inShare0
The right survey is closely related to constitutional nature, and must therefore be analyzed in the context of the legal and institutional context in which the surveys are conducted, making demanding their analysis and understanding of perspective abroad. This is what is observed in particular with the American example of the current election campaign pitting Barack Obama and Mitt Romney.
For a long time there was no attention paid to France in polls in the U.S. presidential election because Obama seemed far ahead and saw Europe and especially in our country, where Democrats have traditionally more understanding than their opponents, this result seemed to satisfy everyone and was therefore not examined.
As we know, the situation has changed, however, after the first televised debate that saw Mitt Romney beat Barack Obama and back sharply in the race or pass - so it seems, unlike the French tradition, the debate between candidates have, there, a real impact, but it is another matter. Since then, the national polls, the French press widely relays, give the two candidates side by side.
The opportunity to look at these surveys in order to produce an analysis: the first law analysis of comparative surveys of this blog, here made of course with great modesty, pending the integration of comparative law in the second edition of our book. It could not be otherwise: the first comparative law could only be spent in the United States.
A first attempt in any case quite interesting that harbinger of exciting research and the validity of the program to build a right of election polls.
The U.S. presidential election through polls: Romney wins the popular vote but Obama wins the election
Le Monde published a number of very interesting articles on American surveys, the most recent, published in its issue of October 23, compares the various U.S. polls published in recent days, with just the last debate between the candidates.
This paper reports an survey for NBC and the Wall Street Journal, published Sunday, October 21, which gave the two candidates with 47% in the polls among potential voters. The previous national poll NBC / WSJ credited the incumbent 49% of the vote, against 46% for the Republican candidate. Another survey, the institute” Public Policy Polling, gave the two candidates for full equality in voting intentions, 48% everywhere, according to interviews conducted over the past three days.
As for the daily Gallup, who previously gave a wide lead in poll Mitt Romney, he reported on the same day an even bigger gap in the polls at the national level: 52% to 45% among likely voters for Republican.
In contrast, another survey, conducted by the Institute IBD / TIPP, gave nearly 6 points ahead to Barack Obama in the polls at the national level (47.9% to 42.2%).
Nevertheless, we must be careful of any of the foregoing analysis of the election results from these surveys. In fact, as the article notes the world, if national polls are useful as they design trends, they do not show the real chances of victory for either candidate for election U.S. decides the number of electors and not the popular vote. The analysis of the surveys should be carried out according to the legal and cultural context specific to the United States.
On this point, but we’ll come back, attached to a New York Times blog is useful. It gives effect to the electoral college 294 votes to 244 for Obama against Romney and 73% chance of winning the election!
Also, a visit to the website of Real Politics is necessary because the site also offers the score at the national level, a score state by state, the results appear more disparate.
A further element of complexity seen in France, where it must be admitted that the election by universal suffrage of the President by all French citizens at the same time (in the marginal reserve the quantitative point of view of French nationals abroad and Overseas) greatly facilitates the analysis.
These elements invite you to return to the peculiarity of the right surveys in the United States and intersecting on the specifics of its electoral system. This study then used to further deepen the fundamental reflection on the relationship between surveys, polls and the right to democracy.
The peculiarities of the law of the American polls heterogeneity and complexity
On the right surveys, it is exercised, it seems the United States as a self-regulation, which may be more differences in methodologies, even if at this stage of our research, the The idea may be issued only as a hypothesis.
Thus, in an article on the differences of opinion polls in the U.S. elections, Le Monde and more precisely one of its journalists, Luc Vinogradoff, highlight differences in method, perhaps, could not happen in France because of increased control of voting intention polls by the Commission surveys and peculiarities of French electoral system.
Three elements of questions emerge: rules [or lack of rules] for the respondents, policy adjustments samples and American specificity due to the size and structure of the state, methods of making websites summaries of election polls.
People registered on the electoral lists or individuals may register to vote?
First, the article notes that unlike France where registration on the electoral roll ends several months before an election, Americans can still register until late in the campaign, allowing the use of two different methods : either test the voting intentions of those listed on the electoral roll, or test the voting intentions only people likely to vote, thus including those not yet registered to vote but could still do it.
However, according to the method used, the results are different: and Gallup, which had hitherto used only criterion inscribed on the electoral lists for daily poll voters changed method on October 9, which increased score Mitt Romney, due to the fact that Republicans are more motivated and more likely to move to vote in this campaign.
What (s) method (s) of political redress samples?
Secondly, it seems that there are also differences in the methods of recovery, including recovery policy. Thus, if we are to believe the article, it appears that the pollsters do their political adjustments vary more significantly than in the French surveys, where the commonly used method is the memory of vote: these adjustments, based not on voting memories but statements of sympathy - in the sense of supporters - would be highly variable.
Thus, a very recent survey by the Pew Center giving Mitt Romney leads with 49% of the vote against 45% for Obama included more voters themselves as Republicans (36%) than Democrats (31%), with plus 30% of “independent” voters, while a survey in September of the same institute, which gave 8 points ahead of Obama (51% to 43%) included a sample of 29% of Republicans 39% of Democrats and 30% independent.
At first glance, such a method therefore appears less stringent than the recovery of the French sample based on a memory of vote, not a statement of sympathy for a candidate. Strange solution that raises questions, so, especially since it seems to vary considerably over time and produce mixed results - in violation of the “essential methodological consistency” chèr the Commission surveys?
This freedom has also been strong criticism in the U.S. media, they Whereas the choice of the recovery of the sample bias the results, and critical differences method. On this point, we can actually query, as demonstrated including the recent controversy over a Gallup poll gives Mitt Romney with a seven-point lead and had been criticized in the United States, which has also made controversial report World account.
Which method to synthesize data from 53 institutes active holes testing the polls at the federal level as at Federated?
The latter section of the World deserves also to be read as it also reveals the complexity of reading surveys in the United States, and by contrast highlights how our political structure makes our particularly simple to achieve political polls and comment.
Thus, the article reports the blog FiveThirtyEight Nate Silver, which compiles the results of the 136 institutions that follow the campaign! Of these, 53 are active, and have published at least one survey from October 1. It is therefore especially sites like RealClearPolitics, FiveThirtyEight and should be consulted for an analysis of the U.S. situation because they are the synthesis and the average of the poll.
But the question of how we make this synthesis there also arises. And it can also vary, giving very different results.
Indeed, the FiveThirtyEight blog gives the two candidates neck and neck in terms of popular vote -% for Barack Obama against% for Mitt Romney, but gives the incumbent president winner in the major constituents as Barack Obama would have 294 votes against 244 for his opponent, giving him a chance to win the election to 73%!
The site will RealPolitics in the same direction but is more nuanced, with significantly different figures: according to this site, Romney has an advantage in terms of popular vote since the place% of the vote against 47% for Obama, who But still the winner with 201 electoral votes against 191 for Mitt Romney - the rest were undecided.
These sites are therefore a synthesis level of complexity and variation still problems of interpretation.
Synthesis
The right of American surveys seems much more heterogeneous and loose than ours and allow more heterogeneity, and the application of methods, views of France, question.
These differences are in any case very interesting and therefore call for further analysis of comparative law. What are the basic rules of the system, what are the differences with ours and how to enjoy it?
This is a set of questions that will be essential to meet, and even more so than differences in law polls adds very special constitutional context of the United States which makes it even more difficult to read the opinion polls.
The peculiarities of the American constitutional system, federal state election by indirect universal suffrage, electoral and election polls
An additional complexity of reading surveys in the United States is of course the federal structure of the state and the peculiarity of the system of election of the President in the United States.
The complexity of the system for electing the U.S. President
The American presidential election is an indirect vote for the election of the electoral college that elects the President of the United States and the Vice President. The electors then meet in each state to formally elect the President and Vice President of the United States. The votes are finally counted before a joint session of Congress in early January.
The appointment of electors and candidate selection are subject to rules established by each State where traditions come from more or less formalized, which reinforces even more complex.
Each of the fifty states elects a number of “electors” equal to the number of Representatives and Senators. The most populous state, California, has 55 votes, while the eight least populous states have only three each.
The consequences of these votes are also different depending on the state and the system itself has inequality: indeed, in all but two states, Maine and Nebraska, the electoral system gives all the voices of the State arrived at the first candidate is the rule of “the winner-take-all.” It is this system that explains the disparity between popular results, which, in recent elections, were neighbors between Republicans and Democrats, and the results of electors who often provide an overwhelming majority of the nominees.
The consequences of this complexity in terms of reading opinion polls
Of course, this particular structure introduces an element of complexity in basic reading surveys in the United States. Thus while most national polls now give Mitt Romney with a slight lead, he could lose the election because of the system of indirect suffrage.
Thus, as we noted above, the reading of the FiveThirtyEight blog is quite fascinating - even if we take these claims, again, with caution, as the reporter here is his own kitchen. Thus, it appears on this site if the two candidates are neck and neck in terms of popular vote -% for Barack Obama against% for Mitt Romney, on the contrary they are in a very different position in terms of electoral votes since Barack Obama had 294 votes against 244 for his opponent, giving him a chance - up - to win the election to 73%!
RealPolitics the site is more nuanced, but going in the same direction, even announcing a defeat Romney with a majority of popular votes. According to this site, Romney has indeed an advantage in terms of popular vote since it places Romney% of the vote against 47% for Obama, which remains the winner with 201 electoral votes against 191 for Mitt Romney.
A superficial reading of the headlines, in fact, could lead to many misinterpretations. A problem that we know very well in France too, and that is without asking the right questions in terms of surveys! That will be one of the topics of the next section of this blog realize that exegesis of the third report of the Commission surveys.
To an elected president without popular support? Surveys, representative democracy and democracy continues
All these elements of analysis finally leads us to wonder, finally, if we are moving to an elected without the majority of popular votes president and what lessons can be learned in terms of democracy.
The discrepancy between the electoral and the popular vote, as we know, is possible: one can quote the presidential election of 1972, when the Republican candidate Richard Nixon was elected with over 95% of the electoral votes so that He had taken only 60% of the popular vote.
And worse, it would not be the first time a minority president would win the election: in the 2000 presidential election, Democratic candidate Al Gore received 550,000 more votes than his Republican opponent George Bush at national, but 550 votes ahead Bush has officially obtained in Florida allowed him to get all the electors of this state and to win the election at the federal level. A democratic scandal in which he is not required to return.
Is this the fate of Barack Obama?
Nobody knows yet, but one thing is certain here, surveys appear to be an instrument of democracy against representative democracy, particularly striking. They could participate in the questioning of the election to the structure so controversial?
Ultimately, therefore, this study reinforces the thesis, widely defended our work, the surveys are provided to be well controlled and well used, a fundamental instrument of democracy. In France … as elsewhere, not just in democratic states!
Stay tuned: highly November 6th!
Romain Rambaud
Share and Enjoy
See more about: best home elliptical