Nov 9, 2012

Worse than expected... this result based on the posterior of pundits

And the circus that was the press coverage of the 2012 presidential election continues on! That ever-so-degrading bastion of news-turned-entertainment, CNN, has now posted this story, the equivalent of the Monday morning water-cooler talk about why the football game was lost on Sunday (in this case, it was the Democrat Donkeys versus the Republican Elephants).

Here's a snippet:
Was it the wrong candidate, low voter turnout, a few dumb comments, a superstorm or falling out of touch with a shifting American demographic? 
Listening to Republicans try to explain what went wrong in their worse-than-expected election thumping reveals a party struggling to define itself amid continuing change in the nation it seeks to lead.
Worse than expected!? What universe do these people live in?

The short story is that we basically knew who was going to win this election back in June. Why? Because an unbiased, statistically-driven accumulation of polls actually IS a good way to estimate something. This is a well-known technique in statistical analysis. If the sample is unbiased compared to the whole, then it is, within statistical uncertainty, a good estimate of the true value. It also gets better with more samples. This is called the "central limit theorem". It's not the "central limit wild-assed guess pulled from the general posterior of television pundits". It's not even a "model". It's a mathematical fact.

Of course, this points to the answer already, but it makes a story that isn't equally rosy to both sides of the question: the Republican party is indeed falling out of touch with a shifting American demographic. The facts are even directly in the news story. Here they are based on one statistic over the abortion issue:
According to exit polls, 59% of voters think abortion should be legal in the United States, compared with 36% who said it should be illegal.
The numbers show a clear story: the pro-choice position is more popular than the anti-abortion position by a margin of 63%. No matter how you spin that, it is a simple fact. How does CNN spin this? See the rest of the quote.
Meanwhile, Marjorie Dannenfelser of the anti-abortion rights group Susan B. Anthony List criticized Romney for failing to make abortion a central campaign issue. Stronger attacks on the administration's policies, such as a requirement that religious organizations provide their employees with health care coverage that she said included abortion-inducing drugs, would have attracted more support from uncommitted voters, Dannenfelser argued.

"He took all the right stances," she said. "The problem was not communicating on the national stage with Obama what (the president's) actual positions were."

Her group endorsed Romney, Dannenfelser said, adding, "We assumed that given who he was, he would make (abortion) more of a national issue."

According to exit polls, 59% of voters think abortion should be legal in the United States, compared with 36% who said it should be illegal.
So facts are hidden in a tiny little sentence at the end of a four-paragraph statement from one right-wingnut about how this would have won the election for Romney, had he just made this a central issue. Well, CNN, there is a very easy way to spin this correctly: "The statements from Dannenfelser contradict the reality from exit polls, where 59% of voters think abortion should be legal in the United States, compared with 36% who said it should be illegal". Easy, right? A simple addition to the statement turns this from giving equal treatment to wingnutters and reality, to a demonstration, based on evidence, that the Republican Right is falling out of touch with reality.

This is yet another example of how numbers and reality can trump entertainment-inpsired mealy-mouthed spins. Let's try to keep that all in perspective, shall we?


Staid Winnow said...

Good to see someone as pissed about this media nonsense

Gene Bernice said...

Well presented and explained.