Algorithmic Misjudgement in Google Search Results: Evidence from Auditing the US Online Electoral Information Environment

Google Search is an important way that people seek information about politics, and Google states that it is ``committed to providing timely and authoritative information on Google Search to help voters understand, navigate, and participate in democratic processes''. In this paper, we interrogate the extent to which government-maintained web domains are represented in the online environment of electoral information of the 2022 US midterm elections, as captured through Google Search results in 3.45 million SERPs for 786 locations across the United States between October and November 2022. Although we find that almost 40% of organic results are contributed by the 40% of government domains, this proportional equilibrium hides the fact that most results either belong to a small number of popular domains or are mistargeted (at a rate of 71.18%) with respect to the location of the search. We consider the frequent omission and mistargeting of non-federal websites a form of algorithmic misjudgement that contributes to civic harm, by obscuring the important role that these institutions play in the election information environment.