I'm specifically interested in electoral volatility. It measures the aggregate change in votes across all parties between two elections. But what if the distance between elections (in years) changes frequently? It seems reasonable to expect that countries with longer distances between elections would have higher levels of volatility. It also seems reasonable to expect that volatility would be different in a country if elections were held at 2 years apart and then at 7 years apart (for hypothetical). So: How could/should I adjust a country's electoral volatility if I want to "normalize" it (to take into account distance between the two elections) for inclusion in a cross-national statistical model? I haven't been able to find any leads on this through Google Scholar or other sources.
Mr. Miguel! The model varies depending on the area, traditions and circumstances. For example, in Eastern European countries may have another thesis – because of the political crises and early elections had higher levels of volatility. In countries with a stable democracy I can be agree with your thesis: “countries with longer distance between elections would have higher levels of volatility” such Hungary, Poland, France, Italy, etc. However, if you help, you can find some interesting analysis in various countries in the addresses:
http://kodu.ut.ee/~allans/ma/ma.doc
http://true-european-voter.eu/sites/default/files/Mapping%20electoral%20volatility_Dassonneville_Hooghe.pdf
https://files.nyu.edu/jat7/public/2014_Ref_Journal_Article_BJPS_Powell_Tucker.pdf
http://www.mwpweb.eu/1/166/resources/publication_1752_1.pdf
Thank you. The Epperly article seems especially promising, but I will look at all the others with great interest.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Statistical Modeling