The world of political data is often confusing for people. However, after the 2012 election, many political operatives opened up on the ways that data were used. Ethan Roeder, the Director of Data of Obama for America penned an opinion piece to explain how the 2012 Obama campaign utilized political data.
Roeder declares, “I am not big brother,” and that political data analysts aren’t emulating George Orwell’s all-seeing government. He continues to explain that the collection and use of political data in contemporary political campaigns are not well understood.
The article further states that the information that political campaigns use is either publicly available or it is freely given up by individuals to the campaign. Roeder says, “You may chafe at how much the online world knows about you, but campaigns don’t know anything more about your online behavior than any retailer, news outlet or savvy blogger.” From there, the campaign analyzes these pieces of information, and in the case of the Obama campaign, created scores to predict how likely it is that individuals would vote and how likely it is that they would support Obama.
This article is compelling to me because it simply states the realities of political targeting. Even with my rudimentary experience with some of the analytic tools that campaigns use, I notice a deep misunderstanding between campaign tactics and journalists.
For example, in December, 2015, the Bernie Sanders’ campaign inappropriately accessed Hillary Clinton’s campaign information and political journalists’ coverage of the story was misguided. The two campaigns use the same database contractor through the Democratic National Committee called NGPVAN. This technology is powerful, but not that difficult to understand.
Regardless, journalists had no knowledge of VAN, the centerpiece of modern democratic campaigns, and kept getting information wrong. For example, The New York Times reported that, “Audit trails of the logs show that people with the Sanders campaign searched and saved multiple files, creating new lists of their own.” This statement is misleading because the audit actually showed that the campaign saved searches, not lists. This distinction is important because it could provide insight into the intent of the people within the Sanders campaign. In VAN, lists are generally used as a static list of people to be exported, but searches are dynamic parameters that return a different set of people depending on new information that has been uploaded into VAN. In the case of the data breach, the Sanders campaign actually saved searches, not lists.
It is all very nuanced, but reporters could have uncovered the differences with a few calls to those with even the most basic training in VAN. The errors go on and on, but the upshot of the coverage is clear: there is a disconnect between politics and political reporters.
Opinion pieces like Ethan Roeder’s are important because they explain to the public how campaigns and governments are operating. I believe the people have the right to know how campaigns are being waged and news agencies should encourage more explanatory pieces by political operatives to fill the knowledge gap.