Category: Data Journalism

Hypothesis testing about the 2016 elections

As a political scientist, there’s nothing like testing theories. One of the prevalent theory is that in a Western liberal democracy, to win a general election, you have to be the closest candidate to the center. Ever since Reagan won over, with the Democrats flocking towards him, and Thatcher did the same in UK, it has never been proved wrong.

Recently, these two new charts came to my notice, from YouGov.

Parties leaders left right spectrum-01

Tony Blair man of the people-01

Since Blair, Labour never won the election, because the Tory candidate was closest to the center.

2016, is a unique year, in both UK and US, and it is perhaps the golden opportunity to find out, if this hypothesis is true. Corbyn is hard left and Trump is hard right. According to this hypothesis, they will never win, and in this case the candidates closer to center, Hillary Clinton in US, and Theresa May in UK will sweep the polls.

We shall find out if that theory is universally true.

Excited?

Feature Photo

Is the media really ignoring Louisiana? No.

There’s a narrative in the media that the media is ignoring the flooding in Louisiana.

The public editor of the New York Times wrote the Times didn’t give it enough coverage. Mike Scott of the Louisiana-based Times-Picayune wrote on August 16 that the Times had only published its first story two days after the rains began.

But is it really true that the Louisiana flooding has been ignored? It’s a major news story, and yet days after the Times published its rebuke of itself, the narrative that Louisiana has been or is being ignored continues to spread down the media stream to columnists and bloggers. If the city of New York were destroyed in a hypothetical super storm, wouldn’t the media cover it?

We actually have precedents we can look at. We don’t need to wonder about hypotheticals. Just months ago in June floods in West Virginia killed 25 people. The floods in Louisiana killed 13. The number of houses that were damaged is reportedly very high–the governor puts it at 40,000. But there are natural disasters that killed more while also causing much, if not as much, damage. (By comparison, West Virginia’s governor said “thousands” of homes were damaged, but CNN reported the significantly lower total of 1,200. It may be too soon to know the property damage in dollars for either.)

Two more large-scale floods: those that killed 21 in Utah in 2015 after Hurricane Linda and the 2010 floods that killed 21 in Tennessee and 10 more in Kentucky and Mississippi while causing over $2 billion in damage.

A theory can be tested: If a dozen or more people were killed in say, Nashville, Tenn. or Hildale, Utah, would the media pay more attention to it than they are doing to the Louisiana floods?

Read More

Powered by WordPress & Theme by Anders Norén

Get the most important and interesting articles right at your inbox. Sign up for B+D periodic emails.