Two surveys of principal investigators conducted between April 2020 and January 2021 reveal that while the COVID-19 pandemic's initial impacts on scientists' research time seem alleviated, there has been a decline in the rate of initiating new projects. This dimension of impact disproportionately affects female scientists and those with young children and appears to be homogeneous across fields. These findings may have implications for understanding the longterm effects of the pandemic on scientific research.The COVID-19 pandemic has disrupted the scientific enterprise [1][2][3] . Researchers in the "bench" sciences, female scientists, and those with young children experienced significant declines in research time and other publication-based metrics, according to data collected before the summer of 2020 (refs. [1][2][3][4][5][6][7][8] ). Now, more than a year into the pandemic and with multiple vaccines developed, circumstances have evolved substantially. This raises an important question: has the pandemic's impact on scientists evolved as well?To answer this question, we distributed a survey in January 2021 by randomly sampling USand Europe-based scientists across a wide range of scientific fields. Importantly, we adopted the same sampling strategy as a previous survey we conducted in April 2020 (ref. 1 ), which allowed us to directly compare the results of the surveys at these two very different stages of the pandemic (Supplementary Note 1 and Supplementary Fig. 1). In the January 2021 survey, we asked scientists many of the same questions from the April 2020 survey, including professional and demographic features. We also added new questions that compare their overall research activity and output in 2020 with 2019, including the number of new research publications, new submissions, new collaborators, and new research projects they started each year. Furthermore, we asked scientists whether or not they conducted any COVID-19-related research in 2020. In total, we collected responses from 6982 respondents across the two surveys who self-identified as faculty or principal investigators (Supplementary Note 2). To supplement our survey findings, we also conducted a series of analyses using a large-scale publication dataset, the Dimensions database, which captures both articles and preprints published up to the beginning of 2021.