Number Ball: Learning Whole Number Arithmetic through Play

• In Event: Experimental Mathematics Interventions in Classrooms and Schoolyards to Promote Early Numeracy and Rational Number Learning

Abstract

Children’s mathematical skills in the early elementary years are strong predictors of later math achievement, emphasizing the importance of early interventions to support children’s math learning trajectories. Play-based approaches guided by adults can effectively support early math competency by leveraging children’s intrinsic motivation to play and incorporating explicit learning goals (Hirsh-Pasek et al., 2020; 2022). This study presents an intervention merging play and math learning for K-2nd graders through adult-facilitated basketball games. Six K-2nd grade teachers participated in three 60-minute sessions to adapt basketball games that effectively supported rational number learning in older children (Authors, 2022) to focus on whole number skills. Figure 1 shows a redesigned basketball court with arcs worth 0 to 5 points (1A), a walkable number line 0 to 20 (1B), and one of the six basketball games from the intervention (1C). Participants were 147 predominantly low-income Latine students (47% female) attending two kindergarten classrooms (n = 49), two first grade classrooms (n = 45), and two second grade classrooms (n = 53). The games were facilitated by six math teachers and one PE teacher with support from the research team. A randomized controlled trial design was used to assign students within classrooms, matching on gender and pre-test scores, to the intervention (n = 72) or control group (n = 75). Students completed three math assessments at pre- and post-test: the Number Sets task (Geary, 2011), a tablet version of the Number Line Estimation task (Xu et al., 2021), and addition and subtraction problems presented on a tablet (Geary et al., 2004). Students’ strategies for solving the additional problems were coded (e.g., counting fingers, verbal counting, decomposition, and retrieval). The tasks were administered individually in English (n = 106) or Spanish (n = 41) based on the child’s preference. The study was pre-registered at https://osf.io/27mhg before post-test data processing and linking to treatment status. Clustering standard errors by classroom and controlling for students’ pre-test scores and grade level, OLS regressions showed small and nonsignificant impacts on standardized scores for an overall composite (b = 0.00, p = .97) and each subtest (-0.16 ≤ b ≤ 0.16, ps > .05). Pre-registered exploratory analysis showed significant impacts on the number line estimation for kindergarteners (b = 0.23, p = .01) and students with below-average math knowledge at pre-test (b = 0.20, p = .03), driven by impacts on the 0 to 10 number line for kindergarteners with low prior knowledge (b = 0.27, p = .03) and on the 0 to 100 number line for first and second graders with low prior knowledge (b = 0.53, p = .02). Unexpectedly, a negative impact was seen for female students in addition problems correctly solved using complex strategies (verbal counting minimum, decomposition, and retrieval; b = - 0.52, p = .03). Implications and possible explanations of the results will be discussed, as well as future directions for the next iteration of the project, including co-design with teachers to align math learning goals from the basketball games to classroom instruction and greater grade level differentiation of content.

Authors

• Vanessa Noemy Bermudez, University of California - Irvine
  Presenting Author

• Kreshnik Begolli, University of California, Irvine

• Daniela Alvarez-Vargas, University of California - Irvine

• Jessica Paola Lopez Perez, University of California - Irvine

• Evelyn Santana, University of California-Irvine

• Drew Bailey, University of California - Irvine

• Lindsey Engle Richland, University of California - Irvine

• Andres Sebastian Bustamante, University of California - Irvine