Expository Case

Other approaches to text-based discussion, not included in the above, are less easy to classify and there is less research on them. These are Conversational Discussion Groups (O'Flahavan, 1989), Dialogical-Reading Thinking Lesson (Commeyras, 1993), Idea Circles (Guthrie & McCann, 1996, and Point-Counterpoint (Rogers, 1990). There are also text-based discussions that have less consistency of application, so they cannot be readily labeled. These include the general class of literature discussion groups based on reader-response theory (see Gambrell & Almasi, 1996), discussion-based envision-ments of literature (Langer, 1993, 1995, 2001), and instructional integrations of writing, reading, and talk (Nystrand, Gamoran, & Carbonaro, 2001; Sperling & Woodlief, 1997). Accountable talk is another approach to conducting intellectually stimulating discussions that, although not specifically designed for discussions about text, has applicability for promoting reading comprehension (Wolf, Crosson, & Resnick, 2004). It comprises a set of standards for productive conversation in academic contexts and forms part of the New Standards Project developed by Lauren Resnick and colleagues at the University of Pittsburgh.

Another dimension on which discussions vary is small-group versus whole-class discussions. In a 1991 study of 58 12th grade students, Sweigart found that student-led small-group discussions produced greater effects on students recall and understanding of essays they had read than did lecture or whole-class discussion. Morrow and Smith, in a 1990 study of kindergarten students who engaged in discussions of stories that were read aloud, reported similar benefits of small-group discussions compared to one-on-one discussions with the teacher or whole-class discussions. Smaller groups provided more opportunities for students to speak, interact, and exchange points of view. Taking into account all available evidence, the best generalization that can be made is that smaller groups are better but they should not be so small as to limit the diversity of ideas necessary for productive discussions (Wiencek & O'Flahavan, 1994).

Yet another dimension is teacher-led versus studentled discussions. The relative merits of these formats have been the subject of debate and some research. On the one hand, the teacher can play an important role in discussion by keeping students on topic and modeling and scaffolding the talk to enhance