Quantitative Content Analysis of Chinese Texts?: A Methodological Note
This paper discusses the quantitative measurement of Chinese texts using hand-coded analysis, Yoshikoder and Wordscores. The paper compares the advantages and disadvantages of each method using the example of the 2008 melamine-tainted milk scandal in China. The policy positions estimated by Yoshikoder are not too different from those using hand-coded analysis. Yoshikoder outperforms Wordscores, but with substantial human intervention.
KeywordsQuantitative Content Analysis Computer-Assisted Content Analysis (CCA) Yoshikoder Wordscores Melamine-Tainted Milk Scandal
This paper was completed during my postdoctoral stay at the Chair of International Politics at the University of Konstanz in Germany and at the Institute for Human Security at La Trobe University in Australia. For financial support for my project on “The Rise of Group Interests and Evidences of Interest Articulation in China” (Taiwan NSC98-2917-I-564-147), I would like to thank Taiwan’s National Science Council. John James Kennedy has commented thoroughly on my earlier draft. I wish to thank him for his assistance.
- 2.Carlson, J.M., and M.S. Hyde. 2003. Doing empirical political research. Boston and New York: Houghton Mifflin Company.Google Scholar
- 7.Lowe, Will. 2006. Yoshikoder: An open source multilingual content analysis tool for social scientists. The American Political Science Association Meeting, Philadelphia, USA. http://www.yoshikoder.org/courses/apsa2006/apsa-yk.pdf (accessed June. 26, 2010).
- 10.Peng, Fuchun, Fangfang Feng, and Andrew McCallum. 2004. Chinese segmentation and new word detection using conditional random fields. Proceedings of the 20th International Conference on Computational Linguistics. Switzerland: Geneva. No 562.Google Scholar
- 11.Schneider, G., and K. Baltz. 2003. The power of specialization: How interest groups influence EU legislation. Revista di Politica Economica 93(1/2): 253–287.Google Scholar