At its most fundamental level, like all instruction, comprehension instruction is composed of moment-to-moment interactions among students and between students and their teacher. Children's competence at making meaning from texts and their motivation to read are shaped by a variety of classroom influences, one of the most important of which may be these interactions. This study listens in on teacher-student interactions with a particular ear toward "conceptual press discourse", an approach to responding to student ideas by pushing students further along the trajectories of their own thinking. Teachers enacting discourse high in conceptual press might ask students to elaborate on their ideas, offer evidence to support those ideas, or provide examples. The hypothesis driving this study is that students whose teachers use high conceptual press discourse moves more frequently will demonstrate stronger comprehension achievement, report greater engagement with reading, and view reading as involving thinking and considering multiple interpretations of texts. The data for this study were collected over the course of a school year in 21 fourth- and fifth-grade classrooms. The 495 students in these classes completed a series of measures keyed to comprehension achievement and engagement once in the fall of 2007 and again in the spring of 2008. Three observations of reading instruction were conducted in-between these two test administrations. These observations yielded quantitative data about teacher-student discourse patterns (in particular, teachers' use of conceptual press discourse). In addition to these data, five qualitative case studies were completed. Additional videotaped observations were conducted in each case study classroom, along with student and teacher interviews. Because students were nested within classrooms, Hierarchical Linear Modeling (HLM) was used to answer questions about associations between teacher level variables and student outcomes. Models were built to predict each of four student outcome measures (the comprehension test and three surveys addressing affective, behavioral, and cognitive facets of reading engagement). Qualitative data was analyzed using a contrasting case approach involving transcription, open- and focused-coding, developing data displays reflecting patterns in coding, and an iterative process of analytic memo writing. The quantitative portion of the study indicates that in classrooms where teachers use a pattern of talk that tended to reduce the cognitive load on students, such as offering hints and narrowing open-ended questions to make them easier, students scored lower on the comprehension achievement measure and rated their affective and cognitive engagement lower. Also, teachers who tended to use this pattern of talk frequently tended to use high conceptual press discourse infrequently. The amount of teacher talk devoted to instruction in comprehension strategies was also a factor predicting student outcomes. The qualitative portion of the study suggests that the patterns of discourse teachers enact are shaped by their beliefs about what it means to comprehend text and how they view their students. However, teachers do not enact discourse patterns that perfectly correspond to their beliefs. Classroom talk is messy. Teachers sometimes talk in ways that mismatch or even contradict their beliefs. Students recognize features of the teacher-student talk that occurs in their classrooms. Despite these "mismatches," students regularly adopt beliefs and perceptions about reading closely aligned to those espoused by their teachers. The findings of this study indicate that teachers should observe and consider the patterns of talk they enact with their students. Teachers should limit practices such as giving hints, narrowing questions to make them easier, answering their own questions, and gathering up student responses without truly examining them. However, teachers need not fixate on enacting a particular pattern of discourse "perfectly;" they should do the best they reasonably can to enact discourse patterns conveying a sense that reading is about thinking, that texts have multiple valid interpretations, and that students are capable of doing important thinking about texts. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://www.proquest.com/en-US/products/dissertations/individuals.shtml.]