a unifying theoretical foundation. A consequence of this, according to Thompson et al., is the inability to develop precise measures of adaptive behavior that would objectively differentiate individuals by disability. An alternative explanation is that adaptive behavior must be understood in the context of the individual’s relevant daily and social life, which is determined by age, culture, and context (Thompson et al., 1999).
Independence of Domains. The 1992 AAMR definition requires that an individual show significant limitations in at least 2 of the 10 adaptive skill areas. A danger of accepting “erroneous domains that are not truly distinct from one another” (Thompson et al., 1999, p. 17) is that it can lead to the inconsistent application of eligibility criteria and unequal treatment across groups of people. Thus, characteristics of the factor structure of a measure of adaptive behavior have important implications for diagnosis.
Thompson et al. (1999) reviewed studies that reported factor analyses of adaptive behavior measures. They made two important points before summarizing their findings: (1) highly correlated factors may indicate that they do not represent independent dimensions and (2) different methods of factor analysis can support different factor structures.
Greenspan (1999) noted that a drawback to the factor analytic approach to determining the dimensional structure of adaptive behavior is that this statistical method cannot determine whether some domains do not make conceptual sense (i.e., items should not have been included on tests in the first place) or whether missing content domains should have been included.
Social Skills Dimension of Social Competence. Most adaptive behavior scales contain factors addressing interpersonal relationships or social skills, but they do not address overall social competence. For indi-