The Efficacy of Sound Partners when Delivered through Peer Tutors

• In Event: The Efficacy of Sound Partners when Implemented by Elementary Tutors: Academic Outcomes and Student Perceptions

Abstract

This randomized controlled trial (RCT) examined the efficacy of the Sound Partners reading intervention (Vadasy, 2005) when delivered through a cross-age peer tutoring model. This instructional method is cost-efficient, broadens access to individualized literacy instruction, and is feasible for use in various environments. Previous research supports cross-age peer tutoring as a successful approach across student populations (Chang et al., 2025; Leung, 2019; Shenderovich et al., 2016). In this model, an older student (the tutor) works with a younger student (the tutee), producing academic and social benefits for both learners (Chang et al., 2025; Leung, 2019).

This study addressed the following research questions:
1. How does Sound Partners delivered by peer tutors compare to adult tutors across multiple reading outcomes?
2. What is the impact of peer-delivered Sound Partners versus business-as-usual (BAU)?
3. What is the impact of adult-delivered Sound Partners versus BAU?

This study took place in 69 YMCA afterschool programs across Texas and Tennessee from Fall 2023 to Spring 2025, involving 770 students in two cohorts. Cohort 1 (n = 268) participated in 2023–2024 across 35 sites; Cohort 2 (n = 502) participated in 2024–2025 across 34 sites. Fifty-five percent of the sample was white, 22% was Black or African-American, 14% of the sample was more than one race, 6% were Asian, and 3% reported other. Twenty-one percent of the sample qualified for special education services and 19% of the sample identified as Hispanic or Latino/a.

Sound Partners, an evidence-based reading intervention (Vadasy et al., 2005), was delivered by either cross-age peer tutors or adults. The program focused on phonemic awareness, phonics, and word reading. Students participated in 20-minute sessions, three times per week, for 17 weeks, receiving between 50 and 110 lessons total. All tutors and tutees received training before lessons began.

This RCT used a multi-cohort design with three conditions: cross-age tutoring, adult-led tutoring, and a BAU control condition. Pre- and posttest assessments were administered by testers blind to condition. Measures included: the Test of Word Reading Efficiency (Phoneme Decoding and Sight Word Efficiency), the Woodcock Johnson IV (Letter Word Recognition, Reading Comprehension, and Spelling subtests), the DIBELS Oral Reading Fluency, and the Test of Silent Reading Efficiency and Comprehension. Attendance and dosage were tracked daily, and implementation fidelity was assessed using audio recordings from weeks 3 and 17.

Data was analyzed using actor-partner interdependence models (APIM; Kenny & Ledermann, 2010). This approach accounts for the interdependent nature of dyadic data (Cook & Kenny, 2005; Kenny, Kashy, & Cook, 2006).
Preliminary findings suggest that cross-age tutors are as effective as adult tutors in delivering Sound Partners. Tutors in grades 3–5 also demonstrated academic gains. Tutees in both treatment groups showed improvements in oral reading fluency, letter-word identification, and sight word reading.

Author

• Elizabeth Swanson, University of Texas at Austin