No abstract

Although sex differences in cognitive abilities have been reported in the experimental literature, there has been a paucity of information about how men and women might perform differently on clinical neuropsychological tests. The present study examines sex differences in verbal learning. Sixtyeight men and 68 women of equal age and education were administered the California Verbal Learning Test. Dependent variables included measures of recall, recognition, learning characteristics, and error types. Women displayed consistently higher levels of immedite and delayed free recall and made greater use of semantic clustering. There were no sex differences on recognition testing or error types. Results suggest that women's superior recall is attributable to better retrieval, which, in turn, is related to their greater use of verbally mediated strategies. This study highlights the importance of investigating sex differences on clinical instruments and reporting separate sex norms when appropriate.

To be able to evaluate the possible diagnostic significance of the magnitude of subtest‐to‐subtest scatter on the Wechsler Adult Intelligence Scale—Revised (WAIS‐R), it is important to know the frequency of various levels of scatter in the general population. To provide this information, the magnitude of scatter across all 11 subtests, and across the subtests of the Verbal and Performance scales, was computed for each of the 1,880 individuals used in the WAIS‐R standardization. The range (difference between highest and lowest subtest scaled scores) was found to be an effective measure of scatter when compared with other, more complex indices. Scaled score scatter, which ranged from 2 to 16 points on the Full Scale, averaged 6.7 points for the Full Scale and 4.7 points each for the Verbal and Performance scales. Scatter had little relationship to age, sex, race, and years of education completed. However, the average amount of scatter increased substantially with IQ level. Normative tables for interpreting scatter are provided, and implications of these findings for the practitioner, especially in neuropsychology, are discussed.

A large number of new or revamped individually administered intelligence tests have been published in the past 2 decades. This article begins by describing the development status of 3 categories of intelligence tests: psychometric-ability instruments, neuropsychologically based instruments, and dynamic assessments, with particular attention to issues that require additional research. It then considers how these categories of tests lend themselves to educational applications. Finally, the article describes recent basic and applied research findings that may influence how intelligence tests continue to evolve.The present scene in intelligence testing is essentially one of stagnation, with much talk but little progress. (Carroll, 1978, p. 93)

The adaptation of individually administered psychological tests from paper to app-based administration formats can present unique threats to the construct and raw-score equivalence of the paper and digital formats. We discuss these potential threats and describe a study evaluating the equivalence of paper and digital versions of the Wechsler Intelligence Scale for Children, fifth edition (WISC–V), which has been adapted to an app-based testing platform called Q-interactive. The present study (N = 350) used an equivalent-groups design to assess the raw-score equivalence of 18 WISC–V subtests that were designed with the intent of minimizing format effect. Effect sizes in standard deviation units were small, ranging from −0.20 to 0.20 (mean absolute value = 0.09), and format effects were unrelated to ability level, age, sex, or parent education. These results suggest that when tests are designed or adapted with digital–paper equivalence as a goal, digital–paper raw-score equivalence can be achieved across a variety of construct domains.