x

Please fill out this short user survey of only 3 questions in order to help us improve the site. We appreciate your feedback!
Parental Interactions and Early Childhood Development at Home
Last registered on January 04, 2021

Pre-Trial

Trial Information
General Information
Title
Parental Interactions and Early Childhood Development at Home
RCT ID
AEARCTR-0006955
Initial registration date
December 26, 2020
Last updated
January 04, 2021 9:16 AM EST
Location(s)
Region
Region
Primary Investigator
Affiliation
Innovations for Poverty Action
Other Primary Investigator(s)
PI Affiliation
Inter-American Development Bank
PI Affiliation
Innovations for Poverty Action
Additional Trial Information
Status
On going
Start date
2020-07-11
End date
2021-03-31
Secondary IDs
Abstract
In this study we conduct a two-stage experiment to estimate the direct and spillover effects of providing text messages to parents with actionable advice on how to support early childhood development at home. We examine whether information is transferred within parents' networks and the impacts on children's cognitive outcomes.
Registration Citation
Citation
Hernandez-Agramonte, Juan Manuel, Olga Namen and Emma Naslund-Hadley. 2021. "Parental Interactions and Early Childhood Development at Home." AEA RCT Registry. January 04. https://doi.org/10.1257/rct.6955-1.0.
Experimental Details
Interventions
Intervention(s)
Using baseline data on parental networks within public childcare services, we study the direct and spillover effects of receiving an informational intervention to improve early childhood development. In particular, we examine whether there is transmission of learning between parents about how to promote children's cognitive development at home. The information is provided as part of a three-month text messaging campaign to support parental engagement in early childhood development during COVID-19. The program consists of more than 50 messages targeted to low-income parents with children aged 4-5 years old. The content of the texts includes information and actionable advice that shows parents how to promote their children's early numeracy and literacy skills in their home environment. In addition, we include motivational messages once a week with positive reinforcements to encourage parents to practice these activities. The messages are designed using behavioral tools to correct for inaccurate beliefs, assist them with goal setting and planning and develop a growth mindset for their role as parents.
Intervention Start Date
2020-08-25
Intervention End Date
2020-12-05
Primary Outcomes
Primary Outcomes (end points)
We estimate direct effects and spillover effects on parental engagement in educational activities, children's numeracy skills and children's literacy skills.
Primary Outcomes (explanation)
Costa Rica: children's numeracy and literacy skills are collected using scores from a standardized remote assessment test on the phone and through a parent survey. Parental engagement is measured by an index constructed using parents' reports about the time spend with their children on educational activities and other productive routines and investments.

El Salvador: parental engagement is constructed in the same way. Children's cognitive outcomes are measured using parents' reports.
Secondary Outcomes
Secondary Outcomes (end points)
Parent-parent and parent-teacher interactions, parental wellbeing, parental perceptions of their role as caregivers, discipline practices at home and children's socio-emotional outcomes.
Secondary Outcomes (explanation)
Parent and teacher interactions are measured using teachers' and parents' surveys. Parental wellbeing, perceptions and discipline practices are captured by parents' self-reported measures. Children's socio-emotional outcomes are collected through parents' reports on children's wellbeing and behaviors.
Experimental Design
Experimental Design
We implement a two-stage randomized experiment by first randomly dividing parental networks into two groups: pure control networks and treatment networks, and then randomly assigning half of the parents of the treatment networks to either receive the text messaging intervention (treated group) or to not receive the intervention (untreated group). This experimental design allows us to estimate spillover effects as the difference in outcomes between untreated parents in treated networks and parents in pure control networks. Parents in the control group and the untreated group receive placebo messages with information about existing resources from the regular services. All groups receive regular virtual programs from the public childcare services.
Experimental Design Details
Not available
Randomization Method
Randomization done in office by a computer
Randomization Unit
Two-stage randomization. First, we conduct a group-level randomization at the parental network level to divide pure control networks and treated networks. Second, we conduct an individual-level randomization within treated networks to randomly assign half of the parents to either receive the text message intervention (treated group) or not receive the intervention (untreated group).
Was the treatment clustered?
Yes
Experiment Characteristics
Sample size: planned number of clusters
691 parental networks (Costa Rica)
1270 parental networks (El Salvador)
Sample size: planned number of observations
4496 parents (Costa Rica) 8431 parents (El Salvador)
Sample size (or number of clusters) by treatment arms
Costa Rica: we randomly assign 338 parental networks (2174 parents) to the pure control group and 353 parental networks to the treatment group (2322 parents). Within the treatment group, 1072 parents receive the text messaging intervention (treated group) and 1250 parents do not receive the intervention (untreated group).

El Salvador: we randomly assign 635 parental networks (4161 parents) to the pure control group and 635 parental networks to the treatment group (4270 parents). Within the treatment group, 2135 parents receive the text messaging intervention (treated group) and 2135 parents do not receive the intervention (untreated group).
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
Costa Rica: our calculations indicate that assuming an intra-cluster correlation of 0.1, a probability of error type I of 0.05, a power of 0.8, 338 control networks in the control group, 353 networks in the treatment group, an average of 6.4 parents per network in the control group and an average of 3.2 parents per network in the treatment group, the minimum detectable effect for the parental engagement index is 1.9 percentage points and 0.12 SD in cognitive outcomes. El Salvador: Our calculations indicate that assuming an intra-cluster correlation of 0.1, a probability of error type I of 0.05, a power of 0.8, 635 control networks in the control group, 635 networks in the treatment group, an average of 6.5 parents per network in the control group and an average of 3.3 parents per network in the treatment group, the minimum detectable effect for the parental engagement index is 1.5 percentage points and 0.087 SD in cognitive outcomes.
IRB
INSTITUTIONAL REVIEW BOARDS (IRBs)
IRB Name
Innovations for Poverty Action
IRB Approval Date
2020-05-15
IRB Approval Number
15517