Encouraging female decision making autonomy: results from household nudges in Afghanistan

A random sample of households benefitting from the UNICEF’s mother and child cash transfer will receive a series of household level nudging sessions consisting of both interactive information sessions and interactive activity sessions conducted separately between women and men in the household in line with cultural norms.
The five home visits cover a range of topics related to mother and child health practices and will be completed by specially trained social mobilisers. These social mobilisers, working in fixed pairs of men and women to comply with Afghanistan-specific social norms, will visit households five times over approximately four months.

A random sample of households benefitting from the UNICEF’s mother and child cash transfer will receive a series of household level nudging sessions consisting of both interactive information sessions and interactive activity sessions played separately between women and men in the household in line with cultural norms.
The five home visits cover a range of topics related to mother and child health practices and will be completed by specially trained social mobilisers. These social mobilisers, working in fixed pairs of men and women to comply with Afghanistan-specific social norms, will visit households five times over approximately four months.
The home visits vary in terms of their pedagogical approach. Two home visits – 1, 3 - will follow a pure informational treatment strategy. Two other visits – 2, 4 - will be interactive and aim to inform and facilitate intra-household discussions between participants with small exercises. The final session (Session 5) combines an informational treatment strategy and an interactive approach. The sequence of the sessions alternates: first we have one informational treatment home visit; then we have the more interactive visit. Specifically:
I. Antenatal and postnatal health care visits (Informational Treatment)
II. The work we do (Interactive Session)
III. Antenatal and postnatal maternal nutrition (Informational Treatment)
IV. Household economy (Interactive Session)
V. Child vaccination (Informational Treatment) and 'a gift from the heart’ (Interactive Session)
The Informational Treatment sessions should last approximately 45 minutes while the interactive sessions and the final session should last approximately between 60 and 75 minutes. Both are composed of three main stages:
• Introduction (up to 15 minutes);
• Core section (up to 45 minutes for the Interactive sessions);
• Farewell (up to 15 minutes).

2025-07-07

2025-10-02

Prenatal care visits, baby delivered in an institution, antenatal care visits, vaccination rate

Primary outcome explanation: What are the impacts of the intervention (household-level nudges designed to encourage the re-interpretation of gender norms) and cash transfers on infant and maternal healthcare seeking practices. Households and individuals in Afghanistan currently face intense macro and micro-level cultural, political and economic vulnerabilities that result in a significant risk of harm to a multi-dimensional spectrum of well-being domains. Mothers and children represent a key vulnerable group in need of humanitarian assistance. There is a large body of literature that estimates the impact of both cash transfer programmes and social and behavioural change communication on similar outcomes. However, there is a limited body of literature that explores this interaction in conflict settings, nor in settings with strict interpretations of gender norms by authorities.

Spending on diverse and nutritious food, nutrition outcomes, uptake of nutrition counselling services, improve knowledge, attitudes and practices and encourage the re-interpretation of gender norms on female decision making in access to healthcare.

The cash transfers and social and behavioural change communication are expected to generate spillover benefits on spending on nutritious and diverse foods, which are assumed to be positively associated with an income shock, and uptake of nutrition service as a by product of the social and behavioural change communication. We also expect that the intervention will affect females` voice and agency in the households, promoting a re-interpretation of gender norms on female decision making in access to healthcare.

For each province, we have identified three areas: one control area, Group A, and two treatment areas, Groups B and C. Both B and C will benefit from the UNICEF’s mother and child cash transfer but only the Group C is eligible to receive intensive household nudging sessions. Our study will therefore focus on this group, within which half of the households will be randomly selected to participate in the intensive household nudging experiment.
A multi-stage sampling strategy is used. Within each area, settlements are randomly selected taking into account the following criteria: population distribution; geographic characteristics; accessibility assessment. In each selected settlement, households are randomly selected using the random walk methodology starting from the community centre. In treated communities, sampling targets are set based on the number of registered UNICEF MCCT beneficiaries.
We will collect baseline information, followed by a midline survey to measure the short-term impact of the intervention on the outcomes of interest, and an endline survey to assess whether the effects persist approximately one year after the intervention concludes. We will leverage the randomized design to estimate the effect of intensive household nudging sessions on our key outcomes. Additionally, we will conduct a mediation analysis to explore the potential channels through which the intervention influences these outcomes.

Randomization done in office by a computer

Households

No

No cluster design

1,200 households

600 households in the treatment group and 600 households in the control group

For the first variable, "Baby received prenatal care (at least once)", the average is 0.78, and the standard deviation is 0.41. Considering a power of 80%, the value of 0.067 is the minimal detectable effect size in a study with these characteristics—meaning an increase of about 8.5% in the variable of interest. For the second variable, "Delivery in an institution", the average is 0.66, and the standard deviation is 0.47. Considering a power of 80%, the value of 0.0761 is the minimal detectable effect size for a study with these characteristics—meaning an increase of more than 10% in the variable of interest.