Digital support systems to improve child development in Peru: A cluster-randomized controlled open label trial
Data files
Feb 26, 2026 version files 1.17 MB
-
digitalECD_26Feb2026.csv
590.79 KB
-
digitalECD_26Feb2026.dta
511.85 KB
-
DigitalECD_codebook_26Feb2026.log
34.03 KB
-
README.md
1.81 KB
-
Trial_Analysis_Replication_File_26Feb2026.do
26.89 KB
Abstract
Digital technologies have the potential to transform early childhood development (ECD) interventions by delivering personalized support at scale. We conducted a cluster-randomized controlled trial in rural Peru to evaluate the effectiveness and cost-effectiveness of an AI-supported digital parenting chatbot as well as traditional home visits as interventions to improve child development. Among 2,461 caregiver-child dyads, both the digital and home-visiting interventions improved child development outcomes at 2.5 years of age, with standardized effect sizes of 0.11 and 0.17, respectively. At one-fifteenth the cost of in-person support, the digital intervention yielded superior cost-effectiveness. These findings suggest digital platforms can be a viable, scalable alternative to support children’s development in resource-constrained settings.
Dataset DOI: 10.5061/dryad.nzs7h4551
Description of the data and file structure
DigitalECD Replication files. The analysis of this trial was conducted using the Stata Statistical Software package. The DO and .DTA files below will create all the figures and tables. The .LOG file contains variable descriptions and value labels. The .csv file is meant as a backup for users not able to read .dta files or Stata code.
Files and variables
File: Trial_Analysis_Replication_File_26Feb2026.do
Description: Stata DO file creating all tables and figures.
File: digitalECD_26Feb2026.dta
Description: Stata file that is called in by the DO file for replicating analysis in Stata.
File: digitalECD_26Feb2026.csv
Description: CSV file containing main data. All analysis was conducted in Stata. We thus encourage users to use the .dta file, which contains a full codebook and a label for each variable. The CSV file is only meant as backup in cases users cannot access the original data.
Empty cells reflect missing data or data that were not collected.
File: DigitalECD_codebook_26Feb2026.log
Description: text file containing variable description and value labels (for categorical variables).
Code/software
The code that creates all figures and tables is in the DO file (and can only be run in Stata).
Access information
Data was derived from the following sources:
- This is the original data de-identifed by the lead author (G Fink).
Human subjects data
All participants have signed a written consent form and provided explicit consent to the publication of de-identified data.