Children and Youth Services Review 34 (2012) 403–408

Contents lists available at SciVerse ScienceDirect

Children and Youth Services Review

j ourna l homepage: www.e lsev ie r .com/ locate /ch i ldyouth

Cross-system barriers to educational success for children in foster care: The frontline perspective

Kathleen Noonan c,⁎,1, Meredith Matone a, Sarah Zlotnik a, Robin Hernandez-Mekonnen d,1,Caroline Watts e, David Rubin a,b, Cynthia Mollen a,b

a PolicyLab, The Children's Hospital of Philadelphia, 34th Street and Civic Center Boulevard, CHOP North - Room 1535, Philadelphia, PA 19104-4399, USAb Department of Pediatrics, The Perelman School of Medicine at the University of Pennsylvania, 3400 Spruce Street, Philadelphia, PA 19104, USAc University of Wisconsin Law School, 975 Bascom Mall, Madison, WI 53706, USAd University of Pennsylvania School of Social Policy and Practice, 3701 Locust Walk, Philadelphia, PA 19104, USAe University of Pennsylvania Graduate School of Education, 3700 Walnut Street, Philadelphia, PA, USA

⁎ Corresponding author at: University of Wisconsin LMadison, WI 53706, USA. Tel.: +1 608 262 2441(Officefax: +1 608 262 5485.

E-mail addresses: kgnoonan@wisc.edu (K. Noonan),(M. Matone), zlotniks@email.chop.edu (S. Zlotnik), rme(R. Hernandez-Mekonnen), wattsca@gse.upenn.edu (C.(D. Rubin), mollenc@email.chop.edu (C. Mollen).

1 Previous full-time affiliation: PolicyLab, The Childr34th Street and Civic Center Boulevard, CHOP North -19104-4399, USA.

0190-7409/$ – see front matter © 2011 Elsevier Ltd. Alldoi:10.1016/j.childyouth.2011.11.006

a b s t r a c t

a r t i c l e i n f o

Article history:Received 14 October 2011Accepted 20 November 2011Available online 8 December 2011

Keywords:Child welfareEducation systemFostering connectionsSystems collaborationPolicy

Children in out-of-home placements are at increased risk for a multitude of poor educational outcomes. Thefederal Fostering Connections to Success and Increasing Adoptions Act of 2008 requires child welfare and edu-cation systems to collaborate towards improved educational outcomes for children in foster care. This studyperformed focus groups with ground-level stakeholders from child welfare and education systems to solicitinformation on the barriers, strategies, and daily experiences of working across systems towards educationalsuccess for children in child welfare in a large, urban area. Ten focus groups were completed with a total of 90study participants. Knowledge and implementation of policies related to the educational needs of children infoster care was highly variable among participants. Across all groups, ineffective and limited cross-systemcommunication, role uncertainty among stakeholders, and prevalence and complexity of behavioral healthneeds among children were identified as barriers to achieving educational success for children in fostercare. Innovative approaches to cross-system collaboration between education and child welfare systemsare needed to affect these positive outcomes. The complexity of mental health service provision for childrenin foster care with behavioral problems will require improved cross-system collaboration.

© 2011 Elsevier Ltd. All rights reserved.

1. Introduction

Evidence has shown that children in foster care experience a rangeof negative health and developmental outcomes (dosReis, Zito, Safer,& Soeken, 2001; Dubowitz, Simms, & Szilagyi, 2000; Halfon,Mendonca, & Berkowitz, 1995; Rubin, Alessandrini, Feudtner,Localio, & Hadley, 2004). When compared to their peers, children inthe child welfare system experience higher rates of grade retention,disciplinary action, behavioral issues, and special education services(Casey Family Programs National Working Group on Foster Care &Education, 2008; Courtney, Roderick, Smithgall, Gladden, &Nagaoka, 2004; Trout, Hagaman, Casey, Reid, & Epstein, 2008).

aw School, 975, Bascom Mall,), +1 215 221 2605(Mobile);

matonem@email.chop.edukonne@sp2.upenn.eduWatts), rubin@email.chop.edu

en's Hospital of Philadelphia,Room 1535, Philadelphia, PA

rights reserved.

Children in foster care also experience educational instability, suchas greater than average delays in enrollment and incomplete academ-ic records, which create obstacles to smooth school transitions andthorough case management (Casey Family Programs NationalWorking Group on Foster Care & Education, 2008). Among a nationalsample of foster care alumni, 68% experienced three or more elemen-tary school placements, while greater than two-thirds reported fiveor more school changes during their time in out-of-home care(Pecora et al., 2005; Smithgall, Gladden, Howard, Goerge, &Courtney, 2004).

The 2008 Fostering Connections to Success and Increasing AdoptionsAct (“Fostering Connections,” or the “Act”) sought to improve educa-tional outcomes for children in foster care (110th United StatesCongress, 2008). Fostering Connections mandates activities for statesto address barriers to educational success for children in foster care.Under the Act – and as a condition to receiving federal child welfarefunds – states must develop a plan to ensure “educational stability”for children in foster care, including assurances that: (1) the place-ment of a child in foster care accounts for the appropriateness ofthe current educational setting and the proximity to the school inwhich the child is enrolled at the time of placement; (2) the Stateagency has coordinated with appropriate local educational agencies

404 K. Noonan et al. / Children and Youth Services Review 34 (2012) 403–408

to ensure that the child remains in the school in which the child is en-rolled at the time of placement; and (3) immediate and appropriateenrollment in a new school if remaining in a home school is not inthe best interests of the child, with all of the educational records ofthe child provided to the school (110th United States Congress,2008).

While public system leaders can adopt new policies, implementa-tion success depends on the will and understanding of stakeholderson the ground (Lipsky, 1980). Recent education reform research hasdemonstrated that how school personnel interprets a policy is criticalto the successful implementation of the policy (Honig, 2006; Spillane,2000). Fostering Connections confounds implementation challengesby requiring two systems – child welfare and education – to align pol-icy goals at the leadership and field levels. Cross-system collaborativemodels developed in response to Fostering Connections will thereforehave to consider implementation challenges in engaging all stake-holders to successfully implement policies in both systems.

States are just beginning to implement the educational require-ments of Fostering Connections (American Bar Association LegalCenter for Foster Care & Education, 2009, 2011). This study takesthe basic – but important – step of soliciting information about thepolicies and processes related to the education of children in fostercare, including cross-system challenges. Focus groups were con-ducted to identify current practices, systemic barriers, and general in-sights into the overall daily experiences from those involved ineducating or supporting the education of children in foster care atthe ground level in a large, urban area. Importantly, focus groupswere not told or prompted about any aspect of Fostering Connections,but asked simply to share their understanding and point of viewabout cross-system issues and challenges. The goal was to distill in-formation from this study for use by policymakers at the local, state,and national levels.

2. Methods

We conducted a focus group study of a purposive sample of adultsinvolved in the foster care and education systems in a large, urbanarea of over one million residents. In 2010, the area school districtserved approximately 150,000 students and employed 11,000teachers. Of the 4100 children in out-of-home placements in 2011,72% were living in foster homes. Of all children in foster care place-ments, 57% or approximately 1700 children were school-aged (ages5–18). Foster care placements included non-kin caregivers (57%)and kinship caregivers (43%). Each child in out-of-home placementis assigned both a city-employed caseworker serving as a case manag-er, as well as a provider agency caseworker to oversee the certifica-tion and supervision of the foster family; there are approximately500 city-employed caseworkers and over 30 contracted provider fos-ter care agencies. The study protocol and consent procedures wereapproved by the institutional review board at the investigators'institution.

2.1. Participants

Subjects were included if they were adults (age 18 years or older)and if they were currently employed in the foster care or educationsystems, including city employed Child Welfare (CW) caseworkers,provider agency caseworkers, teachers, and school counselors. Fosterparents were also included. Additionally, provider agency case-workers were included if they had at least six months experience inthat role; teachers, school counselors, and CW caseworkers were in-cluded if they had at least one year of experience in that role. Fosterparents were included if they had been involved with the child wel-fare system for at least one year. Different levels of experience werechosen to reflect more rapid turnover in the provider agency case-worker group.

Each subgroup had a unique recruitment method, chosen to max-imize recruitment within that group. Provider agency caseworkers,teachers, and school counselors were recruited with flyers placed inthe mailboxes at their places of employment and universal groupemails. CW caseworkers were recruited through their collective bar-gaining unit's labor representative. Foster parents were recruitedthrough their affiliated provider agencies. Interested participantscalled a member of the study team, who conducted a brief phonescreening. If the participant was eligible and interested, the studyprocedures were explained and the consent script was read to the in-dividual. The participants were then contacted at a later time aboutthe specific date and time for the focus group. Focus groups consistedof only members of a particular group (i.e., provider agency case-workers or teachers); foster parent groups were stratified by kinshipcaregivers and non-kinship caregivers. We requested and weregranted a waiver of written consent, as this was a minimal riskstudy of adults. All participants were read a verbal consent scriptand confidentiality statement prior to the beginning of each focusgroup.

2.2. Data collection

First, each participant completed a brief demographic data sheet,including information about age, gender, and years of experience.The consent document was reviewed, and participants were giventhe opportunity to ask questions. Next, the participants were askedto agree to a statement of confidentiality. The focus groups were con-ducted by a trained moderator, using an interview guide that was de-veloped by the study team and tailored to each subgroup. The guideincluded questions about perceptions of the educational stability ofchildren in foster care, discussion of individual roles and responsibil-ities related to the education of children in foster care, children's ex-periences in school, cross-agency collaboration, system barriers, andpotential strategies for improvement. Participants received a gift cer-tificate at the end of the focus group, as well as funds for transporta-tion, if needed. All groups were audio-recorded and transcribed. Inaddition, a member of the study team attended each group and tookrunning notes to identify non-verbal cues and group dynamics thatwould not be clear on the audio-recording. The recordings were tran-scribed by a professional transcription service with knowledge ofconfidentiality issues. All personal information was de-identified inthe transcripts.

2.3. Data analysis

Demographic data were analyzed using Microsoft Excel to obtainranges and means. Transcripts were entered into NVivo 7 (QSR Inter-national, Melbourne, Australia) for organization and to facilitate cod-ing. A modified grounded theory approach was used to analyze thedata (Denzin & Lincoln, 1998). A coding tree was developed a prioribased on the interview guide; codes were then added as neededthrough the analysis process. The first three transcripts were codedby three study team members, and then reviewed for consistency incoding as well as the need for additional codes or better codingdefinitions. Those transcripts were re-coded until consensus wasreached. Each subsequent transcript was coded separately by twostudy team members, who then met to review discrepancies incoding. Any disagreements were resolved through consensus, andmemos of decisions regarding coding were kept in order to ensureconsistency in coding as analysis progressed. Once coding was com-plete, the three-member analysis team met to identify themes thatcrossed all constituent groups. Full study team meetings were alsoconducted to review emerging themes and guide further areas foranalysis.

405K. Noonan et al. / Children and Youth Services Review 34 (2012) 403–408

3. Results

Overall, the focus groups included 90 participants, with six totwelve members per group. The vast majority of participants (90%)were female, and average years of experience with the foster care oreducation systems ranged from seven years among foster parents to19 years among school counselors (Table 1).

The themes that emerged from the focus groups centered on thefollowing major areas: communication, policies and procedures, andissues around caring for children with behavioral problems. Partici-pants also identified several potential strategies for improving the ed-ucational outcomes of children in foster care. In general, there wasconsensus around the identified themes between the different typesof stakeholders; therefore, the themes are presented in aggregatewith illustrative quotes from participants. When there were discrep-ancies in opinion between the groups, we have described them ex-plicitly below.

3.1. Communication

One key issue identified in all of the focus groups was the percep-tion of ineffective and limited cross-system communication.

One school counselor stated, “My experience is of somewhat frus-tration in the lack of communication from the agencies involved…only in emergencies do I get that information, and what's sorelylacking is a phone call to be proactive.”

Similarly, an agency caseworker stated, “You go to the school toget a transfer or to see how the kids were doing in school…andthey're like, ‘Did you know she hasn't been to school in like xamount of days? Did you know that they're failing?’…So whyare you telling me now? What happened before?”

Within the theme of communication emerged discordant view-points between education and child welfare stakeholders about theneed for confidentiality. School counselors and teachers felt that in-formation relevant to a student's schooling was being withheld. Instark contrast, caseworkers and foster parents reported their percep-tion that schools were interested in knowing specific details aboutchildren in foster care that were not related to their educationalneeds. This concept was particularly striking around knowledge ofthe foster care status of children; counselors and teachers felt strong-ly that knowing the child's foster care status is needed for safety andappropriate service delivery, whereas caseworkers and foster parentsfelt that this knowledge could lead to unfair labeling of children ashaving behavioral or educational problems.

Finally, role confusion emerged as a difficult topic for all stake-holders, causing daily frustration for all stakeholder groups. Withinthis theme, participants reported uncertainly about their own role

Table 1Demographics of study participants.

Agencycaseworkers

CWcaseworkers

Fosterparents

Schoolcounselors

Teachers

Participants (N) 15 18 18 18 21Gender, Female 93% 78% 89% 94% 95%Mean years ofexperience

8 10 7 19 12

Age20–29 40% 11% 0% 6% 33%30–39 27% 44% 11% 6% 14%40–49 13% 11% 22% 22% 14%≥50 20% 33% 67% 61% 29%

responsibilities as well as the responsibilities of the school and childwelfare systems. Focus group participants commented frequently onrole assumptions that created discord and hindered communicationbetween systems and affected outcomes for children.

For example, an agency caseworker commented, “But you knowthey want us to do everything. Yet you call the school up andthey're like, ‘Well, you're not authorized to get this information.’”

Another agency caseworker stated, “The schools don't understandthat we need report cards…The state is gonna come and inspect usand if we don't have that in our files, then we're gonna get in bigtrouble.”

3.2. Policies and procedures

Two major concerns related to policies and procedures were iden-tified by all groups. The first was a general but pervasive variabilityaround the knowledge and implementation of policies within thechild welfare and education systems. There was not a single stake-holder group in which all participants demonstrated the same under-standing of procedures related to the child welfare or educationsystems. Second, there is limited knowledge among caseworkers,teachers, counselors, and foster parents about the specific relevantpolicies and procedures, particularly with respect to disclosure of in-formation and behavioral health-related services.

A CW caseworker commented, “Some of the rules and regulationskind of vary depending on the culture of the neighborhood.”

Similarly, according to an agency caseworker, “Putting a policyinto place is one thing — but having the people who actuallyneed to know the policy know it is a totally different situation.”

Specific to mental health services, a school counselor said, “So theremay be inconsistencies between the special ed regulations from thestate and the regulations to get…evaluations, to have the kids getpsychological evaluations for themental health school based service.”

In addition, city-employed and provider agency caseworkersreported that inefficiencies in the school district policies and proce-dures lead to unmet information needs.

An agency caseworker commented, “And they change their tune, likeI could be able to get attendance records and then suddenly I'm call-ing the high school and they're telling me, ‘Oh, we absolutely can'tgive that to you.’ I'm like, that's funny, a few months ago you did.”

Many participants also highlighted variable knowledge of relevantdistrict, state, and federal policies related to children in foster care.For example, at the most general level, there was differing knowledgeabout the policies related to the education of children in foster careand about which person or agency was responsible to implementthem. More specifically, there was confusion with policies regardingthe appropriate school transportation protocols to follow generallyand when a child changed placements.

An agency caseworker shared, “The same thing they said with the[transit passes]…first it was you had to be what, how many, 1.5miles away from the school…Then you got a four week waitingperiod. Then you had to sign up before a certain time…It's justeach week it was something different that they added on.”

A CWcaseworker commented, “They'll say, ‘well who's gonna trans-port the child?’We'll say, ‘Well, you have to.’Well who's gonna pay

406 K. Noonan et al. / Children and Youth Services Review 34 (2012) 403–408

for it? ‘Well, there will be special provisionmadewhere youwill getreimbursed’…They don't make any headway until they're guaran-teed payment or until they're guaranteed there's no liability is-sues…it's not in the best interest of the child but they don't care.”

3.3. Caring for children with behavioral problems

Participants discussed their perception about the association be-tween placement stability of students and behavioral problems; thisissue was particularly important to teachers and foster parents.

For example, a teacher commented, “My kid may not be able toknow her times tables because nobody at home is helping herwith them, and this is why…because I think it's easy as a teacherto kind of say, 'Oh well, she's not trying. She's acting out.' Well,why is she acting out?…because she has no stability in her life”.

Foster parents also noted a relationship between child visits totheir biological parents and an increase in behavioral problems.

One foster parent said, “They can go to a visit on Saturday and…ittakes themmaybe 'til almost Thursday or Friday to get themselvestogether.”

All groups discussed a relationship between knowledge of fostercare status and behavioral problems, either real or perceived, buildingon comments from participants related to communication challenges.

According to a CW caseworker, “Everything's a crisis when itcomes to a foster child, even though the other students may berunning around the room…so they are kind of always singled out.”

An agency caseworker noted, “I did have one of my kids that theteacher disclosed that the child was a foster child and that kind of…the child went downhill from there as far as his behavior in school.”

Finally, the discussion of children's behavioral health issues fo-cused on the availability of appropriate support services. Participantswere concerned with the inability to initiate and maintain servicesthroughout home placement changes, especially related to the diffi-culty in obtaining needed signatures to access services.

A school counselor stated, “The school can meet some of the needsbut cannot meet all of the needs.”

According to a CW caseworker, “It always becomes a problem ofwho is responsible when these kids may need mental health ser-vices or the schools need to be changed. And there's a disconnectbetween who is legally responsible and who is responsible.”

3.4. Identified strategies

Focus group participants discussed strategies they currently use,or would like to use, related to the education of children in fostercare. Current strategies included the use and value of personal rela-tionships, strategic communications up the chain of select agenciesand organizations, and the threat of, or actual use of, court orders aseffective ways to achieve results. For example,

A CW caseworker stated, “I have to build relationships with theseteachers as well as these agencies in order to basically get whatev-er I want in a sense.”

An agency caseworker said, “I find…like going up the ranks hasmost of the time worked for me with [the city child welfare

department] with the school board, and also trying to accessresources.”

Additionally, participants discussed other strategies that could beimplemented to improve results. The most frequently suggestedstrategy was training; almost every group mentioned a need formore training related to cross-system roles and policies. Similarly, arequest for more interagency meetings was cited routinely acrossgroups. On a larger scale, two systems reform strategies were dis-cussed. First, participants suggested that child welfare informationabout students be housed within school system records; including in-formation on foster care status, contact information of caseworkersand foster parents, child guardianship and visitation information,and mental health and community support services received. Partici-pants also expressed a desire for stronger staffing support at theschools.

For example, a teacher commented, “…there's a need for a socialworker to be on site at the school as opposed to just the counsel-or.”

Similarly, a CW caseworker shared, “It'd be a good idea to havelike a liaison sort of between the schools and [the city child wel-fare department] because both of our jobs are centered aroundchildren.”

4. Discussion

Focus groups with caseworkers, teachers, school counselors, andfoster parents highlighted the cross-system challenges related to theeducation of children in foster care. Participants agreed that maneu-vering between the child welfare and education systems to achieveeducational success for children in foster care is rarely easy. Partici-pants identified many challenges that have been previously identifiedas critical to the successful implementation of new policy. Interesting-ly, while we never mentioned the passage of Fostering Connectionsand its cross-system requirements, participant responses suggestthat field-level stakeholders could benefit from policy guidance inthis area.

Our findings suggest that communication, particularly cross-system communication, was laborious and a source of recurrent frus-tration for study participants. Differing concerns and views about pri-vacy appeared to impede communication among participants; thiswas especially true with respect to sharing information about fostercare status, a finding that mirrors previous research (Altshuler,2003). There are several reasons why stakeholders might have diver-gent views on this, including differences related to training, agencyrules and practice norms.

The confidentiality of information about children in foster caremay also be an issue in which different federal and state laws compelconflicting activities, or at minimum, create a confusing implementa-tion environment for field-level stakeholders. For example, thoughthe federal Family Educational Rights and Privacy Act (FERPA) doesnot prohibit child welfare caseworkers from accessing educational in-formation about children in foster care (Family Educational Rights etal., 1974), their ability to do so varies from state to state and evenwithin states, as they are not a “parent” as defined Family EducationalRights and Privacy Act (FERPA) under the Act so do not have automaticaccess to educational records (McNaught, 2005). In addition, moststates have their own confidentiality laws regarding children in fostercare and the type of information and persons to whom informationcan be shared, including what can be shared by the child welfare sys-tem with education stakeholders (McNaught, 2005).

None of these laws intend to thwart what is best for children. Yetthe result may nonetheless be different sets of rules advancing

407K. Noonan et al. / Children and Youth Services Review 34 (2012) 403–408

inconsistent goals. While the aims of protecting the privacy of chil-dren and advancing their educational needs have merit, realistically,it is difficult to balance both. Fostering Connections does not appearto clarify these tensions since it did not modify any of the existingfederal or state confidentiality rules already in place. Fostering Con-nections does, however, expand cross-system collaboration expecta-tions with regard to children in foster care. The burden of thisrequirement will likely fall on frontline stakeholders, even thoughthey may not be equipped or have the requisite authority to addressthem.

Our findings related to communication difficulties are similar tothose of Smithgall and colleagues, who conducted in-depth inter-views with participants comparable to those in our study and foundthat communication between groups was often lacking, and when itdid occur, it was frequently ineffective and complex (Smithgall etal., 2004). A potential solution that policymakers might pursue isthe development of cross-system guidance or rulemaking in whichthe education and child welfare system jointly define how childwelfare and education information will be shared. Legislationenacted in California in 2004 – pre-dating the federal FosteringConnections Act – offers a good foundation regarding information-sharing between education and child welfare agencies ("CaliforniaA.B. 1858, 2004").

While communication was raised primarily as a source of frustra-tion to study participants, the use of communication as a cross-system relationship-building tool was also cited frequently as a suc-cessful strategy. Study participants reported that when they wentout of their way to develop lines of cross-system communication,the outcome they sought (e.g., school admission or school records)would be accomplished. A similar focus group study of students, edu-cators, and caseworkers found that forming collaborative relation-ships among constituents resulted in perceived success for thestudents (Altshuler, 2003), a finding echoed by our participants.This should encourage child welfare administrators who, pursuantto Fostering Connections, are likely put in place additional require-ments about cross-system communication. Our findings imply thatthe intention of a participant to have a positive working relationshiprelated to a child in foster care's education is a good strategy toachieving one.

Unreliable knowledge and inconsistent application of policies andprocedures within the child welfare and education systems was also ashared concern among study participants. Our results indicate verylittle alignment within or across stakeholders about policies relatedto the education of children in foster care. Inconsistent awarenessand application of policies adopted by agency leadership may con-found implementation among those working at the ground-level.This phenomenon was illustrated – as noted above in Sections 3.1and 3.2 – by the differing interpretation and knowledge of policies re-lated to educating children in foster care by study participants. Partic-ipants' differing knowledge about policies related to the education ofchildren in foster care suggests much training is needed for thoseworking closely with children in foster care.

Meeting the cross-system requirements of Fostering Connectionswill require child welfare and education systems to use collaborativemodels to overcome many of the barriers that were identified in ourstudy. The federal government recently reaffirmed its commitmentto these cross-system requirements of Fostering Connections in ajoint statement issued by the Department of Education and the De-partment of Health and Human Services in August 2011. This jointstatement to all Chief State School Officers and State ChildWelfare Di-rectors encouraged “States and [local education agencies] to developor review and, if appropriate, revise their policies and guidelines forserving children in foster care”(Yudin & Samuels, 2011). Many stateshave already enacted new legislation in response to the educationalrequirements of Fostering Connections, explicitly adopting FosteringConnections educational stability goals within state educational code

(American Bar Association Legal Center for Foster Care & Education,2011). Other jurisdictions, such as Philadelphia county, have begunputting in place “cross-system” staff models. Philadelphia recentlydeveloped an Educational Support Center housed within its child wel-fare agency to provide support to both the education and child wel-fare systems for children in foster care (City of Philadelphia, 2010).

While recent collaborative models such as that implemented inPhiladelphia show promise, their impact has yet to be determined.Moreover, some commentators report that collaborative models willbe hard to put in place without a parallel cross-system requirementunder federal education statutes (American Bar Association LegalCenter for Foster Care & Education, 2010). The recent statementfrom the federal education and child welfare agencies emphasizingjoint efforts at the state and local levels may provide additional andnecessary impetus in this regard.

Interestingly, although this study did not specifically set out to ex-plore behavioral issues in education, participants in each focus groupraised concerns around behavioral health spontaneously withoutprompting. Given the prevalence of behavioral health problems forchildren in foster care, reported at greater than 50% (dosReis et al.,2001; Halfon et al., 1995; Jee et al., 2006), and the accompanyingproblem of mental health care access and provision for this popula-tion (Dubowitz et al., 2000), this is not surprising.

Participants' emphasis on behavioral health issues also illustratesthe larger challenge for child welfare agencies that need to coordinatetheir service delivery for children not only across the educational sys-tem, but also with other systems (e.g. behavioral health or juvenilejustice). Based on participants' reports and, more broadly, an under-standing of the development of these systems, they frequently oper-ate independently of one another and thus have limited history ofcollaborating to address the problems of any given child. This historylikely explains many of the lapses in communication and lack of clar-ity in roles that were described in this study.

This study has several limitations. First, as with all qualitative re-search, there may have been selection bias, as those with particularopinions may have been more interested in participating than thosewith other opinions. However, by triangulating our data throughcomparison to that of others', as well as comparison to policy changesalready in place, we believe there is validity to our findings. Anotherpotential limitation is variation in discussion topics among groups;we sought to minimize this possibility through training of moderatorsand the use of a structured interview guide. Finally, these results wereobtained in a single urban child welfare system. However, the types ofbarriers that were observed are likely to be generalizable across sys-tems, even if the nature of the interventions may differ dependenton local organizational and community factors.

5. Conclusions

This study of foster care caseworkers, teachers, school counselors,and foster parents gauged their current understanding of educationalprocesses for children in foster care including opportunities for im-proving outcomes and existing barriers that constrain their daily ef-forts in this regard. Although intended as a study to understandbarriers between two systems, it was readily apparent that the chal-lenges extend to the behavioral health system as well. Our resultsshould help guide child welfare and educational systems as they cre-ate and implement policies for cross-system collaboration pursuantto Fostering Connections, with the ultimate goal of improving educa-tional outcomes for children in foster care.

Acknowledgments

This project was supported by a grant from the William PennFoundation and a fellowship to Dr. Rubin from the StoneleighFoundation. The Mixed Methods Research Lab at the University of

408 K. Noonan et al. / Children and Youth Services Review 34 (2012) 403–408

Pennsylvania Perelman School of Medicine provided technical assis-tance to the project; we are grateful to Eve Weiss in particular forher help with focus group moderation. We thank Amanda O'Reilly,the Philadelphia Department of Human Services, the School Districtof Philadelphia, and the foster care agencies and providers for theircontributions to this study. Cara Curtis provided editorial supportduring the preparation of this manuscript.

References

110th United States Congress (2008). Fostering connections to success and increasingadoptions act of 2008 P.L 110–351, Section 204. Washington, DC.

Altshuler, S. (2003). From barriers to successful collaboration: Public schools and childwelfare working together. Social Work, 48(1), 52–63.

American Bar Association Legal Center for Foster Care and Education (2009). Fosteringconnections act—Implementation checklists for education provisions. RetrievedSeptember 27, 2011, from http://www.americanbar.org/content/dam/aba/migrated/child/education/publications/fc_implementation_checklists_final.authcheckdam.pdf

American Bar Association Legal Center for Foster Care and Education (2010). State andfederal implementation of the education provisions of the fostering connectionsact: Progress and challenges. Retrieved September 30, 2011, from http://www.americanbar.org/groups/child_law/projects_initiatives/education/federal_updates.html

American Bar Association Legal Center for Foster Care and Education (2011). State leg-islation chart: providing school stability and continuity for children in foster care.Retrieved September 27, 2011, from http://www.americanbar.org/groups/child_law/projects_initiatives/education/fosteringconnections.html

California A.B. 1858 (2004 September 29).Casey Family Programs National Working Group on Foster Care and Education (2008).

Educational outcomes for children and youth in foster and out-of-home care:Casey Family Programs.

City of Philadelphia (2010, March 30). City and philanthropic partners work to reducetruancy and improve educational outcomes. Retrieved September 27, 2011, fromhttp://cityofphiladelphia.wordpress.com/2010/03/30/city-and-philanthropic-partners-work-to-reduce-truancy-and-improve-educational-outcomes/

Courtney, M., Roderick, M., Smithgall, C., Gladden, R., & Nagaoka, J. (2004). The educa-tional status of foster children, Vol. 102, Chicago: Chapin Hall Center for Children.

Denzin, N. K., & Lincoln, Y. S. (1998). Grounded theory methodology. In A. Strauss, &J. Corbin (Eds.), Strategies of qualitative inquiry. Walnut Creek: Sage.

dosReis, S., Zito, J.M., Safer, D. J., & Soeken, K. L. (2001).Mental health services for youths infoster care and disabled youths. American Journal of Public Health, 91(7), 1094–1099.

Dubowitz, H., Simms, M. D., & Szilagyi, M. A. (2000/10/). Health care needs of childrenin the foster care system. Pediatrics, 106, 909.

Family Educational Rights and Privacy Act, Pub. L. No. 93–380 § 513 1974.Halfon, N., Mendonca, A., & Berkowitz, G. (1995). Health status of children in foster

care. Archives of Pediatrics & Adolescent Medicine, 149, 386–392.Honig, M. I. (2006). Street-level bureaucracy revisited: Frontline district central-office

administrators as boundary spanners in education policy implementation. Educa-tional Evaluation and Policy Analysis, 28(4), 357–383.

Jee, S. H., Barth, R. P., Szilagyi, M. A., Szilagyi, P. G., Aida, M., & Davis, M. M. (2006). Fac-tors associated with chronic conditions among children in foster care. Journal ofHealth Care for the Poor and Underserved, 17(2), 328–341.

Lipsky, M. (1980). Street level bureaucracy: dilemmas of the individual in public services.New York: Russel Sage Foundation.

McNaught, K. (2005). Mythbusting: breaking down confidentiality and decision-makingbarriers to meet the education needs of children in foster care: American Bar Associa-tion. : Center on Children and the Law.

Pecora, P., Kessler, R., Williams, J., O'Brien, K., Downs, C., English, D., et al. (2005).Improving family foster care: Findings from the Northwest Foster Care AlumniStudy. Seattle: Casey Family Programs.

Rubin, D. M., Alessandrini, E. A., Feudtner, C., Localio, A. R., & Hadley, T. (2004).Placement changes and emergency department visits in the first year of fostercare. Pediatrics, 114(3), e354–e360, doi:10.1542/peds.2003-0594-F.

Smithgall, C., Gladden, R., Howard, E., Goerge, R., & Courtney, M. (2004). Educationalexperiences of children in out-of-home care. Chicago: Chapin Hall Center forChildren.

Spillane, J. P. (2000). Cognition and policy implementation: District policymakers andthe reform of mathematics education. Cognition and Instruction, 18(2), 141–179,doi:10.1207/s1532690xci1802_01.

Trout, A., Hagaman, J., Casey, K., Reid, R., & Epstein, M. (2008). The academic status ofchildren and youth in out-of-home care: A review of the literature. Children andYouth Services Review, 30, 979–994.

Yudin, M., & Samuels, B. (2011). Joint letter to chief state school officers and statechild welfare directors (pp. 2). : U.S. Department of Health and Human Servicesand U.S. Department of Education.