Election Polling Beyond the Ballot

When election polls are released the headline is typically about the candidate ballot test, the so-called “horse race.”  It’s the lead paragraph and primary focus of the story. It’s what’s considered “news” among journalists and the public. Oftentimes, though, these data points are not the story. They are a story, but without context, the interpretation and newsworthiness may be questionable (more on this point in a future post).

Here are four survey questions I believe should be in most election surveys to add that context and aid in the analysis and reporting:

• Generic ballot: This question examines whether the electorate prefers the Democratic or Republican candidate, regardless of who those candidates actually are. It asks, “If the election were held today, for whom would you vote: the Democratic candidate or the Republican candidate.
  Why it’s important: Elections are often personality driven (see Donald Trump) so understanding the predisposition of the electorate compared to how each candidate is polling (e.g., 55% prefer the unnamed Republican yet just 40% are supporting the named GOP candidate) yields insight into both a candidate’s current performance as well as their presumptive floor and ceiling.

• Job approval: If the election includes an incumbent, how do voters perceive the job she or he is doing?
  Why it’s important: An incumbent with a 40% job approval and 42% on the ballot test probably doesn’t have much room to reach the 50% + 1 vote threshold necessary to win most elections. So, “leading on the ballot 42% to 37%” isn’t really a lead, it’s likely the ceiling, or at least a sign the incumbent’s support is weak.

• Issue landscape: This question may be asked open-ended (What do you believe is the most important issue facing your area?) or closed-ended with a list (Which one of the following is the most important issue facing your area?). An even better question reframes it to “What is the most important issue in deciding for whom you will vote for (president, governor, senate, congress, mayor, etc.).
  Why it’s important: Understanding the issues that are important to voters and how well each candidate is addressing these issues in their campaign (or not) helps to foreshadow potential vote support and movement in the ballot numbers. A candidate who is “leading” on the ballot but focused on an issue of low importance may be eclipsed on election day by the trailing candidate who is laser-focused on the most compelling issue on voters’ minds; this reversal is thus less of an “upset” and more of a story on sound strategy and execution.

• Mood of the electorate: This question asks voters whether they believe things in (their community, state, the U.S., etc.) are going in the right direction or off on the wrong track.
  Why it’s important: When I first started polling, incumbents did not generally win re-election when most voters thought things were off wrong track. However, with the perpetual national mood being negative (wrong track), this metric is now generally reserved as an indicator in state and local races. When the mood is positive (right direction), incumbent governors, mayors and others benefit and may outperform polling numbers on the ballot.

Are there questions you look for in election polls? If so, what are they and how do you use them? I’d love to hear from you.