Alcohol, mood and me (not you)

Most psychology research is done by asking a big group of people the same questions at the same time. "So we might get a bunch of Psych 101 undergrads, administer a survey, ask about how much they use alcohol and what their mood is, and just look and see, is there a relationship between those two variables," says Daniel J. Bauer of the University of North Carolina at Chapel Hill, the author of the article.

But a one-time survey of a bunch of college students can only get you so far. For example, it might find that sad people drink more, but it can't tell us whether people drink more at times when they are unhappy, whether the consequences of drinking instead result in a depressed mood, or whether the relationship between mood and alcohol use is stronger for some people than others.

One way psychologists have used to learn more about people is collecting data from people over a longer time period. For example, they might give each subject an electronic device to record blood pressure and stress several times a day, or ask them to log on to a website every night to answer a survey. In one case, Bauer's colleague, Andrea Hussong, asked adolescents to complete daily diaries with ratings of their mood and alcohol use over 21 days. The data showed that the relationship between mood and alcohol use is not the same for everyone. Adolescents with behavioral problems drink more in general, irrespective of mood, but only adolescents without behavioral problems drink more often when feeling depressed.

Analyzing this kind of data requires tougher math than the simple survey data, which is where quantitative psychologists like Bauer come in. "I think even though a lot of researchers are starting to collect this data, I don't think they've taken full advantage of it," he says. In the new paper, Bauer points to other methods that can do a better job of showing how variables relate differently for different people.

The point of all of this is to help people, Bauer says. For example, if psychologists discover that certain kinds of people are more likely to drink when depressed, it would be possible to help those people early. "Ultimately, the idea would be to identify people who might be more at risk and try to help them," he says.