While the No Child Left Behind Act has a detailed formula for bringing students to proficiency on state reading and mathematics tests by the 2013-14 school year, it's much less precise on states' goals for English-language learners.

Under the law, states for the first time must set "annual measurable achievement objectives"—or AMAOs—for how English-language learners are progressing toward learning English. States must also show that they are meeting those goals.

The goals are required in Title III of the No Child Left Behind law, the 2001 revision of the Elementary and Secondary Education Act. Title III supports local programs such as bilingual education or English- as-a-second-language classes that are supposed to help children with limited English proficiency reach the point at which they can comfortably handle classroom work in English. The federal government is expected to spend between $665 million and $686 million on Title III in fiscal 2004.

States that fail to meet the new goals could eventually lose some of their Title III funds.

U.S. Department of Education officials acknowledge that some of the goals states have set are weak, but for the time being, the officials say, they're not rejecting any of the goals because of a lack of rigor.

"The No Child Left Behind Act is political hyperbole," said Christine H. Rossell, a political science professor at Boston University who studies English-language policy. "It's like Lyndon Johnson saying, 'I'm going to alleviate poverty.' Obviously you can't alleviate poverty in a free society, but it sounds good."

Federal officials defend their efforts. A separate provision in the No Child Left Behind law that requires states to include English-language learners in their reading and mathematics tests, the officials say, will push states to help such students learn English.

That provision is in Title I of the law, and it requires that all English-language learners take state standardized academic tests, in English, after they have been in the United States for three years. That period can be extended to five years on a case-by- case basis.

Because many states don't provide tests in languages other than English, most English-language learners must take such tests in English as soon as they enroll in U.S. schools. Previously, many states didn't include English-language learners in statewide testing until they had attended U.S. schools for three years.

Now, states can lose part of their Title I funding, which supports programs for disadvantaged students, if they don't show that English-language learners are making enough progress on state tests, or if they fail to test at least 95 percent of such students.

"The expectation is that the students will become proficient in the English language over a three-to-five-year period of time," said Kathleen Leos, the associate deputy undersecretary for the office of English-language acquisition of the Department of Education.

Gentle Goals

That may be the official expectation, but for many Title III state plans, the objectives and definitions of English proficiency are all over the map.

Michigan's plan seems to be among the most ambitious, promising to bring 95 percent of students who are now at the most basic level of learning English to full proficiency in four years.

Mazin A. Heiderson, an education consultant for the Michigan education department, said the goals were set according to "guesswork on what is known plus what is politically acceptable."

"Our definition of English-language proficiency is that you can make it in all English classes without English-language support," he explained. Such a student would be expected to get about a C average in mainstream classes, he said. "It's not reaching your optimal potential."

On the other end of the spectrum is Minnesota, which divides its English-language learners into three groups, depending on how long they have been in special programs. For those who have studied English for less than three years, the state plans to move 2.5 percent to full proficiency this school year. The goal is only slightly higher for students who have studied from three to five years, or six or more years.

By 2013, Minnesota says, it will have raised the percentage of students who have been in programs for six or more years and who are deemed fully proficient from 3.8 percent to 12 percent.

"That's completely too easy," said Ms. Rossell, who helped lead the successful 2002 ballot-initiative campaign in Massachusetts to curb bilingual education. She argues that the goal should be at least 50 percent for English-language learners who have been in programs for six or more years.

California, which has 1.6 million English- language learners, far more than any other state, falls somewhere in the middle range of states' goals. California says that this school year it will move 30 percent of its students who have been in programs for four or more years (and two other small categories of English-language learners) to full proficiency.

California promises to raise that rate to 46 percent in 10 years.

Jan E. Mayer, the manager of the language-policy-leadership office for the California Department of Education, acknowledged that the state's goals don't account for every English-language learner, and thus don't ensure that California will leave no child behind.

But she noted that the new Title III requirements to report scores of all English-language learners to the federal government will bring more attention to whether each child is learning English.

Like education officials in many other states, the California and Minnesota officials say they set targets based on how well school districts are doing now—and tried to set the bar so that districts would be pushed a bit to improve.

'A First Step'

No one, it seems, wants to come down hard on states for setting low goals for students who are learning English.

"I only give them a large benefit of the doubt because the federal department was pretty slow in issuing the regulations [for Title III]," said Don Soifer, the executive vice president of the Lexington Institute, a conservative think tank based in Arlington, Va. "The initial burden was clearly on the states to design the program first based on what they had in place to begin with."

"This has to be thought of as a first step," added Charlene Rivera, the executive director of George Washington University's Center for Equity and Excellence, and an expert on English-language learners. Many states are grappling for the first time with definitions of English proficiency, she said.

Almost all states said in their Title III plans that they had created or were in the process of drafting standards for English- language development. Many are also commissioning new tests that measure English progress to comply with the No Child Left Behind law. (See chart, this page.)

Ms. Rivera said the new tests likely would focus more on academic English than tests now on the market do. When the tests are in place, the states "will have to recalibrate everything—and redefine their vision of English proficiency," she noted.

Going Easy

Maria Hernandez Ferrier, the director of the office of English-language acquisition and a deputy undersecretary for the federal Department of Education, and Ms. Leos said they were going easy on the states' progress goals until the states have implemented their new accountability systems for English-language learners.

Ms. Leos said the Education Department has approved the Title III plans for 11 states and the District of Columbia, and has told the rest of the states to fix certain aspects of their plans for approval.

For instance, the department has asked a number of states to provide more detail in their test data for English-language learners, she said. If states were asked to fix their AMAOs, it was because they missed including all English-language learners in their projections, or for reasons other than the actual difficulty of the goals.

Ms. Leos said that the states would have to fix any problems with their baseline test data by Nov. 17, and that by April 30, states were expected to turn in standards for English-language proficiency and show how those standards were linked to academic-content standards.

Ms. Leos initially said the letters that the department had sent to the states concerning Title III could be found on the department's Web site. As of late last week, the letters were not on the site, nor had the department released the letters to Education Week. The newspaper obtained each state's Title III plan by filing a request under the federal Freedom of Information Act.

The states that have full approval of Title III sections are: Colorado, Hawaii, Indiana, Louisiana, Michigan, Nevada, New Hampshire, New Mexico, Utah, Vermont, and West Virginia, as well as the District of Columbia.

Several states said in their Title III plans that research shows it takes five to seven years for students to learn English.

Ms. Rivera, though, said such a time frame was reasonable only if one was talking about helping students learn academic English, rather than a command of social, or conversational, English.

Many states have submitted plans that define English proficiency as a score on an English-assessment test, but don't incorporate how students perform on academic tests. So the definition of English proficiency in those states will depend on where the bar is set in the new English-language assessments they are developing.