Maarten Marx gives a presentation about DiLiPaD at the Talk of Europe Creative Camp on October 8, 2014 in Hilversum, The Netherlands.
DiLiPaD’s aim is to enable diachronic comparative research. We compare the three different countries and we compare parties or government vs opposition.
We will use both qualitative and quantitative methods. Below is an example of a qualitative method. We have counted the paragraphs in the Dutch parliamentary proceedings (both houses) containing the phrase bezuiniging or bezuinigen (noun and verb for “budget cuts”). For parliament as a whole and for each party and for speakers from the government we calculated the probability that a paragraph would contain these terms. We did this for each year.
In the figure we give the Kullback Leibler divergence between the probability for each subgroup and the probability for the whole parliament. KL divergence is a measure originating in information theory which measures how bad the probability distribution of the group is in modeling the distribution of the whole.
While these terms are intuitively mostly connected to the right wing, we see that the parties on the left use them more then average. We also see that the Christian party (CDA) has a recent increase in the use of the terms. Probably this has to do with the fact that they are now in opposition…
Click here for a larger image and the data in spreadsheet format.