College size does matter

The number of students in a college class does have an impact on student performance, researchers have found. The effect is most noticeable for class sizes under 10-20 students, with the data showing that the best class size of all is one. Each student added up to a total of 20 students has a significant effect on student grades, but as the numbers increase over that, the effect of the class size variable gradually becomes less pronounced.

Thus, the probability of a student getting a B+ or better in a class with fewer than 20 students is approximately twice as great as in a class with 120 students, but only slightly less in a class with 400 students.

The debate over the effect of class size on performance in K-12 classrooms has raged on for years. Prior to the 1970s, class size was not thought to matter. During the 1990s, the STAR study in Tennessee showed an irrefutable link between small class sizes and student achievement, especially at the lowest elementary school grades. This led a small number of states to allocate money toward reducing class size. Most notably, Florida voters passed an initiative to do so last November.

In higher education, the effect of class size on learning has been harder to pin down. This is mainly because testing to measure student outcomes "is ubiquitous" in the K-12 world, say the three researchers from Binghamton University of the State University of New York who conducted the study. In higher education, "we lack control groups and an agreed upon metric."

Nevertheless, the researchers--E.C. Kokkelenberg, a professor of economics; Michael Dillon, a senior analyst in the Office of Institutional Research; and Sean M. Christy, in the Department of Economics--collected data on 395,000 students from a medium-sized, highly selective public research university. Interestingly, they used an economic model, called wage theory, as their framework for analyzing "the nature of grades from a student's perspective."

This is "one of the innovative portions of the study," says Dillon, who, along with Kokkelenberg, is a member of the United University Professions/AFT. "We are trying to link a traditional economic model to student learning."

Taking into account many variables, such as gender, educational and socioeconomic background, and even faculty attitudes on teaching small and large classes, the researchers came to these findings:

• Small classes are more beneficial to at-risk students than to those not at risk.

• Students who have received Advanced Placement credit are less affected by increasing class size.

• Females do better than males in tolerating increasing class size numbers.

• In general, the impact is greatest for classes with fewer than 40 students. Above that number, the consideration for institutions would be the cost of retention and reputation issues, not so much appreciable differences in individual student performance.

"The study does not show why grades are better or how much students learned," comments Kokkelenberg. "The message is, given this data set, we have strong evidence that class size influences grades. It's not a nice thing to say to university presidents trying to increase class sizes" to save money.

The survey is called, "The Effects of Class Size on Student Achievement in Higher Education: Applying an Earnings Function." The survey can be found online at www.ilr.cornell.edu/cheri/wp/cheri_ wp28.pdf.

American students don't know their geography

A recent survey has left a lot of educators wondering about the sorry state of geographic literacy among young adults in the United States. What in the world is going on?

National Geographic's 2002 Global Geographic Literacy Survey, which examines what students in nine industrialized nations know about geography, showed that Americans ages 18-24 outscored only their counterparts in Mexico when it came to demonstrating knowledge in this key discipline. The results mirror a similar survey conducted by National Geographic in 1988, which placed young Americans dead last in a nine-nation study. The 2002 survey, which was released late last year, provides many eye-opening examples of yawning gaps in Americans' basic knowledge of geography. Consider:

• Almost 30 percent of Americans surveyed could not find the Pacific Ocean.

• Roughly 83 percent could not find Afghanistan on a map.

• More young Americans knew that the island featured in last spring's "Survivor" TV show was located in the South Pacific than could find Israel on a map.

• Americans ranked near the bottom of students surveyed when it came to locating the United States.

• Nearly 30 percent of Americans surveyed believed the United States was home to a billion or more people.

• Americans were the least likely group to know that Afghanistan is where the Taliban and al Qaeda were based.

"If our young people can't find places on a map and lack awareness of current events, how can they understand the world's cultural, economic and natural resource issues?" asked National Geographic Society president John Fahey.

National Geographic is responding to the dismal results by convening an international coalition of education, business and policy leaders to develop strategies to fight geographic ignorance. The American Federation of Teachers will be part of this coalition, along with such groups as the American Society of Newspaper Editors, the Markle Foundation and the National Education Association.

The survey included young adults from Canada, France, Germany, Great Britain, Italy, Japan, Mexico and Sweden. The top scorers on the survey of 3,250 participants were from Sweden, Germany and Italy. Among the factors that influenced performance were educational experience, including taking a course in geography. Other factors were international travel, language skills, varied use of different news sources and Internet use (the survey found that participants who had used the Internet within the past 30 days scored 65 percent higher than those who hadn't).

An online version of the survey is available at www.nationalgeographic .com/geosurvey; visitors can compare their answers to average correct scores posted by young adults in each nation surveyed.