Skip to content

Syllabus

Overview

Course Information

  • 16 sep. 2021 to 31 mar. 2022
  • T 11:30am-1:00pm
  • ScPoBx, Monnet

What you will learn in this course? This course introduces students to the building blocks of data science and machine learning with real world data and applications in political science, sociology or economics.

The ability to collect and analyze data in a sophisticated manner is now a crucial skill set for the modern job market across industries. Learning such methods and tools requires practice. This course follows a “learn-by-doing” approach and will place emphasis on gaining experience by analyzing real-world data and solving problems using cutting edge yet free open source software. It will allow you to apply your theoretical knowledge and articulate the methods learned in an end to end research project throughout the course. The course is designed for social science students that are not necessarily familiar with quantitative methods or computer programming.

Learning support will be provided for at least one programming language, such as C++, Julia, Python, or R, but the choice of language supported may vary between years, depending on judged benefits to students, whether in terms of pedagogy or resulting skills. This year, the default choice is Python.

Goals

  1. Visualize, summarise, and analyse real-world data using reproducible code based on cutting-edge open source tools for data analysis.
  2. Understand the scientific method, critically evaluate existing information, and leverage such methods in a fully reproducible end to end scientific project.
  3. Develop an ethical, theoretical, and methodological understanding of machine learning and artificial intelligence from a social scientist perspective.

Requirements

The course is designed for social science students with no previous experience of quantitative methods, statistics or computer programming.

  • The most important requirement is motivation to work hard on likely unfamiliar material.
  • The second most important requirement is to have access to a computer in order to participate and complete the various activities. From there, we will learn as we go!

In order to fight against digital inequalities among students, Sciences Po Bordeaux sets up a support system under certain conditions (presentation of a purchase invoice; N-1 tax notice of your reference tax household ; notification of CROUS scholarship; certificate of attendance for the current year).

For any questions relating to this device and in order to submit your request, please write only to the following address, specifying your needs:

Structure

This course has 24 class meetings that span over two semester of 12 classes each. During the class meetings students have the opportunity to accomplish various activities. The activities allow students to incrementally assimilate critical concepts that will lead them to deliver a fully reproducible end to end scientific project.

Class Meetings

Live class meetings are held once per week and provide students with many opportunities to perform in-class activities related to the material in that week. Each session is divided into three parts: a live demo, the theoretical discussion, and hack-time which demonstrate how to leverage these concepts programmatically.

The demo is lead by a student. A student is expected to do a 15 minutes presentation. This demo is a hands-on demonstration that introduces and applies a programming concept related to the session and illustrates how to use it through an example within the social sciences.

The theoretical discussion is lead by the instructor. The instructor will usually walk through real world examples that supplement the demo to illustrate how to leverage data science in the real world. The students are expected to follow along with their computers.

The hack-time is where we learn and apply new tools and methods together. The instructor will provide code and data that illustrate the application of data science to political science, sociology or economics. This part is often followed by a "challenge" where students will need to apply what they have learned to new data.

Grading

The course is evaluated over two semesters through various activities allowing students to incrementally improve data science skills. By the end of the course, students will deliver an end to end empirical research project that is fully reproducible and that leverages the acquired data science skills.

Activities Percentage
Participation 30%
Challenges 30%
Project 40%

Communication

This course uses a dedicated open source communication platform named Gitter. When you have questions outside course meetings, click the Chat tab at the top of this page and ask your question in the appropriate channel. The onboarding activity will get you started!

Asking questions publicly and providing answers to your peers' questions allows everyone to learn dynamically. Furthermore, if you have a question on a topic, it is likely that someone else has the same question. Finally, being active on the platform, whether by answering the questions of your peers or by asking questions, is strongly encouraged and will have a positive impact on your participation grade!

Reserve private messages with the instructor for personal matters.