Theme 4: pre-reflection

I was looking for an article in the Journal of Communications, since my last read from that journal proved so interesting. I found a promising candidate in ”Cynics all around? The impact of Election News on Political Cynicism in Comparative Perspective.” (”N=48,872? That seems like a serious quantity. Let’s have a read.”)

The paper investigates the impact of strategy framing in campaign news on levels of political cynicism. Strategy framing is described as news-stories focusing more on the game characteristics of an ongoing political campaign than the substantive issues. For instance, using metaphors from sports, games or war, to explain the actions of parties or candidates in terms of them trying to advance their position or increase their chances of gaining influence, rather than reporting on the actual political issues being discussed.

Using a very large sample size, the study spanned across 21 different EU countries at the time of the EP election in 2009. The method employed was “a multimethod research design including a content analysis and a two-wave panel survey was employed, first, to investigate how the news media in the different EU member states have covered the campaign, and second, to assess the impact of such coverage on the decision of voters to turn out to vote.”

The first part of that multimethod is the content analysis. In that they used a large sample of news stories from all the 21 countries involved (N=48,872) and coded them based on strategic framing. The second part was a two-wave survey, conducted three weeks before and immediately after the election day in respective country.

The authors’ discussion about their data and methods is exhaustive and not very well suited for summary here, but after having read through it and been prompted to revisit the locked compartment in my mind dedicated to mathematical statistic I have substantial confidence in the dependability of the study’s results.

One thing worth commenting on is how the data collected in the surveys as well as the data coded in the content analysis is very much qualitative data, but it is then used in a quantitative way; that makes it important to keep in mind what can be said and what can’t be said about the results. When working with qualitative data in that way, you can use quantitative methods for analyzing trends, but you have to keep in mind that the data-points cannot be considered equidistant. For instance the 7-grade scales the survey participants are asked to rate their agreement with certain propositions on; the data is qualitative, and as such, the scale is in some way arbitrary.

The other paper, Bälter et. al., also shows a very high degree of methodological awareness. It’s interesting to read a paper like this, because a high degree of intellectual honesty shows though in the discussion about what the data could be said to implicate.

What I’ve been thinking mostly about after this weeks reading is how to think about qualitative and quantitative both in terms of methods and in terms of data, and how the two can easily be confused. Also how the two types of methods do a good job of complementing each other, provided you have a good understanding of how they work.

Quantitative methods enable the use of a much larger sample size, because the interesting variables are quantified and easily handled when you compare them to the variables looked at in a qualitative study. In a qualitative method, the focus lies on interpretation which means you may not even know about all the variables you are interested in going into the study.