日本語タイトル:CAI(Computer-aided instruction)は生徒の認知能力を上昇させるのか?

Does Computer-aided Instruction Improve Children's Cognitive and Non-cognitive Skills?: Evidence from Cambodia

執筆者 伊藤 寛武 (慶應義塾大学)/葛西 慧子 (慶應義塾大学SFC研究所)/中室 牧子 (慶應義塾大学)
発行日/NO. 2019年5月  19-E-040
研究プロジェクト 医療・教育サービス産業の資源配分の改善と生産性向上に関する分析
ダウンロード/関連リンク

概要

本論文は、コンピュータによる学習支援が学力や非認知能力に与える影響を、クラスターランダム化比較試験の設計で推定した。カンボジアの5つの公立小学校の1-4年生1,500人を、算数の授業で花まるラボの開発した「Think!Think!」という教育アプリを使用するグループ(介入群20クラス)と、通常の授業を実施するグループ(対照群20クラス)にランダムに割りあて、3ヵ月後の学力や非認知能力への変化を比較した。その結果、小学校3-4年生の学力は0.56-0.67SD上昇し、大学進学への意欲も改善していた。しかし、学習意欲等の非認知能力への変化は観察されなかった。

概要(英語)

This paper examines the causal effect of computer-aided instruction (CAI) on children's cognitive and non-cognitive skills. Closely working with the Cambodian government, we ran clustered-randomized controlled trials at five elementary schools near Phnom Penn for three months. Students were randomly assigned into 20 treatment classes which were allowed to use an app based on CAI instead of regular math classes during the intervention, or 20 control classes. Our empirical results drawn from these experiments suggest that average treatment effect on cognitive skills measured using several types of math achievement tests and IQ tests is positive and statistically significant. The effect is significantly large, especially as compared with prior literature examined in developing countries: the estimated coefficients on student achievements are 0.56-0.67 standard deviations and IQ scores 0.70 standard deviations even after controlling for demographic factors. Furthermore, it is found that that CAI-based app can increase the students' subjective expectation of being able to attend tertiary education in the future. However, there is no significant effect on non-cognitive skills, such as motivation.