ABSTRACT. Objective: Types of college drinkers have been identifi ed using traditional measures (e.g., 12-month drinking frequency). We used an alternative multidimensional approach based on daily reports of alcohol behaviors to identify college drinker statuses, each with a unique behavioral profi le. The current study aimed to (a) identify drinker statuses at the week level across four semesters, (b) examine the predictive utility of drinker status by testing associations with senior-year hazardous drinking and dependence symptoms, and (c) identify concurrent predictors (gender, drinking motivations, hazardous drinking, any dependence symptoms) of senior-year drinker status. We also compared the week-level drinker statuses with drinker statuses identifi ed using traditional measures. Method: A multi-ethnic sample of U.S. college students completed 14-day bursts of daily web surveys across college (91%-96% completed ≥6 daily reports of the sampled week). Analyses focus on nine alcohol-related behaviors (including estimated blood alcohol concentration, pregaming, and drinking games) assessed daily in spring/sophomore year to fall/senior year and drinking motivations, hazardous drinking, and dependence symptoms assessed fall/senior year (n = 569; 56% women). Results: Four week-level drinker statuses were replicated across semesters: Nondrinker, Light Weekend, Heavy Weekend, and Heavy Frequent. Across semesters, drinker status was associated with senior-year hazardous drinking and any dependence symptoms. Senior-year fun/social motivations were also associated with senioryear drinker status. Differences in behavioral profi les between weeklevel drinker statuses and those identifi ed using traditional measures were found. Conclusions: Replicable week-level drinker statuses were identifi ed, suggesting consistency in possible types of drinking weeks. Drinker statuses were predictive of senior-year hazardous drinking and dependence symptoms. (J. Stud. Alcohol Drugs, 77, 38-50, 2016) Received: October 13, 2014. Revision: July 13, 2015. Funding for data collection and work on this study was provided by National Institute on Alcohol Abuse and Alcoholism (NIAAA) Grant R01 AA016016 (to Jennifer L. Maggs). Dr. Fairlie received support from National Institute on Drug Abuse (NIDA) Grant T32 DA017629. Dr. Lanza received support from NIDA Grant P50 DA010075. NIAAA and NIDA had no role in the study design, collection, analysis, or interpretation of the data; writing the manuscript; or the decision to submit the article for publication. The T RADITIONAL ALCOHOL MEASURES ask respondents to aggregate behavior across time (e.g., 12 months). Measures often include typical quantity consumed per occasion, frequency of heavy episodic drinking (HED), and frequency of getting drunk (Dauber et al., 2009;Reboussin et al., 2006). Aggregate estimates usefully describe how people differ from each other and how individuals change over time. The current study takes the alternative approach of using daily reports assessed over 1 week in each ...