[Show abstract][Hide abstract] ABSTRACT: Alternative gold standards to validate an innovative, dialect-neutral language assessment were explored.
Participants were 78 African-American children, ages 5;0 to 6;11. Twenty participants had previously been identified as having language impairment. The Diagnostic Evaluation of Language Variation-Norm Referenced (DELV-NR) was administered and concurrent language samples (LS) collected. Using LS profiles as the gold standard, sensitivity, specificity, and other measures of diagnostic accuracy were compared for 1) diagnoses made from the DELV-NR and 2) participants' clinical status prior to recruitment. A second analysis used results from the first analysis to make evidence-based adjustments in the estimates of DELV-NR diagnostic accuracy.
Accuracy of the DELV-NR relative to LS profiles was greater than accuracy of prior diagnoses, indicating that the DELV-NR was an improvement over pre-existing diagnoses for this group. Specificity met conventional standards, but sensitivity was somewhat low. When reanalyzed using the positive and negative predictive power of the pre-existing diagnosis in a discrepant resolution procedure, estimates for sensitivity and specificity for the DELV-NR were 0.85 and 0.93, respectively.
Making allowances for the imperfection of available gold standards, we conclude that clinical decisions made with the DELV-NR achieved high values on conventional measures of diagnostic accuracy.
Full-text · Article · Oct 2013 · Journal of Speech Language and Hearing Research
[Show abstract][Hide abstract] ABSTRACT: Language difference among speakers of African American English (AAE) has often been considered language deficit, based on a lack of understanding about the AAE variety. Following Labov (1972), Wolfram (1969), Green (2002, 2011), and others, we define AAE as a complex rule-governed linguistic system and briefly discuss language structures that it shares with general American English (GAE) and others that are unique to AAE. We suggest ways in which mistaken ideas about the language variety add to children's difficulties in learning the mainstream dialect and, in effect, deny them the benefits of their educational programs. We propose that a linguistically informed approach that highlights correspondences between AAE and the mainstream dialect and trains students and teachers to understand language varieties at a metalinguistic level creates environments that support the academic achievement of AAE-speaking students. Finally, we present 3 program types that are recommended for helping students achieve the skills they need to be successful in multiple linguistic environments. (PsycINFO Database Record (c) 2012 APA, all rights reserved).
Full-text · Article · May 2012 · Developmental Psychology
[Show abstract][Hide abstract] ABSTRACT: Purpose: The well-known decline in the use of African American English (AAE) features by groups of school-aged AAE-speaking children was reexamined for patterns of overt-, zero-, and mixed-marking for individual features and individual speakers.
Methods: Seven hundred twenty-nine typically developing children between the ages of 4 and 12—511 AAE-speakers learning General American English (GAE) as a second dialect, and 218 GAE-speaking controls—were administered the morphosyntax subtest of the Dialect Sensitive Language Test (Seymour, Roeper, & de Villiers, 2000). Responses to 33 items probing 10 target features were coded for overt marking, zero marking, or neither. A feature-by-feature marking profile for each child allowed us to track how many children at each age were characterized by 100% overt, zero, or mixed marking for different combinations of features.
Results/Conclusions: Findings suggest that no feature was overtly marked for all AAE-first children at any age, and the “mixed” pattern of usage was the most common trend across individual speakers even at age 12 years. Exclusive use of zero marking beyond age 8 years was rare and may serve as a diagnostic indicator.
Full-text · Article · Mar 2010 · Topics in Language Disorders