<p>The science of surveys is very, very sophisticated. It's not a matter of simply drawing up a few questions in your room. Instruments are validated by outside reviewers who have nothing to do with drawing them up, pre-tested multiple times on test populations, and confidence intervals determined accordingly. The colleges pay, and handsomely, to ensure survey reliability. Occasionally, as is the case with the Wechsler surveys, experimental data in "real time" is collected to see whether the results can be validated. (That is the basis for the observation that heavy drinking is generally well underestimated, both in number of drinks, and in the amount of alcohol contained within each one.) Comparative data are also utilized to try to find factors that impact observed rates across campuses. There are also studies conducted to see whether "attention to the problem" on campus impacts the way people answer the surveys, or whether the survey responses then vary further from observable behavior in "real time". </p>
<p>And I too think people should make their own decisions. Many folks LIKE the culture the way it is ("In one striking contrast, 25 percent of non-athletes and 82 percent of male athletes agreed that social life would suffer if drinking were subject to more restrictions on campus."). , while, as Dean Roseman notes:</p>
<p>She also pointed to the pages of comments toward the end [of the Report] as indicative of troubling norms permissive of bad behavior. Its people talking about how their lives are altered by other students drunken behavior, she said. I think people have felt silenced.</p>