Randomized Controlled Trial of a Knowledge-Building Read-Aloud Program in Kindergarten

• In Event: How Implementation Science and Design Thinking Can Be Nurtured in Research That Addresses Wicked Problems

Abstract

Purpose. This study examined effects of the widely used Core Knowledge Language Arts: Knowledge Strand (CKLA) literacy curriculum on kindergarteners’ development of vocabulary and knowledge.
Perspective. The CKLA read-aloud program is a unique approach because it follows a planned sequence of specific topics and views children’s comprehension skills (e.g., forming predictions, answering inferential questions) as resulting from the domain knowledge (e.g., science, social studies) and vocabulary they acquire from hearing texts read aloud. The view has a grounding in cognitive science (Willingham, 2006), positing that children who have a wider breadth of knowledge and vocabulary have an advantage as they have wider and stronger semantic networks of information to draw upon when comprehending a text.

Method and data source. In each of two urban districts, 47 schools were randomly assigned to either the treatment (n=23) or business-as-usual control condition (n=24). Implementation took place in two cohorts (2017-2018 in mid-Atlantic district; 2018-2019 in Southern district). Kindergarten teachers in treatment schools implemented the CKLA read-aloud program for one semester, focusing on science and social studies topics. Students (n=1,205) were tested in the fall (prior to randomization) and spring. The battery of assessments was individually administered and included standardized vocabulary, listening comprehension, and science and social studies knowledge assessments. In addition, the post testing battery also included vocabulary and knowledge measures that were proximal to the intervention, testing the words and knowledge covered in the intervention.

Results. Preliminary results of multilevel modeling with the first cohort of students indicated no discernable main or moderated effects on standardized measures of vocabulary and listening comprehension. There was a non-significant, yet meaningful, main effect on a standardized measure of science knowledge (d = 0.12), which was notable given the intervention’s short duration. There was a significant, medium-sized effect (d = 0.46) on the curriculum-based vocabulary measure, as well as medium- and large-sized effects on children’s knowledge of particular domains, namely plants (d = 0.34) and Native Americans (d = 1.16). There was also a significant interaction such that students who began the year with higher vocabulary scores learned more words. Further analyses will examine the combined results of cohorts and explore site effects.

Significance and wicked problems. This study makes an important contribution to the scarce literature on the effects of a knowledge-building English Language Arts curriculum on kindergarteners’ skills. There are several wicked problems that we faced while conducting this study that speak to the collaboration between researchers and organizational stakeholders. A set of barriers related to classroom implementation emerged from the TDF domains of behavioral self-regulation and professional identity. These included teachers’ reluctance to break from “traditional” conceptions of read-alouds and use what was perceived to be a “scripted” curriculum that they may have viewed as disregard for their professional expertise. Another set of barriers related to the TDF domains of environmental context and resources, as well as social influences and group norms. Specifically, textbook adoption and district/school desires for more rapid change in reading scores presented barriers to longitudinal implementation across multiple years.

Authors

• Sonia Cabell, Florida State University

• James S. Kim, Harvard University

• Thomas G. White, University of Virginia

• HyeJin Hwang, The Florida State University

• Charles Gale, Harvard University