« Adapting Law Lessons for ELLs | Main | ELL Advocate Named to Aspen Commission on NCLB »

Letter to Arne Duncan: 'Race to the Top' Is Unfair to Teachers of ELLs


Schools and the Stimulus

James Crawford, a longtime writer about English-language learners and president of the Institute for Language and Education Policy, has sent a letter to U.S. Secretary of Education Arne Duncan contending that proposed priorities for Race to the Top are a bad idea for teachers of English-language learners.

Crawford has always been a critic of the emphasis on standardized testing in the No Child Left Behind Act, and he tells Duncan that his organization "is especially concerned by your insistence that standardized test scores be used as part of teacher-compensation systems." It's particularly a bad idea, he argues, that teachers of English-language learners be evaluated based on their students' test scores.

"The grant criteria would also place an undue reliance on standardized tests that offer, at best, a blurry snapshot of student progress," Crawford writes. "For English-language learners (ELLs) in particular, such tests are rarely valid or reliable."

In addition, Crawford says, "If teachers are to be penalized for an 'achievement gap' over which they have no control, how many will want to teach ELLs?"

Crawford's position that teachers of ELLs would be harmed if they were to be judged by their students' test scores on states' academic content tests was also expressed this week by Duane Campbell of Choosing Democracy blog, which I blogged about.

Two teachers of ELLs posted comments to that blog item. One said she was OK in being evaluated by how her students score on tests, as long as her ELLs could be tested in the language of their strength, which might be their native language.

Another said her school system's program for ELLs isn't set up in a way that allows her to teach them effectively. She writes: "I am not comfortable with my students' test scores being tied to my 'effectiveness' and/or pay scale. If I was allowed to teach in the manner in which I was trained to do, and based upon research-proven best practices, I would feel differently."

The National Education Association has also written to Duncan in opposition to the measure. See my colleague Stephen Sawchuk's blog entry and story about the union's position.


This argument assumes (perhaps rightly) that test scores alone with be used. But if demographics were included along with regression models, then there's no reason why teachers of ELLs couldn't be evaluated fairly. A recent study of Ontario public schools is an example of an approach which attempts to factor out factors such as average income, first language, etc. http://www.cdhowe.org/pdf/ebrief_85.pdf

All teachers of ELs are in essence English teachers, despite what content or grade we teach. It is our job as teachers to empower students to make gains in reading and writing in English. That said, tests can still be useful in analyzing which teachers were able to most effectively clear the dual hurdles of content and language. Yes, it is an added challenge, but that doesn't imply its unfeasible to reach for this goal.

Its also rather defeatist for any teacher to claim that achievement gaps somehow cannot be closed by them. If not us, then who? There's enough research out there to suggest that good teaching will help close these gaps, so let's not shortchange ourselves and our kids.

We should test for the activity level of the student, the amount of time and type of activity the learner engages in with the language, outside of the classroom.

Is there still talk of using for a growth-based model for student achievement through which a student’s academic growth is measured against a baseline obtained from the previous year or from an initial assessment? This seems to be the only way to show accountability for student achievement of those students who are well-below grade level (which is characteristic of many ELLs). Basing student achievement squarely on grade level assessments of the grade the student is currently in does little to improve the instruction, programming, and policies for these students. Most districts, schools, and classrooms employ formative and interim measures that can provide diagnostic information for teaching and also show annual academic gains. Why not make annual standardized assessments more aligned with district and classrooms measures that seek to identify current levels and the needs/strengths of students rather than only measuring students against current grade standards? Such an approach would also allow students above grade level to demonstrate their current skills and needs.

I can see the down side of growth based models in terms of management. 21st century learning may require 21st century assessments such as well-developed ,computer-based, adaptive assessments that lead students through different grade levels of assessment. Growth-based assessment may also have obscure annual objectives like “Students in 8th grade who start the year at a third grade in math level must make a 2.5 grade level gain to make AYP ” Still, I would take this type of criteria over having students who are progressing in literacy and math at a 2nd or 3rd grade level, have to take an 8th grade test to measure their academic growth. Objectives and goals are important, but they have to be in reach or even take slightly beyond reach.

I don't think academic growth for ELLs can be tied only to teachers even within a growth-based model. Education is a collective process and all participants from the federal policy makers to the classroom teachers should be evaluated in their roles in the process. I may be a better teacher and my students would be more motivated by more tenable contents goals for accountability purposes. It seems the national standards movement might improve the consistency for implementing a growth based models by having a universal criteria on which to base instruction and assessment.

I agree also with the other comments that students should be able to show growth in their stronger language or other languages of instruction (i.e. bilingual programs)and not only English. A 7th student receivig bilingual instruction may show growth from a 6th to an 8th grade level in Math when tested in Spanish, but only shows a 4th grade math score when tested in English. Both assessments provide both accountability and diagnostic information for math and academic language instruction.

Making annual ELL assessments of academic language proficiency a larger thrust in the accountability system for ELLs than it currently occupies should not be promoted. There is too much variability among these standards and assessments across states and they reflect a great variety in the interpretation of academic language proficiency. States created ELP standards and assessments at the advent of NCLB, but most have done little to change and improve them even as research around teaching and assessing academic language has exploded over the past 8 years. This is not to lay blame on states as creating an ELP Standards and assessment system was a huge undertaking. It was great that consortias such as WIDA took shape to assist states in this endeavor.
The federal government is still often unclear about the scope, direction, and role of ELP Standards and assessments. It might be better at this point to try to develop National ELP standards rather than having states retool their ELP systems. Developing national standards for English Language Proficiency may be more daunting than National Content Standards because of the vast differences in the ELP standards and standards-based, annual ELP assessments across the nation, but such an effort may be needed for ELP standards and assessments to evolve and improve. They could also be aligned with National Content Standards as these develop.

When I started teaching ESL I didn't know what was expected of me. I taught first and third grade before, so I knew I could teach. What I didn't know, was how to teach ESL. After 10 years, I now know that we teach everything. I teach them to say bathroom so that they don't have an accident. I teach them to raise their hand when they want something for lunch. I teach them to say hello, thank you, yes, no, please, may I have a piece of paper. We talk about the parts of the body, different types of animals, how much five plus five is. Hopefully, in six months I can have a complete beginner student writing a few sentence about themselves, reading from a basic reader, and commenting on how the new student 'is learn English quick'. This student, could easily be a fifth grader as much as he could be a first grade. So tell me, how, in 4 more months could I get that student to pass the state Math test, Science test, or heck even the NYS ESL Achievement Test. What makes people think that guessing on a few fourth grade leveled reading passages should make a second grade beginner move to intermediate.

When I started my Masters it was in Reading. Taking a course accidentally, I found a passionate group of people that loved their jobs and that motivated me to switch to ESL. It had never occurred to me before then. Years earlier, while going for my BA in Elementary and Special Education, my roommate stated that she was only going for her regular Elementary license because she didn't want to work with those types of kids. I wonder how she is doing now with her inclusion students? The ESL students in her room? The students with disabilities whose parents deny anything is wrong or because the district says the child isn't failing enough to test? Am I the only one that felt the need to get certified in Elementary, Special Edu, TESOL, and Math so that I can help All of my students, whatever the level they are at?.

Yet with all of my certifications, I am incapable of exiting all of my students. How can I focus on the 3 students who still don't know the basic letter sounds after two years with me when they were always in a group of 14-20 students? Is that remediation?

Many of my students have reading difficulties (yet they do not attend classes with the reading specialist), attention and language delays (yet they go untested because they speak another language at home), and worse yet, come from poverished homes where lack of healthy food and health care may have caused developmental delays. No matter how many tutorial sessions I hold, no matter how many kids I cram into my small room, I am but a human, trying the best that I can to educate my students in ALL subject areas, at ALL reading levels, speaking ALL different levels of English and sometimes in MULTIPLE grades at the same time. So no, I don't my paycheck to rest on the test scores of my students. Simply on the dedication I bring with me to work every day to chip down the wall that someone so diligently built back up all night while I was home with my family.

I seems to me that if the Federal Government is going to hold teachers accountable for the effectiveness of their teachings to ELL's, they (the Feds) should supply all schools with the teaching tool. If all teachers are using the same technology to teach ELL's, the teaching (playing) field is leveled and progress is measured on an "apples to apples" basis. Even if the Fed's test for progress is a "standardized" test, scores can be compared among ELL's that are learning from the same platform. Then the success rates amongst teachers can be accurately compared to determine pay incentives.

yes i strongly believe that all the teachers of different schools must be helped with enough tools that can be used teach ELL effectively .
if the teachers are not been given anytools therefore teachers can not take the responsibilty of learners not working good in ELL.

Comments are now closed for this post.

Follow This Blog


Most Viewed on Education Week



Recent Comments