Confirmatory Factor Analysis of the Stanford-Binet Intelligence Scale: Fourth Edition with Preschoolers with Developmental Delays

Abstract

The Stanford-Binet Intelligence Scale: Fourth Edition (Binet IV; Thorndike, Hagen, & Sattler, 1986) has allowed for the evaluation of cognitive abilities across the life span and has been used frequently to assess the cognitive functioning of young children (e.g., 2-year-olds). Traditionally, four cognitive factors are interpreted along with an overall composite score. This study presents results of confirmatory factor analyses of the Binet IV, administered to 441 children between the ages of 2 and 5 years who have developmental delays. Four competing models were tested using iterative, maximum likelihood estimation of parameters. Results are consistent with previous factor analytic studies in that the traditional four-factor model is inferior to a two-factor or three-factor model. Implications for interpretation of Binet IV scores with young children with developmental delays are discussed.

Increasing the Reliability of Ability‐Achievement Difference Scores: An Example Using the Kaufman Assessment Battery for Children

“…Variables that "load strongly" on a particular factor or component are given a weight of one in computing composite scores, and those that do not are given a weight of zero. Factor analyses of different samples, even those using the same extraction and rotation methods, often result in discrepancies regarding which variables should be assigned to which composites (e.g., Gridley & McIntosh, 1991;Kaplan & Alfonso, 1997;Reynolds, Kamphaus, & Rosenthal, 1988;Thorndike, 1990;Thorndike, Hagen, & Sattler, 1986a, 1986b. Therefore, this method creates equal weights that are also sample dependent.…”

Reliable component analysis of the Stanford–Binet: Fourth edition for 2- to 6-year-olds.

“…The more reliable differences resulted in more precise confidence intervals and more powerful significance tests.The authors of the Stanford-Binet: Fourth Edition (SB:FE) originally suggested that four area scores be derived from each administration of the battery (in addition to the overall measure of intelligence) for every age group (Thorndike, Hagen, & Saltier, 1986a, 1986b. Subsequent exploratory and confirmatory factor analyses, however, have demonstrated that this model does a poor job of representing the information contained in the subtests, particularly for young examinees (Gridley & Mclntosh, 1991;Kaplan & Alfonso, 1997;Reynolds, Kamphaus, & Rosenthal, 1988;Thorndike, 1990). For children ages 2 to 6 years, most factor analyses have supported a two-factor interpretation.…”