Educ Stud Math (2014) v86 P 401-429

Educ Stud Math (2014) 86:401–429
DOI 10.1007/s10649-014-9532-8
Teacher support for collective argumentation:

A framework for examining how teachers support students’
engagement in mathematical activities
AnnaMarie Conner & Laura M. Singletary &

Ryan C. Smith & Patty Anne Wagner &
Richard T. Francisco
Published online: 1 February 2014

# Springer Science+Business Media Dordrecht 2014
Abstract We propose a framework for examining how teachers may support collective
argumentation in secondary mathematics classrooms, including teachers’ direct contributions
to arguments, the kinds of questions teachers ask, and teachers’ other supportive actions. We
illustrate our framework with examples from episodes of collective argumentation occurring
across 2 days in a teacher’s classroom. Following from these examples, we discuss how the
framework can be used to examine mathematical aspects of conversations in mathematics
classrooms. We propose that the framework is useful for investigating and possibly enhancing
how teachers support students’ reasoning and argumentation as fundamentally mathematical
activities.
Keywords Argumentation . Reasoning . Questioning . Teaching . Discussions
1 Introduction
Many authors have asserted or implied that participating in discussions is helpful for student
learning of mathematics (Hufferd-Ackles, Fuson, & Sherin, 2004; Krummheuer, 2000; Stein,
Engle, Smith, & Hughes, 2008). Multiple researchers have explored how to facilitate produc-
tive mathematical discussions (e.g., Baxter & Williams, 2009; Hufferd-Ackles et al., 2004;
Staples, 2007). Participating in discussions in a distinctively mathematical way can be framed
as collective argumentation, where collective argumentation involves multiple people arriving
at a conclusion, often by consensus. In particular, we examine collective argumentation in
classrooms in which the teacher and students work together to establish mathematical claims.
In this paper, we introduce the teacher support for collective argumentation framework for
A. Conner (*) : R. C. Smith : P. A. Wagner : R. T. Francisco

Department of Mathematics and Science Education, University of Georgia, 105 Aderhold Hall,
Athens, GA 30602, USA
e-mail: [email protected]
L. M. Singletary
Department of Natural Sciences and Mathematics, Lee University, Cleveland, TN 37311, USA
402 A. Conner et al.
examining how teachers may support collective argumentation in secondary mathematics

classrooms. We begin by acknowledging the difficulty of facilitating mathematical discus-
sions, describing recent work aimed to address some of those difficulties, and examining how
focusing on collective argumentation can provide even more information about the mathe-
matical importance of such discussions. We then describe our framework for teacher support of
collective argumentation1 and illustrate it with examples from our work with student teachers.
Finally, we propose that using the framework allows investigation of the teacher’s role in
supporting students’ reasoning and argumentation as fundamental activities of mathematics.
2 Background
2.1 Facilitating mathematical discussions
Research and accounts of personal experiences suggest that facilitating mathematical discus-
sions is difficult to do well (Hufferd-Ackles et al., 2004; Stein et al., 2008). Lobato, Clarke,
and Ellis (2005) suggested that teachers have been hesitant to say too much during these
discussions, fearing that could be interpreted as “telling” (p. 101), an action that they saw as
having potentially negative consequences (see also Chazan & Ball, 1999). Teachers’ hesitation
to tell—meaning giving broad hints or actual solutions—often leads to discussions in which
the apparent goal is to see multiple solution methods rather than to accomplish a larger
mathematical goal. Other authors have suggested that teachers might experience the dilemma
of balancing needs of the class against faithfulness to the mathematics (e.g., Brodie, 2010).
As suggested by recent policy documents and recommendations for teachers (e.g., Advi-
sory Committee on Mathematics Education, 2011; Martin, 2007; National Governors Associ-
ation Center for Best Practices & Council of Chief State School Officers, 2010), to facilitate
productive mathematical discussions, teachers must engage in behavior that helps students
build from their own understandings toward appropriate understandings of mathematical ideas.
According to Stein et al. (2008), this process begins with choosing appropriate tasks, ones with
high cognitive demand. By high cognitive demand, Stein et al. mean tasks that engage students
in complex thinking processes and have multiple solution paths or entry points. Much has been
written about engaging students in high cognitive demand tasks, including having discussions
as a whole class after small groups of students work on the tasks, but often the parts of the
discussion that are emphasized are the elicitation of multiple solutions or multiple solution
paths rather than the connections between mathematical ideas or the mathematical goal of the
lesson.
Researchers agree that the teacher plays a pivotal role in orchestrating mathematical
discussions, even, or especially, when he or she is not acting as an arbiter of mathematical
truth. Staples (2007), in her description of an experienced teacher establishing inquiry practices
as normative in her classroom, foregrounded one of the most important aspects of the teacher’s
role in a collaborative classroom: “The teacher…has considerable influence over whether or
not a student’s idea is elicited and done so that it can be taken up by the collective” (p. 174).
Boaler and Brodie (2004) argued that the questions a teacher asks influence the nature of a
1
Our framework captures the teacher’s actions in support of mathematical arguments in classrooms. It does not
distinguish actions that might be mathematically productive from those that might not be. Nor does it distinguish
actions that encourage more productive argumentation from others that might limit students’ participation.
However, it does point out the potential actions and contributions of the teacher and allows users of the
framework to draw conclusions about the mathematical and pedagogical potential of such actions.
Teacher support for collective argumentation 403
classroom discussion. Hufferd-Ackles et al. (2004) suggested that the teacher should create a
“math-talk learning community” (p. 81) in a classroom as one way to ameliorate the difficul-
ties of facilitating productive discourse, and Stein et al. (2008) described five practices that are
useful when facilitating discourse, particularly around cognitively demanding tasks.
Hufferd-Ackles et al. (2004) included four components in their framework for a math-talk
learning community: “(a) Questioning, (b) Explaining math thinking, (c) Source of mathe-
matical ideas, and (d) Responsibility for learning” (p. 87). For each of these dimensions, they
described four possible levels of the community. Hufferd-Ackles et al.’s framework provided a
description of a community in which important mathematical ideas could be discussed.
However, their framework does not foreground the importance of appropriate mathematical
reasoning or how the community knows when a mathematical idea has been appropriately
established. Their category of “explaining math thinking” comes closest in level 3, where
students are explaining and defending their thinking, but the actual ideas and reasoning being
used at level 3 are not apparent within the categories of their framework.
Stein et al. (2008) began to address the problem of discussions that do not highlight the
important mathematical ideas by breaking down the work of the teacher into more manageable
parts, which they call the “five practices model” (p. 314). Their model distinctly addressed the
problem of concluding a discussion with recognition that there are multiple ways to solve a
task, replacing that with a carefully sequenced set of student solutions that culminate with a
careful summary that connects the solutions with each other and with the important mathe-
matical ideas that were the intended focus of the task. Stein et al.’s model offloads some of the
decision making to the planning stage of teaching, which is very helpful, particularly for
beginning teachers. Their practices provide useful guidance for teachers to engage their
students in productive discussions. However, we find one element of mathematical thinking
not explicitly present in the practices to be important when thinking about moving toward
conversations that are mathematically productive. That element is understanding and recog-
nizing a mathematically appropriate argument.
Understanding, recognizing, and constructing mathematical arguments are important parts
of the disciplinary practices of mathematics. The National Council of Teachers of Mathematics
(NCTM, 2000) emphasized reasoning and proving as well as communicating in their Princi-
ples and Standards for School Mathematics; their more recent book series describing the
necessary components of the high school curriculum is titled Focus in High School Mathe-
matics: Reasoning and Sense Making (NCTM, 2009), indicating a continued focus on
reasoning, which is an essential part of argumentation. The Common Core State Standards
for Mathematics (CCSSM, National Governors Association Center for Best Practices &
Council of Chief State School Officers, 2010) include among the standards for mathematical
practice, “Construct viable arguments and critique the reasoning of others” (p. 7). In addition,
the authors of the CCSSM assert that being able to justify or derive mathematical statements,
an essential part of mathematical argumentation, is an indication of mathematical understand-
ing. One of the goals of these policy documents is to ensure that students engage in practices
that are foundational to the discipline of mathematics. Reasoning and proof are unequivocally
accepted to be foundational to the discipline of mathematics, but engaging students in formal
deductive proof at an early age may not be developmentally appropriate. Argumentation, as a
precursor to proof, is fundamental to the establishment of mathematical knowledge.
When analyzing a discussion, teachers and other mathematics educators need to think about
the disciplinary practices of mathematics. How did that discussion allow students to participate
in the disciplinary practices of mathematics, particularly with regard to the establishment of
mathematical claims or the reasoning practices that are important in mathematics? How can
teachers foreshadow and eventually lead to thinking about deductive reasoning and proof? We
believe focusing on collective argumentation allows mathematics educators to think about

what makes a discussion distinctively mathematical. In particular, by separating the different
components of an argument, our analysis of collective argumentation allows an examination of
the kinds of reasoning used in establishing claims (by examining the warrants in arguments),
how the students contribute to that reasoning, and what the teacher is doing to support students
in both making and establishing mathematical claims.
2.2 Focusing on collective argumentation
Recent work in mathematics education has highlighted collective argumentation as an impor-

tant part of classroom discourse. Building on Toulmin’s (1958/2003) model of argumentation
in multiple fields, mathematics educators, following Krummheuer (1995), have examined
collective argumentation in classroom settings. This avenue of research is an extension of
Toulmin’s work, as he examined argumentation in the traditional sense of one person
convincing an audience of the validity of a claim. Some work in mathematics education
examines individual construction of arguments (e.g., Hollebrands, Conner, & Smith, 2010;
Inglis, Mejia-Ramos, & Simpson, 2007), but we build on studies addressing collective
argumentation. Current work in collective argumentation involves examining student learning
through that lens (Krummheuer, 2007) as well as examining how ideas become taken “as-if-
shared” (Rasmussen & Stephan, 2008, p. 196).
An argument, as described by Toulmin (1958/2003) and currently used in the field, involves
some combination of claims (statements whose validity is being established), data (support
provided for the claims), warrants (statements that connect data with claims), rebuttals
(statements describing circumstances under which the warrants would not be valid), qualifiers
(statements describing the certainty with which a claim is made), and backings (usually
unstated, dealing with the field in which the argument occurs). Toulmin conceptualized an
argument as occurring with a specific structure (see Fig. 1) in which these parts of arguments
relate to one another in specific ways. In practice, arguments are often more complicated, in
that, for example, statements offered as data may also need defense, thus functioning as both
data in one argument and claim in a subargument.2
In our research, we respond to the call by Yackel (2002) to examine the teacher’s role in
collective argumentation by focusing on both the parts of arguments he or she provides and the
teaching moves that prompt or respond to parts of arguments provided by students. Our
research examines how student teachers, as teachers who are beginning their professional
experience, support their students as they engage in collective argumentation.
We define collective argumentation very broadly to include any instance where students
and teachers make a mathematical claim and provide evidence to support it. An example from
our data can be seen in Fig. 2 where a teacher and her students were discussing the
characteristics of a regular polygon during a whole-class discussion. They used squares as a
prototypical example of regular polygons and made the claim that “the angles are never going
to change in a regular polygon, even if the sides do.” In our discussion of collective
argumentation, we include how teachers support collective argumentation in whole-class
discussions as well as how they facilitate small-group discussions. Our definition of collective
argumentation builds on a distinction introduced by Toulmin (1958/2003) and emphasized by
2
Toulmin (1958/2003) called these preliminary arguments or lemmas (p. 90). We use subargument to indicate
that these may not come temporally before the other argument and to avoid lemma, which has a specific
mathematical meaning.
Fig. 1 Diagram of a generic argu-

ment (adapted from Toulmin, Rebuttal
1958/2003) Unless
So
Data Qualifier Claim
Since
Warrant
On account of
Backing
Krummheuer (1995) between “analytic and substantial arguments” (Toulmin,

1958/2003, p. 125). An analytic argument is perhaps best described as one corre-
sponding with proof in mathematics; it is a “logically correct deduction…[that] contains in its
conclusion nothing that is not already a potential part of the premises” (Krummheuer, 1995, p.
235), while substantial arguments “expand the meaning of such propositions insofar as they
soundly relate a specific case to them by actualization, modification, and/or application”
(Krummheuer, 1995, pp. 235–236). We follow Toulmin and Krummheuer in our rejection
Is it a regular
polygon?
Data Data/Claim
Angles Both squares What’s going to
congruent, are regular stay the same?
sides polygons
congruent
Claim
The angles are
Warrant
never going to
Definition
change in a
of a regular
regular
polygon
polygon, even
Wrote on if the sides do
board
Data Restated: The
Data/Claim
Drawings of angles (are going
You don’t have
2 squares on to have a to stay the same)
the board certain size Warrant
If it is true for
So I guess squares, it is true
Warrant for polygons
what I’m
Only the side
trying to say,
lengths are
do you want to
different
say it to me,
What is the M, you might
difference say it better. - Student
between these - Ms. Bell
two squares? - Both
Fig. 2 Example of argument in Ms. Bell’s class

of analytic arguments (proofs in mathematics) as the only valid arguments and echo
Krummheuer’s contention:
As Toulmin strongly emphasized, a substantial argumentation should not be subordi-
nated or related to an analytic one in the sense that the latter is the ideal type of arguing
and that one can always identify in substantial arguments the logical gulf in comparison
to an analytic one. Substantial argumentation has a right by itself. By substantial
argumentation a statement or decision is gradually supported. This support is not
conducted by a formal, logically necessary conclusion, nor by an arbitrary edict such
as declared self-evidence, but is motivated by the accomplishment of a convincing
presentation of backgrounds, relations, explanations, justification, qualifiers, and so on.
(p. 236)
To examine teachers’ support for collective argumentation, we modified Toulmin’s

(1958/2003) model to include teachers’ actions that were not directly components of the
arguments and to signify who contributed the various components of arguments (as described
in Conner, 2008). We use color and line style to indicate whether the teacher, students, or the
teacher and students interactively contributed each component of an argument, as seen in
Fig. 2 (the teacher contributed the data “angles congruent, sides congruent”; a student
contributed the warrant “only the side lengths are different”; and the teacher and students
jointly contributed the claim “the angles are never going to change in a regular polygon, even
if the sides do”). We also include parts of arguments that are not explicitly stated by the teacher
or her students and therefore must be inferred from the surrounding context of the classroom
and the content of the argument. We call these implicit parts, and we indicate them with the
clouds in the diagrams (the warrant “if it is true for squares, it is true for polygons” was
inferred from the context). This implicit warrant provided the generalization, connecting the
data from specific observations of squares to a more general claim about regular polygons.
Finally, we include contributions by the teacher that are not components of the arguments but
that prompted or responded to parts of the arguments. We represent these by talk bubbles that
are connected to the argument components (the teacher’s question “Is it a regular polygon?” is
an example).
Our adaptation of Toulmin’s (1958/2003) model relies on the construct of subarguments to
build arguments. Toulmin allowed for subarguments in his model, but he did not
elaborate on their status in an episode of argumentation, nor did he illustrate the
structure of an argument that included a subargument. He simply stated that they
could occur when part of the argument was questioned. Figure 2 illustrates an
argument that includes two subarguments. The main claim is “the angles are never
going to change in a regular polygon, even if the sides do.” That claim is supported
by two data that also serve as claims in subarguments: “Both squares are regular
polygons” and “You don’t have to have a certain size square.” Subarguments may
arise in two different ways. In the first, they can be preliminary to a claim, with the
argument building from one claim into another, so that the argument occurs from left
to right in the diagram. In the second, a component, such as data or warrant, is
questioned, resulting in it also becoming a claim. In either case, one component
serves two purposes: data and claim, or claim and warrant. We label these parts as
Data/Claim or Warrant/Claim and examine their functions both separately from the
data, claims, and warrants and combined with the appropriate individual parts. To
streamline the diagrams, we insert the contributions that serve two purposes (such as
both data and claim) only once.
2.3 Focusing on teacher support
To focus on how teachers support collective argumentation, we pay particular attention to the
teacher’s direct contributions of argument components, the questions posed to prompt argu-
ment components, and the other supportive actions used to facilitate the development of an
argument. We differentiate among these different types of support in the same way other
researchers have identified and described particular teaching actions used to orchestrate
mathematical discussions (e.g., Forman, Larreamendy-Joerns, Stein, & Brown, 1998; Lobato
et al., 2005; Staples, 2007). For example, Forman et al. (1998) explored the role of the teacher
primarily in the way the teacher revoiced and framed the argument. Lobato et al. (2005)
provided a broad perspective and examined the kinds of teaching actions used to support the
introduction of new information and mathematical concepts. Among these teaching actions,
she and her colleagues included actions in which the teacher contributed aspects of the
discussion, asked questions, and encouraged student thinking. Similarly, Staples (2007)
characterized the role of the teacher in support of whole-class inquiry as follows: guiding
the mathematics, establishing and monitoring a common ground, and supporting students in
making contributions (p. 172). In addition, Boaler and Brodie (2004) and Franke et al. (2009)
emphasized the role of questions in engaging students in discussions. Although distinct
parallels exist among our research and that of others, we believe our framework builds upon
and extends the actions that other researchers have identified in a way that specifically
characterizes how teachers support collective argumentation as an important subset of their
facilitation of classroom discourse.
3 Context
In the following sections, we describe the teacher support for collective argumentation framework
that arose from our examination of teachers’ support for collective argumentation in mathematics
classrooms. To contextualize our framework, we describe the study, our data collection, and how
we coded our data and conceptualized the framework. We illustrate the framework in the context
of one teacher’s practice, using episodes of argumentation from two class periods of her ninth-
grade class as primary sources for the examples we present. We use those particular lessons
because they contain representative examples of unique features of the framework.
3.1 The study
The framework is based on data collected in a study that explored prospective secondary
mathematics teachers’ beliefs about mathematics, teaching, and proof and how they supported
collective argumentation during their student-teaching experiences. Participants in the study
were students in two of the first author’s teacher preparation courses, one of which was
primarily mathematical and the other primarily pedagogical. These courses emphasized,
among other things, implementing good tasks and facilitating mathematical discourse (for
more information on the courses, see Conner, Edenfield, Gleason, & Ersoz, 2011). The
participants were given broad descriptions of the purpose of the study, because we wanted
to minimize our influence on what was said in interviews or enacted during student teaching.
Our analysis of two participants, Ms. Bell3 and Ms. Carr, during their subsequent student-
teaching experience led to the development of the conceptual framework.
3
All participant names are pseudonyms.
Ms. Bell and Ms. Carr and the school in which they were placed allowed us to video record
a unit of each of their instruction. In the case of each of the other prospective teachers who
participated in the first part of the study, either the participant or the school in which he or she
was placed did not allow video recording. Thus, our data were limited to these two student
teachers. However, Ms. Bell and Ms. Carr were quite different from each other in their beliefs
and preferred methods of teaching, allowing us to see different teaching styles in the two
classes. Ms. Bell’s classes often included small group activities followed by whole class
discussions; in contrast, Ms. Carr usually led whole class discussions interspersed with
lecture-based notes.
Some readers may wonder about the relevance of basing a framework about teaching on
observations of student teachers, arguing that student teachers are only learning to teach and
have not developed the more mature teaching skills of experienced teachers. We acknowledge
that these are novice teachers who might not enact instruction with the facility of some teachers
with more experience. However, while their teaching moves may not have been as mature and
well-developed as some experienced teachers, they did enact a wide range of teaching
behaviors. They particularly asked a wide range of questions, perhaps because of the emphasis
on questioning in their methods course. Thus, we believe that their units of instruction provide
a good foundation for our framework of support moves. This contention is further strengthened
by our application of the framework to published episodes of collective argumentation as
described in Section 3.3.
3.2 Data collection
For their culminating field experience, the student teachers were placed with different mentor
teachers in the same rural high school. We conducted approximately 2 weeks of observations
in each classroom, capturing an entire mathematics unit. To collect our data, a member of the
research team videotaped each of the observed class periods, focusing specifically on the
actions of the student teacher, who acted as the classroom teacher during our observations. We
also took relevant field notes and collected the tasks and worksheets used during instruction.
3.3 Data analysis
We fully transcribed the video recording of each class, also noting actions such as pointing and
other meaningful gestures. The transcripts and video recordings were used as primary data
sources, supplemented by the field notes and artifacts. We divided our research group into
subgroups to analyze the transcripts for arguments using Toulmin’s (1958/2003) model of
argumentation. We will illustrate our methods by describing the process of developing the
diagram depicted in Fig. 2.
Each subgroup worked on developing diagrams for an entire day’s instruction. This was to
ensure that the flow of the classroom interaction was uninterrupted analytically and each
argument was considered within the context of that day’s activity. For example, the diagram in
Fig. 2 was situated in a class day devoted to extending the procedure for finding the sum of the
interior angles of a polygon to the process of determining the measure of a single interior angle
of a regular polygon.
After reading through the transcript, we identified episodes of argumentation and tentative-
ly identified argument components. We purposefully ignored nonmathematical classroom talk
and segments of talk that were largely definitional (such as deciding how many sides a
nonagon has), and we did not attempt to evaluate the mathematical correctness of the
arguments. In focusing on instances where students and teachers made a mathematical claim
and provided evidence to support it (our definition of collective argumentation), we were

interested in the students’ and teacher’s understandings rather than our own. We looked for
natural breaks in the discourse, which helped to identify episodes of argumentation. In our
example, Ms. Bell reiterated the claim and said, “Now how do we find, how do we find these
angles in a regular polygon?” This question prompted a shift in focus, signaling a new episode
of argumentation.
To identify components, we first looked for the main claim of the argument. The main
claim in Fig. 2, the angles in a regular polygon are never going to change, even if the sides do,
was preceded by the linguistic cue, so. Because the diagrams are not temporal representations,
we then examined the transcript both forwards and backwards to determine the information the
students and teacher started with (the data). In our example, Ms. Bell had initiated the
discussion by reiterating the class’s definition for regular polygons and attributed it to a
drawing of a square that she had on the board. She then drew another square and posed the
question, “What is different between these two squares?” The two squares were necessary
components for the discussion that followed, establishing them as data. Ms. Bell’s statement of
the definition ensured that the characteristics, angles congruent, sides congruent, were
foregrounded in the reasoning that followed, establishing her statement also as data. It is
important to note that neither the drawings of squares nor Ms. Bell’s statement about their
characteristics was questioned, further supporting their status as data. We then identified other
statements (oral, written, and even physical actions) within the episode of argumentation,
characterized them as warrants or data/claims, and established the structure of the argument.
We noted instances when there were no warrants bridging the data and claim and made
inferences based upon what the teacher and students might have meant from the other parts of
the argument, including the other warrants explicitly contributed, and contextual details, such
as what was written on the board or the teacher’s questions. In our example, the students and
teacher had established both claims for the squares (both squares are regular polygons and you
don’t have to have a certain size [of a square for it to be a regular polygon]) and then used
those claims as data for the final claim. However, there was no statement that served as a
warrant between that data and the final claim, prompting us to infer the warrant, if it is true for
squares then it is true for polygons.
This pattern held true for all of the diagrams we created during our analysis of the
transcripts. After the subgroup had completed the diagrams from the transcript and each
member of the research team had read the transcript, the entire research team vetted the
proposed diagrams, proposing, debating, and resolving alternative interpretations. This process
resulted in 277 final diagrammed episodes of collective argumentation across the two student
teachers’ units of instruction. As previously described, we coded the diagrams with color and
line style to record whether the teacher, the students, or both teacher and students together
contributed a given argument component (as in Conner, 2008). Shifting our attention to teacher
support, we defined it as any teacher move that elicited or responded to an argument
component4. We captured questions and other supportive actions by reading through the
transcripts and reviewing each class video, carefully watching for gestures and writings on
the board that fit our definition of teacher support. These actions were placed within the
diagram in talk bubbles connected to the relevant component (see Fig. 2). Diagramming the
argument was crucial to identifying a move as supporting argumentation. Not all actions of the
teacher were considered support moves. To illustrate how our definition of teacher support
4
Note that a move did not have to be productive, nor did it have to be part of a productive or mathematically
correct argument to be supportive. We interpreted a move as supporting collective argumentation if it elicited or
responded to a component of an argument.
guided our analysis, we give an example of a teacher move that did not fit within our definition
and was therefore not captured. During the argumentation episode depicted in Fig. 2, after
drawing the second square on the board, Ms. Bell instructed the students to “pretend that’s a
square.” We did not capture this comment in a talk bubble because it did not elicit any
component and it accompanied, rather than responded to, the data component. As such, Ms.
Bell’s comment served to establish the figure as a square, which was captured in the data
component. Thus, in order to be considered to be supportive of collective argumentation, an
action had to elicit or respond to an argument component.
The research group analyzed each type of support separately. We used Toulmin’s (1958/2003)
model to determine the classifications for the direct contributions (e.g., claim, data, and warrant).
Because we were interested specifically in teachers’ support for collective argumentation, we
chose an inductive approach to our analysis of the teachers’ actions beyond direct contributions,
resulting in two types: questions and other supportive actions. Other research has developed
frameworks for teacher questions in the context of mathematical discussions, but we had no
assurance that what we found for collective argumentation would mirror their results. Thus, we
did not start with those questions identified for discourse in general, but searched our data for the
actions that were specific to supporting collective argumentation. We inductively developed
codes for questions and other supportive actions and then collapsed the codes into meaningful
categories. For example, in our analysis of the teachers’ questions, we carefully reviewed each
question that prompted a part of an argument, developed codes for the kinds of questions and
definitions for each, and arranged the collection of questions and codes in multiple ways until we
developed a more refined and coherent picture of the various kinds of questions asked to support
collective argumentation. Then, we looked across the 15 kinds of questions to determine if there
were themes existing among them. Through this process, we collapsed the various kinds of
questions into five categories of questions used to support collective argumentation. The same
process was used to analyze the other supportive actions. After developing our framework, we
examined episodes of collective argumentation at various grade levels and with a variety of
mathematical content reported in the mathematics education research literature (in particular, we
examined arguments reported in Forman et.al, 1998; Knipping, 2003, 2008; Krummheuer, 2007;
McClain, 2002; McCrone, 2005; Weber, Maher, Powell, & Lee, 2008; Wood, 1999; Yackel,
2002), and we found that the teachers’ actions in these episodes could be categorized with our
framework. In addition, when we examined other researchers’ characterizations of the actions of
the teachers in their work, we found that our framework similarly identified specific teaching
actions, and, in some cases, provided more specific insight into the purpose of the move. The
three types of support, along with the categories of each and codes within each category,
constitute the teacher support for collective argumentation framework5.
3.4 Description of the focus class periods
We focus our examples and explication of the framework on one of the teachers, Ms. Bell. We
observed Ms. Bell teaching one geometry unit (seven 90-min classes) of a ninth-grade
accelerated mathematics course. The high school was undergoing a change in its mathematics
curriculum, with the ninth-grade course in its second year of implementation of an integrated
approach to secondary mathematics. Ms. Bell’s instruction usually involved students working
in small groups on a mathematical task. She acted as facilitator of these small-group
5
The three types of support comprise the actual framework. We make the assumption that in order to use the
framework most robustly, a user would diagram the episode(s) of argumentation of interest prior to applying the
categories of the framework.
explorations, rotating between groups. Small-group work on tasks was followed by whole-
class discussions of the task’s mathematical content later in the class period or the next day.
Ms. Bell’s students were motivated and generally considered “good students,” but they were
not considered “gifted.” They had not taken an advanced eighth-grade mathematics course, but
this ninth-grade course would allow them to progress through the high school mathematics
curriculum at a pace that would end with calculus.
We chose episodes of argumentation from two consecutive classes toward the beginning of
the geometry unit. These two classes illustrate most of the kinds of activities in which Ms. Bell
and her students engaged. Because of space considerations, we include complete diagrams of
only two of the arguments that occurred on these days. As in the diagramming of any episode
of argumentation, we present a reconstruction of the final argument; diagrams do not represent
the temporal order in which the argument occurred (see also Krummheuer, 1995).
Day 1 began with time for small groups of students to measure the exterior angles of a set of
polygons provided on a worksheet and reproduced on the board at the front of the class. Each
group of students used a protractor to measure the exterior angles of one polygon and found the
sum of those measures, recording their measures and sums on the board. The class concluded
that the sum of the exterior angles of any polygon was 360°, and Ms. Bell posed the problem of
finding the sum of the measures of the interior angles of any polygon. The students worked on
this problem in small groups, first finding the measure of each interior angle of the given
polygons, then finding the sums, and finally looking across the sums for a general rule or
formula to calculate the sum of the measures of the interior angles of any polygon.
The first episode of collective argumentation that we highlight occurred toward the end of
the time during which the small groups of students were looking for a formula to calculate the
sum. Ms. Bell approached one of the small groups of students to ascertain their progress.
Episode 1 (diagrammed in Fig. 3) began with Ms. Bell asking, “Did you do your table? What
So if it goes up I think that’s a

by the same valid point, that
Start by Restated: 2 sides are going
amount each
finding the Linear to be 0
time, what kind
difference of function is it?
1 Data Rephrased: 2 Data
Chart It’s a linear But what do
(table)
3 Data/Claim 4 Data/Claim Start with 3, Does it
It goes up by function since you we want to get
on It’s a linear as our output? make sense?
paper 180 each time function can’t get a
polygon with
2 or 1 8 Data/Claim 11 Claim
When we The function is
plug in 3, we f(x) = 180x – 360
Keep get 540 but
going we want 180
Warrant 7Data/Claim
Pattern Okay It will be Let’s see if
180x and that works.
5Data/Claim
Warrant Qualifier then
180 is the
Functions that go up I think something
slope 10 Warrant
by the same amount else
To get from 540
each time are linear
Write that to 180, subtract
functions
down. 180 -- 360
All right. These are
Did you do really good – You f(x) is 180x and
your table? guys are making work with this So, how do we
really good steps. get from 540
6 Data/Claim to 180?
From the table,
it is going up
by 1 and then Warrant 9 Warrant
180 Understanding 3(180) = 540
of linear on chart What if we do 3
functions, times 180, what
Warrant f(x) = mx + b are we going to
180 is the Restated: get?
Warrant constant rate of 540
Pattern change
- Student
- Ms. Bell
- Both
Fig. 3 Diagram of day 1, episode 1

is that?” Martin described the table, and there were a few back-and-forth comments about the
table and its contents, which were the number of sides and sums of measures of interior angles
of several polygons. (These are the data labeled 1 in the diagram in Fig. 3.) During this time,
Martin pointed at the table and said, “Two sides equal zero, three sides … I should probably
just ignore that one [referring to two sides].”
Ms. Bell: I think that’s a valid point, that two sides is going to be zero.
Adam: And if you have no sides, it’s not even an angle. [Data labeled 2.]
Ms. Bell: Mmhm, let’s think about it. I think you should start by finding the difference.
Martin: It goes up by 180 each time. [Data/claim labeled 3.]
Ms. Bell: Okay.
Adam: Each one, yeah, and then, uh—.
Ms. Bell: So if it goes up by the same amount each time, what kind of function is it?
Karin: Linear [Data/claim labeled 4.]
Ms. Bell: Linear
Martin: So, it’s something
Ms. Bell: It’s a linear function.
Karin: I think the slope is going to be 180 [Data/claim labeled 5.] because if you look at
the table, it’s going up by one and then by 180. [Data/claim labeled 6.]
Ms. Bell: Okay.
Karin: So it’s 180x, and then something to make it so it’s not—[Data/claim labeled 7.]
When you plug in 3, it doesn’t equal like [inaudible] or whatever.
Ms. Bell: All right. These are really good. You guys are making really good steps. Write
that down, 180−f (x) is 180x, and work with that. Keep going from there, really good
points.
Martin: We’ve got 540, but we want to get 3―I mean 180. [Data/claim labeled 8.]
Ms. Bell: So, what if we do 3 times 180? What are we going to get?
Martin: 3 times 180, 540. [Warrant labeled 9.]
Ms. Bell: 540, but what do we want to get as our output?
Martin: 180.
Ms. Bell: So, how do we get from 540 to 180?
Karin: I think you’ve got a defective one though.
Adam: That was the first one.
Martin: Subtract 360. [Warrant labeled 10.]
Ms. Bell: Let’s see if that works.
Martin: I got it.
Ms. Bell: Does it make sense?
Martin: Mmhm [yes].
Adam: 180x minus 360 [Claim labeled 11.]
Ms. Bell: I don’t want you to just write it down.
Karin: That’s what I got. I was about to graph it.
Ms. Bell: I want you to understand it. Martin, I want you to do me a favor and explain it
to everyone at this table, okay?
After episode 1, Ms. Bell left this small group and worked with a few other groups. At the
end of class, Ms. Bell asked two of the groups to share their formulas for the sum of the
measures of the interior angles of a polygon. The first group shared two versions of the
formula: f(n)=(n−2) 180 and f(n)=180n−360. Their data were the sums of interior angle
measures listed on the board, and they used “trial and error” as their warrant for the formula.
The second group was the group consisting of Karin, Adam, and Martin. They reported their
formula as f(s)=180s−360, and their argument essentially mirrored the one constructed from
their small-group conversation, emphasizing that this function was linear. This first class
period ended with consensus that the sum of the measures of the interior angles of a polygon
was f(s)=180s–360, where s is the number of sides of the polygon. This episode is presented
as an example of the collective argumentation in Ms. Bell’s class. Clearly, the students were
noticing patterns and making conjectures about mathematical relationships. In so doing, they
were engaging in inductive reasoning. We do not claim that this episode in any way
demonstrates the deductive reasoning necessary for mathematical proof. Rather, the students
were engaged in inductive reasoning appropriate for their task, and the teacher supported their
reasoning in constructive ways.
Day 2 began with the students finding the sum of the measures of the angles of a 12-sided
polygon as a quick review of the previous day’s work. Ms. Bell then asked the students to
think about another way to find the sum of the interior angles of any polygon by subdividing
several polygons into the least number of triangles. Together, she and the class connected the
formula from the end of day 1 to the process of subdividing polygons into triangles. After
making that connection, Ms. Bell introduced the term regular polygon and asked the students
to think about what regular meant in everyday language, eventually connecting their intuitive
notions to a class-constructed definition of “all sides congruent, all angles congruent.” From
this definition, the class came to the understanding that all regular quadrilaterals will have
congruent angles, all regular pentagons will have congruent angles, and so on, even though the
side lengths may change.
At this point in the lesson, a student made a comment that led to an investigation by the
class that Ms. Bell had not anticipated. Karin asked a question in which she compared a
rhombus to a square and asked if anyone could predict what the angle measures in a
rhombus would be, acknowledging that they would be different from 90°. Ms. Bell
rephrased Karin’s question for the class, which is the beginning of the following
excerpt. This highlighted episode, episode 2 (diagrammed in Fig. 4), illustrates the
interactions of Ms. Bell and her students when they were investigating a question that had been
posed by a student.
Ms. Bell: So Karin, you’re saying, we know that the interior angle sum is 360. [Data
labeled 1 in the diagram.] We know that these angles are the same and these angles are
the same [points at opposite angles in the rhombus drawn on the board]. [Data labeled 2
in the diagram.] And we know that these angles [points at the same side angles] sum to
what?
Micah: 180 [Data labeled 3 in the diagram.]
Ms. Bell: 180. Do we know anything else? Can we use that information to find what the
interior angle of any rhombus is going to be? [8 s pause]
Pointed to
opposite
Restated
angles
2 Data Wrote on
Opposite board: labeled
angles in a x and y
rhombus are
Restated:
congruent 4 Data/Claim x plus y So what
Diagram Write me equals Wrote does that
an equation Restated (sic) 360 Wrote on Now what on mean?
y x using those board do I have? board
x y 7 Data/Claim 14 Claim
5 Data/Claim 9 Data/Claim We can’t
And we know 2x + 2y = 360 2(x + y) = 360 x + y = 180
these angles determine the
sum to what? specific angles
Wrote on Rephrased:
So can I take 2 in a rhombus
1 Data board x + y = 180
3 Data out? What’s left
Interior on the inside?
Consecutive
angle sum is Pointed back and
angles in a
rhombus sum 360 degrees forth at the 2’s in Now what
to 180 the equation Restated: can I do?
Restated The 2?
6 Warrant 8 Warrant
Pointed to
2 is a common Divide by 2
interior
Warrant factor so we
angles Warrant Restated:
We can add can take it out Can I tell
Definition and Divide by 2
properties with angles to get the specifically
Do these have
parallel lines interior angle sum what x and y
anything in
are?
common?
10 Warrant/Claim
11 Data 12 Data/Claim Can’t determine
Separate x and y = 180 – x general value for y
y on the other (don’t know what x is)
side (in x + y =
So y
180)
equals?
Restated:
On the Warrant Restated:
- Student other side Subtract x 13 Warrant Plug in x
- Ms. Bell Now what am Plug in x
- Both I going to do x = 1, y = 179
with that? x = 2, y = 178
Fig. 4 Diagram of day 2, episode 2

Angela: You should.
Ms. Bell: You should? Should be able to, hmm. [5 s pause] I don’t know. I’m asking
y’all, I really don’t know.
Micah: Let’s test it.
Ms. Bell: You want to test? You want to draw any rhombus on your paper? You guys
want to do that? You might need to use a straight edge. [Students work on drawing
rhombi for 27 s.] So these angles, we could call them x [labels two of the congruent
angles x in the rhombus on the board.] Call these two angles y [writes y in the other two
angles in the rhombus.] [Data/claim labeled 4 in the diagram.] So, write me an equation
using those.
Martin: 2x+2y=360. [Data/claim labeled 5 in the diagram.]
Ms. Bell: 2x+2y=360 [writes 2x+2y=360]. Now, what can I do?
Travis: You could do 2x squared.
Ms. Bell: 2x squared?
Travis: Plus 2xy equals—.
Adam: No, you could just do—.
Travis: Never mind, I was thinking like that.
Ms. Bell: Area?
Adam: a squared.
Travis: a squared plus b squared equals c squared.
Ms. Bell: Oh, oh, okay, interesting.6 Do these have anything in common? [Ms. Bell
indicates 2x and 2y]
Micah: I got a good one.
Karin: Two [Together with Ms. Bell’s contribution, the warrant labeled 6 in the diagram.]
Ms. Bell: The two? So can I take a two out? What’s left on the inside?
Karin: x plus y. [Together with Ms. Bell’s contribution, the data/claim labeled 7 in the
diagram.]
6
This statement and the preceding eight lines did not add to the argument but are included here in the transcript
for the sake of completeness and to illustrate that when diagramming arguments, there are decisions that must be
made concerning what actually contributed to the argument.
Ms. Bell: x plus y equals (sic) 360 [writes 2(x+y) = 360]. Now, what can I do?
Karin: Divide by two [The warrant labeled 8 in the diagram.]
Ms. Bell: Divide by two. Okay. Now what do I have?
Micah: Equals 180. [The data/claim labeled 9 in the diagram.]
Ms. Bell: x plus y equals 180 [writes x+y=180]. But can I tell what x and y specifically are?
Adam: Not so. [Together with Ms. Bell’s contribution and Micah’s later contribution,
part of warrant labeled 10 in the diagram.]7
Martin: You could separate them and get x on the other side. [Data labeled 11 in the
diagram.]
Ms. Bell: x on the other [side], so y equals?
Martin: 180 minus x. That’s what I got. [Data/claim labeled 12 in the diagram.]
Ms. Bell: Now, what am I going to do with that?
Martin: You’re going to plug in x. [Part of warrant labeled 13 in the diagram.]
Ms. Bell: Plug in x.
Micah: But we don’t know x. [Part of warrant labeled 10 in the diagram.]
Ms. Bell: So what does that mean? How many—I can plug in values for x
Martin: x equals 1; y equals 179. [Part of warrant labeled 13 in the diagram.]
Ms. Bell: Umhmm.
Martin: If x equals 2, y equals 178 [Part of warrant labeled 13 in the diagram.]
Ms. Bell: So, I guess what we’re saying is we really can’t, we can’t tell, I don’t think all
rhombuses have the same measures for x and y. [Claim labeled 14 in the diagram.] That’s
what this is saying, right? Okay. We just got really off task, right there. But that was
interesting.
Martin: But I learned something.
After this digression, Ms. Bell again engaged the whole class in investigating the previous
question of finding the measure of one angle in a regular polygon. The students concluded in
7
Although this statement was made by a student, Ms. Bell’s preceding question suggested only one possible
answer. Therefore, the component was attributed to both student and teacher.
specific cases that to find the measure of one angle of a regular n-gon, they should divide the
sum of the measures, found with their previous formula, by n. From this conclusion, Ms. Bell
asked if they could find the measure of one exterior angle of a regular polygon. The students
concluded that they could, using the examples of a regular hexagon and a regular pentagon.
For the rest of the class period, Ms. Bell set small groups of students to the task of proving that
the sum of the measures of the exterior angles of a polygon is 360°.
4 How teachers support collective argumentation
The teachers in our study used three types of support for collective argumentation: directly
contributing argument components, asking questions that elicited parts of arguments, and
using other supportive actions. These three types of support make up the teacher support for
collective argumentation framework (see Table 1) and are described in subsequent sections.
Even though each part of the framework can be used separately and analysis of a teacher’s
practice using each type of support can yield interesting results, we believe the power of the
framework can be seen when the three types of support are analyzed together in the context of
collective argumentation. We assume that to use the framework to analyze the particularly
mathematical aspects of classroom discourse, a researcher would diagram the argumentation
prior to applying the framework.
4.1 Direct contributions to arguments
One of the most obvious ways in which teachers may support collective argumentation is by
providing parts of arguments, which we call direct contributions. In our analysis, we distin-
guish between parts contributed by the teacher, by students, or collaboratively by the teacher
and students together. By direct contributions, we mean those components that were contrib-
uted by the teacher without obvious contribution by students. These specific parts of argu-
ments could be contributed by making a statement verbally, by writing something on the
board, or by presenting a written task. We considered parts of tasks posed by the teacher as part
of the teacher’s direct contributions to arguments when these parts served as data in arguments.
We used Toulmin’s (1958/2003) constructs to classify the teachers’ direct contributions of
argument components. In addition, we analyzed contributions of components that served as
both claims and data (data/claims), both claims and warrants (warrant/claims), and both data
and warrants (data/warrants). To illustrate a teacher’s direct contributions, we refer to episode 2
(Fig. 4) in which Ms. Bell directly contributed a data/claim to the argument by drawing a
diagram on the board. Another illustration can be found in the argument diagrammed in Fig. 2,
in which Ms. Bell directly contributed two data and one warrant by drawing figures and stating
facts about them.
4.2 Asking questions
Our analysis generated 15 different codes for the kinds of questions used to support collective
argumentation. By question, we mean a request for action or information, not simply an
interrogative sentence. Thus, rhetorical questions were not included in this type of support for
collective argumentation, whereas statements requesting an action were included. This meth-
odological decision is consistent with other characterizations of questions in the literature (e.g.,
Boaler & Brodie, 2004). Our analysis of teachers’ questions was done from the viewpoint of
the teacher, so in our analysis, we inferred the teacher’s intent in asking the question rather than
418
Table 1 Teacher support for collective argumentation framework
Direct contributions Questions Other supportive actions
Claims Statements whose validity is Requesting a factual Asks students to provide a Directing Actions that serve to direct the
being established answer mathematical fact students’ attention and/or
the argument
Data Statements provided as support Requesting a method Asks students to demonstrate Promoting Actions that serve to promote
for the claims or describe how they did mathematical exploration
or would do something
Warrants Statements that connect data Requesting an idea Asks students to compare, Evaluating Actions that center on the
with claims coordinate, or generate correctness of the
mathematical ideas mathematics
Rebuttals Statements describing Requesting elaboration Asks students to elaborate on Informing Actions that provide
circumstances under which some idea, statement, or information for the
the warrants would not be diagram argument
valid
Qualifiers Statements describing the Requesting evaluation Asks students to evaluate a Repeating Actions that repeat what has
certainty with which a claim mathematical idea been or is being stated
is made
Backings Usually unstated, dealing with
the field in which the
argument occurs
A. Conner et al.
classifying questions by the types of responses they elicited from students. We organized these
kinds of questions thematically. This process allowed us to characterize the various codes for
the kinds of questions into five general categories: requesting a factual answer, requesting an
idea, requesting a method, requesting elaboration, and requesting evaluation (see Table 2). It is
important to emphasize that our presentation of categories of questions in the framework is not
hierarchical, nor do we value certain categories over others. We present these categories of
questions as a means to describe the ways in which teachers might support collective
argumentation in their classrooms. Throughout the remainder of this section, we explicate
the categories in our framework and illustrate them with examples from Ms. Bell’s teaching.
To support collective argumentation, at times, Ms. Bell asked students questions requesting
a factual answer. In this category, Ms. Bell asked various kinds of questions that prompted her
students to contribute facts by providing an immediate answer or by performing a simple
mathematical action. As examples, there were times when Ms. Bell asked her students to
provide a previous result, the answer to a homework problem or an idea generated in their
small groups. Sometimes the student’s contribution of a previous result was followed by a
longer discussion of mathematical ideas. In day 2, Ms. Bell asked, “What is the sum of the
exterior angles of any polygon?” Ms. Bell’s question requested students to contribute a claim
previously established in day 1, the sum of the exterior angles of any polygon is 360 °.
There were times when Ms. Bell asked students to demonstrate a method or describe a
method used to arrive at a specific answer. We categorized these kinds of questions as those
requesting a method. For instance, when asking for a demonstration, Ms. Bell typically asked a
student to share his or her method with the class by providing work on the front board. In day
Table 2 Categories of questions in support of collective argumentation
Category of question Kind of question with example
Requesting a factual answer: Asks students to Calculation: “10 times 180 is how many degrees?”
provide a mathematical fact Identification: “Which angles are congruent [in the
parallelogram]?”
Previous result: “What is the sum of the exterior angles of
any polygon?”
Recall: “A pentagon has how many sides?”
Term: “What is it called when angles sum to 180?”
Requesting an idea: Asks students to compare, Comparison: “Do these have anything in common? [Ms. Bell
coordinate, or generate mathematical ideas indicates 2x and 2y]”
Conjecture: “If you know how to find a specific interior angle
of a regular polygon, can I find what the exterior angles
are all going to be equal to?”
Construct result: “So can I take a 2 out? What’s left on the
inside?”
Requesting a method: Asks students to Demonstrate a method: “So, will one of y’all go up and show
demonstrate or describe how they did next? So, Travis, tell us what you did.”
something Describe a method: “How did you figure that out?”
Requesting elaboration: Asks students to Explanation: “Can you explain it a different way?”
elaborate on some idea, statement, or diagram Interpretation: “So if it goes up by the same amount each
time, what kind of function is it?”
Justification: “Why?” or “Why doesn’t that work?”
Requesting evaluation: Asks students to evaluate Consensus: “Do you guys agree with that?”
a mathematical idea Reconsider: “Really? Any polygon, huh?”
1, after working in small groups, Ms. Bell asked a student to demonstrate the formula he
derived. She asked, “So, will one of y’all go up and show next? So, Travis, tell us what you
did.” Travis responded to Ms. Bell’s question by sharing his method with the class at the front
board. Using either kind of question in this category, it was possible for Ms. Bell to solicit
multiple responses from students in order to generate a variety of solution methods and further
the class discussion.
Across many episodes of collective argumentation, Ms. Bell posed specific questions
requesting a mathematical idea, asking students to compare, coordinate, or generate mathe-
matical ideas. For example, in episode 2, Ms. Bell asked a sequence of questions that requested
mathematical ideas. First, she asked her students to make a comparison between 2x and 2y,
“Do these have anything in common?” Responding to this question, her students observed that
2 was common in each term; and, in doing so, they contributed part of a warrant in the given
episode. To continue the discussion, Ms. Bell asked a second question. She expected students
to coordinate mathematical ideas to construct a mathematical result. She asked, “So can I take
a 2 out? What’s left on the inside?” With this kind of question, Ms. Bell expected students to
construct a mathematical result by factoring out a 2 from the sum 2x+2y. By asking questions
in this category, she expected students to provide a mathematical idea, which contributed to the
development of the argument.
There were times when Ms. Bell posed questions requesting elaboration. Questions in this
category prompted students to elaborate on their thinking regarding a particular mathematical
idea. This elaboration took the form of asking students to make an interpretation, to contribute
an explanation, or to provide justification. As an example, in day 1, Travis demonstrated his
method for finding the formula for the sum of the measures of the interior angles of a triangle
for the class. Ms. Bell asked him to provide an explanation of the work he was presenting on
the board, saying, “What is this you are writing on the board?” By asking this question, she
expected him to expand upon his written work by providing additional description or
clarification. In this category, Ms. Bell’s requests for elaboration were often addressed to the
entire class, and the elaboration of ideas typically involved students supporting their claims
with data and warrants.
There were times when Ms. Bell asked questions requiring students to either agree or
disagree with a specific mathematical idea. We categorized these kinds of questions as those
requesting evaluation. Sometimes, Ms. Bell asked questions of consensus, to determine if there
was a general agreement among the students about a mathematical claim, whereas others of Ms.
Bell’s questions in this category encouraged students to reconsider a mathematical idea,
implying the idea was not mathematically correct or complete. One particular example occurred
on day 2 when Ms. Bell asked her students to reconsider a student’s claim stating that it was
possible to determine the number of sides of any polygon by dividing 360° by the measure of an
exterior angle of the polygon. Ms. Bell asked, “Really? Any polygon, huh?” Following this
question, students clarified their claim to include “regular” to describe the kind of polygon.
4.3 Other supportive actions
In our analysis, we noticed teachers supported collective argumentation with actions that were
not questions or direct contributions to the argument. Our analysis generated 13 codes for the
different kinds of other supportive actions. We collapsed the codes into five meaningful
categories: directing, promoting, evaluating, informing, and repeating actions (see Table 3).
We present these categories of other supportive actions to assist in describing the ways in
which teachers might support collective argumentation in their classrooms. In this section, we
illustrate each category with an example from Ms. Bell’s teaching.
Table 3 Categories of other supportive actions in support of collective argumentation
Category of other supportive action Kind of supportive action with example or explanation
Directing: Actions that serve to focus Highlights a particular part of the activity, diagram, or procedure
the students’ attention and/or the Hints: “I think you should start by finding the difference.”
argument
Refocuses students’ attention on an important aspect of the activity
Promoting: Actions that serve to Encourages: “These are really good. You guys are making
support mathematical exploration really good steps.”
Suggests: “You might have to try something else.”
Evaluating: Actions that center on the Corrects a student’s incorrect statement
correctness of the mathematics Validates: “Now, I like that definition.”
Verifies the correctness of a contribution to the class
Informing: Actions that provide Clarifies statements with descriptions or gestures
information for the argument Expands on a student’s response to describe the student’s
contribution more fully
Summarizes the main points in a student’s contribution
Repeating: Actions that repeat what Displays a student’s or teacher’s contribution on the board
has been or is being stated Restates a student’s contribution
To support collective argumentation, there were times when Ms. Bell used directing actions
to focus students’ attention to a specific mathematical point or to steer an argument in a
specific direction. For example, in episode 1, Ms. Bell provided a hint to some students
working in a small group. Ms. Bell told the group, “I think you should start by finding the
difference.” With this hint, Ms. Bell focused the argument by directly telling the students what
to do next in order to progress through the mathematical task. In this category, these actions
directed the students’ thinking toward important mathematical aspects in the argument.
Ms. Bell used promoting actions to support students’ exploration of mathematics or their
contributions of parts of arguments. Unlike the directing actions, the actions in this category
tended to allow the argument to move in many possible directions, by encouraging or
suggesting that students continue their reasoning, activity, and/or argument. For example, in
episode 1, Ms. Bell encouraged the small group of students, “These are really good. You guys
are making really good steps.” In this example, Ms. Bell encouraged students to continue in
their thinking. She did not, however, guide the argument in a specific direction.
Ms. Bell used evaluating actions to validate a student’s contribution to an argument, to
verify the correctness of a student’s statement, or to correct a student’s statement. For example,
Ms. Bell often validated a student’s contribution by indicating the contribution was mathe-
matically appropriate or correct. For example, after the class had developed their definition of a
regular polygon on day 2, Ms. Bell stated, “Now, I like that definition.” She validated the
definition without explicitly promoting students to continue their reasoning. These evaluative
actions consistently communicated the soundness of students’ claims or reasoning to the class.
There were times when Ms. Bell used informing actions to assist students in developing
mathematical understanding and problem solving strategies. The information Ms. Bell pro-
vided most often took the form of direct contributions of argument components. There were
times, however, when Ms. Bell provided information that did not directly contribute a
component in an argument. These informative statements supported the argument in other
ways, by clarifying statements or summarizing or expanding on what a student said. For
example, in episode 2, Ms. Bell pointed at the opposite angles in the rhombus to clarify the
angles to which a student referred (see data labeled 2 in episode 2).
Many times, Ms. Bell supported her students’ contributions by announcing to the class
what a student said. These repeating actions generally took the form of restating what a student
said or displaying it on the board without adding or subtracting information. We differentiated
repeating actions from informing actions (such as summarizing) by examining whether the
teacher changed the contributed information in some way. In this category, Ms. Bell’s main
goal was simply to ensure the class heard the information. For example, in episode 1, Ms. Bell
both restated Martin’s equation, “2x plus 2y equals 360” and displayed it on the board using
the appropriate symbols. With these actions, she ensured that the class both heard and saw his
contribution.
5 Contributions of the framework
We propose this framework as a device that allows one to examine conversations in mathe-
matics classrooms in a particularly mathematical way. In the following paragraphs, we
describe how we can use this framework within other frameworks, such as Hufferd-Ackles
et al.’s (2004) math talk framework or Rasmussen and Stephan’s (2008) method for
documenting collective activity to look more specifically at which actions of the teacher are
productive. We examine the connections between our framework and how teachers’ actions in
support of collective argumentation are described in the literature. We also examine how our
framework allows a specific focus on warrants and the ways teachers provide or support their
students in providing warrants. We suggest some ways in which using our framework can
focus one’s attention on the more mathematical aspects of classroom activity. Finally, we
describe some limitations of our framework with respect to its source and applicability.
5.1 Connections to frameworks for classroom activity
Other frameworks, such as Hufferd-Ackles et al.’s (2004) math-talk framework or Stein

et al.’s (2008) five practices, have provided useful ways to examine the growth of mathe-
matical communities and orchestrate discourse. Our framework, whether used in conjunction
with one of these frameworks or separately, allows one to look more closely at specific
mathematical aspects of the conversation. It allows us to view classroom behavior on a more
micro level, categorizing individual actions of the teacher before looking across these to
examine patterns.
As an example, we examined Ms. Bell’s classroom community through the lens of
Hufferd-Ackles et al.’s (2004) math-talk framework. Although we did not have data to look
at the growth of the community over time, we were able to place Ms. Bell’s class on the math-
talk continuum for each of the four categories. Briefly, we placed the interactions in Ms. Bell’s
class at levels 2 or 3 in each of the areas. However, our framework allowed us to provide a
finer-grained analysis of the mathematical conversations. For instance, in Hufferd-Ackles
et al.’s “source of mathematical ideas” category, our framework provided detail that suggests
what kinds of mathematical ideas students learned because the class was at level 3, at least for
the class periods under close scrutiny in this paper.
Hufferd-Ackles et al. (2004) give the following description of level 3 of the math-talk
framework: “Teacher allows for interruptions from students during her explanations; she lets
students explain and ‘own’ new strategies…. Teacher uses student ideas and methods as the
basis for lessons or miniextensions” (p. 90). Our description of Ms. Bell’s classes includes an
example of Ms. Bell’s use of a student’s question as impetus for an unanticipated investigation
when she rephrased Karin’s question, using it and Micah’s request to initiate an investigation
of angle measures in a rhombus. In so doing, she leveraged many of her students’ ideas to
support their eventual conclusion that there was not one solution to one equation containing
two unknown quantities and thus the angles in a rhombus could not be predicted in the same
way as the angles in a square. Although the conclusion about a rhombus was related to the
mathematical ideas being discussed on that day, the conclusion about the number of solutions
(or lack of a unique solution) to an equation with two unknowns was not. Ms. Bell’s
introduction of this important mathematical idea in an otherwise unrelated discussion was
evident through our analysis of her questions and the warrants provided in her class. Thus,
although the math-talk framework might flag this episode as important, our framework specifies
why it is mathematically important: by direct contributions (e.g., providing a diagram of a
rhombus on the board), other support (e.g., labeling the measures of the angles in the rhombus
with variables), and her questions (e.g., “Can I tell what x and y specifically are?”), Ms. Bell
supported her students in using appropriate disciplinary reasoning to draw conclusions about
the geometric content under consideration by using a concept from algebra and functions.
The teacher support for collective argumentation framework goes beyond frameworks
currently in use to allow examination of how teachers impact the development of mathematics
in classrooms and how teachers support the reasoning of students. Rasmussen and Stephan
(2008) have articulated the usefulness of collective argumentation in documenting the pro-
gression of collective knowledge in classrooms by recording how specific ideas progress from
being claims requiring data and warrants to being used as data or warrants in support of other
claims. Our framework allows examination of how the teacher supports students in under-
standing how and when claims that have been previously established can be used in support of
subsequent claims. As a simple example, in several of the day 2 arguments, Ms. Bell asked
questions that prompted students to use the formula they had established on day 1 directly in
finding the sum of the measures of the interior angles of particular polygons and also in finding
a general method for finding the measure of one interior angle of a regular n-gon. The teacher
can also support the students’ use of established claims by other supportive actions such as
validating the contribution of such claims as data or warrants or displaying relevant parts of the
contribution. In some cases, the teacher might model the use of previously established claims
by directly contributing them within the argumentation.
5.2 Connections to other descriptions of teacher moves
Within the mathematics education literature on collective argumentation, several authors have
described the actions of the teacher when facilitating argumentation. Many of those actions fit
within our framework of support for collective argumentation. Our framework serves to
consolidate the variety of moves described by authors, building on their observations of
specific moves. However, many researchers report teacher actions observed at a different
grain size, examining things such as establishing norms (e.g., Weber et al., 2008) or focusing
or funneling (Wood, 1998). Our framework allows identification of the individual moves that
may go into such larger activities, and patterns of the individual moves may allow an
investigation of those larger activities, but we have not yet used our framework to examine
those larger activities. We have, however, examined the moves of teachers described in reports
of collective argumentation and connected them to our framework. Table 4 provides a
summary of some of the moves specifically identified in a subset of the literature that reported
teachers’ actions with respect to collective argumentation and their relation to our proposed
framework of support moves.
Examining patterns within individual teacher moves may give insight into a larger activity,
even outside an explicit examination of collective argumentation. Herbel-Eisenmann and
Table 4 Selected teacher moves from literature reporting teachers’ actions with respect to collective
argumentation
Author(s) Identified teacher moves Category of support for collective

argumentation framework
Forman et al. (1998) Teacher asks a volunteer to “give and Questions, requesting elaboration:
justify his or her answers” (p. 535) justification
Forman et al. (1998) “The teacher revoices Larry’s grounds via Other supportive actions, informing:
repetition and expansion” (p. 536) expands
Forman et al. (1998) “The teacher summarizes the two Other supportive actions, informing:
arguments” (p. 540) summarizes
Knipping (2003) “The teacher encourages the students Questions, requesting an idea:
to formulate conjecture
a conjecture…” (p. 4)
Knipping (2003) “The teacher expresses feelings of Other supportive actions, promoting:
appreciation when the student comes encourages
up with his conjecture” (p. 4)
Knipping (2003) Teacher “encourages the class to confirm Questions, requesting evaluation:
or reject this conjecture” (p. 4) consensus
Wood (1999) Teacher asked a student “to clarify for the Questions, requesting elaboration:
others why she [a student] was explanation
questioning…” (p. 181)
Wood (1999) Teacher “indicated that he [a student] was Questions, requesting elaboration:
to give reasons for his thinking” (p. 181) justification
Yackel (2002) Teacher “capitalized on students’ Other supportive actions, repeating:
contributions” (p. 430) restating
Yackel (2002) “The teacher drew a diagram on the Other supportive actions, repeating:
chalkboard” (p. 432) display
Yackel (2002) Teacher asked a student “to explain Questions, requesting elaboration:
why he added 8 and 8” (p. 436). explanation
“She could facilitate the discussion best by
directly calling for a warrant herself” (p. 436).
Breyfogle (2005) suggest that the patterns of questions posed by teachers provide insights into
the types of interactions teachers have with their students. These interactions include funneling
and focusing interactions. Wood (1998) indicates that funneling interactions occur when a
teacher asks a series of questions that direct students through a procedure or to a desired
solution such that the teacher is doing the majority of the mathematical thinking and the
students are simply providing answers that may seem unconnected. Focusing interactions
occur when a teacher listens to students’ responses and asks questions that direct them based
on what the students are thinking rather than their own solution method (Wood, 1998).
Although these interactions are centered on the patterns of questions, the actions of the teacher
play an important role. In her examples of these interactions, Wood also provides a description
of the teacher’s actions and states how these actions support the specific type of interaction.
Using the teacher support for collective argumentation framework, researchers can examine
the kinds of questions and other supportive actions that compose patterns of funneling and
focusing interactions and the structures of the arguments in which these interactions take place.
Forman et al. (1998) described several teacher actions under the category of revoicing.
These actions involve moves in two different categories of other supportive actions: informing
and repeating. Chapin, O’Conner, and Anderson (2003) include revoicing as an essential math-
talk move. In their discussion, they focus on the importance of clarifying students’
contributions to the discussion and making the contributions of individual students under-
standable to the other members of the class. This talk move is similar to our other supportive
action of informing. However, from our analysis of practice, we found it useful to distinguish
among moves that clarify a student’s contribution, summarize a student’s contribution, or
expand upon a student’s contribution. Examining how and when a teacher clarifies, summa-
rizes, or expands a student’s contribution, whether each argument was mathematically pro-
ductive, and who dominated the contributions to and direction of the argument may allow us to
make further inferences about when it is productive or appropriate to engage in the wide range
of behaviors captured by the framework.
5.3 Focus on warrants
The teacher support for collective argumentation framework allows an explicit focus on the
warrants provided in a class as an indication of the reasoning (see also Conner, 2012) that is
being made public. This framework allows one to examine who contributes the reasoning in
the class and what the teacher does to prompt and respond to the reasoning that is contributed.
The teacher’s direct contributions of warrants provide part of the picture of the reasoning that
he or she values, but it is also important to examine when he or she does not contribute
warrants and whether in those instances he or she prompts a student to provide a warrant, a
student provides a warrant without prompting, or the reasoning is not explicated because the
warrant is implicit. Sometimes when teachers directly contribute parts of arguments, such as
warrants, they may take over the argument, making the argument more a construction of the
teacher than one that is collectively established. However, other teacher contributions serve to
make parts of arguments that would otherwise be left implicit apparent for students or to
emphasize the importance of having and presenting reasons for claims. Because reasoning is a
pivotal part of mathematics, we believe this feature is an important part of the usefulness of our
framework.
One example of the information provided by looking at the warrants in arguments comes
from our analysis of day 1 of Ms. Bell’s class. In the small-group argument highlighted on day
1, the group left implicit the warrant that connects the observation that “it goes up by 180 each
time” to “it’s a linear function.” Ms. Bell implied the warrant in her question that prompted the
data/claim “it’s a linear function,” but did not make it explicit. However, when this small group
presented their argument to the entire class later in that class period, they used essentially the
same data/claims as before (using “interior angles sum changes by 180 each time” and “it is
linear”), but Ms. Bell directly contributed the warrant, saying, “If it changes by the same
amount each time when you are going up by one, it is going to be a linear function.” In so
doing, she explicated the group’s reasoning to the larger class. By not allowing the warrant to
be left implicit, Ms. Bell contributed a valid mathematical reason that connected the pattern to
the relevant mathematical concept. By contributing the warrant, she modeled the activity of
explicating reasoning and made sure that the reasoning was made available to the collective.
5.4 Power and limitations of the framework
One of the aspects of the framework that makes it specifically mathematical is its use in the
context of collective argumentation. Although we believe that it is still useful to look at
questions and other supportive actions without diagramming the argumentation that occurs in a
classroom, it is when the questions and other supportive actions are examined in the context of
the claims, data, and warrants of an argument that the most mathematical aspects of the
classroom conversations come to light. In particular, questions and other supportive actions are
mathematically important when they prompt and reinforce students’ contributions of appro-
priate mathematical claims, data, and warrants. The teacher’s direct contributions, or with-
holding of contributions, or prompting of students to provide contributions to particular parts
of arguments can give insight into what the teacher values or privileges in a mathematical
conversation. A teacher could use the framework to examine his or her practice without
diagramming, but for a robust analysis, a researcher would need to diagram the arguments
as we have illustrated in this paper so that the supportive actions can be connected to the
relevant parts of the framework and conclusions drawn accordingly.
Part of the power of the framework in conjunction with the diagramming of arguments is
that a diagram functions as a sieve. Placing the parts of the argument in the diagram focuses the
researcher on the specific mathematical statements and actions that took place in the class and
excludes the other classroom activities. From the diagrammed components and the accompa-
nying questions and other supportive actions, the actual mathematical ideas that were
foregrounded in the class become apparent. Additionally, the implicit parts of the diagram
highlight the mathematical ideas that may have been present but were not stated or necessarily
accepted by the class. Using the framework, a researcher can then examine how the various
questions and other supportive actions functioned within the mathematical conversation to
highlight or fail to highlight important mathematical ideas.
While our framework does allow an in-depth analysis of the argumentation in a classroom,
it does not provide a value judgment as to the productivity of the argumentation observed or
the usefulness of any particular contribution of a teacher or student. Those judgments are left to
the users of the framework. We specifically did not limit our analysis to those episodes of
argumentation that we judged to be particularly productive, because we wanted our framework
to be applicable to the wide range of argumentation currently present in classrooms. This
means that some arguments may be incomplete or mathematically incorrect. It means that
sometimes the identified support for collective argumentation may actually involve the teacher
“taking over” the argument or providing components of arguments that would have more
productively been left to the student. What the framework does in these cases is allow the user
to identify what the teacher and students did and then to draw conclusions about whether that
would be productive based on the context and his or her mathematical knowledge. An open
question remains concerning what patterns of support moves will support productive argu-
mentation. More research is needed to determine the characteristics of productive argumenta-
tion and how teachers can support such argumentation.
6 Conclusion
Facilitating productive mathematical discussions can be a difficult endeavor for many teachers.
To overcome this difficulty, various researchers have examined practices that assist teachers in
orchestrating these discussions (Hufferd-Ackles et al., 2004; Staples, 2007; Stein et al., 2008).
The teacher actions they propose, such as establishing communities, negotiating responsibility
for actions, and paying attention to student solutions, are very broad. Our framework provides
a finer-grained perspective on the different ways teachers can support collective argumentation
and provides teachers with additional ways to think about their actions during these types of
discussions.
The teacher support for collective argumentation framework can be used to inform teacher
education and professional development activities. Mathematics teacher educators can use the
framework with prospective teachers to engage in a dialog about issues of practice. Prospective
teachers may reflect upon the types of support as they begin to construct their identities as
mathematics teachers. In addition, understanding the different types of supportive actions

could provide practicing teachers with meaningful constructs from which they may examine
and reflect upon their practice. For example, knowledge of the different argument
components may prompt a practicing teacher to examine the extent to which students
contribute components of arguments. This examination of practice may lead the
teacher to reflect upon how the other types of support may be used to prompt
different and better contributions from students.
The framework allows researchers to compare teachers’ practices or to conduct a longitu-
dinal examination of one teacher’s practice over time. For instance, perhaps a teacher is
engaged in a professional development program in which he or she is seeking more student
engagement in mathematical practices such as reasoning. One way to examine the changes in
his or her practice would be to diagram selected instances of argumentation and to focus on the
questions being asked, the kinds of supportive actions employed, and the results of these
actions in terms of student and teacher contributions of warrants and components of
subarguments. The framework allows an examination of the teacher’s supportive role in
engaging students in making and defending mathematical claims.
To continue to address the call by Yackel (2002), the framework allows researchers to
generate cases of teacher support for collective argumentation. Shulman (1986) described
case-based research literature as a powerful educational tool, because cases can be used to
“illuminate both the practical and theoretical” (p. 11). These cases of teacher support
for collective argumentation, developed with the common language provided by our
framework, could move researchers closer to defining what makes collective argu-
mentation productive. Comparisons across cases of teachers’ support would allow for
better understanding of how different combinations of support assist students in
making their reasoning more explicit for others. These comparisons could also give
insight into how teachers can incorporate the students’ reasoning into the structure of
the argument in productive ways.
Mathematics educators can use this framework to monitor their students’ progress toward
engaging in deductive reasoning and eventually toward proof as they examine who is
providing what kinds of reasoning in the arguments they are making. This framework allows
the close examination of the development of sociomathematical norms, particularly in the
areas of justification and explanation, as mathematics educators can trace how students
contribute parts of arguments, particularly warrants, with and without teachers’ support.
Arguably, as students become more mathematically autonomous, teachers should be able to
provide fewer direct contributions and expect their students to provide parts of arguments,
particularly warrants, without needing to be specifically asked for them. Using this
framework, mathematics educators can trace the amounts and kinds of teacher support
for collective argumentation over time with the goal of characterizing the sociomathematical
norms of the class and the disposition of the students with respect to engaging in reasoning and
proof.
The teacher support for collective argumentation framework assisted us in making sense of
the complexities involved in the ways teachers support collective argumentation. With this
framework, we can examine the specifically mathematical aspects of classroom discussions
and trace the reasoning that occurs within the class. The types of support we have outlined and
the interactions between them in the context of collective argumentation allow us to examine
aspects of teachers’ activity that may be important in the development of appropriate disci-
plinary practices. Examining these types of support and their influence on developing disci-
plinary practices may lead to a more definitive characterization of productive collective
argumentation.
Acknowledgments This paper is based on work supported by the University of Georgia Research Foundation
under grant no. FRG772 and the National Science Foundation through the Center for Proficiency in Teaching
Mathematics under grant no. 0227586. Opinions, findings, and conclusions in this paper are those of the authors
and do not necessarily reflect the views of the funding agencies. The authors would like to thank Jeremy
Kilpatrick and Denise Spangler for their helpful comments on earlier versions of this manuscript.
References
Advisory Committee on Mathematics Education. (2011). Mathematical needs: Mathematical needs of learners.
London, UK: Advisory Committee on Mathematics Education.
Baxter, J. A., & Williams, S. (2009). Social and analytic scaffolding in middle school mathematics: Managing the
dilemma of telling. Journal of Mathematics Teacher Education, 13(1), 7–26. doi:10.1007/s10857-009-9121-4.
Boaler, J., & Brodie, K. (2004). The importance, nature and impact of teacher questions. In D. E. McDougall & J.
A. Ross (Eds.), Proceedings of the twenty-sixth annual meeting of the North American Chapter of the
International Group for the Psychology of Mathematics Education (Vol. 2, pp. 773–781). Toronto, Canada:
Ontario Institute for Studies in Education/University of Toronto.
Brodie, K. (2010). Pressing dilemmas: Meaning-making and justification in mathematics teaching. Journal of
Curriculum Studies, 42(1), 27–50.
Chapin, S. H., O’Connor, C., & Anderson, N. C. (2003). Classroom discussions: Using math talk to help students
learn. Sausalito, CA: Math Solutions.
Chazan, D., & Ball, D. L. (1999). Beyond being told not to tell. For the Learning of Mathematics, 19(2), 2–10.
Conner, A. (2012). Warrants as indications of reasoning patterns in secondary mathematics classes. In
Proceedings of the 12th International Congress on Mathematical Education (ICME-12), Topic Study
Group 14 (pp. 2819–2827). Seoul, Korea.
Conner, A. (2008). Expanded Toulmin diagrams: A tool for investigating complex activity in classrooms. In O.
Figueras, J. L. Cortina, S. Alatorre, T. Rojano, & A. Sepulveda (Eds.), Proceedings of the Joint Meeting of
PME 32 and PME-NA XXX (Vol. 2, pp. 361–368). Morelia, Mexico: Cinvestav-UMSNH.
Conner, A., Edenfield, K., Gleason, B., & Ersoz, F. (2011). Impact of a content and methods course sequence on
prospective secondary mathematics teachers’ beliefs. Journal of Mathematics Teacher Education, 14, 483–
504. doi:10.1007/s10857-011-9186-8.
Forman, E. A., Larreamendy-Joerns, J., Stein, M. K., & Brown, C. A. (1998). “You’re going to want to find out
which and prove it”: Collective argumentation in a mathematics classroom. Learning and Instruction, 8,
527–548.
Franke, M. L., Webb, N. M., Chan, A. G., Ing, M., Freund, D., & Battey, D. (2009). Teacher questioning to elicit
students’ mathematical thinking in elementary school classrooms. Journal of Teacher Education, 60, 380–
392. doi:10.1177/0022487109339906.
Herbel-Eisenmann, B., & Breyfogle, M. (2005). Questioning our patterns of questioning. Mathematics Teaching
in the Middle School, 10(9), 484–489.
Hollebrands, K. F., Conner, A., & Smith, R. C. (2010). The nature of arguments provided by college geometry
students with access to technology while solving problems. Journal for Research in Mathematics Education,
41, 324–350.
Hufferd-Ackles, K., Fuson, K., & Sherin, M. G. (2004). Describing levels and components of a math-talk
learning community. Journal for Research in Mathematics Education, 35, 81–116.
Inglis, M., Mejia-Ramos, J. P., & Simpson, A. (2007). Modelling mathematical argumentation: The importance
of qualification. Educational Studies in Mathematics, 66, 3–21. doi:10.1007/s10649-006-9059-8.
Knipping, C. (2003). Argumentation structures in classroom proving situations. Paper presented at the Third
Congress of the European Society for Research in Mathematics Education. Italy: Bellaria.
Knipping, C. (2008). A method for revealing structures of argumentations in classroom proving processes. ZDM:
The International Journal on Mathematics Education, 40(3), 427–441. doi:10.1007/s11858-008-0095-y.
Krummheuer, G. (1995). The ethnography of argumentation. In P. Cobb & H. Bauersfeld (Eds.), The emergence
of mathematical meaning: Interaction in classroom cultures (pp. 229–269). Hillsdale, NJ: Erlbaum.
Krummheuer, G. (2000). Mathematics learning in narrative classroom cultures: Studies of argumentation in
primary mathematics education. For the Learning of Mathematics, 20(1), 22–32.
Krummheuer, G. (2007). Argumentation and participation in the primary mathematics classroom: Two episodes
and related theoretical abductions. Journal of Mathematical Behavior, 26, 60–82. doi:10.1016/j.jmathb.2007.
02.001.
Lobato, J., Clarke, D., & Ellis, A. B. (2005). Initiating and eliciting in teaching: A reformulation of telling.
Journal for Research in Mathematics Education, 36, 101–136.
Martin, T. S. (Ed.). (2007). Mathematics teaching today: Improving practice, improving student learning. Reston,
VA: National Council of Teachers of Mathematics.
McClain, K. (2002). Teacher’s and students’ understanding: The role of tools and inscriptions in supporting
effective communication. Journal of the Learning Sciences, 11(2–3), 217–249.
McCrone, S. S. (2005). The development of mathematical discussions: An investigation in a fifth-grade
classroom. Mathematical Thinking and Learning, 7(2), 111–133.
National Council of Teachers of Mathematics. (2000). Principles and standards for school mathematics. Reston,
VA: Author.
National Council of Teachers of Mathematics. (2009). Focus in high school mathematics: Reasoning and sense
making. Reston, VA: Author.
National Governors Association Center for Best Practices & Council of Chief State School Officers. (2010).
Common core state standards: Mathematics standards. Washington, DC: National Governors Association
Center for Best Practices, Council of Chief State School Officers. Retrieved from http://www.corestandards.
org/the-standards/mathematics.
Rasmussen, C. L., & Stephan, M. (2008). A methodology for documenting collective activity. In A. Kelly, R.
Lesh, & J. Baek (Eds.), Handbook of design research methods in education: Innovations in science,
technology, engineering, and mathematics teaching and learning (pp. 195–215). New York, NY: Routledge.
Shulman, L. S. (1986). Those who understand: Knowledge growth in teaching. Educational Researcher, 15(2),
4–14.
Staples, M. (2007). Supporting whole-class collaborative inquiry in a secondary mathematics classroom.
Cognition and Instruction, 25, 161–217.
Stein, M. K., Engle, R. A., Smith, M. S., & Hughes, E. K. (2008). Orchestrating productive mathematical
discussions: Five practices for helping teachers move beyond show and tell. Mathematical Thinking and
Learning, 10, 313–340.
Toulmin, S. E. (2003). The uses of argument (updated ed.). New York: Cambridge University Press. Original
work published 1958.
Weber, K., Maher, C., Powell, A., & Lee, H. S. (2008). Learning opportunities from group discussions: Warrants
become the objects of debate. Education Studies in Mathematics, 68, 247–261.
Wood, T. (1998). Alternative patterns of communication in mathematics classes: Funneling or focusing? In H.
Steinbring, M. G. Bartolini Bussi, & A. Sierpinska (Eds.), Language and communication in the mathematics
classroom (pp. 167–178). Reston, VA: National Council of Teachers of Mathematics.
Wood, T. (1999). Creating a context for argument in mathematics class. Journal for Research in Mathematics
Education, 30(2), 171–191.
Yackel, E. (2002). What we can learn from analyzing the teacher’s role in collective argumentation? Journal of
Mathematical Behavior, 21, 423–440.

Educ Stud Math (2014) v86 P 401-429

Uploaded by

Copyright:

Available Formats

Educ Stud Math (2014) v86 P 401-429

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Educ Stud Math (2014) v86 P 401-429

Uploaded by

Copyright:

Available Formats

Educ Stud Math (2014) 86:401–429

Teacher support for collective argumentation:

AnnaMarie Conner & Laura M. Singletary &

Published online: 1 February 2014

Keywords Argumentation . Reasoning . Questioning . Teaching . Discussions

A. Conner (*) : R. C. Smith : P. A. Wagner : R. T. Francisco

examining how teachers may support collective argumentation in secondary mathematics

2.1 Facilitating mathematical discussions

believe focusing on collective argumentation allows mathematics educators to think about

2.2 Focusing on collective argumentation

Recent work in mathematics education has highlighted collective argumentation as an impor-

Fig. 1 Diagram of a generic argu-

Krummheuer (1995) between “analytic and substantial arguments” (Toulmin,

Fig. 2 Example of argument in Ms. Bell’s class

To examine teachers’ support for collective argumentation, we modified Toulmin’s

2.3 Focusing on teacher support

3.1 The study

3.2 Data collection

3.3 Data analysis

and provided evidence to support it (our definition of collective argumentation), we were

3.4 Description of the focus class periods

So if it goes up I think that’s a

Fig. 3 Diagram of day 1, episode 1

Martin: It goes up by 180 each time. [Data/claim labeled 3.]

Ms. Bell: Okay.

Adam: Each one, yeah, and then, uh—.

Karin: Linear [Data/claim labeled 4.]

Ms. Bell: Linear

Martin: So, it’s something

Ms. Bell: It’s a linear function.

Ms. Bell: Okay.

Martin: 3 times 180, 540. [Warrant labeled 9.]

Ms. Bell: 540, but what do we want to get as our output?

Ms. Bell: So, how do we get from 540 to 180?

Karin: I think you’ve got a defective one though.

Adam: That was the first one.

Martin: Subtract 360. [Warrant labeled 10.]

Ms. Bell: Let’s see if that works.

Martin: I got it.

Ms. Bell: Does it make sense?

Martin: Mmhm [yes].

Adam: 180x minus 360 [Claim labeled 11.]

Ms. Bell: I don’t want you to just write it down.

Karin: That’s what I got. I was about to graph it.

Micah: 180 [Data labeled 3 in the diagram.]

Fig. 4 Diagram of day 2, episode 2

Angela: You should.

Micah: Let’s test it.

Martin: 2x+2y=360. [Data/claim labeled 5 in the diagram.]

Ms. Bell: 2x+2y=360 [writes 2x+2y=360]. Now, what can I do?

Travis: You could do 2x squared.

Ms. Bell: 2x squared?

Travis: Plus 2xy equals—.

Adam: No, you could just do—.

Travis: Never mind, I was thinking like that.

Ms. Bell: Area?

Travis: a squared plus b squared equals c squared.

Micah: I got a good one.