Not all students drink heavily
       Obviously, college students are not the only people on the planet that occassionally drink more than they should.  However, compared to their non-college peers, they do tend to drink more heavily.  Interestingly, if one looks at the drinking levels of high school students, those bound for college tend to drink less while in high school, but then quickly begin to out-drink their non-college peers once arriving at college.  Alcohol use then tapers off again once college students graduate.  In other words, there really seems to be something about the college environment that promotes, or at least supports, higher than normal levels of alcohol consumption.  The figure below, a modified version of a slide from the NIAAA College Drinking website, illustrates the change in drinking levels over time for college students and their non-college peers. 

       It is important to point out that heavy drinking college students do not represent a random slice of the college demographic. Some students are much more likely than others to drink irresponsibly. The prototypical hard drinking college student is a Caucasian male fraternity member at a homogeneous institution with a high density of alcohol outlets and cheap drink prices. As a group, these individuals drink more heavily than other students. African-American students drink the least and are more likely to be abstainers than Caucasian students. Latino and Asian students typically fall somewhere in between. The type of school that a student attends plays a significant role in influencing drinking behavior. Even Caucasian male fraternity members tend to drink less when they attend schools with a large minority population (Wechsler and Kuo, 2003). Traditionally, female students at all-female schools drink far less than females at coed institutions, though this gap is quickly closing. Indeed, females in general represent a group in which drinking levels really have been on the rise during the past few years (Wechsler et al., 2002).
        Below is a table from the NIAAA College Drinking website that summarizes some of the variables related to drinking levels at college:

      Given the high incidence of heavy drinking among Caucasian male fraternity members, it is quite tempting to conclude that the fraternity environment produces students that drink heavily. This might be partially true, but there is now clear evidence that the fraternity environment is not completely to blame. We have observed (White et al., 2002), as have others, that students who drink heavily during high school are more likely to join fraternities at college and continue to drink heavily. In other words, it appears that many students join fraternities in part because their existing tendency to drink heavily is embraced.
        We have also observed (White et al., 2002) that drinking habits tend to shift as students progress through their college careers. Specifically, freshmen tend to drink less often than seniors, but they drink much more heavily when they do drink. Because drinking is illegal for the vast majority of freshmen, it appears that they drink opportunistically, consuming the alcohol that they have quickly so that they will have a buzz when they go out to parties or bars where they might not have access to more alcohol.  People in the college health field often refer to this as frontloading.
       There are several other variables that influence the drinking habits of students.  For instance, students who attend schools with a high density of alcohol outlets, particularly those offering cheap drinks, are more likely to drink heavily than those at school where alcohol policies are strictly enforced and access to alcohol is limited (Williams et al., 2002; Wechsler et al., 2002; Weitzman et al., 2003). 
        The take home message here should be fairly obvious -- while some college students do drink irresponsibly, not all students do, and those that do drink irresponsibly do not represent a random sample of all college students.