НИС

Qualitative Comparative Analysis aka QCA is a special technique that has gained more popularity among researchers and analysts and offers a number of advantages vis-a-vis 'traditional' case studies and statistical analysis. QCA also known as Boolean algebra is one of the ways how to circumvent the problems of ‘the too few cases and too many variables’ and ‘the small-N problem’. QCA combines strict logic of causal inference without meeting a large number of frequentists statistics assumptions (e.g. sufficient number of cases, degrees of freedom, etc.). QCA draws on deterministic logics instead of stochastic one that opens more opportunities to study complex and intertwined causal relations, mediations and moderation (the ‘third variable’ problems). Thus, QCA is a mid-way between single case studies and large-N analysis. Students will improve their understanding of causal inference and causal mechanisms as well as will learn a package in R for QCA analysis.

Number of students for the block: max. 30.

Social network analysis (SNA) is both a conceptual vision of the world as a complex network of networks and a method of studying social, political, economic structures using network approach and graph theory. Network consists of a set of actors/entitites and ties/relations connecting them. While the concept of social network is quite old, recent advances in statistics and computationl science as well as emergence of World Wide Web and social media revolutionized this approach. We know that we live in a highly connected world. Social network analysis is crucial in studying politics and political action, social movements, social mobility and migration, diffusion of innovations, communication, and organizational studies.

The topics covered in the course are: personal networks and social capital; elites and power structure,  political networks, social movements and collective action, peer effects, influence and selection; structural roles (leaders and brokers); networks in creative world, networks in research, digital networks and social media. You will also learn how to describe, visualize and analyze networks; what are network measures and  what they mean; what are the principles of network formation; random networks and real-life networks ("Small World" phenomenon and preferential attachment ("Matthew Effect").  For thinking about the course you may also look at The Oxford Handbook of Political Networks (2017) available in HSE e-resources.

Number of students for the block: max. 30.

With the advent of new technologies and forms of communication, the modern world began to be characterized by the creation and accumulation of large volumes of very diverse information. Most of the data is stored in text format (for example, scientific articles, posts in social networks, letters, transcripts of speeches, official documents, etc.). Computational methods for text analysis promise to aid at the scale where traditional content analysis is not feasible.
Course attendees will learn how to distinguish typical words and expressions of individuals (for example, the president or the hero of cartoons), conduct topic modelling on a big text data (LDA of > 10k posts), build semantic maps (semantic networks of scientific areas) and etc. This course will be dedicated to the major techniques for mining and analyzing text data to discover different patterns. The goal of the course is to provide a basic understanding of the proper usage text collections in quantitative analysis. R programming environment will be used as a toolbox for text analysis.

Number of students for the block: max. 30.