Polling averages show tight Kansas governor’s race

Political junkies in Kansas are being treated to an unusually large number of polls this election cycle, which gives people who are nerds about statistics a lot to work with.

The latest came out Tuesday night from SurveyUSA, which conducted a poll on behalf of KSN-TV in Wichita. It showed Democrat Paul Davis with an eight-point lead, 48-40 percent, over incumbent Republican Gov. Sam Brownback.

It was the third public poll conducted since the Aug. 5 primary, and all three have shown Davis ahead in the race, giving the Davis camp a boost of confidence. But when all three polls are averaged together, the results still show a close race.

The polls included one by Public Policy Polling, showing Davis ahead by two, 39-37 percent; and one by Rasmussen Reports showing Davis ahead by 10, 51-41 percent. Each poll was conducted differently using different sample sizes, resulting in slightly different margins of error. But when the three of them are combined as if they were all one poll, you come up with this:

Here, we’re looking at the raw number of respondents who indicated how they would vote if the election were held on the day they were polled. SurveyUSA provides those raw numbers. For the other two, the numbers were calculated using the reported percentages and sample size.

Mark Joslyn, a Kansas University political science professor who studies public opinion polling, said those numbers show the race is far from decided.

“The reasonably consistent low support for Brownback does indicate a close race,” Joslyn said in an email when asked to comment on the numbers. “In that regard, how firm voters’ support is and how the undecided break are keys to the eventual outcome.”

An important point about the table above is that we’ve combined the “other/undecided” voters with those who said they support Libertarian candidate Keen Umbehr. Two of the polls, SurveyUSA and PPP, gave people the option of picking Umbehr. The Rasmussen survey did not.

SurveyUSA shows Umbehr polling at 5 percent while PPP showed him at 9 percent. So if we give him an average of 7 percent, out of the 2,213 people sampled, that still leaves about 8.4 percent undecided.

The question for Brownback, then, is what can he do to sway those voters that he hasn’t already done during three and a half years in office. Given his high disapproval ratings (53 percent in Rasmussen; 55 percent in PPP), the answer is twofold: rewrite the narrative of his first term in hopes of changing voters’ assessment of him and launch a negative campaign to convince undecideds that Davis would be worse, both of which the Brownback campaign has been doing.

The latter may be an achievable task, given Davis’ current low name recognition. In the PPP poll, 41 percent said they’re not sure what to think of him. In the Rasmussen poll, 19 percent said they’ve never heard of him.

Thus, Davis’ challenge will be to connect with those undecided voters and make them feel comfortable voting for him before Brownback’s negative message has a chance to sink in.