Sunday’s results leave a scenario far from the clear victory of the PP over the PSOE that many of the polls pointed to. Less than it seems, but far enough away to open one of the most classic debates on post-election Mondays: Why do the polls fail?
Much has been said about the problems of surveys to contact profiles that abstain more or vote for options that are socially frowned upon. A problem that clearly explains a part of the error. Also of the need to understand the margins of error involved in predicting the distribution of seats for 52 constituencies. But we often forget about one of the main problems of the polls to guess the result of the elections: it is difficult to measure what the citizens will vote if the citizens do not know it.
For a long time now, political scientists have wondered how it is possible that the polls vary so much if the factors that explain the vote are, for the most part, stable and predictable months before the elections. Elements that determine the vote such as the economic management of a government, the groups that the candidates have questioned or the popularity of the actors and decisions made throughout the legislature, tend to change little during the 15 days of a campaign. So why are the polls so moving and the vote seem so unpredictable?
The answer reached in 1993 by political scientists Gelman and King is that campaigns are a period in which many voters learn and reflect on what has happened during the legislature and incorporate it into their voting decision. The movements in the polls capture this learning process of what the candidates and the groups they have questioned have done as it occurs. A process that is not linear because campaign messages do not always reach the same intensity or the same voters. The polls move because voters take time to collect and integrate the information that the campaign gives them to make the decision. They take so long that an increasingly large percentage does not do so until hours before going to the polls, which makes it very difficult to predict through questions about voting intentions. It is difficult for a citizen to inform us well of a decision that he has not made.
That’s not to say that surveys aren’t reliable or useful. The surveys show the changes generated by the information that citizens have already learned and managed. Nobody expected results from the PP similar to those of 2019, because the citizens of space had already incorporated the changes in space (the disappearance of Cs or the PP-Vox pacts at a regional and local level) into their decision and they could thus inform the pollsters. But it does mean that the questions about voting intentions will hardly give a stable and reliable picture of what will happen on their own. It is necessary to understand what has changed on the stage, what has already been incorporated by the voters, and what has not. If we do not link the polls to the underlying political movements, our reading will easily leave us unable to anticipate the decision of the entire population that has not yet decided.
