As election day draws closer, political analysts are poring over polling number in order to come up with what they think is the most accurate representation of where the candidates stand. This week, the New York Times took another shot at this in their article, Who Will Be President, by Josh Katz. Katz’s article presents an in-depth analysis of the latest and most creditable polling data, as well as a state-by-state breakdown of voting estimations.
The Times gives Clinton an 85% chance to beat Trump for the highest office of the land. While this estimation might seem confident in Clinton’s ability to shore up the voters she needs to reach 270 electoral votes, the Times analysis follows along the same general trend as other respected news outlets. Sites such as FiveThirtyEight, Princeton Election Consortium, and the Cook Political Report all represent the same forecast that Clinton has around a 75% – 80% chance for victory in November.
While Kuntz does offer a detailed analysis of current polling data and expert prediction, the article does not address criticism concerning the validity of the data under observation. For years polls have been slated because they have the ability to misrepresent the demographics being polled with wide sweeping judgments and general conclusions. Writer Jonathan Bernstein presents a compelling argument against the validity of polls. Bernstein’s main idea is that most polling organizations use two-candidate surveys, which fail to incorporate demographics that support third party candidates. This system also fails to consider the effect of undecided voters across the board.
Most of the polls that are used in the New York Times statistical models are based on incomplete data sets that would lead the researcher to potentially draw misguided conclusions. These polls often fail to compile an accurate representation of the general public. This can happen because not enough people were polled, inconsistent research standards were used, and disproportionate voter demographics were representative of the overall study.
This election has brought with it a large increase in voter turnout. What this will mean for the outcome of the election is still unknown; however, it will certainly change the make-up of demographics, and how political analysts interpret polling data from now on.
(featured image from shutterstock)