The Effects of Technology that Supports the use of IGDIs for Data-driven Intervention Decision Making

• In Event: 3-046 - Use of Individual Growth and Development Indicators to Support Early Intervention and Early Learning

Integrative Statement

Individual Growth and Development Indicators (IGDIs) for infant-toddler (Carta, Greenwood, Walker, & Buzhardt 2010) and PreK education (McConnell et al., 2016) are designed to support progress monitoring, and data-driven decision making by practitioners within an Multi-tiered System of Support (MTSS) model. However, without carefully designed resources and ongoing technical assistance, educators often are not adequately prepared to use child data to make meaningful adaptations of their practices for individual children (Diamond et al., 2013; Akers et al., 2016). Therefore, technology applications have been developed for IGDIs to make data-driven decision making more efficient, feasible, and effective. We will describe these applications and the results of randomized control trials of their efficacy: The Making Online Decisions (MOD) for infants and toddlers, and the Individual Growth and Development Indicators-Automated Performance Evaluation of Early Language and Literacy (IGDI-APEL) for preschool.

The MOD web application is an adaptive intervention support system for infant-toddler service providers to guide their intervention decision making for children with or at risk of language delay. The MOD guides early educators through a five-step decision making process which includes the use of algorithms driven by a child’s performance on the Early Communication Indicator [ECI: Greenwood et al, 2010]) to recommend individualized evidence-based language intervention strategies for parents to use with their children. Using an RCT design, 45 Early Head Start home visitors (HVs) in four states were randomly assigned to two experimental conditions: one that only used evidence-based strategies, and another that also used the MOD to guide their use of those evidence-based strategies with families. Participating families (n=146) served by those HVs had a child between 6-42 months of age who performed below benchmark on the ECI. We hypothesized that: 1) Children whose HV uses the MOD will demonstrate stronger language growth than children whose HV did not use the MOD, and 2) Higher fidelity of MOD implementation will be associated with stronger language growth. Hierarchical Linear Modeling of children’s language growth on the ECI and Preschool Language Scale between groups supported both of these hypotheses.

The APEL iPad application facilitates standardized data collection, supports computer adaptive testing (CAT) so IGDI items are selected and administered to efficiently identify a child’s current skill level, and automatically score performance in real time. This process yields easy-to-use reports, graphs and data analysis tools that allow teachers to evaluate performance and make high-quality instructional decisions about the appropriate tier of instruction needed, and intervention materials to be used within that tier. In a recent RCT, 78 preschool teachers were randomly assigned to three conditions: business as usual, IGDI cards and IGDI-APEL. We hypothesized that children in the IGDI-APEL group would outperform the other groups on Rhyming, Picture Naming, Which One Doesn’t Belong (WODB), Sound Identification, and First Sounds. Analyses confirmed that the IGDI-APEL condition outperformed both conditions for WODB, and Sound ID; they also outperformed the IGDI cards condition on First Sounds, Picture Naming and Rhyming. Implications of using technology to support MTSS in early childhood education will be discussed.

Authors

• Jay Buzhardt, University of Kansas
  Presenting Author

• Alisha Wackerle-Hollman, University of Minnesota
  Non-Presenting Author

• Dale Walker, University of Kansas, Juniper Gardens Children's Project
  Non-Presenting Author

• Charles Greenwood, University of Kansas
  Non-Presenting Author

• Fan Jia, University of Kansas
  Non-Presenting Author

• Scott McConnell, University of Minnesota
  Non-Presenting Author

• Erin Lease, University of Minnesota
  Non-Presenting Author

• Michael Rodriguez, University of Minnesota
  Non-Presenting Author

• Lillian Duran, University of Oregon
  Non-Presenting Author