Read Microsoft Word - studentroles_C&I.doc text version

Students' roles in group work


Running head: Students' roles in group-work

Students' roles in group-work with visual data: A site of science learning

Josh Radinsky Learning Sciences Research Institute, University of Illinois at Chicago Submitted to Cognition and Instruction May 2, 2007. Revised December 30, 2007


Josh Radinsky, Assistant Professor of the Learning Sciences Learning Sciences Research Institute and College of Education University of Illinois at Chicago 1040 W. Harrison St. M/C 147, Chicago, IL 60607 email: [email protected] fax: 312-355-3930

Acknowledgments The author expresses his appreciation to Carlos Rodriguez, Kimberly Alamar, José Trigueros and Jennifer Mundt Leimberer for their collaboration and insights; the students who participated in this research for their patience; Isabel Arias, Sonia Oliva and Luis Sanchez for their tireless work on data collection; Susan Goldman and Melissa Singer for their collaboration in this research and feedback on this paper; and Noel Enyedy for his valuable insights and guidance. This material is based upon work supported by the National Science Foundation under Grant No. 0337598. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author and do not necessarily reflect the views of the National Science Foundation.

Students' roles in group work


ABSTRACT Learning science includes learning to argue with inscriptions: images used to symbolize information about the world and communicate persuasively. The present study examined 6thgrade students learning to invest inscriptions with representational status, such that images became symbols for observing and explaining phenomena in the (represented) world, in the context of a plate tectonics investigation using a geographic information system (GIS). The process of learning to reason with inscriptions was studied in the emergent patterns of participation in small groups, operationalized as roles within group practices. Cross-case comparisons identified ways that two such roles (competitive challenger and quiet bystander) developed along different trajectories for students in different groups. Role development mediated each student's learning of skills and habits of scientific reasoning with inscriptions, including (1) co-assembling a shared "representational state" of the data, and (2) managing the dialectical tensions of argumentation. Rather than treat this mediation as a causal mechanism by which social processes impacted individual learning, role was operationalized as a site of science learning at the intersection of individual and collective processes. Current conceptions of scientific practices as distributed across individuals, communities, and artifacts suggest the value of such a focus for understanding how students learn to do science.

Students' roles in group work


INTRODUCTION Learning to reason with inscriptions Learning to reason with visual information is a central part of learning science. In particular, scientific reasoning includes a facility at communicating with inscriptions: images that are used to symbolize information about the world, with which to persuasively communicate an argument (Latour, 1990; Lehrer & Schauble, 2002; Sfard & McClain, 2002). Inscriptions such as maps, tables, graphs, equations, and diagrams are essential components of the discourse of scientists and science educators, permeating research journals, textbooks, laboratories, and classrooms (Latour & Woolgar, 1986; Latour, 1990; Lemke, 1998). Such visual displays are "more than a simple matter of supplying pictorial illustrations for scientific texts," but rather are "essential to how scientific objects and orderly relationships are revealed and made analyzable" (Lynch, 1990:153). We commonly think of these types of inscriptions as representing aspects of the real world. However, the status of representation does not reside within a material artifact such as a printed map or graph, but rather is a feature of how that artifact is used in practice, i.e. to afford analysis and communication (Goodwin, 2000; Hall, 1999; Hutchins, 1995; Lehrer & Schauble, 2002; Sfard & McClain, 2002). The meanings of inscriptions are not simply facts there to be extracted: they cannot be understood "in isolation from the processes of interaction and work practices through which they are made relevant and meaningful" (Goodwin, 2000:165). People need to learn how to "see" such data images in domain-relevant ways (Hall, 1999; Lynch, 1990). In this sense, the representational status of inscriptions is best conceived neither as self-contained within the artifact, nor as a cognitive product of the mind of the reader or writer, but rather as a feature of the human activity in which it is used.

Students' roles in group work


This study: Investigating plate tectonics with GIS. The present study examines processes by which middle-school students learned to invest inscriptions with representational status, such that they became symbols with which the students could observe and communicate about phenomena in the (represented) world. The context was a 6th grade science unit on plate tectonics called Earth Structures and Processes (Radinsky, Alamar, Leimberer, Rodriguez & Trigueros, 2005; Radinsky, Leimberer & Gomez, 2000), in which the students used a geographic information system (GIS) as one of the tools of their investigation. A GIS is a computer program consisting of a database connected to an interactive map (Audet & Ludwig, 2000; National Research Council, 2006). Information in the database can be viewed on the map as objects of different colors, shapes, and sizes. The GIS is interactive, which means that users can zoom the map in and out, turn layers of data on and off, choose how the information should be displayed, and submit queries to the database to answer specific questions. Increasing availability of such student-accessible data visualization tools, along with widespread access to Internet-based resources in schools, have made the use of rich visual data in classrooms more common (deJong & vanJoolingen, 1998; vanJoolingen, deJong & Dimitrakopoulou, 2007). Inquiry projects in which students make use of these resources are highly valued as opportunities for deep learning about the nature of science, as well as component skills and dispositions for inquiry, such as coordinating multiple goals and activities; making and communicating observations; coordinating observations with explanations of phenomena; and negotiating differences of interpretation within groups (Duschl, Schweingruber & Shouse, 2007; Kuhn, 1991; Lehrer & Schauble, 2002, 2006; Loh, Reiser, Radinsky, Edelson, Gomez & Marshall, 2001; National Research Council, 2006).

Students' roles in group work


Data visualization tools like GIS provide a valuable context for studying the development of scientific reasoning with inscriptions. This is because the dynamic nature of these tools places students in a position in which they are continually both reading and generating new inscriptions. Even the simplest action with a visualization tool, such as zooming in for a closer view of a GIS map, produces a new inscription ­ i.e., a map representing a different geographic area than the previous view, containing a different set of data, and affording different observations. As students manipulate the GIS map over time, they continually create new inscriptions as they go, engaging in activity that is inherently inter-textual and that requires verbal and gestural coordination among group members. These processes of shared sense-making can be analyzed to gain insights into how students' understandings and participation change over time. How do students learn to do this? Representation as practice. If the representational status of visual data is better conceived as a function of sensemaking practices rather than as qualities of the artifacts themselves, as described above, then the learning process is one of developing competence at engaging in these practices. Hall (1999) emphasizes the active nature of representational practices: rather than learning to see what is manifestly "there" in a data image, people jointly engage in "assembling [a] representational state" (p. 201) of an inscription. This is an active process of co-constructing the meanings and purposes of inscriptions, which do not simply "reveal" facts about the world, but rather simultaneously obscure many aspects of the represented world while making others visible. Latour (1990) thus describes inscriptions as "a small window through which [scientists can] read a very few signs from a rather poor repertoire (diagrams, blots, bands, columns)" (p. 22). Yet this "small window," used in the context of a discipline's "professional vision" (Goodwin, 1994), enables a view of phenomena and relationships that would otherwise be inaccessible.

Students' roles in group work


Sense-making with inscriptions thus involves making the "absences" (i.e., the things obscured by the "blots and bands") "present" in discourse (Hall, 1999:194). Even among scientists this process involves negotiation, requiring participants to manage a great deal of ambiguity and uncertainty (Dunbar, 1995; Dunbar & Fugelsang, 2002; Hall, 1999; Latour & Woolgar, 1986). This uncertainty is "endemic in ... scientific practice" (Hall, 1999:199-200), and is still greater among students who are unfamiliar with the domain. Managing this uncertainty is accomplished through discourse in which meanings and goals are co-constructed, contested and negotiated, as a range of communicative resources are brought to bear in assembling a representational state for reasoning about the world. Students need to learn to employ appropriate resources in these conversations with data, in order to make the "absences" of a GIS map (e.g., the earth's crust, an ocean trench, a tectonic plate, an earthquake) present in their conversations. This sense-making process must be both receptive and productive (Enyedy, 2005; Lehrer & Schauble, 2000; Smagorinsky, 2001; Wertsch, 2000): the goal is for students not only to read inscriptions and decode the information that they are meant to represent, but also for students to actively construct them in group discourse as representations of the phenomena they are studying. Small-group inquiry is a valuable context for examining the ways students manage this negotiation of meaning (Radinsky, 2000; Cornelius & Herrenkohl, 2004; Richmond & Striley, 1996). Working in small groups, students have many chances to try to coordinate multiple perspectives (Kuhn, Shaw & Felton,1997; Linn & Hsi, 2000). Unlike whole-class discourse, the relatively more symmetrical power relationships among peers result in discourse that is more fluid, and more easily shaped by the contributions of each member. In the absence of direct interactions with the teacher, the sense-making practices for understanding inscriptions like GIS

Students' roles in group work


images are continually improvised, challenged, and negotiated. For these reasons the present analysis examines learning to reason with visual data in small groups. Learning disciplinary norms and practices of "seeing" data Conceiving of these learning objectives as classroom practices highlights the fact that the site of learning is an interactive one. Cobb (2002) emphasizes that practices consist of normative modes of engagement in collective activity, and can only be observed as "regularities in the ongoing interactions of members of a community" (p. 190). He defines a classroom practice as consisting of three interrelated types of norms: (1) normative purposes for engaging in domain activity; (2) normative standards of argumentation in classroom talk; and (3) normative ways of reasoning with tools and inscriptions (Bowers, Cobb & McClain, 1999). Learning to engage in such a practice, then, involves appropriating these purposes, argumentative standards, and modes of employing inscriptions, as part of participating in the group. Representational practices develop within the goals and activities of a community. As participants co-construct the meanings of visual information, these meanings become part of the larger social negotiation of activity, including emergent goals, collective practices, and evolving relationships. Thus the ways students communicate about visual data involve not only descriptions of what they see, but also social positioning in their relationships with groupmates, other classmates, the class as a whole, and the teacher. When new cultural tools (like GIS images) are introduced into an activity system (such as a classroom or a small group), new patterns of interaction may emerge, with the potential to transform the group's practices and relationships among participants (Cornelius & Herrenkohl, 2004; Wertsch, 1998). The process of constructing the meaning of what is seen together is active, negotiated, and often contested. In this interactive context, multiple perspectives come

Students' roles in group work


together, each bringing some agency to bear on the negotiation. The tension among these perspectives can become a "lever for conceptual change" (Greeno & van de Sande, 2007:15), problematizing the meanings of concepts and data (Radinsky, 2000; Engle & Conant, 2002) and requiring negotiation among competing meanings. Cornelius and Herrenkohl (2004) suggest that students' ownership or authorship of a particular perspective or understanding can be transformed in the negotiation of relationships in small groups. This process can impact students' "conceptual agency" (Greeno, 2006) for constructing domain understandings. This view underscores the fact that becoming adept at reasoning with inscriptions is closely connected to becoming skilled at argumentation: not just representing information as fact, but using it as a tool with which to negotiate understandings with others.1 Argumentation is a discursive form that is widely viewed as a key aspect of knowing, doing, and learning science (Andriessen, 2006; Bell, 2002; Dunbar, 1995; Latour, 1990; National Research Council, 1996). However, the structured model of scientific argumentation proposed by Toulmin (1969) has been shown to have limitations as an analytical tool for examining the emergent negotiations of meaning that occur in classrooms (Cobb, 2002; Leitao, 2000). Leitao (2000) and others have proposed adapted schemes for understanding the ways conflicting ideas are negotiated in everyday discourse, emphasizing the dialectical balance of opposition and agreement, rather than the structural model requiring claims, warrants, backing and data. The value of argumentation for learning lies not primarily in a competition to convince the other of one's point of view, but rather in its affordances for engaging in "cooperative explorations of a dialogical space of


This emphasis on persuasive argumentation resonates with Latour's (1990) sociohistorical analysis of scientific practices. According to Latour, the power of inscriptions lies not simply in their efficiency at graphically displaying quantities and relationships, but in the social affordances of these displays for mobilizing resources for particular purposes: "If you want to understand what draws things together, then look at what draws things together" (p. 60).

Students' roles in group work


solutions" (Andriessen, 2006:445). In so doing, several learning mechanisms are called into play: knowledge is made explicit, new knowledge is co-elaborated, and interpretations of data are employed as evidence in relation to claims and positions (Baker, 2004, cited in Andriessen, 2006; White & Frederiksen, 1998). The coordination of disagreement and agreement, through processes of argument, counter-argument, and reply, can provide the leverage to effect conceptual change (Andriessen, 2005, 2006; Leitao, 2000). For middle-grades students, however, these tensions often are often difficult to manage. There are dangers of tipping too far into a friendly "comfort zone" in which productive disagreement disappears, or too far into a "confrontation zone" in which disagreements become interpersonally hostile and negotiation of understanding ends (Radinsky, 2000; Radinsky, Goldman & Singer, in press; Cornelius & Herrenkohl, 2004; Leitao, 2000). In short, in order to learn to reason with visual data, students must learn to engage productively in argumentation. Where does this learning happen? Individuals' roles in collective practices This study examines the learning processes of students within small groups as they engaged in a scientific inquiry project using a GIS data visualization tool. The focus is on individual development with respect to emergent practices of the group. A methodological challenge is that these individual and social processes cannot be viewed in isolation from one another, as individual accomplishments are contextualized within the larger collective practices in which they develop (Cobb, 2002; Saxe, 2002), and vice versa. Enyedy (2005) frames the methodological challenge for researchers in terms of the agency of individuals vs. that of groups: to understand the accomplishment of the individuals within this ... classroom, we must address the ways that individual agency stands in

Students' roles in group work


relation to participation in a community. Further, we must address the ways that the development of the community stands in relation to the actions and contributions of individuals. (p. 459) This examination of the interface between individual and collective development has emerged as a priority in learning sciences research in recent years (Greeno, 2006), as the field has increasingly combined cognitive and socio-cultural research lenses in efforts to better represent the co-constitutive nature of individual and socially-shared learning (e.g. Cobb, 2002; Enyedy, 2003, 2005; Lehrer & Pritchard, 2002; Tabak, 2004). Where should we locate the learning of students engaging in emergent social processes, such as the co-construction of the meanings of visual data? Constructs of learning in terms of abilities, dispositions and understandings locate what is learned within the individual, but do not characterize that individual's emergent positioning within a social context. Bowers, Cobb & McClain (1999) use the construct of practices as a focal point from which to examine the reflexive relationship between conceptual understandings and abilities at the individual level, and the emergence of norms at the classroom level. Cognitive anthropologists have reconceptualized learning as the development of an identity of mastery within a community of practice (Lave & Wenger, 1991; Wenger, 1998). Sociocultural theorists have recently begun to develop conceptual frameworks for examining the discursive identities as sites of development (Brown, 2004; Gee, 2002; New London Group, 1996), including studying the ways students improvise and negotiate among multiple and often conflicting identities in the process of learning to engage in scientific discourse (Brown, 2004). The present study attempts to synthesize the construct of emergent identities with that of normative practices as lenses for studying everyday classroom talk and activity, by documenting

Students' roles in group work


students' development of roles during small-group, independent inquiry work with visual data. The construct of role is operationalized here as a normative mode of engagement within in a particular activity system ­ an emergent mode of participating that shifts as students adapt to one another, and to the norms that the group develops. This is distinct from other uses of the term "role" that suggest more permanent social identities (e.g. role of parent, student, employee, etc). It is also distinct from the use of the term employed in the cooperative learning literature (Johnson & Johnson, 1982; Kagan, 1992) and related research (Palincsar & Brown, 1984), in which temporary roles are assigned by the teacher to embody particular divisions of labor in group-work (e.g. role of timekeeper, questioner, etc). Instead, the roles identified in the present study are characterizations of the observed patterns of participation by which students construct, with their group-mates, a shared positioning of themselves as a "certain kind of person" (Gee, 2002) in a particular activity system (here, independent small-group work with data). For example, a student may play a role of "Answer Man" in a group (Radinsky, 2000), characterized by norms such as telling groupmates how to answer questions prior to discussion; speaking loudly and authoritatively (in register and cadence) in response to questions or observations of groupmates; and evidencing absolutist warrants for claims (Cornelus & Herrenkohl, 2004). Playing this role in a particular group may emerge from, and contribute to, the student's longer-term, developing identity as a student. However, this student's "Answer Man" role in his small group may shift dramatically over days, or even minutes, depending on the other members of the group (e.g. through negotiations with a more authoritative groupmate), the ways instruction is organized (e.g. as a result of the teacher requiring that each group member be prepared to explain [Kagan, 1992]), or events outside of the group or classroom (e.g. the student's joining a group elsewhere in which

Students' roles in group work


he is positioned as a novice). The role is dynamic, emergent, and is a site of change at the intersection of the individual's identities and the group's norms. Thus the role of "Answer Man," while mediating the ways this student participates in learning and doing science, is also a malleable construct that can be changed through experience fairly quickly, including being a tractable site of instruction as a teacher may challenge or reinforce an emergent role. Clearly such a role has some affordances for engagement in science learning, while also having significant constraints for that learning. As the student adapts his "Answer Man" role in the group, there is the potential to develop (more locally and immediately) new practices and understandings, and (more globally and long-term) revised identities in larger communities, such as brother, scientist, African American, friend, etc (Gee, 2002). In this sense the construct of role is an important site of learning. In the analysis that follows, two roles ­ competitive challenger and quiet bystander ­ were identified in two different groups, and tracked as they changed over the course of an extended investigation. Distinct developmental trajectories were documented for students enacting each role in their respective groups. The analysis identifies the ways role development mediated students' learning of skills and habits of scientific reasoning with inscriptions. The construction of role as a site of learning, rather than a causal mechanism to explain learning in another site, is explored in the Discussion that follows. METHODS Setting and participants This study was conducted in an urban public elementary school, a racially and economically diverse "magnet" school (i.e., drawing students from a range of neighborhoods) that is non-selective in admissions, resulting in academic diversity as well. Participants were

Students' roles in group work


students in two 6th-grade math-science classrooms taught by the same teacher. All students in the teacher's two morning classes were invited to be participants in the study, of whom approximately half returned student assent and parental consent forms. The teacher had 14 years of classroom experience, and was widely respected as an excellent and motivating science teacher: a year after the completion of this study he was recruited by the district to become a "science facilitator," conducting professional development for other teachers throughout the district. The teacher and the researcher have a long-standing collaborative relationship. The two groups selected for analysis in this study were chosen in consultation with the teacher to include students representing a range of academic performance in science (neither the highest- nor lowest-achieving). Curriculum and instructional context The study focused on the learning that occurred during an enactment of Earth Structures and Processes, a middle-school earth science inquiry curriculum unit (Radinsky, Alamar, Leimberer et al, 2005). A central part of the unit involves using GIS software (MyWorld, to investigate three kinds of data: surface elevation, earthquake locations (by date and magnitude) and volcano locations. The goal of the unit is to develop a set of key concepts for understanding plate tectonics, and the skills and habits required to apply those concepts to interpreting large, complex sets of data. In groups of three, students were assigned an "earth structure," one of 15 locations on earth, each presenting a particular configuration of data patterns that students must investigate and explain. This unit was part of the school's regular 6th grade science curriculum, and the teacher had taught it in previous years. The unit lasted 19 instructional days, and included introductory lessons on the earth's crust, the ocean floor, and earthquakes; concept-building activities including reading and

Students' roles in group work


reporting on informational texts, and creating 3D topographic models of their earth structures; and skill-building lessons on using the GIS software and identifying plate boundary zones. The final phase of the unit, which is the focus of the present study, was an independent small-group investigation using the GIS, during which students were assigned the task of identifying the boundaries of the tectonic plate on which their structure sits, and using their understanding of plate boundaries to hypothesize which direction that plate is moving. To complete this task students had to use earthquake data to trace a boundary, and use patterns of elevation, earthquake, and volcano data to infer what kinds of motion were occurring at any given boundary. Three types of plate boundary zones, and the related plate motions and data patterns, were taught in mini-lessons during this phase: subduction zones (where two plates converge, with one plate sliding under another); rift zones (where two plates diverge from a common boundary); and buckling zones (where two plates converge and buckle upwards). The teacher's instructional objective was not for students to necessarily "discover" the correct boundary zones or plate motions, but rather to use data as evidence to support whatever interpretation their group put forward. The five sessions of small-group work with the GIS began with a simplified data map, including only elevation data on the first two days (scaffolded to facilitate learning the GIS tools, and to generate the need for more data), and then introducing the full set of data layers (adding three ranges of earthquake data by magnitude, as well as volcanoes) for Days 3-5 of the investigation. During small-group sessions the students worked independently, while the teacher moved around the room meeting with one group at a time to assess their progress and help them move forward (see Goldman, Radinsky & Rodriguez, 2007, for an analysis of these meetings). Some

Students' roles in group work


of these meetings were "check-in" visits of less than one minute; others were extended (6 to 14 minute) meetings in which he pushed them to articulate their understandings of the concepts and data, clarify differences of interpretation between group members, reflect on how best to support their ideas, and prepare to explain their findings to the class. There were two such extended meetings with each group over the five days' investigation. These teacher meetings with each group are excluded from the present analysis, as they constitute a quite different activity system than that of small-group work without the teacher (e.g., very different distributions of talk, participation in science discourse, emergent goals, etc). Also, the impact of the teacher's instruction on role development is backgrounded in the present analysis. This is not to say that his instruction did not strongly mediate the kinds of development described here, but rather to focus on the developmental process that occurred within the small-group activity system independently of evaluating the multiple influences (inside and outside the classroom) impacting that development. Those analyses are important and relevant, but outside the scope of the present investigation of role development as a site of science learning. Classroom data gathered Two researchers were present in every class session, one operating a video camera and the other taking field notes. During small-group work the video camera followed the target group of three students; in all other situations it followed the teacher and/or the flow of whole-class interaction. Copies of consenting students' work were also recorded electronically. Periodic reflection meetings were held with the teacher, some of which were recorded and transcribed, as a source of triangulation for inferences made about patterns of student participation and learning.

Students' roles in group work


The unit lasted for 19 instructional days, each of which included multiple episodes in distinct classroom activity structures (i.e., whole-class discussion, mini-lesson or lecture, smallgroup work, individual work, or student presentations). Using field notes and video, 40 such episodes were identified across the 19 days. Of these 40 episodes, 13 were small-group work, of which the present analysis focuses on the final five, which were the GIS investigation phase of the project, from days 12 to 17. During these five class sessions the groups worked with the GIS for a total of about 140 minutes together (with slight differences between classes), excluding the meetings with the teacher. Transcription, coding, and reliability The five focal episodes were transcribed and time-coded from videotape in their entirety, including an accuracy check by a second researcher for each transcript. Corrections to the transcript were discussed between transcribers and resolved, when necessary through repeated viewings of the video. Each session transcript included all student talk for the continuous time allocated to small-group work, regardless of the topic of discussion. Discourse that involved any participant other than the three members of the target group (i.e. the teacher, a researcher, or another student) was excluded from the analysis, and marked as a break in the transcript. The verbal unit of analysis was the turn in group talk, defined as an uninterrupted speech burst by an individual student with no pause greater than two seconds. Overlapping speech by two students was treated as two separate turns, with the overlap indicated in the transcript. Each turn was coded for speaker and time (decimal minute derived from time codes in the transcripts). To document what was problematized in discourse by the group and by individual students, all turns were coded for presence or absence of particular categories of speech referents relevant to

Students' roles in group work


reasoning with data images. Code definitions and total applications of each code are included in Table 1 (code categories are not mutually exclusive). Table 1. Discourse referent category codes. Speech Description of code category referent code Data Speech referring to data items or relationships Referent among data items, whether described in the visual language of the inscription (e.g., "dots," "a big line"), or in indeterminate indexical terms (e.g., "look at that one right there," "they're close together") Real-world Speech referring to data items in terms of the Referent inferred real-world referents of the data (e.g. "This one's in the ocean," "It's next to Africa") Domain Speech including any of a list of domain concept Concept words included in science instruction (plate, Referent subduction, buckling, rift, transform, volcano, earthquake, island, magma/lava, elevation/depth, or trench), in English or Spanish, or explanations or descriptions of these concepts using other words Out-ofSpeech referring to people, objects, or events school outside of the classroom, usually (though not Referent necessarily) unrelated to the science classwork. # (%) turns coded (n=3,108) 944 (30.4%)

281 (9.4%) 401 (12.9%)

329 (10.6%)

The coding process was interpretive, requiring the researchers to make inferences about referents of speech bursts that were often ambiguous, especially when considered in isolation. This interpretive process required consideration of students' patterns of participation across multiple turns and even across episodes, as well as consideration of non-verbal actions, especially points, traces, and representational gestures accompanying speech (Singer, Radinsky & Goldman, in press), to determine the referent category. Once the referent category codes were established, a researcher not previously familiar with the transcripts was trained on the coding scheme using two transcripts. After training, two researchers coded three full transcripts (a total of 932 turns, constituting 30.0% of the data corpus) independently, coding each turn for presence

Students' roles in group work


or absence of each of the referent codes listed in Table 1. Cohen's Kappa was used to calculate rater agreement separately for each referent code (Data Referent agreement was 0.7509; Realworld Referent, 0.7222; Concept Referent, 0.7289; Out-of-school Referent, 0.7468). The two researchers then met to resolve differences for each utterance not coded identically. All differences were resolved in these meetings, often through reviewing of the video. One researcher coded the remaining transcripts, and the full set of coded transcripts was used as a primary data source for the case study analyses. OVERVIEW OF THE GROUPS AND THEIR INVESTIGATIONS An initial overview describes the members of each group, the "earth structure" they were assigned to investigate, and the topics that were contested during their small-group GIS work. This overview is followed by comparative case studies (Yin, 2002) for each of two roles, described below in the Results section. CJV group: Cecilia, Juanita and Violeta study Mt. Etna's plate One group consisted of three girls, Cecilia, Juanita, and Violeta, who worked together to investigate the area around Mt. Etna, a volcanic mountain in Italy. This region included southern Europe, the Mediterranean Sea, the northern coast of Africa, and parts of the Middle East and the east Atlantic Ocean (see Figure 1). The elevation patterns in this area rise and fall, but without the kinds of clear, dramatic elevation changes found in other areas (e.g., the steep ascent of the Himalayas, or the deep ocean trenches along the eastern coasts of Asia). There is a great deal of earthquake activity in this region, represented in Figure 1 by small black dots (one for each earthquake location of magnitude 2.0 or greater, recorded in a recent 3-year period2). These dots


This representation of earthquakes is simplified from that used by students in the GIS, in order to improve the print visibility of the figure. In MyWorld the students used three layers of earthquake data sets, divided by ranges of

Students' roles in group work


form a broad, messy band extending across much of southern Europe into the Middle East, and volcanic activity (not shown in Figure 1) is scattered throughout this band. This part of the world is considered by scientists to be a buckling zone, in which two continental plates (African and Eurasian) collide to push up mountains (the Alps). However, unlike other buckling zones such as the Himalayas, the presence of a sea (the Mediterranean) in the center of the plate boundary region can make it difficult to visualize as a meeting of two continental land masses.

Figure 1. GIS map of earthquakes in the Mt. Etna region. Topics that were contested in CJV's argumentation: "connecting" and "how big is a plate." The argumentative discourse involving visual data in this group across the five days developed around two central themes related to plate boundaries and data patterns. The first was the question of whether, and where, elements of plates were "connecting," either in the realworld referents of the data, or in the visible data patterns themselves. This term ("connect") was introduced on Day 1 (Excerpt 2 below), and became the center of the group's intense periods of argumentation in Days 4 and 5 (Excerpts 4-6 below). As such, the analysis focuses in part on

magnitude, with some differentiation of magnitudes by dot size and color. Volcano locations (all known, active or dormant) were represented by uniformly-sized red dots.

Students' roles in group work


how students engaged in discourse about "connects" or "can't connect," and the resources employed by each student in discourse related to this ongoing argument. The second, closelyrelated strand of argumentation in the group focused on debating the appropriate size of a plate ­ whether a particular proposed plate boundary would result in a plate that was too big or too small. In the analyses of emergent roles for this group these two ongoing arguments ­ "can't connect" and "too big to be a plate" ­ constitute the conceptual and procedural issues that were in dispute. EKL group: Eliana, Kerry and Leo study Mt. Fuji's plate The second group included two girls, Eliana and Kerry, and one boy, Leo, who were assigned the region surrounding Mt. Fuji, a volcanic mountain in Japan. Though both groups were assigned crustal formations around a volcano, the data patterns surrounding Mt. Fuji (see Figure 2) are markedly different than those surrounding Mt. Etna. Japan sits at the eastern edge of the Eurasian continental plate, and on the precipice of a steep ocean trench plunging to some of the deepest depths found anywhere on earth. It sits within a clear line of earthquake and volcanic activity (again represented by one black dot for each recent earthquake, as in Figure 1) extending along the edge of this trench, which is part of the "Ring of Fire" pattern of volcanoes surrounding most of the Pacific Ocean. Japan is considered to be located on a subduction zone, at which the oceanic plate of the Pacific Ocean slides underneath the continental plate of Eurasia, creating the pattern of volcanic islands and mountains along one side of the boundary, and a deep ocean trench along the other. Subduction zones are identified by a distinctive pattern of parallel lines of earthquakes, volcanoes, and an ocean trench running along the plate boundary. As it happens, both groups' earth structures (Mt. Etna and Mt. Fuji) are understood by scientists to be located on the same (large Eurasian) plate.

Students' roles in group work


Figure 2. GIS map of earthquakes in the Mt. Fuji region.

Topics that were contested in EKL's argumentation: how plate motion forms a trench. Argumentation in this group focused primarily on a set of contested issues about plate motion at boundary zones, particularly in the form of debate over what kind of plate motion might account for the deep trench near Japan. The trench had become a focus of discussion at the end of the previous, topo-model phase of the unit. The argumentation in the GIS phase employed the new set of boundary-zone concepts introduced in whole class instruction (subduction, rift and buckling), both in speech and in gesture (Singer, Radinsky & Goldman, in press), making it an argument about whether the trench might have been formed by subduction, rift, or buckling. Thus the contested conceptual issues in this group were (1) what happens at different types of plate boundaries, and (2) how trenches are formed. Students' participation in these debates constitutes the discursive terrain in which role development is analyzed for this group, focusing on the resources and approaches employed by students over time in these evolving arguments. Elsewhere (Goldman, Sacay, Singer, Oliva, Allende-Pellot & Radinsky, 2006, as well as analyses in progress) we attend to the science learning that occurred at the conceptual and skill

Students' roles in group work


levels for individual students in these enactments of the Earth Structures and Processes unit, as well as to the teacher's instructional scaffolds with the small groups during the teacher meetings (Goldman, Radinsky & Rodriguez, 2007). For the present analysis, though, the focus is on students' role development as a site of science learning, not as a mechanism for promoting or explaining conceptual or skill-building learning. For this reason the analysis presented here examines the ways students' modes of engaging in argumentation with visual data developed within the small-group context over the course of the investigation. RESULTS The nested case studies in this section examine two different roles students enacted in their respective groups, providing examples of each role from the two groups studied, and then tracing how affordances and constraints of each role manifested themselves differently and contributed to different developmental trajectories for each student's emergent engagement in argumentation with data images. The two roles examined here are those of competitive challenger (Cases 1 and 2), and quiet bystander (Cases 3 and 4). Each of these roles is taken up in turn, presenting nested cases of two students within each role. We first describe each case individually, supporting characterizations of each student's role with discourse and gesture data. We then return to comparative cross-case analyses (Yin, 2002) in the Discussion, addressing how each role developed differently in each case, and illustrating how these developmental trajectories represent sites of science learning. Cases 1 and 2: Competitive challengers It is not uncommon in small groups for one student to assume an authoritative and dominant stance with respect to the completion of the academic task, the science discourse, or both (Radinsky, 2000; Cornelius & Herrenkohl, 2004). When one student positions herself in

Students' roles in group work


this way the others often find themselves responding to her, and constructing their own roles in relation to this bid for authority. One such mode of response that emerged in our data was the role of competitive challenger. This role is operationalized here as normative participation in the group characterized by: · · · · Positioning in opposition to a peer who stakes a claim to authority Resisting being positioned as subservient in group decisions Engaging in argumentation as a mode of competition Coordinating friendly interactions with confrontational ones

Two students who assumed this role are examined in Cases 1 and 2 below: Violeta in the CJV group, and Eliana in the EKL group. Case 1: Violeta as a competitive challenger to Cecilia's authority. Two group members, Cecilia and Violeta, emerged as a co-leaders in quantity of talk (measured in turns), on each day (see Figure 3) and for the project as a whole (see Table 2). In fact their total number of turns during times that both were present was nearly identical (463 and 462 respectively; see Table 2). This co-leadership developed through a process of Violeta challenging Cecilia's bids for authority, engaging in a friendly but persistent competition over decision-making and data interpretations. Table 2. Participation in group talk (CJV Group only 4 sessions: Violeta absent Day 2.) CJV Group Cecilia Juanita Violeta EKL Group Eliana Kerry Leo Total turns (4 sessions) 463 95 462 Total turns (5 sessions) 700 357 867 Turns per minute 7.1 1.5 7.1 Turns per minute 6.1 3.1 7.5 % of turns 45% 9% 45% % of turns 36% 19% 45%

Students' roles in group work


Figure 3. Shares of group talk by day in the CJV group.

Positioning in opposition to a peer who stakes a claim to authority. Excerpts 1 and 2 illustrate how Cecilia's discourse early on had the potential to establish her as an authoritative decision-maker in the group. On Day 1, as they began working with the GIS for the first time, it was Violeta's assigned turn to work the laptop computer shared by the group, but Cecilia tried to assert control over Violeta's actions. Excerpt 1. Cecilia bids for control (Day 1)

2. C: 3. V: 4. C: 5. C: 6. V: 7. C: 8. V: (to Violeta) OK, go go go go (...) Go more over there! Click twice. (I did ...) There. happened? Doop. 9. C: (A little more.) More. More. What Doop. I think Doop. Aah!

((points left))

((map view shifts off screen)) What was that?

I think I (did it) too much! Doop. Doop.




((sound accompanies clicks to move the map view))


you just press really (...) (Hold it Violeta)

((C reaches over and clicks the mouse


Students' roles in group work


This excerpt shows Cecilia exerting her authority over Violeta's actions at the computer (L. 2, 4, 5, 7, 9), and even physically reaching in to take over the task of clicking. Cecilia's authoritative stance also extended to talk in the science domain, as seen in Excerpt 2: Excerpt 2. Cecilia speaks with domain authority (Day 1)

88. C: 89. V: 90. C: 91. V: 92. C: 93. V: 94. C: No, what I was saying, ((waving hand around screen)) cause everything was supposed to be connected before -- go back. (What) Go back! ((pointing toward north on the map)) Oh, to, um You know that story, that everything used to be connected? one big piece? Nuuhh Well then, everything is. And everything is moving to

Here Cecilia's command to "go back!" (L. 88, 90) is linked to her articulation of a domain explanation ("that story", L. 92) that "everything used to be connected ... to one big piece ... [a]nd everything is moving" (L. 92, 94). This appears to be a reference to Pangaea, and the idea that all the continents were once together, but have moved over time ­ a relevant reflection on prior domain understandings. From this beginning, it is easy to imagine Cecilia taking control of the task, the materials, and the authority for science knowledge. However Violeta used both playful and confrontational discourse to challenge Cecilia and assert her own authority, as in Excerpt 3: Excerpt 3. Violeta argues back (Day 1)

66. C: ((to V)) (...) you can't see the plate. ((traces)) But you could see that it's going apart.

((points)) It's breaking.




67. V: 68. C: 69. V: Um, Ceci, that's the bottom. north California. California It's Baja California. I control the mouse today. Yeah. (Oh good for you.)

((points)) Of California.

(see look).

See look, that's north California, (Baja)

Students' roles in group work


70. C: 71. V:

This isn't the mouse ((points to trackpad)) So what, I control the computer today ((pushes C's hands


Here Violeta began by gently interjecting (L. 67), then correcting Cecilia's pronunciation of "Baja" (L. 69), and finally explicitly laying her claim to "control" of the mouse and computer (L. 69, 71), physically pushing Cecilia away. Resisting being positioned as subservient in group decisions. After being absent on Day 2, Violeta asserted her authority in the group again on Day 3, challenging ideas put forward by Cecilia and tentatively supported by Juanita (Excerpt 4). The argument was about where to draw the boundary line between plates in their region. Figure 4 shows the map in the group's science folder on which several possible plates were sketched during this session; in Excerpt 4 Cecilia began by proposing the plate boundary marked "A," ("around Africa," L. 171, 177), to which Violeta objects (L. 178). The discussion went on to raise the possibility that the plate boundary might extend all the way around Eurasia (L. 187)3.

Figure 4. Plate boundary ideas drawn by the CJV group.


The scientifically-accepted boundary of Mt. Etna's plate does include all of Eurasia, a possibility that both Cecilia and Juanita pondered at various points in time, though never with confidence. The plate boundary favored by Violeta, marked "B" in Figure 4, is actually an outline of a part of the boundary (a wide swath of earthquakes), rather than a tracing of a boundary around a plate.

Students' roles in group work


Excerpt 4. Beginning of the "huge plate" argument (Day 3)

170. C: Yeah OK Violeta, we need to like do that whole big one, around Africa and that whole one ((tracing on screen, as V begins to

draw in folder))

171. C: No, it's around Africa ((circling on screen)) 172. V: You want me to do this whole thing? 173. C: No, just that one right there ((tracing)) 174. V: This one? 175. C: Violeta. Look. 176. V: Well that's not around the (...) 177. J: It's around Africa and our region ((tracing)) 178. V: No 179. J: Look, it stops right here ((points)) 180. C: Look, there's some more right there and then ((points)) 181. V: I know but look, OK, start right there. And then it stops right here. ... 185. J: Like that because look it's going around it 186. V: Like this? 187. J: Yeah around Asia, Europe, and here ((points)) 188. V: Look, we can't do a huge one 189. C: Yes we can ... (they) did one 190. J: (...) placa . 191. V: 192. C: 193. V: I don't know It shouldn't be the whole thing


Zoom, zoom zoom zoom!

And a little bit right here

182. C: And then it goes to our region!

That's where the line starts

What are you talking about?

Here Cecilia asserted a hypothesized plate boundary "around Africa and that whole thing" in both speech and gesture (L. 170-171), supported by Juanita (L. 177), but Violeta challenged their proposed plate (L.178). The warrants for Violeta's challenge included a visual image in the GIS map ("start right there ... [a]nd then it stops right here" L. 181); an implicit assumption about the


"Placa tectonica" is Spanish for "tectonic plate."

Students' roles in group work


size of the plate they were supposed to find ("we can't do a huge one," L. 188, 193); and a direct questioning of her groupmates' credibility ("What are you talking about?" L. 191). This supports the characterization of Violeta's resistance to being made subservient in the group's decisions. As a result, Cecilia backed down (L. 192), conceding authority in part to Violeta. Using argumentation as a mode of competition. Beyond simply contradicting Cecilia's statements, Violeta explicitly framed the disagreement as a competition in which she wanted to prevail. She tried several times to recruit Juanita to her side, and during a time when Cecilia was absent at the end of Day 4, she also tried to lobby another student to join her in her opposition to Cecilia's plate hypothesis, seen in Excerpt 5: Excerpt 5. Violeta frames differences as a competition (Day 4)

989. V: Are you, are you like agreeing with me in that it won't be connected? ... 995. V: OK, see so, so I, like, 2½ against 1½, right? Because you are ½ with me and ½ ... 1055. V: YEAH, SO IT'S 3 AGAINST 1, YEAH, YEAH, SO THAT (...) SO THAT with [Cecilia's idea], right?

((dancing and singing)), see, so yeah!

Violeta clearly enjoyed the competitive aspect of the argument over plate boundaries. She eventually prevailed in this competition: her plate outline was the one presented to the class at the end of the project. Coordinating friendly interactions with confrontation. Violeta's competitive challenger role managed a balance of confrontation with comfortable, friendly interaction. This was accomplished two ways: (1) an ongoing stream of silly verbal and physical engagement with Cecilia, such as playful teasing and pushing (e.g., Excerpt 3, L. 71), as well as (2) sudden shifts away from strongly-contested domain argumentation (when threatening to become hostile) to

Students' roles in group work


out-of-school discourse, unrelated to science, that emphasized the girls' personal friendship, such as social gossip or discussions of parties. This pattern can be seen in graphs of coded discourse from Days 4 and 5 (see Figure 5), as the dispute about plates escalated and the deadline for finishing approached. Episodes of argumentation about plate size and patterns "connecting" (labeled argue in Figure 5), marked by clusters of codes referring to data, concepts, and real-world referents, end abruptly at certain points (Day 4 min 1, min 4; Day 5 min 6). Each of these abrupt endings marks an uncomfortable termination of an unsettled argument, threatening the harmony of the group. In each case, what followed was an episode of concentrated "out-of-school" talk (labeled gossip in Figure 5) between Violeta and Cecilia (Day 4 min 2, min 5-6; Day 5 min 8), which reaffirmed their friendship with one another, leaving the argument unsettled but salvaging the comfort zone of interaction (Radinsky, 2000), enabling later resumption of the science discourse.

Figure 5. Periods of argumentation about data are followed by periods of out-of-school talk.

Students' roles in group work


How did Violeta's role develop in practices of argumentation with visual data? As seen above, Violeta's competitive challenger role afforded her opportunities to engage in a number of practices in which she used GIS data images as evidence in scientific argumentation. She frequently marshaled data to support her contention that Cecilia's hypothesized plate was "too big to be a plate," using spoken and gestured references to particular aspects of data images to "assemble a representation" in group discourse (Hall, 1999) that would serve her competitive purposes (Goodwin, 2000). This is seen in Excerpt 6 from Day 5: Excerpt 6. Debate over whether "it connects" (Day 5).

27 28 29 30 31 32 33 34 35 36 37 38 V: Yeah. We say it's all the way up to -- Oh, now I [see] J: [Up to there]. Up to here ((points with cursor)) V: Because look, this is the African plate (that's a good point) C: Yeah, but it connects to that too ((points)). So, it's all the plates, Africa and Sicily and like [part of Europe] J: [But] Mr. Rodriguez, didn't he say [that, that's --] V: [See], but look it wouldn't be able to connect. It's just up to here ((points)) C: Wait. I can't see anything. So, I don't know where you are V: I'm there. ((points)) I'm where you, [I'm where you say to connect] C: [Where?] V: I'm right here ((points)). V: I know but look. down, no, no. there 39 40 41 42 C: I think it's [lots of plates though] V: [See cuz right here]. ((points, see Figure 6)) It wouldn't be able to connect C: Just stop right there (...) ((points)) V: I know, but it still won't be able to connect and it's not that high. (This one is pretty low...) But it can't connect C: There's volcanoes there and everything so

((talking to the computer)) Go down, down,

I like here better (...) OK, we'll go riiiiight

Students' roles in group work


Violeta pointed to the computer screen four times in this excerpt (L. 32, 34, 36, 40), and also carefully crafted the exact GIS image she wanted Cecilia to see, zooming and panning and even talking to the computer in the process (lines 38-40, shown in Figure 6). Her evidence here was a place in the Red Sea where the line of earthquake data showed a gap.

Figure 6. Violeta points to data supporting her explanation.

At the same time, her role constrained her from participating in other practices that could have brought her deeper into domain reasoning with data. Violeta did not engage in episodes in which her groupmates struggled to articulate wonderment and curiosity about the nature of the concept plate itself, as in Excerpt 7: Excerpt 7. Lots of placas (Day 5).

48. J: Yeah, I think it's just that little area ((traces on screen)) 49. C: I still think it's lots of placas 50. V: See cuz look 51. J: Because maybe that could be a placa but it's divided into a [whole bunch of little] ((gesturing)) 52. C: [Or it's starting to be a placa]


53. V: [Weeeee!] ((moving the mouse around in circles)) 54. C: [Is that possible?] ((shrugs her shoulders))

Students' roles in group work


These imaginative hypotheses, such as a plate being "divided into a whole bunch of little" plates (L. 51), or "starting to be a placa ... Is that possible?" (L. 52, 54), illustrate some of the ways in which Cecilia and Juanita problematized the concept of plate, as they tried to come up with explanations to account for the messy data they faced. Violeta, however, sat out these exchanges (here playing with the mouse), perhaps because they did not provide an opportunity for competition in which she might confidently engage. Case 2: Eliana as competitive challenger to Leo's authority A second case study of the competitive challenger role provides a valuable comparison for understanding its affordances and constraints for engaging in the analysis of visual data. In the EKL group, Eliana emerged as a co-leader with Leo in quantity of group talk (see Table 2 above, and Figure 7) early in the investigation. Like Violeta in the CJV group, Eliana carved out her co-leadership through challenges to Leo's bids for authority in group decision-making and science discourse. Though Eliana's parity with Leo in total turns became less apparent on Days 4 and 5 (Figure 7), as her share decreased and Kerry's increased (discussed in Case 4 below), this did not represent a lessening of Eliana's participation in the group's science discourse, as shown in Figure 8.

Figure 7. Distribution of talk in the EKL group by day.

Students' roles in group work


Figure 8. Distribution of turns with domain referents in the EKL group by day.

Positioning in opposition to a peer who stakes a claim to authority. Like Violeta, Eliana played a similar role in her group as a competitive challenger, competing with her loud and opinionated groupmate Leo. Leo frequently made comments in the form of authoritative statements of fact, at times contradicting or mocking Eliana's ideas explicitly, but Eliana did not shy away, as seen in Excerpt 8: Excerpt 8. Eliana engages Leo's authoritative talk (Day 1).

48. L: OK, I think it's a titonic plate because, well, if it is then it would most likely be a subduction zone, as if this were one plate ((points with cursor)) and this were one plate, this one and this one were banging into each other, and this titonic plate is more powerful so pushed this one under and that's why the depth is lower ((moving cursor around)) 49. 50. 51. 52. 53. 54. 55. 56. E: That's why they have tusamies L: That's why they have tu-what? E: Tsunamis? ((changes pronunciation)) L: Tsunamis. No, that has to do with like wind and stuff E: Oh L: I'm going to go (...) ((getting up out of his chair)) E: Oh, look L: OK ((sitting back down))

Students' roles in group work



E: This could be one big plate ((tracing)) and maybe this is another one ((pointing)) and if they hit each other ((points))

Leo's initial statement here (L. 48) employed several domain concept words (plate, subduction zone, and depth, as well as a mispronunciation of tectonic), offering a detailed explanation of the kind of activity he thinks is occurring around Japan. Eliana's response (L. 49) engaged the ideas in Leo's statement, adding another concept to the explanation: that the plate motion causes tsunamis. Leo's mocking of her mispronunciation (L. 50), and his (inaccurate) correction of her (insightful) comment, did not dissuade Eliana from engaging further in the domain discourse. Though she did not reply to his contradiction of her idea here, she did go on to further elaborate the explanation on the table (L. 57). At other times Eliana did directly contradict Leo's arguments, as in Excerpt 9 (L. 363, 365): Excerpt 9. Rift vs. subduction (Day 5).

360. 361. 362. 363. 364. 365. L: This one's obviously moving as a rift ((points)) creating a trench E:

((nods her head))

L: Huuuuh. The Mariana trench ((points)) E: But, I still think it could be going like this ((gestures


L: Mariana Trench E: I still think it could be going like this ((gestures


Resisting being positioned as subservient in group decisions. Leo's authoritative talk might easily have led Eliana to submit to his explanations and decisions, but in her resistance she showed an unwillingness to be positioned as not understanding. For example, Excerpt 10 shows several discourse moves in which Eliana positioned herself as a co-explainer with Leo, rather than one who required help (as Kerry was constructed here).

Students' roles in group work


Excerpt 10. Eliana stakes her claim to explanations of the data (Day 1).

25. L: Okay elevation and depth according to this, look, see this is a gigantic line that's very small very low and here it goes WAY down, [this is most likely] 26. 27. 28. 29. 30. 31. 32. 33. E: [Look wait, don't, look back], it shows it, see right there

((points to legend)), leave it right there

L: This is locations, okay, so Kerry, right here, Kerry pay attention K: Hmm L: Right here it looks normal around 0 to 1000 ft down [but right here] E: [Right there] L: [It goes DOWN!] E: Right [here is the deepest] L: So what do you think happened? Do you think they collided and then they went up to create this one? Do you think they moved up and down like the island? 34. 35. 36. 37. 38. 39. 40. E: They used to be like this ((gestures)) then they just went like that ((hands pull apart)) L: No E: Well if they were like this [and then pssshhh] ((gestures)) L: Well if one part is higher and then the other part is lower [there's two possibilities] K: It went like this ((gestures)) L: Either that it split apart to make the edge of it [where there's very little, right here] E: [That's what I said, it split]

In this exchange Eliana used several moves to avoid Leo's positioning of himself as explaining to her. First she interrupted his explanation (L. 26) and pointed out the legend as a better indicator of depth. As Leo shifted to position Kerry (and not Eliana) as the recipient of his lesson (L. 27), Eliana spoke simultaneously with him (L. 29-32, overlapping speech marked with "[brackets]"), refusing to give him sole access to the floor (and thus sole claim to understanding the data). Finally, when Leo evaluated Eliana's gestured explanation (L. 34) as wrong ("No," L.

Students' roles in group work


35), she responded with a new gesture, and then objected to his framing of "two possibilities" (L. 37, 39), taking ownership of her prior explanation back from him ("That's what I said, it split," L. 40). Using argumentation as a mode of competition. Eliana's role in her group also included explicit competitiveness, as she challenged Leo and showed satisfaction at bettering him. For example, on Day 1 Eliana proclaimed her prowess at finding "titonic plates" to Leo (Excerpt 11). Excerpt 11. Eliana as plate-finder (Days 1 and 2).

457. 458. 459. 460. 461. 462. 463. E: But still, I found the titonic plates, again! ((pointing to

herself)) And I proved them to you!

L: ((sarcastic tone)) Oh, wow! E: Again! L: Big whoop! plates E: E: They could be Well, the other one was a "would"! ((points)) Well, the other L: They could, you found a "could," not a titonic plate ... one was an "is"! How ­ we don't even know if those are titonic

This boast (L. 457, 463), competitive but still playful (e.g., L. 463), was repeated on Day 2, when Eliana corrected Leo's attempt to take credit:

433. 434. 435. 436. L: So these are the titonic plates that I found E: Yeah, I found them L: No, (...) E: I told, I said I thought this, and this, and that one were the titonic plates

Eliana's emphasis on her own role in finding the plates (L. 434, 436) suggests a competitive (while playful) stance toward Leo in group work. Coordinating friendly interactions with confrontation. Like Violeta, Eliana found ways to balance her challenging of Leo's authority with maintaining a comfortable and friendly

Students' roles in group work


dynamic in the group. She accomplished this in part by taking opportunities to explicitly agree with or build on Leo's pronouncements, and then co-constructing them further by adding to or revising them, as seen in Excerpts 8-10 above, and in Excerpt 12: Excerpt 12. Eliana agrees with Leo and adds on (Day 2).

68. 69. 70. 71. 72. 73. L: Oh, oh there we go, there's Japan. Can't you see this is pretty much the deepest part of the entire world? E: It is L: Right here and here K: Isn't this the deepest part? ((points)) L: Well, no I mean it comes deep, I mean it has pretty dark. I mean this is navy blue, I mean seriously E: It is. ((points)) Right here is pretty deep ((points to map)) but this is that color ((points to legend)) that blue ((points to


Here Eliana used a positive strategy of agreeing with Leo ("It is," L. 69, 73) to establish her own position as a knower, and to engage in co-constructing and clarifying the explanation of the data (L. 73) without confrontation. This approach also used the gesture space as a medium in which to co-construct explanations, such as Eliana interacting with Leo's representational gestures, using her hands to add additional elements to his gestured explanations, as shown in Figure 9 (see Singer, Radinsky & Goldman [in press] for a discussion of this and related examples). These co-construction episodes served to offset tensions created in argumentation. How did Eliana's role develop in practices of argumentation with visual data? Eliana's competitive challenger role provided opportunities for her to establish her own agency with respect to making sense of visual data, rather than be constructed as a "student" of Leo's. At the same time her facility with balancing confrontation of Leo with friendly coconstruction of concepts in speech and gesture enabled her to maintain a generative, creative space for exploring domain concepts.

Students' roles in group work


Figure 9. Eliana interacts with Leo's gesture to co-construct what might happen at a rift zone.

Like Violeta, Eliana developed great facility for using visual data in support of a position. When Leo proclaimed "facts" without clear evidence or warrant, Eliana frequently pointed to

Students' roles in group work


specific data, either to clarify the point or to challenge it. For example, on Day 1, Leo asked whether Eliana thought that a particular area might be a plate, and Eliana led their conversation to address specific aspects of the visual data, shown in Excerpt 13: Excerpt 13. Eliana uses data as evidence (Day 1).

42. ... 45. 46. 47. ... 50. 51. 52. 53. 54. E: That doesn't seem right, because right here seems higher L: It shows lines of higher elevation E: So, you're right these two are ((points)) L: What? See ­ E: If there's a higher elevation here ((points)) then they must be hitting each other E: I do believe that's a plate, I mean maybe that's a plate, too, but this one is bigger, that's a plate L: OK, watch it, watch my thing ((moving the cursor)), what do you think, I believe that this is a plate, oooh, ooh E: That looks like it's connected to that ((points)) L: No, don't you believe that this might be a plate, I mean ­

Figure 10. Eliana references GIS data in articulating an explanation.

In response to Leo's statement "I believe that this is a plate" (line 46), Eliana pointed (Figure 10) and made a specific observation of a data pattern ("That looks like it's connected to that"), debated whether the appearance of a higher elevation would be consistent with Leo's idea

Students' roles in group work


(50), and then went on to explicitly connect the data pattern to a domain explanation (54). These turns contrast with the simplicity of Leo's claim, "I believe that this is a plate" (46), in their evidentiary use of data. Furthermore, Eliana introduced an element of uncertainty ("maybe," looks like," "doesn't seem," "must be") into the discussions of data that mitigated Leo's tendency toward absolutist, deproblematized characterizations of the data. Cases 3 and 4: Quiet bystanders Another common orientation of a student to group work, particularly when the group's norms are characterized by competition or challenge, is to observe quietly, disengage, or otherwise pull back from interaction. One such orientation is operationalized here as the role of quiet bystander, characterized by: · · · · Remaining silent for portions of group discourse Withholding one's own observations or opinions Positioning outside of contests for authority Positioning as subservient in group decisions

Two students who assumed this role are examined in Cases 3 and 4 below: Juanita in the CJV group, and Kerry in the EKL group. Case 3: Juanita's quiet bystander role with Cecilia and Violeta. Remaining silent for portions of group discourse. Juanita's participation during the GIS investigation began as very quiet and withdrawn, though she emerged toward the end as a far more active participant. Juanita's silence was most notable on Day 1, when she offered only 1 spoken contribution (L. 85 in Excerpt 15, below), which was cut off and talked over, and otherwise sat silent for all of the 22 minutes of group work. On the other days as well, though, periods of silence of one to three minutes were common for Juanita, even during times when she

Students' roles in group work


was working with only one group mate (Day 2 with Violeta absent, and the end of Day 4 with Cecilia gone). In contrast, neither of her groupmates ever went a full minute in silence. Over all Juanita contributed only 9% of all spoken turns during the 5-day investigation. Withholding observations or opinions. Even on Day 2, as Cecilia tried to bring Juanita into the discourse in Violeta's absence, Juanita appeared to resist putting her own opinions or observations forward, as in Excerpt 14: Excerpt 14. Cecilia recruits Juanita to talk, with Violeta absent (Day 2).

5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. C: OK, which one is it? J: ((points with pen)) That one? C: OK, so how does it get the other thing on? C: Oh, elevation. C: OK. There. There, OK. C: Where is that -- where is it? C: OK. Now what do we do? Or there? Where. Yeah, that's a good one. I don't J: Now we should find the, uh -- rift zones, and -- where (...) C: How 'bout there? know! C: Say. Tell me. Right there? C: You think? J: (All along this) ((points with pen)) C: Oh yeah. OK, so, let's just say -- this one? Yeah? That's not so dark -- right there. (There we go.) That's pretty.

((adjusts monitor))

In this excerpt Cecilia repeatedly asked direct questions of Juanita (L. 5, 7, 10, 11, 13, 14, 15, 17), presented herself as needing Juanita's assistance (L. 11, 13), and explicitly told her to speak (L. 14: "Say. Tell me."). Despite this active recruitment Juanita appeared hesitant, as her three tentative responses here (L. 6, 12, 16) suggest. Positioning outside of contests for authority. During the first three days of the investigation Juanita mainly avoided talk related to decision-making or domain reasoning, the

Students' roles in group work


issues that Cecilia and Violeta were actively debating. On Day 1 Juanita sat quietly for all but one turn (Excerpt 15, L. 85) of a 22-minute session. Excerpt 15. Juanita is talked over (Day 1).

83. V: 84. V: 85. J: 86. C: 87. V: (There's) something right here. This -- oh no, that's an island Wa -- I don't get it. [Go to that] ((points)) Go to this? Or this? (Why do we) -We could go to islands, [because islands] -Right here it's separated.

Juanita's one contribution (85) was talked over by Cecilia (86), and her idea went unheard amidst her groupmates' jockeying and joking (77 - 82). Positioning as subservient in group decisions. Juanita did emerge as a more active participant in the group over time (see Figure 3 above), but she did so mainly by agreeing with positions put forward by one or the other of her groupmates, as in Excerpt 16: Excerpt 16. Juanita orients to her groupmates' positions (Day 4).

526. ... 610. ... 843. ... J: I agree with Violeta ... I agree with you, Violeta, because ­ J: Violeta, ... what Cecilia said, because how is that connected? J: Really I don't know which one is la placa J: Well I think what Cecilia did is not bad, the placa is all that

Juanita still struggled to have her entire opinion heard (note the truncated nature of each of the turns above), and appeared to make her own ideas subservient to those of her peers. How did Juanita's role develop in practices of argumentation with visual data? By Day 5 Juanita's quiet bystander role had begun to develop into a far more participatory mode of work, aided by Cecilia's encouragement on Day 2 and Violeta's attempts

Students' roles in group work


to recruit her as the "swing" vote in the competition with Cecilia. Juanita engaged in a variety of kinds of domain talk. One mode was the kind of generative and creative imagining about the nature of plates (with Cecilia only) seen in Excerpt 7 above. Another mode of talk was problematizing both of her groupmates' propositions in the debate. This was seen in its initial form in her hesitant objections to Violeta and Cecilia (Excerpt 16 above). Excerpt 17 below shows Juanita's emergence as an active debater in the last session on Day 5, as the final presentation approached. Excerpt 17. Juanita problematizes plates (Day 5).

90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. J: ... So what do you (think?). But then if that's the plate C: [Or it could be] up here, but nothing connects there, that's the problem ((points)) J: If that's the [plate that's gonna be] C: [Right there] J: [That's going to be too big] ((points)) C: I know it's cuz it's [lots of plates] J: [And that would be including], another, another area. That would be including the Himalayas or something ((points)) C: (But...) nothing connects right there ((points)) V: [Exactly] J: [Exactly] C: [That's what] I'm saying V: You're saying that the whole, the whole thing is the plate J: And isn't it too big to be a plate? C: No, the other ones are big

Here Juanita raised not only Violeta's objection ("too big to be a plate," L. 94, 102), but also a new one of her own: that a Eurasian plate would include "another area ... the Himalayas" (L. 96). This objection introduces a new warrant for the counter-claim, i.e. the way the task assignment was organized (a different group had been assigned the Himalayas). This marked the emergence of Juanita's own reasoning in the argument, beyond just her alignment with one or

Students' roles in group work


the other of her groupmates. It also showed an impressive coordination of multiple resources for sense-making: spoken and gestured references to data images (L. 90, 94, 96), the meaning of the concept of plate (L. 94, 102), and the logic of the assigned task (L. 96). Case 4: Kerry's quiet bystander role with Eliana and Leo A second case study of the quiet bystander role, Kerry in the EKL group, reveals a similar starting point, but a very different trajectory of role development. Remaining silent for portions of group discourse. Kerry had the lowest share of total talk in her group (19% of turns overall), and though she was not as quiet as Juanita, she also had periods of silence during group work. There were at least three such silent periods lasting a minute or more on each of the five days, and none for either of her groupmates. Withholding observations or opinions. Like Juanita, Kerry rarely offered her own observations or opinions. Her groupmates sometimes solicited her opinion or encouraged her to participate, though she was usually reticent, as seen in Excerpt 18: Excerpt 18. Asking Kerry to speak (Day 1).

L: ... OK, where do we think Kerry? Where do you think we're gonna look for? K: What do you mean? L: Where do you want to look?

What Kerry wanted to do, and what she saw in the data, was often difficult to discern from the content of the group's discourse for these reasons. Positioning outside the contests for authority. With two groupmates engaged in an almost constant stream of talk, much (but by no means all) of it focused on the science investigation, Kerry sat out most of the substantive discussions of plate tectonics. This can be seen in a characteristic segment of coded group discourse on Day 1, in which Eliana and Leo

Students' roles in group work


provide a steady stream of talk referencing concepts, data items, and real-world referents, while Kerry made only occasional contributions (Figure 11), two of them with out-of-school referents (just before minute 17). Domain argumentation mainly passed her by.

Figure 11. Kerry is silent for much of Eliana and Leo's argumentation.

Positioning as subservient in group decisions. Like Juanita's, Kerry's viewpoint was constructed as subservient to those of her groupmates. One common pattern was for the group to construct Kerry's role as that of a student requiring assistance, with Leo or Eliana playing the role of a teacher. This construction of Kerry as a "student" of her groupmates, and her acceptance of that role, is seen in Excerpt 19, in which both Eliana and Leo offered to help her, even competing to explain to her (L. 118-120). Excerpt 19. Explaining to Kerry (Day 1).

118. E:

((to K)) Look, you see this plate right here? ((points)) ((to K))

See right here ((points)),

119. L: I can explain 120. E: No, I want to explain. see how it looks deeper than any other place else in the world

((traces circle)), we think, we think that's a plate, you see

this right here and ... 124. E: Exactly these are two plates that are crashing into one another 125. K: Two plates 126. E: Two plates, two titonic plates under water 127. L: Subduction zone, Kerry, when they go like this, chrssshhhh

Students' roles in group work


Kerry did not resist this construction of her role as "student" of her peers, as seen in her repetition of Eliana's words ("two plates," L. 125), and in this exchange with Leo: Excerpt 20. Kerry as student (Day 2).

14. 15. 16. 17. L: No, yeah, no, um, long before that there was one single island in the entire world, all of these pieces were all connected E: Yeah, I know L: Into one, Kerry K: I'm listening

Here Kerry's comment "I'm listening" (17) suggests that she was willing to go along with her groupmates' instruction in these situations, co-constructing this "student" role. Kerry's own science-domain ideas were constructed as not having much importance in the group's investigation. Though her contributions were sometimes praised by her groupmates, they did not figure into the decision-making leading up to the final presentation. This was seen in the discussion of which direction the plates are moving around Japan. Kerry had suggested (in gesture only) that the plates might be buckling together, rather than rifting apart (Leo's idea) or subducting (Eliana's idea)5. After Kerry made a "buckling" gesture as a suggestion, the group took up this idea (Excerpt 21): Excerpt 21. Kerry's theory (Day 2).

251. L: This is obviously, wait maybe this is a buckling zone, it could, it could be that both of these went like this: prush

((gestures buckling))

252. K: [That's what I said] ... 257. L: Yeah Kerry you were right, we'll call that Kerry's theory 258. E: She is right 259. L: We'll call that Kerry's theory 260. E: Kerry's theory is correct, see look ((tracing))


See Singer, Radinsky & Goldman (in press) for a discussion of the role of gesture in these exchanges.

Students' roles in group work


Here both Leo and Eliana explicitly valued and supported "Kerry's theory" (lines 256-260). However, as the project moved on, this "buckling" theory was never referred to again, and the debate continued to be only about rift and subduction. How did Kerry's role develop in practices of argumentation with visual data? Like Juanita, Kerry's participation in the group did emerge out of the quiet bystander role by the final days, but rather than develop increased participation in domain talk and argumentation, she moved in the opposite direction. Her low level of domain-referenced talk remained relatively constant across days, i.e. less than 10% of domain-referenced turns on each day (Figure 8 above), but her total share of talk increased to nearly the same percentage as Eliana's (Figure 7 above). This was accomplished not through increased engagement in domain argumentation (as it was for Juanita), but instead through far more out-of-school referenced talk, which Kerry initiated more and more as the project increased. Both of her peers participated in this talk frequently with her, maintaining a friendly joking and gossiping throughout the project with Kerry and Leo both leading this talk. But while her groupmates alternated frequently between gossip and science talk, Kerry rarely entered into discussions referencing the visual data. When at the beginning of Day 5 the group engaged in naming visible data patterns according to what they looked like ­ "bat wing," "bird's face," "dolphin" ­ Kerry did participate in this data-referenced talk for three full minutes, the only such period in the project. But as soon as she was challenged in the fast-paced, confrontational talk of Leo and Eliana, she withdrew, as seen in Excerpt 22: Excerpt 22. Kerry withdraws from data-referenced talk (Day 5).

87. 88. 89. 90. L: Hey let's call this one the dolphin! ((traces)) E: Yeah, it does look like a [dolphin] L: [Look], ((traces)) there's it's fin ((traces)) K: (...) I don't see [(...)]

Students' roles in group work


91. 92. 93.

E: [It's a dolphin] L: You're blind Kerry K: I have no creativity ((laughing))

This self-effacing remark (L. 93) follows being interrupted by Eliana (L. 91) and insulted (playfully but rudely) by Leo (L. 92). After this turn, Kerry made no further reference to data for the remaining 18 minutes of this final day (Figure 12, min. 4-22), turning instead to chatty gossip that periodically distracted each of her peers from their focus on the data. This pattern can be seen in the coded data in Figure 12: while all three participated in some out-of-school talk, Kerry employed it exclusively for the final 18 minutes of the project, including some shifts of topic that appeared to intentionally sabotage her peers' focus on the data (min. 8, 13, 19).

Figure 12. Kerry avoids domain discourse, using out-of-school talk (Day 5).

DISCUSSION In this section cross-case comparisons (Yin, 2002) are used to examine the nature of the learning evidenced in each case study. How did role development mediate the process of learning to reason with visual data? Changing participation in scientific practices, and the appropriation of cultural tools for sense-making with visual data, are forms of learning that can productively be viewed through the lens of role development. An advantage of this construct is that it situates this learning at the intersection of individual and social processes of change, a

Students' roles in group work


meeting point of individual identity and collective practices. The normative practices for sensemaking with data differed between the two groups, and the roles identified across groups (competitive challenger and quiet bystander) positioned each student differently within these practices. The different trajectories of role development suggest each role's affordances and constraints for learning. Of particular interest here are the ways these emergent roles oriented these students with respect to two central challenges in scientific inquiry with inscriptions: 1. Engaging in their group's practices of "assembling representational states" (Hall, 1999) with the data for reasoning about plate tectonics; and 2. Managing the dialectical tensions between conflict and agreement that are inherent in argumentation (Leitao, 2000). These two challenges, framed in the Introduction, are used to structure the Discussion of how students' roles mediated their learning to participate in sense-making practices with visual data. Challenge 1: "Assembling representational states" for reasoning about plate tectonics In order for the GIS maps to become meaningful sources of information to be analyzed, the students had to assemble resources with which to make the "absences" (earthquakes, earth's crust, plates, subduction zones) "present" in their discourse (Hall, 1999; Latour, 1990). Different resources were brought to bear in each group, resulting in a unique "semiotic ecology" for each (Enyedy, 2005:461). The role each student played in her group mediated the extent to which she accessed these resources. For example, Violeta participated and excelled in practices in which the group assembled representational states consisting of pointed-to and traced, visible patterns of dots on the screen ­ patterns which were constructed as "connecting" or "not connecting." The significance of these

Students' roles in group work


patterns was taken-as-shared in this group's practices (Bower, Cobb & McClain, 1999); that is, they were accepted as sufficient in argumentation without further explanation, and as such they eventually delivered a victory to Violeta in the debate over plate boundaries. The intersubjectivity underlying these taken-as-shared characterizations was established by using indexical language (e.g., "look," "see," etc), pointing at the screen, and manipulating the map to show particular things (Excerpt 6 above). Violeta's competitive challenger role afforded vigorous participation in these practices, and she developed impressive expertise in constructing data images to support these kinds of claims (e.g., note her clearly increased mastery of the GIS from Excerpt 1[Day 1] to Excerpt 6 [Day 5]). However, this intersubjectivity, established by pointing to and tracing visible patterns, did not address the ambiguity of what "it" was that was not "connecting." Note, for example, the simultaneous utterances of "Exactly!" (Excerpt 17, L. 98-100) in which Cecilia, Violeta and Juanita all implied that the data referenced in Cecilia's gesture (L. 97) constituted evidence of directly opposing positions. What exactly did each student assume she was seeing through the "window" of the data image? In this space of uncertainty, the group needed to construct a representational language in which plate could become a shared concept in common ground (Clark, 1996), associated not only with particular data patterns but also with real-world referents and an abstract explanatory model from which to argue about plate size and boundaries. This was the process Cecilia and Juanita began to engage in on Day 5 (Excerpts 7 and 17), using improvised representational gestures and speech to imagine tiny plates, enormous plates, and an area possibly "starting to be a plate ­ is that possible?" This mode of discourse was an attempt to manage the uncertainty inherent in the investigation (Hall, 1999) by jointly exploring the boundaries of the concept. But in this expansion of the group's assembled

Students' roles in group work


representational state of the data ­ bringing in resources such as imagined processes of plate formation, or vastly different possible sizes of plates, in order to overcome the ambiguity of "can't connect" ­ Violeta's role appeared to constrain her participation. She avoided these ambiguous wonderings, showing no sign of uncertainty, perhaps concerned that it might be a weakness that could undermine her competitive advantage. Juanita's quiet bystander role, on the other hand, mediated her participation in these semiotic practices of the same group quite differently. Discursive practices of managing uncertainty provided opportunities for her role to develop in new directions. She emerged from her early silence, with Cecillia's encouragement (Excerpt 13) and Violeta's recruitment (Excerpt 12), primarily in situations in which concepts or data patterns were problematized and uncertain. This included discourse that constructed the dots on screen as patterns that "can't connect," but it also included talk that constructed the images as "plates," as "the Himalayas," or even as some unknown entity that "could be a placa but it's divided into a whole bunch of little [placas]" (Excerpt 7 L. 51). The challenges of establishing intersubjectivity and assembling a shared representational state created an uncertainty in which Juanita's quiet bystander role gave her opportunities to watch, listen, and allow for elements of possibility in both of her peers' competing hypotheses (Excerpts 16, 17). Her role also afforded her agency as a swing vote, without requiring her to explicitly confront either of her forceful colleagues. The semiotic ecology of the EKL group, on the other hand, employed quite different resources in the representational states it assembled, and they were used for different purposes. What was made most obviously "present" in this group was a trench, an abstract entity freed from the GIS screen and transported to the representational gesture space where visual data were not needed, and an imagined process by which two generic plates might interact to form this

Students' roles in group work


trench.6 These representational practices did afford Kerry an opportunity to participate in domain discourse even within her quiet bystander role: she could put forward her "buckling" theory ­ silently, in gesture only ­ without being required to support it with evidence in the GIS data (Excerpt 21). Given the taken-as-shared practice of hypothesizing freely in this way, her groupmates could ratify her theory as "correct," though they could also proceed to ignore it. When it came to using visual data, however, the EKL group's practices left no room for a quiet bystander to engage in the symbolizing talk. The construction of Kerry's status as that of a "student" of her peers ­ a non-knower ­ minimized her agency. Kerry's quiet bystander role provided no opportunities to develop agency in the group's jocular, confrontational mode of data-referenced talk, and only an occasional opportunity to do so in the representational-gesture space. This withdrawal from science talk should not just be seen as a function of her level of conceptual understanding: Violeta, who like Kerry struggled with some of the key concepts in the post interview, embraced challenge and contestation from her peers, using them as an opportunity to build her skills with the data visualization tools. Eliana's competitive challenger role, on the other hand, proved ideal for her to actively explore a represented world consisting of plates, continents, trenches, earthquakes, volcanoes, and subduction zones, within the group's norms. Uncertainty did not deter her from engaging in extended debate, and her own willingness to be wrong (see Excerpts 8, 13) actually strengthened her hand, as she was able to agree with Leo, extend or amend his explanations with reference to the data, and gradually emphasize her own position. As a competitive challenger, Eliana had opportunities to practice reasoning with data images, representational and indexical gestures,


These purposes are in no way pre-determined by the data: this group might well have questioned how big a plate should be, what plate Mt. Fuji was on, or whether particular parts of their boundary could connect (all would have been relevant, given an ambiguous area near the Marianas Trench that the group spent time discussing). Other groups studying Mt. Fuji in other classes have engaged in these kinds of debates.

Students' roles in group work


unfamiliar science vocabulary (note her adjusted pronunciation of "tsunamis" in Excerpt 8), and a range of argumentation strategies in her contests with Leo. These opportunities for learning were not simply a function of Eliana's greater understanding of the concepts taught in the unit, though her post-interview and test showed that she did have a firmer grasp on all of the key concepts than Kerry did. Even with a high level of individual understanding, it is quite possible for a high-achieving student in Eliana's situation to quickly establish the correct answer to an inquiry question, and then leave the task as "complete," without exercising many of the discursive resources for understanding concepts and reasoning with data (e.g., note the "Answer Man" example above). Other students with similar conceptual understandings to Eliana did not develop their repertoire of symbolizing practices to the same extent. Cecilia, for example, performed similarly to Eliana in the post-test and interview questions on concept meanings, but had fewer opportunities to develop her participation in these discursive practices than Eliana, due to the different practices of her group. In her competitive debate with Leo, on the other hand, Eliana had multiple opportunities to explore the represented world of plates and trenches, coordinating data images with imaginative gestured explanations, responding to challenges and elaborating her explanations. Unlike Violeta, Eliana's competitive challenger role developed along multiple dimensions, expanding from data-referenced competition (where Violeta's role emerged and stayed) to include similarly competitive argumentation in imaginative and abstract areas of domain thinking. Challenge 2: Managing the dialectical tensions of argumentation Both of these groups developed norms that included significant amounts of dispute. Beyond simple disagreement, there were numerous episodes in each group that fit the discursive pattern of argumentation: claim, counter-claim, and reply (Leitao, 2000), often with articulated

Students' roles in group work


warrants and backings that incorporated resources from the science domain (e.g., Excerpts 4, 6, 7, 8, 10, 13, 17, 19). This active contestation and negotiation was part of the learning process envisioned by the teacher: he emphasized that students needed to learn how to argue for an interpretation, and that they would understand the nature of inquiry better if they had to negotiate their purposes and meanings among themselves. Argumentation also played an important part in the ongoing negotiation and development of roles in the groups. For example, we have seen both Eliana and Violeta engage in argumentation as a tool for challenging a dominant group member's authoritative positioning, actions that served both to problematize absolutist knowledge claims in the group's science discourse practices (Cornelius & Herrenkohl, 2004; Wertsch & Rupert, 1993), and also to develop agency within each girl's role as a knower of science. This was evident in Eliana's use of visual data to challenge Leo's unsubstantiated claims about trenches (e.g., Excerpts 9, 13), and in Violeta's use of the absence of data to problematize Cecilia's argument for "one big placa" (Excerpt 6) ­ a discourse move that initiated her groupmates' generative talk about plate formation processes, though that line of inquiry did not have time to come to fruition. These examples show how the social negotiation of authority coincided with the negotiation of the meanings of concepts and data, with implications not only for students' social standing in the group, but also for their "conceptual agency" (Greeno, 2006) with respect to scientific understandings. This can be seen in the example of Kerry's buckling claim (Excerpt 21). It was contributed in gesture only, and was not countered, but still stood as her contribution to the ongoing debate about Japan's plate boundary zone ­ it was a hesitant step into the practices of argumentation for Kerry. In the post-interview Kerry showed agency in making sense with this concept: she reproduced the "buckling" gesture and articulated its meaning

Students' roles in group work


verbally (Singer, Radinsky & Goldman [in press]), something she had not done in the group, and something she did not show evidence of doing for any plate boundary concepts other the one she had tentatively argued for. It appeared at the time that her choice of buckling as her theory arose out of the need to take up a position of her own, different from those of her groupmates (subduction and rift), yet this process of engaging in argumentation seemed to mediate her emergent understanding of the concept, and her agency in reasoning with it weeks later. Students developed agency for reasoning with concepts that played a part in their role development. This suggests the power of argumentation as a potential lever for conceptual change (Andriessen, 2005, 2006; Leitao, 2000). However, there is also an inherent difficulty in managing this mode of learning. Developing argumentative norms places a certain tension on a group of students. In both groups these disputes sometimes strained norms of friendly interaction, as seen in Cecilia's and Violeta's frustration and strain as they tried and failed to find common ground on the final day of the investigation, and in Kerry's withdrawal from discussion of data patterns. In both groups the maintenance of group harmony necessitated some way to moderate the dynamics of disagreement and conflict, a balance that was achieved differently in the two groups, with implications for role development in argumentative practices. Roles that afforded negotiation of meaning through argumentation. Eliana and Juanita both managed the tensions of argumentation in ways that extended, elaborated, and problematized talk about science. Their emergent roles enabled them to engage in contested discussions of concepts and data in ways that kept the string of claims, counterclaims, and replies alive, yet steered the social negotiations of relationships away from hostility. In each case, the student's role developed increased agency in the group through these

Students' roles in group work


argumentative practices. For example, Juanita's emergence from her quiet bystander role was facilitated by Violeta's active recruitment of her as a "swing vote," and as she became more active in these discussions her comments served to problematized and explore, rather than finalize and close debate. Keeping the debate alive served to keep her in the conversation, where her participation grew and deepened. When the discourse turned away from science, Juanita quickly returned to her quiet bystander role, as seen clearly during the periods of out-of-school talk between Cecilia and Violeta on Days 4 and 5 (see Figure 5 above). Similarly, Eliana's responses to Leo's potentially-hostile mode of debate served to extend and complicate exchanges that might otherwise have remained absolutist mini-lectures. With a range of discursive moves, she found ways to agree with parts of his claims, maintain a playful and silly stream of talk and gesture, and at the same time introduce problematizing elements into the discussion. This is visible in her interweaved gestures for describing activity at a boundary (Figure 9), in her playful countering of his authoritative claims (Excerpt 11), and in her use of data to clarify and extend characterizations (Excerpts 10, 12, 13). In these ways Eliana developed a role with significant agency in argumentation and reasoning with data, without falling into a pattern of either hostility or submission, and subtly returning the discourse to the science domain when her peers veered off into gossip or off-task goofing. Roles that constrained negotiation of meaning through argumentation. In contrast, the ways Violeta and Kerry managed the tensions of disagreement served in part to limit opportunities for productive argumentation about science concepts and the meanings of data. For example, Violeta's role afforded intense and extended argumentation up to the point that she might be required to negotiate her position, but it allowed her no option for being persuaded (unlike Cecilia, who allowed for confusion and contradictions within her own

Students' roles in group work


hypotheses). In this sense, though she used data quite persuasively, in the conceptual terrain her argumentation was absolutist: if it can't connect, it can't connect ­ end of story. When competitive challenges threatened to upset the "comfort zone" (Radinsky, 2000) of her friendship with Cecilia, rather than find a way to move the science talk forward through negotiation ­ the essential component for conceptual change ­ Violeta retreated to out-of-school, friendly gossip (Figure 5), keeping the peace but losing the opportunity to learn from the disagreement. Kerry sometimes used the same resource (out-of-school gossip) to opt out of confrontational discourse as well, but significantly, this avoidance of confrontation was not evident during non-science talk: she engaged enthusiastically in disagreeing and poking fun at her peers during out-of-school talk, but reverted to her quiet bystander role in science talk ­ the reverse of Juanita's role development. Thus her avoidance of argumentation was not simply a strategy for avoiding all disagreement. Rather, the way she negotiated the intimidating pace and volume of her groupmates' debates was to construct herself (in collaboration with them) as a non-knower, requiring (or submitting to) explanations. This is not to say that she did not struggle to understand the concepts, which her post-unit assessments suggest she did. Rather, this feature of her role in the group kept her out of argumentation situations that might have helped her understand these concepts (evidenced by the buckling example above), where her peers' teacherly explanations did not appear to afford this learning (evidenced by her inability to use the concepts of subduction or rift comparably well in the post-assessments ­ concepts that her peers had "explained" to her). The non-knower stance of her quiet-bystander role managed the tensions of arguing science with her peers, but in so doing it deprived her of the opportunity to develop conceptual agency.

Students' roles in group work


Balancing the dialectical tensions of argumentation (negotiating and challenging, agreeing and disagreeing) clearly creates challenges and opportunities for role development in any group. As they learn to engage in this kind of talk without being disagreeable ­ to argue with ideas rather than with each other ­ they must navigate uncertain waters, and often they experience annoyance, hurt feelings, or anger in the process. The ways that a group negotiates this tension has significance for the inquiry practices that they develop, and for the roles that emerge for each student with respect to these practices. As the foregoing analysis suggests, understanding the learning that occurs in these contexts requires attention to the interactions among individual role development, conceptual agency, and emergent collective processes. CONCLUSION Learning to reason with inscriptions ­ to construct these images as representations of phenomena in the world, and employ them persuasively in argumentation ­ is a process of developing new modes of participation in communicative processes. We have seen that key challenges in this learning process ­ co-assembling a shared "representational state," and managing the social tensions of argumentation ­ are mediated by each student's emergent role in the group. Rather than treat this mediation as a causal mechanism (e.g. examining the impact of role development on conceptual change, mastery of skills, individual dispositions, etc), this study has examined role development as a site of science learning. What advantages do we gain by attending to this unit of analysis? Roles are improvised, malleable constructs that can change fairly quickly (as Juanita's did), as students position and re-position themselves, and one another, in activity. This plasticity of roles suggests that developing (or trying on) new roles may afford students the chance to explore possibilities for different "ways of being," different conceptions of what "kind of

Students' roles in group work


person" (Gee, 2002) they might be, and might become. At the same time, we have seen that roles have affordances and constraints for developing literacies of scientific reasoning, symbolizing, and other practices. Roles are observable in day-to-day classroom interactions, and as such thay can be tractable in instruction ­ note the use of stylized, highly-structured role assignments in cooperative learning (Johnson & Johnson, 1982; Kagan, 1992) and reciprocal teaching (Palincsar & Brown, 1984) strategies, as methods for scaffolding students' appropriation of specific practices. In contrast to such role assignments, the present study has attended to emergent roles within an activity system over time. Moving our attention from very local, temporarily-assigned roles (e.g. questioner in a reciprocal teaching group) toward more global, stable identities in larger life-worlds (e.g. daughter in a family; Latina in an ethnically-diverse elementary school; or geophyscist in a university department), we can see the emergent roles examined here as an important intermediate site of identity development. Repeated classroom experiences of playing the role of quiet bystander could mediate a student's development, over time, of a more stable identity as one-who-does-not-contribute-an-opinion. By contrast, an opportunity to develop this quiet bystander role toward a more participatory one could open up new possibilities for identity development in the longer term. An important advantage gained by studying the microgenetic emergence of roles lies in the ability to see the agency of students in this learning process. Though larger social forces clearly exert agency in positioning students within particular discursive identities or social categories (Gee, 2002; Holland & Leander, 2004), the analysis presented here shows students' own agency in constructing their roles, and their negotiation of roles in ways that preserved and extended their agency in the group. Each student exerted her influence on the group's emergent

Students' roles in group work


practices, pushing forward those modes of talk that afforded a more active voice (whether inside the science domain or outside), and avoiding those that did not. Specifically with respect to reasoning with inscriptions, students' appropriation of the GIS tool occurred within practices that afforded them agency in the group (Juanita as swing vote; Violeta as competitor), and did not occur within practices that constructed them as without agency (Violeta avoiding brainstorming; Kerry as "student"). Brown (2004) has similarly highlighted students' agency in the development of discursive identities, attending to African American students' complex negotiation of multiple, sometimes conflicting identities in learning to participate in scientific discourse: "[r]esearchers' use of macro perspectives of ethnic identity underemphasize the implications of students' individual agency [in identity development] ... [This understanding] can be enhanced by the microperspectives that research on individual assimilation into science classrooms can provide" (p. 811). The present study advances this goal, focusing on a particular activity system (small groups working with visual data) in order to better view a particular subset of science learning (reasoning with inscriptions). The construct of role development explored here provides a unit of analysis in which to examine such negotiations of identity as they unfold at the microgenetic level, within practices that are developed by the group. This highlights the fact that roles are sites of learning where the individual and the collective meet on common explanatory ground. Hollander & Leander (2004) emphasize the importance of understanding the relationships between social positionings (external impositions of identity categories) and subjectivities (internal conceptions of self in the world), and the need for analytical approaches that reveal "pivotal moments in which social and psychological phenomena come to interanimate and interpenetrate one another" (p. 127), constituting "the

Students' roles in group work


microproduction of a social position" (p. 128). The emergence of roles in the present study constitutes such a focus. Roles are sites in which students can explore possibilities for their emergent identities, which are unfolding on longer scales of time and at larger social geographies. They also can serve as a unit of analysis at which researchers can observe processes of learning that occur across the individual/collective divide. In light of current conceptions of scientific practice as distributed across individuals, communities, practices, and artifacts, this suggests significant value of such a focus for understanding how students learn to do science.

Students' roles in group work


REFERENCES Andriessen, J. (2005). Collaboration in computer conferencing. In A. O'Donnel, C. Hmelo, & G. Erkens (Eds.), Collaboration, reasoning and technology (pp. 277-321). Mahwah, NJ: Lawrence Erlbaum Associates. Andriessen, J. (2006). Arguing to learn. In K. Sawyer (Ed.) Cambridge Handbook of the Learning Sciences (pp. 79-96). New York NY: Cambridge University Press. Audet, R., and G. S. Ludwig. (2000). GIS in Schools. Redlands, CA: ESRI Press. Baker, M. J. (2004). Research on knowledge elaboration in dialogues. Synthese pour l'habilitation a diriger les recherches. Université Nancy 2. Bell, P. (2002). Science is argument: Developing sociocognitive supports for disciplinary argumentation. In T. Koschmann, R. Hall & N. Miyake (Eds.), CSCL2: Carrying forward the conversation (pp. 449-455). Mahwah, NJ: Lawrence Erlbaum Associates. Bowers, J., Cobb, P. & McClain, K. (1999). The evolution of mathematical practices: A case study. Cognition and Instruction, 17(1): 25-64. Brown, B. (2004). Discursive identity: Assimilation into the culture of science and its implications for minority students. Journal of Research in Science Teaching, 41(8): 810834. Clark, H. H. (1996). Using language. New York: Cambridge University Press. Cobb, P. (2002). Reasoning with tools and inscriptions. Journal of the Learning Sciences, 11, 187-216. Cornelius, L. & Herrenkohl, L. (2004). Power in the classroom: How the classroom environment shapes students' relationships with each other and with concepts. Cognition and Instruction, 22(4), 467­498. deJong, T. and W. R. vanJoolingen (1998). Scientific discovery learning with simulations of conceptual domains. Review of Educational Research, 68(2): 179-201. Dunbar, K. (1995). How scientists really reason: Scientific reasoning in real-world laboratories. In R. J. Sternbert & J. E. Davidson (Eds.), The nature of insight (pp. 365-395). Cambridge, MA: MIT Press.

Students' roles in group work


Dunbar, K. & Fugelsang, J. (2005). Scientific thinking and reasoning. In K. Holyoak and R. Morrison (Eds.), The Cambridge Handbook of Thinking and Reasoning (pp. 705-725). New York, NY: Cambridge University Press. Duschl, R. A., Schweingruber, H. A., & Shouse, A. W. (2007). Taking science to school: Learning and teaching science in grade k-8. Washington D.C.: National Academies Press. Engle, R. A., & Conant, F. (2002). Guiding principles for fostering productive disciplinary engagement: Explaining an emergent argument in a community of learners classroom. Cognition and Instruction, 20, 399­483. Enyedy, N. (2003). Knowledge construction and collective practice: At the intersection of learning, talk, and social configurations in a computer-mediated mathematics classroom. The Journal of the Learning Sciences, 12(3) 361-408. Enyedy, N. (2005). Inventing mapping: Creating cultural forms to solve collective problems. Cognition and Instruction, 23(4): 427­466. Gee, J. P. (2002). Identity as an analytic lens for research in education. Review of Research in Education, 25, 99-125. Goldman, S., Radinsky, J., & C. Rodriguez (2007). Teacher interactions with small groups during investigations: Scaffolding the sense-making process and pushing students to construct arguments with data. Paper presented at the Annual Meeting of the American Educational Research Association (AERA), Chicago IL, April 2007. Goldman, S., Sacay, R., Singer, M., Oliva, S., Allende-Pellot, F., Liceaga, I., & J. Radinsky (2006). Emerging conceptual understandings in GIS investigations. Paper presented at the Annual Conference of the American Educational Research Association (AERA), San Francisco CA, April 2006. Goodwin, C. (1994). Professional vision. American Anthropologist, 96(3): 606-633. Goodwin, C. (2000). Practices of seeing: Visual analysis: An ethnomethodological approach. In van Leeuwen, T. & Jewitt, C. (Eds.) Handbook of Visual Analysis (pp. 157-182). London: Sage Publications. Greeno, J. G. (2006). Learning in activity. In K. Sawyer (ed.) Handbook of the Learning Sciences, pp. 79-96. New York NY: Cambridge University Press. Greeno, J. & van de Sande, C. (2007). Perspectival understanding of conceptions and conceptual growth in interaction. Educational Psychologist, 42(1), 9-23.

Students' roles in group work


Hall, R. (1999). The organization and development of discursive practices for "having a theory." Discourse Processes, 27(2): 187-218. Holland, D. & Leander, K. (2004). Ethnographic studies of positioning and subjectivity: An introduction. Ethos, 32(2): 127-139. Hutchins, E. (1995). Cognition in the wild. Cambridge, MA: MIT Press. Johnson, R. T. & Johnson, D. W. (1982). Structuring cooperative learning. New Brighton MN: Interaction Book Company. Kagan, S. (1992). Cooperative learning. Resources for Teachers, Inc: San Juan Capistrano, CA. Kuhn, D. (1991). The skills of argument. Cambridge, MA: Cambridge University Press. Kuhn, D., Shaw, V., & Felton, M. (1997). Effects of dyadic interaction on argumentative reasoning. Cognition and Instruction, 15(3), 287-315. Latour, B. (1990). Drawing things together. In Lynch, M. and Woolgar, S. (eds.) Representation in Scientific Practice,19-68. Cambridge, MA: MIT Press Latour, B., & Woolgar, S. (1986). Laboratory life: The construction of scientific facts. Princeton, NJ: Princeton University Press. Lave, J. & E. Wenger (1991). Situated Learning: Legitimate peripheral participation. New York, NY: Cambridge University Press. Lehrer, R., & Pritchard, C. (2002). Symbolizing space into being. In K. Gravemeijer, R. Lehrer, B. van Oers, & L. Verschaffel (Eds.), Symbolization, modeling and tool use in mathematics education. (pp. 59-86). Dordrecht, Netherlands: Kluwer Academic Press. Lehrer, R., & Schauble, L. (2000). Developing model-based reasoning in mathematics and science. Journal of Applied Developmental Psychology, 21: 39­48. Lehrer, R., & Schauble, L. (2002). Symbolic communication in mathematics and science: Coconstituting inscription and thought. In E. D. Amsel & J. Byrnes (Eds.), Language, literacy, and cognitive development. The development and consequences of symbolic communication (pp. 167­192). Mahwah, NJ: Lawrence Erlbaum Associates, Inc. Lehrer, R., and Schauble, L. (2006). Cultivating model-based reasoning in science education. In K. Sawyer (Eds.), Cambridge handbook of the learning sciences (pp. 371 - 388). New York, NY: Cambridge University Press. Leitao, S. (2000). The potential of argument in knowledge building. Human Development, 43, 332-360.

Students' roles in group work


Lemke, J. L. (1998). Multiplying meaning: Visual and verbal semiotics in scientific text. In JR Martin & R Veel, Eds., Reading Science. London: Routledge (pp.87-113). Linn, M. & Hsi, S. (2000). Computers, teachers, peers: Science learning partners. London: Routledge. Loh, B., Reiser, B. J., Radinsky, J., Edelson, D. C., Gomez, L. M., & S. Marshall. (2001). Developing Reflective Inquiry Practices: A case study of software, the teacher, and students. In K. Crowley, C. Schunn, & T. Okada (Eds.), Designing for Science: Implications from Everyday, Classroom, and Professional Settings. Mahwah, NJ: Erlbaum. Lynch, M. (1990). The externalized retina: Selection and mathematization in the visual documentation of objects in the life sciences. In Lunch, M. & Woolgar, S. (Eds.), Representation in Scientific Practice (pp.153-186). Cambridge MA: MIT Press. National Research Council (1996). National science education standards. Washington, DC: National Academies Press. National Research Council (2006). Learning to think spatially. Washington, DC: National Academies Press New London Group (1996). A pedagogy of multiliteracies: Designing social futures. Harvard Educational Review 66(1): 60-92. Palincsar, A. S. & A. L. Brown (1984). Reciprocal teaching of comprehension-fostering and comprehension-monitoring activities. Cognition and Instruction, 1:117-175. Radinsky, J. (2000). Making sense of complex data: A framework for studying students' development of reflective inquiry dispositions. Learning Sciences. Unpublished doctoral dissertation, Northwestern University, Evanston IL: 349. Radinsky, J., Alamar, K., Leimberer, J., Rodriguez, C., & Trigueros, J. (2005). Science investigations with GIS: Helping students develop the need to know more. Spectrum: Journal of the Illinois Science Teachers' Association, 31(2), 34-42. Radinsky, J., Goldman, S. & Singer, M. (in press). Students' sense-making with visual data in small-group argumentation. Proceedings of the International Conference of the Learning Sciences (ICLS), Utrecht, the Netherlands.

Students' roles in group work


Radinsky, J., Leimberer, J. M., & L. M. Gomez. (2000). Reflective inquiry with complex data: A case study of dispositional learning. Paper presented at the Annual Conference of the American Educational Researchers Association (AERA), New Orleans, LA. Richmond, G. & Striley, J. (1996). Making meaning in classrooms: Social processes in smallgroup discourse and scientific knowledge building. Journal of Research in Science Teaching, 33, 839­858. Saxe, G. B. (2002). Children's developing mathematics in collective practices: A framework for analysis. Journal of the Learning Sciences 11(2-3): 275-300. Sfard, A. & K. McClain (2002). Analyzing tools: Perspectives on the role of designed artifacts in mathematics learning. Journal of the Learning Sciences, 11, 153-161. Singer, M., Radinsky, J. & S. Goldman (in press). The role of gesture in meaning construction. Discourse Processes Special Issue, Eds. Ludo Verhoeven and Arthur Graesser. Smagorinsky, P. (2001). If meaning is constructed, what is it made from? Toward a cultural theory of reading. Review of Educational Research, 71(1): 133-169. Tabak, I., (2004). Synergy: A Complement to Emerging Patterns of Distributed Scaffolding. Journal of the Learning Sciences, 13 (3): 305-355. Toulmin, S. (1969). The uses of argument. New York, NY: Cambridge University Press. vanJoolingen, W.R., deJong, T., & Dimitrakopoulou, A. (2007). Issues in computer supported inquiry learning in science. Journal of Computer Assisted Learning 23 (2), 111­119. Wenger, E. (1998). Communities of practice: Learning, meaning and identity. New York, NY: Cambridge University Press. Wertsch, J. V. (1998). Mind as action. Oxford, England: Oxford University Press. Wertsch, J. (2000). Vygotsky's two minds on the nature of meaning. In C. D. Lee and P. Smagorinsky (Eds.), Vygotskian perspectives on literacy research: constructing meaning through collaborative inquiry (pp.19-30). New York, NY: Cambridge University Press. Wertsch, J. V. & L. J. Rupert (1993). The authority of cultural tools in a sociocultural approach to mediated agency. Cognition and Instruction 11(3 & 4): 227-239. White, B., & Frederickson, J. (1998). Inquiry, modeling, and metacognition: Making science accessible to all students. Cognition & Instruction, 16, 3 ­ 118. Yin, R. K. (2002). Case Study Research, Design and Methods, 3rd ed. Newbury Park: Sage Publications.


Microsoft Word - studentroles_C&I.doc

66 pages

Report File (DMCA)

Our content is added by our users. We aim to remove reported files within 1 working day. Please use this link to notify us:

Report this file as copyright or inappropriate


Notice: fwrite(): send of 207 bytes failed with errno=104 Connection reset by peer in /home/ on line 531