Data Wrangling and Exploration for Social Sciences

Catalog description

The course will teach students how to perform data wrangling and exploratory data analysis in social science contexts using a general-purpose programming language such as Python or Julia. No prior programming experience is assumed. Students will first learn the basics of the programming language used. They will learn about important sources of social science data and how to retrieve data both manually and programmatically using API access to some of these sources. They will learn how to work with data in different file formats (such as CSV, JSON, DTA, RDA XLSX etc.) and about different types of data (numeric, categorical, ordinal, dates and times etc.), and how to work with and express these different types correctly in the chosen programming-language. Students will study how to combine data from multiple sources, identify and deal with missing observations, how to identify outliers, and how to get data in a form that is more amenable for analysis. Students will use various publicly available data sets, such as data on housing prices, macroeconomic indicators and crime. They will identify important characteristics of data by calculating various statistical measures and constructing different types of charts, plots and visualizations. Students will also learn how to work with statistical distributions such as Binomial, Normal, Poisson and Exponential distributions in the chosen programming language to simulate stochastic phenomena in economics and social sciences.