A slew of anti-Obama political robo-text messages made headlines this week, containing text such as: “Obama denies protection to babies that survive abortions. Obama is just wrong,” and, “Medicare goes bankrupt in 4000 days while Obama plays politics with your senior health.”

The texts were eventually traced back to a Virginia political research firm with “deep ties to GOP politicians and conservative causes” (no surprise, there), which is now under scrutiny, as it is illegal to send unsolicited, automated text messages. However, as the texts were sent from an email address, so the firm may be in a legal gray area.

This and other digital marketing strategies are relatively new issues discussed by political scientists. In the last election, Twitter, reddit, Tumblr and Instagram were fledging social networks if they existed at all. Today, all have the capacity to reach voters with unprecedented precision. Fuzzy logic allows companies to aggregate data about purchase and browsing history, credit information, social groups and a myriad of other personal information used to directly target advertising and messaging. These messages are not broadcasted, but rather “narrowcasted” – specifically tailored to an individual and designed to ensure the message reaches that person (Howard).

In the political science field, this fuzzy logic is coined as “indirect-inference public policy polling”, using demographic information, credit card purchases, etc. to make policy preference inferences (Howard). Firms are hired by political campaigns to digitize and gather this personal information, as well as performing tradition survey practices. Your information is then bought, sold and traded in a vast market that ranges from private citizens, lobbying groups, PACs, campaigns and grassroots movements.

This raises innumerable ethical dilemmas about the dissemination of personal information for political gain. Though most of the information used by data-miners is already public record, the buying and selling of our profiles seems to be an intrusion. Groups use this information to solicit donations and support from us, knowing roughly what our income bracket is, the things we’re interested in and how we spend our time. Is it a service that these messages are now more relevant and pertinent to us, or is it intrusive that this data is even utilized? Also, how secure is this data? How many third parties have access to it?

We discuss this topic often in journalism classes to discuss the future of targeting advertising – we’re only a few years away from the eye-scanned, personalized ads featured in the futuristic movie Minority Report. I believe it’s too early to tell the implications of these advances for political campaigns and the electorate, but needless to say the field is changing dramatically.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s