The effectiveness of online tutoring in secondary school: Evidence from a randomized controlled trial

How to cite

Hupkau, Claudia; Roldan, Antonio; Gortazar, Lucas. 2021. "The effectiveness of online tutoring in secondary school: Evidence from a randomized controlled trial." AEA RCT Registry. AEARCTR-0007138. https://doi.org/10.1257/rct.7138-1.2

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Claudia Hupkau* , Antonio Roldan , Lucas Gortazar

* Primary investigator

Registry ID
AEARCTR-0007138
DOI
https://doi.org/10.1257/rct.7138-1.2
First registered
Last updated
Status
Completed
Trial dates
2021-03-16 – 2022-02-05
Countries
Spain (Madrid), Spain (Barcelona)
JEL codes
I2, I24, J24
Registry record
http://www.socialscienceregistry.org/trials/7138

Abstract

Evidence is mounting that there is a substantial socio-economic gap in learning losses experienced by students as a result of widespread school closures in response to the COVID-19 pandemic. This research evaluates the effectiveness of an online tutoring program aimed at students from highly disadvantaged schools in Spain. We will randomly assign students to tutors. Students will be taught in groups of two for a duration of eight weeks. We will evaluate the effect of the program both on cognitive as well as socio-emotional outcomes.

Intervention

The intervention consists of tutoring in maths with a focus on emotional wellbeing, delivered online over the course of 8 weeks (3 hours per week) in groups of 2 students.

Experimental design

Due to limited financial resources to pay additional tutors, among all students who register interest for the program, only some students will be randmly assigned to a mentor.

Design details

Randomization method
Randomization done in office by a computer.
Randomization unit
The unit of randomization are individuals. Randomization will be blocked at the classroom or year group level (depending on the number of students registering per classroom in a given school).
Clusters
We have registered interest to participate by about 380 students from 18 schools. These 380 students will be randomly assigned to one of three treatment arms: 1) tutoring and 2) control group (no tutoring).
Observations
We plan to have 380 pupils in our study.
Arms
The targeted sample size is 200 students for the treatment group, and 180 students in the control group.
Primary outcomes
We will study the impact of the intervention on cognitive as well as non-cognitive outcomes: Cognitive outcomes are measured by scores using a standardized test that will be administered online during school hours. In additional, we will be looking at end-of-year grades and whether the subject was passed, and whether the student had to repeat a school year. Non-cognitive outcomes will be measured via a self-completion survey to students covering socio-emotional skills, locus of control, aspirations, and wellbeing.
Secondary outcomes
Because students are assigned randmly to groups of 2 for their tutoring sessions (within the same classrooms or year group within school) we will further test whether (1) whether the ability match matters (measured through baseline maths test results) (2) whether the gender compisition of the group matters for effectiveness. Due to the fact that we randomly assign tutors to students, we will study heterogeneous effects of the intervention with respect to tutor-student gender match.

Cited by

No citation data available for this registration yet.

Keywords