Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
5 Family, Community, and Peers High schools can increase students' engagement in learning by creating personalized, supportive contexts focused on academic achievement. But schools cannot achieve the high levels of engagement and standards for learning currently asked of them by themselves (e.g., Cohen and Ball, 1999; Comer, 1997; Epstein, 2001b; Gold, Simon, and Brown, 2002; Hill, Campbell, and Harvey, 2000; Shirley, 1997; Steinberg, 1997~. Patricia Gra- ham (1995, p. 22) makes this point vividly: The battleship, the school, cannot do this alone. The rest of the education- al flotilla must assist: families, communities, government, higher educa- tion, and the business community. Only then will all of our children be able to achieve that which by birthright should be theirs: enthusiasm for and accomplishment in learning. In some respects it makes little sense to discuss what schools can do to engage students in learning without considering the settings in which both schools and students live. Ideally, schools would build on the knowledge and interests youth develop at home and in the community and create opportunities for students to extend and apply school- learned skills in contexts outside of school. They would take advantage of resources and supports for learning in the community and be a positive force in the community for developing an environment that supports positive youth development. Elsewhere in this volume we discuss how schools can expand curricu- lum offerings by embedding learning opportunities in authentic work envi- 120
FAMILY, COMMUNITY, AND PEERS 12 ronments (Chapter 6), and how they can collaborate with health and social service providers from the community to address the nonacademic needs of students (Chapter 71. In this chapter, we focus on how schools can increase students' engagement in learning by involving their families and connecting with the community. Also discussed in this chapter are strategies for foster- ing positive peer influences. Peers are often viewed as the enemy when it comes to persuading adolescents to focus on academic pursuits. Research suggests, however, that peer effects on school engagement can be as positive as they can be negative, and that adults can affect the nature of that influ- ence by the way they structure the peer world. SCHOOL-FAMILY-COMMUNITY CONNECTIONS Many efforts to improve schools are "school-centric" in the sense that they focus exclusively on school resources and programs (Honig, Kahne, and McLaughlin, 20011. Although school-focused efforts are certainly nec- essary to improve student engagement, they may not be sufficient. Too many of the factors that shape students' behavior in school are based in their experiences outside of school. This observation does not let schools "off the hook." But it does have important implications for the kinds of school reforms that are likely to have substantial positive effects (Cibulka and Kritek, 1996; Honig et al., 2001; Kahne, O'Brien, Brown, and Quinn, 2001; Wehiage, Smith, and Lipman, 19921. Advocates of school reform initiatives that emphasize families and com- munities as partners generally view schools as part of a wider social ecology that includes neighborhoods, community organizations, and families (Na- tional Research Council and Institute of Medicine, 20021. They seek to marshal the energy, resources, and support of families and communities to promote learning and positive youth development. We will discuss evidence suggesting the value of deep, institutionalized connections among schools, families, and the community. Be forewarned that the kind of connections advocated by many experts require fundamental changes in most schools- in how they are organized, how decisions are made, and even how instruc- tion is implemented (see Comer, 1980; Connell, Gambone, and Smith, 2000; Giles, 1998; Gold, Simon and Brown, 2002; Heckman, 1996; Honig et al., 2001; Lewis, 1996; Murnane and Levy, 1996a; Wehiage et al., 19921. Consistent Messages Adolescents need many sources of support, and they need consistency in the messages they receive from the important people in their lives. Lawrence Cremin (1976) is among the many educators and researchers who over the years have described the school as but one educating institu-
122 ENGAGING SCHOOLS tion among a multiplicity of institutions that educate (see also Deigado- Gaitan, 1987; Mehan, Hubbard, and Villanueva, 1994; Phelan et al., 1998~. Cremin (1976, p. 22) noted, for example, that every family has a curricu- lum "which it teaches quite deliberately and systematically over time." Religious institutions also socialize youth to certain values and beliefs, as do workplaces, radio and television, and youth organizations. The funda- mental question is whether, in Cremin's words, "The relationships among the institutions constituting a configuration of education may be comple- mentary or contradictory, consonant or dissonant" (1976, p. 31; see also Phelan et al., 1998~. A growing body of research underscores the importance of consistent messages emanating from these settings (lessor et al., 1998; for a review, see Aber, Gephart, Brooks-Gunn, and Connell, 1997~. Findings from a study by Steinberg, Darling, Fletcher et al., (1995), for example, suggest the value of congruence between parent and peer support for academic matters. In their study of nine high schools in Wisconsin and Northern California, students who received academic encouragement from both parents and peers performed better in school than those who received encouragement from only one source (see also Lee and Smith, 19991. Darling and Steinberg's (1997) findings on social integration illustrate both the value of coherence in the messages adolescents receive and the importance of positive messages. They measured "social integration" using a questionnaire that assessed the extent to which adolescents had an oppor- tunity for nonteacher adult contact, and parents had the opportunity to meet their adolescents' friends and their parents. The undesirable effects of coherent negative messages were just as strong as the desirable effects of coherent positive messages. The authors explain (pp. 128-129~: Family social integration was associated with adolescent school engage- ment and lower levels of substance use only in communities where other adolescents were engaged in school and were not involved in substance use. In contrast, social integration was associated with lower engagement and more substance use when community adolescents were more deviant and less involved in school. In other words, social integration exaggerated the potentially negative influence of living in a community with poorly adjusted peers. These findings suggest the value of creating social connections between families and school staff and of involving families in school activities that facilitate social interactions with parents who have strong educational val- ues. In addition to promoting consistency in the messages youth receive at school and at home, school-initiated activities can help families build the social networks that have been shown to support youth development and learning.
FAMILY, COMMUNITY, AND PEERS 123 Connections to Families Creating strong linkages between schools and families may support student engagement in other ways, as well as by fostering congruence in the values adolescents are exposed to at home and at school. Although most studies are correlational, making it difficult to establish the direction of causality, research findings are generally consistent with the proposition that children benefit from their parents' involvement in their school (e.g., Ames, Khoju, and Watkins, 1993; Baker and Soden, 1998; Booth and Dunn, 1996; Clark, 1983; Dornbusch and Ritter, 1988; Eccles and Harold, 1996; Epstein, 2001a; Lee, 1994; Lee and Croninger, 1994; Morgan and Sorensen, 1999; Scott-Tones, 1987; Steinberg, Lamborn, Dornbusch, and Darling, 1992; Stevenson and Baker, 1987; Tocci and Englehard, 19911. In one of the few studies that used a longitudinal design, Steinberg (1997) was able to examine cause-effect more directly. He concluded that although parents appeared to be more involved as a consequence of their adolescents doing well in school, their involvement also appeared to contribute to better performance. The value of parent involvement is also suggested by dropout studies. In their study of 14,217 sophomores in 913 public, Catholic, and private schools from the National Education Longitudinal Study of 1988 (NELS 88), Rumberger and Palardy (2002) found lower dropout rates in schools in which parent involvement was high. Parent involvement predicted dropout rates after controlling for the academic and social-class background of students as well as school resources (e.g., student-teacher ratio, proportion of teachers with advanced degrees) and structural characteristics (e.g., size and urbanicity). Parental involvement is almost universally considered desirable by par- ents, educators, and policy makers (Epstein, 2001a), and at least nominal parental involvement has been required in some federally supported educa- tional programs since the mid-1960s through legislation authorizing Title I (Keesling and Melargano, 19831. Parental involvement was listed as a ma- jor national educational priority in the 1994 Goals 2000 Educate America Act, as well as in some local legislation, such as the 1988 Chicago School Reform Act. The Chicago legislation mandated that parents and commu- nity leaders be involved in their local school's budget making, hiring, and firing decisions, and in the development of school improvement plans (Bryk and Schneider, 2002; Bryk, Sebring, Kerbow, Rollow, and Easton, 19981. Studies reporting the benefits to students of parent involvement involve mostly elementary and middle-school children, and often do not differenti- ate among different kinds of parent participation, such as volunteering in activities at school; at-home involvement in children's learning; participat- ing in school decisions through parent-teacher organizations and other
24 ENGAGING SCHOOLS venues; communicating about children's educational progress and plans; obtaining advice from school personnel about child-rearing or parenting practices; and working with the school to collaborate with other commu- nity organizations (Baker and Soden, 1998; Epstein, 2001a, p. 4091. A notable exception is Epstein's work over the past 20 years, in which she differentiates six types of teacher-parent collaboration (e.g., parenting, learning at home, collaboration with the community) and shows that differ- ent parent behaviors are linked to different educational outcomes (Epstein, 2001a). Steinberg (1997) also differentiates the varied forms of involve- ment in his study of adolescents. He found that the most common types of involvement checking homework, encouraging their children to do better, and overseeing their academic program from home did not appear to contribute to performance, perhaps because they were parental responses triggered by academic problems and thus experienced by the student as controlling. A more powerful predictor of student performance was the type of parental involvement that drew parents into the school physically- to attend teacher conferences and school programs such as back-to-school nights and extracurricular activities. Other studies, however, indicate that students who rate their own parents' involvement and encouragement at home as high have higher academic self-concept, greater motivation, and more positive attitudes toward school (e.g., Sanders, 1998; Wiest, Wong, Cervantes, Craik, and Kreil, 2001; Wong, Wiest, and Cusick, 20021. Despite general agreement about the desirability of parental involve- ment, there is a dearth of research at the high school level. Longitudinal research that makes it possible to examine the direction of causality is also needed, and studies that differentiate particular kinds of parent involve- ment would be useful to identify the kinds of involvement that are most beneficial. What is clear from extant research is that parental involvement in schools decreases markedly during middle school and high school (Baker and Stevenson, 1986; Epstein and Dauber, 1991) and that low-income and poorly educated, single and minority parents have relatively low rates of involvement in their children's schools (Comer, 1988; Epstein, 2001a; Hoover-Dempsey, Bassler, and Brissie, 1987; Lareau, 1987,1989; Leitch and Tangri, 1988. There is also considerable evidence that school prac- tices and policies affect the level of parent involvement (Cole and Griffin, 1987; Comer, 1980; Dauber and Epstein, 1993~. One recent comprehensive study of parent involvement, using data from the eighth grade of the NELS study (Sui-Chu and Wilims, 1996), examined what student-level and school- iFor case studies about what schools do and don't do to promote parental involvement of ethnic minorities, see Delgado-Gaitan (1991).
FAMILY, COMMUNITY, AND PEERS 125 level factors influenced four types of parental involvement home discus- sion, home supervision, school communication, and school participation- and the impact of these measures on 8th-grade school achievement. The study found that schools had influence on school participation, although not on other forms of parental involvement that there were substantial differences among ethnic groups in these four parental involvement mea- sures, and that home discussion of school activities was the most strongly associated with student achievement. Epstein (2001a, p. 45) concluded from a review of research: "Teachers' practices to involve families are as or more important than family background variables such as race or ethnicity, social class, marital status, or mother's work status for determining whether and how parents become involved in their children's education." School practices may therefore explain, in part, the relatively low participation rates of parents with children who are most at risk of school failure. Schools may not facilitate the involvement of low-income parents as much as they do more affluent parents. Jordan and Plank (2000) found that parents of low socioeconomic status (SES) were a little more than half as likely as high-SES parents to have been contacted by their adolescent's high school about course selection decisions, postsecondary education, or career plans. This finding suggests that less effort was made to involve low-SES parents than more advantaged parents. But low-SES parents in their study were also less likely to attend a school-sponsored program on postsecondary educational opportunities and financial aid, suggesting that they did not take as much advantage of opportunities the school did provide. We can only speculate about why some low-SES parents might be less likely to take advantage of opportunities for involvement. The evidence does not suggest less interest or concern about their children's school achievement, and most parents want to be more involved (Epstein, 1990; MetLife, 19871. We speculate that language and culture play a role. For example, immigrant parents who do not speak English may fee! less com- fortable in school contexts and have more challenges in participating. So- cial class differences between parents and teachers could produce another psychological barrier. Work schedules, babysitting needs, transportation, and neighborhood safety may also come into play (e.g., Heymann and Earie, 20001. Some of these obstacles can be overcome through various measures, such as having communications to parents written in the languages spoken by parents and providing translators at school events; making the schoo! physically and socially hospitable to the families; and providing transporta- tion and babysitting. Because the obstacles to parent involvement most likely vary among communities, high schools need to assess the needs of their own parent community and identify local resources to address the obstacles they find. One study suggests that including information on family
26 ENGAGING SCHOOLS involvement programs and their implementation, as well as strategies for overcoming barriers to involvement, in teacher education improves teachers' confidence in their ability to involve parents in their children's education (Morris and Taylor, 19981. Communication is probably a first and necessary step to involving parents in their adolescents' education. Teachers' perceptions of the educa- tional values and goals of parents often are different from the values and goals expressed by the parents, and some teachers complain that they do not understand what information parents would like from them (Dauber and Epstein, 19931. Parents, in turn, report that they do not fully under- stand what teachers are trying to do (Epstein, 2001a). Improved communi- cation among parents, teachers, and students about educational objectives and strategies might lead to greater participation by parents as well as foster congruence between messages and supports for learning at home and at school (see Darling and Steinberg, 1997; Lamborn, Brown, Mounts, and Steinberg, 1992; Moll and Greenberg, 1990; Rosenfeld et al., 20001. Special efforts may be required for schools to connect with some par- ents, but the evidence supports the value of such efforts. A number of studies at the elementary and middle school levels have documented the benefits of teachers' efforts to collaborate with single- and low-income parents, and other parents perceived as hard to reach (e.g., Clark, 1983; Comer, 1980; Epstein, 1990; Epstein and Dauber, 1991; Rich, Van Dien, and Maddox, 1979; Scott-Tones, 19871. Moreover, teachers who find ways to work with low-income and single parents are much less likely than teachers who do not work with those parents to hold stereotypic views of them. For example, they are less likely to believe that parents don't care about their children's education or do not want to be involved with the school in activities that will lead to better outcomes for their children . (Becker and Epstein, 1982; Epstein, 19901. Presumably the breakdown of such stereotypes contributes to better communication and connections be- tween teachers and parents. Given the importance of families in adolescents' development, ignoring parents and other adults who have caretaking roles seriously limits the effects of even the most inspired school reform. Most parents want to be involved, and schools that reach out aggressively to parents and reduce obstacles to their involvement amplify their own efforts to improve student engagement and learning.2 Murnane and Levy (1996b) suggest that an outside "agent" (e.g., community organization) can sometimes help schools 2For an example of how schools can connect with parents and other caretaking adults, see the description of First Things First's family advocacy system in Chapters 6 and 8.
FAMILY, COMMUNITY, AND PEERS 127 connect with low-income parents. Some organizations that work with groups of parents to increase connections between parents and the school have created the National Coalition of Advocates for Students (see Murnane and Levy, 1996b). An example of a high school that used a coalition of local churches to help make connections to parents is discussed later in this chapter. Connecting with Communities Several national, and many state and local, organizations now provide support and guidance for school-community partnerships.3 The types of connections vary greatly, as they should to be responsive to the local needs and resources of schools and communities. Our goal here is to give a flavor of the kinds of activities that might be considered. Some of these examples involve tinkering around the edges improving communication and build- ing bridges to community organizations which can serve as the starting point for serious reforms. School-community collaborations that have sub- stantial impact on youth engagement in school, however, involve changes at the core, in how schools define their role and how they function. In addition to linking with community agencies that provide services to meet students' nonacademic needs (see Chapter 7, this volume), schools can develop close connections to after-school and other youth-serving programs. Involving students in well-organized after-school and summer activities may promote positive attitudes about school shared by groups of students because organized activities are consistent with the values and norms of the school. They can provide a positive social context for youth to demonstrate their competence and experience a sense of belonging; furthermore, partici- pants simply have less time to "hang out" with each other in unsupervised activities where they could develop shared beliefs and attitudes that are different from, and possibly in opposition to, those of the school (Lamborn et al., 1992; McLaughlin, 2000; National Research Council and Institute of Medicine, 20021. High schools can also involve community members and create support by serving as a resource for the community. Many high schools provide evening courses for adults in everything from carpentry and computers to Spanish and sewing. Athletic facilities can be made available and perfor- mances can be advertised throughout the community. School personnel and students can show their commitment to the community by participating in 3Examples include Coalition for community Schools (http //www.communityschools.org) and the National community Building Network (http //www.ncbn.org).
28 ENGAGING SCHOOLS clean-up and beautification projects, inviting community members to ac- tivities, such as fairs and pancake breakfasts, and taking their performances out into the community. Parents and other community members can be invited to spend a Saturday helping to improve the physical plant of the school. Community members can be enlisted to advocate for school re- sources, and school personnel and students can assist in efforts to bring resources to the community. School-community connections can also be established to enrich the instructional program. College faculty, students, and local business people can come to the school to work directly with students or participate in classroom instruction. Local artisans and artists can visit classes to demon- strate their craft and to work directly with students. Students can expand their knowledge and learn how to apply new skills by working as interns or volunteers in community settings. More powerful strategies for school-community connections transform schools so that they become one among many coordinated settings in which youth are educated. We will explain what this means through examples. The first example involves a sustained collaboration between a school and a community organization that integrates the academic program with ser- vice to the community. The second is a description of a school-community collaborative in Houston, Texas, which became the basis of a national school reform mode! called Project GRAD. The third example illustrates the growing practice of service learning. All three examples address some or all of the principles of engagement described in this volume and they all promote productive links between schools and their communities. E! Puente Academy for Peace and lustice The E! Puente Academy for Peace and Justice was established in 1993 by the E! Puente community organization with support from New York City's New Visions initiative. An after-school program is so well connected to the academic curriculum of the school that it is difficult to differentiate where one ends and the other begins. Teachers and youth organization staff collaborate with each other and work with youth both during and after the regular school day. Students learn basic skills in part through projects. For example, as part of a math and science unit, they spent afternoons and weekends creating a community garden in a vacant lot. The activity was designed to teach math skills in the context of planning and budgeting and science skills in the process of selecting and growing plants. In another activity students devel- oped writing skills in the context of producing a community newspaper. The program was created to give students opportunities to express their own values and concerns and to build leadership skills by assessing commu-
FAMILY, COMMUNITY, AND PEERS 129 nity needs and making decisions about where to invest their time. One group of students decided to organize an immunization drive for young children. Another group surveyed 500 residents and subsequently devel- oped a walking tour that included historic, economic, and environmental analysis of the south side of the city. Although not a formal evaluation, it is noteworthy that 92 percent of the first graduating class of 33 seniors at- tended college, a figure far above what would be expected given the stu- dents' backgrounds. Project GRAD USA Based on an educational collaborative developed at Jeff Davis High School in Houston, Texas, Project GRAD (Graduation Really Achieves Dreams) is a national program promoting school-community collaboration and the vertical integration of reform efforts, from kindergarten through college. In 2002, Project GRAD was serving 80,000 students attending urban schools with a history of poor performance in eight cities across the United States. Among its goals are to ensure that 80 percent of entering 9th graders in Project GRAD high schools go on to graduate and that 50 percent of graduates enter and complete a program of postsecondary edu- cation.4 A brief history illustrates the many ways a high school can be con- nected to and benefit from the resources of the community. Since 1981 Jeff Davis High School had been involved in a business partnership with Tenneco Corporation, which was located near the school. The partnership provided Davis with tutors, mentors, and some college scholarship money. But when Emily Cole arrived as principal in 1989, Davis High School had all the problems associated with an urban, high-poverty, low-achieving school. Students were wandering the hallways and absenteeism was ram- pant. Nearly 20 percent of Davis students dropped out every year, and fewer than 10 percent of its graduates went on to college. That year, 65 Davis students became pregnant. It was clear that much more needed to be done. Cole made a number of changes at the school, which serves a predomi- nantly Hispanic and low-income community. Internally, she looked for ways to implement research-based programs and practices reflecting the child-centered philosophy of education she previously had found successful as an elementary school principal. To personalize relationships of students 4For further information on Project GRAD, see http //www.projectgradusa.org.
130 ENGAGING SCHOOLS to adults, she adopted block scheduling, restructured the ninth grade to create small learning communities, and restructured the administrative team. To increase students' responsibility for learning, a well-researched coopera- tive discipline program was adopted. But Cole believed that to make a significant difference in education outcomes and the lives of her students, she also would need the help of parents and members of the community. She initiated changes in school governance that allowed parents and teachers to be involved in decision making about the school reform process. With the addition of new partners and new outreach strategies, the high school's business partnership with Tenneco was transformed into a much broader community collaborative. Among the new partners was "Communities in Schools,"5 which pro- vided on-site staff at Davis to help students stay in school by connecting them with whatever social services they needed (see Chapter 7, this vol- ume). Another was the University of Texas Health Science Center, which provided medical services and worked to help cultivate students' interest in the health professions. Also joining the collaborative was The Metropolitan Organization (TMO), a coalition of local churches. Their goal was to help Davis High School to improve communication with parents and also to garner support from Hispanic business professionals, church groups, and other key ele- ments of the local community. With guidance and assistance from TMO, Davis High School implemented its first "Walk for Success" in 1989. Davis staff, including Cole and TMO volunteers, visited the home of every Davis freshman to explain the college scholarship opportunities ($1,000 per year for up to 4 years) and their requirements. Parents were asked to sign an agreement outlining the requirements, one of which was that their high school student attends two summer institutes at the nearby Downtown campus of the University of Houston. TMO volunteers made follow-up contacts with parents to help them monitor their children's progress. Since these collaborative activities with the community were imple- mented, there has been slow but steady improvement in educational out- comes for Davis students. Annual dropout rates from the school have de- creased, and college enrollment has increased markedly. The percentage of students who graduate on time has improved, as have standardized test scores. The coordinated educational reform strategy in Davis' feeder el- ementary and middle schools that was begun in 1993 is gradually increas- 5Communities in Schools is a national model with local affiliates around the country. Its purpose is to connect community resources with schools to promote student learning and healthy development and to prevent students from dropping out (http://www.cisnet.org).
FAMILY, COMMUNITY, AND PEERS 13 ing the number of entering students who have the skills needed to succeed in their 9th-grade classes, and beyond. Davis' efforts demonstrate the prom- ise, the complexity, and the need for patience and persistence in school reform. Service Learning Service learning is another strategy for creating ties between schools and their communities and making the instructional program relevant to students' lives outside of school.6 Students are placed in various commu- nity-serving organizations in ways that are directly connected to the cur- riculum (for example, in science, civics, or social studies). A 1999 study by the National Center for Education Statistics found that nearly half of all U.S. high schools use service learning as part of the curriculum (Pearson, 2002; Skinner and Chapman, 19991. Like work-based learning (see Chapter 6, this volume), service learning provides students with different experiences and information that can make schooling more meaningful and help them to formulate future options. It is based in part on research showing that learners become more motivated when they perceive that what they are learning may be useful to others (Cognition and Learning Technology Group of Vanderbilt University, 1998; McCombs, 1996; National Research Council, 1999; Pintrich and Schunk, 19961. Advocates for service learning focus on four kinds of student engage- ment: civic engagement, social engagement with the communities they serve, personal engagement with individuals they serve, and academic engage- ment, which teachers mention as one of their principal reasons for using service learning (Ammon et al., 20021. The evidence on the effects on school engagement, although sparse, is positive. Melchior's (1997) national study showed greater self-reported engagement (attention, effort, positive emotions) among students whose teachers included service learning than among students whose teachers did not. A large portion of the service learning students (87 percent) reported having learned a new skill that they believed would be useful in the future. In addition, statewide studies in Florida and California showed increased school attendance among students while they were engaged in service learn- ing, compared to the previous semester or year, when they were not partici- pating in service learning (Foliman and Muldoon, 1997; Weiler, LaGoy, 6Service learning is distinguished from community service in that service learning, although it may meet community needs, is deliberately designed to develop skills and understanding of academic content.
32 ENGAGING SCHOOLS Crane, and Rovner, 19981. The authors concluded that service learning makes the curriculum more interesting and more relevant for students and has the potential to motivate students to engage more genuinely in class- . . . room activities. The Teen Outreach Program that combines structured community ser- vice experiences with classroom discussions about careers and relationships has been studied more than most service learning programs. Studies using matched comparison of students in 35 states (Allen, Philliber, and Hoggson, 1990) as well as experimental designs at 25 sites (Allen, Philliber, Herrling, and Kupermine, 1997) found advantages of program participants in rates of school suspension, school dropout, school failure, and even pregnancy. Honig et al. (2001) summarize other studies that provide clues about why service learning might be a particularly effective strategy for engaging youth in school. They suggest that students are more motivated in part because they are applying learning in contexts (hospitals, construction sites) that are personally meaningful. They also suggest that the academic skills youth develop in these settings build their confidence as learners, and that they develop social, emotional, and other nonacademic competencies and greater personal and social responsibility, which facilitate success in school. Three qualities of service learning programs appear to contribute to their success (Honig et al., 20011. First, they are most effective when stu- dents have regular structured time to discuss the content and the process of their practical experiences. Second, successful programs give students op- portunities for personal agency to develop their own ideas and pursue their own interests. Third, effective service learning programs are closely linked to the academic curriculum. It's Not Easy, But It's Worth the Effort Documentation of efforts to develop and sustain school-family-com- munity connections indicates that they are difficult to achieve (Honig et al., 2001; Kahne et al., 2001; White and Wehiage, 19951. Differences in cul- tures, competition for resources and control, limitations in time, mistrust, bureaucratic and funding barriers, and other obstacles need to be over- come. Short-term collaborations and activities are easier, although the pay- off in terms of enhancing youth engagement in school and promoting posi- tive development are not likely to be as great as sustained, coordinated efforts. The evidence suggests that creating such linkages is worth the effort. The effects of the kinds of community partnerships we have described are difficult to isolate for many reasons, including the need to identify an appropriate control group and disentangle the effects of initiatives to col- laborate with families and the community from the effects of other initia-
FAMILY, COMMUNITY, AND PEERS 133 fives being implemented at the same time. Most program evaluations fall short of standards for rigorous scientific evidence, but looking broadly, the relevant evidence suggests the value of such initiatives. The Coalition for Community Schools compiled information on 49 evaluation reports of school-community programs (Dryfoos, 20021. The programs varied substantially in the nature of the activity involved and the level of implementation (e.g., national, state, local school district). Taken together, the programs evaluated showed evidence of achievement gains, improved behavior (with fewer suspensions), reduced high-risk behavior (e.g., drug use), improved parenting practices and parent involvement in school, and lower neighborhood violence rates. Clearly, programs vary in their value and effects on students. Honig et al. (2001, p. 1019) summarize succinctly evidence on the qualities of school- community connections that engage youth and promote positive develop- ment. Effective programs are school. focused on each youth in a holistic sense; focused on all youth; strengths-based, prosocial, and developmental; responsive to specific youth and neighborhoods; youth-centered; and filled with expanded opportunities to learn from adults in and out of Having reviewed the evidence, the committee is convinced that until all organizations, including schools, see themselves as partners in supporting positive youth development, efforts to increase student engagement in school will have limited success. Moreover, organizations, models, and lessons learned from previous efforts are available now to assist schools and com- munities in developing useful collaborations.7 PEERS The social world of peers, like that of the family, can promote engage- ment in learning to the extent that the social relationships in those settings are emotionally supportive and cultivate values consistent with those of the 7A source for such models and lessons learned is the National Network of Partnership Schools at Johns Hopkins University, which has documented many research-based, goal- oriented approaches to improving family and community involvement in the schools. The network offers many ideas for school and district personnel to implement that have success- fully increased family and community participation, such as after-school tutoring programs, parent surveys, and back-to-school picnics (http://www.partnershipschools.org).
34 ENGAGING SCHOOLS school (Comer, 1980; Connell, Gambone, and Smith, 2000; Giles, 1998; Gold et al., 2002; Heckman, 1996; Lewis, 1996; Murnane and Levy, 1996a; National Research Council and Institute of Medicine, 20021. Because peer interactions take place inside of schools as well as in community settings, many if not most peer relationships among adolescents are with other students. Educators, therefore, have some effect on whether peer groups and friendships support or undermine engagement in learning. lust as they can reach out to parents to make the familial environment more supportive of learning, they also can influence the social world of peers. Peer influences are as complex and multilevel as the peer social world itself. To better understand how peers affect adolescents' engagement in learning, we review four lines of research on peer cultures, peer crowds, gangs, and friendship groups. Peer Cultures Casual observations of adolescents' behavior may appear to support Coleman's (1961) claim that they are more concerned about being popular with peers than about achieving academically. Even if this is true, it does not mean that adolescents are not concerned about academic achievement. More recent research suggests that the norms of the peer group in most high schools support, at least at some modest level, academic achievement rather than disengagement (for a review, see Brown, 19901. For example, urban high school students in one study reported that their peers generally encour- aged them to study their school subjects rather than not to do so (Brown, Clasen, and Eicher, 19861. For~ham and Ogbu (1986) have proposed that an antiacademic peer culture is common among Black adolescents.8 They based their thesis on an ethnographic study of a Washington, D.C., high school where nearly all of the students were Black and many were from low-income families. They observed that "peer groups discourage their members from putting forth the time and effort required to do well in school and from adopting the attitudes and standard practices that enhance academic achievement. They oppose adopting appropriate academic attitudes and behaviors because they are considered 'white' " (p. 183; see also Chapter 2, this volume). Steele (1992) also has argued that African-American students believe they have to give up value priorities, preferences, and styles of speech and appearance to master mainstream (predominantly white) culture, and that 80gbu has also applied his notion of oppositional culture to Mexican-Americans (e.g., Ogbu, 1989; see also Romo and Falbo, 1996, and Valenzuela, 1999).
FAMILY, COMMUNITY, AND PEERS 135 some prefer instead to disidentify with mainstream culture. He notes "once Misidentification occurs in a school, it can spread like the common cold.... One's identity as an authentic Black is held hostage, made incompatible with school identification" (p. 751. Other qualitative studies provide further evidence that some Black stu- dents have peer groups that maintain norms against high achievement. In a study of 10 schools engaged in voluntary detracking, some of the Black students interviewed were apprehensive about taking honors classes be- cause they feared being ostracized by Black peers (Yonezawa et al., 20021. A few reported that Black students called Black peers names like "sellout" or "whitewashed" to deter them from developing relationships with white students. For~ham (1996) also found that some Black students in the high school she studied expressed concern about losing their ethnic identity as the result of becoming a successful student. Additional data relevant to the issue of the value students place on academic success is found in two studies of sixth through eighth graders in economically depressed areas of Los Angeles (Graham, Taylor, and Hudley, 19981. When asked to name classmates whom they admired, respected, and wanted to be like, the African-American boys in these studies named more low-achieving classmates than high-achieving classmates. By contrast, African- American girls named more high-achieving than low-achieving classmates as peers whom they admired, respected, and wanted to be like. The data for girls are inconsistent with For~ham and Ogbu's (1986) hypothesis because their ethnographic study suggests that both boys and girls experience a peer culture that devalues academic achievement. Even for boys, the data do not clearly demonstrate that the African-American boys devalued academic achievement, because low-achieving African-American boys might have nominated low-achieving classmates not because they devalued academic achievement, but because they had a positive view of classmates whom they perceived as similar to themselves. Nevertheless, the study suggests that the African-American boys in the sample did emulate peers who were low ace sieving. Survey evidence contradicts For~ham and Ogbu's (1986) hypothesis. Peer culture was indirectly examined in analyses of the data obtained from nearly 17,000 high school sophomores who participated in the first follow- up of NELS (Ainsworth-Darnell and Downey, 1998; see also Downey and Ainsworth-Darnell, 20021. One of the questions the study asked was whether African-American high school students who were high in academic achievement were unpopular with their peers. Most of the students who participated in NELS were white, but the sample included more than 2,000 African-American students. In the sample as a whole, students who re- ported that their classmates regarded them as very good students were also likely to report that they were popular, socially active, and among the social
136 ENGAGING SCHOOLS leaders in the school. Moreover, the relationship between academic achieve- ment (i.e., being a very good student) and popularity was even stronger for African-American students than for white students. Also using NELS data, Cook and Ludwig (1998) found no difference, when controlling for socioeconomic background, in the proportion of Black and white 10th graders who expected to go to college, or the amount of time they spent on homework, or rates of absenteeism variables they claim should have been higher for Black students if they were dispropor- tionately alienated from high school. Ferguson (2002) found in his survey of ethnically mixed suburban schools that more of the Black students than those from any other ethnic group checked "very important" to a question about how strongly friends would agree with the statement, "It's important to study hard to get good grades." In a MetLife (2002) national survey of youth from 7th to 12th grades, more students (64 percent) claimed to have one or more friends who liked school than claimed to have friends who thought that doing well in school was not "cool" (24 percent). The students' ethnicity and their parents' education level were only modestly associated with their responses to these questions. Black students were slightly more likely (67 percent) to report that they had friends who liked school than were white (62 percent) or Hispanic (64 percent) students. Teachers' beliefs about students' friends were, in contrast, strongly associated with the demographic characteristics of the school population. In schools with high minority (more than 66 percent) enrollment, teachers expected about 45 percent of their students to "hang out with people who believe that doing well in school is not 'cool'," compared to 30 percent of the students in schools with 34-66 percent minority students, and 20 percent of the students in schools with less than 34 percent minority students. Taken together, the findings from these and other studies (e.g., Arroyo and Zigler, 1995; Cook and Ludwig, 1997, 1998; Downey and Ainsworth- Darnell, 2002; Farkas, Lleras, and Maczuga, 2002; Spencer, Cross, Harpalani, and Gross, in press; Spencer et al., 2001) suggest that the hy- pothesis of a distinctly oppositional peer culture in high schools with large numbers of students of color is not pervasive. Although such antiacademic subcultures have been observed in ethnographic work, the quantitative survey data suggest that they are not the norm for African-American stu- dents. Viewed from a different perspective, however, the peer culture may be somewhat antiacademic for high school students from all ethnic groups.9 9For examples of cultural subgroups that foster or at least reflect poor school attitudes, see Valenzuela, 1999.
FAMILY, COMMUNITY, AND PEERS 137 in particular, the norms of the high school peer group are likely to be different from the norms of the students' parents. Ethnographic studies suggest that parents of urban high school students think their children should spend more time studying and doing homework than their children think they should spend (e.g., For~ham, 19961. Quantitative studies of differences between adolescents' and their parents' values are rare because few researchers assess both students' and parents' views about academic achievement and even fewer use comparable measures for both. There is, however, some relevant evidence. Coleman (1961) reported that more par- ents wanted their children to be remembered as brilliant students in high school than their children did. In another study conducted in junior high schools (Berndt, Miller, and Park, 1989), students said their parents would be less accepting of misbehavior in class than they were. Ethnographic reports and survey data suggest that many students believe their peers have a negative view of students with high academic achievement (Arroyo and Zigler, 1995; For~ham, 19961. In summary, there is no question that an antiacademic orientation can be found among some students, although large-scale empirical data do not support claims that an antiacademic orientation is pervasive among Afri- can-American students. The existence of such an orientation among some groups of students, especially in urban high schools serving low-income communities, and its negative effects on engagement and learning, suggest the importance of investigating its origins and how schools can address the problem. It is unlikely that those students who demonstrate an antiaca- demic orientation actually wish to fail in school. A possible explanation is that they perceive the demands of academic work as threatening, perhaps because they are asked to carry out academic tasks they do not understand or because they do not perceive the value of carrying out the tasks to be worth the effort required. Whatever the reason and however small the numbers, special attention needs to be given to preventing the development of such values that undermine engagement. A major limitation of the research on peers is that the majority of studies are on white middle-class youth, and few studies examine subgroup differences. In addition, the effect of the school climate, as discussed in Chapter 5, on the peer culture has not been studied extensively. An impor- tant exception is the research of McDill and Rigsby (1973), which showed that whether or not peer influence "mediated" the effects of the normative school climate on student academic behavior depended on the measure of peer influence used, and how the constructs are operationalized. As McDill and Rigsby's work suggests, it is likely that the positive effects of high school climates characterized by high expectations and a press for academic achievement are partly explained by the effect of such a climate on students' own values related to learning and achievement and the values their peers
138 ENGAGING SCHOOLS espouse. More research is needed on how the climate promoted by adults affects peer norms distinguishing among different groups based on gen- der, class, ethnicity, and cultural backgrounds. Peer Crowds When given opportunities to interact with peers, adolescents typically choose peers who are similar in clothing, hairstyles, ethnic background, social class, and other attributes, including academic achievement, school engagement, extracurricular activities, and activities outside of school. Col- lections of friendship groups that share a number of attributes are found across multiple settings. These are referred to as "crowds." Crowds are reputation-based collectives of similarly stereotyped individuals who may or may not spend much time together (Brown, 19901. Thus, a crowd is defined by its reputation in the entire peer group for certain characteristic activities or attributes. Some crowds such as the "draggles" are defined by their primary activities. Other crowds are defined by personal attributes such as ethnicity (in ethnically diverse schools; e.g., "Mexicans"), social status (e.g., "populars"), or academic achievement (e.g., "brains". Because crowds are defined by reputation rather than social interaction, not all members of the same crowd interact socially. In a large high school, some members of the same crowd may not even know one another. Students who belong to different crowds typically differ in their aca- demic achievement. For example, in one major study of peer crowds that included thousands of high school students from ethnically diverse rural and urban high schools (Brown, Mounts, Lamborn, and Steinberg, 1993), students named five crowds to which their classmates belonged: brains, populars, druggies, outcasts, and jocks. Students in the brain and popular crowds generally had high grades; students in the druggie and outcast crowds had low grades, and the jock crowd was in the middle. Belonging to a particular crowd could influence high school students' behavior, but perhaps not exactly as common sense might suggest. Because members of the same crowd do not necessarily interact with one another, there may be no direct influence of crowd members on one another. More- over, students may belong to the same crowd because they have indepen- dently developed certain patterns of behavior, not because other crowd members encouraged them to initiate those behaviors. The most powerful influence of crowd membership may be an indirect result of the treatment received by adults and members of different crowds (for illustrations of this differential treatment, see the case studies by Fine, Valenzuela and Bowditch, 19931. The status hierarchy of crowds within a school is strongly reinforced by the adults there. Principals and teachers favor the members of high-status crowds because those students typically
FAMILY, COMMUNITY, AND PEERS 139 are vocal in their support for school activities and they bring positive forms of attention to the school. For example, adults in the school are usually delighted when athletic teams succeed in high-profile interscholastic ath- letic events. Members of low-status crowds, in contrast, are less revered because they are more difficult to manage and they sometimes bring nega- tive forms of attention to the school. Most students know in which crowds their peers place them and where their crowd falls in the status hierarchy (O'Brien and Bierman, 1988), and the relationships between high-status and low-status crowds are often hos- tile (Coleman, 1961; Eckert, 19891. Consequently, high school generally is not a supportive setting for students in low-status crowds. They often are disliked by their teachers and rejected by their higher status classmates. Not surprisingly, many are disengaged from school or actively hostile toward students and teachers. Stated colloquially, they can see that they are not wanted, and they drop out at the first opportunity (Brown, 1990; Eckert, 19891. The school organization can reinforce the effects of crowd member- ship, or it can minimize them. Academic tracking, for example, makes differences in students' social status highly visible. Tracking can also exac- erbate negative peer influences by isolating students with low academic achievement from peers who are academically engaged and value achieve- ment. Adults can reinforce status difference by being harsher, more authori- tarian, and less supportive and encouraging of low-status students. Rather than reinforce a status hierarchy that adolescents create, schools need to make concerted efforts to counteract a social context that fosters disengagement among low-status students. Eliminating tracking is one strat- egy for making differences among students' skill levels less visible. Another strategy is to create activities that involve students from different crowds that are affirming and decrease the salience of the status hierarchy. Such activities can contribute to a school climate that is cohesive and supportive of individual students and of academic values. Recognizing students for various talents and achievements can also raise the status of students who do not perform well on conventional mea- sures. Some students who are neither high achievers nor jocks have artistic talents (e.g., drawing, break-dancing) and skills (e.g., woodwork, restoring cars) that can be encouraged, publicized, and built on to improve academic engagement. In classes, students' status can be raised by giving them re- sponsibilities in cooperative learning activities, by arranging for them to tutor younger children, and by drawing out their unique experiences and cultural knowledge. In these and other ways, adults in schools can reduce the negative effects of crowd stereotypes on students in low-status crowds, and thus encourage their sense of belonging and commitment to school.
140 ENGAGING SCHOOLS Gangs Gangs are a form of peer group that does not promote academic en- gagement. In many urban high schools, members of youth gangs use their clothing and behaviors to signal their membership to others, especially students in rival gangs (Howell and Lynch, 20001. Even if school adminis- trators make rules to eliminate these signals (e.g., not allowing baseball hats to be worn on school grounds), gang members usually find a way to iden- tify themselves. Gang members are typically described by their peers as belonging to crowds characterized by drugs and violence, but a gang is qualitatively different from a crowd. Gangs are social organizations that have unique rules (e.g., about what clothes to wear) and norms (e.g., about what to do when someone teases you). These organizations can have a powerful influ- ence on adolescents' behavior, particularly by promoting criminal behav- ior, drug abuse, and other deviant behaviors (Bat/in-Pearson, Thornberry, Hawkins, and Krohn, 1998~. In addition to promoting antisocial as well as antiacademic behaviors, gang presence in schools can undermine the ability of all students to be engaged in academic work (Floras-Gonzalez, 2002; Howell, 1998~. Disrup- tions caused by fights, for example, are common, and occasionally nongang members are caught in the crossfire. National surveys have shown that students who report gang activity in their high schools also report that they do not fee! safe at school (Howell, 1998; Howell and Lynch, 2000), a feeling that is hardly conducive to a focus on academic work. Interventions, some implemented in school-based curricula, have been designed to reduce the problems associated with youth gangs, but their effectiveness is rarely evaluated systematically (Howell, 1998~. The focus of some interventions is on decreasing the number of students who join gangs. Other interventions focus on reducing the level of gang violence or encour- aging students to leave a gang and join more socially sanctioned peer groups. The committee speculates that an isolated program intervention ap- proach to preventing youth involvement in gangs is less likely to be success- ful than comprehensive changes in the school climate and instructional program. Adolescents' experiences in school are highly predictive of whether they join a gang. Youth who join gangs are typically low achievers in low- status crowds those least likely to experience high school as a personally supportive social context in which they fee! valued. In a prospective longi- tudinal study of 800 youth in high-crime neighborhood Hill, Lui, and Hawkins (2001) found that the highest risk factor (along with availability of marijuana in the neighborhood) was a learning disability. Students with learning disabilities were 3.6 times more likely to join a gang than were students without learning disabilities; students with low academic achieve-
FAMILY, COMMUNITY, AND PEERS 141 ment were 3.1 times more likely to join gangs compared to better achieving classmates. Not feeling attached (twice as likely) or committed to school (1.8 times as likely) were also strong predictors of joining a gang. These findings suggest that creating high schools that help students learn in a caring, respectful, and supportive social climate may be the best strategy for preventing gang involvement. Adolescents join gangs to meet social and psychological needs that are not being met elsewhere (Padilla, 19921. To the degree that schools can better meet these needs, their students will be less likely to try to meet them in potentially destructive peer groups. Friendship Groups Most adolescents report that they have a few best friends (Berndt, 1996, 19991. An adolescent's best friends are usually friends with each other; thus, they constitute a friendship group. Adolescents spend more time with the members of this close friendship group than with other peers. They also fee! closer and share more intimate information with their best friends than with other peers (Newcomb and Bagwell, 19951. For decades, researchers used estimates of the similarities between friends on specific characteristics, including academic achievement, as an index of how much friends influence one another (e.g., Ide, Parkerson, Haertel, and Walberg, 19811. Influence, however, cannot be assumed from similarity (e.g., Kandel, 1978, 19961. The old saying, "birds of a feather flock together," applies well to friendships (McPherson, Smith-Lovin, and Cook, 20011. Friends have similar characteristics partly because adoles- cents with similar characteristics become friends, not simply because ado- lescents who have become friends influence one another so that they be- . ·. come more similar. More recent longitudinal studies, however, provide credible evidence that adolescents are influenced by their best friends, more so than by any other peers. Berndt and Keefe (1995) used a longitudinal design to examine the influence of best friends on the school adjustment of junior high school students. In the fall of a school year, the students named their best friends and reported their positive involvement in school and their disruptive be- havior in the classroom. Their teachers also reported on their involvement, disruption, and report-card grades. Because most students listed friends who were also participating in the study, the students' scores could be matched with their friends' scores. These assessments were repeated in the spring of the same school year. Analyses showed that students' adjustment was fairly comparable in the fall and the spring. Most students who were high in involvement and academic achievement and low in disruption in the fall showed a comparable profile in the spring. Some students' adjustment changed, however, and the direction of the changes were predicted by the
42 ENGAGING SCHOOLS fall adjustment level of their friends. In particular, students who haci friends who were rated relatively high in the fall in involvement, disruption, or gracles increased on these dimensions from the fall to the spring. These data strongly suggest that the students' adjustment to school was influenced by the adjustment of their friends. Similar data from studies with diverse samples of high school students suggest that their positive engagement and disruptive behavior are influenced by the engagement and behavior of their friends (Epstein, 1983; Hallinan and Williams, 1990; McFarland, 2001; Steinberg, Brown, and Dornbusch, 19961. The longituclinal studies of friends' influence also imply that friends can have either a positive or a negative influence on high school students' engagement, clepencling on the friends' characteristics. In a variant of the usual longituclinal design, Epstein (1983) iclentifieci students in the elemen- tary through high school gracles who were relatively high or low in aca- clemic achievement and who haci friends who were relatively high or low in achievement. One year later, she assessed the students' achievement again. She found that students who initially were high in achievement but whose friends were low in achievement clecreaseci in achievement. Conversely, students who initially were low in achievement but whose friends were high in achievement increased in achievement. Most important, the positive in- fluence of high-achieving friends was roughly equal to the negative influ- ence of low-achieving friends. This large-scale stucly demonstrates that friends' influences on students' engagement in school can be as strong in the . . . . . . . . positive c erection as In t" be negative ~ erection. High school students tend to form friendships within their academic track or set of courses Makes, Gamoran, and Page, 19921. The negative effect of friends suggests the importance of avoiding organizational struc- tures and programs that create concentrated groups of acaclemically clisen- gageci students. Catterall's (1987) observation of a dropout prevention program that increased the dropout rate among program participants is consistent with this view. Attempts to explain this unexpected outcome revealed that students in the program haci many opportunities for social interaction, which leci to the formation of friendships among them. After the program encleci, some students clecicleci to drop out of school, and when they clici so, their friends usually ciroppeci out, too. After reviewing the results of many dropout prevention-intervention programs, Dishion, McCorci, and Poulin (1999, p. 762) conclucleci, "there is reason to be cautious and to avoid aggregating young high-risk acloles- cents into intervention groups." Unless attention is also paid to the friends' characteristics, these interventions may inadvertently strengthen the nega- tive influence of friends with negative characteristics. In particular, inter- ventions of this kind with urban high school students could strengthen
FAMILY, COMMUNITY, AND PEERS 143 negative attitudes toward school, tendencies to drop out of school, and other facets of school disengagement. Again, the research points to potential harmful effects of tracking and other policies that isolate students at risk of becoming disengaged and prevent them from having opportunities to interact and develop friendships with peers who are more committed to schooling. The strength of peer networks and youths' need for peer support, however, make it difficult to promote curricular mixing. In the Yonezawa et al. (2002) study of detracking discussed in Chapter 5, students of color resisted taking ad- vanced and honors courses with white students in part because they did not want to give up the supportive networks of peers they had developed in their lower-level courses. Apparently providing more student choice in the courses they take is not sufficient to achieve substantial mixing of students from different backgrounds and crowds. Concerted efforts need to be made to ensure that students who have previously not had access to a challenging curriculum are made to fee! psychologically safe and comfortable in such courses. CONCLUSIONS Schools can increase adolescents' engagement by harnessing the re- sources of families and the larger community. The evidence reviewed in this chapter suggests the value of efforts to improve communication and coordi- nation among adults in the various settings where adolescents spend their time including schools, homes, religious institutions, as well as the vari- ous organized extracurricular activities sponsored by schools and commu- nity groups. To the extent that adults in these settings collaborate to ensure that the various environments where adolescents interact with each other in the community are inclusive, informed by positive values, and supportive of the healthy development of young people, productive engagement in school- based learning is likely to be promoted (Cornell and Klem, 2000; Connell, Gambone, and Smith, 2000; National Research Council and Institute of Medicine, 20021. Both parent and peer influences can be positive or negative, and school policies and activities affect the direction of these influences. Parents are more likely to be involved, and thus to reinforce the messages from school, in schools that reach out and make proactive efforts to include them. Peer influences likewise do not occur independent of the behavior of adults or the organization of schools, community organizations, and other settings where youths interact. Adults can work to structure school environments that are affirming, supportive, and deliberately designed to make all stu- dents fee! that they are valued members of the school community. This can be achieved by heterogeneous grouping for classes, collaborative learning
44 ENGAGING SCHOOLS activities in and out of class that bring together students with different levels of achievement, activities that are based on shared interests rather than academic competence, and in many other ways. Careful attention needs to be given to how class assignments and activities promote student interactions. Because students will naturally seek others like themselves, self-conscious, proactive strategies to "mix" students are required.