The Evolution of Promising Practices for University Supervisors of Candidates in Yearlong Clinical Experiences

This paper describes the exploration of a team of university supervisors and faculty who were charged with identifying effective practices for university supervisors who evaluated teacher candidates enrolled in yearlong co-taught clinical experiences. The paper discusses the need for greater collaboration in reflective dialogue among the candidate, the collaborating teacher and the supervisors. In addition, the paper discusses the exploration of GROW, a candidate goal-setting structure, as well as 3-way conversations and video-elicited reflections as promising practices for university supervisors. Note: This paper summarizes the content of the co-authors’ presentation at the annual meeting of the The Renaissance Group in San Antonio, Texas, in October of 2015. 28	   T. Strieker, L. Langub & M. Wright Introduction Over the past five years, calls for improvements in teacher effectiveness (Lewin, 2011) have resulted in changes to the accreditation standards (Council for the Accreditation of Educator Preparation (CAEP), 2015), particularly in terms of restructuring clinical experiences by means of school-university partnerships. The most recent draft of CAEP Standard 2.1 requires school-university partners to establish “....mutually agreeable expectations for candidate entry, preparation and exit; ensure that theory and practice are linked; maintain coherence across clinical and academic components of preparation; and shared accountability for candidate outcomes” (CAEP, 2015, p.1). At Kennesaw State University, members of the faculty representing various teacher education programs (e.g., English Education, Secondary and Middle Grades Education, Elementary and Early Childhood Education, and Educational Leadership) formed an interdisciplinary research team to examine the effectiveness of new approaches to yearlong clinical experiences (Gray, Stockdale, & Williams, 2011; Williams, Gray, & Monti, 2010); preservice co-teaching (Heckert, Strieker, & Shaheen, 2012) and instructional coaching (Strieker, Shaheen, Hubbard, Digiovanni, & Lim, 2014). These approaches were aligned with the common variables of highly effective teacher education programs reported by Darling-Hammond (2000), including a common vision of teaching and learning, carefully crafted field experiences, theorization and justification of practice, focus upon the needs of the students, reflection on active pedagogy, engagement of collaborating teachers as lifelong learners, and collaboration with colleagues. While the literature provides definitions of various forms of effective co-teaching during clinical experiences, there is limited research on the pedagogy and practices of university supervisors who evaluate the performance of candidates during their co-taught experiences. Thus, the purpose of our exploration is to explore the pedagogical practices of university supervisors situated in a collegial and reflective model of supervision, developed by, and for, teacher educators and university supervisors who support teacher candidates enrolled in a yearlong, co-taught, clinical experience. Our Journey Since 2008 our university has employed a continuous cycle of program design, implementation, evaluation and research, and revision and refinement by university faculty and school leaders. The redesign of clinical experiences, funded by our Teacher Quality Partnership (TQP) grant (2013, 2014), occurred in three stages (Figure 1). Promising Practices for University Supervisors 29 Stage 1: School-University Partnerships & Yearlong Clinical Experiences Since 2008, school-university teams have developed various levels of partnership agreements that determine the parameters of our yearlong clinical experiences as well as the roles and responsibilities for collaborating teachers and candidates. For example, faculty in Elementary Education established memorandums of understanding (MOUs) with elementary schools based upon specific themes of STEM, global education, and/or social justice. Similarly, faculty in English Education and Science Education formed a variety of partnerships with secondary schools to enlist their former graduates of their programs to serve as collaborating teachers. Later, members of our university faculty, led by our Dean, co-authored successful application for funding through a Teacher Quality Partnership (TQP) (U.S. Department of Education, Teacher Quality Partnership Grant, 2009) with a neighboring school district. The overarching goal of this grant was to develop an Urban Education (UE) option for initial certification that would ultimately lead to improved K-12 student achievement in high-need schools. The funds from this grant were used to support faculty who engaged in the transformation of clinical experiences over the next five years. Stage 2: Pre-service Co-teaching and Co-teaching Coaches Pre-service Co-teaching. In 2011, teams of faculty and administrators at the university and the local schools began to design, develop and research our new approach to pre-service coteaching for candidates enrolled in yearlong clinical experiences. Like other university-based coteaching models (Bacharach, Heck, & Dahlberg, 2010; Badaili & Titus, 2012), our approach allowed the certified teacher to remain in the classroom with the candidate throughout the placement. Thus, both the collaborating teacher and the teacher candidate served as an integral part of planning, instruction, assessment and reflection. Early on, the focus of our research was to better understand how co-teaching in preservice teacher education was different from traditional co-teaching, conducted by two certified teachers (Friend, 1995). In the end, while our new approach was informed by Friend’s original 30	   T. Strieker, L. Langub & M. Wright model, it was also heavily influenced by the work of Roth and Tobin (2012) who researched preservice co-teaching in advanced high school science classes and described co-teaching as copraxis (or a shared experience) that ultimately fosters co-generative dialogue through reflection in practice and reflection on practice. In 2012, we ultimately named our model, Pre-service Coteaching: A Mentor-Model Approach (Center for Educational Placements and Partnerships, 2012-2015). Also in 2012, we reported the benefits of pre-service co-teaching as vehicles for: (a) shifting the power in the classroom from the classroom teacher to the candidate, (b) supporting differentiated instruction, (c) conducting classroom management, and (d) increasing student engagement and learning (Heckert, Strieker, & Shaheen, 2012). Finally, in 2013, we explored ways to align our emerging framework of pre-service co-teaching with that of edTPA; and ultimately, to intentionally support our candidates in demonstrating mastery of the competencies on assessed edTPA (Strieker, Shaheen, Hubbard, Digiovanni, & Lim, 2014). Co-teaching Coaches. From 2011-2012, we also explored ways to situate a specialized form of instructional coaching (coaching co-teaching) to support our candidates and collaborating teachers during their clinical experiences. At that point, all members of our cadre of co-teaching coaches were faculty or former teachers and administrators who were recommended by our local district partners with formal MOUs. Our co-teaching coaches were explicitly employed to offer job-embedded professional development to our candidates and their collaborating teachers. Our co-teaching coaches never evaluated candidate performance. Our EPP did employ university supervisors who were charged with evaluating the teacher candidates based upon a series of observations the supervisors conducted each semester in the yearlong placement. With funding through TQP, our approach to coaching co-teaching was implemented and researched for approximately three years. During that time, we established that our goal-setting framework (adapted from Whitmore, 2002) was effective in guiding our candidates in establishing their personal-improvement goals related to differentiated instruction, classroom and behavior management, and improvement of student learning and/or engagement (Strieker, et al., 2014). Our approach to candidate goal-setting was intentionally designed to support the candidate’s development of the metacognitive skills inherent in reflective practice, theorization of practice, and justification of data-based decision making. In large part, this approach was part of our attempt to align our program with the new CAEP standards and to provide opportunities for our candidates to demonstrate their skills on state and national performance assessments, like edTPA (Figure 2). Promising Practices for University Supervisors 31 Figure 2. GROW Framework for Establishing Candidate Goals Stage 3: Emergence of Developmental Supervision After several years of study, we realized that while the methods used by coaches were extremely valuable, coaching was not sustainable at our university for a number of reasons. First of all, the logistics of deploying cadres of university supervisors and co-teaching coaches was confusing and difficult to manage. In some cases, the departments were not able to find enough qualified people to serve in both roles, so the same individuals served as coaches in some schools and as supervisors in others, leading to role confusion on the part of the new coaches as well as the administrators in the schools, the collaborating teachers, and the candidates. Second, without external funding the university had difficulty finding resources for both coaches and university supervisors. We discussed blending the two functions, but felt that there were crucial differences between the two that precluded combining coaching and supervision (Tschannen-Moran, & Tschannen-Moran, 2011; Tschannen-Moran, & Woolfolk-Hoy, 2001). In the end, our team decided to protect the integrity of coaching and supervision by selecting one approach. Given the difficulties in the logistics and in resourcing coa


Introduction
Over the past five years, calls for improvements in teacher effectiveness (Lewin, 2011) have resulted in changes to the accreditation standards (Council for the Accreditation of Educator Preparation (CAEP), 2015), particularly in terms of restructuring clinical experiences by means of school-university partnerships.The most recent draft of CAEP Standard 2.1 requires school-university partners to establish "….mutually agreeable expectations for candidate entry, preparation and exit; ensure that theory and practice are linked; maintain coherence across clinical and academic components of preparation; and shared accountability for candidate outcomes" (CAEP, 2015, p.1).
At Kennesaw State University, members of the faculty representing various teacher education programs (e.g., English Education, Secondary and Middle Grades Education, Elementary and Early Childhood Education, and Educational Leadership) formed an interdisciplinary research team to examine the effectiveness of new approaches to yearlong clinical experiences (Gray, Stockdale, & Williams, 2011;Williams, Gray, & Monti, 2010); preservice co-teaching (Heckert, Strieker, & Shaheen, 2012) and instructional coaching (Strieker, Shaheen, Hubbard, Digiovanni, & Lim, 2014).These approaches were aligned with the common variables of highly effective teacher education programs reported by Darling-Hammond (2000), including a common vision of teaching and learning, carefully crafted field experiences, theorization and justification of practice, focus upon the needs of the students, reflection on active pedagogy, engagement of collaborating teachers as lifelong learners, and collaboration with colleagues.
While the literature provides definitions of various forms of effective co-teaching during clinical experiences, there is limited research on the pedagogy and practices of university supervisors who evaluate the performance of candidates during their co-taught experiences.Thus, the purpose of our exploration is to explore the pedagogical practices of university supervisors situated in a collegial and reflective model of supervision, developed by, and for, teacher educators and university supervisors who support teacher candidates enrolled in a yearlong, co-taught, clinical experience.

Our Journey
Since 2008 our university has employed a continuous cycle of program design, implementation, evaluation and research, and revision and refinement by university faculty and school leaders.The redesign of clinical experiences, funded by our Teacher Quality Partnership (TQP) grant (2013,2014), occurred in three stages (Figure 1).

Stage 1: School-University Partnerships & Yearlong Clinical Experiences
Since 2008, school-university teams have developed various levels of partnership agreements that determine the parameters of our yearlong clinical experiences as well as the roles and responsibilities for collaborating teachers and candidates.For example, faculty in Elementary Education established memorandums of understanding (MOUs) with elementary schools based upon specific themes of STEM, global education, and/or social justice.Similarly, faculty in English Education and Science Education formed a variety of partnerships with secondary schools to enlist their former graduates of their programs to serve as collaborating teachers.Later, members of our university faculty, led by our Dean, co-authored successful application for funding through a Teacher Quality Partnership (TQP) (U.S.Department of Education, Teacher Quality Partnership Grant, 2009) with a neighboring school district.The overarching goal of this grant was to develop an Urban Education (UE) option for initial certification that would ultimately lead to improved K-12 student achievement in high-need schools.The funds from this grant were used to support faculty who engaged in the transformation of clinical experiences over the next five years.

Stage 2: Pre-service Co-teaching and Co-teaching Coaches
Pre-service Co-teaching.In 2011, teams of faculty and administrators at the university and the local schools began to design, develop and research our new approach to pre-service coteaching for candidates enrolled in yearlong clinical experiences.Like other university-based coteaching models (Bacharach, Heck, & Dahlberg, 2010;Badaili & Titus, 2012), our approach allowed the certified teacher to remain in the classroom with the candidate throughout the placement.Thus, both the collaborating teacher and the teacher candidate served as an integral part of planning, instruction, assessment and reflection.
Early on, the focus of our research was to better understand how co-teaching in preservice teacher education was different from traditional co-teaching, conducted by two certified teachers (Friend, 1995).In the end, while our new approach was informed by Friend's original model, it was also heavily influenced by the work of Roth and Tobin (2012) who researched preservice co-teaching in advanced high school science classes and described co-teaching as copraxis (or a shared experience) that ultimately fosters co-generative dialogue through reflection in practice and reflection on practice.In 2012, we ultimately named our model, Pre-service Coteaching: A Mentor-Model Approach (Center for Educational Placements andPartnerships, 2012-2015).Also in 2012, we reported the benefits of pre-service co-teaching as vehicles for: (a) shifting the power in the classroom from the classroom teacher to the candidate, (b) supporting differentiated instruction, (c) conducting classroom management, and (d) increasing student engagement and learning (Heckert, Strieker, & Shaheen, 2012).Finally, in 2013, we explored ways to align our emerging framework of pre-service co-teaching with that of edTPA; and ultimately, to intentionally support our candidates in demonstrating mastery of the competencies on assessed edTPA (Strieker, Shaheen, Hubbard, Digiovanni, & Lim, 2014).
Co-teaching Coaches.From 2011-2012, we also explored ways to situate a specialized form of instructional coaching (coaching co-teaching) to support our candidates and collaborating teachers during their clinical experiences.At that point, all members of our cadre of co-teaching coaches were faculty or former teachers and administrators who were recommended by our local district partners with formal MOUs.Our co-teaching coaches were explicitly employed to offer job-embedded professional development to our candidates and their collaborating teachers.Our co-teaching coaches never evaluated candidate performance.Our EPP did employ university supervisors who were charged with evaluating the teacher candidates based upon a series of observations the supervisors conducted each semester in the yearlong placement.
With funding through TQP, our approach to coaching co-teaching was implemented and researched for approximately three years.During that time, we established that our goal-setting framework (adapted from Whitmore, 2002) was effective in guiding our candidates in establishing their personal-improvement goals related to differentiated instruction, classroom and behavior management, and improvement of student learning and/or engagement (Strieker, et al., 2014).Our approach to candidate goal-setting was intentionally designed to support the candidate's development of the metacognitive skills inherent in reflective practice, theorization of practice, and justification of data-based decision making.In large part, this approach was part of our attempt to align our program with the new CAEP standards and to provide opportunities for our candidates to demonstrate their skills on state and national performance assessments, like edTPA (Figure 2).

Figure 2. GROW Framework for Establishing Candidate Goals Stage 3: Emergence of Developmental Supervision
After several years of study, we realized that while the methods used by coaches were extremely valuable, coaching was not sustainable at our university for a number of reasons.First of all, the logistics of deploying cadres of university supervisors and co-teaching coaches was confusing and difficult to manage.In some cases, the departments were not able to find enough qualified people to serve in both roles, so the same individuals served as coaches in some schools and as supervisors in others, leading to role confusion on the part of the new coaches as well as the administrators in the schools, the collaborating teachers, and the candidates.Second, without external funding the university had difficulty finding resources for both coaches and university supervisors.We discussed blending the two functions, but felt that there were crucial differences between the two that precluded combining coaching and supervision (Tschannen-Moran, & Tschannen-Moran, 2011; Tschannen-Moran, & Woolfolk-Hoy, 2001).In the end, our team decided to protect the integrity of coaching and supervision by selecting one approach.Given the difficulties in the logistics and in resourcing coaching, and the fact that our institution is accountable for the state and national assessment of the effectiveness of our candidates, our team prioritized supervision.Once that decision was made, our research team shifted its focus to reforming our traditional model of clinical supervision to better meet the needs of candidates enrolled in co-taught, yearlong clinical experiences.To that end, a task force, comprised of faculty and former school administrators, was convened to explore the potential of Developmental SuperVision (Glickman, 1981; Glickman and the Association for Supervision and Curriculum Development, 2002;Glickman, Gordan-Ross, & Ross, 2014).
According to Glickman (1981), Developmental Supervision is a collegial, collaborative and developmental approach to clinical supervision that has evolved over the past 35 years into a highly effective, democratic approach to supervising teachers in P-12 schools (Glickman et al., 2014).Developmental Supervision provides supervisors with a model for adjusting his or her communication and behavioral approaches (ranging from directive to non-directive) to meet the developmental needs of the teacher.Based upon this understanding, our task force felt that Developmental Supervision held promise for implementation in teacher education programs engaged in implementing new models of clinical experiences that incorporated yearlong placements and co-taught instruction.
The increased interest in Developmental Supervision was based upon the perception that candidates need a collaborative and developmental approach to supervision as they develop into mature, effective teachers.The need for collaborative approaches was based, in large part, upon our experiences in moving to yearlong clinical placements, where teacher candidates and collaborating teachers co-taught in the same classrooms for an entire year, and the candidates were supported and evaluated by the same supervisor throughout that year.Given the expanded length of the placements, we felt that all candidates needed a developmental approach to support the trajectory of their growth throughout the yearlong clinical.Developmental approaches appeared to be particularly important for traditional candidates who were in the early stages of adult development and in need of supervisors who were able to adjust their communication and behavioral approaches to support the candidates' developmental needs.
Our move to a model of pre-service co-teaching also seemed to complicate the supervision and evaluation of the candidates.First of all, it appeared that the very nature of our co-teaching cycle, which included co-planning, co-instruction, co-assessment and co-reflection (Center for Educational Placement and Partnerships (2015) created difficulties for supervisors using our original model of supervision.While our supervisors were charged with evaluating the performance of the candidate, they were not charged with evaluating the collaborating teacher who co-taught with the candidate.Furthermore, because the candidate and collaborating teachers co-taught throughout the day, nearly every day, the collaborating teacher knew as much (or more) about the candidate's performance than the supervisor did.Although Bullough & Draper reported in 2004 that triangulation was common among supervisors, collaborating teachers and candidates, we felt a growing need to create a more collaborative, if not collegial model of supervision that would provide the opportunity to access the resources and maximize the contributions of the collaborating teachers in supporting the growth and development of the candidates.
Finally, we felt that Developmental Supervision had the potential to increase the coherence of our teacher education programs by supporting our candidates as they applied the knowledge, skills and dispositions that they had learned in university coursework to their pedagogical practices in schools.Because of the collegial and collaborative nature of Developmental Supervision, we felt that it provided a model for supervisors to adjust not only their approaches, but also their roles.With Developmental Supervision, the supervisors have the flexibility to move in and out of expert and facilitator roles based upon their perceived developmental needs of the candidate.For example, if the candidate demonstrates high needs for information, the supervisor assumes the role of expert and directs the candidate, telling him or her what task to perform, when and how to perform the task and why the task is beneficial to the P-12 students.Conversely, during co-reflection on video, the supervisor has the freedom to assume a facilitation role, guiding 3-way conversations between him or herself, the collaborating teacher, and the candidate.During these sessions, the candidate has the opportunity to actively engage in a dialogue with two experienced professionals and co-generate answers to complex problems of practice.
In the end, three new pedagogical practices emerged from our first two years of exploring ways to adjust and adapt Glickman's Developmental Supervision to a university setting, including the following: • 3-way conversations among the candidate, the collaborating teacher, and the supervisor who facilitates the group; • GROW: Candidate goal-setting instrument; and • Video-elicited reflections.

Promising Practices in Supervision of Teacher Candidates in Co-Taught Placements
Because teaching is a complex activity, the evaluation of teaching and the supervision of teacher candidates are exceedingly complex.To be effective, supervisors must have a broad repertoire of teaching and assessment practices, possess excellent communication and facilitation skills, and understand a myriad of connections that can, and do, occur in the teaching and learning process between the teacher and student.In the current era of accountability, indicators of teacher effectiveness are nearly always based upon student performance and/or engagement in challenging tasks.Therefore, it is incumbent upon the supervisor to use the most efficacious methods of observing student performance.According to scholars in coaching and supervision (Knight, 2014) student work can be studied in two ways: by examining student products, such as writing samples or science projects, and by examining video recordings of the teaching and learning processes.Because our faculty had already devoted a great deal of time and attention to the analysis of student work products, our research team began to explore new and creative ways to use video to evaluate the effectiveness of our candidates, specifically in terms of student performance and/or engagement (Sewall, 2009).
During the 2013-2014 academic year, our second author, then a university supervisor, used pre-post video observations as alternatives to traditional classroom observations.During the final co-reflection on the videos, both the university supervisor and the candidate experienced a significant breakthrough.When the videos of two observations were displayed simultaneously on a split-screen, the differences in the engagement of students became obvious.One video showed merely compliance on the part of the elementary students as they worked quietly at their desks, following the examples of the candidate as she solved math problems on the board.The second video showed the same students, actively engaged, as they worked in teams to solve problems related to various math games.In the second video, the candidate and the collaborating teacher were also actively engaged as they interacted with the small groups of students to foster their problem solving.Thus, when the two video clips, viewed side by side, clearly illustrated the growth of the candidate's skills over that period of time, highlighted the development of the coteaching partnership between the candidate and the collaborating teacher, both of which seemed to increase student engagement in the lesson.Had this lesson not been recorded, it is possible that this growth and professional learning may have gone largely unnoted.
Through this real-life experience, the co-teachers as well as the university supervisor began to understand the real power of video in supporting candidates.To share this information with a broader audience, Langub (2014) engendered discussions among members of the EPP faculty, former coaches, supervisors, and administrators served as a catalyst for today's effort to increase the use of video to record teaching and learning processes.In addition to creating a record of the candidate's professional growth and development, our experimentation with video increased our understanding of how video can be used effectively to benchmark student engagement, particularly in terms of time on task, ratios of interactions, opportunities to respond, and other aspects of the teaching process that are difficult to capture during real-time instruction (Sprick, Knight, Reinke, Skyles, & Barnes, 2010).When video is used to benchmark, the candidate has the advantage of concrete examples to use in the assessment and monitoring of student engagement and learning.These approaches have the potential to assist candidates in improving the engagement of students from simply attending to a task, to compliance, to engagement in challenging tasks.
Our efforts to increase the frequency of, and improve the effectiveness of, video-elicited reflections are based upon the research of numerous scholars, including Schon (2009); Sewall (2009); Calandra & Brantley-Dias (2010); Lokey-Vega & Brantley-Dias (2006); and Knight (2014).Our experiences and understandings are consistent with Sewall (2009) in that we find (anecdotally) that our candidates' reflections on video are more consistent with those of the collaborating teachers and the supervisors.Specifically, it appears that when the candidates reflect upon their own experiences, without the benefit of video, their reflections are neither as nuanced nor as deep as reflections that are elicited from video.Video not only provides the candidate with deeper understanding, it also provides the candidate with the gift of time because the candidate can watch the video repeatedly to observe the responses of his or her students to the pedagogical and classroom management practices used during instruction.

Overall Impact and Future Directions
Over the past year, members of a professional learning community of experienced university supervisors have worked closely with the first author to improve their craft and practice as developmental supervisors.Throughout the year, these individuals expanded their repertoires of communication, behavioral, and interpersonal approaches.For example, some of our supervisors found their candidates to be somewhat immature in the fall of their senior year, and that this immaturity forced the supervisor to assume the role of an expert, with a directive approach of literally telling the candidate what to do and when to do it.However, as the candidates matured, the supervisors shifted their approaches to a more collegial role by facilitating co-reflection and co-generative dialogue with the collaborating teacher.Toward the end of the first year spent in developmental approaches, our supervisors reported that they no longer needed to use directive approaches and that most of the time they facilitated three-way conversations with the candidate and collaborating teachers.The supervisors also reported that they were very comfortable using non-directive approaches, letting the candidate guide his or her own professional learning by means of specific goal-setting and progress-monitoring of selfimprovement.During that same time, the supervisors also worked closely with the first author as well as with the candidates and collaborating teachers, to revise and refine their facilitation of candidate goal-setting, progress monitoring, video observation, and co-reflection.
Based upon our experiences in developing and piloting our EPP's approach to Developmental Supervision, we feel that this approach to supervision shows great promise for supporting our B-12 candidates in this era of high accountability and assessment.Through Developmental Supervision, our candidates engage in ongoing goal setting and progressmonitoring, collaboration and co-teaching, co-generative problem solving, and video-elicited coreflection by the teacher candidate, the collaborating teacher and the university supervisor.
In the summer of 2015, our university supervisors (Wright, Klinger-Maffet, Peterson, Thompson, & Langub) were invited to relate their experiences with the practices described in this paper to a group of current and potential collaborating teachers.Specifically, they stated the following benefits of our model: 1. Goal-setting improves the candidates' focus and increases professional development in areas identified by the candidate, creating greater commitment to his or her personal growth.The university supervisor and the collaborating teacher can then support the candidate in these specific areas, moving to new goals as the year progresses. 2.
Collaborative approaches, particularly 3-way conversations, create a forum for co-reflection and cooperative dialogue.The expertise of the classroom teacher and the university supervisor is developed along with the candidate's, in a mutually supportive environment.The candidate has an equal voice.
3. The use of video and virtual observations provides a record of the candidate's growth and a platform for reflective dialogue on concrete, observable behaviors.The evaluations of the co-teacher and the university supervisor are thus more objective and beneficial to the candidate.Based upon our supervisors' positive perceptions of the new approaches, our future plan is to continue to engage our developmental supervisors in monthly sessions of their professional learning communities where the experienced supervisors will work closely with new ones to: (a) adjust their communication and behavioral approaches to the developmental needs of the candidates; (b) improve the practice of video observation; (c) facilitate candidate goal-setting and progress monitoring; as well as (d) increase student engagement and learning.