Co-Principal Investigators: Downer, Jason; Hall, Colby; Logan, Jessica

Purpose: The purpose of this study is to test the optimal sequence of two teacher-implemented supplemental interventions for kindergarten students at risk for reading difficulties. The first intervention, Read Well Kindergarten (RWK), teaches evidence-based foundational reading skills and the second intervention, Banking Time (BT), is meant to enhance teacher-student relationships to decrease challenging behavior. Both have been shown to be effective and feasible in classroom settings, but they have never been implemented together and their optimal ordering has not been explored. Research has demonstrated a reciprocal relationship between reading and behavior problems that makes treatment for either problem in isolation difficult. Comparing the order of two interventions will allow researchers to determine whether enhancing teacher-child relationships and reducing children's challenging behaviors will enable them to more fully engage with reading intervention or whether improving children's reading skills will facilitate teacher relationships and reduce challenging behavior. This project will use a sequential multiple assignment randomized trial (SMART) to determine the optimal ordering of RWK and BT for improving reading and behavior outcomes for kindergartners at risk for reading difficulties.

Project Activities: Using a SMART design, researchers will randomly assign children at risk for reading difficulties to either the RWK or BT intervention. After 10 weeks of the first intervention, children will be re-screened and assigned to a second-stage treatment when appropriate. This design will allow investigators to examine whether one or both orders are effective and if there is an ideal order.

Products: This project will result in evidence regarding the optimal order of implementing the RWK and BT interventions for improving reading achievement and classroom behavior for kindergarteners at risk for reading difficulties, as well as information on the costs to implement both interventions, their cost effectiveness, and the costs relative to the benefits. The project will also result in a final dataset to be shared, peer-reviewed publications and presentations, and additional dissemination products that reach education stakeholders such as practitioners and policymakers.

Structured Abstract

Setting: The research will take place in elementary schools in Virginia.

Sample: Approximately 72 kindergarten teachers and 720 of their students at risk for reading difficulties will participate in this research.

Intervention: RWK is a published, manualized curriculum used for small group early reading instruction. Lessons provide opportunities for students to build phonemic awareness and knowledge of grapheme-phoneme correspondences, learn decoding strategies, acquire new vocabulary, and develop comprehension skills. BT is a classroom-based dyadic intervention that supports teachers in interacting closely and positively with students in order to develop strong and supportive relationships. It consists of 10-15-minute student-driven sessions, where the teacher follows the student's lead and accepts the student unconditionally. These sessions provide new information to the teacher and student about one another and aim to disrupt the cycle that results in students' escalating disruptive behavior.

Research Design and Methods: This project will use a SMART design to examine the optimal sequencing of RWK and BT. Children who score below an established cut point on an initial reading screener will be randomly assigned to receive either RWK or BT during the first 10 weeks of implementation. RWK students will receive 25 minutes of supplemental reading instruction using the intervention in groups of 4-5 students, 4 days a week for 10 weeks. BT students will receive two 15-minute one-on-one BT sessions with their teacher per week for 5 weeks. After 10 weeks, students will be screened again to assess reading difficulties and assessed to determine the initial effects of the intervention on early reading skills and behavior. All children who received the BT intervention who continue to screen as at risk on the reading screener will move to the RWK intervention. Children who were receiving BT who screen above the risk cut off will return to business-as-usual classroom instruction. Children who were receiving the RWK intervention who score above the risk cut off will be moved to the BT intervention. Children who were receiving RWK who score below the risk cut off will be re-randomized to either BT alone or to BT + RWK. BT + RWK students will receive one BT session and two RWK sessions per week. After 10 weeks, student reading and behavior outcomes will be assessed. This process will be repeated in three cohorts of 240 students each. Information on the costs to implement the interventions will be collected and cost-effectiveness and cost-benefit analyses will be conducted. Implementation fidelity will be observed and assessed throughout implementation.

Control Condition: Due to the nature of the research design, there is no traditional control condition. In the first stage analysis, children who received RWK will serve as the control for those that received BT when examining behavior outcomes, and children who received BT will serve as control for those that received RWK when examining reading outcomes. In the second stage, children who do not receive additional RWK will serve as controls for those who do.

Key Measures: Students will be screened for initial risk of reading difficulties with the Letter Naming Fluency and the Phoneme Segmentation Fluency subtests of the Dynamic Indicators of Basic Early Literacy Skills 8th edition (DIBELS-8). Primary reading outcomes will be assessed with the Comprehensive Test of Phonological Processing, Second Edition (CTOPP-2) and the Letter Word Identification, Word Attack, and Passage Comprehension subtests of the Woodcock-Johnson IV Tests of Achievement (WJ IV). Secondarylanguage outcomes will be assessed with the Clinical Evaluation of Language Fundamentals-5 (CELF-5), the Peabody Picture Vocabulary Test, Fifth Edition (PPVT-5), and the Expressive Vocabulary Test, 3rd Edition (EVT-3). Students will be assessed for initial behavioral difficulties with two teacher-report measures, the Disruptive Behavior Rating Scale and the Sutter-Eyberg Student Behavior Inventory-Revised (SBSEI-R). Primary behavior outcomes will be assessed with the Adapted Leiter-R Assessor Report of Attention/Impulsivity and the Head-Toes-Knees-Shoulders task. Student-teacher relationship outcomes will be measured with the Student-Teacher Relationship Scale (STRS) and the Young Children's Appraisals of Teacher Support (YCATS). Implementation fidelity will be assessed with previously validated researcher-created measures, teacher-reported logs, coach rankings of implementation quality, and weekly video-taped sessions scored by trained evaluators.

Data Analytic Strategy: Primary outcomes will be analyzed with a multi-level regression model where children are nested within classrooms. After the first 10 weeks of intervention, main effects for each intervention will be estimated. After the second 10 weeks of intervention, groups that received both treatments in different orders will be compared. Researchers will use a regression discontinuity analysis to compare children who fall just above the established cut point on the initial reading screener (and thus received no intervention) and those who score just below the established cut point and were randomized into treatment conditions.

Cost Analysis: The research team will estimate costs associated with training and implementation for both interventions, and those associated with sustaining implementation at the school level as well as scaling up implementation to the district level. The estimates will be based on expenses for intervention materials, training, personnel to support implementation, data collection, meetings, utilities, maintenance and other operating costs, and administrative costs. Cost-effectiveness and cost-benefit analyses will be conducted using the returns-on-investment formula.