Evaluating Non-response Rates and Bias in the American Time Use Survey

Abstract

This paper investigates the ongoing issue of non-response in the American Time Use Survey (ATUS) and its impact on the survey's reliability and representativeness. Despite targeted efforts, ATUS has experienced a consistent decrease in response rates, which was exacerbated during the COVID-19 pandemic. This decreasing response rate may link to potential non-response bias, which could distort the survey's depiction of time use patterns across the U.S. population. The study analyzes the complexities associated with non-response rates, emphasizing biases that might emerge from groups with weaker community ties. Challenging traditional assumptions, recent research indicates that high response rates do not automatically mitigate bias. This paper highlights the necessity for innovative survey methods and the adoption of new technologies, such as web-administered diaries and smartphone apps, to address these issues. The analysis underscores the importance of adapting survey strategies to contemporary societal and technological landscapes, aiming to enhance the accuracy, efficiency, and representativeness of ATUS data.