Validity of executive functioning tasks across the WJ III COG, NEPSY, and D-KEFS in a clinical population of children: Applicability to three neurocognitive theories
Avirett, Erin K.
MetadataShow full item record
Inconclusive research regarding the neurocognitive construct of executive functioning has restricted the development of valid pediatric executive functioning assessments (Floyd et al., 2006: Maricle, Johnson, & Avirett, 2010). Misunderstandings in the research have led to divergent executive functioning theories and assessment tasks. Therefore, it cannot be assumed that all executive functioning instruments are measuring the same construct. Given the common inclusion of executive functioning tasks in pediatric neuropsychological evaluations (Stuss & Alexander, 2000), it is important to determine the validity of executive functioning theories and assessment tools. Furthermore, because these evaluations are often administered to children with clinical diagnoses, it is important to assess validity issues with this group. Therefore, this study aimed to determine the concurrent validity of the executive functioning subscales of three commonly utilized neurocognitive instruments: the Woodcock Johnson III Tests of Cognitive Abilities (WJ III COG; Woodcock, McGrew, & Mather, 2001c), the NEPSY (Korkman, Kirk, & Kemp, 1998), and the Delis Kaplan Executive Function System (D-KEFS; Delis, Kaplan, & Kramer, 2001). An associated purpose of this study was to determine the underlying factor structure of the WJ III COG, NEPSY, and D-KEFS, and their fit with three theories of executive functioning. The three theories that were analyzed include the Anderson, Levin, and Jacob (2002) model of executive functioning, the Cattell-Horn-Carroll theory of cognitive abilities (CHC theory; McGrew, 2005), and the Conceptual Model for School Neuropsychological Assessment (SNP model; Miller, 2007, 2010). Archival data was extracted from school neuropsychology case study reports. Children from a clinical sample, aged 8 through 12, were included in the study. Bivariate correlations were conducted in order to determine relationships among executive functioning subtests. These analyses revealed that executive functioning subtests appear to be measuring distinct abilities and are not interchangeable. Furthermore, the reliable use of most of these subtests within a clinical population was indicated. Level of fit between executive functioning models and sample data was depicted using structural equation modeling and analyzed using confirmatory factor analysis. The SNP conceptual model indicated the best fit with sample data.