Fostering Access to Early Childcare - A Randomised Controlled Trial

1-Information-only treatment (T1) (October-December 2022): this involves two clusters of contents
1a) Cluster 1: information aimed at helping families to identify a type of ECS that fits well with their preferences and constraints. Treated families will receive a text message providing access to a short video presenting information on the availability and characteristics of different types of ECS in France and how they may correspond to the different preferences and needs of families. One text message reminder will be sent to encourage mothers to watch this video. In the following days, a second text message with related reminders will give access to a second short video on eligibility to and costs of ECS in France and a third video will present detailed information on a specific type of childcare service (halte-garderies) which may be particularly well-suited to low-SES families (less intensive care, less restrictive criteria of access, easier application procedures and more flexible time schedules).
1b) Cluster 2: information aimed at understanding the application process. Like for 1a), this involves sending text messages and related reminders with links to two videos presenting information on the calendar of applications, the procedures to apply, and tips to maximize the chances of success, such as applying to multiple ECS. The content of the first video of this cluster will vary depending on the district where families live, as each one has a specific application process.
Reminders: Reminders on Cluster 2 will be sent by text messages after child delivery (February 2023) in order to maximize applications for the June commission, where most of the slots get allocated. They will also be sent generic reminders to apply shortly before the deadline for the 2023 applications to ECS (April 2023).
2- Information plus administrative support treatment (T2) (February-April 2022):
This involves a phone call with parents to help them to fill up the forms, as well as personalized application reminders. Parents will receive the same information as T1.

2022-10-16

2023-07-01

Primary outcome: The primary outcome is a dichotomous variable indicating whether parents applied for a slot in at least one ECS. We are interested in both the main effect and the effect on low-SES parents.

Explanation: in public debates, the underrepresentation of low-SES families is often attributed to a preference for informal childcare arrangements. Our intervention primarily aims at showing whether there is an unmet demand for ECS relating to a lack of information and administrative support in the application process.

Secondary outcomes: three dichotomous variables indicating whether families obtained a place in ECS, whether this is a full-time or a part-time attendance and whether this starts in September 2023 or later. Two continuous variables will measure i) the date at which parents applied, as it may impact their likelihood to get a place, and ii) the number of applications submitted. If the intervention impacts childcare attendance, two additional outcomes will be i) a measure of children’s development, and ii) labor market participation.

Explanation: Obtaining a place in ECS is not a primary outcome because the market for daycare is highly rationed to the point that a large treatment effect on applications could reduce the probability of getting accepted due to increased competition. Because the treated sample is rather small relative to the target population, we do not expect massive displacement effects. Therefore a positive effect on this outcome would not imply that, if the intervention was generalized, we would observe the same treatment effect due to general equilibrium effects. However, if we manage to document that an unmet demand for ECS exists, this would be evidence supporting a reduction of rationing and a public investment to strengthen the supply of ECS.

Experimental design
Targeting: we target families who a) wait for a child (5th to 9th month of pregnancy, corresponding to the moment when most mothers visit maternity wards), b) live in the selected area (departments of Paris, Seine Saint Denis and Val de Marne), c) have a telephone with internet connection so that we can send the text messages and videos, and d) understand simple messages in French. It may be noted that even among low-SES and immigrant families, criteria c) and d) are not particularly restrictive as more than nine low-SES families out of ten meet c) and more than nine immigrant individuals out of ten understand French.
Access to the field: we have sent a written presentation of the project to the coordinators of the maternity wards of nine hospitals, who have confirmed their support for the project. During the baseline study (October-December 2022), we will approach mothers in the waiting rooms of these maternity wards using a systematic sampling procedure based on the order of arrival of mothers to the maternity ward. This sampling procedure is stratified by parental education (basic, high school diploma, tertiary diploma). Stopping rule: reaching the predefined sample size and quotas by parental education.
We will then present the study, inform them about confidentiality and data protection rules and ask for their consent to participate in the study. A leaflet containing legal data protection information will also be given to them.
Sample size: 1687 families minimum. This is based on power calculations using the Powerup software with the following parameters: a) minimum detectable effect size for the main effect: 0.15; b) randomization procedure as described above; c) R2=40% owing mainly to baseline information on intention to apply for ECS; d) T1=33%, T2=33% C=34%; e) blocks: 9; f) two-tailed alpha level: 5%; g) statistical power: 80%; h) attrition at the endline: 25%.

Within each maternity ward, we keep individuals who completed the baseline survey and agreed to let their phone number, and then construct stratification variables based on the interaction of the following characteristics from baseline: (i) education level; (ii) the level of coverage in ECS around where the family lives; (iii) a variable with three levels regarding ECS previous use and intention to use it for the child to come a) “intend to use ECS and already has” or b) “intend to use ECS and has not already” or c)“do not intend to use one”. We then randomly assign T1, T2, and control within strata with an expected assignment probability of ⅓ in each arm. Therefore, our subsequent analysis will account for the multi-site block-randomization design. Data collection in each maternity ward takes approximately two weeks.

Individual

No

No cluster

1687 individuals

Control group=563, T1=562, T2=562

0.15 of a standard deviation

PAP