Impacts of TRI on Early Elementary Students’ Reading Fluency

• In Event: A Multifaceted Lens: Perspectives and Impacts of an Effective Elementary Teacher Literacy Professional Development Program

Abstract

Students for whom reading is challenging in the early grades tend to fall further behind their typically developing peers over time (Stanovich, 1986). These diverging trajectories are in part related to difficulties in sight word recognition or decoding accuracy (Torgesen et al., 2001). In addition, early word reading fluency is a robust predictor of young students’ later oral text reading fluency (Kim & Wagner, 2015) and directly and indirectly predicts concurrent reading comprehension (Solari et al., 2018). Disfluent reading can be slow and laborious for both students and teachers (Gangl et al., 2018), resulting in fewer encounters with advanced texts and vocabulary (Hammerschmidt-Snidarich et al., 2019) and further impeding exposure to and practice with more irregular words (Torgesen, 2005).

Interventions supporting students’ early word reading fluency have important ramifications for subsequent reading and schooling success. This study will examine whether TRI impacts students’ sight word and pseudoword reading fluency. Although early reading interventions have improved students’ word reading accuracy (Gersten et al., 2020), fewer have impacted word reading fluency, particularly for younger students. This study will be the first to examine TRI’s impacts on word reading fluency, utilizing a sample of kindergarten and first grade teachers and their students.

This study analyzed data from a three-year RCT of TRI delivered by classroom teachers. The study involved ten Title I rural schools. Across all three years, 50 kindergarten (26 treatment, 24 control) and 50 first grade (29 treatment, 21 control) classrooms at each school were randomized. After accounting for teacher attrition, the final sample included 118 teachers. Following a two-pronged approach to student screening/selection using Dynamic Indicators of Basic Early Literacy Skills (DIBELS) and Woodcock-Johnson (WJ), approximately three students per classroom were selected for the study, with a final sample including 540 students (299 treatment, 240 control); see Table 1. Treatment teachers were provided with training and coaching for two years if they stayed in the same school and grade level. Control teachers served as a “business as usual” counterfactual.

In the fall and spring of each intervention year, trained assessors administered Test of Word Reading Efficiency (TOWRE; Torgesen et al., 1999) Sight Word Efficiency (SWE) and Phonemic Decoding Efficiency (PDE) subtests. Condition was represented by a binary dummy variable. Control variables included fall pretest SWE or PDE scores, fall WJ scores, child gender, child race, maternal education, income, grade, and processing speed. Three-level hierarchical linear models (HLM) accounted for students nested within classrooms and schools. Multiple imputation then deletion (von Hippel, 2007) procedures accounted for missing data. Two HLM models predicted separately to SWE or PDE. Effect sizes were calculated using Hedge’s g.

Results showed that TRI treatment condition positively impacted spring SWE scores (B=2.04, SE=0.77, p=.008, g=0.17) and PDE scores (B=1.36, SE=0.49, p=.006, g=0.23); see Table 2. Results suggest that teacher-delivered reading intervention can promote word-level fluency gains for young students reading below grade level, which is a finding relatively unique to TRI and will be further tested in a current replication project (see study 3).

Authors

• Mary Bratsch-Hines, University of Florida
  Presenting Author

• Sarah Pedonti, Western Carolina University

• Nancy McIntyre, University of Central Florida

• Heather Aiken, University of North Carolina at Chapel Hill