(2nd Update: Adding response from WIDA at University of Wisconsin, and response from author)
We’ve been hearing a lot about problems with new Common Core tests that were designed to be harder than the standardized tests districts used to administer for “accountability” purposes but this post presents a different kind of testing problem: when the test is too easy.
Lauren Kafka, an ESOL teacher in Montgomery County Public Schools for eight years and a journalist for 20 years prior to teaching, writes in the following piece about the “catastrophic academic consequences” for ESOL (English as a Second Language) students when they are evaluated on tests that are too easy, rather than too difficult. In June 2015, she will begin working as an editorial educational consultant.
By Lauren Kafka
Our rigorous testing season in Maryland public high schools is finally coming to a close, as it is in schools all over the country, and teachers, students, parents, administrators, politicians, and pundits are once again debating whether the benefits of high-stakes assessments outweigh the costs and how to lower the achievement gap. As a teacher of English for Speakers of Other Languages (ESOL) at Walt Whitman High School and a parent, I’ve seen first-hand what negative effects test-preparation angst can have on even the most capable students. What’s surprising and ironic is that one standardized test that is too easy for many ESOL students to pass can have even more catastrophic academic consequences than tests that are far more difficult for them.
In this era of Common Core testing in which the bar has been set very high, who would complain about a test with a low bar that helps ESOL students get promoted quickly to mainstream classes? It turns out to be a terrible idea for students in Maryland–a state with about 55,000 ESOL students during the 2014-15 school year.
All of Maryland’s ESOL students are now required to take a test called ACCESS (Assessing Comprehension and Communication in English State-to-State), and many ESOL teachers and administrators believe it is given far too much weight in determining how and when students exit ESOL programs. The results of this test provide the single data point that determines whether students are eligible for ESOL services in the state. No critical decision about education (college admissions, special education eligibility, scholarship awards, etc.) uses just one test score, excluding all other factors.
Even the organization that created this test, the World-class Instructional Design and Assessment (WIDA) Consortium at the University of Wisconsin, didn’t intend for it to be the only criterion. According to their website, no single score should be used as the sole factor for making decisions regarding a student’s English language proficiency. “Students should also be assessed using information about classroom language proficiency, typically provided by teachers,” said WIDA Research Director H. Gary Cook. “We agree that making decisions based on one data point is rarely a good practice.”
In Maryland, students don’t need to pass the ACCESS test to graduate, but they do need to score below a certain threshold (below a 5.0 out of a maximum score of 6.0) to be allowed to stay in ESOL programs. Our ESOL programs offer them accommodations to compensate for their language deficiencies such as sheltered content-area classes (biology, algebra, history, etc.) where the student/teacher ratio is significantly lower than it is in mainstream classes, and they also get extended time on content-area tests and the use of a bilingual dictionary. As soon as students score a 5.0 on the WIDA test, they can no longer enroll in ESOL classes, to fulfill their English requirement, or any of our ESOL-sheltered content-area classes.
Initially, many students and their parents are thrilled to hear that the student is ready to exit ESOL classes. Who wouldn’t be? It sounds like an impressive accomplishment! It’s not until the families realize how misleading the test results are and how they will impact the student’s academic success that they regret the news, and some try to appeal the decision, to no avail.
Here are some reasons why students’ WIDA test results are misleading:
Students take almost the same test each year
The ACCESS test is administered to K-12 English Language Learners when they enter the school system and again each year while they are in ESOL classes. According to WIDA, only “one-third of the test’s items are refreshed annually, which means that many test items are administered for three years prior to their permanent retirement from the test.” This means that approximately 70 percent of the questions are exactly the same as they were the previous year, which allows even the weakest students to improve their scores from one year to the next. Many of our students ask, “Why are we taking the same test we took last year?”
Extremely low threshold
A “passing” grade on the ACCESS test reflects an extremely low threshold of achievement. Most students who have been in ESOL classes for two years and have taken this test more than once can easily score a passing grade, even if all other measures demonstrate their English language proficiency to be quite low. The most challenging reading on the test has an eighth-grade equivalency level, and many students who cannot write two consecutive grammatically correct sentences score a passing grade.
No appeal process
Even if all of an ESOL student’s teachers and counselor agree that students would benefit from taking ESOL classes, they are not permitted to do so after they’ve scored a 5.0 on the ACCESS test. Each year we are asked if students are allowed to continue in our ESOL program because they know they aren’t ready for mainstream English, but we have to say no.
How do these students fare after they exit? Predictably, not well
At Walt Whitman, for example, between 25 and 35 students — about one third of those in our program — exit every June based on this single test score. Most go into mainstream, on-level English classes where they earn low or average grades at best. “Students who have passed this test and are then placed in my on-level English class are ill-equipped to meet the demands of the course,” said Elizabeth Keating, an English teacher at Walt Whitman. “Many of them still struggle to put together a coherent sentence, let alone a paragraph or essay. This is usually not due to a lack of insight or complex thought, but rather a lack of skill. They simply haven’t reached a level of mastery of the English language to allow them to express their ideas. The results are poor grades and their increasing frustration.” Some of these students fail their on-level classes, and, without the testing accommodations offered to ESOL students, several fail the challenging assessments required to graduate. Keating added, “As a regular general education English teacher, I cannot provide the support they’d get in an ESOL class designed to help them gain language skills. We’re doing these students a disservice by placing them out of ESOL courses before they’re ready.”
Maryland graduation rates for ELLs decreased
Our school is not the exception. For the third year in a row, the English Language Learners (ELLs) in Montgomery County Public Schools did not meet the third part of the county’s Annual Measurable Achievement Objectives. According to the Maryland State Department of Education website, as of January 2015, although graduation rates improved in Maryland in 2014 for virtually all demographic subgroups–African American, Hispanic, Asian, American Indian, and white students, as well as special education students and those receiving free- or reduced-price meals–the one group whose graduation rate dipped (from 57.3 to 54.1 percent last year) was ELLs.
In 2014, twelfth-grade students with limited English proficiency met only 36.1 percent of their High School Assessment requirements for graduation. This is more than 20 points lower than the closest subgroup. Many of my colleagues and I believe these students aren’t graduating because they are being pushed through our ESOL programs too quickly and placed in mainstream classes prematurely. “I strongly believe we are dooming many students who are leaving ESOL on the basis of WIDA results, as we are not giving them a chance to compete with native speakers,” said Irene Aluker, an ESOL teacher at Walter Johnson High School. “The decision about whether a student is ready for mainstream English or not has to stay with teachers.”
ACCESS shouldn’t be the only criterion
The ACCESS test, which is aligned with Common Core State Standards, has now been embraced by more than 30 states. It measures many important aspects of students’ progress, but it was never intended to be used as the only gate-keeper. No critical decision about education uses just a single data point, excluding all other factors. A student’s score on the SAT, for example, does not by itself determine whether he or she will be admitted to college. Yet the state of Maryland has determined that a student’s eligibility for ESOL services should be based on an ACCESS score of 5.0, and nothing else. The promotion of students from ESOL to mainstream classes should be based on their speaking, reading, writing, and listening abilities throughout the school year; semester and final exam grades; teacher feedback; and their results on standardized tests.
MCPS wants success for every student. It’s an exceptional school system in which teachers work diligently to help all students, including at-risk students, reach their fullest potential. Before the ACCESS test, many of our exiting ESOL students had the opportunity to enter honors classes and reach notable levels of achievement because their academic English level was so much higher when they left our program. We prepared them to thrive, rather than merely survive. Now we’re not only setting up our students for failure; we’re setting ourselves up for failure. Instead of narrowing the achievement gap, we’re making it even wider.
Here’s a response to the piece from WIDA at the Wisconsin Center for Education Research, University of Wisconsin-Madison
In Ms. Strauss’s article (http://www.washingtonpost.com/blogs/answer-sheet/wp/2015/05/29/why-carlos-cant-graduate-setting-up-esol-students-to-fail-with-a-too-easy-test/) that includes Ms. Kafka’s editorial on the ACCESS for ELLs assessment (ACCESS), there seem to be several misunderstandings. The author correctly quotes Dr. Cook and WIDA’s position regarding the use of ACCESS for reclassification of English Language Learners (ELLs). ACCESS should be but one piece of the information used in schools to determine whether a student requires language instructional services. We believe that teachers should be involved in the decision-making process for reclassifying ELL students as English learners. Regardless, reclassification policy is a state-level decision and one best directed toward those who understand a state’s context and students, in this case the Maryland Department of Education.
Ms. Kafka’s claim that ACCESS is too easy is simply not true. ACCESS assesses a broad range of language proficiency and has been used by many states to monitor academic English language development and ultimately English language proficiency attainment. Of the 1.4 million ELLs who participated on the ACCESS in the 2014-15 school year, less than 20% received an overall proficiency level greater than 5.0. These numbers are consistent with the research on how many students should be proficient annually on these types of English language proficiency assessments. The claim that students taking ACCESS see many of the same test items repeatedly is also not factual. Students seldom take identical forms of the test twice, as the test design supports students responding to new, more challenging items as their proficiency level increases.
Research has shown that ACCESS can indeed be used as part of the evidence in determining whether a student has reached English proficiency. Research also shows that ACCESS strongly predicts ELL students’ proficiency on state academic content assessments. . WIDA believes that every teacher within a school has a role to play in ensuring the success of ELLs both for language and school content learning. This is true when students qualify for special program support and after reclassification.
 Cook, H. G., Boals, T., & Lundberg, T. (2011). Academic achievement for English learners: What can we reasonably expect? Phi Delta Kappan 93(3), 66-69.
Cook, H. G., Linquanti, R., Chinen, M., & Jung, H. (2012). National Evaluation of Title III Implementation Supplemental Report: Exploring approaches to setting English language proficiency and performance criteria and monitoring English learner progress. Washington, DC: U.S. Department of Education.
 Parker, C. E., Louie, J., and O’Dwyer, L. (2009). New measures of English language proficiency and their relationship to performance on large-scale content assessments (Issues & Answers Report, REL 2009–No. 066). Washington, DC: U.S. Department of Education, Institute of Education Sciences, National Center for Education Evaluation and Regional Assistance, Regional Educational Laboratory Northeast and Islands. Retrieved from http://ies.ed.gov/ncee/edlabs.
Here’s a response to that comment by the author of the post, Lauren Kafka:
I was hesitant to respond to WIDA’s descriptions of “misunderstandings” because we agree on so much. As the unidentified WIDA representative noted, we agree that “the ACCESS test should be but one piece of the information used in schools to determine whether a student requires language instructional services.” We also agree, as the WIDA author noted, that “reclassification policy is a state-level decision,” and that’s why many of my colleagues and I hope the Maryland Department of Education will reconsider its use of ACCESS as the only criterion to promote students from ESOL to mainstream classes.
This short-sighted, misguided decision not only affects high school students who exit from ESOL classes prematurely; it also affects elementary and middle school students. I received positive feedback from numerous teachers at all three levels (elementary, middle, and high school) who thanked me for writing my essay. They shared stories with me about at-risk students who earned a passing score on the ACCESS test and then had a horrible time trying to pass their mainstream classes.
As for my claim that students taking the ACCESS test see many test items repeatedly, this information came directly from the WIDA website, and WIDA Research Director H. Gary Cook confirmed it during my accuracy check of the article. According to the WIDA site, “Approximately one-third of the test’s items are refreshed annually, which means that many test items are administered for three years prior to their permanent retirement from the test.” This means that approximately 70 percent of the questions are exactly the same as they were the previous year. Our students routinely ask why they have to take the same test they’ve already taken.
The anonymous WIDA representative also said that my claim that the ACCESS test is too easy is not true. In my last eight years of teaching, I’ve found that students rarely complain about tests that are too easy. The far more common complaint is about tests that are too difficult. After administering this year’s ACCESS test in January, I gave all of my students a brief, anonymous survey, so they could share their honest impressions of this assessment. The vast majority described the test as “too easy” and “not as challenging as the content or assessments in ESOL classes.” If the WIDA staff and the Maryland Department of Education conduct a similar survey among ESOL students at other schools and in other counties in the state, I believe they will find the same results.
The WIDA author wrote, “Of the 1.4 million ELLs who participated on the ACCESS in the 2014-15 school year, less than 20% received an overall proficiency level greater than 5.0.” Even if only 15% received this proficiency level, that’s 210,000 students who are exiting from ESOL classes, and I believe many of them will be completely unprepared for the mainstream classes and Common Core assessments they will be required to pass in order to graduate in the years ahead. Even if ESOL teachers in Maryland continue to administer the ACCESS test, it should not be the single data point used to determine where students are placed the following year. Teachers must be involved in the decision-making process about whether a student exits from ESOL. On this most fundamental point, we clearly agree.