Community Monitoring and Public Service Provision at Scale: An RCT of an Early Childhood Development Program in Nicaragua

How to cite

Boo, Florencia Lopez; Leer, Jane; Kamei, Akito. 2020. "Community Monitoring and Public Service Provision at Scale: An RCT of an Early Childhood Development Program in Nicaragua." AEA RCT Registry. AEARCTR-0006434. https://doi.org/10.1257/rct.6434-1.1

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

Florencia Lopez Boo* , Jane Leer , Akito Kamei

* Primary investigator

Registry ID
AEARCTR-0006434
DOI
https://doi.org/10.1257/rct.6434-1.1
First registered
Last updated
Status
Completed
Trial dates
2013-07-15 – 2018-05-15
Countries
Nicaragua ()
Registry record
http://www.socialscienceregistry.org/trials/6434

Abstract

Expanding small-scale interventions without lowering quality and attenuating impact is a critical policy challenge. Community monitoring offers a low-cost mechanism to ensure quality by making service providers accountable to local citizens, rather than distant administrators. This paper evaluates the impact of a large-scale home visit parenting intervention implemented by the Nicaraguan government, with two types of monitoring: (a) institutional monitoring; and (b) community monitoring.

Intervention

The experimental evaluation of the Home Visit program consists of a sub-sample of 210 communities from 33 municipalities. The 210 communities in the evaluation sample were randomly assigned to intervention and control conditions, with randomization stratified by municipality. Assignment to the control condition implied no home visits until after the study period. Treatment Arms 1 and 2 received home visits, and in Treatment Arm 2, communities were also supported to implement a community-based monitoring strategy in addition to the standard institutional monitoring that both treatment arms received.

Experimental design

The experimental evaluation of the Home Visit program consists of a sub-sample of 210 communities from 33 municipalities. The 210 communities in the evaluation sample were randomly assigned to intervention and control conditions, with randomization stratified by municipality. Assignment to the control condition implied no home visits until after the study period. Treatment Arms 1 and 2 received home visits, and in Treatment Arm 2, communities were also supported to implement a community-based monitoring strategy in addition to the standard institutional monitoring that both treatment arms received.

Design details

Randomization method
Randomization done in office by a computer.
Randomization unit
Community level. Communities in our sample include 150 households on average, of which between 30-50% have children younger than three.
Clusters
215 communities.
Observations
1325 children
Arms
420 children from control communities 442 children from home visit 420 children from home visit + community monitoring
Primary outcomes
Child development outcomes (Denver Language, Denver Fine Motor, and receptive vocabulary as measured by the PPVT)
Secondary outcomes
Child behavior and parenting measure

Cited by

No citation data available for this registration yet.

Keywords