|Title:||Exploring Effective Reading Comprehension Instruction: Classroom Practice, Teacher, and Student Characteristics|
|Principal Investigator:||Connor, Carol M.||Awardee:||University of California, Irvine|
|Program:||Effective Teachers and Effective Teaching [Program Details]|
|Award Period:||4 years (7/1/2013-6/30/2017)||Award Amount:||$1,600,000|
Previous award number: R305A130058
Co-Principal Investigator: Joanne Carlisle (University of Michigan)
Purpose: In general, instructional approaches for improving reading comprehension are not as well developed as are those for decoding. Most recent studies have focused on improving reading comprehension in the upper elementary years and beyond, but these have often not shown significant effects. The purpose of the proposed study is to identify aspects of effective instruction in reading comprehension instruction that predict students’ reading comprehension gains in first, second, and third grade. Researchers plan to explore the complex associations among contexts of instruction, student characteristics and teacher characteristics and to identify how they support or fail to support, as a system, students’ reading comprehension gains. By identifying components of effective comprehension instruction and the contexts that support this instruction, the team will generate a framework that can be used to inform valid and reliable classroom observation systems, develop effective comprehension interventions, and design professional development programs for early elementary teachers.
Project Activities: For this project, the team will analyze archival classroom data, and collect and analyze new observational classroom data. To support these analysis efforts, researchers will refine a system of classroom observation focused on describing reading comprehension lessons. This refined system, the Effective Teaching and Classroom Learning Environments for Reading Comprehension (CLE-RC), will integrate and expand on three existing classroom observation systems that have been used to identify characteristics of effective reading comprehension instruction in early elementary classrooms. Once the CLE-RC system has been refined, it will be used to recode 120 archival video-taped observations of early elementary classrooms, and code 100 new classroom observation videos. The team will analyze the student and teacher assessment data that are linked to these observations to examine moderators and mediators of the effects of instruction on student comprehension gains. This data will then be used to generate a well-developed conceptual framework and theory of action that identifies effective reading comprehension practices and contexts.
Products: The products of this project will be identification of components of effective reading comprehension instruction in grades 1, 2, and 3, and a well-developed conceptual framework that can be used to inform the development of valid and reliable classroom observation systems, effective comprehension interventions, and professional development programs for early elementary teachers. Peer reviewed publications will also be produced.
Setting: The archival videotaped classroom observations used in this study were conducted in Florida and Michigan. These communities are urban, suburban, and rural and are socioeconomically and racial/ethnically diverse. In Years 3 and 4, the researchers will collect additional observation and student data in Arizona, a substantially different setting.
Sample: Eight Florida and Michigan districts and 44 schools participated in the studies from which the videotaped classroom observations were gathered. From these schools, 291 teachers and 2,352 students from Florida, and 40 teachers and 680 students from Michigan participated. In Years 3 and 4, researchers will collect additional observation (n=100) in Arizona. Across samples, approximately 40 percent are African American, 45 percent are White, and the remaining students belong to other ethnicities including Hispanic and Asian. Approximately 40 percent of the Florida students in these samples qualify for the free and reduced lunch program with a higher percentage in Michigan because most of the schools were participating in the Reading First program. The Arizona sample will include a higher proportion of language minority students.
Intervention: Researchers refine a system of classroom observation that integrates and expands on three existing classroom observation systems to explore the salient characteristics of effective reading comprehension instruction. The first is a measure that has been designed to study the kind and quality of instructional practices teachers use in teaching reading comprehension. This measure, Support for Students Learning from Texts (SSLT), has undergone preliminary study. The other two are well-established systems that have been used in prior research: the Individualizing Student Instruction/Pathways (ISI) system and the Video Analysis of Teaching Reading (VATR). The results will be used to create the Effective Teaching and Classroom Learning Environments for Reading Comprehension (CLE-RC) system. Researchers will use this system describe how teachers teach students to understand the texts that they read. The CLE-RC analysis of each reading comprehension lesson will enable an exploration of the relation between teachers’ instructional actions in comprehension lessons, students’ participation, and other contextual factors. In addition, the system will provide data on both the particular instructional moves teachers use and the quality of these moves for individual students and the class as a whole, while considering aspects of the classroom learning environment that predict student outcomes.
Research Design and Methods: The proposed exploratory project will use both new data and archival data. A series of tasks will be carried out to support the goal of exploring the multiple dimensions of teaching and classroom reading comprehension instruction that work together to predict students’ reading comprehension gains. First, the team will make minor adaptations to the three extant observational systems to enable use of the new integrated system to analyze classroom observation data. Second, the team will use an iterative approach to fully refine the Effective Teaching and Classroom Learning Environment for Reading Comprehension (CLE-RC) system. Once the system is stable, reliable, and informative, the team will code 120 archival classroom videos not used in the initial phase. The team will analyze the data from the use of the CLE-RC to determine if this new measure is able to account for additional variance in teachers’ reading comprehension instruction with regard to its impact on students’ reading comprehension gains. When the analysis is complete, the team will conduct two sets of classroom observations using the CLE-RC, once in Year 3 and once in Year 4. The team will assess students’ language and literacy skills, and test the validity and reliability of the CLE-RC with this new sample of teachers and their classrooms. At the conclusion of the study, the team will prepare a fully developed conceptual framework and theory of action based on data collected and analyzed with CLE-RC that will expand our knowledge of effective comprehension instruction.
Control Condition: In the ISI studies, some teachers were randomly assigned to participate in the ISI control condition. This was either a “business-as-usual” condition or an alternative intervention. Therefore, intervention condition will be a covariate in all analyses.
Key Measures: The principal measures will include classroom observation variables in the ISI, VATR and SSLT across three dimensions: Organization, Content, and Context plus other key instructional variables that emerge from the iterative refinement of the CLE-RC. Archival data from Florida include student data from the Woodcock-Johnson II battery, and the Gates-MacGinitie Comprehension Test (Gates). Measures of self-regulation, behavior, and social skills using the Social Skills Rating System were also collected. Teacher background and demographic information, and teacher knowledge about essential concepts about literacy was also collected. Archival data from Michigan includes student performance on the Iowa Tests of Basic Skills (ITBS) reading subtests and Oral Reading Fluency from the Dynamic Indicators of Basic Literacy. Teacher knowledge was assessed using the Teachers Knowledge of Reading and Reading Practices (TKRRP). In the new data to be collected in Arizona, the team will select student and teacher assessments based on the Florida and Michigan protocols. The basic protocol will include student performance on the Gates and the ITBS. These measures include reading and listening comprehension. They will also administer the TKRRP and collect background information on both teacher and students.
Data Analytic Strategies: This project will use advanced methods in psychometric and multi-level multivariate modeling. Researchers will complete psychometric analysis of the new scale and work to establish and study inter-coder reliability. Then, researchers will use value-added analyses to help investigate the validity of CLE-RC. Researchers will use a range of analytic strategies, including a bottom-up iterative approach that makes use of multilevel exploratory measurement models and multilevel structural equation models.