Survey Method Unveils More Risky Behavior
Teenage boys are using illegal drugs, engaging in risky sexual behaviors, and committing violent acts at significantly greater rates than previous studies have estimated, says a report published last week in the journal Science. The researchers attribute the differences to a new survey method they say is more reliable than older methods.
Washington-based researchers from the Urban Institute and the Research Triangle Institute jointly administered questionnaires to 1,600 15- to 19-year-old men in 1995. They used a new computerized technique designed to improve the accuracy of studies on sensitive topics on 80 percent of the men. The remaining 20 percent filled out the self-administered questionnaires using the traditional pencil-and-paper method.
To complete the computerized questionnaires, the youths listened to questions over headphones and answered by pressing keys on a computer keyboard. The technology is called audio-computer-assisted self-interviewing. Both sets of questionnaires, which were identical in content, were administered at the youths' homes or in other locations, such as a neighborhood restaurant.
The study found that 5.5 percent of the youths who answered the computerized questionnaire reported having engaged in homosexual activity, compared with just 1.5 percent of those using the pencil-and-paper version.
Similarly, 6 percent of the group answering the computerized inquiry reported using crack cocaine in the past year, compared with 3.3 percent of the other group; 12 percent of the computerized-testing group said they had carried a gun in the past month, compared with 8 percent of the youths who filled out the paper test. Reports of homosexual sex or intravenous drug use--behaviors associated with risk for HIV infection--were at least two to three times higher among those answering the computerized survey.
While the two groups reported marked differences on the more controversial subjects, their responses were very similar when asked about behavior that was less stigmatized, the study found.
The computerized method may be better at teasing out accurate answers to sensitive questions because of a perception of privacy, the report says. Because the respondents store answers in a computer drive, they may feel they are less likely to be read by others than a paper version, said Laura Lindberg, a research associate at the Urban Institute and an author of the study.
The widely used paper method, while more private than a face-to-face interview, also may pose challenges to youths who are less literate, the article says.
The study suggests that illegal drug use and violence among youths may be more widespread than previously believed, prevention experts said last week.
"If this is correct, this could mean that all the statistics we are basing our [national] policies on are underestimates, and we have more of a problem than we think we do," said Rosalind Brannigan, the vice president of Drug Strategies, a policy and research group in Washington.
Vol. 17, Issue 35, Page 7