IN 1985 I BEGAN COLLECTING BOTH my own data and other people's studies regarding the prevalence of pseudoscientific beliefs in both small samples, and in the public at large. At the time many found the results, which indicated these beliefs were widespread and ubiquitous, to be shocking.
I found myself recently wondering how much progress we have made (if any). And to what factors do we owe our thanks if improvements be found? This curiosity prompted me to organize a research panel at the American Association for the Advancement of Science (AAAS) annual meeting in San Francisco in February, 2007. Overall, there was some good news, some bad news, and a large number of unsuspected findings. What follows is a summary of the most interesting outcomes.
I should note that in 1985 the AAAS established Project 2061 with the intent of helping all Americans to become literate in science. Project 2061 set out recommendations for what all students should know--and be able to do--in science by the time they graduate from high school. The publication of Science for all Americans and Benchmarks for Science Literacy by Project 2061 is the foundation for ongoing efforts to reform curriculum, instruction, and assessment. As a result of Project 2061, in recent decades U.S. secondary schools have greatly changed how they teach science. For example, there is now less emphasis on factual memorization, and instead there is more emphasis on understanding science inquiry (including more hands-on experience and more understanding about the context of science and technology).
Dr. Jon D. Miller, of Michigan State University and one of the panelists, sums up the problem of methodological problems this way: "The core of the problem is that each individual's level of scientific literacy is a result of a combination of formal science classes, work experiences, life experiences, media consumption, and overt information seeking. The strong influence of college-level science courses on adult scientific literacy has been reported in several earlier cross-sectional analyses. However, during the same period of time, the volume of formal science learning opportunities increased significantly. And during the same period the number of science related issues in the public policy agenda has increased in both number and importance." So, the difficulty in assessing any progress seen in science literacy becomes one of disentangling the effects of all these potential causes.
Further, I wondered, do improvements in science literacy lead to reductions in pseudoscientific beliefs? Fortuitously, enough data have now been collected over several decades to allow for some preliminary cohort analyses that might enable us to answer such questions. This therefore became the theme of the panel and basis of invitations to participate.
The first presenter (and co-chairperson) was Dr. Susan Losh, Associate Professor of Education, at Florida State University. Dr. Losh's analyses relied heavily on the U.S. National Science Foundation Surveys of Public Understanding of Science and Technology 1979-2001 (total sample approximately 22,000). Losh used these data to construct a number of synthetic cohorts, and then examine differences in civic science literacy and adherence to pseudoscientific beliefs across these cohorts. She argued that if more recent science teaching in American high schools has indeed contributed to adult science sophistication, then after holding age constant the rejection of pseudoscience should be higher among more recent than among the earlier cohorts.
To assess civic science literacy, Losh utilized an Oxford items index (which measures knowledge of very simple matters of science--for example, "electrons are smaller than atoms"--as a method of assessing civic scientific literacy). Losh subsequently noted that American exposure to high school math and science courses has increased in more recent groups. …