Text-as-Data Analysis of Populist Parties versus Government Parties: To Blend or not to Blend?

Abstract and Appendix.

William Sanger https://www.cirano.qc.ca/en/community/directory/view/2109 (CIRANO and Polytechnique Montréal)https://www.cirano.qc.ca , Nathalie de Marcellis-Warin https://www.nuance-r.com/principalInvestigator.html (CIRANO and Polytechnique Montréal)https://www.cirano.qc.ca/fr/communaute/bottin/view/1477 , Thierry Warin https://www.nuance-r.com/principalInvestigator.html (SKEMA Business School (Raleigh, NC))https://www.skemagloballab.io
2019-09-08

Abstract

Populist parties have always existed. In this article, we look at whether populist parties have evolved recently to assimilate as government parties or on the contrary have distanced themselves from government parties. To capture the notion of populist parties, we limit ourselves to far-right parties as defined in the Manifesto Project. To capture the dynamics of the elements of language used by populist parties, we needed to limit our study to a field of interest. We chose to consider what the manifestos had to say about the European integration as a topic of choice. In order to measure the dynamics of the elements of language, we combine two very original methodological approaches based on Data Science techniques: first, we use a text-as-data approach based on the computation of Jaccard similarity indices and second, we perform LDA analyses on political manifestos in their original language. We use a very unique database from the Manifesto Project starting in 2000 to match the birth of the Economic and Monetary Union.

Appendix

  1. Jaccard Similariry by Country

  2. Pearson’s product-moment correlation between Political Pla

  3. LDA per Country

Citation

For attribution, please cite this work as

Sanger, et al., "SKEMA Global Lab in AI: Text-as-Data Analysis of Populist Parties versus Government Parties: To Blend or not to Blend?", Figshare, 2019

BibTeX citation

@article{sanger2019text-as-data,
  author = {Sanger, William and Marcellis-Warin, Nathalie de and Warin, Thierry},
  title = {SKEMA Global Lab in AI: Text-as-Data Analysis of Populist Parties versus Government Parties: To Blend or not to Blend?},
  journal = {Figshare},
  year = {2019},
  note = {https://skemagloballab.io/research.html/posts/2019-08-21-text-as-data-analysis-of-populist-parties-versus-government-parties/},
  doi = {10.6084/m9.figshare.7781051.v2}
}