With widespread concern for increased alcohol use during the COVID-19 pandemic, there is a pressing need to examine changes in young adults’ alcohol use and to identify antecedents of increased use. We tested the hypothesis that self-reported changes in alcohol use during the pandemic (frequency, quantity, heavy episodic drinking) would relate to perceptions of peers’ changes in alcohol use. In April of 2020, 507 college students self-reported changes in their alcohol use and perceived changes in use for typical students at their university (i.e., norms). Most students in our sample reported decreased alcohol use and perceived decreases in peers’ alcohol use. Perceptions of peers’ changes in alcohol use behavior strongly related to changes in students’ own alcohol use. Findings provide strong support for norms-based strategies that can correct normative misperceptions by highlighting the fact that most college students are not in fact engaging in heavier alcohol use during the COVID-19 pandemic.

Background The contexts in which young adults drink alcohol play a salient role in alcohol‐related outcomes and negative consequences at an event‐level, but less is known about longitudinal risks. We collected longitudinal monthly data across 2 years on (a) daytime drinking, (b) pregaming/pre‐partying, and (c) playing drinking games. We then examined associations between drinking in these contexts and within‐person variability in alcohol consumption, consequences, and simultaneous alcohol and marijuana (SAM) use in a given month. We also examined the extent to which drinking in these contexts predicted hazardous drinking (using Alcohol Use Disorders Identification Test (AUDIT) scores) at a distal follow‐up timepoint, controlling for baseline AUDIT scores. Method A community sample of 759 young adults (Mage = 21.1 years; 56.4% female) completed monthly surveys for 24 consecutive months and a distal 30‐month follow‐up. Multilevel models estimated within‐ and between‐person associations between drinking context frequencies (daytime drinking, pregaming, drinking games) and alcohol‐related outcomes (weekly consumption, consequences, SAM use). A single‐level negative binomial regression tested associations between drinking context frequency averages across a two‐year period and changes in AUDIT scores from baseline to a follow‐up 2.5 years later. Results Over 75% of the sample of non‐abstaining young adults reported drinking in each of the three contexts at least once during the 24‐month period. Within‐persons, young adults reported greater consumption, more negative consequences, and increased likelihood of SAM use during months that they drank in these contexts more often than usual. Each context was associated with negative consequences, even when controlling for alcohol use frequency. More frequent daytime drinking and pregaming, but not drinking games, were associated with increases in AUDIT scores at the 30‐month follow‐up, suggesting that there are potential long‐term risks of drinking in these contexts. Conclusions Findings suggest that daytime drinking, pregaming, and playing drinking games are high‐risk contexts in terms of month‐to‐month and long‐term risks. Additional research is needed on the various contexts in which young adults drink alcohol and the extent to which contextual factors interact with one another to amplify/reduce risks and harms.