Since the late seventies there is a lot of support for the notion that voter volatility is normal. The so-called “frozen cleavages” were a temporary exception. Especially the Netherlands shows high voter volatility and was even called “exceptionally unstable”. A logical mechanism can be formulated that is based on the geographic location of voters. This mechanical approach demonstrates that the perceived volitality is bounded. When investigating three aspects of volatility (vote transfer, number of parties, social cleavages) it becomes clear that volatility can be limited to certain parties in certain situations and during certain periods. Voters can be called “loyal” to parties or within blocks. Although voters may be considered volatile (adrift), objective was to show whether it is possible to find specific voters at zip code level (around 35 people per zip code) that vote according to the campaign objectives “yes”, “no”, or “turnout”. This was tested with 2,144 active campaigners visiting 1,549,323 addresses in 66,040 Dutch zip codes while canvassing or distributing flyers. The result is that this is indeed possible. Flyering has an effect of 1 in 54-62 for the issue “no” and 1 in 22-33 for “turnout”. Canvassing has an effect of 1 in 7 for the issue “no” and 1 in 6 for “turnout”. Effect in this case is a combined effect of voter propensity and campaign effectivity which is not distinguished within the limits of this paper. There were not enough campaigners in the “yes”-campaign to produce valid results, but we do have some anecdotal data.
The application of the logical mechanism to design three referendum campaigns in the Netherlands confirms historic election data can predict the location of voter groups and their susceptibility to campaign messages. Voter volatility is not an obstacle.
Figures - uploaded by
Joost SmitsAuthor contentAll figure content in this area was uploaded by Joost Smits
Content may be subject to copyright.