Psychometric Analysis of the Diagnostic Evaluation of Language Variation–Screening Test: Extension to Low-Income African American Pre-Kindergarteners

Abstract

Seymour, Roeper, & deVilliers, 2003) is a culturally and linguistically unbiased screening tool used to identify children who may be at risk of significant language delays among the larger population of children who speak nonmainstream American English (NMAE) dialects. The DELV-S was developed in response to concerns about the misdiagnosis of African American and other linguistic-minority children with language disorders. In general, language assessments are particularly susceptible to performance bias, especially in linguistically diverse contexts such as the United States (Seymour, Bland-Stewart, & Green, 1998; Washington & Craig, 2004). Specifically, with regard to African American children, many become fluent speakers of African American English (AAE). AAE is a rule-governed, systematic cultural dialect of American English whose form, content, and use are different from more mainstream varieties of American English (Wolfram & Schilling-Estes, 2006). Although its use is particularly stigmatizing in formal contexts (e.g., schools, workplace, written texts), child language and sociolinguistic research have confirmed that AAE is not a deficient form of American English (Stockman, 2010). Despite vast empirical evidence on the nature of child AAE use, African American children continue to be overrepresented in diagnoses of speech language impairments (Klingner et al., 2005). Diagnostic accuracy is particularly complicated within this population, because many AAE features are similar to clinical markers of language impairments (e.g., variable production of grammatical markers such as the regular past tense-ed). Thus, screening tools such as the DELV-S can be very helpful to both clinicians and researchers seeking to characterize African American children's language use. In fact, many researchers have used the DELV-S to describe both spoken dialect use and language ability among young children (e.g., Horton & Apel, 2014; Terry, Connor, Petscher, & Conlin, 2012). However, the psychometric properties of the DELV-S have not been explored extensively. Briefly, the DELV-S has two parts. Part I of the DELV-S is used to evaluate children's NMAE use, with responses allowing children to be characterized as speaking with strong, some, or little to no variation 679402A EIXXX10.

Classification Accuracy of Teacher Ratings When Screening Nonmainstream English-Speaking Kindergartners for Language Impairment in the Rural South

“…Within the test manual and using the high-risk category as the cut score, Seymour et al report that, in a sample of 266 5-yearolds, sensitivity was 73% and specificity was 82%. Use of the DELV-ST-II for clinical practice also has been supported by two other studies, although in both, local norms were established and recommended rather than the criterion-based scores from the test manual (Petscher, Connor, & Al Otaiba, 2012;Terry, Petscher, & Rhodes, 2016).…”

Understanding African American vernacular English and reading achievement: Implications for the science of reading.

“…Petscher et al (2012) investigated the psychometric properties of the DELV screening test for children in kindergarten–second grade and found that when used as a screening assessment, the DELV-ST was more reliable for the assessment of language skills for students who had relatively weaker skills in comparison to those above average language skills. In a follow-up study, Terry et al (2017) extended their sample to 4-year old children and found that the same factor structure was present in the preschool sample (i.e., morphosyntactic factor and nonword repetition factor). However, only partial measurement invariance was documented which means that in comparison to the Petscher et al’s (2012) study, younger children may demonstrate different characteristics of AAVE.…”

“…This discussion of the DELV and the DELV screening provides an example of a tool that can be used by school professionals to understand language variation in Black children. It should be acknowledged that additional research is necessary given the lack of full measurement invariance in Terry et al (2017) because one specific tool is recommended as best practices.…”

Comparison of the Diagnostic Evaluation of Language Variation–Screening Test Risk Subtest to Two Other Screeners for Low-Income Prekindergartners Who Speak African American English and Live in the Urban South

“…Eighty-four percent of the children received either the same risk classification by both examiners or classifications differing by no more than one risk category. Finally, Petscher et al (2012) and Terry et al (2017) examined the factor structure and measurement invariance of the DELV–Screening Test Risk subtest using data from children in pre-K to second grade. These authors found support for the screener, although they recommended a two-factor structure to evaluate the grammar and nonword repetition items separately, the use of norm-referenced scores, and the development of different norms for children in pre-K as compared with K–2.…”