|Title:||The Roles of Instruction and Component Skills in Reading Achievement|
|Principal Investigator:||Branum-Martin, Lee||Awardee:||Georgia State University|
|Program:||Literacy [Program Details]|
|Award Period:||2 years (9/1/2012-8/31/2014)||Award Amount:||$422,549|
Co-Principal Investigator: Pat Taylor and Coleen Carlson (University of Houston)
Purpose: Understanding why some schools are more successful at teaching reading than others requires examination of three key components: student characteristics, school practices such as instruction, and school context effects. Using the Texas Reading First dataset, the research team will complete a secondary data analysis of student reading performance in over 200,000 students in 809 schools in 222 districts over six years. The researchers will examine the nature and growth of student literacy skills over time; the stability of instructional strategies, quality, and how instruction relates to literacy over time; and the extent to which variability between classrooms and campuses predicts differences in how instruction impacts reading achievement.
Project Activities: The researchers will conduct secondary data analysis on the Texas Reading First data set which includes data from multiple sources on more than 200,000 students over six consecutive years. To begin, the team will identify all data pertaining to each student and link data points over time. Data about teachers and schools was also collected, so the team will link students to their classrooms over time. After linking is complete, the team will complete a series of analyses, using both data from standardized tests and observations of teachers' classroom instruction.
Products: The project will produce information regarding the nature and growth over time in reading skills, and preliminary evidence of potentially promising teacher instructional strategies and school contexts. Additionally, the team will prepare peer reviewed publications.
Setting: Data for this study are drawn from students in 222 school districts across the state of Texas.
Sample: Data for this study are drawn from more than 200,000 students in grades K through 3. Also included are data from approximately 4,000 teachers observed in their classrooms during reading instruction.
Intervention: In this exploratory study, the researchers are examining the growth of basic literacy skills in grades K–3, the extent to which various instructional practices predict growth of literacy skills, and the role of school context in literacy growth to identify possible targets for intervention at the instructional or system level.
Research Design and Methods: Texas Reading First was implemented using a quasi-experimental design so that policymakers could examine the impact of district-level funding. This implementation enabled a cohort-sequential design, meaning each student was followed for multiple years, but not the whole six years of the project. Additionally, because multiple companies collected data for Texas Reading First, and they each used their own identification system, the first task of the current project is to link each student's data points across classrooms and over time. Once the data points are linked, the team will be able to answer research questions about student growth and associations with classroom and school level factors. Because students move from classroom to classroom over time (and thus are linked with different teachers over time), and due to the nature of the outcome variables, this team will rely on a newly developed statistical software program, xxM, designed with IES support (R305D090024).
Control Condition: Due to the nature of the research design, there is no control condition.
Key Measures: Texas Reading First used a variety of standardized measures of literacy skills. Schools were allowed to choose between the Iowa Test of Basic Skills or the Stanford Achievement Test, and their respective counterparts, Logramos and Aprenda. Schools were required to administer the Texas Primary Reading Inventory and the Spanish Tejas LEE. Schools were also required to administer the English and Spanish versions of the Texas Assessment of Knowledge and Skills. Instructional strategies were measured using the Texas Teacher Appraisal System, and instructional quality was assessed using the Early Language & Literacy Classroom Observation.
Data Analytic Strategy: Data from Texas Reading First provides a few specific challenges to analysis: the outcome variables are multivariate with numerous measures of literacy skills; the researchers are interested in examining change longitudinally over time; and the students are nested within classrooms within schools, but they each switch classrooms over the course of multiple years. To analyze the data for this secondary data analysis, the researchers will use the xxM statistical software. This software is the first designed to handle latent outcome variables, multilevel models for longitudinal analysis, and cross-classified variables to account for students switching classrooms.
Related IES Projects: Cross-Classified Structural Equations Model: Development of an OpenMX Module and its Application to Multiyear Assessment and Intervention Data in Literacy Research (R305D090024) and Language and Literacy Abilities in Spanish Language Speaking Children (R305A100272)