This project will study scientific argumentation in middle school classrooms. The research team, led by investigators from George Washington, includes linguists, anthropologists, education researchers, and cognitive and social psychologists. Over the course of this project, they will: (1) identify argumentation episodes that naturally occur in a sixth grade science class; (2) analyze contextual factors that prompt such episodes; and (3) relate these findings to student performance on an end-of-unit science test. The investigators hope to determine how contextual factors affect the prevalence and complexity of student argumentation..
The study is based on a corpus of over 250 hours of videotape hyperlinked to 35,000 pages of transcriptions of classroom interaction collected with prior NSF funding during the implementation of an inquiry-based physical science unit called Exploring Motion and Forces. It will also have access to a related quantitative database of student achievement data. The data capture the students during group work at lab tables engaged in spontaneous arguments about the curricular content. The project should provide a methodological and empirical foundation upon which to base further studies that promise to inform science curriculum development as well as teacher training.
This project used video data from one Grade 6 enactment of the EMF physics curriculum unit from the SCALE-uP video database to follow up on a result from the classroom observation study indicating that students’ opportunity to justify their ideas was the greatest single predictor of outcome scores (O’Donnell et al., 2007). Since justification of one’s claims is a hallmark of scientific argumentation (Toulmin 2003 [1958]; Erduran et al., 2004), this project has focused on reliably identifying argumentation episodes and the contextual factors that prompted them. These analyses have set the stage for next step in this discovery process, that is, moving forward toward describing the relationship between student argumentation and conceptual understanding. More than 450 episodes of scientific argumentation were reliably identified in 45 hours of video; students engaged in argumentation for nearly three of those hours (seven percent of their total time in class). Coding results indicate that the trajectory of argumentation increases in a statistically significant way over the course of the curriculum unit. In order to understand the types of argumentation episodes, each episode was coded for the seven target ideas identified as critical to the EMF unit (Ochsendorf, 2009). Argument episodes that included these target science ideas were coded as "conceptual"; other episodes in which students argued about laboratory procedures or their observations were coded as "procedural" or "observational" as appropriate. Coding results indicate that the majority of argumentation episodes were conceptual in nature. Importantly, the percentage of episodes about key science concepts also increased significantly over time. Microanalytic methods offer exciting preliminary results about the contextual "triggers" or prompts for student argumentation. First, the written questions in the unit appear to have prompted most of students’ argumentation in the unit . Furthermore, the arguing students do when prompted by written questions is significantly longer than when otherwise prompted, and more likely to focus on the key concepts of the unit. We know, too, that certain types of questions prompt more, and more conceptually rich, argumentation than others; those that explicitly concern "evidence" play a key role in sustaining argumentation. Close analyses of the linguistic nature of the written question-prompts reveals that ITR questions more frequently nominalize key science ideas. Previous work on a different curriculum unit showed that nominalizations do pedagogical work for the unit, functioning to subtly guide students to focus on certain aspects of their laboratory experiences that are critical to understanding the unit’s key concepts (Viechnicki, 2008). Analysis of the written text of the unit reveals that the key ideas in the unit motion and force are represented as nominalizations in nearly 25% of the written questions put to students, and are statistically more likely to appear in the ITR questions that ask students to reflect on their laboratory experiences, and to understand those experiences in terms of the effect of forces on motion – the key concept of the unit. In addition to text analysis of the written prompts, microgenic techniques applied to the student oral argumentation episodes themselves reveal that claims and subsequent argumentation episodes tend to be launched when students engage in concerted joint attention, a state in which the group is focused together on a common activity. We have isolated cues in the context that trigger argumentation episodes. For example, argumentation is more likely to proceed when students read something out loud to each other, when they shift linguistic registers (using non-colloquial, "scientific" language), and when they orient their bodies in relation to each other, creating a shared physical space. The longest and most socially complex episodes tend to coincide with a register shift, in which one member calls the others to attention by changing tone or speaking style, for example, by reading aloud from the worksheet. Indeed, the act of reading a question out loud is 100 percent predictive of an argumentation episode: if a student read a written question out loud, every single time, an argumentation episode ensued. The local classroom context – how a given class day proceeded, and what activities the students were engaged in at any given time – was also critical in terms of triggering argumentation. "Class period segment" data (which relate to the ways in which the teacher allotted time in the classroom to different learning activities) from a prior-NSF supported research project were brought to bear on these argumentation data (Kuipers, Viechnicki, Massoud, & Wright, 2009), revealing that most argumentation takes place during the "Reflection" time of the class. Coding of the different types of scientific activities in which students engage in a given class day (graphing, observing, recording data, and so on) suggests that after answering written questions, the classroom activity that triggers most argumentation episodes is "observing" during laboratory activities.
