Skip Navigation
Funding Opportunities | Search Funded Research Grants and Contracts

IES Grant

Title: An Efficacy Trial to Evaluate Supporting Paraprofessionals by Advancing Reading Intervention Knowledge and Skill (SPARK)
Center: NCSER Year: 2018
Principal Investigator: Lemons, Christopher Awardee: Stanford University
Program: Reading, Writing, and Language      [Program Details]
Award Period: 4 years (07/01/2018 – 06/30/2022) Award Amount: $3,298,858
Type: Efficacy and Replication Award Number: R324A190240

Previous Award Number: R324A180004
Previous Awardee: Vanderbilt University

Co-Principal Investigators: Preacher, Kristopher; Lambert, Joseph

Purpose: This project will evaluate the efficacy of two models of professional development (PD) designed to enhance paraprofessionals' knowledge and skills related to delivering an early reading intervention to elementary school students with intellectual and developmental disabilities (IDD). The first model, a traditional reading-focused PD (T-PD), will use a training model developed through a prior IES-funded Development and Innovation grant. The second model, enhanced reading-focused PD (E-PD), goes beyond simply training paraprofessionals to implement a specific reading intervention by targeting broader knowledge and skills to improve overall instructional quality. Although paraprofessionals make up half of the special education instructional workforce and often provide instruction to students with disabilities, they receive limited training on basic educational practices. There has been limited research evaluating the efficacy of PD for paraprofessionals; therefore, the current project aims to address this by evaluating the comparative efficacy of two PD models designed to improve paraprofessionals' reading instruction for students with IDD.

Project Activities: A randomized controlled trial will be used to evaluate the efficacy of two models of PD (i.e., T-PD and E-PD) compared to business-as-usual PD provided by schools. Three cohorts of paraprofessional-student pairs will participate in the study across four years. Paraprofessional-student dyads will be randomly assigned to one of two treatment conditions or a business-as-usual control condition. For each cohort, data will be collected at four time points in the fall and spring of paraprofessionals' first year of participation (pre-test and mid-point, respectively), the spring of their second year (post-test) and the spring of their third year (maintenance; Cohorts 1 and 2 only). Data will be analyzed to evaluate the efficacy of the PD models for improving paraprofessionals' knowledge and skills related to reading instruction and students' reading outcomes.

Products: The products of this project will include evidence of the efficacy of T-PD and E-PD for paraprofessionals and their students with IDD and peer-reviewed publications and presentations.

Structured Abstract

Setting: The research will take place in elementary schools in Tennessee, Georgia, and Texas.

Sample: Approximately 135 paraprofessionals and 135 of their students with IDD (ages 5–8) will participate in the study.

Intervention: Both PD models, T-PD and E-PD, are designed to improve paraprofessionals' knowledge and skills in delivering an intervention focused on early reading skills (e.g., phonological awareness, alphabetic principle, decoding) and developed through a previous IES-funded Development and Innovation grant. The T-PD model is a more traditional and simplified model of PD that targets improved fidelity of the reading intervention, whereas the E-PD model targets a broader set of knowledge and skills to improve paraprofessionals' overall reading instruction and students' reading outcomes. Paraprofessionals in both the T-PD and E-PD models of professional development will be trained to implement the reading intervention. In the T-PD model, paraprofessionals will receive ongoing coaching support to increase fidelity of implementation of the student-level reading intervention. In the E-PD model, paraprofessionals will receive the same coaching as the T-PD model, but will also participate in supplemental online training modules focused on reading development, broader features of effective explicit instruction, general behavior management, and targeting problem behavior. Paraprofessionals will receive ongoing coaching support and will deliver the one-one-one reading intervention to their students with IDD for 40 minutes per day for 55 weeks.

Research Design and Methods: Using a randomized controlled trial design, paraprofessional-student dyads will be randomly assigned to one of two intervention conditions (T-PD or E-PD) or the business-as-usual control condition. Three cohorts with approximately 45 paraprofessional-student dyads each will be recruited and randomly assigned each year for the first three years. Paraprofessionals in the intervention conditions will participate in PD and ongoing coaching, and deliver reading intervention to eligible students across two academic years. Data will be collected from each of the three cohorts in the fall and spring of their first year of participation (pre-test and mid-point, respectively), the spring of their second year (post-test) and the spring of their third year (maintenance; only Cohorts 1 and 2 will provide maintenance data due to limits on project duration). Data will be analyzed to examine the effect of each professional development model on paraprofessionals' reading-specific and general knowledge and skills and students' reading outcomes. Data will also be collected on the fidelity of the PD training, implementation of the reading intervention, and instructional practices used by paraprofessionals in the business-as-usual condition.

Control Condition: Paraprofessionals in the business-as-usual condition will participate in regular PD provided by their school. They will deliver reading intervention and support to students in alignment with typical school practice.

Key Measures: Paraprofessional reading-specific knowledge and skills will be measured with a researcher-adapted Praxis exam (i.e., a set of teaching assessments used for certification) using released items from the following assessments: ParaPro, Special Education (Core Knowledge and Applications) and Teaching Reading (Elementary Education). A researcher-generated knowledge assessment will also be used to assess other areas targeted by the PD not addressed by the released items. To measure paraprofessionals' general knowledge and skills, researchers will create a measure using items from the Para Survey of Expectations Tool and the Assessment Checklist for Paras to be completed by paraprofessionals as well as their supervising special education teachers. Student reading outcomes will be measured with the Test of Word Reading Efficiency-2, Woodcock Reading Mastery Test-3, DIBELS, and a researcher-developed reading assessment aligned with the intervention. A survey will be used to collect data on potential paraprofessional-level covariates and moderators (i.e., demographics, prior PD experiences, experience working as a paraprofessional, and job satisfaction). The following measures will provide data on potential student-level moderators: Stanford-Binet Intelligence Scales for EarlyChildhood-5 (nonverbal and verbal visual processing and working memory), Kaufman Brief Intelligence Test-2 (verbal and nonverbal intelligence), a parent survey of student demographics, Expressive Vocabulary Test-2 (expressive vocabulary and word retrieval), and the Peabody Picture Vocabulary Test-4 (receptive vocabulary). Separate researcher-developed fidelity checklists will be used to fidelity of the PD and implementation of the reading intervention in both intervention and business-as-usual classrooms. The Instructional Content Emphasis-Revised measure will be used by coaches to code the content being emphasized during instruction in order to describe differences between instructional practices occurring in each condition.

Data Analytic Strategy: The researchers will use multiple-group structural equation modeling to examine the impact of the PD on paraprofessionals' knowledge and skills. Latent factor models will be used to compare outcomes for each contrast (T-PD vs. BAU, E-PD vs. BAU, and T-PD vs. E-PD). Multiple-group structural equation modeling will also be used to evaluate the impact of the PD on student reading outcomes. Effect sizes will be computed to describe the magnitude of treatment effects for both the paraprofessionals and the students. Moderation analyses will also be conducted to explore whether paraprofessional- and student-level variables moderate treatment effects on paraprofessional and student outcomes.

Related IES Projects: Enhancing Reading Instruction for Children with Down Syndrome: A Behavioral Phenotypic Approach (R324A110162)