Examining the cultural loading and linguistic demand of three neuropsychological assessment batteries for children in a mixed clinical population
There are a disproportionate number of culturally and linguistically diverse (CLD) children and adolescents referred for and placed in special education programs (Guthrie, 2004). These CLD individuals are tested using neurocognitive measures and methods plagued with issues ranging from culturally loaded test content and linguistically loaded test instructions to psychometric issues caused by poor normative reference samples and/or failure to remove error variance attributed to differences in performance for reasons other than cognitive ability level (i.e., level of acculturation or level of language proficiency; Ortiz, Ochoa, & Dynda, 2012). There is a need to examine the influence of cultural loading and linguistic demand on the test performance of CLD individuals. The Culture-Language Test Classifications (C-LTC) and the Culture-Language Interpretive Matrix (C-LIM) are an integrated system that uses classifications of cognitive test batteries on the two dimensions of cultural loading and linguistic demand to determine whether a given test performance reflects differences on these dimensions or a true measurement of ability (Flanagan, Ortiz, & Alfonso, 2007, 2013). However, research on the validity of the C-LTC and C-LIM is limited. A study was conducted using archival data from a mixed clinical sample (n = 520) of children and adolescents from school neuropsychology case studies. This sample included test scores obtained by the participants for selected tests from three standardized neuropsychological test batteries: the Woodcock-Johnson III Tests of Cognitive Abilities, Normative Update (WJ III COG NU; McGrew & Woodcock, 2001); the NEPSY-II: A Developmental Neuropsychological Assessment (NEPSY-II; Korkman, Kirk, & Kemp, 2007); and the Delis-Kaplan Executive Function System (D-KEFS; Delis, Kaplan, & Kramer, 2001a). A multivariate analysis of variance (MANOVA) revealed differences in performance existed between ethnicity groups on selected subtests from the WJ III COG NU, NEPSY-II, and D-KEFS. A discriminant function analysis (DFA) was used as a post-hoc measure. Findings appear to suggest that the C-LTC ratings have some validity for use in CLD individuals. However, there were instances in which subtests rated highly for both cultural loading and linguistic demand were not found to be statistically different across ethnicity in this study. Limitations and implications of this study are presented.