88 (42.7)
 Female 155 (49.2) 37 (33.9) 118 (57.3)
Race/Ethnicity 8.58 (1), p = 0.003
 Student of Color 155 (49.2) 66 (60.6) 89 (43.2)
 White 160 (50.8) 43 (39.4) 117 (56.8)
Disability 6.55 (1), p = 0.010
 Receives Special Education 148 (46.9) 62 (56.9) 86 (41.7)
 No Special Education 167 (53.1) 47 (43.1) 120 (58.3)
Lives with kin 3.47 (2), p = 0.176
 Always 110 (34.9) 31 (28.4) 79 (38.3)
 Sometimes 58 (18.4) 24 (22.0) 34 (16.5)
 Never 147 (46.6) 54 (49.5) 93 (45.1)
Lives with sibling 8.47 (2), p = 0.015
 Always 185 (58.7) 52 (47.7) 133 (64.6)
 Sometimes 68 (21.6) 25 (22.9) 30 (14.6)
 Never 75 (22.87) 32 (29.4) 43 (20.9)
School mobility 9.60 (1), p = 0.002
 Switched schools during academic year 210 (64.0) 85 (78.0) 125 (60.8)
Did not switch schools 105 (36) 24 (22.0) 81 (39.2)
Open in a new tab

Assessments for the SIBS-FC were collected on a 6-month interval timeline beginning at baseline when youth were recruited into the study for a total of four time points. The four time points of SIBS-FC data were used to pull information from statewide educational data from those same date ranges, using enrollment data which included the youth's primary student identification code. After matching youth from the SIBS-FC dataset to their ODE student identification numbers, discipline data were merged into the SIBS-FC dataset.

This study examined discipline events over approximately two academic school years. There was full discipline information for 214 youth across the four time points in this study, 79 youth had discipline information at three time points, 13 youth had discipline information at two time points, and nine youth only had discipline information at one point in time. Youth who were missing discipline data at Time 1 did not significantly differ from youth whose data were not missing when tested by sex, special education status, grade, living with the sibling, living with kin, or being a student of color.

Data for this investigation fell across six academic years 2009–10 to 2014–15, due to the rolling enrollment of the SIBS-FC. The enrollment date for each youth in the SIBS-FC through six months later was used to pull ODE data for Time 1; enrollment date plus 6 to 12 months created Time 2; enrollment date plus 12 to 18 months created Time 3; and enrollment date plus 18 to 24 months created Time 4. There were ODE records for 234 youth at Time 1, for 308 youth at Time 2, for 297 youth at Time 3, and for 291 youth at Time 4, with a total of 1130 records across all time points. Some youth might not have had any observations at a time point from ODE because they were not currently enrolled in public school in Oregon (e.g., being homeschooled or at a crisis center, etc.).

4.3. Measures

4.3.1. Oregon Department of education data and number of discipline events

Number of discipline events was measured as the aggregate of four types of school discipline events: 1.) in-school suspension (n = 32); 2.) truancy/attendance policy violation (n = 20); 3.) out-of-school suspension (n = 109); and 4.) expulsion (n = 4). Fifty-five of all offenses were considered by Oregon Department of Education (ODE) to be for minor offenses (e.g., disruption), 50 were for moderate offenses (e.g., verbal aggression, insubordination), and 60 were for serious offenses (e.g., physical aggression, bringing drugs or weapons to school).

There were 165 unique discipline events across all youth and all time points, representing about 15% of possible events in the timeframe of the study, which is consistent with the range of discipline incidences in prior literature (Scherr, 2007). One hundred and nine youth (33.23%) had at least one discipline event during the time frame of this study. ODE policy requires every disciplinary action (i.e., restraint, confinement, expulsion from premises) to be recorded as discrete incidences, even when multiple disciplinary actions are enacted for a singular misbehavior. In order to avoid overestimating the number of active misbehaviors of youth, school discipline was collapsed into a binary variable (0 = no events; 1 = one or more discipline events).

It should also be noted that because SIBS-FC data were matched to ODE data, two covariates were also included in the models. Time (measurement wave) was included in the models to address the slightly different observation calendar for each youth. Treatment was also included because the larger study was a RCT.

4.3.2. Youth characteristics

Youth's individual personal characteristics included grade which can be considered a proxy for youth's age (developmental time), gender (male/female), race (White/Non-White) and special education enrollment (used as a proxy for disability status). These variables were provided by ODE. Because we intended to run models with multiple predictors and a rare event dependent variable, it was necessary to collapse racial categories into a dichotomous variable, recognizing that white students typically experience fewer institutional barriers. Seventy-five youth (23.8%) were identified as Latino, 52 youth (16.5%) as Black, 11 youth (3.5%) as American Indian or Alaskan Native, 6 youth (1.9%) as Asian Pacific Islanders, and 29 youth (9.2%) as having more than one racial or ethnic identity. At baseline, 131 (41.4%) participants were enrolled in elementary school, 149 (47.3%) were enrolled in middle school, and 35 (11.1%) were enrolled in high school (see grade breakdown in Table 1). One hundred and sixty youth (50.8%) were male. At baseline, 148 youth (46.9%) were enrolled in special education and remained enrolled in special education services throughout the study period. Additional details can be found in Table 1.

An inter-categorical variable was created to examine the second research question regarding discipline events by the intersections of gender, race and disability status (measured by enrollment in special education). Following the guidance of Sen et al. (2009), we created a heuristic matrix and a variable with eight categories (see Table 2). The percentage of students represented by these categories ranged from 6.7% (white females with disabilities) to 18.4% (female students of color with disabilities).

Table 2.

Number and percentage of youth in foster care within the intersectional categories (n = 315).

Total sample
n (%)		 With discipline events
n (%)		 Without discipline events
n (%)
Female, White, without Disabilities 48 (15.2) 6 (5.5) 42 (20.4)
Female, White, with Disabilities 21 (6.7) 6 (5.5) 15 (7.3)
Female, Students of Color without Disabilities 58 (18.4) 15 (13.8) 43 (20.9)
Female Students of Color with Disabilities 28 (8.8) 10 (9.2) 18 (8.7)
Males, White, without Disabilities 30 (9.5) 11 (10.1) 19 (9.2)
Males, White, with Disabilities 43 (13.7) 15 (13.8) 28 (13.6)
Male, Students of Color without Disabilities 31 (9.8) 15 (13.8) 16 (7.8)
Male, Students of Color with Disabilities 56 (17.8) 31 (28.4) 25 (12.1)
Total 315 (100.0) 109 (34.6) 206 (65.4)
Open in a new tab

4.3.3. Contextual characteristics

Contextual characteristics included two key home/placement characteristics as well as one school characteristic. The two placement characteristics included living with kin (0 = no; 1 = yes) and living with sibling in the SIBS-FC (0 = no; 1 = yes). One hundred and ten youth (34.9%) lived with kin at each time point, and 185 youth (58.7%) lived with their sibling during the entire study. School mobility was also examined as a school characteristic, with 210 youth (64.0%) switching schools at least once during the academic year. At each time point, youth were noted as stable if they did not enroll in a new school mid-year.

5. Analytic plan

Hierarchical linear logistic regressions with random effects were estimated to examine the impact of youth characteristics and contextual characteristics on school discipline events.

Hierarchical linear modeling allows for observations (n = 1130; 1 to 4 per youth in this study) to be nested in individuals (n = 315) over time, and for individuals to be nested inside of subpopulations, such as students of color or males. Therefore, HLM can work with data even when the independence assumptions that many statistical analyses rely on are violated by multiple measurements within a person and within sibling pairs. HLM uses all available data to estimate group-level effects, even for groups with small sample sizes without overstating or understating prediction error through partial pooling of parameter estimates (Gelman & Hill, 2009). Group memberships in HLM are assumed to account for some unobserved information; for example, youth who are students of color, female, and have a disability likely have unobserved similarities in their experience of school life that influences their likelihood of having a discipline event, but also share some similarities with male students of color with disabilities.

This modeling approach allowed some predictors (i.e., co-placement with sibling, living with kin, grade, and whether the student had switched schools) to vary over time as repeated measures by individual, while other predictors (i.e., youth gender, disability status, and race) were constant over time and treated as non-level variables. The models used random effects to estimate random intercepts that varied by individual given personal, placement, and school characteristics. Robust standard errors were estimated with the cluster command in Stata 15, specifying the sibling groups as clusters. The estimated coefficients were transformed into odds ratios by Stata 15.

Two models were estimated to predict school discipline events. In the first model, youth characteristics including grade (proxy for youth's age), race, gender and disability status (enrollment in special education), and contextual characteristics including placement with kin, placement with sibling, and school instability were incorporated as separate independent variables. The second model reflected the literature on intersectionality and introduced grouping variables by sex, race, and disability status to test whether youth experienced the intersection of these three identities at school in a way that made them more likely to experience a discipline event.

6. Results

Descriptive findings from this sample of youth in foster care revealed that 17.5% of this sample experienced one or more school discipline events at baseline. However, the prevalence rate was slightly different by time (Time 1 = 17.5%, Time 2 = 15.3%, Time 3 = 10.1%, and Time 4 = 16.2%). Furthermore, 109 youth (33.23% of the sample) experienced one or more discipline events across the two-year study period. Sixty-six percent of school discipline events involved males, and 65.2% of the foster youth disciplined were students of color (Table 1). Ninety-two of the school discipline events (55.8%) were associated with students with disabilities compared to 73 discipline events (44.2%) for students without disabilities, revealing that students with disabilities were significantly more likely to have discipline events [χ2 = 5.87 (1), p = 0.015]. In addition, the descriptive statistics from the inter-categorical variables revealed that there were significant differences between the intersectional groups for number of youth experiencing discipline events [χ2 = 25.99(7); p = 0.001]. Twenty-eight percent of youth with discipline events were male students of color with disabilities, who comprised less than a fifth of this sample (17.8%). Male students of color with disabilities (55.4%) and without disabilities (48.4%) had the highest proportion of youth with discipline events, and white females without disabilities (12.5%) had the lowest proportion of youth with discipline events (Table 3).

Table 3.

Factors predicting discipline events for youth in foster care (n = 315).

Model 1
 Model 2
OR Std. Err. z p [95% Conf. Interval] OR Std. Err. z p [95% Conf. Interval]
Treatment 0.880 0.218 −0.51 0.607 [0.541, 1.432] 0.916 0.224 −0.36 0.719 [0.566, 1.480]
Grade 1.301 0.087 3.96 0.000 [1.142, 1.483] 1.306 0.092 3.79 0.000 [1.138, 1.499]
Male 2.386 0.636 3.26 0.001 [1.415, 4.024] – – – – –
Student of Color 1.752 0.426 2.31 0.021 [1.089, 2.821] – – – –
Special Education 1.236 0.327 0.80 0.424 [0.735, 2.079] – – – – –
Switched Schools 1.531 0.272 2.40 0.017 [1.081, 2.170] 1.549 0.279 2.43 0.015 [1.088, 2.206]
Lives with NonKin 0.769 0.194 −1.04 0.298 [0.469, 1.261] 0.764 0.191 −1.08 0.282 [0.468, 1.247]
Siblings Apart 1.664 0.378 2.25 0.025 [1.067, 2.596] 1.693 0.383 2.32 0.020 [1.086, 2.639]
White, Male, No SpEd – – – – – 5.864 3.343 3.02 0.002 [1.863, 18.454]
SOC, Female, No SpEd – – – – – 2.556 1.311 1.83 0.068 [0.935, 6.987]
SOC, Male, No SpEd – – – – – 4.918 2.572 3.05 0.002 [1.765, 13.706]
White, Female, SpE – – – – – 2.161 1.316 1.26 0.206 [0.655, 7.131]
White, Male, SpE – – – – – 4.211 2.307 2.62 0.009 [1.439, 12.323]
SOC, Female, SpE – – – – – 3.614 2.100 2.21 0.027 [1.157, 11.288]
SOC, Male, SpE – – – – – 7.215 3.554 4.01 0.000 [2.748, 18.944]
Time 0.785 0.077 −2.52 0.011 [0.651, 0.947] 0.783 0.075 −2.57 0.010 [0.650, 0.943]
Cons 0.007 0.005 −7.37 0.000 [0.002, 0.025] 0.007 0.005 −7.24 0.000 [0.002, 0.028]
Wald χ2(df) 51.75(9), p < .000 50.38(13), p < .000
Open in a new tab

Note: Youth Characteristics: SOC = Student of Color, SpEd = Enrolled in Special Education (Proxy for Disability Status).

The Intersectional variables combine youth characteristics; the referent group is white females without disabilities.

Foster youth who had at least one discipline event typically had only one discipline event across all four time points (n = 60; 55.1% of youth with any discipline events). Forty-nine youth had two or more discipline events across all four time points (15.5% of total sample). Most discipline events resulted in out-of-school suspensions or expulsions (n = 113; 68.5%) or in-school suspensions (n = 32; 19.4%).

6.1. Youth and contextual characteristics

Consistent with our hypotheses, the first model revealed that higher grade level, being male, and being a student of color significantly increased the odds of youth experiencing school discipline events. Disability status (measured by enrollment in special education) and treatment status did not approach significance. The time variable controlled for the effect of time of year in which the youth enrolled in the SIBS-FC. In addition, the child welfare-focused contextual variables for the youths' living situations revealed that living apart from one's sibling significantly increased the odds of experiencing a discipline event. Living in a non-kin placement, however, did not significantly increase the odds of experiencing school discipline events. The school-focused contextual variable, school instability, also significantly increased the odds of experiencing school discipline events, as expected.

6.2. Intersectional approach

The second model divided youth into groups based on race, sex and disability status in order to capture who is more likely to be disciplined after taking into account multiple identities: race, sex and disability are predictors that cannot be parceled out within a person. Compared to white females without disabilities (the referent group), who were the least likely to experience a discipline event in this sample, all male groups had significantly higher odds of experiencing discipline events. Five of the seven subgroups compared to the referent group were significantly more likely to experience discipline events. However, female students of color without disabilities, only approached significance (p = 0.07), and white female students with disabilities did not significantly differ from white females without disabilities in odds of experiencing discipline events (p = 0.246).

Post-estimation tests revealed that male students of color with disabilities had significantly higher odds of experiencing discipline than female students of color without disabilities [χ2 = 6.74(1); p = 0.009], and higher odds than white females with disabilities [χ2 = 4.49(1); p = 0.034). Consistent with the previous model, school instability and living apart from one's sibling remained significantly associated with odds of experiencing discipline events, and living away from kin and treatment status did not significantly increase odds. Higher grade level continued to be associated with higher odds of experiencing school discipline events.

7. Discussion

Youth in foster care experience tremendous educational challenges (Trout et al., 2008) and are more likely than their peers to experience school discipline events (Kortenkamp & Ehrle, 2002) which can further exacerbate problematic educational outcomes. Results from this longitudinal investigation add to the limited empirical work that has focused on school discipline events among youth in foster care. Findings demonstrated that key youth (i.e., male gender, non-White race, older grade) and contextual characteristics (i.e., living apart from sibling, instability at school) independently predicted school discipline events among youth in foster care. The intersectional approach also helped to identify which subgroups of youth in foster care were at most risk of experiencing school discipline events. In addition, intersectional findings highlighted that disability status also played a role in the school discipline events foster youth experienced.

7.1. Youth and contextual characteristics

Among youth in this sample, 33.2% experienced one or more discipline events across the 2-year study period. This proportion is on the higher end of the range reported in Scherr's (2007) meta-analysis (99% Confidence Interval range 15–36%); however, it is very similar to the 32% reported in Kortenkamp & Ehrle (2002). The school discipline numbers in this study came from the school district rather than retrospective reports from other informants (e.g., former foster youth or foster parents reporting on youth's experiences).

With respect to youth characteristics and consistent with previous research (Bernedo et al., 2012; Burke & Nishioka, 2014; Zima et al., 2000), the current study found that males, students of color, and students with disabilities experienced higher rates of discipline as compared to other students. However, disability status did not independently predict discipline events as expected. Given that the effect of being male on the odds of experiencing a discipline event was so large, it is possible that some of the predictive power of being a student of color or a student with a disability was overshadowed. In addition, being in an older grade was also significantly predictive of discipline events. Previous research suggests that the level of oppositional or externalizing behaviors tends to increase among older youth, who at the point of the current study were three times more likely in Oregon school districts to receive out-of-school suspensions and expulsions than youth before high school (Burke & Nishioka, 2014). School discipline events peaked in middle school; therefore, future intervention and prevention work may want to focus on this particular educational and developmental period.

Among contextual characteristics, the role of child welfare-focused placement characteristics in predicting discipline events yielded interesting findings. Placement in kinship or non-relative care did not appear to influence the likelihood of experiencing discipline events at school. This non-significant finding might be a function of the school setting being a distinct and somewhat separate dimension of the youth's ecological environment from home (Winokur et al., 2014). However, living apart from one's sibling increased the odds of youth experiencing school discipline events by > 65%. Whereas youth and caregivers tend to interact primarily in household microsystems, sibling dyads and sibling groups may attend the same school and interact within the school settings as well. Sibling contact within the school setting may provide a context for protection, and living together and attending the same school might help siblings maintain some continuity in the face of changing residences, caregivers, and schools. A growing body of research suggests that the sibling relationship may be particularly beneficial to youth in care (e.g., Hegar & Rosenthal, 2009, 2011). Our findings imply that this beneficial effect might be present in school settings, and adds to the growing body of literature focused on the potential benefits of sibling relationships for youth in foster care in educational settings. Future studies should continue to explore the role of contextual characteristics and the protective potential of sibling relationships on the educational outcomes of youth in foster care.

7.2. Intersectional approach

In addition, the intersectional approach appears to add value in understanding the multiple, simultaneous aspects of youth vulnerability on their lived experience. Intersectional variables that combined gender, race, and disability status were included in the second model (instead of entering these youth characteristics independently). Compared to white females without disabilities (the referent group), male and female students of color with disabilities were more likely to be disciplined. In fact, the majority of subgroups were significantly more likely to experience discipline events as compared to the referent group. Students of color with disabilities had significantly higher odds of experiencing school discipline events than white students with disabilities, for both males and females. But male students of color with disabilities were much more likely to be disciplined than male students of color without disabilities. These results comport with those reviewed by McGee (2014), which found that disability and race/ethnicity significantly predicted reports of peer victimization.

These intersectional results suggest that disability and race can, in combination, inform foster youth experiences in school, which might include experiencing more school discipline events than their peers. There is an overrepresentation of students of color in disability categories that are usually undiagnosed until school entry, such as emotional disturbance, mild intellectual disability, and learning disability (National Research Council, 2002). These diagnoses may be based on the judgment of school staff, and can include misidentification of students of color with disabilities that are subjectively-based, such as emotional and behavioral disabilities (Togut, 2011). Race, gender, number of suspensions and socioeconomic status are associated with being identified as having a disability, such as emotionally disturbed or learning disability. Thus, it is not surprising that students of color with disabilities - particularly those identified as having emotional and behavioral disabilities - would have increased odds of school discipline events (Krezmien et al., 2006). These findings may support the thesis by Ferri and Connor (2005) that disablism reinforces racism.

Our approach – taking an intersectional analytical lens – helps us to better understand the complexity involved in increased disciplined rates among male students of color with and without disabilities, as well as male students with disabilities, as the social location of youth in care with multiple non-dominant social identities or statuses (e.g. youth of color with disabilities) is complex. If we had not taken an intersectional approach, for example, we would not have found that disability status matters, particularly in the intersections of gender and race.

Intersectional research (intersections of disability, race/ethnicity and sex) reveals numerous appraisal and coping strategies, varying by social location in response to objective (such as violence) and subjective stressors (McDonald, Keys, & Balcazar, 2007; Mitchell, 2006; O'Toole, 2000; Petersen, 2006; Petersen & Gallagher, 2006). Future research should consider how coping strategies of youth in care vary by social identities and statuses, with particular attention to males of color with and without disabilities, and white disabled males. Future work should also continue to identify what program and supports school staff and child welfare workers need to be culturally competent and equity-literate (see Simmons et al., 2018).

8. Strengths and limitations

This study focuses on school discipline events among youth in foster care, an area that warrants further investigation. Addressing school discipline may be one way to address the educational wellbeing of youth in foster care. In addition, this study highlights the value of two methodological approaches in better understanding the educational experiences of youth in foster care: 1) combining longitudinal multi-method multi-agent data with state administrative data; and 2) the intersectional approach in examining multiple youth characteristics simultaneously. Combining datasets with the intersectional approach provides a nuanced understanding of youth's lived experience, and potentially helps identify systemic inequities. Researchers should continue to utilize these approaches in understanding discipline events as well as the educational well-being of youth in foster care.

This study also had limitations that are important to note. This sample consisted of sibling pairs universally recruited from Oregon Child Welfare, but was limited to data available over time in both datasets, which only included youth who remained in Oregon. It is possible that youth who dropped out had more challenges than youth who were present at all waves. Data from this investigation only included approximately two years of time per youth, not the entirety of these youth's in-school experiences. Analytically, adjustments were not made for the multiple comparisons in the analyses. Another limitation was the use of special education status as a proxy for disability status rather than measuring specific types of disabilities (e.g., learning disabilities). Although special education status has been examined as a proxy for disability in other studies, these two categories are not entirely synonymous.

9. Implications, recommendations, and future directions

The results from the study overall suggest that males and students of color experienced more school discipline events. Results from the intersectional modeling signal a need to examine how and why male students of color with disabilities are more likely to be disciplined.

The results of this study also highlight contextual characteristics that may be important to continue to examine. Specifically, findings indicate the potential educational benefit of living with sibling(s). That is, the protective benefit of sibling co-placement may cross-contexts and be visible in the educational outcomes of youth in care. Perhaps siblings living together experience additional support and this leads to better adaptation to the school setting. Similar to other studies, this study also found that school instability influences school discipline events among youth in foster care. Future studies should continue to examine these and other contextual characteristics to better understand the ways in which contextual differences impact school-related outcomes including school discipline.

Given that youth who are disciplined at school tend to miss learning days and experience other academic challenges, attention should be turned to these youth to determine proper supports and/or interventions necessary early. Since disparities exist, a focus should also be placed on applying an equity lens. Child welfare systems and public education systems need to collaborate to better serve youth who intersect these systems (Mason, 2014; Ferguson & Wolkow, 2012). Special efforts should be made to support cross-system communication for youth in foster care generally, but perhaps particularly for youth experiencing school discipline events. Multiple studies have described that school personnel receive little, if any, training on the unique circumstances of children and youth in foster care (e.g., Ferguson & Wolkow, 2012). Training focused on education advocacy may also be helpful for foster caregivers and caseworkers, so that these key adults are aware of the best ways to support youth in school. These efforts may not only improve educational outcomes for these young people generally, but may also help prevent crossover youth in foster care into the criminal justice system. Teacher and staff training could also promote recognition of and attention to the specific needs of students in foster care before students manifest behaviors that staff interpret as disruptive or threatening, leading to missed learning days and/or lost academic progress.

In addition, child welfare agencies and school systems should determine specific practices that promote educational well-being and better ways to approach school discipline for foster youth that supports connection and engagement. Gallegos and White (2013) highlight the importance of implementing promising practices that promote educational success and therapeutic interventions that prevent delinquency, focusing on placement and school stability, transition services, and implementing alternatives to exclusionary school discipline for youth in foster care. In addition, a recent report focused on the general population identified practices that apply an equity lens to youth's social, emotional and academic development; Restorative Justice Practices (RJP) were identified as an approach specific to school discipline (Simmons et al., 2018). RJPs utilize a problem solving approach to school discipline focused on restitution, resolution and reconciliation (Morrison & Vaandering, 2012). Other opportunities reviewed included promoting racial and socioeconomic integration in schools, trauma-informed approaches, cultural competency and equity literacy training, and school-based social and emotional learning (SEL) and mindfulness approaches (Simmons et al., 2018). Perhaps some of these same strategies would be useful within child welfare offices as well.

In sum, research has clearly documented that school settings can counteract or buffer youth's home related risks (e.g., Hopson & Lee, 2011) and a positive educational experience can mitigate the effects of adversity foster youth face (Pecora, 2012). For these reasons, additional research is needed to understand school discipline events among this educationally vulnerable group. This increased understanding may then be used to promote equity and inclusion in education for children and youth involved in the child welfare system.

Acknowledgements

Research support is gratefully acknowledged from the National Institute of Mental Health for the project, ‘Evaluation of Intervention for Siblings in Foster Care,’ (R01MH085438 Lew Bank, PI). The information reported herein reflects solely the positions of the authors.

Footnotes

Declarations of interest

None.

References

  1. American Bar Association, Education Law Center, and Juvenile Law Center (2014). Legal Center for Foster Care and Education Foster Care & Education Issue Brief, www.fostercareandeducation.org.
  2. Berger LM, Cancian M, Han E, Noyes J, & Rios-Salas V (2015). Children's academic achievement and foster care. Pediatrics, 135(1), e109–e116. 10.1542/peds.2014-2448. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bernedo IM, Salas MD, García-Martín MA, & Fuentes MJ (2012). Teacher assessment of behavior problems in foster care children. Children and Youth Services Review, 34(4), 615–621. [Google Scholar]
  4. Blome WW (1997). What happens to foster kids: Educational experiences of a random sample of foster care youth and a matched group of non-foster care youth. Child & Adolescent Social Work Journal 14(1), 41–53. [Google Scholar]
  5. Burke A, & Nishioka V (2014). Suspension and expulsion patterns in six Oregon school districts. Regional Educational Laboratory at Education Northwest. [Google Scholar]
  6. Castrechini S (2009). Educational outcomes for court dependent youth in San Mateo County. Stanford, CA: John W. Gardner Center for Youth and their Communities.che. [Google Scholar]
  7. Chamberlain P, & Bank L (1989). Toward an integration of macro and micro measurement systems for the researcher and the clinician. Journal of Family Psychology, 3(2), 199–205. [Google Scholar]
  8. Christie AC, Jolivette K, & Nelson CM (2005). Breaking the school to prison pipeline: Identifying school risk and protective factors for youth delinquency. Exceptionality, 13, 69–88. [Google Scholar]
  9. Civil Rights Data Collection (CRDC) (2014). Office of civil rights. U.S. Department of Education; https://www2.ed.gov/about/offices/list/ocr/data.html. [Google Scholar]
  10. Cole ER (2009). Intersectionality and research in psychology. American Psychologist, 64(3), 170–180. [DOI] [PubMed] [Google Scholar]
  11. Costenbader V, & Markson S (1998). School suspension: A study with secondary school students. Journal of School Psychology, 36(1), 59–82. [Google Scholar]
  12. Courtney ME, Roderick M, Smithgall C, Gladden RM, & Nagaoka J (2004). The educational status of Foster children. Chicago, IL: Chapin Hall Center for Children. [Google Scholar]
  13. Crozier JCB, & Richard P (2005). Cognitive and academic functioning in maltreated children. Children & Schools, 27(4), 197–206. [Google Scholar]
  14. Ferguson HB, & Wolkow K (2012). Educating children and youth in care: A review of barriers to school progress and strategies for change. Children and Youth Services Review, 34(6), 1143–1149. [Google Scholar]
  15. Ferri B, & Connor D (2005). Tools of exclusion: Race, disability, and (Re)segregated education. Teachers College Record, 107, 453–474. 10.1111/j.1467-9620.2005.00483.x. [DOI] [Google Scholar]
  16. Gallegos AH, & White CR (2013). Preventing the school-justice connection for youth in foster care. Family Court Review, 51(3), 460–468. [Google Scholar]
  17. Gelman A, & Hill J (2009). Data analysis using regression and multilevel/hierarchical models. Cambridge, NY: Cambridge University Press. [Google Scholar]
  18. Gregory A, Skiba RJ, & Noguera PA (2010). The achievement gap and the discipline gap two sides of the same coin? Educational Researcher, 39(1), 59–68. [Google Scholar]
  19. Grigg J (2012). School enrollment changes and student achievement growth. Sociology of Education, 85(4), 388–404. [Google Scholar]
  20. Gustavsson N, & Ann E (2012). Educational policy and Foster youths: The risks of change. Children & Schools, 34(2), 83–91. [Google Scholar]
  21. Hebers J, Cutuli J, Supkoff L, Heistad D, Chan C, Hinz E, & Masten A (2012). Early reading skills and academic achievement trajectories of students facing poverty, homelessness, and high residential mobility. Educational Researcher, 41(9), 366–374. [Google Scholar]
  22. Hegar, & Rosenthal (2009). Kinship care and sibling placement: Child behavior, family relationships, and school outcomes. Children and Youth Services Review, 31(6), 670–679. [Google Scholar]
  23. Hegar RL, & Rosenthal JA (2011). Foster children placed with or separated from siblings: Outcomes based on a national sample. Children and Youth Services Review, 33(7), 1245–1253. [Google Scholar]
  24. Hopson, & Lee (2011). Mitigating the effect of family poverty on academic and behavioral outcomes: The role of school climate in middle and high school. Children and Youth Services Review, 33(11), 2221–2229. [Google Scholar]
  25. Jacob S,, & Hartshorne TS (2007). Ethics and law for school psychologists (5th ed.). Hoboken, NJ: Wiley. [Google Scholar]
  26. Kirk D (2009). Unraveling the contextual effects on student suspension and juvenile arrest: The independent influences of school, neighborhood, and family social controls. Criminology, 47, 479–520. [Google Scholar]
  27. Kortenkamp K, & Erhle J (2002). The well-being of children involved with the child welfare system: A national overview. Retrieved from http://www.urban.org/url.cfm?ID=310413.
  28. Kothari BH, McBeath B, Lamson-Siu E, Webb SJ, Sorenson P, Bowen H, & Bank L (2014). Development and feasibility of a sibling intervention for youth in foster care. Evaluation and Program Planning 47, 91–99. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Kothari, McBeath, Sorenson, Bank, Waid, Webb, & Steele (2017). An intervention to improve sibling relationship quality among youth in foster care: Results of a randomized clinical trial. Child Abuse & Neglect, 63, 19–29. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Krezmien MP, Leone PE, & Achilles GM (2006). Suspension, race, and disability: Analysis of statewide practices and reporting. Journal of Emotional and Behavioral Disorders, 14(4), 217–226. [Google Scholar]
  31. Lee T, Cornell D, Gregory A, & Fan X (2011). High suspension schools and dropout rates for black and white students. Education and Treatment of Children, 34(2), 167–192. [Google Scholar]
  32. Marrus E (2015). Education in Black America: Is It the New Jim Crow? Arkansas Law Review (1968-Present), 68(1), 27–54. [Google Scholar]
  33. McBeath B, Kothari BH, Blakeslee J, Lamson-Siu E, Bank L, Linares LO, & Shlonsky A (2014). Intervening to improve outcomes for siblings in foster care: Conceptual, substantive, and methodological dimensions of a prevention science framework. Children and Youth Services Review, 39, 1–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. McDonald KE, Keys CB, & Balcazar FE (2007). Disability, race/ethnicity and gender: Themes of cultural oppression, acts of individual resistance. American Journal of Community Psychology, 39(1/2), 145–161. 10.1007/s10464-007-9094-3. [DOI] [PubMed] [Google Scholar]
  35. McGee MG (2014). Lost in the margins? Intersections between disability and other nondominant statuses with regard to peer victimization. Journal of School Violence, 13(4), 396–421. [Google Scholar]
  36. McMillen C, Auslander W,, Elze D, White T, & Thompson R (2003). Educational experiences and aspirations of older youth in foster care. Child Welfare, 32(4), 475–495. [PubMed] [Google Scholar]
  37. Mehana M, & Reynolds AJ (2004). School mobility and achievement: A meta-analysis. Children and Youth Services Review, 26, 93–119. [Google Scholar]
  38. Mitchell DD (2006). Flashcard: Alternating between visible and invisible identities. Equity & Excellence in Education, 39(2), 137–145. [Google Scholar]
  39. Morris EW, & Perry BL (2016). The punishment gap: School suspension and racial disparities in achievement. Social Problems, 63(1), 68–86. [Google Scholar]
  40. Morrison B, & Vaandering D (2012). Restorative justice: Pedagogy, praxis, and discipline. Journal of School Violence, 11, 138–155. [Google Scholar]
  41. National Research Council (2002). In Committee on Minority Representation in Special Education, Donovan MS, & Cross CT (Eds.). Minority students in special and gifted education. Washington, DC: National Academies Press. [Google Scholar]
  42. National Working Group on Foster Care and Education (2006). Educational outcomes for children and youth in foster and out-of-home care.
  43. O'Higgins, Sebba, & Gardner (2017). What are the factors associated with educational achievement for children in kinship or foster care: A systematic review. Children and Youth Services Review, 79, 198–220. [Google Scholar]
  44. O'Toole CJ (2000). The view from below: Developing a knowledge base about an unknown population. Sexuality & Disability, 18(3), 207–224. [Google Scholar]
  45. Pears KC, Kim HK, Buchanan R, & Fisher PA (2015). Adverse consequences of school mobility for children in foster care: A prospective longitudinal study. Child Development, 86(4), 1210–1226. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Pecora PJ (2012). Maximizing educational achievement of youth in foster care and alumni: Factors associated with success. Children and Youth Services Review, 34, 1121–1129. [Google Scholar]
  47. Pecora P, Williams J, Kessler R, Hiripi E, O'Brien K, Emerson J, … Torres D (2006). Assessing the educational achievements of adults who were formerly placed in family foster care. Child & Family Social Work, 11(3), 220–231. [Google Scholar]
  48. Petersen AJ (2006). An African American woman with disabilities: The intersection of gender, race and disability. Disability & Society, 21(7), 721–734. [Google Scholar]
  49. Petersen AJ, & Gallagher DJ (2006). Exploring intersectionality in education: The intersection of gender, race, disability, and class (PhD Doctoral)University of Northern Iowa. [Google Scholar]
  50. Raffaele Mendez LM, & Knoff HM (2003). Who gets suspended from school and why: A demographic analysis of schools and disciplinary infractions in a large school district. Education & Treatment of Children, 26(1), 30–51. [Google Scholar]
  51. Richardson, & Yates (2014). Siblings in foster care: A relational path to resilience for emancipated foster youth. Children and Youth Services Review, 47, 378–388. [Google Scholar]
  52. Rubin DM, Downes KJ, O'Reilly ALR, Mekonnen R, Luan X, & Localio R (2008). The impact of kinship care on behavioral well-being for children in out-of-home care. Archives of Pediatrics & Adolescent Medicine, 162(6), 550–556. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Scherr TG (2007). Educational experiences of children in Foster Care: Meta-analyses of special education, retention and discipline rates. School Psychology International, 28(4), 419–436. [Google Scholar]
  54. Schroeter M, Strolin-Goltzman J, Suter J, Werrbach M, Hayden-West K, Wilkins Z, … Rock J (2015). Foster youth perceptions on educational well-being. Families in Society, 96(4), 227–233. [Google Scholar]
  55. Sen G, Iyer A, & Mukherjee C (2009). A methodology to analyze the intersections of social inequalities in health. Journal of Human Development and Capabilities, 10(3), 397–415. [Google Scholar]
  56. Sheryl AH, Stephanie MP, Herrenkohl TI, Toumbourou JW, & Catalano RF (2014). Student and school factors associated with suspension: A multilevel analysis of students in Victoria, Australia and Washington State, United States. Children and Youth Services Review, 36(1), 187–194. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Shifrer D, Muller C, & Callahan R (2011). Disproportionality and learning disabilities: Parsing apart race, socioeconomic status, and language. Journal of Learning Disabilities, 44(3), 246–257. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Simmons DN, Brackett MA, & Adler N (2018). Applying an equity les to social emotional and academic development Edna Bennett Pierce Prevention Research Center: Pennsylvania State University. [Google Scholar]
  59. Skiba R, Michael J, Nardo R, & Peterson S (2002). The color of discipline: Sources of racial and gender disproportionality in school punishment. The Urban Review, 34(4), 317–342. [Google Scholar]
  60. Smithgall C, Gladden RM, Yang D, & George R (2005). Behavior problems and educational disruptions among children in out-of-home care in Chicago. (Chapin Hall Working Paper).
  61. Stone S (2007). Child maltreatment, out-of-home placement and academic vulnerability: A fifteen-year review of evidence and future directions. Children and Youth Services Review, 29(2), 139–161. [Google Scholar]
  62. Sullivan MJ, Jones L, & Mathiesen S (2010). School change, academic progress, and behavior problems in a sample of foster youth. Children and Youth Services Review, 32(2), 164–170. [Google Scholar]
  63. Togut TD (2011). The gestalt of the school-to-prison pipeline: The duality of overrepresentation of minorities in special education and racial disparity in school discipline on minorities. Journal of Gender Social Policy and Law, 20(1), 163–181. [Google Scholar]
  64. Trout AL, Hagaman J, Casey K, Reid R, & Epstein MH (2008). The academic status of children and youth in out-of-home care: A review of the literature. Children and Youth Services Review, 30(9), 979–994. [Google Scholar]
  65. U.S. Department of Education Office for Civil Rights (2014). Civil rights data collection: Data snapshot (school discipline). Issue Brief No. 1.
  66. Villegas S, Rosenthal J, O'Brien K, & Pecora PJ (2014). Educational outcomes for adults formerly in foster care: The role of ethnicity. Children and Youth Services Review, 36, 42–52. [Google Scholar]
  67. Vincent CG, Sprague JR, & Tobin TJ (2012). Exclusionary discipline practices across students' racial/ethnic backgrounds and disability status: Findings from the Pacific northwest. Education and Treatment of Children, 35(4), 585–601. [Google Scholar]
  68. Winokur M, Holtan A, & Batchelder KE (2014). Kinship care for the safety, permanency, and well-being of children removed from the home for maltreatment. Cochrane Database of Systematic Reviews(1) Art. No.: CD006546. [DOI] [PMC free article] [PubMed] [Google Scholar]
  69. Zima BT, Bussing R, Freeman S, Yang X, Belin TR, & Forness SR (2000). Behavior problems, academic skill delays and school failure among school-aged children in Foster Care: Their relationship to placement characteristics. Journal of Child and Family Studies, 9(1), 87–103. [Google Scholar]

RESOURCES