This paper explores the demographics of Twitter users, focusing on a dataset representing over 1% of the U.S. population. The authors analyze the geographic distribution, gender, and race/ethnicity of Twitter users to understand their representativeness compared to the general U.S. population. They find that Twitter users are more likely to live in populous counties, with a significant male bias and a highly geographically-dependent distribution of race/ethnicity. The study highlights the non-uniform nature of the Twitter population and sets the foundation for future research to improve the use of Twitter data for social science applications.This paper explores the demographics of Twitter users, focusing on a dataset representing over 1% of the U.S. population. The authors analyze the geographic distribution, gender, and race/ethnicity of Twitter users to understand their representativeness compared to the general U.S. population. They find that Twitter users are more likely to live in populous counties, with a significant male bias and a highly geographically-dependent distribution of race/ethnicity. The study highlights the non-uniform nature of the Twitter population and sets the foundation for future research to improve the use of Twitter data for social science applications.