Measuring Childcare Arrangements in El Salvador

How to cite

Contreras, Ivette; Palacios-Lopez, Amparo; Oseni, Gbemisola; Sagesaka, Akiko. 2023. "Measuring Childcare Arrangements in El Salvador." AEA RCT Registry. AEARCTR-0012123. https://doi.org/10.1257/rct.12123-1.0

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

Ivette Contreras* , Amparo Palacios-Lopez , Gbemisola Oseni , Akiko Sagesaka

* Primary investigator

Registry ID
AEARCTR-0012123
DOI
https://doi.org/10.1257/rct.12123-1.0
First registered
Last updated
Status
In development
Trial dates
2023-10-09 – 2023-12-09
Countries
Private
JEL codes
J13, J22, C83
Registry record
http://www.socialscienceregistry.org/trials/12123

Abstract

Expanding access to quality childcare has the potential to yield multi-generational impacts by improving parents’ economic empowerment, child outcomes, family welfare, business productivity and overall economic growth. However, there is very limited data on childcare available and guidelines on how to collect it across low- and middle-income countries, particu¬larly for children below age 3, and this limits the possible assessment of access of childcare services in developing countries.

The objective of this project is twofold. First, it aims to advance the understanding of how to obtain better measures of childcare arrangements in developing countries using household surveys, including those informal childcare arrangements that can be overlooked but are crucial to families. Second, it will explore parents’ preferences on childcare services’ attributes.

The first goal will be addressed by designing and implementing a survey experiment to compare different methods to collect information on household surveys. The evidence on parents’ preferences on childcare services attributes will be obtained from a discrete choice experiment inspired in Datta (2019).

Intervention

Measuring childcare is difficult even in developed countries. For example, when measuring the time invested in childcare is important to consider that there are different levels of care depending on the ability of multi-tasking: primary childcare, secondary childcare, and time spent with children. (Allard, M. et al. 2007). In addition, parental investments are higher in the first years after childbirth; considering the decrease in the fertility rate that these countries had experienced, thus drawing conclusions with a small sample of households with children under 3 years of age could be challenging (Devercelli, A. and Beaton-Day, F. 2020).

Collecting data about childcare in developing countries has specific challenges: (1) more childcare informal arrangements could be present; (2) parental unawareness of childcare options can lead to an imperfect measure of the factors that affect the demand for childcare services (Mateo-Diaz, et al., 2013); and (3) cultural and social norms could play a role on childcare decisions (World Bank, 2022).

There are two interventions:
• Parents or caregivers of children under 8 years of age that live in households assigned to treatment group 1 will be exposed to a detailed survey module to collect information of all the different childcare arrangements that the children needed in the past 4 weeks.
• Parents or caregivers of children under 8 years of age that live in households assigned to treatment group 2 will be exposed to a module that captures all the child activities in the last 24 hours and the people that were with the child at each hour.

Moreover, all households will respond to a discrete choice experiment that aims to measure respondents' willingness to pay for childcare services attributes.

Experimental design

Using Enumeration Areas (EAs) from the Salvadoran Population Census 2007 in rural and peri-urban areas of the provinces of San Salvador and Usulután, 48 EAs were randomly selected to participate in the study. A complete household listing has been conducted in the 48 EAs. In each EA, 21 households will be selected at random, and all households will be assigned to C, T1, or T2. The criterion used to screen the households is that the household includes at least one child under 8 years of age.

In each experiment arm comparable demographic and socio-economic data will be collected at the individual- and household-level. The data will be useful in exploring the heterogeneity of survey treatment effects.

Group status: Control group
Short module (main childcare arrangement)

Group status: Treatment group 1
Detailed modules on childcare (list of childcare providers) + Short module (main arrangement)

Group status: Treatment group 2
Child activities module+ Short module (main arrangement)

Design details

Randomization method
Computer randomization for selection into experiment arms: C, T1, or T2.
Randomization unit
Unit of randomization will be the household.
Clusters
1008 households
Observations
1,512 caregivers
Arms
336 households C, 336 households T1, 336 households T2
Primary outcomes
Number of childcare arrangements per child
Secondary outcomes
Willingness to pay for childcare services attributes.

Cited by

No citation data available for this registration yet.

Keywords