G²LM|LIC - Childcare Vouchers, Maternal Employment and Child Development in Addis Ababa

A parallel survey covers 56 screened childcare providers in the study catchments. 2. Study Design The intervention is a demand-side childcare voucher. Eligible households were randomized within provider catchment, which serves as the randomization stratum and the clustering unit used for standard errors in the analysis.

The trial was designed to measure effects on formal childcare use, maternal labour supply and earnings, maternal wellbeing and intra-household decision-making, and early child development. File Linking and Study Identifiers uid is the study household identifier and is present in every household-level file, allowing study rounds to be linked without SurveyCTO submission keys. member_no identifies members within a household in the endline file, and provider_id identifies childcare centres. treatment1 records assignment to the voucher offer; clust_id identifies the provider catchment and randomization stratum; refugee flags refugee catchments.

The baseline and endline EGRA files contain assessment items only and link to the corresponding household round using uid.