Trump-Card: Married Men vs Unmarried Women, The Skewed Battle Of Sex in USA

What are the prime reasons that generated the married men's bias against Hillary? This question becomes much more significant given that this syndrome was largely absent among the unmarried men, says Subhamoy Maitra.

The curve of interest quotient towards an important election is Bell-shaped with respect to time. Long before the election days, there is little interest. As we close the gap on the time-line, the interest rises. We start getting more and more surveys to guess the uncertainty ahead. The excitement peaks during the election days. Once the counting is over, the uncertainty vanishes. With each passing day, the number of words in various media outlets slowly precipitates. Given the exponential growth in the domain of computer and internet, the political analysis is now mostly quantitative than qualitative. There is less discussion on polity and policy now-a-days, but the focus is more towards data. For example, instead of studying the so-called `social justice’ of Mayawati from political viewpoint, we would be more interested to know what will happen if 2.5% of vote gets transferred from SP to BSP in the upcoming UP election.

 By itself, the US election result of this year presents a classic example of “less vote, more seat” scenario. This is generally a low probability event that happens very rarely. US election statistics have now confirmed that Trump possibly polled around two million votes (this is known as popular vote) less than Hillary (almost two percentage points) but won by a healthy margin of seats (referred to as electoral vote).

When we have a lot of data, one must do something with that instead of just sitting over it. You may be pleased to correlate the gain in vote percentage of ‘Didi’ with the number of researchers produced in China since 1980’s. Don’t get surprised if you discover a nice positive correlation. This is statistics, related to the perpetual theorem: “There are three kinds of lies: lies, damned lies, and statistics”. In fact, now some websites regularly list spurious correlations and one may surprisingly note considerable similarity between the number of “honey producing bee colonies” and “juvenile arrests for possession of marijuana” in US. Old proverb made new, we may now have a statement like “There are three kinds of truths: truths, half truths and statistics”. For the politicians, truth varies with time. For example, the half truths of Trump now become eternal ones after his victory. However, at a given snapshot of time, if the data can be investigated minutely, certain facts might surface. Related to election analysis, the data that you obtain after the election could be a logical platform for such fact finding. It is not possible for an individual to go through each and every data point related to an election. Rather, we mostly depend on summarized result. By itself, the US election result of this year presents a classic example of “less vote, more seat” scenario. This is generally a low probability event that happens very rarely. US election statistics have now confirmed that Trump possibly polled around two million votes (this is known as popular vote) less than Hillary (almost two percentage points) but won by a healthy margin of seats (referred to as electoral vote). There are still some minor legal hassles with this election, but that does not seem to affect the final result.

 

Image via http://theweek.com

Subhamoy Maitra is a Professor at Applied Statistics Unit in Indian Statistical Institute (ISI).

Be first to comment