PP422      One Unit
Data Science for Public Policy

This information is for the 2025/26 session.

Course Convenor

Availability

Requisites

Pre-requisites:

Students must have completed PP407 before taking this course.

Additional requisites:

This will ensure that students have basic fluency in Maths and Statistics along with Python and its main Data Science libraries. 

This course covers the theory and practice of the Data Science project lifecycle in Python for Public Policy, from problem definition and data sourcing/cleaning to exploration, visualization, and modelling. Emphasis will be placed on identifying problems that are suitable for different Data Science techniques and on good practices for managing data. Linear and logistic models and regularization techniques will be covered in the AT and Machine Learning, Clustering and introductory text analysis models will be left for the WT. Key concepts and ideas underlying modelling (bias vs. variance, types of error, training vs. test data) and data ethics and data science ethics will be illustrated and implemented with examples from healthcare, education, urban policy, international development, and other policy areas. By the end of the course, students will have a strong coding workflow and will be able to source and experiment with data for analysis and research, both individually and in a collaborative environment.

30 hours of seminars in the Autumn Term.
30 hours of seminars in the Winter Term.

This course has a reading week in Week 6 of Autumn and Winter Term.

The course will have two 90 minute ‘Harvard style’ lectures/seminars per week. All students taking the course will need to attend both teaching sessions each week.  These are interactive sessions where student participation is expected.  

Students will be expected to produce weekly problem sets throughout the AT and WT.

These books provide an excellent starting point and can be used as the main reference for many topics. A full reading list will be provided at the beginning of the course.

1. James, Gareth, et al. An introduction to statistical learning: With applications in python. Springer Nature, 2023.
2. Chen, Jeffrey C., Edward A. Rubin, and Gary J. Cornwall. Data science for public policy. Springer, 2021.
3. Géron, Aurélien. Hands-on machine learning with Scikit-Learn, Keras, and TensorFlow. " O'Reilly Media, Inc.", 2022.
4. Müller, Andreas C., and Sarah Guido. Introduction to machine learning with Python: a guide for data scientists. " O'Reilly Media, Inc.", 2016.
5. Wilke, Claus O. Fundamentals of data visualization: a primer on making informative and compelling figures. O'Reilly Media, 2019.

Exam (40%), duration: 180 Minutes, reading time: 15 minutes in the Spring exam period

Presentation (40%)

This component of assessment includes an element of group work.

Problem sets (20%)

Students will complete two problem sets (20%), one in the AT and one in the WT. 

Students will also prepare a group presentation and take a final exam for the course.