College scientific literacy courses make a big difference

Art Hobson

Physics teachers, especially physics education researchers, need to know about the science education research of Jon Miller, Professor of Interdisciplinary Studies and Director of the International Center for Scientific Literacy at Michigan State University. He is probably the world's leading expert on the measurement of scientific literacy. This is profoundly important work, because most national and global problems cannot be solved without a scientifically literate populace.

Miller and his colleagues have developed a set of basic science knowledge (concepts such as molecule, laser, DNA, biological evolution) and scientific process (an understanding that science is based on evidence and reason) questions used to study adults in many nations. He has used these questions, periodically updated to reflect new knowledge, in adult scientific literacy tests since 1988. A person scoring above 70 on these tests probably has sufficient knowledge to understand science-related stories in the daily newspapers, and is thus considered to be scientifically literate. By giving his test to a representative sample in each nation, Miller can determine the scientifically literate fraction of that nation's population, called the “scientific literacy rate” (SLR).

The bad news is that global scientific literacy is shockingly low. Among the 34 nations tested in 2005, the SLR rose above 30% in only one nation, Sweden, whose SLR was 35%.

For the United States, the good news is that in all of Miller's results since the beginning of testing in 1988, the U.S. scored above nearly all other nations. In the 2005 tests, for example, the U.S. ranked second with an SLR of 28%; next-ranked were Netherlands, Norway, Finland, and Denmark at 20 to 25%; then 15 European nations including Germany, France, and the United Kingdom scoring between 10 and 19%; and finally 13 other nations including Ireland and Japan at under 10%. In light of American students' mediocre showings in international science tests at the primary and secondary school levels, this is surprising. What happens to Americans after secondary school that accounts for this result?

To investigate this question, Miller asked each U.S. participant in the 2005 tests their age, gender, highest level of education, number of college science courses, number of children present in the household, their use of informal science learning resources (museums, magazines, etc.), and whether their adult occupation is science-related. He found that, over all these variables, the strongest predictor of adult scientific literacy was the number of college science courses taken; 75% of the variability in different people's scientific literacy scores could be predicted simply from this number. In assessing the effect of this variable, the number of college science courses was grouped into just three levels: (1) no courses, (2) one to three one-semester courses, and (3) four or more courses. People falling into the latter two groups were far more likely to be scientifically literate than those in the first group. Note that category (2) represents non-science students who are required to take a few science courses.

Thus the college experience is a strong determinant of scientific literacy in the U.S. The college experience is significantly different in nearly all other nations insofar as science education is concerned. Because other nations focus only on professional training at the college level, they don't require students outside of scientists and engineers to enroll in any science courses at all. Thus they have very few category (2) students. Miller concludes that “the college and university general education requirement to take at least a year of science courses makes a major contribution to the civic scientific literacy of [U.S.] citizens,” and that the surprisingly high U.S. SLR is a result of the positive impact of these college-level science courses for non-science students.*

So it seems likely that all nations could increase their SLR by requiring science courses for non-science college students. Looking only at Europe, we might expect (in light of the superior performance of most European nations relative to the U.S. at the secondary level) such a requirement to raise the median European SLR to at least the 28% U.S. rate. In Miller's 2005 tests, the median European SLR was only 14%. Thus it's plausible that a science literacy course requirement for European non-science college students could double the median SLR in Europe!

But regardless of the precise effects on any nation's SLR, Miller's results certainly underline the importance of college scientific literacy courses. The U.S. should more strongly emphasize these courses, and all other nations should teach them.

I'd be delighted to discuss these ideas with anybody who is interested. Email me at

Art Hobson is Professor Emeritus of Physics at the University of Arkansas in Fayetteville, and author of a scientific literacy textbook Physics: Concepts & Connections, now in its fourth edition. This letter is loosely based on the author's paper “The surprising effectiveness of college scientific literacy courses” appearing in The Physics Teacher, October, 2008.

* Jon D. Miller, "The impact of college science courses for non-science majors on adult scientific literacy," paper presented to a symposium titled "The critical role of college science courses for non-majors" at the annual meeting of the AAAS, 18 Feb 2007, San Francisco.

Disclaimer - The articles and opinion pieces found in this issue of the APS Forum on Education Newsletter are not peer refereed and represent solely the views of the authors and not necessarily the views of APS.