Experimental Evaluation of the Primeira Infância Melhor Program

How to cite

Costa, Gabriel Weber; Souza, André Portela; Carneiro, Pedro; Barros, Ricardo Paes de; Boo, Florencia Lopez; Lima, Lycia; Camargo, Juliana. 2021. "Experimental Evaluation of the Primeira Infância Melhor Program." AEA RCT Registry. AEARCTR-0006799. https://doi.org/10.1257/rct.6799-1.1

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

Gabriel Weber Costa* , André Portela Souza , Pedro Carneiro , Ricardo Paes de Barros , Florencia Lopez Boo , Lycia Lima , Juliana Camargo

* Primary investigator

Registry ID
AEARCTR-0006799
DOI
https://doi.org/10.1257/rct.6799-1.1
First registered
Last updated
Status
On going
Trial dates
2018-05-12 – 2024-06-30
Countries
Brazil ()
Registry record
http://www.socialscienceregistry.org/trials/6799

Abstract

This study aims to evaluate the 'Primeira Infância Melhor' Program (Better Early Childhood, PIM), an ongoing large scale home visiting program targeting socioeconomically disadvantaged families to promote the health, cognitive and socioemotional development of young children during their early years of development in the Brazilian State of Rio Grande do Sul. We explore program expansion within municipalities to randomize individuals (either children aged 0-3 or expectant mothers) into the treatment group or a waiting list whenever there is excess demand. We will assess program impacts on child development, parental practices, parent-child relationship and family access to services. This trial also sets up the building blocks of a longitudinal study design to evaluate medium and long-term effects of PIM on dimensions such as educational achievement and performance, labor market entry and health outcomes.

Experimental design

This study focuses on families with a child aged 0 to 3 or an expectant mother. Eligible individuals are mapped by local program staff. A stratum is defined by child age group and communities/municipalities. For each stratum, we randomize families into two groups: an active treatment and a waiting list. Home visitors then offer the program following the randomization protocol until local supply is met. In case a family is composed of more than one child, all children receive the same treatment status, but we only consider the youngest one for the evaluation sample.

Design details

Randomization method
Randomization done in office by a computer
Randomization unit
Individual (either a child or an expectant mother)
Clusters
Planned total of 3000 families. Currently 2422 in various stages of implementation, divided into 61 strata (mean strata size=39.7).
Observations
3000 children
Arms
Expected sample of 1500 treated x 1500 control. Final numbers are dependent on local program supply capacity. Currently 1192 treated x 1230 waiting list.
Primary outcomes
Motor, cognitive, socioemotional, and communicative development of the child; parenting practices. For future follow-up stages: parent-child interactions; finer measures of child development; educational, health outcomes of the child.
Secondary outcomes
Access to services; mental health of the caregiver. Subgroup analysis: child age, gender, socioeconomic status, educational level of the mother/caregiver, municipality/community, and interactions between these subgroups as sample sizes allow. For future follow-up stages: health outcomes of the child and caregiver/family members and long-run labor market outcomes of the child.

Cited by

No citation data available for this registration yet.

Keywords