For the past 15 years, work/study programs have been provided by many public secondary schools to mildly mentally retarded adolescents (IQ = 55-80). Though the merits of such an approach are generally acknowledged, one critical element has been conspicuously lacking: standardized test evaluation of retarded student achievement (Brolin, 1975). The problem in acquiring such measures is twofold: (1) existing standarized tests typically focus on academic skills rather than the daily living and prevocational skills that are emphasized in work/study programs; and (2) retarded students have reduced ability to respond appropriately to paper and pencil tests because of reading and memory skill deficits (Prehm, 1968). Some recent research and development efforts have focused on assessment that is psychometrically adequate and contentand format-appropriate for retarded adolescents. The Social and Prevocational Information Battery (SPIB) (Halpern, Raffeld, Irvin, & Link, 1975b) has been developed using competencies stressed in training programs as guides for item content. The SPIB consists of nine tests designed to measure knowledge in the areas of job search skills, job related behavior, banking, budgeting, purchasing, home management, physical health care, personal hygiene, and functional signs (survival reading). Most of the SPIB items are presented in a true/false format, and each of the nine tests is orally administered in a group setting. Because reading skills are quite variable within groups of retarded adolescents, oral administration is used in order to minimize the impact of reading ability. Research evidence concerning both the reliability and validity of the SPIB is extremely encouraging (Halpern, Raffeld, Irvin, & Link, 1975a; Irvin & Halpern, 1977). Even though the SPIB is certainly adequate on a number of important dimensions of content and measurement, it would be premature to adopt the true/false SPIB model as a paradigm for the construction of all future standardized tests to be used in measuring the school achievement of retarded adolescents. During the SPIB's development, Prehm's (1968) and other evidence led to the presumption that a verbally presented multiplechoice item would be too difficult for retarded students, due to the large amount of information that would need to be remembered and analyzed in order for the examinee to choose and mark the correct symbol on an answer sheet. This presumption should be empirically examined, since it is possible to construct multiple-choice items with short stems and short alternatives, which will reduce the effects of limited auditory retention on test performance. Such an effort would appear to have substantial merit, given the
Read full abstract