READING COMPREHENSION STRATEGY INSTRUCTION IN UPPER-ELEMENTARY CLASSROOMS WITH ENGLISH LANGUAGE LEARNERS: A STUDY OF PRACTICES AND OUTCOMES
Doyle, Candice Briece
Silverman, Rebecca D
MetadataShow full item record
The goal of this study was to investigate the relationship between reading comprehension strategy instruction (explicit or skills-based) in general education settings and third through fifth grade students' reading comprehension outcomes. In addition, I was interested in whether relationships between instruction and outcomes differed for students from English only (EO) and English language learner (ELL) backgrounds. To address these goals I conducted a secondary data analysis of 59 Reading/Language Arts classroom observation transcripts. These represented observations of 19 teachers at three time points (fall, winter, spring). I analyzed transcripts by employing an iterative coding process including open, axial, and selective coding (Strauss & Corbin, 1990). I coded teacher talk at the utterance (Crookes, 1990) level for either explicit instruction (instruction that included all of the following: introduction, modeling, collaborative practice, guided practice, independent practice) or skills-based practice (teacher practice in which students were asked to apply a comprehension strategy absent of instruction of how to do so). In addition I coded for separate parts of the explicit instruction model (introduction, modeling, collaborative practice, guided practice, independent practice). Then, I quantitized (Tashakori & Tedlie, 1998) the instructional code data into average frequency counts across observations in order to conduct multiple regression analyses with student reading comprehension outcome measures. I found no statistically significant results related to the explicit instruction model (as a whole), or skills-based practice and students' outcomes. However, when analyzing separate parts of explicit instruction, results suggested that more guided practice was associated with higher scores on one outcome measure. In exploring interactions between language background and instructional codes, I found no differences in relationships between instructional codes and reading comprehension for EOs versus ELLs.