What you’ll bring:

• Exceptional alignment with GiveDirectly Values and active demonstration of our core competencies: emotional intelligence, problem-solving, project management, follow-through, and fostering inclusivity. We welcome and strongly encourage applications from candidates who have personal or professional experience in the low-income and/or historically marginalized communities that we serve.

• Bachelor’s degree (or equivalent) in Economics, Statistics, Public Policy, or a related quantitative field

• 2–4 years of experience working with data in applied settings (e.g., experimentation, evaluation, analytics, or program learning), ideally in development, tech, or operations-focused roles

• Solid understanding of experimental design and causal inference concepts (e.g., randomization, treatment/control groups, units of randomization, statistical power, bias) and how to apply them in real-world program contexts

• Experience using R, Python, or Stata to clean, manipulate, and analyze data, including working with multiple data sources (e.g., survey or administrative data)

• Experience collaborating with cross-functional teams (e.g., programs/operations, product, or research) and external partners to implement projects and solve problems

• Fluency in English required

• Comfort working at pace - able to manage multiple workstreams simultaneously, make progress with imperfect data, and iterate quickly based on emerging findings

• Ability to interpret results beyond statistical significance and communicate clear, actionable insights to both technical and non-technical audiences

Execute and manage A/B tests across programs**

• Conduct power calculations to ensure experiments are both statistically rigorous and feasible to implement within program constraints

• Set up pre-specified experimental designs, applying defined experimental groups, outcome measures, and measurement timelines

• Review experiment setups prior to launch and flag execution and measurement risks that may affect interpretability

• Ensure experiments are well-coordinated and executed as designed, aligning implementation with research plans and integrating smoothly into program delivery across Programs and Product teams

• Work at a fast pace across a portfolio of 2–3 live A/B tests at any given time, designed to generate actionable answers quickly and feed rapid iteration of programs and products

Ensure accurate measurement and high-quality data for experiments****

• Collaborate with external Principal Investigators (PIs) to ensure measurement approaches and data collection are aligned with research design and implementation realities

• Ensure experimental outcomes are captured accurately and consistently by applying established indicator and measurement approaches

• Prepare and manage datasets that are clean, well-structured, and ready for analysis using survey, administrative, and product data

• Identify and flag data quality risks (e.g., missingness, inconsistencies, measurement error) that could affect the validity of experimental conclusions

• Conduct targeted literature reviews to ensure measurement approaches are grounded in evidence and aligned with best practices

Analyze experimental data and interpret results to inform decisions****

• Generate reliable and decision-ready analyses of experimental data from A/B tests

• Assess the magnitude and direction of effects and highlight what the results do and do not suggest, noting key limitations

• Ensure results are clearly understood and appropriately interpreted, given data quality, sample size, and implementation considerations

Translate results into actionable insights and learning across experiments**

• Translate experimental results into clear, actionable recommendations to guide program and product decisions for individual country programs, and for the direction of GiveDirectly’s programming as a whole

• Structure results and key learnings so they can be reused to inform future experiments and program design

• Prepare concise learning products (e.g., memos, summaries) that serve multiple audiences, including internal Programs and Product teams, and external academic partners, ensuring findings are clearly communicated for both technical and non-technical stakeholders 

• Prepare and clean de-identified datasets for sharing with external PIs and academic partners, ensuring data is structured, well-documented, and ready for independent analysis

• Contribute to cross-country discussions to ensure learnings from experiments are shared and applied across contexts, and maintain clear documentation so experiments are easy to track, understand, and build on over time