relevance of surveys predicting the results of the elections
Politic surveys really borned in 1936, during the American presidential elections. ... The American press predicts that London is going to win after using what we call the “straw votes” which is the way to realize surveys since the beginning of the 19th century. ... The theoretical and practical way to make a politic survey The political parties, the candidates and the media hold of the surveys to measure the support of the voters for candidates or parties. The surveys are useful, but in condition of being carried out, of being presented, of being interpreted correctly. Moreover the political surveys don’t have to be confused with the electoral estimate. The surveys are based on the declarations of intent of vote whereas the estimate is based on the analysis of the examination of the ballot papers after the close of the poll. When the results of a survey are announced, the methodology used for the creation of the survey must also be made public. ... The surveys, based on declarations of intent, are realized through the different survey’s institutes. ... The politic surveys use in general, for the constitution of the sample to question, the method of the quotas. ... The sample size used for this kind of surveys is about 1000 people. ... An other criteria is very important for the estimation of the reliability of the results: the distribution of the answers. ... But in practice, we consider that this margin is comparable to the surveys realized with a random pulling. ... How to collect the information Nowadays, most of the surveys of intention of vote are realized on the phone which is more often used than the face to face that was present few years ago but which tends to disappear. ... The results are faster and it avoids facing with issues like closed doors or going in dangerous districts. There is one last thing to know before the analysis of the relevance of the surveys through the elections. ... x during the last elections whereas his score was 15%, the intentions of vote for the next one are adjusted in this way. ... First of all we should talk about the errors in the communication of the results that involves a kind of mistrust face to the surveys. The main point is the difference between the politicians who tell the results and the journalists who present them and analyze them. The journalists prefer to clearly, and in a funny way, communicate the results than to let this task to a technical driller who will pay attention to every single detail. The lack of rigor in the interpretation of the results are interpreted in different manners. The surveys may be wrong for different reasons: At a mathematical level, we can notice a bad control of the notion of percentage.