The Tutor is Calling: Phone-based Instruction as a Cheap, Effective Tool for Targeted, Remedial Instruction

• In Event: Remedial Education: Promising Platforms & Technologies

Proposal

This paper will present key insights and evidence from several studies on effective methodologies to bring targeted instructional practices to a phone-based setting.


Basic mobile phones are an inexpensive and widespread tool that has been shown to improve learning in several contexts. An initial phone-tutoring study conducted during COVID-19 in Botswana demonstrated that learning gains were particularly large when instruction was targeted (Angrist, Matsheng, and Bergman 2022). This program, called ConnectEd, has been further tested through replication studies in an additional five trials.
A key element of the ConnectEd model is to use cheap and widespread technology to deliver proven pedagogical models that support learning. Targeting instruction has been shown to work when delivered by teachers or volunteers (Banerjee et al. 2007; Banerjee et al. 2010; Banerjee et al. 2017; Duflo et al. 2020) and when using personalized learning software (Banerjee et al. 2007; Muralidharan, Singh, and Ganimian 2019, Büchel et al., 2022), but are only recently seeing widespread use in remote, one-on-one tutoring settings.


The targeting system behind ConnectEd uses ‘checkpoints,’ or simple problems of the day, to assess student understanding at the conclusion of weekly tutoring sessions. Although simple, accurate targeting can be difficult to administer. This is especially true in school contexts where teachers adhere to a standard curriculum.
To this end, Youth Impact and ConnectEd implementing partners have distilled several best practices to promote effective targeting. These include administering practicum-based training, deploying strong monitoring systems to facilitate timely and effective feedback and working with local partners to develop a context-aligned delivery system.
Finally, data from a 5-country replication trial conducted by Youth Impact and a coalition of partners showed that targeted instruction could improve over time and corresponded to larger treatment effects (Angrist et al., 2023b). This further suggests that testing optimized targeting methods could yield further learning benefits for enrolled children.


The paper will conclude with a future learning agenda around phone-based targeted instruction geared towards cost-effectiveness, impact and scale.

Authors

• Thato Letsomo, Youth Impact

• Colin Crossley, Youth Impact