Individual Submission Summary
Share...

Direct link:

Poster #80 - One-to-One Digital Mathematics Resources in Kindergarten: Results from a District-Wide, Cluster-Randomized Controlled Trial

Fri, March 22, 2:30 to 3:45pm, Baltimore Convention Center, Floor: Level 1, Exhibit Hall B

Integrative Statement

Research on technology in early childhood education has shown that children and teachers enjoy using technological tools (Blackwell, 2014; Brown, 2016) but has not yet documented how digital resources can contribute to learning in classroom environments, especially for low-income children in urban areas (Falloon, 2013; Flewitt et al., 2015). The goal of the current study was to examine the impact of an outreach effort by a local affiliate of the Public Broadcasting Service (PBS), which donated a tablet to every kindergartener in a small urban district in a Midwestern state. Utilizing a cluster-randomized controlled trial design (Puffer, Torgerson, & Watson, 2005), all kindergarten classes in the district were randomly assigned within schools to treatment or control conditions. Treatment classes received tablets loaded with digital games designed by PBS and selected by researchers to promote mathematics learning for kindergarteners (Table 1). Control classrooms received tablets without the PBS math games. Each tablet was provided with a unique ID number and assigned to a specific student, which allowed researchers to track students’ use of the math games remotely. The research questions (RQ) were:

RQ1. Do students in the treatment group demonstrate greater growth on standardized tests of early numeracy and literacy at the end of the year compared to the control group?
RQ2. Do students who began the year with lower standardized test scores demonstrate larger gains from treatment compared to students with higher scores?
RQ3. Do high-volume users of the PBS math games demonstrate larger gains compared to low-volume users within the treatment group?

The sample included 535 kindergarten students in 35 classes in 14 schools (51.4% female, 33.1% White, 28.6% Black, 21.5% Latino, 3.7% Asian, 5-7 years of age, 93.1% receiving free/reduced price lunch). To analyze treatment effects, a series of three-level multilevel models were constructed nesting students within classes within schools using HLM-7 software (Raudenbush, Bryk, Cheong, Congdon, & du Toit, 2011). The outcome variables were spring scores from the district-administered AIMSweb Tests of Early Numeracy and Early Literacy.

The results are summarized in Table 2. After controlling for pretest scores, the treatment group demonstrated significantly higher scores on only AIMSweb subtest, Letter Naming, compared to the control group (RQ1). Thus, treatment appeared to impact literacy rather than math skills, perhaps because literacy skills were used to navigate the games. No interactions between pretest and treatment were significant (RQ2). Overall, the digital games were used infrequently by most of the treatment group, averaging only about 13 minutes per week. Within the treatment group, high-volume game users (top quartile) demonstrated greater growth on one subtest, Missing Number, compared to low-volume users (bottom quartile) (RQ3). These results suggest that access to digital resources, even those created by a trusted provider like PBS, does not necessarily lead to greater academic achievement in the targeted skills. In addition, there is a need for more research that examines how digital resources can be used in classrooms to supplement skills instruction.

Authors