Alternatives to Labeling A number of alternative approaches to classifying exceptional children that focus on educationally relevant variables have been proposed over the years (e.g., Adelman, 1996; Hardman, McDonnell, & Welch, 1997; Iscoe & Payne, 1972; Sontag, Sailor, & Smith, 1977). For example, Reynolds, Zetlin, and Heistad (1996) proposed a system they call “20/20 analysis” as an alternative, nonlabeling approach to the traditional, categorically driven model of special education. The lowest-achieving 20% and the highest-achieving 20% of students would be identified and eligible for broad (noncategorical) approaches to improvement of learning opportunities. In 20/20 Analysis, we begin with measuring progress of students in important areas of learning and identifying those at the margins—those who are not learning well and those showing top rates of learning. . . . At all times the focus is on outcome variables. . . . The idea is to look to the margins...