Adults score low in science literacy.

Most adults in the United States keep abreast of scientific developments through news accounts. Yet only about 23 percent of those adults understand the nature of scientific inquiry well enough to make informed judgments about those science stories, finds a new survey performed for the National Science Foundation. Similarly, though most of the adults polled said they were interested in the environment and pollution, only about one in nine could offer even "a minimally correct scientific explanation" of environmental concepts such as global warming.

In the 12th issue of NSF's biennial Science & Engineering Indicators, published last week, Jon D. Miller and Linda Pifer of the Chicago Academy of Sciences unveil their updated survey on science literacy in the United States. The 1995 data show that about 70 percent of adults think science is valuable, continuing a trend more than 2 decades long (see p. 360). At the same time, Miller points out, understanding of particular concepts-or even of how science is performed-remains low.

For instance, he and Pifer found that fewer than 10 percent of adults can describe a molecule beyond noting that it's small. Only 20 percent can even minimally define DNA, and slightly fewer than half know that Earth rotates around the sun once a year.

Although almost one-quarter could explain correctly how chlorofluorocarbons were believed to contribute to the thinning of stratospheric ozone, only about half of these adults could describe reasonably well where in the atmosphere this thinning is taking place. Moreover, two-thirds were unable to explain the potential health risks of ozone thinning. Even fewer knew how acid rain forms (5 percent) and why it is of concern (2 percent), although most adults said they "oppose" acid rain.

Among the new study's most troubling findings, Miller believes, is a lack of understanding about how science is performed, as exemplified by answers to a pair of questions about a hypothetical experiment testing a new medicine to combat high blood pressure. Question one asked whether it would make better sense to administer the drug to 1,000 individuals or to give the drug to half of them and then compare that group's responses to the reactions of those who got no drug. The second question asked the respondents why they had made their particular choice.

One-third picked the first option, most reasoning that the more people tested, the more reliable the test's findings. Among those who selected the second option-the one considered correct by the scientists-fully 30 percent reasoned that experimenting on only 500 was better "because if the drug kills people, it only kills half as many." Clearly, Miller says, these people do not understand the role and value of a control group.

U.S. adults are no more illiterate about science than their peers in Japan, Canada, and the European Union, Miller says. His new data from 15 nations show that what distinguishes U.S. adults is that they "are the most optimistic about the outcomes of science and have the least reservations about science."