Gender Differences in Math: A Question of Speed?: In high school and college math classes, girls typically get better grades than boys. Yet on tests--especially those geared for top achievers--it's the males who often come out ahead.
The conundrum has puzzled researchers for at least a decade. Some say that boys dominate on tests because they have better spatial abilities. They can, in other words, visualize two- and three-dimensional objects in space and rotate them in their minds--a plus in solving geometry problems. Other theories hold that tests are gender-biased or that girls excel in the classroom because they behave better and write more neatly.
Now comes a group of Massachusetts researchers with another explanation: Boys score higher on tests because they are faster at basic, split-second calculations.
"A small difference of a quarter of a second between two groups over the course of a test that has many items and that has a lot of calculations involved can add up," said Loel N. Tronsky, a researcher from the University of Massachusetts at Amherst. "Someone who can do the arithmetic portions of the test faster will have more time to complete other portions of the test."
In a report published in the July issue of Contemporary Educational Psychology, Mr. Tronsky and his U-Mass colleagues--James M. Royer, Yan Chan, Stanley J. Jackson, and Horace Marchant III--describe nine studies conducted since 1993 with male and female students ranging from 1st grade to college.
Beginning around 4th grade, they found, boys begin to have a slight edge over girls in the speed with which they correctly answer basic addition, subtraction, multiplication, and division problems. The differences show up primarily among top achievers and older students. Among pupils who are all slow at basic mathematics, boys are no faster than girls.
And students who were faster calculators also tended to score higher on the kinds of math tests the investigators gave them. Those tests ranged from quizzes devised by the researchers to standardized basic-skills tests and college-entrance exams.
Besides affording more time on tests, the researchers reason, the split-second-calculation differential may free up mental capacity for deeper thinking about other kinds of math problems.
The researchers hypothesize that top-performing boys may be speedier at math calculations because they practice those skills more often outside school. More typically male pastimes such as computing batting averages and comparing statistical data on sports cards, for example, can hone such skills.
The findings may also have some implications for classroom instruction, the investigators conclude.
"One of the big pushes in math instruction--especially since 1989--has been a focus on a lot more conceptual-type material," Mr. Tronsky said. "I wholeheartedly agree with that. But I think the implication is that you can't let numerical ability fall by the wayside."
Achievement in Chicago: After taking into account factors such as poverty and high student mobility, a group of Illinois researchers has concluded that Chicago's public elementary schools, once labeled the worst in the nation, are outperforming their suburban counterparts.
Sociologist William Rau and his colleagues at Illinois State University analyzed six years of test-score data from the Illinois Goal Assessment Program, or IGAP, a set of reading, writing, math, science, and social studies tests given annually to students across the state in certain grades.
The researchers used sophisticated statistical procedures to predict how well students from Chicago public schools should be expected to do on the tests, given their high poverty rates and the frequency with which they move from school to school. Then the researchers subtracted the schools' actual scores from their predicted scores. The results, which the researchers called "residual" scores, were then compared with those for suburban Cook County schools and for the rest of the state.
On that measure, both Chicago and Cook County started out in 1993 well behind the rest of the state. But city elementary schools began making rapid gains in 1995, and by 1998 they had pulled ahead of their counterparts in the Cook County suburbs and across the state.
The findings are significant because the Chicago district has been engaging in successive waves of administrative change since 1988. Most recently, the state legislature in 1995 set up a mayorally appointed school board and a chief executive officer to oversee the ailing system and to intervene in schools that fail to shape up.
"Given these results, what do we make of the Chicago reforms?" the researchers write. "Our results ... provide fairly convincing evidence that most of the district's schools are moving in a positive direction." The full study is scheduled to be published in the November issue of the journal Sociological Quarterly.
Finding Dyslexia's Source: A group of Massachusetts researchers has added new fuel to the debate over whether dyslexia stems from visual or auditory problems in the brain's pathways.
The researchers from Beth Israel Deaconess Medical Center in Boston and the Harvard Medical School, in post-mortem examinations of human brains, found structural differences between the brains of people who suffered from the common reading disorder and of nondyslexics.
In their study, the results of which were published in July in the journal Annals of Neurology, the researchers found that nondyslexics had similar-sized brain cells in the visual areas on each side of their brains. But in the same areas in the dyslexics' brains, the cells were larger in the left hemisphere than they were on the right.
The new findings aren't likely to settle the long-running debate over the source of dyslexia, and likely won't have immediate ramifications for educators. But the search is important because finding the causes of the disorder could enable researchers to identify it earlier. ("Dealing With Dyslexia," Nov. 12, 1997.)
"Our goal is to really come up with a way for early diagnosis so these kids don't have to suffer through years of failure," said one of the authors of the study, Annette R. Jenner, who is now a postdoctoral fellow in the pediatrics department at Yale University's school of medicine.
Her co-investigators on the project were Glenn D. Rosen and Dr. Albert Galaburda, both from Harvard and Beth Israel. For the study, they examined the brains of five people of varying ages who had been diagnosed with dyslexia early in their lives and of five people who had never had reading difficulties.
Most of the previous anatomical studies of dyslexia, concentrating on the brain's auditory pathways, have also found structural differences. Either lobes or regions of the brain were asymmetrical in size or neuroimaging studies turned up variations in the pathways or the speed with which nerve signals traveled as the brain processes sounds.
The latest findings, however, are among a handful of studies--including a 1991 study by the same group--that have identified differences in the visual pathways as well.
But some researchers also caution against reading too much into such early findings. "Post-mortem information is suggestive but very difficult to interpret," said G. Reid Lyon, the chief of the child-development and -behavior branch at the National Institute of Child Health and Human Development in Bethesda, Md. "The brains you get might not be representative of the population at large that you're interested in."
Examining Dropout Prevention: What does it take to prevent students from dropping out of school? Clues may lie in a massive new evaluation of dropout-prevention programs funded by the federal government between 1991 and 1996.
Researchers from Mathematica Policy Research Inc., a Princeton, N.J.-based company, studied 21 of 85 programs that were part of the U.S. Department of Education's school dropout demonstration-assistance program. Now ended, the program is the largest and longest-lasting dropout-prevention effort underwritten by the federal government.
As part of the $7.3 million study, the researchers collected data on more than 10,000 students, visited high schools and middle schools over a period of two or three years, and talked with parents, teachers, students, and administrators.
They identified three promising dropout-prevention models: alternative middle and high schools for students at risk of dropping out, programs aimed at helping older students earn their General Educational Development credentials, and efforts aimed at restructuring teaching and learning in the entire school so that fewer students will drop out.
Among the less successful efforts, they found, were programs that were designed to build self-esteem or that offered extra help with homework.
Even the most promising programs, however, met with limited success. Restructuring programs, for example, appeared to have little lasting impact in many schools.
"The grant provided a buzz and an initial spurt of energy but, three to four years later, it was common to see a school settling back to the way it was before," said Mark Dynarski, an associate director of research at Mathematica. He wrote the study with senior researcher Philip Gleason.
Yet those kinds of whole-school efforts were often more effective than more targeted programs that sought to address the problem by monitoring attendance, for example, or setting up child-care centers for teenage parents.
A key to preventing dropouts, the researchers conclude, may be to first identify why students in a particular school community are leaving school and then tailor a prevention approach accordingly. And, across all the programs, the researchers say, the most successful efforts tended to be those that worked with students individually.
The full report, "How Can We Help? What We Have Learned From Evaluations of Federal Dropout-Prevention Programs," is also available on Mathematica's World Wide Web site at: www.mathematica-mpr.com/dod-syn.pdf (requires Adobe's Acrobat Reader).
Interesting ideas? Send suggestions for possible research stories to Debra Viadero at Education Week, 6935 Arlington Road, Suite 100, Bethesda, MD 20814; e-mail: firstname.lastname@example.org
Research coverage is underwritten in part by a grant from the Spencer Foundation.
Vol. 19, Issue 2, Page 10