Assistant Professor of Public Policy Michela Carlana, working with Eliana La Ferrara, professor of economics at Bocconi University in Milan, designed and carried out a project this spring to help improve educational outcomes for students across Italy. Hit early and hard by the coronavirus pandemic, Italy found its school year interrupted, with students and teachers sent home to adapt to an online learning environment that no one had anticipated. Carlana, an economist with an interest in education and inequality, developed a program to pair middle-school students from low-income families across Italy with university students acting as volunteer tutors.
More than 500 middle-schoolers from 78 schools across Italy took part in the program. Three hours of tutoring per week improved students’ performance by 4.7 percent—and, even more strikingly, increased well-being by 26 percent and socio-economic skills by 21.1 percent. Academic performance gains were doubled when tutoring was increased to six hours a week.
We spoke to Carlana about this project, which she and her research partner hope will be scaled up next school year.
Q: Why did you decide to undertake this project?
At the beginning of the pandemic, schools were closing. Italy was one of the first countries that decided to shut down schools, but a lot of other countries followed. And we realized that there were a bunch of university students that would be happy and willing to help children. They were at home; they had more time, and a spirit of being willing to volunteer to help others. And on the other hand, children from disadvantaged backgrounds—immigrant or low socio-economic status—needed help with the transition because their parents couldn't easily shift to online teaching, or they couldn't supervise the students during the time in which they were homeschooled. So, we decided to try to match these university students with students from disadvantaged backgrounds.
Students from schools all over Italy participated, but involvement was particularly strong in the north, in areas where there were a lot of cases of coronavirus. Those regions were the most receptive to this program.
We sent out a call, and we got more than 100 Italian schools to sign up for this project. It was more than we had the ability to deal with at that point because we hadn’t raised any grant money. We had over 1,000 applications from parents who applied through the schools. Beneficiaries were selected by the teachers, who made sure we could reach the most disadvantaged kids. A large number of tutors were willing to help, and we were able to match 500 students. We have been helping these kids from the middle of April until the end of May for three to six hours per week. It was intense tutoring.
Now the project is at the end. Many tutors got very attached to the students and are helping them still. They are not doing it for the project itself, but just to keep and build their relationships with the families.
Q: Where are you now in this work?
So, this is the final step, and we are collecting data to evaluate whether this tutoring was effective. Among the 1,000 applicants, we randomly selected the 500 the students that got into the tutoring program. And we kept the other 500 as a control group. We are trying to collect the data from parents of both the treatment and control group. We will be very careful in studying academic, behavioral, and social-emotional components. We are also interviewing the principals of all these schools and the parents and teachers of all these children, in both the treatment and control groups, in order to understand whether the children have been doing more homework, whether they were more engaged in classroom activities, what their grades were, and what happened along different dimensions.
Q: Do the initial results suggest the program is worth continuing next year?
An important aspect is that there was a lot of interest from the regional offices of the Italian Ministry of Education in our project. In some regions, they asked to partner with us for the next school year. So now we are applying for funding. This is a project that could be easily scalable at the international level by matching university students from urban areas with disadvantaged children all over the world.
The key intuition of this project is that, thanks to the virtual nature of the interaction, there will be no need for the tutor and the students to be from the same city or neighborhood. The intervention can expose students in need to tutors from different geographic, social, and cultural backgrounds. For a lot of these kids, the tutoring sessions were the first interactions they had with someone who attended university. Being able to bridge this divide, and being able to offer children exposure to higher aspirations and a different life path, could be very important, even if schools open again during the next academic year. So, this is what we are trying to get funding to do.
The pandemic has exacerbated the differences in opportunities and educational outcomes for children, and this could be a program that could systematically address these inequalities in the future.
Q: How did students interact with tutors? And how did you prepare the volunteer tutors?
The schools have a platform where they do online classes, and they had been providing tablets to some children. Internet connection was still an issue for some of them. We are collecting data about that. Around 20 percent of students did the tutoring sessions using their smartphones. However, tutors were able to share their screens and use digital whiteboards, which made the sessions more effective.
We recruited educational consultants and prepared an online course to help tutors, with slides and videos, but we also had several meetings with groups of tutors, making sure they were addressing all the issues that were coming up. Tutors also had an email address they could write to at any time if they had problems during the tutoring, and an expert in pedagogy would help them immediately. We were particularly concerned because students were learning from home. These are disadvantaged families. And, so, a lot of things could have been going on. We didn't want to leave these university students alone to deal with potential issues they may have faced. There was a lot of support and investment on our side.
Q: How did you tailor the program to help disadvantaged middle-school students (between 11 and 14 years old)?
Yeah, that age is particularly challenging because they need to develop their emotions and interact with others. This is why we pushed the social-emotional component of this program, and we are confident it will have some impact on the academic side.
These parents, from our understanding, tend to work outside the home. Most of them don’t have white-collar jobs where they can work from home. So, we expected to have low involvement from the parents. It’s a measure we collected. We asked the tutors how involved the parents were.
We have gotten very positive emails from principals, teachers, parents, and tutors. So overall, the feeling was very good. But as an economist, I always want to see the data to understand whether it was actually effective.