Advances in Cognition, Education, and Deafness

Analysis

Analysis Linda Delk Educators of deaf children frequently ask, "Are available tests valid for use with deaf students?" Recent developments in psychometric theory and in .the political and social arenas have stimulated renewed interest in this question. As more deaf students are being educated in public school programs, concerns relate to decisions about appropriate placement as well as about the use of minimum competency testing by states to certify students for high school graduation. Issues relating to the assessment of handicapped populations have arisen in response to the requirement for fair testing practices imposed in P.L. 94-142 and in Section 504 of the Rehabilitation Act. At the same time, test validity is receiving renewed attention in the general testing community. The reconceptualization of test validity as a unified construct grows out of the need for clearer understanding of tested constructs, the interpretation of test results, and the consequences of their use. The three-part classification of validity into content, criterion, and construct has given way to the notion of test validation as a continuous process. It is no longer sufficient to think of validity as a property of a test that can be established once and for all by a correlation coefficient..Construct validity has become the central focus and a means of linking test concepts, evidence, social and personal consequences, and values (Cronbach 1988). This unified conception of validity involves "an integrative evaluative judgment of the degree to which empirical evidence and theoretical rationales support the adequacy and appropriateness of inferences and actions based on test scores or other modes of assessment" (Messick 1989). This unified concept of validity provides a useful framework for examining the papers in this chapter. Berkeley explores the need for an integrated, qualitative 96 Analysis 97 approach to assessment of infants and toddlers and discusses assessment in light of its many possible uses and consequences. Aplin examines the adequacy and appropriateness of using a battery of tests to diagnose learning difficulties among deaf students and presents empirical evidence to challenge Van Uden's construct of dyspraxia. Croft et al. examine the utility of a test of study skills for predicting the academic success of deaf college students. Braden focuses on the theoretical limitations of existing research for improving understanding of the construct of intelligence as it applies to deaf persons. At a different level, Rush explores the IQ test profiles of atypical deaf learners. Clark explores the different strategies that deaf and hearing adults may use in visual information-processing tasks. And Sharp-Pucci examines evidence to support selection of a causal theory to explain the simultaneous decline of hearing and cognitive functioning among elderly persons. Berkeley describes a theoretical framework for observing and documenting the dimensions, or salient attributes, of children's responses to the environment. Application of the salient-responses approach may ensure that behaviors at any level of complexity are observed in context, especially when the salient attributes such as utilization and generalization are specified. The reader is left to assume, however, that the choice of theoretical orientation to development will guide the sampling of behaviors to be assessed using the salient attributes of responses framework. While an integrated view of development is desirable, especially for young children, the very nature of the salient-responses approach seems to require that some behavior at some level, such as play, communication, or walking, must be specified. The act of choosing some behaviors to observe necessitates paying less attention to others. If more than one behavior is observed using salient responses, one of the tasks of assessment still remains-that of integrating information across behaviors. Berkeley could help bridge the gap between theory and practice by providing more examples of how the salient-attributes approach could be applied to assessment, intervention, and program evaluation. Aplin's study of the identification of learning difficulties in deaf children begins by questioning Van Uden's assertion that dyspraxia can be identified using a particular battery of tests and that this identification should be used to recommend the type of educational program which deaf children should receive. By using a large, heterogeneous sample of deaf students in public school programs and combining factor analysis with qualitative methods of profile analysis, Aplin concluded that Van Uden's dyspraxia test battery is not adequate for identifying young deaf children for purposes of placement. Aplin's caution speaks to the social and personal consequences of test use. More studies like Aplin's are needed to improve decisions, not only...