Randomised impact evaluation of a CBT-based intervention to foster socioemotional skills in vulnerable youth in Brazil

2024-06-01

2024-12-15

1. Impulsiveness and aggression of students using instruments validated in the liter- ature to assess the behavioural construct of impulsiveness (Barratt Scale/BIS-11, Patton et al. (1995); Malloy-Diniz et al. (2010))
2. Mental health of teachers using instruments validated in the literature to assess emotional problems, hyperactivity, relationship, conduct, and pro-social behaviour (SDQ Adult; Goodman et al. (1998)).
3. Mental health of students using instruments validated in the literature to assess emotional problems, hyperactivity, relationship, conduct, and pro-social behaviour (SDQ Youth; Goodman (1997); Stivanin et al. (2008)).
4. Empathy of students using instruments validated in the literature to assess em- pathy and its several dimensions (IRI/EMRI, Davis (1983); Koller et al. (2001).
5. Grit of students (Duckworth et al. (2007); Noronha and Almeida (2022)).
6. Self-efficacy of students using instruments validated in the literature to assess their beliefs in his or her capacity to learn, relating indicators of production and learning-oriented behaviour (Self-Efficacy Assessment Form, Cristina Medeiros et al. (2000)).
7. Values (Schwartz scale).
8. School Engagement of teachers using instruments validated in the literature to assess their engagement with school activities (Scale of School Engagement, Finn et al. (1995); Silveira and Justi (2018)).
9. Attention control (Attentional Control Scale, Filgueiras et al. 2015).
10. Self-esteem (Rosenberg Self Esteem Scale, Pimentel et al. (2018)).
11. School Engagement of students using instruments validated in the literature to assess their engagement with school activities (Scale of School Engagement).
12. School Violence (administrative data from Brazilian Ministry of Education – ”Prova Brasil”).
13. Educational outcomes on students in primary education: school Performance, educational attainment, probability of enrolling in high school (administrative data from Brazilian Ministry of Education – ”Prova Brasil” and ”Censo Escolar”).
14. Long-term involvement in crime: incarceration (administrative data from Brazil- ian Ministry of Justice).
15. Long-term impacts on health and violence: violent death (administrative data from Brazilian Ministry of Health – ”DataSUS Database”).
16. Long-term impacts on participation in anti-poverty programs (administrative data from Brazilian Ministry of Social Assistance – ”Cadastro U ́nico Database”).
17. Long-term impacts on the labor market: formal employment, wages, use of unem- ployment benefits (administrative data from Brazilian Ministry of Labor – ”RAIS Database”).

We propose an experimental impact evaluation of the SEJA Methodology through a two-level cluster-randomised trial (CRT), where the second level is the school, and first level is the student (Malmberg-Heimonen and Tøge, 2017). By randomising schools into treatment and control groups and following the students over time, this project will examine short and long-term results on a number of outcomes. On the short- term, we look at outcomes such as socioemotional skills, academic performance, school frequency, and enrolment in high school. We will also look at long-term outcomes such as victimisation, labour market participation, wage in the formal market, participation in anti-poverty programs, and criminal engagement.

Not available

Two-level cluster-randomised trial (CRT).

The second level is the school, and first level is the student.

Yes

Our study will involve a sample of schools, with 21 treated schools out of a total of 43 schools in Canoas and 26 treated schools out of 53 total schools in Pelotas.

44,407 students across all schools in both cities.

21 treated schools out of a total of 43 schools in Canoas and 26 treated schools out of 53 total schools in Pelotas.

Pre Analysis Plan