With the recent proliferation of programs designed for at-risk students, it becomes essential to review the available literature in order to define the students we focus upon and determine which practices have proved most effective. Although some published reports are either anecdotal or flawed by poor design (Cox, Davidson, & Bynum, 1995), consensus seems to be emerging regarding best practices for at-risk students.

Definitions of at-risk students

But we must first ask who is at-risk? And at-risk of what? Many organizational reports and even some scientific studies simply assume a definition without articulating it. Even among the available definitions, ambiguity is often the defining characteristic. Thus we find Jan Murdock's understanding, presumably a reflection of the Texas Education Agency's (TEA) official position, that at-risk children have a "greater likelihood of becoming [educationally] disabled because of conditions surrounding their births or home environments" (1999, p. 318). Richard Ashcroft (1999) writes of students "at risk for becoming delinquent." Richard Sagor (1999) cites "a mismatch between learner and learning system." And Susan Bickford (2001) equates at-riskness with violence. Indeed, all of these definitions have legitimacy.

However, based on the preponderance of available research, we can safely say that at riskness is most frequently manifested by poor academic and social skills that promote a general disconnection with the school culture. And we can further note that quite a bit of educational effort and funds are invested in remediating these factors.

Academic failure

Poor academic performance is a clear indication of "at-riskness" (Guerin & Denti, 1999, Karlsson, 1996; Murdock, 1999; Sagor, 1999). The concept seems to be both cyclical and progressive, in that academic failure increases the likelihood that a student will labeled at-risk of the ultimate academic failure: dropping out.

Indeed, one of the prime manifestations of academic failure, grade retention, is identified by Jay Smink (2000), Center Director of the National Dropout Prevention Newsletter, as a cause of dropping out. The research supports him. Slavin and Madden (1989, p. 4) define grade retention as among the "least effective" strategies for dealing with this population and note that "failing more students does have a misleading short-term effect on test percentiles or normal curve equivalents because the students are a year older when they take the tests. However, the long-term effects on student achievement are most often negative." Holmes and Matthews' meta-analytic study (1984, p. 232) concludes that "the potential [of grade retention] for negative effects consistently outweighs positive outcomes." And a more recent meta-analysis reveals a general failure "to demonstrate that grade retention provides greater benefits to students with academic or adjustment difficulties than does promotion to the next grade" (Jimmerson, 2001, Summary).

But even the concept of poor academic performance is not transparent. Mainstream views (Guerin and Denti, 1999; Karlsson, 1996) hold that academic deficiencies are primarily literacy-related. However, some are beginning to challenge that view. David O'Brien (2000, para. 4) argues that at-risk "adolescents are capable and literate if we view them from the perspective of multiliteracies in new times." And R. Neal Shambaugh (2000, Defining Literacy: A Personal Path)argues that "literacy includes personalized and sociocultural ways of reading the world for meaning." This post-modernist suspicion of the traditional dominance of print-literacy suggests that at-riskness is a cultural construct rather than an objective reality.

Disengagement

The second most commonly cited definition of at-riskness, disengagement from school, is also often considered a self-explanatory concept. A typical example is found in Duke and Griesdorn (1999, para. 2) where at-risk youth are identified as "students who just seem to get lost in large secondary schools." However, Husted and Cavalluzzo (2001, pp. 2-3) provide some insight into the syndrome by including disengagement in a constellation of related attitudes. Thus at-risk students are "disengaged from their high school, underachievers, unmotivated, and/or socially isolated disengaged, unmotivated, or otherwise unhappy in the traditional high school environment." Guerin and Denti (1999, para. 3) also provide insight by locating "alienation from school" among factors such as "low self-esteem, limited language proficiency and lack of positive adult role models." And Susan Bickford (2001, p. 3) locates disengagement between "disruption in the classroom" and "youth violence." Disengagement, then, seems to relate to some sort of socio-emotional failure to participate in the school culture.

Deviance

Although most often associated with criminal behavior, deviance more generically refers to any behavior that deviates from mainstream norms. We have already noted Sagor's description of "mismatch" between learner and environment (1999). Most researchers take a more definite stance and identify as a defining characteristic such as behavioral difficulties (Slavin & Madden, 1989; Guerin & Denti, 1999), social isolation manifested by emotional problems and low self-esteem (Duke and Griesdorn, 1999; Guerin and Denti, 1999;. Husted and Cavalluzzo, 2001; Karlsson, 1996), anti-social peer groups (Slavin & Madden, 1989), criminal activity (Bickford, 2001), and low socioeconomic status (Gueirin and Denti, 1999; Husted and Cavalluzzo, 2001; Karlsson, 1996.) Interestingly, Slavin and Madden (1989) actually identify schools with a preponderance of low SES children as a risk-factor in and of itself.

American educational history is rife with examples of alternative educational systems such as private, religious, and home schools. But today's alternative education programs, which 1% of America's schoolchildren (Education Commission of the States, 1999) focus on at-risk children. This review of the literature shall assume the latter definition in the following discussion.

There seems to be no consensus on how alternative programs should be structured. Descriptions include career academies, charter schools, college-based alternative schools, continuation schools, correctional facilities, group homes, hospital classes, juvenile court schools, magnet schools, opportunity schools, residential schools, schools-within-a-school, schools without walls, second-chance schools, separate alternative learning centers, summer schools, residential substance treatment programs, tech-prep schools, and youth camps.

Purpose

The earliest alternative programs seem to have focused on offering alternative learning opportunities to students who did not function well in the mainstream environment (Gregg, 1999). These functioned under the philosophy that "alternative education is a perspective, not a procedure or program. It is based on the belief that there are many ways to become educated, as well as many types of environments and structures within which this may occur" (Morley, 1991). However, alternative programs were soon recognized as appropriate placement for disruptive students. While these two purposes ­ to create alternative learning environments and to remove disruptive students from the traditional learning environment -­ may at first glance seem complementary, further examination reveals a serious conflict. The Southwest Educational Development Laboratory speculates that students who are required to attend such programs may "include a less motivated, and perhaps more at-risk, population" and thus mandatory alternative placement "may jeopardize the culture of the alternative education program" (Implications of Mandates).

Community

And just what is that culture? Or rather, what should it be? The Southwest Educational Development Laboratory and the National Dropout Prevention Center/Network recommend the intentional creation of small communities characterized by student ownership and involvement and by "warm, caring relationships with teachers [and] among students in order to create a supportive school culture" (Successful Program Characteristics). The available research supports this focus (Guerin & Denti, 1999; Husted & Cavalluzzo, 2001; Krovetz, 1999; Leone & Drakeford, 1999; Soleil, 1999).

The National Governors Association Center for Best Practices, the Education Commission of the States, and the National Dropout Prevention Center/Network advocate another approach, recommending community and family involvement in the school as well as student involvement in the community, particularly through service or work-based learning programs. Once again, this recommendation is supported by research (Arnstine & Futernick, 1999; Guerin & Denti, 1999; Jennings, 2001; Meyers, 1999; Schine, 1997; Soleil, 1999; Waldstein & Reiher, 2001; Walker & Golly, 1999).

Behavioral Expectations

Such strong support for interrelationships among students and community speaks to the understanding of at-risk youth as disengaged from the school culture and outside mainstream social norms. If part of the at-risk student's problems stems from an inability to maintain appropriate social connections, then programs which improve social skills, such as service-learning, and maintain high behavioral expectations should prove effective. And indeed, research suggests that they are.

In his beautifully articulated discussion of problems besetting at-risk students, Martin L Krovetz (1999, Nonresilient School Culture, Helping Students Stay Connected) explains that "many alienated students feel hatred and shame when in contact with the 'in' group of a school from whom they felt much disrespect." Kavale & Forness (Author Abstract) speak of "the negative influence of social skill deficits." And Walker & Golly (1999) note that at-risk children "have not been taught basic manners and social conventions [and] often have very atypical views about standards governing appropriate behaviors" (Risk Factors in the Formative Years).

It is thus not surprising to find that alternative programs feature strict discipline, activities that build self-esteem and social skills, and "a positive peer culture" or "a caring, disciplined community that honors civility, service, and diversity and that fosters moral and ethical attitudes and behavior" (Guerin and Denti, 1999, Behavior). Duke and Griesdorn (1999) agree that the social/behavioral dimension is critical enough to be one of the major continuums on which alternative educational programs should be judged.

This is especially interesting in light of a significant body of research that suggests that social skills training has little effect on student behavior. In their 1996 meta-analysis of the available research, Forness and Kavale conclude that a mean effect size of .211 indicates that current methods of addressing social skill deficits are uniformly ineffective. However, the studies analyzed relied on teacher, peer, and self-evaluations of effectiveness. These may not be the best way of measuring behavioral changes.

In their 1997 meta-analysis of "interventions to decrease disruptive classroom behavior," Stage and Quiroz found "that studies using behavioral observation of classroom disruptive behavior were far more likely to evidence a significant decrease in disruptive classroom behavior than studies relying on teacher rating scales" (Effect Size by Design and Instrument). When they omitted studies that relied on subjective measures and reanalyzed the results, these researchers found that all methods had high effect sizes, some close to an ES of 1. Further, the techniques analyzed were effective with all student populations, including regular education and Emotionally Disturbed.

Stage and Quiroz suggest that one cause of the discrepancy between objective and teacher evaluations of student behavior may involve the "halo effect" or the disposition to perceive what we expect. Thus, behavior problems may create perceptions of problem behavior. Walker & Golly (1999, Risk Factors) agree: "Even when such children improve their behavior the reputations they have acquired among their peers and teachers sometimes create a barrier to adults' seeing and accepting the positive changes that have actually occurred." Poor social skills may even create perceptions of disability. In their reanalysis and repositioning of Learning Disabled students, Spear-Swerling and Sternberg (1996, p. 7) note that "a youngster who is a behavior problem in the classroom, and who also has reading difficulties" is significantly more likely to eventually be labeled Learning Disabled than is a "quiet and compliant" student who displays similar difficulties.

So it is not surprising that one of the features of successful alternative education programs identified by the National Dropout Prevention Center/Network involves the clear articulation and consistent maintenance of high behavioral expectations, nor that significant research supports this stance (Denti & Guerin, 1999; Guerin & Denti, 1999; Soleil, 1999Walker & Golly, 1999). By displaying social skills such as respect for students and their interests and learning styles (Arnstein & Futernick, 1999; Cox, Davidson, & Bynum, 1995; Husted, Denti & Guerin, 1999; Guerin & Denti, 1999; Krovetz, 1999; Leone & Drakefield, 1999; Secada, 1999; and Soleil, 1999) effective alternative educational programs allow students to develop the more socially acceptable behaviors that promote academic achievement.

Educational Expectations

Research indicates that high academic expectations are hallmarks of successful alternative programs. Walter Secada (1999) notes that ineffective programs focus on basic skill remediation and vocational training. Slavin and Madden (1989, p. 5) argue that "at best," pull-out remediation programs "may keep at-risk students from falling further behind their agemates, but even this effect is limited to the early grades."

Remediation doesn't work. What does are high academic standards or expectations. Leone and Drakefiled (1999, p. 88) cite "a challenging curriculum" and a "variety of courses" among the hallmarks of successful programs. Duke and Griesdorn (1999) argue that successful schools "shift the focus.. from constant rule enforcement to academic progress" (How Should Discipline be Handled?).

Constructivist techniques also seem to be effective strategies. site Karlsson (1996) suggests that a variety of constructivist strategies are effective. Peer tutoring, cooperative grouping, service learning, answering questions with questions, and a strong focus on student interests are among the hallmarks of successful programs (Denti and Guerin, 1999; Guerin and Denti, 1999; Leone and Drakeford, 1999; Secada, 1999, and Slavin & Madden, 1989).

Caring Adults

A last important feature of successful alternative education involves the choice of staff. Although "high risk youth are among the most difficult students to teach," most alternative education teachers have received little pre-service training "that addresses their students' legal, social and psychological problems"; they tend to be essentially self-educated in this area (Ashcroft, 1999, Aspects of Preservice Preparation, Teacher Survey). And yet, researchers select high levels of teacher training as among the most important components of effective strategies for teaching at-risk youth. Guerin and Denti (1999) stress teacher competence in academic and behavioral instructional strategies. Duke, Griesdorn, and Kraft (1998) suggest that "all teachers in alternative schools should be trained in special education, reading remediation, behavior management strategies, and counseling skills." And Duke and Griesdorn (1999, How should discipline be handled?) recommend that every alternative school should "employ a reading specialist, learning disabilities specialist, [and] guidance counselor."

Site based management; teacher autonomy and control, particularly in regard to curriculum; and continuing professional development in areas specifically related to this special population and setting are also keys to successful programs (Ashcroft, 1999; Denti & Guerin, 1999; Guerin & Denti, 1999; Husted & Cavalluzzo, 2001; Krovetz, 1999; Leone & Drakefield, 1999; Secada, 1999, Soleil, 1999) Many successful programs also feature small classrooms that allow teachers to focus on students and forge strong, supportive relationships (Denti & Guerin, 1999; Gregg, 1999; Husted & Cavalluzzo, 2001, Krovetz, 1999; Secada, 1999).

But why should we limit these best practices to a small percentage of the population? Is it necessary to wait until the hallmark features of academic failure and behavioral dysfunction force students out of the mainstream and into an alternative system? The Southwest Educational Research Laboratory (1995, Conclusion) notes that "traditional schools do not effectively serve many (if not most) students" [and that] "many of the features and goals associated with successful programs would benefit all students." That sentiment is echoed by the National Dropout Prevention Center/Network: "Wouldn't it be wonderful if each traditional school could share these characteristics and operate with the best practices?" Such a shift may be already occurring. The National Governors Association Center for Best Practices (2000, para. 4) notes "a major cultural shift from targeting only at-risk youth to targeting all youth." By directly addressing the core needs of children who fail to thrive in traditional school settings, alternative educational settings may one day prove to be the vanguard of academic reform in the regular education classroom.

Bickford, S. (2001) Alternative education for at-risk youth: An analysis of specific legislation from 1995-1999 in Pennsylvania. U.S. Department of Education. Retrieved on July 11, 2002 from ERIC database.

Cox, S. M., Davidson, W. S., & Bynum, T. S. (1995). A meta-analytic assessment of delinquency-related outcomes of alternative education programs. Crime & Delinquency, fa:2, pp. 219-234.

Denti, L. & Guerin, G. (1999). Issues in Alternative Education. Clearing House, 73:2, pp. 71-72. Retrieved July 11, 2002, from the EBSCOhost database.

Duke, D. J. & Griesdorn, J. (1999). Considerations in the design of alternative schools. Clearing House, 73:2, pp. 89 ­ 92. Retrieved July 11, 2002, from the EBSCOhost database.

Education Commission of the States (1999). Service-learning: An education strategy for preventing school violence. ECS issue paper: Service-Learning/Community Service.

Forness, S. R. & Kavale, K. A. (1996). Treating social skill deficits in children with learning disabilities: A meta-analysis of the research. Learning Disability Quarterly, 19, p. 2-13.

Gregg, S. (1999). Creating effective alternatives for disruptive students. Clearing House, 73:2, pp. 107-113. Retrieved July 27, 2002, from the Wilson Web database.

Guerin, G. & Denti, L. (1999). Alternative education support for youth at-risk. Clearing House, 73:2, pp. 76-78. Retrieved July 11, 2002, from the EBSCOhost database.

Holmes, C. T. & Matthews, K.. M. (1984). The effects of nonpormotion on elementary and junior high school pupils: A meta-analysis. Review of Educational Research, 54, p. 225-36. Retrieved on July 11, 2002 from Wilson Web database.

Husted, T. A. & Cavalluzzo, L. C. (2001). Background paper for New Collaborative Schools (NCS): An overview of at-risk high school students and education programs designed to meet their needs. Alexandria, VA: The CNA Corporation.

Jennings, M. (2001). Two very special service-learning projects. Phi Delta Kappan, 82:6, pp. 474-5.

Jimerson, S. R. (2001). Meta-analysis of grade retention research: Implications for practice in the 21st century. The School Psychology Review, 30:3, pp. 420-437. Retrieved July 11, 2002 from Wilson Web database.

Karlsson, M. R. (1996). Motivating At-Risk Learners. Westminster, CA: Teacher Created Materials.

Krovetz, M. L., (1999). Resiliency: A key element for supporting youth at-risk." Clearing House, 73:2, pp. 121-123. Retrieved July 27, 2002, from the Wilson Web database.

Leone, P. E. & Drakeford, W. (1999). Alternative education: From a 'last chance' to a proactive model. Clearing House, 73:2, pp. 86-88. Retrieved July 11, 2002, from the EBSCOhost database.

Meyers, S. (1999). Service learning in alternative education settings. The Clearning House, 73:2, pp. 114-117. Retrieved July 11, 2002, from the EBSCOhost database.

Morley, R. E. (1991). Alternative education. In Southwest Educational Development Laboratory (1995). Alternative Learning Environments. Insightson education policy and practice, 6. Retrieved on July 25, 2002 from Wilson Web database.

Murdock, J. (1999). Special education in Texas. In Funkhouser, C. W. (Ed.),Education in Texas: Policies, practices, and perspectives (pp. 317 ­ 324).Upper Saddle River, NJ: Merrill.

National Dropout Prevention Center/Network (n.d.). Retrieved on July 1 and July 25, 2002 from http://www.dropoutprevention.org

O'Brian, D. (2000). "At-risk" adolescents: Redefining competence through the multiliteracies of intermediality, visual arts, and representation. International Reading Association. Retrieved July 10, 2002 from http://www.reading.org/focus/adolescent_articles.htm

Sagor, R. (1999). Equity and excellence in public schools: The role of the alternative school. Clearing House, 73:2, 72-75. Retrieved July 11, 2002, from the EBSCOhost database.

Schine, J. (1997). School-based service: Reconnecting schools, communities, and youth at the margin. Theory into Practice, 36, pp. 170-175.

Secada, W. G. (1999). Lessons learned by the Hispanic dropout project. Clearing House, 73:2, pp. 93-95. Retrieved July 11, 2002, from the EBSCOhost database.

Shambaugh, R. N. (2000). What does it mean to be X literate? Literacy definitions as tools for growth. International Reading Association. Retrieved July 10, 2002 from http://www.reading.org/focus/adolescent_articles.htm

Slavin, R. E. & Madden, N. E. (1989). What works for students at risk: A research synthesis. Educational Leadership, 46, 4-13.

Smink, J. (2000). Center makes policy statement on grade retention. National Dropout Prevention Newsletter, 13:2, p. 1.

Soleil, G. (1999). Creating effective alternatives for disruptive students. Clearing House, 73:2, pp. 107-113. Retrieved July 27, 2002, from the Wilson Web database.

Southwest Educational Development Laboratory (1995). Alternative Learning Environments. Insightson education policy and practice, 6. Retrieved on July 25, 2002 from Wilson Web database.

Spear-Swerling, L. & Sternberg, R. J. (1996). Off track: When poor readers become "learning disabled." Boulder, CO: Westview Press.

Stage, S. A. & Quiroz, D. R. (1997). A meta-analysis of interventions to decrease disruptive classroom behavior in public education settings. The School Psychology Review, 26:3, p. 333-368.

Texas Education Agency (1998). Student performance. In Funkhouser, C. W. (Ed.),Education in Texas: Policies, practices, and perspectives (pp. 299 - 316).Upper Saddle River, NJ: Merrill.

Waldstein, F. A. & REiher, T. C. (2001). Service-learning and students' personal and civic development. Journal of Experiential Education, 24:1, pp. 77-13.

Walker, H. M. & Golly, A. (1999). Developing behavioral alternatives for antisocial children at the point of school entry. Clearing House, 73:2, pp. 104-106. Retrieved July 27, 2002, from the Wilson Web database.