Improving remote learning in the context of a developing country: a tutoring intervention.

The learning crisis caused by the covid-19 pandemic is unprecedented. As governments seek to address the learning loss, so far, we know tutoring has been a widely proposed solution because it has proven to have a big positive effect on learning. We propose an RCT which can contribute to the literature of tutoring by looking at the interactions between interventions such as tutoring and tracking, and how socio emotional learning can affect the student’s self-perception, as well as group peer effects. Additionally, we can contribute to the very limited existing causal literature on peer tutoring and its effects on learning. We will be working with k-12 school students from 3rd to 12th grade and focusing on math tutoring. The tutors will be recruited among the universities and normal schools across Mexico.

External Link(s)

Registration Citation

Citation

Aguilar Llanes, Salome et al. 2023. "Improving remote learning in the context of a developing country: a tutoring intervention.." AEA RCT Registry. February 28. https://doi.org/10.1257/rct.7307-2.0

Sponsors & Partners

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information

Experimental Details

Interventions

Intervention(s)

We will recruit tutors (volunteers of universities and normal schools seeking to meet their social service requirements) assign them into WhatsApp groups of 5 kids that will meet 2 times a week. Tutors will be instructed to revise topics that students are studying in class, or past topics that the students are not clear on and needed to understand their curriculum. All tutors will receive initial training on how to teach math and socio-emotional development and we will also create a platform that tutors can use to communicate among each other and share tips and questions. We will try out different versions of these tutoring sessions, some will have an important socio emotional component. In some treatment arms students will be assigned to tutoring groups randomly and in others they will be assigned based on their performance on the baseline evaluation.

Intervention Start Date

2021-03-18

Intervention End Date

2023-07-01

Primary Outcomes

Primary Outcomes (end points)

Outcomes: 1) Learning measured by the Math and Spanish score on standardized evaluations constructed by us based on questions from PLANEA, a national test. We will apply this evaluation before and after the intervention. 2) Self-perception measured by the answers of a survey. 3)Student tutor interactions measured by using a machine learning approach to extract information of the recordings of the tutoring sessions

Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)

We will try to study the differentiated effects of the tutor teaching it is a university student or a normal school student. Additionally, we will look, ex post at the differential success of the intervention depending on the goodness of match between students and tutors for example by gender or region.

Secondary Outcomes (explanation)

Experimental Design

Grades will be randomly assigned to treatments and tutors too. Then they will be randomly matched within each group. Randomization will be stratified based on baseline scores, gender, and past participation in other iterations of tutoring program by the same group.

Experimental Design Details

Randomization Method

We will do a stratified randomization using a computer.

Randomization Unit

The unit of randomization will be at the student level.

Was the treatment clustered?

Yes

Experiment Characteristics

Sample size: planned number of clusters

We have around 16,000 students. Number subject to change due to open enrollment of both tutors and students

Sample size: planned number of observations

Around 16,000 students. Number subject to change due to open enrollment of both tutors and students

Sample size (or number of clusters) by treatment arms

4,000 kids in socio emotional learning (SEL) + tracking
4,000 kids in socio emotional learning (SEL) + no tracking
4,000 kids in just math + no tracking
4,000 kids in just math + tracking

Minimum detectable effect size for main outcomes (accounting for sample design and clustering)

running simmulations with historical data we have an MDE of 0.070 standard deviation in math scores with no attrition, and 0.084 with 30% attrition.

Supporting Documents and Materials

IRB