Home > General > Public Policy Polling: Uhhhh-nalysis

Public Policy Polling: Uhhhh-nalysis

What passes for analysis with the folks over at PPP is stunning. What’s even worse is the Media’s willingness to take what they release at face value, refusing to do even a miniscule amount of analysis, and blast ridiculous non-truths across the lands. I rarely read the PPP “analysis,” or what I will now refer to as uhhhh-nalysis, and head straight for the crosstabs for what occasional useful information can be found.

A certain PPP post caught my attention. A March 22nd release of a Michigan poll lead off their uhhhh-nalysis with this:  “Rick Snyder trounced Democrat Virg Bernero last November to take Michigan’s governorship, but if voters could do it over right now, they would narrowly elect Bernero instead.” First, the poll results show that Bernero’s lead is 47-45, which can hardly be turned into anything other than “a toss-up.” But, if true, this would be a stunning turnaround from November.  In 2010, Snyder walloped Bernero by an impressive 58.1% to 39.9%, or 18.2 points. Yet when looking at whom PPP polled, they had voted for Snyder by a much slimmer +6, 49-43%, undersampling original Snyder voters by almost 10 points while oversampling Bernero’s support by over 3 points. This is why it is important to weight the data to accurately reflect the population, especially when you use very small study samples as PPP does.

When I re-run the numbers using PPP’s data properly we find the following: First, the poll shows that Snyder maintains 84% of the votes he received and picks up 3% of the votes his opponent received.  Fine. Even then, this amount to about 1.71 million votes (down from 1.874 in 2010). His Democrat opponent Bernero keeps 93% of his 2010 vote and picks up 9% of Snyder’s 2010 vote for a total of 1.365 million votes (up slightly from 1.28 million votes).

When accurately calculating the PPP data, the new result is Snyder 56%, Bernero 44%. So, how can I come up with +12% Snyder while PPP comes up with +2% Bernero, all while looking at the same exact data? Simply put, PPP, like most Democrat polling firms, are fully disinterested in portraying accurate public opinion. They don’t care if their polling data had an 80% Democrat share, they’d run the story without a mention. For whatever reason (and the reasons are often debated) polling samples/sizes almost always favor Democrats/liberals. If you fail to acknowledge and account that, you fail your job as an analyst.

Polling outfits like PPP enable some of the most outrageous assertions to be blasted by the Mainstream Media, who are but useless tools in the spread of such disinformation. And the MSM are all to unwilling participants. The Detroit Press and Michigan Live (and other MSM sources in the state), freely publish the claim without the slightest hesitation or curiosity, pushing the non-existent point that Michigan voters want Snyder gone.

Categories: General

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: