Is a New English-Proficiency Test Too Hard? Educators and Experts Debate.
In the wake of a seismic shift in how the nation’s most-used English-proficiency test is scored, a growing number of states have lowered their criteria for students to exit English-learner support services.
Roughly 2 million students took ACCESS 2.0 exams this past school year, encountering new standards that aim to raise the bar for English-language proficiency.
In many of the 35 states that belong to the WIDA consortium—and use ACCESS 2.0, the common test it designed to assess students’ language proficiency—scores plummeted under the more demanding requirements.
For school systems large and small, educating more English-learners than they planned for has meant potential budgeting, scheduling, and staffing crises.
But some districts won’t feel the full brunt of the change right away.
Facing potential bottlenecks in the pipelines that move hundreds of thousands of students from the status of English-language learner to English-proficient each school year, states across the country are re-examining when students make the transition.
Colorado, Maine, Massachusetts, and Tennessee are among the states that have already lowered their exit criteria in response to the new scoring scale.
In Massachusetts, the state education department is evaluating the new scale. Until it reaches a decision, the agency has advised districts to use the old standard while making reclassification decisions.
“We support the consortium’s commitment to rigorous English-language-proficiency standards, but changes to those standards must be vetted carefully, and, if they are adopted, implemented with appropriate advance notice to schools and families,” Jeff Wulfson, the acting commissioner of the Massachusetts Department of Elementary and Secondary Education, said in a statement.
States’ moves to lower the bar for proficiency has ignited debate. Will the thousands of students who fell short of the higher bar be prepared to leave support services, such as English-as-a-second-language classes, and tackle the more-rigorous academic standards in math, English/language arts, and science adopted by many states in recent years?
Some experts maintain that there isn’t a right answer.
“The perception somehow is that when they’re getting a service, an ELL service or in a bilingual program, that somehow they’re not getting a rich academic curriculum. Then somehow you get to the mainstream class and it’s the promised land,” said Kathy Escamilla, an education professor at the University of Colorado and former president of the National Association for Bilingual Education. “If kids aren’t ready ... you’re still putting them in the situation where they don’t know enough English to be successful. It’s especially a problem with redesignation criteria that are too low.”
With the scoring-scale change on the ACCESS 2.0 test, English-learners must demonstrate more-complex language skills in four domains—listening, speaking, reading, and writing—to achieve the same proficiency-level scores.
The WIDA exam has six achievement levels. In the speaking domain, for example, Level 1 students can name objects, people, and pictures, and answer who, what, or when questions.
At Level 5—a score that students must reach to be reclassified as English-proficient in some WIDA-member states—a speaker can engage in debates, listing examples and justifying their responses.
WIDA does not recommend cut scores on the ACCESS test; those decisions fall to states. Prior to the scoring-scale change, most states set their cut scores at 4 or 5.
Anticipating a rough adjustment under the new standards, WIDA’s leadership encouraged states to exercise caution before changing cut scores after the first year, WIDA spokesman Scott Gomer said.
Not all states heeded the advice.
State administrators in Colorado estimate that only 3 percent of the state’s English-learners would have moved out of all language-development courses under the new scale. In past years, between 17 percent and 20 percent of students moved out.
That would mean about 17,000 fewer students had the potential to move out of English-learner support services in the state, where students had to reach an overall score of 5 on a previous version of the ACCESS test to be considered for redesignation as English-proficient.
This summer, the state dropped that requirement to Level 4, the stage where students can give speeches and oral reports, and discuss stories, issues, and concepts. To set the revised criteria, the state studied the relationship between student scores on the ACCESS exams and state tests in other subjects.
“Our expectations are the same, we just had to recalibrate based on the new standard,” said Marie Huchton, a senior statistical consultant in Colorado’s education department.
With academic standards in traditional subjects rising around the country, WIDA had no choice but to increase their own standards.
“They realized that there was a shift on the language demands. They then took a look at their own standards. The standards now follow that shift, better support the kids and what’s needed in the content, not just in English development,” said Magda Chia, the director for strategy, impact, and policy at the Understanding Language/Stanford Center for Assessment, Learning, and Equity. “But, of course, it has a ripple effect that I think they, you know, they met as best that anyone could.”
Without state-level intervention, that ripple effect would have caused major disruption in districts such as Boston and Denver—where 40 to 50 percent of students are English-learners. The new standard created the conditions for a steep drop in the number of students exiting English-learner support services across the nation.
Boston would have gone from having 3,800 students eligible for reclassification two years ago to roughly 730 this past school year—an 80 percent drop.
In Colorado, the Denver schools would have experienced a similar decline, officials estimate.
In Denver, district data show that ELL students who reached Level 5 on the previous ACCESS exam outperformed, on average, their native-English speaking peers on state assessments in reading, writing, and math, and have higher graduation rates.
“It has been a very high bar and students who have reached that bar have been very, very successful,” said Susan Cordova, the district’s deputy superintendent, who began her career as a bilingual teacher.
But that doesn’t always mean that the former English-learners passed state tests. An average of 36 percent of Denver students in 3rd through 9th grades met or exceeded expectations on state English/language arts tests.
Starting with the 2017-18 school year, the performance of English-language learners will have a much greater impact on how schools are judged under the federal Every Student Succeeds Act. The changes in the federal law mandate a more precise measurement of how those students are performing in school.
Plenty of states rely solely on the results of a single test, such as ACCESS 2.0, to determine when students exit support services. The scoring shift in the widely used English-proficiency exam has some states revising the benchmarks they set for English-learner growth.
“There’s very high stakes attached to this,” Escamilla said. “You change the cut score and all of a sudden [a drop happens]. It looks like you’re doing a poor job of educating kids when that might not be it at all.”
Some classroom teachers have criticized the scoring-scale changes, arguing that it places the bar for English-proficiency far too high.
Expecting English-learners students will meet or exceed standards on common core-aligned assessments when they emerge from language-support services is unfair, when native English speakers often struggle to do so, said Jessica Haralson, an ELL teacher in the Malden, Mass., public schools.
“The expectations for English-language learners are actually more arduous than what many monolingual students can actually produce,” Haralson said. “Tons of monolingual students would be classified as ELLs overnight if ACCESS were administered to all.”
Mark Nigolian, a teacher in the Burlington, Vt., school district has participated in state-level discussions on the tougher proficiency standards, but said he often struggles to explain to his own students what they’re being tested on. People working in state education departments don’t often have that experience when setting policy or establishing cut scores, he said.
“Come into the trenches and see what it’s like or actually administer the test and really see that connection,” Nigolian said.
Chia, from Stanford’s Understanding Language initiative, said states and districts will have to shore up their curriculum and teacher training to meet the new WIDA standards.
“The changes in the standards both for the English-language proficiency and the content assessment are incredibly challenging,” Chia said. “The reaction is to blame the test, blame the test scores, because that’s an easy thing to point to. It’s more challenging. That doesn’t mean that the test is invalid; it just means that we’re demanding more of the kids. So the question then is ‘Well, should we be demanding more of the kids? And I think the answer is ‘yes.’”