Download StudyGeneral

Study Overview

Long Term Impact of the Juntos Conditional Cash Transfer Program on Cognitive and Anthropometric Outcomes Among a Cohort of Peruvian Children
Study is 3ie funded:
Study ID:
Initial Registration Date:
Last Update Date:
Study Status:
Change History for Status
Changed On Previous Value
07/15/2022 Ongoing

This paper investigates the impact of participation in the Juntos conditional cash transfer (CCT) program in Peru on child anthropometric and cognitive outcomes. It specifically intends to identify causal pathways that may have helped to facilitate improvements in height-for-age (HAZ) and body mass index-for-age (BAZ) z-scores (see: Andersen et al. 2015) using longitudinal data from the Young Lives cohort study. This study will investigate causal pathways behind those results by using propensity score matching and a triple difference approach to determine whether improvements in anthropometric outcomes may have been driven by the presence of local health clinics that offer nutritional checkups and child health assessments. This study further investigates the longer-term impacts of the program (and modification effects) into data collected in 2014 (4 years after endline data used in Andersen 2015). We hypothesize that the program impacts were modified by a) the availability of (and/or distance to) health facilities and b) the availability of specialty services including pediatrics and nutrition counseling

Registration Citation:

Cameroon, D. and Andersen, C., 2016. Long Term Impact of the Juntos Conditional Cash Transfer Program on Cognitive and Anthropometric Outcomes Among a Cohort of Peruvian Children. Registry for International Development for Impact Evaluations (RIDIE). Available at: 10.23846/ridie094

Health, Nutrition, and Population
Additional Keywords:
conditional cash transfer, health facilities, child anthropometrics, cognition, growth,
Secondary ID Number(s):

Principal Investigator(s)

Name of First PI:
Drew Cameron
UC Berkeley
Name of Second PI:
Chris Andersen
Harvard University

Study Sponsor

Bill & Melinda Gates Foundation; Eunice Kennedy Shriver National Institute of Child Health and Development; Grand Challenges Canada; UK Aid; Netherlands Ministry of Foreign Affairs
Study Sponsor Location:
United Kingdom

Research Partner

Name of Partner Institution:
UC Berkeley
Type of Organization:
Research institute/University
United States

Intervention Overview


The Juntos CCT is an ongoing conditional cash transfer program enacted by the Peruvian government. The program delivers monthly cash transfers of S./ 100 (soles, approx. $30 USD) to eligible families (those below a certain poverty threshold, with at least one child of eligible age (under 14 years), or a pregnant mother). Receipt of the monetary transfer is contingent upon 3 sets of conditions: (1) [For children under 5] Attend regular health and nutrition check-ups (for periodic monitoring of height and weight, complete series of vaccinations, iron and Vitamin A supplements and anti-parasite checks); (2) [For children 6-14] School attendance at least 85% of the school year; (3) [For pregnant and breast-feeding mothers] Attend prenatal and post-natal checks (tetanus vaccination, folic acid and iron supplements and anti-parasite checks). Objectives include: (a) short run reductions in poverty (b) long run reduction in intergenerational poverty through improved access to education and health. For more details see Perova and Vakis (2009).

Private Intervention Details:
Theory of Change:

This evaluation focused on the role of cash transfers, and the intermediary role of proximity to physical infrastructure, as drivers of anthropomorphic change among program recipients. 

Multiple Treatment Arms Evaluated?

Implementing Agency

Name of Organization:
Government of Peru: Presidential Council of Ministers
Type of Organization:
Public Sector, e.g. Government Agency or Ministry

Program Funder

Name of Organization:
Government of Peru
Type of Organization:
Public Sector, e.g. Government Agency or Ministry

Intervention Timing

Intervention or Program Started at time of Registration?
Start Date:
End Date:
Evaluation Method

Evaluation Method Overview

Primary (or First) Evaluation Method:
Other (not Listed) Method:
Additional Evaluation Method (If Any):
Difference in difference/fixed effects
Change History for Additional Evaluation Method (If Any)
Changed On Previous Value
07/15/2022 Other (specify)
Other (not Listed) Method:
Change History for Other Method
Changed On Previous Value
07/15/2022 Difference-in-difference-in-differences (triple-difference estimation)

Method Details

Details of Evaluation Approach:

Following the initial design of Andersen and others, this study will use propensity score matching and double difference estimation to investigate the causal impact of participation in the Juntos conditional cash transfer program on child anthropometric and cognitive outcomes. The analysis will then employ a triple-difference (DDD) estimation to examine the moderating effects (effect modification using interaction terms) of the availability of health facility infrastructure and services. For the DDD estimation, we are interested in estimating how the Juntos program affects child growth differently for children living in communities with and without a health facility. Ideally, we would observe the same child with and without Juntos and with and without a health facility in their communities. Our estimates will measure the difference in the HAZ and TVIP (Peabody Picture Vocabulary Test) scores from 2002-14 for Juntos recipients relative to children that did not receive Juntos (our double-difference (DD) estimate). We then take the difference in our DD estimates for children with and without a health facility, which constitutes our target parameter: the DDD.

Private Details of Evaluation Approach:
Outcomes (Endpoints):

The main outcomes of interest represent changes from round 1 (~2000) to follow up in round 3 (~2010) and round 4 (~2014). Interview dates vary by household These variables are: Difference in height for age z-scores (between round 1 (R1) and R3; and between R1 and R4) Stunting (binary variable based on above HAZ scores around a specific cutoff (-2 zscore)) Difference in BMI for age z-scores (between R1-R3; and R1-R4) Overweight (binary variable based on above BAZ (BMI for age z) around a specific cutoff (+2 zscore)) Difference in weight for age z scores (between R1-R3) - data not available for R4 Difference in TVIP scores (normalized for age) Exposure variables include binary exposure to Juntos (y/n), continuous exposure (in days, months, or years), and binary year cutoff exposures (+/- 2 years; +/- 4 years; etc. depending on logical distribution of exposure times in each of R3 and R4)

Unit of Analysis:
Individual child. We will also consider looking at average outcomes by cluster, though this may not be possible depending on power considerations.

We test multiple hypotheses: H1: Previous results (by Andersen et al. 2015) show statistically significant increases in HAZ among boys exposed to Juntos for less than 2 years, and reduction in overweight among girls. We hypothesize that these results will be similar in the longer term (including 2014 end-line). H2: We do not expect to see significant impacts on TVIP scores among the cohort or either gender group in 2014. H3: We hypothesize that the availability of a health facility within the village of residence will have a modification effect on the HAZ and BAZ scores (effect sizes greater (+ for HAZ and - for BAZ)) for results both in R3 and R4. H3a: We expect that this modification effect will be apparent through various measures of exposure to health facilities (presence; distance; quality; services) as available. H4: We expect that the presence of a health facility will not have a modification effect on TVIP scores in R3 or R4.

Unit of Intervention or Assignment:
Households (families received same amount of cash transfer regardless of number of children)
Number of Clusters in Sample:
Not cluster randomized
Number of Individuals in Sample:
Around 1600 children (accounting for round to round attrition, response error, and child movement)
Size of Treatment, Control, or Comparison Subsamples:
In replication of original results (2010 end-line data): ~300 treatment children, ~1000 controls; In long term analysis (2014 end-line data): ~500 treatment children, ~1000 controls

Supplementary Files

Analysis Plan:
PAP 6-8-16 Cameron et al..docx
Other Documents:

Outcomes Data

The data used for this evaluation is from the Young Lives Cohort Study. This is a secondary data source tracking two age-cohorts of children in each of 4 different countries. We will be using the younger cohort age group from Peru. More information available online at:
Data Already Collected?
Data Previously Used?
Data Access:
Restricted -- Access requires a formal approval process
Data Obtained by the Study Researchers?
Data Approval Process:
Registration with the UK Data Archive and a brief description of the project proposal was required. Majority of data were originally obtained in 2015 for a separate project. Additional datasets that are intended to be for public consumptions (mini-community data for R3 and R4; updated constructed peru file for R1-R4) was not available on the website or web-portal and was instead obtained through contact with data administrators Anne Solon (on 13 July 2016) and Alan Sanchez (on 22 July 2016).
Approval Status:
Yes-obtained approval and have received the data

Treatment Assignment Data

Participation or Assignment Information:
Data Obtained by the Study Researchers?
Data Previously Used?
Data Access:
Data Obtained by the Study Researchers?
Data Approval Process:
Approval Status:

Data Analysis

Data Analysis Status:

Study Materials

Upload Study Materials:

Registration Category

Registration Category:
Non-Prospective, Category 4: Data for measuring impacts have been obtained/collected by the research team and analysis for this evaluation has started

Completion Overview

Intervention Completion Date:
Data Collection Completion Date:
Unit of Analysis:
Clusters in Final Sample:
Total Observations in Final Sample:
Size of Treatment, Control, or Comparison Subsamples:


Preliminary Report:
Preliminary Report URL:
Summary of Findings:
Paper Summary:
Paper Citation:

Data Availability

Data Availability (Primary Data):
Date of Data Availability:
Data URL or Contact:
Access procedure:

Other Materials

Survey Instrument Links or Contact:
Program Files:
Program Files Links or Contact:
External Link:
External Link Description:
Description of Changes:

Study Stopped


Original study was successfully replicated using original young lives data. Updated Young Lives data included a new variable for a wealth index (which was constructed differently than in old iterations of the dataset). This led to non-replicable findings from the original study (using the newly constructed variable), thus original study outcomes were not sensitive to adjustments to the wealth variable in the matching process. Study findings from round 4 of the YL data collection did not substantially depart from the findings from Round 3 originally published in Andersen et al.. Main finding suggests that migration of more than half the study sample make causal claims of the Juntos program's effects on child anthropometry using round 4 of data collection for main outcomes difficult to determine, and even harder to publish.