In recent months, there’s been a lot of talk about the Obama campaign’s vaunted ‘voter probability scoring system’. The purpose of developing such a system relates to improving the targeting of different types of outreach during a campaign. As explained by MIT’s Technology Review in their 2012 election post-mortem:
“Obama’s targeters had assigned every voter in the country a pair of scores based on the probability that the individual would perform two distinct actions that mattered to the campaign: casting a ballot and supporting Obama... This innovation was most valued in the field. There, an almost perfect cycle of microtargeting models directed volunteers to scripted conversations with specific voters at the door or over the phone. ”
In recent months we’ve customized NationBuilder’s Political Capital feature to allow groups to develop increasingly complex supporter engagement measurement, and this got us thinking: what would a simple voting probability scoring system look like?
One concept we came up with was to strip down the scoring system to only social media interaction data, and merge this with eligible voter data and past voter histories. Under this system, you would allocate ‘points’ towards such interactions as Facebook event RSVPs, Likes and Shares along with Twitter following, mentions and retweets for your candidate or party – and that’s all. With some monitoring to filter out negative feedback, over the course of an election you would develop a ‘score’ for a series of potential voters who you may not have been able to reach through field/canvassing or traditional phone banks. To take action on this group of likely supporters you would want to make sure that you’ve added a voter file or list of electors to your Nation, as well as having connected your social media accounts and performed a proper deduplication of your lists.
If you were looking for an additional data source to add to your score, you could use an Advanced Search to filter based upon previous support for your party.
In this use-case, once a supporter eclipsed a predetermined score-level you would update their support level to being a 1 or 2 (political speak for strong supporter). If you were more cautious you could leverage NationBuilder’s inferred support level to track likely supporters acquired this way. Otherwise, on election day you would query your database for any supporters with a score and add them to your list of voter identified throughout the campaign for turnout.