Some pundits were surprised things could shift so dramatically this late in the game. But remember, this is the first presidential election since society became truly hyper-connected through broad use of smartphones and social media. Overnight changes are not just possible but inevitable, especially with mass-audience TV events.
The debate was watched by 67 million people, 42 percent more than watched the last Academy Awards telecast. With that many tuned in, any striking development was guaranteed to have impact.
But how exactly does this happen? How does a single televised discussion alter the entire conversation about the election, and the candidates’ standings in the polls?
Social media offer a massive trove of public reactions to every twist and turn in the campaign. Though the social universe is not a perfect mirror of the electorate, it’s a rich source of insights into what one large group of people is thinking and saying.
To find those insights, we’re introducing a new kind of election analysis we call The Three P’s. It’s very simple. Over the last several decades, traditional news media have increasingly covered the campaign as pure contest - the breathless tracking of who’s up and who’s down sometimes known as horse-race journalism. At the same time, coverage of the candidates as personalities has also grown, a spillover from the broader celebrity-focused culture. Critics traditionally argued that these trends resulted in less coverage of the policy issues that matter most in any election, and a poorly informed public.
But social platforms have radically altered the media landscape. Now, when campaign events are broadcast on TV, viewers respond by doing their own broadcasting to their own audiences. In an age when every citizen with access to technology can join the conversation, is our political process changing? Does the horse race still predominate over the issues? When one mass event reshapes the campaign, which factors are in play?
To answer these questions, we’ve used sophisticated language-based analytic tools to break down the social response to the election into three categories:
• Politics: the campaign itself, the horse race.
• Policy: the issues, from the economy to foreign policy to education.
• Personality: the candidates as people - their life stories, human qualities and quirks.
We determine what statistical proportion of the conversation - a metric called “share of voice” - is devoted to each of the P’s, and then track how the balance fluctuates over time.
We ran our initial Three P’s analysis on the first presidential debate and the results are shown below. The upper graph tracks the overall social conversation over the last few weeks. The lower one shows how the balance between Politics, Policy and Personality shifted in recent weeks.
Over the last several months, Politics generally has had the greatest share of voice, driven by the usual poll numbers and campaign-trail theatrics, as well as unexpected events like Venezuelan President Hugo Chavez’s comment that he would vote for Obama. The latter prompted a brief spike of chatter focused not on Venezuelan-US relations but horse-race speculation and humor.
Then came the very issue-focused debate, which increased Policy’s share of the conversation. As the chart shows, Policy and Politics converged. Although Personality’s share was consistently the lowest throughout this period, it also got a lift from the debate due to heavy tweeting and posting about the candidates’ contrasting energy levels.
Two days after the debate, the government announced the latest jobs data, showing an unexpected increase in U.S. employment. And Policy’s share spiked, briefly dominating the conversation.
Interestingly, the new jobs number was significant because of the political context created by the debate. At a moment when Romney was perceived to be on the rise, here was some apparent good news for Obama. Would the horse race change again? As this dynamic suggests, The Three P’s are closely intertwined.
As we move toward Election Day, the conversation will continue to shift among the three and we’ll update this analysis.