Experimental Design
We will conduct this study in communities across six cities in Kenya: Kajiado, Kiambu, Kisumu, Nairobi, Mombasa, and Nakuru. Within each city, Kidogo will identify 60 communities into which they are planning to expand; all communities are sections of informal settlements. We will then randomly assign Kidogo to enter 30 communities; the remaining communities will serve as the control group.
Within each community, at baseline we will conduct a mapping exercise of all providers within the communities (N 2,022). For each provider found during the mapping, we will conduct a short baseline survey in which we collect provider demographics, opinions on their role of childcare providers, firm operations, and record observations of the facility quality. We anticipate a sample size of roughly 3,600 providers: among home-based and centre-based daycares, we will survey the provider, while in school-based daycares we will interview the school director as well as classroom teachers that provide daycare or playgroup.
Following the baseline survey, Kidogo will conduct the 3-month intervention in treatment communities. Approximately 6 and 12 months after the end of the intervention, we will collect follow-up data on the impacts of Kidogo. During the final follow-up round, we will also randomly sample 2,100 households with at least one child under 6 to determine the impacts of Kidogo for families.
Due to the timing of the planned expansion of Kidogo, we will conduct this evaluation in two “phases”. During the first phase, we will randomly assign 15 communities to receive treatment (and 15 control). During the second phase, we will randomly assign a different set of 15 communities to receive treatment (and 15 control). The first round will occur 2023–2024; The second round will occur approximately one year later (2024–2025).