Teacher Talk: Explorations of Classroom Practices and Links to Literacy Achievement Within Upper Elementary Classrooms

• In Event: The Forgotten Giant: Classroom Talk That Supports Language Development Across Grades

Abstract

This study examines the talk practices of 529 fourth- and fifth-grade literacy teachers and links between classroom talk practices and the reading achievement of 13,197 students. Research and theory suggest experiences with high quality classroom talk would develop the higher level thinking, processing, and language skills involved in reading (Mercer, 1995, 2000; Murphy, Wilkinson, Soter, Hennessey, & Alexander, 2009; Vygotsky, 1978, 1986). An example would be that as students experience academic language, they would build more high quality lexical representations for words to use in their reading (Perfetti, 2007). Research has established guidelines that encourage talk (e.g, Goldenberg, 1992), yet such recommendations are not entering classrooms. In fact, we know little about what is occurring in elementary classrooms because as Nystrand (2006) notes, “Until recently, there have been few large-scale quantitative studies of the effect of classroom discourse on reading comprehension, and these studies have mainly focused on middle and high schools" (p. 403-404). Our study fills this gap and examines the makeup of talk and links to reading achievement in upper elementary classrooms.

The study uses data from the Measures of Effective Teaching (MET) study, which includes assessment data from a large number of classrooms across different school districts teaching varied curricula. Twenty two items related to classroom talk practices were identified from observer protocols (i.e., the CLASS and PLATO) and from the Tripod Student Perception Survey. Two-level exploratory factor modeling indicated that performance could be grouped into three factors of teacher talk at the student level representing teacher explaining, teacher questioning, and promotion of student talk and two factors at the teacher level representing rater and student observation of teachers. Multilevel structural equation modeling suggested that controlling for the other predictors in the model including demographic variables and pretest abilities, both student and rater observations predicted standardized reading gains, although patterns differed. Student perception of teacher explaining but not teacher questioning nor promotion of student talk predicted state standardized reading gains and observer ratings predicted gains on an open-response standardized reading assessment.

Findings deepen understanding of classroom talk practices occurring in upper elementary classrooms and link these practices to reading achievement. Research shows that teachers tend to use different approaches when delivering identical content (Duffy, Roehler, & Rackliffe, 1986). Our study indicates a clear link between reading growth and approaches that use high quality classroom talk versus those that use lower quality talk. Our study also shows that classroom talk is multidimensional such that teacher explanations, teacher questioning, and promotion of student talk represent unique constructs with observer and student ratings representing unique factors at the teacher level. Results related to this dimensionality indicate that high quality explanations are potentially more powerful in promoting reading achievement than questioning and promotion of student talk. Additionally, both independent observers and students provided meaningful information about the talk practices occurring within classrooms as each predicted gains on different standardized reading assessments. Such findings suggest the importance of multiple lenses. Overall, our study confirms that talk matters in upper elementary classrooms.

Authors

• Amanda P. Goodwin, Vanderbilt University

• Sun-Joo Cho, Vanderbilt University

• Stephanie Nunn, University of Maryland

• Rebecca Deffes Silverman, University of Maryland - College Park

• Dan Reynolds, Vanderbilt University - Peabody College