kibedavid01/kenya-worms-rct-replication

# Kenya Deworming RCT Replication This repository contains a replication and critique of the influential randomized controlled trial (RCT) by **Miguel & Kremer (2004)** on school-based deworming in Kenya. The original study evaluates the **impact of mass deworming** on school attendance, health outcomes, and spillover effects in neighboring communities. ## 📍 Study Overview - **Title:** Worms: Identifying Impacts on Education and Health in the Presence of Treatment Externalities - **Researchers:** Edward Miguel, Michael Kremer - **Field Partner:** ICS Africa - **Location:** Busia District, Western Kenya - **Sample:** 75 primary schools, ~30,000 students - **Timeline:** 1997–2001 - **RCT Registry:** AEARCTR-0001081 - **Data:** Replication data ## 🎯 Intervention The **Primary School Deworming Project (PSDP)** provided free deworming drugs every 6 months to children in high-prevalence schools. The treatment was phased across three groups of 25 schools each, randomized at the school level. Treated schools also received health education, charts, and teacher training. ## 📈 Key Findings - **Reduced worm infections** by 76% in treated children - **Increased school attendance** by ~9.3 percentage points - **No effect on test scores** - **Spillovers:** Reduced infections and increased attendance among children up to 3km from treatment schools - **Cost-effectiveness:** US$2.92 per additional year of school participation ## 🧠 Policy Relevance Deworming is one of the **most cost-effective education interventions**, outperforming subsidies and school meals in boosting attendance. These findings have supported nationwide deworming campaigns in Kenya, Ethiopia, India, Nigeria, and Vietnam. ## 📂 Repo Structure kenya-worms-rct-replication/ ├── data/ # Raw and cleaned datasets ├── scripts/ # R scripts for analysis │ └── replication.R ├── docs/ # Critique, replication notes ├── LICENSE └── README.md # This file This repo replicates key results from *Worms: Identifying Impact …