July 21, 2015
Project GLAD — which stands for Guided Language Acquisition Instruction — is a model of sheltered English instruction that school districts around the country have used to prepare their teachers to work with English language learner (ELL) students. ((Project GLAD is owned by the Orange County Department of Education (OCDE), which also houses the program’s National Training Center for professional development services. BE GLAD is endorsed by Marcia Brechtel, the co-creator of Project GLAD, and is now independently run despite its shared origins with the OCDE model. For an explanation of the difference between the OCDE-owned Project GLAD and BE GLAD, click here.)) In fact, almost 300,000 teachers in 18 states have received training in GLAD strategies.
GLAD began as a series of instructional strategies developed in the 1980s by teachers Marcia Brechtel and Linnea Haley to teach academic content to ELLs. In the 90s the project received federal grants to develop into a formal instructional training model. But despite its long-running history, expansive reach, and popularity, little is known about the effects of the model on the educational outcomes of DLLs.
Luckily, researchers from Education Northwest are engaged in a rigorous randomized evaluation of the program — the first experimental study of Project GLAD to date — with funding from the Institute of Education Sciences. The four-year longitudinal study conducted randomized trials in 30 elementary schools in Idaho. In other words, half of the schools were randomly assigned to the treatment group, in which classrooms were instructed with GLAD strategies and the other half were assigned to the control group, in which classrooms were instructed as usual. ((Teachers in control group classrooms carried out “business as usual.” Investigators defined this as the typical instruction ELLs receive in Idaho, which is for the most part a form of sheltered English or ESL instruction.)) Of the 113 teachers in the study, 51 were randomly assigned to receive Project GLAD training and 62 were assigned to a control group that did not receive training.
Treatment group teachers were first trained in GLAD strategies and then observed as they carried out the model in classrooms with fifth grade ELL and non-ELL students. ((As a type of sheltered English instruction model, GLAD is carried out in classrooms with students with varying levels of English language proficiency. Therefore, one of the main goals of the study was to determine whether or not Project GLAD has a different impact on ELL and non-ELL students.)) The model was examined for its impact on the reading comprehension, vocabulary, writing, and science achievement of all students in the experimental classrooms (as compared to students in the control classrooms), and for teachers’ implementation of the model.
So what does the model look like in a classroom setting? As a type of sheltered instruction, Project GLAD is used in mainstream academic classrooms with students at various levels of English language development. Its goal is to seamlessly build in linguistic supports for DLLs during regular content instruction, so that they are better able to understand the subject material at hand. The model goes a step beyond other models, such as SIOP, by giving teachers specific strategies and prescriptive directions on how to implement them in content instruction. Educators are instructed in 35 strategies that are grouped into four goal-oriented categories (focus and motivation; input of new knowledge; guided oral practice; and reading and writing development).
Some of the strategies, such as “group frame” are specifically designed as English language development tools. With “group frame,” the instructor asks a group of students to orally respond to a prompt or to retell a story while he/she records the response on a chart made visible to all students. Oral language development is especially important for DLLs; research shows (see here and here) that strong oral proficiency in English is associated with ELLs’ enhanced reading comprehension, vocabulary knowledge, and writing skills. This means that instruction for DLLs should dedicate a significant amount of time to developing their oral proficiency skills in order to ensure later achievement in reading, writing, and vocabulary.
Initial results from the first two years of the study have been mixed in terms of GLAD’s impact on students’ educational outcomes. The program had a positive impact on the vocabulary, reading comprehension, and on select writing skills of ELLs, although the effects on non-ELLs were inconclusive. ((In the first year of the study, researchers found no significant impact on non-ELLs. However, in the second year, they found a slight negative impact on non-ELLs’ vocabulary knowledge.)) The study also found that the positive impact on ELLs was strongest for those with intermediate levels of English proficiency, ((The study refers to students’ level of English proficiency as measured by the Idaho English Language Assessment (IELA). IELA assigns students the following proficiency levels: beginning, advanced beginning, intermediate, early fluent, and fluent. It defines intermediate as the level where “students begin to expand the complexity and variety of their communication skills but exhibit fairly frequent errors that may impede meaning.”)) but substantially less so for students at other levels of proficiency. According to Theresa Deussen, principal investigator in the study, the use of GLAD “may make more sense in a school made up mostly of intermediate ELLs, compared to a school with just 15 percent ELLs and 85 percent non-ELLs.” In other words, GLAD works best in schools where the majority of students are ELLs and where the majority of ELLs are at an intermediate level of proficiency.
The study also found that GLAD is popular amongst teachers. A resounding 95 percent of the GLAD instructors in the study strongly agreed with the statement, “I enjoy teaching my students using Project GLAD strategies.” However, the study also identified several challenges teachers face, including the lack of time for planning GLAD lessons and preparing materials to use with the strategies, and the difficulty of implementing the lessons as the model intends. For example, many teachers found it difficult to recall all of the strategies and how to properly use them.
But the most significant conclusion of the Education Northwest study is that as a stand-alone measure, Project GLAD is not enough to close the achievement gap between ELLs and non-ELLs. Additionally, the report’s authors cautioned against understanding the model as a replacement for English language development programs for DLLs. Deussen warns that an improper understanding of both GLAD’s purpose and its implementation can be detrimental to DLLs. Citing a case study conducted in Washington state on the use of Project GLAD to instruct ELLs, Deussen explains that districts felt they were serving DLLs by purchasing GLAD training as a sheltered English instruction model, but because teachers “saw their Project GLAD training as providing one more tool in their toolbox, a tool they could use or not use as they saw fit,” DLLs were not receiving either the full implementation of the structured English instruction program or English language development lessons.
Based on the available evidence of how GLAD affects DLLs’ educational performance, schools should probably consider using the model only if the majority of students are intermediate-level DLLs, and if the model is implemented as a supplemental program of language support in mainstream classes for DLLs, not as their only intervention of English language development. Moreover, the results showing the difficulty teachers had in implementing the model should make schools keen on clarifying its purposes and intended form of delivery, to teachers.
The fact that Project GLAD has been in use to instruct DLLs for over 20 years despite the dearth of data on its impacts and implementation makes me wonder, how can a model that has not undergone rigorous evaluation for its effectiveness and implementation, possibly be endorsed by the California Department of Education for use with ELLs and for Title III professional development funding? While its presumed effectiveness may produce praise and satisfied teachers, that doesn’t mean that we know how the model actually affects DLLs. In sum: ignorance may be bliss, but it does not prevent the very real and damaging consequences that may be lurking in the background.
This post is part of New America’s Dual Language Learner National Work Group. Click here for more information on this team’s work. To subscribe to the biweekly newsletter, click here, enter your contact information, and select “Education Policy.”"