Skip to main content
ABC News
Turning Dreams Into Data

By the time he died in 1985, Calvin S. Hall Jr. had collected about 50,000 dream reports. Hall had started off analyzing rats, but he switched to college students because he found their “willingness to answer questions” handy.

So, at Case Western Reserve University, Hall teamed up with psychologist Robert Van de Castle to analyze the dreams of 100 male and 100 female students. Between 1947 and 1950, the researchers collected five dreams per person. But how do you turn dreams into data? Hall and Van de Castle used an innovative quantitative-coding method that “divided dream content into settings, objects, characters, interactions, emotions, misfortunes, and several other categories.”

They discovered that dreams had negative emotions more frequently than positive ones.

Chalabi dreamy numbers

They also found that 1-in-3 dreams involved some form of misfortune, including falling (4 percent of dreams), death (9 percent) and the slightly more ambiguous “bodily misfortunes” (13 percent).

Sex cropped up in about 12 percent of male students’ dreams but only 4 percent of female students’ dreams. The content of those dreams also varied between the genders. For men, 93 percent of sex dreams involved them having sex with someone, and 54 percent of the time it was with someone unfamiliar. For women, they were only having sex themselves in 68 percent of those dreams, and 77 percent of the time, it was with someone they knew.

But maybe college used to be a particularly stressful and sexy time. And maybe the average American dream has changed since 1950.

Fast-forward to 2013 and the latest publication on dreams (PDF download) from the American Psychological Association. Surprisingly, the method to analyze dreams hasn’t changed much since Hall’s. This time, though, the researchers compared not only the dreams of men and women, but also the dreams of Iranians and Americans.

In Iran, as in the U.S., dreams were more likely to feature aggression, misfortune and failure than friendliness, good fortune and success. But cultural differences emerged, too. American men were four times more likely to have dreams with sexual content than their Iranian counterparts. In fact, Iranian men dream about sex about as often as Iranian women do.

Iranians see more dead and imaginary characters in their dreams. Iranians also tended to dream more about characters familiar to them, either family members or people they knew. The researchers looked at similar dream studies in India and Japan and concluded that this “seems very likely to reflect differences in the formation of the self between Eastern and Western societies.” They further explained this by saying, “Eastern cultures are thought to promote an interdependent and embedded self, while Western cultures are often said to encourage an independent and encapsulated self.”

It seems that bad dreams outnumber good ones (if you define nightmares as dreams with negative emotions and scenarios that involve misfortune or failure). And it also looks like that even when they’re unconscious, Americans and Iranians have plenty of differences.

One last thing (and it’s a question you’re probably wondering about): How exactly do you code a dream? Below is an example of a dream collected from a 12- to 13-year-old girl by the University of California, Santa Cruz. It was coded as featuring good fortune and friendliness, as well as an adult character whose occupational identification is known and taking place in a familiar setting.

The last dream I had was I got to go to the Mall and get anything I wanted. There was no one else there. I got four cars filled with clothes and two filled with shoes. I got so much stuff, I got other stuff too like lotion, makeup, Hello Kitty stuff and jewelry. Just when the owner was going to give me 9 million dollars, I woke up. I also got a pager cellular phone and CD’s.

Let us know whether you agree with the dream’s coding. You can also explore the codes interactively and look at other dreams on the site here.

Mona Chalabi is data editor at the Guardian US, and a columnist at New York Magazine. She was previously a lead news writer for FiveThirtyEight.