• A
  • A
  • A
  • ABC
  • ABC
  • ABC
  • А
  • А
  • А
  • А
  • А
Regular version of the site

Working with Different Types of Data in Political Science and Linguistics

Kyle Marquardt is Assistant Professor at the School of Politics and Governance (Faculty of Social Sciences) and Research Fellow at the International Center for the Study of Institutions and Development, and tells HSE Look readers about his research, teaching and previous experience with HSE.

Kyle Marquardt

— What is your main focus of research in political science?

— For the past four years I worked with Varieties of Democracy [V-Dem], which aims  to measure different aspects of democracy across time and pace---more specifically almost every country in the world from 1900 to present, and a large proportion of countries going back to 1800. The idea is that we [the V-Dem Project] gather experts on concepts related to democracy who code both these concepts and other traits of political regimes.

The goal of the project is to gather somewhere around 5 experts to code these different characteristics of democratic governments for each country-year. Since experts give different assessments based on their conceptual differences, the project uses a specific type of statistical modelling to try and aggregate these expert-coded date into a  cross-nationally and cross-temporally comparable dataset. The basic idea is that getting cohesive ratings is hard. Normally the experts give a score from 0 to 5, or something like that, but your ‘one’ maybe someone else’s ‘three’, and it’s difficult to alleviate these discrepancies, even though people are trying to tell you the same thing. This problem becomes even greater when you’ve got scholars from different backgrounds.

I have been trying to develop methods to deal with this problem, and exploring and validating these methods has been one of the main goals in my research with Varieties of Democracy project for the last couple of years.

— What else do you work on?

— The second thing that I’ve done a lot of work on are  issues of identity and language politics, mainly focusing on the former Soviet Union; I’ve done work primarily in Moldova and in Russia. Right now I’m trying to merge these two interests: my work on identity and language politics with the work I’ve been doing on validating expert assessments for analysing political regimes.

One of the things I’m focusing on in particular is trying to measure language change across regions of Russia using various types of data.  Historically, questions about language proficiency have been asked in different ways across time, so trying to come up with a consistent time-series of how, for example, language proficiency in Tatarstan has changed, is difficult. I’m trying to use techniques I worked with at V-Dem and apply them in this context,  creating a consistent dataset from varying measurements.

— Have you had any previous contact with HSE?

— I’ve been to HSE several times prior to this year, mostly at the International Center for the Study of Institutions and Development (ICSID). Three years ago I was at HSE for about a week doing some presentations, and before that I spent several summers in Russia doing field work for my dissertation around 2013-14. Also, in 2006-07 I was a Fulbright scholar in Tatarstan, mainly focusing on learning the Tatar language for my research.

— Any special plans for this academic year?

— I’m very interested in doing more on the linguistic project that I’ve mentioned, and I think it could be very relevant to people working on this issue in Russia, not to mention the greater fields of sociolinguistics and language change. I’m also continuing to work on the surveys I did in Moldova, which are also about the language issues.

In the first two modules I am teaching Bayesian Statistics, which is about a particular branch of statistics which I’ve worked with extensively when aggregating different sorts of data. I’m really excited about teaching that course to a class of very well-trained and motivated students.  I also work with Master’s students in a research seminar and project seminar in the Faculty of Social Sciences.

Finally, I’m also a research advisor for two MA students and two BA students. The projects that these students are working on are really diverse and exciting- one student is working on impediments to American football in Russia, and another is gathering really amazing data about the relationship between refugees and voting patterns in Italy. It’s a great privilege to have the opportunity to help them develop their projects.

December 06, 2019