Does exposure to violence shape people’s attitude, behavior and tolerance towards IPV?

The intervention comprises two treatment arms and a control group. All participants assigned to the treatment group will be exposed to a short video intervention that contains news snippets of incidents on Intimate Partner Violence (IPV) that have happened in Kerala from 2015 to 2021, with a specific emphasis on dowry-related incidents in Kerala. The video intervention will be followed by an interactive group discussion led by a gender expert.
Participants assigned to treatment arm 2 will receive a social media intervention additional to the video and interactive discussion.

2024-11-13

2025-01-31

Individual’s tolerance and attitude towards IPV
Willingness to sign a petition against IPV
Donation experiment (modified dictator games)

We will utilize survey to measure how the intervention has affected participants' attitudes towards IPV. In addition to traditional surveys, we will employ a double list experiment, which includes indirect questions that will help in discreetly measuring people’s attitudes towards sensitive topics.

Individual's willingness to sign a petition addressed to the state government, advocating for the implementation of stringent laws or initiatives aimed at reducing intimate partner violence. This would measure their hypothetical behavior change due to the intervention.

We would like also to measure the decision of participants to donate to an organization combating Intimate Partner Violence (IPV). If participant choose to donate and follow through with this action, it can be considered a genuine change in behavior and provides tangible evidence of the intervention’s impact on their attitudes and actions regarding IPV.

We will recruit participants through a structured, multi-step process, in collaboration with Kudumbashree, a women empowerment program in Kerala, India. Thiruvananthapuram district has been selected for our study based on reports of an increased incidence of IPV in the area. Out of the 124 villages in the district, we plan to randomly stratify villages based on their geographical location and select 40 villages, covering approximately 30% of all villages in the district, based on their geographic location to obtain a sample representative of the broader population. From each selected village, potential participants will be identified using the electoral voters' lists from the Chief Election Commission. Each village comprises multiple polling booths, each serving a relatively consistent number of residents (ranging between 1,000 and 1,500 individuals). Our plan is to randomly select two polling booths from each chosen village and then create a list of people based on our eligibility criteria. We will use Kudumbashree and their community mobilization channels to invite individuals from the list to participate in the study. From each village, we aim to invite 45 participants for the study.

An important condition for participating in our study is that participants have access to and use social media, specifically WhatsApp. Therefore, each potential participant will be sent a personalized invitation to join the study. Invitations will be sent via mail or delivered in person by Kudumbashree Resource Persons (RPs) and Community Development Society (CDS) members from the selected localities.

Following participant identification, the participants will be randomly assigned to one of three treatment arms, and invited to take part in the study. We will conduct a short baseline questionnaire followed by a post-intervention study that measures the outcomes of interest. We will exploit the randomized design to estimate Intention-To-Treat and Average Treatment on the Treated effects on our outcomes of interest. We will run OLS regressions on a treatment dummy (pooled and separated for each treatment arm) our stratifying variables and key demographic control variables.

Not available

The randomization procedure is built into Qualtrics

The randomization will take place at the experimental group level. 45 participants per village will be equally distributed across one of three treatment arms. The treatments will take place at the group level.

Yes

3 groups per villages times 40 villages equals 120 groups.

1800 participants

40 groups for treatment arm 1: the video intervention
40 groups for treatment arm 2: the video intervention plus social media task
40 groups for control

We specify a standardised MDE of 0.2 SD using power of 0.8 and significance level of 0.05