reputation lab

Text Analytics

The media plays an essential role in shaping corporate reputations, and experimental evidence supports the impact of media coverage on customer behavior. The qualitative nature of text, however, limits the study of these processes. To transform "text" into "data" we need transparent and scalable methods, and in this section you will find some of our work that uses tools from computer science to extract meaning from text.

1. Yu, B., Diermeier, D., Kaufmann, S., & Godbout, J.F.  (Forthcoming). Language and Ideology in Congress. British Journal of Political Science.


For a brief overview of this paper, read Kellogg Insight: All Politics Is Cultural - Cultural not economic vocabularies separate liberals and conservatives.


Using our research, ABC News created an online game called “Are You Democrat or Republican? Does Language Give You Away?” Play the game, and discover how you lean based on your word choice.


2.  Beigman Klebanov, B., Beigman, E., Diermeier, D. (2010). Vocabulary Choice as an Indicator of Perspective. Proceedings of the 48th Annual Meeting of the Association for Computational Linguistics: 253-258.


3. Beigman-Klebanov, B., Beigman, E., & Diermeier, D. (2008). Lexical cohesion analysis of political speech. Political Analysis, 16(4): 447-463. Reprinted. (2010) W. Paul Vogt (Ed.) Data Collection. SAGE Benchmarks in Social Research Methods. London, UK: SAGE Publications.

© 2011 Reputations Rules. All rights reserved.
McGraw Hill Professional
Kellogg School of Management, Northwestern University
Privacy Policy