Centering the Oral Language Skills of Minoritized Children: Evidence from Boston’s Universal PreK Program

• In Event: Examining the Language Skills of Racially and Linguistically Marginalized Children in Early Childhood

Abstract

Oral language is a predictor of later literacy skills (Roth et al., 2002; Snowling & Hulme, 2020). Picture book retellings are one way to assess oral language development (Lysaker & Sedburry, 2015). However, few research studies have explored picture book retellings with a specific focus on children from racially and ethnically minoritized groups. To address this need, the current study uses data from the Boston Public Schools’ recent Universal PreK (UPK) expansion to examine indicators of oral language during picture book retelling for Black, Hispanic, and Dual Language Learner (DLL) children at the beginning of the PreK year.

The sample consists of 71 children who attended a community-based PreK program in Boston that participated in UPK during 2021 – 2022. The child sample is 45% Black, 55% Hispanic, and 52% Dual Language Learner (DLL). About half of the sample is female and the average age at test administration was 4.25 years (SD = .32).

As a part of a broader study of the BPS UPK program, students were administered the Renfrew Bus Story narrative retelling task (Renfrew, 1969). Children were told a 186-word story by an examiner. During the telling, the assessor showed each child 12 corresponding illustrations to what was happening in the story. Immediately after hearing the story, children were prompted to retell it using the illustrations as supports. Children’s responses were then scored for inclusion of information, sentence length, and amount of required prompting required, using the examiner’s manual (Cowley & Glasgow, 1994). A small subset of DLL children (10%) demonstrated limited English proficiency on a screener and completed the task in Spanish.

Black, Hispanic, and DLL children both correctly provided more than half of the information from the original study. This score tracks children’s inclusion of particular details in the story retelling. However, for 32% of the 53 possible points, there is no opportunity for partial credit. Nearly half of these instances also require students to produce direct quotes from the story. All children typically performed better on the information tasks that did allow for partial credit, with an average score increase of 17% for these items.

Black students produced sentences that where about 8 words long on average (SD = 2.45) while Hispanic and DLL students used sentences that were about 6 words on average (SD = 3.12 for Hispanic; SD = 3.45 for DLLs). Both Black (67%) and Hispanic (60%) students made unprompted connections to the emotions and motives of the characters, which is a measure outside of the typical assessment metrics. Rather than completely omit components of the story, children in all three groups were likely to unsuccessfully or incorrectly attempt to retell these elements. Such findings highlight creativity and imagination in eliciting responses.

Authors

• Olivia Horne, Harvard University
  Presenting Author

• Catherine E Snow, Harvard University

• Christina J Weiland, University of Michigan

• Meghan Patricia McCormick, MDRC (Manpower Demonstration Research Corporation) - New York City