Saturday, November 23, 2013

"And the survey says?" - Polling the Davidson Electorate

Public polling of our local municipal elections is a rarity.  Polling costs money and takes time - both things that are in short supply when it comes to town elections that see relatively little money raised by candidates and the core campaign period spans only about 30 days.

But what if someone was willing to take the time and do it for free?  What if the person coordinating the polling effort actually had a national reputation in the field of analyzing elections data?  What if the survey could be done with a level of scientific accuracy comparable to the national surveys we all see and hear about in the daily news?

Would it be worth it then?  Yes, absolutely!

In Davidson this election cycle, just such a survey occurred in the run up to the election.  Conducted by Assistant Professor Josh Putnam's Public Opinion class at  Davidson College, a survey was completed in the week before election day.  If you follow politics, you may have heard of Professor Putnam, he writes the Frontloading HQ blog and made a bit of a name for himself during the 2012 election cycle - being one of only 3 nationally known "pundits" to nail the Electoral College results in the election for President of the United States.

Scaling back a bit from the national stage, he and his public policy class conducted a survey during the last week of October hand delivering surveys to randomly selected registered voters, picking them back up, and then tallying the results.  Can you say labor intensive!?!?

The survey covered not only the election, but also some of the policy questions facing the town.  It was remarkably balanced in how close it came to the actual demographics of Davidson - something harder to do with a small population where small numbers can make big percentage differences.  The margin of error was +/- 6.3%.  That's a little outside the bounds of what larger surveys tend to achieve, but still close enough to allow for valuable information to be collected.  It was conducted as a survey of registered voters, not 'likely' voters.

Here are some bits gleaned from the results:
  •  Everyone says they will vote, but most still do not. - 75% of survey respondents said they planned to vote.  That meshes with everything I heard while going door-to-door where just about everyone said they would vote.  Davidson's actual voter turnout? Less than 25%.
  • People still don't like the idea of HOT Lanes on I77. - After a full year of selling the plan to expand I-77 with tolls by installing new HOT lanes, supporters failed to move the dial in their favor at all.  56% of survey respondents said they did not approve of the plan.  Survey results from the National Citizen Survey published in March of 2012 had the number opposed at 53%.  All of Davidson's incumbents supported HOT lanes.  All of the challengers opposed them to varying degrees.
  • Mayor Woods's level of support lower than his vote totals would imply. - 56% of voters approve of the way John Woods handles his job as Mayor.  While he received 95% of votes cast this election, nearly 25% voters left this race blank or wrote in another candidate on their ballot.  These poll numbers imply that a fare number of people voted for John Woods simply because he was the only name on the ballot.  Interestingly, his "approval" number in this poll comes in much closer his voter percentage in the 2011 election when he did have a challenger.  In that election he received 59.67% of the overall vote while actually losing Precinct 127.
  • The race was tight all the way to the end. - Conducted in the last week before the election, four of the candidates ultimately in the top 5 on election day were also in the top five of this poll.  Beth Cashion had the same large lead she enjoyed on election day.  However, at polling time 42% still selected "not sure" - meaning a very large percentage of people made their decisions in the last days before the election.  The below chart shows the difference in percentage of total "votes" received in the poll vs. those on election day.  As you can see a larger chunk of the undecided voters swung toward two of the challengers, Stacey Anderson and myself.  This is a common occurrence in elections where undecided voters choose at the end and go with new faces.  This swing is what pushed Stacey Anderson ahead of Connie Wessner and into the fifth and final spot.

  • Endorsements mattered. - As mentioned in the previous post, several endorsements impacted this election.  That post also mentioned the River Run "endorsement" of neighborhood candidates as the "most important".  Here's why. That endorsement came out on November 1st - after the polling data for this survey was collected.  River Run has over 750 households.  It is not hard to see that this would have some positive impact for the candidates mentioned in that endorsement - Stacey Anderson and Vince Winegardner.  With Ms Anderson having lower name recognition than Mr Winegardner (he is a former River Run POA president), she had more potential upside benefit of the endorsement email on undecided voters. Ultimately, Stacey Anderson edged out incumbent Connie Wessner by only 59 votes.  That's a margin that certainly could have come from this single endorsement in such a large neighborhood.  Causing the first incumbent to lose re-election since 1997 certainly warrants making this endorsement the "most important".  (Incidentally, I also attribute my positive swing in the final numbers to endorsements.   The last major effort of my campaign was to ensure as many people as possible knew of the endorsements I received with one final distribution of campaign literature that focused on these in large part.)
 As you can see, surveys - even in small elections - can be both informative and useful when analyzing how our elections occur and what impacts them.  They provide a window into what the public is thinking, how they receive and digest information, and what tactics work when candidates are trying to communicate.  As these results show, when done right they can also be accurate gauges of what the ultimate outcome will be.

My hope is that this type of survey continues to be done for our elections.  It would also be great if our town embraced using these tools when making large decisions - something I spoke about frequently as part of my own campaign.  Not everything can or should go to a voter referendum, but getting real and meaningful citizen input before making big decisions that impact our quality of life should be a desired goal for all of our elected officials.

To see the detailed data from this survey, click here.

No comments:

Post a Comment