Quinnipiac: The Poll Unskewers Miss the Point

Quinnipiac reports:

Four years ago, supporters of Mitt Romney claimed that the Quinnipiac University Poll and other public polls were skewed because the partisan composition of their samples didn’t line up with exit polls from the 2008 election. They suggested that the only polls that were right were polls showing an electorate with similar demographics, and also conveniently showing Romney in the lead. Thanks to these “polling unskewers,” Romney clung to the mistaken belief that he would win the election right up until election night.

Just as the polling unskewers were wrong in 2012, the polling unskewers of 2016 are wrong again. They assume that the racial composition of current polls must line up with the 2012 exit polls. This is guess work, not scientific polling. Polls should measure what the electorate is, not what somebody thinks it should be.

One example that highlights the folly of relying on past exit polls to predict the future is the partisan shift from 2004 to 2008. In 2004, exit polls showed the electorate was evenly divided between Democrats and Republicans. In 2008, Democrats had a 7 point advantage. Different elections affect the electorate in different ways.

If there was ever an election to be skeptical about assumptions based on exit polls from previous elections, it’s this one. This presidential election already has defied more predictions than any in modern political history. There are so many “firsts” that could affect turn-out, nobody knows what will happen. We will have the first major party candidate with no experience in either public office or a military record, we will have the first woman presidential nominee, and we already have the two most unpopular candidates in history. On top of that, we’ve never seen a candidate campaign as Donald Trump has. It is foolhardy to guess what the electorate will look like 4 months before this unprecedented election.

An article worth your time.