Two studies have used more specific criteria to determine the existence of friendships between children with DD and their typically developing peers. Freeman and Kasari (2002) established systematic standards to determine the existence of friendships for children with Down syndrome, based on behaviours observed between these children and peers in structured play sessions. They reported that at least 30% of all dyads observed did not meet the established criteria for friendship. These data, however, did not necessarily focus on friendships established in school, and were based on very brief observations in an artificially created setting. By contrast, Harry et al. (1998) found that individual features (such as reciprocity, liking, affection, and having fun), identified by Bukowski et al. (1996) as important in the friendships of typically developing children, were also present in the relationship of two girls with disabilities attending a specialised class in an elementary school. This research is one of the few studies that have attempted to describe the friend- ship between two children with DD. However the researchers made their evaluation based on sub- jective judgement, rather than on a systematic measurement of the observed features, and only documented a single relationship.