An interesting, provocative question, asked most pointedly by musician Billy Bragg in the 1980s referring to class struggle in the context of Ronald Reagan and Margaret Thatcher.
But I heard it again in a graduate seminar this week, in a very different context. My background is in humanities-based film studies, but I'm in an empirical, social scientific
Mass Communications program at Texas Tech, so this has involved a lot of getting with the program. The most difficult transition for me is to A.P.A. style (M.L.A. is better, Chicago is my favorite); A.P.A. does not use first names of authors or capitalize most words in book titles.
The big dichotomy in the social sciences generally, and in Mass Comm especially, is between quantitative and qualititative research methods. The clearest way to think about the difference is that quantitative research translates data into numbers so it can be analyzed statistically. The question posed in class this week was whether content analysis is a qualitative or quantitative method.
The consensus is quantitative, and certainly, most content analyses I've read include statistical data analysis. But here's my point: the coding of content in order to make it quantitative is a
qualitative process.
A series of studies examining sexual and violent content in movie trailers by Mary Beth Oliver, et al. illustrates my point. In 2002, the authors conducted extensive content analyses of movie trailers and found that "approximately 76% of the previews in their sample featured at least one act of aggression (with an average of 2.5 aggressive acts per minute), and that 56% of the previews featured at least one sexual scene (with an average of 1.5 sexual scenes per minute)" (p. 597).
This is interesting, useful information, the kind of empirical evidence of commonly believed generalizations in which social science can excel. But these numbers were based on carefully articulated definitions of aggressive and sexual content -- is a character holding a weapon an aggressive act, for instance?
Not only do these numbers require qualitative judgments in order to be collected in the first place, they require qualitative contextualization in order to create any meaningful conclusions. In a subsequent study, Oliver et al. wrote, "Given that movie trailers have only an abbreviated amount of time in which to overview the plot of a film and to pique viewers' interest, violent content may effectively function as a quick and unambiguous indicator that the film will contain dramatic conflict..." (p. 599).
I am new to the social sciences, but when asked whether I'm interested in doing quantitative or qualitative research, my defiant answer is that purely quantitative questions only exist in the world of mathematics and statistics.
* * *
Reference: Mary Beth Oliver, et al., "Sexual and Violent Imagery in Movie Previews: Effects on Viewers' Perceptions and Anticipated Enjoyment,"
Journal of Broadcasting & Electronic Media 51.4 (2007), 596-614.