96359 - DATA SCIENCE FOR POLICY ANALYSIS

Learning outcomes

The aim of the course is to build capacity in using data to inform evidence-based decision-making. The course will provide students with a broad overview of tools and methods for data analysis and applied empirical research, with a particular focus on the estimation of causal effect of public policies. Students will analyze real-word datasets and will be guided through case studies from a variety of policy domains. By the end of the course students will be able to perform a basic – yet rigorous – analysis of data to better understand policy choices. They will gain enough data science literacy to interpret and judge the quality of existing empirical research and to communicate the results to decision makers and the public.

Course contents

The main topics are:

• What is a policy evaluation: ex-ante and ex-post evaluation
• Monitoring, outcome, output and impact evaluation
• Observed, experimental and quasi-experimental data
• Steps in conducting an evaluation analysis
• Counterfactual
• Conditioning and instrumental variables
• Regression Discontinuity Design
• Difference-in-difference methods
• Propensity Score Matching method

 

Note on prerequisites: students are required to have a foundational understanding of the R programming language and basic concepts of simple linear regression models. These prerequisites ensure that participants can effectively engage with the course material and fully benefit from the advanced topics covered.

Readings/Bibliography

The course is mainly based on:

• Gertler, Paul J., Sebastian Martinez, Patrick Premand, Laura B. Rawlings, and Christel M. J. Vermeersch. Impact Evaluation in Practice, second edition. Washington, DC: Inter-American De velopment Bank and World Bank, 2016

Free download available at: https://www.worldbank.org/en/programs/sief-trust-fund/publication/impact-evaluation-in-practice

Other books:

• Angrist, Joshua D., and Jörn-Steffen Pischke. Mastering 'Metrics: The Path from Cause to Effect. Princeton University Press, 2014.
• Cunningham, S. (2021). Causal inference: The mixtape. Yale University Press.

Other useful materials, such as lecture notes and chapters/articles, will be provided through the e-learning platform Virtuale.

Teaching methods

The course consists of a combination of theoretical lectures, applied case studies and quantitative tutorials.

Assessment methods

Students will be expected to conduct a data analysis and deliver an oral presentation of their findings. Students are free to select a topic and dataset that they find interesting and relevant. However, students are encouraged to consult with me before beginning their analysis to ensure that the chosen analysis and dataset are appropriate.

The project must be completed independently by groups of one to three students. Presentations will be held on the official exam day.

Guidelines will be provided on the Virtuale platform.

The final grade will be based on the thoroughness of the analysis, the accuracy and relevance of the terminology used, and the clarity with which the results are presented.

Teaching tools

Additional materials (such as scientific papers, slides, datasets...) will be provided during the lessons and the e-learning platform Virtuale will enable access to these contents.

Office hours

See the website of Silvia De Nicolò

See the website of