Social Data Science

Students are expected to be open-minded and willing to invest in learning the programming language R, which is a powerful language in which a variety of data science methods can be implemented. Furthermore, students are expected to invest a substantial amount of time in acquiring the necessary programming skills both during and outside class so that we can focus on implementing various data science techniques.

In order to understand, produce and consume the next generation of social science studies students need to know how to implement and interpret data science methods.

The purpose of this course is to introduce students to web-based data gathering, text analyses and implementations of machine learning. These tools have proven to be extremely powerful full and some of them are directly responsible for the AI wave that is currently remoulding and restructuring many industries including the public sector. Hence understanding how and what different models do is essential for students who want to be prepared for a labour market, which increasingly demands knowledge of where efficiency can be gained by implementing data science methods.

The course begin by an introduction to R but as already stipulated it is expected that students work on their programing skills outside class.

The course starts with an intensive introduction to R and Rstudio. We shall learn how to import and export data, how to structure data and how to plot data using ggplot2. From there we move to visualization of data using the ggplot2 package.

After this introductory phase we move to a how to gather data from the web using scraping techniques and APIs. The next elements are concerned with introduction to text analyses and to simple introduction to machine learning methods.

The course will be a combination of lectures and labs. A substantial proportion will be hands-on work from the students.

Knowledge that enables the student to:

• Understand the principles of data science and the assumptions behind select statistical methods.
• Discuss assumptions behind select data science methods and if possible provide solutions.
• Interpret results from selected statistical methods.

Skills that enable students to:

• Analyze and critically assess scientific articles that use data science methods taught in the course.
• Create research questions that are relevant in academia and the real world and can be answered using online data and data science methods.
• Select appropriate data science methods for own research questions.

Competences that enables students to:

• Implement data science methods using R.
• Correctly select and interpret the quantities of interest resulting from the application of these methods and link these results to research goals.

The course will use chapters from textbooks, papers and internet-resources, such as:

• Adler (2012) R in a nutshell
• Gareth James, Daniela Witten, Trevor Hastie, Robert Tibshirani (2021) “An Introduction to Statistical Learning – with applications in R, ISBN: 9781071614174
• R for Data Science; Wickham and Grolemund (2015).

The teaching language is English, unless all students (and the course’s teachers and student instructors) are proficient in Danish (in this case the teaching language will be changed to Danish).

Ordinary class discussions (12 classes of 3 hours) where focus is on developing code and apply data science methods to own research question. 

This course is based on active learning, involving a high degree of individual and group interaction and class participation.

Scheduled classes
12 classes of 3 hours

Workload
Face-to-face teaching (lectures):   36 hours
Preparation:                                 210 hours
Examination:                                  24 hours
Total:                                            270 hours