Content uploaded by Jennifer Richards
Author content
All content in this area was uploaded by Jennifer Richards on May 07, 2021
Content may be subject to copyright.
Learning, Culture and Social Interaction 28 (2021) 100490
Available online 27 January 2021
2210-6561/© 2021 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Complexities and opportunities in teachers’ generation of videos
from their own classrooms
☆
Jennifer Richards
*
, Mari Altshuler, Bruce L. Sherin, Miriam Gamoran Sherin,
Christopher J. Leatherwood
Northwestern University, School of Education and Social Policy, Annenberg Hall, 2120 Campus Drive, Evanston, IL 60208, United States
ARTICLE INFO
Keywords:
Classroom video
Video generation
Professional development
Teacher learning
Student thinking
ABSTRACT
Video cameras have become smaller and increasingly embedded in our social lives. This study
investigates the implications of these changes for teachers’ work with video in professional
development. Specically, we focus on complexities and opportunities associated with teachers
generating videos from their own classrooms, in the context of a short online course where
teachers tried several mathematical activities that invited student reasoning and discussion. We
unpack the potential affordances of video generation for teacher learning by examining teachers’
processes — how they captured video during instruction and selected clips to share — and the
characteristics of their nal video products. Analyses demonstrated that generating video clips
was complex work for teachers, requiring coordination of multiple sets of considerations about
their students, the video itself, video viewers, and the course. Further, capturing and selecting
sparked different emphases for teachers. While capturing, teachers foregrounded video quality
considerations while seeking to support students’ experiences. Selecting engaged teachers in
curating (and hence attending to) student thinking within their videos. Teachers’ video products
also demonstrated characteristics known to be supportive of teacher learning. We see these results
as cause for optimism about the potential of teachers’ video generation for supporting teacher
learning.
1. Introduction
About four decades ago, the advent of affordable video recording technology changed teacher professional development (PD).
Teachers began to routinely view videos of teaching, with substantial research exploring the affordances of varied video-based PD
designs for teacher learning (e.g., Gaudin & Chali`
es, 2015; Sherin, 2004; Tripp & Rich, 2012).
Recent technological developments have opened the door for new uses of video. Video cameras have become smaller and truly
ubiquitous; we (literally) carry them in our pockets. This wide availability of video technology has brought about dramatic changes
that we might not have anticipated a decade ago. Most notably, video has come to play an important role in our social lives. Many of us
frequently and readily share videos that capture aspects of our everyday lives and that represent ourselves to others.
In turn, the newly social nature of video has implications for teachers’ use of video in PD. Now, in addition to watching and
☆
This work was supported by funding from the Spencer Foundation (Deeper Learning Labs: Digital Resources for Collaborative Teacher
Learning) in the United States. All ndings, opinions, and recommendations expressed herein are those of the authors.
* Corresponding author.
E-mail address: jrichards@northwestern.edu (J. Richards).
Contents lists available at ScienceDirect
Learning, Culture and Social Interaction
journal homepage: www.elsevier.com/locate/lcsi
https://doi.org/10.1016/j.lcsi.2021.100490
Received 1 September 2020; Received in revised form 21 December 2020; Accepted 31 December 2020
Learning, Culture and Social Interaction 28 (2021) 100490
2
discussing video, teachers can generate video from their own classrooms and share and discuss their videos with their peers. We
anticipate that teachers’ participation in these video-generation activities may present novel opportunities for teacher learning, given
the kinds of thinking and action that go into capturing and selecting video (e.g., Derry et al., 2010).
This paper aims to unpack potential affordances of video generation for teacher learning through two lenses. First, we look closely
at how teachers engage in video-generation processes to explore their potential for teacher learning. Second, we examine the qualities
of teacher-generated videos to determine whether they maintain the characteristics prior work elevates as important for teacher
learning through video viewing. We explore these areas within a designed context: a short online course in which elementary teachers
learned about and tried several mathematics activities that centered student reasoning and discussion. Teachers recorded as they
engaged their students in the activities, then shared video clips of student thinking with their colleagues and commented on each
other’s videos. Within this context, we focused on teachers’ engagement in two video-generation processes: capturing a lesson on video
and selecting a portion of the recorded video to share. Specically, our research questions are as follows:
1. What shaped teachers’ decisions as they captured classroom video and selected clips to share in the PD course?
a. What did teachers consider?
b. How and why did they highlight particular aspects of classroom phenomena — especially student thinking — through their
video generation?
2. What were the characteristics of teachers’ self-generated video clips?
Our analyses suggest that the affordances of the two video-generation processes for teacher learning might be quite different, in
part because they seemed to engage teachers in negotiating different sets of concerns. The process of capturing engaged teachers in
orchestration, in which they sought to balance capturing important features of classroom activity while managing the ongoing expe-
rience of all of their students. In contrast, the process of selecting engaged teachers in curation, with an emphasis on displaying what
they deemed important features of classroom interactions. In this designed context, we also found that teachers produced videos with
characteristics known to be supportive of teacher learning.
2. Conceptual background
2.1. Video and teacher learning
Internationally, video has come to play an important role in supporting teacher learning (Gaudin & Chali`
es, 2015; Major & Watson,
2018). Research has documented impacts of video-based PD on multiple facets of teachers’ knowledge and practice, including their
beliefs about the nature of learning (e.g., Sang et al., 2012), content understandings (e.g., Masats & Dooly, 2011), perceptions of
students (e.g., Weil et al., 2020), collegial conversations about teaching and learning (e.g., Borko et al., 2008), and changes in enacted
pedagogical techniques (e.g., Reisman & Enumah, 2020; Tripp & Rich, 2012). Further, numerous studies highlight how video can be a
supportive tool for facilitating teachers’ noticing and analysis of student thinking (e.g., Barnhart & van Es, 2015; Borko et al., 2008;
Cohen, 2004; Roth et al., 2017; Stockero et al., 2017; van Es & Sherin, 2008, 2010), a focus central to our work, and for supporting
responsive classroom instruction that builds on students’ ideas (Richards & Robertson, 2016).
What makes video such a rich tool for teacher learning? Much of the power has been assumed to come from watching and dis-
cussing video; teachers can examine classroom interactions in detail, multiple times, and potentially from multiple perspectives, which
differs from what is possible during instruction or in other forms of reection (Sherin, 2004). However, video’s potential to support
attention to student thinking depends on the video’s qualities, such as whether it is clear enough to hear students’ contributions
(LeFevre, 2004; Roschelle, 2000), or whether it offers sufcient “windows” into student thinking (through sources like students’ talk,
work, gestures, etc.; see Sherin et al., 2009). Recent studies examining relations between video complexity and teachers’ noticing of
students’ mathematical thinking (Superne & Bragelman, 2018) or teachers’ discussions (Amador et al., 2020) suggest that other
characteristics matter as well, such as the presence of students sharing multiple solution strategies for rich tasks.
2.2. Engaging teachers in video generation
Though prior work has focused on watching and discussing video as the primary vehicles for teacher learning, recently we and
others have begun to explore how the processes of generating video may provide unique opportunities for teacher learning. For
instance, asking teachers to “capture short video clips of student mathematical thinking” (Sherin & Dyer, 2017, p. 482) can invite
teachers to identify and promote student thinking during instruction. A few studies suggest broader positive impacts of recording and
video editing on pre-service teachers, including shifting their reective focus (Calandra et al., 2008; Yerrick et al., 2005). For example,
when Calandra et al. had pre-service teachers record themselves teaching, write reections, and edit accompanying video clips, the
video editing process specically seemed to support increasing attention within reections to students and areas for improvement.
However, recording instruction and selecting excerpts are still relatively novel and potentially complicated tasks for teachers.
Researchers who study and work with video highlight that generating video involves a multitude of value-laden decisions, with varied
affordances and limitations (e.g., Derry et al., 2010; Vossoughi & Escud´
e, 2016). For teachers, such decisions may be particularly
complex — for instance, teachers must decide how to place and point cameras while also facilitating the classroom activity being
lmed. Further, though lming and sharing videos has become increasingly common in everyday life, video production skills are not
central to teaching, in part evidenced by mention of technical challenges in studies that draw on teacher-generated videos (e.g., Nagro
J. Richards et al.
Learning, Culture and Social Interaction 28 (2021) 100490
3
et al., 2017; Zhang et al., 2011). Finally, teachers will always be representing themselves and their practice in self-generated videos,
which can understandably produce anxiety (e.g., Beisiegel et al., 2018; Clark et al., 2019; Fadde et al., 2009).
In this study, we take up Zhang et al.’s (2011) call to “explore teachers’ thinking processes when editing their own video” (p. 9).
Few studies to date have focused on teachers’ video-generation processes as central objects of inquiry, though several (Fadde et al.,
2009; Sherin & Dyer, 2017; Xiao & Eriksson, 2020) have begun to explore teachers’ approaches and thinking when capturing and
selecting video to share in professional learning contexts, with variable results. For instance, Sherin and Dyer found that teachers often
reviewed and reevaluated captured video as part of selecting clips. In contrast, Fadde et al. documented that some teachers did not
seem to watch their full videos while editing — a difference that would likely impact their learning opportunities. Also, teachers in
these studies highlighted different aspects of classroom instruction in their video clips, partly inuenced by their social, professional
contexts. We seek to add to this nascent literature with a systematic analysis of teachers’ video-generation processes (and their
resultant products) in the context of a professional learning course designed to center student thinking in early elementary
mathematics.
2.3. Conceptualizing video generation
In order to conceptualize teachers’ generation of video, we bring to bear two theoretical perspectives. The rst, Goodwin’s (1994)
notion of professional vision, has been used by researchers to understand teacher expertise (e.g., McDonald, 2016; Seidel & Stürmer,
2014; Sherin et al., 2008). Goodwin’s perspective was that members of a profession see and shape the world in socially organized and
negotiated ways. In Goodwin’s words, “Central to the social and cognitive organization of a profession is its ability to shape events in
the domain of its scrutiny into the phenomenal objects around which the discourse of the profession is organized” (p. 626).
Much of the prior work applying professional vision to teacher learning (like that cited above) has focused on how teachers see
classroom events. But for Goodwin (1994), professional vision also had an outwardly-focused meaning — members of a professional
community shape how events are seen by others. He showed how events or phenomena are shaped through practices like highlighting
salient features and producing material representations. In our course, teachers exercised this outward sense of professional vision,
shaping classroom events for each other through choices they made while capturing and selecting videos to share (e.g., what to have in
frame, how to bound clips).
The second theoretical perspective we bring to bear is diSessa’s (1993) notion of an aesthetic. This perspective has not previously, to
our knowledge, been applied to studies of teacher learning. The task of producing a video for others to view required our participating
teachers to make choices — sometimes hard choices, as they often had to judge some portion of a video as more worthy of sharing than
others. As we will show, in some cases teachers made these judgments rapidly and holistically; they seemed to immediately know what
they liked, much as someone might look at a painting and know right away whether they nd it appealing.
This sort of rapid, holistic judgment is precisely what diSessa attempts to capture by the notion of an aesthetic (see diSessa, 1993, pp.
185–189). Broadly, it is not too far off to think of an aesthetic by analogy to the everyday meaning of the term. Just as we might apply
an aesthetic sense for paintings or music, teachers might be said to have an aesthetic sense for classroom phenomena. More precisely,
diSessa uses aesthetic as a technical term for a knowledge system with specic properties: An aesthetic includes a moderately large
number of pieces of tacit knowledge that are only weakly related to one another and tend to be invoked through simple recognition.
Also, an aesthetic does not support extended chains of reasoning. Thus, a teacher might say “that was a good discussion” or “that was
an interesting idea,” but we would not expect them to say much about how they arrived at those judgments. As we will show, teachers’
video selections at times took this form.
3. Study context and methods
3.1. Course and participants
We explored teachers’ video-generation activities within the context of a six-week, online PD course hosted by the Teaching
Channel. Shortened from a longer course originally designed by Lomax et al. (2017), the course was designed to support teachers as
they engaged young students in mathematical thinking and discourse, such as articulating mathematical claims and supporting them
with evidence. The course content was cyclical; each week, teachers learned about an argumentation activity (see Table 1 for a
description of all activities), recorded the activity with their students, uploaded a clip of their recording to the collaborative web
platform, and wrote a reection.
20 Kindergarten through 2nd-grade teachers participated in this course — ten teachers in the spring of 2018, and ten in the spring
Table 1
Recorded argumentation activities.
Week Argumentation activity
1 Getting curious about your students: focal students discussed whether two images were the same or different
2 Which one doesn’t belong?: students discussed what they noticed, what goes together, and what doesn’t belong among a set of four images
3 Equality: students lled in the blank in equations in order to make them equal, for example 3 +_ =8 or 3 +_ =5 +_
4 True or false?: students determined whether or not a given equation was true or false, for example 8 =3 +5 or 3 +5 =8 +2 =10
5 Learning from our focal students: like in week 1, focal students again discussed whether two images were the same or different
J. Richards et al.
Learning, Culture and Social Interaction 28 (2021) 100490
4
of 2019. All teachers taught in one of two neighboring suburban, predominantly white, well-resourced school districts in the United
States. In a pre-course survey, teachers reported having one to 28 years of teaching experience (mean =12.5 years). 85% of teachers
reported previously looking at videos of their teaching, though only 35% of teachers reported sharing videos of their teaching with
peers prior to the course.
3.2. Video capturing and selecting
Teachers recorded videos of ve argumentation activities throughout the course (see Table 1). For the rst and nal videos,
teachers recorded conversations of the same activity with the goal of comparing three focal students’ thinking over time. The inter-
mediary videos were each associated with a different argumentation activity. Though these activities were pre-designed, teachers
chose among several variations. We provided instructions on how to use a platform-specic video-recording application, but teachers
could use whatever tools and processes worked best for them. We also reiterated that teachers were to follow school, district, and
university policies and permissions for recording, which were coordinated prior to the course launch, and we invited teachers to record
in ways that were feasible and respectful given permissions and other considerations in their classrooms.
After recording each activity, teachers were asked to select a clip to upload and share with their colleagues. For the 2019 cohort, the
instructions consistently specied that teachers should upload a 3–4-minute clip and reect on what they noticed about their students’
thinking. In 2018, the time frame was less consistently specied because we had not yet realized the impact this would have on
teachers’ selecting processes. During the week that followed each video upload, teachers responded to each other’s videos and re-
ections. They were encouraged to comment on students’ thinking, the types of evidence students were using to support their ideas,
anything that surprised them, questions they had, or how their ideas were similar to or different from those of their colleagues. We also
engaged with teachers around their classroom videos as facilitators, agreeing as a team to let teachers begin the conversations and to
maintain a focus on student thinking in our comments.
3.3. Data sources
Data sources for the study are described in Table 2.
It is important to note that while our interviews with teachers included questions about particular aspects of capturing and
selecting, we intentionally began with open-ended questions like “What was your general plan for recording today?” or “What will you
do with the video recording?” These allowed us to see what teachers brought up organically.
3.4. Data analysis
3.4.1. Analyses of video processes (RQ 1)
We began our analysis of teachers’ video-generation processes by documenting what teachers did and how they did it when
capturing and selecting. To do so, we triangulated descriptions across classroom observations, teacher think-alouds, and interviews
(Bogdan & Biklen, 2007). Drawing on Goodwin’s (1994) description of highlighting, we paid particular attention to what teachers
marked as salient and how they highlighted these objects or features in their videos. Also, for 2019 teachers for whom we had richer
data on selecting, we created annotated selection maps for two representative argumentation activities (Week 2 and Week 3 or 4) per
teacher to triangulate patterns in what teachers included and excluded in their videos. Resonant with how teachers demarcated
classroom events, we drew on existing characterizations of interaction structures in the literature (e.g., Michaels et al., 2016; West
et al., 2013; see Table 3) and an approach inspired by Colley and Windschitl’s (2020) visualizations to code coarse-grained structures
within 18 full teacher-uploaded videos (nine from Week 2, and nine from Week 3 or 4). Videos ranged in length from 7 min to 32 min
45 s; each was coded by two coders in ve-second increments to reect the dominant structure in play, represented by different colors
(see Fig. 1 for an example; interrater reliability ranged from 73% to 97%, and was above 80% for all but two videos). We discussed each
video until a consensus map was reached, then added an outline to mark teachers’ selections, annotated maps with related think-aloud
and interview data as available, and looked across annotated selection maps for patterns.
We then focused deeper analysis on the rationales underlying teachers’ capturing and selecting actions. As we examined teachers’
interviews and think-alouds, we noticed differences in what teachers considered, to what degree, and when. Through initial coding and
discussion by the rst and second authors (Salda˜
na, 2016), we demarcated four high-level categories of considerations, shown in
Table 2
Data sources.
Data source Number
Classroom observations of teachers’ videotaping set-ups, including eld notes, pictures, and videos 30 (1 per teacher in 2018; 2 per teacher in 2019)
Video-taped semi-structured interviews with teachers following observations and at the end of the
course, focused in part on capturing and selecting video (see protocols in Appendix A)
48 (2 per teacher in 2018 (2 teachers did not participate in
2nd interview); 3 per teacher in 2019)
Video-taped think-alouds where teachers documented and described their selecting process as it
occurred (see instructions to teachers in Appendix B)
26 (2 per teacher in 2019 requested, received 1–5 per
teacher)
Video clips of argumentation activities shared by teachers in course 100 (5 per teacher)
Other teacher-captured videos of argumentation activities, either full videos from which clips were
selected or alternate videos
56 (0–4 uploaded per teacher in 2018; ~5 uploaded per
teacher in 2019)
J. Richards et al.
Learning, Culture and Social Interaction 28 (2021) 100490
5
Table 4. We identied more discrete codes within the student considerations category, given both the range of what we saw teachers
considering and the focus of our research.
We also distinguished between degrees of emphasis. We coded a consideration as “foregrounded” if it drove teachers’ actions and
seemed particularly salient to them, as evidenced by spontaneous mention and/or repetition (e.g., during a selecting think-aloud, a
teacher initiated discussion of video clip time limits (a requirement consideration), repeatedly referenced clip length while working with
their video, and trimmed out a part of student discussion that they wished they could keep because of time). We coded a consideration
as “noted” if there was evidence of teachers contemplating it (including if discussed in response to a specic interview question), but it
was less central in teachers’ thinking or actions (e.g., during an interview, when asked, a teacher mentioned once that they generally
lmed during math block so students wouldn’t be thrown off — a student consideration).
Thus, our nal high-level coding approach involved coding whether a given category of considerations was foregrounded, noted, or
not evident in interview and think-aloud data for both capturing and selecting, for each week available, for each teacher. Categories
were not mutually exclusive; teachers at times foregrounded or noted multiple considerations. At this stage, we removed two 2018
teachers from analysis due to incomplete data sets, leaving a total of 46 capturing data points and 56 selecting data points across 18
teachers. The rst and second authors independently coded all 2019 data, with almost perfect agreement with Cohen’s Kappa on all
high-level categories (Student: 0.816, Video: 0.812, Audience: 0.909, Requirement: 0.933). After discussing and resolving differences,
the authors revisited data points coded as student consideration and independently coded for the more specic student codes,
demonstrating substantial or almost perfect agreement across codes (Sharing: 0.687, Thinking: 0.824, Discussion: 0.820, Experiences:
0.917, Management: 0.959). The rst author then completed coding for the 2018 data, and we examined patterns together across
cohorts.
Finally, we performed one additional, more exploratory analysis because we felt that there were some features of teachers’ thinking
during selecting in particular that our coding did not fully illuminate. First, there were nuanced differences in teachers’ statements that
our coding of considerations did not capture. For example, when teachers considered student thinking, they focused on a variety of
features. Second, our coding did not sufciently portray woven patterns of considerations evident in some teachers’ selections.
Furthermore, teachers sometimes struggled to articulate the rationale behind their judgments; they had the character of impressions
based on tacit knowledge. To capture these features of teachers’ thinking, we explored the use of aesthetics as a lens (described in
Section 2.3). We selected four 2019 teachers for whom these types of aesthetic judgments seemed more prevalent and characterized
the range of aesthetics reected in their selecting, as a way of further specifying the thinking involved in teachers’ video generation.
3.4.2. Analysis of video products (RQ 2)
To assess teachers’ self-generated video clips, we analyzed all clips of argumentation activities that teachers selected to share with
colleagues (N =100) according to qualities described in prior literature. These included technical quality, who and what was visible in
the video frame, and aspects of video complexity (e.g., Amador et al., 2020; LeFevre, 2004; Sherin et al., 2009; Superne & Bragelman,
2018). Our coding scheme can be found in Table 5.
The fourth and fth authors independently coded 20% of the video clips sorted across cohorts, teachers, and weeks of the course,
with interrater agreement of 90% or above across all coding categories and almost perfect agreement with Cohen’s Kappa on the three
categories that required judgments of low, medium, or high levels (Windows: 0.916; Depth: 0.837; Clarity: 0.927). The fth author
then coded the remaining 80 video clips and synthesized overall video characteristics.
4. Results
4.1. Teachers’ capturing and selecting processes (RQ 1)
Capturing video and selecting clips required teachers to make many decisions, some of which happened before teachers even hit
record. Teachers decided when and who to videotape, which argumentation activity, device, and app to use, and how to set up the
camera. Other decisions happened during or after recording, such as deciding when to stop or extend a conversation and whether to
select a continuous clip or multiple discrete chunks. Making these decisions drew on a complex set of considerations.
To frame our ndings, we reiterate that teachers negotiated among four categories of considerations in their video-generation
work: student, video, audience, and requirement considerations. In what follows, we bring these considerations to life with a case of a
Table 3
Coding categories for interaction structures.
Code Color on
map
Description
Teacher instructions Orange Teacher introduces what students will do; includes introducing activities, norm-setting (Michaels et al., 2016), and
logistics or transitions (West et al., 2013)
Wait time Green Time for students to independently think or work (Rowe, 1986; cited in Michaels et al.)
Student partner talk Grey Students talk in pairs simultaneously (Michaels et al.)
Student discussion/open
dialogue
Purple Students contribute ideas and/or build on each other’s ideas (West et al.), and/or teacher invites this through open
questioning or prompting
Teacher explanation/closed
dialogue
Blue Teacher explains content or engages in closed dialogue through leading questions (West et al.); similar to teacher
lecture or recitation (Michaels et al.)
J. Richards et al.
Learning, Culture and Social Interaction 28 (2021) 100490
6
Fig. 1. Annotated selection map for Dorothy, Week 4.
Key (see Table 3): orange =teacher instructions; green =wait time; purple =student discussion/open dialogue; grey =student partner talk; blue =teacher explanation/closed dialogue. Equations for
the week’s argumentation activity are bolded under the “instructions” sections. (For interpretation of the references to color in this gure legend, the reader is referred to the web version of this article.)
J. Richards et al.
Learning, Culture and Social Interaction 28 (2021) 100490
7
teacher from the 2019 cohort, Dorothy, who exemplied common high-level patterns in what teachers considered while capturing and
selecting. Then, we look across our dataset at patterns and variations in how teachers orchestrated video considerations and students’
experiences while capturing, and how they curated clips of student thinking to share.
Table 4
Coding categories for considerations during video generation.
Category Description Examples
Student
considerations
Focus is on students — their actions, experiences
Sharing When or to what degree students talked “Last week I left it till Friday… I wasn’t getting a lot out of them, so I just gured
the earlier in the week I did it, the better.” (capturing)
“I usually started it when they were about to share their thinking.” (selecting)
Thinking The substance, depth, or range of students’ ideas “I wanted to see if they could distinguish between one that had the addition, one
that had the subtraction. Like, is it the same problem?” (capturing)
“I love how he says that it depends on what you’re asking. I love that. So, I’m
gonna go back ’cause I want to highlight all of these different things. And
they’re talking about patterns and arrays and how many shapes are in there.”
(selecting)
Discussion How students interacted with each other “I think also to just hear other kids’ opinions kinda gets my three focus students,
it gave them a little something more to work with.” (capturing)
“But the end, I feel like the other-like once one kid was talking, the other two
were not listening at all.” (selecting)
Experiences Perceptions of students’ experiences or feelings “I was just a little worried that maybe the computer would be, the screen would
be distracting for them.” (capturing)
“I like how condent he was in it.” (selecting)
Management Students’ behavior and/or activity if not being lmed “My thought process was, I wanted to make sure… that other kids were
occupied, like at a task where I can trust what they’re doing.” (capturing)
“Toward the end, they were getting squirrelly… So, I think the beginning.”
(selecting)
Video
considerations
Focus is on video — quality, what is visible in frame,
mechanics of creating desired product
“It was harder to hear the kids… it just wasn’t clear; it wasn’t easy to hear on the
iPad.” (capturing)
“I feel like the setup of the camera was ne, except one- the one boy’s head was
kind of in and out, so maybe nding a different spot or facing it a different way
next time.” (capturing)
“Sometimes honestly it was the volume… so I’d be like, ‘Oh, these kids are a
little louder’.” (selecting)
“When you have a video, it only lets- it doesn’t let you cut things from the
middle part, like, I can’t take out what I want out… there might be a different
way of doing it, I just don’t know.” (selecting)
Audience
considerations
Focus is on viewers — anticipations of needs,
preferences
“I tried to mention the problems just in case they couldn’t- people couldn’t see.”
(capturing)
“I think I would do more of a trimming on this one… cutting out a lot of the part
that they’re really not gonna get a lot of conversation out of.” (selecting)
Requirement
considerations
Focus is on course requirements — work with focal
students, time limits, stated purpose of specic activity
“I normally teach math talks whole class. Um, but they sort of mentioned that
having the three focal students would be good.” (capturing)
“It can only be 2 to 3 minutes, so I’m gonna drag this over until I get to like 3
minutes.” (selecting)
Table 5
Coding categories for characteristics of teacher-generated video clips.
Characteristic Coding category Description
Technical quality Audio quality Whether sound in the clip is audible, inaudible, or mixed (hard to hear at times but not throughout)
Video quality Whether the image is clear, unclear, or mixed (blurry/shaky at times but not throughout)
Visibility in video
frame
Visibility of students Groupings of students that appear in the video (individual students, small group, or whole class)
Visibility of teacher If the teacher is in the video (yes/no)
Visibility of artifact of
argumentation activity
If the argumentation activity (e.g., an image being discussed in Which One Doesn’t Belong?) is visible in
the video (yes/no)
Visibility of student work Whether student work is produced, visible, and/or clear in the video
Video complexity Windows Degree of evidence of student thinking in the video — low, medium or high (low =little evidence;
high =detailed information from one or more sources, such as verbal explanations +written work)
Depth General nature of the mathematical thinking displayed in the video — low, medium, or high (low =
routine or rote, high =sense-making at conceptual level)
Clarity How easy it is to understand student thinking in the video — low, medium or high (low =not clear,
high =clear)
Multiple strategies Whether more than one student strategy is depicted in the video (yes/no)
Student-to-student comments Whether students comment on each other’s contributions (e.g., agreeing or disagreeing, revoicing) in
the video (yes/no)
Student errors Whether errors are present in the video (yes/no)
J. Richards et al.
Learning, Culture and Social Interaction 28 (2021) 100490
8
4.1.1. The case of Dorothy
In her video generation, Dorothy — a kindergarten teacher in the 2019 cohort — exemplied considerations that were common
across the dataset with respect to the capturing and selecting processes. During capturing, Dorothy consistently foregrounded video
considerations. For example, Dorothy highlighted the importance of being able to “see the kids talk to each other and their reactions”
Week 1 Interview, 4/23/19) on video, but she did not know where to locate her camera to best capture her students in the frame. So,
she practiced camera placement with a colleague:
So this morning playing around with it- so I was just trying to see the best spot of how I could view all of the kids… during my
planning time today I just stood up there and I had three chairs out and I literally had my co-workers sit in each three chairs to
see if they were… if you could see them on the camera.
(Week 1 Interview, 4/23/19)
She also marked physical artifacts of the argumentation activity (e.g., a sheet of paper with printed images) as important to make
visible on video, and she “show[ed] the camera the pictures rst” before beginning the activity with students. As she experimented
further with her lming set-up, she noted that it was easier to see the artifact when it was projected onto the board instead of at on a
table — another way of highlighting the argumentation activity on video.
In addition to video considerations, student considerations informed Dorothy’s capturing. For example, during the rst week, Dorothy
explained that she gave her students a “heads up” that she would be recording them later that day so that they wouldn’t be thrown off
by something new (student experiences). Sometimes, student and video considerations seemed entangled. During the fourth week,
Dorothy discussed how whole-group conversations supported richer discussions among her students than small group conversations:
I feel like the conversation is better whole group because I can still focus on my three students but I kind of could get the input of
some of the other kids to kind of keep the conversation going.
(Week 4 Interview, 5/14/19)
This student discussion consideration intermingled with video considerations as Dorothy described her experimentation with
capturing video:
So, depending if I do it in a small group or a whole group, I feel like different things work best… The one time for a whole group
lesson I used my computer, it looked really good, and then I went to go play it and I couldn’t hear anything… From a computer
in a small group, it works because you’re closer, but because the computer was so far, the sound in the video I didn’t think it was
that good.
(Week 4 Interview, 5/14/19)
For Dorothy, focusing on technical quality ran parallel to considering what was happening in the classroom for her students,
illustrating the degree and foci of orchestration frequently seen in capturing.
In contrast, while selecting clips to share, Dorothy consistently foregrounded student considerations, highlighting student thinking in
particular. For example, as Dorothy watched her full 20-min recording during the second week, she explained, “At some point [the
students] start to say [the images] are made up of different shapes, and that’s my favorite part of the conversation” (Week 2 Think-
Aloud, 5/1/19). Once she located that specic part of the conversation, she paused the video, noted the time, and eventually trimmed
her clip starting there. At the end of the course, Dorothy shared that when selecting a clip, she specically highlighted moments in
which students went beyond her expectations in their thinking:
I feel like doing the activities I kind of had in my mind, things that they were going to say, and things that they would notice, and
sometimes they said something completely different that I was like, wow, I didn’t even think of that, so I thought that was
awesome. So, [I wanted to select] things like that where I was just like, that’s an amazing idea that you had, things that surprised
me that I didn’t really think they could do, or that I didn’t think they would notice.
(Final Interview, 5/30/19)
It is important to note here that Dorothy did not just select whatever student thinking was present; she sought to identify and
display — to curate — particular kinds of moments. As we will show, for many teachers, selecting involved the curation of certain kinds
of student thinking or discussion.
While student thinking appeared front and center for Dorothy while selecting, she also coordinated this with other considerations. In
particular, Dorothy maintained some focus on requirement considerations. For example, during each of her selecting think-alouds,
Dorothy tracked whether her focal students were included in potential selections, and she monitored video length in relation to the
instructions. Fig. 1 shows an annotated selection map, layered with data from Dorothy’s Week 4 think-aloud, to demonstrate how
multiple considerations appeared within her selecting process. The bar represents a 14 min recording she uploaded, from which she
selected a segment just under 5 min long to share with colleagues (seen within the rectangle). Dorothy watched most of the recording
once, then re-watched her anticipated selection before trimming from both ends.
The section that Dorothy selected centered a particular region of student discussion about two equations. During Dorothy’s think-
aloud, she regularly commented on specic moments of student thinking (e.g., “That was a good point. It’s just reversed”). She also
seemed to draw on a more holistic sense of where conversation was “better,” tied partly to perceptions of student behavior. With
respect to requirement considerations, Dorothy extended the length of her clip beyond the time allotment (“a little bit longer, but that’s
okay”) to include a part where a student exhibited particular condence, and she noted that she got her “focus students in.” This
J. Richards et al.
Learning, Culture and Social Interaction 28 (2021) 100490
9
example showcases how student considerations, particularly student thinking, were foregrounded but also entangled with other con-
siderations during selecting.
4.1.2. Differing foregrounds across capturing and selecting
Dorothy’s case exemplies broader patterns in what teachers considered while capturing and selecting. Table 6 provides an
overview, looking across all teachers, of foregrounded considerations during capturing and selecting, as well as the prevalence of
particular categories throughout the dataset.
1
Table 6 shows that teachers overwhelmingly foregrounded video considerations during capturing — with 78% of teachers doing so
consistently throughout the course. Student considerations, particularly consideration of students’ experiences, also informed teachers’
capturing work (considered in 89% of data points and foregrounded by 83% of teachers at least once). However, in selecting, teachers
tended to foreground student considerations (61% of teachers consistently, 100% of teachers at least once), specically aspects related to
student thinking. Additionally, like for Dorothy, other sets of considerations were often simultaneously in play — video and requirement
considerations were fairly common during selecting, and audience considerations also appeared more in selecting than in capturing.
In what follows, we unpack these distinct process-level patterns. As with Dorothy’s case, we embed discussion of the “whats” and
“hows” of teachers’ processes, and more ne-grained aspects of highlighting, within teachers’ broader considerations. We also
showcase how categories of teachers’ considerations were not contemplated in isolation, but often intersected or were even in tension
with one another.
4.1.2.1. Capturing: orchestrating video considerations and students’ experiences. When capturing classroom discussions, video consider-
ations — particularly how to enhance video quality and make certain aspects of classroom phenomena visible in frame — were top of
mind for teachers. Video considerations were noted in all capturing data points, with 78% of teachers consistently foregrounding this
category. However, teachers also frequently considered their students’ experiences of lming.
For many teachers, capturing quality video was non-trivial. Teachers described having to gure out the affordances of different
devices and camera positions for what you could see and hear on video. In fact, 75% of teachers exhibited variability or a period of trial
and error in determining their preferred approach to lming. Eight of the 20 teachers landed on propping a camera in front of a small
group of students sitting together, typically at a table. Other approaches included holding the camera, propping the camera behind the
whole class on the carpet, or propping the camera behind a single student at the board.
Navigating what they wanted to make visible in their videos was a common video consideration for teachers. As detailed further in a
prior study of how teachers sought to make student thinking accessible on video in this context (Richards et al., 2020), early in the
course, eight teachers explicitly marked the importance of seeing students’ faces and positioned their cameras accordingly; as one
teacher described, “You can see a lot of their thinking, I think, just through their facial expressions too, so I think that’s very important
to show in the video” (Vicky, Week 1 Interview, 4/25/19). Yet, as the course progressed, some teachers increasingly highlighted
students’ interactions with artifacts of the activity (pointing, writing, etc.). For instance, after the rst week, one teacher wished she
had placed the camera differently “so we could have seen the papers so when they were pointing to things, we would know what
they’re talking about” (Audrey, Week 1 Interview, 4/25/19); the following week she changed her set-up accordingly. Filming set-ups
represented a key way in which teachers highlighted desired windows into student thinking.
Sometimes, teachers found that highlighting particular windows was at odds with another consideration — their perceptions of
students’ experiences. One teacher in the 2019 cohort, Vicky, carefully considered how to position and display the camera to students.
She initially propped it in a corner to minimize its intrusiveness, but when she found that the technical quality of the video was better if
she lmed close to and facing students, she placed a piece of construction paper over the screen to mitigate student distraction or self-
consciousness. Other teachers also mediated students’ experiences of lming by talking with them about the lming, exploring or
negotiating participation and set-up with students, and/or positioning the camera in ways that enhanced their comfort (even when
doing so reduced visibility). About one-third of the teachers addressed students’ stated desires to be involved by lming multiple
groups or engaging the whole class in the same mathematical tasks outside of recording for the course.
Beyond students’ experiences, student considerations tended to focus on management (such as how to support quiet yet meaningful
activity for students who were not being lmed, 49% of capturing data points) and to a lesser degree, student thinking (17% of capturing
data points). Given our interest in identifying opportunities for teachers to learn about student thinking, we explored the latter a bit
more. Capturing data points coded for consideration of student thinking shared an emphasis of teachers wanting to see how students
would respond to carefully chosen argumentation activities. For instance, Kendra described how she picked an equation for Week 3:
So, I wanted to do the multiplication rst because we’ve done a lot of addition and subtraction balancing equations, so I wanted
to see if I threw a curveball with multiplication, could they handle that? Would they be able to apply the same concept?
(Week 3 Interview, 5/9/19)
In this and other examples, teachers’ interests in exploring student thinking shaped their capturing actions; the videos then pro-
vided ways to unpack what happened. Further, teachers often reported “hunches” about portions of student thinking they wanted to
select immediately after lming, suggesting that teachers at least tacitly considered student thinking while capturing, with an eye
toward selecting.
1
Recall that we removed two 2018 teachers from consideration as we did not have complete datasets across multiple weeks for these teachers.
J. Richards et al.
Learning, Culture and Social Interaction 28 (2021) 100490
10
4.1.2.2. Selecting: curating student thinking, in coordination with other considerations. Student thinking, in contrast, was the predominant
consideration for teachers when selecting clips. Across interviews and think-alouds, all teachers foregrounded student considerations at
least once, with 61% of teachers doing so consistently over time — and across the 91% of selecting data points coded as student
consideration, 77% showed attention to student thinking, with student discussion next most common at 28%. Teachers also typically
considered at least one other category, coordinating student considerations with requirements, video production mechanics, and/or
perceptions of audience interests.
Annotated selection maps and think-alouds consistently showed patterns akin to Fig. 1 from Dorothy’s case — selection of a single,
continuous video segment that highlighted regions of student discussion and hence thinking, and typically excluded teachers’ in-
structions and explanations (except when these segments introduced a new problem or occurred in the midst of desired sections of
student discussion), wait time, and student partner talk. While there was variability in the location of selected segments, teachers
tended to trim off either or both ends of full captured videos. Most teachers took this trimming approach without question and quickly
identied specic segments based on recollections and hunches they had while capturing. However, several contemplated trimming
out middle portions and stitching together regions of student thinking but described technical challenges in doing so; only one teacher
spliced together different parts of their video. Such mechanics were the primary video consideration while selecting; issues of technical
quality and visibility prevalent during capturing were less central here.
Teachers primarily narrowed in on segments to share by coordinating the curation of certain aspects of student thinking (e.g., “I
think it’s interesting to be able to see what their- what their misconceptions are over a kid who knows how to do it,” Tracey, Week 3
Interview, 5/9/18) with requirement considerations (70% of selecting data points) and/or audience considerations (36% of selecting data
points). For example, one teacher, Kendra, shifted among considerations during her Week 3 think-aloud (5/13/19). Her full video was
over 9 min long and involved students working with two different equations; her trimmed video was 5 min long and kept portions of
students’ thinking around each equation as seen in the annotated selection map in Fig. 2.
Kendra’s major question was whether to include both equations or focus on one — in her Week 3 interview, she anticipated
showing her colleagues students’ approaches to both (student thinking and audience considerations), but she questioned this in relation to
length (requirement consideration) during the think-aloud. She ultimately decided to keep both even though it took her slightly over
time in order to display a “progression from numbers and how they relate what we did with numbers to letters.” Otherwise, Kendra
trimmed based on her perception of what would be “valuable to my colleagues” (audience consideration). She concluded by making a
holistic judgment of her selection, noting “I think the viewers will get a good idea of kind of the conversation as a whole from this
edited video.”
Such judgments showed up across teachers’ selecting data points and seemed to have the underspecied character of aesthetics. We
briey share examples of several of these aesthetics from our exploratory analyses across four 2019 teachers — describing similar
aesthetics that we saw across multiple teachers in Table 7, and digging into an aesthetic we call “meaty discussion” (following a
Table 6
Considerations during video generation.
Category and metric Capturing Selecting
Student considerations # of teachers foregrounding category 15/18 teachers (83%) at least once; 5
consistently (28%)
18/18 teachers (100%) at least once; 11
consistently (61%)
# of data points in which category
was considered
41/46 data points (89%) 51/56 data points (91%)
Breakdown of subcategories Of 41 data points coded as student
considerations
•Sharing: 4/41 (10%)
•Thinking: 7/41 (17%)
•Discussion: 3/41 (7%)
•Experiences: 27/41 (66%)
•Management: 20/41 (49%)
Of 51 data points coded as student
considerations
•Sharing: 9/51 (18%)
•Thinking: 39/51 (77%)
•Discussion: 14/51 (28%)
•Experiences: 2/51 (4%)
•Management: 8/51 (16%)
Video considerations # of teachers foregrounding category 18/18 teachers (100%) at least once; 14
consistently (78%)
9/18 teachers (50%) at least once; none
consistently
# of data points in which category
was considered
46/46 data points (100%) 32/56 data points (57%)
Audience
considerations
# of teachers foregrounding category 4/18 teachers (22%) at least once; none
consistently
7/18 teachers (39%) at least once; 3
consistently (17%)
# of data points in which category
was considered
11/46 data points (24%) 20/56 data points (36%)
Requirement
considerations
# of teachers foregrounding category 6/18 teachers (33%) at least once; 1
consistently (6%)
12/18 teachers (67%) at least once; 3
consistently (17%)
# of data points in which category
was considered
20/46 data points (44%) 39/56 data points (70%)
Note. We used several metrics to explore considerations across the dataset. The number and proportion of teachers who foregrounded a category
provides insight into its salience for teachers in relation to capturing or selecting, particularly when numerous teachers consistently foregrounded the
category over time. The number and proportion of data points in which a category was coded as considered (either foregrounded or noted) showcases
general prevalence across the dataset. Data points of capturing and selecting differ in number due to the additional think-aloud data from the 2019
cohort.
Proportions over 50% are bolded in the table; proportions over 75% are bolded and underlined.
J. Richards et al.
Learning, Culture and Social Interaction 28 (2021) 100490
11
Fig. 2. Annotated selection map for Kendra, Week 3.
Key (see Table 3): orange =teacher instructions; purple =student discussion/open dialogue; green =wait time; blue =teacher explanation/closed dialogue. Equations for the week’s argumentation
activity are bolded under the “instructions” sections. (For interpretation of the references to color in this gure legend, the reader is referred to the web version of this article.)
J. Richards et al.
Learning, Culture and Social Interaction 28 (2021) 100490
12
teacher’s lead) to demonstrate how aesthetics can capture the interwoven yet hard-to-articulate judgments teachers may use in
highlighting particular aspects of classroom phenomena.
In our exploratory analyses, the “meaty discussion” aesthetic was unique to one teacher, Allison. It arose during our rst interview
with Allison, in which she anticipated approaching trimming by nding parts that were “meatier” (Week 1 Interview, 4/24/19). In a
later interview, the term came up again:
… usually I go for the ending because I feel like that’s where like the meatiest conversations come from, but I feel like this one…
at the beginning the kids were more on track with like 3 +5 =5 +3.
(Week 4 Interview, 5/15/19)
Allison stated that she wanted to extract the “meatiest” conversations, which here seemed to equate to where the kids were more
“on track.” The interviewer asked her what she meant by “meaty,” and she explained:
I: When you say meaty, like what do you- what counts as meaty?
A: So, like the more in-depth conversations where they’re like really able to explain their thinking and then the opportunities
where, when it does make sense, and I have other kids like, “Can you repeat what so and so said”? Like those teachable
moments.
Allison’s comments here could be seen as reecting several of the considerations we coded for, such as student thinking and student
discussion. But, as Allison described it, she was only interested in whether the conversation was “meaty.” The other terms she used to
characterize what counts — in-depth, on track, able to explain their thinking — feel more like attempts to put into words what she
meant by “meaty” rather than separate considerations. Allison applied this aesthetic frequently throughout the course, and her usage
was relatively consistent with the explanation she gave above, except that student discussion seemed to be less important than originally
implied. During her nal interview, for example, Allison explained:
So, I always went for like, the more meatier. Because it just made sense to me… I feel like the conversations that were meatier
were more valuable and I feel like people could see, you know, how different kids think and how different kids explain things
and I think that was more important than like setting the stage for- because sometimes the conversations start off kinda slow and
then toward the middle and the end is where students start to make those connections.
(Final Interview, 5/13/19)
Here, Allison focused more on student thinking (“how different kids think… explain things”) in her depiction of what was meatier.
She also linked audience considerations with her “meaty discussion” aesthetic — meaty portions were most “valuable” for her colleagues
to see. This example demonstrates again how teachers’ selecting processes were driven by student considerations, including nuanced
aesthetic judgments, yet often coordinated with other sets of considerations as well.
Table 7
Video selection aesthetics.
Aesthetic Description Examples
Meaty discussion Select segments with substantive student
talk
“So, I always went for like, the more meatier. Because it just made sense to me. Even the
like introduction would be helpful for my colleagues to see like how the conversation
started, but I feel like the conversations that were meatier were more valuable and I feel
like people could see, you know, how different kids think and how different kids explain
things…” (Allison, End-of-Course Interview, 5/13/19)
Good conversation Select based on whether students listen
and react to each other, and whether group
is on-task
“Watch the video, um, and just kind of… gure out where the best conversation was.”
(Dorothy, Week 1 Interview, 4/23/19)
“I remember by the end, like [student] had to get up a couple times to go get tissues. He’s
got a runny nose. And by the end I feel like the other- like once one kid was talking the
other two were not listening at all. That’s like my hardest part is they don’t listen to each
other.” (Dorothy, Week 1 Interview, 4/23/19)
Interesting idea Select for individual student ideas that are
novel, surprising
“I wanted to make sure I got- had captured a clip that had… any comments that stood out
to me or that I found really interesting or different or that were surprising… I kind of had in
my mind things that they were going to say and things that they would notice, and
sometimes they said something completely different that I was like wow, I didn’t even
think of that.” (Dorothy, End-of-Course Interview, 5/30/19)
Idea development Select to show changes in student thinking “I think I’d like to challenge how [student]- because she had rst chosen to add the two
numbers together… And how she changed her mind because she realized oh wait, just
because it comes at the beginning doesn’t make it different. It could- it’s just like as if it was
at the end, they just moved it to a different spot. And I think that realization was really
important for her.” (Vicky, Week 3 Interview, 5/9/19)
Audience story Select segments to convey to colleagues the
story of what transpired
“So my decision, I’m kind of deciding right now is do I wanna just crop the, um, the part of
the letters equation because it was kind of an interesting conversation, or do I wanna keep
both because there were some really good parts to the other as well. I still think I’m going
to keep both because I think it’s good to see like the progression from numbers and how
they relate what we did from numbers to letters… I think the viewers will get a good idea
of kind of the conversation as a whole from this edited video.” (Kendra, Week 3 Think-
Aloud, 5/13/19)
J. Richards et al.
Learning, Culture and Social Interaction 28 (2021) 100490
13
4.2. Characteristics of teacher-generated video clips (RQ 2)
Given teachers’ processes and considerations, what were their nal video products like? On the whole, teacher-generated clips had
high technical quality. 100% of the video clips were coded as clear video quality; 92% of the clips (92/100) were coded as audible, with
8% mixed (some portions audible, others not). Since teachers used their own devices and largely devised their own capturing set-ups,
we found the high degree of technical quality notable — and likely reective of the degree of attention given to video considerations
during capturing, as well as the remarkable progress in recording capabilities built into widely accessible devices.
With respect to who and what was visible in the videos, the majority of clips (77%) showed students in small groups, followed by
the whole class (16%) and individual students (7%). The teacher was typically not visible, and artifacts of the argumentation activity
were made visible in 66% of clips. Students produced written work in only 25% of clips, clustered in Weeks 3 and 4 where the
argumentation activities involved equations. Overall, clips from eight of the twenty teachers remained fairly consistent and repre-
sentative of these broader patterns over time. Other teachers showed shifts over time in the groupings lmed (such as Dorothy’s
increasing proclivity to lm whole-class conversations) or increasingly made artifacts of the argumentation activity visible (evident for
ve of the twenty teachers).
Most interesting to us was the complexity evident in selected clips. The vast majority of videos were coded as high depth (97%, 96/
99 videos
2
), meaning they generally showed conceptual sense-making as opposed to rote responses; we imagine this particular result is
highly tied to the designed argumentation activities in the course. Most videos were also coded as high in windows (98%, 97/99
videos) and clarity (84%, 83/99 videos), providing access to detailed information about student thinking that could be clearly un-
derstood. Finally, teachers’ videos showed multiple strategies 91% of the time, student-to-student comments 70% of the time, and
student errors (relative to canonical mathematical understandings) 46% of the time — characteristics that were often intentional
choices by teachers during selecting.
5. Discussion
New opportunities and increasing requests for teachers to record videos of their own classrooms as part of PD are sparking
increased attention to what doing so means for teachers. Here, we build from our analyses to discuss takeaways and trade-offs evident
in our PD context, and we consider potential implications for teacher learning.
5.1. Primary takeaways
We begin with two main takeaways in relation to the study’s research questions. First, when working with a) classroom activities
designed to spark student discussion and b) readily available, common technology in the classroom, practicing teachers in this study
generated videos with many of the characteristics noted in the literature as important for promoting noticing and analysis of student
thinking. Specically, teachers’ clips had sufcient technical quality and “windows” into student thinking (Sherin et al., 2009) to
unpack students’ ideas. Further, teachers’ selections often reduced what Superne and Bragelman (2018) would refer to as “non-CMT
[children’s mathematical thinking] moments” in the videos, and included characteristics shown to enhance noticing of student
thinking (e.g., showing multiple strategies, including incorrect ones). This is an important nding with respect to the feasibility of
using teacher-generated videos to support noticing of student thinking in PD contexts.
Second, generating these video clips was complex work for teachers. Unlike one might think, the technical requirements were not
the sole — or even primary — locus of the complexity. Rather, teachers’ interviews and think-alouds demonstrated the need they felt to
coordinate multiple, at times competing, sets of considerations that spanned their students, the video itself, video viewers, and the
course. These considerations resonate with factors elevated in prior work (e.g., technical quality challenges, as noted by Zhang et al.
(2011)), and extend understanding of the range of student considerations teachers may grapple with as they lm while teaching. While
capturing, teachers in our study navigated trade-offs in orchestration, toggling between video considerations (such as what they could
hear and see) and student considerations (such as students’ comfort level or the quality of the conversation). In contrast, selecting
seemed to spark curation, with teachers generally foregrounding student thinking and seeking to display particular forms of thinking in
their videos while coordinating with other considerations. Selecting also seemed to activate and be shaped by largely tacit knowledge
systems that we described as teachers’ aesthetics, which teachers drew on to highlight particular aspects of classroom phenomena over
others.
5.2. Implications for teacher learning and PD design
How teachers engaged in capturing and selecting suggest ways in which these processes, as well as particular aspects of our course
design, may afford or constrain teacher learning opportunities. Here, we focus particularly on learning opportunities with respect to
student thinking — a main focus of the course and numerous PD efforts in the eld.
During the capturing process, teachers’ attention was most drawn to video considerations and students’ experiences — important for
accessible, respectful video generation, but which may reduce explicit attention to student thinking. We anticipate that this is not
2
One video was removed from consideration for video complexity coding due to mixed audio quality.
J. Richards et al.
Learning, Culture and Social Interaction 28 (2021) 100490
14
unique to our context, given other studies’ depictions of teachers’ focus on technical aspects of lming. However, some teachers
treated capturing as an opportunity to explore questions about student thinking. They intentionally chose activities grounded in their
current perceptions of student understanding and anticipated how students may respond — similar to what Sherin and Dyer (2017)
depicted as teachers “preparing to notice” student thinking (p. 487) in other video-generation contexts.
This led us to consider why such exploration and anticipation of student thinking was not more evident in our capturing dataset.
One possibility connects to course design. In our context, teachers were provided with designed argumentation activities to try with
students (though they had a choice in which specic variant(s) to use) and were simply asked to capture the conversation with stu-
dents, with an eye toward later unpacking student thinking. Such parameters and prompts may not have afforded the more open-ended
anticipation and exploration we may have seen if teachers had more latitude in lessons or different prompts — an area for further, more
systematic study. For instance, what would we see if we invited teachers to explicitly identify a question about student thinking that
they wanted to explore through recording? Another possibility relates to our research design; we only interviewed teachers after they
lmed, and we are curious what we might have heard if we talked with them as they planned for lming.
Within the selecting process, asking teachers to select clips that included student thinking productively drew their attention to what
students offered, and teachers generally prioritized student thinking over other considerations when making decisions. Teachers
highlighted student thinking in multiple ways — they pointed out ideas that were novel to them, described the range of thinking they
saw, and singled out specic moments when students changed their thinking. This consistent and multifaceted focus on student
thinking suggests that selecting might have substantial affordances for teacher learning.
Our observations also suggest some limits to potential benets of selecting, at least for some teachers. Teachers in our study
approached selecting in varied ways; some teachers watched their captured videos in full, providing opportunities as seen in Sherin
and Dyer’s (2017) study for reassessing (and potentially revising) their impressions of student thinking. Other teachers, though,
operated from recollections or hunches when selecting, quickly narrowing in on specic video segments without revisiting the full
discussion, which may constrain learning opportunities.
Additionally, if teachers’ video selections depend on aesthetics — largely tacit, hard-to-articulate knowledge systems in which
considerations are woven together — then there is reason to believe that learning may be slow. From an aesthetics perspective,
“learning” may look like small and gradual shifts in the cuing of particular aesthetics in particular contexts, or rening the features
associated with particular aesthetics (e.g., “good conversation” becomes increasingly tied to moments when students initiate inter-
action with the substance of each other’s contributions). Such learning may require repeated experiences of selecting excerpts of videos
over time and may be enhanced in contexts where varied aesthetics are made visible and collaboratively explored.
6. Conclusion
Overall, our exploration of teachers’ engagement with capturing video and selecting clips to share in our particular PD context
leaves us feeling optimistic about the potential for video generation to productively impact the profession of teaching. While the
complexities and dynamics of video generation leave open the possibility of less successful outcomes, and particularly need to be
explored with diverse and/or under-resourced communities, in this study’s context teachers were able to produce videos with
generative qualities largely from their own devices and designs for recording. Further, teachers’ video-generation processes demon-
strated careful orchestration of their lming and classroom activity, and intentional curation when selecting clips. Selecting in
particular seemed to lead teachers to focus on and think about students’ ideas.
It is always difcult in studies like this, where elements of teaching are examined outside of the classroom, to predict the extent to
which there will be an impact on instruction. This remains a direction for further research. However, prior research on video-based PD
has found teachers’ focus on student thinking to carry over to their classroom practice (e.g., Cohen, 2004; van Es & Sherin, 2010).
Moreover, we envision broader possibilities in which video-generation activities are not discrete PD tasks, but rather part and parcel of
the professional work of teaching as teachers inquire into the thinking and learning of their students, with the support of colleagues.
Declaration of competing interest
None.
Acknowledgements
We would like to express our appreciation to the teachers who taught us about what capturing and selecting video entailed in
practice, and to Dr. Tracy Dobie, Sarah Larison, and undergraduate research assistants Jennifer Fishman, Kelsey Morton, and Jordyn
Ricard for their contributions to this work.
Appendix A. Semi-structured interview protocols
This appendix includes the portions of our interview protocols that focused on teachers’ capturing and selecting processes. Protocol
1 was used at the end of classroom observations of teachers’ videotaping set-ups, which occurred once during the 2018 course and
twice during the 2019 course. Protocol 2 was used at the culmination of the course and included additional topics not shown here; we
include several overview questions we asked about teachers’ work with video early in the interview because teachers sometimes
brought up capturing and selecting considerations in response to these questions. Throughout the protocols, "XX" reects places where
J. Richards et al.
Learning, Culture and Social Interaction 28 (2021) 100490
15
interviewers tailored questions based on their observations with specic teachers.
Protocol 1: post-observation interview
1. First, how do you feel the recording process went overall?
2. What was your general plan for recording today?
3. Can you tell me a bit about how you made the decision to record today, in particular?
a. If teacher recorded part of lesson: It looked like you recorded only XX part of the lesson/activity. Why did you decide to record
that part in particular?
b. If teacher recorded whole lesson: It looked like you recorded the whole lesson. Why did you decide to do that?
4. I saw you used XX (computer, iPad, etc.) for recording. Where did you get the device from? Why did you decide to use it?
5. [2019 only] Can you tell me how you decided where to position the camera?
a. Follow-up: What did you hope to get in the frame?
b. If teacher moved the camera during recording: I saw that you moved the camera partway through. What was going on there?
c. If student recorded: What instructions did you give the student?
6. [2018 only] If camera was stationary: I saw that your camera was placed on XX. Can you tell me a bit about your process in
deciding where to position the camera?
7. Was there anything about the recording process that happened that was unexpected?
8. [2018 only] Did anything come up that surprised or concerned you about recording?
9. [2nd observation for 2019 only] What changed in your recording process from the rst time I visited to today? Why did you
make those changes?
10. Is there anything else you want to add about the experience of recording/being recorded in your classroom?
11. For this week, you have to upload a segment of what you just recorded that’s XX minutes long. Can you tell me a bit about what
your next steps might be? What will you do with the video recording?
a. Follow-up: When do you plan to do that?
b. Follow-up: Do you have any ideas about which part(s) you’ll want to be sure to include in your uploaded clip? Why that/those
parts?
12. Do you anticipate facing any challenges during these steps (selecting a clip, uploading, writing a reection)?
Protocol 2: end-of-course interview
Overview questions related to video
1. What are your big takeaways from the course?
a. If teacher doesn’t mention video: Anything you’re taking with you from your work with video?
2. Is there anything from the course you plan to use in the future?
a. [2018 only] Follow-up: And how are you feeling about videotaping in your classroom?
b. If teacher doesn’t mention video: Do you anticipate videotaping at all in the future? If so, for what purpose?
3. Are there any changes you’ve made to your own teaching practice since you started participating in the course?
a. [2019 only] Follow-up: Did you feel like videotaping changed your teaching in any way? Can you give an example?
Capturing and selecting questions
[2018 questions]
1. Can you tell me about the process of videotaping in the classroom?
2. How do you choose when to lm?
3. Who typically lms, or is the device propped up?
4. How do you decide where to point the camera?
a. Follow-up: Does it depend on what you’re recording?
5. How do you select a clip to share?
a. Follow-up: How did you cut your clips?
6. Are there any differences in your process for lming small groups versus the whole class?
7. Did you change your process from the beginning to the end of the course?
8. How was the experience of sharing video with your colleagues?
a. Follow-up: What makes an interesting video clip to share?
[2019 questions]
1. Do you feel like you landed on a “best” method for videotaping in your classroom?
a. If yes: What was it, and why was this best for you?
b. If no: What would you try next, and why?
J. Richards et al.
Learning, Culture and Social Interaction 28 (2021) 100490
16
2. When I visited, you lmed XX (describe set-up). Did you ever lm in any other ways? Why or why not?
3. Did you tend to record the whole discussion or particular parts? Why?
4. Did you lm one time per activity or multiple times? Why?
5. How did you feel videotaping impacted the students who were videotaped? Did this change from the beginning of the course till
now?
6. How did you feel videotaping impacted your classroom as a whole? Did this change from the beginning of the course till now?
7. What were some important criteria to you in terms of selecting portions of video to share?
8. Did you tend to know what you wanted to select before revisiting the video, or did you choose when you revisited the video?
a. If beforehand: Were there any times that your thinking changed after revisiting the video?
9. When you were physically trimming your video, what did you use to identify where to start or stop? (something you could see
onscreen, timestamps, etc.?)
10. How was the experience of sharing video with your colleagues?
Appendix B. Think-aloud instructions for teachers
We asked teachers in the 2019 cohort to record their own selecting/trimming process two times during the course. Fig. B.1 shows
the instructions they received; we shared different ways of screen recording at an in-person kick-off meeting, or teachers could opt to
have us videotape their process instead.
Fig. B.1. Think-aloud instructions for teachers.
J. Richards et al.
Learning, Culture and Social Interaction 28 (2021) 100490
17
References
Amador, J. M., Keehr, J., Wallin, A., & Chilton, C. (2020). Video complexity: Describing videos used for teacher learning. EURASIA Journal of Mathematics, Science and
Technology Education, 16(4), Article em1834. https://doi.org/10.29333/ejmste/113288.
Barnhart, T., & van Es, E. (2015). Studying teacher noticing: Examining the relationship among pre-service science teachers’ ability to attend, analyze and respond to
student thinking. Teaching and Teacher Education, 45, 83–93.
Beisiegel, M., Mitchell, R., & Hill, H. C. (2018). The design of video-based professional development: An exploratory experiment intended to identify effective features.
Journal of Teacher Education, 69(1), 69–89.
Bogdan, R. C., & Biklen, S. K. (2007). Qualitative research for education: An introduction to theory and methods (5th). Boston, MA: Pearson Education, Inc.
Borko, H., Jacobs, J., Eiteljorg, E., & Pittman, M. E. (2008). Video as a tool for fostering productive discussions in mathematics professional development. Teaching and
Teacher Education, 24, 417–436.
Calandra, B., Gurvitch, R., & Lund, J. (2008). An exploratory study of digital video editing as a tool for teacher preparation. Journal of Technology and Teacher
Education, 16(2), 137–153.
Clark, J. C., Chittleborough, G., & Chandler, P. (2019). Video as a second stimulus in developing the professional agency of primary pre-service teachers. In L. Xu,
G. Aranda, W. Widjaja, & D. Clarke (Eds.), Video-based research in education: Cross-disciplinary perspectives (pp. 66–82). New York, NY: Routledge.
Cohen, S. (2004). Teachers’ professional development and the elementary mathematics classroom: Bringing understandings to light. Mahwah, NJ: Lawrence Erlbaum.
Colley, C., & Windschitl, M. (2020). A tool for visualizing and inquiring into whole-class sensemaking discussions. Research in Science Education. https://doi.org/
10.1007/s11165-020-09962-6.
Derry, S. J., et al. (2010). Conducting video research in the learning sciences: Guidance on selection, analysis, technology, and ethics. Journal of the Learning Sciences,
19(1), 3–53.
diSessa, A. A. (1993). Toward an epistemology of physics. Cognition and Instruction, 10(2–3), 105–225.
van Es, E. A., & Sherin, M. G. (2008). Mathematics teachers’ “learning to notice” in the context of a video club. Teaching and Teacher Education, 24(2), 244–276.
van Es, E. A., & Sherin, M. G. (2010). The inuence of video clubs on teachers’ thinking and practice. Journal of Mathematics Teacher Education, 13(2), 155–176.
Fadde, P. J., Aud, S., & Gilbert, S. (2009). Incorporating a video-editing activity in a reective teaching course for preservice teachers. Action in Teacher Education, 31
(1), 75–86.
Gaudin, C., & Chali`
es, S. (2015). Video viewing in teacher education and professional development: A literature review. Educational Research Review, 16, 41–67.
Goodwin, C. (1994). Professional vision. American Anthropologist, 96(3), 606–633.
LeFevre, D. M. (2004). Designing for teacher learning: Video-based curriculum design. In J. Brophy (Ed.), Using video in teacher education (pp. 235–258). San Diego,
CA: Elsevier.
Lomax, K., Fox, A., Kazemi, E., & Shahan, E. (2017). Argumentation in mathematics (An online course developed in partnership with the Teaching Channel).
Major, L., & Watson, S. (2018). Using video to support in-service teacher professional development: The state of the eld, limitations and possibilities. Technology,
Pedagogy and Education, 27(1), 49–68.
Masats, D., & Dooly, M. (2011). Rethinking the use of video in teacher education: A holistic approach. Teaching and Teacher Education, 27(7), 1151–1162.
McDonald, S. P. (2016). The transparent and the invisible in professional pedagogical vision for science teaching. School Science and Mathematics, 116(2), 95–103.
Michaels, S., O’Connor, M. C., Hall, M. W., & Resnick, L. B. (2016). Accountable talk sourcebook: For classroom conversation that works. Pittsburgh, PA: Institute for
Learning, University of Pittsburgh.
Nagro, S. A., deBettencourt, L. U., Rosenberg, M. S., Carran, D. T., & Weiss, M. P. (2017). The effects of guided video analysis on teacher candidates’ reective ability
and instructional skills. Teacher Education and Special Education: The Journal of the Teacher Education Division of the Council for Exceptional Children, 40(1), 7–25.
Reisman, A., & Enumah, L. (2020). Using video to highlight curriculum-embedded opportunities for student discourse. Journal of Teacher Education., 71(15), 551–567.
Richards, J., Altshuler, M., Sherin, B. L., & Sherin, M. G. (2020). Orchestrating for seeing: How teachers see and help others see student thinking when self-capturing
classroom video. In M. Gresal, & I. S. Horn (Eds.), volume 4. The interdisciplinarity of the learning sciences (pp. 1942–1949). Nashville, TN: International Society of
the Learning Sciences.
Richards, J., & Robertson, A. D. (2016). A review of the research on responsive teaching in science and mathematics. In A. D. Robertson, R. E. Scherr, & D. Hammer
(Eds.), Responsive teaching in science and mathematics (pp. 36–55). New York, NY: Routledge.
Roschelle, J. (2000). Choosing and using video equipment for data collection. In A. Kelly, & R. Lesh (Eds.), Handbook of research design in mathematics and science
education (pp. 709–731). Mahwah, NJ: Lawrence Erlbaum.
Roth, K. J., et al. (2017). Design principles for effective video-based professional development. International Journal of STEM Education, 4(1), 31.
Rowe, M. B. (1986). Wait time: Slowing down may be a way of speeding up! Journal of Teacher Education, 37, 43–50.
Salda˜
na, J. (2016). The coding manual for qualitative researchers. Thousand Oaks, CA: SAGE.
Sang, G., Valcke, M., van Braak, J., Zhu, C., Tondeur, J., & Yu, K. (2012). Challenging science teachers’ beliefs and practices through a video-case-based intervention
in China’s primary schools. Asia-Pacic Journal of Teacher Education, 40(4), 363–378.
Seidel, T., & Stürmer, K. (2014). Modeling and measuring the structure of professional vision in preservice teachers. American Educational Research Journal, 51(4),
739–771.
Sherin, M. G. (2004). New perspectives on the role of video in teacher education. In J. Brophy (Ed.), Using video in teacher education (pp. 1–27). New York, NY: Elsevier
Science.
Sherin, M. G., & Dyer, E. B. (2017). Mathematics teachers’ self-captured video and opportunities for learning. Journal of Mathematics Teacher Education, 20(5),
477–495.
Sherin, M. G., Linsenmeier, K., & van Es, E. A. (2009). Selecting video clips to promote mathematics teachers’ discussion of student thinking. Journal of Teacher
Education, 60(3), 213–230.
Sherin, M. G., Russ, R. S., Sherin, B. L., & Colestock, A. (2008). Professional vision in action: An exploratory study. Issues in Teacher Education, 17(2), 27–46.
Stockero, S. L., Rupnow, R. L., & Pascoe, A. E. (2017). Learning to notice important student mathematical thinking in complex classroom interactions. Teaching and
Teacher Education, 63, 384–395.
Superne, A., & Bragelman, J. (2018). Analyzing the impact of video representation complexity on preservice teacher noticing of children’s thinking. Eurasia Journal
of Mathematics, Science and Technology Education, 14(11), Article em1650. https://doi.org/10.29333/ejmste/99501.
Tripp, T. R., & Rich, P. J. (2012). The inuence of video analysis on the process of teacher change. Teaching and Teacher Education, 28(5), 728–739.
Vossoughi, S., & Escud´
e, M. (2016). What does the camera communicate? An inquiry into the politics and possibilities of video research on learning. Anthropology &
Education Quarterly, 47(1), 42–58.
Weil, M., Seidel, T., Schindler, A.-K., & Gr¨
oschner, A. (2020). Opening “windows” for teachers to change classroom discourse. Learning, Culture, and Social Interaction,
26, 100425.
West, E. A., Paul, C. A., Webb, D., & Potter, W. H. (2013). Variation of instructor-student interactions in an introductory interactive physics course. Physical Review
Special Topics - Physics Education Research, 9, Article 010109.
Xiao, B., & Eriksson, M. D. (2020). EDTPA is coming: Video makes embodied practices visible. International Journal of Social Policy and Education, 2(1), 125–133.
Yerrick, R., Ross, D., & Molebash, P. (2005). Too close for comfort: Real-time science teaching reections via digital video editing. Journal of Science Teacher Education,
16, 351–375.
Zhang, M., Lundeberg, M., Koehler, M. J., & Eberhardt, J. (2011). Understanding affordances and challenges of three types of video for teacher professional
development. Teaching and Teacher Education, 27(2), 454–462.
J. Richards et al.
