More or less than five percent - Does it really matter?

In Germany, the wasted vote thesis is quite popular both in political scien ce and among politicians. This widely held view is underpined by empirical evidence which suggests that it makes a difference how a voter perceives th e electoral prospects of his preferred parry. In this contribution, however , we discuss whether the usual approach to test the wasted vote thesis is a dequate. In particular, we argue that seemingly convincing empirical eviden ce in favour of the thesis might be interpreted in terms of psychological r ationalization. Multivariate analyses reveal that this theoretically reason able suspicion is empirically justified because the effect of the perceived electoral prospects vanishes if it is controlled for party identification. These results lead us to the conclusion that political parties need nor pa y as much attention to campaign efforts referring to their electoral prospe cts as they usually do. Additionally, our findings suggest chat it may be p romising to apply the argument presented in this contribution to other elec toral settings.