Child human capital production: A field experiment in the Solomon Islands and East Timor

We introduce and test an early learning intervention on child-directed speech in the Solomon Islands and East Timor, which aims to: (i) estimate the relative contributions of parental inputs in the child human capital production function, and (ii) identify the channels through which early childhood interventions can be effective. We provide parents with information and learning materials intended to stimulate child-directed speech. We then measure short-run (within days of the intervention) and medium-run (after 1 year) effects of the intervention on speech outcomes.

External Link(s)

Registration Citation

Citation

Baranov, Victoria et al. 2023. "Child human capital production: A field experiment in the Solomon Islands and East Timor." AEA RCT Registry. March 07. https://doi.org/10.1257/rct.4116-3.1

Sponsors & Partners

Experimental Details

Interventions

Intervention(s)

We provide parents with information about the importance of child directed speech. They are also given materials they can use with their children to stimulate child directed speech at home. A control group receives no intervention. We measure speech outcomes before and after the intervention and compare them across treatment v. control.

Intervention Start Date

2019-06-06

Intervention End Date

2019-08-20

Primary Outcomes

Primary Outcomes (end points)

Words spoken to the child by adults and children. Utterances by the child. Conversational turns.

Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)

Parental beliefs about the returns to child directed speech. Parental investment in children.

Secondary Outcomes (explanation)

Experimental Design

We assign treatment at the village level. In each village, we ask village leaders to assemble a community meeting, where we randomly select participants via public lottery. The selection criteria for attending the meeting is couples (male and female partners who consider themselves to be in a committed relationship) between the ages of 20 and 45, who have at least one child between the ages of 18 months and 3 years. We then randomly select 5 couples (10 participants) and their respective child to participate in the intervention by pulling numbers out of an urn and announcing them publicly in the community meeting.

We will conduct a follow-up 1 year after the initial intervention.

Experimental Design Details

Randomization Method

Treatment (village level) randomization done in office by a computer. Participants within each village selected via public lottery.

Randomization Unit

Village.

Was the treatment clustered?

Yes

Experiment Characteristics

Sample size: planned number of clusters

50 villages.

Sample size: planned number of observations

50 villages x 5 couples per village = 250 couples. 1 child per couple = 250 children. 2 parents per child = 500 parents.

Sample size (or number of clusters) by treatment arms

14 control villages and 36 treatment villages.

Minimum detectable effect size for main outcomes (accounting for sample design and clustering)

Informed by data from Weber et al. (2017) study in Senegal. We assume 0.8 power, alpha=0.05, and 25% correlation between clusters. First main outcome is words spoken to the child by the parent in a 5-minute session. Baseline mean of 228 words (SD=151). MDE=42% increase. Second main outcome is utterances by the child in a 5-minute session. Baseline mean of 35.5 utterances (SD=15). MDE=27% increase.

Supporting Documents and Materials

IRB