UT Austin researchers sort bilingual language differences from disorders

Feb. 28, 2001

AUSTIN, Texas—A language difference is not a language disorder. Yet every year, some bilingual children in Texas wind up being funneled into classrooms designed for children with learning disabilities.

Dr. Elizabeth Peña, an assistant professor at The University of Texas at Austin¤s department of communication sciences and disorders, is working to develop better ways to assess language skills in bilingual children to prevent misdiagnosis — and to make sure that those bilingual children with genuine learning problems get the help they need.

"If you look at the expected incidence of language impairment, minority children generally are over identified or under identified," Peña said. "Not speaking English well can be interpreted as not having the ability to learn language well. If these kids end up getting placed in special education, expectations tend to go down. They are not getting an education that is consistent with their real abilities."

There's also a serious cost to a child with a genuine learning problem that does not get diagnosed early enough. "If children do not get identified, we have lost some critical time where they could have been getting services. By the third or fourth grade, they might be so far behind they don¤t have a chance," Peña said.

"We are trying to come up with a measure that is better at classifying bilingual children with language problems and those without language problems," she explained.

Peña and Dr. Lisa Bedore, assistant professor of communication sciences and disorders at UT Austin, are collaborating with researchers in California and Pennsylvania to develop a language test for bilingual children 4, 5 and 6 years of age that will measure four important aspects of the child's use of language. These include vocabulary, grammar, the sounds of words and the way children communicate in interactions with others. The research is being funded through a six-year contract from the National Institutes of Health.

In addition, Peña; Dr. Lynda Miller, president of SMART Alternatives Inc.; and Dr. Ronald B. Gillam, an associate professor in UT Austin's department of communication sciences and disorders, have developed a new diagnostic tool called "Dynamic Assessment and Intervention: Improving children's Narrative Abilities." It has been published by Austin-based Pro-Ed.

About 7 percent of all school-age children throughout the United States have unusual difficulty learning and using language, and they struggle with reading, writing and math. Researchers say these children frequently misunderstand what is said to them. They have very limited vocabularies; use short, grammatically incorrect sentences; frequently ask for information to be repeated; and appear reluctant to join conversations.

An educator unfamiliar with Spanish may easily be misled by a bilingual child's struggles with English. But, in general, the mistakes made by a normal child will be consistent, related to the differences in the two languages. For example, adjectives such as "big" come before nouns like "river" in English. But the reverse — "Rio Grande" — is true in Spanish.

Nailing down these consistencies in language use is one of the goals of the researchers, so speech language pathologists who diagnose language learning problems will know for sure what is appropriate for the native language and the new language.

Educators usually determine whether pre-school and elementary students have language learning impairments using traditional tests that check whether a sufficient number of correct answers are given. The tests are seldom in both languages. And they seldom give a complete picture of the child's abilities. Researchers say it's crucial to test bilingual children in both languages because, for example, Spanish-speaking children may have learned the words for colors or numbers in English watching Sesame Street. But they use Spanish words for family members.

The dynamic assessment test developed by Peña's group has sections in both English and Spanish, and must be administered by someone familiar with both languages.

"We give the test to find out where the children are. Then we teach them strategies for performing better on the 'post' test," Peña said. When children are re-tested after the strategy sessions, those who are developing normally pick up the strategies quickly and are able to score well.

"Children who have true language disorders still can't do it," Peña explained.

The researchers did not teach test materials themselves to the children. Instead, the researchers taught the children how to take the test in two or three strategy sessions lasting about 20 minutes each. The children were much more successful on the follow-up test.

Most speech pathologists take about an hour to do an assessment. Peña said dynamic assessment "certainly takes a lot more time — three to five hours. But this could save you a year of providing a service that wasn¤t necessary in the first place.

For more information, contact: Dr. Elizabeth Peña at UT Austin department of communication sciences and disorders at (512) 471-2690, Dr. Ron Gillam at (512) 471-1936 or Dr. Lisa Bedore at (512) 232-5101. Or see the Web site at http://csd.utexas.edu/

For photos, see www.utexas.edu/admin/opa/news/01newsreleases/nr_200102/pena2.html