This paper examines how the Internet is changing ideological segregation among American voters, focusing on online news consumption, offline news consumption, and face-to-face social interactions. The authors define ideological segregation using standard indices from the literature on racial segregation. They find that ideological segregation in online news consumption is low, higher than most offline news consumption, but significantly lower than face-to-face interactions with neighbors, co-workers, or family members. There is no evidence that the Internet is becoming more segregated over time. The paper explores two economic mechanisms that limit online segregation: vertical differentiation, where most news consumption is concentrated in a small number of relatively centrist sites, and the fact that many consumers get news from multiple outlets. The authors also discuss the distribution of ideological exposure across individuals and the changes in online segregation over time. They conclude that ideological segregation on the Internet is lower than many observers have conjectured and that this may be due to the economics of news markets.This paper examines how the Internet is changing ideological segregation among American voters, focusing on online news consumption, offline news consumption, and face-to-face social interactions. The authors define ideological segregation using standard indices from the literature on racial segregation. They find that ideological segregation in online news consumption is low, higher than most offline news consumption, but significantly lower than face-to-face interactions with neighbors, co-workers, or family members. There is no evidence that the Internet is becoming more segregated over time. The paper explores two economic mechanisms that limit online segregation: vertical differentiation, where most news consumption is concentrated in a small number of relatively centrist sites, and the fact that many consumers get news from multiple outlets. The authors also discuss the distribution of ideological exposure across individuals and the changes in online segregation over time. They conclude that ideological segregation on the Internet is lower than many observers have conjectured and that this may be due to the economics of news markets.