Teaching and Teacher Education 122 (2023) 103951

Contents lists available at ScienceDirect

Teaching and Teacher Education

journal homepage: www.elsevier.com/locate/tate

Research paper

Using lesson study to change teacher knowledge and practice: The

role of knowledge sources in teacher change
Amber E. Benedict a, *, Jessica Williams b, Mary T. Brownell b, Lindsey Chapman c,
Alexandria Sweers a, Hyojong Sohn b
a
Mary Lou Fulton Teachers College, Arizona State University, P.O. Box 871811 Tempe, AZ 85287, USA
b
College of Education, University of Florida, USA
c
Wheelock College of Education and Human Development, Boston University, USA

h i g h l i g h t s

Lesson Study is a professional development model that supports teacher learning.

Lesson Study combined curriculum leads to improved teacher knowledge and practice.

Lesson Study helps teachers understand and design lessons to meet support students.

There is a need for more comprehensive, rigorous research on Lesson Study

a r t i c l e i n f o a b s t r a c t

Article history: In this systematic review, we synthesize the findings of 13 studies identifying sources of knowledge
Received 4 December 2020 within Lesson Study professional development (LSPD) and examine influences of knowledge sources on
Received in revised form teachers’ content knowledge, understanding of students, and changes in pedagogical practices. Finally,
2 October 2022
we consider the implications of our findings for design of and future research on LSPD. We find that the
Accepted 7 November 2022
Available online 22 November 2022
literature base for LS is extensive, however, there are few studies that meet criteria for high-quality
research. All studies reviewed reported a positive impact of LS and included a wide variety of LS
structures and knowledge sources.
Keywords:
Esson study
© 2022 Elsevier Ltd. All rights reserved.
Collaboration
Teacher development
Teacher knowledge
Teacher practice

Emerging evidence suggests that teachers’ professional learning collaborative interactions with colleagues in job-embedded con-
is enhanced when professional development (PD) activities are texts to promote teacher learning (Benedict et al., 2013). Initially
collaborative (Kraft & Papay, 2014; Ronfeldt et al., 2015) and occur developed in Japan, LS was introduced to the United States by
within job-embedded contexts (Darling-Hammond, 2017; Stigler and Hiebert (1999) and is now commonly used throughout
Thurlings & den Brok, 2017). Scholars promoting such PD hypoth- the U.S. (Akiba, 2016; Yoshida, 2012) and internationally. LS utilizes
esize that teachers learn through conversations with and obser- a cyclical structure: teachers iteratively analyze student work, use
vations of colleagues related to: (a) expectations for their work, (b) this analysis to inform collaboratively planned lessons, implement
content knowledge for teaching students, (c) how to use new those lessons while observing one another's implementation, and
practices, (d) how to improve practices they are already using, and analyze the lesson's quality (Gersten et al., 2010). These activities
(e) why students might struggle with specific content (Achinstein & are repeated in multiple consecutive cycles over time. Theoretically,
Barrett, 2004). LS enables teachers to learn by engaging in deep analysis of in-
Lesson study (LS) is one PD approach that depends mainly on struction and reflection on student learning through collaborative,
critical discussions that unfold when teachers plan, observe, and
reflect on student learning (Stewart & Brendefur, 2005). Re-
* Corresponding author. searchers suggest that during LS, teachers engage in analysis of
E-mail address: Amber.Benedict@asu.edu (A.E. Benedict).

https://doi.org/10.1016/j.tate.2022.103951
0742-051X/© 2022 Elsevier Ltd. All rights reserved.
A.E. Benedict, J. Williams, M.T. Brownell et al. Teaching and Teacher Education 122 (2023) 103951

student data, learning standards, and curriculum to develop short- Willems and van den Bosche (2019) were the only authors who
and long-term learning goals and improve their capacity to deliver applied quality indicators to included studies. They included five
instruction (Hiebert et al., 2007; Lewis et al., 2006), and studies of experimental or quasi-experimental studies. Two of the five studies
LS demonstrate that teachers do increase their understandings and came from the same data set (Lewis & Perry, 2014, 2017). One of the
pedagogical content knowledge, use of specific practices, ability to studies examined preservice teachers, who may differ from in-
engage in collaborative work, and reflection (e.g., Dudley, 2013; service teachers in their ability to plan for and analyze instruction
Lewis et al., 2009; Norwich & Ylonen, 2013; Puchner & Taylor, because of their novice status. Of these studies, two demonstrated
2006). that LS resulted in changes in teachers' knowledge (Lewis & Perry,
Although findings from studies of LS are encouraging, it must be 2014, 2017; Mutch-Jones et al., 2012), and three demonstrated
recognized that LS approaches are constructed and facilitated in changes in instructional skill (Chizhik et al., 2017; Mutch-Jones
different ways. These differences are likely to influence teacher et al., 2012; Schipper et al., 2017). The authors claim that these
outcomes. For example, traditional LS, first implemented in Japan, studies provide significant evidence for the effectiveness of LS in
is relatively uniform. Teachers collaboratively analyze curriculum increasing knowledge and instructional skill. However, we find that
to plan lessons and the student learning that ensues once lessons claim to be somewhat overstated given the limitations authors
are implemented. Since Japan has a National course of study, noted related to variations in how LS was implemented, instru-
teachers likely have more uniform resources from which to un- mentation issues in two of the studies, and variations in how
derstand content, pedagogy, and student learning. However, in the knowledge and instructional skills were measured. Despite limi-
broader research literature on LS, sources of knowledge vary tations noted by the authors, the studies reviewed show that LS
considerably from study to study. These variations in knowledge holds promise for improving teachers’ knowledge of instructional
sources include teachers (who are often prepared differently, content and pedagogy.
particularly in the United States), researchers or university faculty
who facilitate the LS, researcher-generated, national or district- 2. Rationale and purpose
selected curriculum, and content from PD workshops or institutes
combined with LS. These variations provide teachers access to Findings from literature reviews indicate wide variation in how
different sources of knowledge within the LS process and, in turn, LS is conducted and the quality of implemented studies. These
likely affect what they learn. For instance, when teachers lead LS, variations likely provide teachers with different learning opportu-
what they learn is dependent on the knowledge of those involved, nities, potentially influencing what teachers learn about teaching
the knowledge gained by enacting curriculum, and their ability to content, student learning, and instructional practice (Lewis et al.,
analyze instruction and its impact on student learning from their 2006).
curriculum. In contrast, researcher-led LS combined with PD The purpose of this systematic literature review is to extend and
workshops or institutes afford teachers access to two additional refine previous literature reviews. We examine sources of knowl-
sources of knowledge (i.e., PD content, the researcher). These varied edge available to teachers in LS studies (e.g., participating teachers,
LS configurations offer teachers access to different sources of curriculum, content experts) and identify the changes in teachers'
knowledge, which in turn, likely influence how teachers’ un- knowledge, understanding, and instructional skill. We focus on
derstandings of content, pedagogy, and students evolve. teacher knowledge, understanding of teaching and student
Our review explores how variations in the provision of LS, learning, and instructional practice, as these changes are likely to
specifically the sources of knowledge teachers can access, lead to ultimately influence student learning (Gess-Newsome et al., 2019).
changes in teachers’ understandings of content, pedagogy, and Further, we include only studies that meet quality indicators for
student learning. We intend to extend the results of nine previous qualitative, quantitative, and mixed methods research, as doing so
research syntheses conducted in the last 12 years (Cheung & Wong, supports the veracity of findings and conclusions reported. Spe-
2013; Dudley et al., 2019; Ko, 2019; Norwich et al., 2018, 2021; cifically, we examined the following questions:
Saito, 2012; Seleznyov, 2018; 2019). Additionally, we aim to identify
high-quality studies with robust research designs, methodology 1. What knowledge sources are situated in the different LS ap-
and well-described methods of analysis that can be referred to as proaches described in the reviewed studies?
exemplars for future research in LS. 2. What changes unfold in participating teachers' content knowl-
edge for teaching, understanding of instruction and students,
1. Prior literature syntheses and instructional practice?
3. What are the implications of findings from these studies for
In previous syntheses, researchers examined LS from various constructing LS in the future, particularly for how knowledge
angles, making it difficult to draw conclusions about the research. sources might be incorporated into the LS process?
For example, researchers have examined (a) different ways in
which LS was configured and how those configurations related to
other PD approaches, (Norwich et al., 2018); (b) how processes of 3. Methods
teacher collaboration and teacher learning were framed in Lesson
Study and Learning Study (Ko, 2019); (c) how approaches to LS Relevant studies were selected through an extensive literature
aligned with elements of inclusive practices for students with search, screening and eligibility process. This process is described
disabilities (Norwich et al., 2021); and (d) how LS implemented below and illustrated in Fig. 1.
outside of Japan is structured according to critical components
identified in Japanese LS (Seleznyov, 2018). Four literature syn- 3.1. Literature search
theses examined the influence of LS on teacher and student out-
comes (Cheung & Wong, 2014; Dudley et al., 2019; Seleznyov, 2019; A literature search was conducted using two educational data-
Willems & van den Bosche, 2019); however, quality indicators were bases, ERIC and Academic Search Premier (EBSCO) on May 13, 2021.
not applied as exclusionary criteria, and none of the reviews linked Search terms included primary topic (Lesson Study, collaborative
the structure of LS to the outcomes. Therefore, it is difficult to assess planning, lesson study professional development, and teacher plan-
the veracity of conclusions reported in these literature reviews. ning team), participants and setting (in-service teachers, elementary,
2
A.E. Benedict, J. Williams, M.T. Brownell et al. Teaching and Teacher Education 122 (2023) 103951

Fig. 1. Prisma Flow Chart (Moher et al., 2009) of Literature Review Procedures.

middle grades, secondary, and high school), and outcomes (knowl- evidence of changes to knowledge or practice. If there was a
edge, instruction, quality, collaboration). Terms were combined and disagreement about study eligibility, a third researcher reviewed it
searched using the Boolean operators of OR and AND. The search and made a final decision. After applying the inclusion and exclu-
was restricted to peer-reviewed journal articles, written in English, sion criteria, 409 papers were identified as irrelevant and 70 papers
and published between 2000 and 2021. Researchers also conducted were selected for a full-text review.
a hand search of Teaching and Teacher Education, Review of Educa-
tional Research, Teacher Education and Special Education, Journal of
Teacher Education, Teacher Education Quarterly, International Journal 3.3. Eligibility
of Lesson and Learning studies, and ZDM: The International Journal of
Mathematics Education. The database and hand search generated During the full-text review, two researchers carefully read and
479 articles. coded each study and made a decision based on the selection
criteria described in the screening process. Twenty-four studies
3.2. Article selection were excluded because they: (a) did not measure a change in
teacher practice or knowledge (n ¼ 18), (b) were not an empirical
Titles and abstracts were screened using the Covidence web study (n ¼ 1), (c) included topics other than literacy, math, science,
application. Two researchers reviewed titles and abstracts using the or social studies (n ¼ 3), and (d) did not describe LS cycles or
defined search criteria to determine eligibility. Inclusion criteria described incomplete LS cycles (n ¼ 2). We then applied a quality
required that studies focused on in-service teachers in K-12 settings appraisal coding system from the Mixed-Methods Appraisal Tool
providing literacy, math, science, or social studies instruction and (MMAT; Pluye et al., 2011) to assess the methodological rigor
examined changes in teachers' knowledge and/or pedagogical employed in each study; this system included quality indicators for
practices. Studies were excluded if they included pre-service or qualitative, quantitative, and mixed methods studies. Guiding
higher education teachers, conceptual papers, PD other than LS, or questions helped evaluate specific indicators related to data sour-
did not focus on changes in teachers’ knowledge or instructional ces and collection, analytic processes, and subsequent description
practice. We also excluded studies that employed only teacher in- of relevant findings. After the quality review was completed, only
terviews, as interviews provide perceptions of change but no direct 13 studies (14 articles) met the required criteria.
3
A.E. Benedict, J. Williams, M.T. Brownell et al. Teaching and Teacher Education 122 (2023) 103951

3.4. Analysis 4. Results

Researchers extracted information from articles in three phases: In the following sections, we describe the characteristics of the
(a) extraction of descriptive information (phase 1), (b) in-depth studies reviewed. This description is followed by an analysis of the
assessment of study contexts, goals and knowledge sources studies according to the knowledge sources available to teachers,
(phase 2), and (c) categorization of studies according to knowledge and outcomes documented, including changes in teachers’
sources used (phase 3). In the first phase, researchers documented: knowledge and understanding of students, content, and pedagog-
(a) research purpose/questions, (b) theoretical framework, (c) ical practice along with changes to instruction.
sample/participants, (d) content of LS, (e) grade level(s) taught, (f)
duration of LS, (g) description of LS components, (h) LS facilitator, 4.1. Characteristics of the studies
(i) intended outcomes, (j) measures and data sources, (k) analysis
approach, and (l) results (see summarization in Table 1). In the Studies reviewed varied widely in terms of research design,
second phase, researchers examined: (a) LS structure and compo- content area and LS components, frequency and duration of LS
nents, describing the unique LS features, (b) knowledge sources cycles, and data collection and analysis. Nine studies employed
available in the LS process, (c) knowledge or practice to be changed, qualitative research designs (Benedict et al., 2021; Dudley, 2013;
(d) measures used, and (e) results. Identification of knowledge Gee & Whaley, 2016; Goh & Fang, 2017; Lewis et al., 2009; Murata
sources was a critical component of the second level of analysis. et al., 2012; Pella, 2015; Rock & Wilson, 2005; Suh & Seshaiyer,
Research studies often incorporated multiple sources of knowl- 2015), and three studies employed mixed methods (Collet, 2017;
edge, so we created clear descriptions for each category and one Lewis & Perry, 2014, 2017; Moss et al., 2015). Only two studies
member of the research team organized studies accordingly (see utilized an experimental design (Lewis & Perry, 2014, 2017; Mutch-
Table 2). Studies were first categorized as outside or inside Jones et al., 2012). The Lewis and Perry papers represent one large
knowledge sources. Outside knowledge sources were those study with different analyses.
brought into the school district, such as a researcher-developed The studies reviewed were diverse in topics covered, settings,
curriculum or an outside facilitator. Inside knowledge sources and participants. The majority of papers concentrated on mathe-
were those provided within the school district, such as teacher matical content (n ¼ 9) with the other four focused on writing
teams or school-based instructional coaches. Once studies were (n ¼ 2) and reading (n ¼ 2). Settings of the studies ranged from
labeled as inside or outside knowledge sources, studies were early elementary to middle school. Across all studies, there were
further categorized according to the specific type of knowledge 363 participants, and their number for each study ranged from 3
source (e.g., PD, content area expert, district-provided curriculum) (Murata et al., 2012) to 213 (Lewis & Perry, 2014, 2017). Nine of the
available to teachers. After categorizing the studies, a second and studies had fewer than ten participants. The participants were
third member of the research team reviewed each study's category, elementary school teachers, middle school teachers and school
and disagreements were discussed until a consensus was reached. professionals (e.g., administrators, reading specialists). Only four of

Table 1
Knowledge source categories and descriptions.

Inside Knowledge Sources Description Articles

Teacher teams/peers Teachers engage in self-directed learning with minimal guidance from external Dudley (2013), Pella
sources or district/school-based leaders. (2015)
LS facilitator (coach or school-based leader) A district or school-based facilitator provides support with the LS process, inquiry,
and prompting during planning, teaching, debriefing., Teachers led on identifying
topics to focus on and utilize currently available resources.
Content area expert and/or LS facilitator (coach or school- A district or school-based content expert (ex. Literacy coach) or facilitator provides Collet (2017), Gutierez
based leader) þ PD and/or curriculum and/or content local PD or ongoing support related to current curriculum and instructional (2016), Lewis et al.
resources practices. A facilitator guides teachers through the LS process by 1) providing (2009)
guidance resources 2) facilitating LS meetings 3) attending meetings and supporting
teachers through constructive feedback, inquiry support, and questioning. Teachers
may receive content area resources that are supplemental supports, as opposed to a
new, comprehensive curriculum.
Outside Knowledge Sources Description Articles
Content expert and/or LS facilitator only Outside researcher, facilitator, or content expert guides teachers through the LS Rock and Wilson (2005)
process by 1) providing guidance resources 2) facilitating LS meetings 3) attending
meetings and supporting teachers through constructive feedback, inquiry support,
and questioning. Support is related to academic content, instructional strategies,
and/or pedagogical skills. They may provide content area resources that are
supplemental supports, as opposed to a more comprehensive curriculum.
Curriculum and/or content/LS resources only. A formal curriculum or comprehensive resource package is provided from an Lewis and Perry (2014;
outside source utilized during LS. Teachers independently study and utilize 2017)
curriculum/resources provided
Curriculum þ content expert and/or LS facilitator A formal curriculum or comprehensive resource package is provided from an Murata et al. (2012)
outside source to utilize during LS.A researcher, content expert, or facilitator
provides guidance to support teachers' use of resources.
PD þ curriculum/resources and/or content expert/LS Teachers are provided initial or ongoing formal PD related to content area or LS. The Benedict et al. (2021)
facilitator PD is designed and facilitated by an outside research or content area expert.PD is Gee and Whaley (2016)
provided to all teachers participating in the project, or intervention teachers if it is Rock and Wilson (2005)
an RCT. Examples are: Moss (2015)

Summer institute Mutch-Jones (2012)

Ongoing PD sessions Sherin, Linsenmeier,

Content-focused PD and van Es (2009)

Case students

4
A.E. Benedict, J. Williams, M.T. Brownell et al. Teaching and Teacher Education 122 (2023) 103951

Table 2
Summary of included studies.

Studies LS content focus Setting (grade level Participants Duration of Research design and Results
students) LS and measures
unique
features

Benedict et al. Word study 1 elementary school; 5 GET and 2 SET 5 LS cycles Qualitative; discussion The LS cycles allowed teachers to
(2021) content and 4th and 5th graders across 8 transcripts integrate their understandings of word
pedagogical months study content, pedagogy, struggling
practices across learners, and instructional coordination
RTI tiers within an RTI framework.
Collet (2017) Writing 1 elementary school; 3 GET and 1 SET 20 LS Mixed-method; lesson The LS meetings were effective for
instruction 5th graders meetings plans, field notes, audio improving both teachers' writing
across 5 transcripts, students' instruction and students' writing scores.
months writing scores
Dudley (2013) Mathematical 2 elementary schools; 9 5 GET 1 LS cycle, Qualitative; interviews Teachers reported raised expectations for
thinking year- old students 4h student learning and increased use of
higher-order, open-ended questions.
Gee and Whaley Mathematics 1 elementary school; 16 GET 11 day-long Qualitative; video-recorded Most teachers improved their
(2016) instruction 3rd-5th graders sessions discussions, journal instructional practices in student
during the reflections, interviews discourse, student thinking, and
summer over questioning strategies.
2 years
Studies LS content focus Setting (grade level Participants Duration of Research design and Results
students) LS and measures
unique
features
Goh and Fang A common 1 elementary school; 6 SET and 5 others 2 LS cycles Qualitative; discussion The LS team was able to build a common
(2017) inquiry stance 1st graders (e.g., facilitator) over 10 transcripts inquiry stance and move towards a
and curriculum- weeks curriculum-based deliberation.
based
deliberation in
ELA
Lewis et al. (2009) Algebra 5 elementary and 5 elementary 2 LS cycles Qualitative; videotapes of Teachers improved their knowledge of
instruction middle schools; 5th-6th teachers and 1 over 2 weeks meetings and lessons, mathematics and instructional practices.
graders middle-school written lesson plans, field
teacher notes, records of follow-up
conversations
Lewis and Perry Fractions Not reported; 2nd-5th 13 groups of 1 LS cycle, 91 Qualitative; assessment of Teachers in the treatment group
(2014) instruction graders educators days teacher fraction knowledge, outperformed the control group with
written reflections, LS fractions knowledge gains and
videos, and artifacts improvement in the implementation of
instructional practices.
Lewis and Perry Fractions 27 school districts; 39 groups of 1 LS cycle, Mixed-method; assessment Lesson study was effective for improving
(2017) instruction 2nd-5th graders educators around 90 of teacher/student fraction both teachers' fractions knowledge and
days across knowledge, written students' fractions knowledge.
three reflections, LS videos, lesson
conditions videos, and artifacts
Studies LS content focus Setting (grade level Participants Duration of Research design and Results
students) LS and measures
unique
features
Moss (2015) Geometry and 1 elementary school; K- 4 kindergarten 7 PD sessions Mixed-methods; video/ Teachers increased their content
spatial reasoning 1st graders teachers, one over 5 audio recordings, field knowledge, perceptions of students'
elementary months notes, and focus group mathematical competencies, and
teacher, and 3 interview awareness of high-quality geometry and
others (e.g., spatial reasoning.
principal)
Mutch-Jones, Science 17 middle schools; 5th- 41 science 3 LS cycles in Mixed-methods; Treatment teachers improved their
Puttnick & knowledge and 8th graders teachers and 32 combination knowledge survey abilities to provide meaningful
Minner (2012) accommodations SET with upfront accommodations and plan adaptations
for SWD PD focused for SWD. Longitudinal analysis revealed
on the no difference in science knowledge
features of LS between treatment and control teachers.
Murata et al. (2012) Multidigit 1 elementary school; 3 GET 8 LS Qualitative; LS videos, Teachers were able to develop
subtraction and 2nd-4th graders meetings lesson videos, lesson plans, mathematical pedagogy and knowledge
place value over 6 student work, and interview for teaching.
months notes
Pella (2015) Writing 1 middle school; 6th- 5 ELA teachers 9 LS cycles Qualitative; discussion Teachers were able to make and maintain
instruction 8th graders over 3 years transcripts, field notes, pedagogical shifts towards collaborative
interviews, written writing.
reflections
Studies LS content focus Setting (grade level Participants Duration of Research design and Results
students) LS and measures
unique
features
(continued on next page)

5
A.E. Benedict, J. Williams, M.T. Brownell et al. Teaching and Teacher Education 122 (2023) 103951

Table 2 (continued )

Studies LS content focus Setting (grade level Participants Duration of Research design and Results
students) LS and measures
unique
features

Rock and Wilson Inquiry model 1 elementary school; 6 GET 2 LS cycles Qualitative; interviews, field Through participation in the inquiry
(2005) 3rd-5th graders notes/observations, teacher models of PD, teachers improved their
reflection journals instructional practices in vocabulary and
math instruction.
Suh and Seshaiyer Algebraic 1 Title 1 school; 6th-8th 5 GET and 1 SET 3 LS cycles Qualitative; video clips, Teachers improved their abilities to
(2015) thinking graders over 6 student work, teacher understand the mathematical learning
months reflections, researchers' progression and analyze students'
memos. thinking.

Note. RTI ¼ response to intervention; GET ¼ general education teachers; SET ¼ special education teachers; LS ¼ lesson study; SWD ¼ students with disabilities;
PD ¼ professional development.

the studies included special education teachers (Benedict et al., documented changes in teachers' learning, including changes to
2021; Goh & Fang, 2017; Mutch-Jones et al., 2015; Suh & knowledge and motivation, using video recordings of planning and
Seshaiyer, 2015). debriefing sessions during the LS cycle followed by interviews.
The frequency and duration of the LS experience varied and Dudley also examined how the development of teacher knowledge
were not reported uniformly. Some researchers stated the number led to changes in the support provided to students and collabora-
of LS cycles (e.g., Rock & Wilson, 2005), others described the fre- tive interactions that best supported teacher learning. Data pre-
quency of meetings and their duration (e.g., Pella, 2015), and others sented in the study show that teachers learned to see the benefit of
estimated the number of days for LS (e.g., Lewis & Perry, 2014, asking more open-ended questions, but did not present evidence
2017). for how this helped students develop problem-solving approaches,
The reviewed studies used various outcome measures which or how this changed teachers' responses if students were struggling
included surveys of content knowledge, observations of LS and to articulate a problem-solving approach. Similarly, teachers in the
research lessons, interviews, and collection of artifacts. Only two LS group focused on writing instruction noticed why a student
studies (Lewis & Perry, 2017; Mutch-Jones et al., 2012) used might not be participating and thought that they might need to be
quantitative measures to assess changes in teachers’ knowledge more explicit, but they did not describe what they might do. Thus, it
and the remaining studies used qualitative measures. is difficult to determine the depth of teacher learning and the
impact their newly acquired knowledge might have on the
5. Inside knowledge students.
Pella (2015) investigated how teachers made shifts in peda-
In five studies (i.e., Collet, 2017; Dudley, 2013; Goh & Fang, 2017; gogical reasoning and actions in writing instruction while partici-
Lewis et al., 2009; Pella, 2015), knowledge sources included some pating in LS. Five middle school teachers from five Northern
combination of the LS team, school or district coaches with content California school districts participated in nine LS cycles across a
expertise, a school-based staff development leader, content- three-year period. Participating teachers self-directed the LS while
focused PD, and curriculum and instructional resources. When the researcher observed. During each LS cycle, teachers selected a
new knowledge came from within the LS team, school, and/or process writing strategy to focus on, such as integration of text in
district, we labeled it as an inside knowledge source. written responses and analytic reading of texts. Using situated
cognitive theory, Pella analyzed, using a five-phase approach,
audiotape and transcription of teacher discussions from the plan-
5.1. Teacher teams
ning, observations, debriefing phases, and extensive field notes she
kept for each phase. Evidence from her analysis showed teachers
Teachers' working with their peers and colleagues is a compo-
referenced current curricula, texts, teacher-created lessons, and
nent of the LS process. In some studies, however, peers and the
resources from previous PD activities in their planning sessions. In
curriculum they currently used were the only sources of knowledge
the observation and debriefing sessions, teachers analyzed and
in the process. In Dudley's (2013) and Pella's (2013)studies, LS ap-
discussed students’ responses to the lesson.
proaches involved teacher teams using the district curriculum as
Pella described a rigorous approach to analyzing different data
the primary and only knowledge sources teachers could access to
sources, but the evidence she provided was not always robust or
improve their learning. Dudley (2013) investigated what teachers
convincing. For instance, teachers described how they shifted from
learned from participating in an LS, and how their new knowledge
more teacher-directed instruction to more student-centered,
supported student learning though the latter was not articulated
engaging instruction, but it was unclear how they believed their
explicitly. Participating teachers were five third-grade teachers in
changes better supported student learning. For example, teachers
an urban area of England who formed two LS groups. One group
described strategies they used to engage students and support their
focused on integrating more open-ended questions into their
learning, but she did not discuss specifically how those strategies
mathematics instruction to foster and understand student thinking,
would support student learning. Additionally, researchers provided
and the other on developing self-assessment and oral rehearsal
examples of how teachers knew students were learning, but the
through role play in order to improve student engagement during
examples did not explicate how students were learning. For
writing instruction. During three LS cycles, case pupils, who rep-
example, one teacher observed that students knew about voice
resented types of learners teachers wanted to support, were used to
because they were able to indicate when authors were emotional
focus teachers' planning and debriefing discussions. Researchers

6
A.E. Benedict, J. Williams, M.T. Brownell et al. Teaching and Teacher Education 122 (2023) 103951

because they used tone, word choice, imagery, etc. What was not including asking an independent observer to recode data and
clear was the evidence students were drawing on to identify tone, establishing interrater reliability with research codes, and member
describe how the words and imagery indicated emotions, etc. checking preliminary results with teachers. Quantitative data
included students' standard scores on a state writing assessment.
5.2. Inside experts/facilitators þ curriculum and/or content-focused Throughout the LS process, quotes from LS recordings show that
PD teachers utilized instructional writing strategies introduced by the
coach. The researcher reported that teachers also transformed their
In three studies, experts from the district or school facilitated writing instruction through the increase in use of other process
the LS cycle with the intention of helping teachers deepen their writing strategies (e.g., conferencing with individual students; peer
knowledge or skill. These facilitators were either content experts or discussion, and how it was helping students; and resisting the
staff with deep knowledge of the curriculum. Additionally, teachers temptation to over-scaffold) not addressed specifically by the
participated in content-focused PD and were provided curriculum coach. Although the researcher made appropriate use of quotes to
and instructional resources. Lewis et al. (2009) examined how five support themes, the amount of evidence presented was often
elementary teachers and one middle school teacher's participation limited to one or two brief examples. The limited qualitative evi-
in LS played a role in the development of their content and peda- dence, however, was supported by quantitative evidence. Analysis
gogical knowledge in mathematics. One member of the team was of students' standardized writing scores found statistically signifi-
an instructional coach who also facilitated the LS and previously cant positive results in a year-to-year comparison and in compar-
participated in 15e20 days of professional learning that focused on ison to other students’ yearly gains in the district.
coaching and solving and discussing mathematical problems with Goh and Fang (2017) examined the process of LS experienced by
expert math educators. Teachers participated in a two-week sum- one team of elementary teachers and staff (n ¼ 11) in Singapore.
mer workshop, designed by the LS facilitator and coach, that helped The facilitators included a school-based staff development leader
them collaboratively: (a) investigate math content standards, cur- and two senior teachers, labeled research teachers, who had taught
riculum, and problems students might encounter while solving the planned lessons previously. They facilitated the team of
mathematics problems, (b) plan the lesson based on this investi- teachers in two LS cycles with the aim of exploring how the LS
gation, (c) teach the lesson to a 4th-grade class, (d) debrief, and (e) structure furthered teacher learning. Teachers were implementing
revise and reteach the lesson to a second 4th-grade class. The team a nationally developed curriculum with fidelity, however, the LS
concluded the workshop by debriefing and writing a written was implemented to help them recognize how students were not
reflection about the LS cycle and what they learned from the first always engaged and supported in their learning, and in turn, devise
lesson implementation compared to the second. Researchers video- new strategies and resources to enhance student learning. Teachers
recorded teachers' investigation, planning, instruction, and employed a curriculum developed nationally to support the
debriefing sessions and interviewed teachers. The researchers used development of English Language Learners’ writing and reading
a rigorous analytical approach. First, one researcher selected video skills.
clips she felt captured instances of teachers' knowledge about Data collected included observation of the two LS cycles, pre-
content or understanding of classroom practices; the team met to and post-teacher interviews, student artifacts, and LS plans. The
confirm or disconfirm her choice. A written case was developed of authors provide descriptions of how they open-coded data from
the researchers' collective observations to be reviewed by the the interviews but did not provide many details about the codes,
participants of the LS. how they collapsed codes to form categories, how they used cate-
Qualitative analysis of discourse during LS planning and gories to develop themes, or how they used other data to trian-
debriefing meetings revealed changes in teachers' knowledge of gulate findings. Researchers also provided one example from
mathematics and student thinking (e.g., mathematical patterns and teachers’ discourse during planning that demonstrated how they
reasoning behind these patterns), as well as pedagogical practices. would attempt to assist comprehension of a story by connecting to
In addition, teachers identified the reasoning behind specific background knowledge and then in debriefing sessions provided
mathematical patterns. Researchers provided excerpts from information about how they intended to revise the lesson to better
teachers' discourse that exemplified these changes. For example, address how students were processing the content. Although the
one teacher recognized how a student's thinking was revealed authors demonstrated through this example how the LS process
through his/her counting strategy, and how the teacher decided to helped them identify challenges students faced and identify stra-
eliminate a specific strategy because it did not support the stu- tegies to better support their learning, more evidence, and trian-
dent's conceptual understanding. gulation of evidence would have strengthened their conclusions.
Collet (2017) conducted a mixed-methods study to examine
how participation in LS influenced teachers' writing instruction and 5.3. Outside knowledge
students' performance on writing assignments. A team of three
5th-grade general education teachers and one special education The majority of studies included some type of outside knowl-
teacher participated in LS following a districtwide workshop on LS. edge source, external to the school or school district of the partic-
Twenty LS meetings took place over five months, and teachers ipants. Outside knowledge sources included: (a) content experts or
planned, taught, and reflected on seven lessons. The district literacy facilitators who were often researchers conducting the study, (b)
coach, also the researcher, facilitated the LS by working alongside content-focused PD institutes, and (c) curriculum and resources
the teachers, providing guidance and support through inquiry by related to content area instruction.
asking questions during the planning and debriefing sessions.
Additionally, the coach introduced teachers to specific instructional 5.4. Content expert and/or LS facilitator
practices to utilize during writing instruction (e.g., rubrics, exem-
plar papers, anchor charts, and mentor texts). Qualitative data Rock and Wilson (2005) studied LS facilitated by an outside
collected included a recording of LS meetings, field notes, artifacts, content area expert (i.e., university faculty member) and curricu-
and observations of lessons before, during, and after participation lum coordinator for the district, and its impact on how teachers
in LS. Collet described a detailed process for analyzing these data, perceived LS as a PD process and how engaging in LS affected

7
A.E. Benedict, J. Williams, M.T. Brownell et al. Teaching and Teacher Education 122 (2023) 103951

teachers’ instruction. Facilitators guided teachers through the LS opportunity that involved prescribed instructional materials. Hi-
process based on problems of practice teachers selected. Teachers erarchical linear model (HLM) analyses showed that LS treatment
were introduced to the LS process through a brief presentation contributed a significant portion of variance (0.19) to teachers'
made by the researchers. Next, teachers grouped themselves into knowledge, and had a significant effect on student achievement
two LS teams based on instructional goals in mathematics and lit- (ES ¼ 0.49). Lewis and Perry coded videos of the LS process using 15
eracy. A university faculty member supported teachers with rele- codes that captured how much time teachers spent on different
vant literature and guest lecturers related to general pedagogy (e.g., content in the fractions resource kit. Two additional codes were
differentiation). The school-based facilitator served as a also used to capture discussion of student thinking and work and
“commentator” during LS observation and meetings, asking ques- learning opportunities or a moment in the LS cycle that identified
tions and pressing teachers to think about their experiences or when something happened to influence teachers thinking about
observations in greater detail. Only one LS cycle was implemented. content or students. Analysis of codes used in the LS cycle showed
The researchers used qualitative analyses, specifically textual that teachers used the resource kit in the ways it was intended and
analysis, of interviews, field notes taken during the LS process, and focused on student thinking when expected. Additionally, analysis
teachers' reflective journals. Although Rock and Wilson referenced of teachers' lesson plans and research lessons showed that teachers
a systematic process for coding data, they did not provide specifics who modeled their lessons based on videos provided in the
of how they applied this process to their data, particularly how they resource kit had significantly higher student achievement scores
progressed from coding data to identifying themes. In reporting compared to teachers who did not use the video. Through their
results, the researchers focused on teachers’ perceptions of quantitative and qualitative analysis, Lewis and Perry provided
participating in the LS process, mostly its benefits. The researchers strong evidence showing how knowledge accessed in the LS study
provided one detailed example of how teachers differentiated math process, particularly that provided in the resource kit, influenced
instruction and how students in the lower achieving group pro- teachers’ knowledge and student achievement in fractions.
gressed. The purpose of this study and the implementation of only
one LS cycle reduced the likelihood that Rock and Wilson could
collect strong evidence of change in teacher practice. 5.6. Curriculum þ content expert and/or LS facilitator

5.5. Curriculum and/or Content/LS resources Murata et al. (2012) examined an LS process that engaged an
outside content expert to facilitate LS and provide teachers math-
Lewis and Perry (2014; 2017) used an experimental design to ematics resources, such as curriculum materials and research
examine how LS that incorporated a researcher-developed resource literature, to support instruction (Murata et al., 2012). The facili-
kit impacted facets of teachers' knowledge for teaching fractions tator, one of the researchers, provided content support to teachers
and student achievement. They also examined, using qualitative (e.g., curricular resources, strategies, lesson planning), and logis-
methods, the processes that played a role in developing teachers’ tical support for the LS process (e.g., developing agendas, summa-
knowledge. The study was implemented with 39 school-based rizing previous meetings, guiding discussions during LS meetings).
teams in elementary and middle schools. Of the 213 LS partici- The researchers’ purpose was to determine how teachers partici-
pants, teachers comprised 87 percent of the team members, and pating in an LS process made sense of content, teaching, and stu-
instructional coaches and administrators comprised 13 percent. dent learning as it related to multi-digit subtraction, and how their
The LS treatment plus fractions resource kit was compared to two understandings in these areas changed over time.
comparison groups that received PD on self-selected topics. The Three elementary teachers, representing 2nd, 3rd, and 5th grade
groups were allowed to use an LS format or a locally selected PD participated in eight LS cycles. Unlike typical LS, teachers did not
format. teach the same lesson but planned lessons on the same topic across
Researchers collected the following forms of data: (a) pre and grade levels, but otherwise followed the traditional Japanese LS
post assessment of teachers' fraction knowledge using an instru- cycle. Data collected included video recordings of LS meetings,
ment with previously validated items, (b) pre and post assessment lesson-related materials that teachers created or that they adopted
of students' achievement on fractions, (c) written reflections at the from the curriculum, videos of lessons taught, student work arti-
conclusion of the LS cycle that focused on what teachers learned, facts, and teacher interviews conducted at the conclusion of the
(d) time spent using the fractions resource kit, (e) videos of the LS study. The researchers used a rigorous three-stage process to code
cycle including, investigating, planning, teaching the research teacher talk about content, teaching, and learning related to rep-
lesson, and debriefing, and (f) artifacts from LS meetings. Data resenting multi-digit subtraction. Through the coding process, re-
collected from the study were reported in two different publica- searchers developed four levels of teacher talk that represented
tions, answering somewhat different research questions. In the how teachers were developing understandings of content, teach-
2014 paper, paired t-tests identified significant changes in certain ing, and learning; these levels of talk were triangulated with evi-
facets of teachers' knowledge. Excerpts from the written reflections dence from video-recorded lessons planned during LS. Although
and videos were selected by the researchers to show how sources of teachers' talk progressed in ways that were different for each
knowledge presented in the resource kit and derived from the LS teacher, analysis of their talk at the group level showed evidence of
process played a role in teachers' evolving understandings of teachers' abilities to make connections between content, learning,
fractions and how to teach them. Further, excerpts from teachers’ and teaching and provide support for those connections from their
writing reflections confirm quantitative changes teachers in the instruction or student thinking. Further, evidence from the video-
treatment group made in their knowledge. Lewis and Perry (2014) taped lessons were used to demonstrate how changes in one
do recognize that their analysis of qualitative data was not con- teacher's understanding of representing multi-digit subtraction
ducted in a way that would allow them to draw conclusions about influenced her instruction. By tracing teachers' talk over time in the
how teachers were using knowledge sources across the study. LS process and showing how teachers came to understand and
Data reported within the 2017 study established the efficacy of integrate their knowledge of content, teaching, and students,
LS and showed how aspects of the resource kit contributed to ideas Murata and colleagues demonstrated how teachers were changing
teachers generated in the LS process and teachers' instruction and in response to the LS process and the curricular resources they
captured teachers' perceptions of a professional learning employed during that process.
8
A.E. Benedict, J. Williams, M.T. Brownell et al. Teaching and Teacher Education 122 (2023) 103951

5.7. PD þ curriculum/resources and/or content Expert/LS facilitator resources. Project staff provided coaching during planning and
debriefing about the observed lesson.
In five studies, the LS process was facilitated by a content area Teachers were assessed on a researcher-generated knowledge
expert from outside of the school or district and combined with test designed to assess their knowledge of science content and
additional content-focused PD. Teachers were provided with initial process skills, ability to identify student learning challenges, ability
or ongoing PD related to the content area along with some com- to plan and adapt a lesson, and ability to develop accommodations
bination of other resources, such as a teacher handbook, coaching, for students with learning disabilities. Lessons used in the assess-
and lesson planning resources. ment were taken from a sixth-grade published curriculum. The
Gee and Whaley (2016) investigated how elementary teachers assessment was administered at the beginning, middle, and end of
participating in LS perceived their professional growth and how teachers’ participation in the research. A series of three-level HLM
that growth related to improved teaching of mathematics. Sixteen analyses were used to examine differences in knowledge scores
3rd, 4th, and 5th-grade teachers participated in a two-year PD that across the assessments for individual teachers (level 1), between
included summer institutes focused on increasing knowledge and special and general education teachers on a team (level 2), and
understanding of number sense and problem-based learning, as between teams in the treatment and comparison group (level 3).
well as introducing teachers to LS. In the summer institute, teachers Results showed that teachers in the treatment were not signif-
created a unit of mathematics lessons that they taught using four icantly different than the comparison group for the level of scien-
complete LS cycles each fall of the two-year project. These cycles tific knowledge or student challenges identified though there was a
involved teachers meeting to select a research lesson and discuss positive trend for the treatment group. Treatment teachers did
student needs that might arise in that lesson, teach and observe the increase in their ability to plan adaptations at the midpoint but
lesson, debrief and revise, and reteach the lesson. A university declined in their ability at the end. There was a significant effect for
professor facilitated the LS. treatment on the teacher teams' ability to plan adaptations for the
To achieve their purpose, researchers collected data from the whole class, accounting for 66 percent of the variance; general and
following sources: teacher discussion of video clips of their obser- special education teachers were not significantly different on this
vations that surfaced aspects of instruction they hoped to improve, outcome. Teachers' mean number of accommodations increased at
teachers written reflections on their self-selected video clips, and the midpoint and declined slightly at the end. Within teams, special
recorded semi-structured interviews conducted at the end of each education teachers generated more accommodations than their
year. Data were coded using criteria developed by Sherrin and science education colleagues regardless of their assignment to
colleagues (2009), but it was unclear how researchers moved treatment or comparison. All teachers in the intervention group
from codes to themes. The researchers identified two themes: (5.48 ¼ special education teachers, 4.62 ¼ science teachers)
collaboration and change in practice. Teachers claimed that generated more accommodations than their peers in the compar-
collaboration helped them notice things that they did not always ison group (4.54 ¼ special education teachers, 3.68 ¼ science
see, but mostly in terms of how it affirmed their practice. The re- teachers). These results demonstrate that the LS intervention had a
searchers did not provide evidence that showed how collaboration positive effect on aspects of teacher knowledge; exactly how spe-
improved teachers' thinking about student learning or mathe- cific aspects of the comprehensive LS treatment influenced teach-
matics instruction. Teachers also described how the LS process ers’ knowledge acquisition was not addressed in this study.
helped them see a different way of teaching, make better connec- Moss et al. (2015) conducted a case study of six teachers
tions among mathematical ideas for themselves and their students, participating in a comprehensive innovation that combined
and see different ways of solving problems as a result of student extensive content-focused PD with an elaborated LS process. The
discourse. Quantitative evidence supporting teachers’ claims from project was entitled Math for Young Children (M4YC), which aimed
videotaped lessons were provided; 66 percent of teachers made to develop teachers' knowledge of geometry and broaden aware-
statements that reflected medium to high levels of focusing on ness of students’ mathematical abilities and interests. Moss and
student understanding though these findings were not triangu- colleagues expanded the traditional LS process to engage teachers
lated with specific examples from the qualitative data. Further, in: (1) mathematics challenges related to geometry and spatial
thirty-one percent of teachers seemed to make little progress. They reasoning, (2) the design and implementation of clinical student
made statements about video clips that demonstrated a low level of interviews in order to ascertain students thinking, (3) planning,
focus on student understanding. For example, these teachers implementing, debriefing and revising lessons, and (4) the creation
described needing to help students with “tricks” and memorizing of resources for the public research lesson that is presented at the
steps in order to understand. From the results provided, it was conclusion. The lessons were first planned and implemented with
difficult to determine whether content learned in the PD or three case study students; clinical interviews were conducted after
whether specific ideas acquired through the LS process influenced the research lessons, and teachers discussed information from
the instructional practices teachers utilized. these interviews along with observations of the lesson in debriefing
Mutch-Jones et al. (2012) examined how middle school science meetings. Researcher lessons were revised and re-taught in class-
and special education teachers’ participation in LS impacted their rooms before a public lesson was presented and debriefed. The
knowledge of science content and ability to identify student M4YC project occurred over five months with teachers and re-
learning challenges. Sixteen LS teams, with approximately 5 team searchers working together for five full days and two half-day
members (including at least one special educator), were randomly sessions.
assigned to the LS intervention group (n ¼ 47) or comparison group The purpose of the study was stated in broad terms e to
(n ¼ 46) that received the intervention once the study concluded. determine how adaptation supported teacher learning. Qualitative
Teachers attended a summer PD institute focused on the features of data collected throughout the PD included video and audio re-
LS, research on learning disabilities, investigating content from a cordings of all meetings and lessons, field notes, and focus group
science unit, and providing accommodations to students with interviews before and after the PD. Data were analyzed using a
disabilities. They also engaged in an LS cycle with fourth-grade deductive coding method that employed the four LS adaptations as
students attending summer school. Following the PD, teams of codes. The fourth code, creation of resources, was further analyzed
teachers participated in three LS cycles. Researchers provided for evidence of: changed perceptions of geometric and spatial
teachers with a lesson planning framework and other planning reasoning, changed perceptions of children's thinking and
9
A.E. Benedict, J. Williams, M.T. Brownell et al. Teaching and Teacher Education 122 (2023) 103951

mathematical development, and changed pedagogical approaches. were using to represent problems, even when students made
Evidence was then provided for each code. At the beginning of the mistakes.
training, teachers engaged in geometry and spatial thinking activ- Benedict et al. (2021) conducted a one-year study of Project
ities they would be teaching. Teachers' reflections on this part of the InSync, an innovation that involved content-focused PD combined
PD indicated that they became more aware of how problematic LS. Researchers wanted to understand the role the project played in
their original beliefs about geometry instruction were and of the the development of teachers' collective understandings of word
need to shift to instruction that helped children visualize, mentally study instruction, and how those understandings changed over
rotate, and transpose geometric shapes. These changes prompted time. Participants included teams of fourth and fifth-grade general
teachers, according to one of the principals, to engage with math- and special education teachers (n ¼ 7). The PD and LS meetings
ematics tasks prior to instruction. Through conducting clinical in- were facilitated by a researcher, the first author. Over the course of
terviews and observing students during math tasks, researchers the year, teachers attended five workshops that focused on word
described that teachers increased their knowledge of student per- study involving basic decoding, multisyllabic decoding, and
formance and thinking, asked more probing questions, and used morphology and coordination of tiered instruction within a multi-
their new knowledge to plan instruction. Specific evidence of what tiered support framework. They also participated in an LS training,
teachers did, however, was not provided. Similarly, the researchers three planning cycles where they learned to plan using a specific
did not use evidence from interviews or observations to support framework, and coordinated instruction that incorporated content
their claim that teachers observed students' thinking more instead and strategies from the workshop. Coordinated instruction
of just interviewing students and that doing so resulted in peda- involved developing an overall goal for instruction and then indi-
gogical changes. They did describe one example where the vidual lessons that complimented that goal in terms of content and
exploratory lessons allowed teachers to experiment with a more evidence practices for students receiving tier 1 (whole class in-
teacher-directed versus child-centered approach, and observe struction), tier 2 (small group supplemental instruction with stu-
which approach best supported student learning. Finally, re- dents needing extra support in the general education classroom),
searchers described how teachers generated novel geometry ac- and tier 3 instruction (intensive instruction with students identi-
tivities, but did not provide examples as evidence of the ways in fied as having reading disabilities). After learning to plan coordi-
which teachers' knowledge was changing. The lack of evidence nated instruction, teachers participated in two full LS cycles
supporting study themes draws into question the veracity of the designed to help them implement content from workshops.
authors' claims. Teachers’ debriefings of lesson observations focused on how stu-
Suh and Seshaiyer (2015) examined how content-focused PD dents in tier 2 and 3 benefited from instruction in strategies, and
combined with LS helped six teachers (i.e., five general education how strategies and content might be adjusted to meet their needs.
teachers in grades 3, 6, and 8 and one special education teacher Using a constructivist grounded theory approach, researchers
supporting 5th grade): (a) have meaningful opportunities to analyzed video recordings of teachers’ conversations during LS
articulate learning progression in algebraic thinking across the planning and observation/debriefing sessions. Data were analyzed
grades, and (b) deepen their knowledge for teaching students to using a rigorous and iterative process including the development of
develop algebraic thinking. The assessed innovation involved a initial, focused, and theoretical codes. The authors provided clear
content-focused summer institute followed by teachers’ imple- examples of how they moved through the coding process. The four
menting the LS cycle three times throughout the school year, theoretical codes included: Word Study Content, Pedagogy, Strug-
concluding with a public presentation of research lessons. During gling Learners, and Coordination. Theoretical codes were grouped
the study, teachers engaged in algebraic mathematics tasks, according to when they occurred during the PD and LS cycles and
explored different pedagogical strategies, and used mathematics used to develop the grounded theory.
tools and technology. The LS cycle included: (a) investigating Analysis of teacher talk for all LS sessions showed that teachers
standards, objectives, and curriculum, (b) planning the research evolved in their understanding of PD content and students over
lesson, (c) observing the lesson, and (d) debriefing and revising. time. In the beginning, teachers seemed to focus their conversa-
Researchers completed a qualitative analysis of sources, tions on understanding the content they were learning, relying on
including videos of LS meetings and research lessons, teachers' pedagogical strategies they already knew, and viewing students in
reflection journals, researchers' memos, and artifacts (e.g., lesson terms of their deficits. By the last LS cycle, teachers demonstrated a
plans, teachers' notes, student work samples). The researchers used more integrated understanding of PD content. They were able to
a deductive method to analyze data; they used Confrey's (2012, pp. discuss how they would use evidence-based literacy strategies to
2e20) five elements to categorize data from teachers' dialogue teach content to different students. Instead of viewing students in
related to unpacking of the learning trajectory during LS. These terms of their learning struggles, they began to describe the
included: “(a) underlying concepts and ideas; (b) strategies, rep- intentional ways they could adjust content and strategies across
resentations and misconceptions; (c) models and definitions; (d) instructional tiers to coordinate instruction and better support
complexity, pattern, and development of ideas; and (e) bridging learning. The description of teachers' collective understandings
gaps between standards” (p. 251). Less clear was how these re- over time was well supported by evidence from teachers' discourse
searchers used the categorization of data to develop themes. The during planning and debriefing sessions, showing that teachers
authors did, however, provide evidence from meetings and lessons were able to deploy in integrated fashion information they were
that supported their understanding of a learning progression, learning from the PD workshops to address students’ learning
including how they deliberated about supporting student thinking needs at each tier.
at different grade levels. Further, they used evidence from lesson
plans and observations of lessons to show teachers' use of ques-
6. Discussion
tioning, manipulatives, and role-playing to help students identify
patterns to help students generate multiple ways to represent a
Understanding how professional development influences
problem. Researchers also provided evidence from lesson debrief-
teachers' knowledge and practice is critical to building a research
ings that showed teachers' ability to identify the strategies students
base that informs efforts to improve teaching. Our literature review

10
A.E. Benedict, J. Williams, M.T. Brownell et al. Teaching and Teacher Education 122 (2023) 103951

was designed to improve understanding of the role LS plays in 8. Future research

developing teachers' knowledge and understanding of content,
student learning, and instructional practice. Through our analysis, Through the appraisal of studies that employed high quality
we categorized studies in a novel way that allowed us to examine research design and analysis methods we sought to identify ex-
the similarities and differences between studies that drew upon emplars for how researchers might proceed to further LS research.
comparable sources of knowledge. The primary categories of inside In addition to meeting the initial methodological quality criteria
and outside knowledge sources and identification of types of (Pluye et al.,2015), studies in this review were further examined for
knowledge sources (e.g., content expert, curriculum, PD) provided adherence to additional elements that characterize a rigorous
insight into the various factors that influence teachers’ learning of qualitative study (Leko et al., 2021). Specifically, three studies stood
content knowledge and improvement in pedagogical skills. All but out to us, one mixed method study (Lewis et al. 2014; 2017) and
three studies (Dudley, 2013; Lewis & Perry, 2014, 2017; Pella, 2015) two qualitative studies (Benedict et al., 2021; Murata et al., 2012).
include a content expert or LS facilitator to support teacher teams. Lewis and her colleagues (2014; 2017) provide an example of
Content experts and LS facilitators promoted the creation of indi- using quantitative and qualitative data to investigate how partici-
vidual and collective knowledge through content-specific PD (e.g. pation in LS contributes to building teachers' knowledge, devel-
Mutch-Jones et al., 2012), providing guidance to maintain structure oping improved pedagogical practices, and their influence on
and progress and asking probing and metacognitive questions student learning outcomes. In Lewis et al. (2014), the theoretical
(Benedict et al., 2021; Murata et al., 2012; Rock & Wilson, 2005). model of impact on instruction was clearly described as teachers'
In all studies reviewed researchers reported that teachers participation in LS would lead to intervening changes in teachers'
changed in one or more of the following areas: (a) knowledge for knowledge and practice subsequently impacting student learning.
and understanding student thinking, content, and instructional The quantitative analysis of teachers' fraction knowledge revealed a
practice, and (b) instructional practices employed. However, the significant improvement, and qualitative measures and analyses
evidence provided in most studies included was uneven, as some corroborated this finding. This evidence helps develop a broader
studies provided more convincing evidence than others. In some picture of how teachers' experiences influence knowledge devel-
cases, researchers did not provide sufficient description of how opment. In addition to the measures and analysis used in Lewis
they approached the analysis of their data, and thus it was difficult et al. (2014), Lewis et al. (2017) added a measure to quantify
to assess the veracity of their findings. In other instances, re- changes in students’ fraction knowledge which further adds to the
searchers did not provide sufficiently rich descriptions from robustness of the research and increases its credibility.
teachers’ conversations during the LS process or observations of Benedict et al. (2021) was identified as an exemplar of qualita-
instruction to justify the claims they were making about teacher tive research (Leko et al., 2021). In this study, researchers provided
change (Suh & Seshaiyer, 2015). Moreover, these claims were often a detailed description of the theoretical and epistemological
global and not justified by specific examples that teachers dis- framework guiding the study and an explicit description of the
cussed during the LS process or implemented during observations researchers' positionality. The study's design included features
of instruction. allowing for a comprehensive and nuanced understanding of
Studies in which evidence indicates LS highly impacted changes teachers' development of collective knowledge. Finally, researchers
in knowledge and instructional practice integrated access to some provided a thick description of the qualitative data, including a
source of outside knowledge, often curriculum or additional comprehensive description of teachers' conversations during each
content-focused PD, and in many cases the LS facilitator was an phase of the LS cycles with multiple quotes from teachers.
external expert (Benedict et al., 2021; Rock & Wilson, 2005). In The second example of a high-quality qualitative study in this
these studies, researchers provided sufficient description of their review is Murata et al. (2012). First, researchers described the
knowledge sources and either qualitative and/or quantitative data conceptual framework of how teachers' understanding of in-
that showed how teachers drew on those knowledge sources dur- teractions between teachers/teaching, students/learning, and con-
ing discussions of teaching and observations of instruction tent/material contribute to students' learning outcomes. Data from
(Benedict et al., 2021; Murata et al., 2012), or they made clear multiple sources allowed for triangulation of findings during
connections between the content of the LS, assessments of teacher analysis, and researchers provided a detailed description of their
knowledge, and assessments of student learning (Lewis & Perry, analysis methods. Specifically, lesson videos and student work ex-
2017). These studies suggest that LS combined with a substantial amples were used to corroborate the conclusions drawn from
curriculum and/or PD positively influences what teachers learn teacher discourse during LS and interviews. Careful description of
though more research is needed to draw definitive conclusions the qualitative methods used to analyze teachers’ discourse over
about how LS is configured, and how configurations influence the time provided a solid foundation for the conclusions drawn around
development of teacher knowledge, understanding, and instruc- teacher change. Murata et al. (2012) succeeded in showing how
tional practice. teachers could integrate new knowledge as they made connections
between varied instructional components, student thinking, and
mathematics content.
7. Limitations Careful analysis of these exemplar studies is critical to designing
future LS PD and LS research to maximize the influence of LS in
In this systematic review, the utilization of comprehensive se- improving teachers’ knowledge and practice, as well as conducting
lection criteria and a quality appraisal instrument to identify high- rigorous research that can provide evidence of the influence of LS
quality research resulted in the inclusion of only a fraction of the on changes. For example, Benedict et al. (2021) and Murata et al.
papers screened and reviewed. Ensuring studies were aligned with (2012) show how analysis of teacher discourse over time is an
our research questions and met methodological quality criteria effective means of examining change in LS. These analyses lead to a
prevented us from capturing the results of the LS research field as a more nuanced understanding of how teachers deepen their
whole. The stringent criteria used in our review prevented a knowledge due to participation in LS. Lewis and Perry (2014; 2017)
comprehensive examination of the existing body of LS literature as provide a model for designing a rigorous study that provides strong
well as analysis of how LS research has evolved and its continued evidence that can be used to reproduce, scale up, and evaluate an LS
progress. PD experience.
11
A.E. Benedict, J. Williams, M.T. Brownell et al. Teaching and Teacher Education 122 (2023) 103951

These exemplary studies, along with the other studies reviewed, Education and Development, 18(1), 87e99. https://doi.org/10.3316/aeipt.214245
Gersten, R., Dimino, J., Jayanthi, M., Kim, J. S., & Edwards, L. (2010). Teacher study
show the promise of LS when paired with additional knowledge
group: Impact of the professional development model on reading instruction
sources to positively influence teacher learning, practice, and stu- and student outcomes in first grade classrooms. American Educational Research
dent outcomes. Moving forward, researchers need to conduct more Journal, 47(3), 694e739. https://doi.org/10.3102/0002831209361208
rigorous studies to further identify the circumstances and compo- Gess-Newsome, J., Taylor, J. A., Carlson, J., Gardner, A. L., Wilson, C. D., &
Stuhlsatz, M. A. (2019). Teacher pedagogical content knowledge, practice, and
nents that are most impactful to inform the design of high-quality student achievement. International Journal of Science Education, 41(7), 944e963.
LS. https://doi.org/10.1080/09500693.2016.1265158
Goh, R., & Fang, Y. (2017). Improving English language teaching through lesson
study: Case study of teacher learning in a Singapore primary school grade level
9. Conclusion team. International Journal for Lesson and Learning Studies, 6(2), 135e150.
https://doi.org/10.1108/IJLLS-11-2015-0037
This systematic literature review contributes to the growing Gutierez, S. (2016). Building a classroom-based professional learning community
through lesson study: Insights from elementary school science teachers. Pro-
body of empirical research designed to understand if and how LS fessional Development in Education, 42(5), 801e817. https://doi.org/10.1080/
impacts teachers' knowledge and instructional practices and, ulti- 19415257.2015.1119709
mately, improves students’ academic outcomes. By identifying how Hiebert, J., Morris, A. K., Berk, D., & Jansen, A. (2007). Preparing teachers to learn
from teaching. Journal of Teacher Education, 58(1), 47e61. https://doi.org/
knowledge is accessible to teachers in LS and examining ap- 10.1177/0022487106295726
proaches that deepen teachers' knowledge and improve instruc- Ko, P. Y. (2019). Beyond labels: What are the salient features of lesson study and
tional practices, we hope for teacher educators to design more learning study? Educational Action Research, 27(4), 543e563. https://doi.org/
10.1080/09650792.2018.1530126
impactful professional learning experiences, the kind of PD that
Kraft, M. A., & Papay, J. P. (2014). Can professional environments in schools promote
impacts student achievement. Our findings indicate that LS is a teacher development? Explaining heterogeneity in returns to teaching experi-
promising professional learning experience for teachers, especially ence. Educational Evaluation and Policy Analysis, 36(4), 476e500. https://doi.org/
10.3102/0162373713519496
when coupled with additional knowledge catalysts (e.g., PD, con-
Leko, M. M., Cook, B. G., & Cook, L. (2021). Qualitative methods in special education
tent area expert, curriculum, or instructional resources). Taken research. Learning Disabilities Research & Practice, 36(4), 278e286. https://
together, this research demonstrates there is great potential for LS doi.org/10.1111/ldrp.12268
to be leveraged as a mechanism for teachers to take newly acquired Lewis, C., & Perry, R. (2014). Lesson study with mathematical resources : A sus-
tainable model for locally-led teacher professional learning. Mathematics
knowledge and situate it within their daily instructional Teacher Education and Development, 16(1), 22e42. https://doi.org/10.3316/
repertoires. aeipt.205652
Lewis, C., & Perry, R. (2017). Lesson study to scale up research-based knowledge: A
randomized, controlled trial of fractions learning. Journal for Research in
Data availability Mathematics Education, 48(3), 261e299. https://doi.org/10.5951/
jresematheduc.48.3.0261
No data was used for the research described in the article. Lewis, C. C., Perry, R. R., & Hurd, J. (2009). Improving mathematics instruction
through LS: A theoretical model and north American case. Journal of Mathe-
matics Teacher Education, 12(1), 285e304. https://doi.org/10.1007/s10857-009-
References 9102-7
Lewis, C. C., Perry, R., Hurd, J., & O'Connell, M. P. (2006). Lesson study comes of age
Achinstein, B., & Barrett, A. (2004). (Re)framing classroom contexts: How new in North America. Phi Delta Kappan, 88(4), 273e281. https://doi.org/10.1177/
teachers and mentors view diverse learners and challenges of practice. Teachers 003172170608800406
College Record, 106(4), 716e746. https://eric.ed.gov/?id¼EJ687642. Moher, D., Liberati, A., Tetzlaff, J., Altman, D. G., & PRISMA Group*.. (2009). Preferred
Akiba, M. (2016). Lesson study in Florida: A longitudinal survey of district policy and reporting items for systematic reviews and meta-analyses: The PRISMA state-
practice from 2013 to 2015. https://fsu.digital.flvc.org/islandora/object/fsu: ment. Annals of Internal Medicine, 151(4), 264e269. https://doi.org/10.7326/
277459/datastream/PDF/view. 0003-4819-151-4-200908180-00135
Benedict, A. E., Brownell, M., Bettini, E., & Sohn, H. (2021). Learning together: Moss, J., Hawes, Z., Naqvi, S., & Caswell, B. (2015). Adapting Japanese lesson study to
Teachers' evolving understanding of coordinated word study instruction within enhance the teaching and learning of geometry and spatial reasoning in early
an RTI framework. Teacher Education and Special Education, 44(2), 134e159. years classrooms: A case study. ZDM, 47(3), 377e390. https://doi.org/10.1007/
https://doi.org/10.1177/0888406420930686 s11858-015-0679-2
Benedict, A., Parks, Y., Brownell, T. M., Lauterbach, A. A., & Kiely, M. T. (2013). Using Murata, A., Bofferding, L., Pothen, B. E., Taylor, M. W., & Wischnia, S. (2012). Making
lesson study to align elementary literacy instruction within the RTI framework. connections among student learning, content, and teaching: Teacher talk paths
Teaching Exceptional Children, 45(5), 22e31. in elementary mathematics lesson study. Journal for Research in Mathematics
Cheung, W. M., & Wong, W. Y. (2014). Does lesson study work?: A systematic review Education, 43(5), 616e650. https://doi.org/10.5951/jresematheduc.43.5.0616
on the effects of lesson study and learning study on teachers and students. Mutch-Jones, K., Puttick, G., & Minner, D. (2012). Lesson study for accessible sci-
International Journal for Lesson and Learning Studies, 3(2), 137e149. https:// ence: Building expertise to improve practice in inclusive science classrooms.
doi.org/10.1108/IJLLS-05-2013-0024 Journal of Research in Science Teaching, 49(8), 1012e1034. https://doi.org/
Chizhik, E. W., Chizhik, A. W., Close, C., & Gallego, M. (2017). SMILE (Shared Men- 10.1002/tea.21034
toring in Instructional Learning Environments) effectiveness of a Lesson-Study Norwich, B., Benham-Clarke, S., & Goei, S. L. (2021). Review of research literature
approach to student-teaching supervision on a teacher-education performance about the use of lesson study and lesson study-related practices relevant to the
assessment. Teacher Education Quarterly, 44(2), 27e47. https://www.jstor.org/ field of special needs and inclusive education. European Journal of Special Needs
stable/90010517. Education, 36(3), 309e328. https://doi.org/10.1080/08856257.2020.1755929
Collet, V. S. (2017). Lesson study in a turnaround school: Local knowledge as a Norwich, B., Fujita, T., Adlam, A., Milton, F., & Edwards-Jones, A. (2018). Lesson
pressure-balanced valve for improved instruction. Teachers College Record, study: an inter-professional collaboration approach for educational psycholo-
119(6), 1e58. https://doi.org/10.1177/016146811711900605 gists to improve teaching and learning. Educational Psychology in Practice, 34(4),
Confrey, J. (2012). Articulating a learning sciences foundation for learning trajectories 370e385. https://doi.org/10.1080/02667363.2018.1468733
in the CCSS-M. North American Chapter of the International Group for the Norwich, B., & Ylonen, A. (2013). Design based research to develop the teaching of
Psychology of Mathematics Education. https://eric.ed.gov/?id¼ED584933. pupils with moderate learning difficulties (MLD): Evaluating lesson study in
Darling-Hammond, L. (2017). Teacher education around the world: What can we terms of pupil, teacher and school outcomes. Teaching and Teacher Education,
learn from international practice? European Journal of Teacher Education, 40(3), 34(1), 162e173. https://doi.org/10.1016/j.tate.2013.04.012
291e309. https://doi.org/10.1080/02619768.2017.1315399 Pella, S. (2015). Pedagogical reasoning and action: Affordances of practice-based
Dudley, P. (2013). Teacher learning in lesson study: What interaction-level teacher professional development. Teacher Education Quarterly, 42(3), 81e101.
discourse analysis revealed about how teachers utilized imagination, tacit http://www.jstor.org/stable/teaceducquar.42.3.81.
knowledge of teaching and fresh evidence of pupils learning, to develop Pluye, P., Robert, E., Cargo, M., Bartlett, G., O’cathain, A., Griffiths, F., &
practice knowledge and so enhance their pupils' learning. Teaching and Teacher Rousseau, M. C. (2011). Proposal: A mixed methods appraisal tool for systematic
Education, 34(1), 107e121. https://doi.org/10.1016/j.tate.2013.04.006 mixed studies reviews. http://mixedmethodsappraisaltoolpublic.pbworks.com.
Dudley, P., Xu, H., Vermunt, J. D., & Lang, J. (2019). Empirical evidence of the impact Puchner, L. D., & Taylor, A. R. (2006). Lesson study, collaboration and teacher effi-
of lesson study on students' achievement, teachers' professional learning and cacy: Stories from two school-based math lesson study groups. Teaching and
on institutional and system evolution. European Journal of Education, 54(2), Teacher Education, 22(7), 922e934. https://doi.org/10.1016/j.tate.2006.04.011
202e217. https://doi.org/10.1111/ejed.12337 Rock, T. C., & Wilson, C. (2005). Improving teaching through lesson study. Teacher
Gee, D., & Whaley, J. (2016). Learning together : Practice-centred professional Education Quarterly, 32(1), 77e92. https://www.jstor.org/stable/23478690.
development to enhance mathematics instruction. Mathematics Teacher Ronfeldt, M., Farmer, S. O., McQueen, K., & Grissom, J. A. (2015). Teacher

12
A.E. Benedict, J. Williams, M.T. Brownell et al. Teaching and Teacher Education 122 (2023) 103951

collaboration in instructional teams and student achievement. American Stewart, R. A., & Brendefur, J. L. (2005). Fusing lesson study and authentic
Educational Research Journal, 52(3), 475e514. https://doi.org/10.3102/ achievement: A model for teacher collaboration. Phi Delta Kappan, 86(9),
0002831215585562 681e687. https://doi.org/10.1177/003172170508600912
Saito, E. (2012). Key issues of lesson study in Japan and the United States: Literature Stigler, J. W., & Hiebert, J. (1999). The teaching gap: Best ideas from the world's
review. Professional Development in Education, 38(5), 777e789. https://doi.org/ teachers for improving education in the classroom. The Free Press.
10.1080/19415257.2012.668857 Suh, J., & Seshaiyer, P. (2015). Examining teachers' understanding of the mathe-
Schipper, T., Goei, S. L., de Vries, S., & van Veen, K. (2017). Professional growth in matical learning progression through vertical articulation during lesson study.
adaptive teaching competence as a result of Lesson Study. Teaching and Teacher Journal of Mathematics Teacher Education, 18(3), 207e229. https://doi.org/
Education, 68, 289e303. https://doi.org/10.1016/j.tate.2017.09.015 10.1007/s10857-014-9282-7
Seleznyov, S. (2018). Lesson study: An exploration of its translation beyond Japan. Thurlings, M., & den Brok, P. (2017). Learning outcomes of teacher professional
International Journal for Lesson and Learning Studies, 7(3), 217e229. https:// development activities: A meta-study. Educational Review, 69(5), 554e576.
doi.org/10.1108/IJLLS-04-2018-0020 https://doi.org/10.1080/13803611.2018.1509719
Seleznyov, S. (2019). Lesson study beyond Japan. Evaluating impact, International Willems, I., & Van den Bossche, P. (2019). Lesson study effectiveness for teachers'
Journal for Lesson and Learning Studies, 8(1), 2e18. https://doi.org/10.1108/IJLLS- professional learning: A best evidence synthesis. International Journal for Lesson
09-2018-0061 and Learning Studies, 8(4), 1e19. https://doi.org/10.1108/IJLLS-04-2019-0031
Sherin, M. G., Linsenmeier, K. A., & van Es, E. A. (2009). Selecting video clips to Yoshida, M. (2012). Mathematics lesson study in the United States. International
promote mathematics teachers' discussion of student thinking. Journal of Journal for Lesson and Learning Studies, 1(2), 140e152. https://doi.org/10.1108/
Teacher Education, 60(3), 213e230. 20468251211224181

13