Estimating the Treatment Effect of New Device Deployment on Uruguayan Students’ Online Learning Activity

Abstract

When implementing large-scale educational computing initiatives (e.g., One Laptop Per Child) it is vital to allocate resources for training, support, and device deployment judiciously. One question that arises is how learners’ engagement with online educational resources is affected by receiving a new computer; do the benefits justify the costs? In this paper, we perform a quasi-experimental analysis to measure the effect of new device deployment on students’ online learning activity, operationalized as either the number of interaction events with their LMS, or the number of attempted exercises in their math ITS. The focus is on 6th-grade learners in Uruguay, which to-date has delivered over 750,000 computers to pupils nationwide. Our results suggest that, relative to learners’ online learning activity before device deployment, the absolute effects are small but the relative effect are stat. sig. and surprisingly strong: the estimated relative increase on 2016 students’ overall LMS activity is 49%. The effects are positive for both 2015 and 2016 and persist several months after device delivery. Moreover, we find that students attempt to solve stat. sig. more (88%) math problems during the month after they receive a new device. We discuss possible reasons and implications for large-scale educational computing programs.