Improving Student Conceptual Math Knowledge Through an Online Transmedia Learning Platform

• In Event: A Structured Approach for Testing Effectiveness of Educational Technology Products

Abstract

Research has shown that U.S. students continue to struggle immensely in mathematics, especially compared with the performance of students from many Asian countries (OECD, 2016). Part of why U.S. students lag behind their international peers relates to the continued emphasis on procedural learning and repeated practice in mathematics curricula, rather than deeper, more complex math thinking (Kilpatrick et al., 2001). The current research study examined how an online fractions software helps 3rd and 4th graders deepen their knowledge of math practices and misconceptions across different math domains.
[company] has been developing engaging math software focused on deep learning and strategy development for a decade. The math program is focused on single-digit operations and fractions and a significant component of the program has been the development of characters and stories that engage students’ imaginations and make abstract concepts, such as strategy use, explicit. Transmedia learning offers an exciting way to engage students in difficult and abstract mathematics topics. Transmedia learning involves engaging with a narrative that develops across different media (Raybourn, 2014). Research has shown some initially promising results (McCarthy, Tiu, Li, 2018; Pasnik & Liorente, 2013). While most current educational software and edutainment employ ineffective pedagogy and focus solely on skill-building (Bray & Tangney, 2017; EdReports, 2017), novel transmedia approaches aimed at improving conceptual understanding have the potential to positively impact student learning.
To test the impact of [product] on students’ fraction learning, the current study employed an underpowered randomized control trial design with twenty 3rd grade teachers and 403 students. Teachers were randomly assigned to either a treatment group or a business-as-usual group. Treatment teachers implemented 16 Fractions Boost Intervention lessons in whole class and small group settings for 13 weeks. Students who received small group instruction were identified as low-performing on the study’s pre-assessment. Teachers in the control condition taught their usual curriculum and were instructed to provide some additional support to low-performing students in their class. Data sources were informed by the products’ logic model and included: teacher pre and post surveys, teacher logs, student pre assessment, and student pre-post attitude survey. The final analytic sample included 206 students in the treatment condition and 197 in the control condition.
Researchers conducted a two-level hierarchical model with students nested within teachers, which included fixed effects for pre-test and condition. This model was tested on the sample overall, and on low-performing students. The results of the sample overall showed small effects (g = .24; p = .48) but the size of effects with the sample of low-performing students showed large effects (g = .60; p = .27). These results are promising as they indicate that the [product] platform and resources may play a role in bolstering low-performing math students’ learning gains.

Author

• Ruchita Patel, WestEd