Ok, Google: 'What is the Role of Google Ranking for Selective Exposure to Political Information in a Google Search Media Environment?'

This article examines the role of ranking versus source and story headline for political information seeking strategy in a Google search media environment. With the emergence of the Internet, individuals utilize search engines as a compass to navigate the Internet and to tailor information seeking. Due to time and cognitive constraints, individuals are neither willing nor able to consume the whole amount of information provided by search engines. They rest on heuristic cues to select political information seeking. Departing from a traditional media environment where mostly source matters for selecting political information, I consider that additional heuristic cues such as a story headline (content) and ranking (bandwagon) provide additional cues to drive political information seeking strategy. This article contributes to the debate about selective exposure in a changing media environment. The transition from a low to a high-choice and from a “push” to a “pull” information environment allows extensive and easy access to an almost unlimited information environment. These characteristics force scholars to reassess the selective exposure debate. In this changing media environment, search engines are potentially the most relevant devices for contemporary information seeking. However, to the best of my knowledge, no study examines the role of a search engine as a device to seek political information. This article is a first attempt to close this gap. This article reports the results of an online survey experiment during a campaign for a referendum vote which combines a tax and pension reform in Switzerland in May 2019. The goal is to study contemporary information seeking behavior in a search engine media environment. To do so, I mimic a Google web interface. First, respondents are instructed to type what kind of information they want to obtain regarding the referendum in a Google search bar. Second, respondents are exposed to 10 Google news content predetermined. They are instructed to pick as many sources as they want to form their opinion. Respondents are randomly assigned to one out of four conditions: 10 news content randomly assigned (control group) (a); 10 news content with fixed ranking (b); 5 news content randomly ranked 1 to 5, and 5 others news content randomly ranked 6 to 10 (c); 2 news content randomly ranked 1 to 2, and 8 others news content randomly ranked 3 to 10 (d). The impact of the ads heuristic cue is also measured. To maximize real-world accuracy, I reproduce a detailed Google web interface (e.g. layout and colors) and extract Google news story headline from repeated real-world Google search during the campaign. Additionally, I postulate that nobody goes to the second page on Google. The results permit to assess the role of ranking (bandwagon) versus source and story headline in driving selective exposure to political information in a Google information environment.