Improving human capital by reducing menstrual stigma in Madagascar: the role of young female leaders

Building on earlier projects by CARE in rural Madagascar, CARE will recruit, train and support Jeune Fille Leaders (JFLs, “Young Girl Leaders”): school girls who are willing to speak out against harmful social norms and act as advocates for menstrual hygiene in their school.

The JFLs will be a group of between 3 and 6 adolescent girls per school, who will be selected, trained, given information regarding optimal menstrual practices, and encouraged to discuss menstruation in an open and positive way with their female peers at school. They will be encouraged to (i) directly share information on menstrual hygiene with their peers, and (ii) reduce menstrual stigma by promoting positive and open discussion of the topic. Through both of these channels, the JFL program could transform the human capital of girls in schools by reducing school absenteeism, by improving health behaviors and health outcomes, as well as socio-emotional wellbeing.

2022-02-21

2023-03-31

School absenteeism measured using unannounced spot-checks and survey data

Psychosocial wellbeing

Take-up of reusable menstrual pads (vouchers used, and incentivized measure of willingness to pay for pads)

Knowledge on facts about menstruation and appropriate health behaviors

Self-reported exposure to discussions on hygiene/menstruation.

Measures of social stigma and social norms, including (i) incentivized lab-in-the-field experiments (measuring willingness to break taboos/discuss menstruation), (ii) measures of second order beliefs, (iii) measures of bullying and teasing

Outcomes of caregivers: (i) knowledge index about menstruation of the mother (ii) Willingness to pay for menstrual hygiene products

The study will evaluate the effectiveness of the JFL program for reducing stigma and adolescent absenteeism.

The sample includes 140 schools in 3 rural districts in Madagascar. Randomization is at the school level, stratifying by (i) whether the school has functional infrastructure providing access to water at baseline; (ii) whether the school is primary or secondary; and (iii) the number of girls in the school aged 12 or greater. The treatment arms are:
(1) JFL (structured sessions) + Base Treatment (35 schools): receive the JFL program where JFLs are trained, and run structured classroom sessions about menstrual hygiene and WASH, plus the “base treatment” (described below).
(2) JFL (organic) + Base Treatment (35 schools): receive the JFL program where JFLs are trained, and are encouraged to have informal conversations with their peers about menstrual hygiene and WASH, plus the “base treatment”.
(3) Base treatment (35 schools). Only receive base treatment before endline, and receive JFL program after endline data collection.
(4) Control (35 schools). Receive no interventions.

The “base treatment”, delivered to groups (1), (2) and (3), includes (i) the construction of WASH infrastructure at schools, (ii) a voucher system providing access to reusable menstrual pads from local providers, (iii) the creation of “WASH committees” at the school level involving students and parents, and (iv) teacher training on WASH subjects.

There will also be district-wide interventions targeting girls and parents of all schools (treatment and control), including (i) a local newspaper encouraging positive WASH behaviors, (ii) training of local seamstresses to make menstrual pads, (iii) community-level celebrations of WASH-related world days.

In addition to primary data collection through baseline, midline and endline surveys, we will carry out unannounced spot-checks in schools during the first and second academic year of the program, with 5 spot checks per school, and 1 school-level survey at endline. These spot-checks will involve a roll-call in classrooms carried out by enumerators to provide a reliable measure of absenteeism at the school and individual level. The main primary data collection are a baseline (July-September 2021), midline (after 1 year) and endline (early 2023), using surveys of the teenage girls attending the targeted schools at baseline, as well as their parents.

Not available

Randomization done in office by a computer

Randomization at school level

Yes

140 schools

The sample includes 17 girls per school in each school, giving a total sample of 2,380. In addition, surveying JFLs and girls in their network will add an estimated 8 additional girls in the 70 schools with the JFL intervention, yielding a total predicted sample size of 2,940.

35 schools in the JFL arm with structured sessions + base treatment, 35 in the JFL arm with organic diffusion + base treatment, 35 with the base treatment only (construction of WASH infrastructure, voucher system for sanitary pads, creating of WASH committees at the school level, teacher training on WASH subjects), and 35 in control.

The design allows for an MDE of 0.2 standard deviations for main treatment effect when comparing 70 JFL treatment schools to 35 base treatment or control schools. This analysis assumes a power of 80%, significance threshold of 5%, intra-cluster correlation of 0.1, and an ANCOVA analysis with assumed correlation between baseline and endline outcomes of 0.5.