Machine Learning for Decision-Making in Development Evaluations
Seminar | Online
-
Organized by:
International Initiative for Impact Evaluation
About the Event
In this event, 3ie will present a machine learning method, called Causal Forests (CF), for treatment effect heterogeneity analysis in a school-based gender attitude change program in Haryana, India to advance equity and evidence in policy-making.
Methods: Combining CF analysis with qualitative methods to investigate how participant characteristics, including gender, interact with program outcomes. CF analysis identifies heterogeneous sub-groups, enhancing understanding of contextual factors impacting program success.
Results: CF analysis, supplemented with qualitative insights, reveals nuanced treatment effect variations, including gender-related aspects, often missed in conventional analyses.
Conclusions: This study introduces a novel approach in development interventions, emphasizing gender and equity considerations. CF analysis aids policymakers in targeting sub-groups effectively, fostering more inclusive and informed decision-making.
Event details:
The session begins at 8 AM US Pacific Time/ 11 AM US Eastern Time on June 4th, 2024. Please use the Zoom link provided to join the session. The passcode to access the event: 959740
Methods: Combining CF analysis with qualitative methods to investigate how participant characteristics, including gender, interact with program outcomes. CF analysis identifies heterogeneous sub-groups, enhancing understanding of contextual factors impacting program success.
Results: CF analysis, supplemented with qualitative insights, reveals nuanced treatment effect variations, including gender-related aspects, often missed in conventional analyses.
Conclusions: This study introduces a novel approach in development interventions, emphasizing gender and equity considerations. CF analysis aids policymakers in targeting sub-groups effectively, fostering more inclusive and informed decision-making.
Event details:
The session begins at 8 AM US Pacific Time/ 11 AM US Eastern Time on June 4th, 2024. Please use the Zoom link provided to join the session. The passcode to access the event: 959740
Speakers
| Name | Title | Biography |
|---|---|---|
| Fiona Kastel | Senior Research Associate | Fiona Kastel engages in research on the application of innovative data sources and methods to impact evaluations and evidence synthesis. She holds a Master’s in Public Affairs from Brown University, and has experience in sectors like climate, agriculture, education, and conflict and security. |
| Geetika Pandya | Senior Research Associate | Geetika Pandya specializes in mixed-method research to assess intervention impacts on marginalized communities, with a focus on gender equity. She holds a Masters in Development Practice from UC Berkeley, and has led projects on WASH, climate change, economic empowerment, and financial inclusion. |
| Devika Lakhote | Data Scientist | Devika Lakhote specializes in data science applications for evidence generation and synthesis, with expertise in causal inference and machine learning techniques. She holds a PhD in Public Policy from the University of Chicago, and over 8 years of experience working in international development. |