Abstract

The aim of this article is to present and apply to the Italian context a method, alternative to those typically used, which is able to detect and quantify the capacity of different research institutes to propose unbiased estimates of electoral preferences. This is an innovative tool mainly because it combines the results from different polls conducted by various research organizations throughout the election campaign. The application of this method - proposed by Jackman in 2005 - to the Italian political elections of 2008 shows that voting intentions for the various political parties have been systematically distorted by most research institutions. Specifically, the tendency to distortion is accentuated at the two parties Sinistra arcobaleno and Leghe (Lega Nord and Mpa).