I know that I want to do a content analysis of Chinese media. I find it intriguing that Chinese media can be used to examine Chinese government policies. But I do not know what specific topic I want to study. I have considered China's views toward Taiwan, China's territorial claims in the South China Sea, the Chinese media itself, or potentially something that combines all of these topics. There are problems with making the study too broad, and also with studying a topic that is not really covered by the chinese media source (the Xinhua News Agency) that will form the basis of my content analysis.
Thinking through these issues has allowed me to consider two key aspects of my ALM research: the topic itself, and the methodology.
I've already talked a bit about the topic question. Other than the topics listed above, the field is really wide open -- the Xinhua English service has covered thousands of news items relating to foreign and domestic policy since the archives were digitized beginning on January 1, 1977. For the class I took last semester, Prof. Kuhn's seminar on Chinese emigration in modern times, I was able to research China's policies regarding overseas Chinese in Vietnam and Cambodia in the late 1970s, using Xinhua news coverage as a barometer of Beijing's policies. There is really a lot of other fodder in the Xinhua archives that could be subjected to similar research. Another possibility: Making the topic Xinhua itself -- for intance, examining how the news agency covered major subjects during the 1980s, or 1990s.
There's also the methodology issue. To recap, I used the following methodology in my research paper for Prof. Kuhn's class:
1) Used simple frequency counts to draw out patterns in the data
2) Used a traditional historical approach to explain the patterns as well as analyze qualitatively the news articles themselves.
Prof. Kuhn suggests that rather than using my hybrid approach, I could have taken the quantitative approach a step further, or several steps further:
Your reasoning is plausible in view of the evidence you cite from non-numerical sources. I wonder, though, if there could be a way to use a finer-meshed array of frequency counts to support one or more of these hypotheses? Having gone this far, it should be possible to construct plausible strings or phrases to check out -- to support the Indonesian angle, for instance, or the Soviet angle. .... In short, I think you have made an excellent start thought quantitative methods. Can it be refined past the hypotheses stage, or will it always require verification from the non-quantititative side?
Hmmm. This got me thinking ... could I have applied a finer mesh of searches to support my hypotheses? To tease out the Soviet angle, certainly: I could take the results of searches I performed on "overseas Chinese", "Vietnam" and "Kampuchea" and seen how many of the results also mentioned the USSR.
But to tease out the editorial angle of these stories, I am not sure how much simple searches would have helped ... that is, a search on the Xinhua archive for Vietnam and the Soviet Union might turn up lots of results, but I think only a human review could determine if those stories were hostile to Vietnam and the USSR vs. ambivalent and even friendly. It is incorrect to assume that Xinhua would naturally be hostile toward the USSR, as Chinese policy toward the USSR at one timehttp://www.blogger.com/img/gl.link.gif was very friendly, and conceivably there were alternating periods of friendliness/hostility after 1977, depending on personal or policy issues between the two countries.
Additionally, there are degrees of friendliness and hostility that frequency counts wouldn't be able to tease out, except under the most contrived scenarios.
After thinking about these issues in the car, and at home, I have come to the conclusion that my methodologies might take two routes:
1) A content analysis consisting of a frequency count followed my manual coding for meaning, like Scharrer's content analyis of news coverage of Hillary Rodham Clinton
2) A two-stage frequency count content analysis ... one to determine basic trends in Xinhua coverage, followed by analysys of the results and the formation of more hypotheses, which could then be tested by another round of frequency counts.
More on this soon ...
No comments:
Post a Comment