Research explores test-driven decision making that can limit opportunities, keep students reading below grade level.
As an English teacher at Pasadena High School and at the Rudas Secondary School of Economics in Dunaújváros, Hungary, Patty Carroll was a first-hand witness to what she calls a “reading crisis” in schools. Today, the Ph.D. candidate is in a joint doctoral program in Special Education which is overseen by UCLA School of Education’s Human Development and Psychology Division and California State University, Los Angeles. Her research, which is conducted with her dissertation advisor Professor Alison Bailey, explores the assessments that schools use to place children into English learner (EL) or special education classes. Carroll says that her research was fueled by her interest in reading and language development.
“When I was teaching English overseas, I saw that students acquired English as a second language at different rates,” Carroll says. “There was certainly no one-size-fits-all method of language teaching, especially with a non-phonetic language like English. If there was a reading crisis, it was caused by the language itself. But what I found in teaching high school in the states is that I had bright, eager ninth graders who still couldn’t read fluently – some were even stuck at a 2nd or 3rd grade reading level. These were kids that were classified as English learners, but they were born in the United States. I thought, ‘Something has gone wrong here.’”
After earning her master’s degree at Cal State LA in TESOL, Carroll became a consultant for the Berkeley-based Consortium on Reading Excellence (CORE), working with schools nationwide to implement reading programs and to guide teachers through assessment-based decision making in K-12.
“Even through these initiatives, I saw that we were still not doing a strong enough job helping teachers know the difference between typical and atypical delays in reading development, especial for English learners,” she says. “Something that I noticed when working with literacy coaches, teachers, and principals is that there is still a research-to-practice gap. There are policies in place to help schools meet the needs of students who have other languages spoken at home or who may have learning disabilities, and there is professional development for teachers on how to administer and interpret tests. But the training tends to stop outside the classroom door. Once teachers step inside, they’re on their own. How could they use what they learned for split second decisions in full inclusion classrooms of 20, or 40 students?”
Through her work with CORE, Carroll became interested in how assessments are developed and used in schools to decide which students get extra help with reading and math, and which students are placed in special education. She wanted to know more about the laws and policies behind special education, particularly as they relate to English learners, which led to pursuing her doctorate at UCLA. She says that assessment results can obscure the actual development needs of a young reader, whether a language other than English is a primary influence or not.
“When kids are enrolled for school in the United States, every family fills out a home language survey,” Carroll says. “If a language other than English is reported to be spoken in the home, the child is immediately given an assessment to measure their English language abilities. Students without sufficient English skills will perform poorly on this assessment, but so will students with language delays or learning disabilities, especially at age four. A child may have a family member who speaks another language, but that doesn’t necessarily mean that differences or delays are related to the presence of another language. There could be a developmental delay, a cognitive delay, or just the fact that the child is four years old and doesn’t test well. So the burning question is, when a student labeled as an English learner gets to middle school or high school and still isn’t reading effectively, what is the root cause?
“As a teacher, I didn’t have a lot of access to information that would help me answer this question. What I saw in working with students every day is that they wanted to be successful in high school, to graduate, and to move on to better things, but some were stuck in an underperforming track of classes just because they were still classified as an English learner. Other classes that I was teaching had native English speakers who were also still struggling to read fluently.
“It’s understandable when parents express concern about what’s happening in the EL system. If kids are acquiring language and learning to read at the same time, there is a specific kind of instruction we can provide. If kids need individualized educational programs, we have the tools to make that happen. But if struggling readers are just struggling readers, should we be segregating them into different tracks based on an assessment we gave them when they were four?
Carroll’s research seeks to find out if students who spend extended time classified as English learners do so because they need to acquire more language, or if they cannot exit EL programs because of the policies and practices around how test data are interpreted. Her dissertation, titled “A Mixed Methods Study of Data Use and Policy Implementation: English Language Proficiency Classification, Reclassification, and Educational Programming Decisions for Kindergarten-Twelfth Grade Language Minority Students in One California School District,” is being written from interviews of district administrators, principals, vice principals, ELD teachers, and special education providers, as well as six years of performance data on the district’s EL students.
“I’m looking at how data is used to make decisions about students, exploring the data to see why students may be staying in EL programs too long, and then analyzing EL system policies to see if they are doing the job we expect them to do,” Carroll says. “Schools use multiple sources of data to make these decisions, but how the data are combined in making a decision – the decision rules – can make a big difference. Under some rules, students who no longer need EL services may have to wait a full year to access college-preparation courses because of missing just one point on one subtest. Under other rules, students may have to wait until they have mastered reading before their schedule will allow time to receive extra help in math. Decision rules have a direct impact on EL students’ outcomes. My work aims to equip districts and policymakers with a greater understanding of how all these pieces fit together.”
Carroll says that pressure on schools from local and federal government on how EL programs are funded makes teachers and administrators feel that their hands are tied.
“Schools would like to mainstream a student as soon as he or she is ready,” she says. “But when there is no funding allocated to monitor the progress of students who exit, monitoring which is part of the federal mandate, some districts choose to err on the side of waiting to reclassify just so students will have some sort of support. In California as in much of the nation, we’re moving toward a common definition of English learner. To achieve this, we’re trying to standardize how multiple measures are combined and interpreted nationwide. We also have this big push towards using multiple measures for all kinds of decisions, not only in EL and special education, but also for gifted programs and how we assess teachers. Using multiple measures is easy in theory, yet difficult in practice. That is why research like this, research on data use in schools, is so important.”
Carroll’s research is timely, in view of the recent wave of test development in the United States, including an overhaul of state Common Core assessments that are being piloted in California and throughout the nation.
“This means that we’re going to be measuring students’ abilities slightly differently in the coming years based on a new rigorous set of standards,” says Carroll. “Also, the English language proficiency assessments are being rewritten, not only in California but for all states, and researchers here at CRESST are involved in these efforts. So right now, we have a prime opportunity at UCLA to provide research and leadership to ensure these assessments produce the best possible inferences for decision making and instruction for all students.
“The language assessment policies that have been put into place were designed to ensure equal opportunities for students who come from what we call language minority homes,” she says. “The question remains, is policy implementation producing the outcomes we had hoped for, and are data being used to create equal opportunity? Or are labels just creating unintended limitations? We want students to get the support they need, for as long as they need it.
“I believe that bilingual, dual immersion, and EL programs are a positive,” says Carroll. “Knowing more than one language is a tremendous resource, and our schools should be doing more to encourage bilingualism. Language learning enriches our society and makes us more competitive globally. For that reason, I want to see assessments used to the best of their intent and interpreted in ways that reduce limitations. No child should have to wait until they learn to read to get extra help in math. And no child should have to wait until the ninth grade to learn how to read.”
Carroll is a Teaching Fellow in Writing Programs at UCLA, and earned her master’s degree in education from UCLA in 2012. She has recently had her articles, “Determining English language proficiency for students with disabilities: Why decision rules matter,” accepted for a special issue of the National Society for the Study of Education yearbook, and “Do decision rules matter? A descriptive study of English language proficiency assessment classifications for English-language learners and native English speakers in fifth grade” (with A.L. Bailey) accepted by the journal Language Testing. Carroll is a member of the American Educational Research Association (AERA) and the National Council on Measurement in Education (NCME).