|Title:||Writing for English Language Learners (WELLS): Exploring the Relationship Between Writing Instruction and Student Outcomes|
|Principal Investigator:||Gort, Mileidis||Awardee:||University of Colorado, Boulder|
|Program:||English Learners [Program Details]|
|Award Period:||3 years (9/1/2014–8/31/2017)||Award Amount:||$1,420,910|
Co-Investigators: Linda Caswell (Abt Associates); Elizabeth Howard (University of Connecticut); Mary Avalos (University of Miami)
Previous Grant Number: R305A140695
Previous University Affiliation: Ohio State University
Purpose: The ability to write well is a foundational skill necessary for success at school and beyond; however, English language learners (ELLs) consistently perform at lower levels in writing than proficient English speakers. Little is known about how best to improve the writing performance of ELLs. In this project, the research team aims to explore two key malleable factors that have the potential to improve ELLs writing outcomes — the quality of sheltered instruction which is designed to provide additional support for ELLs and the quality of writing instruction. Researchers will investigate the relationship between instruction and writing (with a focus on argument genre) among Latino students in fourth grade. The researchers will also investigate potential differences in these relationships for students with different levels of English and Spanish language and literacy proficiency.
Project Activities: Researchers will examine the relationship between classroom writing instruction and writing outcomes for three groups of Latino students: those who are currently identified as ELL; those who were formerly identified as ELL; and those never identified as ELL. The team will develop a new observation tool to measure the quality of classroom writing instruction as well as a rubric for scoring writing samples in the first year of the study. In the second year, the team will observe writing instruction and gather writing samples from fourth-grade students in 20 classrooms. Analyses will be conducted on data collected in the second year of the grant on sheltered instruction, classroom instruction in writing, student assessments of vocabulary (in both English and Spanish), teacher experience in teaching ELLs, and language practices in the home.
Products: Products of this grant include preliminary evidence of potentially promising practices and strategies to improve ELLs' writing abilities. In addition, researchers will disseminate findings through peer-reviewed publications, publications for practitioners, and presentations at conferences for researchers and practitioners.
Setting: This project will take place in elementary schools in an urban district in Florida. In these schools, approximately 20% of students are ELLs and primarily Spanish-speaking.
Sample: Approximately 340 fourth-grade Latino students in 20 classrooms will participate in this study.
Intervention: The research team will study malleable factors in writing and sheltered instruction to provide future guidance for improving classroom instruction to improve ELL writing performance. Writing instruction will be characterized using a researcher-developed tool (called ‘WIQ-OP') focused on practices thought to increase the use of academic language. Sheltered instruction will be characterized using the Sheltered Instruction Observation Protocol (SIOP). Systemic functional linguistics theory will inform identification of features of academic language that students need to use to write effectively, with a focus on argument writing.
Research Design and Methods: In Year 1, researchers will develop a new observation protocol to measure the quality of classroom writing instruction (WIQ-OP) and a rubric for scoring argument writing samples with a focus on use of academic language (WELLs-AL). The WIQ-OP will capture elements of writing instruction as well as supports for acquisition of academic language. The WELLS-AL will focus on linguistic and genre-specific elements of academic language. In Year 2, the team will observe classroom writing instruction and gather writing samples from students in each of the three classification groups. Data collection will include student vocabulary assessments in the fall, classroom observations of writing instruction using the WIQ-OP and SIOP protocols; a teacher questionnaire to obtain information on prior experience teaching ELs, writing, and in using the SIOP model; student-produced argument writing samples; and student background data obtained from school records. Student writing samples will be scored in the summer of Year 2 using both the Florida Standardized Assessment (FSA) rubric and the WELLs-AL rubric. Data analyses will take place in Year 3.
Control Condition: Due to the nature of the research design, there is no control condition.
Key Measures: Key student outcome measures for this project include the Expressive One-Word Picture Vocabulary Test — 4: Spanish-Bilingual Edition (EOWPVT); FSA writing score; and argument writing score (using a researcher-developed scoring rubric called WELLs-AL). Measures of classroom instruction include the WIQ-OP and SIOP protocols.
Data Analytic Strategy: Researchers will describe the quality of writing instruction and sheltered instruction using descriptive statistics. Hierarchical linear modeling will be used to account for clustering of students in classrooms and to predict student FSA writing scores from the WELLs-AL rubric score. EL status, prior year (third grade) FSA scores, EOWPVT score, and home language and literacy practices will be included as potential moderators of student performance. In the analyses, researchers will control for teacher characteristics and student demographic variables, including gender, free/reduced lunch eligibility, attendance, and participation in special education.