Polling and Accuracy Measure: Evidence from the Italian Case

Abstract

Over the last two decades, the use of pre-electoral polls during election campaigns has become increasingly commonplace. Moreover, their failure to predict the winner has caused concern about their reliability and accuracy. This raises the question underlies this paper: to what extent does polling accurately record real change in electoral preference? To answer this question, we analyze the last three Italian general elections (2006, 2008, and 2013) and their outcomes to see what evidence they provide concerning reliability and accuracy in the Italian case. Therefore, we employ a revised measure of A in order to fit better with the Italian political system. The main conclusion is that Italian pollsters were constantly wrong over the three general elections. There is a high percentage of polls classes as inaccurate in each election and, especially, in the last election.