Abstract Using mixed methods, this case study explored outcomes associated with the adoption and implementation of a community schools approach in four Title I schools using the Community Collaboration Model for School Improvement. Trends in school data demonstrate academic achievement improvements in three of the four schools. Absenteeism and the number of office discipline referrals dropped from pre- to two years postimplementation. Teacher and staff perceptions of school climate and the learning support system were more favorable two years postimplementation. Interview and focus group data involving 10 stakeholders highlight facilitators and barriers, and demonstrate contextual factors supporting or hindering implementation efforts. Findings showcase the promise of community schools and other partnership approaches for ensuring healthy youth development and learning, especially for students in high-impact schools. Each year in the United States, nearly 400,000 students drop out of public high schools (Stark & Noel, 2015). Among high school graduates, approximately one-third complete no postsecondary education or training (Ruppert, 2003). Of those who do, many are unprepared. Forty percent of four-year and 63 percent of two-year college students require some type of remediation (Callan, Finney, Kirst, Usdan, & Venezia, 2005). Contributing to these alarming trends is a host of complex, interwoven issues both within and external to schools. Young people, especially those living in poverty, often bring with them to school multiple barriers impeding their learning and development. To address these needs, schools need new strategies to ensure that all students fluidly matriculate and successfully transition to adulthood. In light of these trends, the American Academy of Social Work and Social Welfare Grand Challenges for Social Work prioritize the healthy development of all youths through prevention (Hawkins et al., 2015). Preventing school dropout is a central priority within this grand challenge, along with other emphases such as increasing access to behavioral health care and promoting collaboration (Hawkins et al., 2015). Schools can and should lead efforts toward the vision laid out by the grand challenges for several reasons. First, youths spend the majority of their time in schools, and schools provide unmatched access to health and social services (Anderson-Butcher, Mellin, Iachini, & Ball, 2014; Flaherty, Weist, & Warner, 1997). In addition, students form important relationships with teachers, coaches, intervention assistants, and other school staff, and these adults serve as caring role models and key referral agents (Anderson-Butcher et al., 2014). Last, school administrators, school social workers, and other stakeholders often facilitate partnerships with outside agencies to promote student outcomes. Schools can maximize these linkages to further connect students and families to community resources and supports. In short, school–family–community partnership models that respond to the complex, diverse needs of students and families have the potential to help ensure healthy development for all youths and address dropout (Anderson-Butcher, 2004; Blank, Melaville, & Shah, 2003; Dryfoos, Quinn, & Barkin, 2005). To advance research in this area, this study examined outcomes associated with the implementation of community schools in four elementary schools over a two-year period. School–Family–Community Partnerships and Community Schools A variety of partnership models exist to support schools in expanding improvement priorities beyond just academic learning, such as the coordinated school health model (Hurwitz & Weston, 2010), the “whole school, whole community, whole child” approach (Lewallen, Hunt, Potts-Datema, Zaza, & Giles, 2015), and comprehensive systems of learning supports frameworks (Adelman & Taylor, 2002). Schools using these models develop collaborative partnerships with community-based organizations, parents and families, the private sector, government, and others. By doing so, school communities secure resources to enhance services during the school day and extend opportunities for learning and development through out-of-school time programming (Anderson-Butcher et al., 2008). Outcomes from partnership approaches include improved academic outcomes, enhanced student perceptions of school climate, and improved system-level capacities such as improved linkage and referral processes (Anderson-Butcher, Iachini, Ball, Barker, & Martin, 2016; Anderson-Butcher, Lawson, Iachini, Bean, et al., 2010; Anderson-Butcher, Lawson, Iachini, Flaspohler, et al., 2010; Lawson, 2013; Leone & Bartolotta, 2010). Community schools, one specific partnership model, originated from the pioneering work of Jane Addams and John Dewey in Chicago during the early 20th century. Community schools integrate the settlement house idea with expanded visions of the neighborhood school, exploring how schools can be hubs of child development, family support, and neighborhood cohesion. The best models combine school-based and -linked services, and strive to maximize family, school, and community resources. Evidence is growing in relation to the role of community schools in promoting positive school and student outcomes (Blank et al., 2003; Moore & Emig, 2014). Most of these studies, however, are descriptive in nature, and few explore impacts over time. In addition, little research examines behavioral outcomes, such as attendance and discipline, or the value of these approaches for varying stakeholders. This mixed-methods case study explores outcomes associated with the implementation of a community schools framework using the Community Collaboration Model for School Improvement (CCMSI) process (Anderson-Butcher et al., 2008). This study explores school-level outcomes, including academic and behavioral indicators, from baseline to two years postimplementation. Changes in teacher and staff perceptions of factors central to learning and climate are explored. Last, stakeholder- perceived facilitators and barriers affecting implementation efforts are distilled. Findings can inform school social work practice as the field addresses grand challenges focused on promoting healthy youth development and preventing school dropout and other problem behaviors. Method Four Title I elementary schools in a large, urban, Intermountain West District provided the context for this study. During the 2011–2012 academic year, all four schools were in No Child Left Behind (NCLB) school improvement status, with two identified as focus schools (that is, a designation given to the lowest 15 percent of schools in the state). About 50 percent of the students in these schools were racial and ethnic minorities, and over 80 percent resided in poverty (see Table 1). Initial CCMSI efforts began in spring 2012, and new programs and services were implemented starting in fall of the 2012–2013 academic year. Table 1: School Demographic Data in 2013–2014 School 1 School 2 School 3 School 4 Enrollment 606 596 762 599 Grades served K–5 K–5 K–5 K–5 Race and ethnicity (%) White 51.5 59.5 46.5 63.4 Minority 48.5 40.5 53.5 36.6 Hispanic 35.5 25.2 36.0 22.7 American Indian 7.9 4.9 11.0 7.9 Asian 1.4 3.0 0.8 1.9 African American 2.1 3.8 3.6 2.5 Pacific Islander 1.6 2.7 2.0 1.5 % living in poverty 88 76 92 71 % special education 7 8 4 7 % English language learners 40 22 42 17 School 1 School 2 School 3 School 4 Enrollment 606 596 762 599 Grades served K–5 K–5 K–5 K–5 Race and ethnicity (%) White 51.5 59.5 46.5 63.4 Minority 48.5 40.5 53.5 36.6 Hispanic 35.5 25.2 36.0 22.7 American Indian 7.9 4.9 11.0 7.9 Asian 1.4 3.0 0.8 1.9 African American 2.1 3.8 3.6 2.5 Pacific Islander 1.6 2.7 2.0 1.5 % living in poverty 88 76 92 71 % special education 7 8 4 7 % English language learners 40 22 42 17 Sources: Canyons School District. (2014). Academic and behavioral data records. Sandy, UT: Author. Utah State Office of Education. (2014). School report card data. Salt Lake City: Author. At the time, the four schools of interest here were involved in districtwide improvement efforts aligned to NCLB accountabilities. More specifically, the four schools and overall district were adopting curriculum-based measures in reading and math, reading comprehension assessments, and common formative assessments to monitor learning and progress. In addition, positive behavioral interventions and supports (PBIS) were being implemented districtwide, and schools were focused on creating common expectations and norms, implementing behavioral interventions, and rewarding students for good behavior. Professional development experiences for teachers and staff also were underway, especially ones to improve the quality of instruction in classrooms. To broaden and deepen these traditional school improvement efforts, as well as address other conditions affecting learning, each school followed the CCMSI model, as depicted in Figure 1 (Anderson-Butcher, Lawson, Iachini, Bean, et al., 2010). Figure 1: View largeDownload slide Ohio Community Collaboration Model for School Improvement Source: Anderson-Butcher, D., Lawson, H. A., Bean, J., Flaspohler, P., Boone, B., & Kwiatkowski, A. (2008). Community collaboration to improve schools: Introducing a new model from Ohio. Children & Schools, 30, 161–172. Figure 1: View largeDownload slide Ohio Community Collaboration Model for School Improvement Source: Anderson-Butcher, D., Lawson, H. A., Bean, J., Flaspohler, P., Boone, B., & Kwiatkowski, A. (2008). Community collaboration to improve schools: Introducing a new model from Ohio. Children & Schools, 30, 161–172. Efforts followed CCMSI implementation milestones. Specifically, one of the first steps involved expanding the school improvement team membership and adding parents and caregivers, local after-school program providers, nonprofit leaders, and other stakeholders to the “planning table.” This expanded school improvement team led the school community in a comprehensive needs assessment, resource mapping, and gap analysis process. During the initial stages, various stakeholders were surveyed to assess their perceptions of the school climate, school engagement, student nonacademic barriers to learning, parent and community involvement, and other school experience factors. Several needs were identified. For instance, students were experiencing multiple stressors. Specifically, survey data found that 47.6 percent of students reported having trouble sleeping in the week prior to the survey administration, and 32.3 percent felt worried. One in four youths reported no involvement in out-of-school-time positive youth development experiences. Survey data also suggested that parents and caregivers were dealing with stressors such as unemployment, underemployment, and challenges with meeting basic needs. Likewise, only 43.4 percent of parents and caregivers reported that the schools “help families get the services we need in the community.” Other challenges were noted, such as significant internalizing symptomology among students. Next, school leaders, in partnership with parents and caregivers, community partners, and other stakeholders, began mapping school- and community-based resources across the five CCMSI pathways. Gaps in programs and services were noted, especially as top priority needs were identified based on the academic, behavioral, and stakeholder survey data. In the end, expanded school improvement plans resulted and included academic priorities (for example, targeting after-school tutoring), as well as ones focused on youth development and school climate (for example, opening Boys & Girls Clubs on site), parent/family engagement and support (for example, housing Family Literacy Centers), health and social services (for example, adding social work interns), and community partnerships (for example, partnering with the local United Way and faith-based organizations). In response to the needs assessment, resource mapping, and gap analyses, new or expanded evidence-based strategies were implemented according to the PBIS framework, an approach also embedded in the grand challenge focus area on enhancing youth development (Hawkins et al., 2015). Tier 1 universal strategies for all students focus on improving schoolwide climate, managing classroom behaviors, and providing quality instruction (for example, PlayWorks! recess interventions). Tier 2 early identification and intervention strategies target at-risk and gifted and talented students (for example, social skills groups, after-school tutoring). Tier 3 indicated interventions focused on students and families with highly individualized needs (for example, school-based mental health services). Priority was given to the development and enhancement of services both during the school day and during out-of-school time. In addition to these intervention priorities, system-level innovations were created as the building leadership teams focused on improving infrastructure, another key step in the CCMSI adoption process. New community school coordinators (individuals prepared as school social workers) led partnership development and facilitated overall efforts. School social work interns were added to provide individual and group skill building and case management. School psychologists began doing more indirect practice as they consulted with teachers. Infrastructure was improved, through key activities such as the enhancement of the student learning support system. More specifically, teachers were provided professional development in relation to the early identification of academic challenges and other barriers to learning. Referral procedures were created and included the assignment of a specific person in the building to be the single-point-of-contact person for all referrals. In addition, service coordination wraparound teams, called Care Teams, were created at each school to case individual students in need of additional learning supports to address academic, social, emotional, physical health, and other basic needs. Data on students identified for services were monitored regularly to assess progress through a response-to-intervention process. Teachers and staff also received professional development and consultation to infuse PBIS strategies in their classrooms, support students having experienced trauma, and use data to provide differentiated instruction. Procedures and Measures Mixed methods included records reviews, survey data collection with teachers and staff, and interviews and focus groups with key stakeholders. Procedures were approved by the institutional review boards at a large midwestern university and the school district. Records Review School and program records from 2012 to 2014, along with associated secondary data, were collected from the district and partnering agencies. Each school’s state report card was reviewed. Two school performance metrics were of interest: achievement and growth points. The state’s standardized testing procedures changed during the course of this study, so scores could not be compared across time. Instead, the overall letter grades served as the main indicator of academic performance. Attendance and office discipline referrals (ODRs) data were examined in aggregate counts for each school year. Because the schools stopped serving sixth grade in 2013–2014, analyses included only kindergarten through fifth grade. Please note that variation in enrollment was not substantial enough to influence overall trends, so unstandardized absenteeism and ODR counts are reported. Last, participation records were collected to describe rates of service utilization. Teacher Perception Surveys District administrators collected teacher and staff perceptual data in January 2012 and January 2014 using the Community and Youth Collaborative Institute’s School Experience Survey (CAYCI SES) (Anderson-Butcher, Amorose, Iachini, & Ball, 2013), a psychometrically sound tool designed to assess how different audiences experience their school and community. Eighty-seven percent of school employees (n = 212) across the four schools completed the survey in 2012, whereas 85 percent completed them in 2014 (n = 286). Approximately two-thirds of the respondents in each sample were classroom teachers (67 percent in 2012 and 70 percent in 2014). CAYCI SES data were collected electronically, without individual identifiers in both 2012 to 2014. Table 2 provides a summary of the scales and example items. In the current sample, Cronbach’s alphas were acceptable (α = .73 to .94) for all scales during both measurement years, except for the school climate scale in 2012 (α = .68 in 2012). Scale scores were calculated for those with complete data. Select individual items on the CAYCI SES also were examined by calculating the percentage selecting either “agree” or “strongly agree.” Respondents also could select “don’t know,” which was treated as missing data. Table 2: Teacher Perception Data in 2012 and 2014 Scale Sample Item Range 2012a 2014b M (SD) M (SD) Academic Motivation Students like the challenge of learning new things in school. 1–5 3.79 (0.82) 3.85 (0.87) Basic Needs My students’ families or caregivers have their basic needs met (that is, food, shelter, clothing). 1–5 3.11 (0.92) 3.21 (1.02) Externalizing Behaviorsc My students demonstrate hyperactivity or restlessness. 1–5 3.55 (0.78) 3.66 (0.79) Internalizing Behaviorsc My students are sad or depressed. 1–5 4.17 (0.49) 4.24 (0.52) Perceived Learning Supports There is a system in place at my school where teachers and staff can refer students and families who are in need of additional learning. 1–5 3.90 (0.83) 4.13 (0.68)d Psychological Well-Being My students have positive self-esteem. 1–5 3.63 (0.87) 3.74 (0.89) Safety My students feel safe at school. 1–5 3.90 (0.84) 4.02 (0.81) School Climate My school has strong relationships among teachers and students. 1–4 3.11 (0.44) 3.25 (0.45)d Social Skills My students are good at making friends. 1–5 3.67 (0.78) 3.78 (0.76) Scale Sample Item Range 2012a 2014b M (SD) M (SD) Academic Motivation Students like the challenge of learning new things in school. 1–5 3.79 (0.82) 3.85 (0.87) Basic Needs My students’ families or caregivers have their basic needs met (that is, food, shelter, clothing). 1–5 3.11 (0.92) 3.21 (1.02) Externalizing Behaviorsc My students demonstrate hyperactivity or restlessness. 1–5 3.55 (0.78) 3.66 (0.79) Internalizing Behaviorsc My students are sad or depressed. 1–5 4.17 (0.49) 4.24 (0.52) Perceived Learning Supports There is a system in place at my school where teachers and staff can refer students and families who are in need of additional learning. 1–5 3.90 (0.83) 4.13 (0.68)d Psychological Well-Being My students have positive self-esteem. 1–5 3.63 (0.87) 3.74 (0.89) Safety My students feel safe at school. 1–5 3.90 (0.84) 4.02 (0.81) School Climate My school has strong relationships among teachers and students. 1–4 3.11 (0.44) 3.25 (0.45)d Social Skills My students are good at making friends. 1–5 3.67 (0.78) 3.78 (0.76) aMean scores calculated for respondents completing all scale items; sample sizes for 2012 ranged from 169 (Basic Needs) to 210 (School Climate). bSample sizes for 2014 ranged from 225 (Basic Needs) to 285 (School Climate). CAll negatively worded items were reverse coded for purpose of mean score calculations. dDifference between 2012 and 2014 was statistically significant (p = .001). Qualitative Interviews and Focus Groups Key informant interviews were conducted in person with the four school principals. A focus group was conducted with six stakeholders identified by district leadership as critical informants, including a school-based mental health provider, two district administrators, an after-school program coordinator, a university partner, and a community school coordinator. All participants provided consent for participation. Interviews and the focus group took place in meeting rooms in the schools approximately 18 months into the community school implementation. A semistructured format was used. Example interview and focus group questions included the following: “What barriers have you encountered?” and “How has implementation of the community schools affected your school or organization as a whole?” The interviewer used probing questions to elicit further detail. Each interview or focus group lasted approximately 60 minutes and was transcribed. Transcripts were reviewed for overall themes using inductive techniques (Patton, 1990). Throughout the coding process, a peer reviewer was consulted, to discuss emergent themes and enhance validity. As recommended by Barker and Pistrang (2005), two key leaders also validated emergent themes through a member checking process. Results After implementing initial CCMSI steps, the four schools prioritized new or enhanced programs and strategies across the five CCMSI pathways in 2012–2013. Participation numbers served as initial indicators of impact. During each implementation year, an additional 260 youths received after-school tutoring and another 100 youths were served in prekindergarten classrooms. New programs were added and served large numbers of students. For instance, an average of 818 youths participated in the after-school program, 150 students received Care Team wraparound supports, and 175 youths received school-based mental health services. School-Level Outcomes Associated with Adoption and Implementation A number of data sources were examined to assess school-level outcomes, including secondary data collected from various state and local sources, as well as teacher and staff perceptual survey data. Secondary Academic and Behavioral Data Over the course of the two years, indicators on statewide academic report cards improved at three of the four schools. Trends at each school building in various report card outcomes are showcased in Table 3. Please note that the school showing no report card improvements (school 4) was the highest achieving school at baseline, as well as the school serving the least diverse student population. Figure 2 showcases trends in ODRs. The number of ODRs declined steadily over time at school 4. At the other schools, ODRs rose initially, which stakeholders attributed to the adoption of a formal system to track ODRs for the first time in 2012–2013. By the 2013–2014 school year, the number of ODRs across the four schools had fallen 22.5 percent from baseline. School 2 showed the most improvement in reductions in ODRs. As also demonstrated in Figure 2, absenteeism in all four schools was reduced. In fact, an overall 37 percent drop in absenteeism occurred between the 2011–2012 and 2013–2014 school years. The buildings with the greatest reductions in absenteeism were schools 3 and 4. Table 3: Three Years of Report Card Data across Four Schools School 1 School 2 School 3 School 4 School year Year 1 Year 2a Year 3 Year 1 Year 2a Year 3 Year 1 Year 2a Year 3 Year 1 Year 2a Year 3 Achievement/ proficiencyb 137 42 68 207 118 126 152 51 54 211 113 127 Growthb 167 172 183 186 195 197 132 155 160 206 220 187 Total scoreb 304 214 251 393 313 323 284 206 214 417 333 314 Overall grade D D C C B B F D D B B B School 1 School 2 School 3 School 4 School year Year 1 Year 2a Year 3 Year 1 Year 2a Year 3 Year 1 Year 2a Year 3 Year 1 Year 2a Year 3 Achievement/ proficiencyb 137 42 68 207 118 126 152 51 54 211 113 127 Growthb 167 172 183 186 195 197 132 155 160 206 220 187 Total scoreb 304 214 251 393 313 323 284 206 214 417 333 314 Overall grade D D C C B B F D D B B B Source: Utah State Office of Education. (2014). School report card data. Salt Lake City: Author. aIn the 2013–2014 school year, Utah adopted a new standardized test that contributed to drops in achievement scores not only in these four schools, but statewide. bThe maximum achievement/proficiency score and growth score are both 300, for a maximum total score of 600. Figure 2: View largeDownload slide Behavioral Indicators across Three School Years Figure 2: View largeDownload slide Behavioral Indicators across Three School Years Teacher and Staff Perceptions Teacher and staff perceptions on all scales were more favorable in 2014 than 2012 (see Table 2). In particular, the differences in perceived learning supports and school climate were statistically significant (p = .001). Examination of individual scale items revealed specific areas of growth. For instance, 76.3 percent of teachers and staff at baseline, compared with 93.9 percent at two years postadoption, reported there was “a system in place where teachers/staff can refer students and families who are in need of additional learning supports.” Likewise, 77.0 percent of teachers and staff at baseline versus 86.1 percent at post reported, “Students in need of extra learning supports in my school are able to get them”; 79.2 percent at baseline compared with 88.9 percent in 2014 reported, “Teachers and staff work closely with school counselors, social workers, and other support staff in my school.” Factors Affecting Adoption and Implementation Facilitators Stakeholders mentioned the importance of infrastructure, including the value of a strong, clearly defined organizational structure. They highlighted the need for regular partnership meetings to review progress, examine challenges, explore emergent needs, and brainstorm solutions. School-level teaming structures also were identified as essential, especially the Care Team wraparound structures that allowed for team problem solving related to individual student needs. Key programs and partners were noted, particularly ones involving academic interventions, school-based mental health services, and parent–family educational programs. Several stakeholders also mentioned the value of data, including the importance of the CAYCI SES data for focusing on priorities during the needs assessment, resource mapping, and gap analysis process. In addition, participants mentioned how data were important for fostering a sense of shared ownership among partners. Key institutional leaders and staff at both the district and schools were viewed as important. In fact, stakeholders pointed out that the schools with the more experienced principals were also the ones with the strongest organizational structure. The role of the community school coordinators was mentioned by several participants, as they said that individuals in these roles “stayed positive,” were “persistent,” and were extremely “patient.” Flexibility also was important to the community school design. As one stakeholder stated, there was “flexibility in the ways we can do the community school; the way we can schedule programs and design things to meet the needs of the students in our schools; as well as to fit what resources and opportunities we are dealing with into the programs.” Professional development and consultation were mentioned as important for building the knowledge and skills of individuals as they explored “new ways of doing things.” Professional development opportunities related to PBIS implementation and trauma-informed classrooms were identified as especially important, and schools where teachers implemented these evidence-based strategies with fidelity tended to have more significant reductions in ODRs. Last, stakeholders identified policy facilitators. Federal accountabilities and state-level testing frameworks showcased how the schools were underperforming, putting pressure on the district to adopt the community schools model. In addition, existing state-level initiatives and priorities (such as PBIS) made it easier to promote the adoption of new innovations. Facilitators and Barriers Several themes emerged as both facilitators and barriers. Challenges were noted in relation to garnering buy-in for the community schools work, especially among classroom teachers and principals. Efforts moved more smoothly once stakeholders “got on board,” and stakeholders reported on how the CCMSI milestones and logic model (that is, Figure 1) helped build knowledge and gain commitments. In addition, buy-in seemed to be an important facilitator in relation to strengthening the learning support system. In fact, CAYCI SES data significantly improved over the course of implementation, most likely due to the buy-in teachers and staff had in relation to the need for improving the identification and referral system. Pacing (or lack thereof) also was noted as both a facilitator and barrier. Stakeholders perceived that “we grew so fast” and there was not always time to “reflect” and “process” what was happening in “real time.” Stakeholders, especially ones representing the more high-impact schools (that is, schools 1 and 3), described how adding new programs and staff simultaneously and quickly (for instance, during the first year of adoption) caused additional stressors. Yet the fast growth and addition of various new strategies made it easy for stakeholders to see success and want more. One example is noteworthy. After-school programs were implemented in 2012 and initially served an average 120 students in each site. There was a rush to serve as many students as possible, but the adoption of such a large program the first year was challenging. Stakeholders suggested that retrospectively, the programs should have started small, “got their feet wet,” and then expanded to serve more youths. Significant needs among youths and families facilitated a sense of urgency to improve conditions but also were a challenge because stakeholders felt overwhelmed. This was especially the case in school 3, the building with the most at-risk student population. High needs among students in this school created a sense of urgency related to the adoption of new programs, but the addition of more programs and interventions added multiple new stressors to the system. Finally, funding was a key facilitator, as diverse funding streams were used to support implementation efforts. The lack of dollars, however, often limited progress. Barriers Interview and focus group respondents mentioned how some teachers and staff, district leaders, and local partners “lacked awareness” of the significant needs among students and in the community. They reported how some individuals did not know the extent to which families were struggling, and in turn had less value placed among adopting interventions to address nonacademic barriers. This seemed to especially pertain to schools 1 and 3, where there were the most needs. In addition, participants shared how they sometimes had differing expectations for what a community school should be and how much time and resources they thought were needed to implement expanded strategies. Stakeholders mentioned ways in which turf hindered implementation. Whether among internal school partners or between school and community leaders, tensions arose when negotiating how to make decisions regarding what was perceived as “my school,” my program, or my student. Challenges were especially evident in regard to the relationships among school employees and after-school program staff, and in schools with the largest after-school programs. One major challenge involved who “owned” the space. Teachers were frustrated by the after-school program using “my class” to run its program. Teachers and after-school program staff had different philosophies in relation to the primary focus of the program (that is, academic only versus whole-child focus) and the degree of structure desired. Additional turf issues and tensions resulted in ongoing conflicts and involved issues related to funding, staffing, performance expectations, communication, poor implementation efforts, and interpersonal relationships. Finally, stakeholders reported that many people working in the community school experienced burnout and struggled with whether they could indeed make an impact. Turnover among district and school leaders, teachers and staff, and students also was seen as a barrier. This theme was noted primarily in relation to schools 1 and 3, where there were higher percentages of students who were English language learners and living in poverty. Discussion This mixed methods study provides support for the adoption and implementation of a community schools model. Initial indicators of the impact were discernible in the four Title I schools two years postadoption. Foremost, large numbers of youths and families were served through the partnership agenda, particularly in the after-school programs and school-based mental health services. School-level outcomes improved over the two-year period. Specifically, school report card scores improved in three of the four schools (and the one school not improving had the most favorable school report card at baseline and had less room to grow in relation to improvements). Behavioral data improved in all four schools, as both absenteeism and ODRs decreased from baseline to two years postimplementation. Schools with the least amount of improvement were the ones most high impact (schools 1 and 3), as there were high levels of needs among students and families, as well as additional barriers related to mobility and family stability. Marked improvements were noted in teacher and staff perceptions, especially in relation to school-level constructs related to the learning support system and school climate. Improvements in these areas may be attributed to the strong focus initially placed on adopting PBIS-related universal strategies and improving tier 2 supports through targeted after-school tutoring program, social skills groups, and Care Team wraparound supports. As the four Title I schools moved through the CCMSI milestones, new ways of “doing business” were developed. New systems, policies, and partnerships facilitated adoption and implementation, and people working in and with the schools took on new or expanded roles and responsibilities. Schools with solid organizational structures, strong principal and community school coordinators, and effective Care Teams seemed to have the most progress. Academic data, however, revealed that more work is needed, as student academic performance in all schools was still well below benchmarks and state averages. The barriers noted here, as well as other challenges within high-impact schools, must be addressed to promote further progress and turnaround. Data comparisons across the schools further demonstrate this issue. The two schools with the most challenging student populations (schools 1 and 3) had less amplified behavioral improvements over time than the other two schools in the case study. Stakeholders also accentuated more barriers to adoption in these settings. School reforms are more difficult given the complexities of these settings. Findings should be considered in light of the study’s limitations. The adoption and implementation of the community schools agenda took place in real time and real place, instead of in a controlled environment. Although ecological validity was promoted, the design limited generalizability and implications related to causal effect. Likewise, many innovations occurred during the two-year implementation period, making it difficult to determine which strategies made the most difference. Future research should explore the adoption of the CCMSI using more rigorous research designs, and continue to explore variabilities in implementation efforts based on local school and community contexts. Also, only perceptions of teachers and staff were explored, and these data were collected in the aggregate (so the degree to which individual changes occurred is unknown). Additional studies exploring different stakeholder perspectives are needed. In the end, the challenges of applied research in complex schools are evident and point to the need for a cautious interpretation. That said, it is important to remember that moving school-level academic indicators takes time, up to five to 10 years in some cases (Fullan, 2001). Implications for School Social Work Practice Taken together, lessons learned from this case study can inform other efforts to ensure the healthy development of all youths through school–family–community partnership models such as community schools. Learning support system elements discussed here—such as new roles (for example, community schools coordinators), expanded planning processes, new data usages, redefined roles, and new teaming structures (for example, Care Teams and building leadership teams)—may be adopted elsewhere to facilitate implementation efforts. In addition, the case for the role of schools as hubs of partnership development is evident, especially in relation to promoting school climate and improving academic and behavioral outcomes (ones known to influence graduation and dropout). One key implication centers on the need for social workers who can take on new roles to facilitate school–family–community partnerships to support the overall grant challenges agenda. In these schools, the community school coordinators used indirect practice skills to build partnerships, facilitate relationships, coordinate services, oversee the learning support system, and design and implement programs and maximize resources. In addition, community mental health workers (most of whom were licensed clinical social workers) colocated their services to the school. These school-based mental health providers were more than outsiders in host agencies; they were integral parts of the Care Teams and consulted regularly with school personnel. These new and expanded roles of social workers working in schools are central to assisting schools in promoting learning and healthy youth development, especially among students who are falling behind or have nonacademic barriers to learning. Given these evolving roles and responsibilities, preservice training programs and professional development opportunities are needed to ensure that school social workers possess the core competencies required to do this complex work. Competencies such as leading and working in teams, building trust, and facilitating collaborations are critical for success (Ball, Anderson-Butcher, Mellin, & Green, 2010; Mellin, Anderson-Butcher, & Bronstein, 2011). Conclusion A community schools approach, especially one guided by CCMSI, provides a framework for integrating nonacademic strategies into traditional school improvement efforts. Schools can become community hubs and help to maximize school, family, and community resources to support academic learning, healthy development, and overall school success. This work is critical in light of the complexity of factors affecting youths and contributing to high dropout rates, and has implications for how social work as a discipline mobilizes to promote the healthy development of all youths as part of the grand challenges agenda. As social workers rise to meet the needs of today’s youths, community schools and other school–family–community partnership models should be considered a promising strategy. References Adelman, H. S., & Taylor, L. ( 2002). Building comprehensive, multifaceted, and integrated approaches to address barriers to student learning. Childhood Education, 78( 5), 261– 268. Google Scholar CrossRef Search ADS Anderson-Butcher, D. ( 2004). Transforming schools into 21st century community learning centers [Trends & Issues]. Children & Schools, 26, 248– 252. Google Scholar CrossRef Search ADS Anderson-Butcher, D., Amorose, A. J., Iachini, A., & Ball, A. ( 2013). Community and Youth Collaborative Institute School Experience Surveys . Columbus: Ohio State University, College of Social Work. Anderson-Butcher, D., Iachini, A. L., Ball, A., Barker, S., & Martin, L. D. ( 2016). A university–school partnership to examine the adoption and implementation of the Ohio community collaboration model in one urban school district: A mixed-method case study. Journal of Education for Students Placed at Risk, 2, 190– 204. Google Scholar CrossRef Search ADS Anderson-Butcher, D., Lawson, H. A., Bean, J., Flaspohler, P., Boone, B., & Kwiatkowski, A. ( 2008). Community collaboration to improve schools: Introducing a new model from Ohio. Children & Schools, 30, 161– 172. Google Scholar CrossRef Search ADS Anderson-Butcher, D., Lawson, H. A., Iachini, A., Bean, J., Flaspohler, P., & Zullig, K. ( 2010). Capacity-related innovations resulting from pilot school and district implementation of a Community Collaboration Model for School Improvement. Journal of Educational and Psychological Consultation, 20, 257– 287. Google Scholar CrossRef Search ADS Anderson-Butcher, D., Lawson, H., Iachini, A. L., Flaspohler, P., Bean, J., & Wade-Mdivanian, R. ( 2010). Emergent evidence in support of a Community Collaboration Model for School Improvement. Children & Schools, 32, 160– 171. Google Scholar CrossRef Search ADS Anderson-Butcher, D., Mellin, E., Iachini, A., & Ball, A. ( 2014). Promoting mental health in schools. In B. Melnyk & P. Jensen (Eds.), A practical guide to child and adolescent mental health screening, early intervention, and health promotion ( 2nd ed., pp. 357– 373). Cherry Hill, NJ: National Association of Pediatric Nurse Practitioners. Ball, A., Anderson-Butcher, D., Mellin, E., & Green, J. ( 2010). Developing interdisciplinary competencies for expanded school mental health practice: An exploratory study. School Mental Health, 2, 114– 125. Google Scholar CrossRef Search ADS Barker, C., & Pistrang, N. ( 2005). Quality criteria under methodological pluralism: Implications for conducting and evaluating research. American Journal of Community Psychology, 35, 201– 212. Google Scholar CrossRef Search ADS Blank, M., Melaville, A., & Shah, B. ( 2003). Making the difference: Research and practice in community schools . Washington, DC: Coalition for Community Schools and the Institute for Educational Leadership. Callan, P. M., Finney, J. E., Kirst, M. W., Usdan, M. D., & Venezia, A. ( 2005). The governance divide: A report on a four state study on improving college readiness and success . San Jose: National Center for Public Policy and Higher Education. Canyons School District. ( 2014). Academic and behavioral data records . Sandy, UT: Author. Dryfoos, J., Quinn, J., & Barkin, C. (Eds.). ( 2005). Community schools in action: Lessons from a decade of practice . New York: Oxford University Press. Flaherty, L. T., Weist, M. D., & Warner, B. S. ( 1997). School-based mental health services in the United States: History, current models, and needs. Community Mental Health Journal, 32, 341– 352. Google Scholar CrossRef Search ADS Fullan, M. ( 2001). The new meaning of educational change ( 3rd ed.). New York: Teachers College Press. Hawkins, J. D., Jenson, J. M., Catalano, R., Fraser, M. W., Botvin, G. J., Shapiro, V., et al. . ( 2015). Unleashing the power of prevention [Discussion Paper] . Washington, DC: Institute of Medicine and National Research Council. Hurwitz, L., & Weston, K. ( 2010). Using coordinated school health to promote mental health for all students . Washington, DC: National Assembly on School-Based Health Care. Lawson, H. A. ( 2013). Third-generation partnerships for P-16 pipelines and cradle-through-career education systems. Peabody Journal of Education, 88, 637– 656. Google Scholar CrossRef Search ADS Leone, P. E., & Bartolotta, R. ( 2010). Community and interagency partnerships. In R. Algozzine, A. P. Daunic, & S. W. Smith (Eds.), Preventing problem behaviors: Schoolwide programs and classroom practices ( 2nd ed., pp. 167– 180). Thousand Oaks, CA: Corwin Press. Lewallen, T. C., Hunt, H., Potts‐Datema, W., Zaza, S., & Giles, W. ( 2015). The whole school, whole community, whole child model: A new approach for improving educational attainment and healthy development for students. Journal of School Health, 85, 729– 739. Google Scholar CrossRef Search ADS Mellin, E. A., Anderson-Butcher, D., & Bronstein, L. R. ( 2011). Strengthening interprofessional team collaboration: Potential roles for school mental health professionals. Advances in School Mental Health Promotion, 4( 2), 51– 61. Google Scholar CrossRef Search ADS Moore, K. A., & Emig, C. ( 2014). Integrated student supports: A summary of the evidence base for policymakers . Bethesda, MD: Child Trends. Patton, M. Q. ( 1990). Qualitative evaluation and research methods (2nd ed.). Newbury Park, CA: Sage Publications. Ruppert, S. S. ( 2003). Closing the participation gap: A national summary . Denver: Education Commission of the States. Stark, P., & Noel, A. M. ( 2015). Trends in high school dropout and completion rates in the United States: 1972–2012 (NCES 2015-015). Washington, DC: U.S. Department of Education, National Center for Education Statistics. Utah State Office of Education. ( 2014). School report card data . Salt Lake City: Author. © 2017 National Association of Social Workers
Children & Schools – Oxford University Press
Published: Jan 1, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera