The Language of Science
By the turn of the century, experts predict that every science teacher in the country will have students in their classrooms who are not native English speakers.
Yet, when Elizabeth B. Bernhardt and her former colleagues at the National Center for Science Teaching and Learning looked for information on how science teachers could best teach those students, they came up practically empty-handed. Of the nearly 8,000 pages of science-teaching-methods textbooks they examined, only 16 pages specifically mentioned second-language learners. And 13 of those pages came from the same book.
Even more discouraging, some of the science teachers they interviewed told them they weren't interested in the subject.
"'Our job is to teach science,"' Bernhardt recalls the teachers saying. "'Send them when they're ready for science, and then everything will be OK."'
But to the researchers at the Columbus, Ohio-based center, that response was unacceptable. They knew from other studies that students need six to eight years to become completely fluent in a second language. And, English-as-a-second-language teachers, as hard as they try, cannot all be experts on photosynthesis.
"If we're sitting around waiting six to eight years, then kids are 14 or 16 at best before they can get into a science class," says Bernhardt, who now directs the language center at Stanford University. Figuring out how to solve that problem has been the focus of a two-year study she and her colleagues recently completed.
The project had a twofold mission. First, the researchers had to determine how science teachers could best address the needs of the second-language learners in their charge. Second, they had to get the message into the hands of those teachers.
In the process, however, the research team discovered something else: For students learning a new language, many of the methods now used to teach science may do more harm than good.
The National Center for Science Teaching and Learning is one of some 20-plus research centers the U.S. Department of Education now supports. According to Michael D. Aiello, the center's program manager, teaching science to culturally diverse students has been part of its mission fromthe start.
"Science knowledge is clearly the knowledge that is going to give kids the greatest access to the greatest number of opportunities in their lives," Bernhardt says. "If kids don't get turned on to science in the early grades, they're lost."
Moreover, the vocabulary of science presents special problems for second-language learners. Words like "force" and "mass" that have both a common meaning and a scientific one can be especially perplexing. Precise technical terms that aren't used in everyday life pose other challenges.
To carry out its research mission, the center hired Bernhardt, whose previous work at the Ontario Institute for Studies in Education in Canada had centered on language learning. The center directors reasoned that her perspective could offer new insights for science educators.
But, says Bernhardt, "it's not so easy to walk into somebody else's field and say, 'Let me give you my perspective."'
Besides combing the professional literature and interviewing science teachers, the researchers made long-term observations in two model classrooms: a Spanish-language immersion classroom in a magnet elementary school in Columbus, Ohio; and a high school program in Santa Cruz, Calif., for Spanish-speaking students considered at risk of failing.
They visited the classrooms over a period of months, videotaping lessons, analyzing the classroom dialogue, and testing the students. From those efforts, they drew several conclusions.
First, while most science teachers they interviewed seemed to think that the way to help struggling English speakers was to simplify their teaching, the teachers at these schools--Gladstone Elementary School in Ohio and Santa Cruz High School in California--succeeded by taking a different tack. They elaborated on their science talk, emphasizing the proper vocabulary and giving students more information, if needed, to help make sense of the content.
"When we use language that's chopped out or broken in some way," Bernhardt says, "we immediately deny kids exactly what they need, which is access to scientific vocabulary."
The teachers also made what the researchers called "on-line adjustments"--tailoring their lessons as they went along to plug in the gaps they saw in students' knowledge or to draw out what students knew but could not express.
The study team also concluded that students often knew more than they could express in an unfamiliar language. Allowing students to talk with one another in their native tongues helped capitalize on that existing knowledge base.
And almost all the students were better at writing about science than they were at speaking about it. "There's more time to think, write, and compose," Bernhardt explains. "Plus, affectively, it's a more private act."
The researchers' most controversial findings, however, are those that run counter to the current movement in science education to give students more hands-on experience.
Eight months into their nine-month observation of the 42 Gladstone 4th and 5th graders, investigators decided to give 20 of the students an independent assessment. They asked the youngsters to read four different text selections in Spanish, their second language. One was narrative, and the other three were expository texts about the scientific method, the classification of animals, and the planets. The students were asked to recall the texts in the language they felt most comfortable with.
The researchers also considered those children's scores in reading, vocabulary, and mathematics on a national standardized test. And, finally, during the last month of the experiment, they gave the children a performance assessment on the scientific method and the use of variables. Their task was to figure out how to determine the absorbency of a paper towel.
They found that, for the most part, students' scores on the Spanish reading tasks intercorrelated with the standardized-test measures. In other words, students who did well on the English vocabulary test also did well on the Spanish tests and vice versa.
But none of the measures--neither the Spanish-reading scores nor the standardized scores--could be linked to the performance assessment.
"What we think that tells us is that the performance assessment does indeed tap behaviors other than those generally tapped by more conventional measures," Bernhardt says. "At the same time, however, that indicates that the performance assessment doesn't tap literacy ability--the ability to glean information from text materials."
At Santa Cruz High School, where the students were learning English, the researchers made a similar observation. Two weeks after the 12 to 16 10th graders in that program gave science oral reports in English, many of them had trouble recalling what their presentations had been about.
"The kids were so concerned about their performance in English," Bernhardt says, "that they had apparently done a lot of memorizing to get through the report and do a good job."
What all this means, Bernhardt and her colleagues have concluded, is that the pendulum in the movement to reform science education may be swinging too far for students who struggle with a second language.
Even in exemplary second-language programs like the ones the center researchers followed, teaching and testing that focus on the "doing" of science can have a cost. Performance assessments can give the false impression that students know more--or less--than theydo. And they leave out scientificliteracy skills.
"Much of the scientific endeavor is discussion and experimentation, which is what performance assessment focuses on," Bernhardt says. "But another part of the scientific endeavor is being informed about science knowledge, and that is important, too."
Such findings, Bernhardt admits, are not popular.
"You'd have to be cautious about any recommendations that would say to reduce the amount of time that kids have experiences with scientific phenomena," says Kenneth G. Tobin, a professor of science education at Florida State University. "Clearly, if you do science, you're going to be involved in some equipment and problem-solving. The problem is, what is the teacher's role?"
Tobin questions whether the center's evidence is strong enough to support all of its conclusions. Still, he adds, "If students are doing it and not learning then, we have to say, 'OK, how can we develop the scientific discourse?"'
Despite the skeptical reception they have received, the researchers have tried to get their message out to science educators and reformers. They produced a teacher-friendly audiotape on their findings and mailed copies to principal investigators in all the state systemic initiative programs supported by the National Science Foundation.
"In some sense, this issue was not on the radar screen at all when we first started," says Aiello of the center. "Maybe now we're a blip on the edge."