News Release

College students likely drink much more alcohol than they think they do

Peer-Reviewed Publication

Alcoholism: Clinical & Experimental Research

The embargo on this release has changed since it originally posted.
  • Previous research has found that college students underestimate their alcohol consumption.
  • Recent findings confirm that college students tend to overestimate volumes, over-pour drinks, and under-report levels of consumption.
  • Educational feedback appears to improve understanding of actual consumption levels.

Previous research has shown that college students tend to put too much alcohol into what is considered a "drink," likely leading to inaccuracies in self-reported consumption, which is a mainstay of surveys on college drinking. A study in the April issue of Alcoholism: Clinical & Experimental Research confirms that college students tend to overestimate volumes, over-pour drinks, and under-report levels of consumption. However, educational feedback regarding definitions of standard drinks appears to improve students' understanding of how much they actually drink.

"Just about everything we know about how much college students drink comes from survey studies," said Aaron White, assistant professor in the department of psychiatry at Duke University Medical Center and first author of the study. "We ask them to tell us how much they drink and we assume that their answers are totally accurate. In order for students to be accurate, they have to know how much alcohol constitutes a single serving. It turns out that they don't. When we ask them to pour us drinks of different types, they tend to pour too much. When we ask them to simply define how many ounces there should be in a single drink, they tend to give us numbers that are way too big. This tells us a few things. The first is that we have totally failed to teach students some of the most basic information about alcohol - what a single serving is. It also tells us that students' answers on alcohol surveys are probably inaccurate. How could they be [accurate] if researchers and students have different definitions of how big a single serving is?"

A "standard drink" is generally defined as a beverage that contains 0.6 ounces of pure alcohol. The most commonly used standard drinks are 12 ounces of five-percent beer, five ounces of 12-percent wine, or 1.5 ounces of 40-percent (80 proof) distilled spirits.

"The problem is that a typical drinker probably doesn't think this way," said White. "To many people, a 'drink' is one serving regardless of how big it is. Someone who has had four gin and tonics would probably give the same answer on an alcohol survey as someone who has had four Long Island iced teas, when in fact the latter could contain three times as much alcohol as the former."

"A lot of health information about alcohol is conveyed in terms of 'drinks,'" added Susan Tapert, assistant professor of psychiatry at the VA San Diego Healthcare System and the University of California at San Diego. "For example, cardiovascular benefits have been associated with one drink per day for females and one to two drinks per day for males. It is critical for people to understand that when these messages say 'a drink,' they do not mean a Big Gulp filled with alcohol."

For this study, researchers recruited 133 (78 males, 55 females) undergraduate students to complete an alcohol survey as well as "free pour" a single beer, glass of wine, shot of liquor, or the amount of liquor in a mixed drink. Roughly half of the subjects received feedback regarding their definitions of standard drinks. All participants were then resurveyed about their recent alcohol consumption.

"We found that there is a relationship between how big someone thinks a drink should be and how much alcohol they pour when asked to pour single drinks," said White. "The bigger the definition of a drink, the bigger the drink they pour."

"Interestingly, after receiving detailed feedback and instruction on the definition and volume of 'a drink,'" said Tapert, "the college students became more accurate in determining drink size and consequently reported that they typically consumed a larger number of beverages than they had previously reported. This suggests that college students, and perhaps other groups, may underreport the amount of alcohol they consume, but that education about the definition of 'a drink' can improve the accuracy of self-reported drinking."

"If we tell students that binge drinking – four drinks for females and five for males at a time – is dangerous and we want them to stay under this limit, then we absolutely must do a better job of teaching them what a drink is," added White. "Not doing so would be like telling people that it's OK to have one order of french fries per day without telling them whether we mean small or super-sized."

White suggested that college itself would be a good place to start, either during freshman orientation or via a school-wide e-mail. He also suggested that manufacturers of alcoholic beverages step up to the proverbial plate. "Why they are not required to place serving-size information on their containers, or why they do not do so voluntarily, is way beyond my understanding," he said. "If you buy grape juice in a grocery store, all you have to do is look at the label to find out how many ounces are in a serving and how many servings are in the entire container. If that information is on grape juice, then why isn't it on wine, which is just fermented grape juice?"

Both Tapert and White noted the implications of these findings for how alcohol researchers assess drinking behavior. "The findings certainly do not suggest that we should stop using surveys to study college drinking," said White. "Surveys are still one of the most powerful tools we have. The data simply suggest that we shouldn't take students' answers to questions about how much they drink at face value until we are certain that all students are being taught how to better define standard drinks."

"We need to improve the accuracy of self-reported drinking, so that our research can be as precise as possible," added Tapert. "Furthermore, I would be very interested to know the extent to which older adults and adolescents make similar estimations of their alcohol consumption."

"Teaching people what constitutes a single drink … applies to the general public, too," said White. "Most people are aware that a drink or two per day could convey benefits for the heart. This might be true, but it's also clear that going over that limit can increase a person's risk of a variety of diseases, particularly in women. So, it's sort of dangerous, in my opinion, to tell people that a drink or two per day is safe without also teaching them what we mean by the term 'drink.' My guess is that lots of people think they only have one or two drinks per day, when in fact they have much more."

###

Alcoholism: Clinical & Experimental Research (ACER) is the official journal of the Research Society on Alcoholism and the International Society for Biomedical Research on Alcoholism. Co-authors of the ACER paper, "College Students Lack Knowledge of Standard Drink Volumes: Implications for Definitions of Risky Drinking Based on Survey Data," were: Courtney L. Kraus, Julie D. Flom, Lori A. Kestenbaum, Jamie R. Mitchell and Kunal Shah of the Department of Psychiatry at Duke University Medical Center; and H. Scott Swartzwelder of the Department of Psychiatry at Duke University Medical Center and Neurology Research Laboratories at Durham VA Medical Center. The study was funded by the Institute for Medical Research, and the National Institute on Alcohol Abuse and Alcoholism.


Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.