Poster #45 - Development and Initial Validation of an Early Childhood Practitioner-Administered Language Comprehension Screener

• In Event: Poster Session 15

Abstract

Objective: The Simple View of Reading (Gough & Tunmer, 1986) conceptualizes reading comprehension as the product of word decoding and language comprehension. Accordingly, children with lower levels of oral language may also experience challenges with reading comprehension (e.g., Catts et al., 2006; Justice et al., 2013). As such, there have been calls to assess oral language early and often, and to facilitate the language skills that support reading comprehension (e.g., Adlof & Hogan, 2019; LARRC, 2017). Language sample analysis (LSA) is regarded as the gold standard for culturally-sensitive, ecologically-valid measurement of oral language; however, it is time-consuming and requires specialized training and software, limiting its viability as a screener by early childhood (EC) practitioners (e.g., Bowles et al., 2020).

The aim of this research is to describe the development and initial validation of an EC practitioner-administered language comprehension screener for grades Pre-K3—Kindergarten. Specifically, we ask: To what extent might an oral language screener serve as a proxy for language abilities in the areas of semantics and morphosyntax as measured by language sample analysis (LSA)?

Method: We developed a language comprehension screening instrument for grades Pre-K3, Pre-K4, and Kindergarten (Grades 1-3 currently in development) consisting of the following components: 1) illustrated passages and an accompanying passage retell scoring rubric to assess aspects of semantics (e.g., key passage details; vocabulary) and morphosyntax (e.g., coherence; sentence structure); 2) expressive comprehension questions; and 3) receptive comprehension questions. We piloted the measure with 1,273 children in grades Pre-K3—Kindergarten, across 10 divisions, 46 schools, and 227 classrooms in one state in the Southeastern US between Mar-Jun 2022. Passage retells were recorded for later analysis, and expressive and receptive comprehension questions were scored live.

To date, researchers have transcribed 263 passage retells (~20% of sample) and applied SALT transcription conventions (SALT Software, LLC, 2020). LSA is being conducted using CLANc V12 MOR and KIDEVAL programs (MacWhinney, 2000; 2022). Additional planned analyses include using passage retell scoring rubrics to assess aspects of semantics and morphosyntax (Figure 1). We will also examine item distributions for passage retell LSA indices, passage retell rubric scores, and expressive and receptive comprehension items, and calculate psychometric properties of the screening measure using Item Response Theory and factor analytic techniques in addition to validating the screener using existing standardized language measures (e.g., CELF Preschool-3 Screening Test; Wiig et al., 2020).

Results: LSA has been conducted for 78 Pre-K3, 46 Pre-K4, and 139 Grade K children, to date (Table 1). Preliminary descriptive findings for LSA indices reveal expected higher levels of language productivity (e.g., mean length of utterance in words; type/token ratio) and complexity (e.g., number of verbs per utterance; use of Brown’s 14 morphemes) for children in higher grade levels than lower grade levels.

Conclusion: Language comprehension screening is critical to supporting the development of reading comprehension during the early childhood years. This research describes the development and initial validation of an EC practitioner-administered screener with promise to be implemented at scale as part of a comprehensive language and literacy screener.

Authors

• Khara Lynn Pence Turnbull, University of Virginia - Charlottesville
  Presenting Author

• Jissel Anaya, University of Virginia

• Christa Haring, University of South Florida

• Carlin Connor, University of Virginia

• Emily J. Solari, University of Virginia - Charlottesville