It appears that I am not the first researcher to consider joining SCT and Twitter; Bobby Rozzell (I’m guessing an undergraduate at the Elliott School of Communication, Wichita State University) joined the two in 2008.
“Wichita “Twitters” about the 2008 Presidential Election: Fantasy Theme Analysis of Messages During Three Election Night Time Phase” was a paper delivered by Rozzell to the 5th Annual GRASP Symposium, Wichita State University, in 2009.
In the paper Rozzell notes that he/she used Glaser’s (1965) Constant Comparative method of qualitative analysis. However, Rozzell doesn’t go on to explain why that particular analytic method was chosen and what other methods were originally considered and ultimately rejected in favour of Glaser’s method.
Rozzell took three distinct time periods as units of analysis; I wonder if Glaser’s method will support a different unit, that of a fortnight’s Twitter output. Because as Wu found (2011), traditional CAQDAS tools aren’t ready for Twitter:
We found it difficult to use currently available CAQDAS tools for such interactive analysis as they are all based on dealing with long files, but not one line messages.
Trying to fit one-line messages into Nvivo made no sense and Wu resorted to using Excel for analysis.
Glaser, B.G. 1965. The Constant Comparative Method of Qualitative Analysis. Social Problems, Vol. 12, No. 4 (Spring, 1965), pp. 436-445
Rozzell, B.L. 2009. Wichita “Twitters” about the 2008 Presidential Election: Fantasy Theme Analysis of Messages During Three Election Night Time Phase. Proceedings of the 5th Annual GRASP Symposium, Wichita State University.
Wu, M. 2011. Symbolic Convergence Theory and Analysis of Online Discussion. Proceedings of QUALITATIVE RESEARCH IN WEB 2.0
ASIA-PACIFIC, Macau, Feb 25.