Does Computer-Aided Instruction Improve Children’s Cognitive and Noncognitive Skills? (Asian Development Review: Volume 38, Number 1

Despite the relatively short period of intervention of 3 months, the students were enthusiastic about using Think!Think! The drawbacks of our study may be the presence of evaluation-driven behavioral changes in the treatment group called the Hawthorn effect and/or in the control group called the John-Henry effect. [...] There is no statistically significant difference in the results of the NAT between the G3 students assigned to treatment classes and those assigned to control classes, although the G4 students in the control classes performed slightly better on the TIMSS than those in the treatment classes, even after controlling for school-by-grade fixed effects, following Bruhn and McKenzie (2009). [...] According to the descriptive statistics, the mean of the Tanaka B-type IQ test score is 78.612 with a standard deviation of 13.451, and the mean of the DAM type IQ score is 0.692 with a standard deviation of 0.207. [...] Our primary focus is the estimated effect of access to Think!Think! on the NAT for G3 students and on the TIMSS for G4 students in the first row of the table. [...] Although the difference in the DAM scores for the entire sample and even the interaction term with grades are not statistically significant, the skills of younger students seem to improve.