SRCD 2019 Biennial Conference
March 21-23, 2019
Title: Benefits of Playing Tablet Games on Kindergarten Children’s Numerical Knowledge
Session: Innovative Approaches at Home and School to Improve Children’s Early Math Skills
Authors: Geetha Ramani, Emily Daubert, Grace Lin, Snigdha Kamarsu, Alaina Wodzinski, Susanne Jaeggi
Abstract: Mathematics knowledge in early childhood can have a large impact on children’s later mathematical achievement even through adolescence (Watts, Duncan, Siegler, & Davis-Kean, 2014). Contributors to variation in mathematical achievement include both numerical knowledge and underlying cognitive processing abilities (Geary & Hoard, 2005). The current study tested the benefits of tablet-based training games that targeted each of these areas for improving the mathematical knowledge of kindergarten-age children from diverse backgrounds. We hypothesized that playing a number-based game targeting numerical magnitude knowledge would improve children’s broader numerical knowledge, and playing a working memory (WM) game also would improve children’s numerical knowledge given its important underlying role in mathematics achievement.
We randomly assigned kindergarteners into one of three conditions (n=148; 46% boys; Mage=6 years 0 months; 40% of household incomes less than $45,000 a year). In the domain-specific condition, children played a board game with numbers 0-100, resembling Chutes and Ladders, against a computer opponent. In the domain-general condition, children played a WM game that involved recalling increasingly long sequences of stimuli. In the active control condition, children played a color-version of the board game. Children played the games for ten 15-minute sessions at their schools. Five numerical knowledge assessments served as the pretest and posttest measures: counting principles (“Can you count these stars and tell us how many there are?”), 0-100 number line estimation (“Where does N go on the number line?”), symbolic magnitude comparison (“Which is more?”), numeral identification (“What number is this?’’), and arithmetic problems (“What is 4 + 2?”). Children were administered the same measures at a one-month follow-up visit. Structural equation modeling (SEM) was used to model learning gains across time. A multi-group comparison using a latent change score model was conducted using repeated measures of children’s numerical knowledge. The two-step approach using Full Information Maximum Likelihood (FIML) to handle missing data was utilized. In the model, the five numerical knowledge variables measured at pretest, posttest, and follow-up were used to create three latent factors indicating children’s general math performance at the three time points.
Controlling for pretest numerical knowledge, preliminary analyses revealed that children who played the number game significantly improved their numerical knowledge from pretest to posttest above and beyond the improvements during the same time period compared to those in the control group, p < .001 (Figure 1A). They maintained their benefits one month later as indicated by the non-significant difference in the amount of change in numerical knowledge from posttest to follow-up, ps = .494. Children in the WM condition did not improve significantly more than children in the control condition between pretest and posttest, ps = .180. However, they showed significantly greater gains than children in the control group between posttest and follow-up, p < .01 (Figure 1B). Overall, our results suggest that playing the number game improved children’s numerical knowledge, and these improvements remained stable for at least a month. Importantly, playing WM games also benefited children’s numerical knowledge, but there may be a delay in the emergence of these benefits.