Automated collection and coding of online campaign ads

This project aims to develop a novel software data collection and analysis platform that enables studies of microtargeting in online election campaigns for the first time. This microtargeting has serious implications for how our elections and our democracy function. We will configure various profiles of our web bots to test how the messaging of campaigns differs based on the perception of the voter receiving the ad. In particular, we manipulate the age, gender, language, partisanship, and geography of the perceived voters to test the campaigns abilities to microtarget. The automated analysis aims to extract answers of interests to political science scholars from the streaming ad content.  Four basic research questions will be examined: 1) Can campaigns effectively microtarget voters when advertising online? 2) Do campaigns engage in issue convergence in online advertising? 3) Are candidate more likely to engage in negative advertising in online advertisements? 4) Do these strategies vary based on their ability to microtarget?  The results will advance our understanding of how campaigns function and the ability of candidates and other political actors to target voters with online ads. Tools developed in this project will be available for public access.

Results:

Funding Organization: National Science Foundation

Duration: 08/31/2020

Award Amount: $529,776.00

Award Number: 1729775

Principal Investigator(s): David A. PetersonWallapak TavanapongAdisak Sukul, Olga Chyzh