Search

With the number of new and revised nonverbal and language-reduced tests of cognitive ability, selection and interpretation of appropriate measures can be complicated. Seven nonverbal or language-reduced tests with normative data collected within the last 15 years were evaluated. Besides evaluating technical adequacy, other variables affecting test selection and interpretation including adequacy of floors and item gradients, provision of data for subgroups of students, percent of timed items, and response mode are described. Eight additional tests with language-reduced components also were reviewed. Implications for practice and research are presented.

Three studies examine the validity of the Preschool Form of the Cognitive Abilities Scale—Second Edition (CAS-2). Significant high concurrent criterion-related validity correlations, corrected for restricted range, are found between the CAS-2 and the Detroit Test of Learning Ability—Primary: Third Edition for 26 three-year-olds (rc= .86) and the Bayley Scales of Infant Development—Second Edition for 20 two-year-olds (rc= .63). Significant correlations also are evident for the CAS-2 and the Test of Early Language Development—Third Edition for 26 three-year-olds (rc= .77) and the Adaptive Behavior Assessment System—Second Edition for 20 two-year-olds (rc= .65). Predictive power is critical when making interpretations regarding future performance. A high, significant correlation (rc= .72) is observed for the CAS-2 for 37 children at ages 2 and 3 and the Wechsler Intelligence Scale for Children—Third Edition administered 6 years later. Results expand validity evidence for the CAS-2, increasing confidence in interpretations regarding young children's performance.

Autism is a low-incidence disability that is complex to assess and for which rates continue to increase. Assessment options for autism are reviewed and presented in the context of recent research and a comprehensive, multidisciplinary assessment model. The model involves three levels that yield data progressing from more subjective and general to more objective and specific. Each level provides unique information critical to verifying eligibility, planning instruction, and monitoring progress. Because of the complexity of autism, a systematic and comprehensive assessment approach is critical to reducing error in decisions regarding eligibility and treatment.

Ignoring the adequacy of floors and item gradients for tests used with young children can have serious consequences. Thus, because of the importance of early intervention for reading and math problems, we used the criteria suggested by Bracken for adequate floors and item gradients, and reviewed 15 reading tests and 12 math tests for ages 4-0 through 7-11. Results indicate that few of these tests have adequate floors at the age at which their norms begin. Although most of the reading tests have adequate item gradients, only 4 of the 12 math tests did not have item gradient problems. If test floors and item gradients are not considered, examiners risk misdiagnosing young children and drawing invalid conclusions for research and program evaluation, especially for low-performing children.

The construct stability of the Infant Form of the Cognitive Abilities Scale-Second Edition (CAS-2) was examined with 54 children who were assessed when they were between 3 and 23 months old. They were assessed again with the Preschool Form of CAS-2 when they were 2- or 3-year-olds. The two forms have no overlapping items and employ different formats. The correlation for children assessed initially when less than 1 year of age was .28 (.46 corrected for restricted range). For those assessed initially at 12 to 23 months, a correlation of .65 (.76 corrected) was obtained. Results for the majority of individuals remained stable over time. Implications for practice and suggestions for future research are discussed.

AbstractPurposes of the study were to determine the type and format of information pediatricians receive from schools regarding diagnosis and treatment of Attention Deficit-Hyperactivity Disorder (ADHD), their preferences for this information, and their opinions on referral rates and treatment. Results from a random sample of members of the American Academy of Pediatrics indicate that for both diagnostic and treatment information, significant discrepancies exist between what pediatricians prefer and the information schools always or usually provide. The majority of pediatricians believe schools overrefer and over one-third believe parents overrefer for ADHD. Pediatricians recommend behavior management at home and school more often than stimulant medication. To improve referral, diagnosis, and treatment, discussion of discrepancies in expectations between school psychologists and pediatricians could be helpful. Additional training in some areas for both groups also may be beneficial.

Current norm-referenced cognitive measures (the Battelle Developmental Inventory, Bayley Scales of Infant Development-Second Edition, Cognitive Abilities Scale, Leiter International Performance Scale-Revised, Mullen Scales of Early Learning, and Stanford-Binet-Fourth Edition) were reviewed and evaluated in terms of their use with children from birth through 2 years of age. Also, relevant literature on issues critical to appropriate use of the tests is discussed. Careful evaluation of the strengths and problems of each test is critical to accurate diagnosis and evaluation of children's progress. To aid psychologists responsible for assessing the youngest children, variables relevant to test selection and interpretation are discussed.

Two studies were conducted to evaluate the longterm predictive validity of the Cognitive Ability Scales (CAS), a test designed specifically for 2and 3-year-old children. The ability of the CAS to predict intelligence and reading and math achievement was evaluated with two samples of children (N = 20 and N = 30) enrolled in grades one through five. An interval of approximately 5 years was used. The CAS predicted performance on the Binet-IV and the WISC-R with correlations in the moderate range. Correlations of CAS Math results with the Test of Early Mathematical Ability and the K-TEA Math were also in the moderate range. Correlations of CAS Reading results with the Test of Early Reading Ability-2 and the K-TEA Reading, however, were in the low to moderate range. Thus, adequate long-term predictive validity for the CAS was demonstrated for intelligence and math performance.

In Study 1, the effect of making tokens contingent on correct performance of low social position preschoolers on the McCarthy Scales of Children's Abilitieswas examined. Preschoolers in a token reinforcement group scored significantly higher (mean=8 IQ points) than subjects in a control group. In Study 2, the effect of tokens on McCarthyresults was examined as a function of social position in a 2×2 design. The high social position control group scored significantly above the low social position control. The low social position token reinforcement group, however, performed as well as both the high social position control and token reinforcement groups. Use of systematic reinforcement contingencies to reduce test error is discussed.

Thirteen American Indian students in the fifth through ninth grades were administered the WISC-R under standardized conditions; another twelve were tested under standardized procedures, but received token reinforcement immediately following each correct response. Tokens were exchangeable for a variety of back-up rewards. Performance IQ scores were significantly higher than verbal IQ scores for both groups. The token reinforcement group scored significantly higher (12 points) than the control group on the full scale IQ. Implications for research and practice are discussed.

AbstractThe reliability and validity of curriculum-based measurement (CBM) for Braille readers was evaluated. Passage sample lengths of 1-, 2-, and 3-minutes were examined for 15 students in Grades 3 through 7. Internal consistency coefficients ranged from.97 to.99, alternate-form reliability coefficients from.88 to.98, and percentage of agreement between independent examiners ranged from.97 to 98%. Test-retest reliability coefficients with a 14-to 16-day retest interval ranged from.95 to.98. Criterion-related validity coefficients with the Diagnostic Reading Scales(Spache, 1981) ranged from.60 to.82. Overall teacher ratings of students' Braille reading performance also correlated significantly with CBM results ranging from.80 to.85. A 2-minute passage length was found to be reliable, valid, and efficient with assessing and monitoring progress of Braille readers. The findings from this study were very similar to the data on technical adequacy for CBM when used with sighted readers.

Ratings by 23 parents and 21 regular education teachers on the Social Skills Rating Scale (SSRS) for elementary-age students who were blind were compared to ratings for the SSRS sighted norm group. No significant differences were found in the overall ratings of social skills or on the Self-Control or Responsibility subscales, but several important differences were noted on the Assertion and Cooperation subscales. Students who were blind were rated as less academically competent and higher on problem behaviors than the sighted norm group, even though stereotypical behaviors were not assessed by the SSRS. Implications for instruction are discussed.

The level of object-permanence was assessed for 16 infants (age, 7 months) when three types of stimuli were used. A familiar object (a toy brought from home), novel objects, and a significant object (baby's bottle) were used to examine how each object type would affect the infants' levels of search. Results indicated that when presented with a novel object, 7-month-old infants, as a group, would exhibit significantly higher levels of search behavior than they would for either the significant or the familiar stimuli. The babies' bottle elicited significantly higher levels of search from all infants than did a familiar object. In three cases it was observed that the significant object elicited a higher level of search than did novel objects. Definitions of novelty and significance are examined. Implications of the findings in terms of assessment are discussed.

Abstract:The survey reported on in this paper replicated and modified Levine's (1974) nationwide survey of the psychological services provided for deaf and hard of hearing students. The purpose of our survey was to up date the information in light of the requirements of PL 94-142 (1975) and current practice. We noted several changes in practice since the first survey, but unfortunately, little change has taken place in many important areas. We suggest possible areas for expanding the role of school psychologists and discuss implications for training.