I needed a quick and easy way to compare changes in policy documents and quantify the changes. However, I didn´t want to go into large-scale text analysis, because I also wanted to read the documents (or at least skim them and see what the changes are all about). The process was less „Give me the numbers!“ but more „I want to read the documents, have the changes highlighted, and oh, by the way, could you tell me how many words were changed?“ I used the compare documents function of word to show me the changes, and this little macro was extremely helpful in counting (or at least approximating) the number of changed words.
At the end of the week, I´m on my way to the ECPR general conference in Prague. In the panel Politics of Energy transformation, I will present some work that I do together with Eva Ruffing on public participation in German energy grid planning. And I have the great pleasure to be discussant in the Religious factor in Morality Politics panel. There are some exciting new developments in the field. I hope I can give some valuable (or at least not completely nonsensical) comments.
I have a new blog post for the ivory tower blog. In a nutshell: One of my favourite topics when teaching international organizations is the Weberian view of international organizations as bureaucracies articulated by Martha Finnemore and Michael Barnett. The UNHCR field manuals provide a very nice real-world illustration of this perspective and are very helpful to illustrate the argument that definitions and categories are extremely powerful.
Now, this may seem an odd way to use stata, but I found it even more cumbersome to do the same thing in Excel. What I wanted to do: To make an argument about institutional procedures, I need some kind of project timeline graph showing how long certain procedures take (and where they overlap).
I inserted my data into stata in the following way:
|FNA approval procedure / EIA||15aug2012||2|
|FNA approval procedure / EIA||26nov2012||2|
Each part of the procedure has two observations, one for the starting time, one for the end of the procedure. The value variable is just to have a nice ordering of the different procedures on the y-axis.
Now, we can make a series of connected or line graphs, that connect the start- and the end-date for each procedure.
twoway (line value time if value == 4, lwidth(vvthick) lpattern(solid)) (scatter value time if value == 3, msize(vlarge) msymbol(circle)) (line value time if value == 2, lwidth(vvthick) lpattern(solid)) (line value time if value == 1, lwidth(vvthick) lpattern(solid)), ytitle("") ylabel(0(1)4, nolabels) xtitle("") xlabel(#7) legend(order(1 "TSO consultation" 3 "FNA approval procedure / EIA" 2 "Second draft published" 4 "FNA consultation")) scheme(colorblind)
And here it is (as always: colorblind scheme by Daniel Bischof)
Could be better, but this is just a first shot. Basically, the argument I want to make stands out: The German federal network agency (FNA) can hardly use the information it gets during its own consultation, because its own consultation starts when the whole approval procedure is already running. Thus, the FNA uses the information gained in the consultations done by the transmission system operators (TSOs).
Am Mittwoch den 25.5. halte ich einen Vortrag am Institut für Ethik und Geschichte der Medizin der Universität Göttingen. Es geht dabei um das deutsche Embryonenschutzgesetz im Lichte international vergleichender Daten und die Frage, wie speziell der deutsche Fall eigentlich ist.
Im wesentlichen basiert der Vortrag auf den folgenden Publikationen:
Fink, Simon (2007): Ein deutscher Sonderweg? Die deutsche Embryonenforschungspolitik im Licht international vergleichender Daten. Leviathan 1/2007, 107-128 (Abstract)
Fink, Simon (2008): Forschungspolitik zwischen Innovation und Lebensschutz. Die Determinanten von Embryonenforschungspolitiken im internationalen Vergleich. Baden-Baden: Nomos (Reihe Staatslehre und politische Verwaltung). (Abstract)