Showing papers by "Susan E. Embretson published in 1997"

The Factorial Validity of Scores from a Cognitively Designed Test: The Spatial Learning Ability Test

Abstract: The Spatial Learning Ability Test (SLAT) was designed from a cognitive processing theory to measure specified aspects of spatial processing. Mathematical modeling of item difficulties and response times have supported SLAT as involving primarily spatial processing. Here, four studies on the factorial validity of SLAT are summarized to elaborate nomothetic span. SLAT was the highest loading test on the spatial ability factor in the context of either simple or complex spatial tests. Further, SLAT was less related to verbal-analytic coding skills than many other spatial tests, including a test that contained the same item type. Thus consistent with the construct representation of SLAT as involving spatial processing, the factorial validity studies indicate SLAT as a more pure measure of spatial ability.

Using confirmatory factor analysis to identify newborn behavior structure with the NBAS

Abstract: The Neonatal Behavioral Assessment Scale (NBAS) (Brazelton, 1973, 1984) is a 27-item scale representing individual differences in the newborn infants' behavioral repertoire. The NBAS items are typically collapsed into six clusters (Lester, 1984). Although the NBAS has been examined using exploratory factor analysis methods, no confirmatory factor analysis study has been published in which the behavior structure of full-term, healthy infants is examined. In the current research, competing models assessing variations of the Lester behavior clusters were performed using confirmatory factor analysis. Models comparing six independent factors, six correlated factors, and models estimating a general adaptiveness factor plus six factors were assessed. Examination of the results suggests continued investigations of neonatal behaviors are needed.