Cofounder Pulse Poll – Finding a cure for the common poll, Round 2


Key Takeaway: BOTH campaigns for US Senate are Correct!! Senator Paul holds a narrow lead, but Mayor Gray is very, very close.
Several months ago, we launched a poll here at Babbage Cofounder to get a “pulse” of what Kentuckians thought about the campaigns of the two leading candidates in the race for U.S. Senate here in Kentucky – sitting U.S. Senator Rand Paul and Lexington Mayor Jim Gray. While some respected pundits, political thinkers and seasoned operatives disagreed with our poll’s methodology, we found that Senator Paul held a 2 point advantage over his challenger.
After months of dormancy, political junkies (if you’re reading this, you know who you are) are beginning to see signs of an actual U.S. Senate race here in Kentucky. We’ve even seen the first traces of internal polling from at least one of the candidates – Mayor Jim Gray, who announced recently that his internals show him in a dead heat with Sen. Paul.  Washington experts ranking upset possibilities don’t mention Kentucky.

So, we decided, let’s run our own little poll again and see what it says! And hey, since we’re not “real” pollsters anyway, it’s all in good fun!

The latest Cofounder Pulse Poll, run from Wednesday, June 22 to Friday, June 24, shows Senator Paul with a 1 point lead over Mayor Gray. But LOTS of folks still haven’t made up their minds.

As was the case with the first poll we ran back in April, this is a representative sample of how the internet population in Kentucky feels about a particular issue or political race. We again used Google Consumer Surveys,which makes use of inferred demographic and location information to employ stratified sampling method by distributing the surveys based on the targeted audience to Google’s publisher network and/or Android smartphone users. Google infers demographics through respondents’ browsing history (DoubleClick cookies for age & gender and IP address for geography), then they match them against existing government statistical data. Google Consumer Surveys uses post-stratification weighting to compensate for sample deficiencies to remove bias among the survey sample. This gives a more accurate result with lower root mean square error (RMSE) which also makes the results better represent the Current Population Survey (CPS).

Google Consumer Surveys has its limitations. However, in 2012, Nate Silver, then of the New York Times FiveThirtyEight blog fame, concluded that Google Consumer Surveys was the #1 most accurate poll online and the #2 most accurate poll overall.

And, as mentioned before, with what we’ve seen from recent problems with polling in other election cycles, we’re just trying to find a cure for the common poll.  We certainly hope there are other polls in the works to clarify U.S. Senate Contest.  Only 132 days until November 8th.
You’ll find below a breakdown of various cross tabs for your reading enjoyment.

Overview of the June 22, 2016 Cofounder Pulse Poll with results tab

Inferred gender breakout


admincoCofounder Pulse Poll – Finding a cure for the common poll, Round 2