Abstract
There is an accumulating body of data indicating that handicapped mainstreamed students are not socially acceptable to their nonhandicapped peers. However, this determination has typically been based on data derived from sociometric techniques and could therefore be artifactually derived from sociometric procedures. The current research compares three methods used to evaluate social ability: teacher ratings, sociometric ratings, and direct observation. These data reveal that handicapped children (a) are viewed as less socially acceptable by both teachers and peers but (b) do not differ from their nonhandicapped peers in actual amounts of positive and negative social interaction. Implications of those findings for reinterpretation of past research and future research direction are discussed.