# Kazemi & Hubbard (2008)

New Directions for the Design and Study of Professional Development: Attending to the Coevolution of Teachers' Participation Across Contexts

## Abstract

Research on professional development (PD) typically focuses on what teachers learn as a result of their participation in PD. Questions are framed unidirectionally: To what extent does participation in PD affect teachers' classroom practice? The authors challenge this unidirectional conceptualization of teacher learning and instead argue for understanding the multidirectional influences between teachers' participation across the PD and classroom settings. Drawing on research in mathematics education, they argue that researchers should examine what teachers are learning during and after PD, looking at the coevolution of participation between classroom practice and PD. To advance studies of and designs for PD, the authors' perspective leads them to argue for the importance of better understanding how teachers come to make sense of primary artifacts, depictions, and enactments in and through PD.

• Attending to "Knowledge" and "Knowing" Across Contexts in Research and Development Programs
• Practice-Based Professional Education (PBPE)
• What Role Does Teachers' Instructional Context Play in the Design of PD?
• From Unidirectional Analyses to Coevolution of Participation in PD and Classroom Practice
• Illustrating the Coevolution Between PD and Practice
• Adaptive PD: Teachers Studying Their Own Work
• Commercially Available PD: Teachers Studying Written Cases
• Implications for Studying and Designing PD
• Understanding and Eliciting the Diversity of Teachers' Experimentation and Incorporating Depictions of That Work in PD
• Examining the Situated Nature of Primary Artifacts
• Exploring How Enactments of Routine Activities Can Support the Generation of New Knowledge and Ways of Knowing
• Conclusion

## Summary

Because professional development (PD) has an uneven effect on teachers, Kazemi & Hubbard sought to rethink how we study and understand PD by focusing how teacher learning operates multidirectionally across both PD and classroom contexts. There is relatively little evidence with which to understand how PD acts to support teaching (Borko, 2004; Lewis, Perry, & Murata, 2006; Wilson & Berne, 1999) despite calls to focus PD on inquiry and experimentation (Ball, 1997; Cochran-Smith & Lytle, 1999; Franke, Carpenter, Levi, & Fennema, 2001; Lampert & Ball, 1998). Using the concept of activity settings (Wertsch, 1985) to identify the different contexts, such as PD, Kazemi and Hubbard distinguish between the knowledge teachers possess and the knowing they put into action (Cook & Brown, 1999). While knowledge is seen as essential, it alone does not explain a teacher's actions and the quality of their teaching. In addition to knowledge of content, of students, and of teaching (Ball & Bass, 2000; Shulman, 1986), teachers need to demonstrate their ability to teach in their specific classroom context. Professional development needs to consider how those contexts affect both the effects of PD on teaching and how teachers learn in PD.

Ball & Cohen (1999) characterized practice-based professional education (PBPE) as a collective professional effort grounded in the activities, questions, analysis, and criticisms of teaching. Practice-based teacher education should simulate classroom activity in ways accessible to teachers. While some commercially available PD programs are meant to generalize to many teachers, adaptive PD (Horn, 2008) attends to teachers' classrooms and challenges. To bridge the PD activity setting and the classroom activity setting, Kazemi and Hubbard use the idea of boundary objects (Star & Griesemer, 1989) and brokering, the negotiations of meaning across settings (Wenger, 1998).

Kazemi and Hubbard's main question was, "What role does the classroom play in the design of PD?" This was supported by three subquestions (pp. 430-431):

1. Whose practice is represented in the PD context and with what kinds of records?
2. What aspects (e.g., persons, tools) of PD appear in the teachers' classroom contexts?
3. When and how are these PD initiatives accounting for individual and collective teacher learning?

Most commercial PD is designed to help teachers learn about the practices of other teachers (e.g., Dolk, Fosnot, Cameron, Teig, & Hersch, 2005; Schifter, Bastable, & Russell, 1999), while some on-site PD incorporates the practices of the participating teachers (e.g., Borko, Jacobs, Eiteljorg, & Pittman, 2008; Cobb, Dean, & Zhao, 2006; Jacobs, Franke, Carpenter, Levi, & Battey, 2007; Kazemi & Franke, 2004; Lewis, Perry, Murata, 2006; Sherin & Han, 2004) or uses a hybrid approach (Schifter, Bastable, & Russell, 1999). PD uses many records of practice, such as student work (e.g., Cobb, Dean, & Zhao, 2006; Kazemi & Franke, 2004), written cases (e.g., Schifter, Bastable, & Russell, 1999), lesson plans (Lewis, Perry, Murata, 2006), and video clips (e.g., Borko, Jacobs, Eiteljorg, & Pittman, 2008; Dolk, Fosnot, Cameron, Teig, & Hersch, 2005; Sherin & Han, 2004). In the PD described by Cobb, Dean, & Zhao (2006) student work was described as a boundary object which struggled to connect PD to practice, as the facilitators saw PD as evidence upon which to design future instruction while teachers saw student work as a product to be evaluated. Other PD has claimed more success using student work as a boundary object (e.g. Crespo, 2002; Kazemi & Franke, 2004).

In the PD program Developing Mathematical Ideas (DMI) the assignments include opportunities for teachers to take what they've learned, try applying it in their classrooms, then report back in the PD. Some PD has attempted to create boundary objects that move across contexts, such as index cards with problems (Jacobs, Franke, Carpenter, Levi, & Battey, 2007), math tasks (Borko, Jacobs, Eiteljorg, & Pittman, 2008), or lesson plans (Lewis, Perry, Murata, 2006) developed in PD then utilized in the classroom. Kazemi and Hubbard expressed a concern with these approaches because while the boundary object may traverse from PD to the classroom, the role of the professional developer in the classroom is either unclear or nonexistant. The lack of follow-up or sustained involvement from professional developers leaves gaps in our understanding about individual and collective teacher learning. Teachers in DMI write reflections as artifacts of their learning, and in Cognitively Guided Instruction a tool was used to assess changes in teachers' beliefs and practices (Franke, Fennema, & Carpenter, 1997). Other PD materials were developed to mainly support the PD itself (e.g., Boaler & Humphreys, 2005; Seago, Mumme, & Branca, 2004; Smith, Silver, Stein, Boston, Henningsen, Hillen, 2005). Adaptive PD has focused more on evolving norms (e.g., Cobb, Dean, & Zhao, 2006; Kazemi & Franke, 2004; Sherin & Han, 2004) but is typically focused on the duration of the PD and not long-term changes in practice. As some have recommended focusing adaptive or design-based PD on a collective learning trajectory (Cobb, Dean, & Zhao, 2006; Cobb, Stephan, McClain, & Gravemeijer, 2001), Kazemi and Hubbard sought to support those efforts by examining how such collective trajectories and paths of individual teacher learners, especially after the PD has concluded.

### Coevolution of Participation in PD and Practice

Instead of simply evaluating whether PD has an effect on practice (Lewis, Perry, & Murata, 2006), Kazeimi and Hubbard suggest using a transformation of participation view of teacher engagement in interpersonal and cultural-historical activities (Lave, 1996; Rogoff, 1997). Prior work by Kazemi (1999) and Kazemi & Franke (2004) showed how teacher experimentation affected their engagement in PD, and the PD in turn affected their experimentation in classrooms, and Kazemi and Hubbard warn "to be careful not to simplistically talk about PD as the realm of knowledge and classroom practice as the realm of knowing" (p. 432). To illustrate, Kazemi and Hubbard use two examples of PD, one adaptive and the other commercial.

Kazemi and Hubbard's adaptive PD example comes from Franke, Kazemi, Shih, Biagetti, & Battey (2005). Teachers brought student work to the PD to discuss and researchers tracked an improvement over time in teachers' attention to student thinking. This attention led to new instructional designs that promoted computational fluency, the analysis of which was reported in Franke, Kazemi, Shih, Biagetti, & Battey (2005), Kazemi (1999), and Kazemi & Franke (2004). While some teachers appeared to view the PD as separate from classroom activity, some used ideas from the PD in their classroom and a third group of teachers used the PD as a forum for collaboration around problems they encountered in their practice.

The commercial PD example comes from DMI and a five-year study of several hundred elementary teachers (Schifter, Bastable, & Russell, 1999). DMI seminars focused on written and video cases of student activity that would be analyzed by the teachers. Ten teachers were visited in their classrooms by researchers studying the effects of the PD (Kazemi, Lenges, Jackson, & Stimpson, 2003). Teachers reported that their DMI experience helped them recognize more kinds of student thinking and the demands presented by various mathematical tasks. The researchers recognized that while the DMI seminars were valuable for initiating this shift in practice, the seminars were not responsive to the specific contexts and experiences of the teachers. Additionally, a variety of observed practices and teaching styles was observed by the researchers, leading to questions about what, exactly, the PD had done or not done for participating teachers. The realities of the classroom sometimes conflicted with PD goals (Cobb & Smith, 2007; Coburn, 2004), conflicts the DMI seminars were not always able to accommodate in their designs for teacher learning.

Kazemi and Hubbard recognize the need for both the design and study of PD to be aware of the diverse contexts of practice and to bring teacher experiences back into the PD. To do this, they focus on three types of records of teacher practice: depictions, artifacts, and enactments. Each of these can reveal shifts in both individual and collective teacher learning that help relate teachers' practices to their PD experiences. Kazemi and Hubbard start by recommending that researchers identify a focus of the PD that they can observe in teacher practice, then plan to reason with that practice when teachers return to the PD. The ways teacher learning are depicted across time need some standardization across the field of teacher education, perhaps beginning with research descriptions suggested by Staples (2008) AERA.

In addition to bringing teachers' real-world problems from the classrooms into PD, Kazemi and Hubbard also suggest that depictions of practice could be used in PD to push teacher thinking. This will require skill on the part of PD facilitators, and researchers studying teacher learning could attend to the "face" and "transparency" of practice (Little, 2002) as well as norms of interaction within teacher communities (Little & Horn, 2007). Horn (2010) tracked teacher participation in PD as replays of past experiences and rehearsals of possible future experiences. Positive signs of teacher learning in replays and rehearsals were teachers' requests for more detailed information and active connection-making between a specific replay and a more generalized rehearsal.

Kazemi and Hubbard argue that classroom artifacts should be thought of as boundary objects and used to negotiate meaning across PD and classroom contexts. If the classroom artifact is student work, teachers need support in evaluating the range of thinking and strategies seen in the work, instead of viewing the artifact as something to be graded. Researchers should evaluate teacher learning as they interact with the artifacts, both in the context of the classroom as well as PD.

Finally, Kazemi and Hubbard agree with Grossman & McDonald (2008) and Grossman, Compton, Igra, Ronfeldt, Shahan, & Williamson (2009) in the assertion that pedagogies of investigation need to be coupled with pedagogies of enactment. Enactment here refers not to actual enactment with students, but a practice-focused rehearsal that considers varying complexity of the possible teaching experience. Kazemi and Hubbard specifically recommend an enactment they call routine instructional activities and are informed by Lampert (2005) and Graziani (2005). These routine instructional activities represent a discrete teaching activity that can be studied and practiced, but larger than similar routines described by Leinhardt & Steele (2005). The activity serves as a focus for a group of teachers in PD and has enough predictable features to limit improvisation in each enactment. Kazemi and Hubbard stress that the activities are structured, not scripted, and do not eliminate the need for professional judgment. The routines can be used like a recipe (Schoenfeld, 1998) that gets refined with practice. Exactly how these routines are defined and introduced in PD requires more work, but Leinhardt & Steele (2005) give a promising start. While not as simple as "practice makes perfect", Kazemi and Hubbard sees the kinds of knowledge needed for "disciplined improvisation" (Sawyer, 2004).

## Corrolary

APA
Kazemi, E., & Hubbard, A. (2008). New directions for the design and study of professional development: Attending to the coevolution of teachers' participation across contexts. Journal of Teacher Education, 59(5), 428–441. doi:10.1177/0022487108324330
BibTeX
@article{Kazemi2008,
author = {Kazemi, Elham and Hubbard, Amanda},
doi = {10.1177/0022487108324330},
journal = {Journal of Teacher Education},
number = {5},
pages = {428--441},
title = {{New directions for the design and study of professional development: Attending to the coevolution of teachers' participation across contexts}},
url = {http://jte.sagepub.com/content/59/5/428.short},
volume = {59},
year = {2008}
}