Philip Leifeld
Philip Leifeld


Job opportunity: Data Scientist with emphasis on network analysis and graph databases. Deadline: 6 January 2021.

I am a professor (Chair of Comparative Politics) in the Department of Government at the University of Essex. My disciplinary background is in political science and public policy, with a secondary focus on research methodology and network science.

My research is mainly about political networks. I have published extensively on how policy networks operate, the role of interest groups and advocacy in policy formulation and decision making, and institutional design in international organisations. To this end, I have also contributed methodological innovations in the areas of network science and statistics, including network models for longitudinal data and associated statistical software packages.

My work on discourse network analysis has created a new branch of the literature. A recent thematic issue of the journal Politics and Governance includes a dozen original articles that employ discourse network analysis in different subfields of political science and public policy. My software Discourse Network Analyzer (DNA) and its associated R package rDNA have been used in many projects, theses, and publications. The basic idea of DNA is that policy debates among elite actors (e.g., interest groups, legislators, government agencies etc) in text data such as newspaper articles can be analysed as temporal networks. The DNA software allows you to code positive or negative statements actors make about concepts in a qualitative way and then export them into various kinds of networks. We can then throw the whole arsenal of network-analytic methods at the problem in order to identify changing coalitions in the policy debate, polarisation of debates, opinion leadership, and other things of interest. I have spent much time figuring out the right transformations and methods that operationalise theories in politics/policy.

I am also the author and maintainer of the texreg R package, which creates regression tables in various formats. The package is in the 96th percentile of R package downloads. It takes more than 100 statistical models and creates highly customisable regression tables for LaTeX, HTML, Word, ASCII, and inclusion in Markdown documents. I have recently added innovative ways to gauge user feedback through the use of a "praise" function.

Before joining Essex, I was a professor (Chair of Research Methods) at the University of Glasgow in the School of Social and Political Sciences. I am still serving as external supervisor for a couple of PhD students at Glasgow, both in the social sciences and in the School of Mathematics and Statistics. At Glasgow, I gained extensive leadership experience by serving as Director of Graduate Training for eleven Departments. I introduced a new methods programme into the College-wide Graduate School and served in various associated administrative and leadership roles. In my current position at Essex, I am responsible for PGT recruitment in the Department of Government, and I am also affiliated with the Institute for Analytics and Data Science (IADS) through a knowledge transfer partnership in the finance sector in London.

At Essex, I am teaching GV903 Advanced Research Methods, a year-long statistics sequence for political scientists that includes maximum likelihood and generalised linear models and some advanced models for time series, panel, multilevel, network, and causal inference settings. Previously I taught a large, College-wide Research Design course with > 300 students at the postgraduate level at the University of Glasgow, and I have also taught substantive politics modules on Political Parties and Public Choice, the Governance of Collective Goods, Pension Politics, Political Networks, and other topics. I occasionally teach summer school courses (for example, at the ECPR Winter School or at the IADS Summer School in Data Science and Analytics), mostly either on statistical network analysis or more specifically on discourse networks. I am also often invited to deliver workshops and occasionally keynotes on computational social science topics.

Tweets by @PhilipLeifeld