The language gap

Today, Ali and Rothstein discuss ways to educate students who aren’t fluent in English. Previously, they debated devoting resources to closing the achievement gap, the No Child Left Behind Act and reasons for lagging minority performance. Tomorrow, they’ll discuss policies to boost students’ achievement.

If it works for all students, it can work for English learners
By Russlynn Ali

Richard,

Since the last census, the Los Angeles Unified School District has reclassified as English proficient 13% of its formerly English language-learning students. In truth, that number compares favorably with a 9% reclassification rate for California public schools as a whole.

That said, nobody should be happy with reclassification rates that low or with the huge achievement gaps between L.A. Unified’s English learners and their English-speaking counterparts in reading and math. On the 2007 fourth-grade California Standards Tests, L.A. Unified’s English learners were 41 points behind their English-speaking peers in English language arts and 30 points behind in math.

Mind you, L.A. Unified isn’t doing very well for its African American students either — only 38% of its African American students are proficient or advanced in fourth-grade math, compared with 34% of fourth-grade English-learner students. White students, too, don’t perform as well as their counterparts in other urban districts. Yes, the district is improving — faster than the state as a whole. But clearly, L.A. Unified still needs to do far better by all groups of students.

All of my experience tells me that these poor results are not occurring because L.A. Unified’s teachers don’t care about their students, including their English language learners. Most care a lot. But California public schools educate almost one-third of all English-learning students in the U.S., and we’ve devoted far too few resources — as a state and as a country — to studying and measuring what works, then helping teachers do more of it.

Teachers of English-learning students are hungry for tools that will help them succeed. But it’s not even clear how long such students who enter school at different ages typically should take to become proficient in English. And evaluations on the effectiveness of alternative approaches are spotty at best.

Last year, however, California took an important step forward by releasing school-level Academic Performance Index scores for English learners. With these new data, we can start identifying the schools throughout the state that are doing a better job of educating English-learning students to high levels, and we can learn more about their strategies for success. Despite the challenges such students bring with them, success is possible.

Emerging analyses (PDF) provide some helpful hints (PDF). It turns out that schools that work well for English-learning students look quite similar to schools that work well for other groups. They have strong teachers who are well equipped to teach English-learning students, and they have high expectations for their students to master academic English and work hard not just to meet but to exceed state and federal accountability goals for student achievement.

Schools that do best with English learners consistently measure student performance, not just on assessments designed to measure mastery of English but in core academic subjects. Principals and teachers use that data to drive instructional change, develop strategies to help their students and themselves succeed, and better inform and engage parents. These higher-performing schools provide their English learners with strong supports, including extra instructional time and quick interventions for students who start to slip.

Schools that work for English language learners deliver strong, standards-aligned curricula. They include a framework for building academic vocabulary, including — as wonderful work by Phil Daro and Uri Treisman has shown — specific advice on the words in each content domain that English learners typically don’t know. These schools give their teachers the support and professional development they need. Most important, they don’t see English learners as impossible to teach.

This last point is especially important. Many people believe that most English-learning students are new immigrants who “can’t possibly be expected to speak the language and catch up quickly.” This is a destructive stereotype, and it is wrong. The majority of students enrolled in our state’s English-learner programs are not immigrants, let alone new immigrants.

More than 1.5 million students in California K-12 school system are classified as English learners. Yet according to the Migration Policy Institute, only about 637,000 children in California between the ages of 5 and 17 are foreign-born. That means that more than half of the students enrolled in English-learning programs were born and raised in the U.S. This isn’t about whether we can educate new immigrants to higher levels; this is about whether we can educate all students, new immigrants included, to higher levels.

Even when English learners are in our schools for a long time, we don’t teach them what they need to know. Indeed, in L.A. Unified’s graduating class of 2006, two-thirds of English learners who didn’t pass the California High School Exit Exam had been languishing in English-learning programs for more than 10 years.

We can do better. For our state’s future, we must do better.

Russlynn Ali is the executive director of The Education Trust-West, an Oakland-based think tank focused on closing the achievement gaps separating low-income students and students of color from other young Californians.

The politics of reclassification
By Richard Rothstein

Russlynn, you properly state that schools doing better with English-learning students have invested in better curriculum, teacher training, extra instructional time and quick interventions for students who slip.

I also agree when you say, “It’s not even clear how long such students who enter school at different ages typically should take to become proficient in English.” But then how can you conclude that a 13% reclassification rate is too low? How do the age distributions of immigrant children in Los Angeles compare to those in the rest of the state or the nation? Wouldn’t time needed for fluency also depend on how much English reinforcement immigrant children get at home and from peers, how much English is spoken in their communities and how much access they have to interesting English reading materials outside school? We should answer these questions before jumping to conclusions that reclassification rates are too low. They could be too high when schools are pressured to reclassify children before they are ready to study in English without additional support.

California districts have different reclassification criteria. Some use score cut-offs on California’s standardized test. Others do not. Earlier this week, you agreed that using cut-off points on standardized tests can be “statistical trickery and gaming.” Does Los Angeles’ higher (than statewide) reclassification rate stem from better instruction or lower criteria? We should know this before jumping to conclusions about where reclassification rates are better.

Reclassification is, itself, an arbitrary standard. If children get the support they need after reclassification, higher reclassification rates should be expected. But if they don’t, reclassification can be disastrous. Nationwide, reclassification rates differ widely, partly because of different instructional quality, but also partly because of different arbitrary reclassification criteria.

You express shock that more than half of the students enrolled in English-learning programs were born and raised in the U.S. This statistic seems shocking, but the reality is that many immigrants (especially from Mexico and Latin America) come to California to work as young adults, with children born soon after they arrive. These children, born and raised in homes of recent immigrants who don’t speak English and in communities made up mostly other recent immigrants, are indistinguishable for educational purposes from immigrant children.

To know how well our schools assimilate Latino children, we need better data. Schools should record mothers’ country of birth and educational attainment. We’d then likely find that few, if any, children born to non-immigrant mothers were enrolled in English-learning programs. How well these children do in school is probably comparable to how well non-Latino children do — those who come from families with higher literacy levels perform, on average, better.

There’s a widespread myth that schools did better with earlier immigrant generations. Were it true that schools were once more successful, panicked reactions about today’s English-learning students might be more understandable. But in the first decades of the 20th century, there were enormous achievement gaps between immigrants and the native-born. Italians, for example, like many Mexicans today, came here for unskilled work, with little literacy in their own language. In big cities 100 years ago, 80% of native white children, but only 58% of Italian children, stayed in school another year. As late as 1931, only 11% of Italian students who entered high school graduated, compared to more than 40% for all students. This was a much bigger native-immigrant gap than we have today.

Instruction in students’ home languages was once commonplace, abolished not because it wasn’t working but because of public hostility to immigrants after we went to war with Germany in 1917. Have you ever wondered why we use a German word, “kindergarten,” to describe early education? American kindergartens were first established in the 1870s and taught solely in German to give immigrants a head start in school. English was introduced gradually in later grades.

Public schools teaching only in German were everywhere. In Texas, there were even Czech-language public schools.

San Francisco established segregated public schools taught in Chinese, Indian, Mongolian and Japanese. German, Italian and French immigrants, on the other hand, were taught in their native languages in regular public schools. But an “anti-immigrant” school board majority in 1873 fired all French- and German-speaking teachers. After immigrant protests, bilingual instruction was reestablished in 1874. In 1877, the California Legislature enacted a prohibition on bilingual education, but the governor refused to sign it.

I’ve reviewed this history to show that fights about when to reclassify students have been historically more about politics than pedagogy. Russlynn, you are right about the kinds of quality instruction children need, but until we do better at answering your question about how long it should take immigrants of different ages and family literacy levels to become English-fluent, politics will continue to trump pedagogy.

Richard Rothstein is a research associate of the Economic Policy Institute in Washington and author of “Class and Schools: Using Social, Economic, and Educational Reform to Close the Black-White Achievement Gap” He was formerly the national education columnist for the New York Times.